Purification and Gene cloning of an enantioselective thioesterification enzyme from Brevibacterium Ketoglutamicum KU1073, a deracemization bacterium of 2-(4-Chlorophenoxy)propanoic acid

Article abstract of DOI:10.1271/bbb.100431

We succeeded in the purification and gene cloning of a new enzyme, α-methyl carboxylic acid deracemizing enzyme 1 (MCAD1) from Brevibacterium ketoglutamicum KU1073, which catalyzes the (S)-enantioselective thioesterification reaction of 2-(4-chlorophenoxy)propanoic acid (CPPA). The cloned gene of MCAD1 contained an ORF of 1,623 bp, encoding a polypeptide of 540 amino acids. In combination with cofactors ATP, coenzyme A (CoASH), and Mg ²±, MCAD1 demonstrated perfect enantioselectivity toward CPPA. The optimal pH and temperature for reaction were found to be 7.25 and 30 °C. Under these conditions, the K_m and k_cat values for (S)-CPPA were 0:92 ± 0:17mM and 0:28 ± 0:026 s^-1 respectively. The results for substrate specificity revealed that MCAD1 had highest activity toward fatty acid tails with a medium chain-length (C8). This result indicates that MCAD1 should be classified into a family of medium-chain acyl-CoA synthetase. This novel activity has never been reported for this family.

Full text of DOI:10.1271/bbb.100431

Biosci. Biotechnol. Biochem., 74 (12), 2405–2412, 2010
Purification and Gene Cloning of an Enantioselective Thioesterification Enzyme
from Brevibacterium ketoglutamicum KU1073, a Deracemization Bacterium
of 2-(4-Chlorophenoxy)propanoic Acid
y
2
Dai-ichiro KATO,1; Hiromitsu YOSHIDA,1 Masahiro TAKEO,1 Seiji NEGORO,1 and Hiromichi OHTA
1Graduate School of Engineering, University of Hyogo, 2167 Shosha, Himeji, Hyogo 671-2280, Japan
2Department of Biosciences and Informatics, Keio University, 3-14-1 Hiyoshi, Kohoku-ku, Yokohama 223-8522, Japan
Received June 8, 2010; Accepted August 27, 2010; Online Publication, December 7, 2010
[doi:10.1271/bbb.100431]
We succeeded in the purification and gene cloning of
a new enzyme, ꢀ-methyl carboxylic acid deracemizing
enzyme 1 (MCAD1) from Brevibacterium ketoglutamicum
KU1073, which catalyzes the (S)-enantioselective thio-
esterification reaction of 2-(4-chlorophenoxy)propanoic
acid (CPPA). The cloned gene of MCAD1 contained an
ORF of 1,623 bp, encoding a polypeptide of 540 amino
acids. In combination with cofactors ATP, coenzyme A
(CoASH), and Mg2þ, MCAD1 demonstrated perfect
enantioselectivity toward CPPA. The optimal pH and
temperature for reaction were found to be 7.25 and
30 ꢀC. Under these conditions, the Km and kcat values for
(S)-CPPA were 0:92 ꢁ 0:17 mM and 0:28 ꢁ 0:026 sꢂ1
respectively. The results for substrate specificity re-
vealed that MCAD1 had highest activity toward fatty
acid tails with a medium chain-length (C8). This result
indicates that MCAD1 should be classified into a family
of medium-chain acyl-CoA synthetase. This novel
activity has never been reported for this family.
In this case, the starting material and the product are
the same compound except for their chirality. Successful
application of this process yields a pure, optically active
compound that is approximately equal to the originally
synthesized racemate. Thus, deracemization is a novel
strategy for the preparation of optically active compounds.
There have been only a few reports on biocatalytic
systems, including the use of rat liver and fungi,
Cordyceps militaris and Verticillium lecanii, that cause
the deracemization reaction.7–10) These are capable of
inverting the chirality of the (R)-enantiomers of 2-
arylpropanoic acids to the corresponding (S)-antipode.
Through thorough investigations of rat-related enzymes
it became clear that three enzymes are involved in this
biotransformation system (Fig. 1).11–14) In these three
mechanistic reactions, only the long-chain acyl-CoA
synthetase-catalyzed reaction proceeds in an enantiose-
lective manner. Thus, the enantiomeric ratio of the acid
shifts to the (S)-form with the progress of the reaction.
Recently, we reported a novel biocatalytic deracem-
ization of CPPA and its derivatives using the whole-
cell system of Nocardia diaphanozonaria JCM3208.15)
Substrate specificity investigation and the detection of
metabolic intermediates suggested that this deracemiza-
tion process competitively inhibits the ꢀ-oxidation
pathway. This indicates that an acyl-CoA thioester is
involved in this process.16) Although we tried enzyme
purification on the assumption that an enantiose-
lective thioesterification enzyme is also involved in this
deracemization process, the trial was not successful due
to the instability of the enzymes in the cell-free extract.
In this report, we describe the isolation of several new
microorganisms that exhibited deracemization activity
of CPPA. In addition, we succeeded in the purification
and gene cloning of an enantioselective thioesterifica-
tion enzyme, ꢁ-methyl carboxylic acid deracemizing
enzyme 1 (MCAD1), from Brevibacterium ketoglutami-
cum KU1073.
Key words: Brevibacterium ketoglutamicum; enantiose-
lective thioesterification; acyl-CoA synthe-
tase; deracemization
The preparation of optically pure enantiomers is very
important, for the physiological activity of a compound
often depends on the absolute configuration of chiral
centers in the molecule. In the case of 2-aryloxypropa-
noic acids, only the (R)-enantiomers show herbicide
activity.1,2) Additionally, whereas (R)-2-(4-chlorophe-
noxy)propanoic acid (CPPA) lowers the level of serum
cholesterol and prevents platelet aggregation, the (S)-
antipode leads to muscle irritability and spasms by
inhibiting chloride channels.3) Hence a number of
approaches to prepare optically active 2-aryloxypropa-
noic acid have been made.4)
The usual biocatalytic method to obtain enantiomeri-
cally enriched compounds from racemates is kinetic
resolution, but the maximum yield of the desired
enantiomer is limited theoretically to 50%. To overcome
this disadvantage, the deracemization process has been
deeply investigated.5,6) Deracemization inverts the chir-
ality of one enantiomer in a racemic mixture to the other
antipode, resulting in one optically active compound.
Materials and Methods
Chemicals were purchased from standard suppliers and were used
without further purification unless otherwise noted. (R)- and (S)-CPPA
were prepared from a racemic mixture by a diastereomeric salt
y
To whom correspondence should be addressed. Tel/Fax: +81-79-267-4969; E-mail: kato@eng.u-hyogo.ac.jp
Abbreviations: CoASH, coenzyme A; CPPA, 2-(4-chlorophenoxy)propanoic acid; DTT, dithiothreitol; MCAD1, ꢁ-methyl carboxylic acid
deracemizing enzyme 1; PMSF, phenylmethylsulfonyl fluoride; PPB, potassium phosphate buffer

2406
D. KATO et al.
substrate.19) The unreacted carboxylic acid was prepared by the method
Long-chain acyl-CoA
synthetase
CH
CH
3
3
described in the section on MCAD1 activity assay by HPLC method.
The ee was confirmed by chiral phase HPLC using a Daicel Chiralcel
OJ column. Separation was achieved by employing an isocratic mobile
phase consisting of hexane/2-propanol/trifloroacetic acid (TFA) (95/
5/0.1) at a flow rate of 0.5 ml/min at 254 nm. Under these conditions,
the elution profile for CPPA was 25.9 and 31.9 min for the (R)- and
(S)-forms respectively.
CO H
COSCoA
2
X
X
Hydrolase
(R )-Acid
(R)-CoA thioester
Net Reaction
= Deracemization
2-Arylpropionyl-CoA
epimerase
The cell pellet was harvested from 2 liters of culture broth
(100 ml ꢁ 20), suspended in 100 ml Buffer A (100 mM MOPS-NaOH
buffer, pH 7.25 containing 1 mM DTT), and subjected to disruption by
French press. The cell debris was removed by centrifugation
(14;500 ꢁ g, 10 min, 4 ꢀC). To remove nucleic acid, 112 ml of 2%
protamine sulfate was added to the cell-free extract (112 ml). After
stirring for 15 min, the precipitate was removed by centrifugation
(14;500 ꢁ g, 20 min, 4 ꢀC). Solid ammonium sulfate was carefully
added to the supernatant with gentle stirring until the solution reached
40% saturation. After equilibration for 30 min, the mixture was
centrifuged, and the supernatant was separated. To this supernatant,
ammonium sulfate was again added to 70% saturation. The solution
was allowed to equilibrate for 30 min, and the resulting precipitate was
collected by centrifugation (14;500 ꢁ g, 20 min, 4 ꢀC). The precipitate
was dissolved in Buffer A and dialyzed for 4 h. The dialyzed fraction
CH
CH
3
3
CO H
COSCoA
2
X
X
Hydrolase
(S)-CoA thioester
(S )-Acid
Fig. 1. Proposed Reaction Mechanism for the Deracemization
Reaction of 2-Arylpropanoic Acid in Rat Liver.
formation method using optically active 1-(1-naphthyl)ethylamine as a
counter amine. Authentic samples of 2-aryloxypropanoyl-CoA were
chemically synthesized by a previously described method.17) 2-Methyl-
3-phenylpropanoic acid (MPPA) and 2-phenylthiopropanoic acid were
synthesized by a published procedure.15) A Bradford assay was done
to determine protein concentrations using bovine serum albumin as
standard.18)
was loaded onto
a pre-equilibrated Toyopearl HW-75F column
(250 ml) and eluted with Buffer A. The fractions were collected in
7 ml portions. To the collected active fractions, ammonium sulfate was
added to 20% saturation. The ammonium sulfate solution was applied
to a Toyopearl Butyl-650M column (100 ml), pre-equilibrated with
Buffer A containing a 20% saturation of ammonium sulfate. After
washing of the column, the enzyme was eluted using a linear gradient
from 20 to 0% saturation of ammonium sulfate. The fractions were
collected in 12 ml portions. The active fractions were combined,
concentrated to 22 ml with an amicon filter, and applied to a Toyopearl
DEAE-650M column (50 ml) pre-equilibrated with Buffer A. After
washing of the column, the enzyme was eluted using a linear gradient
from 0 to 300 mM NaCl. The fractions were collected in 12 ml portions.
The active fractions were combined, concentrated to 20 ml with an
amicon filter, and loaded onto a Macro-Prep Ceramic Hydroxyapatite
Type I column (10 ml), pre-equilibrated with a mixture of Buffer A and
100 mM PPB (pH 7.0) (volume ratio, 3:1). After washing of the
column, the enzyme was eluted using a linear gradient (volume ratio,
3:1 to 1:3) of a mixture of Buffer A and 100 mM PPB (pH 7.0). The
fractions were collected in 12 ml portions. The combined active
fractions were dialyzed for 4 h against Buffer A containing 200 mM
NaCl. The volume of the solution was concentrated to 22 ml with an
amicon filter. The dialyzed enzyme solution was loaded onto a DEAE-
Sepharose CL-6B column (50 ml), pre-equilibrated with Buffer A
containing 200 mM NaCl. After washing of the column, the enzyme
was eluted using a linear gradient from 200 to 300 mM NaCl. The
fractions were collected in 12 ml portions. The active fractions were
combined, and the solution was concentrated to 8 ml with an amicon
filter to obtain the purified enzyme.
Screening of microorganisms that deracemize CPPA. Microbial
strains were obtained from a glycerol stock solution of cells at Keio
University. The microorganism was cultured with shaking in 10 ml
of N. diaphanozonaria medium for 48 h at 30 ꢀC.16) The cells were
harvested by centrifugation (4;500 ꢁ g, 10 min, 4 ꢀC) and then washed
with 0.1 M potassium phosphate buffer (PPB) (pH 7.0). They were
resuspended in 5 ml of 0.1 M PPB containing (ꢂ)-CPPA (5 mg,
0.1% w/v) in a test tube. The tube was shaken at 30 ꢀC for 72 h. The
reaction mixture was then filtered through celite. The celite was washed
with ethyl acetate (EtOAc), and the washing was combined with the
filtrate. After acidification with 2 M HCl, the mixture was extracted with
EtOAc. The organic layer was washed with brine, dried over anhydrous
sodium sulfate, and concentrated in vacuo. The residual acid was
converted to the corresponding methyl ester with diazomethane, and
this was followed by a conventional work-up. The residue was purified
by pTLC (hexane/EtOAc = 9/1) to yield the methyl ester of the
starting acid as a colorless oil. The enantiomeric excess (ee) of the ester
was determined by high-performance liquid chromatography (HPLC)
with a Daicel Chiralcel OJ column (Daicel, Japan). The conditions and
elution profiles for HPLC analyses were described previously.16)
General procedure for the deracemization reaction of CPPA by cell-
free extract from isolated bacteria. The microorganism was cultured
with shaking in 10 ml of N. diaphanozonaria medium for 48 h at 30 ꢀC.
These pre-cultured cells were then added to 90 ml of fresh medium
and cultivated at 30 ꢀC for 48 h (second cultivation). In the case of
B. ketoglutamicum, the second incubation lasted for 24 h. The cells
were then harvested and suspended in 9.1 ml of 100 mM MOPS-NaOH
buffer (4 ꢀC, pH 7.25) containing 1 mM DTT. They were disrupted by
French press. The cell debris was removed by centrifugation
(14;500 ꢁ g, 10 min, 4 ꢀC). Three hundred ml of ethylene glycol (final
concentration, 3% v/v), 360 ml of cofactor solution (final concentration,
10 mM of ATP, 2 mM of CoASH, and 10 mM of MgCl2), and 240 ml of
substrate solution (final concentration, 1 mM of (ꢂ)-CPPA, pH 7.0)
were added (total volume, 10 ml) to this cell-free extract, and the
mixture was incubated for 24 h at 30 ꢀC. The reaction was quenched by
the addition of 1 ml of 2 M HCl, and was then extracted by diethyl ether
(Et2O). The ee and yield of the product were confirmed after
conversion to a methyl ester by treatment with diazomethane.
The N-terminal amino acid sequence of the purified enzyme was
performed by Edman degradation using a protein sequencer (Applied
Biosystems, Model 491, USA).
Cloning and sequencing of the MCAD1 gene. Total genomic DNA
was isolated from B. ketoglutamicum by the method reported by Saito
and Miura.20) To identify a partial, internal sequence of the MCAD1
gene, PCR was performed, using genomic DNA as a template with
forward (ATGTTGAGCACCATGCAGGACGCGC) and reverse
(GCTTCTGTTCGAACTTGCCGACGCTG) primers. The forward
primer was designed according to the revealed N-terminal amino acid
sequence. The reverse primer was designed based on the internal
sequence of a putative acyl-CoA synthetase from Nocardia farcinica
IFM10152 (GenBank accession no. BAD55557). The N-terminal
amino acid sequence of this protein showed high homology (84.6%) to
that of MCAD1, according to a FASTA search.
Purification of enantioselective thioesterification enzyme from
B. ketoglutamicum. All purification steps were carried out at 4 ꢀC
and were monitored by sodium dodecyl sulfate polyacrylamide gel
electrophoresis (SDS–PAGE). Enzyme active fractions were deter-
mined by the index of ee improvement of unreacted carboxylic acid
using a previously reported principle and (ꢂ)-CPPA as the starting
The amplified 1.5 kbp PCR product was purified, inserted into
pGEM-T easy vector (Promega, USA), and used for the transformation
of E. coli XL10-Gold ultracompetent cells (Stratagene, USA). The
positive clones were isolated and sequenced using an ABI Prism 3100
DNA sequencer (Applied Biosystems, Japan).

Enantioselective Thioesterification Enzyme of Brevibacterium
2407
For hybridization, 80 ng of genomic DNA was digested using Acc I,
Kpn I, Sac I, Hind III, Xba I, Pst I, EcoR I, Sal I, BamH I, Xho I,
Hinc II, and Sph I. This mixture was then transferred to a Hybond-N+
membrane using a G capillary blotter (TAITEC, Japan). Hybridization
was done at 60 ꢀC for 13 h using the AlkPhos Direct labeling and
detection System with CDP-Star (GE Healthcare, UK) using the
1.5 kbp PCR product described above as probe. The 0.9 and 1.0 kbp
fragments, digested with Hinc II and Sph I respectively, showed strong
signal intensities. These fragments were isolated and inserted into a
pUC19 Hinc II or Sph I/BAP vector and used in the transformation of
E. coli XL10-Gold ultracompetent cells. The positive clones were
isolated and sequenced to confirm the overall ORF sequences.
To measure the yield of thioester product, to the separated aqueous
layer was added 120 ml of 2 M NaOH and this was incubated at room
temperature for 30 min. The reaction mixture was then acidified by
adding 120 ml of 2 M HCl and extracted the hydrolyzed product by
700 ml of Et2O. The separated organic layer was concentrated in vacuo.
The yield, which was taken for the yield of thioester product, was
analyzed by HPLC as described above.
MCAD1 activity assay by the spectrophotometric method. The
activity of MCAD1 was also spectrophotometrically observed. This
assay measures the initial rate of AMP formation by coupling the
reaction of MCAD1 with adenylate kinase, pyruvate kinase, and lactate
dehydrogenase, following NADH oxidation at 340 nm with a spec-
trophotometer (U-2800A, Hitachi, Japan).21–23) The standard reaction
mixture for this assay contained 100 mM MOPS-NaOH (pH 7.25),
1 mM DTT, 10 mM ATP, 10 mM MgCl2, 1 mM (S)-CPPA, 2 mM
CoASH, 1 mM phosphoenolpyruvic acid, 0.2 mM NADH, 40 mg/mL
adenylate kinase, 20 mg/mL pyruvate kinase, 20 mg/mL lactate
dehydrogenase, and 120 mg of MCAD1. The total volume was brought
to 500 ml. The mixture, containing all components except the MCAD1
protein, was preincubated at room temperature (27 ꢂ 2 ꢀC) for 3 min.
The reaction was then initiated by addition of the enzyme.
Nucleotide sequence data analysis and alignment were conducted
using a BioEdit program. Sequence similarity to MCAD1 was
analyzed using the NCBI FASTA program.
Plasmid construction and expression of MCAD1. The MCAD1
gene was amplified from genomic DNA by PCR with forward
(GGAATTCCATATGTTGAGCACCATGCAGGACGCGC) and reverse
(TGCTCTAGAAGTACGTCGACATCGCTTTTCACACGG) primers.
These primers were designed to contain artificially introduced Nde I
and Xba I sites (italic form) in the 50 and 30 ends of the MCAD1 gene
respectively. Underlined ATG represents the initial codon. The
amplified 1.6 kbp fragment was subcloned to pGEM T-easy vector.
The subcloned vector was first digested with Nde I and Xba I, and then
purified and inserted into a pColdI expression vector (Takara, Japan).
The resulting plasmid, pColdI-MCAD1, was used to express the
recombinant MCAD1 protein. E. coli BL21(DE3) cells were trans-
formed with pColdI-MCAD1, cultivated, and treated to express
MCAD1 following the manufacturer’s instructions.
Kinetic analysis of recombinant MCAD1. The kinetic parameters
were calculated by a spectrophotometric method. The kcat and Km
values were evaluated by Michaelis–Menten analysis using GraphPad
prism, version 5.01 (GraphPad, USA). The kcat values were expressed
as turnover numbers per subunit (Mr of the subunit, 61,700). Kinetic
studies were performed with various concentrations of the substrate
(0.1–1.25mM). Each assay was repeated five times.
After recombinant gene expression, the cells were collected,
disrupted by sonication (20 kHz, 30 s ꢁ 10 times) in 5 ml of 50 mM
PPB (pH 7.0) containing 300 mM NaCl, and centrifuged (14;500 ꢁ g
for 10 min, 4 ꢀC). The enzyme was purified from the supernatant using
a TALON Metal Affinity Resin (2 ml) (Clontech, USA) following the
manufacturer’s instructions, and identified by SDS–PAGE. The active
fractions were combined and dialyzed overnight against Buffer A.
Purification and identification of 2-(4-chlorophenoxy)propanoyl-
CoA. 2-(4-Chlorophenoxy)propanoyl-CoA was purified using solid
phase extraction cartridges (Chromabond C 18 ec, Macherey-Nagel,
Germany),24) and were confirmed by HPLC and TOF-MS analysis.
HPLC analysis was performed using a Senshu Pak ODS-1251
(250 mm ꢁ 4:6 mm) column at room temperature, and was detected at
232 nm. Separation was achieved by employing an isocratic mobile
phase consisting of a 50 mM solution of KH2PO4 (pH 5.5)/methanol
(1/1) at a flow rate of 0.6 ml/min. Under these conditions, the retention
time of (S)-2-(4-chlorophenoxy)propanoyl-CoA was 22.6 min.
Mass spectra were obtained using a LCT-Premier (Waters, USA).
Data were acquired in negative ESI mode, and leucine enkephalin was
selected as a reference compound for high-accuracy, exact mass
measurement.
Molecular weight determination of recombinant MCAD1. The
molecular weight of the native enzyme was measured by gel-filtration
¨
chromatography. The enzyme was subjected to AKTA FPLC (Pharmacia
Biotech, USA) on a Superdex 75 10/300 GL at a flow rate of 0.5
ml/min, using 50 mM PPB (pH 7.2) containing 150 mM NaCl at room
temperature. Eluent absorbance was monitored at a wavelength of
280 nm. The relative mobility of the enzyme was compared to those
of the proteins from the Gel Filtration Standard kit (Bio-Rad, USA).
The molecular weight of the subunit of the enzyme was estimated by
SDS–PAGE on the basis of relative mobility as compared with those of
standard proteins from Perfect Protein Markers (Novagen, USA) and
Pre-stained SDS–PAGE standards (Bio-Rad, USA).
Results and Discussion
Screening of microorganisms capable of deracemizing
CPPA and the enzyme stability of cell-free extracts
Microorganisms deracemizing CPPA were screened
and obtained from stock cultures at Keio University. Of
approximately 300 cultured species, only Mycobacterium
smegmatis KU1047, B. ketoglutamicum, and Pseudomo-
nas aeruginosa KU1097 were found to show deracem-
ization activity. The yields and ee values of the recovered
substrate were compared and displayed (Fig. 2, Table 1).
The enantioselectivity of the microorganisms obtained
were found to be similar; the racemate of CPPA was
deracemized to the (R)-enantiomer.
MCAD1 activity assay by the HPLC method. The rate of thioester
formation was determined from the yields of unreacted carboxylic acid
and thioester product. Because the yield of the thioester in the aqueous
layer could not be determined as they were, the thioester bond was
hydrolyzed and the yield of the resulting free acid was measured. The
reaction mixture contained 100 mM MOPS-NaOH (pH 7.25), 1 mM
(ꢂ)-CPPA, 10 mM ATP, 2 mM CoASH, 10 mM MgCl2, 3% w/v
ethylene glycol, and an appropriate amount of MCAD1 protein. The
total volume was brought to 500 ml. The mixture, containing all
components except the MCAD1 protein, was preincubated at 30 ꢀC for
5 min. The reaction was initiated by the addition of the enzyme and
incubated at 30 ꢀC for 1 h. The reaction was quenched by adding 100 ml
of 2 M HCl and 700 ml of Et2O. After shaking, the reaction mixture was
centrifuged at 17;610 ꢁ g for 10 min, and two layers were separated.
To measure the yield of unreacted carboxylic acid, the separated
organic layer was concentrated in vacuo. The yield was determined by
HPLC using a COSMOSIL 5C18-ARII (150 mm ꢁ 4:6 mm) column at
room temperature and (ꢂ)-MPPA as internal standard. Separation was
achieved by employing an isocratic mobile phase consisting of H2O/
acetonitrile/TFA (2/1/0.05%) at a flow rate of 0.5 ml/min, detected at
254 nm. Under these conditions, the retention times of MPPA and
CPPA were 33.8 and 40.9 min respectively.
Deracemization activity was also observed in the cell-
free extract of B. ketoglutamicum when the necessary
cofactors for the acyl-CoA synthetase-catalyzed forma-
tion of acyl-CoA thioester (ATP, CoASH, and Mg2þ
)
were added to the reaction mixture. Notably, 50% of
maximum activity was still observed after 7 d, but no
activity was detected when any one of the above
cofactors was absent. In other microorganisms, however,
deracemization activity were not detected even imme-
diately after cell disruption.

2408
D. KATO et al.
Cl
Cl
Deracemizing
CH
CH
3
3
microorganisms
O
CO H
O
CO H
2
2
(
)- form
(R )-form
Fig. 2. Deracemization Reaction of CPPA by Selected Microorgan-
isms: N. diaphanozonaria JCM3208, M. smegmatis KU1047,
B. ketoglutamicum KU1073, and P. aeruginosa KU1097.
Table 1. Deracemization Reaction of CPPA by Screened Micro-
organisms
Whole cellsa
Cell-free extracta
yield (%) ee (%) yield (%) ee (%)
N. diaphanozonaria JCM3208
M. smegmatis KU1047
95
98
90
85
97 (R)
47 (R)
72 (R)
36 (R)
95
95
98
97
0 (rac.)
0 (rac.)
24 (R)
Fig. 3. SDS–PAGE Analysis of an Enantioselective Thioesterifica-
tion Enzyme, MCAD1, from B. ketoglutamicum KU1073.
Proteins were separated on a 10% polyacrylamide gel in the
presence of 0.01% SDS. Lane 1, molecular weight standards; lane 2,
purified MCAD1.
B. ketoglutamicum KU1073
P. aeruginosa KU1097
0 (rac.)
aReaction conditions are described under ‘‘Materials and Methods.’’
Table 2. Purification of MCAD1 from B. ketoglutamicum KU1073
Protein
(mg)
Total activitya
(nmol/min)
Yield
(%)
Specific activitya
(nmol/min/mg)
Purification
(fold)
Purification step
Cell-free extract
Ammonium sulfate precipitation
HW-75F
1660
839
605
24
240
317
250
25.3
17.3
9.4
100
132
104
11
0.145
0.338
0.414
1.06
1.0
2.3
2.9
7.3
Butyl Toyopearl
DEAE Toyopearl
Hydroxyapatite
2.9
7.2
3.9
0.7
6.06
41.8
111
109
0.58
0.11
16.1
15.8
DEAE Sepharose
1.77
aThe activity of thioester formation was determined by the method of HPLC assay described under ‘‘Materials and Methods.’’
Purification of enantioselective thioesterification
enzyme (MCAD1) from B. ketoglutamicum
hydrophobic, ion-exchange, and hydroxyapatite column
chromatography. The optimal pH and temperature for the
reaction were found to be 7.25 and 30 ꢀC. The specific
activity of the purified enzyme was analyzed by HPLC,
and was found to be 15.8 nmol/min/mg. A solution
containing 1 mM (ꢂ)-CPPA, cofactors (10 mM ATP,
2 mM CoASH, 10 mM MgCl2), and 3% w/v ethylene
glycol in Buffer A at 30 ꢀC (standard condition) was
used. Under these conditions, the enzyme catalyzed (S)-
enantioselective thioester formation. The purified enzyme
required cofactors (ATP, CoASH, Mg2þ) since no activity
was detected in the absence of any one of the cofactors.
The final product afforded a single band by SDS–
PAGE, and the molecular weight was determined to
be approximately 60,000 Da (Fig. 3). The first 13 N-
terminal amino acid residues of the purified enzyme
were sequenced to be MLSTMQDAPLSIA, and these
were then used in the cloning of the gene.
In the case of rat liver studies, long-chain acyl-CoA
synthetase catalyzed enantioselective thioester forma-
tion is the first step in the deracemization process.14) We
have proposed that a similar enzyme should work in
B. ketoglutamicum, because the deracemization activity
in the cell-free extract was detected only in the presence
of cofactors for the formation of acyl-CoA synthetase
(ATP, CoASH, and Mg2þ). Hence the enantioselective
thioesterification enzyme for CPPA was selected as the
target protein for purification. This protein was tenta-
tively named ꢁ-methyl carboxylic acid deracemizing
enzyme 1 (MCAD1), to the effect that it is an enzyme
relating to the deracemization reaction of ꢁ-methyl
carboxylic acid.
For enzyme purification, a simple procedure to assay
the index of ee improvement of unreacted acid was used.
This reported method was first employed to determine
whether firefly luciferase has a capacity for enantiose-
lective thioester formation toward 2-arylpropanoic
acid.19) The racemate is used as a starting material,
and if the enzyme has the ability to distinguish between
the absolute configuration of R and S, the ee of the
unreacted carboxylic acid increases. Based on this
principle, collected fractions exhibiting activity can be
selected easily during enzyme purification.
Cloning and sequencing of MCAD1
The complete ORF of MCAD1 contained a DNA
sequence of 1,623 bp, encoding 540 amino acids. The
ORF initiated and terminated with an ATG and a TGA
codon respectively. The nucleotide sequence of the
MCAD1 gene was submitted to the DDBJ database
(accession no. AB546128). After comparison, a putative
AMP-binding domain signature (PROSITE accession
no. PS00455) was found in the sequences at positions
182 to 193 (Fig. 4).
As summarized in Table 2, MCAD1 from the cell-free
extract of B. ketoglutamicum was purified 109-fold to
homogeneity with an overall yield of 0.7%. Purification
included ammonium sulfate precipitation, gel-filtration,
The entire amino acid sequence of MCAD1 was used
to search for protein sequence homology in the FASTA

Enantioselective Thioesterification Enzyme of Brevibacterium
2409
Fig. 4. Amino Acid Sequence Alignments of MCAD1 and Predicted Acyl-CoA Synthetases, EEN87350 (predicted medium-chain fatty acid-CoA
ligase from Rhodococcus erythropolis SK121), BAH36329 (predicted fatty acid-CoA ligase (fadD) from Rhodococcus erythropolis PR4),
ACY19667 (predicted AMP-dependent synthetase and ligase protein from Gordonia bronchialis DSM 43247), ACU98289 (predicted acyl-CoA
synthetase from Saccharomonospora viridis P101), EEP72933 (predicted Acyl-CoA synthetase from Micromonospora sp. ATCC 39149), and
BAD55557 (putative acyl-CoA synthetase protein from Nocardia farcinica IFM 10152).
The conserved residues among all the sequences are indicated by asterisks. The underlined regions represent the putative AMP-binding
domain signature in MCAD1 (PROSITE accession no. PS00455).

2410
D. KATO et al.
CH
CH
3
program and the Genbank database. The results of
amino acid sequence alignment are shown in Fig. 4. The
MCAD1 protein sequence showed 64–99% identity to
the family of acyl-CoA synthetases, which have not yet
been analyzed in detail. In contrast, MCAD1 showed
low homology to rat long-chain acyl-CoA synthetase
(5% identity), which is involved in the deracemization
process of 2-arylpropanoic acid.14) These results imply
that MCAD1 and its homology proteins constitute a new
family of acyl-CoA synthetases.
CH
3
MCAD1
3
+
Ar
Ar
Ar
2+
O
(
CO H
O
CO H
2
O
COSCoA
ATP, Mg
CoASH
2
(S)-Thioester
)-Acid
(R)-Acid
Ar = p-Cl-phenyl
Fig. 5. MCAD1-Catalyzed Enantioselective Thioesterification of
CPPA.
MCAD1 exhibited perfect enantioselectivity, and only the (S)-form
was transformed to the thioester product in the presence of ATP,
Mg2þ, and CoASH.
Expression and enzymatic properties of recombinant
MCAD1
14
12
10
8
100
80
60
40
20
0
A
MCAD1 was cloned in cold-inducible vector and
expressed in E. coli with an N-terminal, fused His-tag
(MNHKVHHHHHHIGGRH). As determined by SDS–
PAGE, the molecular weight of the recombinant enzyme
was calculated to be 61,700 Da, but the weight of
the native enzyme was found to be 121,578 Da by
high performance gel-filtration column chromatography.
These results indicate that the native enzyme was an
active dimer. In the case of the His-tag removed protein
at the factor Xa site, the same result was obtained.
Hence we believe that the native enzyme purified from
bacterial cells and the cloned recombinant protein have
the same character, and we examined using the His-tag
fused recombinant proteins here.
The specific activity of purified recombinant MCAD1
toward 1 mM (ꢂ)-CPPA was assayed by HPLC, and was
calculated to be 13.4 nmol/min/mg. Cofactors (ATP,
CoASH, Mg2þ) were also required in this process.
Additionally, the enzyme showed no activity toward
0.5 mM (R)-CPPA within the experimental error (<0:01
nmol/min/mg), while the specific activity toward 0.5
mM (S)-CPPA was found to be 12.9 nmol/min/mg.
These findings indicate that MCAD1 shows perfect
enantioselectivity in the thioesterification of CPPA, and
the (R)-form does not affect the catalytic activity of
MCAD1-catalyzed thioester formation (Fig. 5).
The thioesters were purified by solid-phase extrac-
tion,24) and were identified by ESI-MS based on
orthogonal TOF-MS. The mass spectra of (S)-2-(4-
chlorophenoxy)propanoyl-CoA were dominated by ions
that were in close agreement with the calculated mass
of singly charged thioester (½M ꢃ Hꢄꢃ as observed
948.1208, and predicted 948.1208).25) HPLC analysis
was also carried out, and the retention time of the purified
product agreed with the chemically synthesized one.
The effect of pH on enzyme activity was examined
(Fig. 6A). The enzyme exhibited maximum activity at
pH 7.25 in a MOPS-NaOH buffer. The optimal reaction
temperature was found to be approximately 30 ꢀC
(Fig. 6B). These results are in good agreement with
the wild-type enzyme purified from the bacterial cells.
Various compounds were added to the reaction
system, and enzymatic activity was compared (Table 3).
The enzyme was inhibited completely by solutions
containing 1 mM Hg2þ and 10 mM Ni2þ, Fe2þ, Co2þ, and
Cu2þ. p-Chloromercuribenzoic acid (PCMB) and Zn2þ
also strongly affected activity. Activity was slightly
decreased by the addition of N-ethylmaleimide, 5,50-
dithiobis(2-nitrobenzoic acid), and certain metal ions,
including Agþ, Ca2þ, and Mn2þ. PMSF, NaN3, and
iodoacetic acid had no effect on enzyme activity.
6
4
2
0
10
6
7
8
9
pH
14
12
10
8
100
80
60
40
20
0
B
6
4
2
0
50
0
10
20
30
40
Temperature (°C)
Fig. 6. Effects of pH (A) and Temperature (B) on MCAD1-Catalyzed
Thioester Formation of (ꢂ)-CPPA.
Relative activity was calculated by comparing with the specific
activity, using the value obtained in 100 mM MOPS-NaOH
(pH 7.25) at 30 ꢀC as standard. The specific activity of thioester
formation was determined by HPLC analysis taking the final
concentration of (ꢂ)-CPPA to be 1 mM. In experiments on the pH
activity profile, the following buffers (100 mM) were used: pH 6.0–
7.0 MOPS-NaOH; 7.0–8.0 PPB; 8.0–9.0 Tris–HCl. In the case of the
temperature-effect experiments, 100 mM MOPS-NaOH (pH 7.25)
was used.
Substrate specificity of recombinant MCAD1 was
analyzed by a spectrophotometric method (Table 4).
Likewise, MCAD1 showed perfect enantioselectivity
toward CPPA. Fatty acids of various chain lengths were
also subjected to this assay. As shown in Table 4,
MCAD1 exhibited thioesterification activity toward
fatty acids with 4 to 18 carbons. Among these
compounds, the highest activity was detected in n-
octanoic acid (C8), a medium chain-length fatty acid.
These results indicate that MCAD1 is to be classified
into a family of medium-chain acyl-CoA synthetases.
This is the first report of enantioselective thioesterifica-
tion activity of a medium-chain acyl-CoA synthetase
toward 2-aryloxypropanoic acid. The enzyme can also
accept 6-mercapthexanoic acid, 6-hydroxyhexanoic
acid, and 11-mercaptundecanoic acid, which have a
thiol or a hydroxyl group at the !-position. In the case of

Enantioselective Thioesterification Enzyme of Brevibacterium
2411
Table 3. Effects of Various Reagents on the MCAD1-Catalyzed
Thioesterification Reaction of (ꢂ)-CPPA
Table 4. Substrate Specificity of MCAD1-Catalyzed Thioesterifica-
tion Reaction
Conc.
(mM)
Relative activityb
(%)
Relative activityb
Reagent
Carboxylic acida
(%)
Controla
PMSF
NaN3
—
10
10
1
100
97
(S)-CPPA
(R)-CPPA
Acetic acid (C2)
7
N.D.
N.D.
0.4
34
94
92
Iodoacetic acid
N-Ethylmaleimide
5,50-Dithiobis(2-nitrobenzoic acid)
n-Propanoic acid (C3)
n-Butanoic acid (C4)
n-Hexanoic acid (C6)
n-Octanoic acid (C8)
n-Decanoic acid (C10)
n-Dodecanoic acid (C12)
n-Butyldecanoic acid (C14)c
n-Hexadecanoic acid (C16)c
n-Octadecanoic acid (C18)c
6-Mercaptohexanoic acid
6-Hydroxyhexanoic acid
Ahx
1
69
47
1
46
100
PCMB
AgNO3
ZnCl2
HgCl2
NiSO4
FeSO4
CoCl2
CuCl2
CaCl2
MnCl2
1
27
53
1
67
43
30
26
15
89
71
1
26
1
N.D.
N.D.
N.D.
N.D.
N.D.
80
10
10
10
10
10
10
0.5
75
11-Mercaptoundecanoic acid
Boc-Ahx
Cbz-Ahx
59
6
N.D., not detected; Relative activity could not be calculated because the
specific activity was under the limit of detection.
58
82
36
4
ꢁ-Lipoic acid
aControl contained enzyme solution without any reagents.
bThe activity under the control condition (13.8 nmol/min/mg) was taken to
be 100%. The specific activity of thioester formation was determined by the
method of HPLC assay with the final concentration of (ꢂ)-CPPA taken to be
1 mM.
Hydrocinnamic acid
(ꢂ)-2-Phenoxypropanoic acid
(ꢂ)-2-Phenylthiopropanoic acid
(ꢂ)-MPPA
4
38
(S)-2-Phenylpropanoic acid
(R)-2-Phenylpropanoic acid
Benzoic acid
0.6
0.3
0.3
6-aminohexanoic acid (Ahx), however, thioesterification
activity was decreased dramatically. Interestingly the
catalytic activity was recovered when the !-amine was
protected by Boc or Cbz protecting group. This result
indicates that the substitution of !-position is important
to determine the substrate accessibility of this enzyme.
Additionally, hydrocinnamic acid and ꢁ-lipoic acid were
also accepted by this enzyme with high efficiency. In
addition, (ꢂ)-2-phenoxypropanoic acid, (ꢂ)-2-phenyl-
thiopropanoic acid, and (ꢂ)-MPPA were also converted
to thioester, which are accepted by the whole-cell
deracemization system of N. diaphanozonaria JCM
3208.16) In the cases of 2-phenoxypropanoic acid and
2-phenylthiopropanoic acid, the results of chiral phase
HPLC analysis revealed that its enantioselectivity was
the same as that of CPPA. In the case of MPPA, both
enantiomers were converted to thioesters speedily with
little precedence over (R)-enantioselectivity. These
results indicate that the substrate specificity of MCAD1
might be similar to that of the thioesterification enzyme
of N. diaphanozonaria, though its enzyme was not yet
been isolated. MCAD1 also catalyzed the thioesterifica-
tion reaction of 2-phenylpropanoic acid (toward R),
though the specific activity was one-tenth of that of (S)-
CPPA and the enantiodifferentiational activity toward
2-phenylpropanoic acid was not sufficient. These results
differ from those reported for rat long-chain acyl-CoA
synthetase.
Anthraquinone-2-carboxylic acid
2-Anthracenecarboxylic acid
1-Pyrenebutyric acid
Ferrocenecarboxylic acid
Retinoic acid
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
N.D.
trans-Cinnamic acid
Caffeic acid
Malonic acid
Phenylalanine
N.D., not detected; Relative activity could not be calculated because the
specific activity was under the limit of detection.
aThe final concentration of carboxylic acid was 1 mM.
bRelative activity was calculated by comparison with specific activity,
taking the value toward n-octanoic acid (C8) (248 nmol/min/mg) as
standard. The specific activity of thioester formation was determined by
spectrophotometric assay.
cTo dissolve the substrate, activity was measured when 0.2% Triton X-100
was added to the reaction mixture.
Table 5. Kinetic Parameters of MCAD1 Catalyzed Thioesterification
Reaction
Km
kcat
(sꢃ1
kcat=Km
(sꢃ1ꢅmM
Substrate
ꢃ1
(mM)
)
)
(S)-CPPA
ꢁ-Lipoic acid
0:92 ꢂ 0:17 0:28 ꢂ 0:026
0:30 ꢂ 0:027 2:1 ꢂ 0:080
0.28
7.0
n-Octanoic acid (C8) 0:077 ꢂ 0:009 6:6 ꢂ 0:27
86.0
The specific activity of thioester formation was determined by spectropho-
tometric assay.
Anthraquinone-2-carboxylic acid, 2-anthracenecar-
boxylic acid, 1-pyrenebutyric acid, ferrocenecarboxylic
acid, retinoic acid, trans-cinnamic acid, caffeic acid,
malonic acid, and phenylalanine were ineffective sub-
strates.
Kinetic studies of (S)-CPPA, n-octanoic acid, and
ꢁ-lipoic acid were also conducted. The efficiency of
thioester formation was calculated from of the data
obtained by spectrophotometric analysis. The Km and
kcat values were determined by Michaelis-Menten fitting
(Table 5). These results also confirmed that MCAD1 is
a suitable enzyme to catalyze the thioester formation
of fatty acids with a medium chain-length.
In this report, we describe the isolation of several new
microorganisms that exhibited deracemization activity
of CPPA. Additionally, we purified and cloned a new
enzyme, MCAD1, from B. ketoglutamicum. This protein
was found to catalyze the (S)-enantioselective thioester-
ification reaction of CPPA in a practically perfect

2412
D. KATO et al.
8) Hall SD and Xiaotao Q, Chem. Biol. Interact., 90, 235–251
(1994).
manner. Investigation of substrate specificity revealed
that this enzyme shows high enzymatic activity toward
fatty acids possessing a medium chain-length. MCAD1
shares very low homology with long-chain acyl-CoA
synthetase, which has been found to catalyze the
enantioselective thioester formation of 2-arylpropanoic
acid in rats. Specific activity toward 2-arylpropanoic
acid was also relatively low. Thus, MCAD1 is to be
classified in a family of medium-chain acyl-CoA
synthetases. This activity has not previously been
reported.
9) Rhys-Williams W, Thomason MJ, Hung Y-F, Hanolon GW, and
Lloyd AW, Chirality, 10, 528–534 (1998).
10) Rhys-Williams W, Thomason MJ, Lloyd AW, and Hanlon GW,
Pharm. Sci., 2, 537–540 (1996).
11) Knadler MP and Hall SD, Chirality, 2, 67–73 (1990).
´
12) Brugger R, Alıa BG, Reichel C, Waibel R, Menzel S, Brune K,
and Geisslinger G, Biochem. Pharmacol., 52, 1007–1013
(1996).
13) Sevoz C, Benoit E, and Buronfosse T, Drug Metab. Dispos., 28,
398–402 (2000).
14) Brugger R, Reichel C, Alia BG, Brune K, Yamamoto T,
Tegeder I, and Geissinger G, Biochem. Pharmacol., 61, 651–
656 (2001).
Acknowledgment
15) Kato D, Mitsuda S, and Ohta H, J. Org. Chem., 68, 7234–7242
(2003).
This work was partially supported by a Grant-in-Aid
for Young Scientists (B) (19750144) from the Japan
Society for the Promotion of Science (JSPS).
16) Kato D, Miyamoto K, and Ohta H, Tetrahedron: Asymmetry, 15,
2965–2973 (2004).
17) Kawaguchi A, Yoshimura T, and Okuda S, J. Biochem., 89,
337–339 (1981).
References
18) Bradford MM, Anal. Biochem., 78, 248–254 (1976).
19) Kato D, Teruya K, Yoshida H, Takeo M, Negoro S, and Ohta H,
FEBS J., 274, 3877–3885 (2007).
1) Feller DR, Kamanna VS, Newman HAI, Romstedt KJ, Witiak
DT, Bettoni G, Bryant SH, Conte-Camerino D, Loiodice F, and
Tortorella V, J. Med. Chem., 30, 1265–1267 (1987).
2) Bettoni G, Loiodice F, Tortorella V, Conte-Camerino D,
Mambrini M, Ferrannini E, and Bryant SH, J. Med. Chem.,
30, 1267–1270 (1987).
20) Saito H and Miura K, Biochim, Biophys. Acta, 72, 619–629
(1963).
21) Horecker BL and Kornberg A, J. Biol. Chem., 175, 385–390
(1948).
22) Wu M-X and Hill KAW, Anal. Biochem., 211, 320–323 (1993).
23) Hisanaga Y, Ago H, Nakagawa N, Hamada K, Ida K,
Yamamoto M, Hori T, Arii Y, Sugahara M, Kuramitsu S,
Yokoyama S, and Miyano M, J. Biol. Chem., 279, 31717–31726
(2004).
3) Colton IJ, Ahmed SN, and Kazlauskas RJ, J. Org. Chem., 60,
212–217 (1995).
4) Bornscheuer UT and Kazlauskas RJ, ‘‘Hydrolases in Organic
Synthesis’’ 2nd ed., Wiley-VCH (2006).
5) Turner NJ, Curr. Opin. Chem. Biol., 8, 114–119 (2004).
6) Faber K and Kroutil W, Curr. Opin. Chem. Biol., 9, 181–187
(2005).
24) Beuerle T and Pichersky E, Anal. Biochem., 302, 305–312
(2002).
25) Yonekubo J, Inoue T, and Sasaki H, Chromatography, 27, 85–
89 (2006).
7) Knihinicki RD, Williams KM, and Day RO, Biochem. Pharma-
col., 38, 4389–4395 (1989).