Does vitamin C have a pro-oxidant effect?

Abstract

The conclusions of the Scientific Correspondence by Podmore et al.¹ are undermined by some unaddressed scientific issues. Podmore et al. gave vitamin C supplements of 500 milligrams to healthy volunteers, and measured levels of vitamin C in plasma and of oxidative DNA adducts in lymphocytes before, during and after supplementation.

Main

Because the reported changes in levels of 8-oxoguanine and 8-oxoadenine occurred in lymphocytes, the relevant vitamin C concentrations are not those in plasma but those in lymphocytes; however, these data were not presented. Because millimolar concentrations of ascorbate are found in lymphocytes, measurements of vitamin C are practical using as few as 0.25×10⁶ cells and a sensitive electrochemical high-performance liquid chromatography assay^2,3,4. Available cell numbers should not have been a limiting problem because few cells are needed for the measurement compared with the number that can be isolated from peripheral blood²,³,⁵.

Although the data on plasma vitamin C concentrations presented by Podmore et al.¹ were expressed as percentage changes only, true intracellular vitamin C concentrations can still be calculated. Because the vitamin C supplement produced a 60% increase in plasma concentrations of vitamin C, and because 500 mg is a saturating dose of vitamin C, the initial plasma vitamin C concentration was roughly 50 μM (refs 5,6). However, at this concentration lymphocytes are already saturated, with an intracellular ascorbate concentration of approximately 3 mM (refs 5,6). Increasing the extracellular plasma concentration of ascorbate above 50 μM will not affect the intracellular concentration further.

If the reported changes in oxidative DNA adducts could not be due to changes in intracellular lymphocyte vitamin C concentrations, what is responsible? The concentrations of 8-oxoguanine reported by Podmore et al. are 25-120 times more than those reported by others^7,8,9. These measurements are notoriously difficult to make and are easily increased artefactually by oxidation⁹. The possibility of oxidation is increased further if mononuclear cells rather than pure lymphocytes were isolated, because activated contaminating monocytes can produce superoxide and other oxidants. The purity of the isolated lymphocytes was not indicated by Podmore et al. In addition, without proper precautions ascorbate can act as a pro-oxidant. By increasing plasma ascorbate levels with supplements, the authors may have increased the probability of an oxidation artefact occurring.

Because Podmore et al. used an experimental design that did not take into account the fact that cells become saturated with vitamin C at lower doses than does plasma, an inappropriate choice of tissue sampled for vitamin C, unclear lymphocyte-isolation procedures, and assays that may be biased by oxidation artefacts, we believe that their conclusions are not justified by the data.