Characterization of the Distribution of Nitrite in Human Blood.

Several groups have recently shown that blood can transport nitric oxide (NO) bioactivity so that NO can act as an endocrine signal transducer. The mechanisms of this process are uncertain. We have recently shown that nitrite ions can generate NO in the presence of deoxygenated hemoglobin (Cosby et al. Nature Medicine 2003). Thus, nitrite appears to be a major intravascular NO storage molecule and hemoglobin a nitrite reductase contributing to erythrocyte-dependent hypoxic vasodilation. Such a model requires a reaction of nitrite with intra-erythrocytic hemoglobin, however, published uptake rates of nitrite from plasma into erythrocytes are very slow and dynamic blood flow regulation requires more rapid responses. We therefore hypothesized that the red blood cell itself might be a reservoir for nitrite, with compartmentalization of nitrite and deoxyhemoglobin at the erythrocyte membrane; such a metabolon would allow for rapid nitrite reductase chemistry and NO export as hemoglobin deoxygenates. Nitrite was measured using reductive chemiluminescence in conjunction with a novel ferricyanide-based hemoglobin oxidation assay to rapidly preserve nitrite in the presence of hemoglobin following red cell lysis. We find that measurement of whole blood nitrite using ferricyanide-based hemoglobin oxidation to stabilize nitrite is a valid, sensitive, linear, and reproducible surrogate for RBC nitrite. Using the stabilization solution, nitrite in whole blood remained constant over 12 h at room temperature, whereas without it the half life of nitrite was 12 ± 4 min. Using this novel methodology, the nitrite levels measured in plasma, RBC, and calculated and measured in whole blood in normal volunteers were 121 ± 9, 620 ± 170, 245 ± 58, and 323 ± 44 nmol/L. The difference between measured and calculated blood nitrite levels reflects destruction of RBC nitrite by hemoglobin during sample processing. There was an artery-to-vein gradient for blood nitrite (A. vs V.: 169 ± 12 vs. 149 ± 11 nmol/L; p=0.03). Intraperitoneal injection of L-NAME (a potent NOS-inhibitor; 100 mg/kg) in mice caused a time dependent decrease of whole blood nitrite (0 h: 650 ± 29, 2h: 556 ± 26, 4h: 455 ± 11, 421 ± 19, 24 h: 551 ± 30 nmol/L; p=0.002). Differences of blood nitrite across the human forearm circulation correlated significantly (r²=0.94) with forearm bloodflow after infusion of acetylcholine (7.5 ug/min; n=5), consistent with nitrite formation across the vasculature with eNOS activation. We conclude that in humans, the RBC nitrite concentration is 5 times greater than plasma nitrite concentrations; A-V gradients in blood nitrite are consistent with nitrite utilization as a source of bioactive NO in the microvasculature and levels of blood nitrite are controlled by endogenous NO production. Levels of nitrite in the RBC may function as a locally available source of nitrite for deoxyhemoglobin-dependent nitrite reduction to NO, thus allowing for more efficient nitrite-deoxyhemoglobin mediated hypoxic vasodilation. Modulation of intraerythrocytic nitrite may be of therapeutic interest in diseases with a lack of bioavailable NO, such as sickle cell disease.