Occurrence of agmatine pathway for putrescine synthesis in Selenomonas ruminatium

Article abstract of DOI:10.1271/bbb.70550

Selenomonas ruminantium synthesizes cadaverine and putrescine from L-lysine and L-ornithine as the essential constituents of its peptidoglycan by a constitutive lysine/ornithine decarboxylase (LDC/ODC). S. ruminantium grew normally in the presence of the specific inhibitor for LDC/ODC, DL-α-difluoromethylornithine, when arginine was supplied in the medium. In this study, we discovered the presence of arginine decarboxylase (ADC), the key enzyme in agmatine pathway for putrescine synthesis, in S. ruminantium. We purified and characterized ADC and cloned its gene (adc) from S. ruminantium chromosomal DNA. ADC showed more than 60% identity with those of LDC/ODC/ADCs from Gram-positive bacteria, but no similarity to that from Gram-negative bacteria. In this study, we also cloned the aguA and aguB genes, encoding agmatine deiminase (AguA) and N-carbamoyl-putrescine amidohydrolase (AguB), both of which are involved in conversion from agmatine into putrescine. AguA and AguB were expressed in S. ruminantium. Hence, we concluded that S. ruminantium has both ornithine and agmatine pathways for the synthesis of putrescine.

Full text of DOI:10.1271/bbb.70550

Biosci. Biotechnol. Biochem., 72 (2), 445–455, 2008
Occurrence of Agmatine Pathway for Putrescine Synthesis
in Selenomonas ruminatium
1;**
1;***
1;****
Shaofu LIAO,1; Phuntip POONPAIROJ,
*
Kyong-Cheol KO,
Yumiko TAKATUSKA,
y
1;*****
1
1
2;
Yoshihiro YAMAGUCHI,
Naoki ABE, Jun KANEKO, and Yoshiyuki KAMIO
1Laboratory of Applied Microbiology, Department of Microbial Biotechnology,
Graduate School of Agricultural Science, Tohoku University,
1-1 Amamiyamachi, Aobaku, Sendai 981-8555, Japan
2Department of Human Health and Nutrition, Graduate School of Comprehensive Human Sciences,
Shokei Gakuin University, 4-10-1 Yurigaoka, Natori 981-1295, Japan
Received August 28, 2007; Accepted November 16, 2007; Online Publication, February 7, 2008
[doi:10.1271/bbb.70550]
Selenomonas ruminantium synthesizes cadaverine and
putrescine from L-lysine and L-ornithine as the essential
constituents of its peptidoglycan by a constitutive lysine/
ornithine decarboxylase (LDC/ODC). S. ruminantium
grew normally in the presence of the specific inhibitor
for LDC/ODC, DL-ꢀ-difluoromethylornithine, when
arginine was supplied in the medium. In this study,
we discovered the presence of arginine decarboxylase
(ADC), the key enzyme in agmatine pathway for
putrescine synthesis, in S. ruminantium. We purified
and characterized ADC and cloned its gene (adc) from
S. ruminantium chromosomal DNA. ADC showed more
than 60% identity with those of LDC/ODC/ADCs from
Gram-positive bacteria, but no similarity to that from
Gram-negative bacteria. In this study, we also cloned
the aguA and aguB genes, encoding agmatine deiminase
(AguA) and N-carbamoyl-putrescine amidohydrolase
(AguB), both of which are involved in conversion from
agmatine into putrescine. AguA and AguB were ex-
pressed in S. ruminantium. Hence, we concluded that
S. ruminantium has both ornithine and agmatine path-
ways for the synthesis of putrescine.
Polyamines such as putrescine, cadaverine, and
spermidine are essential constituents of peptidoglycan
and they play a significant role in the maintenance of the
integrity of the cell envelope in Selenomonas ruminan-
tium, Veillonella parvulla, V. alcalescens, and Anaero-
vibrio lipolytica.1–3) When S. ruminantium and two
species of Veillonella are grown in a medium supple-
mented with putrescine or cadaverine, putrescine and
cadaverine respectively link covalently to the ꢀ-carbox-
yl group of the D-glutamic acid residue of the peptido-
glycan, which is catalyzed by diamine:lipid intermediate
transferase, in normal cell growth.3–7) In S. ruminanti-
um, both putrescine and cadaverine are constitutively
synthesized from L-ornithine and L-lysine by lysine/
ornithine decarboxylase (LDC/ODC), which has decar-
boxylase activities towards both L-lysine and L-ornithine
with similar Km.8) Hence, our LDC was designated
lysine/ornithine decarbozylase (LDC/ODC).9) LDC/
ODC activities are prevented irreversibly by DL-ꢀ-
difluoromethyllysine (DFML) or DL-ꢀ-difluoromethylor-
nithine (DFMO).8) Since S. ruminantium has no L-
ornithine specific decarboxylase,10) LDC/ODC is
thought to be an important enzyme supplying putrescine
and cadaverine in the construction of diamine-contain-
ing peptidoglycan.8) In S. ruminantium, the production
of LDC/ODC is highly regulated and strictly linked to
the growth phase, i.e., a drastic decrease in LDC/ODC
Key words: Selenomonas ruminantium; agmatine path-
way for putrescine synthesis; arginine de-
carboxylase; agmatine deiminase; N-carba-
moyl-putrescine amidohydrolase
The first and the second authors contributed equally.
y
To whom correspondence should be addressed. Tel/Fax: +81-22-381-3347; E-mail: ykamio@shokei.ac.jp
Present address: Department of Food Science and Nutrition, Meiho Institute of Technology, 23 Ping Kuang Rd., Nei Pu Pingtung, 91212,
Taiwan
*
**
Present address: Biotech Central Research Unit, National Center for Genetic Engineering and Biotechnology, 113 Paholyothin Rd., Klong 1,
Klong Luang, Pathumthani, 12120, Thailand
***
Present address: Advanced Radiation Technology Institute, Korea Atomic Energy Research Institute, 1266 Sinjeong-dong, Jeongeup-shi,
Jeolabuk-do, 580-185, Republic of Korea
****
Present address: Department of Molecular and Cell Biology, University of California, Berkeley, California 94720-3202, USA
Present address: Department of Biochemistry, Robert Wood Johnson Medical School, Piscataway, New Jersey 08854, USA
*****
Abbreviations: LDC, L-lysine decarboxylase; ODC, L-ornithine decarboxylase; ADC, L-arginine decarboxylase; AguA, agmatine deiminase; AguB,
N-carbamoyl-putrescine amidohydrolase; NCP, N-carbamoyl-putrescine; DFML, DL-ꢀ-difluoromethyllysine; DFMO, DL-ꢀ-difluoromethyl-ornithine;
DFMA, DL-ꢀ-difluoromethylarginine; PALP, pyridoxal-50-phosphate

446
S. LIAO et al.
activities occurs on entry into the stationary phase of cell
growth, due to rapid degradation of LDC/ODC.8) Rapid
degradation of LDC/ODC caused a drastic decrease in
the intracellular free cadaverine content in the stationary
phase, whereas the putrescine, spermidine, and spermine
contents in the stationary phase were on a level with that
in the exponential phase (unpublished data). In the
current study, we isolated a new protein of 22 kDa (P22),
which is induced on entry into the stationary phase of
putrescine-grown S. ruminantium cells, as a regulatory
factor in LDC/ODC degradation by the ATP-dependent
proteolysis system.11) In the preceding study, we
characterized P22 and found that it is a direct counter-
part of eukaryotic antizyme (AZ) in the acceleration of
the degradation of S. ruminantium LDC/ODC due to its
binding to the P22-binding site in the LDC/ODC
molecule.9,12) We also found that P22 is a ribosomal
protein of S. ruminantium with no similarity to mouse
AZ in amino acid sequence.12) Recently, we observed
that S. ruminantium grew normally even in the presence
of 5 mM DFMO when 10 mM L-arginine was supplied in
a chemically defined medium (CD medium).9) When
S. ruminantium was grown with 14C-labeled L-arginine
in the presence of DMFO, the radioactive L-arginine was
incorporated into the cells and converted into agmatine
and putrescine. These findings suggest that S. ruminan-
tium has an agmatine pathway besides the ornithine
pathway for the synthesis of putrescine. We discovered
the presence of L-arginine decarboxylase (ADC), the
first enzyme in the agmatine pathway. We purified and
characterized ADC, and cloned the ADC gene (adc)
from S. ruminantium chromosomal DNA. Furthermore,
we identified agmatine deiminase (aguA) and N-carba-
moylputrescine (NCP) amidohydrolase (aguB) genes
from a flanking region of adc. They are involved in the
conversion of agmatine to putrescine. These genes were
cloned and expressed in Escherichia coli, and recombi-
nant AguA and AguB (rAguA and rAguB) were purified
and characterized. Expression of AguA and AguB in
S. ruminantium was confirmed immunologically using
anti-rAguA and -rAguB antisera, respectively.
vant was from Wako Pure Chemical Industry (Osaka,
Japan). Anti-rabbit and anti-mouse IgG (Fc)-alkaline
phosphatase conjugate were from Promega (Madison,
WI). ABI PRIZM BigDye Terminator Cycle Sequencing
Ready Reaction kit was from Applied Biosystems
(Foster city, CA). Unless otherwise stated, chemicals
used in this study were of the best grade commercially
available.
Bacterial strains and plasmids, and culture condi-
tions. The strains used in this study included S. rumi-
nantium subsp. lactilytica,9) and E. coli DH5 and E. coli
Rosetta (DE3) (Novagen) which were used as host
strains. Culture media included a Bacto-tryptone (1%)-
Bacto-yeast extract (1%)-glucose (0.5%) medium sup-
plemented with 0.01% caproic acid (TYG) and CD
medium10) for S. ruminantium, and an LB medium for
E. coli strains. S. ruminantium was grown under anae-
robic conditions, as described previously.9)
Reaction mixture and enzyme assay. ADC, AguA, and
AguB were assayed by measuring agmatine, NCP, and
putrescine, which were converted from L-arginine,
agmatine, and NCP respectively by ADC, AguA, and
AguB. For ADC assay, the reaction mixture contained
50 mM potassium phosphate buffer (pH 6.5), 10 mM L-
arginine, 50 mM PALP, 1 mM dithiothreitol (DTT), and
an enzyme peparation in a total volume of 100 ml. AguA
and AguB activities were assayed in 50 mM PIPES
buffer (pH 6.5) containing 1 mM DTT, 25 mM agmatine
for AguA and 10 mM NCP for AguB, and the enzyme
preparation in a total volume of 100 ml. The reaction
mixtures were incubated for 10 min at 50 ꢀC for ADC
and 40 ꢀC for AguA and AguB. The reaction was
stopped by adding TCA at a final concentration of 5%,
and the precipitate was removed by centrifugation at
10;000 ꢁ g for 5 min. The amounts of agmatine, NCP,
and putrescine were measured by HPLC using a TSKgel
Polyaminepak column (Tosoh, Tokyo) and cellulose
thin-layer electrophoresis using a cellulose plate 5716
(Merk, Darmstadt, Germany) by the method described
previously.14) One katal (kat) of each enzyme activity
was defined as the formation of 1 mole each of
agmatine, NCP, and putrescine per second. Specific
activity is given as kat/kg of enzyme. The kinetic
parameters, Km, Vmax, and Ki were calculated from
initial rate measurements for the decarboxylation of
L-arginine, the formation of NCP, and putrescine by
fitting to the Michaelis-Menten equation using a non-
linear regression algorithm.
Materials and Methods
Materials. DFML, DFMO, and DL-ꢀ-difluorometh-
ylarginine (DFMA) were kind gifts from Merrel Daw
Research Institute (Cincinnati, OH). Agmatine was
purchased from Aldrich (St. Louis, MO). NCP was
synthesized chemically according to the method describ-
ed by Smith and Garraway.13) Plasmids pMW118 and
pMW119 were purchased from Nippon gene (Tokyo)
and plasmids pET15b and pET28b were from Novagen
(Madison, WI). Restriction enzymes, T4 DNA ligase,
and Taq DNA polymerase were from Takara, (Otsu,
Japan). A HiTrap chelating HP column, Hybond ECL
membrane, and L-[U-14C]arginine (11 GBq/mmol) were
from GE Healthcare Bio-Science (Chalfont St. Giles,
UK). Lysylendopeptidase and Freund’s complete adju-
Purification of S. ruminantium ADC. S. ruminantium
cells were grown in 100 liters of TYG medium
supplemented with 20 mM L-arginine. Cells were har-
vested at mid-exponential phase by continuous centri-
fugation and washed once with 100 mM potassium
phosphate buffer (pH 7.0) containing 5 mM MgCl2.
The cells (wet weight, 134.7 g) were suspended in

Agmatine Pathway for Putrescine Synthesis in S. ruminantium
447
300 ml of 50 mM potassium phosphate buffer, pH 7.0,
containing 10 mM 2-mercaptoethanol and 0.1 mM PALP
(buffer A), and disrupted in a French pressure cell (SLM
Instrument, Rochester, NY) at 1,000 kg/cm2. After
centrifugation at 20;000 ꢁ g for 20 min, the supernatant
obtained was dialyzed against 250 mM potassium phos-
phate buffer (pH 7.0) containing 10 mM 2-mercapto-
ethanol and 0.1 mM PALP, and heated at 62 ꢀC for
10 min with gentle swirling. After centrifugation at
20;000 ꢁ g for 30 min, the supernatant was collected and
then put on a TSKgel DEAE-Toyopearl 650M column
(40 ꢁ 240 mm; Tosoh, Tokyo), which was equilibrated
with 50 mM potassium phosphate buffer, pH 7.0 (buf-
fer B). The column was washed with buffer B and then
eluted with a linear gradient of NaCl in buffer B (0 to
500 mM). Active fractions, eluted from 250 to 350 mM
NaCl, were collected and dialyzed against buffer A,
and then put on a TSKgel DEAE-5PW column (21:5 ꢁ
150 mm; Tosoh, Tokyo), which was equilibrated with
buffer B. The column was washed with buffer B and
eluted with a linear gradient of NaCl (0 to 400 mM). The
active fractions eluted from 220 to 250 mM NaCl in
buffer B were pooled. After it was dialyzed against
buffer A, the enzyme fraction was put on a TSKgel HA-
1000 column (7:5 ꢁ 75 mm; Tosoh) equilibrated with
buffer B. The column was washed with buffer B. The
enzyme was then eluted with a linear gradient of
potassium phosphate from 50 to 200 mM, pH 7.0. The
fractions eluted from 60 to 80 mM potassium phosphate
were pooled. After dialysis against buffer A, the enzyme
fraction was put on a Mono Q HR 5/5 column
(5 ꢁ 50 mm; Pharmacia, Uppsala) equilibrated with
buffer B. The column was washed with buffer B and
eluted with a liner gradient of NaCl in buffer B (0 to
500 mM). The fractions eluted from 280 to 330 mM NaCl
were pooled and dialyzed against buffer A.
sequencing of adc, the amino acid sequences corre-
sponding to the N-terminal 27-residue and 23-, 7-, 11-,
and 10-residue segments were found.
Preparation of oligonucleotide primers for cloning of
adc from S. rumiantium. According to the N-terminal
(Met1-Asp-Arg-Ile-Asn-Gln-His7) amino acid sequence
of intact ADC and its lysylendopeptidase fragment,
NH2-Asp-Glu-Ala-His-Gly, the oligonucleotide primers
Fw (50-ATGGA(TC)CG(ATGC)AT(TCA)AA(TC)CA-
(AG)CA-30) and Rv (50-GT(ATGC)CC(AG)TG(ATG-
C)GC(TC)TC(AC)TC-30) were designed. In combina-
tion with each primer, an approximately. 600-bp frag-
ment was amplified from the chromosomal DNA of
S. ruminantium by PCR using a Takara Ex Taq
polymerase. The amino acid sequence predicted from
the DNA sequence in the amplified DNA fragment
corresponded to that of the 16-, 23-, 7-, 11-, and 4-
residue segments H7TAPVYEAMLELRKRR22, E71AEE-
LTADAFGAQHAFFMVHGTT92, V99LSTVRA105, A149
-
LSDVERAIRQ159, and K193VLV196 respectively, of
intact ADC. Hence, the 600-bp PCR fragment was
used as a DNA probe in the cloning of adc from the
S. ruminantium chromosomal DNA.
Cloning of S. ruminantium ADC gene (adc) and its
DNA sequencing. A 3.5-kbp EcoRI fragment of chro-
mosomal DNA of S. ruminantium was strongly hybri-
dized with the probe. We amplified the 3.5-kb fragment
by inverse PCR using internal primers of a 605-bp
fragment. As a result of inverse PCR walking, the
nucleotide sequence of a 10.5-kb region containing adc
and its flanking region was determined.
ORF identification, homology search, and alignment
of multiple nucleotide and amino acid sequences.
Protein and nucleotide sequences were compared with
those on databases using FASTA and BLAST programs
implemented at the EMBL/GenBank/DDBJ nucleotide
sequence databases and the SWISSPROT/NBRF-PIR
protein sequence databases. Multiple-sequence align-
ment was done using a GENETYX program (Software
Development, Tokyo).
Determination of the N-terminal and internal amino
acid sequences of S. ruminantium ADC preparation.
The purified ADC preparation (200 mg) was dissolved in
1 ml of 70% formic acid containing 400 mg of CNBr and
kept in the dark under N2 gas at 24 ꢀC for 24 h.
Alternatively, the purified ADC preparation (100 mg)
was completely digested with 1 mg of lysylendopepti-
dase at 37 ꢀC for 12 h in 0.4 ml of 100 mM Tris–HCl
buffer (pH 9.0). Peptide fragments were subjected to
SDS–PAGE in a Tris-tricine buffer system, and then the
separated peptides were transferred to an Immobilon-P
membrane (Millipore, Bedford, MA) by the method
described previously.14) The N-terminal amino acid
sequences of the peptides were analyzed with an
Applied Biosystems protein sequencer Model 491. The
N-terminal sequence of the purified ADC was deter-
mined to be M1DRINQHTAPVYEAMLELRKRRVV-
PFD27. The amino acid sequences of four internal
peptide fragments, that were isolated were determined to
be EAEELTADAFGAQHAFFMVHGTT, VLSTVRA,
ALSDVERAIRQ, and KVLVDEAHGT. From the DNA
Construction of plasmids pET28b-adc, -orf6, and
-orf7 for expression of S. ruminantium ADC, ORFs 6
and 7 in E. coli. To construct an expression system for
rADC, ORFs-6 and -7, DNA fragments, including a
cording region of ADC, ORFs-6 and -7, were amplified
by PCR and ligated into plasmid pET28b. The resulting
plasmids were designated pET28b-adc, -orf6, and -orf7.
Each plasmid was transformed into E. coli Rosetta
(DE3). Cells carrying each plasmid was grown in LB
medium containing kanamycin (20 mg/ml) at 37 ꢀC with
shaking to an optical density of 0.6 at 600 nm, IPTG was
added at a final concentration of 0.4 mM, followed by
incubation with shaking for an additional 2 h. The cells
were collected by centrifugation at 4 ꢀC and suspended

448
S. LIAO et al.
in 50 mM potassium phosphate buffer (pH 7.4) contain-
ing 50 mM PALP, and disrupted in a French pressure cell.
The cell lysate was centrifuged at 8;000 ꢁ g for 20 min
at 4 ꢀC, and the supernatant was used for purification of
rADC and ORFs 6 and 7.
Putrescine
Agmatine
Arginine
Site-directed mutagenesis of adc and orfs 6 and 7.
Site directed mutagenesis was done by the overlapping
extension method.15,16) The mutation of each mutant was
confirmed by DNA sequencing.
Agm
0
2
4
6
8
10 12 24
Time (h)
Fig. 1. Conversion of 14C-Arginine into Agmatine and Putrescine in
S. ruminatium.
Purification of the recombinant proteins. Crude
recombinant ADC and ORFs 6 and 7 with His-Tag at
the C-terminal were purified with a HiTrap chelating HP
column using a linear gradient from 0 to 500 mM
imidazole in 20 mM sodium phosphate, pH 7.2, contain-
ing 500 mM NaCl. ADC and ORFs 6, and 7 were eluted
in a fraction containing 150 mM imidazole. The active
fractions were collected and dialyzed against 50 mM
PIPES buffer, pH 7.0, containing 1 mM DTT (buffer C)
and then applied to a DEAE-5PW column equilibrated
with buffer C (Tosoh, Tokyo). After washing of column
with the buffer C, the enzymes were eluted with a linear
gradient of NaCl from 0 to 500 mM in buffer C. The
active fractions from the DEAE-5PW column was
dialyzed against buffer C and then ammonium sulfate
was added at a final concentration of 1,000 mM and
applied to a phenyl-5PW column (Tosoh), which was
equilibrated with buffer C containing 1,000 mM ammo-
nium sulfate. After the column was washed with
buffer C containing 1,000 mM ammonium sulfate, en-
zyme fractions were eluted with a linear gradient of
ammonium sulfate from 1,000 mM to 0 mM in buffer C.
The molecular masses of the recombinant proteins were
measured with a TSK gel G3000SW gel filtration
column (Tosoh; diameter, 0.75 cm; height, 30 cm)
equilibrated with 50 mM potassium phosphate buffer
(pH 6.0) containing 200 mM NaCl, using albumin (64.7
kDa), ovalbumin (45.8 kDa), chymotrypsinogen (19.9
kDa), and ribonuclease A (15.4 kDa) as molecular
standards.
S. ruminantium cells were cultured in CD medium supplemented
with 10 mM L-[U-14C]arginine (0.18 MBq) and 5 mM DFMO. At the
times indicated, cells were harvested and extracted with 5% TCA.
After the precipitate was discarded by centrifugation at 10;000 ꢁ g,
the supernatant was analyzed by cellulose thin-layer electrophoresis
to detect 14C-arginine, -agmatine, and -putresine.
Results and Discussion
Conversion of L-arginine into agmatine and putres-
cine
S. ruminantium cells were grown at 37 ꢀC in CD
medium supplemented with 10 mM 14C-labeled L-argi-
nine in the presence of 5 mM DFMO. At the times
indicated in Fig. 1, cells were harvested and treated with
10% cold TCA. After centrifugation, the TCA soluble
fraction was analyzed by cellulose thin-layer electro-
phoresis (Fig. 1). As shown, the labeled L-arginine was
incorporated into the cells up to 4-h after incubation.
Thereafter, the amount of L-arginine decreased between
0.5 and 8, accompanied by an increase in agmatine and
putrescine. After 10 h of incubation, no L-arginine was
detected in the TCA soluble fraction. On the other hand,
the accumulation of putrescine in the TCA soluble
fraction proceeded up to 12 h after incubation. The data
clearly showed that 14C-lableled L-arginene incorporated
into the cells was converted into agmatine and putres-
cine in S. ruminantium in the presence of 5 mM DMFO.
Therefore, it is suggested that S. ruminantium has ADC
pathway for the synthesis of putrescine.
Preparation of anti-S. ruminantium ADC, -rAguA,
and -rAguB antisera. Mice antiserum raised against
rADC, rAguA, and rAguB was prepared by the method
described previously.8)
Detection of ADC in the crude extract from S. rumi-
nantium
The ADC activity during growth of the S. ruminan-
tium cells in TYG medium supplemented with 10 mM
L-arginine and 0.01% capric acid was examined.
Sonicated extracts from the cells were taken at various
times after inoculation and ADC activity was assayed.
ADC activity was detected, and the activity in all the
cells in culture increased up to 6 h, and reached ap-
proximately 3:9 ꢁ 10ꢂ9 kat/liter of culture, and decreas-
ed thereafter. Accordingly, the timing of cell harvest for
preparation of crude extract from the cells was taken to
be during the period, 3.5–4.0 h (O.D.660 = 1.0) after
inoculation. When the cells were grown in CD medium
without L-arginine, no detectable ADC activity was
found in the crude extract prepared from the cells.
Miscellaneous. Quantitative assay of protein and
SDS–PAGE were performed by the methods described
previously.8)
Nucleotide sequence accession numbers. The nucleo-
tide sequences of adc, orfs-1, -2, -3, -4, -5, -6, and -7
have been deposited in the DDBJ/EMBL/GenBank
nucleotide sequence databases under accession Nos.
AB198397, AB198398, AB196518, AB198392, AB198-
393, AB198394, AB198395, and AB198396.

Agmatine Pathway for Putrescine Synthesis in S. ruminantium
Table 1. Purification of S. ruminantium ADC
449
Total volume
(ml)
Total protein
(mg)
Total activity
(ꢁ10ꢂ9 kat)
Specific activity
(ꢁ10ꢂ3 kat/kg)
Yield
(%)
Purification
(fold)
Purification step
Crude extract
318
310
232
25
3200
2310
331
339
425
132
12.5
7.18
29.1
0.123
0.184
0.399
0.635
17.1
100
108
33.5
3.2
1
1.5
Heat treatment
DEAE-Toyopearl 650M
DEAE-5PW
3.25
5.16
139
19.7
0.42
0.117
HA-1000
MonoQ HR 5/5
21
1.8
7.4
6.3
249
2024
Purification and characteristics of S. ruminantium
ADC
Purification
by the ADC preparation with Ki values of 0.8 mM and
2.1 mM respectively (Fig. 2A and B). In the range from 0
to 0.2 mM DFMA, plots of the reciprocals of reaction
velocities against L-arginine and L-lysine gave a series of
straight lines intercepting a single point on the 1/v axis,
indicating that the inhibition type of DFMA is compet-
itive and that the catalytic domains of ADC towards
the two substrates are identical. When the enzyme was
incubated for 60 min at 4 ꢀC with DFMA (10 mM) and
dialyzed against buffer A, the decarboxylating activities
towards both substrates were irreversibly lost (data not
shown), indicating that DFMA is an irreversible inhibitor
of S. ruminanitum ADC. On the other hand, neither
DFML nor DFMO affected ADC activities.
Through the five purification steps shown in Table 1,
ADC was purified about 2,030-fold to electrophoretic
homogeneity with a specific activity of 0.25 kat/kg of
protein. At the final step of purification, total specific
ADC activity increased to about 15 times that at the
semi-final step (HA-1000 column chromatography).
After MonoQ HR 5/5 chromatography, some contam-
inants, which prevented the ADC activity, have been
removed. Though the inhibitors of ADC were not
identified, it is assumed that the inhibitors are entity with
high molecular mass, because they did not pass through
a cellulose dialyzing bag. The N-terminal amino acid
sequence of the enzyme was M1DRINQHTAP10VY-
EAMLELRKRR22. The purified ADC can be stored at
ꢂ80 ꢀC for at least 6 months without any appreciable
loss in the presence of 1 mM PALP.
Optimal conditions for enzyme activity
The optimal pH for enzyme activity was 6.5, same as
that for an E. coli inducible ADC, but not for constit-
utive ADC. At pH 7.0, 8.0, and 9.0, the activity de-
creased to 80%, 70%, and 50% as compared to that at
pH 6.5, respectively. The optimal temperature for the
enzyme activity was estimated to be 60 ꢀC. The enzyme
activities at 10 min of incubation at 50, 55, 60, and 70 ꢀC
were about 4, 7, 10, and 1.2 times higher respectively
than at 40 ꢀC.
The presence of the enzyme molecule in dimeric form
with ADC activity
The apparent molecular mass of the purified ADC
preparation was 120 kDa as judged by gel filtration
analysis, and was 58 kDa on SDS–PAGE (Table 2). On
gel filtration, the monomeric form of ADC of 58 kDa
was also separated with no enzyme activity (data not
shown). The data show that the active ADC comprised
two identical monomeric subunits.
Enzyme stability
The purified enzyme preparation, which was stored at
pH 6.0 to 7.5 at 4 ꢀC, maintained activity for at least 1
month without loss. Full enzyme activities and approx-
imately 50% of them were retained at 50 and 60 ꢀC
respectively for 60 min in the presence of 1 mM PALP.
On the other hand, at 50 ꢀC in the absence of PALP,
90% of the activity was lost within 10 min. No enzyme
activity was observed in the enzyme preparation heated
at 70 ꢀC for 60 min even in the presence of 1 mM PALP.
Comparison of properties of S. ruminantium ADC
with other ADCs
Table 2 shows a comparison of properties of S. ru-
minantium ADC with that of ADCs from E. coli,17–19)
soybean,20) and rat.21) S. ruminantium ADC was mark-
edly different from the other ADCs listed in this table in
enzyme properties, except for a similarity in the cofactor
requirement for the enzyme activity. S. ruminantium
ADC of 120 kDa, consisted on two identical subunits of
58 kDa, whereas the E. coli constitutive (296 kDa) and
inducible (820 kDa) ADCs consisted of three identical
subunits of 72 kDa18) and 10 identical subunits of
820 kDa22) respectively. It is also of interest that
Substrate specificity and inhibitor
To determine the substrate specificity of the enzyme
preparation, the enzyme concentrations used in the
enzyme reaction were held at about 1.0–100 nM. L-
Arginine was the preferred substrate for this enzyme,
whereas L-lysine was attacked at a rate of 10% of that
for L-arginine. Neither D-lysine, L-ornithine, nor L-
histidine acted as substrate. The kinetics of decarboxyl-
ation of L-arginine and L-lysine by the ADC preparation
were analyzed by measuring initial velocities over the
range of substrate concentrations (1 to 8 mM). Over the
concentration range used, both L-arginine and L-lysine
behaved as Michaelis-Menten-type substrates (Fig. 2A
and B). A double-reciprocal plot of the initial velocities
demonstrated that the Km values for L-arginine and
L-lysine were 5.6 mM and 50 mM respectively, and that
the Vmax values were 12.5 and 5.9 mmol minꢂ1 mgꢂ1 of
protein for L-arginine and L-lysine respectively. DMFA
inhibited the decarboxylation of L-arginine and L-lysine

450
S. LIAO et al.
Table 2. Comparison of S. ruminantium ADC with ADCs from E. coli, Plant, and Rat
E. coli
Plant
Property
S. ruminantium
Rat
(Glycine max)
Inducible (AdiA)
Constitutive (SpeA)
Decarboxylase type
Subunit Mr
type I
58,000
120,000
(Dimer)
PLP
type I
82,000
type III
72,000
type III
74,000
240,000
(Trimer)
PLP
type III
55,000
—
820,000
(Decamer)
PLP
296,000
(Tetramer)
PLP
Native Mr
Cofactor
Substrate specificity
PLP
0.03
(with Mg2þ
Km L-arginine (mM)
5.6
0.065
0.0461
0.75
)
Km L-ornitine (mM)
Km L-lysine (mM)
—
50
—
—
—
—
—
—
0.25
—
L-ꢀ-Chloro-ꢁ-
guanidovaleric acid
Inhibitor
DFMA
0.8
Putrescine, Spermidine
—
Agmatine, Putrescine
—
Ca2þ, Co2þ
Ki for ADC activity
(mM)
0.026
—
Ki for L-LDC activity
(mM)
2.1
—
5.2
—
8.4
—
—
—
—
—
—
—
8.3
25
—
Optimal pH
Optimal temperature
(ꢀC)
6.5
60
37–40
—
pH stability
Thermal stability for
30 min
6.5–7.5 (>80%)
With PLP
Without PLP
60 ꢀC (54%)
35 ꢀC (50%)
—
—
—
—
30–50 ꢀC
—
—
—
A
B
0.1 mM DFMA
0.05 mM DFMA
0.2 mM DFMA
0.1 mM DFMA
1.0
1.0
No inhibitor
No inhibitor
0.5
µ
0.5
µ
0.1
0
1.0
0
-1
-1
1/[L-arginine] (mM )
1/[L-lysine] (mM )
Fig. 2. Lineweaver-Burk Plots of L-Arginine Decarboxylase (A) and L-Lysine Decarboxylase (B) Activities of the Purified S. ruminantium ADC
Preparation with and without DFMA.
S. ruminantium ADC activity was not inhibited by
putrescine, spermidine, agmatine, or ꢀ-hydroxy-ꢁ-gua-
nidovaleric acid, which are known to be strong
A
comparison of the fold type of S. ruminantium ADC and
others is discussed below.
gene and 7 ORFs (Fig. 3A). The ORF for the adc gene
(solid arrow in Fig. 3A) encoded a 485-amino acid
protein of 53,313 Da whose N-terminal sequence match-
ed that determined for purified ADC. The adc gene,
spanning positions 2,961–4,410 within the cloned
genomic sequence, started with an ATG codon and
ended with a TAA stop codon. Putative ꢂ35 (TTGCTT)
and ꢂ10 (TAAAAT) sequences were found 84 and
61 bp upstream of the translation initiation codon of adc
respectively (Fig. 3B). Ten base pairs upstream of the
inhibitors of the other ADCs shown in Table 2.23,24)
Nucleotide sequence of the S. ruminantium ADC gene
(adc)
The determined 11.5-kbp sequence included the adc

Agmatine Pathway for Putrescine Synthesis in S. ruminantium
451
0
5
10
kbp
adc
orf1
2
orf3 orf4
orf5
orf6 orf7
Fig. 3. Genetic Organization of adc and Its Flanking Regions, and Nucleotide Sequence of the adc of S. ruminantium.
A, Genetic organization of a 10.5-kbp fragment from S. ruminantium chromosomal DNA, including the adc and its flanking regions. The solid
and open arrows show the adc gene and seven other predicted ORFs and the directions of their transcription. B, Nucleotide sequence of the adc.
The strand shown is in the 50 to 30 direction, and numbers of nucleotides were counted from the first nucleotide of the 10.5-kbp fragment. The
deduced amino acid sequence is indicated by the single-letter code under the nucleotide sequence. The amino acid sequence determined
chemically, putative ribosomal binding site, predicted promoter sequence, and translation termination codon are indicated by single underlines,
double underline, black-shaded boxes, and a single asterisk respectively.
ATG codon, a ribosomal binding site consensus se-
quence (GGAGG), was found at position 2,951–2,955.
No termination loop was found downstream of the TGA
stop codon.
To confirm that adc was the gene encoding S. rumi-
nantium ADC, the cloned adc was expressed in E. coli.
The recombinant ADC (rADC) was purified and its
characteristics were compared with that of ADC
preparation. The purified rADC protein had character-
istics identical to those of the native ADC protein.
even in the presence of PALP. On the other hand,
S. ruminantium ADC showed no homology with an
E. coli constitutive ADC,28) which belongs to fold
type III as do eukaryotic ODC and plant ADC.25) The
data suggested that S. ruminanitum ADC belongs to fold
type I decarboxylase.
Nucleotide sequencing analysis of the flanking region
of adc
Two orfs (orf6 and orf7 in Fig. 3A) that showed high
homology with the genes encoding agmatine utilization
enzymes, were found downstream of the adc gene.
ORF6, which consists of 372 amino acids, had 54% and
43% identity in amino acid sequence with agmatine
deiminase of Peudomomas aeruginosa29) and Strepto-
coccus mutans30) respectively, and 48% identity with
peptidyl-arginine deiminase-like protein of Listeria
monocytogenes (accession no. AAG03681). ORF7,
which is composed of 292 amino acids, had 57%
identity in amino acid sequence with NCP amidohy-
drolase of Peudomomas aeruginosa. ORF7 also showed
71, 58, 57, and 55% identities with carbon-nitrogen
hydrolase family proteins from Streptococcus pneumo-
niae (accession no. AAK75046), Erwinia carotovora
(accession no. CAG77171), Caulobacter crescentus
(accession no. AAK22199), and Burkholderia pseudo-
mallei (accession no. CHA34094) respectively. These
results suggest that S. ruminantium has an agmatine
pathway for putrescine synthesis that includes ADC,
agmatine deiminase, and NCP amidohydrolase.
ORFs-1, -2, -3, -4, and -5, encoding 467-, 207-, 286-,
439-, and 391-amino acid proteins respectively, were
also identified. The amino acid sequences of ORFs-1, -2,
-3, -4, and -5 exhibited 61, 31, 67, 59, and 61% overall
identify with putative Naþ/melibiose symport or related
transporter (accession no. ZP 00732149), putative cyti-
dylate kinase (accession No. EAA23705), putative
spermidine synthase (accession no. AAK75042), puta-
tive saccharopine dehydrogenase (accession no. AAB37-
373), and putative carboxynorspermidine decarboxylase
of Streptococcus pneumoniae (accession No. BAB07-
677) respectively. At positions 10 to 15 bp upstream of
the first ATG codons of orfs4, -5, -6, and -7, putative
ribosomal binding sites were observed. An inverted-
repeat sequence consisting of a stem-loop with a 12-bp
Amino acid sequence homologies of S. ruminantium
ADC with other bacterial decarboxylases
A comparison of the deduced amino acid sequence
of S. ruminantium ADC with those of other possible
bacterial decarboxylases revealed that S. ruminantium
ADC showed 40 to 68% identities with possible
arginine, ornithine, and lysine decarboxylases from
genome sequencing of Streptococcus pneumoniae, Clos-
tridium perfringens, Fusobacterium nucleatum, Bacillus
cereus, and putative decarboxylases from Symbiobacte-
rium thermophilium and Thermoanaerobacter tengcon-
gesis (Fig. 4). Grishin et al. have proposed a topo-
graphic model of the PALP-binding domain of
eukaryotic and prokaryotic ODCs by the analysis of
known barrel structures of ODCs.25) They classified the
PALP-utilizing enzymes into seven fold types from the
predicted secondary structures, and showed that the
prokaryotic ODCs, prokaryotic LDCs, and bacterial
inducible ADCs belong to fold type I. One of these,
inducible ODC from Lactobacillus 30a, has three amino
acids involved in PALP binding in its active center. Lys-
355 in the Ser-X-His-Lys motif, which is commonly
found in PALP-dependent decarboxylase, forms a Schiff
base with PALP. An invariant Asp-316 residue is also a
common motif found in many PALP-dependent en-
zymes.25,26) Although S. ruminanitum ADC showed less
than 20% identity in amino acid sequence of inducible
ODC from Lactobacillus 30a, E. coli constitutive LDC,
and E. coli inducible ADC (AdiA),27) which belongs to
fold type I (Table 2), S. ruminanitum ADC and pre-
dicted enzymes listed in Fig. 4 shared three residues
(His-117, Asp-197, and Lys-228) that are involved in
PALP binding in fold type I decarboxylases (Fig. 4). An
H117A mutant of S. ruminantium ADC lost its activity

452
S. LIAO et al.
Fig. 4. Sequence Data of the Predicted ADC/ODC/LDC from Genome Sequencing.
The sequence data were obtained from GenBank (http://www.ncbi.nlm.nih.gov): S. pneumoniae (Spn) LDC (accession No. AAK75040),
Cl. perfringens (Cpe) LDC (accession no. BAB80255), F. nucleatum (Fnu) ODC (accession no. EAA24580), S. thermophilum (Sta) ODC
(accession no. BAD42250), T. tengcongesis (Ttn) LDC (accession no. AAM24281), B. cereus (Bce) ADC (accession no. AAP10882), E. coli
(Eco) constitutive LDC (accession no. BAA21656), E. coli inducible ADC (AdiA) (accession no. AAA97017), and Lactobacillus 30a (L30)
ODC (accession no. AAA64830).

Agmatine Pathway for Putrescine Synthesis in S. ruminantium
453
ORF7
reaction
ORF6
reaction
Control
A
Putrescine -
Agmatine -
NCP
-
B
ORF6
H2O NH3
ORF7
H2N
NH C NH
NH2
H2N
NH C NH2
O
H2N
NH2
CO2+NH3
Agmatine
NCP
Putrescine
Fig. 5. Cellulose Thin-Layer Electrophoresis of the Reaction Products from Agmatine and NCP by the Recombinant ORF-6 and -7 (A), and a
Schematic Representation of the Enzyme Reactions by ORF-6 and -7 (B).
Thin-layer electrophoresis was done as described in ‘‘Materials and Methods.’’ The reaction products were detected by spraying with
ninhydrin reagent.2)
arm appeared at positions 290 bp to 321 bp downstream
of the stop codon of orf7.
10
1
Cloning of the orf6 and orf7 genes, and preparation
of recombinant ORF6 and ORF7
The orf 6 and -7 genes were amplified by PCR and
cloned to pET, and recombinant ORF -6 and -7 proteins
were expressed as C-terminal His6-tag fusion proteins in
E. coli Rosetta (DE3). Recombinant proteins were
purified by Ni2þ chelating affinity, DEAE-5PW, and
phenyl-5PW column chromatographies to electropho-
retic homogeneity. The enzyme activities of the purified
recombinant ORF6 and ORF7 preparations were exam-
ined using agmatine and NCP respectively as substrates
(Fig. 5). Agmatine was converted to a ninhydrin-posi-
tive product, showing migration similar to that of
authentic NCP, by rORF6 (Fig. 5). The reaction product
from agmatine by rORF6 was analyzed with an API
2000 mass spectrophotometer. Its molecular mass was
131.8 Da identical to that of authentic NCP. NCP was
converted to putrescine by ORF7 (Fig. 5). From these
results, ORFs 6 and -7 were confirmed as agmatine
deiminase and NCP amidohydrolase respectively.
Hence, orfs6 and -7 were designated the agmatine
deiminase gene (aguA) and the NCP amidohydrolase
gene (aguB), respectively.
0.1
AguA(ORF6)
AguB(ORF7)
0.01
0
1 2 5 7 8 9 10 11 12
3 4 6
Time (h)
Fig. 6. Western Blotting Analysis of Crude Extract from S. rumi-
nantium for the Presence of AguA and AguB.
S. ruminantium cells were grown in TYG medium. At the times
indicated in this figure, cells were harvested and disrupted by
sonication. The sonicated extract was analyzed by SDS–PAGE,
followed by Western blotting for the presence of AguA and AguB,
using anti-rAguA and -rAguB antisa respectively.
indicating that both aguA and aguB were expressed in
S. ruminantium.
Properties of rAguA and rAguB
The molecular masses of S. ruminantium AguA and
AguB were calculated to be 41,391 Da and 32,532 Da
respectively from their predicted amino acid sequences.
The molecular mass of the rAguA preparation as
measured by SDS–PAGE and gel filtration analysis
was 45 kDa. On the other hand, the determined molecu-
lar mass of the rAguB preparation was 37 kDa by SDS–
PAGE and the 80 kDa by gel filtration analysis. The data
show that S. ruminantium rAguA and rAguB consisted
Expression of aguA and aguB in S. ruminantium
Expression of aguA and aguB in S. ruminantium cell
was examined by Western blotting analysis using anti-
rAguA and -rAguB antisera (Fig. 6). Both AguA and
AguB were detected in the cells grown in TYG medium,

454
S. LIAO et al.
of a monomer and a homodimer respectively. AguAs
from Pseudomonas aeruginosa,31) Arabidopsis thali-
ana,32) maize shoots,33) and rice34) have been reported to
be homodimers consisting of 43, 43, 44, and 95 kDa
subunits respectiverly, except that from soy bean, with a
monomer of 70 kDa.35) AguBs from P. aeruginosa and
A. thaliana have been reported to consist of homohex-
amers consisting of subunits of 33 kDa, and 35 kDa
respectively.3,31,36)
The optimum pH and temperature for rAguA were 6.5
and 45 ꢀC, respectively. The enzyme was stable for
30 min in a pH range of 6.0 to 9.0 and at temperatures
below 40 ꢀC. Optimum pH for rAguB activity was about
7.0, and the enzyme was stable from pH 6.0 to 8.0. The
optimum temperature for rAguB was 45 ꢀC. AguB
activity was stable at up to 70 ꢀC for 30 min in 50 mM
PIPES buffer (pH 6.5).
The Km and Vmax values obtained from a Line-
weaver-Burk plot were 6.67 mM and 1.22 nmol/min/mg
protein of rAguA towards agmatine and 0.22 mM and
0.33 nmol/min/mg protein of rAguB towards NCP
respectively. AguA from rice seedlings showed high
Km (15 mM),4) while the Km values for AguA from
P. aeruginosa27) A. thaliana,3) maize shoots,33) soy-
bean,35) and cucumber37) were lower than 1 mM. The
Km value for S. ruminantium rAguB was almost the
same as those for AguB from P. aeruginosa,31) A. thali-
ana,3) and maize shoots.33)
The effects of metal ions on rAguA and rAguB
activities were also examined. Metal ions, including
Al3þ, Co2þ, Cu2þ, Fe2þ, Fe3þ, Ni2þ, and Zn2þ, inhibited
rAguA activity at 1 mM, but Mn2þ did not. This was
similar to AguA from soy bean,35) ice seedling,34) and
corn.38) On the other hand, rAguB was strongly inhibited
by Co2þ, Fe2þ, Mn2þ, and Zn2þ. Iodoacetamide and
Hg2þ were strong inhibitors of both rAguA and rAguB
activities. These results suggest that cysteine residues in
rAguA and rAguB molecules are pivotal in the catalysis
of both enzymes.
Acknowledgments
This work was supported in part by the Foundation of
the Noda Institute for Scientific Research. S. Liao and
K-C. Ko were supported by pre-doctoral fellowship
from Yoneyama Foundation and Waroh Suzuki Foun-
dation respectively. P. Poonpairoi was the recipient of a
fellowship from the Ministry of Education, Cultures,
Sports, Science and Technology of Japan. Y. Takatsuka
and Y. Yamaguchi were the recipients of postdoctoral
fellowships from the Japan Society for Promotion of
Science.
References
1) Hirao, T., Sato, M., Shirahata, A., and Kamio, Y.,
Covalent linkage of polyamines to peptidoglycan in
Anaerovibrio lypolytica. J. Bacteriol., 182, 1154–1157
(2000).
2) Kamio, Y., Itoh, Y., Terawaki, Y., and Kusano, T.,
Cadaverine is covalently linked to peptidoglycan in
Selenomonas ruminantium. J. Bacteriol., 145, 122–128
(1981).
3) Kamio, Y., and Nakamura, K., Putrescine and cadaver-
ine are constituents of peptidoglycan in Veillonella
alcalescens and Veillonella parvula. J. Bacteriol., 169,
2881–2884 (1987).
4) Kamio, Y., Structural specificity of diamines covalently
linked to peptidoglycan for cell growth of Veillonella
alcalescens and Selenomonas ruminantium. J. Bacteriol.,
169, 4837–4840 (1987).
5) Kamio, Y., Itoh, Y., and Terawaki, Y., Chemical
structure of peptidoglycan in Selenomonas ruminantium:
cadaverine links covalently to the D-glutamic acid
residue of peptidoglycan. J. Bacteriol., 146, 49–53
(1981).
6) Kamio, Y., Poso, H., Terawaki, Y., and Paulin, L.,
Cadaverine covalently linked to a peptidoglycan is an
essential constituent of the peptidoglycan necessary for
normal growth in Selenomonas ruminantium. J. Biol.
Chem., 261, 6585–6589 (1986).
Effect of site-directed mutagenesis on the activities of
rAguA and rAguB
7) Kamio, Y., Terawaki, Y., and Izaki, K., Biosynthesis of
cadaverine-containing peptidoglycan in Selenomonas
ruminantium. J. Biol. Chem., 257, 3326–3333 (1982).
8) Takatsuka, Y., Onoda, M., Sugiyma, T., Muramoto, K.,
Tomita, T., and Kamio, Y., Novel characteristics of
Selenomonas ruminantium lysine decarboxylase capable
of decarboxylating both L-lysine and L-ornithine. Biosci.
Biotechnol. Biochem., 63, 1063–1069 (1999).
9) Yamaguchi, Y., Takatsuka, Y., Matsufuji, S., Murakami,
Y., and Kamio, Y., Characterization of a counterpart to
mammalian ornithine decarboxylase antizyme in prokar-
yotes. J. Biol. Chem., 281, 3995–4001 (2006).
10) Kamio, Y., and Terawaki, Y., Purification and properties
of Selenomonas ruminantium lysine decarboxylase. J.
Bacteriol., 153, 658–664 (1983).
Alignment of AguA among several homologs ob-
tained from GenBank revealed that C361 residue
wasconserved. On the other hand, triad conserved
regions containing glutamate, lysine, and cysteine
residues in the nitrilase family, and K147, C153, and
E160 residues, were found in S. ruminantium rAguB.
Hence, the roles of these conserved cysteine residues in
rAguA and rAguB were examined by site-directed
mutagenesis. The C361A mutant of rAguA showed
20% of the activity of wild-type rAguA. On the
other hand, the C153A and E160A mutants of rAguB
completely lost activity. These results confirm the
inhibitory effect of the thiol-modifying agent, iodoace-
tamide, and Hg2þ on the cysteine residues of AguA
and AguB.
11) Yamaguchi, Y., Takatsuka, Y., and Kamio, Y., Identi-
fication of a 22-kDa protein required for the degradation
of Selenomonas ruminantium lysine decarboxylase by
ATP-dependent protease. Biosci. Biotechnol. Biochem.,

Agmatine Pathway for Putrescine Synthesis in S. ruminantium
455
66, 1431–1434 (2002).
26) Momany, C., Ernst, S., Ghosh, R., Chang, M.-L., and
Hackert, M. L., Crystallographic structure of a PLP-
dependent ornithine decarboxylase from Lactobacillus
12) Takatsuka, Y., Yamaguchi, Y., Ono, M., and Kamio, Y.,
Gene cloning and molecular characterization of lysine
decarboxylase from Selenomonas ruminantium delineate
its evolutionary relationship to ornithine decarboxylases
from eukaryotes. J. Bacteriol., 182, 6732–6741 (2000).
13) Smith, T. A., and Garraway, J. L., N-Carbamoylputr-
scine: an intermediate in the formation of putrescine by
barley. Phytochem., 3, 23–26 (1964).
˚
30a to 3.0 A resolution. J. Mol. Biol., 252, 643–655
(1995).
27) Stim, K. P., and Bennett, G. N., Nucleotide sequence of
the adi gene, which encodes the biodegradative acid-
induced arginine decarboxylase of Escherichia coli. J.
Bacteriol., 175, 1221–1234 (1993).
14) Matsuda, E., Abe, N., Tamagawa, H., Kaneko, J., and
Kamio, Y., Gene cloning and molecular characterization
of an extracellular poly(L-lactic acid) depolymerase
from Amycolatopsis sp. strain K104-1. J. Bacteriol., 187,
7333–7340 (2005).
15) Ho, S. N., Hunt, H. D., Horton, R. M., Pullen, J. K., and
Pease, L. R., Site-directed mutagenesis by overlap
extension using the polymerase chain reaction. Gene,
77, 51–59 (1989).
16) Horton, R. M., Hunt, H. D., Ho, S. N., Pullen, J. K., and
Pease, L. R., Engineering hybrid genes without the use
of restriction enzymes: gene splicing by overlap exten-
sion. Gene, 77, 61–68 (1989).
17) Blethen, S. L., Boeker, E. A., and Esmond, E. E.,
Arginine decarboxylase from Escherichia coli. I. Puri-
fication and specificity for substrates and coenzyme. J.
Biol. Chem., 243, 1671–1677 (1968).
18) Buch, J. K., and Boyle, S. M., Biosynthetic arginine
decarboxylase in Escherichia coli is synthesized as a
precursor and located in the cell envelope. J. Bacteriol.,
163, 522–527 (1985).
19) Wu, W. H., and Morris, D. R., Biosynthetic arginine
decarboxylase from Escherichia coli: purification and
properties. J. Biol. Chem., 248, 1687–1695 (1973).
20) Nam, K. H., Lee, S. H., and Lee, J., Purification and
characterization of arginine decarboxylase from soybean
(Glycine max) hypocotyls. Plant Cell Physiol., 10,
1150–1155 (1997).
21) Regunathan, S., and Reis, D. J., Characterization of
arginine decarboxylase in rat brain and liver: distinction
from ornithine decarboxylase. J. Neuro. Chem., 74,
2201–2208 (2000).
22) Nowak, S., and Boeker, E. A., The inducible arginine
decarboxylase of Escherichia coli B: activity of the
dimer and the decamer. Arch. Biochem. Biophys., 270,
110–116 (1981).
28) Moore, R., and Boyle, S. M., Nucleotide sequence and
analysis of the speA gene encoding biosynthetic arginine
decarboxylase in Escherichia coli. J. Bacteriol., 172,
4631–4640 (1990).
29) Stover, C. K., Pham, X.-Q., Erwin, A. L., Mizoguchi, S.
D., Warrener, P., Hickey, M. J., Brinkman, F. S. L.,
Hufnagle, W. O., Kowalik, D. J., Lagrou, M., Garber, R.
L., Goltry, L., Tolentino, E., Westbrook-Wadman, S.,
Yuan, Y., Brody, L. L., Coulter, S. N., Folger, K. R.,
Kas, A., Larbig, K., Lim, R. M., Smith, K. A., Spencer,
D. H., Wong, G. K.-S., Wu, Z., Paulsen, I. T., Reizer, J.,
Saier, M. H., Hancock, R. E. W., Lory, S., and Olson, M.
V., Complete genome sequence of Pseudomonas aeru-
ginosa PA01, an opportunistic pathogen. Nature, 406,
959–964 (2000).
30) Griswold, A. R., Chen, Y. Y., and Burne, R. A., Analysis
of an agmatine deiminase gene cluster in Streptococcus
mutans UA159. J. Bacteriol., 186, 1902–1904 (2004).
31) Nakada, Y., and Itoh, Y., Identification of the putrescine
biosynthesis genes in Psuedomonas aeruginosa and
characterization of agmatine deiminase and N-caba-
moylputrescine amidohydrolase of the arginine decar-
boxylase pathway. Microbiol., 149, 701–714 (2003).
32) Janowi, T., Kneifel, H., and Piotrowski, M., Identifica-
tion and characterization of plant agmatine imidohydro-
lase, the last missing link in polyamine biosynthesis of
plants. FEBS Lett., 544, 258–261 (2003).
33) Yanagisawa, H., Agmatine deiminase from maize
shoots: purification and properties. Pytochem., 56, 643–
647 (2001).
34) Chaudhuri, M. M., and Ghosh, B., Agmatine deiminase
in rice seedlings. Phytochem., 24, 2433–2435 (1985).
35) Park, K. H., and Cho, Y. D., Purification of monomeric
agmatine iminohydrolase from soy bean. Biochem.
Biophy. Res. Commun., 174, 32–36 (1991).
36) Piotrowski, M., Janowitz, T., and Kneifel, H., Plant C-N
hydrolases and the identification of a plant N-carba-
moylputrescine amidohydrolase involved in polyamine
biosynthesis. J. Biol. Chem., 278, 1708–1712 (2003).
37) Sakakibara, Y., and Yanagisawa, H., Agmatine deimi-
nase from cucumber seedling is a monospecific enzyme:
purification and characteristics. Protein Expr. Purif., 30,
88–93 (2003).
23) Bitonti, A. J., Casara, P. J., McCann, P. P., and Bey, P.,
Catalytic irreversible inhibition of bacterial and plant
arginine decarboxylase activities by novel substrate and
product analogues. Biochem. J., 242, 69–74 (1987).
24) Tabor, C. W., and Tabor, H., Polyamines in micro-
organisms. Microbiol. Rev., 49, 81–99 (1985).
25) Grishin, N. V., Phillips, M. A., and Goldsmith, E. J.,
Modeling of the spatial structure of eukaryotic ornithine
decarboxylases. Protein Sci., 4, 1291–1304 (1995).
38) Yanagisawa, H., and Suzuki, Y., Corn agmatine imido-
hydrolase. Plant Physiol., 67, 697–700 (1981).