BIOSYNTHESIS OF NUCLEIC ACIDS 381 



Exogenous adenine »• "Active" *- Polynucleotide 



adenine adenine 



// '^ 



Synthesis *■ "Incomplete" V-'-"-V-^ Common 



de novo purine intermediate 



Inosinic \\ 



acid \_ . 



t 



Exogenous guanine *■ "Active" *■ Polynucleotide 



guanine guanine 



Fig. 5. Possible relationships of purines arising de novo, of exogenous purines, and 

 of the interconversion of the purines. 



This is borne out by the low dilution factors which can be encountered. From the 

 available estimates of the half-time of the average PNA of the liver of about 8 days, 

 determined by the retention of the isotope incorporated from adenine'^ or from 

 formate, '' a daily renewal of about 8 to 9% per day is calculated, and it is obvious 

 that the observed renewals of liver PNA adenine of 12% in 36 hours, ^^ and of over 

 22% in 5 days,"'"' represent the derivation of most of the adenine of the newly 

 synthesized nucleic acids from the administered adenine. Low dilution factors are 

 also encountered with orotic acid and with some nucleosides and nucleotides. 



b. Interconversions of the Purines 



These could be accounted for by reversal of the reactions leading back to 

 a common intermediate at the point of bifurcation of the pathways (Fig. 5). 

 The alternative of a metabolic bridge between the pathways for a more di- 

 rect interconversion of adenine and guanine derivatives may not be re- 

 quired. However, its postulation offers certain advantages since there are 

 suggestions that in the course of normal synthesis de novo at least a part of 

 either purine may arise via the other, and this cannot be rationahzed if 

 newly formed adenine must return through the common intermediate en 

 route to guanine. 



The inference that part of either purine may arise via the other is derived 

 from several observations. The relative incorporation of formate into ade- 

 nine and guanine (the A/G ratio) is generally greater at shorter times (cf. 

 1-day and 3-day values, Table II, columns 11 and 12). In L. casei^^ such a 

 tendency also appears under certain conditions. 



In the rat, with glycine-2-C" i'* the A/G ratio was also considerably greater than 

 1, although with glycine-1-C" ^^''^^ it was almost exactly 1. With glycine-N'* the 

 A/G ratio is always loi^is? less than 1, due partly to the rather specific incorporation 

 of glycine-N into the 2-amino group of guanine; howeveri"^ in some instances (but 

 not always), the purine ring of the guanine may be more extensively labeled than 

 that of adenine (Table IV), lines 4 and 5). On the other hand, with rat liver cell sus- 

 pensions in vitro the incorporation of glycine-2-C" led to an A/G ratio less than 1,'** 



1" R. Abrams, E. Hammarsten, and D. Shemin, /. Biol. Chem. 173, 429 (1948). 

 i«8 G. A. LePage, Cancer Research 13, 178 (1953). 



