366 RICHARO S( IIWEKT AND JOHN BISHOP 



1960). Tlie .same trctitnicnt followed hy ])i-ccipitatioii of the lihosomes 

 with magnesium and dialysis has been used by Takanami (19()ll. 



Using such purified particulate preparations, several assay systems 

 have been developed, and some progress has been made in the purifica- 

 tion of the nondialyzable factors required for the transfer reaction. 

 Solubility properties, column chromatography, and heat lability indicate 

 that the transfer factor or factors are protein and presumably an 

 enzymatic reaction is involved (Nathans and Lipmann, 1960; Bishop 

 and Schweet, 1961b). A heat-stable, nondialyzable transfer factor has 

 also been reported (Grossi and Moldave, 1960). Nathans and Lipmann 

 (1960) reported a 30- fold purification of an enzyme fraction from the 

 rat liver 105.000 g supernatant by ammonium sulfate and acetone frac- 

 tionation. Takanami (1961) has purified 8-fold an enzyme fraction 

 from the same source by DEAE-cellulose chromatography. Transfer 

 enzymes have also been purified from pea seedlings (Webster, 1961), 

 from E. coli (Nathans and Lipmann, 1961), and from rabbit reticulo- 

 cytes (Bishop and Schweet, 1961c). 



It may be asked whether the transfer function is served by the same 

 group of enzymes which catalyze the activation of the amino acids. 

 This is an extremely difficult question to answer rigorously. It may be 

 that quite different concentrations of the same group of enzymes are 

 required for the different types of catalytic activity. Thus, the demon- 

 stration that a certain enzyme fraction will catalyze the transfer reac- 

 tion, but not incorporation from free amino acids (Bishop et nl., 1960a; 

 Takanami, 1961), may simply mean that more enzyme is required to 

 activate than to transfer the amino acids. However, relative purification 

 data (von der Decken and Hultin, 1960) and the failure to form 

 leucyl-RNA by a fraction which transferred from C**-leucyl-RNA to 

 protein (Nathans et al., 1962) is highly suggestive that transfer is not a 

 function of the activating enzymes. The final answer, however, may be 

 found in working with the artificial system of Nirenberg et al. (1962) in 

 which a mixture of polyuridylic acid and DNasc-trcated E. coli ribo- 

 somes will accept only the amino acid phenylalanine from amino 

 acyl-RNA. 



The first suggestion that more than one enzyme was involved in the 

 transfer reaction came from von der Decken and Hultin (1960), who 

 fractionated the pH 5 supernatant of rat liver by batch treatment with 

 hydroxylapatite, and followed the activity of various eluted fractions in 

 the transfer of valine and tyrosine. In one particular instance a relative 

 separation of valine and tyrosine transfer activities was observed. 

 However, there was a strong correlation between low tyrosine transfer 

 and hiqh amount of tyrosine activation, suggesting that Ww low transfer 



