OPEN DISCUSSION 613 
relatively insensitive to external osmotic pressure or to external pH over the range from 6 to 8. 
That AIB entered the cell by normal pathways for amino acids was shown by the specific competi- 
tion of alanine and glycine, the relative affinities being AIB > alanine > glycine. 
The extent of leakage of internally accumulated AIB into distilled water was not great. How- 
ever, if cells were lysed there was an immediate release of most of the AIB, a small fraction 
remaining bound with the cell debris. Accumulated AIB was also released when aerobic metabo- 
lism was inhibited. AIB was taken up rapidly by protoplasts with resultant swelling. This result, 
together with the lytic release and the exchange (below), suggested the existence of the amino 
acid free in an internal pool. 
Exchange reactions between intracellular and extracellular AIB were discovered. These reac- 
tions depended on both internal and external concentrations of AIB, even at low levels of accumula- 
tion, and complete exchangeability was demonstrated. It was postulated that mobile carriers 
mediated the reaction. 
The accompanying figure summarizes diagrammatically the postulated mechanisms of AIB 
entry and exit and their relationship to surface structures of a bacillus. 


ACTIVE 
MEDIATED TRANSPORT 
EXCHANGE seal DIFFUSION 
FREE AIB 
BOUND AIB 
CELL meh IN CELL WALL 
ABSORPTION CHEMADSORPTION 
Fig. 1. Postulated mechanisms of the permeability of Bacillus megaterium to a-aminoisobutyric 
acid and the relationship to the cell wall and cytoplasmic membrane. 
CHRISTENSEN: Dr. ZABIN’s studies also have been mentioned. I wonder if we can ask him to 
comment on his work in Paris. 
ZABIN: Permease has been mentioned here. It has been defined by Dr. COHEN AND Dr. Monop in 
the case of galactoside permease as a specific inducible substance which is responsible for the 
accumulation of galactosides by certain strains of F. coli. This name implies, of course, the existence 
of an enzyme or enzyme system which carries out this process. 
The original studies of RICKENBERG, COHEN, BuTTIN AND Monop which gave the observations 
that led to this concept, used thiogalactosides as substrates because these materials were not me- 
tabolized by the enzyme /-galactosidase. Actually, it turned out that thiogalactosides are not com- 
pletely inert. Under some conditions one can show with whole cells the formation of a derivative 
which accumulates to the extent of 5 per cent or less of the total amount of thiogalactoside 
concentrated by the bacterial cell. This derivative was isolated by Dr. HERZENBERG and was shown 
to be a 6-O-acetyl thiogalactoside. One very interesting finding which was obtained by Dr. HERZEN- 
BERG and others was that only those mutant strains which could carry out the accumulation 
reaction could form the derivatives. At that time it was not possible to decide whether or not 
this was simply a coincidence. 
Kepes, Monop and I were able to demonstrate in cell-free extracts of various mutant strains 
the formation of this acetyl thiogalactoside. This was a simple reaction which required an enzyme, 
acetyl coenzyme A, and an acceptor. With this system we were then able to test a variety of 
mutants for the presence or absence of the enzymatic activity forming the acetyl galactoside in 
extracts and to compare this with the presence or absence of the permease reaction in the whole 
cells from which these extracts were derived. 
We found that there was a 1 : 1 correlation. Every time permease was induced the acetylation 
activity was present. If the strain was not induced, and, therefore, did not have permease, there 
was no acetylation activity. Since certain of these mutants were single step mutations, this sug- 
gests very strongly that the acetylation activity is very closely associated with the permease 
activity. Yet the acetyl compound cannot be an intermediate in the accumulation process, be- 
