V. ESTIMATION 253 



less could he used satisfactorily only in the presence of sodium cyanide. 

 jMapson and Ingram, applying this modified E. coli method, could not 

 obtain complete recover^' of dohydroascorhic acid, owing to a small con- 

 current conversion to diketogulonic acid.'"^ They showed, however, that 

 tliis can be allowed for by calculation and is minimized by the use of so- 

 dium azide in place of the cyanide. These workers traced the failure of the 

 E. coli method in vegetable juices to the formation of nitrous acid from 

 nitrates by the nitratase enzyme system of the bacteria and were able to 

 overcome this interference by using amidosulfonic acid. D-Isoascorbic 

 acid, reductic, and hydroxytetronic acids, if present, would cause inter- 

 ference in this method. 



c. Reducing Substances Not Ascorbic Acid 



The 2 , 6-dichlorophenolmdophenol method, if carried out with the pre- 

 cautions already described, is usually not subject to interference by reducing 

 substances other than ascorbic acid. Glutathione, for instance, and some 

 phenolic substances are prevented from interfering by the fact that the 

 titration is carried out in strongly acid solution .2^- ^^ Sulfur dioxide, fer- 

 rous salts, and tannins are, however, among the reducing substances for 

 which suitable modifications may have to be made in the determination 

 of ascorbic acid by the indophenol method. Two approaches to this problem 

 have been suggested. One is to titrate the sample extract before and after 

 the vitamin has been destroyed by ascorbic acid oxidase.'"^- '"' This prin- 

 ciple has received relatively little application, mainly because the enzyme 

 is not fully specific for ascorbic acid, and attention has been concentrated 

 on the alternative procedure, which is adjustment of conditions of assay so 

 that substances which may normally reduce the dye are inhibited or allowed 

 for by difference. 



(1) Tannins, Cysteine, Thiosulfate, and Sulfhydryl Compounds. Van 

 Eekelen and Emmerie removed tannins, cysteine, thiosulfate, and sulf- 

 hydryl compounds ^vith mercuric acetate, followed by precipitation of the 

 excess mercury salt with hydrogen sulfide which, it was claimed, also 

 allowed the estimation of dehydroascorbic acid.^- This procedure was also 

 found to have the added advantage of removing color likely to interfere 

 in the determination of the end point when titratmg plant extracts. 



In the urine of subjects on low intakes of vitamin C, a considerable pro- 

 portion of the indophenol titer may be due to thiosulfate rather than to 

 ascorbic acid itself. With higher intakes of the vitamin, however, the indo- 

 phenol-reducing substance is mostly ascorbic acid. This consideration does 



»<" L. W. Mapson and M. Ingram, Biochem. J. 48, 551 (1951). 



'«8 iM. Srinivasan, Biochem. J. 31, 1524 (1937). 



los G. T. Meiklejohn and C. P. Stewart, Biochem. J. 35, 761 (1941). 



