Carbon Nanotube-Modified Carbon Fiber Microelectrodes for In Vivo

Aug 4, 2007 - Analytical Chemistry 2017 89 (1), 300-313 ... Measurements with Easy Precalibration and Interference-Free from Proteins .... Langmuir 0 ...
0 downloads 0 Views 172KB Size
Download PDF
Anal. Chem. 2007, 79, 6559-6565

Carbon Nanotube-Modified Carbon Fiber Microelectrodes for In Vivo Voltammetric Measurement of Ascorbic Acid in Rat Brain Meining Zhang,†,‡ Kun Liu,†,‡ Ling Xiang,†,‡ Yuqing Lin,†,‡ Lei Su,† and Lanqun Mao*,†

Beijing National Laboratory for Molecular Sciences, Institute of Chemistry, the Chinese Academy of Sciences (CAS), Beijing 100080, China and Graduate School of the CAS, Beijing 100039, China

This study demonstrates a new electrochemical method for in vivo measurements of ascorbic acid (AA) in rat brain with multiwalled carbon nanotube (MWNT)-modified carbon fiber microelectrodes (CFMEs) based on the electrochemical property of MWNTs for facilitating AA oxidation. Cyclic voltammetry results indicate that the prepared MWNT-modified CFMEs possess a marked electrocatalytic activity toward AA oxidation and can be used for its selective measurement in the presence of other kinds of electroactive species coexisting in rat brain, such as 3,4dihydroxyphenylacetic acid, uric acid, and 5-hydroxytryptamine. The selectivity of the MWNT-modified CFMEs toward AA measurement is further studied in vivo by exogenously infusing ascorbate oxidase into the brain, and the results confirm that the prepared electrodes are selective and can thus be used for reliable in vivo measurements of AA in rat brain, combined with their good stability during in vivo measurements. The basal level of striatum AA is determined to be 0.20 ( 0.05 mM (n ) 3). The application of the voltammetric method with the MWNT-modified CFMEs is preliminarily demonstrated for in vivo observation of homeostatic regulation of striatum AA with exogenous infusion of AA into the brain. Increasing interest has developed in the measurement of ascorbic acid (AA) in rat brain1 because, although it is not a kind of neurotransmitter directly involved in the neurotransmission process, effort so far has revealed that the chemical properties of AA substantially endow it with multifunctional physiological functions.2 For example, as an electron donor, AA serves as one of most important small-molecular-weight antioxidants and free* Corresponding author: (fax) +86-10-62559373; (Phone) +86-10-62646525; (e-mail) [email protected]. † Institute of Chemistry, CAS. ‡ Graduate School of the CAS. (1) For a review, see: (a) Gru ¨ newald, R. A. Brain. Res. Rev. 1993, 18, 123. (b) Kulagina, N. V.; Shankar, L.; Michael, A. C. Anal. Chem. 1999, 71, 5093. (c) Bossi, A.; Piletsky, S. A.; Piletska, E. V.; Righetti, P. G.; Turner, A. P. F. Anal. Chem. 2000, 72, 4296. (d) Kalakodimi, R. P.; Nookala, M. Anal. Chem. 2002, 74, 5531. (2) For reviews, see: (a) Hediger, M. A. Nat. Med. 2002, 8, 445. (b) Kohen, R.; Nyska, A. Toxicol. Pathol. 2002, 30, 620. (c) Rice, M. E. Trends Neurosci. 2000, 23, 209. (d) Rebec, G. V.; Pierce, R. C. Prog. Neurobiol. 1994, 43, 537. 10.1021/ac0705871 CCC: $37.00 Published on Web 08/04/2007

© 2007 American Chemical Society

radical scavengers and as such is normally neuroprotective.3 Such a property also enables AA to act as an essential cofactor for biosynthesis of, for example, neuropeptides.4 Moreover, AA has been recognized as a neuromodulator for dopamine (DA) and glutamate involved in physiological processes.5 Therefore, a facile method for in vivo monitoring of AA in rat brain is highly desired for understanding its physiological and pathological functions. AA is a water-soluble and hexonic sugar acid with two dissociable protons (pKa 4.04 and 11.34). Under physiological conditions, it occurs as an ascorbate anion and is found throughout the body. In most kinds of animals, such as the rat, biosynthesis of AA from glucose mainly occurs in the liver or kidney with intermediate formation of L-gulonic acid and L-gulonolactone.6 Rather than from this biosynthetic route, in human, nonhuman primates, or guinea pigs AA is available from dietary sources, is absorbed from the gut, and is subsequently distributed to all other tissues through the blood.2c In all animals, including mammal and human, AA enters into the central nervous system (CNS) by active transport at the choroids plexus via blood.7 In CNS, AA first diffuses into brain extracellular fluid (ECF) with a concentration ranging from 200 to 400 µM and then transports from ECF to neurons, in which its concentration reaches several millimolar.2c In ECF, the concentration of AA can be homeostatically regulated and dynamically modulated by, for instance, glutamate-mediated activity via glutamate-ascorbate heteroexchange, as well as be affected by drugs, electrical or behavioral stimulation, and excitation.2 In electrochemistry, AA can be easily oxidized into dehydroascorbic acid through a two-electron and one-proton process followed by an irreversible hydrolysis to finally produce an electroinactive product of 2,3-diketogulanic acid.8 As reported (3) Kontos, H. A. Stroke 2001, 11, 2712. (4) Glembotski, C. C. Ann. N. Y. Acad. Sci. 1987, 498, 54. (5) Glgelashvili, G.; Schousboe, A. Brain Res. Bull. 1998, 45, 233. (6) Nishikimi, M.; Fukuyama, R.; Minoshima, S.; Shimizu, N.; Yagi, K. J. Biol. Chem. 1994, 269, 13685. (7) (a) Spector, R.; Lorenzo, A. V. Am. J. Physiol. 1973, 225, 757. (b) Sotiriou, S.; Gispert, S.; Cheng, J.; Wang, Y.; Chen, A.; Hoogstraten-Miller, S.; Miller, G. F.; Kwon, O.; Levine, M.; Guttentag, S. H.; Nussbaum, R. L. Nat. Med. 2002, 8, 514. (8) (a) Ruiz, J. J.; Aldaz, A.; Dominguez, M. Can. J. Chem. 1977, 55, 2799. (b) Ruiz, J. J.; Aldaz, A.; Dominguez, M. Can. J. Chem. 1978, 56, 1533. (c) Dryhurst, G.; Kadish, K. M.; Scheller, F.; Renneberg, R. Biological Electrochemistry; Academic Press: New York, 1982. (d) Hu, I.; Kuwana, T. Anal. Chem. 1986, 58, 3235. (e) Prieto, F.; Coles, B. A.; Compton, R. G. J. Phys. Chem. B 1998, 102, 7442.

Analytical Chemistry, Vol. 79, No. 17, September 1, 2007 6559

previously,9 this product is readily adsorbed on the electrode surface, resulting in electrode fouling and thereby a high overpotential for AA oxidation. The high-potential oxidation of AA essentially renders difficulties in exploring the electrochemical property of AA to constitute an electrochemical protocol for its selective measurement in cerebral systems because of the great interference from other kinds of electroactive species coexisting in the brain, such as 3,4-dihydroxyphenylacetic acid (DOPAC) and uric acid (UA). To improve the selectivity for in vivo voltammetric measurement of AA in the brain, two strategies have so far been employed in most cases. One is to use carbon fiber microelectrodes with careful electrode pretreatment mainly through electrochemical approaches.10 The other is to use ascorbate oxidase (AAox) to abstract the net current response for AA from the total current response recorded for the electroactive species in the brain including DOPAC, DA, and UA.11 Motivated by development of a selective and reliable electrochemical method for in vivo measurement of AA in rat brain, we have recently been searching for new electrochemical systems capable of facilitating AA oxidation and thereby useful for its selective measurement in rat brain.12 To this end, we have found that the uses of carbon nanotubes (CNTs), one kind of carbonbased nanostructure, can essentially accomplish such a pursuit and the as-established CNT-based selective oxidation of AA has recently been exploited to constitute a new electrochemical protocol for on-line and continuous measurement of AA in rat brain coupled with in vivo microdialysis.13 By taking advantage of the excellent electrochemical properties of CNTs reported previously,14 we demonstrate here a new electrochemical method for in vivo measurement of AA in rat brain with multiwalled carbon nanotube (MWNT)-modified carbon fiber microelectrodes. Compared with the existing voltammetric methods reported for in vivo measurements of AA,10,11,15 the method demonstrated here (9) (a) Raj, C. R.; Tokuda, K.; Ohsaka, T. Bioelectrochemistry 2001, 53, 183. (b) Raj, C. R.; Ohsaka, T. J. Electroanal. Chem. 2001, 496, 44. (c) Raj, C. R.; Okajima, T.; Ohsaka, T. J. Electroanal. Chem. 2003, 543, 127. (10) (a) Wightman, R. M.; Kennedy, R. T. In Microelectrode: Theory and Application; Montenegro, M. I., Queiro´s, M. A., Daschbach, J. L., Eds.; Kluwer Academic Publishers: Dordrecht, The Netherlands, 1991; Chapter 7. (b) Adams, R. N. Anal. Chem. 1976, 48, 1128A; and references cited therein. (c) Stamford, J. A.; Justice, J. B., Jr. Anal. Chem. 1996, 68, 359A; and references cited therein. (d) Zhang, X.; Zhang, W.; Zhou, X.; Ogorevc, B. Anal. Chem. 1996, 68, 3338. (e) Feng, X.; Brazell, M.; Renner, K.; Kasser, R.; Adams, R. N. Anal. Chem. 1987, 59, 1863. (11) (a) Schenk, J. O.; Miller, E.; Adams, R. N. Anal. Chem. 1982, 54, 1452. (b) Ghasemzedah, B.; Cammack, J.; Adams, R. N. Brain Res. 1991, 547, 162. (c) Brazell, M. P.; Marsden, C. A. Brain Res. 1982, 249, 167. (12) (a) Zhang, M.; Gong, K.; Zhang, H.; Mao, L. Biosens. Bioelectron. 2005, 20, 1270. (b) Gong, K.; Zhang, M.; Yan, Y.; Su, L.; Mao, L.; Xiong, S.; Chen, Y. Anal. Chem. 2004, 76, 6500. (c) Yu, P.; Lin, Y.; Xiang, L.; Su, L.; Zhang, J.; Mao, L. Langmuir 2005, 21, 9000. (13) Zhang, M.; Liu, K.; Gong, K.; Su, L.; Chen, Y.; Mao, L. Anal. Chem. 2005, 77, 6234. (14) (a) Gong, K.; Yan, Y.; Zhang, M.; Xiong, S.; Mao, L. Anal. Sci. 2005, 21, 1383. (b) Gooding, J. J. Electrochim. Acta 2005, 50, 3049. (c) Zhao, Q.; Gan, Z.; Zhuang, Q. Electroanalysis 2002, 14, 1609. (d) Wang, J. Electroanalysis 2005, 17, 7. (e) Dai, L.; Soundarrajan, P.; Kim, T. Pure Appl. Chem. 2002, 74, 1753. (f) Chen, R.; Huang, W.; Tong, H.; Wang, Z.; Cheng, J. Anal. Chem. 2003, 75, 6341. (g) Su, L.; Gao, F.; Mao, L. Anal. Chem. 2006, 78, 2651. (h) Yan, Y.; Zheng, W.; Su, L.; Mao, L. Adv. Mater. 2006, 18, 2639. (15) (a) Gonon, F.; Buda, M.; Cespuglio, R.; Jouvet, M.; Pujol, J. F. Brain Res. 1981, 223, 69. (b) Stamford, J. A.; Kruk, Z. L.; Millar, J. Brain Res. 1984, 299, 289.

6560

Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

requires less technical demand and could find some interesting applications in physiological and pathological investigations. EXPERIMENTAL SECTION Reagents and Solutions. DA, sodium ascorbate, UA, DOPAC, 5-hydroxytryptamine (5-HT), and AAox (Cucurbita species, EC 1.10.3.3) were all purchased from Sigma and used as supplied. A stock solution of AA (1.0 mM) was prepared just before use. Artificial cerebrospinal fluid (aCSF) was prepared by mixing NaCl (126 mM), KCl (2.4 mM), KH2PO4 (0.5 mM), MgCl2 (0.85 mM), NaHCO3 (27.5 mM), Na2SO4 (0.5 mM), and CaCl2 (1.1 mM) into doubly distilled water, and the solution pH was adjusted pH 7.4. MWNTs (10-30 nm in diameter and 0.5-50 µm in length) were purchased from Shenzhen Nanotech Port Co., Ltd. (Shenzhen, China). Prior to use, MWNTs were purified by refluxing the asreceived MWNTs in 2.6 M nitric acid for 5 h followed by centrifugation, resuspension, filtration, and air-drying to evaporate the solvent. Other chemicals were of at least analytical reagent grade and used without further purification. Aqueous solutions were prepared with doubly distilled water. Preparation and Modification of Carbon Fiber Microelectrodes (CFMEs). CFMEs were fabricated as reported previously.16 Briefly, a glass capillary (i.d. 1.5 mm, length 100 mm) was pulled on a microelectrode puller (WD-1, Chengdu Instrument Factory, Sichuan, China) into two capillaries; the fine tip of each was broken into 30-50 µm in diameter. The pulled capillary was used as the sheath of CFMEs. A single carbon fiber (7 µm in diameter, Tokai Carbon Co., Tokai, Japan) was attached to a copper wire with silver conducting paste. Then, the carbon fiberattached copper wire was carefully inserted into the capillary with carbon fiber exposed to the fine open end of the capillary and Cu wire exposed to the other end of the capillary. Both open ends of the capillary were sealed with epoxy resin with 1:1 ethylendiamine as the harder and the excess epoxy on the fiber was carefully removed with acetone. After that, the CFMEs were dried at 100 °C for 2 h. The exposed carbon fiber was cut to 5001000 µm in length under a microscopy. Prior to modification with MWNTs, the fabricated CFMEs were first sequentially sonicated in acetone, 3 M HNO3, 1.0 M KOH, and distilled water each for 3 min. Then, the electrodes were subjected to electrochemical activation, first with potential-controlled amperometry at +2.0 V for 30 s, at -1.0 V for 10 s, and then with cyclic voltammetry in 0.5 M H2SO4 within a potential range from 0 to 1.0 V at a scan rate of 0.1 V s-1 until a stable cyclic voltammogram was obtained. To prepare the MWNT-modified CFMEs, 2 mg mL-1 MWNTs was dispersed into N,N-dimethylformamide, and the mixture was sonicated to give a homogeneous dispersion. One drop of the dispersion was applied onto a smooth glassy plate, and the CFMEs were modified with MWNTs by carefully immersing and rolling the electrodes into the droplet for ∼1 min under a microscope. Close attention should be paid not to break the carbon fibers during this process. The electrodes were taken out from the droplet, air-dried, and thoroughly rinsed with distilled water before use. Apparatus and Measurements. Electrochemical measurements were performed with a computer-controlled electrochemical (16) (a) Mao, L.; Jin, J.; Song, L.; Yamamoto, K.; Jin, L. Electroanalysis 1999, 11, 499. (b) Tian, Y.; Mao, L.; Okajima, T.; Ohsaka, T. Biosens. Bioelectron. 2005, 21, 557.

Scheme 1. Schematic Illustration of Experimental Setup for in Vivo Voltammetric Measurement of Striatum AAa

a With the MWNT-modified electrode and for exogenous infusion of standard solutions of AA and AAox into rat striatum through a silicon capillary and FEP tubing pumped with a microinjection pump.

analyzer (CHI660B, CHI Instrument). The MWNT-modified CFMEs were used as the working electrode and a platinum wire as the counter electrode. For both in vitro and in vivo electrochemical measurements, an implantable microsized Ag/AgCl electrode was used as reference electrode. The electrode was prepared by first polarizing Ag wire (diameter, 1 mm) at +0.6 V in 0.1 M hydrochloride acid for ∼30 min to produce an Ag/AgCl wire and then inserting the as-prepared Ag/AgCl wire into a pulled glass capillary, in which aCSF was aspirated from the fine end of the capillary and used as the inner solution for the reference electrode. The other end of the capillary with Ag wire exposed was sealed with epoxy. Scanning electron microscopy used for characterization of MWNT modification onto the CFMEs was performed on an Hitachi S4300-F microscope (Hitachi Inc., Tokyo, Japan). In Vivo Experiments. Adult male Sprague-Dawley rats (350-400 g) were purchased from Center for Health Science, Peking University. The animals were housed on a 12:12 h lightdark schedule with food and water ad libitum. Animal experiments were performed as reported previously.13 Briefly, the animals were anaesthetized with chloral hydrate (345 mg/kg, ip) and positioned onto a stereotaxic frame. The MWNT-modified CFME was implanted into striatum (AP ) 0 mm, L ) 3 mm from bregma, V ) 4 mm from dura) using standard stereotaxic procedures.17 The prepared microsized Ag/AgCl reference electrode was positioned into the dura of brain and secured with tooth cement. Stainless steel wire embedded in subcutaneous tissue on the brain was used as the counter electrode. Differential pulse voltammetry (DPV) was employed for in vivo voltammetric measurements of AA in rat brain. In order to examine the selectivity of the prepared MWNT-modified CFME for in vivo voltammetric measurements of AA and the possible application of the method for in vivo observation of homeostatic regulation of striatum AA in the brain, (17) Swanson, L. W. Brain Maps: Structure of the Rat Brain. A Laboratory Guide with Printed and Electronic Templates for Data Models and Schematics; Elsevier: Amsterdam, 1998.

a fine silicon capillary (i.d. 75 µm) was in parallel combined with the MWNT-modified CFME under a microscope with the outlet of the capillary ∼400 µm higher than the tip of microelectrode and a 50-100-µm spacing between the capillary and the microelectrode (Scheme 1). The capillary was implanted into the striatum together with the MWNT-modified CFME and was used to exogenously infuse the standard solutions of ascorbate oxidase and AA into the brain. These solutions were delivered from gasimpermeable syringes and pumped through tetrafluoroethylene hexafluoropropene (FEP) tubing by a microinjection pump (CMA 100, CMA Microdialysis AB, Stockholm, Sweden), as schematically shown in Scheme 1. RESULTS AND DISCUSSION Electrochemical Properties of MWNT-Modified CFMEs. Figure 1 depicts typical cyclic voltammograms (CVs) obtained at bare and the MWNT-modified CFMEs in aCSF containing 5 mM Fe(CN)63-. Well-defined, sigmoid-shaped voltammograms were achieved on both kinds of electrodes, demonstrating that a nonlinear diffusion process was involved in the electrochemical process on the electrodes and thereby revealing the microsized property of the prepared MWNT-modified CFMEs. The small charging current of the fabricated CFME essentially indicates a good insulation and sealing of the CFMEs, while the larger charging current of the MWNT-modified CFME, relative to that of the CFME, confirms the modification of the MWNTs onto the CFMEs since the modification of MWNTs onto the CFME increases the electrode area and thus increases the interfacial capacitance of the prepared MWNT-modified CFME. The MWNTs confined onto CFMEs are found to be relatively stable, which could be evident from the almost unchanged steady-state current recorded for the redox process of Fe(CN)63-/4- couple at the MWNT-modified electrode after the electrode was consecutively scanned in aCSF containing 5 mM K3Fe(CN)6 for at least 50 cycles (data not shown). Moreover, the MWNTs could still be clearly Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

6561

Figure 1. Typical CVs at bare (dotted line) and MWNT-modified (solid line) CFMEs in aCSF containing 5 mM Fe(CN)63-. Scan rate, 10 mV s-1.

Figure 2. CVs at the MWNT-modified (A) and (B) bare CFMEs in aCSF (pH 7.4) in the absence (dotted lines) and presence (solid lines) of 0.50 mM AA. Scan rate, 10 mV s-1.

seen on the CFMEs under a microscope after in vivo measurements. Figure 2 compares AA oxidation at MWNT-modified (A) and bare (B) CFMEs in aCSF. At bare CFME, AA is oxidized with an ill-defined voltammetric response, indicating a slow electrontransfer process of AA at the CFME. Such a sluggish electrontransfer kinetics is possibly due to the inactivated surface conditions of the CFMEs under the present conditions or electrode fouling caused by the deposition of an oxidation product of AA.9 In contrast, at the MWNT-modified CFME, AA oxidation reaches a well-defined steady state at 0.0 V with a half-peak potential of ∼ -0.06 V. The tailed response that commences at -0.15 V at the MWNT-modified CFME was ascribed to the reduction of the dissolved O2. Such a comparison of AA oxidation at bare and MWNT-modified CFMEs essentially reveals that the electron-transfer kinetics of AA oxidation is largely enhanced at the latter electrode, which is consistent with our early works.12a,b,13 6562 Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

Figure 3. CVs at the MWNT-modified CFME in aCSF in the absence (dotted lines) and presence (solid lines) of 5-HT (A), UA (B), DA (C), and DOPAC (D). Concentration of each species was 0.50 mM. Scan rate, 10 mV s-1.

The observed kinetic enhancement is possibly due to the surface property of the CNTs14,18 and to the electrochemical property of AA19 and substantially makes it possible to differentiate AA oxidation from the electrochemical processes of other electroactive species coexisting in the cerebral systems. This property essentially forms a strong basis for selective measurement of AA with the MWNT-modified CFMEs, as displayed in Figure 3. At the MWNT-modified CFME, the half-peak potentials for 5-HT (A), UA (B), DA (C), and DOPAC (D) and are 0.26, 0.28, 0.22, and 0.19 V, respectively. These potentials are more positive than and well separated from that for AA oxidation at the same electrode, revealing the possibility of in vivo selective measurement of AA in rat brain with the prepared MWNT-modified CFMEs, as demonstrated below. Selectivity, Linearity, Stability, and Reproducibility. As displayed in Figure 3, among all species examined as the possible interferents for AA measurement, DOPAC remains to have the most potential in terms of its higher concentration in rat brain and its lower oxidation potential, as compared with those of DA and 5-HT. We thus use such species as an example to the in vitro investigation of the selectivity of the MWNT-modified CFMEs toward AA measurement by using DPV, as shown in Figure 4 A. With the DPV method, the current for AA oxidation is readily measurable even though the followed oxidation of DOPAC is partially overlapped with that of AA. This property essentially makes it possible to selectively determine AA virtually interference-free from DOPAC and other electroactive species coexisting (18) (a) Wang, J.; Musameh, M.; Lin, Y. J. Am. Chem. Soc. 2003, 125, 2408. (b) Wang, J.; Musameh, M. Anal. Chem. 2003, 75, 2075. (c) Gong, K.; Dong, Y.; Xiong, S.; Chen, Y.; Mao, L. Biosens. Bioelectron. 2004, 20, 253. (d) Moore, R. R.; Banks, C. E.; Compton, R. G. Anal. Chem. 2004, 76, 2677. (e) Chou, A.; Bo ¨cking, T.; Singh, N. K.; Gooding, J. J. Chem. Commun. 2005, 42. (19) (a) DeClements, R.; Swain, G. M.; Dallas, T.; Holtz, M. W.; Herrick, R. D., II; Stickney, J. L. Langmuir 1996, 12, 6578. (b) Pontikos, N. M.; McCreery, R. L. J. Electroanal. Chem. 1992, 324, 229. (c) Hu, I. F.; Karweik, D. H.; Kuwana, T. J. Electroanal. Chem. 1985, 188, 59. (d) Evans, J. F.; Kuwana, T. Anal. Chem. 1977, 49, 1632. (e) McCreery, R. L. In Electroananlytical Chemistry; Bard, A. J., Ed.; Dekker: New York, 1991.

Figure 5. Typical DPVs at the MWNT-modified CFME in aCSF containing AA with different concentrations of 0.0, 0.1, 0.25, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, and 1.0 mM.

Figure 4. DPVs at the MWNT-modified CFMEs in aCSF (A) containing AA (0.3 mM) (curve 1) and AA (0.3 mM) + DOPAC (20 µM) (curve 2) and in the striatum (B) before (curve 1) and after (curve 2) exogenous infusion of AAox (39.3 units mL-1) into the striatum for 8 min at a rate of 3 µL min-1. DPV conditions, 4 mV step height, 50 mV amplitude, 0.05 s pulse width, 0.2 s pulse period, and 2 s quite time.

in the brain, as further confirmed in vivo with exogenous infusion of AAox to the vicinity of the electrode with the method shown in Scheme 1. Figure 4B depicts typical DPVs obtained at the MWNT-modified CFME in the striatum before (curve 1) and after (curve 2) exogenous infusion of AAox into the striatum. AAox is known as an enzyme that specifically catalyzes AA oxidation, and its infusion into rat brain will thus lead to AA depletion. As expected, the current response recorded for AA (Figure 4B, curve 1) almost completely disappeared with continuous exogenous infusion of 39.3 units mL-1 AAox into the striatum for 8 min (Figure 4B, curve 2). The AAox-induced disappearance of the DPV response at the MWNT-modified CFMEs further suggests that the as-developed electrochemical method with the MWNTmodified CFMEs could be used for in vivo selective measurements of AA in rat brain. In addition to its high selectivity, the MWNTmodified CFMEs show a good linearity for AA measurement, as depicted in Figure 5. The currents recorded clearly increase with increasing AA concentration in solution and are linear with AA concentration within a range from 0.1 to 1.0 mM (I/µA ) 0.48C /mM + 0.33, γ ) 0.9987). The detection limit, based on a signalto-noise ratio of 3, is calculated to be 0.04 mM. We have further studied in vivo the sensitivity and stability of the prepared MWNT-modified CFMEs and found that the sensitivity of the MWNT-modified CFME is decreased to 66% of the initial value after its implantation into rat brain for 10 min. Afterward, the current response becomes stable, allowing reliable measurements of AA in rat brain. This time period is remarkably shorter than those routinely required for other kinds of microelectrodes during in vivo measurements of AA, at which the electrode fouling caused by the slow nonspecific adsorption of

Figure 6. DPV responses consecutively recorded in vivo every 2 min at the MWNT-modified CFME in the striatum of the anesthetized rats after the electrode was implanted into rat brain for 10 min.

biomacromolecules, such as proteins,10a-c essentially results in a longer time to stabilize the current response recorded in vivo.15a This problem could be reasonably expected to become more serious for AA measurement since, in addition to the adsorption of biomacromolecules, the product of AA oxidation has been reported to be able to adsorb onto electrode surface,9 further resulting in decrease in electrode sensitivity and a long time for stabilizing the current response recorded in vivo. The relatively short equilibration time of the MWNT-modified CFME is likely attributed to the marked ability of MWNTs against electrode fouling, as previously observed by Wang et al. 18a,b,20 and our group,13 or possibly to the rapid protein adsorption onto MWNTs, as compared with that on other kinds of carbon materials such as carbon fibers. Figure 6 displays typical DPVs recorded every 2 min with the MWNT-modified CFME in the striatum of the anesthetized rats after the electrode was implanted into the brain for 10 min. The MWNT-modified CFME shows good stability and reproducibility for the measurements of endogenous AA, and the relative standard deviation is calculated to be 5.3% (n ) 6). According to the postcalibration, the basal level of striatum AA of the anesthetized rats under normal conditions is estimated to be 0.20 ( 0.05 mM (n ) 3), which is almost in agreement with the reported values.2 (20) Musameh, M.; Wang, J.; Merkoci, A.; Lin, Y. Electrochem. Commun. 2002, 4, 743.

Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

6563

Although some methods have been reported for in vivo measurements of AA,10,11 the method demonstrated here with the MWNT-modified CFMEs remains remarkable in terms of its facility and, as a result, the ready adoptability by the nonelectrochemists. As well studied previously,19e at most kinds of carbonbased electrodes, AA is an inner-sphere redox species with surface-sensitive, not oxides-sensitive, electrochemical properties. This feature essentially constitutes substantial consequences for the facilitation of AA oxidation at most kinds of carbon-based electrodes including CFMEs after electrode preactivation with, for instance, electrochemical activation in acidic media.10 Although the preactivation procedure, in particular with electrochemical activation, essentially endows the as-formed electrodes with a high selectivity toward AA oxidation and thereby enables them to be capable of selective measurements of AA in rat brain, it remains to be very essential to optimize the conditions for electrode preactivation, because such conditions were reported to be closely linked to the electrochemical properties of the as-formed electrodes10e,21 and to largely vary with the source of the CFMEs.21 Such a source-sensitive feature essentially suggests that the conditions used for electrode preactivation should be carefully optimized through a time-consuming trial and error. On the other hand, although the direct usage of CFMEs, without such critical electrode preactivation procedures, could largely simplify the experimental protocols for in vivo measurement of AA in rat brain and thus be readily adopted by the physiologists and pathologists, this method unfortunately suffers from a selectivity problem and often requires in vivo exogenous infusion of AAox to sort out the electroactive species coexisting in the brain and being co-oxidized with AA after the measurements.11 Compared with those existing CFMEs-based electrochemical methods that necessitated either careful electrode preactivation or exogenous infusion of AAox, the method demonstrated here for in vivo measurements of AA requires less technical demand, obviating the critical electrode pretreatment since the CFMEs used here simply serve as the conducting substrate for MWNTs, and the heterogeneous electron transfer for AA oxidation actually occurs at the confined MWNTs. Moreover, the electrode-to-electrode variation was found to be negligible provided the prepared CFMEs were carefully trimmed almost into the same length. These advantages, along with the high selectivity and good reproducibility of the MWNT-modified CFMEs as demonstrated above, substantially enable the asdeveloped voltammetric method to be relatively useful for physiological and pathological investigations of AA, as demonstrated below, with in vivo observation of homeostatic regulation of striatum AA as an example. In Vivo Observation of Homeostatic Regulation of Striatum AA. As mentioned above, the concentration of AA is sitedependent in the CNS; average AA concentration in neurons is several millimolar, but only 1 mM in glia and 0.2-0.4 mM in ECF.2 Such a physiological phenomenon is indicative of the compartmentalization of AA, which essentially makes the intracellular store of AA act as a reservoir to compensate the changes in extracellular AA, resulting in the homeostatic regulation of the extracellular (21) (a) Wang, J.; Peng, T.; Villa, V. J. Electroanal. Chem. 1987, 234, 119. (b) Runnel, P. L.; Joseph, J. D.; Logman, M. J.; Wightman, R. M. Anal. Chem. 1999, 71, 2782. (c) O’Shea, T. J.; Garcia, A. C.; Blano, P. T. J. Electroanal. Chem. 1991, 307, 63.

6564 Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

Figure 7. DPVs recorded at the MWNT-modified CFME in the striatum of the anesthetized rats after continuous infusion of 1.0 mM AA for 6 min with 3 µL min-1 (curve 1) and 10 min after stopping AA infusion (curve 2). Curve 3 represents DPV recorded with the MWNTmodified CFME in the striatum under normal physiological conditions.

concentration of AA.2 Meile and Fillenz previously observed the homeostatic regulation of extracellular AA by using a potentialcontrolled amperometric method with a carbon paste electrode implanted into the rat striatum through exogenously infusing different concentrations of AA into the brain.22 By using that method, they found that current response increased rapidly with AA infusion and retuned to basal level when AA infusion was stopped. Such a homeostatic phenomenon was also observed in slices when the slice was put into the media containing different concentrations of AA.23 Figure 7 displays the DPV response recorded with the MWNTmodified CFME in the rat striatum after 1.0 mM AA was exogenously infused into the vicinity of the implanted electrode through the capillary at a flow rate of 3 µL min-1, as shown in Scheme 1. We can see that the current increased by 100% after AA was consecutively infused into the rat brain for 6 min (curve 1), as compared with the physiological level (curve 3). After AA infusion was stopped for 10 min, the DPV response was almost restored to the initial value, as shown in curve 2, indicating that AA concentration was gradually recovered to the basal level. The observed changes in the striatum AA recorded with the MWNTmodified CFMEs induced by AA infusion is consistent with the previous reports and might be due to the homeostatic regulation of AA, as demonstrated previously.22,23 This demonstration substantially suggests that the developed electrochemical method with the MWNT-modified CFMEs could be used for in vivo reliable measurements of AA in rat brain and could thus be potentially useful for physiological and pathological investigations. CONCLUSIONS By taking advantage of the excellent electrochemical properties of carbon nanotubes for facilitating the oxidation of AA, we have successfully developed a new electrochemical method for in vivo voltammetric measurement of AA with MWNT-modified carbon fiber microelectrodes. In vitro and in vivo experiments demonstrated that the electrodes possess a high selectivity and a good reproducibility and could thus be used for selective and reliable measurement of AA in rat brain. This study essentially offers a (22) Miele, M.; Fillenz, M. J. Neurosci. Methods 1996, 70, 15. (23) Rice, M. E.; Perezpinzon, M. A.; Lee, E. J. K. J. Neurophysiol. 1994, 71, 1591.

new and facile method for in vivo measurement of AA in rat brain that could find some interesting applications in physiological and pathological investigations. ACKNOWLEDGMENT The work is financially supported by NSF of China (Grant 20575071 and 20435030), Chinese Academy of Sciences, and

Center for Molecular Science, Institute of Chemistry. L.M. thanks NSF of China for Distinguished Young Scholars (Grant 20625515). Received for review March 24, 2007. Accepted June 21, 2007. AC0705871

Analytical Chemistry, Vol. 79, No. 17, September 1, 2007

6565