Electrochemistry and Electrocatalytic Properties of Hemoglobin in

Apr 17, 2007 - Analytical Chemistry 2009 81 (19), 8249-8255 ... Detection of Extracellular H2O2 Released from Human Liver Cancer Cells Based on TiO2 N...
0 downloads 0 Views 387KB Size
Download PDF
Anal. Chem. 2007, 79, 3581-3588

Electrochemistry and Electrocatalytic Properties of Hemoglobin in Layer-by-Layer Films of SiO2 with Vapor-Surface Sol-Gel Deposition Guoyue Shi,* Zhiyu Sun, Meichuan Liu, Li Zhang, Ye Liu, Yunhe Qu, and Litong Jin*

Department of Chemistry, East China Normal University, 3663 Zhongshan Road(N), Shanghai, 200062, People’s Republic of China In this paper, layer-by-layer {Hb/SiO2}n films assembled by alternate adsorption of positively charged hemoglobin (Hb) and vapor-surface sol-gel deposition of silica at 50 °C onto a glassy carbon electrode were reported. The result films were characterized with cyclic voltametery, electrochemical impedance spectroscopy, UV-vis spectroscopy, and SEM, and the direct electrochemical and electrocatalytic properties of Hb in these layer-by-layer films were investigated. A pair of well-defined quasireversible cyclic voltammetric peaks were observed, and the formal potential of the heme FeIII/FeII redox couple was found to be -0.330 V(vs SCE). The electron-transfer behavior of Hb in {Hb/SiO2}n films was dependent on the vapor temperature, the number of layers, and the pH of the Hb solution, based on which a set of optimized conditions for film fabrication was inferred. The hemoglobin in{Hb/SiO2}n films displayed good electrocatalytic activity to the reduction of hydrogen peroxide, and H2O2 had linear current response from 1.0 × 10-6 to 2.0 × 10-4 M with a detection limit of 5.0 × 10-7 M (S/N ) 3). The apparent heterogeneous electron-transfer rate constant (ks) was 1.02 ( 0.03 s-1, and the apparent MichaeliMenten constant (Kmapp) was 0.155 mM, indicating a potential application in the third-generation biosensor. Hemoglobin (Hb), a protein that stores and transports molecular oxygen in mammalian blood, comprises two heme-containing domains, R domain and β domain. It is probably the most thoroughly studied protein and is an excellent mode for research on the direct electron transfer between proteins and underlying electrodes.1 Direct electrochemistry of heme proteins has attracted considerable attention because it provides fundamental knowledge of the redox behavior of proteins in biological system and is a more effective way of fabricating of biosensors, bioreactors, and biomedical devices than using mediators.2-4 Nevertheless, the facilitation of electron transfer between the macromolecule Hb and the electrodes is challenging. The inaccessibility of the heme buried in the large three-dimensional structure and the subsequent * Corresponding authors: (tel) 86-21-62237105; (fax) 86-21-62232627; (e-mail) [email protected], [email protected]. (1) Ohno, H. Electrochim. Acta 1998, 43, 1581-1587. (2) Armstrong, F. A.; Wilson, G. S. Electrochim. Acta 2000, 45, 2623-2645. (3) Hamachi, I.; Fujita, A.; Kunitake, T. J. Am. Chem. Soc. 1997, 119, 90969102. (4) Armstrong, F. A.; Hill, H. A. O.; Walton, N. J. Acc. Chem. Res. 1988, 21, 407-413. 10.1021/ac062034g CCC: $37.00 Published on Web 04/17/2007

© 2007 American Chemical Society

electrode passivation due to protein adsorption make it generally difficult to establish direct electron transfer between hemoglobin and conventional electrodes.5 The layer-by-layer (LBL) films assembled by alternate adsorption of positively charged Hb at pH 5.0 and oppositely charged species from their solutions through electrostatic interactions have been reported as a new method for facilitating electron transfer between the macromolecule of Hb and the electrodes.6 Under appropriate conditions, hemoglobin can be assembled with many oppositely charged materials, such as polyions,7-9 inorganic nanoparticles,10-13 sol-gel-derived inorganic materials,14,15 and so on. Among these oppositely charged materials, sol-gel-derived inorganic materials have emerged as a more attractive matrix for the immobilization of protein because of their porous film architectures, which provide better mass transport and allow larger protein loading per unit area.16 However, the traditional procedures of depositing many sol-gel materials are very complex and need strong acidic solutions, which lead to denaturization of the immobilized protein and limit the sensitivity of result biosensors.17,18 Yu and Lu19,20 reported a titania sol-gel matrix with a new vapor-surface deposition strategy that retained the advantages of conventional sol-gel materials and simplified the procedure of the sol-gel synthetic process. It also removed the drawbacks caused by acid medium and organic solvent needed in traditional sol-gel processes. Proteins immobilized in these films retained their native structures and showed direct, reversible electrochemistry. In addition, the layers’ composition and thickness can be precisely controlled. (5) Stellwagen, E. Nature 1978, 275, 73-74. (6) Decher, G. Science 1997, 227, 1232-1237. (7) Lvov, Y.; Lu, Z.; Schenkman, J.; Zu, X.; Rusling, J. F. J. Am. Chem. Soc. 1998, 120, 4073-4080. (8) Pei, R. J. Biomacromolecules 2001, 2, 463-468. (9) Ma, H.; Hu, N.; Rusling, J. F. Langmuir 2000, 16, 4969-4975. (10) Banerjee, S.; Wong, S. S. J. Am. Chem. Soc. 2003, 125, 10342-10350. (11) Gao, Q.; Suib, S. L.; Rusling, J. F. Chem. Commun. 2002, 19, 2254-2255. (12) Wang, Q.; Lu, G.; Yang, B. Langmuir 2004, 20, 1342-1347. (13) He, P. L.; Hu, N. F.; Rusling, J. F. Langmuir 2004, 20, 722-729. (14) Huang, J. G.; Ichinose, I.; Kunitake, T.; Nakao, A. Langmuir 2002, 18, 90489053. (15) Yu, J. H. Anal. Chem. 2002, 74, 3579-3583. (16) Wang, Q. L.; Lu, G. X.; Yang, B. J. Sens. Actuators, B 2004, 99, 50-57. (17) Lei, C. H.; Shin, Y. S.; Liu, J.; Ackerman, E. J. J. Am. Chem. Soc. 2002, 124, 11242-11243. (18) Lee, C. H.; Lang, J.; Yen, C. W.; Shin, P. C.; Lin, T. S.; Mou, C. Y. J. Phys. Chem. B 2005, 109, 12277-12286. (19) Yu, J. H. Ana. Chim. Acta 2003, 486, 209-216. (20) Lu, H. Y. Electroanalysis 2006, 18, No. 4, 379-390.

Analytical Chemistry, Vol. 79, No. 10, May 15, 2007 3581

Figure 1. Diagram of construction of the {Hb/SiO2}n/PDDA films on the GC electrode surface combining vapor-surface sol-gel SiO2 deposition and protein adsorption.

SiO2 sol-gel, like the sol-gel-derived TiO2, has excellent properties such as biocompatibility, tunable porosity, high thermal stability, and chemical inertness, which is necessary for maintaining the high activities of enzymes and their accessibility to substrates.21 However, the biosensors or electrodes made from this material have displayed limited lifetime because of its fragility, shrinkage, and cracking. In addition, the requirement of a high acidic hydrolysis condition, which is detrimental to the proteins, limited the sensitivity of such biosensors.22 The vapor-surface deposition method might be able to overcome these shortcomings. However, according to the best knowledge of the author, the application of vapor-surface deposition technique to the SiO2 solgel has not been reported. Due to the lower vapor pressure, it is not feasible to apply the vapor-surface deposition technique with tetraethyl orthosilicate (TEOS) as SiO2 source at the room temperature. In the present work, for the first time, the novel vapor-surface deposition technique is employed to develop a SiO2 sol-gel matrix in weak acidic (pH 5.0) medium at 50 °C. The elevated temperature increases the vapor pressure of TEOS and improves its hydrolysis rate, leading to a short hydrolysis time of 20 min. A glassy carbon electrode (GCE) was first coated with a SiO2 solgel matrix and dried. The modified GCE was soaked in a Hb solution (pH 5.0), allowing the adsorption of the protein to form a SiO2/Hb bilayer. Repeating the above procedures, layer-by-layer {SiO2/Hb}n films were fabricated with a controllable thickness. Electrochemical impedance spectroscopy (EIS), cyclic voltammetry (CV), and UV-vis were utilized to monitor the layer-by-layer assembling process. A Soret band appears in the UV-vis spectra of the protein in {SiO2/Hb}n films at the same position as in the spectra of the native protein in solution phase, indicating that the surface-immobilized protein assumed a nearly native conformation even after being heated to 50 °C for several times during the hydrolysis of TEOS. A good catalytic activity to the H2O2 reduction was achieved by the {SiO2/Hb}n films modified electrode with an apparent heterogeneous electron-transfer rate constant (ks) of (21) Li, J. Anal. Chim. Acta 1996, 335, 137-145. (22) Deere, J.; Magner, E.; Wall, J. G.; Hodnett, B. K. J. Phys. Chem. B. 2002, 106, 7340-7347.

3582 Analytical Chemistry, Vol. 79, No. 10, May 15, 2007

1.02 ( 0.03 s-1 and a Michaeli-Menten constant (Kmapp) of 0.155 mM. EXPERIMENTAL SECTION Chemicals and Materials. All chemicals were of analytical grade and were used without further purification. Bovine Hb was purchased from Sigma (St. Louis, MO) and used as received. A 1.0 × 10-4 M Hb solution was prepared in 0.02 M phosphate buffer, pH 5.0 and stored at a temperature of 4 °C. TEOS (98%) was purchased from United Chemical Technologies (Bristol, PA). Poly(diallyldimethylammonium) (PDDA) and hydrogen peroxide (30%) were all obtained from Sigma. The H2O2 solutions with different concentrations were freshly prepared for each experiment. Phosphate buffer solutions with various pH values were prepared by mixed stock standard solutions of Na2HPO4 and KH2PO4; the pH was adjusted with 0.1 M H3PO4 or NaOH. All solutions were made from twice-distilled water. Electrode Fabrication. The glassy carbon electrode was polished with 0.05-µm alumina slurry followed by thorough flushing with twice-distilled water and then ultrasonically in 1:1 nitric acid, 1 M NaOH, acetone, and twice-distilled water. Last, the electrode was dried in air. The {Hb/SiO2}n/PDDA film was deposited onto the clean, dry GCE following a layer-by-layer deposition strategy shown in Figure 1. First, the clean GC electrode was soaked in the aqueous solution of PDDA (0.5%) for 20 min to deposit a positively charged precursor layer of PDDA and then washed and dried under ambient conditions. Second, after an aliquot of 5 µL of 0.2 M potassium phosphate buffer solution (PBS; pH 5.0) was dropped onto the PDDA-coated surface, the GCE was fixed upside down above liquid TEOS in a sealed plastic tubule at 50 °C for 20 min. The SiO2 sol-gel layer was formed slowly through hydrolysis of TEOS. Third, the electrode was washed and immersed into a hemoglobin (1 × 10-4 M) solution of pH 5.0 for 20 min, where the positively charged Hb (pI ) 7.4) was adsorbed to the negatively charged SiO2 surface (pI ∼ 2-3) through electrostatic interactions. The loosely attached Hb molecules were removed from the surface by a thorough rinse with twice-distilled water. The second and third steps can be repeated several times to obtain

the desired number of Hb/SiO2 layers. The electrode was kept at 4 °C in a refrigerator for storage. Instruments and Characterizations. Electrochemical experiments were carried out on a CHI 660A electrochemical workstation (CH Instruments) with a bare GCE (diameter 3 mm, BAS Co.) or a {Hb/SiO2}n/PDDA-modified GCE as the working electrode, a saturated calomel electrode (SCE; Jiangsu Electroanalytical Instruments Factory) as the reference electrode and a platinum electrode as the auxiliary electrode. All electrochemical experiments were performed in PBS, pH 7.0. All solutions were deoxygenated by bubbling with highly pure nitrogen for at least 20 min and maintained under nitrogen atmosphere during the measurements. Scanning electron microscopy (SEM) images for { Hb/SiO2}6/ PDDA films deposited on cleaned glass slides were obtained with a Hitachi S-4700 scanning electron microanalyzer to estimate the thickness. The UV-vis spectra of the Hb samples were measured at the wavelength ranging from 300 to 600 nm at room temperature using Cary 50 Conc UV-vis spectrophotometer. The assembly process of {Hb/SiO2}n/PDDA films was also monitored by UV-vis spectroscopy. RESULTS AND DISCUSSION Spectroscopic Characterization of Hb in Hb/SiO2 Bilayer. The position of the Soret absorption band of iron heme provides information about conformation of heme proteins.23,24 The Soret band shifts or disappears when Hb is denatured.25 Figure 2 illustrates the UV-vis spectra of Hb in solution phase (Figure 2A) and in SiO2 sol-gel matrix (Figure 2B) at different temperatures. A narrow Soret band appears in the UV-vis spectrum (Figure 2a) of the solution-phase Hb (pH 5.0) at 407 nm at room temperature. While the absorbance of this band decreases and the peak becomes broader after the solution temperature of Hb is raised to 50 °C (Figure 2b) for 20 min, this suggests a structural variation in the vicinity of the heme site and that the protein might be denatured at 50 °C. However, an interesting and different phenomenon was observed on the dry Hb/SiO2 film of the glass slides at temperatures higher than 50 °C. Figure 2c-h represents the UV-vis absorbance of SiO2 glass slide (c) and glass slide coated with Hb/SiO2 film at different temperatures from 50 to 90 °C, respectively. The Soret bands of SiO2/Hb are located at 408 nm and are stable for more than 20 min at 50 or 60 °C. No obvious shift was observed compared to that of Hb in PBS (pH 5.0), which indicated that no significant Hb denaturation occurred in the presence of SiO2 sol-gel at temperatures higher than the denaturizing temperature for the native protein. These findings show that SiO2 sol-gel matrix stabilizes the immobilized protein, and the Hb molecule retains its native conformation and thus its biological activity. It has been proposed that the sol-gel matrix limits conformational fluctuations of large amplitude26,27 and the (23) George, P.; Hanania, G. J. Biochem. 1953, 55, 236-243. (24) Theorell, H.; Ehrenberg, A. Acta Chem. Scand. 1951, 5, 823-848. (25) Nassar, A. E. F.; Willis, W. S.; Rusling, J. F. Anal. Chem. 1995, 67, 23862392. (26) Das, T. K.; Khan, I.; Rousseau, D. L.; Friedman, J. M. J. Am. Chem. Soc. 1998, 120, 10238-10269. (27) Hu, S.; Smith, K. M.; Spiro, T. G. J. Am. Chem. Soc. 1996, 118, 1263812646.

Figure 2. (A) UV-vis spectra of Hb in PBS (pH 5.0) at 20 °C (a) and after treatment at 50 °C (b). (B) UV-vis spectra of SiO2/glassy slide (c) and Hb/SiO2/glassy slide after treatment at 50 (d), 60 (e), 70 (f), 80 (g), and 90 °C (h). Table 1. Relationship between Vapor Pressure of TEOS and Temperature temperature/°C vapor pressure/kPa

20 0.13

30 0.25

40 0.46

50 0.84

60 1.45

70 2.42

intermolecular protein-protein interactions.28,29 Consistently, the SiO2 sol-gel matrix prepared in this study shows the potential of dramatically enhancing the thermal stability of the resulting biosensor. Optimization of {Hb/SiO2}n/PDDA Electrode. Hu and Yu20,30 reported a well-behaved TiO2 sol-gel matrix by surfacevapor deposition of titanium isopropoxide at room temperature. The same condition is not applicable to our case because of the much less vapor pressure and hydrolysis activity of tetraethyl orthosilicate than those of titanium isopropoxide. Since the thermal stability of Hb can be greatly enhanced in the {Hb/SiO2}n films and Hb retains its activity at a temperature as high as 70 °C, it is reasonable to raise the preparation temperature above room temperature to reach a good efficiency. There is a simple relationship between the amount of TEOS deposited onto PDDAmodified electrode and the vapor pressure of TEOS. The vapor pressure increases with the increase of vapor temperature and thus the amount of deposited TEOS. At 20 °C, the amount of TEOS deposited on the electrode surface was limited, even when the vapor time was prolonged to 6 h, and was far less than the amount (28) Wang, J.; Liu, Jj.; Cepra, G. Anal. Chem. 1997, 69, 3124-3127. (29) Lioyd, C. R. Langmuir 2000, 16, 9092-9094. (30) Yu, J. H.; Liu, S. Q.; Ju, H. X. Biosens. Bioelectron. 2003, 19, 509-514.

Analytical Chemistry, Vol. 79, No. 10, May 15, 2007

3583

Figure 3. Relationship between the current ip.c of Hb and the immersion time in Hb solution with (a) bare eldectrode and (b) {Hb/ SiO2}6/GCE.

adsorbed at 50 °C for 20 min. Table 1 shows the relationship between vapor pressure of TEOS and temperature. The vapor pressure of TEOS at 50 °C (0.84 kPa) is almost 7 times higher than that at 20 °C (0.13 kPa). Considering the bioactivity of proteins, we choose 50 °C as the optimized temperature for TEOS deposition in this paper. The hydrolysis of TEOS is usually catalyzed by strong acid or strong base, a condition that is generally detrimental to the electroactivity proteins.

Figure 4. EIS in 0.1 mol/L KCl solution containing 5.0 mM Fe(CN)63-/Fe(CN)64- with (a) bare GCE, (b) SiO2/GCE, (c) {Hb/ SiO2}1/GCE, (d) {Hb/SiO2}2/GCE, and (e) {Hb/SiO2}6/GCE.

A neutral condition is optimum for Hb activity, but the hydrolysis rate is limited. Therefore, a compromised condition, pH 5.0, was chosen in this work to reach a high hydrolysis rate and retain the protein’s electroactivity during the vaporization process. Figure 3 shows the effect of Hb immersion time on the reduction peak current ip,c for both the Hb immobilized on the bare electrode (a) and the Hb immobilized on the vapor SiO2 filmmodified electrode (b). We can see that the cathodic peak currents of the Hb on the bare electrode is almost independent of the Hb immersion time. In plot b of the Figure 3, there is a rapid increase of the ip.c when the immersion time of Hb is increased from 5 to 20 min and then almost reaches a plateau from 20 to 60 min. Therefore, 20 min was chosen as the optimal immersion time for Hb adsorption in this work. Previous studies show that stable and consistent layer-by-layer growth of self-assembled films is observed only when the deposition time is long enough. It is speculated that the assembly of polyions on charged surfaces is a two-stage process: macromolecular chains are anchored to the surface by some segments during the short initial stage and then relax to a low energy and stable bound state during the long second stage of self-assembly(>10 min).31 In our case, it was possible that the positively charged Hb macromolecules tended to adsorb on

selected defect sites of negatively charged hydrophobic SiO2 solgel matrix during the first several minutes of deposition; after ∼10 min, the macromolecule underwent a conformational adjustment and realized close contact to the oppositely charged matrix surface. Electrochemical Impedance Spectroscopy of {Hb/SiO2}n/ PDDA film. Electrochemical impedance spectroscopy can provide useful information on the impedance changes of the electrode surface during the fabrication process. For the sake of simplicity, the equivalent electrical circuit (inset in Figure 4) was employed as a mode describing the electrical network of the electrochemical interface, in which RΩ represents the electrolyte resistance between the working electrode and the reference electrode, Cd is the double-layer capacitance, Rct, the charge-transfer resistance, is related the electron-transfer kinetics of the redox probe(Fe(CN)63-/ Fe(CN)64-) at the electrode interface and can be estimated from the diameter of the semicircular part (higher frequency) of the EIS curve, and Rw, the mass transport resistance, can be estimated from the linear part (lower frequency) of the EIS curve.32,33 Figure 4 illustrates typical Nuquist plots obtained from bare GCE (a), SiO2/GCE (b), {Hb/SiO2}1/GCE (c), {Hb/SiO2}2/GCE (d), and {Hb/SiO2}6/GCE (e) in 0.1 M KCl solution containing 5.0 mM Fe(CN)63-/Fe(CN)64-. The Rct of the SiO2/GCE is a little higher than that of the bare electrode (its Rct is near to nil), suggesting that the SiO2 sol-gel slowed down the electron transfer of the redox probe due to the electrostatic repulsion between the negatively charged SiO2 sol-gel and Fe(CN)63-/ Fe(CN)64-. The higher Rct of SiO2/GCE is the direct experimental evidence of the successful deposition of SiO2 sol-gel on the surface of GCE. When Hb is adsorbed onto SiO2/GCE to form {Hb/SiO2}1 film, the positively charged Hb attracts Fe(CN)63-/ Fe(CN)64-. This effect was canceled by the bulky volume of the Hb molecules, which block further the access of electrode surface

(31) Tsukruk, V. V.; Bliznyuk, V. N.; Visser, D.; Campbell, A. L.; Bunning, T. J.; Adams, W. W. Macromolecules 1997, 30, 6615-6625.

(32) Zhou, A. H. J. Colloid Interface Sci. 2000, 229, 12-20. (33) He, X. Y.; Zhu, L. Electrochem. Commun. 2006, 8, 615-620.

H+

(CH3CH2O)4Si + 2H2O 98SiO2 + 4C2H5OH

3584 Analytical Chemistry, Vol. 79, No. 10, May 15, 2007

to the redox pair Fe(CN)63-/Fe(CN)64-. Therefore, an increased Rct is observed for {Hb/SiO2}1/GCE as shown in plot c in Figure 4, indicating the successful adsorption of Hb onto SiO2/GCE. Furthermore, Figure 4c-e shows that the electron-transfer resistance increases as the number of Hb/SiO2 bilayers (n) increases, consistent with larger blocking effect of thicker film.34 Monitoring of {Hb/SiO2}n/PDDA film Growth by UV-vis and CV. The amount of Hb-contained multilayer film was monitored by UV-vis during the electrode fabrication. Figure 5A shows the adsorption spectra of the layer-by-layer {Hb/SiO2}n films with n ) 1, 2, 3, 4, 5, and 6. The Soret bands of these spectra have consistent peak wavelength, 408 nm, and the maximum absorbance of the film increases linearly, indicating that a nearly constant amount of Hb was absorbed in each {Hb/SiO2} bilayer. Cyclic voltammetry was employed to characterize the electrontransfer behavior of Hb during the growth of the {Hb/SiO2}n/ PDDA films. Typical voltammograms were shown in Figure 5B. A pair of well-defined, nearly reversible redox peaks appear in each of the cyclic voltammergrams obtained at the {Hb/SiO2}n/ GCE for different n values. The peak currents of the cyclic voltammograms increased monotonously with the increase of the number of {Hb/SiO2} bilayers. Those voltammograms show that a good electronic communication is established between the Hb molecules and the glassy carbon electrode surface in the presence of SiO2 sol-gel. The reasons are probably because of the porous and three-dimensional structure of vapor SiO2 sol-gel material. This structure may provide a favorable microenvironment for the proteins, favor the orientation of the proteins on the electrode surface, and reduce the distance between the protein’s heme center and the electrode surface. At the scan rate of 100 mV/s of the {Hb/SiO2}6/GCE, the anodic peak potential (Ep,a) and the cathodic peak potential(Ep,c) of Hb are -0.292 (vs SCE) and -0.368 V (vs SCE), respectively, corresponding to a peak-to-peak separation (∆Ep) of 76 mV and a formal potential (E0) of -0.330 V calculated with E0 ) 1/2(Ep,a+ Ep,c). The small peak-to-peak separation indicates a fast electron-transfer rate for {Hb/SiO2}6/ GCE. Figure 5B(a) and (b), the voltammograms for the bare electrode and the SiO2 /GCE, contain only nonfaradic current in the potential window of interest, which indicates that the direct electron transfer is achieved through the entrapment of Hb in the SiO2 membrane. Hb molecules in different {Hb/SiO2} bilayers assume different distances from the surface of the GCE and have different electrontransfer rate constants. Therefore, even though each bilayer contains a constant number of Hb molecules, the bilayer that is closer to the electrode surface contains more electrochemically detectable Hb molecules. This behavior could be verified by cyclic voltammetry. The average surface concentration of electroactive Hb (Γ, mol‚cm-2) can be estimated from the charge integration of the reduction peak of the cyclic voltammogram of {Hb/SiO2}n films, using the formula of Γ ) Q/nFA, where Q is the charge (C), n is the number of transfer electrons, F is the Faraday constant, and A is the geometric area of the GC electrode (cm-2). Figure 5C plots the values of Γ (mol‚cm-2) as a function of the number of the {Hb/SiO2} bilayers. The nonlinear dependence of the amount of detected Hb on the number of n is consistent with the hypothesis that the Hb molecules were immobilized layer by (34) Feng, J. J.; Xu, J. J.; Chen, H. Y. Biosens. Bioelectron. 2006 (in press).

Figure 5. (A) UV-vis spectra on glass slides for {Hb/SiO2}n/PDDA films with different numbers of bilayers; (a-f) n ) 1-6. (B) Cyclic voltammograms of (a) bare GC electrode and GC electrode coated with (b) SiO2/PDDA, (c) {Hb/SiO2}1/PDDA,, (d) {Hb/SiO2}2/PDDA, (e) {Hb/SiO2}3/PDDA, (f) {Hb/SiO2}4/PDDA, and (g) {Hb/SiO2}6/ PDDA in PBS (pH 7.0) (C) Influence of the number of bilayers (n) on average surface concentration of electroactive Hb (Γ, mol‚cm-2).

layer onto the GCE. As shown in Figure 5C, a good communication between Hb molecules and the electrode surface was maintained even when n is as high as 6. Films of six {Hb/SiO2} bilayers, designated as {Hb/SiO2}6, were chosen in most of the electrochemical and electrocatalytic experiments. The thickness of {Hb/SiO2}6/PDDA films was characterized with SEM. One SEM image for {Hb/SiO2}6/PDDA films is shown Analytical Chemistry, Vol. 79, No. 10, May 15, 2007

3585

Figure 6. SEM of {Hb/SiO2}6/PDDA film.

Figure 8. (A) Cyclic voltammograms of {Hb/SiO2}6/GC electrode in 0.2 M PBS with pH values of (a) 4.3, (b) 5.0, (c) 6.0, (d) 8.0, and (e) 9.0. (B) Plot of E0’ vs pH value.

Figure 7. (A) Cyclic voltammograms of {Hb/SiO2}6/PDDA electrode in pH 7.0 PBS at scan rates of (a) 50, (b) 100, (c) 150, (d) 200, (e) 250, (f) 300, (g) 350, (h) 400, (i) 450, and (j) 500 mV/s (a-j, from inner to outer). (B) Plot of peak current vs scan rate.

in Figure 6, and the total thickness of {Hb/SiO2}6/PDDA films is about 120-150 nm. Electron Transfer of {Hb/SiO2}n/PDDA Films. Figure 7 shows the cyclic voltammograms of the {Hb/SiO2}6/PDDA electrode in PBS (pH 7.0) at different scan rates from 50 to 500 mV/s. Each voltammograms are symmetric and each has a pair of well-defined redox peaks. The cathodic and anodic peak currents are linearly proportional to the scan rates in the range 3586 Analytical Chemistry, Vol. 79, No. 10, May 15, 2007

of 50-500 mV/s, a characteristic of a surface-confined electrochemical process.35,36 According to Laviron’s equation,37the electrontransfer rate constant ks was estimated as 1.02 ( 0.03 s-1, comparable to that of Hb immobilized on electrodeposited mesoporous tungsten oxide of 0.97 ( 0.06 s-1,38 two times larger than that of Hb immobilized on a Au colloid-cysteamine-modified gold electrode of 0.49 s-1.39 Therefore, the vaporized SiO2 solgel film facilitates the interfacial electron transfer between Hb heme center and the electrode surface. The pH of the buffer solution affects the electrochemical behavior of Hb in the {Hb/SiO2}6/PDDA film. As shown in Figure 8, the CVs with stable and well-defined peaks were observed from pH 4.3 to 9.0; however, the voltammograms shift toward the (35) Yoon, H. C.; Hong, M. Y.; Kim, H. S. Anal. Chem. 2000, 72, 4420-4427. (36) Dai, Z. H.; Liu, S. Q.; Ju, H. X.; Chen, H. Y. Biosens. Bioelectron. 2004, 19, 861-867. (37) Laviron, E. J. Electroanal. Chem. 1974, 52, 355-393. (38) Feng, J. J.; Xu, J. J.; Chen, H. Y. Electrochem. Commun. 2006, 8, 77-82. (39) Gu, H. Y.; Yu, A. M.; Chen, H. Y. J. Electroanal. Chem. 2001, 516, 119126.

Figure 9. Cyclic voltammograms of (a) bare GC electrode, (c) SiO2/ PDDA electrode, and (e) {Hb/SiO2}6/PDDA electrode in blank solution and (b) bare GC electrode, (d) SiO2/PDDA electrode, and (f) {Hb/ SiO2}6/PDDA electrode in 10 µM H2O2.

negative direction with the increase of pH. Quantitative analysis show that the formal potential (E0) depends linearly on the buffer pH, with a slope of -51.4 mV/pH (the correlative coefficient is 0.996), which is close to the expected value of -58.0 mV/pH for the one-proton-transfer coupled one-electron-transfer reaction, which is written as40,41

HbFe(III) + H+ + e- h HbFe(II) In addition, the voltammograms collected in the PBS of pH 4 before and after immersion in a PBS of pH 9 using the {Hb/ SiO2}6/PDDA electrode were exactly the same. Electrocatalytical Activity of the {Hb/SiO2}6/PDDA Electrode to H2O2. Electrocatalytic reductions of hydrogen peroxide were examined using cyclic voltammetry using the {Hb/SiO2}6/ PDDA electrode. Figure 9 shows the voltammograms of a bare electrode, SiO2/PDDA/GCE, and {Hb/SiO2}6/PDDA/GCE in the absence or in the presence of H2O2. Only nonfaradic signals are observed with a bare electrode and SiO2/PDDA/GCE in blank solution and 1.0 × 10-5 M H2O2 solution (Figure 9a-d). In contrast, when the {Hb/SiO2}6/PDDA electrode is used, there is a pair of well-defined peaks at -0.3 to -0.4 V in the blank solution, which can be attributed to the heme FeIII/FeII redox couple. In addition, the reduction peak current increases while the oxidation peak current decreases in the presence of 1.0 × 10-5 M H2O2 solution in Figure 9e and f, indicating the reduction of H2O2 was catalyzed by the Hb immobilized in the {Hb/SiO2}6 film. Figure 10 shows typical amperometric responses of SiO2/ PDDA/GCE (a) and {Hb/SiO2}6/PDDA/GCE (b) with the successive additions of 10 µM H2O2 at -300 mV. No response is observed at the SiO2/PDDA electrode, but the reduction currents rise step by step at the {Hb/SiO2}6/PDDA/GCE. The responses of H2O2 are linearly proportional to the H2O2 concentration from (40) Nassar, A. E.; Zhang, Z.; Hu, N.; Rusling, J. F.; Kumosinski, T. K. J. Phys. Chem. 1997, 101, 2224-2231. (41) Feng, J. J.; Zhao, G.; Xu, J. J.; Chen, H. Y. Anal. Biochem. 2005, 342, 280286.

Figure 10. (A) Amperometric responses of (a) SiO2/PDDA electrode and (b) {Hb/SiO2}6/PDDA electrode at an operating potential of -300 mV upon successive additions of 10 µM H2O2. (B) Plot of catalytic current of {Hb/SiO2}6/PDDA electrode vs H2O2 concentration.

1.0 × 10-6 to 2.0 × 10-4 M with a detection limit of 5.0 × 10-7 M (S/N ) 3). The apparent Michaelis-Menten constant (Kmapp), an indicator of enzyme-substrate kinetics, can be calculated using to the Lineweaver-Burk equation:

1/Icat ) 1/Imax+ Kmapp/ImaxCH2O2 where Icat is the electrocatalytic current, Imax is the maximum current under saturated substrate conditions, and CH2O2 is the bulk concentration of H2O2.42 In this work, Kmapp is ∼0.155 mM, smaller than that of Hb immobilized on LBL films assembled with SiO2 nanoparticles of 0.219 mM,43 indicating the higher catalytic activity of the {Hb/SiO2}6/PDDA film toward the reduction of H2O2 (the smaller Kmapp shows the higher catalytic ability). Stability and Reproducibility of the {Hb/SiO2}6/PDDA Electrode. The electrocatalytic activity of Hb/SiO2}6/PDDA filmmodified electrode, stored in 0.2 M pH 7.0 PBS, was tested in the 5.0 × 10-5 M H2O2 solution for eight times each day for a (42) Kamin, R. A.; Wilson, G. S. Anal. Chem. 1980, 52, 1198-1205. (43) He, P. L.; Hu, N. F. Electroanal. 2004, 16 (No. 13-14), 1122-1131.

Analytical Chemistry, Vol. 79, No. 10, May 15, 2007

3587

continuous five days; the RSD of the current response is 4.5%. After two weeks, the sensor retains 85% of initial current response to the 5.0 × 10-5 M H2O2. In addition, the modified electrode also showed good reproducibility. The five electrodes, fabricated independently, showed an acceptable reproducibility in the current response to the 5.0 × 10-5 M H2O2 with a RSD of 3.9%. These results show that the vapor-deposition SiO2 sol-gel matrix and the LBL method are very efficient for retaining the electrocatalytic activity of the hemoglobin and preventing it from leaking out of the electrode. CONCLUSION In this paper, vapor-surface sol-gel deposition of silica combined with alternate adsorption of Hb was employed to construct {Hb/SiO2}n films in a weak acidic solution (pH 5.0) at elevated temperature (50 °C). Hb can be immobilized on the SiO2 sol-gel matrix due to the electrostatic interactions between the two parts. The properties of Hb in these {Hb/SiO2}n films were characterized spectroscopically and electrochemically, and it was found that the Hb retained a nearly native secondary structure at

3588

Analytical Chemistry, Vol. 79, No. 10, May 15, 2007

a wide range of temperatures. The apparent heterogeneous electron-transfer rate constant (ks) of Hb was found to be 1.02 ( 0.03 s-1 based on Laviron’s formulism. In addition, the Hb immobilized in the {Hb/SiO2}n films displayed excellent eletrocatalytical activity to the reduction of hydrogen peroxide with a Michaelis-Menten constant (Kmapp) of 0.155 mM. Therefore, the layer-by-layer {Hb/SiO2}n films are promising in the fabrication of the third-generation biosensors based on the direct electrochemistry of enzymes. ACKNOWLEDGMENT This work is supported by the National Natural Science Foundation of China (20675032 and 20475017), the Natural Science Foundation of Shanghai (05ZR14043), and Shanghai Rising-Star program (06QH14004).

Received for review March 17, 2007. AC062034G

October

31,

2006.

Accepted