Isolation and characterization of racemase from Ensifer sp. 23-3 that acts on α-aminolactams and α-amino acid amides

Article abstract of DOI:10.1007/s10295-017-1981-5

Limited information is available on α-amino-ε-caprolactam (ACL) racemase (ACLR), a pyridoxal 5′-phosphate-dependent enzyme that acts on ACL and α-amino acid amides. In the present study, eight bacterial strains with the ability to racemize α-amino-ε-caprolactam were isolated and one of them was identified as Ensifer sp. strain 23-3. The gene for ACLR from Ensifer sp. 23-3 was cloned and expressed in Escherichia coli. The recombinant ACLR was then purified to homogeneity from the E. coli transformant harboring the ACLR gene from Ensifer sp. 23-3, and its properties were characterized. This enzyme acted not only on ACL but also on α-amino-δ-valerolactam, α-amino-ω-octalactam, α-aminobutyric acid amide, and alanine amide.

Full text of DOI:10.1007/s10295-017-1981-5

J Ind Microbiol Biotechnol (2017) 44:1503–1510
DOI 10.1007/s10295-017-1981-5
BIOCATALYSIS - ORIGINAL PAPER
Isolation and characterization of racemase from Ensifer sp. 23‑3
that acts on α‑aminolactams and α‑amino acid amides
Daisuke Matsui1,2 · Ken‑ichi Fuhshuku1,3 · Shingo Nagamori1 · Momoko Takata1 ·
Yasuhisa Asano1,2
Received: 20 June 2017 / Accepted: 10 September 2017 / Published online: 19 September 2017
© Society for Industrial Microbiology and Biotechnology 2017
Abstract Limited information is available on α-amino-ε-
caprolactam (ACL) racemase (ACLR), a pyridoxal 5′-phos-
phate-dependent enzyme that acts on ACL and α-amino acid
amides. In the present study, eight bacterial strains with the
ability to racemize α-amino-ε-caprolactam were isolated
and one of them was identifed as Ensifer sp. strain 23-3.
The gene for ACLR from Ensifer sp. 23-3 was cloned and
expressed in Escherichia coli. The recombinant ACLR was
then purifed to homogeneity from the E. coli transformant
harboring the ACLR gene from Ensifer sp. 23-3, and its
properties were characterized. This enzyme acted not only
on ACL but also on α-amino-δ-valerolactam, α-amino-ω-
octalactam, α-aminobutyric acid amide, and alanine amide.
Introduction
Cyclic α-amino acid amides are called α-aminolactams, and
they contain an amino group at the α-position. Of these,
α-amino-ε-caprolactam (ACL), which is synthesized from
lysine through a dehydration reaction [19], is one of the sev-
eral biologically active compounds such as capuramycins
[21] that signifcantly inhibit mycobacterial growth [16].
The cells of Cryptococcus laurentii which contain l-ACL
hydrolase [9, 11, 13] have been used in combination with the
cells of a soil bacterium, Achromobacter obae, which pro-
duces ACL racemase (ACLR, EC 5.1.1.15) in the cytoplasm
[1, 17] for the industrial production of l-lysine from dl-ACL
[10]. ACLR from A. obae has been characterized in detail
and has been found to be a pyridoxal 5′-phosphate (PLP)
enzyme. Achromobacter cycloclastes, Alcaligenes faecalis,
and Flavobacterium arborescens isolated from soil are also
considered to be producers of ACLR [10–12]. We have pre-
viously discovered α-amino acid amides as new substrates
for racemization by ACLR; when used in combination
with d- or l-enantioselective amino acid amide hydrolases,
ACLR is very efective for the dynamic kinetic resolution
of α-amino acid amides [3, 4, 22, 23]. We have reported
the synthesis of a series of optically active medium-sized
(seven- to nine-membered) α-aminolactams [7]. To prepare
d-α-aminolactams under mild conditions, we extended the
screening for microorganisms with the ability to enantiose-
lectively hydrolyze l-α-aminolactams [8].
Keywords α-Amino-ε-caprolactam · α-Aminolactam ·
α-Amino acid amide · Racemase · Racemization
Daisuke Matsui and Ken-ichi Fuhshuku contributed equally to
this work.
Electronic supplementary material The online version of this
article (doi:10.1007/s10295-017-1981-5) contains supplementary
material, which is available to authorized users.
* Yasuhisa Asano
asano@pu-toyama.ac.jp
1
Biotechnology Research Center and Department
To date, enzymes catalyzing the racemization of ACL
have been reported in some bacteria. However, only two
ACLRs from A. obae and Citreicella sp. SE45 had been
characterized in detail, including their substrate specifci-
ties [1, 18]. Accordingly, we screened for microorganisms
in soil that produced ACLR and found eight such strains. In
the present report, we describe its heterologous expression
of Biotechnology, Toyama Prefectural University, 5180
Kurokawa, Imizu, Toyama 939-0398, Japan
2
Asano Active Enzyme Molecule Project, ERATO, JST, 5180
Kurokawa, Imizu, Toyama 939-0398, Japan
3
Present Address: Department of Interdisciplinary Science
and Engineering, Meisei University, 2-1-1 Hodokubo, Hino,
Tokyo 191-8506, Japan
Vol.:(0123456789)
1 3

1504
J Ind Microbiol Biotechnol (2017) 44:1503–1510
in Escherichia coli BL21(DE3), and enzymatic racemization
30 °C. To purify the enzyme, the wet cells were obtained by
cultures of 40 fasks for three times (60 liters total).
of amino acid amides using the recombinant.
The E. coli BL21(DE3) harboring pET28a was sub-
cultured at 37 °C for 12 h in a test tube containing 5 mL
of LB medium supplemented with 50 μg/mL kanamycin.
The subculture was then transferred to 500 mL of the same
medium at 37 °C for 5 h in 8.0 L of LB medium supple-
mented with 50 μg/mL kanamycin. Then 0.5 mM isopropyl
β-d-thiogalactopyranoside was added to the culture, and the
culture was incubated at 37 °C for an additional 5 h.
Materials and methods
Materials
Both enantiomers of α-aminolactams [7, 20] and α-amino
acid amides [2] were prepared according to previously
described procedures. Restriction endonucleases and the
ligation reaction mixture were obtained from Toyobo Co.,
Ltd. (Osaka, Japan) and Takara Bio Inc. (Shiga, Japan). All
other chemicals were purchased from Sigma-Aldrich Japan
K.K. (Tokyo, Japan), Kanto Chemical Co., Inc. (Tokyo,
Japan), Nacalai Tesque, Inc. (Kyoto, Japan), or Tokyo
Chemical Industry Co., Ltd. (Tokyo, Japan), unless other-
wise stated, and were of the highest commercially available
grade.
Enzyme assay
The enzyme activity of ACLR was assayed by measuring
the conversion of one enantiomer of ACL to another. ACLR
solution (100 μL) was added to 100 mM KPB (pH 7.0) con-
taining 100 μmol of d- or l-ACL, 2.0 pmol of PLP (a total
volume of 1000 μL), and incubated at 30 °C for 5 min under
shaking at 1000 rpm. To terminate the reaction, the reac-
tion mixture was added to 400 μL of 300 mM HClO4. After
centrifugation (15,000×g, 10 min, 4 °C), a 300-μL aliquot
was withdrawn and mixed with 700 μL of water. Then, after
centrifugation (15,000×g, 10 min, 4 °C), the supernatant
was analyzed by HPLC using a Crownpak CR(+) column
(5 µm, 4 mm × 150 mm; Daicel Corp.) to detect recovered
ACL. One unit of enzyme activity was defned as the amount
of enzyme required to convert 1 μmol of d-ACL to l-ACL
per minute under standard conditions.
Screening method for ACL racemization activity
The ACL-assimilating microorganisms were isolated accord-
ing to our previous reports [8]. The cells from 2.0 mL of the
ACL medium were harvested by centrifugation (15,000×g,
5 min, 4 °C), and suspended in 990 μL of 20 mM potassium
phosphate bufer (KPB) (pH 7.0) and washed with saline
(0.85%). To this suspension, 10 μL of 20% d- or l-ACL
aqueous solution was added as a substrate. After incubat-
ing under shaking at 1000 rpm for 24 h at 30 °C, a 200-μL
aliquot was withdrawn and mixed with 400 μL of 300 mM
HClO4. After centrifugation (15,000×g, 10 min, 4 °C), a
300-μL aliquot was withdrawn and mixed with 700 μL of
water. Next, after centrifugation (15,000×g, 10 min, 4 °C),
the supernatant was analyzed by high-performance liquid
chromatography (HPLC) using a Crownpak CR(+) col-
umn (5 µm, 4 mm × 150 mm; Daicel Corp., Osaka, Japan)
(solvent, 60 mM HClO4; fow rate, 0.30 mL/min; column
temperature, 30 °C; detection wavelength, 200 nm) for the
detection of ACL (tR = 5.8 min for the d-form and 8.2 min
for the l-form).
For the study of substrate specifcity, the yield and enan-
tiomeric excess of all compounds were determined by HPLC
using a Crownpak CR(+) column (5 µm, 4 mm × 150 mm;
Daicel Corp.), as described below
α‑Amino‑δ‑valerolactam (AVL)
Solvent, 60 mM HClO4; fow rate, 0.10 mL/min; column
temperature, 5 °C; detection wavelength, 200 nm; and tR,
24.4 min for the l-form and 25.8 min for the d-form.
α‑Amino‑ω‑octalactam (AOL)
Culture conditions
Solvent, 60 mM HClO4; fow rate, 0.30 mL/min; column
temperature, 5 °C; detection wavelength, 200 nm; and tR,
7.4 min for the d-form and 9.1 min for the l-form.
The cells of Ensifer sp. 23-3 (TPU 6133) were inoculated
into 5.0 mL of TGY medium consisting of 5.0 g of peptone,
5.0 g of yeast extract, 1.0 g of glucose, and 1.0 g of K2HPO4
in 1000 mL of water and incubated under reciprocal shak-
ing at 300 strokes/min for 22 h at 30 °C. The cells were
then transferred into 500 mL of SACL medium in a 2-L baf-
fed fask, which was the same as the above-described ACL
medium, except for the addition of 0.50% sodium succinate,
and were incubated under shaking at 150 rpm for 24 h at
Alanine amide
2-Aminobutyric acid amide: solvent, 60 mM HClO4; fow
rate, 0.30 mL/min; column temperature, 5 °C; detection
wavelength, 200 nm; and tR, 6.0 min for the d-form and
9.3 min for the l-form.
1 3

J Ind Microbiol Biotechnol (2017) 44:1503–1510
1505
Purifcation of racemase from Ensifer sp. 23‑3
Analytical methods
and recombinant enzyme
Protein concentrations were measured using the Bio-Rad
protein assay kit (Bio-Rad Laboratories, Hercules, CA,
USA) with bovine serum albumin as a standard [5].
All purifcation procedures were performed at temperatures
below 4 °C. Wet cells (313 g) from 60 L of the culture were
suspended in 1.5 L of 20 mM KPB (pH 7.0) and disrupted
by Insonator 201 M (19 kHz; Kubota Corp., Tokyo, Japan)
for 20 min. After centrifugation (20,000×g, 30 min), the
supernatant was used as the cell-free extract.
SDS-PAGE was performed with an electrophoresis unit
(ATTO Corp., Tokyo, Japan). The molecular weight of an
enzyme subunit was estimated by SDS-PAGE on the basis of
relative mobility compared with those of standard proteins
(Bio-Rad Laboratories): phosphorylase b (97,400), bovine
serum albumin (66,200), ovalbumin (45,000), carbonic
anhydrase (31,000), trypsin inhibitor (21,500), and lysozyme
(14,400). The protein bands obtained in SDS-PAGE were
stained with Brilliant Blue G (Tokyo Chemical Industry
Co., Ltd.) and destained in MeOH/AcOH/water (3/1/6). The
molecular weight of the native enzyme was estimated by gel-
fltration HPLC using a Superdex 200 10/300 GL column
with 20 mM KPB (pH 7.0) containing 150 mM NaCl as the
eluent at 0.45 mL/min on the basis of relative mobility com-
pared with those of standard proteins (Oriental Yeast Co.,
Ltd., Tokyo, Japan): glutamate dehydrogenase (290,000),
lactate dehydrogenase (142,000), enolase (67,000), ade-
nylate kinase (32,000), and cytochrome C (12,400).
The cell-free extract was applied to a Q Sepharose (GE
Healthcare UK Ltd., Buckinghamshire, England) column
(300 mL), and the adsorbed enzyme was eluted with a
linear gradient of NaCl (0–300 mM) in 20 mM KPB (pH
7.0). The active fractions were brought to 30% (NH4)2SO4
saturation, and applied to a Butyl-Toyopearl 650 M (Tosoh
Corp., Tokyo, Japan) column (100 mL) equilibrated with
20 mM KPB (pH 7.0) containing 30% saturated (NH4)2SO4.
The adsorbed enzyme was eluted with a linear gradient of
(NH4)2SO4 (30–0% saturation) in 20 mM KPB (pH 7.0).
The active fractions were dialyzed against 20 mM KPB (pH
7.0) and applied to a DEAE-Toyopearl 650 M (Tosoh Corp.,
Tokyo, Japan) column (85 mL) equilibrated with 20 mM
KPB (pH 7.0). The adsorbed enzyme was eluted with a lin-
ear gradient of NaCl (0–500 mM) in 20 mM KPB (pH 7.0).
Then, the active fractions were applied to a Mono Q 10/100
GL (GE Healthcare UK Ltd.) column (8 mL), and eluted by
some conditions as DEAE-Toyopearl. The active fractions
were further applied to a Resource PHE (GE Healthcare
UK Ltd.) column (1 mL) and eluted by the same condi-
tions as Butyl-Toyopearl 650 M. The active fractions were
concentrated using Amicon Ultra-4 (Merck KGaA, Darm-
stadt, Germany), and applied to a Superdex 200 10/300 GL
(GE Healthcare UK Ltd.) column (24 mL) equilibrated with
20 mM KPB (pH 7.0).
Cloning, sequence analysis and expression of gene
encoding ACLR from Ensifer sp. 23‑3
Chromosomal DNA was isolated using a cell culture of
Ensifer sp. 23-3, and an overlapping region containing the
5′-teminal and 3′-terminal sequence of the ACLR gene (aclr)
was amplifed from the DNA of Ensifer sp. 23-3 using the
following primers, 5′-GGCGACAACGAGCCCTACAAC
ATCC-3′ and 5′-GATCGCGAAACGCTGGAGCCCGCG
-3′, which were based on alignments with other homologous
enzymes with an N-terminal amino acid sequence. The prod-
uct was purifed, cloned into the Novagen pT7 Blue T-vector
(Merck KGaA), and sequenced using an ABI PRISM 3500
genetic analyzer (Thermo Fisher Scientifc Inc., MA, USA).
The pET protein expression system, which is induced by
isopropyl-β-thiogalactopyranoside, was employed for the
overexpression of gene encoding ACLR from Ensifer sp.
23-3. The structural gene was amplifed from genomic DNA
by PCR using two primers, 5′-CGCGGCAGCCATATGGTG
ACCGAAGGCCTTTACGCCAGGG-3′ and 5′-CTCGAA
TTCGGATCCTCACCAGCCGGCGAATTCTGAAAGC-
3′, with Gfex DNA polymerase (Takara Bio Inc., Shiga,
Japan) for the construction of expression system. After the
digestion of pET28a by NdeI and BamHI, the amplifed gene
was ligated into pET28a using the In-Fusion HD cloning kit
(Takara Bio Inc.). The resulting construct (pET28a-aclr) was
introduced into E. coli BL21(DE3), and the recombinant
strain was cultured as described above.
The recombinant ACLR from the E. coli transformant
harboring the ACLR gene from Ensifer sp. 23-3 was purifed
according to previously described procedures [17, 18]. The
cells were harvested by centrifugation at 8000 × g for 10 min
at 4 °C and washed with saline (0.85%). The cells were
resuspended in 20 mM KPB (pH 7.0), containing 250 mM
sucrose and 20 μM PLP, and disrupted by sonication for
10 min. The lysate was centrifuged at 8000×g for 10 min
at 4 °C, and the supernatant was purifed by nickel afnity
chromatography using His GraviTrap (GE Healthcare UK
Ltd.) according to the manual. The active fraction was used
for the characterization of recombinant ACLR.
Determination of N‑terminal amino acid sequence
For the determination of the N-terminal amino acid sequence
of racemase from Ensifer sp. 23-3, the active fraction was
analyzed using the Procise 494 HT protein sequence system
at APRO Life Science Institute Inc. (Tokushima, Japan).
1 3

1506
J Ind Microbiol Biotechnol (2017) 44:1503–1510
Efects of metal ions and inhibitors
culture, using ion-exchange, hydrophobic, and gel-fltration
column chromatographies. The fnal step of chromatog-
raphy provided a single band with a molecular weight of
approximately 45,000 on SDS-PAGE, which corresponded
to the only peak on the chromatogram exhibiting activity
toward ACL, as shown in Fig. 1a. The molecular weight of
the native enzyme was determined to be 49,800 by high-
performance gel-fltration chromatography, which indicated
that the native enzyme is a monomer.
Enzyme activity was measured under standard conditions,
after pre-incubating recombinant ACLR from Ensifer sp.
23-3 with various compounds at a fnal concentration of
1.0 mM for 10 min at 30 °C.
Results and discussion
Since the purifcation to homogeneity of racemase from
Ensifer sp. 23-3 was very difcult, we cloned the gene for
racemase based on the N- terminal amino acid sequence of
purifed racemase enzyme from Ensifer sp. 23-3 and express
its gene in E. coli for the characterization of the ACLR.
Isolation and identifcation of microorganisms
We have previously described the isolation of microorgan-
isms that could assimilate ACL as the sole source of carbon
and nitrogen from 115 soil samples and have identifed Mes‑
orhizobium sp. L88 and Aneurinibacillus migulanus L168
as producers of ACL hydrolase [8]. After observing growth
on a screening medium containing 0.20% dl-ACL, Ensifer
sp. 23-3 was selected.
Identifcation of nucleotide sequence
and overexpression of gene in E. coli
The N-terminal amino acid sequence of the enzyme was
Met-Thr-Glu-Gly-Leu-Tyr-Ala-Arg using the above puri-
fed enzyme from Ensifer sp. 23-3. The gene for ACLR was
amplifed from the chromosomal DNA of Ensifer sp. 23-3
by PCR, and the sequence of the entire replicon length was
determined. The gene consisted of 1320 bp encoding a pro-
tein of 439 amino acids. The predicted molecular weight of
the ORF was in agreement with that of a subunit of ACLR
The growth of Ensifer sp. 23-3 and 64-1 in a large-scale
incubation was relatively high among the other strains,
which were obtained during this screening (Table S1). Since
the purifcation procedure from Ensifer sp. 23-3 was com-
paratively easier than that from 64-1, we selected Ensifer sp.
23-3 as a candidate to purify and characterize ACLR.
The taxonomic characteristics of strain 23-3 were as
follows: rod-shaped, Gram-negative, non-spore-forming,
motile, glucose oxidation-negative, catalase-positive, and
oxidase-positive [14]. The most variable region of the 16S
rRNA sequence revealed >99% identity to Ensifer sp. [6].
These characteristics indicated that strain 23-3 belonged to
the genus Ensifer.
Cultivation of Ensifer sp. 23‑3
To purify racemase from Ensifer sp. 23-3, the culture condi-
tions suitable for inducing potent ACL racemization were
explored. With regard to carbon sources, sodium succi-
nate induced good racemization activity (data not shown).
Although racemase from Ensifer sp. 23-3 could be expressed
in an ACL-free medium such as TGY medium, with regard
to nitrogen sources, ACL was most efective for cell growth
and the induction of racemase activity. To minimize the
amount of ACL used, diferent concentrations of dl-ACL
were examined, and 0.10% was found to be suitable for the
large-scale incubation, since ACL concentrations lower than
0.10% afected the induction of racemization activity.
Fig. 1 SDS-PAGE of purifed racemase from Ensifer sp. 23-3. Lane
1 and 3, standards: phosphorylase b (97,400), bovine serum albumin
(66,200), ovalbumin (45,000), carbonic anhydrase (31,000), trypsin
inhibitor (21,500), and lysozyme (14,400). Lane 2, purifed racemase
from Ensifer sp. 23-3. Lane 4 and 5, purifed racemase from E. coli
BL21(DE3) without and with additives (10−4 M PLP, 0.1% 2-mercap-
toethanol, and 0.03 M sucrose), respectively
Purifcation of racemase from Ensifer sp. 23‑3
As summarized in Table S2, racemase from Ensifer sp.
23-3 was purifed 31.9-fold to homogeneity, with an overall
yield of 0.0749% from the cell-free extracts from the 60-L
1 3

J Ind Microbiol Biotechnol (2017) 44:1503–1510
1507
Table 1 Substrate specifcity of ACLR from Ensifer sp. 23-3 (n = 3)
from Ensifer sp. 23-3. In addition, the deduced amino acid
sequence of the N-terminal region of ORF was identical to
the amino acid sequence obtained from the purifed pro-
tein. Therefore, we concluded that the ORF is the gene of
ACLR from Ensifer sp. 23-3. After the culture of the E. coli
BL21(DE3) transformant harboring the ACLR gene from
Ensifer sp. 23-3, the recombinant ACLR protein was puri-
fed by nickel afnity column chromatography [18, 19]. The
molecular weight of the purifed recombinant enzyme with
the stabilizer was the same as that of the enzyme from Ensi‑
fer sp. 23-3, but the purifed enzyme without the stabilizers
was not stable and resolved, as shown in Fig. 1b. The recom-
binant enzyme showed similar ACL racemization activity to
the wild-type enzyme obtained from Ensifer sp. 23-3.
The ORF has a sequence identical to that of aspartate
aminotransferase family protein from Ensifer adhaerens
OV14 in the DDBJ nucleotide sequence database (acces-
sion number, LC171379). The amino acid sequence was
used to search for homologous sequences using the BLAST
program. The deduced amino acid sequence of ACLR from
Ensifer sp. 23-3 exhibited high similarity (99%) to that of
aspartate aminotransferase family protein (WP_025424688)
from Ensifer. The deduced amino acid sequence and charac-
terized ACLRs were aligned (Fig. S1), and a phylogenetic
tree was drawn (Fig. S2). ACLR from Ensifer sp. 23-3 also
has residues that are conserved in other ACLRs [19]. The
search also indicated that the amino acid sequence for the
PLP-binding pocket [15] was conserved and the PLP cofac-
tor may bind to the conserved Lys268 residue, forming a
Schif’s base. The predicted structure constructed using
SWISS-MODEL (https://swissmodel.expasy.org/) was com-
pared with a crystal structure of ACLR from A. obae (PDB:
3DXV), and the position of Lys268 residue was the similar
to that of Lys267 residue which forms a Schif’s base with
PLP (Fig. S3).
Substrate
Relative activity (%)
d-ACL
l-ACL
100 4.08 (365 U/mg)
97.1 3.12
d-AVL
l-AVL
111 2.71
17.2 1.90
d-AOL
l-AOL
1.78 0.021
1.41 0.074
2.12 0.051
0.768 0.022
0.192 0.020
0.082 0.005
d-Alanine amide
l-Alanine amide
d-2-Aminobutyric acid amide
l-2-Aminobutyric acid amide
None of the enantiomers of the following compounds were substrates
for ACLR from Ensifer sp. 23-3: glutamic acid amide, glutamine
amide, lysine amide, phenylalanine amide, valine amide, and their
corresponding α-amino acids
Table 2 Kinetic characterization of ACLR from Ensifer sp. 23-3
Substrate
kcat (s−1
)
Km (mM)
kcat/Km
(s−1/mM)
l-ACL
d-Alanine amide
l-Alanine amide
d-2-Aminobutyric acid amide
l-2-Aminobutyric acid amide
371
10.2
3.3
1.1
0.7
11.2
3.8
1.3
3.5
2.1
33.1
2.68
2.54
0.31
0.33
their corresponding α-amino acids were inert as substrates
for ACLR from Ensifer sp. 23-3.
Kinetic parameters of the d- and l-forms of two amino
acid amides (2-aminobutyric acid amide and alanine amide)
and l-ACL were calculated using the Lineweaver–Burk plots
(Table 2). The kcat/Km values for the d-form were similar to
those for the l-form.
PLP may be tightly bound to the enzyme, because the
racemization activity did not decrease signifcantly follow-
ing dialysis against the PLP-free bufer (data not shown).
The stability with some additives was examined using puri-
fed recombinant enzymes, and found PLP, sucrose, and
2-mercaptoethanol function as stabilizers (Fig. S4).
Figure 2 shows the time course of the racemization
reaction of l-ACL using recombinant ACLR from Ensifer
sp. 23-3. The reaction mixture contained 2.0 μM of PLP,
100 mM of KPB (pH 7.0), 40 mM of substrate, and ACLR.
The reaction mixture was incubated at 30 °C, and control
reactions were performed without enzyme under the same
conditions. After 70 min, the l-form was completely race-
mized by recombinant ACLR from Ensifer sp. 23-3.
Substrate specifcity and kinetic characterization
of ACLR from Ensifer sp. 23‑3
To investigate substrate specificity, purified ACLR
from Ensifer sp. 23-3 was used to racemize various
α-aminolactams and α-amino acid amides, and its activ-
ity was assayed (Table 1). In addition to ACL, the enzyme
acted on various α-aminolactams and α-amino acid amides,
including AVL, AOL, alanine amide, and 2-aminobutyric
acid amide. However, glutamic acid amide, glutamine
amide, lysine amide, phenylalanine amide, valine amide, and
Efects of pH and temperature on activity and stability
of ACLR from Ensifer sp. 23‑3
The efects of pH and temperature on the activity of ACLR
from Ensifer sp. 23-3 were examined. Activity was measured
at various pH values in several 100 mM bufers at 30 °C
and at various temperatures in 100 mM KPB (pH 7.0). The
1 3

1508
J Ind Microbiol Biotechnol (2017) 44:1503–1510
Fig. 2 Enzymatic racemization of l-ACL (open circle) to d-ACL
(flled circle) by recombinant ACLR from Ensifer sp. 23-3. The
reaction mixture (0.50 mL) contained 50 mmol of KPB (pH 7.0),
1.0 nmol of PLP, 20 mmol of l-ACL, and 10 mg of recombinant
ACLR from Ensifer sp. 23-3. The reaction was followed by HPLC at
the point shown in the graphics
enzyme exhibited maximum activity between pH 6.5 and 10
and at 65 °C, as shown in Fig. 3.
The stability of the enzyme was examined at various pH
values and temperatures. The enzyme was pre-incubated at
various pH values in several 20 mM bufers for 30 min at
30 °C. An aliquot of each enzyme solution was then taken,
and the remaining activity was assayed using d-ACL as a
substrate under standard conditions. As shown in Fig. 4a,
the enzyme was stable between pH 6.0 and 9.0. After pre-
incubation of the enzyme at various temperatures in 20 mM
KPB (pH 7.0) for 30 min, the remaining activity of each
enzyme was measured. The enzyme was stable up to 60 °C
and was inactivated above 70 °C, as shown in Fig. 4b.
Fig. 3 Optimal pH and temperature on activity of ACLR from Ensi‑
fer sp. 23-3. a Optimal pH: activity was measured at various pH val-
ues in the following 100 mM bufers at 30 °C: sodium citrate bufer
(pH 3.0–6.0), KPB (pH 6.0–7.5), Tris–HCl bufer (pH 7.5–9.0), and
sodium bicarbonate bufer (pH 9.5–10.5). b Optimal temperature:
activity was measured at various temperatures (30–95 °C) in 100 mM
KPB (pH 7.0)
by d-penicillamine, 59% inhibition by 8-hydroxyquinoline,
and 21% inhibition by d-cycloserine, which as carbonyl
reagents, are typical inhibitors of PLP-dependent enzymes.
Further, 36% of racemization activity could be recovered by
the subsequent addition of 1.0 mM PLP. The inhibition by
PLP was obtained. These results were consistent with those
obtained using recombinant ACLR, which is a PLP-depend-
ent enzyme and as with those obtained using ACLR from
A. obae in the inhibition type and recovery with PLP [1].
Efects of metal ions and inhibitors
Each ACLR activity of the recombinant enzyme was meas-
ured after incubation with various compounds at a fnal
concentration of 1.0 mM for 10 min at 30 °C. The enzyme
was completely inhibited by the addition of CuCl2 and
HgCl2, and showed 88% inhibition by AgNO3, 62% inhibi-
tion by MnCl2, 58% inhibition by NiCl2, 57% inhibition by
AlCl3, 57% inhibition by CdCl3, 43% inhibition by ZnCl2,
32% inhibition by CoCl2, and 23% inhibition by CrCl3. The
enzyme lost its activity when incubated with chelating rea-
gents such as 5,5′-dithiobis(2-nitrobenzoic acid) (100%)
and N-ethylmaleimide (50%) but was only weakly afected
by iodoacetic acid (12%). The serine protease inhibitor
phenylmethane sulfonyl fuoride also inhibited the ACLR
activity (59%). The enzyme was completely inhibited by
the addition of hydroxylamine, and showed 88% inhibition
Conclusions
Strains with the ability to racemize ACL were isolated in the
present study, and one of them was identifed as Ensifer sp.
strain 23-3. The gene for ACLR from Ensifer sp. 23-3 was
cloned and expressed in E. coli. The recombinant ACLR
was purifed to homogeneity from an E. coli transformant,
and its properties were characterized. The enzyme activi-
ties for ACL (100% toward d-ACL as a substrate) and AVL
1 3

J Ind Microbiol Biotechnol (2017) 44:1503–1510
1509
for the Promotion of Science (JSPS), and JST ERATO Asano Active
Enzyme Molecule Project (Grant No. JPMJER1102), Japan. The
authors appreciate Dr. Kimiyasu Isobe for his critical reading of the
manuscript and many helpful comments.
Compliance with ethical standards
Confict of interest The authors declare that they have no confict
of interest.
Ethical approval This article does not contain any studies with
human participants or animals performed by any of the authors.
References
1. Ahmed SA, Esaki N, Tanaka H, Soda K (1983) Properties of
α-amino-ε-caprolactam racemase from Achromobacter obae.
Agric Biol Chem 47:1887–1893
2. Asano Y, Nakazawa A, Kato Y, Kondo K (1989) Properties of
a novel d-stereospecifc aminopeptidase from Ochrobactrum
anthropi. J Biol Chem 264:14233–14239
3. Asano Y, Yamaguchi S (2005) Discovery of amino acid amides as
new substrates for α-amino-ε-caprolactam racemase from Achro‑
mobacter obae. J Mol Catal B 36:22–29
4. Asano Y, Yamaguchi S (2005) Dynamic kinetic resolution of
amino acid amide catalyzed by d-aminopeptidase and α-amino-
ε-caprolactam racemase. J Am Chem Soc 127:7696–7697
5. Bradford MM (1976) A rapid and sensitive method for the quan-
titation of microgram quantities of protein utilizing the principle
of protein-dye binding. Anal Biochem 72:248–254
Fig. 4 Efects of pH and temperature on stability of ACLR from
Ensifer sp. 23-3. a pH stability: activity was measured after a pre-
incubation at various pH values in the following 20 mM bufers for
30 min at 30 °C: sodium citrate bufer (pH 3.0–6.0), KPB (pH 6.0–
7.5), Tris–HCl bufer (pH 7.5–9.0), and sodium bicarbonate bufer
(pH 9.5–10.5). b Thermostability: activity was measured after a pre-
incubation at various temperatures (30–95 °C) in 20 mM KPB (pH
7.0) for 30 min
6. Casida L Jr (1982) Ensifer adhaerens gen. nov., sp. nov.: a bacte-
rial predator of bacteria in soil. Int J Syst Bacteriol 32:339–345
7. Fuhshuku K, Asano Y (2012) Synthesis of optically active
medium-sized α-aminolactams via ring-closing metathesis. Tet-
rahedron 68:6651–6655
8. Fuhshuku K, Takata M, Iwatsubo H, Asano Y (2014) Prepara-
tion of d-α-aminolactams by l-enantioselective degradation of
α-aminolactam mediated by Mesorhizobium sp. L88. Biocatal
Agric Biotechnol 3:42–47
9. Fukumura T (1976) Enzymatic conversion of dl-α-amino-ε-
caprolactam into l-lysine. I. Screening, classifcation and distri-
bution of l-α-amino-ε-caprolactam-hydrolyzing yeasts. Agric Biol
Chem 40:1687–1693
(111% of the activity toward d-AVL as a substrate) were
high, and the enzyme also acted on AOL, alanine amide,
and α-aminobutyric acid amide slightly (1.78, 2.12, and
0.192% of the activity toward d-AOL, d-alanine amide, and
d-2-aminobutyric acid amide as substrates, respectively,
Table 1). This is the third case of the purifcation of ACLR
and its characterization in detail. The information such as
substrate specifcity and stability are very important for
application. Furthermore, we are interested in the metabo-
lism of ACL and want to know how microorganisms evolved
specifc enzymes that can racemize ACL. We could isolate a
candidate bacterium for further study, because we only had
genetic imformation of the other strains such as A. obae and
Citreicella sp. SE45 and their ACLRs.
10. Fukumura T (1977) Enzymatic conversion of dl-α-amino-ε-
caprolactam into l-lysine. Part III. Bacterial racemization of
α-amino-ε-caprolactam. Agric Biol Chem 41:1321–1325
11. Fukumura T (1977) Enzymatic conversion of dl-α-amino-ε-
caprolactam into l-lysine. Part VI. Conversion of d- and dl-α-
amino-ε-caprolactam into l-lysine using both yeast cells and
bacterial cells. Agric Biol Chem 41:1327–1330
12. Fukumura T (1977) Partial purifcation and some properties of
α-amino-ε-caprolactam-racemizing enzyme from Achromobacter
obae. Agric Biol Chem 41:1509–1510
13. Fukumura T, Talbot G, Misono H, Teramura Y, Kato K, Soda
K (1978) Purifcation and properties of a novel enzyme, l-α-
amino-ε-caprolactamase from Cryptococcus laurentii. FEBS Lett
89:298–300
14. Garrity G, Staley JT, Boone DR, De Vos P, Goodfellow M, Rainey
FA, Schleifer K-H (2006) Bergey’s manual of systematic bacteri-
ology, vol 2. The proteobacteria. Springer, Berlin
Acknowledgements This research was supported by a Grant-in-Aid
for Scientifc Research (A) (23248015) to Y. Asano, Grant-in-Aid for
Young Scientists (B) (25850053) to K. Fuhshuku from Japan Society
15. John RA (1995) Pyridoxal phosphate-dependent enzymes. Bio-
chim Biophys Acta 1248:81–96
1 3

1510
J Ind Microbiol Biotechnol (2017) 44:1503–1510
16. Koga T, Fukuoka T, Doi N, Harasaki T, Inoue H, Hotoda H,
Kakuta M, Muramatsu Y, Yamamura N, Hoshi M (2004) Activ-
ity of capuramycin analogues against Mycobacterium tuberculosis
Mycobacterium avium and Mycobacterium intracellulare in vitro
and in vivo. J Antimicrob Chemother 54:755–760
20. Pellegata R, Pinza M, Piferi G (1978) An improved synthesis of
γ-, δ-, and ε-lactams. Synthesis 1978:614–616
21. Yamaguchi H, Sato S, Yoshida S, Takada K, Itoh M, Seto H,
Otake N (1986) Capuramycin, a new nucleoside antibiotic. Tax-
onomy, fermentation, isolation and characterization. J Antibiot
39:1047–1053
17. Okazaki S, Suzuki A, Mizushima T, Kawano T, Komeda H, Asano
Y, Yamane T (2009) The novel structure of a pyridoxal 5′-phos-
phate-dependent fold-type I racemase, α-amino-ε-caprolactam
racemase from Achromobacter obae. Biochemistry 48:941–950
18. Payoungkiattikun W, Okazaki S, Ina A, Aran H, Asano Y (2017)
Characterization of an α-amino-ε-caprolactam racemase with
broad substrate specifcity from Citreicella sp. SE45. J Ind Micro-
biol Biotechnol 44:677–685
22. Yamaguchi S, Komeda H, Asano Y (2007) New enzymatic method
of chiral amino acid synthesis by dynamic kinetic resolution of
amino acid amides: use of stereoselective amino acid amidases in
the presence of α-amino-ε-caprolactam racemase. Appl Environ
Microbiol 73:5370–5373
23. Yasukawa K, Asano Y (2012) Enzymatic synthesis of chiral phe-
nylalanine derivatives by a dynamic kinetic resolution of corre-
sponding amide and nitrile substrates with a multi-enzyme sys-
tem. Adv Synth Catal 354:3327–3332
19. Payoungkiattikun W, Okazaki S, Nakano S, Ina A, Aran H, Asano
Y (2015) In silico identifcation for α-amino-ε-caprolactam race-
mases by using information on the structure and function relation-
ship. Appl Biochem Biotechnol 176:1303–1314
1 3