G. S. STENT AND C. R. FUERST 455 



recoil does not rupture the phosphate ester linkage, i.e. if the phosphorus 

 nucleus remained in place after all, then the two deoxyribose residues are forth- 

 with linked by a sulfate diester, which should undergo spontaneous hydrolysis 

 in aqueous medium (Kremann, 1907). Inspection of the structure shown in 

 Fig. 5 indicates, however, that breakage of one ester link would not neces- 

 sarily lead to the disruption of the DNA molecule, since the multitude of 

 hydrogen bonds still hold the two sister strands together. This has recently 

 been pointed out by Dekker and Schachman (1954), who propose on the 

 basis of physicochemical evidence that the polynucleotide strands of "native" 

 DNA are not actually continuous throughout the length of the macromolecule 

 but are already interrupted in such a fashion that on the average one out 

 of twenty to fifty phosphate links is singly instead of doubly esterified, as 

 indicated in Fig. 5. Thus, if there already exist spontaneous breaks within 

 intact DNA, it is not unreasonable to suppose that the low efficiency of killing 

 per P^- disintegration means that the DNA molecule can continue to function 

 even after a few additional interruptions of the polynucleotide chains have 

 been generated by radioactive decay. 



An event secondary to the disruption of the phosphate diester must then 

 attend the lethal fraction a of P^- disintegrations. The most reasonable hy- 

 pothesis would appear to be that inactivation is caused by a complete cut of 

 the DNA double helix. One way in which this could occur is that enough energy 

 liberated by the decaying P^- atom has been transmitted by a sequence of elastic 

 collisions to the other strand to also cause a break there. Another possibility, in 

 view of the proposal by Dekker and Schachman, would be that the lethal de- 

 cay takes place in an atom situated nearly in apposition to one of the few in- 

 complete ester links on the other strand. In either case, a complete cut results 

 because few or no hydrogen bonds remain between the spots where both sister 

 strands are broken to oppose the dissociation of the macromolecule into two 

 smaller pieces. The effect of heat on the efficiency a is readily explained in 

 terms of this hypothesis. The rapid rise of a above 55°C. must be due to the 

 dissociation of the hydrogen bonds at these temperatures {cf. Dekker and 

 Schachman), thus causing less and less resistance to separation of the two 

 strands by the energy of the radioactive disintegration. A greater and greater 

 fraction of the P^- decays can, therefore, result in a complete cut of the double 

 helix. The effect of freezing and of low temperatures on reducing a might be 

 explained by the increase of viscosity of the medium in which the two pieces 

 involved in the break have to move; i.e., that when the DNA is embedded in 

 ice there exists a greater chance that the energy of the P^' transmutation has 

 already been dissipated before the cut has actually taken place. 



Action of Ionizing Radiations. — 



It would be possible, though technically rather difficult, to ascertain whether, 

 in agreement with the hypothesis just proposed, decay of incorporated P^^ 



294 



