Dithiobissuccinimidyl Propionate as an Anchor for Assembling

Nov 5, 1999 - Preventing Nonspecific Adsorption on Polymer Brush Covered Gold Electrodes Using a Modified ATRP Initiator. Biomacromolecules ,. Abstrac...
64 downloads 7 Views 96KB Size
Download PDF
Anal. Chem. 1999, 71, 5530-5537

Dithiobissuccinimidyl Propionate as an Anchor for Assembling Peroxidases at Electrodes Surfaces and Its Application in a H2O2 Biosensor M. Darder,† K. Takada,‡ F. Pariente,† E. Lorenzo,† and H. D. Abrun˜a*,‡

Department of Chemistry and Chemical Biology, Baker Laboratory, Cornell University, Ithaca, New York 14853-1301, and Departamento de Quı´mica Analı´tica, Universidad Auto´ noma de Madrid, Canto Blanco 28049, Madrid, Spain

Exposure of gold surfaces to solutions of dithiobis Nsuccinimidyl propionate (DTSP) gives rise to the modification of the surface with N-succinimidyl-3-thiopropionate (NSTP) which can, in turn, react with amino groups allowing for the covalent immobilization of enzymes such as horseradish peroxidase (HRP). The coverage of NSTP has been estimated to be of the order of 1.3 × 10-10 from the charge consumed during its reductive desorption. The binding reaction of HRP with NSTP modified gold surfaces has been studied with the quartz crystal microbalance, and the results suggest that the immobilization process involves two steps in which the first (faster) appears to correspond to the rapid incorporation of the enzyme whereas the second is likely due to the slow incorporation of additional enzyme and/or reorganization of the immobilized layer. Spectrophotometric and electrochemical assays indicate that the immobilized HRP retains its enzymatic activity after immobilization onto the DTSP modified gold surface. The amount of immobilized (and active) HRP was estimated from QCM and spectrophotometric measurements to be of the order of 1.5 × 10-11 mol/cm2. A peroxide biosensor was developed making use of a gold surface modified with DTSP and HRP employing Os and Ru complexes of 1,10-phenanthroline 5,6-dione (phen-dione) of the type [M(phendione)x(L)3-x]+2 (where L ) 1,10-phenanthroline or 2,2′bipyridine, x ) 1-3) as mediators with the quinone moieties being the active component. The efficiency of the mediators increased with increasing number of phendione ligands. The determination of hydrogen peroxide is of great relevance, ascribable to both the fact that it is the product of the reactions catalyzed by a large number of oxidase enzymes and that it is essential in food, pharmaceutical, and environmental analysis.1-4 * Corresponding author: (tel.) 607-225-4720; (fax) 607-255-9864; (e-mail) [email protected]. † Universidad Auto ´ noma de Madrid. ‡ Cornell University. (1) Somasundrum, M.; Kirtikara, K.; Tanticharoen, M. Anal. Chim. Acta 1996, 319, 59-70. (2) Kulys, J. J.; Pesliakine, M. V.; Samalius, A. S. Bioelectrochem. Bioenerg. 1981, 8, 81-88. (3) Wang, J.; Lin, Y.; Chen, L. Analyst (Cambridge, U.K.) 1993, 118, 277-280. (4) Tatsuma, T.; Watanabe, T. Anal. Chim. Acta 1991, 242, 85-89.

5530 Analytical Chemistry, Vol. 71, No. 24, December 15, 1999

The direct electrochemical detection of H2O2 often requires relatively high overpotentials.1,5 However, hydrogen peroxide can be detected enzymatically at low applied potentials by employing peroxidases as bioelectrocatalysts for its electrochemical reduction. Peroxidases catalyze the oxidation of a variety of organic and inorganic compounds by hydrogen peroxide or other peroxo species of the ROOH type, with the resulting reduction of the peroxide to water. Among peroxidases, horseradish peroxidase (HRP) has been the most widely studied in the development of enzyme-based amperometric biosensors. It has been widely applied in monoenzyme biosensors for the determination of peroxides and HRP inhibitors and in bienzyme biosensors, for the determination of substrates such as glucose, alcohols, and D-amino acids, in which hydrogen peroxide is formed as the product of the enzymatic reaction.6-10 A highly sensitive H2O2 amperometric sensor would be valuable in the development of biosensors for a wide range of molecules by combining it with hydrogen peroxide-producing oxidases.11 It could also be used for the determination of activators or inhibitors of reactions catalyzed by peroxidases, such as potassium cyanide12 and nitric oxide.13 The organization of biomaterials as monolayers on electrode surfaces and the application of these monolayer electrodes as amperometric biosensors are areas of intense research activity.14-18 There is also a great deal of interest in the development of novel approaches to electrode modification, in general, and (bio)sensor applications, in particular. Dithiobis(N-succinimidyl propionate) (5) Liu, H.; Ying, T.; Sun, K.; Qi, D. J. Electroanal. Chem. 1996, 417, 59-64. (6) Liu, H.; Zhang, Z.; Zhang, X.; Qi, D.; Liu, Y.; Yu, T.; Deng, J. Electrochim. Acta 1997, 42, 349-355. (7) Cso ¨regi, E.; Gorton L.; Marko-Varga, G. Electroanalysis 1994, 6, 925-933. (8) Charpentier, L.; El Murr, N. Anal. Chim. Acta 1995, 318, 89-93. (9) Shin, M.; Yoon, H. C.; Kim, H. Anal. Chim. Acta 1996, 329, 223-230. (10) Kalcher, K. Electroanalysis 1990, 2, 419-433. (11) Tatsuma, T.; Gondaira, M.; Watanabe, T. Anal. Chem. 1992, 64, 11831187. (12) Tatsuma, T.; Oyama, N. Anal. Chem. 1996, 68, 1612-1615. (13) Casero, E.; Darder, M.; Pariente F.; Lorenzo, E. Anal. Chim. Acta, in press. (14) Li, J.; Yan, J.; Deng, Q.; Cheng, G.; Dong, S. Electrochim. Acta 1997, 42, 961-967. (15) Ruan, C.; Yang, F.; Lei, C.; Deng, J. Anal. Chem. 1998, 70, 1721-1725. (16) Willner, I.; Katz, E.; Riklin, A.; Kasher, R. J. Am. Chem. Soc. 1992, 114, 10965-10966. (17) Willner, I.; Lapidot, N.; Riklin, A.; Kasher, R.; Zahavy, E.; Katz, E. J. Am. Chem. Soc. 1994, 116, 1428-1441. (18) Katz, E.; Willner, I. Langmuir 1997, 13, 3364-3373. 10.1021/ac990759x CCC: $18.00

© 1999 American Chemical Society Published on Web 11/05/1999

(DTSP) also known as Lomant’s reagent19 is used as a protein cross-linking reagent through acylation of free primary or secondary aliphatic amino groups. DTSP also adsorbs onto gold surfaces, through the disulfide group,20 so that the terminal succinimidyl groups allow further covalent immobilization of amino-containing biomolecules.21,22 In this paper we describe the use of gold electrodes modified with a monolayer of DTSP for the covalent immobilization of HRP with the intent of developing a peroxidase-modified electrode for the amperometric determination of peroxide. Os and Ru complexes of 1,10-phenanthroline-5,6-dione (phen-dione) were used as diffusional mediators between the active site of the immobilized enzyme and the electrode surface. The influence on the mediation reaction of the number of phen-dione ligands present in the metal complex has also been studied. Moreover, QCM (quartz crystal microbalance) experiments have been carried out in order to study the time evolution of the immobilization of HRP through the N-succinimide ester groups present in DTSP modified gold electrodes. EXPERIMENTAL SECTION Materials. Peroxidase type I (EC 1.11.1.7) from horseradish (120 units/mg solid, molecular weight 44 000) was purchased from Sigma Chemical Co. and stored as received at -20 °C. Hydrogen peroxide solutions were obtained by diluting a 30% solution from Carlo Erba. The solution concentration of hydrogen peroxide was determined from its absorbance at 240 nm using an extinction coefficient of 39.4 mM-1 cm-1.23 3,3′,5,5′-Tetramethylbenzidine (TMB) from Sigma was stored at 4 °C. Dithiobis-N-succinimidylpropionate (DTSP) was purchased from Fluka and stored at 4 °C. [Os(phen-dione)(phen)2](PF6)2, (phen ) 1,10-phenanthroline), [Ru(phen-dione)(bpy)2](PF6)2, (bpy ) 2,2′-bipyridine), [Ru(phendione)2bpy](PF6)2, and [Ru(phen-dione)3](PF6)2 were synthesized as previously described.24 Sodium phosphate (Sigma Chemical Co.) was used in the preparation of buffer solutions. Water was purified with a Millipore Milli-Q-System. All other reagents were of at least analytical-grade and were used as received. Apparatus. Voltammetric Measurements. Cyclic voltammetric studies were performed with an IBM EC/225 potentiostat connected to a Soltec VP-64235 X-Y recorder and with an Autolab/ PGSTAT10 potentiostat from Eco-Chemie. The electrochemical experiments were carried out in three compartment electrochemical cells with standard-taper joints so that all three-compartments could be hermetically sealed with Teflon adapters. Gold disk electrodes (2-mm-diameter, 0.031 cm2 geometric area, and 0.090 cm2 microscopic area), sealed in soft glass, were used as working electrodes. A large-area coiled platinum wire was employed as a counter electrode, and all potentials are reported against a sodiumsaturated calomel electrode (SSCE) without regard for the liquid junction. All solutions were deaerated with nitrogen gas for at least 15 min before use, and the gas flow was kept over the solution during experiments. (19) Lomant, A. J.; Fairbanks, G. J. Mol. Biol. 1976, 104, 243-261. (20) Katz, E. J. Electroanal. Chem. 1990, 291, 257-260. (21) Wagner, P.; Hegner, M.; Kernen, P.; Zaugg, F.; Semenza, G. Biophys. J. 1996, 70, 2052-2066. (22) Sehgal, D.; Vijay, I. K. Anal. Biochem. 1994, 218, 87-91. (23) Nelson, D. P.; Kiesov, L. A. Anal. Biochem. 1972, 49, 474-478. (24) Goss, C. A.; Abrun ˜a, H. D. Inorg. Chem. 1985, 24, 4263-4267.

Spectrophotometric Measurements. Absorbance measurements were carried out with a Milton Roy Spectronic 3000 Array spectrophotometer. Quartz cuvettes (path length ) 1 cm) were used, and the measurements were performed at room temperature. EQCM Measurements. AT-cut quartz crystals (5 MHz) of 2.54-cm diameter with Au electrodes deposited over a Ti adhesion layer (Maxtek Co.) were used for EQCM measurements. An asymmetric keyhole electrode arrangement was used, in which the circular electrode geometrical areas were 1.370 cm2 (front side) and 0.317 cm2 (backside). The electrode surfaces were overtone-polished, giving rise to smooth surfaces. Prior to use, the quartz crystals were cleaned by immersion in piranha solution, H2SO4/H2O2 (3:1). Caution: Piranha Solution is extremely reactive! They were subsequently rinsed with water and acetone and dried in air. The quartz crystal resonator was set in a probe (TPS550, Maxtek) made of Teflon in which the oscillator circuit was included and the quartz crystal was held vertically. The probe was connected to a water-jacketed, conventional, three-chamber electrochemical cell by a homemade Teflon joint, and the cell was thermostated at 25.0 ( 0.1 °C with a thermostated bath (Digital Temperature Controller 9101, Fisher Scientific). On short-time experiments (time required for a CV), the stability of the oscillator was of the order of (0.1 Hz. Nitrogen gas was used to degas the solutions before use, and it flowed over the solutions during experiments. One of the electrodes of the quartz crystal resonator, in contact with the solution, was also used as the working electrode. In these experiments, a silver wire was used as a quasireference electrode to avoid potential interferences in frequencychange measurements due to adsorption of Cl- ions (from a SSCE reference electrode) given their known propensity to adsorb onto gold surfaces. The potential of the working electrode was controlled with a potentiostat (CV-27, BAS). The frequency, which was measured with a plating monitor (PM-740, Maxtek Inc.) and the current, measured with the potentiostat, were simultaneously recorded by a personal computer which was interfaced to the above instruments using LabView. The admittance of the quartz crystal resonator was measured near its resonant frequency by an impedance analyzer (HP4194A, Hewlett-Packard) equipped with a test lead (HP16048A). A probe similar to that used in the QCM measurements, but which did not include an oscillator circuit inside, was used to accomplish a direct connection of the quartz crystal resonator to the impedance analyzer. Procedures. Electrode Conditioning. Polycrystalline golddisk electrodes were polished with 1.0-µm diamond paste (Buehler), rinsed with water, and sonicated for 10 min in distilled water. The electrodes were activated by holding the potential in 0.1 M H2SO4 at +2.0 V for 5 s and then at -0.35 V for 10 s, followed by potential cycling between -0.35 and +1.5 V at 4 V/s for 1 min. Finally, the cyclic voltammogram characteristic of a clean polycrystalline gold electrode was recorded at 100 mV/s and used to calculate the microscopic area by integration of the cathodic peak associated with the reduction of the gold oxide. The electrode was subsequently rinsed with water and acetone and used immediately in the monolayer preparation. Adsorption of DTSP and Enzyme Immobilization. The conditioned electrode was immersed in a 4 mM DTSP solution in dimethyl sulfoxide (DMSO) for 1 h at room temperature. Analytical Chemistry, Vol. 71, No. 24, December 15, 1999

5531

Afterward, the electrode was thoroughly rinsed with acetone and finally with 5 mM phosphate buffer (pH 7.0). Electrodes were employed immediately after preparation. HRP was immobilized by immersion of the DTSP-modified gold electrode in a 5 mM pH 7.0 phosphate buffer solution containing HRP at a concentration of 1.0 mg/mL. The electrode was kept in the enzyme solution for 24 h at 4 °C and then rinsed with phosphate buffer. Determination of Heme Content in HRP Stock Solutions. The heme content of HRP (and hence, the concentration of HRP) solutions can be estimated from the absorbance at 401 nm using an extinction coefficient of 9.5 × 104 M-1 cm-1.25 Absorbance measurements of a 1.0 mg/mL (23 µM) HRP solution in phosphate buffer resulted in a heme concentration of 9.7 µM. This indicates that the preparation contains 43% enzyme. This value was employed in all further calculations which involved the use HRP stock solutions. Determination of Peroxidase Activity. Peroxidase activity was determined spectrophotometrically by following the change in absorbance at 373 nm due to the cation radical of TMB (TMB•+) and specifically to its oxidation in the presence of hydrogen peroxide and enzyme. In these determinations, solutions containing 100 µL of 10 mM TMB, 10 µL of 10 mM H2O2 and aliquots of a 100 nM HRP solution were prepared with 5.0 mM phosphate buffer (to a total volume of 2.00 mL). The absorbance was measured at 373 nm 30 s after the addition of enzyme. The resulting plot of absorbance vs amount of enzyme present in the assay was linear with a slope of 0.155 absorbance units/pmol of enzyme. RESULTS AND DISCUSSION Modification of the Electrode Surface by Adsorption of DTSP. Soaking of a gold electrode in a 4 mM solution of DTSP in DMSO resulted in the adsorption of, ostensibly, N-succinimidyl3-thiopropionate (NSTP). We base this assignment on the fact that, on gold surfaces, disulfides undergo dissociative chemisorption to give rise to the corresponding (adsorbed) thiolates.26 Cyclic voltammetric measurements of the modified electrode were carried out in 0.1 M phosphate buffer (pH 7.0) at 50 mV/s, and the results are presented in Figure 1. First of all, no Faradaic processes were observed over the potential range of 0.0 to -0.4 V for either the bare (Figure 1 inset; curve a) or the DTSP modified gold electrode (Figure 1 inset; curve b). However, in the latter case, there was a significant decrease in the capacitative current as would be anticipated for an electrode covered with a low dielectric layer. This establishes that upon immersion of a gold electrode in a DTSP solution, an adsorbed layer is formed and that it is stable to through-rinsing with aqueous (phosphate buffer) as well as nonaqueous solvent (acetone). On the first cathodic scan to potentials negative of -0.40 V (Figure 1 solid line; main panel) a well-defined chemically irreversible wave was observed with a peak potential value of -0.75 V. This is in very good agreement with a report by Imabayashi et al.27 for the reductive desorption of a 3-mercaptopropane monolayer from a gold-electrode surface. On the second (Figure 1 main panel; dashed line) and subsequent scans, the peak was essentially absent. Given the similarity of 3-mercapto-propane to N-succinimidyl-3-thiopropionate, we ascribe the aforementioned (25) Shannon, L. M.; Kay, E.; Lew, J. Y. J. Biol. Chem. 1966, 241, 2166-2172. (26) Dubois, L. H.; Nuzzo, R. G. Annu. Rev. Phys. Chem. 1992, 43, 437-463. (27) Imabayashi, S.; Hobara, D.; Kakiuchi, T. Langmuir 1997, 13, 4502-4504.

5532

Analytical Chemistry, Vol. 71, No. 24, December 15, 1999

Figure 1. Linear sweep voltammograms at 50 mV s-1 of a gold electrode modified with a DTSP monolayer in 0.1 M pH 7.0 phosphate buffer over the potential range from 0.0 to -0.90 V. The inset shows cyclic voltammograms at 50 mV s-1 of (a) a bare gold electrode and (b) a gold electrode modified with a DTSP monolayer over the potential range from 0.0 to -0.4 V.

Scheme 1. Depiction of the formation of a DTSP modified gold surface and subsequent immobilization of HRP

process to the reductive desorption of the monolayer from the electrode surface. This is also in accord with the general propensity of adsorbed thiols(ates) to undergo reductive desorption from gold-electrode surfaces.28 The surface coverage, calculated by integration of the charge under the voltammetric wave, was found to be 1.3 × 10-10 mol cm-2, assuming that the process involves the transfer of two electrons per mole. Measurement of Heme Concentration and Peroxidase Activity of HRP Immobilized onto a DTSP Layer. The immobilization of HRP on DTSP modified gold surfaces ostensibly takes place by nucleophilic attack of primary amino groups of the enzyme on the terminal N-succinimidyl esters in the monolayer as depicted in Scheme 1. At this time, we have no evidence as to which amino group(s) in HRP is(are) responsible for binding. To determine the amount of active HRP immobilized on the DTSP modified gold electrode, a variant of the spectrophotometric method described in the Experimental Section was employed. For this purpose, a DTSP-HRP modified electrode was immersed in (28) Widrig, C. A.; Chung, C.; Porter, M. D. J. Electroanal. Chem. 1991, 310, 335-359.

Figure 2. Absorption spectra of solutions containing 500 µM TMB and 50 µM H2O2 in 5.0 mM pH 7.0 phosphate buffer in the absence of HRP (a) and at HRP concentrations of (b) 0.5 (c) 1.0 (d) 1.5, and (e) 2.0 nM. Spectra were recorded 30 s after the addition of the enzyme. Trace (f) corresponds to the spectrum of a DTSP-HRP modified gold electrode after it had been immersed for 30 s in a solution containing 500 µM TMB and 50 µM H2O2 in 5.0 mM pH 7.0 phosphate buffer. The inset presents a calibration curve of absorbance vs amount of HRP. The open circle represents the value for the modified electrode in trace (f) of the main panel.

a spectrophotometric cuvette containing TMB (0.5 mM) and H2O2 (50 µM). The solution was stirred for 30s, the electrode removed from the solution, and the absorption due to the generation of oxidized TMB was recorded. The absorption spectra of solutions containing TMB and H2O2 under the same conditions with increasing (through the addition of a stock HRP solution) amounts of HRP were also recorded (as described in the Experimental Section), and the results are shown in Figure 2. In all cases there was an absorption band at 373 nm due to formation of the oxidized form of TMB whose concentration could be correlated to the amount of active enzyme present. By plotting the absorbance (measured after 30 s) at 373 nm vs added HRP, the calibration curve shown in the inset to Figure 2 was obtained. From the calibration curve (Figure 2 inset) and using the absorbance resulting from the HRP immobilized on the electrode (0.239), the amount of immobilized active enzyme was determined to be 1.6 × 10-12 mol. This value corresponds to a surface coverage of active enzyme of 1.8 × 10-11 mol/cm2. When compared to the surface coverage value of N-succinimidyl-3-thiopropionate of 1.3 × 10-10 mol/cm2, each HRP molecule accounts for about 7 equiv of the former. This would suggest that the HRP molecules react with a number of N-succinimidyl-3-thiopropionate molecules or that the “foot-print” of HRP is such that it effectively covers the area of 7 N-succinimidyl-3-thiopropionate molecules. However, on the basis of our data, we cannot establish which of the two (or both) factors is present and/or dominant. To serve as a comparison to the spectrophotometric measurements, QCM measurements were also carried out. The QCM technique allows the measurement of mass changes at surfaces by changes in the resonant frequency of the quartz crystal. The frequency and mass changes are related by the Sauerbrey

Figure 3. Time dependence of the frequency changes of a DTSP modified quartz crystal resonator in 5.0 mM phosphate buffer solution at 25 °C, upon the addition of HRP (to a concentration of 0.23 µM). The solid line represents a double exponential fit to the data with the vertical dashed line separating the two intervals. Inset A: Logarithmic plot of data in the main panel; Inset B: Time dependence of the resistance parameter of the quartz crystal resonator.

equation29

∆m ) -Cf ∆F

(1)

where ∆F is the change in frequency (Hz), ∆m is the mass change (ng cm-2), and Cf (17.7 ng Hz-1 cm-2) is a proportionality constant for the 5 MHz crystals used in this study. To study the mass changes associated with the immobilization of the enzyme (HRP) onto a DTSP monolayer, a DTSP modified QCM Au resonator was immersed in a thermostated solution (50 mL of 5.0 mM phosphate buffer, pH 7.0) and the frequency monitored as a function of time. After the temperature and frequency had stabilized, 50 µL of the HRP stock solution (230 µM) were added to the solution, so that the final heme concentration was 0.23 µM. Figure 3 shows the resulting frequency changes as a function of time. As can be seen, upon the addition of the enzyme, the frequency decreased rapidly during the first 30 min and subsequently (Figure 3; vertical dashed line) decreased more slowly until a steady state was reached. Assuming that the immobilization process is kinetically controlled, we attempted to fit these data but were unable to obtain an acceptable fit with a simple exponential. On the other hand, excellent results were obtained by a double exponential fit, as shown in Figure 3 (solid line) and in inset A to Figure 3. The vertical dashed line in the main panel indicates the two separate regions. In addition, in inset A, where the data are plotted on a logarithmic scale, the presence of two separate linear segments is clearly evident. From the fits, values of 0.08 and 0.02 min-1 were obtained for the first and second processes, respectively. Although largely speculative on (29) Sauerbrey, G. Z. Phys. 1959, 155, 206-222.

Analytical Chemistry, Vol. 71, No. 24, December 15, 1999

5533

our part, we believe that the initial (faster) process corresponds to the reaction between surface-bound N-succinimide groups and amino groups of the enzyme. The second process could be due to reorganization of the enzyme, with the possible incorporation of additional enzyme and/or water and supporting electrolyte. The overall decrease in frequency was 27 Hz. From this change in frequency and using eq 1, the mass of HRP deposited on the modified electrode surface was estimated to be 655 ng (after taking into account the area of the oscillator: 1.37 cm2) which, if due only to the immobilized HRP, would correspond to a surface coverage of about 1.5 × 10-11 mol cm-2. This is in excellent agreement with the value calculated by the spectrophotometric method. This also suggests that for the second (slower) process alluded to above, most of the frequency change appears to be associated with additional incorporation of enzyme (HRP). The above-mentioned value of the surface coverage is also comparable to that reported by Oyama et al. for HRP adsorbed on graphite.30 Since the frequency of the quartz crystal resonators is sensitive not only to mass but also to the rigidity, roughness, and viscoelasticity of the film surface and this sensitivity becomes more significant as the film thickness is increased, it is important to establish the main factors giving rise to the frequency changes during the immobilization on the quartz crystal resonator.31 Admittance measurements of the quartz crystals were carried out during the immobilization of the HRP, following the same procedure as the one used to measure the frequency decrease. The resistance parameter as a function of time is presented in inset B to Figure 3, where it can be observed that its value remains essentially constant during the experiment. This demonstrates that the main factor giving rise to the frequency changes is a change in the mass of the resonator, so, therefore, frequency and mass changes can be correlated and eq 1 is applicable. Electrocatalytic Reduction of Hydrogen Peroxide. The electrochemical activity of redox enzymes is of particular importance in the development of amperometric biosensors. The electrical communication/connection between redox proteins and electrode surfaces provides a general means of enhancing the activity of redox-active biocatalysts.32 Direct contact between the protein’s redox center and the electrode surface is generally ineffective since the former is often buried inside the protein matrix. Various methodologies have been employed to enhance the interaction between redox proteins and electrodes. One of the more common approaches involves modification of the electrode surface with promoter molecules that bind and/or align the protein in such a way so as to facilitate electron transfer.33,34 To ascertain if this was the case for the DTSP-HRP modified gold electrodes, their activity toward H2O2 was examined. However, no electrocatalytic activity was observed, suggesting that the DTSP had little, if any, effectiveness as a promoter. In general, it is found that when promoters are used the heterogeneous charge-transfer rate is typically much lower than it is when redox (30) Tatsuma, T.; Ariyama, K.; Oyama, N. J. Electroanal. Chem. 1998, 446, 205209. (31) Buttry, D. A.; Ward, M. D. Chem. Rev. 1992, 92, 1355-1379. (32) Bartlett, P. N.; Tebbut, P.; Whitaker, R. G. Prog. React. Kinet. 1991, 16, 55-60. (33) (a) Armstrong, F. A.; Hill, H. A. O.; Walton, N. J. Acc. Chem. Res. 1988, 21, 407-413. (b) Frew, J. E.; Hill, H. A. O. Eur. J. Biochem. 1988, 172, 261-269. (34) Ruzgas, T.; Cso ¨regi, E.; Emne´us, J.; Gorton, L.; Marko-Varga, G. Anal. Chim. Acta 1996, 330, 123-138.

5534 Analytical Chemistry, Vol. 71, No. 24, December 15, 1999

Table 1. Formal Potentials, in Volts vs SSCE, for the Transition-Metal Complexes Acting as Diffusional Mediators in Aqueous Solution (pH 7.0) mediator

E°′ (quinone-based)

E°′ (metal-based)a

[Os(phen-dione)(phen)2](PF6)2 [Ru(phen-dione)(bpy)2](PF6)2 [Ru(phen-dione)2(bpy)](PF6)2 [Ru(phen-dione)3](PF6)2

-0.035 -0.040 -0.045 -0.020

+0.93 +1.35 +1.43 +1.40

a

In acetonitrile/0.1M TBAP.

Figure 4. Cyclic voltammetric response at 10 mV/s for a DTSPHRP modified gold electrode in a deareated solution of 0.2 mM [Os(phen-dione)(phen)2](PF6)2 in 0.1 M phosphate buffer (pH 7.0) as a function of H2O2 concentration: (a) 0.0, (b) 0.12, (c) 0.25, (d) 0.38, (e) 0.50, and (f) 0.75 mM.

mediators are used.35 Thus, an alternative approach, based on the use of redox mediators, was pursued. In solution, electron-transfer mediators operate by a diffusional route facilitating electrical contact between the enzyme’s redox center and the electrode surface. In the case of peroxidases, it has been shown that both one- and two-electron mediators can be employed. Typical of the first would be ferrocenyl derivatives, hexacyanoferrates, and other transition metal complexes.36 In the case of the latter, quinones/hydroquinones and related materials are often used. In this case, the kinetics are typically much faster and the mediator also provides the necessary protons.37 In the present work, the transition metal complex [Os(phen-dione)(phen)2](PF6)2 was used as an electron-transfer mediator in solution. The intent was to employ the quinone moiety within the phen-dione ligand as redox mediator. It was anticipated that the metal-localized (e.g., Os(II/III)) redox process within the complex would not exhibit any catalytic activity due to its rather positive formal potential (Table 1 and ref 24). The cyclic voltammetric responses (at slow sweep rate) of the DTSP-HRP modified gold electrodes with the metal complex in solution and in the presence and absence of hydrogen peroxide were used to assess the activity of the enzymatic electrode. Curve a in Figure 4 shows the cyclic voltammetric response from +0.2 to -0.2 V at 10 mV/s for a DTSP-HRP modified gold electrode in contact with a 0.1 M pH 7.0 phosphate buffer solution containing 0.2 mM of the osmium complex but in the absence of hydrogen peroxide. The well(35) Paddock, P. M.; Bowden, E. F. J. Electroanal. Chem. 1989, 260, 487. (36) Schubert, F.; Saini, S. Anal. Chim. Acta 1991, 245, 133. (37) (a) Foulds, N. C.; Lowe, C. R. J. Chem. Soc., Faraday Trans. 1986, 82, 1259-1264. (b) Pantano, P.; Morton, T. H.; Kuhr, W. G. J. Am. Chem. Soc. 1991, 113, 1832.

Figure 5. Catalytic current (b) and peak potential (0) vs H2O2 concentration for a DTSP-HRP modified gold electrode, using [Os(phen-dione)(phen)2](PF6)2 as a diffusional mediator. Data were obtained from the cyclic voltammograms in Figure 4.

behaved redox response of the phen-dione Os complex in aqueous media is readily apparent.24,38 Under these conditions, transition metal complexes of phen-dione exhibit pH-dependent redox responses centered on the quinone moieties, and in fact, such a dependence has previously been established.24,38 Upon the addition of hydrogen peroxide (to a final concentration of 0.12 mM), an enhancement of the cathodic current (Figure 4b) is clearly noted. Additional increases in the concentration of hydrogen peroxide to (c) 0.25, (d) 0.38, (e) 0.50, and (f) 0.75 mM resulted in concomitant increases in the cathodic peak current. At hydrogen peroxide concentrations above 0.2 mM, there was no peak in the reverse (anodic) potential sweep, consistent with a high degree of electrocatalytic activity. The response of the biosensor to hydrogen peroxide was linear to 0.40 mM. For higher hydrogen peroxide concentrations (see Figure 5) the response tended to level off and eventually remained constant, indicating a saturation response. It can also be observed in Figure 5 that the peak potential of the catalytic wave shifted to more negative values as the hydrogen peroxide concentration increased, again suggesting a kinetic limitation. It should also be mentioned that no electrocatalytic activity was observed when the potential was scanned over the Os(II/III) wave centered at about +0.93 V. As mentioned above, this would be the anticipated result given the high value of the formal potential. The electrocatalytic activity of the mediator toward the reduction of hydrogen peroxide in the absence of HRP was also tested. For this purpose, a DTSP modified gold electrode (without HRP) was immersed in a buffer solution containing 0.2 mM [Os(phendione)(phen)2](PF6)2, and no electrocatalytic response was observed upon the addition of H2O2 to a concentration of 2 mM. The EQCM could also be used to follow the electrochemical response of the soluble mediator ([Os(phen-dione)(phen)2](PF6)2) at a DTSP-HRP modified gold electrode in the absence and in the presence of hydrogen peroxide. Figure 6 presents the cyclic voltammogram and the frequency responses as a function of applied potential in a 0.1 M pH 7.0 phosphate buffer solution (38) Wu, Q.; Maskus, M.; Pariente, F.; Tobalina, F.; Ferna´ndez, V. M.; Lorenzo, E.; Abrun ˜a, H. D. Anal. Chem. 1996, 68, 3688-3696.

Figure 6. Cyclic voltammogram and frequency-potential curves for a DTSP-HRP modified quartz crystal resonator in 0.1 mM [Os(phen-dione)(phen)2](PF6)2 in 0.1 M pH 7.0 phosphate buffer (in the absence of O2). The potential was swept between +0.20 and -0.50 V at 10 mV/s, in the absence (A) and in the presence of 0.25 mM of H2O2 (B). A silver wire was used as the reference electrode (Ag° vs SCE; ∆E ) 170 mV).

containing 0.1 mM ([Os(phen-dione)(phen)2](PF6)2) at a DTSPHRP modified electrode in the absence (A) and in the presence (B) of hydrogen peroxide. As mentioned previously, since the frequency of the quartz crystal resonator is sensitive not only to mass but also to rigidity and other properties of the film surface, admittance measurements of the quartz crystal resonators were carried out. No difference was observed in the resistance parameter between the oxidized (+0.20 V) and reduced (-0.50 V) states, indicating that the use of eq 1 was justified. As can be seen in Figure 6A, in the absence of substrate (H2O2), the wellbehaved redox response of the osmium complex was observed. The reduction causes a decrease in the frequency, corresponding to an increase in the mass of the quartz crystal resonator, while the oxidation reaction causes an increase in the frequency, corresponding to a decrease in the mass. This fact indicates that the mediator is adsorbed onto the DTSP-HRP modified electrode, since, otherwise, no changes in frequency with applied potential would be observed. Given that the redox reaction involves 2 protons, the difference in mass would be 2 amu, which would give rise to a frequency change well below our resolution. However, as can be ascertained from Figure 6, the overall frequency decrease upon reduction was -5.7 Hz. We thus ascribe these frequency changes as arising from changes in the degree of solvation of the quinone and hydroquinone groups, with the latter being significantly more solvated than the former, as a result of the enhanced ability of hydroquinone to form multiple hydrogen bonds with solvent molecules. When EQCM experiments were performed using a bare gold quartz crystal resonator in contact with a 0.1 M phosphate buffer solution (pH 7.0) containing 0.1 mM [Os(phen-dione)(phen)2]2+, a similar CV and similar frequency-potential curves (∆F ) -6.1 Hz) to those obtained using the DTSP-HRP modified EQCM electrode were obtained. This also clearly indicates that the decrease in the frequency upon reduction of the quinone sites of the Os complex is due to adsorbed Os complex. Further, the shape of the CV did not appear to be that resulting from deposition/ stripping processes, indicating that the changes in the frequency Analytical Chemistry, Vol. 71, No. 24, December 15, 1999

5535

are unlikely to have been caused by deposition and dissolution of the Os complex itself upon reduction and oxidation, respectively. In addition, Meq (g mol-1), which gives an indication of the change in mass per adsorbed Os complex molecule during the redox reaction, was calculated to be 136. This value would be too small if the adsorption/desorption of the Os complex itself was responsible for the changes in frequency given its much higher molecular weight (MW of cation ) 760.8). Thus, assuming that the contribution to changes in the frequency from diffusional species is negligible, and also that only protons move across the interface between the film and the solution upon the redox reaction, the number of water molecules (solvation) accompanying each proton was calculated, from the surface coverage of the Os complex (4.0 × 10-10 mol cm-2) and the frequency change (6.1 Hz), to be 7.5. Similar results have been previously reported by Scherson et al., who found (also using an ECQM) that, for thiolsubstituted quinones adsorbed on gold, the redox reaction gave rise to a change in solvation of about 3 water molecules.39 The addition of hydrogen peroxide to a concentration of 0.25 mM gave rise to an electrocatalytic current (Figure 6B), indicating that the enzymatic reduction of hydrogen peroxide proceeds at the DTSP-HRP modified gold surface of the quartz crystal resonator. The overall frequency decrease accompanying the redox reaction was -4.8 Hz. This value is also much larger than that which would correspond to the exchange of two protons. This, again, suggests that the frequency changes are due to changes in the degree of solvation of the quinone and hydroquinone groups. In the presence of hydrogen peroxide the overall frequency change was slightly lower than that observed in the absence of hydrogen peroxide. This may be due to the fact that, upon the addition of peroxide, the electrochemically generated hydroquinone groups are rapidly reoxidized while the enzyme is returned to its native state. Thus, less hydrogen bonds can be formed, giving rise to a lower frequency decrease. Effect of Oxygen. In preliminary experiments, the cyclic voltammetric response of a DTSP-HRP modified gold electrode was obtained between +0.15 and -0.25 V at 10 mV/s, in a 0.1 M pH 7.0 phosphate buffer containing 0.2 mM [Os(phen-dione)(phen)2](PF6)2 and in the presence of oxygen. Under these conditions, an electrocatalytic current can be observed in the absence of hydrogen peroxide, indicating that oxygen represents a significant interferrent. To ascertain whether the oxygen interference was due to its effect on the mediator or on the HRP, a similar experiment was carried out with an unmodified gold electrode. Under these conditions, the same interference was observed, indicating that the complex is involved in the electrocatalytic reduction of oxygen. To avoid this interference, all the experiments were carried out in the absence of oxygen by degassing. Effects of Metal Center and Nature of Ligands. To test the influence of the transition metal of the complex as well as the effect of the number of phen-dione ligands, we have also determined the electrocatalytic activity of DTSP-HRP modified electrodes using [Ru(phen-dione)(bpy)2](PF6)2, [Ru(phen-dione)2(bpy)](PF6)2, and [Ru(phen-dione)3](PF6)2 as mediators in solution. As has been noted for the osmium complex, the voltammetric (39) Mo, Y.; Sandifer, M.; Sukenik, C.; Barrida, R. J.; Soriaga, M. P.; Scherson, D. Langmuir 1995, 11, 4626.

5536 Analytical Chemistry, Vol. 71, No. 24, December 15, 1999

Figure 7. Calibration curve for an DTSP-HRP modified gold electrode using 0.1 mM [Ru(phen-dione)(bpy)2](PF6)2 (9), 0.1 mM [Ru(phen-dione)2(bpy)](PF6)2 ([), and 0.1 mM [Ru(phen-dione)3](PF6)2 (O) as diffusional mediators in 0.1 M pH 7.0 phosphate buffer solution.

responses in all cases prior to the addition of hydrogen peroxide were similar to those previously described for these complexes and are ascribed to the reversible two-electron/two-proton oxidation/reduction of the hydroquinone/quinone group on the phendione ligands. The values of the formal potentials for these complexes, in aqueous solution, are summarized in Table 1. The addition of hydrogen peroxide gave rise to large increases in the cathodic current while the anodic wave virtually disappeared, which is characteristic of a strong electrocatalytic effect. Figure 7 presents plots of the catalytic current vs H2O2 concentration over the range of 0.10-0.80 mM for the different complexes used as soluble mediators. To normalize the responses to the different electrodes employed, the currents are plotted as (ic - id)/id where ic is the current under catalytic conditions (i.e., in the presence of substrate; H2O2) and id is the current due to the mediator in the absence of substrate. As can be seen, in all cases initially the cathodic peak current increased linearly with hydrogen peroxide concentration and then leveled off, suggesting a saturation response. However, the current densities at the same hydrogen peroxide concentration for complexes with different numbers of phen-dione are different, suggesting that the kinetics of regeneration of the enzyme’s active form, mediated by the complex in solution, are likely rate-limiting. Regeneration of the active form of HRP is believed to take place in two one-electron steps where both one- and two-electron donors are active with, as mentioned above, the latter being much more effective.34 Thus, one would expect a higher catalytic signal (at the same peroxide concentration) when mediators with greater numbers of phendione ligands are used. Such an anticipated response was indeed observed so that the activity of [Ru(phen-dione)3](PF6)2 was higher than that of [Ru(phen-dione)2(bpy)](PF6)2 which, in turn, was higher than that of [Ru(phen-dione)(bpy)2](PF6)2. However, it should also be mentioned that their relative reactivity was not in the ratio of 3:2:1. Although at this time we are not certain as to the origin of this effect, it might arise from hydrogen-bonding effects which would “tie-up” the reactive sites and would also be expected to increase with increasing number of phen-dione

ligands. Thus, the effective activity of a given mediator would be a balance between the number of phen-dione ligands, on one hand, and hydrogen-bonding effects on the other. CONCLUSIONS Dithiobis-N-succinimidyl propionate (DTSP) undergoes dissociative chemisorption on gold to give rise to the modification of the surface with N-succinimidyl-3-thiopropionate (NSTP) at a coverage of about 1.3 × 10-10. The pendant succinimidyl groups in NSTP can react with amino groups of horseradish peroxidase (HRP) so that it becomes covalently bound to the NSTP modified gold surface. The amount of active HRP immobilized has been estimated from spectrophotometric and QCM measurements giving values of 1.8 × 10-11 and 1.5 × 10-11 mol/cm2, respectively. On the basis of QCM measurements, the binding reaction of HRP with NSTP modified gold surfaces appears to involve two steps with the first corresponding to the rapid binding of HRP and the second to incorporation of additional enzyme and/or reorganization of the immobilized layer. Electrochemical and spectrophotometric assays indicate that the immobilized HRP retains its

enzymatic activity after immobilization. A peroxide biosensor was developed making use of a gold surface modified with DTSP and HRP employing Os and Ru complexes of 1,10-phenanthroline 5,6dione (phen-dione) of the type [M(phen-dione)x(L)3-x]+2 (where L ) 1,10-phenanthroline or 2,2′-bipyridine, x ) 1-3). The efficiency of the mediator was dependent on the nature of the ligands within the complex, with higher activity being exhibited by complexes with a higher number of phen-dione ligands. ACKNOWLEDGMENT This work was supported by the DGICYT of Spain through Grants BIO 961016-C02-02 and PB-97-0037, the Comunidad Auto´noma de Madrid (Grant 06M/044/96), and the National Science Foundation. M.D. also acknowledges support by a Fellowship from the Comunidad Auto´noma de Madrid.

Received for review July 12, 1999. Accepted September 23, 1999. AC990759X

Analytical Chemistry, Vol. 71, No. 24, December 15, 1999

5537