108 PHYSIOLOGICAL CHEMISTRY 



method lies in the fact that the chain can be extended only at 

 one end. 



3. By treating an ammo acid or a peptid with phosphorus 

 pentachloride, the carboxyl group is converted into an acid 

 chloride group which then may be caused to interact with the 

 amino group of an amino acid or a peptid in a manner analagous 

 to that described in the second method of synthesis. Thus gly- 

 cocoll will be converted into NH 2 CH 2 CO Cl which may be com- 

 bined with a second glycocoll molecule or other amino acid to 

 form a dipeptid as in 2. 



The compounds prepared in these ways show many of the 

 reactions characteristic of the products of partial hydrolysis of 

 proteins, thus some give the biuret reaction, those containing 

 tyrosine give the Millon test, those containing tryptophane give 

 the Adamkiewicz test, and certain of the more complex of them 

 are precipitated by some of the protein precipitants. Many of 

 them even are split by enzymes found in the body which are 

 known to split protein decomposition products. These synthetic 

 products thus are identical with those obtained in protein hy- 

 drolysis, a fact which is one of the strongest pieces of evidence 

 that the' proteins are constructed on the general plan of the 

 polypeptids. Other evidence for this form of linkage in the 

 protein is given by various facts. The proteins react only to a 

 very limited extent with reagents which interact with free NH 2 

 groups. Very little hydrochloric acid is taken up by them, the 

 formaldehyde reaction described above shows relatively few 

 free amino groups, as does also the Van Slyke method also de- 

 scribed above. In alkaline hydrolysis the amino acids become 

 racemic for the most part. Only those whose carboxyl groups 

 are combined, as is the case in the amid linking, will become 

 racemic on hydrolysis. All these results are explained if the 

 amino acids making up the proteins are assumed to be united 

 by the "peptid linking." 



The behavior of polypeptids with enzymes is especially in- 

 teresting, and demonstrates the extremely specific character of 

 enzyme action. For example, it has been shown that the di- 



