THE PROBLEM OF NUCLEOTIDE SEQUENCE IN DEOXYRIBONUCLEIC ACIDS 93 



which the thymine is arranged; and a similar thing happens with cytosine versus 

 methylcytosine. In the latter case, one cannot speak of "replacement", because 

 you don't replace one natural compound by another, but must infer that they 

 occupy very different places in the chain. 



Dische: These changed sequences of pyrimidines and purines which are 

 brought about by bromouracil appear to be very difficult to reconcile with the 

 function of the nucleotide sequences as a coding system. If you get a complete 

 redistribution of the solitary pyrimidines then you get also a complete change in the 

 sequence of the purines. It is difficult to see how this organism can have any 

 similarity to its parents if you make such a complete change in the sequence of 

 practically all molecules of the nucleic acids. I think this is really a most remarkable 

 phenomenon. The second question I should like to ask is whether the pairing is 

 found also in the one strand DNA ? 



Chargaff : To answer your first question : on the one hand you have a hypo- 

 thesis and on the other you have some meagre findings. I don't know whether I 

 should be called upon to reconcile the experimental findings with a hypothesis. It 

 should be the other way around. The fact is that in normal E. coli DNA i5"o of 

 the thymine is solitary. This is the only 6-ketopyrimidine present in DXA. E. coli 

 treated with bromouracil gives i8"o of the thymine as solitary plus 33*^0 of the 

 bromouracil which has been incorporated as solitary unit. In other words, whereas 

 in the first case 1 5*^*0 of the ketopyrimidines are flanked by purines, almost 27",, are 

 so arranged in the second case. You are perfectly right that if there is any informa- 

 tion coded into the DNA, looking at the figures you would believe that we are 

 dealing with two entirely different organisms. I leave it to you to draw your own 

 conclusion. I don't want to speculate too much at this early stage. It could, of 

 course, be that not all the DNA is functional in a strict sense. Those fractions that 

 really carry information may be left relatively unattacked. In answer to your 

 second question, there is no obvious pairing discernible in Sinsheimer's analysis. 

 In any event, none of the regularities which one encounters in the other DNAs 

 is found in this. 



