750 R. W. HENDLER 
conditions. The column was light tan in color and iron did not wash off. Free amino 
acids however, could be removed with water. Amino acid hydroxamates formed a 
red band at the top of the column and were not eluted with water but were eluted 
with 0.3 N HCl. Palmitic hydroxamate, due to its insolubility, stayed at the top of 
the column, produced no color and was not eluted. An alcoholic HCI solution later 
revealed the fatty acid hydroxamate as a purple band which could easily be eluted 
from the column. Table II shows how some control mixtures behaved in this system. 
When various fractions of amino acid—lipid complexes (obtained from chromato- 
graphy on silicic acid) were subjected to this treatment, it was found that a high per- 
centage of the radioactivity was not in the free amino acid form, (Table III). The 
amount of radioactivity in the water wash is an upper limit since a certain amount of 
lipoidal material could be washed through in this fraction. Furthermore, there was 
a considerable amount of red material which could be eluted with 0.3 N HCl from the 
most non-polar peak from silicic acid. Upon paper-chromatographic analysis with 
authentic samples of the amino acid hydroxamates, it was found that the red Fe 
hydroxamate obtained from this tissue fraction was not in the form of free amino 
acid hydroxamates. A peptide hydroxamate was considered as a possibility, especially 
since the dilution experiments previously described indicated multisite positions for 
the amino acid valine. The presence of peptide material in this fraction was further 
indicated by the Lowry modified biuret reaction and the liberation of a complex 
spectrum of amino acids by 6 N HCI hydrolysis overnight. It was found that after 
a 15-min incubation with radioactive alanine and proline, the specific activity of the 
average alanine and proline liberated from this total fraction by hydrolysis was 15 
times higher than the average specific activity for each of these amino acids liberated 
from the total proteins. One other finding of interest was that a radioactive sugar 
moiety was also isolated from the hydrolyzate of this fraction. 
Further evidence was sought to determine if it is the carboxyl group of the amino 
acid derivative which is linked to the lipid by an ester bond. To study this question, 
we quantitatively determined the amount of hydroxamate material and amount of 
Lowry modified biuret material present in two successive pooled cuts from chromatog- 
raphy on silicic acid. It was found that the ratio of hydroxamate to biuret — FoLIN 
material was the same for both of these chromatographic fractions and we take this 
as further evidence that it is the amino acid derivative which is linked by its carboxy] 
group to the lipids. Since this material in the past has been obtained by many 
transfers in a counter-current distribution system between an aqueous acid phase 
and hexane, the amino group must also be bound. Reaction with dinitrofluoro- 
benzene followed by hydrolysis and chromatography have revealed the absence of 
free amino groups for the amino acids in general. DN P-serine was observed as would 
be expected since this lipid fraction contained cephalin. Very faint DNP-aspartic 
and glutamic acid were also observed. Free amino acids were revealed by ninhydrin 
after electrophoresis. It thus appears that one way in which the highly polar char- 
acter of amino acid compounds is suppressed involves binding both polar ends of the 
molecule. 
Certain studies are in progress using Escherichia coli, and they are still in an 
incomplete form. Since these studies represent a collaborative effort and_ will 
be reported separately when ready, I will not go into detail at this time. Ho- 
wever, in rounding out the current picture with respect to the metabolic role of 
References p. 758 
