Anal. Chem. 1996, 68, 2621-2628
o-Phenylenediamine-Modified Carbon Fiber Electrodes for the Detection of Nitric Oxide Marilyn N. Friedemann,‡,⊥ Scott W. Robinson,‡,⊥ and Greg A. Gerhardt*,†,‡,§,⊥
Departments of Psychiatry and Pharmacology, Neuroscience Training Program, and Rocky Mountain Center for Sensor Technology, Box C268-71, University of Colorado Health Sciences Center, 4200 East Ninth Avenue, Denver, Colorado 80262
Nitric oxide (NO•) sensors were prepared using o-phenylenediamine (o-PD) and Nafion to modify the surface of 30 µm diameter carbon fiber electrodes. These electrodes were compared with nickel porphyrin-type NO• sensors that have already been described. High-speed chronoamperometry, amperometry, and differential pulse voltammetry were used to compare the performance of sensors modified with various combinations of Nafion, o-PD, or nickel(II) meso-tetrakis(3-methoxy-4-hydroxyphenyl)porphyrin (Ni-TMPP), in order to determine which electrodes had the most sensitivity and selectivity for NO•. Our findings showed that electrodes treated with Nafion first, followed by o-PD, were very sensitive to NO•, with a detection limit of 35 ( 7 nM. In addition, o-PD electrodes were also very selective against ascorbate (>600:1), dopamine (>300:1), and nitrite (>900:1). Moreover, in the range of 0-6 µM NO•, o-PD electrodes displayed excellent linearity (R2 g 0.997). In contrast, Ni-TMPP electrodes (with Nafion) had significantly poorer detection limits (76 ( 12 nM) and were less selective against dopamine (20 000:1) that these data points were raising the average selectivities much higher than the ratios exhibited by a majority of the electrodes. For example, the typical range for NO:AA was 500-1000:1. Therefore, these outliers were also designated to have a selectivity value of 1000: 1. Sensitivity (Detection Limit). The detection limits (defined as a signal-to-noise ratio of 3) of the electrodes were determined with chronoamperometry and amperometry. The signal-to-noise ratios were calculated by determining the variation in the background current after the electrodes had stabilized 5-10 min in solution. Data collected for 15 s prior to detection of NO• were (28) Gevantman, L. H. In Handbook of Chemistry and Physics, 76th ed.; Lide, D. R., Ed.; CRC Press: Boca Raton, FL, 1995; p 6-3.
Analytical Chemistry, Vol. 68, No. 15, August 1, 1996
2623
averaged, and the standard deviation was calculated. From this, the nanomolar detection limit for NO• was determined. Linearity. Electrodes were calibrated with 3-7 additions of NO• in 0.1 M PBS (pH 7.4). The PBS solution was bubbled with nitrogen prior to all calibrations, in order to minimize the reaction of NO• with oxygen. Approximately 20 s elapsed between addition of NO• and its measurement at a particular concentration. Linear regression calculations were performed on the data. Response Time. Comparisons of o-PD-, Ni-TMPP-, and Nafion-coated and bare carbon fiber electrode response times were accomplished in a flow stream using 10 Hz chronoamperometric recording. Similar to other NO• recordings, the applied potential was +0.9 V, and the resting potential was 0.0 V. A syringe pump (Razel Scientific Instruments, Stamford, CT) was used to infuse PBS at a rate of 1 mL/min through an injection valve (Altex 210A, Berkeley, CA) fitted with a 50 µL loop. The outflow of PBS was through a stainless steel tube (0.02 in. i.d.; 0.062 in. o.d.). The carbon fiber electrode was carefully centered in the tube using a stereotaxic apparatus and a dissection microscope. A standard Ag/AgCl electrode was placed in contact with the outflow stream. Once flow was initiated, a stable baseline recording was attained, and then 100 µL of 20 µM NO• was injected into the tube. The half-maximal (50%) response time was calculated by finding the midpoint between the time at which the sensor first began to detect NO• and the time for the signal to reach maximum height. This point was rounded to the nearest 100 ms. Direct Measurement of NO• Release from Renal Arterioles. Preliminary attempts to measure endogenous NO• release in a biological system were performed using isolated rat renal arterioles. Glomeruli with arterioles attached were dissected and isolated as previously described.29,30 The vessel preparation was placed in a temperature-controlled chamber (1 mL volume) on a microscope stage and was perfused with 37 °C Krebs-Ringer bicarbonate buffer (pH 7.4) consisting of 115 mM NaCl, 25 mM NaHCO3, 2.5 mM K2PO4, 1.2 mM MgSO4, 1.8 mM CaCl2, 5.5 mM glucose, 2.0 mM pyruvic acid, and 1 g/dL dialyzed bovine serum albumin (fraction V; Sigma). Electrodes were positioned ∼50100 µm from the microvessel. Amperometry was used to measure ACh-evoked NO• release. In one experiment, ACh was applied to the preparation at concentrations of 10-5, 10-6, and 10-7, in order to study dose-dependent NO• release. In another experiment, ACh-evoked NO• release was blocked using NG-nitro-L-arginine methyl ester (L-NAME, Sigma), a nitric oxide synthase inhibitor. For this study, L-NAME (5.6 µM) was bath applied over a 10 min period, prior to ACh stimulation, to inhibit the release of NO• elicited by a single dose of 10-5 M ACh. NO• Diffusion in the Rat Brain. Male Sprague-Dawley rats were anesthetized with urethane (1.25 g/kg ip) and prepared for in vivo electrochemical recording as previously described.24,27 A miniature Ag/AgCl reference electrode (200 µm diameter) was implanted into the brain at a site remote from the recording areas. NO• (2 mM in PBS with 2000 units/mL SOD, pH 7.4) was loaded into a single-barrel micropipet with a 10-15 µm o.d. tip diameter. (29) Conger, J. D.; Falk, S. A. Am. J. Physiol. 1993, 264, F134-F140. (30) Conger, J. D.; Falk, S. A.; Robinette, J. B. J. Am. Soc. Nephrol. 1993, 3, 1792-1803. (31) Rice, M. E.; Gerhardt, G. A.; Nagy, G.; Hierl, P. M.; Adams, R. N. Neuroscience 1985, 15, 891-902. (32) Rice, M. E.; Nicholson, C. Anal. Chem. 1989, 61, 1805-1810. (33) Wink, D. A.; Christodoulou, D.; Ho, M.; Krishna, M. C.; Cook, J. A.; Haut, H.; Randolph, J. K.; Sullivan, M.; Coia, G.; Murray, R.; Meyer, T. METHODS: Companion Methods Enzymology 1995, 7, 71-77.
2624
Analytical Chemistry, Vol. 68, No. 15, August 1, 1996
The micropipets were attached to o-PD electrodes with a tip separation of 325-375 µm. The exact distance between the pipet and electrode was carefully measured in order to calculate a diffusion coefficient based on the equation
Dapp ) d2/6T where d is the distance (cm) between the electrode and pipet tip and T is the time (s) for the signal to reach maximal amplitude.31,32 The electrode-pipet assembly was then inserted into discrete areas of the cortex and striatum of the rat brain. Small volumes (
