Characterization of glutamate decarboxylase from a high γ-aminobutyric acid (GABA)-producer, Lactobacillus paracasei

Article abstract of DOI:10.1271/bbb.70163

γ-Aminobutyric acid (GABA) has several physiological functions in humans. We have reported that Lactobacillus paracasei NFRI 7415 produces high levels of GABA. To gain insight into the higher GABA-producing ability of this strain, we analyzed glutamate decarboxylase (GAD), which catalyzes the decarboxylation of L-glutamate to GABA. The molecular weight of the purified GAD was estimated to be 57 kDa by SDS-PAGE and 110 kDa by gel filtration, suggesting that GAD forms the dimer under native conditions. GAD activity was optimal at pH 5.0 at 50°C. The Km value for the catalysis of glutamate was 5.0 mM, and the maximum rate of catalysis was 7.5 μmol min^-1 mg^-1. The N-terminal amino acid sequence of GAD was determined, and the gene encoding GAD from genomic DNA was cloned. The findings suggest that the ability of Lb. paracasei to produce high levels of GABA results from two characteristics of GAD, viz., a low Km value and activity at low pH.

Full text of DOI:10.1271/bbb.70163

Biosci. Biotechnol. Biochem., 72 (2), 278–285, 2008
Characterization of Glutamate Decarboxylase from a High ꢀ-Aminobutyric
Acid (GABA)-Producer, Lactobacillus paracasei
y
1;
Noriko KOMATSUZAKI,1;2 Toshihide NAKAMURA,1 Toshinori KIMURA,2 and Jun SHIMA
1National Food Research Institute, 2-1-2 Kannondai, Tsukuba, Ibaraki 305-8642, Japan
2Graduate School of Agriculture, Hokkaido University, 9 Kita 9-jo nishi, Kita-ku, Sapporo,
Hokkaido 060-8589, Japan
Received March 20, 2007; Accepted October 23, 2007; Online Publication, February 7, 2008
[doi:10.1271/bbb.70163]
ꢀ-Aminobutyric acid (GABA) has several physiolog-
ical functions in humans. We have reported that
Lactobacillus paracasei NFRI 7415 produces high levels
of GABA. To gain insight into the higher GABA-
producing ability of this strain, we analyzed glutamate
decarboxylase (GAD), which catalyzes the decarboxyl-
ation of L-glutamate to GABA. The molecular weight of
the purified GAD was estimated to be 57 kDa by SDS–
PAGE and 110 kDa by gel filtration, suggesting that
GAD forms the dimer under native conditions. GAD
activity was optimal at pH 5.0 at 50 ꢀC. The Km value
for the catalysis of glutamate was 5.0 mM, and the max-
imum rate of catalysis was 7.5 ꢀmol minÀ1 mgÀ1. The N-
terminal amino acid sequence of GAD was determined,
and the gene encoding GAD from genomic DNA was
cloned. The findings suggest that the ability of Lb. par-
acasei to produce high levels of GABA results from two
characteristics of GAD, viz., a low Km value and activity
at low pH.
The production of GABA in Lb. paracasei was superior
to that in Lb. brevis, a known GABA producer, but
GABA production by Lb. paracasei has not yet been
characterized. We consider Lb. paracasei to be a good
model for investigating the high levels of GABA
production in LAB.
Glutamate decarboxylase (GAD; EC 4.1.1.15) is the
enzyme that catalyzes the decarboxylation of L-gluta-
mate to GABA. It has been reported that GAD is present
in the mammalian brain,11,12) plants,13,14) Escherichia
coli,15) Aspergillus,16) and LAB.17) The purification and
enzymatic properties of GAD in such LAB species as
Lactobacillus and Lactococcus have been reported.
Ueno et al.18) purified GAD from Lb. brevis and then
determined its biochemical characteristics. Park and
Oh19) reported the cloning and sequencing of the GAD
gene from a newly isolated Lb. brevis. Nomura et al.20)
characterized GAD and cloned the gene responsible for
GAD activity in L. lactis, and found that L. lactis
contained one gene encoding GAD.
Key words: glutamate decarboxylase (GAD); Lacto-
bacillus paracasei; ꢀ-aminobutyric acid
(GABA)
In this report, we describe a four-step procedure for
the purification of Lb. paracasei GAD to homogeneity,
and we discuss our investigation of its biochemical
characteristics. The biochemical characteristics of
Lb. paracasei GAD were compared with those of
GAD from other LAB species. We also explain the
cloning and sequence of the gene encoding GAD.
ꢀ-Aminobutyric acid (GABA) is a nonprotein amino
acid widely distributed in nature.1) It has several
physiological functions, such as neurotransmission and
the induction of hypotensive, diuretic, and tranquilizer
effects.2,3) Due to the physiological functions of GABA,
the development of functional foods containing GABA
in high concentrations has been actively pursued.4,5)
GABA production by various microorganisms has been
reported, including bacteria,6,7) fungi,8) and yeasts.9)
We focused on the GABA production ability of lactic
acid bacteria (LAB) because LAB possess commercial
potential as a starter of production in fermented foods
such as pickled vegetables and fermented meats and
fishes. We have reported that Lactobacillus paracasei
isolated from a traditional Japanese fermented fish
( funa-sushi) showed high GABA-producing ability.10)
Materials and Methods
Strains and media. Lb. paracasei NFRI7415 was used
in enzyme production and gene cloning. Escherichia
coli DH5ꢁ was used in plasmid manipulations.
Lactobacilli MRS medium (Difco Laboratories,
Detroit, MI) was used in the cultivation of Lb. paraca-
sei. LB medium (0.5% yeast extract, 1% tryptone, and
1% NaCl) was used in the cultivation of E. coli.
Purification of GAD. A pre-culture of Lb. paracasei
was grown to the stationary phase at 37 ꢀC for 20 h in
y
To whom correspondence should be addressed. Fax: +81-29-838-7996; E-mail: shimaj@affrc.go.jp

Characterization of Glutamate Decarboxylase of Lb. paracasei
279
MRS medium. Two ml of the pre-culture was inoculated
into 1 liter of MRS medium containing 100 mM of
glutamate, and the inoculate was incubated at 37 ꢀC for
72 h. Our previous study showed that GABA production
occurred in MRS medium after a cultivation of period of
at least 72 h.10) After cultivation, the cells were collected
by centrifugation (8;000 ꢁ g for 10 min at 4 ꢀC) and
washed twice with phosphate-buffered saline (PBS;
pH 7.0) containing 8 g of NaCl, 0.2 g of KCl, 1.44 g of
Na2HPO4, and 0.24 g of KH2PO4 per liter. The cells
were suspended in 50 ml of PB (phosphate buffer (PB):
20 mM sodium phosphate pH 7.0), containing 0.1 mM
pyridoxal 50-phosphate (PLP) and 0.1 mM 2-mercaptoe-
thanol. The suspension was incubated with 0.2 mg/ml of
lysozyme at 37 ꢀC for 15 min. After incubation, 10 g of
glass beads (100 mm in diameter, Merck, Darmstadt,
Germany) were added to the cell suspension, and the
cells were disrupted by shaking for 30 s (10 times) using
a homogenizer (type MSK, Braun, Frankfurt, Germany)
at 4 ꢀC. Crude extracts were obtained by centrifugation
at 10;000 ꢁ g for 10 min at 4 ꢀC.
system: CH3OH: CHCl3 (9.5:0.5, v/v). An amino acid
analyzer was used specifically to measure GAD activity
as the amount of GABA produced. One unit of enzyme
activity was defined as the amount of enzyme that
produced 1 mmol of GABA in 1 min. The protein
concentration was determined with BCA protein assay
reagent (Pierce, Rockford, IL).
SDS–PAGE of purified GAD. SDS–PAGE of fractions
eluted by column chromatography was performed by the
method of Laemmli21) using 10% polyacrylamide gel.
The standard marker of molecular weight was obtained
from Bio-Rad Laboratories (Hercules, CA). After
electrophoresis, the gels were stained with 0.1% Coo-
massie brilliant blue R-250 (CBB) and destained with
30% methanol.
N-Terminal amino acid sequence. The purified GAD
was separated on 10% SDS–PAGE gel and electro-
transferred to a PVDF membrane. The membrane was
stained with 0.1% CBB, destained with 60% methanol,
and washed in distilled water. The band of GAD was
sliced from the membrane, and its N-terminal sequence
was determined using a protein sequencer (HP G1005A
protein sequencing system, Hewlett Packard, Palo Alto,
CA).
Ammonium sulfate was added to the crude extract to
give a final concentration that was 80% saturated. The
precipitate was collected by centrifugation (10;000 ꢁ g
for 10 min at 4 ꢀC) and dissolved in 5 ml of PB. The
enzyme solution was dialyzed against PB at 4 ꢀC
overnight. The dialysate sample was adjusted to 0.4 M
ammonium sulfate by adding PB containing 0.8 M
ammonium sulfate. The sample was applied to a column
(4 ꢁ 80 cm) of Butyl-Toyopearl 650M (Tosoh, Tokyo)
equilibrated with PB containing 0.8 M ammonium
sulfate. The enzyme was eluted with a linear gradient
of ammonium sulfate (0.8 to 0 M). The fractions
containing GAD activity (18 ml) were pooled, and
ultrafiltration was performed at 4 ꢀC for the concen-
tration of proteins. Then the samples were loaded onto
DEAE-Sepharose FF columns (GE Healthcare Bio-
Science, Piscataway, NJ) equilibrated with PB. Each
column was washed with PB, and the enzyme was eluted
with a linear gradient of NaCl (0 to 1 M). The fractions
containing GAD activity (16 ml) were pooled and
dialyzed against PB. The dialysate was loaded onto
Mono Q 5/50 GL columns (GE Healthcare Bio-Science)
equilibrated with PB, and the samples were eluted with a
linear gradient of NaCl (0.4 to 0.6 M). The fraction
containing GAD activity (2 ml) was used in further
analyses.
Analytical gel filtration. The purified GAD (0.2 ml)
was loaded on a column of Sephacryl S-100 HR 16/60
equilibrated with PB containing 0.15 M NaCl and EDTA,
and eluted with the same buffer at a flow rate of 0.5
ml/min. ꢂ-Amylase (200 kDa), alcohol dehydrogenase
(150 kDa), bovine serum albumin (66 kDa), carbonic
anhydrase (29 kDa), and cytochrome c (12.4 kDa) were
used as molecular weight standards. All standards were
purchased from Sigma-Aldrich (St. Louis, MO).
Effect of temperature and pH on GAD activity. The
optimum temperature of GAD activity was determined
during a 30 min exposure to a temperature range of
20 ꢀC to 60 ꢀC in 0.2 M pyridine-HCl buffer (pH 5.0)
with 50 mM of glutamate and 10 mM of PLP. The thermal
stability of the enzyme was determined by incubating
the enzyme in a temperature range of 20 ꢀC to 60 ꢀC for
1 h at pH 5.0, and the residual activity of the enzyme
was measured at 37 ꢀC.
The optimum pH of pH stability of GAD activity was
determined within a range of pH 3.5 to 6.0 in 0.2 M
pyridine-HCl buffer supplemented with 50 mM of glu-
tamate and 10 mM PLP at 37 ꢀC. The pH stability of the
enzyme was determined by incubating it at each pH with
0.2 M pyridine-HCl buffer for 1 h at 4 ꢀC, and its residual
activity was measured.
Enzyme and protein assays. An enzyme solution was
incubated with 50 mM of glutamate and 10 mM of PLP in
0.2 M pyridine-HCl buffer (pH 5.0) at 37 ꢀC for 10 min.
The enzyme reactions were stopped by boiling the
samples for 5 min. The reaction mixture was then
analyzed by thin-layer chromatography (TLC) or with
an amino acid analyzer (LC-11A, Yanagimoto, Kyoto,
Japan). To determine the activity in the fractions eluted
by column chromatography, TLC using Silica gel 60
F245 (Merck) was carried out with the following solvent
Cloning of the gene encoding GAD. Genomic DNA
was isolated from Lb. paracasei NFRI 7415 using an
InstaGene kit (Bio-Rad) and was then used as the
template DNA in PCR cloning. To amplify the internal

280
N. KOMATSUZAKI et al.
fragments of the gene encoding GAD, primers designed
from highly conserved regions of GAD were used. The
nucleotide sequences of these degenerate primers were
as follows: forward (F3) 50-CAR(A/G)GTN(A/C/G/T)-
TGY(C/T)TGGGAR(A/G)AA-30 and reverse (R6) 50-
GGR(A/G)TAN(A/C/G/T)ACN(A/C/G/T)AR(A/G)N-
(A/C/G/T)CCR(A/G)TAY(C/T)TTR(A/G)TG-30. PCR
was initiated with 4 min of denaturation at 94 ꢀC, fol-
lowed by 30 cycles of 1 min at 94 ꢀC, 1 min at 40 ꢀC, and
1 min at 72 ꢀC using a Gene Amp PCR System (Applied
Biosystems, Lincoln, CA). The PCR product was
purified using a gel extraction kit (Qiagen, Hilden,
Germany). The purified product was ligated into the
pTA2 vector (Toyobo, Osaka, Japan) using T4 ligase
(Toyobo), and the ligated vector was then introduced
into E. coli DH5ꢁ. The cloned DNA was sequenced as
described by Sanger et al.22) using an ABI PRIZM 310
genetic analyzer with a BigDye Terminator kit (Applied
Biosystems).
kDa
150
M
F14 F15 F16 F17
100
75
GAD
50
37
Fig. 1. SDS–PAGE of Factions (F14-F17) Exhibiting GAD Activity
Eluted with Mono-Q Column Chromatography.
The fractions (F14-F17) eluted by Mono-Q column chromatog-
raphy were subjected to SDS–PAGE. M, molecular mass standard.
To isolate the full-length GAD gene, inverse PCR was
carried out.23) Restriction enzymes suitable for digestion
of genomic DNA were selected by Southern blot
analysis. Self-ligation was carried out using the genomic
DNA digested with PstI.
proximately 57 kDa, in fraction 16; this result indicated
that GAD was purified to homogeneity (Fig. 1). The
purification of GAD in Lb. paracasei is summarized in
Table 1. The data revealed that the specific activity of
GAD increased approximately 27-fold, with a yield of
6.2%. We used the eluted active fraction (fraction 16)
from Mono-Q column chromatography as the purified
enzyme preparation for the subsequent studies.
The molecular weight of native GAD was determined
by gel filtration analysis. A single peak with GAD
activity was detected at a molecular weight of approx-
imately 110,000 (data not shown). These results, taken
together with those of the SDS–PAGE analysis, reveal
that the Lb. paracasei GAD forms a dimer under native
conditions.
Primers for the inverse PCR primers were designed
based on the nucleotide sequence of the cloned internal
fragment and the N-terminal amino acid sequence
determined by protein sequencing. The sequences of
the primers were as follows: gF3, 50-GACTTGGAA-
GGTGAACTTGC-30; gN1-2, 50-GCAATACTAGCTG-
GCATTGG-30; gF6, 50-GGAGGATCTAAAGATGGC-
CA-30; and gN1-1, 50-GGCAACTCTTCAGTTGGCAA-
30. The first PCR was carried out using gF3 and gN1-2
primers and self-ligated genomic DNA as the template.
The PCR cycling conditions were 94 ꢀC for 5 min, fol-
lowed by 30 cycles of 94 ꢀC for 1 min, 55 ꢀC for 1 min,
and 72 ꢀC for 3 min. The products of the first PCR were
used as a template for the second PCR. The second PCR
was carried out using gF6 and gN1-1 primers. The
second PCR product was purified and ligated into
pGEM-T Easy vector (Promega, Madison WI). The
DNA sequence of cloned fragments was determined, and
the sequence was analyzed with Genetyx software
(Software Development, Tokyo) or Clustal W software.
Properties of GAD in Lb. paracasei
The pH and temperature dependencies of GAD
activity were examined, and the results are shown in
Fig. 2. The findings revealed optimum activity at pH 5.0
(Fig. 2A), and the optimum temperature for GAD
activity was observed at 50 ꢀC with 30-min reactions
(Fig. 2B). The optimum pH for GAD activity in
Lb. paracasei differed from that previously reported
for GAD from other LAB.18,20) The enzyme was active
within a broad pH range as compared with GADs from
other LAB species.
Results
The pH and thermal stability of GAD activity are
shown in Fig. 2C and D, respectively. The enzyme was
stable within a pH range of 4.5 to 5.0, but was unstable
below pH 4.0 and above pH 5.5. It was found to be
stable within a temperature range of 20 to 40 ꢀC, as
shown in Fig. 2D.
The influence of salts on the activity of Lb. paracasei
GAD was also investigated. We found that 10 mM of
ammonium ions and 10 mM of calcium ions enhanced
GAD activity by 132% and 114% respectively, as
compared with the activity in the no-addition control.
However, 10 mM of sodium chloride inhibited GAD
Purification of GAD from Lb. paracasei
Purification of GAD in Lb. paracasei was attempted
using the following approaches: ammonium sulfate
precipitation, hydrophobic chromatography, and anion-
exchange chromatography (viz., DEAE-Sepharose and
Mono-Q chromatography). GAD was eluted by Mono-Q
chromatography from the fraction in which the NaCl
concentration was 0.44 M. The purity of fractions
containing GAD activity was confirmed by SDS–PAGE.
The SDS–PAGE analysis of the fractions detected a
single band, the molecular weight of which was ap-

Characterization of Glutamate Decarboxylase of Lb. paracasei
Table 1. Purification of GAD from Lb. paracasei
281
Total
protein
(mg)
Total
activity
(U)
Specific
activity
(U/mg)
Yield
(%)
Purification
(fold)
Purification step
Crude extract
199
33
124
52
0.62
1.55
3.86
11.6
16.7
100
41.9
26.8
9.2
1.0
2.5
Ammonium sulfate
Butyl-Toyopearl
DEAE-Sepharose
Mono-Q
8.6
1.0
0.2
33
6.2
11.4
3.2
18.7
26.9
6.2
A
B
100
100
80
60
40
20
0
80
60
40
20
0
20
30
40
50
60
3.5 4.0 4.5 5.0 5.5 6.0
pH
Temperature (°C)
C
D
100
100
80
60
40
20
0
80
60
40
20
0
20
30
40
50
60
3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0
Temperature (°C)
pH
Fig. 2. Effects of pH and Temperature on GAD Activity.
A, Optimum pH. Enzymatic activity was measured in 0.2 M pyridine-HCl buffer at various pHs. B, Optimum temperature. The activity of the
enzyme was measured in 0.2 M pyridine-HCl buffer for 30 min, pH 5.0, at various temperatures. C, Effects of pH. The activity of the enzyme was
measured in 0.2 M pyridine-HCl buffer at 37 ꢀC after preincubation of GAD at different pHs at 4 ꢀC for 1 h. D, Effects of temperature. The
activity of the enzyme was measured in 0.2 M pyridine-HCl buffer at 37 ꢀC after preincubation of GAD at different temperatures and at pH 5.0
for 1 h. Relative GAD activities are shown, with the maximum level of activity represented as 100%. Data from three independent experiments
are expressed as mean ꢃ SD.
activity, and the residual activity was 38.5%. GAD
activity was also inhibited by 1 mM and 10 mM of
EDTA; residual activities of 66.8% and 44.5% respec-
tively were observed. These results suggest that other
ions were involved in GAD activity. The Km and Vmax
values were determined for the reaction using glutamate
as a substrate, and the values were calculated from a
Lineweaver-Burk plot. The Km and Vmax of the enzyme
were calculated to be 5.0 mM and 7.5 mmol minꢂ1 mgꢂ1
respectively.
homology with the N-terminal sequence of GAD were
found.
Cloning and sequencing of the gene encoding GAD
Cloning of the gene encoding the GAD of Lb. par-
acasei was attempted. First an internal fragment of the
gene was obtained by PCR using degenerate primers
designed based on the conserved sequences in GAD
from Lb. brevis, L. lactis, and Lb. plantarum. A 360-bp
fragment was obtained and sequenced (data not shown).
The amino acid sequence deduced from the DNA
sequence of the internal fragment showed high similarity
with the sequence GAD of L. lactis and Lb. plantarum.
To isolate the full-length gene encoding GAD, inverse
PCR was carried out using primers designed based on
the determined N-terminal amino acid sequence and the
nucleotide sequences of the internal fragments. The PCR
Sequencing of N-terminal amino acids
The N-terminal sequence of GAD produced by
Lb. paracasei was determined with a protein sequencer.
It appeared that the N-terminal sequence was SEKN-
DEQMI. Protein homology searches using BLAST were
carried out, but no sequences that showed higher

282
N. KOMATSUZAKI et al.
Fig. 3. Comparison of the Deduced Amino Acid Sequence of GadB of Lb. paracasei with the GAD Sequences of Other Lactic Acid Bacteria.
The deduced amino acid sequence was analyzed using Clustal W (1.83). Residues conserved in GAD from LABs are marked by asterisks. The
arrow indicates the active site for PLP binding.
product containing the full coding region of the gene
was obtained. The nucleotide sequencing suggested that
this PCR product contained the full length of the gene
encoding GAD (designated gadB). The results indicate
that the gadB gene consists of 1,443 bases, which
encode a protein of 481 amino acid residues. The
predicted molecular mass of the protein was 54.3 kDa.
The N-terminal sequence deduced from the sequence of
gadB and the sequence determined by protein sequenc-
ing were identical. Comparison of the amino acid
sequence deduced from gadB with the protein sequences
of other forms of GadB from other LAB demonstrated
that the gadB product shared a high level of homology
with GadB from other LAB (Fig. 3). A computer-based
homology search using BLAST revealed that the
sequence of GadB shared 79% and 52% homology with
the GAD of Lb. brevis ATCC 367 and the GAD of
L. lactis subsp. lactis 01-7 respectively. The product of
gadB of Lb. paracasei has the consensus lysyl residue
known to be the active site of PLP binding (Fig. 3).
The sequence of the gadB gene of Lb. paracasei was
deposited in the DDBJ database under accession
no. AB295641.
Discussion
We have reported that Lb. paracasei NFRI 7415
produces relatively high levels of GABA.10) In this

Characterization of Glutamate Decarboxylase of Lb. paracasei
283
Table 2. Properties of GAD from Lb. paracasei, Lb. brevis, and L. lactis
Lb. paracaseia
Lb. brevisb
L. lactisc
Molecular weight (kDa)
SDS–PAGE
Gel filtration
57
110
60
120
54
ND
Optimum pH
Optimum temperature
Km value (mM)
5.0
4.2
4.7
ND
50 ꢀC
5.0
30 ꢀC
9.3
0.51
ND
Activation by 10 mM NaCl
Activation by 10 mM (NH4)2SO4
Activation by 10 mM CaCl2
Production of GABA (mg/ml)
[glutamate (mM)d]
ꢂ
+
ꢂ
+
ND
ND
ND
69.6 [0]25)
54.6 [20]29)
+
6180 [100]10)
950 [10]10)
5090 [100]10)
ND, not determined.
aData obtained in this study.
bData from Ueno et al.18)
cData from Nomura et al.20)
dGlutamate concentrations in the growth medium.
study, we examined the biochemical characteristics
of glutamate decarboxylase (GAD) in Lb. paracasei.
Table 2 summarizes the biochemical characteristics of
GAD from Lb. paracasei and describes its differences
from Lb. brevis (a high GABA-producing strain) and
L. lactis (a low GABA-producing strain) GAD. The
molecular weight of GAD was estimated to be 57 kDa
by SDS–PAGE and 110 kDa by gel filtration. These data
strongly suggest that the GAD of Lb. paracasei forms a
dimer under native conditions. GAD dimer formation
has also been reported in Lb. brevis.18) In Lactobacillus
species, the dimer formation of GAD might be con-
served, whereas the GAD of E. coli forms a hexamer.24)
The optimum temperature for the GAD activity of
Lb. paracasei was 50 ꢀC, much higher than that of
Lb. brevis. This optimum temperature is potentially of
advantage in commercial applications of this enzyme,
because the higher temperature prevents the growth of
contaminants. The enzyme thus might be useful in
GABA production using a bio-reactor or batch reaction.
However, because the enzyme was not stable at 50 ꢀC,
further improvement by modification or mutation may
be necessary to achieve greater stabilization at 50 ꢀC.
The optimal pH of GAD in Lb. paracasei was pH 5.0,
somewhat higher than that of either Lb. brevis or
L. lactis, but the pH range for the maintenance of
activity was relatively broad. The GAD activity showed
a bell-shaped pH profile from pH 4.5 to 5.5, but it lacked
pH dependence between 4.0 and 4.5. Because GAD
activity at pH 4.0 is relatively high, GAD does not
exhibit a typical bell-shaped pH profile. In high GABA-
producing strains (Lb. paracasei and Lb. brevis), GAD
activity was still observed at pH 4.0,18) but very low
levels of GAD activity have been observed at pH 4.0 in
a low GABA-producing strain (L. lactis).25) These
results suggest that low-pH GAD activity might be
important for producing high levels of GABA in LAB.
In LAB and E. coli, GABA production is among the
most important mechanisms for achieving acid resist-
ance.26–28) Since Lb. paracasei increases acidity via
growth, it has been suggested that it produces GABA in
order to increase the pH of the growth medium.10)
The Km value of GAD from Lb. paracasei was much
lower than that from Lb. brevis. This characteristic
might account for the effective conversion from gluta-
mate to GABA, but the Km value of GAD from L. lactis
was lower than that from Lb. paracasei. In L. lactis, the
pH range for GAD activity was relatively narrow. Hence
high GAD activity at optimum pH might be required for
GABA synthesis in L. lactis. To produce high amounts
of GABA, both low pH activity and a low Km value
might be necessary GAD characteristics.
We cloned the gene encoding GAD (gadB) of
Lb. paracasei, and found that the predicted molecular
mass (54.3 kDa) was similar to that of the dimerized
native form (110 kDa). The deduced amino acid se-
quence was compared with those of GAD from other
LAB. Although these genes shared a high level of
homology, the N-terminal and C-terminal regions were
not conserved (Fig. 3). It is possible that differences in
primary structure might contribute to the high GABA-
producing ability of some species. We searched the
conserved sequence upstream of gadB and found a
ribosome binding sequence (GGAGG), but we did not
find any possible promoter sequences. We speculated
that gadB and the other genes located upstream of it
might have an operon structure. It has been reported that
gadB and gadC (encoding the antiporter that exchanges
glutamate for GABA) form an operon structure in
L. lactis.20) At present, it is unknown whether gadC
exists upstream of gadB in Lb. paracasei.
In this study, we identified the biochemical character-
istics and putative primary structure of GAD. Various
forms of GAD from high GABA-producing strains
(Lb. paracasei and Lb. brevis) were found to have
similar characteristics, but there were some notable
differences in terms of optimum pH and temperature.
Possible reasons for the high levels of GABA production

284
N. KOMATSUZAKI et al.
J. Y., Purification and characterization of a novel form of
brain L-glutamate decarboxylase. J. Biol. Chem., 269,
7249–7254 (1994).
by Lb. paracasei include GAD properties such as low
pH activity and high affinity to glutamate. Expression of
gadB might be important for achieving relatively high
GABA production. To gain further insight into the high
GABA productivity of this strain, a gene expression
analysis of gadB and analysis of the transcriptional and
translational regulation of GAD are currently underway.
13) Johnson, B. S., Singh, N. K., Cherry, J. H., and Locy,
R. D., Purification and characterization of glutamate
decarboxylase from cowpea. Phytochemistry, 46, 39–44
(1997).
14) Oh, S. H., Choi, W. G., Lee, I. T., and Yun, S. J.,
Cloning and characterization of a rice cDNA encoding
glutamate decarboxylase. J. Biochem. Mol. Biol., 38,
595–601 (2005).
Acknowledgment
This study was supported in part by a grant from a
Ministry of Agriculture, Forestry, and Fisheries (MAFF)
Food Research Project: ‘‘Integrated Research on Safety
and Physiological Function of Food.’’
15) Rice, E. W., Johnson, C. H., Dunnigan, M. E., and
Reasoner, D. J., Rapid glutamate decarboxylase assay
for detection of Escherichia coli. Appl. Environ. Micro-
biol., 59, 4347–4349 (1993).
16) 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).
References
1) Manyam, B. V., Katz, L., Hare, T. A., Kanifefski, K.,
and Tremblay, R. D., Isoniazid-induced elevation of
cerebrospinal fluid (CSF) GABA levels and effects on
chorea in Huntington’s disease. Ann. Neurol., 10, 35–37
(1981).
2) Jakobs, C., Jaeken, J., and Gibson, K. M., Inherited
disorders of GABA metabolism. J. Inherit. Metab. Dis.,
16, 704–715 (1993).
17) Bertoldi, M., Carbone, V., and Voltattorni, C. B.,
Ornithine and glutamate decarboxylase catalyse an
oxidative deamination of their ꢁ-methyl substrates.
Biochem. J., 342, 509–512 (1999).
18) Ueno, Y., Hayakawa, K., Takahashi, S., and Oda, K.,
Purification and characterization of glutamate decarbox-
ylase from Lactobacillus brevis IFO 12005. Biosci.
Biotechnol. Biochem., 61, 1168–1171 (1997).
3) Gran, H. M., Gadaga, H. T., and Narvhus, J. A.,
Utilization of various starter cultures in the production of
amasi, a Zimbabwean naturally fermented raw milk
product. Int. J. Food Microbial., 88, 19–28 (2003).
4) Tsushida, T., and Murai, T., Conversion of glutamic acid
to ꢀ-aminobutyric acid in tea leaves under anaerobic
conditions. Agric. Biol. Chem., 51, 2865–2871 (1987).
5) Saikusa, T., Horino, T., and Mori, Y., Accumulation of
ꢀ-aminobutyric acid (GABA) in the rice germ during
water soaking. Biosci. Biotechnol. Biochem., 58, 2291–
2292 (1994).
19) 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. Bioresource Technol., 98, 312–319
(2006).
20) Nomura, M., Nakajima, I., Fujita, Y., Kobayashi, M.,
Kimoto, H., Suzuki, I., and Aso, H., Lactococcus lactis
contains only one glutamate decarboxylase gene. Micro-
biology, 145, 1375–1380 (1999).
21) Laemmli, U. K., Cleavage of structural proteins during
the assembly of the head of bacteriophage T4. Nature,
227, 680–685 (1970).
22) Sanger, F., Nicklen, S., and Coulson, A. R., DNA
sequencing with chain-terminating inhibitors. Proc. Natl.
Acad. Sci. USA, 74, 5463–5467 (1977).
6) Smith, D. K., Kassam, T., Singh, B., and Elliot, J. F.,
Escherichia coli has two homologous glutamate decar-
boxylase genes that map to distinct loci. J. Bacteriol.,
174, 5820–5826 (1992).
7) Maras, B., Sweeney, G., Barra, D., Bossa, F., and John,
R. A., The amino acid sequence of glutamate decarbox-
ylase from Escherichia coli. Eur. J. Biochem., 204, 93–
98 (1992).
8) Kono, I., and Himeno, K., Changes in ꢀ-aminobutyric
acid content during beni-koji making. Biosci. Biotechnol.
Biochem., 64, 617–619 (2000).
23) Stephens, S. K., Floriano, B., Cathcart, D. P., Bayley,
S. A., Witt, V. F., Jimenez-Diaz, R., Warner, P. J.,
and Ruiz-Barba, J., Molecular analysis of the locus
responsible for production of plantaricin S, a two-peptide
bacteriocin produced by Lactobacillus plantarum
LPCO10. Appl. Environ. Microbiol., 64, 1871–1877
(1998).
9) Hao, R., and Schmit, J. C., Cloning of the gene for
glutamate decarboxylase and its expression during
conidiation in Neurospora crassa. Biochem. J., 293,
735–738 (1993).
24) Capitani, G., De Biase, D., Aurizi, C., Gut, H., Bossa, F.,
and Grutter, M. G., Crystal structure and functional
¨
analysis of Escherichia coli glutamate decarboxylase.
EMBO J., 22, 4027–4037 (2003).
10) Komatsuzaki, N., Shima, J., Kawamoto, S., Momose, H.,
and Kimura, T., Production of ꢀ-aminobutyric acid
(GABA) by Lactobacillus paracasei isolated from tradi-
tional fermented foods. Food Microbiol., 22, 497–504
(2005).
11) Denner, L. A., and Wu, J. Y., Two forms of rat brain
glutamic acid decarboxylase differ in their dependence
on free pyridoxal phosphate. J. Neurochem., 44, 957–
965 (1985).
25) Nomura, M., Kimoto, H., Someya, Y., Furukawa, S., and
Suzuki, I., Production of ꢀ-aminobutyric acid by cheese
starter during cheese ripening. J. Dairy Sci., 81, 1486–
1491 (1998).
26) Gut, H., Pennacchietti, E., John, R. A., Bossa, F.,
Capitani, G., De Biase, D., and Grutter, M. G., Esche-
¨
richia coli acid resistance: pH-sensing, activation by
chloride and autoinhibition in GadB. EMBO J., 25,
2643–2651 (2006).
12) Nathan, B., Hsu, C. C., Bao, J., Rosemary, W., and Wu,
27) Sayed, A. K., Odom, C., and Foster, J. W., The

Characterization of Glutamate Decarboxylase of Lb. paracasei
285
Escherichia coli AraC-family regulators GadX and
GadW activate gadE, the central activator of gluta-
mate-dependent acid resistance. Microbiol., 153, 2584–
2592 (2007).
resistance mechanism in Lactococcus lactis and its
regulation. Mol. Microbiol., 27, 299–310 (1998).
29) Nomura, M., Kimoto, H., Someya, Y., and Suzuki, I.,
Novel characteristic for distinguishing Lactococcus
lactis subsp. lactis from subsp. cremoris. Int. J. Syst.
Bacteriol., 49, 163–166 (1999).
28) Sanders, J. W., Leenhouts, K., Burghoorn, J., Brands,
J. R., Venema, G., and Kok, J., A chloride-inducible acid