320 Discussion 



monofunctional compound. The stronger the positive charge, the more 

 mutagenic it is. 



Pontecorvo: Dr. Alexander, you showed us electron micrographs of 

 drastically modified DNA molecules. Those were molecules having a 

 molecular weight of 6 million ; suppose that instead of numerous mole- 

 cules of mustard per molecule of DNA you had only one, then they 

 would probably be modified in a less drastic way ? 



Alexander: The physicochemical methods are of course much cruder 

 than any biological method, but even physicochemically one can show 

 that these molecules are coiled up, i.e. internally crosslinked after only 

 something like 60 reactions have occurred. This is already sufficient to 

 lower the light-scattering curve substantially, showing that the shape 

 of the molecule has been restricted internally. The cell can probably 

 distinguish very much finer changes. So far as forming these gels goes, 

 when we treat the complete nucleoprotein something like 20 reactions 

 is already sufficient to change the DNA from being a soluble material 

 into a crosslinked gel. Therefore, if one makes allowance for the much 

 less refined techniques, one gets down to the orders of magnitude which 

 one might encounter in biology. Few reactions can so effectively modify 

 a big molecule like DNA as an internal crosslink which enforces an 

 unnatural shape on it. 



Davis : If agents with polyfunctional groups are mutagenic bj^ virtue 

 of a crosslinking reaction with DNA, I agree with Dr. Alexander that it 

 seems difficult to see how they could lead not only to loss of an enzyme- 

 forming capacity but also to reversion of this loss. However, the 

 difference in mechanism between a loss and a gain may not be so great 

 if we consider that a crosslinking group might distort rather than 

 destroy the function of the altered region of DNA. Because, though the 

 gene normally reduplicates itself precisely, DNA with an added cross- 

 linking group can hardly be expected to duplicate that group. Instead, 

 a mutation might arise if the added group so distorted the parent DNA 

 that it served as a template for forming a third DNA, neither crosslinked 

 nor identical with the original. This DNA could perpetuate its alteration 

 in future generations. In turn, one might imagine that on further 

 mutagenic treatment a crosslinking group could so distort the mutant 

 gene that its progeny turned out to be the original gene again. 



Alexander: I think this is quite feasible. I am impressed by the subtlety 

 of these refined biological reactions and I felt that to modify a code, this 

 mechanism is too crude. That was why I suggested that these hidden 

 breaks might form a better starting point for looking at this problem. 

 I must also admit that I had in mind that the great superiority of the 

 polyfunctional agent is usually associated with toxic effects; e.g. Dr. 

 Fahmy finds that the polyfunctional agents are vastly superior to the 

 monofunctional when he scores lethals in his Drosophila. Yet when we 

 come to something more decisive, like a back mutation. Prof. Wester- 

 gaard does not find this difference. 



Hayes: When you put your DNA into urea, it breaks down. Do you 

 mean that the two helices come apart, and if so how do they come 

 apart? Surely they don't come apart without breaks? 



