AMINO ACID POOL FORMATION IN Escherichia colt 599 
The state of the subject at this writing is that the two simple models that have 
been widely discussed, termed by MoNnop permease and stoichiometric site, both 
fail in a qualitative way to explain the presently known facts. Following a discus- 
sion of the reasons for these failures, a new model is proposed which we have called 
the carrier model. The deductions from this model are consistent with practically all 
of the observations. 
The permease model 
The permease model is shown schematically in Fig. 6. COHEN AND MONOD state’: 
“Obviously the actual mechanism of specific permeation must be more complex than 
the deliberately bare and abstract model we have set up”. Further, they use the 
term “permease” nearly synonymously with “specific permeation mechanism”. 
However, for reasons of clarity, in this discussion, the term “permease model” will 
be restricted to a model with the essential features listed in Fig. 6. The list has been 
abstracted from ref. 8 and is written for the case of an energy-dependent amino 
acid-concentrating system. 
The permease model accounts for a great many of the observations regarding the 
concentration of substances by bacteria. There is no need here to repeat the ex- 
tensive review of the evidence given in ref. 8. It is sufficient, for the moment, to say 
that the evidence clearly shows that there are stereospecific sites which act as cata- 
lysts for the concentration of compounds by the cell. In other words, there are sites 
which participate in the early steps of the process but do not, by themselves, hold the 
compounds in the cell. Also some evidence indicates that the catalytic sites are pro- 
tein in nature. Such sites will certainly have to be retained as a part of any adequate 
model for the pool-forming process. 
The permease model predicts a number of features of the concentration process 
which are contradicted by the evidence. In order to visualize these contradictions, 
we will first consider the way in which a pool is formed according to the permease 
model, and then discuss the experimental observations listed in Table I. 
An external amino acid first forms a complex with the permease (y) in the osmotic 
barrier. In the presence of an energy supply, the complex dissociates on the inside 
of the barrier. The free amino acid within the cell may then pass out of the cell by 
means of a non-specific leak (z). Thus there will be a continuous circulation of amino 
acid through the pool at equilibrium when the two rates balance. For a given external 
concentration, the pool size will be determined by the rate at which the permease 
can pump amino acid into the cell. At the external concentration at which this rate 
saturates, the pool size will saturate. 
The observations (Table I, items 1 and 2) that glucose is not required to maintain 
a large internal concentration but is required to form a pool at normal rates are 
clearly not consistent with the above description of the pool-forming process. Simi- 
larly, the permease model fails to explain the observation (items 3 and 4) that, 
while pools are formed very slowly at 0°, pools formed before chilling are main- 
tained indefinitely (if not too large a size). From the point of view of the permease 
model these results imply that when glucose is exhausted or the temperature reduced 
to 0°, not only is the active permeation process suppressed but also the “passive” 
leak is suppressed to an identical degree. Further, the pool is maintained at 25° in the 
References p. 609 
