This is a continuation of U.S. Application Serial No. 07/871 ,463, entitled "Nitric Oxide Sensor," filed April 21 , 1992, by Tadeusz Malinski, incorporated herein by reference.

FIELD OF THE INVENTION The present invention generally relates to sensors and sensing techniques which can selectively and quantitatively detect NO in solution in both biological and chemical media. More specifically, the present invention relates to NO sensors which utilize conductive catalytic materials deposited on microfibers or other supports to monitor the presence or release of NO using amperometric, voltammetric or coulometric methods.

BACKGROUND OF THE INVENTION

Nitric oxide (NO) has recently been shown to be a key bioregulatory molecule in a number of physiological processes. For example, NO plays a major role in the biological activity of endothelium derived relaxing factor (EDRF), abnormalities in which are associated with acute hypertension, diabetes, ischaemia and atherosclerosis. NO is also considered a retrograde messenger in the central nervous system, appears to be involved in the regulation of macrophage cytotoxic activity and platelet aggregation inhibition, and has been implicated in endotoxic shock and genetic mutations. In addition, a number of drugs and other xenobiotics are metabolized to produce NO as either the effector molecule or as a harmful metabolite. See e.g. Furchgott, R.F. et al.,

Nature 288:373-376 (1980); Palmer, R.M. et al., Nature 327:524-526 (1987); Furchgott,

R.F., "Mechanism of Vasodilation", IV:401-414 (ed. Vanhoutte, P.M.) (Raven, NY) (1988); Ignaro, LJ. et al., PNAS (USA) 84:9265-9269 (1987); Wei, E.P. et al., Cir. Res.

57:781-787 (1985); Piper, G.M. et al., J. Am. J. Physiol. 24:4825-4833 (1988);

Vanbethuysen, K.M. et al., J. Clin. Invest. 79:265-274 (1987); Freiman, P.C. et al., Circ.

Res. 58:783-789 (1986); and Schuman, E.M. et al., Science 254:1503 (1991).

Several different methods have been employed in the past to measure NO concentration in aqueous solution. For example, analysis of the ultimate aerobic oxidation products of NO, i.e. nitrite/nitrate (NO ₂ 7NO ₃ ^" ), has been used as a measure of NO. Monitoring of UV-vis spectral changes resulting from the conversion of oxyhemoglobin to met-hemoglobin has also been used as an indicator of NO concentration. These methods, however, provide only an indirect and thus less accurate measurement of NO. NO has also been measured in biological systems

using a Thermal Energy Analyzer (TEA), in which NO reacts with ozone to produce a characteristic chemiluminescent response. Downes, M.J. et al., Analyst 101 :742-748 (1978). This approach, however, requires a lengthy regeneration time, the isolation of NO from solution, and cannot be miniaturized for in situ monitoring of NO release. Mass spectrometry has also been attempted, but has problems similar to the TEA approach.

Recently, a modified oxygen electrode for the detection of nitric oxide has also been reported. Shibuki, K., Neurosa Res.9:69-76 (1990). This electrode, however, has a relatively large diameter (0.25 mm), a slow response time and a narrow concentration range (1-3 μM). Although this method is advantageous in that it discriminates against the NO ₂ ^~ produced in the outer solution of the electrode, it is not selective for NO in the presence of any NO ₂ ^" produced in the electrode inner solution. There is also some question of the validity of this technique owing to the small current observed and the lack of standards done in less than μM concentration. Although the above-described methods can be used to measure NO in biological or chemical media, they are not sufficiently sensitive nor specific to provide a direct and accurate quantitative measurement of NO, particularly at low NO concentrations. Furthermore, none of the methods or sensors employed to date can rapidly and selectively measure NO release by the cell in situ in the presence of oxygen and/or NO ₂ ^" . Development of this methodology is crucial in order to evaluate endogenous NO release, distribution and reactivity on molecular level in biological systems.

Thus, there exists a need for a sensitive and selective sensor for direct quant ^' rtive measurements of NO. An optimal sensor for monitoring NO release should be sturdy and capable of sufficient miniaturization for in situ measurement in a single cell. The sensor should also be sensitive enough to produce an adequate signal to be observable at the low levels of NO secreted in biological environments. Due to the variation in the amount of NO secreted by different types of cells (e.g. from nanomoles/10 ⁶ cells in macrophages to picomoles in endothelial cells), the signal produced by the sensor should also change linearly over a wide range of concentrations. See Marietta, M.A., Trends Biochem. ScL 14:488-492 (1989). The short half-life of NO in biological systems, on the order of about 3 - 50 seconds, also mandates a fast response time. See Moncada, S. et al., Pharmacol. Rev. 43:109-142 (1991). The NO sensor and method of the present invention exhibit these desirable characteristics.

SUMMARY OF THE INVENTION

The NO sensor and method of the present invention provide a direct and accurate measurement of NO in biological and chemical media. A sensor of the present invention generally comprises an electrode having a catalytic material capable of catalyzing oxidation of NO coated with a cationic exchanger. The sensor provides a direct measurement of NO through the redox reaction of NO - NO ⁺ + e ^" and is selective for NO through the discrimination of the cationic exchanger against nitrite

(NO ₂ ^" ). Although the sensor can be fabricated on any scale, it can be miniaturized to provide a microsensor which can accurately measure NO in situ at the cell level. in one preferred embodiment of the invention, the amount of NO released from a single cell can be selectively measured in situ by a microsensor with a response time better than about 10 msec. The microsensor comprises a thermally-sharpened conductive carbon fiber with a tip diameter of about 0.5-0.7 μm covered with several layers of polymeric porphyrin capable of catalyzing NO oxidaiton with a cationic exchanger deposited thereon. Using either a two or three electrode arrangement, the microsensor can be operated in either the amperometric, voltammetric or coulometric mode. The microsensor is characterized by a linear response up to about 300 μM and a detection limit of about 10 nM NO concentration, which allows detection of NO release at the levels present in a single biological cell. The sensor also discriminates against NO ₂ ^" , the most problematic interferant with current NO sensing techniques.

In other embodiments of the present invention, larger scale NO sensors are used to measure NO concentrations in chemical media, cell culture, extracellular fluids and tissue, rather than in single cells. For example, a carbon electrode with a larger tip diameter, platinum mesh or a tin indium oxide layered plate is coated with a conductive catalytic polymeric porphyrin and a cationic exchanger. A linear response and low detection limits similar to the NO microsensor are observed.

Other features and advantages of the present invention will become apparent from the following description and appended claims, taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

Figures 1A and B depict preferred monomeric porphyrin structures used in sensors of the present invention.

Figure 2 is a differential pulse voltammogram of NO at various concentrations. Figure 3 is a graph showing nitric oxide response (nA) of NO solutions measured by a sensor of the invention.

Figure 4A is a microscopic photograph of a carbon fiber microsensor of the present invention.

Figure 4B is an electron scanning micrograph of the portion of the microsensor covered with a coat of isolating wax-resin mixture. Figure 4C is an electron scanning micrograph of the thermally-sharpened tip of the microsensor covered with conductive polymeric porphyrin.

Figure 5 is a scan showing the growth patterns for poly-TMHPPNi, deposited from 5x1 O^M TMHPPNi, 0.1 M NaOH solution by continuous scan cyclic voltammetry on a carbon fiber microelectrode. Figures 6A, B, C and D are voltammograms showing the response of the microsensor in the differential pulse voltametric mode.

Figures 7A, B, C and D are scans showing the response of the microsensor in the amperometric mode.

Figure 8A - C are schematic overviews of macrosensors of the present invention used in cell culture.

Figure 9 shows the response of a platinum mesh macrosensor to NO release by a cell culture grown directly on the sensor surface.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

The basic strategy used in the design of a preferred embodiment of the NO sensor is based on catalytic oxidation of NO which uses a specific potential unique to NO - NO ⁺ + e ^" . The normal oxidation potential for NO is about 1.0 V vs SCE on a standard platinum electrode, which potential can be lowered with various materials capable of catalytically oxidizing NO. The current or charge generated thereby is high enough to be used as an analytical signal in microsystem. In accordance with the principles of the present invention, a working electrode of a sensor of the present comprise a conductive solid support with a catalytic surface for NO oxidation. A catalytic surface on a conductive support can be provided using several approaches. For example a conductive catalytic material capable of catalyzing

NO oxidation can be layered on a conductive solid support; the conductive catalytic material can be layered on a conductive material coated on a conductive or nonconductϊve base material; or the conductive catalytic material can itself comprise the conductive support. The third approach can be accomplished by fashioning the electrode directly from the conductive catalytic material or by incorporating or doping a catalyst into the support material. A working electrode of a sensor of the described embodiment of this invention preferably comprises a solid conductive support coated

with one or more layers of a conductive material capable of catalyzing oxidation of NO, hereinafter referred to as catalytic material.

It will be appreciated that several types of catalytic materials can be used in a sensor of the present invention, as long as the catalytic material exhibits electronic, ionic or redox conductivity or semiconductivity, collectively referred to herein as conductivity. Such materials include, but are not limited to, polymeric porphyrins and polypthalocyanines. The above-mentioned materials can contain central metals, preferably transition or amphoteric metals. Polymers which can also be used but require doping include, for example, polyvinylmetallocenes (e.g. ferrocene), polyacetylene doped with different metal redox centers and polypyrroline doped with different redox centers such as, e.g. methyl viologen.

Preferred catalytic conductive materials for a sensor of the present invention are polymeric metalloporphyrins, which are organic p-type semiconductors with relatively high conductivity and which can be successfully deposited on a supporting conductive material. Polymeric metalloporphyrins have been shown to have high catalytic effect for the electrochemical oxidation of several small organic and inorganic molecules. Bennett, J.E. et al., Chem. Materials 3:490-495 (1991). Polymeric porphyrins polymerized and copolymerized from monomeric porphyrins N,N'-di(5-p- phenylene-10,15,20-tri(3-methoxy-4-hydroxyphenyl)porphyrin;1 ,10,-phenantroline-4,7- diamine, and 5-p-(pyrole-1 -yl) phenylene-1 0,1 5,20-tri-(3-methoxy-4- hydroxyphenyl) porphyrin with Fe, Mn, Co and Ni as central metals are more preferred given their high catalytic effect for selective electrochemical oxidation of NO. Even more preferred compounds include tetrakis(3-methoxy-4-hydroxyphenyl) porphyrin (TMHPP) and meso-5'-O-p-phenylene-2',3'-O-isopropylidene uridine-tri(n-methyl-4- pyridinium)porphyrin (PUP), shown in Figures 1A and B.

In order to discriminate against NO ₂ ^" , the porphyrinic catalysts used in the present invention are also preferably covered with a thin layer of a cationic exchanger to prevent anion diffusion to the catalytic surface. Suitable cationic exchangers include AQ55D available from Kodak and Nafion. Nation, which is used in the Specific Examples, is a negatively charged cationic exchange polymer which prevents diffusion of anions like NO ₂ ^" to the electroactive surface of the polymeric porphyrin, but is highly permeable to NO.

The thin layer of polymeric porphyrin film can be electrochemically deposited, as described in detail below, on any solid conductive support. As previously discussed, a conductive support can comprise a material that in itself is conductive

or a conductive or nonconductive base material coated with a conductive material. Conductive materials which do not need to be coated with additional conductive materials are preferred. It will also be appreciated that, although the catalytic component of the invention is preferably layered on a conductive support, the conductive catalytic material can also comprise the conductive support. Conductive support materials particularly suitable for smaller scale sensors of the invention include carbon fibers, and gold or platinum wire. Due to their mechanical properties as well as the possibility for controlled miniaturization, carbon fibers are preferable support materials for microsensors in single cell applications. See e.g. Maiinski, T. et al. Anal. Chem. Ada. 249:35-41 (1991); Bailey, F. et al., Anal. Chem. 63:395-398 (1991).

It will be appreciated that the dimension of the sensor of the invention can be varied to produce virtually any size sensor, including microsensors with a tip diameter of about 1 μm or less and macrosensors, including fibers with a larger tip diameter (e.g. about 1 - 10 mm) and metallic mesh and conductive layered plates. Thus, while Specific Examples Il-V describe the production and use of a microsensor for use in small environments such as single cells or synapses, the same techniques can be applied to a larger support, such as described in Specific Examples I and IV , to produce convenient macrosensors for tissue, cell culture or chemical media studies.

In measuring NO, a two or preferably three electrode system can be employed. The working electrode, comprising the coated carbon fiber, with mesh or plate, is connected to a conductive lead wire (e.g. copper) with conductive (e.g. silver) epoxy, with the lead wire connecting to the voltammetric analyzer, potentiostat or coulometric measuring instrument. The auxiliary or counterelectrode generally comprises a chemically inert conductive material such as a nobel metal (e.g. platinum wire), carbon or tin indium oxide which is also connected to the measuring instrument with a lead wire. In a three electrode system, a reference electrode, such as a standard calomel electrode (SCE), is also employed and connected to the measuring instrument with a third conductive lead wire. in use, the working electrode, with the other electrode(s) in proximity, is placed into the analytic solution. It will be appreciated that by "analytic solution" is meant any aqueous or nonaqueous solution in which NO is to be detected or measured. The term thus includes both chemical and biological media, including tissue fluids and extracellular and cellular fluids. It will also be appreciated that the sensor of the invention can be used quantitatively to detect the presence of NO and also quantitatively to measure the levels of NO present in the analytic solution. To detect

or measure NO release in a single biological cell, a microsensor of appropriate dimension can be either inserted into or placed close to the cell membrane.

The cell membrane surface concentration of NO is influenced by the following factors: release of NO due to the action of bradykinin, adsorption and chemisorption of NO on the surface of the cell membrane, oxidation by O ₂ and organic molecules, and diffusion into other cells and to the bulk solution. The decay in NO response following the addition of standard amounts of oxygen was studied. Decreases of only 22% and 35% were observed after 4 and 10 min respectively, following the addition of 100 μM O ₂ to a 20 μM NO solution in the absence of biological material. These measurements in the presence of O ₂ indicate that its role in the oxidation of NO may have been overestimated, and that NO oxidation is due mainly to the biological material as was previously suggested by Monocada. See Moncada, S. et al., Pharmacol. Rev. 43:109-142 (1991).

SPECIFIC EXAMPLE I A carbon macroelectrode covered with conductive porphyrin polymer was prepared as follows. A glassy carbon electrode (GCE) (diameter about 2 mm) was coated with conductive polymeric porphyrins by cyclic voltammetry or controlled potential oxidation (4 min) at 0.7 V vs SCE of the monomeric porphyrin in 0.1 M NaOH solution (5 ml). The auxiliary electrode was a platinum (Pt) rod and the reference electrode was a standard calomel electrode (SCE). The porphyrin-coated (about 0.8 - 1.5 nm/cm ² ) electrode was removed from the solution and stored in 0.1 M base. The porphyrins used were Ni ²⁺ , Co ²⁺ , or Fe ³⁺ TMHPP or PUP as shown in Figures 1A and B. The porphyrin-coated electrodes were then further coated with 4 μl of 5% Nation solution. A stock solution of saturated nitric oxide was prepared anaerobically in pH =

7.4 (0.1 M) phosphate buffer. This stock solution was then added in the correct volume to obtain the desired final concentration of NO (10, 20 and 40 μM). The electrochemical cell had a Pt rod counter electrode and SCE reference electrode and the working electrode was the glassy carbon electrode coated with polymeric porphyrin film and Nation, as described above. It will, also be appreciated that a two electrode arrangement, i.e. the working and auxiliary electrode, can be utilized. Measurements were performed in 5 ml phosphate buffer (pH = 7.4, 0.1 M) which served as the supporting electrolyte. All solutions were degassed prior to use and kept under nitrogen. A base line scan was taken using linear sweep voltammetry (range = 0 to +0.9 V vs SCE) or differential pulse voltammetry (range = +0.4 to 0.9

V vs SCE). Aliquots of the NO stock solution were introduced to the cell via a gas- tight syringe. The final dilution was taken as the final NO concentration. A typical NO response by differential pulse voltammetry is shown in Figure 2 (Epa = 0.7 V vs SCE), which shows the results using a GCE/Ni(ll)TMPP Nafion-coated (4 μL) electrode. As shown in Figure 3, the electrode gave a linear response (as did all three electrodes) in the range of [NO] = 1 = 100 μM (0.7 V vs SCE).

SPECIFIC EXAMPLE II Carbon microfiber conductive supports for the microsensor were produced by threading an individual carbon fiber (7 μm) through the pulled end of a capillary tube with approximately 1 cm left protruding. Non-conductive epoxy was put at the glass/fiber interface. When the epoxy that was drawn into the tip of the capillary dried, the carbon fiber was seared in place. The carbon fiber was sharpened following standard procedure using a microburner. See Bailey, F. et al., Anal. Chem. 63:395- 398 (1991). The sharpened fiber was immersed in melted wax-resin (5:1) at controlled temperature for 5 - 15 sec. After cooling to room temperature, the fiber was sharpened again. During burning, the flame temperature and the distance of the fiber from the center of the flame need to be carefully controlled. While the diameter of the sharpened lip is smaller, the tip length is larger, with the overall effect of the resulting electrode being a slim cylinder with a small diameter rather than a short taper. This geometry aids in implantation and increases the active surface area. Scanning electron microscopy of the fiber produced shows that the wax is burned approximately to the top of the sharpened tip. The area of the tip, controllably fabricated with appropriate dimensions, is the only part of the carbon fiber where electrochemical processes can occur. A typical length of the electrochemically active tip is between 4-6 μm. For the sensor to be implanted into a cell, this length must be smaller than the thickness of the cell. The unsharpened end of the carbon fiber was attached to a copper wire lead with silver epoxy.

Referring now to the Figures, Figure 4A is a microscopic photograph of a complete NO microsensor of the present invention. Figure 4B is an electron scanning micrograph illustrating the part of the microsensor covered with the coat of isolating wax-resin mixture. Figure 4C is an electron scanning micrograph of the thermally- sharpened tip of the microsensor covered with conductive polymeric porphyrin and Nation as described below.

SPECIFIC EXAMPLE III

The growth patterns for poly-TMHPPNi were examined. Poly-TMHPPNi was deposited from a solution of 0.1 M NaOH containing 5 x 10 ^"4 M monomeric tetrakis(3- methoxy-4-hydroxy-phenyl)porphyrin (TMHPP), with Ni as a central metal (TMHPPNi) by continuous scan cyclic voltammetry from 0.0 to 1.1 V, on a carbon fiber microelectrode (16 μm ² surface area), as generally described in ¹⁶ . As shown in Figure 5, peaks la and lc correspond to the oxidation of Ni(ll) to Ni(lll) and reduction of Ni(lll) to Ni(ll), respectively, in the film. The Ni(ll)/Ni(lll) redox couple observed at 0.5 V allows porphyrin surface coverage, (r), to be monitored (optimal r = 0.7-1.2 nmol cm ² ). Surface coverage is calculated from the charge transferred under process la (r = 0.8 nM/cm ² ). The surface coverage depends upon the initial concentration of TMHPPNi, electrolysis time and potential.

Following deposit on the fiber, the porphyrin film was conditioned by 5 - 10 scans from 0.4 to 0.9 V. At this stage, the electrode should be stored in 0.1 M NaOH. Sensor fabrication was completed by dipping in the Nation solution (5%) for 15 - 20 sec and left to dry (5 min) and stored in pH 7.4 buffer. Since the Ni(ll)/Ni(lll) reaction requires diffusion of OH ^" to neutralize a charge generated in the poly-TMHPPNi and OH ^" cannot diffuse through Nation, the later absence of the Ni(ll)/Ni(lll) voltammetric peaks in 0.1 M NaOH demonstrated the integrity of the Nation film coverage. SPECIFIC EXAMPLE IV

NO monitoring was done by differential pulse voltammetry using a classic three electrode system: the sensor as the working electrode, a saturated calomel electrode (SCE) reference electrode and a platinum (PE) wire auxiliary electrode. The pulse amplitude was 40 mV and the phosphate buffer solution was pH 7.4. Differential pulse voltammograms were obtained for oxidation of NO on poly-TMHPPNi without Nation (depicted as A in Figure 6) and with Nation (depicted as C in Figure 6) and for 1 μM NO in the presence of 20 μM NO ₂ ^" on poly-TMHPPNi without Nation (depicted as B in Figure 6) and with Nation (depicted as D in Figure 6).

DPV of NO on poly-TMHPPNi without Nation showed a peak at 0.63 V in buffer pH 7.4 (see Figure 6A). DPV of a solution of 1 μM NO and 20 μM NO ₂ ^" showed a single peak at 0.80 V (see Figure 6B). The peak current was thus three times higher than that observed at 0.63 V for NO alone. This indicated that the oxidation of NO ₂ ^" and NO occur at a similar potential, but that the current increase is not proportional to the concentration of NO ₂ ^" . The NO peak current with the Nafion-coated sensor was observed at 0.64 V (see Figure 6C). Although the observed current is lower, Nation

coverage provides high selectivity against NO ₂ ^" . Only a 1 % increase in current and no change of potential was observed for oxidation of 1 μM NO in the presence of 20 μM NO ₂ ^" (see Figure 6D). Thus the porphyrinic microsensor was selective for NO and insensitive for NO ₂ ^" up to a ratio of at least 1 :20. A linear relationship was observed between current and NO concentration up to 300 μM (r = 0.994; slope = 2.05 nA/μM; n = 21). The response time (time for the signal increase from 10% to 75%) in the amperometric mode was less than 10 milliseconds. The detection limit calculated at a signal/noise ration = 3 was 20 nM for DPV and 10 nM for the amperometric method. Since, in a volume equivalent to that of an average single cell (10 ^"12 L), about 10 ^"2 attomoles (10 ^"20 moles) of NO can be detected, the detection limit of the sensor is 2-4 orders of magnitude lower than the estimated amount of NO released per single cell (1-200 attomol/cel!) ^{1 ,13} .

SPECIFIC EXAMPLE V Amperometric detection of NO by the microsensor under various biological conditions was also studied. Ring segments from porcine aorta (about 2-3 mm wide) and porcine aorta endotheiial cell culture were prepared according to previously described procedures. Using a computer-controlled micropositioner (0.2 mm X-Y-Z resolution), the microsensor could be implanted into a single cell, or placed on the surface of the cell membrane, or kept at a controlled distance from the cell membrane. Alternating current was measured in three electrode systems, as described above, at constant potential of 0.75 V modulated with 40 mV pulse in time intervals of 0.5 sec. The background, shown in Figure 7A, was measured in cell culture medium at 37°C (DMEM-Dulbecco's Modified Eagle Medium, 100 mg/L D-glucose, 2 mM glutamine, 110 mg/L sodium pyruvate, 15% controlled process serum replacement TYPE I). No change of the background was observed after the addition of 50 nM of bradykinin to 5 ml of cell culture medium. As shown in Figure 7B, 2 nm of NO were injected by microsyringe into the cell culture medium, a 5 mm distance from the microsensor. As shown in Figure 7C, one microsensor was placed on the surface of the single endotheiial cell in the aortic ring, and another was implanted into the smooth muscle cell. 2 nm of bradykinin was injected into the medium near the endotheiial cell. After 3 ± 0.5 sec (n = 7), NO release was detected and a steady increase of surface concentration to a plateau at 450 ± 40 nM was observed after 200 sec (see Figure 7C). After 16 min, the surface concentration of NO decreased to zero. No significant difference in NO surface concentration was found for the endotheiial cell from cell culture (430 ± 40 nM, n = 7). NO was detected in a single smooth muscle

cell within 6.0 ± 0.5 sec (n = 7) after injection of bradykinin and a maximum concentration (130 ± 10 nM, n = 7) was observed after 90 sec (see Figure 7D). The observed current indicates that the initial concentration around the sensor is 230 nM and decreases to 40 nM after 17 sec due mainly to depletion of NO by diffusion and also reaction with O _2>

SPECIFIC EXAMPLE VI Sensors utilizing a layer of tin indium oxide on a glass plate base used as either a counterelectrode or as the conductive layer of a working electrode were also constructed. Figure 8A illustrates the use of a layer of indium oxide (14) as a counterelectrode (10), (10), whereas Figure 8B illustrates its use as a conductive layer of the working electrode of a macrosensor (12). BCH1 myocytes (16) were grown under standard culture conditions at 2 x 10 ⁷ cell/cm ² on a glass plate (18) (Figures 8A and C) or on a plate layered with catalytic polymeric iron porphyrin with Nation coated thereon (20) (Figure 8B). As shown in Figures 8A and B the tin indium oxide semiconductor layer in both cases was attached to the measuring instrument by a copper wire lead (22) with silver epoxy (24).

The schematic of Figure 8C depicts the set up for NO measurements of the cell culture in Figure 8A. Cells were grown on a tin indium oxide (14) layered glass plate (18) placed in a Petri dish (20) with standard culture media (26). A microsensor working electrode (34) constructed as described in previous Examples was then used to measure NO release in situ. The culture was microscopically monitored (30 - inverted microscope) and the working electrode positioned with a micromanipulator (32). As shown in Figure 8C, microsensor was attached to a measuring instrument such as a voltammetric analyzer (36) with the results fed to a computer (38) connected to a plotter (40) and printer (42) for result readout. NO response results observed were on the order of those in the previously described Specific Examples. Similar results were also obtained in cell cultures grown on fine platinum mesh and are shown in Figure 9.

The foregoing discussion discloses and describes merely exemplary embodiments of the present invention. One skilled in the art will readily recognize from such discussion, and from the accompanying drawings and claims, that various changes, modifications and variations can be made therein without departing from the spirit and scope of the invention as described herein and defined in the following claims. All publications cited herein are incorporated by reference.