244 ORIGIN OF STRUCTURES AND FUNCTIONS 



the structure of insulin as follows: the molecular weight of 

 soluble insulin is about 48,000. It varies, however, with the 

 concentration and pH of the solution. When the pH is less 

 than 4 or more than 7-5 the insulin molecule dissociates into 

 parts with a molecular weight of 12,000. These parts are each 

 composed of four open polypeptide chains in two of which 

 (the A chains) the terminal amino group belongs to a glycine 

 residue and the terminal carboxyl group to an aspartic acid 

 residue. The corresponding terminal residues in the other 

 two chains (the B chains) are phenylalanine with a free amino 

 group and alanine with a free carboxyl group. 



NH, I S S 1 NH, NHj NH; 



II I II 



Gly lieu Val.Glu.Glu Cy.Cy.AIa Ser . Val . Cy . Sor . Leu Tyr Glu Lou Glu. Asp.Tvr.Cv Asp 



I ■ r 



s s 



NH.NH. S S ' 



III I 



Phe Val.Anp.Glu.His.Lcu.Cy.Cly .Ser. His . Leu. Val . Glu . Ala. Leu Tyr Leu . Val .Cy Gly Glu. Arg Gly Phe.Phe.Tyr Thr Pro Lys Ala 



Fig. 15. Formula of ox insulin. 



Sanger and colleagues" consider that, strictly speaking, the 

 basic unit of insulin is a particle with a molecular weight of 

 6,000 consisting of one A chain and one B chain joined 

 together by disulphide bridges. Their complete formula for 

 ox insulin is shown in Fig. 15. According to C. Tanford 

 and J. Epstein,*^ two such particles are joined together by 

 means of zinc atoms to form a particle with a molecular 

 weight of 12,000. 



These data as to the sequences of amino acid residues in 

 the polypeptide chains of insulin do not show the periodicity 

 in the arrangement of amino acids suggested by Bergmann. 

 The arrangement here is far more complicated. Two identi- 

 cal amino acid radicals may be side by side or may be 

 separated from one another by any number of other residues. 

 There is no obvious regularity or rhythm in these sequences. 

 Moreover, a definite sequence must be present, at least in 

 some part of the molecule, if the protein is to exercise its 

 physiological functions. We still do not know why this is so, 

 we cannot explain the immediate cause of this specificity, 

 but facts which have been obtained recently demonstrate 

 beyond doubt that the specificity exists both for insulin and 

 for other analogous hormones. For example, the two hor- 



