Kinetic analysis of horseradish peroxidase ‘‘wiring’’ in redoxpolyelectrolyte–peroxidase multilayer assemblies

Victor Rosca 1, Ionel Catalin Popescu *

Department of Physical Chemistry, ‘‘Babes-Bolyai’’ University, Arany Janos 11, RO-3400 Cluj-Napoca, Romania

Received 30 September 2002; accepted 7 October 2002

Abstract

A hydrogen peroxide sensitive interface based on a self-deposited redox polyelectrolyte–peroxidase assembly was used as model

system for investigation of the electron transfer between chemically modified horseradish peroxidase and the redox active centers of

poly[(vinylpyridine)OsðbpyÞ2Cl] redox polymer. The assembly was built up by sequential adsorption of the redox polymer and theperoxidase on gold electrodes modified with a self-assembled monolayer of 3-mercapto-1-propane sulfonic acid. Making use of a

kinetic model, formulated for steady-state conditions, and of the rotating disc electrode technique it was possible to estimate: (i) the

second order rate constant of reaction between the modified peroxidase and hydrogen peroxide (2:74� 105 M�1 s�1) and (ii) theapparent turnover number of the oxidized enzyme regeneration by the osmium redox center (2:14 s�1), i.e., two parameters that

reflect the efficiency of the electrical communication within the investigated interface.

� 2002 Elsevier Science B.V. All rights reserved.

Keywords: Multilayer self-assembly; Redox polymer; Horseradish peroxidase; Hydrogen peroxide; Reagentless amperometric biosensors; Rate

constants

1. Introduction

Amperometric biosensors based on peroxidase-modi-

fied electrodes have been developed for detection of

hydrogen peroxide and organic hydroperoxides, as well

as phenols, aromatic amines, potassium cyanide, etc., i.e.,

of molecules that are substrates, activators, or inhibitorsof the enzyme catalyzed reaction [1]. On the other hand,

peroxidase-modified electrodes can be used for the

immuno-detection of analytes [2–4] or can be coupled

with different hydrogen peroxide producing oxidases

resulting in sensors for many other analytes of interest

[5–8]. Horseradish peroxidase (HRP, EC 1.11.1.7.) – the

model enzyme for whole class – is the most commonly

used peroxidase for bioelectrode construction.Electrodes modified by coverage with redox or elec-

tron conducting polymers, incorporating physically or

chemically attached enzyme molecules, represent an

important and extensively studied group of peroxidase-

based electrodes [1]. Thus, electrodes based on HRP,

bound to [OsIII=IIðbpyÞ2Cl]-containing polyvinyl pyri-

dine or polyvinyl imidazole [9–11], are among the most

successful peroxidase-based modified electrodes, if such

criteria as sensitivity, lower detection limit and stabilityare considered (for comparison see Table 2 in reference

[1]).

The above-mentioned polymer materials are com-

monly referred to as redox ‘‘wire’’ or redox hydrogels.

Redox ‘‘wire’’ is a redox macromolecule designed to: (i)

physically attach the enzyme to the electrode; (ii)

strongly complex the enzyme protein; (iii) electrically

connect the enzyme�s redox center to the electrode, allthese without serious deactivation of the enzyme [12].

When an oxidoreductase is integrated in such a three-

dimensional polymer matrix, electrons are transported

via the redox polymer network between the enzyme�sredox center and electrode. Extended experimental data

on the electrical communication between the enzyme

and redox polymers is now available [9,12–15], though

Electrochemistry Communications 4 (2002) 904–911

www.elsevier.com/locate/elecom

*Corresponding author. Tel.: +40-641-938-33; fax: +40-641-908-18.

E-mail address: cpopescu@chem.ubbcluj.ro (I. Catalin Popescu).1 Current address: Schuit Institute of Catalysis, Eindhoven Univer-

sity of Technology, 5600 MB Eindhoven, The Netherlands.

1388-2481/02/$ - see front matter � 2002 Elsevier Science B.V. All rights reserved.

PII: S1388-2481(02 )00486 -1

the subject is far to be exhausted. Thus, of particularinterest are kinetic aspects of the enzymes ‘‘wiring’’ by

redox polymers [12].

Recently it was demonstrated that uniform thin

polymeric films can be assembled on a variety of sub-

strates by alternating electrostatically controlled ad-

sorption of polycations and polyanions on charged

surfaces [16]. In a short time, layer-by-layer assembly

technique proved to be an efficient strategy for thefabrication of multicomponent nanostructured films on

various solid supports [17].

Much of the recent work on these polyelectrolyte

multilayers has focused on generalizing and expanding

the technique to various combinations of charged

components, including various synthetic polyelectrolytes

[17], biopolymers, such as proteins [18,19] and enzymes

[20–29]. Most importantly, building-up of multilayerredox polyelectrolyte-oxidoreductase architectures,

based on electrostatic interactions, allows fine control of

the film thickness and the amount of the biomolecule at

the electrode surface and, from these considerations, the

technique could present a promising approach to elec-

trode coating with thick redox hydrogel films loaded

with enzyme molecules. Nevertheless, only few papers

[27,28] dealt with a kinetic analysis of these sensingstructures, in order to estimate quantitatively the effi-

ciency of the enzyme electrical ‘‘wiring’’ as well as the

main kinetic parameters involved in electrical signal

generation. The approach is essentially based on the

expression describing the dependence of the catalytic

current on the substrate concentration, when the

substrate diffusion is neglected [30,31]. Combining the

kinetic parameters for the soluble enzyme (the Micha-elis–Menten constant and the turnover number) with

those furnished by the non-linear fit of the above-men-

tioned equation to the experimental data, an estimation

of the enzyme surface concentration and the pseudo-

first-order rate constant corresponding to the enzyme/

redox polymer reaction in the investigated architecture

was possible. However, it is worthy to mention that this

approach disregards the inherent variation of the en-zyme kinetic parameters due to its immobilization on

the electrode surface.

In a previous paper [8], the potentiality of multilayer

self-deposition technique based on electrostatic interac-

tions for the enzyme bioelectrodes design was analyzed.

Thus, it was described the use of layer-by-layer self-de-

position of a positively charged poly[(vinylpyri-

dine)OsIII=IIðbpyÞ2Cl] redox polymer (RP) and someoxidoreductases on a negatively charged alkanethiolate-

modified gold electrode surface. In this way a reagentless

biosensor, incorporating a chemically modified horse-

radish peroxidase (HRPmod), for hydrogen peroxide has

been produced. The sensitivity of 58:1 mA M�1 cm�2

obtained for one RP-HRPmod pair of layers was much

lower than that of 1 A M�1 cm�2 – theoretically maxi-

mal sensitivity of a peroxidase electrode [1] and actuallythe highest sensitivity ever reported for hydrogen per-

oxide detection at HRP-based enzyme bioelectrodes [9].

A straightforward way to increase the sensitivity is fur-

ther alternate deposition of new biocatalytic and trans-

ducing layers, resulted in a multilayer assembly. This

could bring about an increase of the importance of both

electron ‘‘diffusion’’ and the substrate diffusion limita-

tions. Consequently, only a thorough kinetic analysis ofthe resulted multilayer RP-HRP assembly, as well as its

structural elements, could provide further insight into

the interface and could be of help in identification of

certain criteria for optimization of the bioelectrode re-

sponse. In this sense, a simple approach for quantitative

estimation of the rate of electron transfer between the

enzyme active center and the transducing macromole-

cule, as well as of the immobilized enzyme catalytic ac-tivity, is proposed and discussed.

2. Experimental

Sodium salt of 3-mercapto-1-propane sulfonic acid

(MPS) was purchased from Aldrich. The cationic

poly[(vinylpyridine)OsIII=IIðbpyÞ2Cl] redox polymer(RP) partially quaternised with bromoethylamine was

synthesized as described elsewhere [32]. Horseradish

peroxidase (HRP, EC 1.11.1.7, 290 U mg�1 (solid); typeVI) was supplied by Sigma. The enzyme was chemically

modified according to procedure described elsewhere [8].

The purpose of the modification was manipulation of

the enzyme charge by introducing sulfonate groups co-

valently bound to the enzyme [8]. The activity of themodified enzyme was nearly the same as that measured

before the modification, when calculated per protein

content (U mg�1). The H2O2 solutions were prepared

daily by dissolving 30% (w/w) H2O2 (Merck, Darmstadt,

Germany) in a phosphate buffer solution. All other re-

agents were of analytical grade. Aqueous solutions were

prepared with water purified with a Milli-Q system

(Millipore, Milford, USA).Electrochemical measurements were performed in a

conventional single-compartment three-electrode cell

using a computer-controlled voltammetric analyzer

AutoLab-PGSTAT10 (Eco Chemie, Utrecht, The

Netherlands). The reference electrode was a potassium

chloride saturated silver/silver chloride electrode

(AgjAgCljKClsat:), to which all the electrode potentials

are referred. A coiled platinum wire served as auxiliaryelectrode. Steady-state amperometric measurements at

different rotation speeds of the working electrode were

performed using a modulated speed rotator (model

AMS-FRX, Pine, Grove City, PA, USA).

Gold disc electrode (Ø 5 mm) was used as working

electrode throughout the experiments. The electrode was

polished with 0.3 lm a-alumina and 0.05 lm c-alumina

V. Rosca, I. Catalin Popescu / Electrochemistry Communications 4 (2002) 904–911 905

(Buehler, IL, USA) to a mirror-like finish, followed bycleaning for 5 min with ethanol and water in ultrasonic

bath. Subsequently, the electrode was etched electro-

chemically by potentiodynamic cycling (20 mV s�1 scanrate) between +0.44 and +1.5 V for 30 min in 0.1 M

sulfuric acid containing 10 mM potassium chloride.

During this time, an up to 100 nm thick layer of gold is

dissolved by chloroaurate complex formation [33]. The

real surface area was estimated from oxygen adsorptionmeasurements, according to the method proposed by

Rand and Woods [34].

The negatively charged gold surface (Au/MPS) was

prepared by immersing the electrode into 1 mM MPS

solution in ethanol for about 12 h, followed by thorough

rinse with pure ethanol and re-immersing into the de-

position solution for 1 h more. The growth of the Au/

MPS/RP/HRPmod/RP multilayer assembly was achievedby sequential deposition of two RP layers and one in-

tercalating HRPmod monolayer by alternative immersion

of the MPS monolayer covered Au electrode into cor-

responding aqueous solutions. RP was deposited for 2 h

from its 100 lg ml�1 aqueous solution at room tem-

perature while stirring vigorously. Biocatalytic layer was

obtained by self-deposition of the modified peroxidase

(HRPmod) for 2 h at 4 �C from 100 lg ml�1 enzymeaqueous solution, prepared in 0.1 M phosphate buffer

solution (pH 6.0). Between deposition steps the elec-

trode was rinsed with pure water and then was allowed

to dry at room temperature for 30 min. Before deposi-

tion of the biocatalytic layer no significant amperomet-

ric response at the working potential (0.1 V) was

observed upon addition of hydrogen peroxide.

3. Kinetic model

In Fig. 1 is depicted the idealized picture of the in-

terface in study. It is commonly accepted that in mul-

tilayer assemblies based on electrostatic interactions,

regardless of the components� nature (polyelectrolyte,

charged colloidal particles, large organic or inorganic

molecules), the deposition is restricted to formation of a

nearly single molecular layer [17]. As electrostatic forcesare perceived as the forces that control the assembly

growth [8], HRPmod adsorption ought to be limited to a

nearly single molecular layer. In order to place the

enzyme molecule into the same environment as that in

ðRP=HRPmod=RPÞn multilayer film and to eliminate or

diminish the enzyme orientation effect on the signal

generation, the RP/HRPmod/RP assembly was investi-

gated.The following reaction sequence can be proposed for

the case of mediated bioelectrocatalytic reduction of

H2O2 at Au/MPS/RP/HRPmod/RP electrode:

HRPðFe3þÞ þH2O2!k1Comp: IþH2O ð1Þ

Comp: IþOsIIðþHþÞ!k2 Comp: IIþOsIII ð2Þ

Comp: IIþOsIIðþHþÞ!k3 HRPðFe3þÞ þOsIII þH2O

ð3Þ

ðOsIIÞ1 þ ðOsIIIÞ2!ke ðOsIIIÞ1 þ ðOsIIÞ2 ð4Þ

OsIII þ e� !kS;M

OsII ð5Þ

Fig. 1. Schematic representation of the investigated interface.

906 V. Rosca, I. Catalin Popescu / Electrochemistry Communications 4 (2002) 904–911

where HRP(Fe3þ) is the ferric enzyme; Comp. I andComp. II are oxidized intermediates (the so-called

compounds I and II [35]); OsII and OsIII designate the

reduced and, respectively, oxidized forms of

[OsIII=IIðbpyÞ2Cl] complex attached to the polymer ma-

trix. Reaction (1) involves a two-electron oxidation of

the ferrihaem prosthetic group of the enzyme by H2O2.

The resulted intermediate consists of oxyferryl iron

(Fe4þ@O) and a porphyrin p cation radical [35]. As aresult of two consecutive one-electron transfers from the

electron donor substrate (OsII), the enzyme, after pass-

ing through an intermediate species (Comp. II), is re-

turned to its resting state. Reaction (4) describes

transport of the oxidizing equivalents to the electrode

surface by the self-exchange reaction (electron hopping).

Rapid electron transfer from the electrode surface to the

oxidized mediator closes the electrocatalytic cycle.In case of rotating disc electrode (RDE) technique,

H2O2 reduction current measured at the Au/MPS/RP/

HRPmod/RP electrode is a combination of the mass-

transport-limited current IL and the reaction-rate-lim-

ited current IK [36]:

1

I¼ 1

IKþ 1

IL: ð6Þ

Theoretically, the mass-transport-limited current

could consist of two components: one characterizing thesubstrate diffusion in solution (through the diffusion

layer) and the other one – through the external RP layer.

Though, as it will be shown in Section 4 the term de-

scribing the last component can be disregarded.

In accordance with the Levich equation [36], the

mass-transport-limited current is a function of the hy-

drogen peroxide bulk concentration [H2O2] and the

angular rotation speed x of the electrode:

IL ¼ 0:62nFAD2=3x1=2m�1=6½H2O2; ð7Þ

where n is the number of electrons transferred to the

enzyme in one catalytic cycle (theoretically n ¼ 2); F is

the Faraday constant; A (cm2) is the geometric area of

the electrode; D (cm2 s�1) is the hydrogen peroxide

diffusion coefficient; m ðcm2 s�1Þ is the kinematic vis-

cosity of water; [H2O2] (mol l�1) is the hydrogen per-

oxide concentration.

For the interface in study the kinetic limitations arenot obvious, but, based on comparative analysis and

experimental tests, reactions (1)–(3) were identified as

the most probable rate-limiting steps. Firstly, the rate of

reaction between native HRP and H2O2 is known to be

high (15� 106 M�1 s�1 [37]). Though effect of the en-

zyme immobilization matrix [38], as well as possible ef-

fect of the enzyme modification, could bring about a

significant decrease of the enzyme activity. Secondly,[OsIIðbpyÞ2Cl] complex reacts slowly with oxidized na-

tive HRP in solution (7:8� 102 M�1 s�1 [10]), being aquite sluggish reductive substrate for HRP. As it will be

shown in Section 4 the rate of heterogeneous electrontransfer between the polymer�s redox centers and elec-

trode surface as well as electron ‘‘diffusion’’ through the

transducing polymer can be disregarded as rate limiting

steps. Based on these considerations, it was assumed

that kinetic limitation comes from reactions (1)–(3) only.

Formally, in case of the investigated system the re-

action-rate-limited current can be expressed analogously

to that for mediated bioelectrocatalytic reduction ofH2O2 at (sub)monolayer HRP-modified electrode in

presence of a freely diffusing electron donor substrate

[39]:

1

IK¼ 1

2n1FACHRP

1

k1½H2O2

�þ k2 þ k3

k2k3

1

½Os

�; ð8Þ

where CHRP (mol cm�2) is the surface concentration of

the modified enzyme; [Os] (mol l�1) is concentration of

the Os redox centers attached to the polymer matrix

formed by the two RP transducing layers; n1 is the

number of electrons transferred per each electron donor

species (n1 ¼ 1). This approach is quite simplistic as it is

unlikely the Os redox centers within the RP layers adopt

a uniform three-dimensional distribution.In spite of those, the model allows estimation of the

turnover number of the oxidized enzyme reduction by

[OsIIðbpyÞ2Cl] redox centers. Thus, Eq. (8) can be re-

written as follows:

1

IK¼ 1

2n1FACHRP

1

k1½H2O2

�þ 1

kS;ET

�: ð9Þ

The (apparent) turnover number kS;ET ðs�1Þ could beinformative from at least two reasons: (i) it presents a

parameter characterizing the efficiency of the electrical

communication between enzyme and redox polymer,

and, consequently, could be used as a criterion for op-

timization of the bioelectrode response; (ii) it allows

direct comparison with other peroxidase-based bioelec-

trodes, particularly those based on direct electron

transfer (ET) between the enzyme active center and theelectrode surface.

By combining Eqs. (6) and (7) the Koutecky–Levich

equation for the observed bioelectrocatalytic H2O2 re-

duction current can be derived:

1

I¼ 1

IKþ 1

0:62nFAD2=3x1=2m�1=6½H2O2: ð10Þ

By plotting I�1 versus x�1=2 the reaction-rate-limitedcurrent can be separated from the mass-transport-lim-

ited one. From the slope and intercept of the dependenceof IK on H2O2 concentration in doubly reciprocal plot,

k1 and kS;ET, respectively, can be calculated.It is worth to mention that the Koutecky–Levich

formalism requires both diffusional and kinetic pro-

cesses to be of the first order. This is not the case of the

mechanism of H2O2 reduction at the interface in study:

Eq. (9) is non-linear in hydrogen peroxide concentration

V. Rosca, I. Catalin Popescu / Electrochemistry Communications 4 (2002) 904–911 907

and, consequently, Eq. (10) is not strictly correct.

Though, in conditions of the enzyme under-saturation

the kinetic process is of the first order in hydrogen

peroxide concentration (see also Section 4). Conse-

quently, the application of the Koutecky–Levich treat-ment is justified.

4. Results and discussions

As it can be seen from Fig. 2 the second RP layer is

electrically connected to the electrode surface, thus

forming with the first RP layer a more or less continuousredox network. In accordance with the data presented in

Table 1, the electrochemical behavior of the two RP

layers as a whole presents somehow unexpected char-

acteristics when compared to that of the first RP layer.

Thus, the binding of the second layer of redox polymer

does not double the electrode coverage by redox active

centers. It results that the amount of redox polymer in

the outer layer is not enough to assure a full sand-wiching of the HRP molecules between two redox

polymer layers. Consequently, the transduction effi-

ciency of the investigated architecture will be dimin-

ished. In the same time, the presence of the second redox

polymer layer slightly increases the peak separation(DEp) as well as the peak width at half maximum

(EFWHM). This behavior could be assigned partially to

the manifestation of electron hopping mechanism of the

charge transport through the polymer network formed

by the two RP layers. However, the electron diffusion

pathways are too short to make an important difference

to the surface-confined redox couple behavior. On the

other hand, the interposed enzyme layer could also exertan effect on the RP electrochemical response.

On the other hand, the electrochemical parameters

values (DEp and EFWHM) of both voltammograms

present relatively small discrepancies in comparison with

their theoretical values. These differences should reflect

the interactions (attractive for the anodic wave, and

repulsive for the cathodic one) expected to exist between

the redox active centers and their microenvironment.As it was stated above, heterogeneous electron

transfer was disregarded as a possible rate-limiting step.

This assumption is supported by the fact that observed

electrocatalytic current (in conditions of high rotation

speed and at saturating hydrogen peroxide concentra-

tions) does not depend on the applied potential in the

range between 0 and 300 mV (results not shown). This is

not surprising at all as the redox couple is known to befast.

H2O2 reduction currents measured at the investigated

bioelectrode at different H2O2 concentrations and rota-

tion speeds were plotted in Levich coordinates (Fig. 3).

It was concluded that the system in study shows a well-

behaved mixed control (diffusional-kinetic) current re-

sponse. From Fig. 4 it can be seen that increase of the

H2O2 concentration brings about decrease of the slopeof the Koutecky–Levich plot and, consequently, de-

crease of the mass-transport-limited current contribu-

tion to the observed current up to negligibly small

values.

The slope of the Koutecky–Levich plot depends on

the bulk concentration of H2O2 and is proportional to

the number of electrons transferred per H2O2 molecule

(see Eq. (7)). The experimental number of electrons(1.93), calculated from the slope of the Koutecky–Le-

vich plot versus [H2O�12 ] (linear least-square regression

analysis, Fig. 5(A)), is close to the theoretical value of 2.

Hence an important criterion of the Koutecky–Levich

treatment applicability to the investigated system is

Fig. 2. Cyclic voltammograms recorded at Au/MPS/RP (solid line) and

Au/MPS/RP/HRPmod/RP (dashed line) modified electrodes. Experi-

mental conditions: 0.1 M phosphate buffer solution containing 0.15 M

sodium chloride (pH 7.0), potential scan rate 20 mV s�1.

Table 1

Electrochemical parameters of the voltammetric response for RP-modified electrodes in two different configurationsa

Interface E00 (mV) DEp (mV) EHWFM (mV) Ipa=Ipc C � 1010

(mol cm�2)Anodic Cathodic

Au/MPS/RP/ 324 14 76 108 1.41 1.2

Au/MPS/RP/

HRPmod/RP

333 18 88 119 1.37 1.6

a See Fig. 2 for experimental conditions.

908 V. Rosca, I. Catalin Popescu / Electrochemistry Communications 4 (2002) 904–911

fulfilled. It is worth mentioning that results of the Stu-

dent�s test (tcalc ¼ 3:338, ttab ¼ 3:499, significance level

0.01, k ¼ 7) indicate that the intercept of the regression

line is not significantly different from zero. Conse-

quently, the current characterizing the hydrogen per-

oxide diffusion through the outer RP layer does not have

any important contribution to the mass-transport-lim-ited current [40].

In Fig. 5(B) the intercepts of the Koutecky–Levich

plot (I�1K ) are plotted against ½H2O2�1. In accordance

with Eq. (9), estimation of both k1 and kS;ET is possibleprovided that the surface concentration of HRPmod is

known. In this paper the enzyme surface concentration

CHRP was approximated to 8 pmol cm�2, corresponding

to the closely packed HRP monolayer (HRP is a glob-

ular protein of about 50 �AA diameter [41]). We expect this

value to be an overestimation of the enzyme surface

concentration. This approximation induces an impor-

tant uncertainty in estimation of both rate constants.However, this drawback could be overcome by experi-

mental estimation of the peroxidase surface concentra-

tion, making use of such a sensitive gravimetric

technique as quartz crystal microbalance (14 pmol cm�2,on a graphite coating [42]) or from UV-vis spectra of the

fluorescein-labeled HRP (1.2 pmol cm�2, in a multilayerassembly formed by layer-by-layer deposition of con-

canavalin A and HRP [43]). Another related problem, offundamental importance, is the enzyme layer structure

(monolayer vs. multilayer adsorption) and the fraction

of the catalytically active enzyme effectively connected

to the electrode. In Section 3 we argue for a monolayer

deposition of the enzyme, but that is the only informa-

tion that we can provide on the enzyme layer structure

and composition for the moment (see Fig. 1).

Using the above-mentioned approximate value forCHRP, the rate constant k1 of the reaction between H2O2

and HRPmod was found to be 2:74� 105 M�1 s�1 andthe apparent turnover number kS;ET was equal to

2:14 s�1. As it was mentioned above, the non-linearity

of Eq. (9) in hydrogen peroxide concentration could be a

source of the errors in estimation of the rate constants.

The errors induced should probably be small, as the

experimental points at lowest, i.e., the enzyme under-saturating, concentrations are dominating ones in the

linear regression analysis.

The rate constant k1 for HRPmod in RP/HRPmod/RParchitecture is about 54 times lower than that for the

wild type HRP in solution (15� 106 M�1 s�1 [37]). Thisconsiderable difference could be assigned to the enzyme

immobilization effect and probably to the enzyme par-

tial deactivation upon its chemical modification.It is worthy noting that, kS;ET allows a direct compar-

ison between the investigated transducing scheme and

that based on direct ET between the enzyme active center

and electrode. Thus, the value calculated for HRPmod(2:14 s�1) is close, but somehow higher than, to the

turnover number of the heterogeneous electron transfer

for wild type HRP adsorbed on spectroscopic graphite

(0:66 s�1 [44], pH 7) and is considerably higher than thatat a graphite coating (2:0� 10�4 s�1 [34], pH 6.4). In this

sense to be also compared the sensitivity of Au/MPS/RP/

HRPmod/ interface (58:1 mA M�1 cm�2 [8]) and that ofpyrolytic graphite electrodes modified with physically

adsorbed native HRP (60 mA M�1 cm�2 [45]), in both

cases the enzyme being most probably presented in a

nearly single molecular layer. These facts suggest that the

Fig. 3. Levich plot for hydrogen peroxide bioelectrocatalytic reduction

at Au/MPS/RP/HRPmod/RP-modified RDE. Experimental conditions:

0.1 M phosphate buffer solution, 0.15 M sodium chloride (pH 7.0),

containing hydrogen peroxide at different concentrations: (j) 2.2 lM;

(s) 5.4 lM; (N) 10.8 lM; (.) 21.6 lM; (r) 53.9 lM; (d) 107.8 lM;

applied potential 0.1 V vs. AgjAgCljKClsat.

Fig. 4. Koutecky–Levich plot for hydrogen peroxide bioelectrocata-

lytic reduction at Au/MPS/RP/HRPmod/RP-modified RDE. Experi-

mental conditions: 0.1 M phosphate buffer solution, 0.15 M sodium

chloride (pH 7.0), containing hydrogen peroxide at different concen-

trations: (j) 2.2 lM; (s) 5.4 lM; (N) 10.8 lM; (.) 21.6 lM; (r) 53.9

lM; (d) 107.8 lM; applied potential 0.1 V vs. AgjAgCljKClsat.

V. Rosca, I. Catalin Popescu / Electrochemistry Communications 4 (2002) 904–911 909

efficiency of the peroxidase regeneration by the trans-

ducing redox polymer is most probably superior but still

comparable to that provided by direct electron transfer.

At the same time, to be recalled and emphasized thefundamental advantage of using redox polymer net-

works. These conducting immobilization matrices pro-

vide electrical communication for enzyme molecules

situated far from the surface, effectively increasing the

‘‘virtual’’ surface of the electrode [46].

5. Conclusions

New results on the electrical communication in redox

polyelectrolyte–peroxidase assemblies, built up by elec-

trostatically controlled self-deposition technique were

presented and discussed.

It was shown that, using a simple kinetic model,

formulated for steady-state conditions, it was possible to

estimate the immobilized horseradish peroxidase activityas well as the rate of the oxidized peroxidase regenera-

tion by the osmium redox polymer, expressed by the

apparent turnover number of the reaction. The last

constant could be useful for investigation of the factors

controlling the electrical communication between the

enzyme and the transducing macromolecule redox cen-

ters, eventually serving as a criterion for optimization of

the bioelecrode response.

Acknowledgements

The Romanian Academy supported this project fi-

nancially (Grant 21/2002). Authors acknowledge prof.

E. Dominguez for providing the modified enzyme and

redox polymer as well as for helpful discussions.

References

[1] T. Ruzgas, E. Cs€ooregi, J. Emn�eeus, L. Gorton, G. Marko-Varga,

Anal. Chim. Acta 330 (1996) 123.

[2] D.M. Ivnitskii, R.A. Sitdikov, V.E. Kurochkin, Anal. Chim. Acta

310 (1992) 45.

[3] W.O. Ho, D. Athey, C.J. McNeil, Biosens. Bioelectron. 10 (1995)

683.

[4] W.O. Ho, D. Athey, C.J. McNeil, H.J. Hager, G.P. Evans, W.H.

Mullen, J. Electroanal. Chem. 351 (1993) 185.

[5] J. Kulys, M. Pesliakiene, A. Samalius, Bioelectrochem. Bioenerg.

8 (1981) 81.

[6] L. Gorton, G. J€oonsson-Pettersson, E. Cs€ooregi, K. Johanson, E.

Dom�ıınquez, G. Marko-Varga, Analyst 117 (1992) 1235.

[7] L. Boguslavsky, H. Kalash, Z. Xu, D. Beckles, L. Geng, T.

Scotheim, V. Laurinavicius, H.S. Lee, Anal. Chim. Acta 311

(1995) 15.

[8] A. Narv�aaez, G. Su�aarez, I.C. Popescu, I. Katakis, E. Dom�iinguez,Biosens. Bioelectron. 15 (2000) 43.

[9] M. Vreeke, R. Maidan, A. Heller, Anal. Chem. 64 (1992) 3084.

[10] M.G. Garguilo, N. Huynh, A. Proctor, A.C. Michael, Anal.

Chem. 65 (1993) 523.

[11] B.R. Horrock, D. Schmidtke, A. Heller, A.J. Bard, Anal. Chem.

65 (1993) 3605.

[12] A. Heller, J. Phys. Chem. 96 (1992) 3579.

[13] B.A. Gregg, A. Heller, J. Phys. Chem. 95 (1991) 5970.

[14] B.A. Gregg, A. Heller, J. Phys. Chem. 95 (1991) 5976.

[15] I. Katakis, A. Heller, in: F.W. Sheller, F. Schubett, J. Fedrowitz

(Eds.), Frontiers in Biosensorics. Fundamental Aspects, Bukhau-

sen Verlag, Basel, 1997, p. 229.

[16] G. Decher, J. Schmitt, Prog. Colloid Polym. Sci. 89 (1992) 160.

[17] G. Decher, Science 277 (1997) 1232.

[18] Y. Lvov, G. Decher, G. Sukhorukov, Macromolecules 26 (1993)

5396.

[19] Y.M. Lvov, Z. Lu, J.B. Schenkman, J.F. Rusling, J. Am. Chem.

Soc. 120 (1998) 4073.

[20] K.T. Kinnear, H.G. Monbouquette, Langmuir 9 (1993) 2255.

[21] Y. Lvov, K. Ariga, I. Ichinose, T. Kunitake, J. Am. Chem. Soc.

117 (1995) 6117.

[22] F. Mizutani, Y. Sato, S. Yabuki, Y. Hirata, Chem. Lett. 4 (1996)

251.

[23] M. Onda, Y. Lvov, K. Ariga, T. Kunitake, Biotechnol. Bioeng. 51

(1996) 163.

Fig. 5. Dependence of the slopes (A) and the intercepts (B) of the Koutecky–Levich plot on hydrogen peroxide concentration. Experimental con-

ditions: as in Fig. 4.

910 V. Rosca, I. Catalin Popescu / Electrochemistry Communications 4 (2002) 904–911

[24] K.T. Kinnear, H.G. Monbouquette, Anal. Chem. 69 (1997)

1771.

[25] J. Hodak, R. Etchenique, E.J. Calvo, K. Singhal, P.N. Bartlett,

Langmuir 13 (1997) 2708.

[26] S.-F. Hou, K.-S. Yang, H.-Q. Fang, H.-Y. Chen, Talanta 47

(1998) 561.

[27] E.J. Calvo, F. Battaglini, C. Danilowicz, A. Wolosiuk, M. Otero,

Faraday Discuss. 116 (2000) 47.

[28] E.J. Calvo, R. Etchenique, L. Pietrasanta, A. Wolosiuk, Anal.

Chem. 73 (2001) 1161.

[29] E.S. Forzani, M. Otero, M.A. Perez, M. Lopez Teijelo, E.J.

Calvo, Langmuir 18 (2002) 4020.

[30] C. Bourdillon, C. Demaille, J. Moiroux, J.M. Saveant, J. Am.

Chem. Soc. 116 (1994) 10328.

[31] P.N. Bartlett, K.F.E. Pratt, J. Electroanal. Chem. 397 (1995) 61.

[32] I. Katakis, A. Heller, Anal. Chem. 64 (1992) 1008.

[33] U. Oesch, J. Janata, Electrochim. Acta 28 (1983) 1237.

[34] R. Woods, in: A.J. Bard (Ed.), Electroanalytical Chemistry, vol. 9,

Marcel Dekker, New York, 1976, p. 124.

[35] H.B. Dunford, in: J. Everse, K.E. Everse, M.B. Grisham (Eds.),

Peroxidases in Chemistry and Biology, vol. 2, CRC Press, Boca

Raton, FL, 1991.

[36] A.J. Bard, L.R. Faulkner, Electrochemical Methods: Fundamen-

tals and Applications, Wiley, New York, 1980 (Chapter 8).

[37] H. Anni, T. Yonetani, in: H. Siegel, A. Siegel (Eds.), Metal Ions in

Biological Systems, Mechanism of Action of Peroxidases, Marcel

Dekker, New York, 1992, p. 220.

[38] L. Cook, R. Larter, P. Shen, T. Geest, J. Phys. Chem. 97 (1993)

9060.

[39] L. Gorton, A. Lindgren, T. Larsson, F.D. Munteanu, T. Ruzgas,

I. Gazaryan, Anal. Chim. Acta 400 (1999) 91.

[40] C.P. Andrieux, J.-M. Saveant, in: R.W. Murray (Ed.), Molecular

Design of Electrode Surfaces, Wiley, New York, 1992, pp. 218–

227.

[41] T. Asakura, T. Yonetani, J. Biol. Chem. 244 (1969) 537.

[42] T. Tatsuma, K. Ariyama, N. Oyama, J. Electroanal. Chem. 446

(1998) 205.

[43] Y. Kobayashi, J. Anzai, J. Electroanal. Chem. 507 (2001) 250.

[44] T. Ruzgas, L. Gorton, J. Emneus, G. Marko-Varga, J. Electro-

anal. Chem. 391 (1995) 41.

[45] U. Wollenberger, V. Bogdanovskaya, S. Bobrin, F. Scheller, M.

Tarasevich, Anal. Lett. 23 (1990) 1795.

[46] K. Habermuller, M. Mosbach, W. Schumann, Fresenius J. Anal.

Chem. 366 (2000) 560.

V. Rosca, I. Catalin Popescu / Electrochemistry Communications 4 (2002) 904–911 911