90 PTEROYLGLUTAMIC ACID 



ment and a diazotizable aromatic amine which could be estimated }\v the 

 method of Bratton and Marshall.'-' In the absence of oxygen no diazotizable 

 amine or fluorescent pigment was produced by alkaline hydrolysis. The 

 fluorescent pigment proved to be a dibasic acid having pKa values of 3.9 

 and 7.7. Elementary analysis suggested the empirical formula CtHsNsOs. 

 Decarboxylation of the fluorescent dibasic pigment at 300° resulted in the 

 liberation of approximately 1 mole of carbon dioxide and the formation of 

 a fluorescent monobasic acid with a pKa of 8.0. Oxidation of the original 

 dibasic acid with chlorine water, followed by hydrolysis with 0.1 N HCl at 

 140°, yielded a compound which gave positive color tests for guanidine. The 

 formation of guanidine under such conditions constitutes evidence for a 

 pyrimidine ring with an amino group in the 2 position. The fluorescent 

 dibasic acid showed characteristic absorption spectra in 0.1 N sodium hy- 

 droxide with maxima at 253 and 365 m/x. The empirical formula C7H5N5O3, 

 the titration data, and formation of guanidine suggested a 2-aminopurine 

 or a 2-aminopteridine. The absorption spectra, however, eliminated the 

 possibility of a purine, because purines do not have absorption maxima 

 above 300 m/x. Thus the available evidence pointed toward a 2-aminopteri- 

 dine with an enolic and a carboxy group. With this evidence available, 

 attempts were made to synthesize pteridines having these functional 

 groups. The compound was identified as 2-amino-4-hydroxypteridine-6- 

 carboxylic acid by comparison with the synthetic compound, and the 

 monobasic fluorescent pigment produced by decarboxylation was identified 

 as 2-amino-4-hydroxypteridine. 



The structure of these two pteridines was established by Mowat et al}^ 

 by the following series of reactions. Diethylmesoxalate was condensed with 

 2 , 4 , 5-triamino-6-hydroxypyrimidine (II) to yield isoxanthopterincarboxy- 

 lic acid (III). The structure of isoxanthopterincarboxylic acid (III) is sho^^^l 

 by formula III, although it had not previously been definitely established 

 whether the carboxyl group occupied the 6 or 7 position. On chlorination 

 of isoxanthopterincarboxylic acid (III) and subsequent reduction with hj- 

 drogen iodide, one of the hydroxyl groups was removed to give compound 

 V, which was identical with the dibasic fluorescent pigment. Presumably 

 either the 4- or the 7-hydroxyl could have been removed by this proce- 

 dure. The presence of the 4-hydroxyl in compound V was shown in two 

 ways. First decarboxylation of 2-amino-4-hydroxypteridine-6-carboxylic 

 acid (V) gave 2-amino-4-hydroxypteridine (VI), the structure of which was 

 established by its synthesis from glyoxal and 2,4,5-triamino-6-hydroxypyr- 

 imidine (II). Its formation by this method demands a hydroxyl group in 



22 A. C. Bratton and E. K. Marshall, Jr., /. Biol. Chem. 128, 537 (1939). 



" J. H. Mowat, J. H. Boothe, B. L. Hutchings, E. L. R. Stokstad, C. W. Waller, 



R. B. Angier, J. Semb, D. B. Cosulich, and Y. SubbaRow, /. Am. Chem. Soc. 70, 



14 (1948). 



