132 S. S. COHEN 



However, in the microbial systems, it has not only been possible in many 

 instances to define the site of the metabolic block but also to describe an 

 enzymatic reaction with known substrates and products that exists in the 

 parent organism and is modified m the mutant. Several types of such modi- 

 fication have been found. In most auxotrophic mutants, the enzyme is 

 missing or at least the reactivity characteristic of the enzyme cannot be 

 detected. Obviously, this distinction is of fundamental importance for the 

 problem of the nature of gene action; in these instances it must be determined 

 whether the enzyme has not been made or if it is inhibited. In the case of 

 the system synthesizing pantothenate in Neurospora, it has been shown that 

 the enzyme is present in an inhibited state in cell-free extracts of certain 

 mutants but not in extracts of the parental strain. The enz5niie may be 

 purified away from the inhibitor in this instance. It is evident that if this 

 separation were not possible, it would be a very difficult matter to determine 

 if the enzyme were really not made or were merely hidden from detection.^ 

 Several approaches to this difficult problem are available. For example, one 

 might prepare highly purified enzyme and antisera to this protein, and test 

 extracts of mutant organisms for cross-reacting proteins. Many possible 

 complications could obscure the results of such an analysis. In a more 

 biological approach, used to analyze the apparent absence of glutamic 

 dehydrogenase in Neurospora mutants, Fincham (1951, 1954) prepared 

 heterocaryons of varying numbers of dehydrogenase-deficient nuclei. The 

 dehydrogenase activities of the various heterocaryotic mycelia were related 

 to the numbers of nondeficient nuclei and indicated that the mutant nuclei 

 did not facilitate the production of inhibitors. It was concluded as a result of 

 this and other analyses that m these systems the mutation of a single gene 

 did mdeed prevent the formation of a single enzyme. 



That mutant genes did not always result in complete metabohc blocks was 

 clearly demonstrated by Bonner et al. (1952) in the case of tryptophan- 

 requirmg mutants of Neurospora. When N^ ^-labeled precursors of tryptophan 

 were fed to these mutants, in addition to an unlabeled carbon and nitrogen 

 source, and various accumulation products of tryptophan metabolism were 

 isolated, these compounds were found to contain only a portion of their 

 nitrogen in the form provided by the fed precursor. This indicated that a 

 portion of the tryjjtophan coidd be synthesized through the apparent block. 

 Such mutants have been said to possess "leaky" genes and have been 



^ It will be evident to most readers that the problem of detecting the presence in un- 

 infected bacteria of the enzyme for generating hydroxymethyl cytosine is of this type. 

 A similar question may be raised concerning the origin of neuraminidase in influenza- 

 infected animal cells. If such enzymes should prove to be inhibited in normal cells, as is 

 deoxyribonuclease in many, the problem may then be posed as to the mechanism by 

 which a virus releases such an inhibition. 



