The chemical identification of ascorbic acid in urine.

DURING the past 3 years several methods have been proposed, and widely used, for the estimation of ascorbic acid in urine. They all depend on the reducing properties of ascorbic acid and are all non-specific. By careful limitation of conditions it is possible, with these methods, to exclude interference by a number of substances known to be present in urine and to reduce the reagents employed. There is, however, no proof that any of the methods is really estimating ascorbic acid, and the lack ofunanimity among the various protagonists suggests that none of the methods, even with carefully defined conditions, is really specific. Indeed there is no complete proof that ascorbic acid is normally present in urine, though indirect evidence obtained by analysis of urine from normal and scorbutic subjects suggests that it is. However, attempts at biological identification have failed, though the failures may reasonably be attributed to toxicity on the part of the urine [Ahmad, 1936; Wieters, 1935]. More serious was the failure of Hinsberg & Ammon [1936] to isolate ascorbic acid from urine. They attempted to obtain it in the form of the 2:4-dinitrophenylhydrazine derivative of dehydroascorbic acid, and concluded, from their failure with normal urine and their success with added ascorbic acid, that urine could not contain, at most, more than 0 3 mg. ascorbic acid per 100 ml., i.e. less than a third of the amount indicated by titration with dichlorophenolindophenol. Some time ago two of us [Scarborough & Stewart, 1937] found that the indophenol-reducing substance in normal urine was partly, though not completely, oxidized aerobically in the presence of the cauliflower hexoxidase described by Szent Gy6rgyi [1928] and by Hopkins & Morgan [1936]. This rather strengthened the view that ascorbic acid is excreted normally, but did not amount to proof, since the enzyme has not been shown to be completely specific. It was therefore decided to attempt the actual isolation of a derivative of ascorbic acid from normal urine, obtained from persons receiving an ordinary mixed diet without ascorbic acid supplements. A preliminary note on this attempt, which was successful, has appeared [Stewart et al. 1937]; while this note was in the press, a similar success was announced by Meuwissen & Noyons [1937] who used a very similar method. As, however, the identification of the isolated substance by the present authors seems to be more rigorous than that of Meuwissen & Noyons, and since another point of interest arose during the work, it seems desirable to give a fuller account of the work than is possible in the columns of Nature.