136 PTEROYLGLUTAMIC ACID 



of PGA in sulfanilamide-inhibited E. coli is similar to that observed with 

 E. coli mutants (Lampen et al}'^) . 



In summary, it can be said that in many organisms PGA can replace the 

 requirement for p-aminobenzoic acid. This especially seems true where the 

 function of p-aminobenzoic acid is concerned only with synthesis of purines 

 and pyrimidines. In E. coli, where p-aminobenzoic acid is concerned in the 

 synthesis of methionine, serine, purine, and pyrimidine, PGA has been 

 implicated directly only in the synthesis of thymine and indirectly in that 

 of adenine. How then does p-aminobenzoic acid function in the synthesis 

 of methionine and serine? Three hypotheses seem probable: (a) p-Amino- 

 benzoic acid is converted efficiently to a PGA-like type of coenzyme, but 

 the conversion of preformed PGA to this coenzyme is very inefficient; (b) 

 p-aminobenzoic acid functions as a coenzyme without being incorporated 

 into a PGA-like type of molecule; (c) p-aminobenzoic acid functions via 

 vitamin B12 in the synthesis of methionine and serine. 



Unfortunately little direct experimental data can be brought to bear on 

 this point. However, it is significant that the synthesis of methionine, 

 serine, and adenine, which cannot be taken over by PGA in E. coli, can be 

 ascribed to PGA function in other types of organisms. This will be discussed 

 in greater detail in the next section. 



5. Role of PGA in Synthesis of Amino Acids 



PGA promotes serine synthesis in the growth of S. faecalis R.^^' ^^ It has 

 been observed^^ that pteroylglutamic acid is essential for the synthesis of 

 serine from glycine and formate by resting washed cells of S. faecalis R. 

 In this particular system N'"-formyl PGA is 30 % more active than PGA. 

 Tetrahydro PGA and citrovorum factor are no more active than PGA 

 when added simultaneously with other components. However, when the 

 cells are first incubated with citrovorum factor in glucose-phosphate, and 

 the glycine and formate added subsequently, serine synthesis is more rapid 

 than when PGA is similarly used. Also, in the above systems formation of 

 citrovorum factor accompanies the synthesis of serine promoted by PGA. 

 These same workers^^ also found that the serine requirement for growth by 

 Leuconostoc mcsenteroides can be replaced by high concentrations of glycine 

 plus a high carbon dioxide tension and that the citrovorum factor greatly 

 reduces the carbon dioxide requirement. It thus appears that carbon dioxide 

 may serve as the source of the single-carbon fragment in this experimental 

 system. The very small amount of citrovorum factor used precludes the 

 possibility of its serving as a stoichiometric source of the single-carbon frag- 

 ment, and this factor must function instead as a more efficient catalyst 



" H. P. linxiuist, Thesis, University of Wisconsin, 1949. 



■'" H. R. irollfuid and W. W. Meinko, ./. Biol. Chcm. 178, 7 (1949). 



