Improved methods to measure end products of nitric oxide in biological fluids: nitrite, nitrate, and S-nitrosothiols.

The aim of this study was to compare and improve standard methods to determine nitrite (NO2-), nitrate (NO3-) and S-nitrosothiol (RSNO) levels in cell culture supernatants, sera, and urine. We modified the conventional Griess reaction by replacing sulfanilamide with dapsone (4,4'-diamino-diphenylsulfone) and compared the NO2- levels in our study samples with a commercially available NO2- assay kit. Our modification, along with ultrafiltration of the samples, resulted in an enhanced sensitivity to measure NO2- down to 0.2 microM. The detection limit was further improved to 0.02 microM when NO2- was identified by the fluorochrome 2,3-diaminonaphthalene (DAN). To measure the stable end product NO3- by the Griess reaction or the DAN method, this anion must be reduced to NO2-. We compared the capacity of bacterial nitrate reductase with the reducing metal cadmium to convert NO3- to NO2-. After reduction, NO2- levels were determined either by the DAN method or by our modified Griess reaction. We found that there was a high correlation (r2 = 0.998) in total NO2- concentrations in the study samples using both methods for reducing NO3- to NO2-. The simultaneous determination of NO2- and NO3- was achieved by using anion-exchange chromatography (HPLC; Polyspher IC AN-1 column). The detection limit of this assay for each anion is 0.5 microM, and it can be applied equally well to sera, urine, and culture media. We also adapted the DAN method to determine RSNO levels in our study samples. Using this approach, we were able to measure RSNO levels down to 0.15 microM. As result we discovered that RSNO levels were markedly increased in urine from septic patients and in supernatants from cytokine-stimulated human tumor cell lines. L-Citrulline, a coproduct of NO biosynthesis, was measured using a colorimetric assay with a sensitivity limit of 3.0 microM. Increased L-citrulline levels in media from cultured cells, but not in sera or urine, correlated with increased NO production. Although all methods studied were suitable for quantifying end products of NO in biological fluids and media, the use of bacterial reductase and the modified Griess reaction proved successful to provide the greatest sensitivity and linear range for routine measurements of NO2- and NO3-.