Role of the active site residues arginine-216 and arginine-237 in the substrate specificity of mammalian D-aspartate oxidase

Article abstract of DOI:10.1007/s00726-010-0658-4

d-Aspartate oxidase (DDO) and d-amino acid oxidase (DAO) are flavin adenine dinucleotide-containing flavoproteins that catalyze the oxidative deamination of d-amino acids. Unlike DAO, which acts on several neutral and basic d-amino acids, DDO is highly specific for acidic d-amino acids. Based on molecular modeling and simulated annealing docking analyses, a recombinant mouse DDO carrying two substitutions (Arg-216 to Leu and Arg-237 to Tyr) was generated (R216L-R237Y variant). This variant and two previously constructed single-point mutants of mouse DDO (R216L and R237Y variants) were characterized to investigate the role of Arg-216 and Arg-237 in the substrate specificity of mouse DDO. The R216L-R237Y and R216L variants acquired a broad specificity for several neutral and basic d-amino acids, and showed a considerable decrease in activity against acidic d-amino acids. The R237Y variant, however, did not show any additional specificity for neutral or basic d-amino acids and its activity against acidic d-amino acids was greatly reduced. The kinetic properties of these variants indicated that the Arg-216 residue is important for the catalytic activity and substrate specificity of mouse DDO. However, Arg-237 is, apparently, only marginally involved in substrate recognition, but is important for catalytic activity. Notably, the substrate specificity of the R216L-R237Y variant differed significantly from that of the R216L variant, suggesting that Arg-237 has subsidiary effects on substrate specificity. Additional experiments using several DDO and DAO inhibitors also suggested the involvement of Arg-216 in the substrate specificity and catalytic activity of mouse DDO and that Arg-237 is possibly involved in substrate recognition by this enzyme. Collectively, these results indicate that Arg-216 and Arg-237 play crucial and subsidiary role(s), respectively, in the substrate specificity of mouse DDO.

Full text of DOI:10.1007/s00726-010-0658-4

Amino Acids (2011) 40:467–476
DOI 10.1007/s00726-010-0658-4
ORIGINAL ARTICLE
Role of the active site residues arginine-216 and arginine-237
in the substrate specificity of mammalian D-aspartate oxidase
•
Masumi Katane Yasuaki Saitoh Kazuhiro Maeda
•
•
•
•
•
Toshihiko Hanai Masae Sekine Takemitsu Furuchi
Hiroshi Homma
Received: 20 May 2010 / Accepted: 8 June 2010 / Published online: 22 June 2010
Ó Springer-Verlag 2010
Abstract D-Aspartate oxidase (DDO) and D-amino acid
oxidase (DAO) are flavin adenine dinucleotide-containing
flavoproteins that catalyze the oxidative deamination of
D-amino acids. Unlike DAO, which acts on several neutral
and basic D-amino acids, DDO is highly specific for acidic
D-amino acids. Based on molecular modeling and simu-
lated annealing docking analyses, a recombinant mouse
DDO carrying two substitutions (Arg-216 to Leu and Arg-
237 to Tyr) was generated (R216L-R237Y variant). This
variant and two previously constructed single-point
mutants of mouse DDO (R216L and R237Y variants) were
characterized to investigate the role of Arg-216 and Arg-
237 in the substrate specificity of mouse DDO. The R216L-
R237Y and R216L variants acquired a broad specificity for
several neutral and basic D-amino acids, and showed a
considerable decrease in activity against acidic D-amino
acids. The R237Y variant, however, did not show any
additional specificity for neutral or basic D-amino acids and
its activity against acidic D-amino acids was greatly
reduced. The kinetic properties of these variants indicated
that the Arg-216 residue is important for the catalytic
activity and substrate specificity of mouse DDO. However,
Arg-237 is, apparently, only marginally involved in sub-
strate recognition, but is important for catalytic activity.
Notably, the substrate specificity of the R216L-R237Y
variant differed significantly from that of the R216L vari-
ant, suggesting that Arg-237 has subsidiary effects on
substrate specificity. Additional experiments using several
DDO and DAO inhibitors also suggested the involvement
of Arg-216 in the substrate specificity and catalytic activity
of mouse DDO and that Arg-237 is possibly involved in
substrate recognition by this enzyme. Collectively, these
results indicate that Arg-216 and Arg-237 play crucial and
subsidiary role(s), respectively, in the substrate specificity
of mouse DDO.
Keywords D-Aspartate oxidase ꢀ D-Amino acid oxidase ꢀ
Flavoprotein ꢀ Site-directed mutagenesis ꢀ
Substrate specificity
Abbreviations
DAO
DDO
FAD
IC50
D-Amino acid oxidase
D-Aspartate oxidase
Flavin adenine dinucleotide
50% Inhibitory concentration
Isopropyl-b-D-thiogalactopyranoside
IPTG
NMDA N-Methyl-D-aspartate
Introduction
Of the free D-amino acids that are present in mammals,
D-serine (D-Ser) and D-aspartate (D-Asp) are the most
studied. D-Ser is found in the mammalian forebrain, where
it persists at high concentrations over the lifetime of the
animal. It binds to the glycine (Gly)-binding site of
the N-methyl-D-Asp (NMDA) receptor, a subtype of the
L-glutamate (L-Glu) receptor, and potentiates glutamatergic
neurotransmission in the central nervous system (Matsui
et al. 1995; Mothet et al. 2000). Accordingly, it has been
proposed that D-Ser regulates the L-Glu-mediated activation
M. Katane ꢀ Y. Saitoh ꢀ K. Maeda ꢀ T. Hanai ꢀ M. Sekine ꢀ
T. Furuchi ꢀ H. Homma (&)
Laboratory of Biomolecular Science, Department
of Pharmaceutical Life Sciences, Kitasato University,
5-9-1 Shirokane, Minato-ku, Tokyo 108-8641, Japan
e-mail: hommah@pharm.kitasato-u.ac.jp
123

468
M. Katane et al.
of the NMDA receptor by acting as a co-agonist (for
reviews see Nishikawa 2005; Scolari and Acosta 2007). In
contrast, substantial free D-Asp is found in a wide variety of
mammalian tissues, particularly in the central nervous,
neuroendocrine, and endocrine systems. Several lines of
evidence suggest that D-Asp plays an important role in
regulating developmental processes, hormone secretion,
and steroidogenesis (for reviews see D’Aniello 2007;
Homma 2007).
model for mouse DDO was proposed based on the deduced
amino acid sequence. In the proposed structure, the side
chains of Arg-216, Tyr-223, Arg-237, Arg-278, and Ser-308
were oriented toward the predicted active site binding pocket
(Katane et al. 2007a), implicating the involvement of these
residues in mouse DDO catalysis. Mutagenesis analysis
enabled the identification of Arg-216, Arg-237, and Ser-308
of mouse DDO as functionally important residues (Katane
et al. 2007a, 2008). However, the precise roles of these
residues in catalysis remain to be clarified.
In mammalian tissues, two types of degradative
enzymes that are stereospecific for D-amino acids have
been identified: D-amino acid oxidase (DAO, also abbre-
viated to DAAO, EC 1.4.3.3) and D-Asp oxidase (DDO,
also abbreviated to DASPO, EC 1.4.3.1). DAO and DDO
are flavin adenine dinucleotide (FAD)-containing flavo-
proteins that catalyze the oxidative deamination of D-amino
acids to generate the corresponding 2-oxo acids, along with
hydrogen peroxide and ammonia. DAO displays broad
substrate specificity and acts on several neutral and basic
D-amino acids, such as D-Ser, D-Ala, and D-Arg. DDO is
highly specific for acidic D-amino acids, such as D-Asp,
NMDA and D-Glu; none of which are substrates for DAO.
DAO and DDO have been found in various organisms and
their physiological roles in vivo are now being extensively
investigated. Mammalian DAO and DDO reportedly reg-
ulate endogenous D-Ser and D-Asp levels, respectively, and
mediate the elimination of accumulated exogenous
D-amino acids from various organs (for reviews see Katane
and Homma 2010; Pollegioni et al. 2007).
In the present study, a similar approach was used to
investigate the roles of Arg-216 and Arg-237 of mouse
DDO in the catalytic process, particularly with respect to
substrate specificity. A double-point mutant of mouse DDO
(R216L-R237Y variant) was used, in which the Arg-216
and Arg-237 codons were mutated to Leu and Tyr,
respectively. This variant, along with previously con-
structed single-point mutants of mouse DDO (R216L and
R237Y variants; Katane et al. 2007a), was characterized.
This study reports a crucial role for Arg-216, and a sub-
sidiary role for Arg-237, in the substrate specificity of
mouse DDO and provides novel and useful insights into the
structure–function relationships of mammalian DDO.
Materials and methods
Chemicals
Recently, it was shown that the amount of D-Asp in
human seminal plasma and spermatozoa is significantly
reduced in oligoasthenoteratospermic and/or azoospermic
donors compared with normospermic donors (D’Aniello
et al. 2005). In human patients undergoing in vitro fertil-
ization, the D-Asp content of the pre-ovulatory follicular
fluid is relatively lower in older patients than in younger
ones (D’Aniello et al. 2007). This decrease in D-Asp con-
tent appears to be related to a decrease in the oocyte quality
and fertilization competence. These reports support the
physiological significance of DDO in regulating the D-Asp
content in vivo. However, little is known about the regu-
latory mechanisms controlling DDO activity, or the
molecular details of its structure–function relationships.
X-ray crystallographic analysis, in vitro mutagenesis,
and/or the use of specific enzyme inhibitor(s) are helpful and
effective methods for detailed investigation of the structure–
function relationships of the enzyme. These approaches have
been used to identify several amino acid residues within
DAO that are functionally and structurally important (for
reviews see Pilone 2000; Tishkov and Khoronenkova 2005).
Previously, we cloned a cDNA from mouse kidney encoding
DDO (Katane et al. 2007a). Since the three-dimensional
(3D) structure of DDO has not been determined, a structural
Ampicillin, D- and L-amino acids, bovine serum albumin,
and catalase from Aspergillus niger were purchased from
Sigma–Aldrich (St Louis, MO, USA). FAD, isopropyl-b-
D-thiogalactopyranoside (IPTG), imidazole, 2,4-dinitro-
phenylhydrazine, benzoate, crotonate, malonate, and
meso-tartrate were purchased from Wako Pure Chemical
Ind. (Osaka, Japan). All other chemicals were of the
highest grade available and purchased from commercial
sources.
Construction of recombinant protein expression
plasmids
The construction of expression plasmids for the N-termi-
nally His-tagged mouse DDO (pRSET-His-mDASPO),
N-terminally His-tagged Arg-216-to-Leu (R216L) variant
of mouse DDO (pRSET-His-mDASPO-R216L), N-termi-
nally His-tagged Arg-237-to-Tyr (R237Y) variant of
mouse DDO (pRSET-His-mDASPO-R237Y), and N-ter-
minally His-tagged human DAO (pRSET-His-hDAO) has
been described previously (Katane et al. 2007a, 2010). The
expression plasmid for N-terminally His-tagged mouse
DAO (pET15-DAO), in which the cDNA fragment corre-
sponding to the entire mouse DAO-coding sequence was
123

Role of the active site residues arginine-216 and arginine-237
469
cloned into pET-15b (Novagen, Madison, WI, USA), was
kindly provided by Prof. R. Konno (International Univer-
sity of Health and Welfare, Tochigi, Japan).
recombinant proteins were purified to near-homogeneity
when examined by SDS-PAGE. From 1 L of fermentation
broth, we obtained the following amounts of purified
N-terminally His-tagged proteins: 0.3 mg mouse DDO,
0.3 mg human DAO, 0.4 mg mouse DAO, 0.2 mg R216L
variant of mouse DDO, 0.2 mg R237Y variant of mouse
DDO, and 0.1 mg R216L-R237Y variant of mouse DDO.
An R237Y substitution was introduced into pRSET-His-
mDASPO-R216L by site-directed mutagenesis using the
QuikChange II Site-Directed Mutagenesis Kit (Stratagene,
La Jolla, CA, USA) to generate an expression plasmid for
the N-terminally His-tagged R216L-R237Y variant of
mouse DDO (pRSET-His-mDASPO-R216L-R237Y). The
following mutagenic oligonucleotides were used (mutated
codons are underlined): 50-GTA ACC CTG GGA GGA
ACT TAC CAG AAA GGA GAC TGG AAT CG-30 and
50-CGA TTC CAG TCT CCT TTC TGG TAA GTT CCT
CCC AGG GTT AC-30. Introduction of the desired muta-
tions was confirmed by sequencing.
Enzymatic activity assays
The protein concentrations of the purified enzyme prepa-
rations were determined using the Bio-Rad Protein Assay
Kit (Bio-Rad Laboratories, Hercules, CA, USA) with
bovine serum albumin as a standard.
The activity of DDO and DAO was determined using a
colorimetric assay for 2-oxo acid production as previously
described (Katane et al. 2007b). Briefly, appropriate amounts
(0.25–28 lg) of the purified enzymes were added to a reaction
mixture of air-saturated 40 mM sodium pyrophosphate buffer
(pH 8.3), 23 U A. niger catalase, 60 lM FAD, and 20 mM
amino acids in a final volume of 150 lL. The mixture was
incubated at 37°C for 15 min, and then 10 lL of 100% (w/v)
trichloroacetic acid was added to stop the reaction. The 2-oxo
acid product was reacted with 2,4-dinitrophenylhydrazine and
quantitated by measuring the A445 against a blank mixture
lacking amino acids. One unit of enzyme activity was defined
as the production of 1 lmol 2-oxo acid per min, under the
assay conditions described above.
Expression and purification of recombinant proteins
Escherichia coli strain BL21(DE3)pLysS cells were trans-
formed with expression plasmids and cultured at 37°C with
shaking in Luria–Bertani medium containing ampicillin
(100 lg/mL). Crude extracts were prepared from cells trans-
formed with pRSET-His-mDASPO, -hDAO, and pET15-
DAO, as previously described (Katane et al. 2008, 2010).
For cells transformed with pRSET-His-mDASPO-R216L,
-R237Y, or -R216L-R237Y, the culture was grown to
A620 = 0.5, incubated for an additional 30 min at a reduced
temperature (30°C), and maintained at 26°C for an extra
30 min. After adding 0.1 mM IPTG, the culture was further
incubated at 26°C for 16 h. Cells were then centrifuged at
10,000 9 g for 10 min at 4°C, and crude extract was prepared
using BugBusterÒ Protein Extraction Reagent and LysonaseTM
Bioprocessing Reagent (Novagen) in the presence of 50 lM
FAD and protease inhibitors (Roche Applied Science, Mann-
heim, Germany), according to the manufacturer’s instructions.
All recombinant proteins were purified by affinity
chromatography using a chelating column as described
below. Crude extracts were applied to a His GraviTrapTM
column (GE Healthcare Bio-Sciences, Piscataway, NJ,
USA) equilibrated with 20 mM sodium phosphate buffer
(pH 7.4) containing 0.5 M NaCl and 10 mM imidazole.
The column was then washed with the same buffer, and
bound proteins were eluted using a step-wise gradient of
50–500 mM imidazole. Each fraction (2.0–2.5 mL) con-
taining recombinant protein was mixed with 100 lL of
1 mM FAD and dialyzed twice for 3 h at 4°C against 1 L
of 10 mM sodium pyrophosphate buffer (pH 8.3) con-
taining 2 mM EDTA, 5 mM 2-mercaptoethanol, and 10%
glycerol. The dialyzed fractions were recovered and cen-
trifuged at 10,000 9 g for 10 min at 4°C to pellet the
proteins denatured during dialysis. The supernatants were
pooled as purified enzyme and used immediately for
enzyme assays or stored frozen at -80°C until use. All
The kinetic parameters of the enzyme reaction were
determined under conditions in which the production of
2-oxo acids was linear with incubation time and exhibited
Michaelis–Menten-type properties. The maximal velocity
(Vmax) and Michaelis constant (Km) values for the substrates
D-Asp, NMDA, D-Glu, D-Phe, and D-His were determined
using different final concentrations (2–40, 1–40, 10–80,
2–40, and 2–40 mM, respectively). The Vmax and Km values
were calculated according to the method of Wilkinson
(1961). The molecular activity (kcat) values were calculated
from the Vmax values and the estimated molecular masses of
the recombinant proteins (42,074, 43,631, 40,862, 42,031,
42,081, and 42,038 Da, for N-terminally His-tagged mouse
DDO, human DAO, mouse DAO, R216L variant of mouse
DDO, R237Y variant of mouse DDO, and R216L-R237Y
variant of mouse DDO, respectively).
Results and discussion
Computational modeling and structural comparison
of DAO/substrate and DDO/substrate complexes
To determine the preferred substrate for mammalian DAO,
the enzymatic activity of purified recombinant human and
123

470
M. Katane et al.
mouse DAO against various amino acids was determined.
Human and mouse DAO displayed deamination activity
against several neutral and basic D-amino acids. In both
enzymes, the highest activity was detected when using
D-Phe as the substrate (Table 1). None of the recombinant
proteins demonstrated activity against any protein L-amino
acid or N-methyl-L-Asp (data not shown). The 3D X-ray
crystallographic structures of porcine and human DAO are
known (Kawazoe et al. 2006, 2007; Mattevi et al. 1996;
Miura et al. 1997; Mizutani et al. 1996; Todone et al.
1997). In those structures complexed with the DAO sub-
strate analogue o-aminobenzoate, the carboxyl group of
o-aminobenzoate interacts with the side-chain guanidino
group of Arg-283 and with the side-chain hydroxyl group
of Tyr-228, while the amino group of o-aminobenzoate
interacts with the backbone carbonyl of Gly-313 (Fig. 1a)
(Kawazoe et al. 2007; Miura et al. 1997). Importantly, a
loop formed by residues 216–228 has been postulated to
act as an ‘‘active-site lid’’ that opens and closes upon
substrate/product migration in and out of the active site
(Todone et al. 1997). The side chain of Tyr-224 within the
‘‘active-site lid’’ covers the bound ligand (Fig. 1a). This
residue is presumably important in the hydrophobic envi-
ronment of the active site in the closed conformation
(Mattevi et al. 1996; Miura et al. 1997; Mizutani et al.
1996).
was modeled using the Lamarckian Genetic Algorithm and
AutoDock 4.0 software (Fig. 1b) (Huey et al. 2004, 2007).
D-Phe was used as the bound ligand, since recombinant
human DAO showed the highest activity against this
D-amino acid (Table 1). The 3D X-ray crystallographic
structure of human DAO (Protein Data Bank ID: 2E48)
(Kawazoe et al. 2007) was used as the docking protein, and
a theoretical model of mouse DAO complexed with D-Phe
was constructed (Fig. 1c). Since the 3D structure of mouse
DAO has not been determined, a structural model for
mouse DAO was proposed using the homology modeling
method and SWISS-MODEL on the ExPASy web server
(http://swissmodel.expasy.org/) (Arnold et al. 2006; Peitsch
1995). The 3D X-ray crystallographic structures of porcine
DAO (Protein Data Bank ID: 1KIF; 1DDO) (Mattevi et al.
1996; Todone et al. 1997) were used as template structures.
A theoretical model of the mouse DAO/D-Phe complex was
subsequently determined using AutoDock 4.0 software
(Huey et al. 2004, 2007).
In both the human DAO/D-Phe and mouse DAO/D-Phe
complex models (Fig. 1b, c), the orientation of the D-Phe
a-carboxyl and a-amino groups in the active sites was
similar to that of o-aminobenzoate bound to human
DAO. In addition, the side chain of D-Phe was oriented
toward a space where the side chain of a Leu residue
was located (Leu-215 or Leu-213 of human or mouse
DAO, respectively). These modes are consistent with
those in previously reported structural models of the
porcine DAO/D-Leu complex (Miura et al. 1997;
Setoyama et al. 2006).
To gain insight into the binding of a substrate D-amino
acid to mammalian DAO, the human DAO/D-Phe complex
To investigate the binding of a substrate D-amino acid to
mammalian DDO, a theoretical model of mouse DDO
complexed with D-Asp was constructed (Fig. 1d). Using a
mouse DDO structure previously proposed by the authors
(Katane et al. 2007a), the annealing docking of D-Asp at the
mouse DDO active site was simulated using AutoDock 4.0
software (Huey et al. 2004, 2007). The orientation of the a-
carboxyl and a-amino groups of D-Asp in the active site
was similar to that of o-aminobenzoate and/or D-Phe bound
to human and mouse DAOs (Fig. 1d). In addition, the side
chain (b-carboxyl group) of D-Asp was oriented toward a
space where the side chain of Arg-216 was located. These
binding modes are consistent with those in the previously
reported structural model of the bovine DDO/D-Asp com-
plex (Sacchi et al. 2002). Thus, Arg-216 of mouse DDO is
structurally equivalent to Leu-215 or Leu-213 of human or
mouse DAO, respectively, and Arg-216 likely interacts
with the b-carboxyl group of an acidic D-amino acid.
Notably, the loop structure corresponding to the ‘‘active-
site lid’’ of mammalian DAOs was missing in the proposed
structure of mouse DDO, and Arg-237 of mouse DDO was
structurally equivalent to Tyr-224 or Tyr-222 of human or
mouse DAO, respectively.
Table 1 Enzymatic activity of purified recombinant human and
mouse DAO against various D-amino acids
Substrates
Specific activity (U/mg protein)
Human DAO
Mouse DAO
D-Asn
D-Gln
D-Ala
D-Phe
D-Met
D-Val
D-Leu
D-Ser
D-Thr
D-Trp
D-Cys
D-Arg
D-Lys
D-His
1.01 0.08
ND
0.069 0.013
ND
5.13 0.46
11.7 0.7
0.83 0.13
2.00 0.21
4.50 0.31
1.49 0.23
0.81 0.03
5.89 0.29
0.39 0.06
0.30 0.05
ND
3.60 0.74
4.19 0.28
0.16 0.01
1.98 0.37
1.42 0.57
0.32 0.04
0.085 0.015
1.51 0.22
0.12 0.03
0.083 0.019
ND
0.74 0.05
0.23 0.05
Data are shown as the mean standard deviation of three to four
independent assays
ND not detected, 1 U = 1 lmol substrate oxidized per min
123

Role of the active site residues arginine-216 and arginine-237
471
A
B
Tyr-228
Tyr-228
Tyr-224
Tyr-224
Leu-215
Leu-215
Arg-283
Arg-283
o-Aminobenzoate
D-Phe
Gly-313
Gly-313
Arg-237
FAD
FAD
C
D
Tyr-223
Tyr-226
Leu-213
Tyr-222
Arg-216
Arg-278
Arg-281
D-Asp
Ser-308
D-Phe
FAD
Gly-311
FAD
Fig. 1 Structural mapping and comparison of the key catalytic
residues in the active sites of mammalian DAO and DDO. The
experimentally determined structure of human DAO complexed with
o-aminobenzoate (Protein Data Bank ID: 2E48) (Kawazoe et al.
2007) is shown in a. The proposed structural models of human DAO
complexed with D-Phe, mouse DAO complexed with D-Phe, and
mouse DDO complexed with D-Asp are shown in b, c, and d,
respectively. In these models, the simulated annealing docking of D-
Phe or D-Asp at the active site was performed using AutoDock 4.0
software (Huey et al. 2004, 2007) (see text for details). FAD
molecules are shown in yellow, while bound ligands and side chains
of amino acid residues that have been demonstrated, or are believed,
to be catalytically important are colored according to the atom type:
green carbon, blue nitrogen, red oxygen
Structure-based design of mouse DDO variants
or mouse DAO, or for D-Asp bound to wild-type mouse
DDO (Fig. 2b). The theoretical models were also obtained
with two corresponding single-point mutants (R216L and
R237Y variants) and D-Phe as the bound ligand. In the
active site of the R216L variant (Fig. 2c), D-Phe docks in
the correct orientation, similar to that observed for D-Phe in
human or mouse DAO, for D-Asp in wild-type mouse
DDO, and for D-Phe in the R216L-R237Y variant. In
contrast, D-Phe is docked in an aberrant orientation in the
R237Y variant (Fig. 2d), similar to the wild-type mouse
DDO/D-Phe complex. Collectively, these models predict
that Arg-216 in mouse DDO plays crucial role(s) in its
substrate specificity, and that this residue is more critical
than Arg-237 for substrate recognition by mouse DDO.
Simulated annealing docking analysis was subsequently
used to analyze the roles of two Arg residues of mouse
DDO (Arg-216 and Arg-237) in the catalytic process,
particularly with respect to substrate specificity. D-Phe was
used as the bound ligand and the model with the lowest
energy conformation contained this amino acid in an
aberrant orientation compared with those observed for
human DAO/D-Phe, mouse DAO/D-Phe, or mouse DDO/
D-Asp complexes (Fig. 2a). This aberrant orientation is
consistent with the observation that the mouse DDO has
very low, or undetectable, activity against several neutral
and basic D-amino acids.
The mutant mouse DDO carrying R216L and R237Y
substitutions (R216L-R237Y variant) used in this study
was designed so that its active site was similar to that of
mammalian DAO, and so that D-Phe could be bound in the
correct orientation at the active site. A theoretical model of
the R216L-R237Y variant with D-Phe as the ligand dem-
onstrated that this amino acid docks in the correct orien-
tation, similar to that observed for D-Phe bound to human
Substrate specificity of the mouse DDO variants
In this section, the substrate specificity of the purified
recombinant proteins (mouse DDO and its R216L-R237Y,
R216L and R237Y variants) was determined using various
D-amino acids and their enantiomers as substrates
(Table 2). Consistent with our previous report (Katane
123

472
M. Katane et al.
Fig. 2 Structural comparison of
the orientation of D-Phe bound
to the active sites of mouse
DDO and its variant derivatives.
The theoretical models of the
simulated annealing docking of
D-Phe at the active sites of
mouse DDO, its R216L-R237Y,
R216L, and R237Y variants are
shown in a, b, c, and d,
A
B
Tyr-237
Arg-237
Tyr-223
Tyr-223
Arg-216
Leu-216
Arg-278
Arg-278
respectively (see text for
details). The color scheme is the
same as in Fig. 1
D-Phe
FAD
D-Phe
FAD
Ser-308
Ser-308
C
D
Tyr-237
Arg-237
Tyr-223
Tyr-223
Leu-216
Arg-216
Arg-278
Arg-278
D-Phe
FAD
D-Phe
FAD
Ser-308
Ser-308
et al. 2007a, 2008), wild-type mouse DDO displayed
similar activity levels against D-Asp and NMDA, and rel-
atively low activity levels against D-Glu (Table 2). No
activity was detected against any of the neutral or basic
D-amino acids examined, with the exception of D-Asn.
The R216L-R237Y variant displayed lower activity
against D-Asp and NMDA and higher activity against
D-Asn than the wild-type enzyme, and no activity against
D-Glu (Table 2). Importantly, this variant acquired an
expanded substrate specificity, displaying activity against
D-Phe, D-Trp, D-Cys, and D-His. The activity against D-His
was significantly higher than that of human or mouse DAO,
while activity against D-Phe, D-Trp, and D-Cys was
relatively low, or equivalent, to that of mouse DAO
(Tables 1, 2).
The R216L variant showed relatively low levels of
activity against D-Asp and NMDA compared with the wild-
type enzyme, while no activity was detected against D-Glu
or D-Asn (Table 2). Interestingly, this variant also showed
activity against D-Phe, D-Trp, D-Cys, and D-His, although
the activity level was lower than that of the R216L-R237Y
variant. The R216L-R237Y and R216L variants showed no
Table 2 Enzymatic activity of purified recombinant mouse DDO and its variant derivatives against various D-amino acids
Substrates
Specific activity (U/mg protein)
Wild-type mouse DDO
R216L-R237Y variant
R216L variant
R237Y variant
D-Asp
NMDA
D-Glu
D-Asn
D-Phe
D-Trp
D-Cys
D-His
12.2 0.9
13.1 0.6
0.46 0.07
0.16 0.02
ND
0.36 0.03
0.77 0.11
ND
0.59 0.03
0.69 0.05
ND
1.42 0.09
1.75 0.11
0.13 0.03
ND
0.43 0.04
1.37 0.13
0.77 0.08
0.15 0.09
2.03 0.26
ND
0.37 0.03
0.098 0.024
0.023 0.002
0.29 0.03
ND
ND
ND
ND
ND
ND
ND
Data are shown as the mean standard deviation of three independent assays. None of the recombinant proteins showed any activity against
D-Gln, D-Ala, D-Met, D-Val, D-Leu, D-Ser, D-Thr, D-Arg and D-Lys
ND not detected, 1 U = 1 lmol substrate oxidized per min
123

Role of the active site residues arginine-216 and arginine-237
473
activity against neutral or basic D-amino acids other than
D-Phe, D-Trp, D-Cys, and D-His (data not shown). The
R237Y variant displayed activity against D-Asp, NMDA,
and D-Glu, but at a significantly lower level than that of the
wild-type enzyme (Table 2). In contrast to the R216L-
R237Y and R216L variants, this variant did not display
detectable activity against any of the neutral or basic
D-amino acids tested. None of the recombinant proteins
showed any activity against protein L-amino acids or
N-methyl-L-Asp (data not shown).
Table 3 Apparent steady-state kinetic parameters of purified
recombinant proteins using several acidic D-amino acids as substrates
Enzymes and
substrates
kcat (s-1
)
Km (mM)
kcat/Km
(s-1 M-1
)
Wild-type mouse DDO
D-Asp
NMDA
D-Glu
12.2 0.3
13.0 0.4
1.60 0.11
8.85 0.60
9.73 0.94
79.4 9.6
1,378
1,331
20
R216L-R237Y variant
D-Asp
0.53 0.02
83.7 3.3
71.7 3.0
6
Taken together, these results indicate that Arg-216 plays
critical role(s) in the substrate specificity of mouse DDO,
and are consistent with the simulated annealing docking
results described above. In addition, and consistent with
our previous study (Katane et al. 2007a), Arg-216 and Arg-
237 act as key functional residues for its full enzymatic
activity.
NMDA
0.86 0.03
12
R216L variant
D-Asp
0.91 0.02
1.00 0.02
22.9 0.9
19.1 0.8
40
52
NMDA
R237Y variant
D-Asp
1.49 0.05
1.85 0.07
0.36 0.01
8.90 0.86
7.00 0.87
80.5 3.7
167
265
4
NMDA
D-Glu
Characterization of the kinetic properties of the mouse
DDO variants
The apparent kcat and Km values for D-Asp, NMDA, or D-Glu were
determined as described in ‘‘Materials and methods’’ and are listed
with the standard deviation
The apparent kinetic parameters (kcat and Km) of the mouse
DDO variants were determined with D-Asp, NMDA and
D-Glu as substrates (Table 3), and were compared with
those of wild-type mouse DDO. Wild-type mouse DDO
showed similar levels of catalytic efficiency (kcat/Km) for
D-Asp and NMDA, but a relatively low efficiency for
D-Glu. The kcat values for the R216L-R237Y and R216L
variants against D-Asp and NMDA were significantly lower
than those of the wild-type enzyme, but the Km values were
higher (Table 3). Therefore, the catalytic efficiency of
these variants for D-Asp and NMDA was markedly lower
than that of the wild-type enzyme. However, the kcat values
for the R237Y variant against D-Asp, NMDA, and D-Glu
were significantly lower than those of the wild-type
enzyme (Table 3), although the Km values for this variant
against D-Asp, NMDA, or D-Glu were comparable to, or
slightly lower than, those of the wild-type enzyme, sug-
gesting that this variant and the wild-type enzyme have
relatively similar affinities for acidic D-amino acids.
mouse DAO, while this variant was less efficient at D-His
deamination than human DAO.
Taken together, the results of the kinetic analysis
described above suggest that Arg-216 of mouse DDO plays
crucial role(s) in both substrate specificity and catalytic
activity, while Arg-237 appears to be only marginally
involved in the binding affinity for acidic D-amino acids, but
is important for catalytic activity. However, the Km values of
the R216L-R237Y variant for D-Asp and NMDA were sig-
nificantly higher than those of the R216L variant (Table 3),
suggesting that the introduction of the R237Y substitution
into the R216L variant reduces its binding affinity for acidic
D-amino acids. In addition, the Km values of the R216L-
R237Y variant for D-Phe and D-His were significantly lower
than those of the R216L variant (Table 4). Thus, the intro-
duction of the R237Y substitution affects the substrate-
binding affinity of the R216L variant more appreciably than
that of the wild-type enzyme. It is conceivable that the side-
chain guanidino group of Arg-216 interacts with the b-car-
boxyl group of acidic D-amino acids and is essential for the
correct substrate orientation within the active site (Fig. 1d).
In the R237Y variant, the Arg-216 residue presumably plays
a critical role in substrate recognition, and the R237Y sub-
stitution may not significantly affect the enzyme interaction
with acidic D-amino acids, while in the R216L-R237Y
variant, in which the side-chain guanidino group at position
216 is missing and therefore the strict specificity for acidic
D-amino acids is decreased, introduction of the R237Y
substitution presumably reduces the substrate-binding
affinity.
Next, the kcat and Km values for the mouse DDO variants
were determined using D-Phe and D-His as the substrates
and compared with those of human and mouse DAOs
(Table 4). The catalytic efficiency (kcat/Km) of the R216L-
R237Y variant for D-Phe was relatively lower than that of
the human or mouse DAO. In contrast, the catalytic effi-
ciency of this variant for D-His (447 s-1 M-1) was about
2.8 or 45 times higher than that of human or mouse DAO
(162 and 10 s-1 M-1), respectively. Similarly to the
R216L-R237Y variant, the catalytic efficiency of the
R216L variant for D-Phe was lower than that of the human
or mouse DAO (Table 4), and the catalytic efficiency for
D-His (22 s-1 M-1) was about 2.2 times higher than that of
123

474
M. Katane et al.
meso-tartrate more efficiently than that of the wild-type
enzyme. In addition, the activity was also inhibited by
10 mM benzoate (a mammalian DAO inhibitor) suggesting
the involvement of Arg-237 in substrate recognition. To
further investigate the role of Arg-237 in substrate speci-
ficity, the effects of malonate and crotonate on the enzy-
matic activity of the R237Y variant were compared with
those of the wild-type enzyme. Malonate and crotonate
reportedly inhibit mammalian DDO and DAO, respec-
tively, by substrate competition (Dixon and Kenworthy
1967; Fonda and Anderson 1968). The 50% inhibitory
concentration (IC50) values of wild-type mouse DDO and
the R237Y variant with meso-tartrate, malonate, benzoate,
and crotonate as inhibitors are listed in Table 5. The
R237Y variant was more sensitive to meso-tartrate, ben-
zoate and crotonate, while its sensitivity to malonate was
comparable to that of the wild-type enzyme. These results
indicate that Arg-237 is possibly involved in substrate
recognition by mouse DDO.
Table 4 Apparent steady-state kinetic parameters of purified
recombinant proteins with D-Phe and D-His as substrates
Enzymes and
substrates
kcat (s-1
)
Km (mM)
kcat/Km
(s-1 M-1
)
Human DAOa
D-Phe
9.89 0.22
0.68 0.02
2.59 0.21
4.18 0.54
3,823
162
D-His
Mouse DAO
D-Phe
5.42 0.45
0.46 0.01
24.2 4.1
45.8 1.0
224
10
D-His
R216L-R237Y variant
D-Phe
D-His
1.70 0.05
13.5 0.8
126
447
1.50 0.01
3.36 0.10
R216L variant
D-Phe
0.75 0.05
0.36 0.01
44.4 4.6
16.4 1.5
17
22
D-His
The apparent kcat and Km values for D-Phe and D-His were determined
as described in the ‘‘Materials and methods’’ and are listed with the
standard deviation
The present study demonstrates that the Arg-216 residue
plays crucial role(s) in the substrate specificity of mouse
DDO. Both Arg-216 and Arg-237 are conserved in all the
mammalian DDO that has been cloned to date. The struc-
turally equivalent residue to Arg-216 in mammalian DAOs is
Leu or Ile. The simulated annealing docking results of D-Asp
in the wild-type mouse DDO active site suggest that the
side-chain guanidino group of Arg-216 interacts with the
b-carboxyl group of acidic D-amino acids, and is essential for
the correct substrate orientation in the active site (Fig. 1d). It
is proposed that D-Phe can bind to the mouse DDO active site
in the correct orientation in case a side-chain guanidino
group is missing at position 216 (Fig. 2). With these results
in mind, an R216L substitution was made in mouse DDO that
permitted the modification of its substrate specificity. The
Arg-216 residue of mouse DDO may function as an essential
scaffold for substrate recognition and binding, and therefore
is catalytically important for the full enzymatic activity
against acidic D-amino acids. This is consistent with a recent
report in which the short stretch region (Ile-215–Asn-225) of
pig DAO was substituted with the corresponding human
DDO short stretch region (Arg-216–Gly-220) (Setoyama
et al. 2006). This pig DAO variant acquired activity against
D-Asp, together with considerably decreased normal activity
against D-Ala and D-Arg.
a
Katane et al. (2010)
Characterization of the sensitivity of the mouse DDO
variants to DDO and DAO inhibitors
To further study the role played by Arg-216 and Arg-237 in
the substrate specificity of mouse DDO, the effects of
meso-tartrate and benzoate on the enzymatic activities of
the wild-type and variant proteins were examined. Ben-
zoate and meso-tartrate reportedly inhibit mammalian
DAO and DDO, respectively, by competing with their
`
substrates (De Marco and Crifo 1967; Klein 1960). As
expected, meso-tartrate dose-dependently inhibited the
activity of wild-type mouse DDO against D-Asp, while
benzoate had no effect, even at a concentration of 10 mM
(Fig. 3a). In contrast, and as expected, benzoate dose-
dependently inhibited the activity of both human and
mouse DAO against D-Phe, while meso-tartrate had no
effect, even at a concentration of 10 mM (Fig. 3b, c).
Figure 3d–g indicate that the activity of the R216L-
R237Y and R216L variants against D-Asp and D-Phe was
dose-dependently inhibited by meso-tartrate. Furthermore,
the activity of these variants against D-Asp and D-Phe was
also dose-dependently inhibited by benzoate, although the
sensitivity was significantly less than that of human or
mouse DAO. Thus, benzoate sensitivity was increased in
the R216L-R237Y and R216L variants compared to the
wild-type enzyme, consistent with the expanded substrate
specificities of the R216L-R237Y and R216L variants for
several neutral and basic D-amino acids (Table 2).
The present study also suggests that Arg-237 is cata-
lytically important for the full enzymatic activity of mouse
DDO against acidic D-amino acids, but is only marginally
involved in the substrate-binding affinity. Nevertheless, the
introduction of the R237Y substitution into mouse DDO
made it more sensitive to several DDO and DAO inhibitors
(Fig. 3; Table 5). The additional introduction of the R237Y
substitution into the R216L variant increased its catalytic
efficiency for D-Phe and D-His (Table 4), and decreased its
As shown in Fig. 3h, the activity of the R237Y
variant against D-Asp was dose-dependently inhibited by
123

Role of the active site residues arginine-216 and arginine-237
475
meso-Tartrate
meso-Tartrate
meso-Tartrate
meso-Tartrate
A
B
D
C
Benzoate
Benzoate
Benzoate
Benzoate
Wild-type mouse DDO
(D-Asp)
Human DAO
(D-Phe)
Mouse DAO
(D-Phe)
R216L-R237Y variant
(D-Asp)
140
120
100
80
140
120
100
80
140
120
100
80
140
120
100
80
*
***
***
60
60
60
60
***
***
40
40
40
40
20
20
20
20
***
***
***
0
0
0
0
0
1
10
0
1
10
0
1
10
0
1
10
[ Inhibitor ] (mM)
[ Inhibitor ] (mM)
[ Inhibitor ] (mM)
[ Inhibitor ] (mM)
meso-Tartrate
meso-Tartrate
meso-Tartrate
meso-Tartrate
H
E
F
G
Benzoate
Benzoate
Benzoate
Benzoate
R216L variant
(D-Asp)
R237Y variant
(D-Asp)
R216L variant
(D-Phe)
R216L-R237Y variant
(D-Phe)
140
120
100
80
140
120
100
80
140
120
100
80
140
120
100
80
*
***
**
**
*
***
60
60
60
60
***
***
40
40
40
40
***
20
20
20
20
0
0
0
0
0
1
10
0
1
10
0
1
10
0
1
10
[ Inhibitor ] (mM)
[ Inhibitor ] (mM)
[ Inhibitor ] (mM)
[ Inhibitor ] (mM)
Fig. 3 Effects of inhibitors on the enzymatic activities of recombi-
nant mouse DDO and its variant derivatives. Enzymatic activities
were assayed using recombinant mouse DDO (a), recombinant human
DAO (b), recombinant mouse DAO (c), the R216L-R237Y variant of
mouse DDO (d and e), the R216L variant of mouse DDO (f and g),
and the R237Y variant of mouse DDO (h), in the presence of the
indicated concentrations of meso-tartrate (open bar) and benzoate
(gray bar). The substrates used were: D-Asp for a, d, f and h; and
D-Phe for b, c, e and g. The activities of the recombinant proteins in
the presence of the inhibitors are shown as a percentage of their
activities in the absence of inhibitors. Data represent the
mean standard deviation of three to four independent assays.
*p \ 0.05; **p \ 0.01; ***p \ 0.001 (Student’s
activity in the absence of inhibitor
t test) versus
Table 5 Comparison of IC50 values of purified recombinant mouse
DDO and the R237Y variant with several monocarboxylic and
dicarboxylic acids as inhibitors
role(s) in catalysis remains undefined. Additional studies to
elucidate the crystal structure of wild-type mouse DDO and
its variants are certainly needed.
Inhibitors
IC50 (mM)
Acknowledgments The authors would like to thank Prof. Ryuichi
Konno (International University of Health and Welfare) for his kind
gift of the mouse DAO expression plasmid, and Mr. Yutaka Tanig-
uchi and Ms. Mayumi Tetsuka, Midori Nitta, and Suno Masakawa for
their technical assistance. This work was supported in part by a
Grant-in-Aid for Scientific Research (21590071) from the Ministry of
Education, Culture, Sports, Science, and Technology of Japan, as well
as by a Kitasato University Research Grant for Young Researchers (to
M.K.).
Wild-type mouse DDO
R237Y variant
meso-Tartrate
Malonate
8.99 0.41
5.31 0.25
69.5 6.8
25.3 2.3
6.39 0.11***
5.60 0.14
Benzoate
43.3 6.6**
19.7 1.5*
Crotonate
Data are shown as the mean standard deviation of three indepen-
dent assays. The substrate used was D-Asp
* p \ 0.05; ** p \ 0.01; *** p \ 0.001 (Student’s t test) versus
wild-type mouse DDO
References
catalytic efficiency for D-Asp and NMDA (Table 3). Thus,
the Arg-237 residue in mouse DDO is likely to play sub-
sidiary role(s) in substrate specificity, although its precise
Arnold K, Bordoli L, Kopp J, Schwede T (2006) The SWISS-
MODEL workspace: a web-based environment for protein
structure homology modeling. Bioinformatics 22:195–201
123

476
M. Katane et al.
D’Aniello A (2007) D-Aspartic acid: an endogenous amino acid with
an important neuroendocrine role. Brain Res Rev 53:215–234
D’Aniello G, Ronsini S, Guida F, Spinelli P, D’Aniello A (2005)
Occurrence of D-aspartic acid in human seminal plasma and
spermatozoa: possible role in reproduction. Fertil Steril
84:1444–1449
D’Aniello G, Grieco N, Di Filippo MA, Cappiello F, Topo E,
D’Aniello E, Ronsini S (2007) Reproductive implication of D-
aspartic acid in human pre-ovulatory follicular fluid. Hum
Reprod 22:3178–3183
Klein JR (1960) Competitive inhibition of D-amino acid oxidase by
benzoate as a function of substrate. Biochim Biophys Acta
37:534–537
Matsui T, Sekiguchi M, Hashimoto A, Tomita U, Nishikawa T, Wada
K (1995) Functional comparison of D-serine and glycine in
rodents: the effect on cloned NMDA receptors and the extracel-
lular concentration. J Neurochem 65:454–458
Mattevi A, Vanoni MA, Todone F, Rizzi M, Teplyakov A, Coda A,
Bolognesi M, Curti B (1996) Crystal structure of D-amino acid
oxidase: a case of active site mirror-image convergent evolution
with flavocytochrome b2. Proc Natl Acad Sci USA 93:7496–
7501
Miura R, Setoyama C, Nishina Y, Shiga K, Mizutani H, Miyahara I,
Hirotsu K (1997) Structural and mechanistic studies on D-amino
acid oxidase-substrate complex: implications of the crystal
structure of enzyme-substrate analog complex. J Biochem
122:825–833
`
De Marco C, Crifo C (1967) D-Aspartate oxidase from pig kidney. III.
Competitive inhibition by dicarboxylic hydroxyacids. Enzymo-
logia 33:325–330
Dixon M, Kenworthy P (1967) D-Aspartate oxidase of kidney.
Biochim Biophys Acta 146:54–76
Fonda ML, Anderson BM (1968) D-Amino acid oxidase. II. Studies of
substrate-competitive inhibitors. J Biol Chem 243:1931–1935
Homma H (2007) Biochemistry of D-aspartate in mammalian cells.
Amino Acids 32:3–11
Mizutani H, Miyahara I, Hirotsu K, Nishina Y, Shiga K, Setoyama C,
Miura R (1996) Three-dimensional structure of porcine kidney
˚
D-amino acid oxidase at 3.0 A resolution. J Biochem 120:14–17
Huey R, Goodsell DS, Morris GM, Olson AJ (2004) Grid-based
hydrogen bond potentials with improved directionality. Lett
Drug Des Discov 1:178–183
Huey R, Morris GM, Olson AJ, Goodsell DS (2007) A semiempirical
free energy force field with charge-based desolvation. J Comput
Chem 28:1145–1152
Katane M, Homma H (2010) D-Aspartate oxidase: the sole catabolic
enzyme acting on free D-aspartate in mammals. Chem Biodivers.
doi:10.1002/cbdv.200900250
Mothet JP, Parent AT, Wolosker H, Brady RO Jr, Linden DJ, Ferris
CD, Rogawski MA, Snyder SH (2000) D-Serine is an endoge-
nous ligand for the glycine site of the N-methyl-D-aspartate
receptor. Proc Natl Acad Sci USA 97:4926–4931
Nishikawa T (2005) Metabolism and functional roles of endogenous
D-serine in mammalian brains. Biol Pharm Bull 28:1561–1565
Peitsch MC (1995) Protein modeling by E-mail. Biotechnology (N Y)
13:658–660
Katane M, Furuchi T, Sekine M, Homma H (2007a) Molecular
cloning of a cDNA encoding mouse D-aspartate oxidase and
functional characterization of its recombinant proteins by site-
directed mutagenesis. Amino Acids 32:69–78
Katane M, Seida Y, Sekine M, Furuchi T, Homma H (2007b)
Caenorhabditis elegans has two genes encoding functional
D-aspartate oxidases. FEBS J 274:137–149
Katane M, Hanai T, Furuchi T, Sekine M, Homma H (2008)
Hyperactive mutants of mouse D-aspartate oxidase: mutagenesis
of the active site residue serine 308. Amino Acids 35:75–82
Katane M, Saitoh Y, Seida Y, Sekine M, Furuchi T, Homma H (2010)
Comparative characterization of three D-aspartate oxidases and
one D-amino acid oxidase from Caenorhabditis elegans. Chem
Biodivers. doi:10.1002/cbdv.200900294
Pilone MS (2000) D-Amino acid oxidase: new findings. Cell Mol Life
Sci 57:1732–1747
Pollegioni L, Piubelli L, Sacchi S, Pilone MS, Molla G (2007)
Physiological functions of D-amino acid oxidases: from yeast to
humans. Cell Mol Life Sci 64:1373–1394
Sacchi S, Lorenzi S, Molla G, Pilone MS, Rossetti C, Pollegioni L
(2002) Engineering the substrate specificity of D-amino-acid
oxidase. J Biol Chem 277:27510–27516
Scolari MJ, Acosta GB (2007) D-Serine: a new word in the
glutamatergic neuro-glial language. Amino Acids 33:563–574
Setoyama C, Nishina Y, Mizutani H, Miyahara I, Hirotsu K, Kamiya
N, Shiga K, Miura R (2006) Engineering the substrate specificity
of porcine kidney D-amino acid oxidase by mutagenesis of the
‘‘active-site lid’’. J Biochem 139:873–879
Kawazoe T, Tsuge H, Pilone MS, Fukui K (2006) Crystal structure of
human D-amino acid oxidase: context-dependent variability of
the backbone conformation of the VAAGL hydrophobic stretch
located at the si-face of the flavin ring. Protein Sci 15:2708–2717
Kawazoe T, Tsuge H, Imagawa T, Aki K, Kuramitsu S, Fukui K
(2007) Structural basis of D-DOPA oxidation by D-amino acid
oxidase: alternative pathway for dopamine biosynthesis. Bio-
chem Biophys Res Commun 355:385–391
Tishkov VI, Khoronenkova SV (2005) D-Amino acid oxidase:
structure, catalytic mechanism, and practical application. Bio-
chemistry (Mosc) 70:40–54
Todone F, Vanoni MA, Mozzarelli A, Bolognesi M, Coda A, Curti B,
Mattevi A (1997) Active site plasticity in D-amino acid oxidase:
a crystallographic analysis. Biochemistry 36:5853–5860
Wilkinson GN (1961) Statistical estimations in enzyme kinetics.
Biochem J 80:324–332
123