380 B. MAGASANIK 



On the other hand, nucleoproteins may be well-defined chemical struc- 

 tures in which nucleic acid and protein are arranged in a definite pattern, 

 and these entities may be the ultimate carriers of biological properties. 

 The detachment of the PNA from the protein would then present us with 

 fragments valuable for chemical and physical study but without biological 

 interest. 



The question of the nature of PNA-protein is thus largely a question of 

 the nature of the bonds which hold PNA and protein together. Certain 

 relevant conclusions may be drawn by examining the conditions which 

 cause these bonds to break. 



The early nucleic acid chemists used dilute alkali to separate PNA from 

 protein, but with the recognition of PNA as an alkali-labile polymer, other 

 methods came into use. The most generally applicable of these methods is 

 treatment with hot 10% sodium chloride.^^ Recently milder procedures, 

 such as treatment with guanidine hydrochloride^^ or with sodium dodecyl 

 sulfate^^ in the cold, have been found effective in separating PNA from 

 protein. In the special case of the plant viruses even milder treatment 

 suffices: the PNA of tobacco mosaic virus was liberated by very short 

 heating at neutral pH;-^ the PNA of turnip yellow mosaic virus could be 

 detached from the protein by treatment with 30 % ethanol in the cold.^^ 



It is quite evident from the ease with which nucleoprotein can be split 

 that the two moieties are not joined by covalent bonds. The agents used 

 for the separation of PNA from protein are all eff'ective protein dena- 

 turants, and, indeed, the removal of PNA is always accompanied by the 

 denaturation of the protein. It seems likely, therefore, that the forces 

 which are responsible for keeping protein molecules in their native folded 

 configuration are also responsible for the binding of PNA to protein. These 

 forces are the coulombic attractions of oppositely charged ions, the attrac- 

 tion of dipoles, and hydrogen bonds. The polar groups on protein and PNA 

 are well suited for such mutual attractions. 



These forces of attraction may be quite nonspecific and cause the forma- 

 tion of nucleoprotein whenever PNA and protein are in close proximity. 

 Alternatively, nucleoproteins may be formed just like native proteins, 

 under the influence of specific directing forces active in the cell. Recent 

 work sheds some light on this problem, although sufficient evidence which 

 would enable us to choose between these alternatives has not been ob- 

 tained. Szafarz'" observed that the supernatant fluid of cytoplasmic ex- 

 tracts centrifuged at 60,000 g did not contain a component corresponding 



2* J. N. Davidson and C. Waymouth, Biochem. J. 38, 375 (1944). 

 26 E. Volkin and C. E. Carter, J. Am. Chem. Soc. 73, 1516 (1951). 



26 E. R. M. Kay and A. L. Bounce, J. Am. Chem. Soc. 75, 4041 (1953). 



27 S. S. Cohen and W. M. Stanley, J. Biol. Chem. 144, 589 (1942). 



