WALDO E. COHN 



The application of paper and ion-exchange chromato- 

 graphic techniques to the ribonuclease digests of RNA (49,62) 

 permitted the isolation of many of the fragments whose properties 

 had hitherto been known only en bloc. It was soon shown that 

 (7) the pyrimidine mononucleotides are all of the 3' variety, (2) 

 the polynucleotides end in 3' pyrimidine nucleotides, and {3) 

 the intervening linkages are of the 3 ',5' (or 2 ',5') variety. An 

 extension of great significance was that ribonuclease degradation 

 proceeds through or concomitantly with intermediate 2':3' 

 cyclization of pyrimidine nucleoside phosphates which are 

 subsequently degraded by the enzyme to the end products, 3' 

 nucleotides (48). Synthetic substrates for ribonuclease showed 

 that the enzyme hydrolyzes only the esters of 3' pyrimidine 

 nucleotides through the intermediate 2':3' cyclic phosphate, 

 which indicates that the 3 ' form probably exists in at least the py- 

 rimidine nucleotides of the original chain (7). 



The question as to whether the internucleotide linkage is 

 3 ',5' or' 2 ',5' or a mixture of both was finally settled by the use 

 of an enzyme from spleen which hydrolyzed RNA to 3 ' nucleo- 

 tides (of both purines and pyrimidines) but without the formation 

 of cyclic nucleotides (29,30,63). The latter observation 

 eliminated consideration of intermediate cyclization, and thus 

 the appearance of 3' nucleotides (of purines and pyrimidines 

 alike) could be accepted as evidence of the 3 ',5' linkage to the 

 exclusion of 2 ',5'. 



At the time of the initial postulate of a 3 ',5' (or 2 ',5') 

 linkage, necessitated primarily by the degradation of RNA to 

 5 ' nucleotides by enzymes, it was pointed out that the possibility 

 of branching, by phosphorylation of the 2' (or 3') hydroxy! 

 giving rise to triply esterified ribose, must be considered (10). 

 Attempts to demonstrate, by methylation experiments, the 

 existence in RNA of ribose without free hydroxyls failed to be 

 convincing because of the difficulty of proving that the un- 

 methylated ribose found was not due to failure to methylate a 

 free position (1). Another approach lay in considering the 

 products released from RNA by the action of the diesterase of 



466 



