A sensitive LC-MS/MS method to quantitate nitrite in human plasma

Purpose: Numerous methods are available to quantitate nitrite in biological samples, including fluorescence, chemi-luminescence, capillary electrophoresis, colorimetric and ultraviolet (UV) spectrophotometry and gas chromatography-mass spectrometry (GC-MS). However, limitations associated with these techniques (e.g., lack of specificity, low sensitivity, etc.) prompted us to develop a LC-MS/MS method to quantitate nitrite in biological samples. Methods: We compared two derivatization methods which are used to convert the endogenous nitrite to a more stable organic compound that is able to be analyzed by LC-MS/MS. The first S-nitrosoglutathione (GSNO)-based method was performed by selected-reaction monitoring specific mass transition m/z 337 ([M+H]+)→m/z 307 ([M+H-14NO]+•) for GS14NO and GS15NO as internal standard (IS) by reaction of nitrite and glutathione (GSH). The second 2,3-naphthotriazole (NAT)-based method was performed by measuring NAT m/z 170 ([M+H]+)→m/z 115 ([C9H7+] and NAT-N15 as IS following the reaction of 2,3-diaminonaphthalene (DAN) with nitrite to produce NAT. We found that the latter NAT-based derivatization method is reproducible, stable, and 100 times more sensitive than the GSNO method. In addition, we validated the NAT-based method for precision and accuracy, recovery, stability and derivatization time and temperature. We then utilized the optimized method to quantitate nitrite in human plasma. Results: We found that nitrite stability in human plasma filtrate can be affected by freeze-thaw cycles and storage temperature which were evidenced by a decrease in nitrite levels over 50% after 24 hours at -20°C. However, the derivatized and extracted samples were stable for 24 hours at room temperature. When derivatization is fully complete and performed at 370C and 45 minute incubation, the recovery is 108%. The precision and accuracy are 15%, and 106%, respectively. The linearity range is 0.13-16 µM with linear regression, 1/x weighing correlation coefficients = (r), R2>0.9990. The same amount of labeled isotope as IS was added to each sample to keep track of signal fluctuation between study samples, In doing so, it was confirmed that the quantitation of endogenous nitrite was indeed affected by individual complex sample content. And the use of isotope-labeled nitrite reduced the impact of high background levels of nitrite in biological matrices, which other methods cannot achieve. Conclusion: We have developed and validated a reproducible and highly sensitive LC-MS/MS method for quantitation of nitrite in biological samples. In addition, we anticipate that our method can be utilized in other fluidic human biological samples.