II. CHEMISTRY 



227 



COOII 



CHsO-^^COOH 

 CHi 



COOH 



CHsO^^COOll 



CH, 



(1) (2) 



Since the Gibbs test with 2,6-dichloroquin()iu> chlorimide was positive 

 for pyridoxine, the compound could not be para substituted as rej^ards 

 the hyth'oxyl, and therefore position G in the ring was unsul)stituted. The 

 structure of the methoxydicarboxyhc acid should then be structure 1, 

 which w^as confirmed by synthesis by Harris et aZ.^" 







II II II CH.OH 



CHj-C-CH^-C-CHOC^Hj + CXCH^-C-NH 



CH2OC5HS 



H,cl^^>0 

 H 



CH2O 



HClor 

 HjSOj 



at 120° 



H3C 



\ 



C = 

 



02N 

 H3C 



CH20 



\ 



C = Hj + Pt 



^N 



CI 



H2N 

 H3C 



CH20 



\ 



^N 



c=o 



CI 



H: + Pd 



BaCOa 



CH2O 



H2NffS-C = XaNO, 



HaCk^^ 



H2SO4 



CH2O 



H0r|'''\-C = 



CH,0 



CHjNi 



I \ COOH 



CH30|pS-C = Ba(MnO,), ^ CHjOfT^COOH 



HcCl^.J 



HaCkj^^ 



Synthesis of Dicarboxy Acid derived from Pvridoxine Methyl lather-" 



The vitamin was therefore 2-methyl-3-hydroxy-4,5-di(hydroxymethyl) 

 pyridine. 



CH2OH 



HO 

 CH 



CH.,OH 



Pyridoxine 



/). Knhn, Wendt, and Westphal 



In a series of three papers, Kuhn, Wendt, and Westphal''"'^ independ- 

 ently established the structure of pyridoxine. The line of attack was also 



2« S. A. Harris, K. T. Stiller, and K. P^olkers, J. Am. Chem. Soc. 61, 1242 (1939). 



