462 D. O. JORDAN 



mately eight nucleotides. In view of the more recent developments it should 

 be pointed out that Astbury^^ did not completely reject the possibility that 

 the nucleotides might be disposed spiralwise around the long axis of the 

 molecule, but he concluded that if the nucleotides did not lie closely on top 

 of one another, then the state of packing must be dimensionally equivalent 

 to such an arrangement so that the neighboring nucleotide molecules were 

 closely interleaved in a surprisingly regular pattern. This, Astbury con- 

 sidered, was unlikely. 



The Astbury structure rested, in part, on the assumption that the nucleo- 

 tides were flat or approximately flat molecules, i.e., the purine or pyrimi- 

 dine ring systems were in the same plane as that of the sugar. This was 

 shown to be wrong by Furberg,'*-'^ who showed that in the nucleoside 

 cytidine the two ring systems instead of being parallel were almost per- 

 pendicular. This observation is of fundamental importance in the de- 

 velopment of the structure of nucleic acids and has been confirmed for 

 2' ,3'-isopropylidene-3 ,5'-cycloadenosine iodide by Clark et aZ.** and by Zuss- 

 man.'*'*^ As has been pointed out above, the bond between sugar and purine 

 or pyrimidine (Ng or 3 — Ci) is a single bond and so rotation about this 

 bond is possible. However, Furberg** finds that not all orientations of the 

 sugar and purine or pyrimidine are equally feasible and the most favorable 

 position is considered that shown in Fig. 5 which is that found in the crystal 

 structure of cytidine.'*''^ In view of these observations on the structure of 

 the nucleosides and nucleotides, Furberg'*^ revised the Astbury structures 

 and proposed two possible models which are given in Fig. 6. In contrast to 

 the Astbury model, these two modifications have the planes of the sugar 

 rings, as well as the P — O3 bonds (Fig. 5), approximately parallel to the 

 long axis of the molecule. Most atoms, including the phosphorus atoms, 

 lie in planes 3.4 A. apart, thus explaining the strong 3.4-A. reflection. 



In model 1 (Fig. 6) the pyrimidine and purine rings are piled almost on top of each 

 other; they cannot be piled directly on top of each other without bringing some of 

 the atoms in successive nucleotides too near together. Van der Waals attraction be- 

 tween the rings will stabilize the structure. The ribose rings and the phosphate groups 

 form a spiral enclosing the column, the spiral repeating itself after eight nucleotides 

 as required by the Astbury model. In model 2, the ribose rings and phosphate groups 

 form a central column from which the purines and pyrimidines stand out perpendicu- 

 larly. This model has the disadvantage that there are no intramolecular Van der 

 Waals forces between the purines and pyrimidines, but the rigidity of the molecule 

 will depend on such forces between the sugar molecules. 



Following the successful formulation of the structure of some of the pro- 



** V. M. Clark, A. R. Todd, and J. Zussman, J. Chern. Soc. 1951, 2952. 



^^^ J. Zussman, Acta Cryst. 6, 504 (1953). 



" S. Furberg, Acta Chem. Scand. 6, 634 (1952). 



