Anal. Chem. 1997, 69, 968-971
An Electrochemical Method for Making Enzyme Microsensors. Application to the Detection of Dopamine and Glutamate Serge Cosnier,*,† Christophe Innocent,† Laurence Allien,‡ Serge Poitry,‡ and Marco Tsacopoulos‡
Laboratoire d’Electrochimie Organique et de Photochimie Redox, UMR CNRS 5630, Universite´ Joseph Fourier BP 53, 38041 Grenoble Cedex 9, France, and Experimental Ophthalmology Laboratory, Centre Me´ dical Universitaire de Gene´ ve, 1 rue Michel-Servet, 1211 Geneva 4, Switzerland
A novel method of microbiosensor fabrication is described. It is based on the electrochemical polymerization of an enzyme-amphiphilic pyrroleammonium solution on the surface of a microelectrode in the absence of supporting electrolyte. By trapping glutamate oxidase (GMO) or polyphenol oxidase (PPO) in such polypyrrole films, we made microbiosensors for the amperometric determination of glutamate or dopamine, respectively. The response of the GMO microelectrode to glutamate was based on the amperometric detection of the enzymically generated hydrogen peroxide at 0.6 V vs SCE. The detection limit and sensitivity of this microbiosensor were 1 µM and 32 mA M-1 cm-2, respectively. The response of the PPO microelectrode to dopamine was based on the amperometric detection of the enzymically generated quinoid product at -0.2 V. The calibration range for dopamine measurement was 5 × 10-8-8 × 10-5 M and the detection limit and sensitivity were 5 × 10-8 M and 59 mA M-1 cm-2, respectively.
attractive way for fabricating microsensors.5 One major advantage of electrochemical deposition procedures over more conventional methods is the possibility to precisely electrogenerate a polymer coating over small electrode surfaces of complex geometry. Furthermore, this method enables exact control of the thickness of the polymer layer based on measurement of the electrical charge passed during the electrochemical polymerization.6,7 However, the electrical entrapment of an enzyme in conducting polymer films requires high concentrations of monomer, enzyme, and supporting electrolyte during the electropolymerization process.6-8 This is because enzyme incorporation in the growing polymer film is solely due to the presence of enzyme in the immediate vicinity of the electrode surface and does not result from specific electrostatic interactions between the electrogenerated polymer and the enzyme molecules. In order to improve the amount of entrapped enzymes in polymer matrices, we report here a slightly different approach of electrochemical enzyme entrapment based on the amphiphilic pyrrole monomer 1: CH3
The application of amperometric enzyme electrodes in chemical analysis continues to be the subject of considerable research interest.1-3 In particular, the development of miniaturized enzyme electrodes is an attractive avenue for in vivo measurements of metabolites such as glucose, hormones, and neurotransmitters.4 Since metabolites can vary substancially over a period of a few minutes, such microbiosensors that can function in the tissues or bloodstream may become a powerful tool for clinical and neurochemical monitoring. In addition, they should provide the means to characterize the neurochemical microenvironment during nerve cell stimulation and pharmacological manipulations. Many strategies including direct adsorption, cross-linking with glutaraldehyde, covalent binding, and entrapment in polymeric gels or carbon paste have been developed for the immobilization of enzymes on electrode microsurfaces. Among these various immobilization procedures, the entrapment of enzymes in electrogenerated conducting polymer films appears as a simple and †
Universite´ Joseph Fourier. Centre Me´dical Universitaire de Gene`ve. (1) Turner, A. P. F., Karube, I., Wilson, G. S., Eds. Biosensors: Fundamentals and Applications; Oxford University Press: New York, 1987. (2) Guilbault, G. G., Mascini, M., Eds. Uses of Immobilized Biological Compounds; NATO ASI Series 252; Kluwer Academic Publishers: New York, 1993. (3) Wang, J. Anal. Chem. 1995, 67, 487R-492R. (4) See, for example: Abe, T.; Lau, Y. Y.; Ewing, A. G. Anal. Chem. 1992, 64, 2160-2163. ‡
968 Analytical Chemistry, Vol. 69, No. 5, March 1, 1997
N
(CH2)12
N
CH3 BF4– CH3
1
In contrast to the main classical monomers such as pyrrole, 1,2diaminobenzene, aniline, or phenol, this amphiphilic monomer possesses a net positive charge and can be electropolymerized in aqueous solution without a supporting electrolyte to promote the incorporation of negatively charged enzymes during polymer formation. The capacity of the cationic monomer 1 to be used in microbiosensor fabrication was studied via the immobilization of glutamate oxidase (GMO) and polyphenol oxidase (PPO) on platinum and carbon microelectrodes, respectively. The analytical characteristics of the resulting microbiosensors toward the measurement of glutamate and dopamine is described. EXPERIMENTAL SECTION Reagents. Polyphenol oxidase (EC 1.14.18.1, from mushroom, 3800 units mg-1) was purchased from Sigma. Glutamate oxidase (5) Bartlett, P. N.; Cooper, J. M. J. Electroanal. Chem. 1993, 362, 1-12. (6) Trojanowicz, M.; Krawczynski Vel Krawczyk, T. Mikrochim. Acta 1995, 121, 167-181 and references cited therein. (7) Schumann, W. Mikrochim. Acta 1995, 121, 1-29 and references cited therein. (8) Coche-Guerente, L.; Cosnier, S.; Innocent, C.; Mailley, P.; Moutet, J. C.; Morelis, R. M.; Leca, B.; Coulet, P. R. Electroanalysis 1993, 5, 647-652. S0003-2700(96)00841-4 CCC: $14.00
© 1997 American Chemical Society
(EC 1.4.3.11, from Streptomyces species, 8 units mg-1) was kindly donated by Dr. Hitoshi Kusakabe of Yamasa Shoyu Ltd., Japan. The amphiphilic pyrrole derivative [12-(pyrrol-1-yl)dodecyl]trimethylammonium tetrafluoroborate (1) was synthetized as previously reported.9 Water was doubly distilled in a quartz apparatus and adjusted to pH 7 by adding NaOH. All other reagents used were of analytical reagent grade. Electrochemical Measurements. Glass-insulated platinum microelectrodes with a tip diameter of ∼30 µm were fabricated as described by Donati et al.10 Briefly: A micropipet was made by pulling a glass capillary on a microelectrode puller. With a diamond cutter, the tip of the micropipet was then broken to an inner diameter of about ∼30 µm. A Pt wire (Pt 90%, Ir 10%) was etched by electrolysis until its tip diameter was only about 2-5 µm. This Pt wire was then introduced into the micropipet and advanced until its tip protruded from the opening of the micropipet. Insulation of the Pt wire by glass coating was obtained by gently melting the tip of the micropipet; this was achieved either by approaching a heating element or by passing electrical current through the Pt-wire. Finally, the protruding tip of the Pt wire was etched by electrolysis until the Pt was flush with the glass. Carbon fiber microelectrodes (active tip of 500 µm in length and 8 µm in diameter) were purchased from Radiometer Analytical S.A. The batch electrochemical measurements were carried out in a conventional three-electrode cell thermostated at 20 ( 0.1 °C. A saturated calomel electrode (SCE) and a platinum wire served as reference and counter electrodes, respectively. All potentials are reported with respect to the SCE. Electrochemical experiments were performed with Tacussel PRG-DEL potentiostat in conjunction with a Kipp and Zonen BD 91 XY/t recorder. Amperometric measurements of glutamate and dopamine were carried out under stirred conditions in a 0.05 M HEPES buffer solution (pH 7.4) containing 0.1 M NaCl or in a 0.1 M phosphate buffer solution (pH 7.4), respectively. Reproducibility experiments were carried out with a microelectrode which was cleaned with HNO3, washed with water, and reused eight times. Enzyme Immobilization. The poly-1-GMO and poly-1PPO microelectrodes were prepared as follows. Polymerization of 2 mM monomer 1 aqueous solution containing 8 mg mL-1 enzyme was carried out by controlled-potential electrolysis for 10 min at 0.75 V. The resulting poly-1-enzyme microelectrodes were thoroughly rinsed with water and soaked for 20 min in stirred 0.1 M phosphate buffer (pH 7.4) in order to remove the adsorbed enzyme molecules. In Vivo Measurements. Male Wistar rats weighing ∼280 g were anesthetized with 4% chloral hydrate and placed in a stereotaxic apparatus. A 250 µm chloridized silver wire and a 250 µm platinum wire, which were implanted into the cortex under the dura, were used as reference and counter electrodes, respectively. The modified and naked microelectrodes were stereotaxically implanted into the cortex and advanced so that their active tip was located in the striatum. RESULTS AND DISCUSSION Electropolymerization of monomer 1 in aqueous solution can be carried out at a lower concentration (2 mM) than those usually (9) Coche-Guerente, L.; Deronzier, A.; Galland, B.; Moutet, J. C.; Labbe, P.; Reverdy, G.; Chevalier, Y.; Amhrar, J. Langmuir 1994, 10, 602-610. (10) Donati, G.; Pournaras, C. J.; Munoz, J. L.; Poitry, S.; Poitry-Yamate, C. L.; Tsacopoulos, M. Invest. Ophthalmol. Visual Sci. 1995, 36, 2228-2237.
Figure 1. (A) Steady-state current-time responses of (a) a poly1-GMO microelectrode and (b) a poly-1 microelectrode (diameter of exposed platinum surface 30 µm) for increasing glutamate concentration in 5 µM steps; (B) calibration plot for glutamate. Applied potential 0.6 V vs SCE; air-saturated 0.05 M HEPES buffer (pH 7.4) with 0.1 M NaCl.
used for pyrrole.6-8 In contrast to that with pyrrole, the electropolymerization process can be performed in the absence of supporting electrolyte, thus inducing electrostatic interactions between negatively charged enzyme molecules and the cationic monomers 1. Furthermore, poly-1 presents an improved and potential-independent anion-exchange capacity as compared to regular polypyrrole, which exhibits an anion-exchange capacity restricted to one negative charge per three pyrrole rings.11 Owing to the lipophilic character of the alkyl chain, monomers 1 are expected to also display hydrophobic interactions with enzyme molecules.12,13 Our first example based on the entrapment of GMO has been developed for the detection of glutamate, which is one of the important neurotransmitters involved in a variety of neurological diseases.14,15 Since GMO has an isoelectric point of 6,16 it carries negative charges at neutral pH and thus can develop electrostatic interactions with the amphiphilic monomers 1. Therefore the immobilization of GMO on a platinum microelectrode (30 µm diameter) was accomplished in the following way. Polymerization of 1 (2 mM) in an aqueous solution (pH 7) containing GMO was carried out by controlled-potential oxidation at 0.75 V. Since GMO catalyzes the aerobic oxidation of glutamate with concomitant production of hydrogen peroxide, the amperometric detection of glutamate was assayed in air-saturated HEPES buffer (pH 7.4) by potentiostating the poly-1-GMO electrode at 0.6 V. Figure 1A presents the steady-state current response of the microbiosensor to successive increments (5 µM) of glutamate concentration, demonstrating the successful entrapment of GMO in the resulting polypyrrole matrix. The responses were stable, and the response time (determined as the time required to reach a new current value indistinguishable from the final steady-state current, after a substrate injection) was within 10 s, illustrating (11) (12) (13) (14)
Deronzier, A.; Moutet, J. C. Acc. Chem. Res. 1989, 22, 249-255. Bourdillon, C.; Majda, M. J. Am. Chem. Soc. 1990, 112, 1795-1799. Cosnier, S.; Innocent, C. Anal. Lett. 1994, 27, 1429-1442. Faden, A. I.; Demediuk, P.; Pander S. S.; Vink, R. Science 1989, 244, 798800. (15) Olney, J. W. Retina 1992, 2, 341-359. (16) Kusakabe, H.; Midorikawa, Y.; Fujishima, A.; Yoshino, H. Agric. Biol. Chem. 1983, 47, 1323-1328.
Analytical Chemistry, Vol. 69, No. 5, March 1, 1997
969
Figure 2. Effect of GMO concentration in the growth solution on the glutamate sensitivity of the poly-1-GMO electrode.
the good permeability of the host polymer. A control experiment was carried out with a poly-1 electrode without immobilized GMO, and no amperometric response was observed for successive injections of glutamate (Figure 1A). Figure 1B shows the amperometric response current of the poly-1-GMO electrode as a function of glutamate concentration. The calibration curve was quasi-linear with glutamate concentration up to 0.3 mM and curved gradually at higher concentrations. The sensitivity of the bioelectrode (determined as the slope of the initial linear part of the calibration curve) and its detection limit are 32 mA M-1 cm-2 and 1 µM, respectively. This detection limit is similar to those recorded with different glutamate macro- and microsensors17-22 but is markedly lower than the value (100 µM) recently obtained with a 25 µm polypyrrole-GMO electrode.23 By comparing the sensitivities of bioelectrodes made from aqueous solutions containing 2 mM monomer 1 and various GMO concentrations (Figure 2), the optimum enzyme concentration was found to be 8 mg mL-1. The microbiosensor construction was also quite reproducible: eight poly-1-GMO electrodes were prepared by following identical electropolymerization steps in an aqueous solution containing 2 mM monomer 1 and 8 mg mL-1 GMO, and their responses toward glutamate were investigated. The comparison of the sensitivity determined from the resulting calibration plots indicates that the relative standard deviation is only 9%. The microbiosensors were also examined for the storage and operational stabilities. Poly-1-GMO electrodes were stored dry at 4 °C, and their calibration curves in the glutamate concentration range 0-0.4 mM were recorded after 1, 5, 12, and 16 days. Only a 5% decay of initial sensitivity was observed after 16 days. However, the bioelectrode sensitivity dropped to 26% of its initial (17) Villarta, R. L.; Cunningham, D. D.; Guilbault, G. G. Talanta 1991, 38, 4955. (18) Hale, P. D.; Lee, H. S.; Okamoto, Y.; Skotheim, T. A. Anal. Lett. 1991, 24, 345-356. (19) Kar, S.; Arnold, M. A. Anal Chem. 1992, 64, 2438-2443. (20) Tamiya, E.; Sugiura, Y.; Takeuchi, T.; Suzuki, M.; Karube, I.; Akiyama, A. Sens. Actuators B 1993, 10, 179-184. (21) Obrenovitch, T. P.; Koshy, A.; Zilkha, E.; Richards, D. A.; Bennetto, H. P. Curr. Sep. 1993, 12, 48-49. (22) Hu, Y.; Mitchell, K. M.; Albahadily, F. N.; Michaelis, E. K.; Wilson, G. S. Brain Res. 1994, 659, 117-125. (23) Cooper, J. M.; Foreman, P. L.; Glidle, A.; Ling, T. W.; Pritchard, D. J. J. Electroanal. Chem. 1995, 388, 143-149.
970 Analytical Chemistry, Vol. 69, No. 5, March 1, 1997
Figure 3. Calibration plots for dopamine obtained with (a) a poly1-PPO microelectrode and (b) a poly-1 microelectrode (carbon fiber tip, 500 µm in length and 8 µm in diameter). The linear response region is given in the inset (correlation coefficient 0.9996). Applied potential -0.2V vs SCE; air-saturated 0.1 M phosphate buffer (pH 7.4).
value after 50 days. This decrease may be attributed to some GMO denaturation by the enzymically generated hydrogen peroxide, the bioelectrode sensitivity being periodically determined. The operational stability was evaluated by recording the steady-state current response with increasing concentration of glutamate in the linear part of the calibration curve. The permanent loss of activity was 7% after 50 analyses performed over 115 min. The second example describes the immobilization of PPO on a carbon fiber microelectrode in order to detect dopamine, another important neurotransmitter. The entrapment of PPO, which is negatively charged (isoelectric point 4.724) at pH 7 used for the enzyme-monomer 1 solution, was performed by following the same operating conditions as those described for GMO. PPO catalyzes the oxidation of several monophenols and o-diphenols to o-quinones while molecular oxygen is reduced to water. Therefore, the analytical capabilities of the resulting poly-1-PPO electrodes toward dopamine determination were investigated in air-saturated 0.1 M phosphate buffer (pH 7.4) by potentiostating the microelectrode at -0.2 V in order to detect amperometrically the generated quinoid product.25 Figure 3 shows a plot of the current response as a function of the dopamine concentration. A linear relationship was observed for dopamine concentrations ranging from 5 × 10-8 to 8 × 10-5 M, indicating a biosensor sensitivity of 59 mA M-1 cm-2 for dopamine while no amperometric response was observed with a poly-1 electrode without PPO. It should be noted that the same procedure of biosensor elaboration performed in the presence of 0.1 M LiClO4 as supporting electrolyte leads to a poly-1-PPO microelectrode exhibiting a markedly reduced dopamine sensitivity (38 mA M-1 cm-2). In addition, the response of the poly-1-PPO microelectrode to 10 µM ascorbate as a representative example of endogenous interferent was examined. No appreciable change in the steady-state current response of the biosensor was observed, illustrating the absence of interference. Furthermore the microbiosensor provides an attractive detection limit, namely, 50 nM, (24) Kertesz, D.; Zito, R. Biochim. Biophys. Acta 1965, 96, 447-462. (25) Cosnier, S.; Innocent, C. Bioelectrochem. Bioenerg. 1993, 31, 147-160.
which is the lowest concentration value that can be detected using microsensors based on PPO.26,27 This very low value is probably due to an electrochemical recycling phenomenon of the enzyme substrate (dopamine) inducing an amplification of the biosensor response.25,28-30 Indeed, the quinoı¨d product enzymically generated by the immobilized PPO diffuses quickly to the electrode surface where the PPO substrate (dopamine) is electrochemically regenerated. Thus, dopamine can undergo successive cycles of enzymatic oxidation-electrochemical reduction. It should be noted that detection limits of the same order of magnitude have recently been obtained with bare or polypyrrole-coated microelectrodes involving high-pass filtering combined with fast-scan cyclic voltammetry.31,32 Preliminary in vivo experiments have been performed with this microbiosensor in the brain of an anesthetized rat placed on a stereotaxic frame. A poly-1-PPO electrode was implanted into the striatum for 50 min, then removed, washed carefully with phosphate buffer, and transferred into a cell containing 0.1 M phosphate buffer. The current response to dopamine was recorded before and after the implantation step, and no change of the biosensor sensitivity was observed. This demonstrates the mechanical stability of the poly-1-PPO coating toward the penetration of the brain tissue as well as the absence of biosensor fouling by brain components (lipids). In addition, the movement of the poly-1-PPO electrode from the cortex to the striatum, (26) Zhihong, L.; Wenjian, Q.; Meng, W. Anal. Lett. 1992, 25, 1171-1181. (27) Wang, J.; Chen, Q. Anal. Chem. 1994, 66, 1988-1992. (28) Cosnier, S.; Popescu, I. Anal. Chim. Acta 1996, 319, 145-151 and references therein. (29) Besombes, J. L.; Cosnier, S.; Labbe, P. Talanta 1996, 43, 1615-1619. (30) Onnerfjord, P.; Emme´us, J.; Marko-Varga, G.; Gorton, L.; Ortega, F.; Dominguez, E. Biosens. Bioelectron. 1995, 10, 607 and references therein. (31) Pihel, K. ; Walker, Q. D.; Wightman, R. M. Anal. Chem. 1996, 68, 20842089. (32) Cahill, P. S.; Walker, D.; Finnegan, J. M.; Mickelson, G. E.; Travis, E. R.; Wightman, R. M. Anal. Chem. 1996, 68, 3180-3186 and references therein.
which contains a higher density of dopaminergic terminals, induces an increase (2.7 pA) in the steady-state current response of the microbiosensor, which could be attributed to increased dopamine levels in the striatum. A control experiment was carried out with a naked carbon fiber microelectrode potentiostated at -0.2 V and placed in the cortex and then in the striatum. No significant change in the steady-state current response of the microelectrode was observed between the two structures. Although the identity of dopamine remains to be pharmacologically confirmed, these results strongly suggest that the increase in current response recorded with the microbiosensor was due to increased PPO activity. CONCLUSION In this report we have illustrated, in connection with polyphenol oxidase and glutamate oxidase, the attractive potentialities offered by the fabrication method of microbiosensors based on the electrochemical polymerization of an aqueous amphiphilic pyrrole-enzyme solution. It is expected that this simple enzyme immobilization method will be useful for the development of in vitro and in vivo measurement techniques involving enzyme-based microelectrochemical devices. ACKNOWLEDGMENT We thank Prof. A. L. Benabid and Dr. A. Deronzier for fruitful discussions and laboratory facilities. Acknowledgment is also made to COTRAO (France) for providing a fellowship and FNRS 31-37587.93 for financial support. Received for review August 19, 1996. Accepted December 8, 1996.X AC960841H X
Abstract published in Advance ACS Abstracts, February 1; 1997.
Analytical Chemistry, Vol. 69, No. 5, March 1, 1997
971
