STRUCTURE OF INSULIN 



be regarded as part of the functional hormone is open to ques- 

 tion. 



N -Terminal Residues 



After the amino acid composition of a protein has been 

 ascertained, the next problem is to determine their sequence in 

 the polypeptide chains. One amino acid to which a definite 

 position in the molecule may be assigned is the residue that is 

 situated at the end of the chain, and bears a free amino group. 

 This is known as the iV-terminal residue and may be distin- 

 guished from other residues by suitable chemical reactions in- 

 volving its free amino group. Thus an initial attack on the 

 insulin molecule was made by developing a general method for 

 the identification and estimation of the A^-terminal residues of 

 proteins and peptides (14). The reagent used was 1,2,4- 

 fluorodinitrobenzene, which reacts with amino groups to form 

 yellow DNP (dinitrophenyl) compounds. Hydrolysis of a 

 DNP-protein yields the DNP derivatives of those amino acids 

 which were present as i\^-terminal residues. Thus in a hydrol- 

 yzate of DNP-insulin it was possible to detect DNP-glycine and 

 DNP-phenylalanine, and allowing for hydrolytic loss there was 

 one equivalent of each present. This indicated that there were 

 two polypeptide chains present, a glycyl chain and a phenyl- 

 alanyl chain. 



Oxidation 



Insulin contains a high proportion of cystine, and it seemed 

 probable that the two different chains were joined together 

 through the disulfide bridges of these residues, especially since 

 no other type of covalent cross-linkage is known to occur in 

 proteins. Thus in order to separate the two chains, insulin was 

 oxidized with performic acid (15), splitting the cystine residues 

 and converting them to cysteic acid. From the oxidation mix- 

 ture two fractions were isolated, which appeared to represent 

 the whole of the insulin. Fraction A was acidic and had only 



437 



