THE CHEMISTRY OF THE PROTEINS 107 



This compound is glycylglycine, as termed by Fischer, the 

 group NH 2 .CH 2 .CO being called the glycyl group. It might 

 be mentioned that the benzoyl derivative of this compound 

 was obtained by Curtius in 1882. 



The ester of this compound was formed when the glycine 

 anhydride was boiled with alcoholic hydrochloric acid. Both 

 free acid and the ester have a great tendency to change back 

 into the anhydride, a property which has been utilised in 

 isolating such compounds from the products of hydrolysis of 

 proteins. The amino group is also very reactive, e.g. with 

 chlorcarbonic acid ester it gives carbethoxyglycylglycinester, 

 from which, on hydrolysis, glycylglycinecarboxylic acid is 

 obtained. When heated with leucinester, carbethoxyglycyl- 

 glycinester undergoes a similar condensation, forming carb- 

 ethoxyglycylglycylleucine, where three molecules of amino 

 acids are coupled together. 



The diketopiperazine of alanine, and that of leucine, the 

 so-called leucinimide, both gave the corresponding dipeptides 

 alanylalanine and leucylleucine, 



H,N . CH(C 4 H 9 ). CO . NH . CH(C 4 H 9 ) . COOH. 



Difficulties were, however, experienced with the condensation 

 of the other amino acid esters, and with the more complex 

 esters of the dipeptides, as also with the condensation of 

 dissimilar amino acids, where several products may result, 

 so that progress by these means became almost impossible. 

 The compounds so prepared, as expected from their constitu- 

 tion, gave the red biuret reaction, which was shown by Schiff 

 to be given by compounds having CO — NH groups joined 

 together directly, or by a C atom or by a N atom. 



About this time (1902) Curtius and Benrath prepared benzoyl- 

 pentaglycinaminoacetic acid, and two years later Curtius showed 

 that his biuret base was triglycylglycinester. This compound 

 is the ester of Fischer's triglycylglycine. Triglycylglycin- 

 ester, according to Curtius and Gumlich, changes into an 

 anhydride, which is an octapeptide. 



The method of preparing dipeptides from the diketopipera- 

 zines a few years later again proved useful, when it was found 

 that by heating the esters of the amino acids to 150-180 combina- 

 tion occurred, and instead of splitting open the ring by boiling 

 with concentrated hydrochloric acid, dilute alkali at the ordinary 



