82 MUTATIONS 



is obvious that if one is interested in using mutagens as a decoding 

 tool, one should limit oneself to those which are most likely to act 

 directly on the nucleotides, and this is the thing I want to discuss next. 

 But, as I said, I have a more biological approach to it, and I personally 

 am very interested in the many steps which go on in a cell when a 

 mutation occurs. 



What I wish to discuss now is how do these mutagens act? I think 

 that nitrous acid is one substance for which even I must assume that, 

 at least in the tobacco mosaic virus, in the polio virus, and in the 

 transforming principle, it acts in the way in which it was predicted it 

 would act, namely, by deamination of the nucleic acid bases. Viel- 

 metter and Schuster (76) in Tiibingen think they can determine which 

 base transformations are the actual mutagenic ones. The principle they 

 use is this. They establish a correlation between the effect of pH on 

 deamination of the bases in vitro and mutation or killing of T2, and 

 they come to the conclusion that deamination of adenine and hydroxy- 

 methyl cytosine acts mutagenically and deamination of guanine is 

 lethal. I cannot judge this work because of my lack of chemical knowl- 

 edge. I wonder whether Dr. Freese knows it. 



Freese: Yes, I do. 



Auerbach: What do you feel about it? 



Freese: It shows a correlation between the pH dependence of the 

 mutagenic effect and the pH dependence of the deamination of the 

 three bases, and that correlation is very good. 



Auerbach: It seems to be all right, also, because the phage work 

 really suggests this action of nitrous acid is directly on the DNA. I 

 think that is very nicely suggested by comparison of the action of 

 nitrous acid on the two-stranded DNA of T4 (77) and the one- 

 stranded one of Tessman's (74) phage. 



Freese: That is X 174. There is only one DNA strand in this phage. 



Auerbach: Well, anyway, when the one-stranded phage was treated 

 with nitrous acid, there were only complete mutants, but when the two- 

 stranded ones were treated, there was a high proportion of mosaics, as 

 one would expect. In fact, the proportion in Vielmetter's work was 

 rather low, something like 50 per cent mottled plaques and 50 per cent 

 complete plaques, whereas one would expect, of course, a much higher 

 percentage of mottled ones. But I think that Vielmetter has a plausible 

 explanation in assuming that, in many cases where he gets a complete 

 plaque, the one strand of the DNA has been knocked out by a lethal 

 hit, so the mutant is produced only from the other one. 



Freese: Perhaps, I should say something here about some experi- 



