A8YMMETR Y OF PRO TO PLA^M AND CANCER 131 



1( + ) giutainic acid from malignant cells. Kogl and Erx- 

 leben (1939) inmiodiately pointed out that their oppo- 

 nents did not pay sufficient attention to the different sol- 

 ubilities of the optical isomers. Racemic dl-glutamic 

 acid, in the form of both chlorhydrate and barium salt, is 

 two times more soluble than 1-glutamic acid. The pure nat- 

 ural isomer consequently cristallizes first and, if the crys- 

 tallization is not complete, the racemic isomer will be left 

 in the mother liquid. 



Lipmann and his collaborators (1940) also opposed 

 their findings to Kogl's and Erxleben's data. On ac- 

 count of some difficulties in the ordinary isolation pro- 

 cedures, Lipmann attempted to determine the total d- 

 amino acid content of the human tumors and of normal 

 tissues by means of d-amino acid oxidase with the aid of 

 the Krebs enzyme. He found 1.85% of d-amino acids in 

 hydrolvzates of normal tissues and 1.84% of d-isomers in 

 those of cancer tissues, that is, practically the same value 

 in the two cases. However, Lipmann himself admits that 

 the accuracy of his method is not great. Moreover, since 

 each hydrolysis inevitably leads to a partial racemization 

 of such labile amino acids as serine and proline, the deter- 

 mination of total d-amino acid content loses some of its 

 significance. 



On the other hand Kogl's data have been confirmed by 

 Arnow and Opsahl (1939). The glutamic acid which they 

 isolated from normal tissues had an optical rotation of 

 a — --h 31.0^, and that isolated from malignant tissue had 

 an optical rotation of a = H- 5.5°. 



If one assumes that Kogl's data are correct, it is, how- 

 ever, not clear whether the unusual optical isomer of glu- 

 tamic acid pre-exists in the cancer cell, or whether the par- 

 tial racemization observed is of a factitious nature. If 

 glutamic acid, in the protein molecule of cancer cells, en- 

 ters into some special labile compound different from that 

 in which it exists in usual protein molecules, it is con- 

 ceivable that hydrolysis could lead to a partial racemiza- 

 tion in the cancer cell but not in the normal cell. This 



