| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
a author for correspondence: fax 34-1-3908001, e-mail jarenas{at}h12o.es
The nitric oxide radical (NO·) plays an important role as a physiological messenger (1). NO is formed from L-arginine (2) by NO synthase (NOS; EC 1.14.13.39), which exists in several isoforms (3). Constitutive calcium-dependent isoforms (cNOS) modulate the control of vascular tone in endothelial cells or the neurotransmission in neurons, whereas inducible calcium-independent isoforms (iNOS) are located in macrophages, chondrocytes, and hepatocytes and are induced by cytokines and endotoxin (4)(5). Pathological conditions associated with increased release of cytokines and endotoxin, e.g., inflammation or sepsis (6), can therefore increase NO production.
NO is a very unstable, short half-life gas that breaks down rapidly into the stable products nitrate and nitrite (7). Upon coming into the bloodstream, nitrite reacts immediately with oxyhemoglobin to form methemoglobin. Consequently, most NO produced is detected in serum as the remaining product, nitrate (8). Recently, several reports focused on methods to measure nitrate concentrations in biological fluids (9)(10)(11). One of the most commonly used methods is based on the reduction of nitrate to nitrite by cadmium or nitrate reductase, the nitrite produced being determined by Griess reaction (9)(12)(13). Other methods for monitoring NO production are based on chemiluminescence (11)(14), enzymatic assay with an internal standard (10), or chromatographic procedures (8) for nitrate (15)—all of which are time-consuming for routine application in clinical chemistry laboratories.
We describe a rapid semiautomated method based on the Griess reaction, involving a shortened incubation period of nitrate with cadmium. The method is applicable to several types of biological fluids.
Serum or plasma samples from healthy individuals after a 12-h fast were obtained in accordance with the Medical Ethical Committee of our hospital. Samples were stored at -20 °C and were stable for at least 6 months. The method was adapted to a Hitachi 717 Analyzer (Boehringer Mannheim). Cadmium granules were from Fluka. Copper sulfate, zinc sulfate, sulfanilic acid, glycine, ascorbic acid, sodium nitrite, and potassium nitrate were from Merck; sodium hydroxide, phosphoric acid, and hydrochloric acid from Panreac; and naphthylethylenediamine and lithium nitrate from Carlo Erba.
We deproteinized 300 µL of serum or plasma by adding 250 µL of 75 mmol/L ZnSO4 solution, stirring, and centrifuging at 10 000g for at least 1 min at room temperature, after which 350 µL of 55 mmol/L NaOH was added. Again, the solution was stirred and centrifuged at 10 000g for 3 min and the supernatant was recovered (the supernatant must be free of turbidity for measuring nitrate concentrations). We diluted 750 µL of supernatant with 250 µL of glycine buffer (45 g/L, pH 9.7).
For assay of urines, 100 µL of urine was diluted with 900 µL of the glycine buffer; for cerebrospinal fluid (CSF), 750 µL was mixed with 250 µL of the glycine buffer.
Cadmium granules (2–2.5 g) were rinsed three times with deionized distilled water and swirled in a 5 mmol/L CuSO4 solution in glycine-NaOH buffer (15 g/L, pH 9.7) for 5 min. The copper-coated granules are to be used within 10 min. After use, the granules are rinsed and stored in 100 mmol/L H2SO4 solution; they can be regenerated by repeating these steps.
Calibrators at various concentrations were prepared by diluting stock 20 mmol/L solutions of NaNO2, KNO3, or LiNO3 with deionized distilled water. The nitrate calibrators were diluted with glycine buffer just as the serum samples were. Calibration curves were made over a linear range of nitrate between 0 and 100 µmol/L.
Freshly activated cadmium granules (2–2.5 g) were added to 1 mL of pretreated deproteinized serum, urine, CSF, or calibrator. After continuous stirring for 10 min, the samples were transferred to appropriately labeled tubes for nitrite determination.
Reagent 1 consisted of 50 mg of N-naphthylethylenediamine
dissolved in 250 mL of distilled water. Reagent 2 was prepared by
dissolving 5 g of sulfanilic acid in 500 mL of 3 mol/L HCl. Both
solutions are stable for at least a year at 4 °C. Results were
expressed as the mean of duplicate determinations. New calibration
curves were calculated for each batch of samples. The instrument
settings for the method, shown in Table 1
, were used with 20 µL of sample, 75 µL of reagent 1, and 80
µL of reagent 2. Final concentrations were 1.47 mmol/L (0.43 g/L)
N-naphthylethylenediamine, 26.4 mmol/L (4.57 g/L) sulfanilic
acid, and 1.37 mmol/L HCl.
|
The reduction of nitrate at concentrations of 10, 30, and 50 µmol/L
was complete within 10 min at room temperature (Fig. 1
). To confirm the completion of the reduction of nitrate to
nitrite, we added nitrite, 15 µmol/L, to the three nitrate solutions.
Before reduction, nitrite concentrations were 15.5, 15.1, and 15.8
µmol/L, respectively. After 10 min of reduction, concentrations were
25.8, 45.7, and 66 µmol/L, respectively, indicating that the
reduction was complete.
|
The reaction was linear with concentrations of nitrate calibrator up to
600 µmol/L: y = 0.0019x 0.0022
(r = 0.9999, Sy
x =
0.001). The detection limit of the assay, i.e., the concentration
corresponding to the mean ± 3 SD (n = 20) for the 0 µmol/L
(nitrate-free) calibrator solution was 0.8 µmol/L.
Analytical recovery of nitrate (0, 25, 50, 75, and 100 µmol/L) added to plasma samples from apparently healthy subjects was 98.8–102.6% (mean 100.4%, n = 8) over the concentration range tested (0–150 µmol/L)
Within-run precision was calculated by 20 replicate determinations of two serum pools with normal and high nitrate concentrations. Between-day precision (n = 20 days) was determined from replicate measurements of each pool (between each experiment serum pools were stored at -20 °C). Within-run and between-day imprecisions were 4.2–6.2% and 5.7–8.5%, respectively.
Solutions of nitrate from 10 to 100 µmol/L were aliquoted and stored at -20 °C. Aliquots were assayed for nitrate before and after storage every day for as long as 45 days. Between-day imprecisions were <10% over the concentration range tested. We found no correlation between the nitrate concentration and the time of storage. These data confirm the stability of the nitrate solutions.
For interference studies, we used nitrate calibration curves made with hyperbilirubinemic, hemolytic, and hypertrigliceridemic plasma samples. Recovery of nitrate in these plasma samples after deproteinization was 98–104% (mean 101.1%, n = 4). Because ascorbate and phosphate inhibited the reduction of nitrate by 20% and 34%, respectively, we tested the possibility of interference by phosphate or ascorbate after deproteinization. Nitrate calibration curves were performed with otherwise normal plasma samples to which had been added 10 mmol/L phosphate or 2 mmol/L ascorbate. Recovery of nitrate from deproteinized phosphate- or ascorbate-supplemented plasma samples was similar to that in the nonsupplemented plasma samples: 98.6–102.4% (mean 100.3%, n = 4).
We used this method to measure the nitrate concentration in sera from 40 healthy subjects who had fasted for 12 h, 25 men and 15 women, ages 18–60 years (mean ± SD 35 ± 13 years), and in CSF from 38 patients who had undergone lumbar puncture to rule out the diagnosis of subarachnoid hemorrhage, pseudotumor cerebri, or other indications in the usual neurological survey, 23 women and 15 men, ages 35–80 years (mean ± SD 60.6 ± 15.4 years), who subsequently were proved to be free of disease. The mean nitrate concentrations were 27.3 µmol/L (median 24.8 µmol/L; range 5.5–58.2 µmol/L) for serum and 5.9 µmol/L (median 4.8 µmol/L; range 1.5–16.1 µmol/L) for CSF.
In methods that rely on reduction of nitrate to nitrite by a metal or nitrate reductase, with subsequent determination of nitrite by diazotization (9)(12)(13), deproteinization is essential to eliminate artifactually high results in most plasma or serum samples (9)(13). In the kinetic cadmium assay, the conversion of nitrate to nitrite can be best achieved with a combination of a reducing agent and a catalyst (13). The main drawbacks of this method are that some metals carry the reduction beyond nitrite, that some constituents of biological material such as ascorbate or phosphate interfere, and that the addition of an internal standard is needed to overcome the effects of inhibitors (13). In their nitrate reductase-based reduction method, however, Moshage et al. (9) documented near-quantitative and reproducible recoveries for nitrate in normal, hemolytic, or hyperbilirubinemic plasma samples. Consistent with their findings, our method shows similar high (98–104%) recovery values for nitrate in normal, hemolytic, hyperbilirubinemic, or hypertrigliceridemic plasma samples after deproteinization. Moreover, we found that deproteinization removes the interference of ascorbate and phosphate. Our data suggest that substances in plasma/serum neither interfere with nor enhance the quantitative reduction of nitrate and development of Griess chromophore.
In this method, the reduction of nitrate to nitrite is complete within 10 min, much shorter than the 90 min for the kinetic cadmium reduction method (13), 45 min for the enzymatic one-step assay (10), and 20 min for the enzymatic method of Moshage et al. (9). Green et al. (12), by using a high-pressure reduction column coupled with a continuous-flow analyzer, reported a shorter reduction time (within 1 min), and an analysis rate of 30 samples per hour. Although our reduction time is longer, the analysis rate, considering the overall time spent on deproteinization, reduction, and measurement in a Hitachi 717 analyzer, amounts to >150 samples per hour. Recently, Yang et al. (11) described a very sensitive method for chemiluminescent detection of nitrate, but estimates for the analysis rate were 15–30 samples per hour. Moreover, the automated device by Green et al. (12) was specifically designed for analyzing nitrate concentrations and is not available to the majority of clinical chemistry laboratories. By contrast, Hitachi analyzers are widely used and allow simultaneous performance of other analyses. An additional advantage of this method is its low cost, because the cadmium may be regenerated and the reagents are cheaper than those used in enzymatic methods (nitrate reductase or ß-NADPH).
The semiautomated method described here responds in a quantitative linear fashion to nitrate within the range of 0.8 to 600 µmol/L, whereas the other methods are linear only up to 200–300 µmol/L. The detection limits and the within-run and between-day imprecisions for this method are similar to those reported in the other methods (9)(10)(12)(13). Values reported for serum are similar to those documented by Moshage et al. (9).
We conclude that this semiautomated procedure provides an easy, less time-consuming, inexpensive, and reliable method of analysis for nitrate concentration in biological fluids, and is therefore suitable for routine performance in clinical chemistry laboratories.
Acknowledgments
We are indebted to José Antonio Molina for providing CSF samples. J.A.N.-G. is supported by a grant (96/5567) from FIS, Ministry of Health, Spain.
Footnotes
Serv. de Bioquím. Clín., Hosp. 12 de Octubre, Carretera de Andalucía, km 5400, 28041 Madrid, Spain
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  M. A. Hladunewich, G. C. Derby, R. A. Lafayette, K. L. Blouch, M. L. Druzin, and B. D. Myers Effect of l-Arginine Therapy on the Glomerular Injury of Preeclampsia: A Randomized Controlled Trial. Obstet. Gynecol., April 1, 2006; 107(4): 886 - 895. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), 30 µmol/L (
]) and 10 µmol/L (
).