Measuring Nitric Oxide in Single Neurons by Capillary Electrophoresis

Home · Browse the Journal ... As a family of novel fluorescent indicators for nitric oxide (NO), the diaminofluoresceins .... Alicia Beltrán , M. Isa...
0 downloads 0 Views 137KB Size
Download PDF
Anal. Chem. 2006, 78, 1859-1865

Measuring Nitric Oxide in Single Neurons by Capillary Electrophoresis with Laser-Induced Fluorescence: Use of Ascorbate Oxidase in Diaminofluorescein Measurements Won-Suk Kim,† Xiaoying Ye, Stanislav S. Rubakhin, and Jonathan V. Sweedler*

Department of Chemistry and the Beckman Institute, University of Illinois, Urbana, Illinois 61801

As a family of novel fluorescent indicators for nitric oxide (NO), the diaminofluoresceins (DAFs) have allowed realtime measurement of neuronal NO, an important gaseous neurotransmitter. However, the measurement of NO by the most commonly used NO sensor, 4,5-diaminofluorescein (DAF-2), is altered by two processes: the interaction of DAF-2 with intracellular dehydroascorbic acid (DHA) and the impact of ascorbic acid (AA) on the levels of N2O3, the intermediate product of the oxidation of NO that reacts with DAF-2. Similar AA/DHA effects are observed with other DAF probes, including DAF-FM and DAR-4M. To overcome these limitations, we use a specific enzymatic reaction to eliminate the confounding effect of AA on DAF quantitation of NO and then use capillary electrophoresis (CE) with laser-induced fluorescence (LIF) detection to distinguish the various reaction products. First, the enzyme ascorbate oxidase (AO) is used to catalyze the oxidation of AA to DHA. Next, CE-LIF separates the fluorescent products of the reaction of DAF-2 with NO and DHA. Control experiments, including standard mixtures and single neurons with added NO donor, successfully demonstrate the utility of this approach. This protocol is further tested with homogenates of the mouth area from the sea slug Aplysia californica, previously shown to be NO-positive, and individual nitric oxide synthase-containing buccal neurons from the freshwater snail, Lymnaea stagnalis. In each case, significant amounts of NO are detected. This AO DAF methodology is specific, effective, simple, and allows NO to be measured in single cells without detectable interference from other compounds. Nitric oxide (NO), an important messenger molecule in the cardiovascular, immune, and nervous systems,1,2 is involved in the regulation of diverse physiological and pathological mecha* Corresponding author. Address: Department of Chemistry, University of Illinois, 600 South Mathews Ave. 63-5, Urbana, IL 61801. Voice: 217-244-7359. Fax: 217-244-8068. E-mail:[email protected]. † Current Address: Analytical Science Center, Corporate R&D, LG Chem., Ltd./Research Park, Daejeon 305-380, Korea. (1) Schmidt, H. H.; Walter, U. Cell 1994, 78, 919-925. (2) Kerwin, J. F. J.; Lancaster, J. R. J.; Feldman, P. L. J. Med. Chem. 1995, 38, 4343-4362. 10.1021/ac051877p CCC: $33.50 Published on Web 01/10/2006

© 2006 American Chemical Society

nisms in biological systems.3-5 To elucidate the functional role of NO in both normal and pathologically modified biological systems, precise detection of in vivo and in situ NO is required. However, examination of total NO release from groups of cells can be misleading. For example, a decrease in NO generation by one subpopulation of cells may be masked by an increase of NO production by another subpopulation, and often, these subpopulations can be functionally distinct. Individual cells, upon exposure to a number of intercellular and extracellular signals, have the ability to produce NO in a broad diapason of concentrations under different physiological conditions. The biological importance of this molecule proffers the challenge to develop reliable NO detection methods at the single-cell level. In 1998, Nagano et al.6 developed a series of novel fluorescent indicators, diaminofluoresceins (DAFs), to measure NO generated under physiological conditions. Since then, DAFs have offered researchers the combined advantage of specificity, noncytotoxicity, sensitivity, and simplicity of use for direct detection of NO. Subsequent articles describing the utilization of DAFs for bioimaging NO have appeared in the literature.7-11 However, the specificity of 4,5-diaminofluorescein (DAF-2), a commonly used form of DAF, has been questioned.12-14 We recently reported that DAF-2 rapidly reacts with dehydroascorbic acid (DHA) under physiologically relevant conditions to generate compounds (DAF-2-DHAs) that have fluorescence emission profiles similar to that of DAF-2 triazole (DAF-2T), the reaction product of DAF-2 and N2O3.15 N2O3 is the intermediate product of NO oxidation, and it is N2O3 that reacts with DAF-2. In addition, (3) Moncada, S.; Palmer, R. M. J.; Higgs, E. A. Pharmacol. Rev. 1991, 43, 109142. (4) Bredt, D. S.; Snyder, S. H. Annu. Rev. Biochem. 1994, 63, 175-195. (5) Bogdan, C. Nat. Immunol. 2001, 2, 907-916. (6) Kojima, H.; Nakatsubo, N.; Kikuchi, K.; Kawahara, S.; Kirino, Y.; Nagoshi, H.; Hirata, Y.; Nagano, T. Anal. Chem. 1998, 70, 2446-2453. (7) Nagano, T.; Yoshimura, T. Chem. Rev. 2002, 102, 1235-1269. (8) Kuo, R. C.; Baxter, G. T.; Thompson, S. H.; Stricker, S. A.; Patton, C.; Bonaventura, J.; Epel, D. Nature 2000, 406, 633-636. (9) Itoh, Y.; Ma, F. H.; Hoshi, H.; Oka, M.; Noda, K.; Ukai, Y.; Kojima, H.; Nagano, T.; Toda, N. Anal. Biochem. 2000, 287, 203-209. (10) Broillet, M. C.; Randin, O.; Chatton, J. Y. FEBS Lett. 2001, 491, 227-232. (11) Kojima, H.; Hirata, M.; Kudo, Y.; Kikuchi, K.; Nagano, T. J. Neurochem. 2001, 76, 1404-1410. (12) Roychowdhury, S.; Luthe, A.; Keilhoff, G.; Wolf, G.; Horn, T. F. W. Glia 2002, 38, 103-114. (13) Nagata, N.; Momose, K.; Ishida, Y. J. Biochem. 1999, 125, 658-661. (14) Jourd’heuil, D. Free Radical Biol. Med. 2002, 33, 676-684.

Analytical Chemistry, Vol. 78, No. 6, March 15, 2006 1859

fluorescence microscopy and fluorometer data may be biased, because fluorescence generated from the DAF-2/DHA reaction cannot be resolved from that of DAF-2T (from NO) without chromatographic or electrophoretic separation. Capillary electrophoresis (CE) with laser-induced fluorescence (LIF) is one of the best choices for single cell analysis; it separates DAF-2 reaction products so that DAF-2-DHAs and DHA-2T can be detected in one CE run.16,17 It is feasible to assay the amount of NO, despite the competitive reaction of DHA, if the reaction time and amount of DAF-2 used is controlled. However, the DAF-2 reaction with NO is inhibited in the presence of ascorbic acid (AA) to a larger extent than by the direct reaction of DAF-2 with DHA, possibly because AA affects the amount of available N2O3.15 Supporting this mechanism are several reports showing that AA inhibits nitrosation by competing for the nitrosating agents formed from nitrite.18,19 AA tends to be ubiquitously distributed in biological systems, with a concentration range of 0.1-10 mM, which is often 3 orders of magnitude higher than the expected NO level (10 nM-10 µM).20-22 As an example, the measured AA concentration in Aplysia californica MCC neurons is ∼1 mM,23 making quantitation of the expected level of NO in these single cells difficult. We recently reported a DAF-2 method that detects NO in the presence of AA and DHA by freezing the sample and spatially separating the NO source from the DAF-2 sensor.24 As a result, only gaseous NO diffuses out of the frozen sample to react with DAF-2. Using this approach, significant NO production in the mouth area of the sea slug was reported. Although this method permits rapid analysis of NO in large tissues using CE-LIF, a fluorometer, or both, it is not optimal for single cell NO detection due to the small sample size and reagent volumes inherent to single cell analysis. Here we describe an enzymatic method that uses the enzyme ascorbate oxidase (AO) to eliminate the effect of AA on the measured values of NO, and the method can be applied to single cell assays of NO with CE-LIF. AO, one of the multicopper oxidases, catalyzes the four-electron reduction of dioxygen to water. By using L-ascorbate as the reducing substrate, within minutes, AO catalyzes the oxidation of AA with oxygen present in the air to form L-dehydroascorbic acid and water.25-28 AO is (15) Zhang, X.; Kim, W. S.; Hatcher, N.; Potgieter, K.; Moroz, L. L.; Gillette, R.; Sweedler, J. V. J. Biol. Chem. 2002, 277, 48472-48478. (16) Leikert, J. F.; Rathel, T. R.; Muller, C.; Vollmar, A. M.; Dirsch, V. M. FEBS Lett. 2001, 506, 131-134. (17) Rathel, T. R.; Leikert, J. F.; Vollmar, A. M.; Dirsch, V. M. Biol. Proced. Online 2003, 5, 136-142. (18) Licht, W. R.; Tannenbaum, S. R.; Deen, W. M. Carcinogenesis 1988, 9, 365372. (19) Tannenbaum, S. R.; Wishnok, J. S.; Leaf, C. D. Am. J. Clin. Nutr. 1991, 53, 247S-250S. (20) Grunewald, R. A. Brain Res. Rev. 1993, 18, 123-133. (21) Washko, P. W.; Wang, Y. H.; Levine, M. J. Biol. Chem. 1993, 268, 1553115535. (22) Rice, M. E.; Russo-Menna, I. Neuroscience 1998, 82, 1213-1223. (23) Kim, W. S.; Dahlgren, R. L.; Moroz, L. L.; Sweedler, J. V. Anal. Chem. 2002, 74, 5614-5620. (24) Ye, X. Y.; Kim, W. S.; Rubakhin, S. S.; Sweedler, J. V. Analyst 2004, 129, 1200-1205. (25) Clark, E. E.; Poillon, W. N.; Dawson, C. R. Biochim. Biophys. Acta 1966, 118, 72-81. (26) Cole, J. L.; Avigliano, L.; Morpurgo, L.; Solomon, E. I. J. Am. Chem. Soc. 1991, 113, 9080-9089. (27) Marchesini, A.; Kroneck, P. M. H. Eur. J. Biochem. 1979, 101, 65-76. (28) Meyer, T. E.; Marchesini, A.; Cusanovich, M. A.; Tollin, G. Biochemistry 1991, 30, 4619-4623.

1860

Analytical Chemistry, Vol. 78, No. 6, March 15, 2006

currently used to eliminate the interference of AA in oxidative colorimetric reactions in the clinical;29,30 food;31,32 and biological sciences, including neuroscience.33-35 Interestingly, Rebec et al.35 used AO in live animals to inactivate extracellular AA by infusing AO bilaterally into the striatum of rats to study the relationship between endogenous striatal AA and behavior. Their results showed that infusion of AO worked without undesirable side or cytotoxic effects. The inhibitory effect of AA on the formation of DAF-2T prevents NO from being detected in single cells, even after the addition of NO donor. Therefore, the aim of this study is to investigate whether the use of AO to eliminate the inhibiting effect of AA allows measurement of NO in single neurons. This AO approach is first tested using standards and biological in vitro systems, followed by evaluation of the protocol with a homogenate of mouth area tissue from A. californica, known to be nitric oxide synthase (NOS)-positive.36 Finally, this method is successfully used to detect NO in the B2 neuron, an NO-synthesizing cell, from Lymnaea stagnalis. EXPERIMENTAL SECTION Chemicals and Reagents. All reagents were of the highest available purity and were obtained from Sigma-Aldrich (St. Louis, MO), unless otherwise noted. The CE run buffer, 30 mM sodium borate, contained 0.03 g of boric acid (H3BO3) and 0.97 g of sodium borate decahydrate (Na2B4O7‚10H2O) in 100 mL of ultrapure Milli-Q water (Millipore, Bedford, MA), adjusted to pH 9.85 ( 0.05 with 3 M NaOH (Fisher Scientific, Fairlawn, NJ). Two different sample buffers were used in this work: 0.1 M phosphate buffer (pH 7.4 ( 0.1) was prepared using 0.26 g of monobasic sodium phosphate (NaH2PO4‚H2O) and 2.17 g of dibasic sodium phosphate (Na2HPO4‚7H2O) in 100 mL of ultrapure Milli-Q water; and phosphate-buffered saline (PBS) (pH 7.4 ( 0.1) containing 1.47 mM KH2PO4, 2.68 mM KCl, 137 mM NaCl, 0.90 mM CaCl2, 0.49 mM MgSO4, and 9.57 mM Na2HPO4‚7H2O. DAF-2 was purchased from EMD Biosciences, Inc. (Calbiochem Brand, San Diego, CA). DAF-FM and DAR-4M were purchased from Molecular Probes, Inc. (Eugene, OR). The original fluorescent reagents, dissolved in dimethyl sulfoxide (DMSO), were diluted into the desired buffer. The final concentration of DMSO in the DAF-2 and DAF-FM solutions was less than 1% (v/v). The L-ascorbic and L-dehydroascorbic acids were purchased from Fisher Scientific. DHA exists as a dimer in the crystalline form, but it spontaneously converts to a hydrated monomer in aqueous solution. The NO donor, diethylamine nitric oxide sodium salt (DEANO) (Molecular Probes, Inc.) decomposes to release free NO into solution with a half-life of 2.5 min at neutral pH and 37 °C.37 The AO spatulas (29) Speek, A. J.; Schrijver, J.; Schreurs, W. H. P. J. Chromatogr. 1984, 305, 53-60. (30) Esteban, M. R.; Ho, C. N. Microchem. J. 1997, 56, 122-129. (31) Saari, N. B.; Osman, A.; Selamat, J.; Fujita, S. Food Chem. 1999, 66, 5761. (32) Antonelli, M. L.; D’Ascenzo, G.; Lagana, A.; Pusceddu, P. Talanta 2002, 58, 961-967. (33) Mehlhorn, R. J. J. Biol. Chem. 1991, 266, 2724-2731. (34) Ebersole, B. J.; Molinoff, P. B. J. Neurochem. 1992, 58, 1300-1307. (35) Rebec, G. V.; Wang, Z. J. Neurosci. 2001, 21, 668-675. (36) Bodnarova, M.; Sadreev, R. I.; Panchin, Y. V.; Uvarov, P.; Jezzini, S.; Lovell, P.; Martasek, P.; Moroz, L. L. Program No. 44.2, Abstract Viewer and Itinerary Planner. Society for Neuroscience: Washington, DC, 2003; online. (37) Hrabie, J. A.; Klose, J. R.; Wink, D. A.; Keefer, L. K. J. Org. Chem. 1993, 58, 1472-1476.

were obtained from Roche Diagnostics GmbH Inc. (Mannheim, Germany). The paper zone of an AO spatula contains ∼17 units of AO. To make the AO solution, a spatula was kept in 2 mL of the desired buffer for 10 min and then stirred vigorously for 5 s, every 2 min. The resultant solution was filtered by a 0.2-µmdiameter, pore-sized syringe filter (Fisher Scientific) before use. CE-LIF Analysis and Fluorometry. A laboratory-assembled CE system, with 350-356-nm excitation from an Ar/Kr mixed gas laser, Coherent Innova 70 Spectrum (Coherent Inc., Palo Alto, CA), and a photomultiplier tube (PMT) HC125-03 (Hamamatsu Corporation, Bridgewater, NJ), was employed for separation and monitoring the reaction products of DAF-2 with other compounds. This CE system has been previously described.23 Injections took place at 8 kV (12 µA) for 8 s, corresponding to a 7.6-nL injection volume. Separations were performed at 20 kV (∼40-42 µA). Detection took place on-capillary, 60 cm beyond the point of injection. The fluorescence was collected at 90° by an all-reflective microscope objective (Ealing Electrooptical, Holliston, MA) and filtered spatially by a machined 3-mm pinhole and spectrally by a 80-nm fwhm, 500-nm interference filter, 03FIB004 (Melles Griot, Irvine, CA), and a high-pass filter, 400EFLP (Janos Technology, Townshend, VT). The PMT signal, consisting of a series of TTL pulses, was read by a data acquisition card, 6024E (National Instruments, Austin, TX). All counting, instrumentation, and voltages were controlled by a custom-tailored program written in Labview 5.0.1 (National Instruments). Fluorescence measurements were performed with a spectrophotometer, F-3010 (Hitachi, Tokyo, Japan). Emission fluorescence (scan range 500-600 nm) of 0.2 mL of reaction mixture was monitored upon excitation at 495 nm. The excitation wavelength was chosen by scanning the excitation wavelength in a range of 410-500 nm, with the emission detection set at 515 nm. Both the excitation and emission band-passes were 5 nm. Scan speed and response time were 60 nm/min and 2 s, respectively. The fluorescence counts were calculated with Microsoft Excel software (Microsoft Corporation, Redmond, WA). Reaction Protocols. Due to the short lifetime of NO donor, the DEANO powder was kept on dry ice before use. After DEANO dissolution and dilution to desired concentrations in buffers at 4 °C, DEANO solutions were immediately mixed with the reaction solutions at room temperature, as described below. For purposes of evaluation and calibration, a series of known concentrations of DEANO solutions were added to and mixed with DAF-2 aliquots. If the effects of DHA or AA were being monitored, the DEANO solution was combined with DHA or AA just prior to the addition of the DAF-2 solution. Unless otherwise stated, all reactions were at room temperature for 20 min. Due to the instability of DHA and AA, all solutions were freshly made just before the experiment in a nitrogen-purged buffer solution containing 1 mM ethylenediamine-tetraacetic acid (EDTA). If any stock solution was used for more than one protocol, it was kept at 4 °C and used within 30 min. The CE experiment with AO used an incubation of AO with 10 µM DAF-2 (final concentration) and analytes in a 0.2-mL PCR tube. Cellular Analysis. Aplysia californica (Anaspidea aplysiidae) (100-200 g) were obtained from the Aplysia Research Facility (University of Miami, Miami, FL) and kept in an aquarium containing continuously circulating, aerated and filtered artificial

seawater (ASW) at 14-15 °C until use. Animals were anesthetized by injection of isotonic MgCl2 (∼30-50% of body weight) into the body cavity. The mouth areas were cut and placed in polypropylene vials for homogenization with a Teflon rod. The homogenized material was sonicated (∼1 min) and briefly centrifuged (2 min at 14000g) to remove cell membranes and any remaining sheath material. The supernatant was carefully removed and diluted with sample buffer, if needed, before CE analysis. The cerebral ganglia were dissected and placed in ASW containing (in mM) 460 NaCl, 10 KCl, 10 CaCl2, 22 MgCl2, 6 MgSO4, and 10 HEPES, pH 7.7, or in ASW-antibiotic solution: ASW containing 100 units/mL of penicillin G, 100 µg/mL of streptomycin, and 100 µg/mL of gentamicin, pH 7.7. The ganglionic sheaths were digested enzymatically by incubating the ganglia in ASW-antibiotic solution containing 1% of protease (Type IX: Bacterial; Sigma-Aldrich) at 36 °C for 1-2 h, depending on animal size. Next, the cerebral ganglia were washed in fresh ASW, and the dorsal side of each was desheathed. Using a 0.38µm-diameter tungsten wire (WPI, Sarasota, FL), the ganglia were pinned dorsal side up to a silicone elastomer (Sylgard, Dow Corning, Midland, MI) layer, then placed in a recording chamber containing 3-4 mL of ASW-antibiotic media. Identified neurons and neuronal groups were manually dissected using sharp tungsten needles. Specimen of Lymnaea stagnalis (Pulmonata, Basommatophora), freshwater pond snails, were from an animal colony maintained in our laboratory for seven years in aerated freshwater aquariums at room temperature and fed lettuce. All animals selected for experimentation were adults of at least 25 mm in shell length. After surgical dissection, the central nervous systems (CNS) were placed in an extracellular physiological solution consisting of the following compounds (in mM): 44 NaCl, 1.7 KCl, 4 CaCl2, 1.5 MgCl2, and 10 HEPES (pH 7.4 ( 0.2, adjusted with 3 M NaOH). The thick connective tissue sheaths were removed, and the ganglia treated with 0.2% protease (type XIV, Sigma-Aldrich) for 30 min; the thin connective tissue sheaths were removed after washing out the protease. Neurons were visually identified according to earlier published maps of the CNS and by their morphological characteristics.38-40 Single neurons were mechanically isolated using two fine tungsten wires. In both animal experiments, the dissected neurons and neuronal groups were transferred into 0.75-mL polypropylene tubes (Fisher Scientific). Derivatization without AO application was achieved by adding 2 µL of 30 µM DAF-2 solution (final concentration: 10 µM) and 4 µL of buffer. Derivatization with AO application was achieved by adding 2 µL of 30 µM DAF-2 solution, 2 µL of AO solution, and 2 µL of buffer. Data Analysis. Peak heights in standards and in cellular and homogenate samples were determined by software Grams 386 (Thermo Galactic, Salem, NH). Statistical analyses were performed using the t-test procedure of SigmaPlot 2000 (SPSS Inc., Chicago, IL). (38) Benjamin, P. R.; Winlow, W. Comp. Biochem. Physiol., Part A: Mol. Integr. Physiol. 1981, 70, 293-307. (39) McCrohan, C. R.; Benjamin, P. R. J. Exp. Biol. 1980, 85, 149-168. (40) McCrohan, C. R.; Benjamin, P. R. J. Exp. Biol. 1980, 85, 169-186.

Analytical Chemistry, Vol. 78, No. 6, March 15, 2006

1861

Figure 1. Electropherograms of DAF-2 derivatized samples: (a) MCC; (b) F-cluster; (c) E-cluster; (d) upper labial nerve; and (e) secretory vesicles from A. californica. Conditions: CE run buffer, 30 mM borate at pH 9.85; capillary dimensions, 80 cm × 50 µm i.d. with 60-cm effective length; electrokinetic injection, 8 s at 8 kV; electrophoresis voltage, 20 kV; fluorescence excitation, 350-356 nm from an Ar/Kr mixed gas laser.

RESULTS AND DISCUSSION DHA/AA Interference with NO Detection by DAF-2 in Biological Samples. DHA reacts with DAF-2 under physiologically relevant conditions to form a unique series of fluorescent products, from which we identified two major compounds, DAF2-DHA-500 and DAF-2-DHA-518.15 The two primary amines of DAF-2 act as electron-donating groups to react with the 1,2carbonyl groups of DHA to form a conjugated system containing the -CdN moieties. Reaction products between these two reactants have been previously suggested to contain either a quinoxaline structure with the five-member ring of ascorbate remaining closed41 or a 1H-quinoxaline-2-one structure with the ascorbate ring opened.42,43 Indeed, these have been found to be the major peaks using electrospray ionization mass spectrometric analysis, and further tandem mass analysis confirmed these assignments. To confirm the cross-reaction between DHA and DAF-2 in biological specimens, five different cellular samples from A. californica are analyzed here (Figure 1). DAF-2-DHA peaks and DAF-2T peaks are identified by a comparison to standards run immediately before or after CE assays. As expected, there is DHA in all samples and in varying amounts. DHA/AA exists in a (41) Parish, H. A.; Gilliom, R. D. Carbohydr. Res. 1982, 102, 302-307. (42) Awad, L. F. Carbohydr. Res. 2000, 326, 34-42. (43) Somogyi, L. Liebigs Ann. Chem. 1995, 4, 721-724.

1862 Analytical Chemistry, Vol. 78, No. 6, March 15, 2006

number of partially oxidized forms and so is often detected as multiple peaks. We confirm the source of these peaks by the analysis of a series of standards and a variety of oxidizing and reducing agents. Although CE-LIF separates DAF-2T (migrates at ∼27 min) from DAF-2-DHAs, it is still difficult to judge whether there is NO produced in the cell because the DAF-2T peak is so small that there is little difference between it and the blank DAF2T signal. What is the origin of the blank signal? This blank signal may arise from preexisting NO or N2O3 in the environment, or it may be from the solution storage containers. In fact, the presence of NO in the exhaled air of healthy humans has been measured at 30 ppb44 Commercially available DAF-2 from different distributors and from different batches from the same distributor exhibit different DAF-2T blank peak intensities, and the intensity of this peak increases as the DAF-2 stock solution ages. Other researchers have also reported that commercial DAF-2 has minor fluorescent components, including DAF-2T, when using HPLC.45 Because of this signal, it is important to run control solutions with every experiment. DHA/AA Interference with NO Detection by DAF-FM and DAR-4M. AA and DHA interfere with DAF-2-assisted NO measurements, adding significant complexity to the interpretation of experimental results, especially those obtained from live cells and tissues. One possible solution for this problem is to use other DAF reagents. Although DAF-2 is the most commonly employed fluorescent reagent for NO detection, it is not the only one; DAFFM and DAR-4M are available, each with specific advantages. For example, when compared to DAF-2, DAF-FM is more tolerant of pH changes, and DAR-4M uses lower energy excitation, which results in less damage to biological samples. Although these fluorescent indicators have distinct structures, their reaction mechanisms with NO are similar. Therefore, it is reasonable to investigate whether DAF-FM and DAR-4M also react with DHA to form fluorescent products similar to those described above and whether AA reduces DAF-2 nitrosation. The reaction of DAF-FM and DAR-4M with DHA, as well as the effects of AA, are shown in Figure 2. The results clearly demonstrate that DAF-FM also cross-reacts with DHA. Combining 1 mM of DHA and 10 µM of DAF-FM in solution (pH 7.4) for 20 min produces one broad peak, observed with CE-LIF (Figure 2A,c). DAF-FM combined with both DHA and DEANO generates DAF-FM-DHA as well as DAFFMT (Figure 2A,d). In the presence of AA (Figure 2A,e), a decrease in the DAF-FMT intensity is observed, even though the 64% decrease is less than that observed for DAF-2. Since the excitation wavelength for DAF-4M is 560 nm and our CE-LIF system uses 352 nm excitation, the DAR-4M reaction with DHA and the AA effects on the reaction are only tested with the fluorometer (Figure 2B). DAR-4M reacts with DHA more easily than DAF-2, and AA also partially inhibits DAR-4M triazole formation. Thus, we conclude that DHA/AA influences NO measurement by DAF-FM and DAR-4M. Methods to reduce this effect are required and described below. Use of Ascorbate Oxidase. The effects of DHA on the measured NO concentrations are due to the competition between NO and DHA for DAF-2. These effects can be reduced by adding (44) Alving, K.; Weitzberg, E.; Lundberg, J. M. Eur. Respir. J. 1993, 6, 13681370. (45) Takahama, U.; Hirota, S.; Oniki, T. Free Radical Res. 2005, 39, 737-745.

Figure 3. Application of AO eliminates the inhibitory effect on DAF2T formation by AA. Representative electropherograms include: (a) 10 µM DAF-2 + 10 µM DEANO; (b) 10 µM DAF-2 + 10 µM DEANO + 1 mM AA; (c) 10 µM DAF-2 + 10 µM DEANO + 1 mM AA + AO.

Figure 2. A. Electropherograms of DAF-FM with DHA, AA, and DEANO standards: (a) 10 µM DAF-FM; (b) 10 µM DAF-FM + 1 mM DEANO; (c) 10 µM DAF-FM + 1 mM DHA; (d) 10 µM DAF-FM + 1 mM DEANO + 1mM DHA; (e) 10 µM DAF-FM + 1 mM DEANO + 1 mM AA. The reagents are dissolved in pH 7.4 ( 0.1 sodium phosphate buffer, and the reaction time is 20 min. B. Fluorescence spectra of DAR-4M with DHA, AA, and DEANO. Excitation wavelength is 560 nm. All reagents are dissolved in pH 7.4 ( 0.1 sodium phosphate buffer, and the reaction time is 35 min.

excess fluorescence reagent, followed by CE separation; however, AA interference on NO measurement occurs by inhibiting the DAF-2 + NO reaction. This is attributed to an additional reaction between AA and N2O3 to form NO, which decreases the amount of N2O3 available to react with DAF-2 and form DAF-2T.13 As shown in Figure 3, in the presence of 1 mM AA, even the addition of 10 µM DEANO to the DAF-2 solution does not result in an

Figure 4. In vitro test of AO application with metacerebral cells (MCC) from A. californica. Representative electropherograms are shown: (a) MCC; (b) 10 µM DAF-2; (c) 10 µM DAF-2 + MCC with DEANO added (final concentration of 750 nM) and not treated with AO; (d) 10 µM DAF-2 + MCC with DEANO added (final concentration of 750 nM), after treatment with AO.

increase in the intensity of the DAF-2T peak as compared to the blank peak. The small DAF-2-DHA peaks detected (Figure 3b) are likely due to the rapid autoxidation of AA. In an effort to eliminate the inhibitory effect of AA, AO is applied. AA is converted to DHA by AO and, thus, is no longer available to inhibit the reaction between NO and DAF-2. The conversion efficiency is 99.4 Analytical Chemistry, Vol. 78, No. 6, March 15, 2006

1863

Figure 5. (A) Representative electropherograms of DAF-2-reagent-derivatized B2, B1, and B4 neurons from L. stagnalis after AO application as well as a blank DAF-2 electropherogram. (B) Summary of measurements of NO in the identified individual neuron electropherograms. Average values ( standard deviations are shown.

( 8.4% (n ) 12) based on nine experiments in the absence of NO donor with different AA levels (three repeats at 5 µM, 1 µM, and 0.2 µM each) and three experiments with AA at 1 mM in the presence of NO donor. As shown in Figure 3, while the DAF-2T peak is almost undetectable in the presence of AA (Figure 3b), this peak is detectable at the expected (no AA level) fluorescence intensity after the addition of AO (Figure 3c). Furthermore, there are no other peaks generated from AO addition, which shows AO specifically catalyzes the oxidation of AA. On the basis of standard experiments, we performed >20 control experiments, such as neurons spiked with NO donors, NO inhibitors, and other compounds, as well as dilution experiments. The AO DAF/CE-LIF methodology allows NO to be measured without interference from other compounds. One of the in vitro experiments (shown in Figure 4) uses the metacerebral cell (MCC) from A. californica. The MCC is a good model system for examining AA/DHA effects on DAF-2T formation because it has high amounts of both AA and DHA.23 The reported concentrations are 1.5 ( 0.2 mM for AA and 0.28 ( 0.06 mM for DHA. However, there is ambiguity about whether MCC is NO-positive.36,46 For this reason, NO donor was added to test AO-assisted NO detection in this biological model system. The importance of (46) Moroz, L. L. Microsc. Res. Tech. 2000, 49, 557-569.

1864 Analytical Chemistry, Vol. 78, No. 6, March 15, 2006

using a separation in these measurements is obvious (Figure 4), because the NO-related peak (DAF-2T), the DHA related peaks, and several neuron-specific, unidentified peaks (Figure 4a) are well-separated from each other. These extraneous, unidentified compounds would certainly complicate detection in fluorometeronly or fluorescence-microscope-based approaches. When examining the MCC cell without the application of AO (Figure 4c), the DAF-2T peak observed is not different from the DAF-2T blank peak (Figure 4b), even though the final concentration of NO donor is 750 nM. The results shown here use low microliter derivatization reactions so that a picoliter biological sample is diluted about 1000-fold before being analyzed; obviously, by using smaller volume reactions, our ability to detect low concentration NO products in small-volume samples, such as single cells, increases greatly. NO Measurements in Tissue Homogenate. NO plays an important role in the control of the feeding networks in different animal models. Therefore, we tested homogenate from the mouth areas of A. californica to determine the feasibility of the AO approach in DAF-2 NO measurements. NOS requires the reduced form of NADPH as a cofactor for its function. Thus, fixativeresistant NADPH diaphorase (NADPH-d) staining is used as an

indicator of NOS presence.47 The mouth areas are positively NADPH-d-stained and contain sensory cells known as a source of NO.24,36 Significant amounts of NO are observed with AO addition to the mouth area tissue homogenate (P < 0.05, n ) 5; data not shown). Without AO application, NO production is also observed, but in much lower amounts because AA inhibition of NO-induced DAF-2 nitrosation is concentration-dependent. Only at high concentrations (e.g., 1 mM), does AA completely attenuate DAF-2T formation; lower AA concentrations also attenuate DAF-2 nitrosation, but to a lesser extent. This effect has been confirmed with standard experiments (data not shown). NO Measurements in Single Cells. L. stagnalis is a wellestablished animal model for the study of NO biology.46,48-52 To validate our AO approach for smaller, single cell NO detection, NO levels are measured in the single B2, B1, and B4 neurons of L. stagnalis (Figure 5), because these remain some of the best examples of identifiable and easily isolatable NOS-positive and -negative neurons. B2 neurons of the buccal ganglia contain NOS, but B4 and B1 do not. In Figure 5A, we can see that a DAF-2T signal is detected in B1 and B2 neurons. DAF-2-DHA peaks are small because the buccal neuron cytoplasm is diluted several thousand-fold during DAF addition and reaction so that the effect of native DHA is reduced. Our data show that B2 cells produce significant amounts of NO, as compared to B4 cells (P < 0.05, n ) 4), which produce a DAF-2T signal with an intensity similar to that detected in the control blank sample (Figure 5B). Measurements of B1 neuron contents do not produce significant differences as compared to B4 measurements. In contrast to the uniformly low NO signal in B4 neurons, in some cases, the DAF-2T peaks in B1 neurons are more intense than in control blank experiments, suggesting the presence of detectable amounts of NO. The origin (47) Moroz, L. L.; Gillette, R.; Sweedler, J. V. J. Exp. Biol. 1999, 202, 333-341. (48) Korneev, S. A.; Piper, M. R.; Picot, J.; Phillips, R.; Korneeva, E. I.; O’Shea, M. J. Neurobiol. 1998, 35, 65-76. (49) Korneev, S. A.; Straub, V.; Kemenes, I.; Korneeva, E. I.; Ott, S. R.; Benjamin, P. R.; O’Shea, M. J. Neurosci. 2005, 25, 1188-1192. (50) Moroz, L. L.; Dahlgren, R. L.; Boudko, D.; Sweedler, J. V.; Lovell, P. J. Inorg. Biochem. 2005, 99, 929-939. (51) Serfozo, Z.; Vereb, Z.; Roszer, T.; Kemenes, G.; Elekes, K. J. Neurocytol. 2002, 31, 131-147. (52) Taylor, B. E.; Harris, M. B.; Burk, M.; Smyth, K.; Lukowiak, K.; Remmers, J. E. J. Exp. Zool., Part A 2003, 295A, 37-46.

of this NO is under investigation and may include NO produced by unknown L. stagnalis NOS, by diffusion of NO from B4 neurons, or other possibilities. It is worth noting that the standard deviation of NO measured in single cells is quite large, which may indicate that even individual cells of the same type may produce different amounts of NO. CONCLUSIONS We find that the commonly used fluorescent reagents for NO detection, such as DAF-2, DAF-FM, and DAR-4M, cross-react with DHA and that AA inhibits the nitrosation of these fluorescent reagents, resulting in fluorescence signal attenuation. We introduce a novel approach for solving this problem by using AO to quickly convert AA to DHA, followed by CE to separate the fluorescent products from DHA and NO. This combined AO/CELIF approach enables the simultaneous measurement of DHA and NO. We test the viability of this methodology with standard solutions, in vitro single cell analyses, and cell homogenate measurements, thereby confirming its application to single cell analysis by measuring NO production in identified buccal neurons from L. stagnalis. Significant NO is detected in NOS-positive B2 neurons and not detected in NOS-negative B4 neurons. Therefore, we conclude that this simple, yet specific, detection method effectively measures NO in single cells without interference from other compounds. ACKNOWLEDGMENT We thank Andrew Shaw for his assistance with the fluorometer measurements. The financial support of NSF CHE 98-77071 and NIH Grants DK 070285 and NS 031609 to J.V.S. is gratefully acknowledged. We thank Ms. Stephanie Baker for critical reading and editing of this manuscript.

Received for review December 3, 2005.

October

19,

2005.

Accepted

AC051877P

Analytical Chemistry, Vol. 78, No. 6, March 15, 2006

1865