702 HH: EAGLEVANDEK. Aq PIEZ 
for metabolites the cells can synthesize. Cultured human cells can be adapted to 
grow in the absence of added glutamine if they are exposed for varying periods of 
time to extremely high and non-physiological levels of glutamic acid!®. As DEMars 
showed!* the cells then formed increased amounts of glutamine synthetase; having 
‘adapted’, they would then grow at relatively low levels of glutamic acid. However, 
it has recently been found that such adapted cells can grow in the absence of exo- 
genous glutamine only at high population densities. If the population density is 
reduced below a critical level, even the glutamic acid-adapted cells have a glutamine 
requirement. 
Perhaps the most striking example we have encountered of a population-dependent 
requirement is the case of cystine. Cystine is one of the five amino acids required for 
the growth of mammalian cells, even though it is not required by the whole organism. 
It has developed that every serially propagated human cell so far examined can 
indeed synthesize cystine from methionine and glucose by the classic pathway 
which involves the biosynthesis of homocysteine and serine as intermediates, their 
condensation to cystathionine, and the cleavage of the latter to homoserine and 
cystine!’. However, if the cells are grown in a medium containing only methionine 
and glucose, in which the cell is under the necessity of making and retaining meta- 
bolically effective levels of homocystine, serine, cystathionine and cystine, it will 
not survive unless the population density is maintained in excess of about 500 000 
cells/ml. If the cells are given pre-formed homocystine, so that they must now 
make and retain only serine and cystathionine en route to cysteine, they will grow 
with inocula of approx. 50 000-100 000/ml. If both homocystine and serine are pro- 
vided, the critical population density becomes 50-500/ml; and if given both cystine 
and serine, one cell will grow in ro! vols. of fluid. 
The general explanation for these population-dependent requirements is probably 
the fact that at low population densities, the cells are unable to make enough of 
the specific compound to “condition” the medium, 7.e. to bring its concentration in 
the medium up to a level in equilibrium with a metabolically effective intracellular 
pool, before the cells die of what is in effect a specific amino acid deficiency. At the 
critical population density, the task of conditioning the medium with respect to 
the product is shared by enough cells so that the necessary levels can be reached 
before the cells die; and in all the situations so far examined, at that critical popula- 
tion (but not at lower levels) one does indeed find that the concentration in the 
medium attains the minimum level necessary for the growth of smaller inocula’. 
3. Closely related to the foregoing is the size of the intracellular amino acid pool 
required for protein synthesis. To determine this point, the cells had to be grown 
at extremely low external concentrations under steady state conditions, such that 
the concentration of amino acids inside and outside the cell was not changing. This 
proved possible under cloning conditions, with an extremely large volume of fluid 
relative to the volume of cells, and a large amount of amino acid relative to that 
used for protein synthesis!*. Under these circumstances, the rate of protein synthesis, 
measured by the reciprocal of the generation time, was a function of the external 
concentration of amino acids, and increased sharply with relatively small increments 
in the concentration of the external amino acid. From the data of Fig. 1 showing 
the degree to which the individual amino acids were concentrated by specific cell 
lines, one could obtain the internal concentration of amino acid corresponding to 
References p. 705 
