386 G. ROUSER ef al. 
and the host can readily reach a balance close to that of the normal individual. 
It seems probable that the deviations that have been observed from the normal 
levels on occasions in untreated patients represent changes in the equilibrium 
established between the metabolism of myeloid cells and the remainder of the body, 
and that, through various metabolic adaptations on the part of the host, the changing 
conditions produced by growth and metabolism of the leukemic cells is counter- 
balanced with a return to the normal level. When the host is unable to balance the 
changes, marked deviations and death appear to follow. Normal or nearly normal 
levels of free amino acids appear to be maintained in many of the patients until 
shortly before death when the levels may rise. This has been observed by other in- 
vestigators in experimental animals bearing tumors!*. The finding that plasma and 
cellular glutamic acid levels are frequently elevated and cellular glutamine levels 
may be quite low in some types of leukemia points to the metabolism of these com- 
pounds as being of considerable importance. The fact that aspartic acid tends to be 
elevated along with glutamic acid in leukocytes and platelets suggests that trans- 
amination to form these dicarboxylic acids may be increased in leukemia. 
The findings in patients with a generalized increase in myeloid cells illustrate 
certain basic relationships existing between cells and plasma. Erythrocytes may 
fail to show an elevation seen in plasma, although a more direct correspondence 
has been observed with dog erythrocytes. A generalized increase in plasma amino 
acids is associated with an overall elevation of leukocyte and platelet free amino 
acids without appreciable alteration of the characteristic relative amounts of the 
free amino acids. Urine, like erythrocytes, may not reflect the changes seen in plasma. 
One substance of special interest is taurine. Since taurine is a metabolic end-pro- 
duct (not metabolized further), taurine levels can be used to deduce some relation- 
ships between the levels of plasma and cellular free amino acids. The plasma of 
patients with granulocytic leukemia may show a characteristically high or low 
taurine level as compared to normal individuals or other patients with leukemia. 
This characteristic plasma level is reflected in a comparable leukocyte level. Taurine 
was not detected on chromatograms from plasma of some of the patients with 
chronic granulocytic leukemia who were studied over a period of many months. The 
leukocytes and platelets of these individuals contained approximately one-half as 
much taurine as leukocytes obtained from patients whose plasmas showed a distinct 
spot for taurine. The erythrocytes from patients whose plasma does not contain 
taurine are usually devoid of taurine. Evidently leukocytes concentrate taurine and 
contain more taurine when plasma taurine is highest. The basis for these differences 
in the taurine levels of plasma is not known, although it does appear to be character- 
istic for a patient and may change during the course of the disease. It probably re- 
presents a decreased rate of formation of taurine. 
The general lack of correspondence of the erythrocyte and plasma levels of amino 
acids is perhaps to be expected upon the basis of a relatively poor permeability for 
amino acids shown by erythrocytes?®. When normal humans ingest glycine or alanine 
the plasma levels rise, but the levels in erythrocytes do not rise as rapidly and equili- 
bration takes place over a period of several hours!®, 2°, A similar slow equilibration 
of plasma and erythrocyte glutamine has been observed (see part VI). Dog erythro- 
cytes appear to equilibrate more rapidly with plasma since HANDLER e¢ al.?” ob- 
served a rapid equilibration of glycine between dog plasma and erythrocytes, and 
References p. 447/448 
