CHEMICAL BONDS IN NUCLEIC ACIDS 425 



monoesterase treatment has, on the other hand, been confirmed in several 

 laboratories."'^^ Brown and Todd®- pointed out that an uptake of periodate 

 after ribonuclease action but without phosphomonoesterase treatment 

 would only be possible if Cz- — — P linkages were split. The picture of 

 ribonuclease action which emerges from all these observations is that the 

 enzyme cleaves Cs- — — P linkages in the postulated structure XVI giving, 

 in addition to pyrimidine-2'- or -3 '-mononucleotides, fragments containing 

 terminal pyrimidine nucleotide residues, bearing monoesterified phosphoryl 

 groups at C2' or C3.; enzymic removal of these phosphomonoester groups 

 then yields periodate-oxidizable pyrimidine nucleoside residues linked 

 through Cs' to the rest of the (purine) polynucleotide chain. 



When ribonucleic acid is treated for short periods with ribonuclease, 

 two substances are produced^"^ which are converted, respectively, to uri- 

 dylic and cy tidy lie acids by further action of the same enzyme. These two 

 substances have been shown to be identical®^ '^"^ with the cyclic phosphates, 

 uridine-2', 3 '-phosphate and cytidine-2', 3 '-phosphate, synthesized by 

 Brown, Magrath, and Todd.^^ It is clear that these cyclic phosphates are 

 the precursors of the pyrimidine mononucleotides found in ribonuclease 

 digests and that they must have arisen by a mechanism akin to that postu- 

 lated for the chemical hydrolysis of the nucleic acids.^'- This was confirmed 

 by Brown, Dekker, and Todd^^* in a further investigation of the action 

 of ribonuclease on the cyclic phosphates. Volkin and Cohn"° had shown 

 that the pyrimidine nucleotide fraction in ribonuclease digests consists 

 solely of uridylic acid b and cytidylic acid b (i.e., the 3 '-phosphates), and 

 Brown, Dekker, and Todd provided an adequate explanation of this fact 

 by showing that ribonuclease effects a unidirectional cleavage of the cyclic 

 2 ',3 '-phosphates of uridine and cytidine yielding exclusively the 6(3')- 

 isomers; ribonuclease, however, had no action on the cyclic 2 ',3 '-phosphates 

 of adenosine and guanosine, a fact which supports the specificity of ribo- 

 nuclease for pyrimidine nucleotide derivatives. In the course of this inves- 

 tigation it was shown that the phosphoryl group in cytidjdic acid b oc- 

 cupied the same position (probably 3') as in uridylic acid b by converting 

 the former to the latter by deamination under alkaline conditions which pre- 

 cluded phosphoryl migration. In both the cyclic pyrimidine nucleotides, 

 then, it was clear that ribonuclease cleaved the C2' — ^O — P linkage. It 

 is of some interest to note in connection with the demonstration that cyclic 

 phosphates are intermediates in ribonuclease digestion that in 1946 

 Schramm, Bergold, and Flammersfeld''^ had suggested the presence of 



1" R. Markham and J. D. Smith, Research 4, 344 (1951). 



108 R. Markham and J. D. Smith, Nature 168, 406 (1951). 



">» D. M. Brown, C. A. Dekker, and A. R. Todd, J. Chem. Soc. 1952, 2715. 



i^« E. Volkin and W. E. Cohn, Federation Proc. 11, 303 (1952). 



I'l G. Schramm, G. Bergold, and H. Flammersfeld, Z. Naturforsch. 1, 328 (1946). 



