296 ZACHARIAS DISCHE 



times that of DNA, so that the reaction could be used to determine DNA 

 in nucleoproteins. 



c. Reaction of DNA with Tryptophan and Perchloric Acid 



This reaction, described by Cohen in 1944/^ represents a special case of 

 a more general reaction of carbohydrates with tryptophan in acid solution 

 described by Thomas.^* 



Procedure: — To 1 cc. of a solution containing 100 to 500 fig. of DNA are added 0.2 

 cc. of 1% tryptophan solution and 1.2 cc. of 60% perchloric acid. The solution is 

 heated in a boiling water bath for 10 minutes, cooled in tap water, and read in the 

 spectrophotometer 5 minutes later. A purple color appears. The absorption spectrum 

 has so far only been determined with a Klett photoelectric colorimeter. The maximum 

 is obtained with a filter having a transmission range between 485 and 550 m/x- Accord- 

 ing to our own observation, an absorption maximum at 500 m// is obtained with the 

 Beckman spectrophotometer. 



Specificity of the reaction: — Deoxyguanosine was found to react with a 

 molar extinction coefficient identical with that of DNA. This suggests that 

 the pyrimidine nucleotides of DNA react in this reaction to the same extent 

 as the purine nucleotides. This conclusion, however, should be further 

 checked by direct determinations on isolated pyrimidine nucleotides as 

 well as breakdown products in which the purine sugar is destroyed. Aldo- 

 pentoses, aldo- and ketohexoses, trioses, ascorbic acid, furfural, acetalde- 

 hyde, benzaldehyde, glyceraldehyde, and palmitaldehyde give color reac- 

 tions with the reagent which, however, differ in their absorption maximum 

 when determined with the Klett photoelectric colorimeter. Rhamnose 

 apparently showed the same maximum as DNA, while the intensity of 

 the color corresponded only to 4 % of that of an equal amount of DNA. 



Mechanism of the reaction: — Cohen^® suggested that the colored product 

 is a Schiff base produced by the condensation of 2-deoxyribose with the 

 nitrogen of the pyrrol ring. This suggestion is based on the assumption that 

 2-deoxyribose, differing from other carbohydrates with a hydroxyl at carbon 

 2, would not be able to participate in the Amadori reaction and thus an 

 intermediate containing conjugated double bonds would accumulate. 

 Observations in our laboratory have shown that furfuryl alcohol reacts 

 with Cohen's reagent producing a pink color with an absorption maximum 

 at 500 van. The absorption curve is like that of arabinal and not identical 

 with that of DNA, and the molar extinction coefficient at 500 mn 40% 

 lower than that for DNA. This and the reactivity of hydroxy aldehydes 

 with the reagent suggest that here, as in the diphenylamine reaction, not 

 the aldehydic form of 2-deoxypentose, but an aldehydic intermediate which 

 can be formed from the sugar as well as from furfuryl alcohol, is responsible 

 for the color reaction. 



" P. Thomas, Z. Physiol. Chem. 199, 10 (1931). 



