Glutamate decarboxylase from Lactobacillus brevis: Activation by ammonium sulfate

Article abstract of DOI:10.1271/bbb.70782

In this study, the glutamate decarboxylase (GAD) gene from Lactobacillus brevis IFO12005 (Biosci. Biotechnol. Biochem., 61, 1168-1171 (1997)), was cloned and expressed. The deduced amino acid sequence showed 99.6% and 53.1% identity with GAD of L. brevis ATCC367 and L. lactis respectively. The His-tagged recombinant GAD showed an optimum pH of 4.5-5.0, and 54 kDa on SDS-PAGE. The GAD activity and stability was significantly dependent on the ammonium sulfate concentration, as observed in authentic GAD. Gel filtration showed that the inactive form of the GAD was a dimer. In contrast, the ammonium sulfate-activated form was a tetramer. CD spectral analyses at pH 5.5 revealed that the structures of the tetramer and the dimer were similar. Treatment of the GAD with high concentrations of ammonium sulfate and subsequent dilution with sodium glutamate was essential for tetramer formation and its activation. Thus the biochemical properties of the GAD from L. brevis IFO12005 were significantly different from those from other sources.

Full text of DOI:10.1271/bbb.70782

Biosci. Biotechnol. Biochem., 72 (5), 1299–1306, 2008
Glutamate Decarboxylase from Lactobacillus brevis:
Activation by Ammonium Sulfate
1
1;2
1;y
Kazumi HIRAGA, Yoshie UENO, and Kohei ODA
1
Department of Applied Biology, Graduate School of Science and Technology,
Kyoto Institute of Technology, Goshokaido-cho, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan
2
Kyoto Prefectural Technology Center for Small and Medium Enterprise,
1
34 Chudoji, Minamimachi, Shimogyo-ku, Kyoto 600-8813, Japan
Received December 4, 2007; Accepted January 13, 2008; Online Publication, May 7, 2008
[doi:10.1271/bbb.70782]
11–15)
In this study, the glutamate decarboxylase (GAD)
(Gad B),7–10) Lactococcus lactis gad B (Gad B),
1
6)
gene from Lactobacillus brevis IFO12005 (Biosci. Bio-
technol. Biochem., 61, 1168–1171 (1997)), was cloned and
expressed. The deduced amino acid sequence showed
Neurospora crassa (Gad protein), Listeria monocyto-
1
7)
genes gad A (Gad A), gad B (Gad B),
flexneri gad A (Gad A), gad B (Gad B),
Aspergillus oryzae gad A (Gad A)
Shigella
8)
1
and
have been
1
9,20)
9
9.6% and 53.1% identity with GAD of L. brevis
ATCC367 and L. lactis respectively. The His-tagged
recombinant GAD showed an optimum pH of 4.5–5.0,
and 54 kDa on SDS–PAGE. The GAD activity and
stability was significantly dependent on the ammonium
sulfate concentration, as observed in authentic GAD. Gel
filtration showed that the inactive form of the GAD was a
dimer. In contrast, the ammonium sulfate-activated
form was a tetramer. CD spectral analyses at pH 5.5
revealed that the structures of the tetramer and the
dimer were similar. Treatment of the GAD with high
concentrations of ammonium sulfate and subsequent
dilution with sodium glutamate was essential for tet-
ramer formation and its activation. Thus the biochem-
ical properties of the GAD from L. brevis IFO12005 were
significantly different from those from other sources.
reported so far, but only a few reports on the
biochemical properties of GAD have been published,
due to its intrinsically unstable nature.
Among these, E. coli GADs (two isoforms, Gad
A and Gad B) have been characterized in detail. One
of the unique points is that activity was stimulated
2
1,22)
by NaCl.
acidification of culture medium by converting glutamic
The GADs were shown to prevent
2
3)
acid to GABA (an acid-tolerance mechanism).
The crystal structures of GADs have been elucidat-
ed.
We have screened for GABA-producing microorgan-
isms from fermented foods and other sources, and have
2
2,24)
2
5)
succeeded in isolating several lactic acid bacteria. In
parallel with this screening, we selected Lactobacillus
brevis IFO12005 as a standard strain in order to compare
with the biochemical properties of GAD from other
sources, but the productivity of GAD was not sufficient
to study biochemical properties. In addition, GAD was
also unstable during purification by several column
chromatographies. Hence we carried out cloning of the
GAD gene and its expression using E. coli in order to
get large amounts of GAD. Using the recombinant
GAD, we elucidated some unique characteristics of the
GAD as activation by ammonium sulfate.
Key words: ammonium sulfate; glutamate decarbox-
ylase (GAD); gamma-aminobutyric acid
(GABA); Lactobacillus brevis
0
Glutamate decarboxylase (GAD) is a pyridoxal 5 -
phosphate (PLP) dependent enzyme that catalyzes the
irreversible alpha-decarboxylation of L-glutamate to
gamma-aminobutyric acid (GABA). GAD is widely
distributed in nature, in microorganisms, plants, and
1)
animals. GABA is an amino acid not found in proteins,
a major inhibitory neurotransmitter with hypotensive
Quite recently, gene cloning and expression of GAD
from Lactobacillus brevis OPK-3 in E. coli was report-
2
,3)
26)
and diuretic effects in animals.
)
Some GABA-con-
5)
ed, but the recombinant GAD was not purified and
4
taining foods, such as tea, beni-koji, and germinated
brown rice have been developed.
characterized. One of the reasons why that team did not
study the biochemical properties of a purified enzyme
might be the unstable nature mentioned above.
6)
Enzymatic properties and molecular cloning of GAD
genes in Escherichia coli gad A (Gad A), gad B
y
To whom correspondence should be addressed. Fax: +81-75-724-7710; E-mail: bika@kit.ac.jp
Abbreviations: GAD, glutamate decarboxylase; GABA, gamma-aminobutyric acid; PLP, pyridoxal 5 -phosphate
0

1300
K. HIRAGA et al.
Materials and Methods
was digested with Bgl II, and after treatment with
T4 polynucleotide kinase (Takara) it was ligated
into pQE-70 vector, prepared as described above, to
express a C-terminally (His)6-tagged fusion protein in
E. coli.
Restriction enzymes and DNA-modifying enzymes
were purchased from Nippon Gene (Tokyo), New
England Biolabs (Ipswich, MA), and Toyobo (Osaka,
Japan). A DIG DNA labeling kit was purchased from
Roche Diagnostics (Basel, Switzerland). A nylon mem-
brane was from Pall (East Hills, NY) and a PVDF
membrane was from Millipore (Billerica, MA). Ni-NTA
agarose and pQE-70 were from Qiagen (Hilden, Ger-
many). Other reagents were purchased from Wako Pure
Chemicals (Osaka, Japan) or Nacalai Tesque (Kyoto,
Japan).
Purification of GAD-(His)6 fusion protein. E. coli
Rosetta-gami B (DE3) pLac I cells harboring the
expression plasmid were shaken at 110 strokes/min
ꢀ
overnight at 37 C in LB medium containing 50 mg/ml
of ampicillin and 15 mg/ml of kanamycin. The overnight
culture was inoculated into the same medium. The
ꢀ
cultivation was continued with shaking at 25 C. The
fusion protein was induced by the addition of 1 mM
IPTG when the absorbance at 660 nm reached 0.6, and
Strains and culture conditions. Lactobacillus brevis
IFO 12005 was cultivated under static conditions in
glucose-yeast extract-peptone (GYP) medium contain-
ꢀ
then it was cultured overnight at 25 C. The cells,
harvested by centrifugation at 10;000 ꢁ g for 15 min at
ꢀ
ꢀ
ing 1% sodium glutamate for 24 h at 30 C. E. coli
JM109 harboring an expression plasmid was cultivated
4 C, were washed once with chilled distilled water, and
then suspended in phosphate-buffered saline (PBS)
containing protease inhibitor cocktails (Roche Diagnos-
ꢀ
in LB broth containing 50 mg/ml of ampicillin at 37 C
ꢀ
for 24 h. E. coli Rosetta-gami B (DE3) pLac I harboring
an expression plasmid was cultivated under similar
conditions with 50 mg/ml of ampicillin and 15 mg/ml of
kanamycin.
tics). The cells were disrupted by ultrasonication at 4 C,
and then triton X-100 was added (final concentration,
1%), then it was left on ice for 1 h. The supernatant
obtained after centrifugation at 15;000 ꢁ g for 20 min at
ꢀ
4
C was applied to the Ni-NTA-agarose column. The
GAD gene cloning. Genomic DNA was purified from
ꢀ
column was washed with PBS, and then the fusion
protein was eluted stepwise with 20 to 250 mM imi-
dazole-containing PBS buffer.
L. brevis IFO 12005 using the Wizard SV Genomic
DNA Purification System (Promega, Madison, WI), and
was subjected for PCR. A partial GAD gene was
amplified by PCR using KOD-plus (Toyobo) and the
Enzyme assay. Enzyme solution (0.1 ml) and 0.1 ml of
4 M ammonium sulfate were mixed. After incubation for
several min at room temperature, 1.3 ml of substrate
solution (20 mM sodium glutamate, 0.2 mM PLP, 0.2 M
pyridine-HCl, pH 4.6) were added to the enzyme solu-
tion, and then the reaction mixture was incubated at
0
following primers: 5 -CAR ATG GAR CCN CAR
0
0
GCN GAY GA-3 and 5 -GG RTA NGC NGG NAC
0
YTG CCA-3 . The amplified DNA fragment was subcl-
oned into pPCR-Script Amp predigested cloning vector
with a PCR-Script Amp cloning kit (Stratagene, La
Jolla, CA). The plasmid obtained was digested with Eco
RI, and the insert DNA (about 1 kbp) was labeled with a
DIG-labeling kit (Roche Diagnostics) and used as a
probe in screening.
A Pst I/Sal I digested genomic DNA fragment of
L. brevis IFO 12005 was ligated into the Pst I/Sal I site
of pUC18, then transformed into E. coli JM109 to
construct a partial genomic DNA library. About 500
colonies were screened by colony hybridization using a
DIG-labeled DNA probe. A 5-kbp DNA fragment was
sequenced by gene walking. A Taq dye deoxyꢁ
terminator cycle sequencing kit (Perkin Elmer) and
Perkin Elmer model 373A DNA sequencer were used in
sequencing.
ꢀ
37 C for 60 min. The reaction was stopped by boiling
for 5 min, and GABA was analyzed by TLC (Merck,
Darmstadt, Germany; solvent, n-butanol:acetic acid:
water 3:2:1 (v/v/v); detection, ninhydrin), and quanti-
fied with an amino acid analyzer (Shimazdu LC-9A,
using a Shim-pack Isc-07 Na column, Shimazdu, Kyoto,
Japan).
One unit of enzyme activity (U) was defined as
the amount of enzyme forming 1 micromole of GABA
per min.
CD spectrum. The CD spectrum (200–250 nm) of
the purified GAD (dissolved in 10 mM sodium McIlvain
buffer, pH 5.5 or 7.0) was measured with a Jasco
J-720-A spectropolarimeter (Nihon Bunko, Tokyo) at
ꢀ
Construction of expression plasmid. An expression
vector, pQE-70, was initially digested with Sph I, blunt-
ended with a DNA-blunting kit (Takara, Kyoto, Japan),
and digested with Bgl II. A GAD gene of L. brevis
IFO 12005 was amplified by PCR using the following
25 C using a quartz cuvette with a 0.1-cm path.
Gel-filtration column chromatography. A Superdex
200 10/300 GL (GE Healthcare, Uppsala, Sweden) size-
exclusion column was equilibrated with buffer (50 mM
sodium acetate buffer, pH 5.5, 0.5 M ammonium sulfate,
0.2 mM PLP, and 1 mM sodium glutamate) and calibrated
with standards (1.35–67 kDa; Bio-Rad, Hercules, CA)
0
primers: 5 -C ATG ATG AAT AAA AAC GAT CAG
0
0
GAA AC-3
CTT GTT ATC TTG-3 . The DNA fragment obtained
and 5 -GAAGATCT AAC GGT GGT
0

Glutamate Decarboxylase from L. brevis
1301
Fig. 1. Nucleotide and Deduced Amino Acid Sequences of gad B Gene from Lactobacillus brevis IFO 12005.
The nucleotide sequence and the deduced amino acid sequence are numbered on the right. The assumed ribosome-binding site is shown
by underline. The presumed catalytic amino acid residues (Asp256, His288, and Lys289) are shown by boxes. The accession number of the
gene is AB258458.
at a flow rate of 0.2 ml/min. Purified recombinant GAD
120 mg protein in the same buffer) were loaded onto
the column, and the apparent molecular weight was
that of L. brevis OPK-3.26) A deduced amino acid
sequence of GAD from L. brevis IFO 12005 showed
99.6% identity with that of L. brevis ATCC367, 53.1%
identity with that of L. lactis, and 38.1% identity with
that of E. coli (Gad B). By multiple alignment analysis
(
estimated from the standard curve.
2
7)
Results
reported by Sukhareva et al.,
highly conserved
catalytic amino acid residues were also found in the
L. brevis GAD. Therefore, Asp 256, His 288, and Lys
289 were assumed to be the catalytic amino acids. They
are shown in Fig. 1 (boxed).
GAD gene cloning
After screening of a partial DNA library, containing
about 4 to 6 kbp Pst I/Sal I genomic DNA fragments, a
5
-kbp DNA fragment containing about 1.6 kbp GAD
gene was obtained. The GAD gene of L. brevis IFO
2005 was 1,440 bp (ORF), encoding a protein of 480
Expression and purification of the GAD protein
The expression level of His-tagged GAD protein
(GAD-6xHis) in E. coli JM109 was very low (Fig. 2A),
and low GAD activity was detected. In contrast, the
expression level of GAD-6xHis increased drastically
1
amino acids (Fig. 1, accession no., AB258458). The
nucleotide sequence showed 99% identity with that of
Lactobacillus brevis ATCC367, but no matching with

1302
K. HIRAGA et al.
+
IPTG
+ IPTG
0
h 2h 4h
0h 2h 4h
(
A)
(B)
(C)
JM109
Rosetta-gami B
Fig. 2. SDS-Polyacrylamide Gel- and Western Blotting-Analyses of Recombinant GAD Protein.
A 10% polyacrylamide gel was used in analysis. Bovine serum albumin (66.2 kDa), ovalbumin (45.0 kDa), and carbonic anhydrase (31.0 kDa)
were used as protein markers. A, SDS–PAGE analysis of total cell proteins in E. coli JM109 after 0, 2, and 4 h, induction with 1 mM IPTG. B,
SDS–PAGE analysis of total cell proteins in E. coli Rosetta-gami B after 0, 2, and 4 h induction with 1 mM IPTG, sonic supernatant of E. coli
Rosetta-gami B after overnight induction with 1 mM IPTG, and recombinant GAD purified with a Ni-NTA-agarose column. C, Western blotting
analysis of purified recombinant GAD from E. coli Rosetta-gami B and purified authentic GAD from L. brevis IFO 12005.
when E. coli Rosetta-gami B (DE3) pLac I was used as
the host strain (Fig. 2B), and higher GAD activity was
detected.
A cell-free extract of the cells from 1 liter of culture
was applied to the Ni-NTA-agarose column, and about
A
1
00
8
6
4
2
0
0
0
0
0
1
0 mg of the fusion protein was purified as a single band
by Coomassie staining after SDS–PAGE (Fig. 2B). The
N-terminal amino acid sequence (five amino acid
residues) of the recombinant GAD protein was found
to be identical with that of authentic GAD. Western
blotting analysis revealed that the purified authentic and
recombinant GAD proteins were recognized by anti-
GAD antiserum raised against authentic GAD of
L. brevis IFO12005 (Fig. 2C).
0
.0
1.0
2.0
3.0 (initial)*
0.0 0.2 0.4 0.6 0.8 1.0 1.2 (final)**
(
NH
4
)
2
SO
4
conc. (M)
B
Dilution with
Na Glu soln.
+
(NH4)2 SO4
GAD
GAD*
GAD**
State 1
State 2
State 3
Characteristics of the recombinant GAD
The recombinant GAD, purified as a His-tagged
fusion protein, showed optimum pH at 4.5–5.0, and was
fully activated after initial treatment with ammonium
sulfate (> about 1.8 M) (Fig. 3A, and in Fig. 3B, state 1
to state 2) and the following addition of substrate
Fig. 3. Effects of Ammonium Sulfate on GAD Activity.
The GAD activity of the authentic GAD ( ) and recombinant
GAD ( ) was measured at various concentrations of ammonium
sulfate. The ‘‘initial’’ scale shows initial concentrations of ammo-
nium sulfate added to the enzyme solution before the addition of the
other materials, and the ‘‘final’’ scale shows final concentrations of
ammonium sulfate in the reaction mixture after the addition of all
materials, such as GAD, PLP, ammonium sulfate, and sodium
glutamate. Maximal GAD activity is shown as 100%. Values are
averages of three replicates, and standard deviations are indicated by
vertical bars.
(
Fig. 3B, state 2 to state 3), and the Km value was 1.4
mM. In contrast, authentic GAD protein showed opti-
mum pH at 4.2–4.6, and was fully activated by similar
procedures, with a Km value of 1.0 mM (Table 1).28)
The thermal stability of the recombinant GAD was
enhanced in the presence of ammonium sulfate at 0.5 M
and higher concentrations (Fig. 4).
as a single peak (Fig. 5A) corresponding to the dimer
(apparent molecular weight, 110 kDa) at pH 7.0 (a
similar single peak was also detected at pH 5.5),
Gel-filtration column chromatography
The recombinant GAD was eluted from the column

Glutamate Decarboxylase from L. brevis
1303
Table 1. Some Characteristics of Glutamate Decarboxylases from Microorganisms
Origin
Subunit
kDa)
Number of
subunit
Opt. pH
Opt. Temp.
ꢀ
Km
kcat=Km
ꢂ1 ꢂ1
References
(
( C)
(mM)
(M
s )
2
Lactobacillus brevis
(
54
4
4.2–4.6
(4.5–5.0)
30
1.0
(1.4)
0.51
1.0
1:4 ꢁ 10
28)
2
recombinant)
Lactococcus lactis
Escherichia coli
(2:6 ꢁ 10 )
54
50
33
48
—
6
4.7
4.4
5
—
—
—
60
—
—
—
—
15)
21)
29)
19)
Neurospora crassa
Aspergillus oryzae
1
2.2
13.3
6
—
4
0
1
1
20
00
A
30
8
6
4
2
0
0
0
0
0
2
0
0
0
1
0
20
40
60
80
Retention time (min)
4
3
0
0
0
5
10
B
Time (h)
Fig. 4. Effect of Ammonium Sulfate on the Stability of Purified
Recombinant GAD.
Purified recombinant GAD was adjusted to various concentrations
20
of ammonium sulfate: , 0 M; , 0.25 M; , 0.5 M; ꢁ, 1 M; and
,
M. After treatment at 37 C at pH 4.6 for the indicated periods, 4 M
1
0
ꢀ
2
ammonium sulfate solution was added to each sample in order
to adjust the ammonium sulfate concentration to 2.0 M. Then the
substrate solution was added, and the remaining activity was
measured at pH 4.6 for 1 h. Maximal GAD activity is shown as
0
0
20
40
60
80
Retention time (min)
100%. Values are the averages of three replicates, and standard
deviations are indicated by vertical bars.
Fig. 5. Gel Filtration Analysis of GADs before (A) and after (B)
Treatment with Ammonium Sulfate.
Purified recombinant GAD was analyzed by Superdex 200
1
0/300 GL column before (A) and after (B) treatment with 2 M
whereas the GAD after ammonium sulfate activation
and the following dilution with sodium glutamate
was eluted as a tetramer (apparent molecular weight,
ammonium sulfate and subsequent dilution with sodium glutamate-
containing buffer to give 0.5 M ammonium sulfate and 1 mM sodium
glutamate. For details, see text.
2
50 kDa) at pH 5.5 (Fig. 5B).
CD spectrum analysis
CD spectra of the recombinant GAD were measured
hexametric except for the GAD of Neurospora crassa
1
5,19,21,28,29)
(Table 1).
it has been reported that GAD is stimulated by NaCl.
In the case of Gad B from E. coli,
2
2)
in the far UV region (200–250 nm). The CD spectrum of
the GAD dimer (Fig. 6c; the sample used was the minor
peak in Fig. 5B) and that of the tetramer (active form, in
Fig. 6b; the sample used was the major peak in Fig. 5B)
at pH 5.5 were similar to each other, but clearly
different from that of the GAD dimer at pH 7.0 (the
inactive form, in Fig. 6a).
The halide-binding site was identified to be the bases
of the N-terminal helices, which are necessary to form
2
2)
two triple-helix bundles of the Gad B hexamer.
In order to characterize the GAD from L. brevis
IFO12005 in detail, we isolated and cloned its gene.
The deduced amino acid sequence showed considerable
similarity to Gad B of other microbial species, with
3
0)
Discussion
L. brevis ATCC367 GAD, 99.6%,
with L. lactis
1
4)
GAD, 53.1% and with E. coli Gad B, 38.1%. How-
ever, detailed biochemical data have not been reported
for the GAD from L. brevis ATCC367 or L. lactis,
except for that from E. coli.
We have reported some enzymatic properties of
2
8)
authentic GAD from L. brevis IFO12005.
The
presence of sulfate ions and dimer formation were
essential to the enzymatic activity. GADs of microbial
origin, such as E. coli and Aspergillus oryzae, are
In E. coli chromosome, two GAD isoforms, gad A
and gad B, are located separately. Another gene, gad C,

1304
K. HIRAGA et al.
4
3
2
1
0
.E+07
.E+07
.E+07
.E+07
.E+00
a
200
210
220
230
240
250
-
-
-
-
1.E+07
2.E+07
3.E+07
4.E+07
b
c
Wavelength (nm)
Fig. 6. Circular Dichroism Spectra of Purified Recombinant GAD.
Line a, Purified recombinant GAD (30 ml, 0.56 nmole) and 20 mM phosphate buffer, pH 7.0 (270 ml) were mixed, then the CD spectrum was
measured. The final concentration of the enzyme was 100 mg/ml. Line b, The GAD tetramer (0.5 M ammonium sulfate, 0.2 mM PLP, 1 mM
sodium glutamate, 50 mM acetate buffer, pH 5.5) fractionated by gel filtration was subjected to measure the CD spectrum. The final concentration
of the enzyme was 46 mg/ml. Line c, The GAD dimer (0.5 M ammonium sulfate, 0.2 mM PLP, 1 mM sodium glutamate, 50 mM acetate buffer,
pH 5.5) fractionated by gel filtration was subjected to measure the CD spectrum. The final concentration of the enzyme was 10 mg/ml.
encodes a putative glutamate/GABA antiporter. This
GAD system of E. coli is assumed to control the
acidification of cytosol by decarboxylating a glutamate
IFO12005 GAD, treatment with a high concentration
of ammonium sulfate and subsequent dilution with
glutamate solution are essential.
þ
23)
into a neutral GABA through the incorporation of H .
In Fig. 5B, the major peak of the tetramer (retention
time, 58 min) showed GAD activity, and in contrast, the
minor peak of the dimer (retention time, 68 min) did not
show any GAD activity. The dimer was not activated
by the addition of ammonium sulfate and subsequent
dilution with sodium glutamate. These results suggest
that the tetramer is an active form and the dimer is an
inactive form. We have reported that dimer formation
L. brevis IFO12005 had only a single copy gene for gad
B (data not shown), as did L. lactis. A gene similar to
gad C (accession no., AB258459) was also located very
0
close to the 5 -side of the gad B gene (42.5% identity
with L. lactis gad C). These results suggest that
L. brevis IFO 12005 has an acid tolerance mechanism
1
5)
similar to L. lactis.
2
8)
The total amount of authentic GAD purified from
L. brevis IFO12005 was 0.45 mg/l of culture, whereas
the recombinant GAD purified from E. coli Rosetta
gami B harboring the expression plasmid yielded 10
mg/l of culture. Thus we succeeded in getting enough
enzyme to study the biochemical properties of GAD.
The Km value of recombinant GAD (1.4 mM) was
slightly higher than that of authentic GAD (1.0 mM).
This difference might be due to interference of the C-
terminal His-tag of the recombinant GAD of L. brevis
IFO12005 with the creation of a proper substrate-
binding pocket.
was essential to activity, but that proved to be a
mistake. As far as we know, this is the first report
of a tetramer form of GAD from microorganisms
1
5,19,21,28,31)
(Table 1).
CD spectral analyses revealed that the structure of the
active form of GAD at pH 5.5 was different from that of
the inactive enzyme at pH 7.0. The addition of ammo-
nium sulfate did not cause any significant structural
changes at either pH level, suggesting that it did not
induce overall structural changes, but did induce subtle
structural changes at the active site, probably in the
vicinity of the presumed catalytic residues, D256, H288,
and K289, of L. brevis 12005 GAD. These residues are
well conserved among all GADs, even those with low
overall similarity.
Consequently we can say that hydrophobic interac-
tions between subunits under a high concentration of
ammonium sulfate and proper binding of sodium
glutamate to the GAD facilitate making an active
tetramer from an inactive dimer. It is very doubtful to
expect such a high hydrophobic situation in nature. In
order to understand the activation mechanism of the
GAD, further study including crystal structural analysis
is necessary.
Activation of enzyme activity by a high concentration
of ammonium sulfate (1.6 M) has been reported for
phospholipase D,3 but the mechanism has not yet been
clarified. Initially, we thought that the high concentra-
tion of ammonium sulfate might lower the pH of the
reaction mixture, and that this low pH might induce
conformational changes in L. brevis GAD. However,
acidic pH treatment (pH 4.0–5.5) of GAD without the
addition of ammonium sulfate did not result in activa-
tion even after subsequent addition of sodium glutamate.
At the second step of activation, dilution of the
ammonium sulfate-treated GAD with sodium glutamate
solution was essential, because this step was not
replaced with distilled water, 50 mM sodium acetate
1)
Acknowledgments
0
buffer (pH 5.5), or pyridoxal 5 -phosphate solution.
Thus it was found that in order to activate L. brevis
We are grateful to Dr. Alexander Wlodawer (National

Glutamate Decarboxylase from L. brevis
biol., 145, 1375–1380 (1999).
1305
Cancer Institute, USA) for critical reading, corrections,
and helpful comments on a draft of the manuscript.
1
6) Hao, R., and Schmit, J. C., Cloning of the gene for
glutamate decarboxylase and its expression during
conidiation in Neurospora crassa. Biochem. J., 293,
References
7
35–738 (1993).
1
)
Ueno, H., Enzymatic and structural aspects on glutamate
decarboxylase. J. Mol. Cat. B: Enzymatic, 10, 67–79
17) Cotter, P. D., Gahan, C. G. M., and Hill, C., A glutamate
decarboxylase system protects Listeria monocytogenes
in gastric fluid. Mol. Microbiol., 40, 465–475 (2001).
18) Waterman, S. R., and Small, P. L. C., Identification of
(
2000).
2
)
)
Roberts, E., and Frankel, S., Glutamic acid decarbox-
ylase in brain. J. Biol. Chem., 188, 789–795 (1951).
Stanton, H. C., Mode of action of gamma aminobutyric
acid on the cardiovascular system. Arch. Int. Pharma-
codyn. Ther., 143, 195–204 (1963).
s
the promoter regions and ꢁ -dependent regulation of the
3
gadA and gadBC genes associated with glutamate-
dependent acid resistance in Shigella flexneri. FEMS
Microbiol. Lett., 8, 155–160 (2003).
4
)
Lin, Z., Saito, H., Omori, M., Inomata, T., Kato, M.,
Sawai, Y., Fukatu, S., and Hakamata, K., Effect of
gabaron tea on the blood pressure and kidney function of
rats loaded with saline. J. Home Econ. Jpn. (in
Japanese), 51, 265–271 (2000).
Kono, I., and Himeno, K., Changes in ꢀ-aminobutyric
acid content during beni-koji making. Biosci. Biotech-
nol. Biochem., 64, 617–619 (2000).
19) Tsuchiya, K., Nishimura, K., and Iwahara, M., Purifica-
tion and characterization of glutamate decarboxylase
from Aspergillus oryzae. Food Sci. Technol. Res., 9,
283–287 (2003).
20) Kato, Y., Kato, Y., Furukawa, K., and Hara, S., Cloning
and nucleotide sequence of the glutamate decarboxylase-
encoding gene gadA from Aspergillus oryzae. Biosci.
Biotechnol. Biochem., 66, 2600–2605 (2002).
5
)
)
6
Saikusa, T., Horino, T., and Mori, Y., Accumulation of
gamma-aminobutyric acid (gaba) in the rice germ during
water soaking. Biosci. Biotechnol. Biochem., 58, 2291–
21) Fonda, M. L., L-Glutamate decarboxylase from bacteria.
Meth. Enzymol., 113, 11–16 (1985).
22) Gut, H., Pennacchietti, E., John, R. A., Bossa, F.,
Capitani, G., Biase, D. De, and Grutter, M. G., Esche-
richia coli acid resistance: pH-sensing, activation by
chloride and autoinhibition in GadB. EMBO J., 25,
2643–2651 (2006).
23) Hersh, B. M., Farooq, F. T., Barstad, D. N.,
Blankenhorn, D. L., and Slonczewski, J. L., A gluta-
mate-dependent acid resistance gene in Escherichia coli.
J. Bacteriol., 178, 3978–3981 (1996).
24) Dutyshev, D. I., Darii, E. L., Fomenkova, N. P., Pechik,
I. V., Polyakov, K. M., Nikonov, S. V., Andreeva,
N. S., and Sukhareva, B. S., Structure of Escherichia coli
glutamate decarboxylase (GADꢂ) in complex with
2292 (1994).
7
)
)
Tramonti, A., Biase, D. De, Giartosio, A., Bossa, F., and
John, R. A., The Roles of His-167 and His-275 in the
reaction catalyzed by glutamate decarboxylase from
Escherichia coli. J. Biol. Chem., 273, 1939–1945 (1998).
Biase, D. De, Tramonti, A., John, R. A., and Bossa, F.,
Isolation, overexpression, and biochemical characteriza-
tion of the two isoforms of glutamic acid decarboxylase
from Escherichia coli. Protein Expr. Purif., 8, 430–438
8
(
1996).
9
)
Smith, D. K., Kassam, T., Singh, B., and Elliott, J. F.,
Escherichia coli has two homologous glutamate decar-
boxylase genes that map to distinct loci. J. Bacteriol.,
˚
glutarate at 2.05 A resolution. Acta Cryst., D61, 230–
235 (2005).
174, 5820–5826 (1992).
1
0) Maras, B., Sweeney, G., Barra, D., Bossa, F., and John,
R. A., The amino acid sequence of glutamate decarbox-
ylase from Escherichia coli: evolutionary relationship
between mammalian and bacterial enzymes. Eur. J.
Biochem., 204, 93–98 (1992).
25) Hayakawa, K., Ueno, Y., Kawamura, S., Taniguchi, R.,
and Oda, K., Production of ꢀ-aminobutyric acid by lactic
acid bacteria. Seibutsu-kogakukaishi (in Japanese), 75,
239–244 (1997).
26) Park, K. B., and Oh, S. H., Cloning, sequencing and
expression of a novel glutamate decarboxylase gene
from a newly isolated lactic acid bacterium, Lactoba-
cillus brevis OPK-3. Bioresour. Technol., 98, 312–319
(2007).
1
1) Sanders, J. W., Leenhouts, K., Burghoorn, J., Brands,
J. R., Venema, G., and Kok, J., A chloride-inducible acid
resistance mechanism in Lactococcus lactis and its
regulation. Mol. Microbiol., 27, 299–310 (1998).
1
2) Sanders, J. W., Venema, G., Kok, J., and Leenhouts, K.,
Identification of a sodium chloride-regulated promoter in
Lactococcus lactis by single-copy chromosomal fusion
with a reporter gene. Mol. Gen. Genet., 257, 681–685
27) Sukhareva, B. S., and Mamaeva, O. K., Glutamate
decarboxylase: computer studies of enzyme evolution.
Biochemistry (Moscow), 67, 1428–1436 (2002).
28) Ueno, Y., Hayakawa, K., Takahashi, S., and Oda, K.,
Purification and characterization of glutamate decarbox-
ylase from Lactobacillus brevis IFO12005. Biosci.
Biotechnol. Biochem., 61, 1168–1171 (1997).
29) Hao, R., and Schmit, J. C., Purification and character-
ization of glutamate decarboxylase from Neurospora
crassa conidia. J. Biol. Chem., 266, 5135–5139 (1991).
30) Makarova, K., Slesarev, A., Wolf, Y., Sorokin, A.,
Mirkin, B., Koonin, E., Pavlov, A., Pavlova, N.,
Karamychev, V., Polouchine, N., Shakhova, V.,
Grigoriev, I., Lou, Y., Rohksar, D., Lucas, S., Huang,
K., Goodstein, D. M., Hawkins, T., Plengvidhya, V.,
Welker, D., Hughes, J., Goh, Y., Benson, A., Baldwin,
(
1998).
1
1
1
3) Sanders, J. W., Venema, G., and Kok, J., A chloride-
inducible gene expression cassette and its use in induced
lysis of Lactococcus lactis. Appl. Environ. Microbiol.,
63, 4877–4882 (1997).
4) Nomura, M., Kobayashi, M., Ohmomo, S., and
Okamoto, T., Inactivation of the glutamate decarbox-
ylase gene Lactococcus lactis subsp. cremoris. Appl.
Environ. Microbiol., 66, 2235–2237 (2000).
5) Nomura, M., Nakajima, I., Fujita, Y., Kobayashi, M.,
Kimoto, H., Suzuki, I., and Aso, H., Lactococcus lactis
contains only one glutamate decarboxylase gene. Micro-

1306
K. HIRAGA et al.
K., Lee, J.-H., D ´ı az-Mu n˜ iz, I., Dosti, B., Smeianov, V.,
the lactic acid bacteria. Proc. Natl. Acad. Sci. USA, 103,
15611–15616 (2006).
Wechter, W., Barabote, R., Lorca, G., Altermann, E.,
Barrangou, R., Ganesan, B., Xie, Y., Rawsthorne, H.,
Tamir, D., Parker, C., Breidt, F., Broadbent, J., Hutkins,
R., O’Sullivan, D., Steele, J., Unlu, G., Saier, M.,
Klaenhammer, T., Richardson, P., Kozyavkin, S.,
Weimer, B., and Mills, D., Comparative genomics of
31) Nakamura, S., Shimooku, K., Akisue, T., Jinnai, H.,
Hitomi, T., Kiyohara, Y., Ogino, C., Yoshida, K., and
Nishizuka, Y., Mammalian phospholipase D: activation
by ammonium sulfate and nucleotides. Proc. Natl. Acad.
Sci. USA, 92, 12319–12322 (1995).