PHYSIOLOGICALLY ACTIVE COMPOUNDS 243 



the next by 15 residues of other amino acids and each of the 

 12 residues of proline will be separated from the next by 47 

 other residues, etc." 



Hormones, enzymes, antibiotics and 

 antigens. 



However, this hypothesis that the structure is fundament- 

 ally related to the ratios between the numbers of different 

 amino acid residues in any particular protein was not con- 

 firmed by the direct study of the breakdown products ob- 

 tained by partial hydrolysis of proteins and polypeptides. 

 On the contrary, the application of this tedious but very 

 reliable method has actually enabled people to elucidate the 

 very complicated arrangement of the amino acid residues in 

 the peptide chains of a number of physiologically important 

 compounds. This applied, in the first place, to toxic sub- 

 stances produced by bacteria (antibiotics), such as gramicidin. 

 By partial hydrolysis of the simplest of these, gramicidin S,^* 

 which is a cyclopolypeptide," it has been possible to isolate 

 numerous dipeptides and tripeptides by paper chromato- 

 graphy. By comparing these, the whole sequence of amino 

 acids in this cyclic peptide has been established.^^ Further- 

 more, the use of similar methods has led to the elucidation 

 of the sequence of the amino acid residues in tyrocidines 

 A and B^^ and other polypeptide antibiotics.'^^ 



Corresponding studies on proteins are naturally of special 

 interest. The one which has now been most thoroughly 

 studied is insulin. This hormone, which is particularly 

 important on account of its physiological activity, has been 

 studied by many workers, both in respect of its amino acid 

 composition and in respect of the arrangement of the amino 

 acid residues in the polypeptide chain. ^^ 



The work of F. Sanger and his colleagues'^'' has given a 

 clear picture of this structure. Sanger marked the terminal 

 amino groups of the polypeptides present in insulin by 

 condensing them with 2 : 4-dinitro-i-fluorobenzene ; he then 

 submitted this derivative of insulin to partial hydrolysis and 

 studied the breakdown products obtained in this way. On 

 the basis of the results thus obtained Sanger then arrived at 



