GENETIC CONTROL OF CELL INTEGRATION 323 



much enzyme is produced in nondividing cells as in growing cells within 

 a short time interval. From the sigmoid shape of these curves, one 

 might suppose that they reflect a population heterogeneity — that initially 

 only a small fraction of the cells could utilize galactose, and that the 

 curve really represents the increase of that fraction in the population, 

 as it becomes the predominant cell type. This possibility is effectively 

 ruled out by the similarity of the curve found in the absence of a 

 nitrogen source. It is seen that a new enzymatic activity can arise in 

 scarcely dividing yeast cells, which also are not dying, as shown by 

 plating at intervals to determine their viability. In this experiment, 

 the appearance of the enzyme galactozymase has been induced by galac- 

 tose, which is termed the inducer. 



Similar results have been obtained with a number of enzyme systems 

 in yeast and in bacteria. With more sophisticated techniques, it has 

 been possible to demonstrate in two systems (/3-galactosidase formation 

 and penicillinase formation) that virtually every cell in the population 

 behaves uniformly in altering its enzyme content in response to the new 

 environment. Thus, enzyme adaptation in these systems is a cellular 

 phenomenon, not a population shift. 



Further evidence in support of this conclusion comes from studies of 

 the deadaptation process, in which the enzyme content per cell decreases 

 in the absence of an inducer, at a rate far greater than could be at- 

 tributed to differential growth rates of a heterogeneous population of 

 cells. 



To analyze the system further, it is important to investigate the origin 

 of the enzyme itself. Does it appear by some subtle change in a pre- 

 formed protein, or is it synthesized de novo from its constituent amino 

 acids? This question has been decisively answered with the enzyme 

 jS-galactosidase of E. coli, and galactozymase of yeast, employing a 

 number of different methods. In one type of experiment, cells were 

 grown on glucose in a medium containing S^^ until all proteins were 

 effectively radioactive. Then the cells were washed and resuspended in 

 a nonradioactive medium containing an inducer of the enzyme. After 

 appearance of enzyme activity, /3-gaIactosidase was extracted from the 

 cells and purified. As purification proceeded, the radioactivity of the 

 preparation decreased, until finally the highly purified enzyme, charac- 

 terized both enzymatically and chemically, was virtually free of all S ^. 

 This experiment is a stringent test of the source of the sulfur-containing 

 amino acids of the enzyme, showing definitely that they come from the 

 nonradioactive amino acid precursors in the medium at the time of 

 induction, and not from pre-existing proteins within the heavily radio- 

 active cells. It may therefore be concluded that in this system at least. 



