Increasing the storage and oxidation stabilities of N-acyl-d-amino acid amidohydrolase by site-directed mutagenesis of critical methionine residues

Article abstract of DOI:10.1016/j.procbio.2012.06.003

The recombinant N-acyl-d-amino acid amidohydrolase (N-d-AAase) of Variovorax paradoxus Iso1 was unstable during protein purification and storage at 4 °C. Since the methionine oxidation might be the artificial factor leading to the inactivation of N-d-AAase, eight potential oxidation sensitive methionine residues of the enzyme were individually substituted with leucine utilizing site-directed mutagenesis. Among them, five mutants, M39L, M56L, M221L, M254L, and M352L remained at least 70% of wild-type specific activity. The enzyme kinetic parameters of M221L revealed a 44% decrease in K_m, and finally reflected a 2.4-fold increase in k_cat/K_m. Moreover, its half-life at 4 °C increased up to 6-fold longer than that of the wild-type. Structural analysis of each methionine substitution was carried out based on the crystal structure of N-d-AAase from Alcaligenes faecalis DA1. Met²²¹ spatial closeness to the zinc-assistant catalytic center is highly potential as the primary site for oxidative inactivation. We conclude that the replacement of methionine M221 with leucine in N-d-AAase successfully enhances the oxidative resistance, half-life, and enzyme activity. This finding provides a promising basis for the engineering the stability and activity of N-d-AAase.

Full text of DOI:10.1016/j.procbio.2012.06.003

Process Biochemistry 47 (2012) 1785–1790
Contents lists available at SciVerse ScienceDirect
Process Biochemistry
journal homepage: www.elsevier.com/locate/procbio
Increasing the storage and oxidation stabilities of N-acyl-d-amino acid
amidohydrolase by site-directed mutagenesis of critical methionine residues
I-Chen Peng, Kai-Yin Lo, Chun-Hua Hsu, Chia-Yin Lee∗
Department of Agricultural Chemistry, National Taiwan University, Taipei 10617, Taiwan
a r t i c l e i n f o
a b s t r a c t
Article history:
The recombinant N-acyl-d-amino acid amidohydrolase (N-d-AAase) of Variovorax paradoxus Iso1 was
unstable during protein purification and storage at 4 ◦C. Since the methionine oxidation might be the
artificial factor leading to the inactivation of N-d-AAase, eight potential oxidation sensitive methionine
residues of the enzyme were individually substituted with leucine utilizing site-directed mutagenesis.
Among them, five mutants, M39L, M56L, M221L, M254L, and M352L remained at least 70% of wild-type
specific activity. The enzyme kinetic parameters of M221L revealed a 44% decrease in Km, and finally
reflected a 2.4-fold increase in kcat/Km. Moreover, its half-life at 4 ◦C increased up to 6-fold longer than
that of the wild-type. Structural analysis of each methionine substitution was carried out based on the
crystal structure of N-d-AAase from Alcaligenes faecalis DA1. Met221 spatial closeness to the zinc-assistant
catalytic center is highly potential as the primary site for oxidative inactivation. We conclude that the
replacement of methionine M221 with leucine in N-d-AAase successfully enhances the oxidative resis-
tance, half-life, and enzyme activity. This finding provides a promising basis for the engineering the
stability and activity of N-d-AAase.
Received 8 February 2012
Received in revised form 8 May 2012
Accepted 4 June 2012
Available online 22 June 2012
Keywords:
N-acyl-d-amino acid amidohydrolase
N-d-AAase
Variovorax paradoxus Iso1
Site-directed mutagenesis
Methionine oxidation
Enzyme stability
© 2012 Elsevier Ltd. All rights reserved.
1. Introduction
faecalis DA1 comprises of a small ␤-barrel and a catalytic (␤␣)8-
barrel with a 63-residue insertion. The A. faecalis DA1 N-d-AAase
N-acyl-d-amino acid amidohydrolase (EC 3.5.1.81, N-d-AAase) is
useful for optical resolution of dl-amino acid and highly promising
for commercial application to d-amino acid production, due to its
ability to enantioselectively catalyzing the zinc-assisted deacety-
lation of N-acyl-d-amino acids and producing the corresponding
d-amino acids. d-amino acids are important intermediates for
chiral chemical synthesis, such as pesticides, sweeteners, bioac-
tive peptide [1,2] and semi-synthetic antibiotics (review in [3,4]).
Although little is known about the physiological affects of N-d-
AAase on microorganisms, it has been intensively studied for its
promising application to resolving dl-amino acids, such as con-
tinuously producing d-amino acid coupling with N-acyl amino
acid racemase [5–7]. Many N-d-AAases have been purified and
characterized from a number of bacteria, such as Alcaligenes
[8–11], Pseudomonas [12], Stenotrophomonas [13], Streptomyces
[14], Microbacterium [15], and several have been cloned and
expressed in the Escherichia coli system [7,16–19]. Crystal struc-
ture of the 483-residue N-d-AAase from Alcaligenes faecalis DA1 is
reported [20,21]. The protein structure of the N-d-AAase from A.
binds two zinc ions on ␣ and ␤ sites with widely different affinities.
A catalytically essential Zn2+, tightly bound on the ␤-site coordi-
nated by Cys96, His220 and His250 activated the enzyme [20,21].
The surface-exposed methionine residues of proteins are sus-
ceptible to oxidative stress, and can be oxidized into methionine
sulfoxide by reactive oxygen species [22]. Methionine sulfoxide
reductase, capable of reducing the methionine sulfoxide residue
of protein back to methionine, provides an antioxidant mech-
anism for scavenging reactive species through the oxidation of
NAD(P)H in the cell [23]. If the oxidized methionine were not
reduced, that will lead to change the hydrophobicity of residue
[24], alter the protein conformation [25], and lose the biological
activity [22]. According to Bordo and Argos’s statistical results,
exchanging methionine with leucine will minimize disturbance on
the protein structure [26]. Replacement of methionine with non-
oxidizable leucine in several industrially useful enzymes, such as
subtilisin [27] and N-carbamoyl d-amino acid amidohydrolase [28],
successfully enhances the oxidative resistance.
Variovorax paradoxus is a microorganism with diverse metabolic
capabilities and has mutual beneficial interaction with plants.
Its whole genome was decoded [29]. Our previous report indi-
cated that the N-d-AAase of V. paradoxus Iso1 purified from
E. coliTop10 (pTrc2A-damA3) expressing N-d-AAase gene showed
high stereospecificity to substrates N-acyl-d-amino acids [17]. This
V. paradoxus N-d-AAase enzyme is suitable for application to d-
∗
Corresponding author at: Department of Agricultural Chemistry, National
Taiwan University, 1 Sec. 4, Roosevelt Road, Taipei 10617, Taiwan. Tel.: +886 2 2363
2393; fax: +886 2 2366 0581.
E-mail address: clee@ntu.edu.tw (C.-Y. Lee).
1359-5113/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.procbio.2012.06.003

1786
I.-C. Peng et al. / Process Biochemistry 47 (2012) 1785–1790
of 95 ◦C for 30 s and then 12 cycles of 95 ◦C for 30 s, 55 ◦C 1 min, 68 ◦C for 12 min and
Table 1
a final incubation at 4 ◦C were performed on a GeneAmp PCR system 9700 (Applied
Biosystems). Thermocycled reaction mixture was immediately digested by adding
10 U of reaction enzyme DpnI, incubated on 37 ◦C for 1 h, and transformed into
E. coli XL-1 blue. The DNA sequence of transformant was determined by the mod-
ified dideoxy chain termination method [30] with a 3100 DNA analyzer (Applied
Biosystems).
Oligonucleotide primers used in this study.
Primer
Nucleotide sequencea
N-d-AAase gene specific
Primer E
Primer H
Site-directed mutagenesis
M39L
M56L
M79L
M171L
M221L
M254L
M273L
5ꢀ-CGGCGAATTCACCAGTAAGGAGGAGAATC-3ꢀ
5ꢀ-GCGCAAGCTTCGGCTACGGGC-3ꢀ
5ꢀ-GTATCGCTGCGCTGGGCGATCTGCG-3ꢀ
5ꢀ-GTGGACGCGACGGGCCTGATCGTGG-3ꢀ
5ꢀ-CCCAGGCGGAGCTGAGTTACAAGGTTTCG-3ꢀ
5ꢀ-GAGGTGGCCCAGCTGCGCCTGCTGC-3ꢀ
5ꢀ-ATGTGACCCACCTGCGGCACGAGGACC-3ꢀ
5ꢀ-CACAAGGTCCTGCACAGCGCCAACTTCG-3ꢀ
5ꢀ-CATCCTCGAGACCCTGAAGAAGCAGCCG-3ꢀ
5ꢀ-CGACGAGACCGAACTGCAGAAGATCATGGC-3ꢀ
2.4. Expression and purification of recombinant N-d-AAase
A single colony of E. coli JM109 harboring plasmid pKK-dam3 was pre-cultured
in 3 ml 2xYT broth supplemented with 200 ␮g/ml ampicillin at 25 ◦C for 18 h. The
overnight culture, 10 ml, was seeded into a 500 ml flask containing 200 ml of the
same medium and incubated at 10 ◦C with 150 rpm shaking. At OD600 0.4–0.6, a
final concentration 0.1 mM of IPTG was added and additional 80 h of incubation time
was achieved to induce the expression of target protein. The cells were collected and
resuspended in buffer A (50 mM Tris–HCl, 10% glycerol, 0.01% 2-mercaptoehtanol,
pH 7.5) and disrupted by sonication. After heated at 40 ◦C for 15 min, cell debris
and denatured proteins were removed by centrifugation and filtration with a 0.45-
␮m filter. The clarified crude lysate was then applied to an AKTA Prime system
(Amersham Pharmacia Biotech) equipped with a HiPrep Q Sepharose Fast Flow ion
exchange column pre-equilibrated with buffer A. The adsorbed protein was eluted
with a buffered 0–0.4 M NaCl linear gradient. The active fractions were collected,
brought to 20% ammonium sulfate saturation, filtrated again, and applied to a HiPrep
Butyl Sepharose Fast Flow column pre-equilibrated with buffer B (buffer A contain-
ing 20% ammonium sulfate). The absorbed enzyme was eluted with a buffered 20–0%
ammonium sulfate linear gradient. All fractions were monitored with the enzyme
activity assay. The purified enzymes were concentrated with Amicon Ultra (30 kDa
MWCO) (Millipore) to approximately 1 mg/ml prior to the following experiments.
The purified enzymes were analyzed by SDS-PAGE (10% polyacrylamide) [31] and
Western blotting.
M352L
a
Restriction enzyme recognition sites (EcoRI and HindIII) are underlined; the
mutation site (Leu code) is in bold type and underlined; the ribosome binding site
is in bold type and italicized.
amino acid production [17]. However its unstable characteristics
during protein purification and storage at 4 ◦C will limits the prac-
tical applications. The peptide sequencing results of LC/MS/MS
analysis provided preliminary evidence that some methionine
residues of the recombinant d-aminoacylase were oxidized into
methionine sulfoxide revealed by the increase 16 of molecular mass
[17]. In this study, site-directed mutagenesis, methionine replaced
by leucine, was employed to investigate whether methionine oxi-
dation was the critical factor that leads to enzyme inactivation
during storage at 4 ◦C. Furthermore, structural analysis was carried
out to interpret the influence of each mutation on the oxidation
stability of the enzyme.
2.5. Enzyme activity assay and determination of kinetic parameters
N-d-AAase activity was measured using N-acetyl-d-methionine as a substrate.
Concentration of reaction product was measured by applying a d-amino acid oxidase
coupling with horse radish peroxidase. The detailed condition was as previously
described [17]. Basically, N-d-AAase and N-acetyl-d-methionine were incubated at
2. Materials and methods
4
◦C. Then the mixture of d-amino acid oxidase and colorimetric solution containing
2.1. Enzymes, chemicals, plasmids and bacterial strains
peroxidase (10 U/ml), 4-aminoantipyrine (0.04 ␮g/ml) and 0.8 ␮l phenol was added.
The OD at wavelength 520 nm was measured. One unit was defined as generation of
1 ␮mol d-methionine in 1 min. Protein concentration was determined by the Brad-
ford method and bovine serum albumin as the standard [32]. Kinetic parameters kcat
and Km values were determined by fitting the initial rates as a function of substrate
concentration to the Michaelis–Menten equation.
Catalase, d-amino acid oxidase from porcine kidney, horse radish peroxidase
and enzyme substrate N-acetyl-d-methionine were purchased form Sigma Chemical
Co. VIP peroxidase substrate kit was from Vector lab. Restriction enzymes and T4
DNA ligase were from Invitrogen. PfuTurbo® DNA polymerase and QuikChange®
Site-Directed Mutagenesis Kit were from Stratagene. DyNAzyme II DNA polymerase
was from Finnzymes. Pfu DNA polymerase was from Promega. Protein assay kit and
30% Acrylamide/Bis solution were from Bio-Rad. Plasmid pKK233-3 (GE Healthcare
Co.) and E. coli JM109 were used for constructing a protein expression vector and
expression host, respectively. E. coli XL-1 blue was used for DNA manipulation and
plasmid maintenance. QIAprep spin miniprep kit and QIAquick gel extraction kit for
plasmid preparation and DNA gel purification, respectively, were from QIAGEN.
2.6. Storage stability and oxidative resistance experiments
Enzymes were dialyzed against a 50 mM Tris–HCl buffer (containing 10% glyc-
erol, 0.01% 2-mercaptoehtanol, pH 7.5) and concentrated to approximately 1 mg/ml
using the Amicon Ultra (30 kDa MWCO). The enzyme solutions were stored at 4 ◦
C
and enzyme activity was periodically assayed. The half-life of each enzyme was
calculated according to the equation Y = 100 exp{f(x)} + ε. Y, x, and ε mean relative
enzyme specific activity (%), time (h), and associated standard error, respectively.
For oxidation stability experiment, the purified enzymes (1 mg/ml) were incubated
with various concentrations of hydrogen peroxide at 25 ◦C for 10 min and terminated
the oxidation reaction by adding catalase (246 U) for removal of residual hydrogen
peroxide. The stability of each enzyme was estimated by measuring the ratio of
residual activity to the initial activity of N-d-AAase.
2.2. Construction of the expression vector for producing recombinant N-d-AAase
The N-d-AAase gene of V. paradoxus Iso1 was amplified from plasmid pBK-
damH1 [17] using forward primer E and reverse primer H, in which a restriction
enzyme EcoRI recognition sequence and an E. coli-recognizing ribosome-binding
site (AAGGAG) were introduced in front of the start codon with the primer E, and
HindIII recognition sequence was introduced with primer H (Table 1). The reaction
mixture (50 ␮l) contained 1× reaction buffer (Finnzymes provided), 1 ␮M primers
(E and H) each, 200 ␮M dNTP, 1 ␮l plasmid DNA pBK-damH1 and 1 U DyNAzyme
II DNA polymerase (Finnzymes) combining 0.1 U of Pfu DNA polymerase. PCR pro-
gram consisted of 94 ◦C for 5 min then 25 cycles of 94 ◦C for 40 s, 37 ◦C 40 s, 72 ◦C for
3 min and finally stopped at 4 ◦C were performed on a GeneAmp PCR system 9700
(Applied Biosystems). The amplified gene fragment was gel purified, digested with
restriction enzymes (EcoRI and HindIII), gel purified once again, and ligated with
the same enzyme-treated vector, pKK233-3, to construct an N-d-AAase expression
plasmid pKK-dam3.
2.7. Homology modeling of V. paradoxus Iso1 N-d-AAase
The template used for the homology modeling was the crystal structure of A.
faecalis DA1 d-aminoacylase (PDB code: 1M7J) [21]. The structural model was built
using the HOMOLOGY module of the INSIGHT2005 (Accelrys, San Diego, CA, USA),
based on pairwise alignment. The structure with lowest violation score and lowest
energy score was chosen as the candidate. The candidate model was minimized in
200 steps using the DISCOVER3 module with a consistent valence force field. Finally,
the model was validated by PROCHECK program.
2.3. Site-directed mutagenesis
2.8. Circular dichroism
The site-directed mutagenesis of V. paradoxus Iso1 N-d-AAase was performed
directly on the plasmid pKK-dam3, containing the N-d-AAase gene, by using
QuikChange® Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA, USA). The OPC
purified mutagenic primers, cloning primers and sequencing primers were pur-
chased from MB Biotech (Taipei, Taiwan) (Table 1). Mutagenesis was performed
as described in the QuikChange kit manual. Reaction mixture (50 ␮l) contained 1×
reaction buffer (Stratagene provided), 1 ␮M primers each, 200 ␮M dNTP, 50 ng plas-
mid DNA (pKK-dam3) and 2.5 U PfuTurbo® DNA polymerase. PCR program consisted
The purified enzyme was concentrated with Amicon Ultra (30,000 MWCO) (Mil-
lipore) and exchanged into 50 mM sodium phosphate buffer (pH 7.5) in a final
concentration of 2 ␮M. CD spectra were taken over a range of 190–250 nm at 0.1 nm
increments and a 0.25 s acquisition time on a Jasco J-715 CD spectropolarimeter
(Jasco co., Japan). Each protein sample was scanned in a 0.1-cm quartz cuvette three
times with 50 mdeg sensitivity.

I.-C. Peng et al. / Process Biochemistry 47 (2012) 1785–1790
1787
Fig. 2. Circular dichroism spectra of wild-type and mutants in far UV region at
25 ◦C. Each spectrum was obtained from wild-type, M39L, M56L, M221L, M254L,
and M352L. The protein concentration was 2 ␮M and the cell length was 0.1 cm. The
buffer was 50 mM phosphate buffer (pH 7.5).
N-acetyl-d-methionine as substrates. After the N-acetyl group
was removed by N-d-AAase, the concentration of d-methionine
was measured by applying d-amino acid oxidase coupling with
horse radish peroxidase. In this study, one unit was defined as
generation of 1 ␮mol d-methionine in 1 min.
M39L and M221L mutants maintained the activities close to that
of the wild type enzyme, while M56L, M254L and M352L mutants
only retained about 70% wild type enzyme activity (Fig. 3). Inter-
estingly, substitution of M79, M171 and M273 with leucine totally
abolished the enzyme activity (data not shown). Although M171
and M273 were found oxidized in LC/MS/MS analysis, replacement
of leucine residue did not help and was even detrimental to enzyme
activity. This suggests that these three methionine residues may be
critical for enzyme activity.
Fig. 1. 10% SDS-PAGE (A) and Western blotting (B) of purified wild-type and mutant
N-d-AAase. Lane 1: Wild type; Lane 2: M39L; Lane 3: M56L; Lane 4: M79L; Lane 5:
M171L; Lane 6: M221L; Lane 7: M254L; Lane 8: M273L; Lane 9: M352L.
3. Results and discussion
3.1. Mutagenesis and expression of the N-d-AAase
The N-d-AAase enzyme of V. paradoxus Iso1 purified from
E. coli JM109 was unstable during protein storage at 4 ◦C. The
activity of the enzymes can only last less than four days. This
N-d-AAase peptide contains seventeen methionine residues [17].
LC/MS/MS peptide sequencing identified 79% of the N-d-AAase pro-
tein sequence, and showed that five methionine residues of the
enzyme (Met39, Met171, Met254, Met273, and Met352) were oxi-
dized into methionine sulfoxide by the increase 16 of molecular
mass. Nine methionine residues were not oxidized and identi-
fied in LC/MS/MS analysis. However, three methionine residues
(Met56, Met79 and Met221) within the peptides were not sequenced
successfully in LC/MS/MS analysis, and the oxidation of these
residues were not known. Peptide sequencing results suggested
that these oxidative methionine residues of the N-d-AAase may
lead to enzyme malfunction.
3.2. Storage stabilities of the N-d-AAase mutants
Since M79L, M171L and M273L mutants lost the enzyme activ-
ities, only wild-type and 5 mutants (M39L, M56L, M221L, M254L
and M352L) were subjected to enzyme storage experiment at 4 ◦C.
One mg/ml purified enzymes were stored at 4 ◦C and the activities
were measured every day.
Three mutants (M39L, M221L, and M254L) showed great
improvement of the enzyme stability. The half-lives were extended
to 156.2 h, 255.6 h and 174.8 h, respectively, as compared with the
Since oxidation was suspected as the main reason how puri-
fied N-d-AAase lost its activity during the storage, we tried to use
various concentrations of reducing reagents, including 0.01–0.1%
of ␤-mercaptoethanol and 1–10 mM of dithiothreitol to increase
the stability of N-d-AAase. However, the reducing reagents we
tested did not prolong the enzyme stability compared to the con-
trol (data not shown). To avoid the oxidation of methionine, we
applied site-directed mutagenesis to substitute methionine with
non-oxidizable leucine. We selected eight Met residues of the N-
d-AAase including Met39, Met56, Met79, Met171, Met221, Met254
,
Met273, and Met352 for further investigations.
Wild-type and eight mutant enzymes were expressed and puri-
fied from E. coli JM109 (Fig. 1, upper gel). Each mutant enzyme
showed the same apparent molecular mass (53Kda) as that of
the wild-type enzyme in Western blotting analysis (Fig. 1, lower
gel). To detect the potential structure change from the introduced
mutation, 1–2 ␮M wild-type and mutant enzymes were scanned in
circular dichroism at wavelength 190 to 250 nm. The similar spec-
tra indicated that the mutations do not affect the global structure
of mutant enzymes (Fig. 2).
Fig. 3. Relative enzyme activities of purified wile-type and mutant N-d-AAases.
N-d-AAase activity was measured using N-acetyl-d-methionine as substrates. The
specific activity of wild type enzyme was set as 100%.
Next, we measured the enzyme activities of these
mutants. Purified N-d-AAase activity was measured using

1788
I.-C. Peng et al. / Process Biochemistry 47 (2012) 1785–1790
Table 2
Table 3
Half-life estimates and associated standard errors of N-d-AAase mutants at 4 ◦C. Data
were fitted by an exponential model of the form y = 100 exp [15] + ε. Half-lives were
decided by the coefficients of f(t) and calibration method.
Specific activities and kinetics analysis of N-d-AAase mutants.
a
Enzyme
Specific activity (U/mg) kcat (s−1
)
Km (mM) kcat/Km (M−1 s−1
)
Wild-type 76.58
57.53
61.31
25.20
76.08
54.06
34.56
27.89
19.64
26.30
15.52
13.73
14.36
2.06 × 103
3.12 × 103
9.58 × 102
4.90 × 103
3.94 × 103
2.41 × 103
Enzyme
Half-life (h)
Estimate
M39L
M56L
M221L
M254L
M352L
71.06
51.27
85.78
52.83
54.42
S.E.
Wild-type
M39L
M56L
M221L
M254L
M352L
41.9
156.2
50.6
255.6
174.8
58.1
2.6
2.7
1.3
8.7
4.7
2.3
a
N-acetyl-d-methionine was used as reaction substrate. The initial velocity was
determined at various concentrations of N-acetyl-d-methionine (0.1, 1, 5, 10, 25, 50,
75, 100, 125 mM).
opposite ways in resistance to H2O2. Similar to M39L, M56L only
showed slight improvement whereas M352L even decreased the
resistance at 50 ␮M and 100 ␮M H2O2 (Fig. 4).
M221L and M254L mutants, who show the longest half-lives at
4 ◦C, also exhibited more resistance to 100 ␮M hydrogen perox-
ide oxidation than wild type enzyme (Fig. 4). While wild-type and
other mutants almost lost entire enzyme activities, mutant M221L
could even retain about 25% and 12% of its initial activity following
treatment with 500 ␮M and 1 mM H2O2 for 10 min, respectively
(Fig. 4).
41.9 h of wild-type enzyme. The half-lives of the other 2 mutants
(M56L and M352L) were almost similar to wild-type (Table 2).
Mutations occurred on M39L, M221L and M254L could actually
prolong storage life at 4 ◦C beyond the fixed limit of wild-type. Thus
this result emphasized the fact that oxidation occurring on criti-
cal methionine residues was the main factor that led to enzyme
inactivation.
3.3. Oxidative resistance of the N-d-AAase mutants
The results strongly supported that the d-aminoacylase rapidly
inactivated during protein purification and storage at 4 ◦C were
caused by the oxidation of methionine residues on protein, and
Met221 was the primary residue which oxidation would remark-
ably impact the enzyme activity. Although mutant M221L resisted
1 mM H2O2 inactivation, the low retained activity may suggest that
some residues except Met were also be oxidized and resulted in
partially decreased enzyme activity. Oxidation in proteins is com-
mon occurred on histidine, lysine, proline, cysteine, arginine and
methionine residues [34].
Hydrogen peroxide treatment was performed for evaluation of
the oxidative resistance of wild-type and mutant enzymes. The oxi-
dation of enzymes caused by hydrogen peroxide correlated to the
production of methionine sulfoxide and led to enzyme inactivation
[27,33]. Different concentrations of H2O2 were treated on wild-type
and mutant enzymes (M39L, M56L, M221L, M254L, and M352L) for
10 min. After that, we added catalase into the reaction mixture for
decomposition of hydrogen peroxide to water and oxygen. Then
the remained enzyme activity was measured after the action of
catalase.
Although mutant M39L exhibit almost 4 times longer stability
at 4 ◦C than that of the wild type, it showed slight improve-
ment of enzyme resistance to H2O2 at 50 ␮M and 100 ␮M (Fig. 4)
but retained trace enzyme activity (under 1% activity relative to
its initial activity) after a 500 ␮M or 1 mM hydrogen peroxide
treatment. While both M56L and M352L mutants did not signifi-
cantly extent the stability at 4 ◦C storage experiment, they showed
3.4. Kinetic characterization
The kinetic constants, kcat and Km, were determined for the wild-
type and mutant enzymes as shown in Table 3. Although Met39 was
far away from the Zn2+ catalytic center, replacement of Met with
Leu resulted in a 29.6% decrease in Km and an almost equal kcat value
(Table 3). These results suggested that the small ␤-barrel structure
of the enzyme should be related to the substrate binding but not
related to enzyme catalysis.
Mutant enzyme M221L revealed not only a 44% decrease in Km
,
but also a 32% increase in turn-over rate (kcat). The kcat/Km value
increased 2.4 fold compared with value for the wild-type enzyme
(Table 3). The experimental results emphasized that substitution
Met221 with non-oxidizable leucine could enhance not only the
substrate affinity but also its catalytic efficiency.
The mutant enzyme M254L showed a 50% decrease in Km, an
almost equal turn-over rate (kcat), and a 50% increase in kcat/Km
(Table 3). The experimental results suggested that Met254 corre-
lated directly with substrate binding but not with catalytic reaction.
3.5. Structural analysis
A crystal structure of N-d-AAase derived from A. faecalis DA1,
its protein sequence shares 54.6% identity and 62.5% similarity
with that of N-d-AAase (accession no. AY127900) from V. paradoxus
Iso1, was recently reported [21], and was input as the template for
homology modeling on the Swiss-Model server [35]. The positions
of 8 mutated Met residues on the structural model of V. paradoxus
Iso1 N-d-AAase was shown in Fig. 5A. Met39 located at the small
␤-barrel, which is related to its structural stability. Met221 and
Met254 located at the catalytic (␤␣)8-barrel structure, and were
Fig. 4. Oxidative stability of the wild-type and mutant N-d-AAase. The purified
enzymes were concentrated to 1 mg/ml in 10% glycerol, 0.01% 2-mercaptoethanol,
and 50 mM Tris–HCl buffer (pH 7.5). Oxidative stabilities were determined by incu-
bation of buffered enzymes with different concentrations of H2O2 for 10 min at
room temperature. The residual activities were measured under the standard assay
conditions. Data represent the averages of triplicate experiments.
˚
˚
spatially close to the zinc-assist catalytic site (10.49 A and 9.8 A).

I.-C. Peng et al. / Process Biochemistry 47 (2012) 1785–1790
1789
Fig. 5. (A) Ribbon diagram of the V. paradoxus Iso1 N-d-AAase structure model. The positions of Met39, Met56, Met79, Met171, Met221, Met254, Met273 and Met352 are shown in
ball-and-stick structure. (B) Location of Met39 and the surrounding amino acid residues. (C) The partial view surrounding Met221 in which neighboring amino acid residues
are located. (D) Met254 and the neighboring amino acid residues Trp191, Tyr192, Lys252, Met298, Tyr344, Phe346 and Ile347 constitute the substrate binding pocket.
Met56 and Met352 locate at the surface of the small ␤-barrel and TIM
barrel structure, respectively. Mutagenesis on the surface-exposed
residues Met56 and Met352 did not make an improvement in half-
into methionine sulfoxide, the polar side-chain of that would inter-
fere the structural stability of the hydrophobic region, and would
lead to enzyme inactivation.
life, H2O2 resistance or enzyme kinetics. The side chains of Met39
,
According to the structural model of V. paradoxus Iso1 N-d-
AAase, the side chains of residues Met254 with Trp191, Tyr192, Lys252
,
Val16, Phe25, Val29, Ile36 and Ile421 form a hydrophobic packing
(Fig. 5B). When the residue Met39 was oxidized into methionine
sulfoxide, the hydrophobic packing and interaction in the region
may be changed and the close contact between Met39 S␦ and Val29
Met298, Tyr344, Phe346, and Ile347 pack a hydrophobic pocket, in
which substrate was recognized (Fig. 5D). When the residue Met254
was substituted with the higher hydrophobic residue Leu, the H2O2
resistance ability and substrate binding ability (Km) of N-d-AAase
were both enhanced (Table 3). This result supported that main-
taining hydrophobicity at Met residue 254 is required for substrate
binding, and higher hydrophobic amino acid residue substituted
at this position would enhance the substrate binding ability but
not the catalytic ability. Once the non-polar residue Met254 was
oxidized into polar residue methionine sulfoxide, the hydrophobic
environment of the substrate pocket would be changed and that
would disturb the substrate recognition and lead to losing enzyme
activity.
C␥2 (3.1 A) may lead to some steric hindrance.
˚
The protein model of V. paradoxus Iso1 N-d-AAase demonstrated
that a catalytically essential Zn2+ tightly bound on ␤ site with
coordination by Cys95, His220 and His250 that activate the enzyme
(Fig. 5C). Residue Met221 is just adjacent to the Zn2+ coordinat-
ing residue His220 in the protein sequence. Based on the protein
model, the Met221 side chain S␦ is close to Zn2+ at ␤ site (10.49 A)
˚
232
and has close contacts with Val228 C (6.08 A), Leu
C␦2 (3.35 A),
˚
˚
Thr235 O␥1 (3.88 A) and Val
C␤ (4.65 A) that together form a
248
˚
˚
stable hydrophobic packing structure (Fig. 5C). The hydrophobic-
ity index of Leu (3.8 in Kyte-Doolittle scale) is higher than that of
Met (1.9 in Kyte-Doolittle scale). Replacement of Met residue with
more hydrophobic residue Leu on position Met221 would further
maintain the structural stability of this region. Furthermore, the
improvement of kcat value and Km value actually supported the
inference that position 221 required hydrophobic residue for sub-
strate recognition and catalytic reaction. If Met221 were oxidized
Based on the results of site-directed mutagenesis analysis,
kinetic characterization, the storage stability and the H2O2 oxi-
dation experiments, we concluded that oxidation of specific
methionine residues caused the rapid lose of V. paradoxus Iso1 N-
d-AAase enzyme activity during protein purification and storage at
4 ◦C. Oxidation that occurred on Met39, Met221 and Met254 impacts
the maintenance of enzyme activity. Met221 close proximity to the

1790
I.-C. Peng et al. / Process Biochemistry 47 (2012) 1785–1790
active site was the primary residue for oxidative inactivation of V.
paradoxus Iso1 N-d-AAase.
[12] Sakai K, Oshima K, Moriguchi M. Production and characterization of N-acyl-
d-glutamate amidohydrolase from Pseudomonas sp. strain 5f-1. Appl Environ
Microbiol 1991;57:2540–3.
[13] Muniz-Lozano FE, Dominguez-Sanchez F, Diaz-Viveros Y, Barradas-Dermitz
DM. d-Aminoacylase from a novel producer: Stenotrophomonas maltophilia ITV-
0595. J Ind Microbiol Biotechnol 1998;21:296–9.
4. Conclusion
[14] Sugi M, Suzuki H. Purification and properties of d-aminoacylase of Streptomyces
olivaceus. Agric Biol Chem 1978;42:107–13.
[15] Liu J, Nakayama T, Hemmi H, Asano Y, Tsuruoka N, Shimomura K, et al. Microbac-
In this article, the N-d-AAase gene from V. paradoxus Iso1 was
cloned and expressed in E. coli. The Iso1 N-d-AAase is suitable for
optical resolution of N-acyl-dl amino acids. To improve the enzyme
oxidative resistance, site-directed mutagenesis was performed, and
each mutant protein was purified and characterized. Based on the
structural analysis of the stabilizing effects of each amino acid sub-
stitution, we found that methionine oxidation was the critical factor
leading to the inactivation of N-d-AAase, and Met221 spatial close-
ness to the Zinc-assistant center is the primary site for oxidative
inactivation. Our study will contribute significantly to figure out
amino acid residues in enhancing N-d-AAase activity and oxidative
resistance.
terium natoriense sp. nov.,
a novel d-aminoacylase-producing bacterium
isolated from soil in Natori, Japan. Int J Syst Evol Microbiol 2005;55:661–5.
[16] Hsu CS, Lai WL, Chang WW, Liaw SH, Tsai YC. Structural-based muta-
tional analysis of d-aminoacylase from Alcaligenes faecalis DA1. Protein Sci
2002;11:2545–50.
[17] Lin PH, Su SC, Tsai YC, Lee CY. Identification and characterization of a new gene
from Variovorax paradoxus Iso1 encoding N-acyl-d-amino acid amidohydrolase
responsible for d-amino acid production. Eur J Biochem 2002;269:4868–78.
[18] Wakayama M, Ashika T, Miyamoto Y, Yoshikawa T, Sonoda Y, Sakai K, et al.
Primary structure of N-acyl-d-glutamate amidohydrolase from Alcaligenes
xylosoxydans subsp. xylosoxydans A-6. J Biochem 1995;118:204–9.
[19] Wakayama M, Katsuno Y, Hayashi S, Miyamoto Y, Sakai K, Moriguchi M. Cloning
and sequencing of a gene encoding d-aminoacylase from Alcaligenes xylosoxy-
dans subsp. xylosoxydans A-6 and expression of the gene in Escherichia coli.
Biosci Biotechnol Biochem 1995;59:2115–9.
Acknowledgments
[20] Lai WL, Chou LY, Ting CY, Kirby R, Tsai YC, Wang AH, et al. The functional
role of the binuclear metal center in d-aminoacylase: one-metal activation and
second-metal attenuation. J Biol Chem 2004;279:13962–7.
This work was supported by Grant NSC-92-2317-B-002-010
from the National Science Council, Taipei, Taiwan. The authors
would like to thank professors SH Liaw for the valuable discus-
sion regarding protein homology modeling, TC Chang for assistance
with obtaining the CD spectra, and YM Pong for the calculation of
enzyme half-lives.
[21] Liaw SH, Chen SJ, Ko TP, Hsu CS, Chen CJ, Wang AH, et al. Crystal structure of d-
aminoacylase from Alcaligenes faecalis DA1. A novel subset of amidohydrolases
and insights into the enzyme mechanism. J Biol Chem 2003;278:4957–62.
[22] Stadtman ER, Moskovitz J, Levine RL. Oxidation of methionine residues of pro-
teins: biological consequences. Antioxid Redox Signal 2003;5:577–82.
[23] Levine RL, Berlett BS, Moskovitz J, Mosoni L, Stadtman ER. Methionine residues
may protect proteins from critical oxidative damage. Mech Ageing Dev
1999;107:323–32.
[24] Gellman SH. On the role of methionine residues in the sequence-independent
recognition of nonpolar protein surfaces. Biochemistry 1991;30:6633–6.
[25] Harndahl U, Kokke BP, Gustavsson N, Linse S, Berggren K, Tjerneld F, et al. The
chaperone-like activity of a small heat shock protein is lost after sulfoxidation of
conserved methionines in a surface-exposed amphipathic alpha-helix. Biochim
Biophys Acta 2001;1545:227–37.
[26] Bordo D, Argos P. Suggestions for safe residue substitutions in site-directed
mutagenesis. J Mol Biol 1991;217:721–9.
[27] Estell DA, Graycar TP, Wells JA. Engineering an enzyme by site-directed
mutagenesis to be resistant to chemical oxidation. J Biol Chem 1985;260:
6518–21.
References
[1] Bodanszky M, Perlman D. Peptide antibiotics. Science 1969;163:352–8.
[2] Asano Y, Lubbehusen TL. Enzymes acting on peptides containing d-amino acid.
J Biosci Bioeng 2000;89:295–306.
[3] Martinez-Rodriguez S, Martinez-Gomez AI, Rodriguez-Vico F, Clemente-
Jimenez JM, Las Heras-Vazquez FJ. Natural occurrence and industrial
applications of d-amino acids: an overview. Chem Biodivers 2010;7:1531–48.
[4] Volpato G, Rodrigues RC, Fernandez-Lafuente R. Use of enzymes in the pro-
duction of semi-synthetic penicillins and cephalosporins: drawbacks and
perspectives. Curr Med Chem 2010;17:3855–73.
[28] Chien HR, Hsu CL, Hu HY, Wang WC, Hsu WH. Enhancing oxidative resis-
tance of Agrobacterium radiobacter N-carbamoyl d-amino acid amidohydrolase
by engineering solvent-accessible methionine residues. Biochem Biophys Res
Commun 2002;297:282–7.
[5] Su SC, Lee CY. Cloning of the N-acylamino acid racemase gene from Amyco-
latopsis azurea and biochemical characterization of the gene product. Enzyme
Microb Technol 2002;30:647–55.
[6] Tokuyama S, Hatano K. Overexpression of the gene for N-acylamino acid
racemase from Amycolatopsis sp. TS-1-60 in Escherichia coli and continuous
production of optically active methionine by a bioreactor. Appl Microbiol
Biotechnol 1996;44:774–7.
[7] Wakayama M, Watanabe E, Takenaka Y, Miyamoto Y, Tau Y, Sakai K, et al.
Cloning, expression, and nucleotide sequence of the N-acyl-d-aspartate ami-
dohydrolase gene from Alcaligenes xylosoxydans subsp. xylosoxydans A-6. J
Ferment Bioeng 1995;80:311–7.
[8] Moriguchi M, Ideta K. Production of d-aminoacylase from Alcaligenes denitrifi-
cans subsp. xylosoxydans MI-4. Appl Environ Microbiol 1988;54:2767–70.
[9] Tsai YC, Tseng CP, Hsiao KM, Chen LY. Production and purification of d-
aminoacylase from Alcaligenes denitrificans and taxonomic study of the strain.
Appl Environ Microbiol 1988;54:984–9.
[29] Han JI, Choi HK, Lee SW, Orwin PM, Kim J, Laroe SL, et al. Complete genome
sequence of the metabolically versatile plant growth-promoting endophyte
Variovorax paradoxus S110. J Bacteriol 2011;193:1183–90.
[30] Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating
inhibitors. Proc Natl Acad Sci U S A 1977;74:5463–7.
[31] Laemmli UK. Cleavage of structural proteins during the assembly of the head
of bacteriophage T4. Nature 1970;227:680–5.
[32] Bradford MM. A rapid and sensitive method for the quantitation of micro-
gram quantities of protein utilizing the principle of protein–dye binding. Anal
Biochem 1976;72:248–54.
[33] Miyazaki C, Takahashi H. Engineering of the H2O2-binding pocket region
of a recombinant manganese peroxidase to be resistant to H2O2. FEBS Lett
2001;509:111–4.
[10] Yang YB, Hsiao KM, Li H, Yano H, Tsugita A, Tsai YC. Characterization of d-
aminoacylase from Alcaligenes denitrificans DA181. Biosci Biotechnol Biochem
1992;56:1392–5.
[34] Stadtman ER. Oxidation of free amino acids and amino acid residues in
proteins by radiolysis and by metal-catalyzed reactions. Annu Rev Biochem
1993;62:797–821.
[11] Chen HP, Wu SH, Wang KT. d-Aminoacylase from Alcaligenes faecalis possesses
novel activities on d-methionine. Bioorg Med Chem 1994;2:1–5.
[35] Schwede T, Kopp J, Guex N, Peitsch MC. SWISS-MODEL: an automated protein
homology-modeling server. Nucleic Acids Res 2003;31:3381–5.