General Discussion 343 



Lederberg: We have spent these days in discussing issues of 

 fundamental importance but have not often touched on drug 

 resistance as a practical problem which, as Sir Charles points out, 

 was the original motivation for this symposium. Perhaps now is 

 the time to revert to prospects for practical control, based on the 

 principles we have been elaborating. 



Two approaches present themselves: to cope with either the 

 genotype or the phenotype of resistance, though neither offers a 

 panacea. As regards the genotype, we might ask either to prevent 

 the occurrence of resistance mutations, or induce their reversion 

 to sensitive alleles. Neither is within the grasp of present-day 

 genetic technology. It is true that we have fragmentary informa- 

 tion on mutagenic and antimutagenic chemicals, but the effects 

 now known would have an inappreciable influence on resistance as a 

 practical matter. Some investigators have hoped that DNA-trans- 

 duction of sensitivity might be used in a therapeutic sense. But 

 this process is only the transfer of information, not the imposition 

 of it: in all known cases of transduction (excepting lysogenic, 

 colicinogenic and F conversions) several per cent of the recipient 

 cells are the most that can be transformed, and this ratio is often 

 much less. Furthermore, the progeny of a transformed cell are 

 characteristically a mixture of the old and new types. This is not 

 to dampen hopes for the control of the mutational process, but these 

 hopes can be realized only by dint of vigorous attacks on funda- 

 mental issues of microbial, indeed of general, genetics. 



The control of the resistant phenotype may afford more im- 

 mediate prospects. Where resistance is physiologically adaptive, 

 as in the developments of penicilhnase, we might hope to mtervene 

 in the bacterial response, for example if we would find a specific 

 antagonist for it. Or we might equally look for specific antagonists 

 of the resistant phenotype, e.g. an anti-penicillinase. Furthermore, 

 in some instances, resistance may be due to biochemical deletions 

 (of entities which would then be called receptors ; in many cases these 

 may be specific permeases) the restoration of which would confer 

 phenotypic sensitivity. (This is one possible explanation of Eagle's 

 co-killing phenomenon discussed earlier.) 



We may also reflect that the very pattern of sensitivity of exist- 

 ing pathogens (by which antibiosis itself is defined) must reflect 

 some biological disadvantage, however slight, on the part of resis- 

 tant genotypes. Our neatest trick would be to turn the tables on 

 resistant types by discovering, exaggerating and exploiting this 

 defect. On the whole, little was said about resistance phenotypes 

 at the symposium, and little more is known. The propects for 

 the achievement of the necessary understanding are, however, 



