Spraying Enzymes in Microemulsions of AOT in Nonpolar Organic

Web Lab, Novosibirsk IT Centre, Voskhod 26a, 630102 Novosibirsk, Russia, Department of ... Chemical Faculty, Moscow State University, 119992 Moscow, R...
0 downloads 0 Views 104KB Size
Download PDF
Anal. Chem. 2005, 77, 7074-7079

Technical Notes

Spraying Enzymes in Microemulsions of AOT in Nonpolar Organic Solvents for Fabrication of Enzyme Electrodes Stepan Shipovskov,†,‡ Daria Trofimova,§ Eduard Saprykin,‡ Andreas Christenson,⊥ Tautgirdas Ruzgas,⊥,| Andrey V. Levashov,§ and Elena E. Ferapontova*,‡,⊥,#

Department of Molecular Biophysics, Lund University, P.O. Box 124, SE-221 00 Lund, Sweden, Group of Bioinformatics, Web Lab, Novosibirsk IT Centre, Voskhod 26a, 630102 Novosibirsk, Russia, Department of Chemical Enzymology, Chemical Faculty, Moscow State University, 119992 Moscow, Russia, Department of Analytical Chemistry, Lund University, P.O. Box 124, SE-221 00 Lund, Sweden, Faculty of Health and Society, Malmo¨ University, SE-205 06 Malmo¨, Sweden

A new technique suitable for automated, large-scale fabrication of enzyme electrodes by air-spraying enzymes in organic inks is presented. Model oxidoreductases, tyrosinase (Tyr) and glucose oxidase (GOx), were adapted to octane-based ink by entrapment in a system of reverse micelles (RM) of surfactant AOT in octane to separate and stabilize the catalytically active forms of the enzymes in nonpolar organic media. Nonpolar caoutchouk polymer was also used to create a kind of “dry micelles” at the electrode/solution interface. Enzyme/RM/polymer-containing organic inks were air-brushed onto conductive supports and were subsequently covered by sprayed Nafion membranes. The air-brushed enzyme electrodes exhibited relevant bioelectrocatalytic activity toward catechol and glucose, with a linear detection range of 0.1100 µM catechol and 0.5-7 mM glucose; the sensitivities were 2.41 A M-1 cm-2 and 2.98 mA M-1 cm-2 for Tyr and GOx electrodes, respectively. The proposed technique of air-brushing enzymes in organic inks enables automated construction of disposable enzyme electrodes of various designs on a mass-production scale. Enzyme electrodes provide a specific and accurate detection of relevant analytes and nowadays are widely used in clinical practice,1,2 in the food and drink industries,3,4 and in agricultural5 * Corresponding author. E-mails: [email protected]; elena.ferapontova@ analykem.lu.se. † Department of Molecular Biophysics, Lund University. ‡ Novosibirsk IT Centre. § Moscow State University. ⊥ Department of Analytical Chemistry, Lund University. | Malmo ¨ University. # Present address: School of Chemistry, The University of Edinburgh, Joseph Black Building, West Mains Road, Edinburgh, EH9 3JJ, U.K. (1) Wang, J. J. Pharm. Biomed. Anal. 1999, 19, 47-53. (2) Wilson, G. S.; Hu, Y. Chem. Rev. 2000, 100, 2693-2704. (3) Mello, D.; Kubota, L. T. Food Chem. 2002, 77, 237-256. (4) Prodromidis, M. I.; Karayannis, M. I. Electroanalysis 2002, 14, 241-261. (5) Trojanowicz, M. Electroanalysis 2002, 14, 1311-1328.

7074 Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

and environmental analysis.6,7 The main requirements for currently designed enzyme electrodes are low cost, ease of assembly and capability of mass production, portability, stability under operation and during storage, and possibility for miniaturization. In many cases disposable, single-use test electrodes are required.8,9 Presently, thick-film screen-printing techniques are used as a main machine method for construction of disposable enzyme electrodes.8,10-12 Screen-printed thick-film electrodes comprise layers of metal- or carbon/graphite-based inks/pastes, which either may themselves contain an enzyme or may be covered by a layer of enzyme that is stepwise assembled onto the top layer of the electrode.8,10,12-17 In addition to traditional screen-printing, the benefits of piezoelectric and thermal ink-jet printing technologies18-20 and electrochemical co-deposition of enzymes with conductive and nonconductive polymers21-25 were demonstrated. (6) Ruzgas, T.; Csoeregi, E.; Emmneus, J.; Gorton, L.; Marko-Varga, G. Anal. Chim. Acta 1996, 330, 123-138. (7) Hanrahan, G.; Patil, D. G.; Wang, J. J. Environ. Monit. 2004, 6, 657-664. (8) Bilitewski, U.; Chemnitius, G. C.; Rueger, P.; Schmid, R. D. Sens. Actuators, B 1992, B7, 351-355. (9) Magner, E. Analyst 1998, 123, 1967-1970. (10) Alvarez-Icaza, M.; Bilitewski, U. Anal. Chem. 1993, 65, 525A-533A. (11) Rohm, I.; Kuennecke, W.; Bilitewski, U. Anal. Chem. 1995, 67, 23042307. (12) Hart, A. L.; Collier, W. A.; Janssen, D. Biosens. Bioelectron. 1997, 12, 645654. (13) Cardosi, M. F.; Birch, S. W. Anal. Chim. Acta 1993, 276, 69-74. (14) Schmidt, A.; Rohm, I.; Rueger, P.; Weise, W.; Bilitewski, U. Fresenius’ J. Anal. Chem. 1994, 349, 607-612. (15) Nagat, R.; Yokoyama, K.; Clark, S. A.; Karube, I. Biosens. Bioelectron. 1995, 10, 261-267. (16) Stiene, M.; Bilitewski, U. Anal. Bioanal. Chem. 2002, 372, 240-247. (17) Mersal, G. A. M.; Khodari, M.; Bilitewski, U. Biosens. Bioelectron. 2004, 20, 305-314. (18) Newman, J. D.; Turner, A. P. F. Anal. Chim. Acta 1992, 262, 13-17. (19) Allain, L. R.; Stratis-Cullum, D. N.; Vo-Dinh, T. Anal. Chim. Acta 2004, 518, 77-85. (20) Setti, L.; Piana, C.; Bonazzi, S.; Ballarin, B.; Frascaro, D.; Fraleoni-Morgera, A.; Giuliani, S. Anal. Lett. 2004, 37, 1559-1570. (21) Schuhmann, W. Immobilized Biomol. Anal. 1998, 187-210. (22) Schuhmann, W. Mikrochim. Acta 1995, 121, 1-29. (23) Cosnier, S. Biosens. Bioelectron. 1999, 14, 443-456. (24) Kurzawa, C.; Hengstenberg, A.; Schuhmann, W. Anal. Chem. 2002, 74, 355-361. 10.1021/ac050505d CCC: $30.25

© 2005 American Chemical Society Published on Web 09/30/2005

The applicability of any of these manufacturing procedures can be restricted either by insufficient stability of the enzyme in the ink/paste medium or by inadequate technical control of the electrode performance.26 As an alternative, direct spraying of enzyme-containing organic inks onto solid supports is of evident interest for industrial production of enzyme electrodes. This procedure simplifies electrode construction and offers the possibility of a minimized thickness of the applied enzyme layer, economizing on the amount of the enzyme used; quick evaporation of the organic ink phase shortens the time of the electrode preparation. On the other hand, the spraying technique has a number of limitations, namely, stability of enzymes both in dispersed state and in nonpolar organic media used as a base phase of a sprayed aerosol. Most enzymes rapidly lose their catalytic activity at relatively low concentrations of the majority of the organic solvents.27-30 A certain minimal amount of water, playing mainly a structural role, is essential to maintain biocatalysis.29,31 Systems of organic solvents containing enzymes entrapped in reverse micelles (RM) formed by surfactant molecules have been shown to be promising as media for efficient bioseparation and biocatalysis.29,32-34 The RM represent a water microemulsion and can be considered as a pool of water stabilized by amphiphilic surfactant molecules with their hydrophilic polar “heads” facing water inside the micelles and their hydrophobic nonpolar “tails” oriented toward surrounding nonpolar organic media. For a variety of enzymes, for example, R-chymotrypsin,35,36 laccases,37 peroxidases,38 and aldehyde oxidoreductase,39 the efficient separation and stabilization of enzyme catalytic activity in nonpolar organic media by enzyme entrapping in the inner water-containing cavities of the surfactant RM were reported.32,33,40,41 The enzymes tyrosinase (Tyr)42,43 and glucose oxidase (GOx),44 when entrapped in RM of surfactants in organic (25) Neugebauer, S.; Isik, S.; Schulte, A.; Schuhmann, W. Anal. Lett. 2003, 36, 2005-2020. (26) Albareda-Sirvent, M.; Merkoci, A.; Alegret, S. Sens. Actuators, B 2000, 69, 153-163. (27) Carrea, G.; Riva, S. Angew. Chem., Int. Ed. 2000, 39, 2226-2254. (28) Gupta, M. N. Eur. J. Biochem. 1992, 20, 25-32. (29) Faber, K. Biotransformation in Organic Chemistry; Springer: Germany, 1997. (30) Klibanov, A. Chemtech 1986, 16, 354-359. (31) Deng, Q.; Dong, S. Anal. Chem. 1995, 67, 1357-1360. (32) Martinek, K.; Levashov, A. V.; Khmelnitsky, Y. L.; Klyachko, N. L.; Berezin, I. V. Science 1982, 218, 889-891. (33) Klyachko, N. L.; Levashov, A. V.; Kabanov, A. V.; Khmelnitsky, Y. L.; Martinek, K. In Kinetics and Catalysis in Microheterogeneous Systems; Gra¨tzel, M., Kalyanasundaram, K., Eds.; Marcel Dekker, Inc.: New York, 1991; pp 135-181. (34) Klyachko, N. L.; Levashov, A. V. Curr. Opin. Colloid Interface Sci. 2003, 8, 179-186. (35) Levashov, A. V.; Klyachko, N. L.; Bogdanova, N. G.; Martinek, K. FEBS Lett. 1990, 268, 238-240. (36) Biswas, R.; Pal, S.-K. Chem. Phys. Lett. 2004, 387, 221-226. (37) Khmelnitsky, Y. L.; Gladilin, A. K.; Roubailo, V. L.; Martinek, K.; Levashov, A. V. Eur. J. Biochem. 1992, 206, 737-745. (38) Gazaryan, I. G.; Klyachko, N. L.; Dulkis, Y. K.; Ouporov, I. V.; Levashov, A. V. Biochem. J. 1997, 328, 643-647. (39) Andrade, S. L. A.; Brondino, C. D.; Kamenskaya, E. O.; Levashov, A. V.; Moura, J. J. G. Biochem. Biophys. Res. Commun. 2003, 308, 73-78. (40) Levashov, A. V. Pure Appl. Chem. 1992, 64, 1125-1128. (41) Vinogradov, A. A.; Kudryashova, E. V.; Levashov, A. V.; van Dongen, W. M. A. M. Anal. Biochem. 2003, 20, 234-238. (42) Shipovskov, S.; Ferapontova, E.; Ruzgas, T.; Levashov, A. Biochim. Biophys. Acta 2003, 1620, 119-124. (43) Shipovskov, S.; Levashov, A. Biocatal. Biotransform. 2004, 22, 57-60. (44) Kamyshny, A.; Trofimova, D.; Magdassi, S.; Levashov, A. Colloids Surf., B 2002, 24, 177-183.

solvents, can exhibit catalytic activities comparable to those exhibited in the aqueous state. Tyr is a bifunctional tetrameric copper enzyme with a molecular weight of 128-133 kDa.45 Tyr catalyzes the oxidation of phenolic compounds in two steps, first through the orthohydroxylation of monophenols and second through the dehydrogenation of further formed o-diphenols (catechols) to the related o-quinones as final reaction products, while reducing molecular oxygen to water.46 Tyr is used intensively in construction of enzyme electrodes for detection of (poly)phenolics. The majority of Tyr sensors exploit principles of facile biological recognition of phenolic compounds by Tyr and electrochemical recycling of the resulting quinone products at electrodes.47-51 GOx is a homodimer of molecular weight 150-160 kDa, each monomer unit containing one FAD center.52 GOx catalyzes oxidation of β-D-glucose by dioxygen, producing D-gluconolactone, which in turn hydrolyzes to gluconic acid and H2O2. GOx is the most extensively studied in glucose enzyme electrodes due to both the high practical relevance of glucose determinations in clinical practice and the food and microbiological industries and relatively high durability of the enzyme.2,53 The first enzyme electrodes for glucose analysis used membrane- or gel-entrapped GOx coupled to a Clark oxygen electrode.54-56 The majority of commercial GOxbased sensors still exploit the basic principle of glucose determination by amperometrically measuring oxygen consumption or, alternatively, the formation of H2O2 during enzymatic transformation of glucose.4,9,54-56 In the present work, an innovative method is presented for fabrication of enzyme electrodes by direct spraying of enzymes in microemulsions of a surfactant in nonpolar organic inks. Tyr and GOx were chosen as model enzymes for their development due to their numerous applications in amperometric enzyme electrodes. Additionally, the stability of the catalytic function of these enzymes varies substantially between differing media. For example, because of inactivation by organic solvents, Tyr rapidly loses its catalytic activity in dry organic media30,31 and is, thus, highly relevant for studies of the enzyme-stabilization possibilities yielded by the proposed spraying technique. In contrast to Tyr, GOx was shown to be relatively stable in many organic solvents.57-59 This enables further investigation of the working behavior of sprayed enzyme electrodes as a function of the spraying procedure. The enzyme-containing organic inks were air-brushed onto the surface of disk metal electrodes, and the function of the (45) Kertesz, D.; Zito, R. Biochim. Biophys. Acta 1965, 96, 447-462. (46) Duckworth, H. W.; Coleman, J. E. J. Biol. Chem. 1970, 247 (245), 16131617. (47) Zacharial, K.; Mottola, H. A. Anal. Lett. 1989, 22, 1145-1158. (48) Kulys, J.; Schmid, R. D. Anal. Lett. 1990, 23, 589-597. (49) Skladal, P. Collect. Czech. Chem. Commun. 1991, 56, 1427-1433. (50) Cosnier, S.; Innocent, C. Bioelectrochem. Bioenerg. 1993, 31, 147-160. (51) Wang, J.; Fang, L.; Lopez, D. Analyst 1994, 455-458. (52) Hecht, H. J.; Kalisz, M. H.; Hendle, J.; Schmid, R. D.; Schomburg, D. J. Mol. Biol. 1993, 229, 153-172. (53) Wang, J. Electroanalysis 2001, 13, 983-988. (54) Clark, L. C. J.; Lyons, C. Ann. N.Y. Acad. Sci. 1962, 102, 29-45. (55) Updike, S. J.; Hicks, G. P. Nature 1967, 214, 986-988. (56) Clark, L. C. J.; Sachs, G. Ann. N.Y. Acad. Sci. 1968, 148, 133. (57) Iwuoha, E. I.; Smyth, M. R. NATO ASI Ser., Ser. E 1993, 252, 245-254. (58) Iwuoha, E. I.; Smyth, M. R.; Lyons, M. E. G. J. Electroanal. Chem. 1995, 390, 34-45. (59) Campanella, L.; Favero, G.; Sammartino, M. P.; Tomassetti, M. Talanta 1998, 46, 595-606.

Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

7075

enzymes electrodes so obtained was examined in the presence of the corresponding enzyme substrates. EXPERIMENTAL SECTION Materials. Tyrosinase from mushrooms (polyphenoloxidase, EC 1.14.18.1, Tyr), the anionic surfactant dioctyl sulfosuccinate sodium salt (Aerosol OT, AOT), and catechol were obtained from Sigma (St. Louis, MO). Graphite powder (1-2 µm, synthetic) was from Aldrich (Gillingham, Dorset, UK). Glucose oxidase from Aspargillus niger (EC 1.1.3.4, GOx) was acquired from Biozyme Laboratories (Blaenavon, South Wales, UK). Caoutchouk (1,4polyisoprene standard 600 000) was from Fluka, and a 20% solution of Nafion in ethanol was from Aldrich (both Steinheim, Germany). Octane, ethanol, acetonitrile, and components of buffer solutions were from Merck (Darmstadt, Germany). All reagents were of analytical grade and were used as received. The solutions were prepared with deionized Milli-Q water (Millipore, Bedford, MA). Preparation of Reverse Micelles. Preparation of RM containing Tyr or GOx was described elsewhere.42,44 Briefly, enzyme microemulsions were prepared by adding either 10 µL of a 0.4 µM solution of Tyr in a citrate-phosphate buffer solution or 10 µL of a 30 µM solution of GOx in an acetate-phosphate buffer solution (aqueous systems) to 2 mL of 0.1 M AOT solution in octane (solvent-surfactant solution). For preparation of RM stabilized by caoutchouk, a 2.5% solution of caoutchouk in octane was mixed with 0.1 M AOT in octane in the proportion 1:2, and this solvent-surfactant solution was used for preparation of RM. The required values of the water/surfactant molar ratio w0 ) [H2O]/[AOT] were achieved by addition of a certain amount of water (or an aqueous enzyme solution) to the solvent-surfactant solution. The mixtures were thoroughly shaken until homogeneous, optically transparent micellar solutions were obtained. Activity Assay. The catalytic activity of Tyr in different environments was followed by monitoring spectrophotometrically the conversion of catechol to o-quinone. O-Quinone is characterized by molar absorptivity  ) 760 M-1 cm-1 at wavelength of 405 nm. A 10-µL portion of 0.4 µM Tyr in citrate-phosphate buffer, along with 5 µL of various concentrations of catechol (2-4000 mM) in acetonitrile were added to 2 mL of a recipient solution. To monitor Tyr activity in aqueous buffer, the recipient solution was citrate-phosphate buffer held at various pH values (3.07.5). To monitor Tyr activity in an RM system, the recipient solution was 0.1 M AOT in octane with various amounts of pH 6.5 citrate-phosphate buffer to yield different degrees of hydration, w0. The catalytic activity of GOx in different environments was followed by monitoring spectrophotometrically the conversion of pyrogallol to the corresponding quinone product, which is characterized by molar absorptivity  ) 4400 M-1 cm-1 at a wavelength of 420 nm. In the assay, 10 µL of 30 µM GOx in acetate-phosphate buffer (pH 7.0), 5 µL of 0.4 M pyrogallol solution in acetone, and 5 µL of 25 µM peroxidase solution were together added to 2 mL of a recipient solution. The enzymatic GOx reaction was then initiated by addition of 5-50 µM of 40 mM β-D(+)glucose. To monitor GOx activity in aqueous buffer, the recipient solution was 20 mM acetate-phosphate buffer held at various pH values (3.0-6.0). To monitor GOx activity in an RM system, the recipient solution was 0.1 M AOT in octane, pH 5.5, with various amounts of 20 mM acetate-phosphate buffer, 5-200 7076

Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

µL, to achieve different degrees of hydration, w0. Before the measurements, the reaction mixtures were thoroughly shaken in a tightly closed cuvette until optically transparent homogeneous solutions were obtained. An Ultrospec II spectrophotometer (LKB, Bromma, Sweden) was used. The kinetic analysis was carried out according to the classical Michaelis-Menten equation by using nonlinear regression: v ) (kappE0S0)/(Km + S0), where v is the reaction rate; E0 and S0 are the enzyme and substrate initial molar concentrations, respectively; kapp is the first-order apparent catalytic constant; and Km is the Michaelis constant. Electrode Preparation and Electrochemical Measurements. The Au or Pt disk electrodes (L 0.2 cm, CH-Instruments, Austin, TX) were polished first with fine emery paper (Tufbak Durite, P1200) and then to a mirror luster with alumina slurry (0.1 µm, Stuers, Copenhagen, Denmark) using a Microcloth pad (Bauer, Germany). After a short rinse with water and ultrasonication in acetone and water, the electrodes were electrochemically cleaned in 1 M H2SO4 by potential cycling between -0.2 and 1.7 V, the final voltammograms being similar to those reported for clean Au and Pt.60 Enzyme-containing organic inks were sprayed onto the electrodes through a steel-foil mask of an appropriate size by using an HP-100B airbrush (0.2-mm nozzle, Olympos, Japan), illustrated in the Supporting Information. The airbrush was driven by compressed air (1.6 bar), providing an airflow of 5-10 L min-1. An airflow valve relay (Bu¨rkert, Germany) was employed to switch the spray on and off at determined intervals. Enzyme- and graphite-containing organic inks were prepared by mixing the enzyme/RM/caoutchouk solution in octane (w0 of 25) with graphite powder; the final concentration of graphite in the inks was 6% (w/w). The inks were sprayed onto the surface of Au (Tyr) and Pt (GOx) disk electrodes and allowed several seconds to dry. The enzyme layers were finally covered with Nafion films by further air-spraying of a 0.1% Nafion solution in ethanol. Cyclic voltammetry (CV) and chronoamperometric (CA) measurements were performed in a standard three-electrode cell connected to an Autolab potentiostat (PGSTAT 30, Eco Chemie, The Netherlands) equipped with GPES 4.9 software. An Ag/AgCl(KClsat) electrode and a Pt wire were used as the reference and auxiliary electrodes, respectively. The enzyme electrode served as the working electrode. CV and CA with enzyme electrodes were done either in 4 mM citrate-phosphate buffer solution (Tyr) or in 0.1 M phosphate buffer containing 0.15 M NaCl (GOx) at pH 6.0. CA with Tyr electrodes was performed at -0.05 V and with GOx electrodes at +0.6 V. For 10 similarly produced air-sprayed electrodes, the relative standard deviation of their response was < 7%. All experiments were done at ambient temperature (22 ( 1 °C). The apparent Km values were obtained by direct fitting of the calibration plot data to the MichaelisMenten equation, Imeasured ) ImaxS0/(Km + S0). The sensitivity of the enzyme electrodes was then calculated as Imax/Km related to the geometric surface area of the electrodes. RESULTS AND DISCUSSION The proposed procedure for fabrication of enzyme electrodes involved air-spraying of enzyme solutions in organic inks through (60) Hoare, J. P. Electrochem. Soc.: Electrochem. Sci. Technol. 1984, 18081819.

Table 1. Apparent Km Values, mM, for Tyr (pH 6.5) and GOx (pH 5.5) in Aqueous Media, in RM/Octane Media (w0 of 25), and after Air-Spraying in Organic Inks (Two Sprayed Layers) enzyme/media

aqueous solution

RM/octane

ink/electrode

tyrosinase glucose oxidase

4.1 15.7

2.5 1.2

0.14 6.2

Figure 1. Dependence of the catalytic activity kapp of (a) Tyr and (b) GOx on the degree of hydration w0 of RM of 0.1 M AOT in octane. Dashed lines correspond to the catalytic activities of the enzymes in the aqueous solutions, at (a) pH 6.5 and (b) pH 5.5, respectively.

appropriate masks onto solid supports. The procedure was found to be reproducible and quick (because of fast evaporation of organic phase from the electrode surface); it provided efficient and stable bioelectrocatalytic function of enzymes. To stabilize enzymes in organic inks, the entrapment of Tyr and GOx in a system of RM of surfactant AOT in a nonpolar organic solvent was employed. Catalytic Activity of Enzymes in Aqueous and Micellar Systems. pH-Dependence of the Catalytic Activities of Tyr and GOx in Aqueous Media. The specific enzymatic activity kapp of both Tyr and GOx exhibited a characteristic bell-shaped dependence on pH, with a maximum at pH 6.5 for Tyr (the enzyme still displays high values of activity over the range of pH 6.5-7.5) and at pH 5.5 for GOx, in good agreement with existing data for Tyr61 and GOx.62 Further studies of micellar catalysis were performed in buffer solutions at pH 6.5 (Tyr) and pH 5.5 (GOx). Catalytic Activity of Tyr and GOx in a System of Reversed Micelles. A particular type of bioassembly is constituted when any protein is solubilized in RM. In a first approximation, the structure of these aggregates depends on the water content and the nature of the enzyme; under optimum conditions, the enzyme activity may approach that in an aqueous medium. Microemulsions of Tyr and GOx in RM of AOT in octane were examined as a function of the degree of hydration, w0. Both enzymes remained catalytically active in the RM/octane media. The dependence of the specific enzymatic activity, kapp, on w0 for Tyr in RM of 0.1 M AOT in octane is presented in Figure 1a. Although in dry octane enzymatic activity of Tyr is drastically depressed due to the enzyme inactivation by the organic solvent, the activity of Tyr in RM increases with an increasing water/ surfactant molar ratio and saturates at a w0 higher than 20, the limiting value still being lower than the activity of Tyr in aqueous (61) Brown, R. S.; Male, K. B.; Luong, J. H. T. Anal. Biochem. 1994, 222, 131139. (62) Boyer, P. D.; Lardy, H.; Myrback, K.; Bentley, R. The Enzymes; Academic Press: New York, 1963.

Figure 2. Time dependences of the apparent catalytic activities kapp of Tyr in the aqueous (pH 6.5, b) and RM/octane (O) systems, and those of GOx in the aqueous (pH 5.5, 9) and RM/octane (0) systems. The degree of hydration w0 is 25.

solutions. At w0’s of 12 and 25, the profile of kapp displays extremes, the maximum value at w0 ) 25 approaching the value characteristic for the aqueous media. w0 ) 25 corresponds to the size of a tetrameric form of Tyr,45 whereas w0 ) 12 corresponds to a catalytically active monomeric form of Tyr.43 The catalytic activity of GOx (Figure 1b) is not so much depressed by the organic solvent as in the case of Tyr, but it also increases with increasing w0 and approaches the level characteristic of the aqueous media at w0 g 15. Dynamic light-scattering studies44 evidenced that the mean size of RM at w0 ) 14-25 is optimal for entrapment of the GOx dimer molecule, ∼7 nm in diameter,63 whereas at higher values of w0, RM may incorporate several GOx molecules. It can be seen that both Tyr and GOx retain their catalytic activity in the RM/octane systems and that this activity depends on the water/surfactant molar ratio, w0. Low values of w0 are compatible with incorporation of single molecules of enzymes in the RM system, whereas higher values of w0 may provide conditions for entrapment of multimolecular assemblies. For both enzymes, depending on the varying solubility of their substrates in organic media, the apparent values of Km decreased from 1.6 to 12 times when changing from the aqueous solutions to the RM/ octane media (Table 1). For follow-up studies, RM systems with w0 ) 25 were chosen. Stability of Tyr and GOx in the Aqueous and Micellar Systems. Both Tyr and GOx lose from 20 to 25% of their catalytic activity when stored for 3-4 days in aqueous solutions under ambient conditions (Figure 2). The loss of their catalytic activity mainly occurs within the first 5-7 h, with more stable values reached on a longer time scale. Entrapment of GOx in the RM system (63) Baszkin, A.; Boissonnade, M. M.; Rosilio, V.; Kamyshny, A.; Magdassi, S. J. Colloid Interface Sci. 1997, 190, 313-317.

Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

7077

enabled somehow the stabilization of the activity of GOx, which decreased only 10% during 55 h. At extended times, the enzyme activity in the RM system became coincident with the activity in the aqueous medium. Compared to GOx, stability of Tyr in the micellar system was not so high (Figure 2) and decreased more than 50% within first 55 h, with a tendency for further decrease at longer times. Implementation of the Air-Spraying Technique for Construction of Enzyme Electrodes. Several schemes for preparation of enzyme electrodes by spraying enzyme-containing organic inks have been developed. All of them involved development of inks; air-spraying of the inks through a mask onto an electrode surface; and after evaporation of the organic-phase component from the enzyme layer, final air-spraying of a polymer solution to cover the enzyme layer with a membrane film. Tyr and GOx entrapped in RM of AOT in mixed nonpolar lowand high-molecular-weight organic phase, namely, octane and caoutchouk, were used to develop inks. As was previously shown,42 films created from RM of AOT in low-weight organic phase exhibited low stability at the electrode/solution interface after organic solvent evaporation. RM were sequentially degraded, and the enzyme was gradually desorbed from the electrode surface, resulting in diminishing activity of the enzyme electrodes over time. Caoutchouk was used to obtain a kind of “dry micelles” at the electrode surface after evaporation of a low-weight organic phase. This stabilized the enzyme-RM system by providing a nonpolar organic phase matrix.42 Graphite was alternatively added, as well. Inks prepared in these ways were sprayed onto the surfaces of metal electrodes and successively covered by the Nafion membrane to prevent the enzyme leaking into the polar bulk solution.42 CVs of the air-sprayed Tyr electrodes in the presence of catechol are presented in Figure 3. Catalytic waves of dioxygen reduction mediated by catechol were observed with the electrodes obtained by spraying Tyr/RM/caoutchouk organic inks without (Figure 3a) and with graphite (Figure 3b). Despite differences in the background signals, particularly pronounced in the case of sprayed graphite electrodes, both Tyr/Au and Tyr/graphite/Au electrodes demonstrated bioelectrocatalytic responses to catechol within the microampere current range. However, during measurements in stirred solutions, enzyme films without graphite were eventually desorbed from the Au surfaces. In contrast, the films obtained by spraying graphite inks demonstrated good mechanical stability, with maximal activity of Tyr electrodes achieved after two consecutive spraying procedures. When Tyr was not entrapped in RM, no bioelectrocatalytic activity of sprayed enzyme electrodes could be followed. The response of Tyr electrodes was calibrated against the concentration of catechol (Figure 3, insets). The sensitivity of the graphite electrodes was 2.41 ( 0.22 A M-1 cm-2 (Imax ) 10.4 µA), and the detection limit was 0.1 µM of catechol. Electrochemical recycling of catechol, directly involved in the Tyr catalytic cycle, provided Km values at least an order of magnitude lower, as compared to the solution chemistry of Tyr (Table 1). Linearity of the response was observed within a concentration range of 0.1100 µM; the electrodes did not lose their initial activity over the course of a 2-h operation. Keeping the enzyme electrodes in ambient conditions (from 24 to 48 h) resulted in 30% loss of the 7078 Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

Figure 3. A. CVs of (1, 2) the RM/caoutchouk Au electrode and (3) the Tyr/RM/caoutchouk Au electrode covered with Nafion membranes; (2, 3) in the presence of 50 µM catechol (pH 6.0, scan rate 50 mV s-1). Inset: Catechol calibration curve constructed using the current response in the maximum of the reduction peak in CVs at -0.15 V upon the increasing catechol concentration. B. CVs of the Tyr/RM/caoutchouk graphite electrode covered with a Nafion membrane (1) in the absence and (2) in the presence of 10 µM catechol in magnetically stirred solutions (pH 6.0, scan rate 50 mV s-1). Inset: Amperometric responses of the Tyr electrode to successive additions of catechol (applied potential -50 mV).

initial bioelectrocatalytic activity of the electrodes, which correlated with the inherent instability of Tyr itself (Figure 2). Enzyme electrodes were also fabricated using GOx as a model enzyme possessing much higher stability compared to Tyr (Figures 1, 2). Electrodes prepared by spraying GOx/RM/ caoutchouk- and graphite-containing organic inks onto the surface of Pt were probed during glucose determination by anodically oxidizing the enzymatically formed H2O2 at +0.6 V. Practically, the amperometric response of GOx electrodes depended both on the amount of the sprayed enzyme and on the diffusion of substrates into the enzyme layer. Similarly to Tyr electrodes, the activity of GOx electrodes depended on the number of sprayed layers (see Supporting Information). The maximal activity was achieved with two sprayed layers, providing an increased sensitivity for glucose from 0.79 (one layer, Imax ) 0.44 µA, Km ) 17.9 mM) to 2.98 mA M-1 cm-2 (two layers, Imax ) 0.58 µA, Km ) 6.2 mM). Due to oxygen diffusion limitations, glucose calibration curves were linear only up to 7 mM of glucose. The apparent Km values ranged between those obtained in the aqueous phase and in the RM/octane system (Table 1). Upon further spraying, the

increasing thickness of the enzyme film resulted in decreased responses due to the restricted diffusion of the reactants through the film. The response of electrodes stored under ambient conditions correlated with a 10% decreasing activity of GOx (Figure 2). In conclusion, the proposed technique combining RM systems and airbrush spray deposition was shown to be efficient for fabrication of enzyme electrodes: both Tyr and GOx retained their catalytic activity after air-brushing of organic inks, which contained the enzymes in microemulsions of AOT in nonpolar organic media. By air-spraying oxidoreductases catalytically stabilized by RM in a high-weight nonpolar organic polymer matrix, enzyme-containing films with controllable bioelectrocatalytic activity may be produced at metal conductive supports. The sensitivity of the enzyme electrodes and the apparent values of Km depended both on the spraying procedure and on the detection mode. The electrochemical recovery of a substrate considerably reduced the Km; otherwise, values characteristic for catalysis in solution were observed. The stability of the sprayed electrodes under storage directly correlated with the inherent stability of the enzymes in RM. The

spraying technique can be extended to a variety of other oxidoreductases, primarily including the enzymes possessing low stability toward inactivation in organic media. The proposed technique allows extending the application of enzymes in organic solvents from laboratory in vitro experiments to industrial-scale enzyme-electrode production. ACKNOWLEDGMENT The work was supported by the EU Project INTELLISENS (QLK3-CT-2000-01481), the Russian Educational Program, the Swedish Institute, and the Wenner-Gren Foundations. SUPPORTING INFORMATION AVAILABLE Additional material as described in the text. This material is available free of charge via the Internet at http://pubs.acs.org.

Received for review March 25, 2005. Accepted August 29, 2005. AC050505D

Analytical Chemistry, Vol. 77, No. 21, November 1, 2005

7079