Bienzymatic amperometric biosensor for glucose based

Analytica Chimica Acta 451(2002)251–258Bienzymatic amperometric biosensor for glucose based on polypyrrole/ceramic carbon as electrode materialFaming Tian,Guoyi Zhu ∗Laboratory of Electroanalytical Chemistry,Changchun Institute of Applied Chemistry,Chinese Academy of Sciences,Changchun 130022,China Received 6April 2001;received in revised form 10September 2001;accepted 10September 2001AbstractA novel amperometric biosensor utilizing two enzymes,glucose oxidase (GOD)and horseradish peroxidase (HRP),was developed for the cathodic detection of glucose.The glucose biosensor was constructed by electrochemical formation of a polypyrrole (PPy)membrane in the presence of GOD on the surface of a HRP-modified sol–gel derived-mediated ceramic carbon electrode.Ferrocenecarboxylic acid (FCA)was used as mediator to transfer electron between enzyme and electrode.In the hetero-bilayer configuration of electrode,all enzymes were well immobilized in electrode matrices and showed favorable enzymatic activities.The amperometric detection of glucose was carried out at +0.16V (versus saturated calomel reference electrode (SCE))in 0.1M phosphate buffer solution (pH 6.9)with a linear response range between 8.0×10−5and 1.3×10−3M glucose.The biosensor showed a good suppression of interference in the amperometric detection.©2002Elsevier Science B.V .All rights reserved.Keywords:Biosensor;Glucose;Polypyrrole;Sol–gel;Ceramic carbon electrode1.IntroductionThe sol–gel technique has attracted much attention in the field of immobilization of different reagents [1–5].The excellent properties of sol–gel materi-als,such as chemical inertness,optical transparency,simplicity of preparation,negligible swelling in aque-ous,low temperature encapsulation and mechanical stability,endued these materials with an extensive role in preparation of chemical sensors and biosen-sors [6–12].The entrapped species such as chemical materials and biomolecules magnificently preserved their chemical properties or bioactivities.The most∗Corresponding author.Tel.:+86-431-5262071;fax:+86-431-5685653.E-mail address:zhuguoyi@ (G.Zhu).interesting thing is that the leaching of these en-trapped species did not occur or just occurred very slowly.The reason for this is that the sol–gel materials are usually formed by hydrolysis of an alkoxide precursor followed by condensation to yield a poly-meric oxo-bridged SiO 2network and the immobilized species are encapsulated within the physically rigid network [6].The enzymes and proteins can be immobilized within sol–gel matrices still maintaining their native properties and reactivities,this makes the technique a potential tool for the development of new biosensors.Sol–gel-derived electrochemical biosensors mainly relied on two basic configurations:conductive ceramic composites [13–15]and electrode surface coatings [8,16–18].Since the pioneering work of Lev and co-workers sol–gel derived composite carbon electrodes0003-2670/02/$–see front matter ©2002Elsevier Science B.V .All rights reserved.PII:S 0003-2670(01)01405-2252F.Tian,G.Zhu/Analytica Chimica Acta451(2002)251–258(CCEs)[13,19–22]have been widely used to deve-lop all kinds of amperometric biosensors.Catalysts, enzymes,electron transfer mediators can be readily incorporated into the matrices of CCEs for the deve-lopment of surface renewable amperometric ually mediators or electrocatalysts are used in CCEs to improve the performance of oxidase-based biosensor.Generally,the detection mode involved in oxidase-based biosensors is often based on the electrochemical detection of hydrogen peroxide,which is produced in the course of the enzyme-catalyzed oxidation of substrates by dissolved oxygen.However,the direct oxidation of hydrogen peroxide requires a relative high working potential(exceeding ca.0.6V versus saturated calomel reference electrode(SCE))in order to obtain a sufficiently high sensitivity.At such a potential many substances,usually present in biologic samples(such as urate,acetaminophen and ascorbic acid)can also be electrochemically oxidized leading to interfering response in quantitation of substrate concentration.General methods used to suppress the interferences have involved utilizing selective elec-trocatalysts as electron mediators[23]to reduce the overpotential of electrochemical oxidation of hydro-gen peroxide,or constructing an additional membrane [24–27]on the tip of the electrode to prevent the diffusion of interferences.An attractive alternative approach to suppress the interferences has been developed by the construction of bienzymatic peroxidase/hydrogen peroxide produ-cing oxidase amperometric biosensors in the past few years[26,28–31].In such configurations,hydrogen peroxide is selectively reduced by peroxidase.The mechanism of the bienzymatic biosensors may be expressed as followssubstrate+O2oxidase→product+H2O2H2O2+(POD)red→H2O+(POD)oxwhere(POD)red and(POD)ox are the reduced and oxi-dized forms of peroxidase,respectively.The(POD)ox is directly electroreduced by electrode in mediatorless biosensor or by the oxidized form of mediator in me-diated biosensor.In this paper,we takes advantages of sol–gel process to construct a bienzymatic amper-ometric biosensor for glucose by coating a glucose oxidase/polypyrrole(GOD/PPy)membrane on the surface of ferrocenecarboxylic acid(FCA)-mediated horseradish peroxidase(HRP)-modified CCE.2.Experimental2.1.ChemicalsMethyltrimethoxysilane(MTMOS)and FCA were purchased from Acros.Graphite powder was obtained from Aldrich.GOD(EC1.1.3.4,type VII-S,245.9 units mg−1from Aspergillus niger)was obtained from Sigma.HRP was purchased from Shanghai Lizhu Dongfeng Biotechnology Co.Ltd.Pyrrole was freshly distilled prior to use and stored under nitrogen atmosphere.Double distilled water was used for the preparation of all phosphate buffer solutions(0.1M, containing0.1M KCl).Fresh stock solution of urate, lactate and oxalate were prepared before use and kept in the dark.Ascorbate was prepared immediately be-fore use.All other reagents were of analytical grade and used without further purification.2.2.ApparatusAn EG&G PARC model273potentiostat driven by an IBM PC with270software was used for electro-polymerization,voltammetric and amperometric mea-surements.A three-electrode cell with a SCE and a platinum foil counter electrode was used.The solu-tions were deaerated thoroughly for at least15min with pure N2and kept under a positive pressure of this gas during all experiments except the ampero-metric detection experiment.In the amperometric measurements,the buffer solution was magnetically stirred.All experiments were carried out at ambi-ent temperature.All potentials were measured and reported versus SCE.2.3.Preparation of HRP-modifiedFCA-mediated CCEGenerally high concentrations of methanol and/or strong acid are used in conventional sol–gel proce-dures employed in electroanalytical chemistry,which can cause enzymes to become denatured.Thus, the silica sols are usually prepared without adding methanol and with low concentration of acid as catalyst in constructing sol–gel derived biosensors.F .Tian,G.Zhu /Analytica Chimica Acta 451(2002)251–258253The silica sol–gel solutions used to construct CCEs were prepared as follows.A mixture of 1ml MTMOS,0.5ml water and 0.1ml HCl (0.01M)was sonicated for 10min,then,the homogeneous silica solution was stored over night at 4◦C in refrigerator.Twenty milligrams of HRP was dissolved in 0.5ml water and subsequently mixed with 160mg graphite powder in a mortar.The mix-ture was allowed to dry in a desiccator at 4◦C.The enzyme-modified graphite carbon,20mg FCA and 0.3ml silica sol–gel solution were thoroughly mixed.The mixture was packed into one end of a 3mm i.d.glass tube to a length of 5mm,subsequently allowed to dry and gel at least 5days at 4◦C.After being polished on weighing paper the surface of resulting HRP-modified FCA-mediated CCEs were smooth and shiny.Copper wire,inserted from the other end of glass tube,provided the electrical contact.2.4.Coating GOD/PPy on the surface of HRP-modified CCEThere are usually several electrochemical tech-niques used for direct electropolymerization including potentiostatic,galvanostatic and potentiodynamic.In the experiment,the galvanostatic mode was employed with a current density of 0.06mA cm −2in 0.1M phosphate buffer solution (pH 7.0)containing 0.25MFig.1.Cyclic voltammograms of HRP-modified FCA-mediated CCE in phosphate buffer solution (pH 7.0)at different scan rates:(1)5mV s −1;(2)20mV s −1and (3)50mV s −1.The inset shows the plot of anodic peak current vs.the square root of scan rate.pyrrole and 3.5mg ml −1GOD,which yielded films with more uniform thickness than in the potentiostatic mode.The solution was deaerated with N 2for 15min and left unstirred prior to electropolymerization.The bienzymatic biosensors were thoroughly washed after preparation and stored in phosphate buffer at 4◦C when not in use.3.Results and discussion3.1.Characteristics of bienzymatic glucose biosensor 3.1.1.The electrochemistry of HRP-modified FCA-mediated CCEThe performance of enzyme doped CCE was affected by the ratio of water to silicon in silica sol and the concentration of mediator.Under the optimized conditions in preparing enzyme doped CCE [32],the electrochemistry of HRP-modified FCA-mediated CCE was studied by cyclic voltam-metry.In the original several cycles,the peak to peak separation E p of FCA-mediated CCE was more than 350mV .With the diffusion of supporting electrolyte into the HRP-modified CCE,a thin active layer on the electrode surface was gradually come into being and reached equilibrium.Correspondingly,the redox couple of FCA became more reversible.Fig.1shows254F .Tian,G.Zhu /Analytica Chimica Acta 451(2002)251–258the cyclic voltammograms of HRP-modified FCA-mediated CCE at different scan rates.There are well-defined anodic and cathodic waves with mean peak potential E 1/2=(E pa +E pc )/2of 278mV .Inset shows the plot of peak current versus the square root of scan rate.A linear relation is recorded and the straight line does not pass though the origin,which suggests that the system is poorly diffusion controlled.3.1.2.Effect of PPy film thickness on the response of bienzymatic biosensorThe enzyme entrapment within polymer films can be accomplished non-covalently and covalently.Non-covalent entrapment is easily carried out by elec-tropolymerization of the monomers in the presence of enzyme.A relative large enzyme concentration is usually required in these non-covalent entrapments.As is known,PPy carries a net positive charge when being oxidatively polymerized and GOD is nega-tively charged in solutions with pH >4.2[33].Thus,GOD is prone to be entrapped within PPy membrane relying on the charge complimentarity between them.The amperometric response of PPy-based biosen-sor is largely affected by the amount of enzyme entrapped within the film and the thickness of film.These parameters can be controlled by manipulating the concentrations of enzyme,pyrrole in solution and the amount of charge duringelectropolymerization.Fig.2.Effect of passing charge during electropolymerization on the response of bienzymatic glucose biosensor towards 1.0×10−4M glucose.In this experiment,the electropolymerization was carried out in phosphate buffer solution (pH 7.0)con-taining 0.25M pyrrole and 3.5mg ml −1GOD accord-ing previous report [34].Fig.2shows the effect of electropolymerization charge on the response of bien-zymatic biosensor toward 10−4M glucose.It can be seen that a maximum response was achieved when the charge for electropolymerization was 10.18mC cm −2.The response increased with increasing of charge for electropolymerization during range from 4.60to 10.18mC cm −2.This may be due to the increasing of GOD within polymer membrane.When the passing charge was large than 10.18mC cm −2,because of the suppression of polymer membrane to the diffusion of analyte,the response decreased even though the amount of GOD entrapped within polymer membrane is known to be increased.Thus,all bienzymatic glu-cose biosensors were fabricated by coating GOD/PPy membrane on the surface of HRP-modified CCE with electropolymerization charge of 10.18mC cm −2in phosphate buffer solution (pH 7.0)containing 3.5mg ml −1GOD and 0.25M pyrrole.3.2.Response mechanism of bienzymatic biosensor for glucose detectionFig.3describes the proposed reaction mechanism of bienzymatic amperometric biosensor for the electro-F .Tian,G.Zhu /Analytica Chimica Acta 451(2002)251–258255Fig.3.Response mechanism of bienzymatic biosensor for glucose detection.enzymatic detection of glucose.Glucose,which dif-fuses from bulk solution into the GOD/PPy mem-brane,is catalyzed by GOD and dissolved dioxygen.The resulting hydrogen peroxide diffuses either to the surface of HRP-modified FCA-mediated CCE or back to the bulk buffer solution.The oxidized form of HRP (HRP ox )is produced by the enzymatic reaction of hydrogen peroxide and HRP,and subsequently reduced by the FCA to regenerate HRP.The resulting ferricinium ion (FCA +)is electroreduced to regene-rate FCA,producing the response current at the same time.The process involved in the glucose detection at the bienzymatic amperometric biosensor could be detailed as follows according to commonly recog-nized mechanism [35]glucose +GOD (FAD )→gluconolactone +GOD (FADH 2)Fig.4.Dependence of the bienzymatic biosensor response on applied potential to 1.0×10−4M glucose in phosphate buffer solution at pH 7.0.GOD (FADH 2)+O 2→GOD (FAD )+H 2O 2HRP +2H 2O 2→HRP ox +2H 2O HRP ox +2FCA →2FCA ++HRP 2FCA ++2e −→2FCAwhere FAD and FADH 2are the oxidized and reduced forms of flavine adenine dinucleotide,the active center of GOD,respectively.3.3.Effect of applied potential and buffer pH on the response of bienzymatic biosensorFig.4shows the effect of applied potential on the response of the biosensor.The amperometric detec-tion was carried out in 0.1M phosphate buffer solu-tion (pH 7.0)in the presence of 1.0×10−4M glucose.256F.Tian,G.Zhu/Analytica Chimica Acta451(2002)251–258Fig.5.Effect of pH on the bienzymatic glucose biosensor response to1.0×10−4M glucose.Applied potential160mV vs.SCE.It was found that the sensitivity of bienzymatic glu-cose biosensor increased with increasing potential from130to160mV and decreased during the po-tential range from160to190mV.The increased sensitivity with applied potential can be attributed to the increased driving force for the electroreduction of ferricinium ion,which is produced in enzymatic reac-tion.The highest sensitivity is acquired at potential of 160mV.The effect of phosphate buffer solution(con-taining1.0×10−4M glucose)pH on the response of bienzymatic glucose biosensor is shown in Fig.5, which shows the highest amperometric response at pH 6.9.On the basis of above results,the pH of6.9and potential of160mV were selected for the fol-lowing experiments in order to obtain the highest sensitivity.3.4.Selectivity of the bienzymaticglucose biosensorThe response from the bienzymatic glucose bio-sensor was also measured in the presence of some possible interfering substances(such as urate,lactate, oxalate,ascorbic acid,citrate and lactose).Table1 summarizes the effect of interferences on the biosen-sor response.The ratio of the amperometric response of mixtures of1.0×10−4M each interfering substance in the presence of1.0×10−4M glucose compared to that of1.0×10−4M glucose alone is taken as the criterion for the selectivity of the glucose biosensor. It can be seen that those substances cause hardly any interference on the response of the biosensor.In the hetero-bilayer configuration of the biosensor,the diffusion of interfering species from bulk solution to the electrode surface of CCE is slowed down by the PPy membrane.Especially,the selectively enzymatic reduction of hydrogen peroxide by HRP and the low operating potential for amperometric detection mainly contribute to the high selectivity of the glucose biosensor.Table1Effect of possible interfering substances on the bienzymatic glucose biosensor responsePossible interfering substance Current ratios aUrea0.99Lactate0.98Oxalate 1.00Ascorbic acid 1.05Citrate0.99Lactose 1.00a Current ratio=I mix/I glu.I mix:current of mixture of10−4M interference substance and10−4M glucose;I glu:current of10−4M glucose alone.F .Tian,G.Zhu /Analytica Chimica Acta 451(2002)251–258257Fig.6.Calibration curve of bienzymatic glucose biosensor in phosphate buffer solution (pH 6.9)at 160mV vs.SCE.The inset shows the linear range of the calibration curve.3.5.Calibration and stabilityThe calibration curve for the bienzymatic glu-cose biosensor (Fig.6)obtained by an amperometric response under optimized experimental conditions shows a linear response range between 8.0×10−5and 1.3×10−3M glucose with a sensitivity of 1.11␮A mM −1and a correlation coefficient of 0.998.It can be seen from Fig.6that the bienzymaticglucoseFig.7.Stability of the glucose biosensor stored at 4◦C.biosensor shows a high background current.This can be attributed to the hetero-bilayer configuration of the glucose ually the enzyme doped CCE shows a large background current because that the doped enzyme altered the hydrophilicity of the CCE surface [36].With the coating of PPy on the surface of enzyme doped CCE,the glucose biosen-sor shows a high background current.The apparentMichaelis–Menten constant (K appM ),which shows an258F.Tian,G.Zhu/Analytica Chimica Acta451(2002)251–258 indication of the enzyme–substrate kinetics,for theglucose biosensor is calculated to be0.25mM fromthe linear part of the calibration curve according to theEadie–Hoffstee form of the Michaelis–Menten equa-tionI ss=I max−K app MI ss Cwhere I ss is the steady-state current,I max maximum current under saturated substrate conditions and C bulk concentration of the substrate.The response time of biosensor towards glucose is less than25s.The detection limit,estimated as three-times the noise,is 1.0×10−5M for glucose.The stability of the bienzymatic glucose biosen-sor was studied by amperometric detection of1.0×10−4M glucose for every2–3days.Fig.7shows the effect of storage time on the relative response of the glucose biosensor.After1week,the biosensor retained 92%of the initial sensitivity when stored at4◦C.The biosensor showed a steady sensitivity during the fol-lowing time.After3weeks detection,the biosensor retained68%of initial sensitivity.Because of the rapid deactivation of enzymes and PPy membrane,the glu-cose biosensor lost its sensitivity rapidly when stored in air.4.ConclusionsA novel bienzymatic amperometric biosensor for glucose has been developed by electrodepositing GOD/PPy membrane on the surface of HRP-modified FCA-mediated CCE.The enzymes were well im-mobilized within the electrode matrices and retained satisfactory enzymatic catalytic activities.The hetero-bilayer glucose biosensor showed a good suppression of interferences(such as urate,oxalate,ascorbic acid, etc.)with favorable sensitivity towards glucose.The method combining electropolymerization with sol–gel derived CCE provides feasible choice for developing novel biosensors.References[1]L.L.Hench,J.K.West,Chem.Rev.90(1990)33.[2]M.D.Petit-Dominguez,H.Shen,W.R.Heineman, C.J.Seliskar,Anal.Chem.69(1997)703.[3]K.Kimura,T.Sunagawa,M.Yokoyama,Anal.Chem.69(1997)2379.[4]C.Y.Shen,N.M.Kostic,J.Am.Chem.Soc.119(1997)1304.[5]W.Song,N.Lu,Y.Jiang,Y.Liu,X.Wang,J.Liu,H.Xu,C.Sun,Microchem.J.62(1999)344.[6]B.Dave,B.Dunn,J.S.Valentine,J.Zink,Anal.Chem.66(1994)1120A.[7]O.Lev,M.Tsionsky,L.Rabinovich,V.Glezer,S.Sampath,I.Pankratov,J.Gun,Anal.Chem.67(1995)22A.[8]J.Lin,C.W.Brown,Trends Anal.Chem.16(1997)200.[9]M.M.Collinson, A.R.Howells,Anal.Chem.72(2000)702A.[10]D.Avnir,S.Braun,O.Lev,M.Ottolenghi,Chem.Mater.6(1994)1605.[11]J.Wang,Anal.Chim.Acta399(1999)21.[12]A.Walcarius,Electroanalysis10(1998)1217.[13]M.Tsionsky,G.Gun,V.Glezer,O.Lev,Anal.Chem.66(1994)1747.[14]J.Wang,P.V.A.Pamidi,Anal.Chem.69(1997)4490.[15]G.Oskam,P.C.Searson,J.Phys.Chem.B102(1998)2464.[16]J.Wang,P.V.A.Pamidi,D.R.Zanette,J.Am.Chem.Soc.120(1998)5852.[17]B.Wang,B.Li,Z.Wang,G.Xu,Q.Wang,S.Dong,Anal.Chem.71(1999)1935.[18]P.C.Pandy,S.Upadhyay,H.C.Pathak,Electroanalysis11(1999)59.[19]S.Sampath,O.Lev,Electroanalysis8(1996)1112.[20]S.Sampath,O.Lev,J.Electroanal.Chem.446(1998)57.[21]J.Wang,D.S.Park,P.V.A.Pamidi,J.Electroanal.Chem.434(1997)185.[22]J.Gun,O.Lev,Anal.Chim.Acta336(1996)95.[23]S.A.Wring,J.P.Hart,Analyst117(1992)1215.[24]G.G.Wallace,M.Smyth,H.Zhao,Trends Anal.Chem.18(1999)245.[25]S.Cosnier,A.Senillou,M.Grätzle,te,N.Vlacho-poulos,N.J.Renault,C.Martelet,J.Electroanal.Chem.469 (1999)176.[26]M.C.Shin,H.C.Yoon,H.S.Kim,Anal.Chim.Acta329(1996)223.[27]S.Hu,C.Xu,J.Luo,J.Luo,D.Cui,Anal.Chim.Acta412(2000)55.[28]L.Mao,K.Yamamoto,Anal.Chim.Acta415(2000)143.[29]E.F.Perez,G.Oliveira Neto,L.T.Kubota,Sens.Actuators B72(2001)80.[30]S.Cosnier,mbert,M.Stoytcheva,Electroanalysis12(2000)356.[31]M.Niculescu,C.Nistor,I.Frébort,P.Peˆc,B.Mattiasson,E.Csöregi,Anal.Chem.72(2000)1591.[32]J.Li,L.S.Chia,N.K.Goh,S.N.Tan,J.Electroanal.Chem.460(1999)234.[33]P.N.Barlett,J.M.Cooper,J.Electroanal.Chem.362(1993)1.[34]M.Umaña,J.Waller,Anal.Chem.58(1986)2979.[35]E.Csöregi,L.Gorton,G.Marko-Varga,Electroanalysis6(1994)925.[36]I.Pankratov,O.Lev,J.Electroanal.Chem.393(1995)35.。