HOROWITZ AND LEUPOLD 



of gene functions are indispensable, 

 and if dispensable and indispensable 

 functions are randomly distributed 

 among the genes, then the probability 

 of detecting a mutation in a gene with 

 two primary functions is only one per 

 cent. On the other hand, if the fre- 

 quency of indispensable functions is 

 low, then the chance of detecting 

 multifunctional genes will be much 

 better. The determination of the pro- 

 portion of indispensable functions is 

 thus critical for the one gene-one en- 

 zyme concept. The question is how 

 this quantity is to be determined. It 

 w ould seem almost by definition to be 

 unknowable, in which case the one 

 gene-one enzyme idea must be ban- 

 ished to the purgatory of untestable 

 hypotheses, along with the proposition 

 that a blue unicorn lives on the other 

 side of the moon. 



THE FREQUENCY OF INDISPENSABLE 

 FUNCTIONS IN NEUROSPORA 



What is needed is a method for de- 

 tecting mutations which result in loss 

 of an indispensable function and for 

 comparing their frequency to that of 

 mutations which cause loss of a dis- 

 pensable function. It occurred to one 

 of us (Horowitz, 1948, 1950) that the 

 so-called "temperature mutants" of 

 Neiirospora might form the basis of 

 such a method. Temperature mutants 

 are a class in which the mutant pheno- 

 type is fully expressed only in a par- 

 ticular temperature range. Generally, 

 such mutants exhibit a growth factor 

 requirement when cultured at 35°, 

 but grow in its absence at 25°; in a 

 few cases this relationship is reversed 

 —i.e., the growth factor is required at 

 the lower, but not at the higher, tem- 

 perature. In three instances it has been 

 found that particular temperature mu- 

 tations behave as alleles of mutations of 

 the usual, temperature-independent 



213 



sort, and it seems not unlikely that this 

 will be found to be generally true. 



The usefulness of these mutants for 

 the present problem is based on the 

 expectation that the mutant will be 

 recoverable in the temperature range 

 within which it has no growth factor 

 requirement, regardless of whether a 

 dispensable or an indispensable func- 

 tion has been lost. This expectation is 

 borne out by the fact that a group of 

 temperature mutants which fails to 

 grow on complete medium at the mu- 

 tant temperature is, in fact, known. 

 Of the 26 temperature mutants known 

 in Nejirospora, 12 are of this type, 

 while 14 grow on complete medium 

 in the temperature range within which 

 they have a requirement. In other 

 words, roughly one-half of these mu- 

 tants has lost an indispensable function. 



In using the temperature mutants as 

 a sampling device it is assumed that 

 genes controlling indispensable func- 

 tions are just as likely to yield tem- 

 perature alleles as those controlling 

 dispensable functions. This assumption 

 is supported by two considerations. 

 In the first place, the two classes of 

 functions are in no sense natural cate- 

 gories, but depend largely on the com- 

 position of the particular complete 

 medium which is employed. There is 

 thus no reason to assume that the genes 

 governing these functions differ from 

 one another in any fundamental way. 

 In the second place, among the tem- 

 perature mutants whose specific re- 

 quirement is known there is no indica- 

 tion that any one kind of nutritional 

 requirement is favored over others. 

 Mutation to temperature alleles ap- 

 pears to occur at random among genes 

 controlling known biochemical syn- 

 theses (Horowitz, 1950). 



With the information that the fre- 

 quency of indispensable gene func- 

 tions constitutes approximately 50 



