Electrochemical Control of Protein Monolayers at Indium Tin Oxide

Langmuir 2004, 20, 1901-1908

1901

Electrochemical Control of Protein Monolayers at Indium Tin Oxide Surfaces for the Reagentless Optical Biosensing of Nitric Oxide Duncan H. P. Hedges,† David J. Richardson,‡ and David A. Russell*,† School of Chemical Sciences and Pharmacy and School of Biological Sciences, University of East Anglia, Norwich NR4 7TJ, United Kingdom Received September 25, 2003. In Final Form: December 15, 2003 Cytochrome c has been immobilized onto functionalized, optically transparent indium tin oxide (ITO) electrodes by covalent and electrostatic techniques. Covalent immobilization was achieved by the formation of a disulfide bond between N-succinimidyl 3-(2-pyridyldithio)propionate- (SPDP-) modified cytochrome c and SPDP-silanized ITO. Additionally, ITO electrodes have been modified with the bifunctional reagent 1,12-dodecanedicarboxylic acid (DDCA), resulting in formation of a carboxylic acid-terminated monolayer. Covalent protein attachment to the DDCA-functionalized ITO was achieved with the cross-linker 1-[3(dimethylamino)propyl]-3-ethylcarbodiimide hydrochloride. Electrostatic attachment of the protein involved ion-pair and hydrogen-bond interactions between the terminating carboxylic acid groups of the DDCAfunctionalized ITO and the primary amine groups of the lysine residues of cytochrome c. The electrostatic interaction between the cytochrome c and the functionalized ITO resulted in greater rotational mobility of the protein at the electrode surface, leading to ca. 63% electroactivity, as compared to ca. 41% electroactivity for the covalently immobilized protein. The redox state of the electrostatically bound cytochrome c monolayers could be electrochemically switched between ferric and ferrous forms. Electrochemical control of the bound protein was used to regenerate the biosensing surface following binding of nitric oxide (NO). Ligation of NO with the cytochrome c was monitored by measurement of the change of absorbance intensity at 416 nm. Through application of a negative potential, the cytochrome c was reduced from the ferric to the ferrous form, which led to the removal of the ligated NO. Application of a positive potential regenerated the ferric cytochrome c, enabling multiple repeat measurements of NO. Such electrochemical control of proteins immobilized on transparent electrodes enables the optical biosensing of analyte targets without recourse to exogenous reagents.

Introduction The use of optically clear electrodes enables the coupling of spectroscopic techniques with electrochemistry.1 Typically, such coupling configurations have been applied to the elucidation of reaction rates and mechanisms, although applications in other fields such as biosensing are also apparent. A limited number of reports exist for the development of optical biosensors with electrochemically controlled regeneration of the sensing element, thus enabling multiple analyte measurements.2-4 Such schemes are attractive since they eliminate the need for an additional reagent to regenerate the molecular recognition element through removal of the analyte species and/or changing the redox state of the binding center. The difficulty of analyte removal has led to many biosensors being developed with a view to single-use disposable systems. The development of an efficient mechanism for the removal of bound analyte will therefore advance biosensor technology by providing multiple-use sensors. For biosensing systems, the immobilization of the biological sensing element in a stable configuration is an * Corresponding author: e-mail [email protected]. † School of Chemical Sciences and Pharmacy. ‡ School of Biological Sciences. (1) Heineman, W.; Hawkridge, F.; Blout, H. Spectroelectrochemistry at Optically Transparent Electrodes. In Electroanalytical Chemistry, Vol 13; Bard, A. J., Ed.; Marcel Dekker: New York, 1986. (2) Asanov, A. N.; Wilson, W. W.; Oldham, P. B. Anal. Chem. 1998, 70, 1156-1163. (3) Topoglidis, E.; Campbell, C. J.; Cass, A. E. G.; Durrant, J. R. Langmuir 2001, 17, 7899-7906. (4) Topoglidis, E.; Lutz, T.; Willis, R. L.; Barnett, C. J.; Cass, A. E. G.; Durrant, J. R. Faraday Discuss. 2000, 116, 35-46.

integral part of a successful design. Immobilization of biological molecules onto an optically transparent indium tin oxide (ITO) electrode is a relatively new development. ITO can be pretreated to provide a surface of hydroxyl groups to which various linkers can be attached to enable protein immobilization. A commonly used strategy is to silanize the ITO,2,5-12 which is then used to covalently immobilize biological molecules.2,5-8 The ITO surface can also be modified with bifunctional reagents,13-22 such as dicarboxylic acids,14 thus producing a monolayer with a (5) Willner, I.; Blonder, R. Thin Solid Films 1995, 266, 254-257. (6) Xu, J.; Zhu, J.-J.; Huang, Q.; Chen, H.-Y. Electrochem. Commun. 2001, 3, 665-669. (7) Wilson, R.; Schiffrin, D. J. Analyst 1995, 120, 175-178. (8) Wilson, R.; Kremeskotter, J.; Schiffrin, D. J.; Wilkinson, J. S. Biosens. Bioelectron. 1996, 11, 805-810. (9) Hillebrandt, H.; Tanaka, M. J. Phys. Chem. B 2001, 105, 42704276. (10) Markovich, I.; Mandler, D. J. Electroanal. Chem. 2001, 500, 453-460. (11) Chen, K.; Caldwell, W. B.; Mirkin, C. A. J. Am. Chem. Soc. 1993, 115, 1193-1194. (12) Wei, T.-X.; Zhai, J.; Ge, J.-H.; Gan, L.-B.; Huang, C.-H.; Luo, G.-B.; Ying, L.-M.; Liu, T.-T.; Zhao, X.-S. Appl. Surf. Sci. 1999, 151, 153-158. (13) Liron, Z.; Tender, L. M.; Golden, J. P.; Ligler, F. S. Biosens. Bioelectron. 2002, 17, 489-494. (14) Napier, M. E.; Thorp, H. H. Langmuir 1997, 13, 6342-6344. (15) Popovich, N. D.; Eckhardt, A. E.; Mikulecky, J. C.; Napier, M. E.; Thomas, R. S. Talanta 2002, 56, 821-828. (16) He, P.-G.; Takahashi, T.; Hoshi, T.; Anzai, J.; Suzuki, Y.; Osa, T. Mater. Sci. Eng. C 1994, C2, 103-106. (17) Yan, C.; Zharnikov, M.; Golzhauser, A.; Grunze, M. Langmuir 2000, 16, 6208-6215. (18) Tanaka, T.; Honda, Y.; Sugi, M. Jpn. J. Appl. Phys. 1995, 34, 3250-3254. (19) Gardner, T. J.; Frisbie, C. D.; Wrighton, M. S. J. Am. Chem. Soc. 1995, 117, 6927-6933.

10.1021/la035795c CCC: $27.50 © 2004 American Chemical Society Published on Web 01/21/2004

1902

Langmuir, Vol. 20, No. 5, 2004

reactive terminating group to which the biomolecule can be bound, either electrostatically16 or covalently.13-15 Other immobilization methods on ITO include adsorption to the surface hydroxyls23-29 and the formation of LangmuirBlodgett films.30-33 Recently, Durrant and co-workers34 have absorbed cytochrome c and hemoglobin onto nanocrystalline tin oxide and titanium dioxide electrodes to study the electrochemical behavior of these electrodes. Further, these workers demonstrated the use of hemoglobin on the tin oxide electrode for the electrochemical biosensing of nitric oxide (NO). We are interested in developing optical biosensors and in this article report the use of cytochrome c, as the biological sensing molecule, deposited onto ITO electrodes for the optical detection of NO. Previously we have encapsulated cytochrome c35,36 and another heme protein, cytochrome c′,37 within silica sol-gel matrixes for the measurement of NO. Such biosensing measurements relied on the use of chemical reductants and oxidants to regenerate the heme center for multiple use. The reactivity of cytochrome c toward NO is governed by the protein’s oxidation state.38 At pH 7, ferric cytochrome c reacts rapidly with NO. However, under the same pH conditions, reaction of ferrous cytochrome c with NO proceeds slowly. Therefore, by electrochemically switching between the redox forms of the cytochrome c, the removal of NO and subsequent regeneration of the ferric form should provide a multiple-use, reagentless, biosensing platform. The reaction scheme for the operation of the electrochemically controlled optical biosensor is shown in Scheme 1. In this paper, three different immobilization strategies are investigated to determine the optimum attachment of cytochrome c to an ITO electrode surface for maximal electroactivity. The optimized configuration was then used to establish that the redox state of the cytochrome c can be electrochemically controlled on the ITO surface. The stable form of cytochrome c is the ferric form, which readily binds NO. Upon application of a negative potential, the (20) Oh, S.; Yun, Y.; Kim, D.; Han, S. Langmuir 1999, 15, 46904692. (21) Oh, S.-Y.; Han, S.-Y. Langmuir 2000, 16, 6777-6779. (22) Breen, T. L.; Fryer, P. M.; Nunes, R. W.; Rothwell, M. E. Langmuir 2002, 18, 194-197. (23) El Kasmi, A.; Leopold, M. C.; Galligan, R.; Robertson, R. T.; Saavedra, S. S.; Kacemi, K. E.; Bowden, E. F. Electrochem. Commun. 2002, 4, 177-181. (24) Bowden, E. F.; Hawkridge, F. M.; Blount, H. N. J. Electroanal. Chem. 1984, 161, 355-376. (25) Daido, T.; Akaike, T. J. Electroanal. Chem. 1993, 344, 91-106. (26) Fang, A.; Ng, H. T.; Su, X.; Li, S. F. Y. Langmuir 2000, 16, 5221-5226. (27) Fang, A.; Ng, H. T.; Li, S. F. Y. Langmuir 2001, 17, 4360-4366. (28) Pyon, M.-S.; Cherry, R. J.; Bjornsen, A. J.; Zapien, D. C. Langmuir 1999, 15, 7040-7046. (29) Armistead, P. M.; Thorp, H. H. Anal. Chem. 2000, 72, 37643770. (30) Qian, D.; Nakamura, C.; Noda, K.; Zorin, N. A.; Miyake, J. Appl. Biochem. Biotechnol. 2000, 84-86, 409-418. (31) Noda, K.; Zorin, N. A.; Nakamura, C.; Miyake, M.; Gogotov, I. N.; Asada, Y.; Akutsu, H.; Miyake, J. Thin Solid Films 1998, 327-329, 639-642. (32) Ramanathan, K.; Ram, M. K.; Malhotra, B. D.; Murthy, A. S. N. Mater. Sci. Eng. C 1995, 3, 159-163. (33) Lee, S.; Anzai, J.-I.; Osa, T. Sens. Actuators B 1993, 12, 153158. (34) Topoglidis, E.; Astuti, Y.; Duriaux, F.; Gra¨tzel, M.; Durrant, J. R. Langmuir 2003, 19, 6894-6900. (35) Blyth, D. J.; Aylott, J. W.; Richardson, D. J.; Russell, D. A. Analyst 1995, 120, 2725-2730. (36) Aylott, J. W.; Richardson, D. J.; Russell, D. A. Chem. Mater. 1997, 9, 2261-2263. (37) Blyth, D. J.; Aylott, J. W.; Moir, J. W. B.; Richardson, D. J.; Russell, D. A. Analyst 1999, 124, 129-134. (38) Yoshimura, T.; Suzuki, S. Inorg. Chim. Acta 1988, 152, 241249.

Hedges et al. Scheme 1 Reaction Pathway of the Cytochrome c-Based NO Biosensor

ferrous form of the cytochrome c was produced, releasing NO into the surrounding solution. The NO was then removed by flushing the biosensing cell with buffer. A positive potential was applied to regenerate the ferric cytochrome c, enabling further NO binding measurements to be made. The binding of NO to the cytochrome c was monitored through measurement of the change in absorbance intensity at the Soret band wavelength (416 nm) of the protein. Subsequently the cytochrome c-ITO electrode system was used to make multiple measurements of NO via the reagentless regeneration cycle. Experimental Section Reagents. Horse heart cytochrome c, 1,12-dodecanedicarboxylic acid (DDCA), (3-aminopropyl)triethoxysilane (APTS), dithiothreitol (DTT), N-[3-(dimethylamino)propyl]-N′-ethylcarbodiimide hydrochloride (EDC), nitric oxide, N-succinimidyl 3-(2pyridyldithio)propionate (SPDP), sodium dithionite, and potassium hexacyanoferrate(II) trihydrate were all obtained from Sigma-Aldrich (Gillingham, Dorset, U.K.). Indium tin oxide(ITO-) coated glass was obtained from Crystran Ltd. (Poole, Dorset, U.K.). All other reagents were obtained from Fisher Scientific U.K. Ltd. (Loughborough, Leicestershire, U.K.). Doubly distilled, deionized water was used for all solutions. Prior to use, the cytochrome c was chromatographically purified to remove the deamidated forms, as previously described.39 Toluene was dried by distillation. All other chemicals were used as received. Substrate Preparation. ITO substrates were pretreated to ensure an active hydroxyl surface layer was present. Pretreatment involved immersion of the substrates in a mixture of 5:1:1 H2O/H2O2(30%)/NH3(25%) for 1 h at 70 °C.10 The glass substrates were then washed with copious amounts of water and blown dry with nitrogen. Immobilization of Cytochrome c onto ITO Substrates. Three methods were investigated to bind the cytochrome c to the ITO substrates to produce the protein-bound electrodes. These involved (i) silanization of the ITO surface and the subsequent covalent attachment of the modified protein via disulfide bond formation or (ii) electrostatic attachment of dodecanedicarboxylic acid to the ITO surface, followed by either (a) covalent attachment of the protein to the carboxylic acid monolayer or (b) direct electrostatic attachment of the protein to the carboxylic acid monolayer. (i) Silanization of ITO and Subsequent Cytochrome c Immobilization. The pretreated ITO substrates were silanized in a refluxing 10% solution of APTS in dry toluene for 30 min. Substrates were then washed with dry toluene and acetone and allowed to dry at room temperature. The silanized ITO surface was further modified through the introduction of pyridyl disulfide moieties following the method of Jonsson et al.40 The silanized (39) Brautigan, D. L.; Ferguson-Miller, S.; Margoliash, E. Methods Enzymol. 1978, 53D, 128-164.

Protein-Modified ITO Surfaces as Optical Biosensors

Langmuir, Vol. 20, No. 5, 2004 1903

Scheme 2. Covalent Immobilization of Cytochrome c onto Silanized ITO Substratesa

a (A) Silanization of ITO surface with APTS; (B) further surface modification with SPDP and reduction with DTT; (C) cytochrome c modification with SPDP; (D) covalent attachment of the SPDP-modified cytochrome c to the functionalized ITO substrate surface.

substrates were immersed in a solution containing 5-10 mM SPDP, 100 mM NaCl, and 1 M EDTA in 100 mM sodium phosphate buffer, pH 7.5, for 30 min. The substrates were rinsed and then reduced with 100 mM DTT in water for 30 min, after which the substrates were again rinsed. The cytochrome c protein was modified with SPDP according to the method of Carlsson et al.41 A SPDP:cytochrome c ratio of 4:1 was used. The SPDPbound protein was then covalently attached to the modified ITO surface by disulfide bond formation, following the method of Jonsson et al.40 This series of surface modifications is shown in Scheme 2. (ii) Electrostatic Attachment of Carboxylic Acid Monolayer on ITO Substrates and Subsequent Cytochrome c Immobilization. Carboxylic acid monolayers were electrostatically bound by immersion of the pretreated ITO substrates in 5 mM DDCA in ethanol for 24-36 h. The functionalized substrates were thor(40) Jonsson, U.; Malmqvist, M.; Olofsson, G.; Ronnberg, I. Methods Enzymol. 1988, 137, 381-388. (41) Carlsson, J.; Drevin, H.; Axen, R. Biochem. J. 1978, 173, 723737.

oughly rinsed in ethanol and 5 mM sodium phosphate buffer, pH 7, to remove weakly bound species. The cytochrome c was then bound to the carboxylic acid monolayer by either covalent or electrostatic attachment strategies. (a) Covalent attachment of the protein was achieved following a method similar to that used by Collinson and Bowden42 for the attachment of proteins to carboxylic acid-terminated self-assembled monolayers (SAMs) on gold. The DDCA-functionalized ITO substrates were immersed in a cytochrome c solution for 15 min. An aliquot of 100 mM EDC in 5 mM sodium phosphate buffer, pH 7, was added. The covalent attachment of the protein to the surface took 30 min. The final protein and EDC concentrations were 30 µM and 5 mM, respectively. (b) Electrostatic attachment of the protein was readily achieved through the immersion of the DDCA-functionalized substrates in a 30 µM cytochrome c solution in 5 mM sodium phosphate buffer, pH 7, for 30 min.42 The reaction schemes for the covalent and electrostatic attachment of the cytochrome c to the DDCA-modified ITO substrates are shown in Schemes (42) Collinson, M.; Bowden, E. F. Langmuir 1992, 8, 1247-1250.

1904

Langmuir, Vol. 20, No. 5, 2004

Hedges et al.

Scheme 3. Covalent Attachment of Cytochrome c to the DDCA-Functionalized ITO Surfacea

a (A) Electrostatic attachment of DDCA onto the ITO surface; (B) activation of the DDCA monolayer with EDC; (C) covalent binding of cytochrome c.

Scheme 4. Electrostatic Attachment of Cytochrome c to the DDCA-Functionalized ITO Surface

3 and 4, respectively. Prior to analytical experimentation, all protein-bound substrates were washed with buffer. Contact Angle Measurements. Pretreated and functionalized ITO substrates were characterized through water contact angles measurements of static drops. At least three drops were placed on each sample. Values reported are the averages of at least four samples. Angle variation from spot to spot did not exceed 4°. Instrumentation. The spectroelectrochemical cell was an inhouse, 1.5 mL three-electrode cell with optically transparent faces. The cell was sealed with a rubber septum that provided entrance slits for the protein-bound ITO working electrode, Ag/ AgCl reference electrode (3 M KCl), and platinum wire counterelectrode. Electrochemical potentials were applied with an µAutolab Type II potentiostat (Eco Chemie, Utrecht, The Netherlands). UV/visible absorption spectra of the protein-coated electrode, under applied potential, were recorded on a Hitachi U-3000 spectrometer (Tokyo, Japan) with the spectroelectrochemical cell fitted in the sample compartment. Redox Change of Immobilized Cytochrome c. Both chemical and electrochemical methods were used to control the redox state of the surface-bound cytochrome c on the ITO electrode. (i) Chemical Control. Addition of 100 µL of sodium dithionite solution (ca. 10 mg mL-1) produced the ferrous form of the cytochrome c protein. Flushing the cell with 5 mM sodium phosphate buffer, pH 7, followed by addition of 100 µL of potassium ferricyanide solution (ca. 10 mg mL-1), resulted in formation of the ferric form. UV/visible absorption spectra were recorded from

400 to 450 nm. The results obtained were displayed as a difference spectrum (ferrous cytochrome c - ferric cytochrome c). (ii) Electrochemical Control. Electrochemically controlled redox changes were achieved by applying potentials of -0.18 and 0.35 V vs Ag/AgCl, for 50 s, for the generation of the ferrous and ferric species, respectively. UV/visible absorption spectra were recorded from 400 to 450 nm. Reference spectra were then recorded at the same two applied voltages after irreversible oxidation of the protein at 1 V for 500 s.42 Reference spectra were subtracted from the corresponding ferrous or ferric protein spectrum. The difference spectrum (ferrous cytochrome c - ferric cytochrome c) was then calculated from the corrected spectra. Thus, the difference spectra from both chemical and electrochemical methods could be compared. NO Binding to the Protein-Coated ITO Substrates. The protein-bound ITO electrode was immersed in 1.3 mL of deoxygenated 5 mM phosphate buffer, pH 7, in the electrochemical cell. NO binding to the surface-bound cytochrome c was achieved by syringe injection of 100 µL of saturated NO solution (giving an approximate 2 mM NO solution)44 through the rubber seal. Reversal of NO binding was achieved by flushing the cell via a gravity-driven flow system from a deoxygenated buffer reservoir. A potential of -0.18 V vs Ag/AgCl was applied on flushing followed by a subsequent potential of 0.35 V for ferric cytochrome c regeneration. (43) Tarlov, M. J.; Bowden, E. F. J. Am. Chem. Soc. 1991, 113, 18471849. (44) Feelisch, M. J. Cardiovasc. Pharmacol. 1991, 17, S25.

Protein-Modified ITO Surfaces as Optical Biosensors

Langmuir, Vol. 20, No. 5, 2004 1905

Table 1. Water Contact Angle Measurements for Pretreated and Functionalized ITO Surfaces substrate

WCA (deg)

pretreated ITO DDCA-functionalized ITO silanized ITO