The first thermophilic α-oxoamine synthase family enzyme that has activities of 2-amino-3-ketobutyrate CoA ligase and 7-keto-8-aminopelargonic acid synthase: Cloning and overexpression of the gene from an extreme thermophile, Thermus thermophilus, and characterization of its gene product

Article abstract of DOI:10.1271/bbb.70438

The first thermophilic α-oxoamine synthase family enzyme was identified. The gene (ORF TTHA1582), which is annotated to code putative α-oxoamine synthase family enzymes, 7-keto-8-aminopelargonic acid (KAPA) synthase (BioF, 8-amino-7-oxononanoate synthase, EC 2.3.1.47) and 2-amino-3-ketobutyrate CoA ligase (KBL, EC 2.3.1.29), in a genomic database, was cloned from an extreme thermophile, Thermus thermophilus, and overexpressed in Escherichia coli. The recombinant TTHA1582 protein was purified and characterized. It exhibited activity of BioF, which catalyzes the condensation of pimeloyl-CoA and L-alanine to produce a biotin intermediate KAPA, CoASH, and CO₂ with pyridoxal 5′-phosphate as a cofactor. The protein is a dimer with a subunit of 43 kDa that shows an amino acid sequence identity of 35% with E. coli BioF. The optimum temperature and pH were about 70 °C and about 6.0. The enzyme showed high thermostability at temperatures of up to 70 °C for 1 h, and a half-life of 1 h at 80 °C. Thus the TTHA1582 protein was found to have the highest optimum temperature and thermostablility of the α-oxoamine synthase family enzymes so far reported. Substrate specificity experiments revealed that it was also able to catalyze the KBL reaction, which used acetyl-CoA and glycine as substrates, and that enzyme activity was seen with the following combinations of substrates: acetyl-CoA and glycine, L-alanine, or L-serine; pimeloyl-CoA and L-alanine, glycine, or L-serine; palmitoyl-CoA and L-alanine. This suggests that the recombinant TTHA1582 protein has broad substrate specificity, unlike the reported mesophilic enzymes of the α-oxoamine synthase family.

Full text of DOI:10.1271/bbb.70438

Biosci. Biotechnol. Biochem., 71 (12), 3033–3040, 2007
The First Thermophilic ꢀ-Oxoamine Synthase Family Enzyme
That Has Activities of 2-Amino-3-ketobutyrate CoA Ligase
and 7-Keto-8-aminopelargonic Acid Synthase: Cloning
and Overexpression of the Gene from an Extreme Thermophile,
Thermus thermophilus, and Characterization of Its Gene Product
y
Takaaki KUBOTA, Jyunpei SHIMONO, Chie KANAMEDA, and Yoshikazu IZUMI
Department of Biotechnology, Tottori University, 4-101 Koyamacho-minami, Tottori 680-8552, Japan
Received July 11, 2007; Accepted August 20, 2007; Online Publication, December 7, 2007
[doi:10.1271/bbb.70438]
The first thermophilic ꢀ-oxoamine synthase family
Biotin (vitamin H) is an essential cofactor required
for a number of carboxylation, transcarboxylation, and
decarboxylation reactions in all living organisms. Its
participation in gluconeogenesis, the metabolism of
amino acids, and the initiation of fatty acid biosynthesis
makes it an indispensable entity in cellular metabolism.
In the microbial production of glutamic acid by
Corynebacterium glutamicum, biotin is known to be
an important compound that can markedly regulate the
production level.
enzyme was identified. The gene (ORF TTHA1582),
which is annotated to code putative ꢀ-oxoamine syn-
thase family enzymes, 7-keto-8-aminopelargonic acid
(
KAPA) synthase (BioF, 8-amino-7-oxononanoate syn-
thase, EC 2.3.1.47) and 2-amino-3-ketobutyrate CoA
ligase (KBL, EC 2.3.1.29), in a genomic database, was
cloned from an extreme thermophile, Thermus thermo-
philus, and overexpressed in Escherichia coli. The re-
combinant TTHA1582 protein was purified and char-
acterized. It exhibited activity of BioF, which catalyzes
the condensation of pimeloyl-CoA and L-alanine to pro-
duce a biotin intermediate KAPA, CoASH, and CO2
1
,2)
As for the biotin biosynthetic pathway, the cofactor is
synthesized from pimeloyl-CoA through 7-keto-8-ami-
nopelargonic acid (KAPA), 7,8-diaminopelargonic acid
(DAPA), and dethiobiotin (DTB) by the catalysis of four
0
with pyridoxal 5 -phosphate as a cofactor. The protein is
a dimer with a subunit of 43 kDa that shows an amino
3
,4)
5)
enzymes: KAPA synthase, DAPA aminotransferase,
)
6
7–11)
acid sequence identity of 35% with E. coli BioF. The
ꢀ
DTB synthase, and biotin synthase,
bioF, bioA, bioD, and bioB genes respectively.
encoded by the
optimum temperature and pH were about 70 C and
about 6.0. The enzyme showed high thermostability at
All studies of the biosynthetic pathway of biotin, and
related enzymes their regulatory mechanisms, and
microbial biotin production have so far been performed
exclusively using mesophilic organisms, such as micro-
ꢀ
temperatures of up to 70 C for 1 h, and a half-life of 1 h
ꢀ
at 80 C. Thus the TTHA1582 protein was found to have
the highest optimum temperature and thermostablility
of the ꢀ-oxoamine synthase family enzymes so far re-
ported. Substrate specificity experiments revealed that it
was also able to catalyze the KBL reaction, which used
acetyl-CoA and glycine as substrates, and that enzyme
activity was seen with the following combinations
of substrates: acetyl-CoA and glycine, L-alanine, or L-
serine; pimeloyl-CoA and L-alanine, glycine, or L-serine;
palmitoyl-CoA and L-alanine. This suggests that the
recombinant TTHA1582 protein has broad substrate
specificity, unlike the reported mesophilic enzymes of
the ꢀ-oxoamine synthase family.
3
,6–12)
organisms, including Bacillus sphaericus,
1
Esche-
richia coli, B. subtilis, and Saccharomyces cerevisiae,
3)
1
4)
and plants, including Arabidopsis thaliana and Lav-
andula vera. Moreover, from the standpoint of applied
microbiology, there has been no report on the synthesis
of biotin by combining enzymes, which have synthetic
activities of biotin intermediates even if they are not
originally biotin biosynthetic enzymes, from different
organisms on a genetic level.
Thus, with regard to both fundamental and applied
aspects, we have decided to find and classify (as a
library) various thermophilic enzymes that have activ-
ities to synthesize of biotin and its intermediates, based
on the known genome databases of thermophilic
organisms.
Key words: biotin; 7-keto-8-aminopelargonic acid syn-
thase; thermophilic enzyme; ꢀ-oxoamine
synthase family; 2-amino-3-ketobutyrate
CoA ligase
Enzymes from thermophiles are attractive in terms of
y
To whom correspondence should be addressed. Fax: +81-857-31-5267; E-mail: izumi@bio.tottori-u.ac.jp
Abbreviations: KAPA, 7-keto-8-aminopelargonic acid; BioF, KAPA synthase; KBL, 2-amino-3-ketobutyrate CoA ligase

3034
T. KUBOTA et al.
cultivation of recombinant E. coli. The other chemicals
used were commercial products.
NH2
O
OH
CoA-S
OH
H C
3
+
Construction of TTHA1582-overproducing strains.
The TTHA1582 gene was amplified by PCR from the
genomic DNA of T. thermophilus HB8 (which was
isolated using a Gene TLE kit, TakaraBio, Otsu, Japan)
O
O
L-Alanine
Pimeloyl-CoA
0
using primers 5 -GGAATTCCATATGAGCCTGGAT-
0
Pyridoxal
phosphate
0
CT-3 and 5 -ATAAGAATGCGGCCGCTTAGCGGA-
0
TTATACGGAG-3 . The amplified products were cloned
between the NdeI and NotI sites in pET-21a vector
CoA-SH + CO2
(Novagen, Madison, WI). Protein expression was carried
out in E. coli BL21 (DE3) cells (Novagen).
NH2
O
OH
O
Expression and purification of the TTHA1582 protein
from recombinant E. coli. E. coli BL21 (DE3) was
transformed with pET-21a/ttha1582, and the transform-
ant was grown at 30 C in 32 l of the medium described
above. When the cultures reached an optical density at
H C
3
ꢀ
7
-Keto-8-aminopelargonic acid (KAPA)
6
60 nm of 0.4, isopropyl-ꢁ-D-thiogalactopyranoside was
Fig. 1. Reaction of KAPA Synthase.
ꢀ
added to 0.5 mM. The cells were grown at 30 C for 12 h
after induction, harvested by centrifugation (12;000 ꢁ g,
ꢀ
5
min, 4 C), and washed with 10 mM of potassium
higher resistance to physical/chemical denaturation than
mesophiles. Also, the production of useful compounds at
higher temperatures by organisms having thermophilic
enzymes should have advantages in terms of low risk of
contamination and reactor cooling requirements.
phosphate buffer (pH 7.0). The precipitated cells (121 g
wet weight) were suspended in basal buffer (10 mM
potassium phosphate buffer, pH 7.0). Cell disruption
was carried out by sonication (Sonifier 450; Branson,
Danbury, CT), and cell debris was removed by centri-
ꢀ
In this study, we examined such enzymes from an
extreme thermophile, Thermus thermophilus (growth
fugation (19;000 ꢁ g, 30 min, 4 C). For purification,
ꢀ
crude cell extract was heat-treated (70 C, 60 min),
ꢀ
temperature,1 47–85 C), for which a gene manipula-
5)
ꢀ
followed by centrifugation (19;000 ꢁ g, 30 min, 4 C).
tion system is well established, and from which many
thermophilic proteins have been extensively character-
The supernatant fraction was applied to a DEAE-
Sepharose FF column equilibrated with the basal buffer.
After the column was washed with the basal buffer,
the bound proteins were eluted with the basal buffer
containing 0.1 M NaCl at a flow rate of 80 ml/h. The
active fractions were combined and concentrated by
ultrafiltration. The concentrated fraction was heat-
ized.1
6–21)
We started by investigating the genome
sequence of T. thermophilus HB8 to search for homo-
logs of E. coli bioF, bioA, bioD, and bioB in a genomic
database (GenBank accession no. AP008226). We found
that the TTHA1582 gene codes putative ꢀ-oxoamine
synthase family enzymes, KAPA synthase (BioF, 8-
amino-7-oxononanoate synthase, EC 2.3.1.47) and 2-
amino-3-ketobutyrate CoA ligase (KBL, EC 2.3.1.29).
The TTHA1582 gene product was overexpressed in
E. coli and found to have the activity of BioF, which
catalyzes the condensation of pimeloyl-CoA and L-
alanine to produce KAPA, CoASH, and CO2, with the
ꢀ
treated (65 C, 60 min), followed by centrifugation
ꢀ
(19;000 ꢁ g, 30 min, 4 C). The supernatant fraction
was dialyzed against the basal buffer containing 2 M
(NH ) SO , and was applied to a Phenyl-Toyopearl
4
2
4
column equilibrated with the buffer. After the column
was washed with the same buffer, the bound proteins
were eluted at a rate of 4.5 ml/h with a linear (NH4)2-
SO4 concentration gradient (2–0 M) formed in the buffer.
The active fractions were combined and concentrated by
ultrafiltration, and the purity of the protein was judged
by SDS–polyacrylamide electrophoresis (PAGE).
0
3,4)
participation of pyridoxal 5 -phosphate (Fig. 1).
Here we report the cloning and overexpression of the
TTHA1582 gene, and also the purification and enzymo-
logical characterization of the recombinant protein.
Materials and Methods
Assay of enzyme activity. KAPA synthase activity was
measured mainly using the bioassay described previ-
3
,4,22)
Materials. Pimeloyl-CoA was chemically synthesized
from pimelic acid as the starting material and purified
essentially as described by Ploux and Marquet.1 The
KAPA and DAPA used were kind gifts of Ajinomoto
Co., Inc., Tokyo. Luria-Bertani (LB) medium supple-
mented with 100 mg/ml ampicillin was used in the
ously.
The standard assay mixture, consisting of
100 mM of potassium phosphate buffer (pH 6.0), 50
mM of L-alanine, 0.2 mM of pimeloyl-CoA, 0.1 mM of
2)
0
pyridoxal 5 -phosphate, and enzyme solution in a final
ꢀ
volume of 50 ml, was incubated at 70 C for 1 h. The
amount of KAPA synthesized was determined micro-

Thermophilic KAPA Synthase from an Extreme Thermophile, T. thermophilus
3035
Fig. 2. Alignment of the Amino Acid Sequence of TTHA1582.
Alignment of the deduced amino acid sequence of TTHA1582 with the E. coli KBL (DDBJ protein ID: AAC76641) and E. coli BioF
(AAC73863) proteins. The alignment was generated with the ClustalW program. An asterisk indicates conserved residues, a colon indicates
similar residues, and a period indicates weakly similar residues. Residues that are highly conserved between the proteins are shown in bold, and
those numbered 1–5 have functions associated with E. coli BioF and KBL. That is, residues 1–4 are active site components. Residue 5 is part of a
0
potential substrate-binding site, and residue 4 is the pyridoxal 5 -phosphate cofactor-binding site.
biologically by the paper disk plate method using
Saccharomyces cerevisiae. One unit of enzyme activity
was defined as the amount of enzyme necessary to
synthesize 1 nmol of KAPA under the conditions de-
scribed.
previously.24) The reaction mixture (1–5 ml) was spotted
onto Toyo no. 51B filter paper (20 ꢁ 400 mm), and the
compounds were separated using a developing solvent
of n-butanol:1 N HCl (6:1, v/v). After drying, the
chromatograph was blotted for 1 h onto a biotin-free
agar plate impregnated with S. cerevisiae as the test
organism, as in the microbiological assay described
above. After the paper was removed, the agar plate was
Alternatively, enzyme activity was measured by
continuous spectrophotometric assay based on monitor-
ing the release of CoASH from acyl-CoA with Ellman’s
0
23)
ꢀ
reagent, 5,5 -dithiobis-2-nitrobenzoic acid (DTNB).
incubated for 20 h at 30 C. Rf values of growth spots for
KAPA, DAPA, DTB, and biotin were determined and
The standard assay mixture, in a final volume of 400 ml,
was basically the same as above, except that 0.1 mM
DTNB was added. The reaction was initiated by the
addition of L-alanine, and absorbance changes at 412 nm
due to formation of the thiophenolate (" ¼ 13;600
2
5)
compared with known values.
Results
412
ꢂ1
ꢂ1
ꢀ
M
cm ) were monitored for 3 min at 70 C in a V-
Identification and cloning of the bioF homolog in
T. thermophilus HB8
1700 UV–visible spectrophotometer (Shimadzu, Kyoto,
Japan).
Searches of the T. thermophilus HB8 genome data-
base (DDBJ accession no. AP008226) detected three
putative biotin metabolism-related genes, TTHA0607
(protein ID: BAD70430), TTHA0904 (protein ID:
BAD70727), and TTHA1582 (protein ID: BAD71405),
Identification of the reaction product of KAPA
synthase. The reaction product of KAPA synthase was
identified by bioautography using a procedure described

3036
T. KUBOTA et al.
encoding predicted proteins with 37%, 42%, and 35%
identity to the protein sequences of E. coli biotin
synthase (BioB), Rhizobium etli biotin transporter
protein (BioY), and E. coli KAPA synthase (BioF)
respectively.
1
2
3
4
5
6
9
7
The newly identified T. thermophilus putative bioF
TTHA1582) gene was used in a database search by
66
(
BLAST. Various sequences obtained from the BLAST
search were designated a putative ꢀ-oxoamine synthase
family enzyme bioF gene or a 2-amino-3-ketobutyrate
CoA ligase (KBL) gene. Alignment of the deduced
amino acid sequence of the TTHA1582 gene with
E. coli BioF and KBL (Fig. 2) indicated that TTHA1582
is 41% identical to E. coli KBL and 35% to E. coli BioF
based on sequence data retrieved from DDBJ. Hence we
cloned the TTHA1582 gene by PCR amplification of
T. thermophilus HB8 total DNA.
4
5
3
0
2
1
0
4
Overexpression and purification of the TTHA1582
protein
The TTHA1582 protein was overexpressed using the
pET-21a plasmid and the E. coli host strain BL21
(
DE3). When the E. coli transformant was grown under
optimum overexpression conditions, as described above,
the protein was expressed in soluble form. Then it was
purified essentially to homogeneity through four steps
Fig. 3. SDS–Polyacrylamide Gel Electrophoresis of Recombinant
TTHA1582.
Lane 1, marker proteins; lane 2, cell-free extract; lane 3, 1st heat-
treated extract; lane 4, pooled fraction after DEAE-Sepharose; lane
(
Fig. 3), as described in ‘‘Materials and Methods.’’
5, 2nd heat-treated extract; lane 6, purified enzyme after Phenyl-
Toyopearl.
Typically, about 37 mg of pure enzyme was obtained
from 121 g of wet cells. The molecular weight deter-
mined by SDS–PAGE was 43 kDa, which was consistent
with the molecular weight of 43.4 kDa calculated from
the deduced amino acid sequence. The apparent mo-
lecular mass of the native recombinant protein, as
estimated by Blue Native-PAGE (NativePAGE Bis-Tris
Gels; Invitrogen, Carlsbad, CA,26,27) was found to be
about 85 kD. These results suggest that the recombinant
enzyme has a dimeric structure in solution.
Table 1. Bioautographic Analysis of the Reaction Product Synthe-
sized by TTHA1582 Protein
ꢃ
Rf value
(Reference )
KAPA
DAPA
DTB
0.40
0.13
0.84
0.75
(0.35)
(0.10)
(0.82)
(0.70)
Biotin
Product
0.39
KAPA synthase activity of the TTHA1582 protein
Under the standard assay conditions, as described in
The reaction was carried out under the standard assay conditions. After the
reaction, the reaction mixture was subjected to ascending paper chromatog-
raphy followed by bioautography, as described in ‘‘Materials and Methods.’’
‘‘Materials and Methods,’’ the purified TTHA1582
protein exhibited KAPA synthase activity with a specific
ꢃData from reference 25
ꢂ1
activity of about 2,700 units mg protein. The reaction
product, KAPA, was identified by bioautography analy-
sis by comparison of its Rf value with that of the
authentic KAPA sample (Table 1). Then we examined
the enzymological properties of the KAPA synthase
activity of the recombinant TTHA1582 protein, as
described below.
temperatures (Fig. 5A). The enzyme showed a half-life
ꢀ
of 1 h at 80 C. It was also stable at pH 4.5–5.0, and a
loss of less than 10% of the initial activity was observed
after 1 h of incubation at these pH (Fig. 5B).
Substrate specificity and kinetic parameters
Effects of pH and temperature on activity and stability
When the enzyme activity was measured at various
pHs and temperatures, the optimum pH was found to be
The substrate specificity of the KAPA synthase
activity of the recombinant TTHA1582 was tested using
various acyl-CoAs other than pimeloyl-CoA and various
amino acids other than L-alanine, and a continuous
spectrophotometric assay based on monitoring the
release of CoASH, as described in ‘‘Materials and
Methods.’’ The results obtained are shown in Table 2.
When pimeloyl-CoA was used as an acyl-CoA substrate,
ꢀ
about 6.0 and the optimum temperature about 70 C
(Fig. 4). The KAPA synthase activity of the TTHA1582
protein was found to be relatively stable at temperatures
ꢀ
of up to 70 C, with a loss of less than 10% of the initial
activity observed after 1 h of incubation at these

Thermophilic KAPA Synthase from an Extreme Thermophile, T. thermophilus
3037
A
B
100
100
8
6
0
0
80
60
40
20
0
40
20
0
30
40 50 60 70 80 90 100
0
3
6
9
12
Temperature (°C)
pH
Fig. 4. Optimum pH and Temperature of KAPA Synthase Activity of TTHA1582.
Enzyme activity was measured at the indicated temperatures (A) and pHs (B). The enzyme reactions were carried out as described in
‘
‘Materials and Methods,’’ except that the temperature and pH were varied as indicated. In (B), the buffers used were citric acid-K2HPO4 ( ),
potassium phosphate ( ), and glycine-NaOH ( ). The total reaction volume was 0.05 ml, and 0.55 mM of purified enzyme was added to each
reaction mixture.
A
B
1
00
100
80
60
40
20
0
8
6
4
2
0
0
0
0
0
3
0
40 50 60 70 80 90 100
0
3
6
9
12
Temperature (°C)
pH
Fig. 5. Thermostability and pH Stability of KAPA Synthase Activity of TTHA1582.
Enzyme stability was measured at the indicated temperatures (A) and pHs (B). Samples of the purified enzyme were preincubated at each
temperature and pH for 1 h, and were tested for residual activity under the standard conditions. The enzyme reactions were carried out as
described in ‘‘Materials and Methods.’’ In (B), the buffers used were citric acid-K2HPO4 ( ), potassium phosphate ( ), and glycine-NaOH ( ).
The total reaction volume was 0.05 ml, and 0.55 mM of purified enzyme was added to each reaction mixture.
L-alanine was the most active amino acid substrate, and
glycine and L-serine were 50% as active as L-alanine.
When acetyl-CoA was used as the substrate, glycine was
Table 2. Substrate Specificity of TTHA1582 Protein
the most active, and L-alanine and L-serine were about
0% and 15% as active respectively as glycine. How-
Specific activity (nmol/min/mg)
Amino acida
6
acetyl-CoA
pimeloyl-CoA
palmitoyl-CoA
ever, when palmitoyl-CoA was used, only L-alanine
served as a substrate. In the reaction with either acyl-
CoA, no activity was seen with the L-alanine isomers, D-
alanine, or ꢁ-alanine, or the other amino acids listed in
Table 2.
None
0
190
109
0
0
Glycine
L-Alanine
L-Serine
1.8
3.6
1.8
0
6.4
0
27.3
The enzyme activity was measured by continuous spectrophotometric assay
according to the procedures described in ‘‘Materials and Methods.’’
aIn the reaction with acyl-CoA, no activity was detected with the following
amino acids: D-alanine, ꢁ-alanine, L-valine, L-leucine, L-isoleucine, L-glu-
tamic acid, L-aspartic acid, L-arginine, L-threonine, L-cysteine, L-methionine,
L-proline, L-phenylalanine, L-tyrosine, L-tryptophan, or L-histidine.
Kinetic parameters
The enzyme obeyed Michaelis-Menten kinetics
with respect to pimeloyl-CoA, acetyl-CoA, L-alanine,

3038
T. KUBOTA et al.
Table 3. Comparison of Steady-State Rate Constants for TTHA1582 Protein
pimeloyl-CoA
acetyl-CoA
L-alanine
L-glycine
Enzymes
Km
kcat=Km
Km
kcat=Km
Km
kcat=Km
ꢂ1 ꢂ1
Km
kcat=Km
ꢂ1 ꢂ1
3
ꢂ1 ꢂ1
3
ꢂ1 ꢂ1
3
(
mM)
(s
M
ꢄ10 )
(mM)
(s
M
ꢄ10 )
(mM)
(s
M
)
(mM)
(s
M
ꢄ10 )
TTHA1582
EC BioF28)
EC KBL29)
80 ꢅ 13
25 ꢅ 2
—
1:73 ꢅ 0:2
2:4 ꢅ 0:43
—
30
—
59
3.16
—
11:8 ꢅ 0:8
0:5 ꢅ 0:04
—
7:0 ꢅ 0:4
120 ꢅ 21
—
0.25
—
4.21
—
—
12
—
Kinetic parameters (ꢅSD) were determined by nonlinear regression analysis of the data using the Michaelis-Menten equation. Enzyme activity was measured by the
same procedures as described in Table 2, except that the substrate concentrations were changed.
Data from references 28, 29
ꢀ
15)
and glycine (Table 3). The Km values determined for
pimeloyl-CoA and L-alanine were about 3-fold and
temperature (65–72 C) of T. thermophilus HB8, and
the enzyme showed high thermostability (> 90%)
ꢀ
2
00-fold higher respectively than those determined
during incubation at temperatures of up to 70 C. KAPA
2
8)
for the E. coli BioF, indicating that TTHA1582 has
a lower affinity for the substrates of KAPA synthase.
However, the Km value determined for glycine when
acetyl-CoA was used as the acyl-CoA substrate of the
KBL reaction was much lower than that with the
corresponding E. coli KBL (about 1/50), indicating
that the TTHA1582 protein has a higher affinity for
glycine. The kcat=Km value determined for pimeloyl-
CoA was almost the same as that of E. coli BioF, while
that for L-alanine was about 20-fold lower.
synthase (EC 2.3.1.47), KBL (EC 2.3.1.29), 5-amino-
levulinate synthase (ALAS, EC 2.3.1.37), and serine
C-palmitoyltransferase (SPT, EC 2.3.1.50) form the ꢀ-
oxoamine synthase family. The TTHA1582 protein
had the highest optimum temperature of the enzymes of
the family so far reported: B. sphaericus BioF (60 C),
Rhodopseudomonas palustris ALAS (35–45 C), and
Rattus norvegicus (rat) liver SPT (50 C). Moreover,
the TTHA1582 protein also had the highest thermo-
stability of the enzymes so far reported: B. sphaericus
BioF (approximately 50% after 10 min of incubation at
3
1)
2
9)
ꢀ
32)
3)
ꢀ
ꢀ
33)
ꢀ
3)
Discussion
70 C) and human peripheral blood ALAS (approx-
ꢀ
34)
imately 50% after 30 min of incubation at 40 C).
0
With regard to both fundamental and applied aspects,
we initiated the present study with the aim of finding and
classifying as a library various thermophilic enzymes
that have activities to synthesize of biotin and its
intermediates, based on known genome databases of
thermophilic organisms. We were able to identify a
thermophilic enzyme that has KAPA synthase activity
for the first time. This is of significance in developing
the new research field of thermophilic biotin biosyn-
thesis.
From detailed BLAST searches of the T. thermophi-
lus genome database, three ORFs with similarities to bio
genes of other microorganisms were found: TTHA0607,
TTHA0904, and TTHA1582 (putative bioB, bioY, and
bioF respectively). No bioA or bioD genes were found in
BLASTX and BLASTP searches using homologs from a
diverse range of other microbes, and the ORFs identified
as putative bioB, bioF, and bioY were not organized in
an operon.
These pyridoxal 5 -phosphate-dependent enzymes
catalyze condensations between amino acids and car-
boxylic acid-CoA thioesters with concomitant decar-
boxylation of the amino acid, but no protein respon-
sible for multifunctional activity has so far been
3
5)
1
5,29,36)
characterized in this family.
The substrate speci-
ficity experiments on the recombinant TTHA1582
protein revealed it to have broad substrate specificity
as compared with such enzymes reported in mesophiles
(BioF, KBL). The protein was able to catalyze the
reaction of KBL, which uses acetyl-CoA and glycine as
substrates (Table 2). Furthermore, the activity was seen
with the following combinations of substrates: acetyl-
CoA and either L-alanine or L-serine; pimeloyl-CoA and
either glycine or L-serine; palmitoyl-CoA and L-alanine.
Recently, it was reported that Mycobacterium tuber-
culosis KAPA synthase shows a unique substrate
stereospecificity: it uses both L- and D-alanine,3 but
the recombinant TTHA1582 protein did not utilize D-
alanine.
From the facts described above, the TTHA1582
protein should be considered to be the first thermophilic
enzyme of the ꢀ-oxoamine synthase family. Hence it
can also be employed as a model for future studies of the
mechanisms of the thermostabilization and the substrate
specificity, and so on of ꢀ-oxoamine synthase family
enzymes. This requires further investigation of the
structure-function relationship using such approaches
as site-directed mutagenesis and 3-D structure determi-
nation. Furthermore, the TTHA1582 protein reported
7)
As for TTHA1582, which was chosen for the present
study from the above three ORFs, the recombinant
TTHA1582 protein exhibited KAPA synthase activity.
The purified TTHA1582 protein is probably a homo-
dimer of two identical subunits, with a molecular weight
of 43 kDa for each. This is similar to the structural
composition previously reported for KAPA synthase
3
0)
from E. coli (41.7 kDa for each subunit), but different
1
2)
from that of B. sphaericus (a monomer of 46 kDa).
The optimum temperature of the TTHA1582 protein
ꢀ
was about 70 C, as high as the optimum growth

Thermophilic KAPA Synthase from an Extreme Thermophile, T. thermophilus
3039
here might be applicable in the future not only to
microbial biotin production but also to the enzymatic
production of various ꢀ-oxoamine compounds at high
temperatures.
Studies of other thermophilic enzymes showing
biosynthetic activities of biotin and its intermediates
from various thermophilic or hyperthermophilic micro-
organisms are now in progress in our laboratory.
sphaericus bioB transformant. Biosci. Biotechnol. Bio-
chem., 58, 1738–1741 (1994).
1
0) M e´ jean, A., Bui, B. T. S., Florentin, D., Ploux, O., Izumi,
Y., and Marquet, A., Highly purified biotin synthase can
transform dethiobiotin into biotin in the absence of any
other protein, in the presence of photoreduced deaza-
flavin. Biochem. Biophys. Res. Commun., 217, 1231–
1
237 (1995).
1
1) Ohshiro, T., Kishimoto, T., Arase, M., and Izumi, Y.,
Characterization of the biotin synthase reaction from
Bacillus sphaericus using the photoreduced deazaflavin
system. J. Fement. Biotechnol., 86, 446–450 (1998).
Acknowledgments
We thank Professor S. Kuramitsu and Professor R.
Masui, Osaka University, Osaka, Japan, for providing
the T. thermophilus HB8 strain and for valuable advice.
We are also grateful to Professor A. Marquet and
Professor O. Ploux of the University of Pierre et Marie
Curie, Paris, France, to Professor T. Hoshino and
Professor A. Nakamura of Tsukuba University, Tsuku-
ba, Japan, and to Lecturer T. Ohshiro of Tottori
University, Tottori, Japan, for providing encouragement
and helpful suggestions.
12) Ploux, O., and Marquet, A., The 8-amino-7-oxopelarg-
onate synthase from Bacillus sphaericus: purification
and preliminary characterization of the cloned enzyme
overproduced in Escherichia coli. Biochem. J., 283,
3
27–331 (1992).
1
3) Ohshiro, T., Yamamoto, M., Bui, B. T. S., Florentin, D.,
Marquet, A., and Izumi, Y., Stimulatory factors for
enzymatic biotin synthesis from dethiobiotin in cell-free
extract of Escherichia coli. Biosci. Biotechnol. Bio-
chem., 59, 943–944 (1995).
14) Pinon, V., Ravanel, S., Douce, R., and Alban, C., Biotin
synthesis in plants: the first committed step of the
pathway is catalyzed by a cytosolic 7-keto-8-amino-
pelargonic acid synthase. Plant Physiol., 139, 1666–
1676 (2005).
15) Oshima, T., and Imahori, K., Description of Thermus
thermophilus (Yoshida and Oshima) comb. nov., a
nonsporulating thermophilic bacterium from a Japanese
thermal spa. Int. J. Syst. Bacteriol., 24, 102–112 (1974).
16) Yamagishi, A., Tanimoto, N., Suzuki, T., and Oshima,
T., Pyrimidine biosynthesis genes (pyrE and pyrF) of an
extreme thermophile, Thermus thermophilus. Appl.
Environ. Microbiol., 62, 2191–2194 (1996).
17) Hoseki, J., Yano, T., Koyama, Y., Kuramitsu, S., and
Kagamiyama, H., Directed evolution of thermostable
kanamycin-resistance gene: a convenient selection mark-
er for Thermus thermophilus. J. Biochem., 126, 951–956
(1999).
18) Shimizu, H., Yamagata, S., Masui, R., Inoue, Y.,
Shibata, T., Yokoyama, S., Kuramitsu, S., and Iwama,
T., Cloning and overexpression of the oah1 gene
encoding O-acetyl-L-homoserine sulfhydrylase of Ther-
mus thermophilus HB8 and characterization of the gene
product. Biochim. Biophys. Acta, 1549, 61–72 (2001).
19) Okamoto, A., Kato, R., Masui, R., Yamagishi, A.,
Oshima, T., and Kuramitsu, S., An aspartate amino-
transferase from an extremely thermophilic bacterium,
Thermus thermophilus HB8. J. Biochem., 119, 135–144
(1996).
References
1)
2)
3)
4)
5)
6)
Tanaka, K., Iwasaki, T., and Kinoshita, S., Studies on
L-glutamic acid fermentation. V. Biotin and L-glutamic
acid accumulation by bacteria. Nippon N o¯ geikagaku
Kaishi (in Japanese), 34, 593–600 (1960).
Aida, K., Glutamic acid. In ‘‘Biotechnology of Amino
Acid Production,’’ eds. Aida, K., Chibata, I., Nakayama,
K., Takinami, K., and Yamada, H., Kodansha Scientific,
Tokyo, pp. 120–124 (1986).
Izumi, Y., Morita, H., Tani, Y., and Ogata, K., Partial
purification and some properties of 7-keto-8-amino-
pelargonic acid synthetase, an enzyme involved in biotin
biosynthesis. Agric. Biol. Chem., 37, 1327–1333 (1973).
Izumi, Y., Sato, K., Tani, Y., and Ogata, K., Distribution
of 7-keto-8-aminopelargonic acid synthetase in bacteria
and the control mechanism of the enzyme activity.
Agric. Biol. Chem., 37, 1335–1340 (1973).
Izumi, Y., Morita, H., Tani, Y., and Ogata, K., 7,8-
Diaminopelargonic acid aminotransferase, an enzyme
involved in biotin biosynthesis by microorganisms.
Agric. Biol. Chem., 39, 175–181 (1975).
Izumi, Y., Kano, Y., Inagaki, K., Kawase, N., Tani, Y.,
and Yamada, H., Characterization of biotin biosynthetic
enzymes of Bacillus sphaericus: a dethiobiotin produc-
ing bacterium. Agric. Biol. Chem., 45, 1983–1989
(
1981).
7
)
Fujisawa, A., Abe, T., Ohsawa, I., Shiozaki, S.,
Kamogawa, K., and Izumi, Y., Bioconversion of
dethiobiotin into biotin by resting cells and protoplast
of Bacillus sphaericus bioB transformant. Biosci. Bio-
technol. Biochem., 57, 740–744 (1993).
Fujisawa, A., Abe, T., Ohsawa, I., Kamogawa, K., and
Izumi, Y., Bioconversion of dethiobiotin to biotin by a
cell-free system of a bioYB transformant of Bacillus
sphaericus. FEMS Microbiol. Lett., 110, 1–4 (1993).
Ohshiro, T., Yamamoto, M., Izumi, Y., Florentin, D.,
Bui, B. T. S., and Marquet, A., Enzymatic conversion of
dethiobiotin to biotin in cell-free extract of Bacillus
20) Hiramatsu, Y., Kato, R., Kawaguchi, S., and Kuramitsu,
S., Cloning and characterization of the uvrD gene from
an extremely thermophilic bacterium, Thermus thermo-
philus HB8. Gene, 199, 77–82 (1997).
21) Takamatsu, S., Kato, R., and Kuramitsu, S., Mismatch
DNA recognition protein from an extremely thermo-
philic bacterium, Thermus thermophilus HB8. Nucleic
Acids Res., 24, 640–647 (1996).
8
)
)
22) Tanaka, M., Izumi, Y., and Yamada, H., Biotin assay
using lyophilized and glycerol-suspended cultures. J.
Microbiol. Methods, 6, 237–246 (1987).
9
23) Ellman, G. L., Tissue sulfhydryl groups. Arch. Biochem.

3040
T. KUBOTA et al.
Biophys., 82, 70–77 (1959).
4) Ogata, K., Izumi, Y., and Tani, Y., The controlling
action of actithiazic acid on the biosynthesis of biotin-
Mechanism of ꢀ-oxoamine synthases: identification of
the intermediate Claisen product in the 8-amino-7-
oxononanoate synthase reaction. Biochemistry, 46,
5972–5981 (2007).
2
vitamers by various microorganisms. Agric. Biol. Chem.,
37, 1079–1085 (1973).
32) Choi, H. P., Hong, J. W., Rhee, K. H., and Sung, H. C.,
Cloning, expression, and characterization of 5-amino-
levulinic acid synthase from Rhodopseudomonas pal-
ustris KUGB306. FEMS Microbiol. Lett., 236, 175–181
(2004).
33) Williams, R. D., Wang, E., and Merrill, A. H., Jr.,
Enzymology of long-chain base synthesis by liver:
characterization of serine palmitoyltransferase in rat
liver microsomes. Arch. Biochem. Biophys., 228, 282–
291 (1984).
34) Furuyama, K., Harigae, H., Heller, T., Hamel, B. C. J.,
Minder, E. I., Shimizu, T., Kuribara, T., Blijlevens, N.,
Shibahara, S., and Sassa, S., Arg452 substitution of the
erythroid-specific 5-aminolaevulinate synthase, a hot
spot mutation in X-linked sideroblastic anaemia, does
not itself affect enzyme activity. Eur. J. Haematol., 76,
33–41 (2006).
35) Eliot, A. C., and Kirsch, J. F., Pyridoxal phosphate
enzymes: mechanistic, structural, and evolutionary con-
siderations. Annu. Rev. Biochem., 73, 383–415 (2004).
36) Fubara, B., Eckenrode, F., Tressel, T., and Davis, L.,
Purification and properties of aminoacetone synthetase
from beef liver mitochondria. J. Biol. Chem., 261,
12189–12196 (1986).
2
2
5) Iwahara, S., and Kanemaru, Y., Biosynthesis of biotin
from dethiobiotin by intact cells of biotin producing
bacterium. Agric. Biol. Chem., 39, 779–784 (1975).
6) Sch a¨ gger, H., and Von Jagow, G., Blue native electro-
phoresis for isolation of membrane protein complexes in
enzymatically active form. Anal. Biochem., 199, 223–
231 (1991).
2
7) Sch a¨ gger, H., Cramer, W. A., and Von Jagow, G.,
Analysis of molecular masses and oligomeric states of
protein complexes by blue native electrophoresis and
isolation of membrane protein complexes by two-
dimensional native electrophoresis. Anal. Biochem.,
217, 220–230 (1994).
2
2
3
3
8) Webster, P., Alexeev, D., Campopiano, D. J., Watt, R.
M., Alexeeva, M., Sawyer, L., and Baxter, R. L.,
Mechanism of 8-amino-7-oxononanoate synthase: spec-
troscopic, kinetic, and crystallographic studies. Bio-
chemistry, 39, 516–528 (2000).
9) Mukherjee, J. J., and Dekker, E. E., Purification,
properties, and N-terminal amino acid sequence of
homogeneous Escherichia coli 2-amino-3-ketobutyrate
CoA ligase, a pyridoxal phosphate-dependent enzyme.
J. Biol. Chem., 262, 14441–14447 (1987).
0) Alexeev, D., Alexeeva, M., Baxter, R. L., Campopiano,
D. J., Webster, S. P., and Sawyer, L., The crystal
structure of 8-amino-7-oxononanoate synthase: a bacte-
rial PLP-dependent, acyl-CoA-condensing enzyme. J.
Mol. Biol., 284, 401–419 (1998).
37) Bhor, V. M., Dev, S., Vasanthakumar, G. R., Kumar, P.,
Sinha, S., and Surolia, A., Broad substrate stereospeci-
ficity of the Mycobacterium tuberculosis 7-keto-8-ami-
nopelargonic acid synthase: spectroscopic and kinetic
studies. J. Biol. Chem., 281, 25076–25088 (2006).
1) Kerbarh, O., Campopiano, D. J., and Baxter, R. L.,