An Addressable Microelectrode Array for Electrochemical Detection

Jul 30, 2008 - Diana Bonilla , Maria Mallén , Roberto de la Rica , César Fernández-Sánchez , and Antonio Baldi. Analytical Chemistry 2011 83 (5), ...
0 downloads 0 Views 865KB Size
Download PDF
Anal. Chem. 2008, 80, 6830–6833

An Addressable Microelectrode Array for Electrochemical Detection Zhenyu Lin,†,‡ Yasufumi Takahashi,† Yuusuke Kitagawa,† Taizo Umemura,† Hitoshi Shiku,† and Tomokazu Matsue*,† Graduate School of Environmental Studies, Tohoku University, 6-6-11 Aoba, Aramaki, Sendai 980-8579, Japan, and Ministry of Education Key Laboratory of Analysis and Detection Technology for Food Safety, and Department of Chemistry, Fuzhou University, Fuzhou, Fujian 350002, China Column and row electrodes on two different glass substrates were orthogonally arranged in order to assemble an addressable microelectrode device for the purpose of comprehensive electrochemical detection. Amperometric signal at the individual crossing point of the column and row electrodes was detected separately on the basis of redox cycling of localized electroactive species occurring between the electrodes. The addressable microelectrode device was simple and could be easily assembled; however, it comprised as many as 10 × 10 addressable detection points on a single chip. The basic electrochemical performance of the device was investigated by using the ferricyanide/ferrocyanide redox couple. Electrochemical responses at 100 individual points could be collected within 22 s. The present device was successfully used for imaging the spots of alkaline phosphatase on the array substrate. The results indicate that the device can be applied to comprehensive and high-throughput detection and imaging of biochemical species. Recently, the development of array-based biosensors has received considerable attention due to the strong demand for rapid, comprehensive, and high-throughput analyses.1,2 Many array-based biosensors developed so far are based on fluorescence detection since fluorescence measurement usually has a high sensitivity and a variety of tools for performing the measurements are commercially available.3 However, fluorescence detection has some drawbacks such as undesired fluctuations due to quenching or emission from other materials, shielding by a turbid solution or vessels, and need for labeling nonfluorescent species, which may cause toxic side effects in the analytes. Another form of detection utilized in array-based biosensors is amperometric detection, which offers sufficient sensitivity as well as specificity. In the past decade, various arrays of amperometric microelectrodes have been designed by using sophisticated microfabrication * To whom correspondence should be addressed. E-mail: matsue@ bioinfo.che.tohoku.ac.jp. Tel. and Fax: +81-22-795-7209. † Tohoku University. ‡ Fuzhou University. (1) Davis, C. B.; Shamansky, L. M.; Rosenwald, S.; Stuart, J. K.; Kuhr, W. G.; Brazill, S. A. Biosens. Bioelectron. 2003, 18, 1299–1307. (2) Ng, J. K.; Selamat, E. S.; Liu, W. T. Biosens. Bioelectron. 2008, 23, 803– 810. (3) Zoski, C. G.; Simjee, N.; Guenat, O.; Koudelka-Hep, M. Anal. Chem. 2004, 76, 62–72.

6830

Analytical Chemistry, Vol. 80, No. 17, September 1, 2008

technologies and applied to chemical and biological analyses. The electrochemical device with microelectrode arrays has substantial advantages, including rapid response time, steady-state diffusion, and small iR drop. An important feature of an array-based biosensor is that it enables simultaneous analysis of a complex sample to yield qualitative and quantitative data about the substances in the sample.4-9 For developing a reliable and easyto-use analytical procedure involving microelectrode arrays, individually addressable measurements are required, especially for comprehensive screening purposes.10-12 The conventional way to individually address each electrode of an array is to connect the electrode line to a corresponding bond pad. The electrochemical measurements are then carried out on each electrode sequentially.13-17 This method is easy to implement from a technological point of view; however, the number of individually addressable electrodes is limited since sufficient space for bond pads is not available on the chip border. Another way to realize the individual addressability of array electrodes is to use an integrated circuit (IC).18,19 Huge microelectrodes can be integrated in an IC; however, the IC-based microelectrode array is difficult to fabricate and is expensive. (4) Cheung, K. C.; Renaud, P.; Tanila, H.; Djupsund, K. Biosens. Bioelectron. 2007, 22, 1783–1790. (5) Gunning, D. E.; Chichilnisky, E. J.; Litke, A. M.; Shea, V. O.; Smith, K. M.; Mathieson, K. Nucl. Instrum. Methods Phys. Res. A 2007, 576, 215–219. (6) Aguiar, F. A.; Gallant, A. J.; Rosamond, M. C.; Kataky, R. Electrochem. Commun. 2007, 9, 879–885. (7) Brumlik, C. K.; Martin, C. R.; Tokuda, K. Anal. Chem. 1992, 64, 1201– 1203. (8) Menon, V. P.; Martin, C. R. Anal. Chem. 1995, 67, 1920–1928. (9) Bond, A. M.; Oldham, K. B.; Zoski, C. G. Anal. Chim. Acta 1989, 216, 177–230. (10) Grancharov, G.; Khosravi, E.; Wood, D.; Turton, A.; Kataky, R. Analyst 2005, 130, 1351–1357. (11) Dill, K.; Montgomery, D. D.; Ghindilis, A. L.; Schwarzkopf, K. R.; Ragsdale, S. R.; Oleinikov, A. V. Biosens. Bioelectron. 2004, 20, 736–742. (12) Pei, J.; Tercier-Waeber, M.-L.; Buffle, J.; Fiaccabrino, G. C.; Koudelka-Hep, M. Anal. Chem. 2001, 73, 2273–2281. (13) Yu, P.; George, S. W. Faraday Discuss. 2000, 116, 305–317. (14) Aoki, A.; Matsue, T.; Uchida, I. Anal. Chem. 1990, 62, 2206–2210. (15) Connoly, P.; Moores, G. R.; Monaghan, W.; Shen, J.; Britland, S.; Clark, P. Sens. Actuators, B 1992, 6, 113–121. (16) Huang, Y.; Ewalt, K. L.; Tirado, M.; Haigis, R.; Forster, A.; Ackley, D.; Heller, M. J.; O’Connell, J. P.; Krihak, M. Anal. Chem. 2001, 73, 1549–1559. (17) Hayashi, K.; Horiuchi, T.; Kurita, R.; Torimitsu, K.; Niwa, O. Biosens. Bioelectron. 2000, 15, 523–529. (18) Kakerow, R.; Manoli, Y.; Mokwa, W.; Rospert, M.; Meyer, H.; Drewer, H.; Krause, J.; Cammann, K. Sens. Actuators, A 1994, 43, 296–301. (19) Meyer, H.; Drewer, H.; Gruendig, B.; Cammann, K.; Kakerow, R.; Manoli, Y.; Mokwa, W.; Rospert, M. Anal. Chem. 1995, 67, 1164–1170. 10.1021/ac801389d CCC: $40.75  2008 American Chemical Society Published on Web 07/30/2008

Figure 1. (a) Scheme of the working area of the device and (b) addressable microelectrode array detection system. A, Top view; B, side view; 1, double-sided adhesive paper; 2, row electrode; 3, column electrode; 4, glass substrate; 5, gap of the substrate; 6, gap between the two microelectrodes; 7, width of the microelectrode.

Usually, it is difficult to detect very low currents (pA-nA) with an IC-based array since fabrication of a number of high resistances in a chip is difficult. In the present report, we propose a novel method to realize individually addressable electrochemical measurements using a device consisting of an array of microelectrodes. In the device, band microelectrodes are arranged orthogonally to fabricate an n × n array of measurement points with only 2n bonding pads for external connection. The crossing points of the column and row electrodes are deemed as the addressable elements, which can easily address by connecting and setting the potential at the column and row electrodes. Although this device is easy to fabricate, a high density of sensor points can be integrated in one chip. EXPERIMENTAL SECTION Chemicals. Alkaline phosphatase (ALP, 65 units/mg) and 25% glutaraldehyde solution (GA) were produced by Wako Pure Chemical Industries, Ltd. 4-Aminophenyl phosphate (pAPP) was obtained from LKT Laboratories, Inc. All the chemicals were used as received and solutions were prepared using Milli-Q reagent water (Milli-Q, Millipore, 18.2-MΩ resistivity). Microelectrode Array Fabrication. Figure 1a shows the scheme of the addressable microelectrode array. An electrode array with 10 platinum microbands was constructed on a glass substrate by using a photolithographic method. Two glass substrates with the microelectrode array were orthogonally conglutinated face to face by a double-sided adhesive paper (10 µm thick) and formed 100 orthogonally crossing points. Instrumentation. The detection system is shown in Figure 1b. Three channels (W1-W3) of a multichannel potentiostat (HA1010 mM4, Hokuto Denko, Corp., Tokyo, Japan) were used for potential control and current acquisition. The potentiostat was connected to the electrodes by a multiplexer, and the data were controlled and collected by a program developed by Labview

Figure 2. Effect of the gap between the substrates on the amplification of the current. (a) 30 µm; (b) 10 µm. Solution: 4 mM K3[Fe(CN)6] in 0.1 M KCl solution. Column electrode voltage: 0.5 V. Row electrode voltage: 0 V.

through an AD/DA converter (PXI-2529 and PXI-6723, respectively, National Instrument, Austin, TX). An Ag/AgCl and a Pt wire were used as the reference and counter electrodes, respectively. Amperometric Scanning Procedure. All the electrodes were cleaned before the measurements. The device was found to show no detectable cross talk. The scheme of the amperometric measurement procedure is shown in Figure S1 (In Supporting Information). Before the amperometric scan, a voltage (VC) was applied to the all the column electrodes through W1 of the potentiostat for 2 s in order to stabilize the current on the electrodes. Then, the column electrode 1 (C1) was connected to W2 set at VC, and the current data were transferred to a PC. This readout process was sequentially repeated from C1 to C10 with Analytical Chemistry, Vol. 80, No. 17, September 1, 2008

6831

Figure 3. (a) Scheme of the “E” imaging device and (b) the imaging of the current readout on the column electrodes. Solution: 4 mM K3[Fe(CN)6] in 0.1 M KCl solution. Column electrode voltage: 0.5 V. Row electrode voltage: 0 V.

Figure 4. (a) Scheme of the points modified with the enzyme and (b) the imaging of the amplified current at each point. Solution: 4 mM pAPP in pH 9.5 HEPES buffer solution. Column electrode voltage: 0 V, Row electrode voltage: 0.3 V.

an interval of 20 ms for the readout time in each step. All the current responses were readout through W2, while W1 was used only for the potential control. During scanning, the other column electrodes not used for the readout and all the row electrodes were connected to W1 set at VC. Since all the column and row electrodes were set at VC, no redox cycling was expected at the crossing points. The sequential data from C1 to C10 through the above process were used as the background signals. Next, a voltage (VR) was applied to row electrode 1 (R1) through W3 of the potentiostat for 2 s for preconditioning purposes to stabilize the current, while the other row electrodes (R2-R10) were set at VC through W1. Then, the currents at C1-C10 were detected sequentially one by one (20 ms for readout time). The change in the current compared with the background signal at the same column electrode was considered as the current response at the crossing point between the column electrode and R1. The same measurements were sequentially repeated for the other row electrodes (R2-R10) to address and acquire all the responses at every crossing point. 6832

Analytical Chemistry, Vol. 80, No. 17, September 1, 2008

Enzyme Modification. A mixture of 0.2 mg of ALP, 1.0 mg of bovine serum albumin, 15 µL of water, and 2 µL of 5% GA was prepared and modified on the crossing points of the lower array electrodes with a capillary tube. The electrode was then dried at room temperature for 10 min, followed by washing with a 0.05 M phosphate buffer (pH 7.0) to remove the excess GA. The thickness of the modified layer was ∼2 µm. RESULTS AND DISCUSSION Principle of Detection. If the compound showing reversible electrochemical behavior is present in the interspaces between the column and row electrodes that are held at appropriate potentials, the reduction and oxidation of the compound proceeds at the crossing points.20-27 Since the distances between the (20) Matsue, T.; Aoki, A.; Abe, T.; Uchida, I. Chem. Lett. 1989, 133–136. (21) Hermes, T.; Buhner, M.; Bucher, S.; Sundermeier, C.; Dumschat, C.; Borchardt, M.; Cammann, K.; Knoll, M. Sens. Actuators, B 1994, 21, 33– 37. (22) Bard, A. J.; Crayston, J. A.; Kittlesen, G. P.; Shea, T. V.; Wrighton, M. S. Anal. Chem. 1986, 58, 2321–2331.

column and the row electrodes are only 10 or 30 µm and the electrochemical reaction at adjacent electrodes has no influence, as described in the previous section, the amplification of the electrochemical signal occurs only at the crossing point of interest. By sequentially changing the potential applied to the column and row electrodes and the readout column electrode, we collected the electrochemical responses at the entire crossing points, allowing the system to be addressable. This device architecture reduces the number of complicated bonding pads for n2 elements (in an n × n array) to 2n for the multiplexer. Therefore, a highly integrated device with massive electrochemical addressable points can be realized on the basis of the present architecture. Effects of Gap between the Substrates on the Current Amplification. Figure 2 shows the current responses of 4.0 mM [Fe(CN)6]3- observed at the C1 electrode of the device with interspace distances of 30 (a) and 10 µm (b) between the column and the row array electrodes. At the point indicated by the arrow, the potential applied to the row electrodes was changed from 0.5 to 0 V. When the distance was 30 µm (Figure 4a), the amplified signal was small and the current did not reach a steady state in several seconds. When the distance was 10 µm (Figure 4b), significant amplification of the electrochemical response was observed and the current reached a steady state within 2 s. Therefore, in this study, the period for precondition (tp) was set at 2 s (tp ) 2 s). It should be noted that if the interspace distance decreases further, a larger amplification and a shorter response time to reach the steady state would be observed. Electrochemical Imaging. We partly covered the row electrodes with 5-µm-thick SU-8 indicating the character “E” (Figure 3a). Then, a 4.0 mM [Fe(CN)6]3- solution was injected into the device with an interspace distance of 10 µm. The electrochemical responses at 10 × 10 crossing points were sequentially collected according to the amperometric scanning procedure (VC ) 0.5 V, VR ) 0.0 V). The current image at every addressable point shows a clear image E (Figure 3b). At the cross points covered by SU-8, the redox cycling was blocked and the current was not amplified, thereby showing smaller electrochemical responses. The total scan time to obtain this image was 22 s. A large part of this time is mostly consumed by the preconditioning of the row electrodes (23) Niwa, O.; Tabei, H.; Solomon, B. P.; Xie, F. M.; Kissinger, P. T. J. Chromatogr., B 1995, 670, 21–28. (24) Morita, M.; Niwa, O.; Horiuchi, T. Electrochim. Acta 1997, 42, 3177–3183. (25) Tomcik, P.; Bustin, D.; Fresenius, J. Anal. Chem. 2001, 371, 562–564. (26) Svir, I. B.; Oleinick, A. I.; Compton, R. G. J. Electroanal. Chem. 2003, 560, 117–126. (27) Liu, Z.; Niwa, O.; Kurita, R.; Horiuchi, T. Anal. Chem. 2000, 72, 1315– 1321.

(20 s, 2 s for each row electrode). Improving preconditioning will significantly reduce the time required for imaging. Next, we immobilized ALP at several cross points (Figure 4a) and injected 4.0 mM pAPP into the device. Figure 4b shows the current image of 10 × 10 crossing points (VC ) 0.0 V and VR ) 0.3 V). Significant responses are clearly observed at the points modified with the enzyme, while hardly any amplified current is observed at the other crossing points. ALP that was immobilized at the crossing point hydrolyzed pAPP to generate 4-aminophenol (PAP). PAP thus formed was oxidized at 0.3 V to produce p-quinoneimine, which could be reduced at 0 V to produce PAP, thereby exhibiting redox cycling.28 Therefore, the amplification by redox cycling proceeded only at the crossing points modified with ALP. Since ALP is widely used as a labeled enzyme for enzyme-linked immunosorbent assay (ELISA), a combination of the present addressable measurement and ELISA will provide a novel way for the high-throughput screening of biologically important materials CONCLUSION In this study, we propose an amperometric microdevice with addressable microelectrode arrays for comprehensive and highthroughout electrochemical detection. The device is simple but provides electrochemical responses with redox cycling at hundreds of measurement points in a very short time (less than 1 min). The device has a wide variety of potential applications, including the two-dimensional imaging of the distributions of chemical species and small particles. The most important application is the comprehensive detection of biological materials such as proteins, DNA, and living cells for drug screening, genome/ proteome analysis, and food/environmental monitoring. ACKNOWLEDGMENT This study was partly supported by Special Coordination Funds for Promoting Science and Technology, Formation of Innovation Center for Fusion of Advanced Technologies, from Japan Science and Technology Agency. SUPPORTING INFORMATION AVAILABLE Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org. Received for review July 7, 2008. Accepted July 20, 2008. AC801389D (28) Gabig-Ciminska, M.; Holmgren, A.; Andresen, H. Biosens. Bioelectron. 2004, 19, 537–546.

Analytical Chemistry, Vol. 80, No. 17, September 1, 2008

6833