526 G. H. BEAVEN, E. R. HOLIDAY, AND E. A. JOHNSON 



the concentration of sodium chloride in such a solution is subsequently 

 raised, the «(?) value is reduced, but not to the value that would have been 

 found had the DNA not been previously exposed to the lower concentration. 

 It may be concluded that the irreversible increase in e(P) which occurs at 

 low salt concentration must be due to irreversible structural changes in the 

 DNA molecule. The possibility that such a change is due to contamination 

 by depolymerizing enzyme has been suggested by Gilbert et al?^ to explain 

 the spontaneous changes in the viscosity of DNA dissolved in distilled water. 

 In the light of these results some doubt exists about the interpretation 

 that can be placed on the careful work of Shack et al}^ on the effects of 

 varying metal-ion concentrations on the absorptivity of DNA. Since their 

 solutions were initially made up in distilled water, the DNA may have 

 been irreversibly changed to an unknown extent, though evidently not 

 sufficiently to obscure the effects under study. Fig. 25 contains information 

 derived from their paper,^^ showing the lowering of e(P) of DNA, initially 

 dissolved in distilled water, following the addition of monovalent cations. 

 It also demonstrates the much greater activity in this respect of divalent 

 cations. In this connection Lawley*^ has shown that cations compete for 

 the DNA anion in the order H+, Mg++, Na+, K+. Cavalieri^^ has shown that 

 adding a large excess of magnesium sulfate to a solution of sodium DNA 

 lowers the pH. Fig. 24 shows that in the presence of metal cations the pH 

 value at which the increase in «(?) commences is lowered. This figure also 

 includes Frick's data^'* in molar sodium chloride solution. On the alkaline 

 side of neutrality metal cations have very little effect. In Fig. 26 are shown 

 titration curves illustrating the combined effects of variations in salt con- 

 centration and pH on the e(P) value of DNA. The effect of exposure to low 

 salt concentration is to increase the c(P) value at pH 6, although the final 

 value at low pH is independent of this prior treatment. Denaturation in 

 this manner thus reduces the eventual increase in e(P) on acidification. 

 This behavior suggests itself as a useful test for the state of denaturation 

 of a given sample of DNA. It is dissolved in 10"^ M sodium chloride and its 

 absorbance determined at pH 3.0 and pH 7.0. The ratio {Ra) of these two 

 values may be taken as a measure of the extent of denaturation of the 

 sample. From Fig. 26, Ra is seen to be 1.30 for Thomas' undenatured 

 material and 1.12 for the material used to obtain the other two curves. 

 Both Frick'^ and Shack et al.^* have suggested that the c(P) value of a 

 DNA preparation is a good indication of its state of denaturation. The test 

 given here has the added advantage that phosphorus determinations are 

 unnecessary. It is evident from the results of this work that metal cations 

 play an important role in the stabilization of the structure of native DNA. 



" L. Gilbert, W. G. Overend, and M. Webb, Exptl. Cell Research 2, 349 (1951). 



