THE SYNTHESIS OF THE PROTEINS 51 



mode of combination other than a polypeptide linking is present in 

 arginine, and other modes are still possible. Bergell and Feigl have 

 prepared combinations where two amino acids are combined to the 

 ammonia by both their carboxyl groups, and these are not attacked by 

 trypsin nor by pepsin. The oxyamino acids may be combined in the 

 form of ethers with one another, or in the form of esters with other 

 amino acids, and anhydrides of amino acids are possible. Further, 

 diketopiperazine rings may occur, certain of which are easily hydrolysed 

 by alkali. 



Proteins, according to Cohnheim, are not hydrolysed by erepsin, the 

 enzyme of the intestinal mucous membrane, which hydrolyses only the 

 proteoses and peptones, converting them into amino acids ; if proteins 

 are previously hydrolysed by pepsin, they are then converted into 

 amino acids by erepsin. Pepsin would therefore appear to have a 

 special function rather than act like trypsin and the other enzymes, 

 and it may attack one of the other possible linkings of amino acids. If 

 it only produces some five or six products, there would only be the 

 same number of such linkages. 



The optically active dipeptides, d-alanyl-d-alanine and d-alanyl-1- 

 leucine, have been employed by Abderhalden and Koelker for compar- 

 ing the activity of various enzymes, such as pancreatic trypsin, yeast 

 endotryptase and intestinal erepsin ; by observing the change in rotation 

 they were able to determine the rate at which these polypeptides were 

 hydrolysed, and they found that yeast endotryptase was the most active, 

 erepsin attacked the dipeptide more slowly, and trypsin in forty-eight 

 hours had scarcely hydrolysed it at all. Not only can the rate of 

 change in rotation, which property was made use of for this purpose, be 

 used to show differences in the various enzymes, but also it can be used 

 for determining the rate of action (see monograph by W. M. Bayliss, 

 F.R.S., on Enzyme Action) of the enzyme under various conditions 

 as Abderhalden, in conjunction with Michaelis and Gigon, has 

 shown. 



By means of the change in rotation Abderhalden and Koelker have 

 also attempted to determine at which point a tripeptide is first attacked 

 by an enzyme. The specific rotation of 1-leucyl-glycyl-d-alanine is 

 + 20, that of 1-leucyl-glycine is + 85, and that of glycyl-d-alanine is 

 - 50. An increase in rotation would show that d-alanine was first 

 separated and 1-leucyl-glycine formed; a decrease in rotation would 

 point to a separation of 1-leucine and the formation of glycyl-d-alanine 

 according to the following scheme : 



4* 



