THE OXIDATION OF GLUTAMIC ACID BY BRUCELLA ABORTUS'
A. G. MARR,2 C. B. OLSEN,3 H. S. UNGER, AND J. B. WILSON
Department of Bacteriology, University of Wisconsin, Madison, Wisconsin
Received for publication June 5, 1953
The oxidation of glutamic acid by avirulent
strains of Brucella abortus is the most rapid
respiratory process thus far studied in this
species (Gerhardt et al., 1950a). Furthermore,
glutamic acid will serve as the sole source of
both carbon and nitrogen for all strains of B.
abortus which have been tested (Gerhardt
et al., 1950b). Glutamic acid is important in the
synthesis of other amino acids by transamination
(Altenbern and Housewright, 1951), and the
evidence thus far indicates that the reductive
amination of a-ketoglutaric acid to glutamic
acid may be the chief means of amino group
synthesis.
The purpose of this study was to determine
the products of the oxidation of glutamic acid
and to investigate the intermediate steps.
METHODS
The conditions used in this laboratory for
growing and harvesting cells of B. abortus
have been reported in detail previously (Gerhardt
et al., 1950a). B. abortus, strain 11, a relatively
avirulent smooth strain closely related meta-
bolically to strain 19, was used throughout this
investigation. Metabolic reactions were followed
manometrically at 34 C and at pH 5.5 or 6.8
buffered with 0.067 M potassium phosphate.
The cells were killed either with flowing steam or
sulfuric acid and were removed by centrifugation.
Amino acids were determined by the chromato-
graphic methods of Housewright and Thorne
(1950) and of Stein and Moore (1950), ammonia
by direct nesslerization, and total nitrogen
by a modification of the method of Johnson
(1941). Total carbon was determined by wet
combustion (van Slyke and Folch, 1940, as
modified by Stutz and Burris, 1951) and citric
1 Published with the approval of the Director
of the Wisconsin Agricultural Experiment Station.
2 Present address: Department of Bacteriology,
University of California, Davis, California.
3 Present address: Dugway Proving Ground,
Tooele, Utah.
606
acid by colorimetric determination of the
pentabromo-acetone derivative (Perlman et al.,
Downloaded from http://jb.asm.org/ on December 24, 2019 by guest
1944). Pyruvic acid was determined by the
method of Friedemann and Haugen (1943) and
was identified by chromatography of the 2,4-
dinitrophenylhydrazone on paper (Warburton
et al., 1951).
RESULTS
Products of the oxidation of glutamic acid.
From a comparison of gas exchange and
ammonia production during the oxidation of
glutamic acid with the theoretical equation for
complete oxidation, it is obvious that the oxida-
tion of glutamic acid is incomplete:
(1) 1 glutamic acid + 1.8 02 --
2.3 CO2 + 0.7 NH3 (observed)
(2) 1 glutamic acid + 4.5 02 --
5.0 CO2 + 1.0 NH3 (theory)
Since the oxidation of glutamic acid is incomplete
and since both 2,4-dinitrophenol and sodium
azide stimulate the rate of oxidation, it appeared
that glutamic acid was being assimilated oxi-
datively. Both sodium azide and 2,4-dinitro-
phenol are known to inhibit oxidative assimilation
and thus stimulate oxygen uptake. Chemical
analyses of the cells and of the supernatant
liquid after oxidation of glutamic acid by washed
cells demonstrated no appreciable assimilation of
carbon or nitrogen by the cells but did show the
presence of carbon and organic nitrogen in
solution. Chromatographic analysis for amino
acids showed that all of the glutamic acid had
disappeared and the only ninhydrin positive spot
had an Rf corresponding to alanine. Since glu-
tamine has an Rf similar to alanine and thus
might be confused with alanine, the supernatant
liquid was hydrolyzed with 6 N HCl and again
analyzed. The Rf of the ninhydrin positive spot
was unchanged, and there was no evidence of
glutamic acid which would have been formed
from glutamine. Although the determination of
Rf's on paper chromatograms and on columns of
19531 OXIDATION OF GLUTAMIC ACID BY B. ABORTUS
"dowex-50" is the extent of the characterization,
it seems safe to conclude that alanine is one end
product of the oxidation of glutamic acid.
Estimations of the amounts of alanine, am-
monia, and carbon dioxide produced accounted
for almost all of the nitrogen, but a large portion
of the carbon of glutamic acid was not recovered.
This result indicated the presence of some
nonnitrogenous compound. Determination of
keto acids demonstrated the formation of
approximately 0.3 moles of pyruvic acid per
607
the gas phase. Each vessel contained approxi-
mately 1 mg cell nitrogen and 500 micromoles of
substrate in a final volume of 3.0 ml. After
incubation the supernatants were analyzed
chromatographically for amino acids. The results
demonstrated an active glutamic-alanine trans-
aminase confirming the findings of Altenbern and
Housewright (1951) with strain 19.
TABLE 1
Products of oxidation of L-glutamic acid by

Downloaded from http://jb.asm.org/ on December 24, 2019 by guest

mole of glutamic acid. The pyruvic acid was Brucella abortus, strain 11
characterized by the isolation of the 2,4-dinitro- MOLES PER MOLE GLUTAMIC ACID OXIDIZED
phenylhydrazone and by chromatographic com-
02 C02 Pyruvate Alanine* NHs
parison of this derivative with authentic keto
acid 2,4-dinitrophenylhydrazones. Carbon di- 1.9 2.1 0.28 0.45 0.64
oxide, ammonia, pyruvic acid, and alanine 1.8 _ 0.31 0.41 0.64
account for almost all of the carbon and
nitrogen of the glutamic acid oxidized (table 1). * Alanine was estimated as total nitrogen minus
Smaller amounts of other products also may be ammonia. The assumption that all of the organic
formed. nitrogen remaining after the disappearance of
glutamic acid is alanine has been confirmed by
After the products of oxidation of glutamic acid direct alanine analyses.
were identified, the following scheme for the Vessels contained 15 AM L-glutamic acid, pH
mechanism of the oxidation was postulated: 6.8.
NH3 CO2
+ +
Glutamic acid
-2H
~--> Ketoglutaric acid -211 K
-*Succinic acid
-2H
I
Alanine Fumaric acid
T
I
M1a
Pyruvic acid +- Oxalacetic acid -2H Malic acid

CO2
Most of our work on the mechanism of glutamic
acid oxidation has been designed to test this
scheme.
Transamination reactions of B. abortus. Since it
appeared possible that alanine is formed by
transamination between glutamic acid and
pyruvic acid, an investigation was made of the
transamination reactions catalyzed by resting
cells of B. abortus. The system used was essen-
tially the same as that of Altenbern and House-
wright (1951) with the exception that experiments
were performed in Warburg vessels with N2 as
Origin of the amino group of alanine. The
immediate source of the amino group of alanine
formed during the oxidation of glutamic acid
was investigated by allowing resting cells to
oxidize glutamic acid in the presence of N15-
ammonia until approximately half of the glutamic
acid had disappeared. After removing cells, the
ammonia was recovered by distillation. The
glutamic acid and alanine were separated on
"dowex-50" ion exchange resin, and the nitrogen
of the amino acids was converted to ammonia by
Kjeldahl digestion. The ammonia from all of the
6508 MARR, OLSEN, UNGER, AND WILSON [VOL. 66
samples was converted to N2 and the N'5 deter- oxidation of these two acids to pyruvic acid
mined. Controls were included to determine the with a concomitant oxidation of part of the
exchange between the amino group of alanine pyruvic acid before the break in the rate of
and ammonia. N'5 concentration in the various oxygen consumption. Unequivocal demonstra-
compounds is given in table 2. The atom per cent
excess N'5 in alanine is below the atom per cent
excess in ammonia but is considerably higher
than the atom per cent excess N15 in glutamic 15 ~ ~ 7
acid. This means that free ammonium ions
cannot be the sole source of the amino group of
o A~/°

Downloaded from http://jb.asm.org/ on December 24, 2019 by guest

alanine, but that direct amination of pyruvic
acid is probably an important source of alanine
in this rieaction. After correcting for the exchange n~~
o~~**
~ ~ ~~~~~
0~~~~~
25 ./°
TABLE 2
Incorporation of N"IH3 into glutamic acid and
alanine during the oxidation of glutamic
acid by Brucella abortus, strain 11
FRACTION 0 HR 4.0 HR 5.5 HR
50 100 150
Ammonia ............. 31.218 21.428 18.628 MINUTES
Glutamic acid. 0.188 1.193 2.923 Figure 1. The oxidation of succinic acid by
Alanine ............... 6.353 9.043 Brucella abortus. Flasks contained 10 Mm succinic
acid and 1 ml cell suspension (ca 0.5 mg nitrogen).
Ammonia, control .. - 30.218 Final pH 5.5; 0.067 M phosphate.
Alanine, control. - - 1.123
All data are in atom-per cent-excess N'5. Flasks
contained 0.5 mM L-glutamic acid. Control con- /cop
tained 0.2 mM L-alanine. All flasks contained
0.5 mm N'5H4Cl and 5 mg cell nitrogen in a final
volume of 14 ml buffered at pH 6.8.

between alanine and ammonia (alanine, control), 0

one can calculate from the concentration of 0~~~~~~~~~
0

isotope in glutamic acid and ammonia that 105 10 ISO 0

approximately 75 per cent of the alanine is
formed by transamination between pyruvic acid
and glutamic acid and 25 per cent by some other
mechanism, presumably direct amination. Ex-
change between alanine and ammonia accounts
for only a small amount of the incorporation
of N'5.
Production of pyruvic acid from proposed
intermediates. Resting cells of B. abortus, strain 11,
oxidized a-ketoglutaric, succinic, and malic
acids until the break in rate of oxygen con-
sumption, and the supernatants then were
analyzed for pyruvic acid. Approximately 0.3
mole of pyruvic acid accumulated per mole of
succinic or malic oxidized. The manometric
data on oxygen consumption and carbon dioxide
evolution (figures 1 and 2) also suggest an
50 900 ISO
MINUTES
Figure 2. The oxidation of L-malic acid by
Brucella abortus. Flasks contained 10 MM L-malic
acid and 1.0 ml cell suspension (equivalent to
ca 0.5 mg nitrogen); pH 5.5; 0.067 M phosphate.
tion of the production of pyruvic acid during the
oxidation of a-ketoglutaric acid has not been
possible.
Direct amination of a-ketoglutaric acid. If
a-ketoglutaric acid is oxidized in the presence of
0.0017 M ammonium sulfate, traces of both
19531 OXIDATION OF GLUTAMIC ACID BY B. ABORTUS
glutamic acid and alanine are produced. If,
however, succinic or malic acid is oxidized in the
presence of ammonium sulfate, there is no
demonstrable production of alanine.
Inhibitors of the oxidation of glutamic acid and
of the oxidation of proposed intermediate com-
pounds. Both arsenite and bisulfite strongly
inhibit the oxidation of glutamic acid. These
tof
-~
I
~ ~
/0'o-~~~~~~~~oX
0 1
ec
of 0 change in the presence of the inhibitor is quite
Figure S. The effect of fluoroacetic acid on the
oxidation of glutamic acid. Flasks contained
10 tAM L-glutamic acid, 0.003 M sodium fluoroace-
tate, and 1"ml cell suspension (ea 0.5 mg nitrogen).
Final pH 6.8; 0.067 M phosphate. A: C02 produc-
tion, control. B: C02 production with fluoro-
acetate. C: 02 uptake, control. D: 02 uptake with
fluoroacetate.-
inhibitor-s were used in an attempt to block the
oxidation of a-ketoglutaric acid so that the first
step in the oxidation of glutamiec acid could be
studied; however, ther-e is almost complete
inhibition of both oxygen uptake and ammonia
production. Concentr-ations of arsenite which
cause only partial inhibition did not result in the
accumulation of a-ketoglutarie acid. Apparently
the fiirst step in the reaction is most sensitive to
these inhibitors.
Inhibition of the oxidlation of succinic acid, the
6i09
next step in the proposed scheme, also has been
attempted. The dehydrogenation of succinic
acid by intact cells of B. abortus is not inhibited
by malonic acid; however, after extraction of the
cells with cold acetone or toluene, preparations
are obtained which rapidly reduce methylene
blue with glutamic or succinic acids as substrates.
The dehydrogenation of succinic acid by extracted
cells is inhibited by malonic acid. Unfortunately,
these preparations appear to be incapable of
using oxygen as a hydrogen acceptor.
Downloaded from http://jb.asm.org/ on December 24, 2019 by guest
Recently, Altenbern and Housewright (1952)
have shown that fluoroacetic acid causes the
accumulation of citric acid during the oxidation of
intermediates of the tricarboxylic acid cycle by
resting cells of B. abortus, strain 19. Figure 3
shows the effect of fluoroacetic acid on the
oxidation of glutamic acid by B. abortus, strain 1 1.
In the presence of the inhibitor the rate of
oxidation is slightly depressed; however, after
the break in initial rate of oxidation, gas ex-
low. Citric acid accumulates, and the accumula-
tion of pyruvic acid increases. It appears that
fluoroacetate partially inhibits the oxidation
of accumulated C3 acids. This contention is
supported by the finding that the oxidation of
lactic acid is greatly suppressed by fluoroacetic
acid with the accumulation of citric acid.
DISCUSSION
The scheme proposed for the oxidation of
glutamic acid by B. abortus, strain 11, contains
the assumption that at least part of the tri-
carboxylic acid cycle is operative in B. abortus.
Although none of the intermediate steps in the
degradation of the carbon chain of glutamic
acid has been demonstrated with certainty,
cells of B. abortus have been shown to oxidize
the proposed intermediates at reasonably rapid
rates. A consideration of the products of the
oxidation of glutamic acid leads to the conclusion
that the degradation of this C5 acid to the C3
acids, pyruvic acid and alanine, is more rapid
than the further metabolism of the C3 compounds.
One explanation is that pyruvic acid is oxidized
to CO2 and water via a tricarboxylic acid cycle,
but that compounds entering the cycle at the
C5 stage "jam" the cyclic process causing the
accumulation of pyruvic acid even under highly
aerobic conditions. Unpublished experiments
have shown that both glucose and lactic acid
are oxidized far beyond the stage of pyruvic acid.
 610 MARR, OLSEN, UNGER, AND WILSON
Apparently the accumulation of large amounts of
pyruvic acid occurs only during the oxidation of
members of the tricarboxylic acid cycle.
The accumulation of pyruvic acid during the
oxidation of glutamic acid and the acids of the
tricarboxylic acid cycle by B. abortus may
explain the origin of the carbon chain of alanine
which has been shown to accumulate in cultures
of B. abortus (Goodlow et al., 1950).
Transamination between glutamic acid and
pyruvic acid is one means of alanine formation
by resting cells of B. abortus (Altenbern and
Housewright, 1951). This reaction has been
demonstrated in B. abortus, strain 11, under
anaerobic conditions with both amino and keto
acids added. The data obtained by carrying out
the oxidation of glutamic acid in the presence
of isotopic ammonia indicate that both glutamic
acid and free ammonium ions are precursors of
the amino group of alanine. The incorporation of
isotopic ammonia into alanine occurs both
directly and by exchange between ammonia
and glutamic acid followed by transamination
between glutamic acid and pyruvic acid.
The synthesis of glutamic acid during the
oxidation of a-ketoglutaric in the presence of
ammonia strongly suggests direct amination of
this keto acid. Still et al. (1950) obtained synthesis
of glutamic acid with mitochondrial preparations
under similar conditions. These investigators
also reported no amino group synthesis during
the oxidation of succinic acid in the presence of
ammonia. Although we failed to detect the
formation of alanine during the oxidation of
precursors of pyruvic acid in the presence of
ammonia, the evidence obtained with isotopic
ammonia and the report of Altenbern and
Housewright (1951) indicate that direct amina-
tion of pyruvic acid occurs in B. abortus.
SUMMARY
Glutamic acid is oxidized to pyruvic acid,
alanine, carbon dioxide, and ammonia by Brucella
abortus, strain 11. Pyruvic acid also accumulates
during the oxidation of succinic and malic
acids which may be intermediates in the oxida-
tion of glutamic acid. Evidence obtained with
N"5-ammonia shows that alanine formed during
the oxidation of glutamic acid contains more
N1' than glutamic acid but much less N'5 than
ammonia. This is presumptive evidence for two
mechanisms of alanine synthesis: transamina-
tion and direct amination.
[VOL. 66.
REFERENCES
ALTENBERN, R. A., AND HOUSEWRIGHT, R. D..
1951 Alanine synthesis and carbohydrate
oxidation by smooth Brucella abortus. J.
Bact., 62, 97-105.
ALTENBERN, R. A., AND HOUSEWRIGHT, R. D.
1952 Carbohydrate oxidation and citric
acid synthesis by smooth Brucella abortus,
strain 19. Arch. Biochem. Biophys., 36,
345-356.
FRIEDEMANN, T. E., AND HAUGEN, G. E. 1943:
Downloaded from http://jb.asm.org/ on December 24, 2019 by guest
Pyruvic acid. II. The determination of
keto acids in blood and urine. J. Biol.
Chem., 147, 415-442.
GERHARDT, P. G., LEVINE, H. B., AND WILSON,
J. B. 1950a The oxidative dissimilation of
amino acids and related comnpounds by
Brucella abortus. J. Bact., 60, 459-467.
GERHARDT, P. G., TUCKER, L. A., AND WILSON,
J. B. 1950b The nutrition of brucellae:.
Utilization of single amino acids for growth.
J. Bact., 59, 777-782.
GoODLOW, R. J., MIKA, L. A., AND BRAUN, W.
1950 The effect of metabolites upon growth
and variation of Brucella abortus. J. Bact.,
60, 291-300.
HOUSEWRIGHT, R. D., AND THORNE, C. B. 1950,
Synthesis of glutamic acid and glutamyl
polypeptidebyBacillusanthracis. I. Forma-
tion of glutamic acid by transamination.
J. Bact., 60, 89-100.
JOHNSON, M. J. 1941 Isolation and properties-
of a pure yeast polypeptidase. J. Biol.
Chem., 137, 575-586.
PERLMAN, D., LARDY, H. A., AND JOHNSON, M. J.
1944 Determination of citric acid in fermen-
tation media and biological materials. Ind.
Eng. Chem., Anal. Ed., 16, 515-516.
STEIN, W. H., AND MOORE, S. 1950 Chromato-
graphic determination of the amino acict
composition of proteins. Cold Spring Har--
bor Symposia Quant. Biol., 14, 179-190.
STILL, J. L., BUELL, M. V., AND GREEN, D. E.
1950 Studies on the cyclophorase system.
VIII. Oxidation of L-glutamate. Arch. Bio-
chem., 26, 406-412.
STUTZ, R. E., AND BURRIS, R. H. 1951 Photo-
synthesis and metabolism of organic acids in
higher plants. Plant Physiol., 26, 226-243.
VAN SLYKE, D. D., AND FOLCH, J. 1940 Mano-
metric carbon determination. J. Biol. Chem.,
136, 509-546.
WARBURTON, R. H., EAGLES, B. A., AND CAMP-
BELL, J. J. R. 1951 The intermediate
metabolism of Pseudomonas aeruginosa. V.
The identification of pyruvate as an inter-
mediate in glucose oxidation. Can. J. Bot-
any, 29, 143-146.