II. CHEMISTRY 541 



This compound possesses no centers of asymmetry, since in its formation 

 the asymmetry of the two carbon atoms bearing the amino groups is 

 destroyed. In contrast, the other possible desthiodiaminocarboxyUc acid 

 should still possess an asymmetric carbon atom. In fact the quinoxaline 

 prepared from desthiobiotin proved to be optically inactive and in its 

 chemical and physical characteristics identical with the quinoxaline pre- 

 pared from the corresponding synthetic diamonopelargonic acid. It is ob- 

 vious from these data that of the two possible structures (VII and X), only 

 structure VII agreed with the fact observed in the study of desthiobiotin 

 (p. 536). 



6. The Structure of Biotin 



Final proof for the correct structure of biotin was obtained by ex- 

 haustive methylation studies on the diaminocarboxylic acid from biotin.^" 

 Treatment with dimethylsulfate and alkali, followed by decomposition 

 with strong hydrochloric acid, resulted in the formation of a sulfur-con- 

 taining acid which was isolated in crystalline form (m.p. 40 to 41°) and 

 identified as 5-(thienyl-2)-valeric acid. A mixture of this compound with 



-CHo— CH2— CH2— CH2— COOH 



the corresponding synthetic 5- (thienyl-2) -valeric acid showed no depression 

 of the melting point and gave identical color reactions. As shown in Fig. 4, 

 in which the wavelength is plotted against the specific extinction coefficient, 

 both compounds showed identical ultraviolet absorption curves, with a 

 maximum absorption at 234 m^t. t, 



In view of all these data, obtained through different analytical methods, 

 the correct formula of biotin must be represented by the following structure. 



CO 



HNi' ,'CH 



I I 



HC4 aCH 



I I 



H2C5 2CH«CH2'CH2'CH2«CH2*COOH 



s 



Biotin 

 2'-Keto-3,4-iiiiidazolido-2-tetrahydrothiophene-n-valeric acid 



^"D. B. Melville, A. W. Moyer, K. Hofmann, V. du Vigneaud, /. Biol. Chem. 146, 

 487 (1942). 



