CHEMICAL BONDS IN NUCLEIC ACIDS 



429 



and WooUey" from acid hydroly sates of yeast ribonucleic acid. Neverthe- 

 less, fractionations of oligonucleotide mixtures in ribonuclease digests have 

 been achieved by paper chromatography and electrophoresis,"^ and by 

 ion-exchange chromatography,^^ which have given solutions of apparently 

 homogeneous substances. Using ion-exchange chromatography, as many 

 as 30-40 distinct elution peaks have been observed, and many of these 

 have been shown to represent solutions of virtually pure oligonucleotides. 

 Using such solutions, the structure of a number of oligonucleotides has been 

 determined. The following example,^^ set out schematically below, will 

 suffice as an example of such a determination carried out on a trinucleotide 

 containing the bases uracil, adenine, and guanine in the ratio 1:1:1 ob- 

 tained from a ribonuclease digest (in the formulas U, A, and G represent 

 the uracil, adenine, and guanine residues). 



OPOjHj 

 U-C2.-C3.-C5. 



P 



/ 

 A C2-^C3: -Cj. 



P 



/ 

 G— C2— C3.— Cj. 



U-C2.-C3.--C5 



barley 6 (3')- " J y 



/ 



venom 

 diesterase 



Uridylic acid b and 



idenylic and guanylic 



acids a and b 



G_C2.-C3.-C5. 

 XXII 



alkali 



Uridine and adenylic 



and guanylic acids 



a and 6 



Uridine-5'-phosphate 

 and 



Adenosine-5'-phosphate 

 and 



Guanosine 



Alkaline hydrolysis yielded uridylic acid h, together with the mixed a and 

 b isomers of adenylic and guanylic acid. By removal of the terminal phos- 

 phoryl group with phosphomonesterase or barley 6(3')-niit-leotidase,"^ a 

 product XXII was produced whose molecular weight could be deduced 

 from the ratio of inorganic to total phosphorus. With snake venom diester- 

 ase XXII underwent fission at the Cs- — ^O — ^P linkages, giving guanosine 

 and the 5 '-phosphates of uridine and adenosine, while with alkali it gave 

 uridine and the mixed purine a and h nucleotides. Treatment of other oli- 

 gonucleotides first with phosphomonoesterase, and then with snake venom 

 diesterase, always yielded only nucleoside-5 '-phosphates and a nucleoside 

 representing the terminal residue, indicating that all were linear structures 

 since chain-branching would have been expected to lead to other types of 

 breakdown products by the action of snake venom diesterase. 



The earlier work of Merrifield and Woolley," in which they isolated and 



"» L. Shuster and N. O. Kaplan, Federation Proc. 11, 286 (1952). 



