452

Anal. Chem. 1990, 62, 452-458

(2) Pons, S.; Fleischmann, M. Anal. Chem. 1987, 5 9 , 1391A. (3) Johnson, D. C.; Ryan, M. D.; Wllson, G. S. Anal. Chem. 1988, 6 0 , 147R. (4) Howell, J. 0; Wightman. R. M. Anal. Chem. 1984, 5 6 , 524. (5) Howell, J. 0.; Wightman, R. M. J. Phys. Chem. 1984, 88, 3915. (6) Andrieux, C. P.; Garreau, D.; Hapiot, P.; Pinson, J.; SavBant, J. M. J. Elechoanal. Chem. 1988, 243, 321. (7) Andrieux, C. P.; Garreau, D.; Hapiot, P.; SavCnt, J. M. J. Electroanal. Chem. 1988, 248, 447. (8) Amatore, C.; Kelly, R. S.; Dristensen. E. W.; Kuhr. W. G.; Wightman, R . M. J. Electroanal. Chem. 1988, 273, 31. (9) Ponchon, J.-L.; Cespugiio, R.; a n o n , F.; Jouvet, M.; Pujol, J.-F. Anal. Chem. 1979, 5 1 , 1483. (10) Geng. L.; Murray, R. W. Inorg. Chem. 1986, 2 5 , 3115. (11) Janata, J.; Bezegh, A. Anal. Chem. 1988, 6 0 , 62R. (12) Ikariyama, Y.; Yamauchi, S.; Yukiashi, T.; Ushioda, H. Anal. Lett. 1987, 2 0 , 1791. (13) Ikariyama, Y.; Yamauchi, S.; Yukiashi, T.; Ushioda, H. Anal. Lett. 1987, 20, 1407. (14) Brlna, R.; Pons, S.; Fieischmann, M. J . Electroanal. Chem. 1988, 244, 81. (15) Chidsey, C. E.; Feidman, 9. J.; Lundgren, C.; Murray, R. W. Anal. Chem. 1986, 5 8 , 601. (16) Murakami, T.; Nakamoto, S.; Kimura. J.; Kuriyama, T.; Karube, I. Anal. Lett. 1988, 79, 1973. (17) White, H. S.; Kittlesen, G. P.: Wrighton, M. S. J. Am. Chem. Soc. 1984, 106, 5375. (18) Kittlesen, G. P.; White, H. S.; Wrighton, M. S. J. Am. Chem. SOC. 1984, 106, 7389. (19) Paul, E. W.; Ricco. A. J.; Wrighton. M. S. J. Phys. Chem. 1985, 8 9 , 1441.

(20) Thackeray, J. W.; White, H. S.; Wrighton, M. S. J. phys. Chem. 1985, 8 9 , 5133. (21) Bhnger, D.; Wrighton, M. S. Anal. Chem. 1987, 5 9 , 1426. (22) Natan, M. J.; Mallouk, T. E.;Wrighton, M. S. J. Wys. Chem. 1987, 9 1 , 648. (23) Natan, M. J.; BQlanger, D.; Carpenter, M. K.; Wrighton, M. S. J. Phys. Chem. 1987, 91, 1834. (24) Jones. E. T. T.: Chvan, 0. M.; Wriohton. M. S. J. Am. Chem. SOC. 1987, 109, 5526. Sanderson, D. G.; Anderson, L. B. Anal. Chem. 1985, 5 7 , 2388. Bard, A. J.; Crayston, J. A.; Kittlesen, G. P.; Shea, T. V.; Wrighton, M. S. Anal. Chem. 1988, 5 8 , 2321. Shea, T. V.; Bard, A. J. Anal. Chem. 1987, 5 9 , 2101. Licht, S . ; Cammarata, V.; Wrighton, M. S. Science 1988, 243, 1176. Feldman, B. J.; Feldberg. S. W.; Murray, R. W. J. Phys. Chem. 1987, 9 1 , 6558. Aoki, K.; Morita, M.; Niwa, 0.; Tabei, H. J. Electroanal. Chem. 1988, 256, 269. Morita, M.; Longmlre, M. L.; Murray, R. W. Anal. Chem. 1988, 6 0 ,

-

-. . _. 3770

(32) Aoki. K.; Tokuda, K.; Matsuda, H. J. €/ectroanal. Chem. 1987, 225, 19. (33) Aoki, K.; Tokuda, K.; Matsuda, H. J. Electroanal. Chem. 1987, 230, 61. (34) Coen, S.; Cope, D. K.; Taliman, D. E. J. Eiectroanal. Chem. 1988, 215. 29.

RECEIVED for review August 14, 1989. Accepted November 29, 1989.

Mediated, Anaerobic Voltammetry of Sulfite Oxidase L. A. Coury, Jr., B. N. Oliver,' J. 0. Egekeze,*C. S. Sosnoff, J. C. Brumfield, R. P. Buck, and R. W. Murray* Venable and Kenan Laboratories of Chemistry, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 2 7599- 3290

The anaerobic voltammetry of the Mo/Fe enzyme, sulfite oxldase (SO), is descrlbed for the mediators cytochrome c, [Ru( NH3)O]3+12+, TMPD"', and [Co( bpy)3]3+12+.Theory derived for steadydate vdtamnetric catalysis correctly predicts the observed concentration and scan-rate dependencies of the catalytic waves. The instances for whlch existing EC,, theories may be applied to two catalytic reacttons coupled to an Interfacial charge transfer are considered. The blmolecuiar with reduced rate constant for the reactlon of [C~(bpy),]~+ SO is calculated and determlned to be approximately 5 X lo4 L.moi-'*s-". The appearance of catalytic prepeaks at low sulfite concentrations Is noted and the shape of corresponding 111 curves from chronoamperometry is examined. The analytical Implications of the novel time dependence of the cataiytlc current under these conditions are discussed.

INTRODUCTION Increasing interest in electrochemical sensors capable of selective response to species of analytical importance has motivated evaluation of various enzyme systems as sensor components (1-3). Perhaps the most widely studied enzymes for this application are the oxidase enzymes (4). Under aerobic conditions, many oxidase enzyme-based sensors seek to a m perometrically monitor hydrogen peroxide generated from the 'Current address: British Gas,London Research Station, Fulham, London, UK SW6 2AD. *Permanent address: Department of Chemistry a n d Physics, Augusta College, Augusta, GA 30910. * Author t o w h o m correspondence should be addressed.

reduction of dioxygen during the enzymatic reaction sequence. Recently, ferrocene derivatives have been explored as synthetic electron acceptors for some oxidase enzymes under anaerobic conditions (5-7). Electron transfer mediators often offer the advantage of driving enzyme turnover at less extreme potentials than would otherwise be necessary if hydrogen peroxide were the species to be monitored electrometrically. Decreased operating potentials may increase selectivity by eliminating contributions to the measured response from other species undergoing redox reactions at more extreme potentials. In addition, mediators allow experiments under anaerobic conditions, circumventing problems arising from direct (nonenzymatic) oxidation of the enzymatic substrate by dissolved oxygen. The enzyme sulfite oxidase [EC 1.8.2.11 has previously been incorporated into membranes and used in Clark-type oxygen electrode based sensors (8,9). There are several disadvantages of the previous approaches. In order to oxidize enzymatically generated peroxide at an appreciable rate, it is necessary to polarize the working electrode a t highly positive potentials, where sulfite may itself be oxidized directly. Oxidation of sulfite can lead to passivation of solid electrode surfaces, limiting their useful lifetime (10). For sulfite sensor work, it is common to stabilize aqueous sulfite solutions from air oxidation by adding formaldehyde or glycerol as stabilizers (8,111 to the analyte solution. These stabilizers form adducts with sulfite that are not oxygen sensitive (12), thus overcoming the inherent contradiction of employing analyte solutions containing readily oxidizable sulfite but yet saturated with oxygen. Such sensing strategies rely on dissociation of these adducts to supply sulfite for the

0003-2700/90/0362-0452$02.50/00 1990 American Chemical Society

ANALYTICAL CHEMISTRY, VOL. 62, NO. 5, MARCH 1, 1990

reaction of interest, however. Also, when air-equilibrated, aqueous solutions are used in amperometric sensors, the concentration of the electron acceptor (dioxygen) cannot be easily systematically varied to alter the sensitivity of the electrode reaction. All of the above complications could be avoided if synthetic electron acceptors were used with sulfite oxidase under anaerobic solution conditions. We present here the first report of steady-state, catalytic voltammetry of sulfite oxidase (SO) under anaerobic conditions, using several different mediators. From available theoretical treatments for steady-state electrocatalytic processes, we demonstrate the rate law for the mediated, enzymatic oxidation of sulfite to sulfate and determine the bimolecular rate constant for the reaction between the reduced enzyme and the mediator [ C ~ ( b p y ) ~ ]We ~ + .also report the observation of voltammetric prepeaks at low sulfite concentrations and discuss the implications of this effect on the time-dependence of catalytic currents monitored in an amperometric mode. The goal of this research is fundamental characterization of the reactions of a particular enzyme to aid in the rational design of new biosensors.

EXPERIMENTAL SECTION Sulfite Oxidase. Several sources of SO were used. A commercially available preparation from chicken livers (Sigma Chemicals) was diluted 15 with 0.02 M Tris buffer, pH 7.5, and stored at 5 "C. The enzyme was also isolated from chicken (13) and rat livers (14) as previously described and stored frozen in small portions until needed. The purity of each batch of enzyme was determined by calculating the ratio of the W-vis absorbance obtained at 414 nm to that at 280 nm. The absorbance at 414 nm is a Soret band arising from the cytochrome b5 center in SO (15) while the 280-nm band is a composite absorbance from all aromatic amino acid residues in solution proteins, so this ratio is a measure of the fraction of protein in solution which is SO. For all preparations used this ratio was greater than 0.6. Molar concentrations of enzyme solutions were calculated as previously described (16). The values of the molecular weights are 1.08 X lo6gmol-' for the chicken enzyme (13) and 1.14 x lo5 gmol-' for the rat enzyme (16). Enzymatic activity was monitored periodically at pH 8.5 by assaying the ferricyanide reductase activity of SO (15) at saturating concentrations of sflite (0.4 mM) with respect to the enzyme concentration (0.8 nM). The rate of bleaching of ferricyanide (0.4 mM) was monitored at 420 nm for ca. 300 s. In these experiments, the concentration of enzyme was too low for its Soret band to interfere with the nearly coincident ferricyanide absorbance. Enzymatic activity (reported as the decrease in absorbance at 420 nm) after thawing was found to remain invariant over the course of several days (dA/dt = 3 2 . 2 0.2) X lo4 AUK') when the enzyme solution was stored at 5 "C between experiments. Reagents. Cytochrome c (type VI, horse heart, prepared without trichloroacetic acid, Sigma Chemicals) was found to be electroactive at edge-plane pyrolytic graphite electrodes without further purification when used shortly after receipt. Tris(2,2'~ + prepared , by bipyridine)cobalt(II) complex, or [ C ~ ( b p y ) ~ ]was adding an appropriate amount of an aqueous CoC12solution to a 4-fold molar excess of the ligand (17). The excess ligand in solution, at the concentrations employed for voltammetric experiments (typically < 50 pM), had no effect on catalytic currents, as determined by comparisons to solutions prepared from solid [Co(bpy),]C12. All other chemicals utilized were high-purity, commercially available materials. Solutions for electrochemical experiments (with the exception of stock sulfite solutions) were prepared daily in 0.02 M Tris/O.l M KCl buffer, pH 7.5. Sodium sulfite solutions were prepared fresh every few hours in Ar-saturated buffer and kept on ice in a septum-capped vial under a blanket of Ar. Apparatus. Spectrophotometry was conducted with an HP8452A photodiode-array spectrometer interfaced to a laboratory PC. Locally constructed instruments for cyclic voltammetry (CV) and chronoamperometry (CAI were of conventional design (18). Data from CA experimenta were digitized with a Summa Graphics MM1103 Bit Pad Two data tablet interfaced to a PC and ma-

*

453

re//'cvt-c SO,=tH,O 4 n 2

f 'I

cyt -c

Flgure 1. Catalytic cycle proposed operative in vivo for the SO catalyzed oxidation of sulfite to sulfate. Mo and Fe represent the enzymic molybdenum and cytochrome b Oxidation states.

nipulated with customized software written in Borland Turbo Pascal. The coplanar arrangement of edge-plane pyrolytic graphite (EPG) working, Ag-wire quasi-reference and Pt-ring auxiliary electrodes has been described previously (19). A (0.1 M KCl) Ag/AgCl reference electrode, isolated from the protein-containing solution, replaced the internal Ag-wire reference for later experiments. The area of the EPG working electrode was determined to be 0.0671 0.0008 cm2 (10 replicate measurements; electrode repolished with 0.05-pm alumina between determinations) by CA measurements on 10.0 mM potassium ferricyanide/l.O M KCl solutions, assuming Do = 7.63 X lo4 c m 2 d (20). The electrochemical cell was a single-compartment tube with a threaded top into which the electrode assembly could be sealed with an O-ring bushing and which could accommodate solution volumes as small as 400 1L. One port in the cell was fitted with a septum through which solutions could be introduced via syringe (with Teflon-tipped plunger) under anaerobic conditions, and another was fitted with a degassing tube to maintain a blanket of Ar over the solution during experiments. Care was taken to avoid bubbling Ar through solutions after introduction of enzyme (or cytochrome c ) aliquots, since this type of agitation produced frothing of the protein and resulted in loss of enzyme activity. Cell solution volumes of between 500 and 600 p L were most typically employed. The duration of our experiments, for example that in Figure 3 (vide infra),is sufficiently long that