Full Paper

Silica-Polyaniline Based Bienzyme Cholesterol Biosensor: Fabrication and Characterization Kalayil Manian Manesh, Padmanabhan Santhosh, Anantha Iyengar Gopalan, Kwang-Pill Lee* Department of Chemistry Graduate School, Kyungpook National University, Daegu 702-701, South Korea *e-mail: [email protected] Received: February 20, 2010;& Accepted: May 21, 2010 Abstract In the present investigation, silica-polyaniline based bienzyme cholesterol biosensor is fabricated through a simple one-step electrochemical method. The one-step fabrication process involves electrochemical polymerization of N[3(trimethoxysilyl)propyl]aniline to result poly(N[3-(trimethoxysilyl)propyl]aniline) (PTMSPA) and simultaneous immobilization of two enzymes, horseradish peroxidase (HRP) and cholesterol oxidase (ChOx) into PTMSPA matrix. The modified electrode is designated as PTMSPA-HRP/ChOx-ME. PTMSPA facilitates direct electron transfer between the electrode surface and the active redox centers of HRP. This enables the operation of a biosensor at a low working potential of about 150 mV (vs. Ag/AgCl) for the detection of hydrogen peroxide. The PTMSPA-HRP/ ChOx-ME demonstrates excellent analytical performance for the detection of cholesterol between 1 and 25 mM with high sensitivity and selectivity. PTMSPA possesses features suited for the fabrication of third-generation biosensors. Keywords: HRP, ChOx, Direct electron transfer, Third-generation biosensors, Amperometric cholesterol biosensor, Biosensors

DOI: 10.1002/elan.201000138

1. Introduction Cholesterol is an important lipid (steroid alcohol) and often found to be esterified with a fatty acid in the human body. It is one of the essential components of nerve cells and a precursor for many bio-materials, such as bile acid and steroid hormones [1]. The cholesterol or plaque build-up causes arteries to become thicker, harder and less flexible, slowing down and sometimes blocking blood flow to the heart. Excessive presence of cholesterol in plasma can increase risk of developing clinical disorders such as hypertension, arteriosclerosis and myocardial infarction [2]. The determination of cholesterol is therefore important in clinical diagnosis. Various analytical methods such as calorimetric, spectrophotometric and electrochemical methods have been developed for the quantification of cholesterol [3, 4]. The constraints in achieving good specificity and selectivity while using chemical methods could be overcome by enzymatic assays [5]. Enzyme based sensors (biosensors) involve two important stages, immobilization of biorecognition element in a matrix and transduction of signal from the analyte by a suitable technique. Development of electrochemical amperometric biosensors involves immobilization of enzymes onto the electrode surface and transduction of the response of the analyte in the form of signal (current). The most widely used cholesElectroanalysis 2010, 22, No. 20, 2467 – 2474

terol biosensors are based on oxidase enzymes, cholesterol oxidase (ChOx) that generates H2O2 in presence of oxygen, which is detected electrochemically [6]. Electrochemical signal transduction offers advantages such as rapid analysis, stability, good linearity and specificity. A serious problem that hampers the performance of electrochemical biosensors for the analysis of biological samples is the presence of metabolites or other electroactive compounds that interfere with the electrochemical signal of the target molecule. These interfering species complicate the signal from the analyte, as they are oxidize/reduce at the potential close to the electrochemical detection of H2O2 [7]. Moreover, electrochemical cholesterol biosensor still needs improvement in terms of sensitivity, reproducibility and life time. A new and alternative approach is proposed that utilizes bienzymes (oxidase and peroxidase) to circumvent the interference problem and improve the sensitivity of the biosensor [8]. Such a bienzyme based biosensor operates at relatively low applied potentials where the noise level and interferences from oxygen and other electrooxidizable species are minimal. In addition, enzyme cascades, where an enzyme is catalytically linked to another enzyme, can produce signal amplification and therefore increase the biosensor efficiency [9]. Bienzyme electrodes for the detection of various metabolites/analytes, such as alcohol, D-/L-amino acid, choline, cholesterol and glucose

 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

2467

Full Paper

K. M. Manesh et al.

were developed using peroxidase (from Horseradish/ Fungal) and the respective oxidase enzymes [10–13]. Cholesterol biosensor uses ChOx that produces H2O2 from cholesterol in presence of molecular oxygen and the enzymatically formed H2O2 is detected by an immobilized enzyme, horseradish peroxidase (HRP). HRP contains heme as a prosthetic group that catalyzes the reduction of H2O2 and can facilitate direct electron transfer (DET) to the electrode [8]. The DET between active site of HRP and electrode is relatively faster, compared with other reductases, possibly due to 3D configuration of the protein, where the heme group is in the outer region and assists the electron transfer. The biocatalytic process at the electrode surface can be augmented with a suitable matrix that could accommodate/immobilize higher loading of enzymes and provides adequate stability and activity to the enzymes. Silica based materials are suited for enzyme immobilization due to their large surface area, large sorption capacity, thermal stability, adaptability to host molecules of various sizes and shapes and biocompatibility [14, 15]. In addition, silica matrices protect the peroxidase activity from H2O2 inactivation even at relatively high H2O2 concentrations and prevent further degradation [16]. However, for effective immobilization of enzymes, silica needs to be functionalized with suitable structural moieties. Electroactive conducing polymers (CP) are extensively used in electrochemical sensors as they offer considerable flexibility through available monomeric structures [17, 18]. Inorganic (silica)–organic (CP) composite materials can be prepared using organic molecule to obtain functionalized silica as the precursor. Post-synthetic reactions can be subsequently exploited with the functional groups in silica for the covalent attachment of CP units to obtain silica-CP structures [19]. In the present investigation, we have fabricated a bienzyme cholesterol biosensor using a silica-polyaniline (PANI) composite material. The silica-PANI composite was prepared through a simple one-step electrochemical approach. The methodology used in the present work minimizes the tediousness in the multi-step procedures [20], which are generally encountered for the fabrication of sol-gel based bienzyme biosensors. The silica-PANI composite (poly(N[3-(trimethoxysilyl)propyl]aniline); PTMSPA) was formed as film on the surface of the electrode by one step electrochemical method and simultaneously used to immobilize the two enzymes, HRP and ChOx. Typically, TMSPA was electropolymerized through sol-gel process in the presence of two enzymes to result PTMSPA–enzyme composite, in which the enzymes (HRP and ChOx) are in-situ entrapped into the sol-gel PTMSPA matrix. PTMSPA contains silica network with interconnected PANI chains and offers synergistic advantages of both silica and PANI for the fabricated cholesterol biosensor.

2468

www.electroanalysis.wiley-vch.de

2. Materials and Methods 2.1. Reagents HRP (E.C 1.11.1.7, 250 U mg1, from Horseradish), ChOx (EC 1.1.3.6, 50 U mg1, from Brevibacterium sp.), N[3-(trimethoxysilyl)propyl]aniline (TMSPA), 2-naphthalene sulfonic acid (NSA), hydrogen peroxide, guanidine chloride, cholesterol, Triton X-100, isopropanol, dopamine, uric acid, acetaminophen, ascorbic acid, lactic acid, cystein, albumin and glucose were analytical grade samples from Sigma-Aldrich. Deionized water (18.2 MW cm, Millipore) was used throughout the experiments. A stock solution of cholesterol was prepared in Triton X-100/isopropanol mixture at 60 8C and then diluted with 0.1 M phosphate buffer saline (PBS; pH 7). All other solutions were prepared with 0.1 M PBS (pH 7) afresh at the time of experiments. 2.2. Electrochemical Measurements All the electrochemical measurements were performed by using EG&G PAR 283 Potentiostat/Galvanostat. Ag/ AgCl and platinum wire were used as reference and counter electrodes, respectively. Indium-doped tin oxide (ITO)-coated glass plate (1 cm  1 cm area; specific surface resistance 10 W) and platinum (Pt) disc electrode (2 mm diameter) were used for making the sensor electrodes. Prior to experiments, the ITO plate was rinsed with acetone and washed with distilled water. Pt electrode was cleaned electrochemically by cycling the potential between 0.5 and + 1.2 V in 2 M sulfuric acid. Note that ITO electrode was used for all the electrochemical experiments except for hydrodynamic voltammetry. 2.3. Fabrication of PTMSPA-HRP/ChOx-ME Electrochemical polymerization of TMSPA and simultaneous immobilization of enzymes were performed at the working electrode to fabricate PTMSPA-HRP/ChOx biosensor. Typically, a solution of TMSPA (50 mM) in NSA (1 M) containing fixed mass ratio of the two enzymes (HRP/ChOx) was electropolymerized by applying a potential of + 1.0 V (vs. Ag/AgCl) at the indium tin oxide (ITO) coated glass electrode for 10 min. Similarly, biosensor electrodes were fabricated with different solutions containing the two enzymes in varying mass ratios (HRP/ ChOx) between 1 : 7 to 7 : 1. The total amount of two enzymes was maintained as 8 mg mL1. 2.4. Characterization The morphology of PTMSPA was examined by high resolution transmission electron microscopy (HRTEM, Jeol JEM-ARM1300S, Japan). The Fluorescence measurements were performed using Perkin-Elmer LS 55 Luminescence Spectrometer. The excitation was made at a wavelength of 285 nm at a 458 angle of the film plane.

 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Electroanalysis 2010, 22, No. 20, 2467 – 2474

Silica-Polyaniline Based Bienzyme Cholesterol Biosensor

This wavelength was chosen in order to avoid contribution of the five-tyrosine residues present in HRP.

3. Results and Discussion 3.1. Fabrication of the Modified Electrode The fabrication of the modified electrode involves the potentiostatic deposition of sol-gel PTMSPA with simultaneous co-immobilization of HRP and ChOx. The electrolyte, NSA with its amphiphillic character forms spherical micelles in aqueous solution. The electrostatic interactions between the NH2 group in TMSPA and SO3  group in NSA induces self-assembly of TMSPA molecule in the spherical micelles environment. This self-assembled structure was subjected to electropolymerization by the application of + 1.0 V. The polymerization of TMSPA involves the oxidation of NH2 group in TMSPA to initiate the polymerization of aniline units. Simultaneously, the hydrolysis of OCH3 groups in TMSPA and condensation of OH group occur and sol-gel derived PTMSPA was obtained. PTMSPA was deposited as green colored thin film on the surface of the working electrode. The sol-gel process is also accompanied by the encapsulation of the two enzymes, HRP and ChOx, to result PTMSPA- HRP/ ChOx-ME. The mechanism of formation of PTMSPA is detailed in our previous report [13]. Figure 1 presents a HRTEM image of PTMSPA. HRTEM image reveals that PTMSPA comprised of stacked spherical (silica) particles. The average diameter of the spherical particle is in the range of ca. 1.0 mm. Also, well defined, uniform and ordered nanometer scale pores could be seen through the HRTEM image (Figure 1). The existence of PANI within the pores of silica could be visualized through contrast images. Thus, the HRTEM image reveals the presence of both silica network and PANI chains in PTMSPA matrix. 3.2. Electrochemistry of H2O2 at PTMSPA-HRP-ME Before evaluating the performance of the bi-enzyme electrode, PTMSPA-HRP-ME was fabricated and the electrochemical performances of PTMSPA electrode (containing only HRP) towards H2O2 was evaluated. 3.2.1. Cyclic Voltammetry The electrochemical behavior of PTMSPA-HRP-ME was investigated using cyclic voltammetry to validate the electron transfer between HRP and electrode surface. CVs were recorded at PTMSPA-HRP-ME in PBS at a scan rate of 25 mV s1. The PTMSPA-HRP-ME displays a pair of redox peaks at 153 mV (Epc) and 106 mV (Epa), Figure 2A(ii). These peaks are assigned for the redox reaction of active sites (Fe(III)/Fe(II)) in HRP. The formal potential (E8’) is found to be 130  5 mV. This potential is close to the E8’ of the Fe(II)/Fe(III) redox couple of the electroactive heme center of HRP [21], and informs Electroanalysis 2010, 22, No. 20, 2467 – 2474

Fig. 1. HRTEM image of PTMSPA; scale bar: 0.5 mm.

that DET is facilitated between HRP and electrode. PTMSPA electrode was also fabricated without HRP. In such a case, redox peaks were not observed at PTMSPAME, (Figure 2Ai). HRP at PTMSPA matrix showed a typical property of a surface-confined electrochemical process [22, 23]. The bioelectrocatalytic activity of the PTMSPA-HRPME toward 300 mM H2O2 was evaluated by cyclic voltammetry and the corresponding CV is shown in Figure 2A(iii). The reduction peak current at about 150 mV significantly increased upon addition of 300 mM H2O2 in PBS, which was accompanied by the (almost) disappearance of the oxidation peak for HRP-Fe(II). The bio-electrocatalytic reduction of H2O2 occurred at a potential of 150 mV (vs. Ag/AgCl) at PTMSPA-HRP-ME, which is quite different from the formal potential of compound-I/-II and compound-II/HRP(III) noticed around 700 mV [24]. In the typical bio-catalytic reaction, H2O2 oxidizes the heme iron of HRP to the ferryl [Fe(IV)=O] state and produces either a porphyrin cation radical (compound I) or transient protein radicals (intermediate I) or both in the reactive intermediate [25, 26]. However, in most of the cases (based on the nature of modified surfaces), DET involves an inter-conversion between the Fe(III) and Fe(II) oxidation states of the enzyme to the electrode, which can be inferred from the location of the DET current at rather negative potential values, typically at 100 mV (vs. Ag/AgCl) [27]. The difference in catalytic mechanism observed in the present investigation is due to the specific conformational changes of HRP in the PTMSPA layer or variations in the heme microenvironment. Moreover, the peroxidase/PTMSPA interface offers

 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.electroanalysis.wiley-vch.de

2469

Full Paper

K. M. Manesh et al.

Fig. 2. A) Cyclic voltammograms of (i) PTMSPA electrode in PBS, PTMSPA-HRP-ME in (ii) absence and (iii) presence of 300 mM H2O2 in PBS; scan rate: 25 mV s1. B) Fluorescence emission spectra of (i) PTMSPA-HRP, (ii) 0.88 mM HRP in PBS and (iii) PTMSPA-HRP treated with 6 M GndCl. Inset shows the magnified spectra of (i) and (ii).

principally different peroxidase molecular surface and electrode surface characteristics [28]. It is apparent from the voltammetric measurements that DET was observed between the electrode and HRPFe(II)/Fe(III) redox couple. However, it is possible that the free heme groups, which most probably were extracted during immobilization of HRP, would undergo DET. Fluorescence measurements were carried out to complement the voltammetric results and to authenticate the observed DET is only from the HRP-heme, but not from free heme groups. Fluorescence spectroscopy is an ideal tool to study the mechanism of protein inactivation upon immobilization. It provides insight into the conformational changes, unfolding processes, as well as the detachment of the active group of the entrapped protein. HRP possesses a single buried Tryptophan (Trp) residue at posi2470

www.electroanalysis.wiley-vch.de

tion 117 and the fluorescence of Trp offers information of the tertiary structure of the protein [29]. Figure 2Bi shows the fluorescence emission spectrum of PTMSPA-HRP. Special care was taken to dedope the sample to result emeraldine base form of polyaniline PANI-EB. It is to be noted that in PANI-EB, the benzonoid groups present adjacent to a quinonoid structure do not result any fluorescence emission at this range [30]. In order to have a better comparison, emission spectrum of 0.88 mM of HRP was recorded, Figure 2Bii. Good correlation between the spectra was noticed. Both spectra showed a broad emission peak around 335 nm with low intensities. This indicates that the hydrophobic environment of the Trp remains unaltered in both the cases. In general, the fluorescence intensity of Trp is quenched in the native protein due to the close presence of the heme group. However, in the denatured protein the fluorescent of Trp is strong and red shifted [31]. The similarity between emission spectrum of native HRP and PTMSPAHRP suggests that the HRP retains its native configuration in PTMSPA matrix. Otherwise, the similarity in the emission spectrum could not be observed if the heme groups are extorted during immobilization. The emission spectrum of PTMSPA-HRP, treated with 6 M guanidinium chloride (GndCl) is shown in Figure 2Biii. GndCl is one of the strongest denaturants which unfolds the native structure of all the proteins. A very intense, red shifted band around 342 nm was observed for PTMSPA-HRP treated with 6 M GndCl. The increase in the fluorescence intensity with red shifts in the position compared with PTMSPA-HRP is due to the removal of heme groups from the enzyme and consequent loss of enzyme nativity. It is thus well demonstrated that HRP retain its biocatalytic activity in the PTMSPA matrix and undergoes DET with the electrode.

3.2.2. Hydrodynamic Voltammetry Hydrodynamic voltammetric measurements were performed at PTMSPA-HRP-ME in order to determine the optimum potential required for the amperometric detection of H2O2 [32]. The current values were extracted from the amperograms recorded after the addition of 500 mM H2O2 to PBS (Figure 3). The response current was found to be negligible between + 50 and 0 mV. However, when the potential is moved towards negative direction, an increase in cathodic current was noticed from 50 to 200 mV. A highest current response was observed at 150 mV. Hence, a working electrode potential of 150 mV vs. Ag/AgCl was chosen for subsequent investigations to achieve high sensitivity and also to minimize possible interferences in the biosensing of H2O2. Such a potential dependence profile observed here is in good agreement with the CV results shown in Figure 2A. No significant reduction current was observed between 200 and + 50 mV at PTMSPA electrode (without HRP).

 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Electroanalysis 2010, 22, No. 20, 2467 – 2474

Silica-Polyaniline Based Bienzyme Cholesterol Biosensor

Fig. 3. Hydrodynamic voltammograms obtained at PTMSPAHRP-ME to addition of 500 mM H2O2 in PBS by varying the potential between 200 and + 50 mV at a rotation rate of 1700 rpm.

Fig. 4. Amperometric responses of PTMSPA-HRP-ME to the successive addition of 100 mM H2O2 at the potential of 150 mV in PBS.

3.2.3. Optimization of Experimental Variables on H2O2 Detection

excellent electron transducer for the quantification of H2O2. This facilitates us to fabricate a cholesterol biosensor with co-immobilization of ChOx (and HRP) in the PTMSPA matrix. In the typical bio-catalytic reaction, cholesterol is hydrolyzed to cholestenone and forms H2O2, in the presence of the enzyme, ChOx. Thus, the enzymatically formed H2O2 is reduced by HRP. The regeneration of HRP from oxidized to reduced state and vice versa, is attained by DET from the heme group of HRP at the PTMSPA-HRP/ChOx-ME. The biochemical reaction involved in electrochemical measurement of cholesterol is,

Amperometric measurements were performed to study the influence of various experimental variables (pH, film thickness, enzyme loading, temperature) on the sensitivity of 500 mM H2O2 at PTMSPA-HRP-ME [13]. Sensitive current response for the reduction of H2O2 at PTMSPAHRP-ME was achieved for a film thickness of 4 mm with a HRP loading of 65 % in pH 7 at 298 K. Hence, further amperometric measurements were performed with these optimized conditions.

CHOx Cholesterol þ O2 ƒƒ ƒ! Cholesterone þ H2 O2

3.2.4. Amperometric Measurements Under the optimized assay conditions, amperometric responses of PTMSPA-HRP-ME to successive concentration changes of 100 mM H2O2 in stirred PBS were conducted at the potential of 150 mV, and the corresponding current-time profile is shown in Figure 4. At 150 mV, the spiked H2O2 caused an increase in reduction current. A good linear relationship between current and H2O2 concentration was obtained up to 500 mM H2O2 with a sensitivity of 1.62 mA M1 (r = 0.9936), and the detection limit as 1 mM (3s). The response time (achieving 90 % of the maximum response) was about 2 s, which indicated a fast electron transfer process at the electrode. The response time observed here is comparable to the one measured at HRP-ZrO2 composite thin films (10 s) [33]. Thus, the PTMSPA-HRP-ME exhibited a rapid and sensitive amperometric response to successive stepwise changes to the H2O2 concentrations. 3.3. PTMSPA-HRP/ChOx-ME as a Cholesterol Sensor PTMSPA-HRP-ME showed enhanced sensitivity and selectivity for the detection of H2O2 and proved to be an Electroanalysis 2010, 22, No. 20, 2467 – 2474

The non-ionic surfactant, Triton X-100 is generally used as a solvent for dissolving cholesterol. The effect of Triton X-100 concentration on the detection of cholesterol was examined. The increase of Triton X-100 resulted in the increase of cholesterol solubility and as well as viscosity of the solution, which resulted in reduced response. Hence, isopropanol was used as a cosolvent along with Triton X-100. The optimized concentration of Triton X100 and isopropanol was found to be 0.3 and 0.25 % (v/v), respectively. PTMSPA-HRP/ChOx-MEs were fabricated with different ratios of loading of HRP and ChOx and tested for its biocatalytic activity towards cholesterol. It was found that the electrode with HRP (4 U/mg) and ChOx (2.5 U/mg) yields the optimal sensor response and affords high sensitivity. 3.3.1. Amperometric Measurements Figure 5A illustrates a typical amperometric response on successive additions of cholesterol (1 mM) at the PTMSPA-HRP/ChOx-ME, when the potential was kept as 150 mV. When an aliquot of cholesterol is added into

 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.electroanalysis.wiley-vch.de

2471

Full Paper

K. M. Manesh et al.

Fig. 6. Flow injection amperometric response at PTMSPAHRP/ChOx-ME with (A) increasing concentration of cholesterol (1–10 mM), (B) 10 repetitive injections of a 5 mM cholesterol in PBS; E = 150 mV and flow rate: 2.0 mL min1.

Fig. 5. (A) Amperometric responses of PTMSPA-HRP/ChOxME to the successive addition of 1 mM cholesterol at the potential of 150 mV in PBS; (B) calibration curve.

the PBS, a sharp increase in current was noticed and the current instantaneously reached a stable value. The sensor attains 90 % of steady-state current within 6 s. The response time is much lower than that reported with ChOx at sol-gel derived tetraethylorthosilicate films (20 s) [34], ChOx at Prussian blue-SiO2 composite (60 s) [35], ChOx at sol-gel film (51 s) [36] and polypyrrole-hydrogel film (30s) [37]. Figure 5B shows the plot of current vs. concentration of cholesterol. The PTMSPA-HRP/ ChOx-ME shows a linear response to current for concentrations of cholesterol between 1 and 25 mM with a sensitivity of 0.123 mA M1 and a correlation coefficient of 0.9983. The detection limit for cholesterol is estimated to be 500 mM with a signal-to-noise ratio of 3. 3.3.2. Flow Injection Analysis The attractive performance of the flow-injection amperometric detection of cholesterol at PTMSPA-HRP/ChOxME is also demonstrated. The operating parameters such as flow rate and applied potential were optimized in order to have better sensitivity at PTMSPA-HRP/ChOxME. Figure 6A shows a series of duplicate injections of cholesterol standard solutions (1–10 mM) into PBS at a 2472

www.electroanalysis.wiley-vch.de

flow rate of 2.0 mL min1, E: 150 mV vs. Ag/AgCl. Well-defined current signals are observed for these changes in the level of cholesterol. The response is highly reversible, without any noticeable carry-over. Interestingly, lower signal-to-noise ratio with very stable base line was obtained for cholesterol at PTMSPA-HRP/ChOxME. Another attractive feature is the reproducibility of the detection in the flow stream. Figure 6B shows the high reproducibility of the amperometric response for 10 repetitive injections of a 5 mM cholesterol solution at a flow rate of 2.0 mL min1. 3.3.3. Interference Studies It is well documented [38] that the use of low detection potential of a sensor electrode would be effective to avoid the interference from the generally known electroactive species (DA, UA, AP and AA) that coexist in biological fluids. In the present investigation, DA, UA, AP and AA were checked individually for their electroactivity at PTMSPA-HRP/ChOx-ME by varying the potentials between 200 and 50 mV in PBS. A well-defined current response (ca. 1.51 mA) was observed for 10 mM cholesterol at 150 mV (vs. Ag/AgCl), whereas, no significant current responses were observed for the other electroactive species at this potential. On the other hand, at a higher potential, for instance at + 150 mV, DA, UA, AP and AA show enhanced electroactivity than cholesterol, which are higher than the response observed at 150 mV. Further, selectivity of PTMSPA-HRP/ChOx-ME towards cholesterol was ascertained by performing amperometric measurements in the presence of the aforementioned electroactive substances. Figure 7 represents the amperometric response obtained at PTMSPA-HRP/

 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Electroanalysis 2010, 22, No. 20, 2467 – 2474

Silica-Polyaniline Based Bienzyme Cholesterol Biosensor

Fig. 7. Effect of interfering signals from dopamine (b), uric acid (c) and acetaminophen (d) of 50 mM each on the performance of PTMSPA-HRP/ChOx-ME towards 5 mM cholesterol (a) in PBS; E = 150 mV.

ChOx-ME when the electrode was biased at 150 mV vs. Ag/AgCl. Initially, cholesterol along with DA was spiked into the stirred PBS and the amperometric response at PTMSPA-HRP/ChOx-ME was recorded. A sudden increase in the current response (0.76 mA) was observed after the addition of mixture of cholesterol and DA. In order to validate that the current response was only from the reduction of enzymatically formed H2O2 and not from the combined influence of DA, cholesterol alone was spiked to the same solution and the corresponding amperogram was recorded. An increase in current of 0.75 mA was observed. In a separate experiment, 50 mM DA was spiked into PBS and the amperometric response was recorded at PTMSPA-HRP/ChOx-ME. Feeble, if any, distinction in current response was observed upon addition of DA into PBS. These collective observations corroborate that the current response resulted only from cholesterol. Similar measurements were performed with UA and AP at PTMSPA-HRP/ChOx-ME. The results indicate that either DA or UA or AP did not cause any observable interference in the current signal at PTMSPAHRP/ChOx-ME during the detection of cholesterol. It is well known that some of these compounds are good reducing substrates for peroxidase compound I and II [39]. Conversely, in the present investigation, we could not observe any noticeable interference effect from such compounds during the detection of cholesterol at PTMSPA-

HRP/ChOx-ME. This is probably due to the lower reaction rates of the respective electroactive species [40] and the low concentration used in the present investigation. On the other hand, AA constitutes a major interferent during the detection of cholesterol at PTMSPA-HRP/ ChOx-ME. This is due to the fact that AA can either consume a part of the enzymatically generated H2O2 to form dehydroascorbate or reduces the oxidized states of HRP. However, the interference from AA can be eliminated by using Nafion permselective membrane or by surface-adsorbing ascorbate oxidase onto the electrode [41]. Besides the electroactive interferents, the other interferents such as lactic acid, cystein, albumin (0.5 mM each) and glucose (5 mM) were also been investigated and no significant effects were found during the detection of cholesterol at PTMSPA-HRP/ChOx-ME. Thus, it is clear, at a potential of 150 mV, electrochemical detection of cholesterol is free from other interference species, indicating high selectivity of PTMSPA-HRP/ChOx-ME towards cholesterol.

3.3.4. Real Sample Analysis The feasibility of using PTMSPA-HRP/ChOx-ME for the determination of free cholesterol in a human serum sample was assessed under optimized experimental conditions. Serum sample was diluted using PBS and analyzed without any pretreatment. Table 1 shows the results for the detection of free cholesterol for serum samples. The results are comparable to reliable spectrophotometric values (measured using o-dianisidine). Thus, the PTMSPA-HRP/ChOx-ME proved to be highly sensitive for the given cholesterol concentrations in serum samples and showed better selectivity with exclusion of common interferences.

3.3.5. Sensor Repeatability, Reproducibility and Stability At first, the repeatability of the current response towards 1 mM cholesterol of a single PTMSPA-HRP/ChOx-ME was examined and the relative standard deviation, RSD was found to be 1.82 % for nine successive assays. The reproducibility of the electrode fabrication was measured with 1 mM cholesterol in PBS at nine different electrodes fabricated concurrently. RSD was found to be 3.7 %. The long-time storage stability of the PTMSPA-HRP/ChOxME was measured. The electrode retains 90 % of the ini-

Table 1. Detection of cholesterol at PTMSPA-HRP/ChOx-ME modified electrode in human serum samples. Samples

Human serum 1 Human serum 2

Detected before spike (mM) Reference method [a]

at PTMSPA-HRP/ ChOx-ME

1.02 0.53

1.08 (RSD = 3.5 %) 0.58 (RSD = 1.3 %)

Spiked (mM)

3 3

Detected after spike (mM) Reference method [a]

at PTMSPA-HRP/ ChOx-ME

4.07 3.52

4.06 (RSD = 2.2 %) 3.50 (RSD = 1.6 %)

Human serum 1: as received. Human serum 2: 50 % diluted using PBS. [a] through spectroscopic method using o-dianisidine, R.S.D. was calculated for three replicates. Electroanalysis 2010, 22, No. 20, 2467 – 2474

 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.electroanalysis.wiley-vch.de

2473

Full Paper

K. M. Manesh et al.

tial value after 4 weeks. After 50 days, current response decreases to 55 % of the initial value.

4. Conclusions A mediatorless cholesterol biosensor was fabricated through a simple one-step procedure using a silica-conducting polymer matrix. HRP immobilized at PTMSPA (PTMSPA-HRP-ME) exhibits direct electron transfer to the electrode and bio-electrocatalytic activity for the reduction of H2O2. At 150 mV (vs. Ag/AgCl), PTMSPAHRP-ME showed sensitive and selective detection of H2O2. These features enable us to fabricate the cholesterol biosensor by incorporation of the additional enzyme, ChOx, for the detection of free cholesterol. The resulted biosensor (PTMSPA-HRP/ChOx-ME) exhibited fast amperometric response, low detection limit (500 mM) and a broad linear concentration range to cholesterol (1 and 25 mM). Moreover, the PTMSPA-HRP/ChOx-ME displayed high sensitivity, good reproducibility, and longterm stability. The attractive feature of the present biosensor is the DET between the electrode and the redox center of protein that eliminates the need of any redox mediators for the signal transduction. The utility of such a biosensor for the non-mediated catalytic detection of free cholesterol shows great promise for the use in invitro measurements.

Acknowledgement This work was supported by Priority Research Centers Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology (2009-0093819). The authors acknowledge the Korea Basic Science Institute (Daegu, Korea) and Kyungpook National University Center for Scientific Instrumentation.

References [1] S. Singh, A. Chaubey, B. D. Malhotra, Anal. Chim. Acta 2004, 502, 229. [2] D. S. Fredrickson, R. I. Levy in: The metabolic Basis of Inherited Disease (Eds: J. B. Stanbury, J. B. Wyngaarden, D. S. Fredrickson), McGraw-Hill, New York 1972, p. 545. [3] A. Krug, R. Gobel, R. Kellber, Anal. Chim. Acta 1994, 287, 59. [4] T. Yao, M. Satomura, T. Nakahara, Electroanalysis 1995, 7, 143. [5] C. C. Allain, L. S. Poon, C. S. G. Chan, W. Richmond, P. C. Fu, Clin. Chem. 1974, 20, 470. [6] A. I. Gopalan, K. P. Lee, D. Ragupathy, Biosens. Bioelectron. 2009, 24, 2211.

2474

www.electroanalysis.wiley-vch.de

[7] P. Santhosh, K. M. Manesh, K. P. Lee, A. I. Gopalan, Electroanalysis 2006, 18, 894. [8] C. Bongiovanni, T. Ferri, A. Poscia, M. Varalli, R. Santucci, A. Desideri, Bioelectrochemistry 2001, 54, 17. [9] J. Kulys, Enzyme Microb. Technol. 1981, 3, 344. [10] J. Kulys, R. D. Schmid, Biosens. Bioelectron. 1991, 6, 43. [11] J. Kulys, U. Bilitewski, R. D. Schmidt, Sens. Actuators B 1991, 3, 227. [12] J. Kulys, R. D. Schmid, Bioelectrochem. Bioenergetics 1990, 24, 305. [13] K. M. Manesh, P. Santhosh, S. Uthayakumar, A. I. Gopalan, K. P. Lee, Biosens. Bioelectron. 2010, 25, 1579. [14] J. Wang, Anal. Chim. Acta 1999, 399, 21. [15] S. Wu, H. X. Ju, Y. Liu, Adv. Funct. Mater. 2007, 17, 585. [16] C. R. Lloyd, E. M. Eyring, Langmuir 2000, 16, 9092. [17] P. Santhosh, K. M. Manesh, A. Gopalan, K. P. Lee, Sens. Actuators B 2007, 125, 92. [18] K. M. Manesh, P. Santhosh, A. Gopalan, K. P. Lee, Anal. Bioanal. Chem. 2007, 360, 189. [19] K. P. Lee, A. M. Showkat, A. I. Gopalan, S. H. Kim, S. H. Choi, Macromolecules 2005, 38, 364. [20] J. Dong, Y. Hu, J. Xu, X. Qu, C. Zhao, Electroanalysis 2009, 21, 1792. [21] H. A. Harbury, J. Biol. Chem. 1957, 225, 1009. [22] K. M. Manesh, H. T. Kim, P. Santhosh, A. I. Gopalan, K. P. Lee, Biosens. Bioelectron. 2008, 23, 771. [23] P. Santhosh, K. M. Manesh, S. Uthayakumar, A. I. Gopalan, K. P. Lee, Biosens. Bioelectron. 2009, 24, 2008. [24] Z. S. Farhangrazi, M. E. Fossett, L. S. Powers, W. R. Ellis, Biochemistry 1995, 34, 2866. [25] C. E. Schulz, R. Rutter, J. T. Sage, P. G. Debrunner, L. P. Hager, Biochemistry 1984, 23, 4743. [26] C. E. Catalano, Y. S. Choe, P. R. Ortiz de Montellano, J. Biol. Chem. 1989, 264, 10534. [27] T. Ruzgas, A. Lindgren, L. Gorton, H. Hecht, J. Reichelt, U. Bilitewski, in Electroanalytical Methods for Biological Materials (Eds: A. Brajter-Toth, J. Q. Chambers), Marcel Dekker, New York 2002; pp. 233 – 254. [28] E. E. Ferapontova, Electroanalysis 2004, 16, 1101. [29] P. Asuri, S. S. Bale, R. C. Pangule, D. A. Shah, R. S. Kane, J. S. Dordick, Langmuir 2007, 23, 12318. [30] J. R. G. Thorne, J. G. Masters, S. A. Williams, A. G. MacDiarmid, R. M. Hochstrasser, Synth. Met. 1992, 49, 159. [31] J. E. Brunet, G. A. Gonzalez, C. P. Sotomayor, Photochem. Photobio. 1983, 38, 253. [32] P. Santhosh, K. M. Manesh, S. Uthayakumar, S. Komathi, A. I. Gopalan, K. P. Lee, Bioelectrochemistry 2009, 75, 61. [33] Z. Tong, R. Yuan, Y. Chai, Y. Xie, S. Chen, J. Biotechnol. 2007, 128, 567. [34] A. Kumar, R. R. Pandey, B. Brantley, Talanta 2006, 69, 700. [35] J. Li, T. Peng, Y. Peng, Electroanalysis 2003, 15, 1031. [36] T. Yao, K. Takashima, Biosens. Bioelectron. 1998, 13, 67. [37] S. Brahim, D. Narinesingh, A. G. Elie, Anal. Chim. Acta 2001, 448, 27. [38] P. Santhosh, K. M. Manesh, A. Gopalan, K. P. Lee, Anal. Chim. Acta 2006, 575, 32. [39] Electroanalytical Methods for Biological Materials (Eds: A. B. Toth, J. Q. Chambers), Marcel Dekker, New York 2002, p. 233 – 254. [40] O. V. Lebedeva, N. N. Ugarova, Russ. Chem. Bull. 1996, 45, 18. [41] K. V. Inamdar, K. G. Raghavan, D. S. Pradhan, Clin. Chem. 1991, 37/6, 864.

 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Electroanalysis 2010, 22, No. 20, 2467 – 2474