INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

NITRIC OXIDE (NO) is a small, diffusible molecule with a short half-life (8) with various biological functions (4, 12, 14). It was first discovered by Furchgott (5) to be an endothelium-derived relaxing factor, which was identified as being identical to NO by Palmer et al. (15) and Ignarro et al. (9) in 1987. Up until now, three different isoforms of NO synthases (NOS) have been identified (4, 14). The neuronal NOS, which is calcium/calmodulin dependent, produces NO that acts as a neurotransmitter. After stimulation, macrophages express large amounts of a calcium/calmodulin-independent NOS-releasing NO in high concentrations for bactericidal purposes. The constitutive endothelial NOS causes vascular relaxation via cGMP-mediated mechanisms, and, moreover, albuminally released NO inhibits platelet aggregation and leukocyte adhesion (14). Because NO has a short half-life (2-30 s) (8), it is difficult to measure authentic NO. It rapidly decomposes to nitrate and nitrite, which may accumulate in the sample. There are different methods of determining NO, which have been described in detail elsewhere (1). Briefly, it is possible to measure NO by using a bioassay (6), an oxyhemoglobin assay (10), electron paramagnetic resonance (7), chemiluminescence (15), HPLC (11), the Griess reaction (17), or different electrochemical electrodes (3, 13, 19). Moreover, it is possible to measure NO indirectly via its second messenger, cGMP.

Up until now, the Griess reaction has been used to measure nitrite/nitrate, although the detection limit of this technique is in a micromolar range, which is usually only reached by induction of the inducible NOS.

This study presents a new, sensitive method to determine nitrite/NO with a commercially available electrochemical NO electrode (WPI, Sarasota, FL). We will describe that it is possible to measure NO in aqueous solutions from different biological models by a conversion of nitrite to NO. A big advantage of this method is that the total (basal and agonist-induced) NO release of cells in culture can be measured because the stable degradation products of NO (nitrate/nitrite) accumulate in the incubation buffer.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Calibration of the NO electrode. The amperometric ISO-NO electrode (WPI) measures NO by oxidizing it at the working electrode. The resulting current is displayed in picoamperes and is proportional to NO concentrations (Fig. 1). To prevent nitrate and nitrite from interfering with the electrode, it is coated with a specialized membrane (WPI); other diffusable gases such as CO are not measured because of the potential at the electrode. The electrode was calibrated by using a modification of the protocol (3, 19) given by the manufacturer. A chemical titration calibration was performed by using an acidic iodide solution (0.1 mol/l H₂SO₄, 0.14 mol/l K₂SO₄, 0.1 mol/l KI), to which equal volumes of different KNO₂ solutions were added (Figs. 2 and 3). NO is formed stoichiometrically and is measured directly. Briefly, 500 µl of KNO₂ were added to 10 ml of the acidic solution at room temperature while being stirred (100 rpm). The same calibration was performed with increasing volumes of a KNO₂ solution of a definite concentration (5 µmol/l). Instantly the nitrite is reduced to NO, which is constantly measured by the electrode that is placed in the solution. The detection limit of the electrode is 2 nmol/l NO (2, 3) in an aqueous solution, which means that a minimum of 50 µl of a solution that contains 2 µmol/l nitrite has to be added to the 10-ml bath to be detectable. Therefore, in an experiment, the added nitrite sample is diluted according to its volume in the 10-ml bath, resulting in a much lower concentration. It is important for the volume of the sample containing nitrite to be greater than the detection limit in the 10-ml bath. It is possible to add solutions up to a volume of 500 µl (single addition) to the 10-ml solution without disturbing the measurement. When performing an experiment, we did not add more than 1 ml (total sum of serial injections) into the 10-ml bath; afterward we replaced the bath with a fresh solution to perform further experiments. We always took the peak currents to quantify our values.

View larger version (14K): [in this window] [in a new window]   Fig. 1.   Electric current [picoampere (pA)] induced by the reduction of nitric oxide (NO) at the electrode is directly proportional to the NO concentration (nmol/l).

View larger version (13K): [in this window] [in a new window]   Fig. 2.   Typical original recording of a calibration with KNO2 in which increasing concentrations of KNO2 (500 µl) were added to a 10-ml solution (stirred, 100 rpm) in which the electrode was placed.

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Calibration curve with increasing KNO2 concentrations (500 µl), which were added to a 10-ml bath. Linear regression yielded a well-correlated straight line. Thin lines represent the 95% confidence interval; arrows show the detection limit of a colorimetric assay and a chemiluminescent Griess assay (data according to the manufacturers).

Preparation of porcine coronary arteries. Coronary arteries of female pigs were prepared immediately after excision of the hearts and stored in oxygenated Tyrode solution. Special care was taken to preserve the endothelium during removal of fat and connective tissue. The artery was cut into rings of 0.8 cm (30-50 mg wet wt, 2-mm diameter), opened longitudinally, fixed on customized needles, and placed in an organ bath.

Cell culture of porcine aortic endothelial cells. Porcine aortic endothelial cells (PAEC) were harvested and cultured as described elsewhere (3). Briefly, porcine aorta were obtained fresh from a local abattoir, and the endothelial cells were isolated enzymatically and cultured in medium 199 with 10% fetal calf serum. On confluence (2-3 days), the cells were detached by treatment with trypsin, seeded onto sterile glass coverslips, and grown to subconfluence (70-80%). Only cells from passage 2 were used for the experiments. The purity of the PAEC culture was tested with different immunochemical markers.

Preparation of guinea pig hearts according to Langendorff. Hearts of adult guinea pigs were prepared as previously described (16). Briefly, the guinea pigs were heparinized (1,000 IU heparin/kg), and the hearts were excised. They were perfused at a constant perfusion pressure of 60 mmHg with an oxygenated Tyrode solution at 37°C. The hearts were beating spontaneously, and the left ventricular pressure was recorded isovolumetrically.

Measurement of the nitrite/NO production. The vessel pieces (30-50 mg wet wt) were fixed on a plastic rod with the endothelium outside (3-mm diameter) and incubated for 30 min in 500 µl of HEPES buffer (in mmol/l: 5 HEPES, 140 NaCl, 5 KCl, 2 CaCl₂, 1 MgCl₂, and 10 glucose, adjusted to a pH of 7.4) at 25°C (without stirring). The incubate was then treated with 50 µl of a NADPH-FAD mixture (1.5:0.03 mg/ml) and 1 min later with 15 µl of nitrate reductase solution (10 U/ml; Roche; final concentrations of 0.3 U/ml nitrate reductase). The resulting mixture was incubated at room temperature for another 30 min. The nitrite/NO concentration was estimated as described under Calibration of the NO electrode.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Calibration. A calibration with a KNO₂ solution (500 µl) in increasing concentrations showed a concentration-dependent linear increase in NO with a detection limit of 2 nmol/l in the 10-ml bath (Figs. 2 and 3). Increasing volumes of a fixed concentration of KNO₂ resulted in a linear signal too (2). A sharp increase in NO was detected after application of the KNO₂ solution, which slowly decreased down to basal levels, depending on the height of the signal (Fig. 2). The baseline stability did not change when several reductants were injected (up to 8 times). The detection limit using our method was 0.2 µmol/l nitrite in 500 µl and, therefore, was lower than the detection limit of a colorimetric assay, as well as a chemiluminescent Griess assay (detection limits according to the manufacturers, Roche and R&D) (Fig. 3).

Measurement of NO from native vessels. Isolated porcine coronary arteries released nitrate/NO in a time-dependent manner (Fig. 4) (n = 4). We standardized our experiments to a 30-min incubation of the vessels. Basal nitrite/NO levels (per ring) released by the constitutive NOS after 30 min of incubation were 350 ± 17 pmol · 10 mg¹ · 30 min¹ (n = 25) (Fig. 5). Additionally, we induced an enhanced NO release of the vessels by using carbachol as a conventional agonist. After a period of 30 min, carbachol (1 µmol/l) significantly elevated nitrite/NO levels (+50 ± 9%; n = 6; Fig. 5). The basal and the carbachol-induced NO formation could be blocked by the NOS inhibitor N^G-nitro-L-arginine methyl ester (0.1 mmol/l), which shows the specificity of our measurements (data not shown).

View larger version (10K): [in this window] [in a new window]   Fig. 4.   Segments (10 mg) of porcine coronary vessels showed a time-dependent basal release of nitrite/NO. Values are means ± SE; n = 4 segments.

View larger version (13K): [in this window] [in a new window]   Fig. 5.   Original recording shows the nitrite/NO peak elicited by 100 µl of the incubate solution in which the vessel pieces had been placed (30 min). This aliquot was added to 10 ml of the acidic iodide solution. After incubation with the agonist carbachol (1 µmol/l) under the same conditions, the nitrite/NO content increased significantly.

Measurement of NO from endothelial cell cultures. Cultured endothelial monolayers showed a basal NO release of 186 ± 17 pmol · 10 cm² · 30 min¹. We used shear stress to induce an increased NO formation, which resulted in stimulation of the NO release (658 ± 23 pmol · 10 cm² · 30 min¹; n = 3). To compare our method with conventional Griess assays, we performed measurements with a colorimetric assay (Roche) and a chemiluminescent assay (R&D). We were not able to detect nitrite in the supernatant of endothelial cell cultures with these methods; thus it seemed that the sensitivity in these systems was too low.

Measurement of NO in heart perfusate. The effluent of a Langendorff-perused guinea pig heart was analyzed to measure nitrite/NO. The electrode was sensitive enough to monitor basal NO release after 30 min of equilibration in the effluent. The basal NO concentration was 700 pmol/min at a flow of 11 ml/min.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

We could show that it is possible to measure basal and agonist-induced NO release of endothelial cells with an electrochemical electrode by converting nitrite back to NO. This method is very sensitive, with a detection limit of 2 nmol/l, which means that, in our system, a sample (500 µl) with a nitrite content of 0.2 µmol/l would be detected. It even provides the ability to measure NO in heart perfusate. The method is based on the conversion of nitrite to NO in acidic iodide solutions, which can be detected by the electrode. Nitrite is one of the two biological degradation products of NO. The other product is nitrate. The enzymatic conversion of nitrate to nitrite is necessary to measure the total amount of NO formed by the cells because the iodide solution is not strong enough to reduce nitrate to NO. Without a conversion, significantly lower levels of NO are detected (data not shown).

Because the endothelial cells are incubated for a fixed time period, we cannot monitor the actual NO release online at a defined time, which is possible with the electrode (3, 19). On the other hand, it is a big advantage to be able to measure basal NO release, which is hardly possible if the electrode is placed above the endothelial cells and NO is measured directly, because in this case the basal NO liberation is incorporated into the baseline and only a stimulated NO release can be detected. Additionally, the total NO amount formed would be obtained, even if the NO is inactivated by oxygen radicals, because, at the end, nitrite will be generated. Compared with the Griess reaction, the typical assay used to measure nitrite and nitrate, our detection limit is significantly lower (Fig. 3) (18). Moreover, the Griess reaction might interfere with thiols, proteins, and plasma constituents, whereas our methods seems not to be affected up to very high unphysiological concentrations (18). The chemiluminescent NO assay offers a detection limit of the same range but is significantly more expensive. Additionally, because nitrate and nitrite are stable compounds, one does not need to worry about the short half-life of NO, and it is possible to freeze biological samples and measure them later. Additionally, the volume that is needed to perform a measurement is low (50-500 µl), and the time to perform a measurement is short. To work with this highly sensitive electrode, it is necessary to ground the experimenter and the recording devices thoroughly, otherwise small changes in electric currents will interfere with the recording. The same refers to changes in pH, flow, and temperature.

In summary, we are presenting a highly sensitive method to measure NO in biological samples by conversion of nitrite with a commercial NO electrode.