288 ZACHARIAS DISCHE 



tration shows a faint blue color and an absorption maximum at 595 m/x. 

 The extinction coefficient of this nucleotide, however, is less than one- 

 sixtieth of that of DNA, while cy tidy lie acid shows no significant color at 

 all. The reaction of DNA, therefore, represents the reaction of its purine 

 nucleotides only. (Brady and McEvoy-Bowe^°'^ reported that it is possible 

 also to determine the deoxypentose of the pyrimidine nucleotides of DNA 

 with the diphenylamine reaction after preceding bromination which makes 

 the pyrimidine-sugar link more easily hydrolyzable.) It is obvious that the 

 splitting of the glycosidic linkage of the sugar is necessary for the reaction. 

 On the other hand, substitution in position 5 and 3 of the sugar by phos- 

 phoric acid apparently does not significantly affect the intensity of the 

 reaction. The blue color with the absorption maximum at 595 m/x is not 

 produced by any of the normal sugars or hexals, hexuronic acid, aliphatic 

 and aromatic aldehydes, glycolic aldehyde, trioses, aldo- and ketotetrose, 

 nor furfuraldehyde.^ Arabinal, on the other hand, produces the blue color 

 with the identical absorption spectrum,^ and the extinction coefficient, 

 as measured in the Spekker absorptiometer, is about one-third higher than 

 that of 2-deoxyribose. Finally, AUerton et al}^ recently repoited that 3- 

 deoxyxylose, as well as 2,3-deoxyxylose, produces a blue color with the 

 diphenylamine reagent. The extinction coefficient at the maximum of 

 absorption, however, is for both substances about one-thirtieth of that of 

 2-deoxyribose, and the color develops much more slowly. No data about 

 the absorption maximum and the form of the absorption curve were 

 reported for these two deoxysugars, and it is therefore not established 

 whether the blue color produced is identical with that derived from 2- 

 deoxypentoses. 



Mechanism of the reaction: — The mechanism of the diphenylamine re- 

 action was extensively studied by Deriaz et al}'^ It was found that co-hy- 

 droxylaevulic aldehyde, but not laevulic acid, gives with the reagent a 

 blue color with an absorption spectrum identical or very similar to that 

 derived from 2-deoxyribose. They furthermore found that furfuryl alcohol 

 also gives the same reaction. Further studies showed that it is possible to 

 obtain co-hydroxylaevulic aldehyde by heating with methanolic HCl in lar- 

 ger quantities from furfural alcohol and arabinal than from 2-deoxyribose. 

 In agreement with that, the molar extinction coefficient of the first two 

 compounds in the diphenylamine reaction is considerably higher than of 

 the last one. The formation of co-hydroxylsevulic acid from 2-deoxyribose 

 and arabinal involves an inner dismutation between carbon 3 and 4 in which 

 the hydroxy 1 of carbon 3 is shifted to carbon 4, while in the case of furfuryl 

 alcohol this shift is brought about simply by hydrolysis of the oxygen ring. 



i"* T. G. Brady and E. McEvoy-Bowe, Nature 168, 299 (1951). 



" R. AUerton, E. G. Overend, and M. Stacey, J. Chem. Soc. 1952, 213. 



