Stable-isotope dilution LC–MS/MS measurement of nitrite in human plasma after its conversion to S-nitrosoglutathione

• Circulating nitrite is a biomarker of endothelium-derived nitric oxide (NO).
• Conversion of nitrite to S-nitrosoglutathione (GSNO) greatly improves its LC–MS/MS analysis.
• Matrix-effects are entirely overcome by the internal standard 15N-nitrite.
• LC–MS/MS and GC–MS methods from our group provide similar plasma nitrite concentrations.
• Healthy fasting humans have plasma nitrite concentrations of the order of 1 μM.

A specific, sensitive and fast LC–MS/MS method with positive electrospray ionization for the quantitative determination of nitrite in human plasma is reported. Added [15N]nitrite served as the internal standard (IS). Endogenous nitrite and IS were converted to their S-nitrosoglutathione (GSNO) derivatives, i.e., GS14NO and GS15NO, respectively, by using excess glutathione (GSH) and HCl. For plasmatic nitrite, fresh plasma (0.5 mL) was spiked with the IS (1000 nM) and ultrafiltered (cut-off 10 kDa). Ultrafiltrate aliquots (100 μL) were treated with aqueous GSH at a final concentration of 1 mM and 1 μL of 5 M HCl for 5 min. After final sample dilution (1:1, v/v) with acetonitrile–water (70:30, v/v), 2 μL aliquots were injected via a thermostated (4 °C) autosampler. The mobile phase was acetonitrile–water (70:30, v/v), contained 20 mM ammonium formate, had a pH value of 7, and was pumped isocratically at 0.5 mL/min. A Nucleoshell column was used for LC separation. The retention time of GSNO was about 0.8 min and the total analysis time 5 min. Quantification was performed by selected-reaction monitoring the specific mass transition m/z 337 ([M+H]+) → m/z 307 ([M+H−14NO]+) for GS14NO (i.e., for endogenous nitrite) and m/z 338([M+H]+) → m/z 307 ([M+H−15NO]+) for GS15NO (i.e., for the IS). The method was thoroughly validated in human plasma (range, 0–2000 nM). The LOD and LOQ values of the LC–MS/MS method were determined to be 1 fmol and 5 nM [15N]nitrite, respectively. The relative matrix-effect of about 21% was outweighed entirely by the IS. In freshly prepared plasma samples from heparinized blood donated by three healthy subjects, nitrite concentration was determined by LC–MS/MS to be 516, 199 and 369 nM. These concentrations were confirmed by using a previously reported GC–MS method and agree with those measured previously by HPLC–UV (334 nm) after nitrite conversion to S-nitroso-N-acetylcysteine (SNAC) by N-acetylcysteine (NAC). Measurement of nitrite by LC–MS/MS as GSNO is about 1000 times more sensitive than by HPLC–UV as SNAC. The applicability of the method to microdialysate, urine, and saliva samples from humans was demonstrated. The agreement of two orthogonal MS-based methods indicates that the concentration of nitrite in freshly prepared, non-frozen plasma from heparinized blood of fasted healthy humans is of the order of 400 nM.