antioxidant protection. In addition, only one study (Levine et al., 1996a) with seven apparently healthy males reported plasma, neutrophil, and urinary ascorbate concentrations during vitamin C depletion and repletion to steady state. Thus, there are wide uncertainties in the data utilized to estimate the vitamin C requirements. However, in the absence of other data, maximal neutrophil concentration with minimal urinary loss appears to be the best biomarker at the present time. It must be emphasized that research is urgently needed to explore the use of other biomarkers to assess vitamin C requirements.
The evidence summarized in the preceding sections indicates that vitamin C functions in vivo to scavenge reactive oxidants in activated leukocytes, lung, and gastric mucosa, and to protect against lipid peroxidation. Therefore, the determination of an EAR for vitamin C is based on an amount estimated to provide antioxidant protection. Evidence summarized in the earlier section “Antioxidant Functions in Leukocytes” indicates that the vitamin's antioxidant function in leukocytes, which includes neutrophils, lymphocytes, and monocytes, is especially important. In addition, studies with guinea pigs and monkeys show that the concentration of ascorbate in the leukocytes more accurately reflects liver and body pool ascorbate than does the concentration in plasma or erythrocytes (Omaye et al., 1987). The vitamin is transported into leukocytes by an energy-dependent transport system that concentrates the vitamin some 25, 40, and 80 times higher than plasma levels in neutrophils, platelets, and lymphocytes, respectively (Evans et al., 1982; Jacob et al., 1992; Levine et al., 1996a).
The cells actively concentrate the vitamin, which serves as a cellular reservoir of reducing capacity and scavenges damaging phagocyte-derived oxidants such as superoxide and myeloperoxidase-derived hypochlorus acid (HOCl) and reactive nitrogen species (RNS). In both the cell-free and the activated neutrophil systems described earlier, the protection of α-1-antiprotease against inactivation by HOCl (Halliwell et al., 1987) and the inhibition of super-oxide production (Anderson and Lukey, 1987) were directly proportional to ascorbate concentrations within the normal range of plasma ascorbate concentrations (22 to 85 µmol/L [0.4 to 1.5 mg/dL]). Data plotted in Figure 5-2 show that superoxide production by activated neutrophils was inhibited 29, 44, 52, and 55 percent by extracellular ascorbate concentrations of 28, 57, 114, and 284 µmol/