CHEMICAL BONDS IN NUCLEIC ACIDS 



433 



the main, discussion of branching from the main cliain by incorporation of 

 phosphotriester linkages.'" However, with the advent of an adequate ex- 

 planation of the hydrolytic behavior of ribonucleic acids it became possible 

 to define with some certainty the possible types of branching which might 

 be considered. As has already been briefly mentioned, Brown and Todd,®^ 

 starting from the requirement that ribonucleic acids are hydrolyzed 

 to mononucleotides under mild alkaline conditions, pointed out that in 

 addition to branching on phosphorus, branching by attachment of polynu- 

 cleotide chains to C2' in one of the nucleoside residues in a main polynu- 

 cleotide chain of type XVI could be considered, provided that the attach- 

 ment was through a normal phosphodiester group attached to C3. (or C2') 

 in the first nucleoside residue of the branching chain; only in this way 

 could alkali-lability be maintained. C5. could not be considered as the point 

 of linkage in the first residue of the branch as the result would be an alkali- 

 stable system, cyclization either in that residue or in the residue forming 

 the branch-point in the main chain being impossible. These authors there- 

 fore proposed a general structure of the type XXIII for branched ribo- 

 nucleic acids. In this structure the branches can be extended following the 

 normal C3. — C5. .sequence as hi the main chain. The incorporation of a 

 certain number of branches involving phosphotriester linkages into such a 

 structure would present no problem provided, again, that the Cs- — Cs- 

 sequence was followed in the branches. 



C^. C3. C5. 



X 



C. /C,.— 03,-05. 



\ 



■ Co. O c 



Oo. Oo. Oi 



"p 



\ 



XXIII 



Some support for the branched-chain structure XXIII was provided by 

 methylation studies.'- '^-^ Methylation of yeast ribonucleic acid, followed 

 by hydrolysis and separation of the mixture of sugar derivatives obtained, 

 yielded ribose as well as its mono- and dimethyl derivatives. The presence 

 of ribose was taken to indicate triply-substituted ribofuranose residues in 

 the nucleic acid.''^^ It is, however, doubtful if the method of methylation 

 can be safely applied in this field on account of the lability of the internu- 



'" A. S. Anderson, G. R. Barker, and K. R. Farrar, Nature 163, 445 (1949). 

 •" A. S. Anderson, G. R. Barker, J. M. Gulland, and M. V. Lock, ./. Chevi. Soc. 1952, 

 369. 



