Catalytic mechanism of serine racemase from Dictyostelium discoideum

Article abstract of DOI:10.1007/s00726-012-1442-4

The eukaryotic serine racemase from Dictyostelium discoideum is a fold-type II pyridoxal 5′-phosphate (PLP)-dependent enzyme that catalyzes racemization and dehydration of both isomers of serine. In the present study, the catalytic mechanism and role of the active site residues of the enzyme were examined by site-directed mutagenesis. Mutation of the PLP-binding lysine (K56) to alanine abolished both serine racemase and dehydrase activities. Incubation of d- and l-serine with the resultant mutant enzyme, K56A, resulted in the accumulation of PLP-serine external aldimine, while less amounts of pyruvate, α-aminoacrylate, antipodal serine and quinonoid intermediate were formed. An alanine mutation of Ser81 (S81) located on the opposite side of K56 against the PLP plane converted the enzyme from serine racemase to l-serine dehydrase; S81A showed no racemase activity and had significantly reduced d-serine dehydrase activity, but it completely retained its l-serine dehydrase activity. Water molecule(s) at the active site of the S81A mutant enzyme probably drove d-serine dehydration by abstracting the α-hydrogen in d-serine. Our data suggest that the abstraction and addition of α-hydrogen to l- and d-serine are conducted by K56 and S81 at the si- and re-sides, respectively, of PLP.

Full text of DOI:10.1007/s00726-012-1442-4

Amino Acids (2013) 44:1073–1084
DOI 10.1007/s00726-012-1442-4
ORIGINAL ARTICLE
Catalytic mechanism of serine racemase from
Dictyostelium discoideum
•
Tomokazu Ito Motoki Maekawa Shuhei Hayashi
•
•
•
Masaru Goto Hisashi Hemmi Tohru Yoshimura
•
Received: 1 September 2012 / Accepted: 30 November 2012 / Published online: 27 December 2012
Ó Springer-Verlag Wien 2012
Abstract The eukaryotic serine racemase from Dictyoste-
lium discoideum is a fold-type II pyridoxal 50-phosphate
(PLP)-dependent enzyme that catalyzes racemization and
dehydration of both isomers of serine. In the present study,
the catalytic mechanism and role of the active site residues of
the enzyme were examined by site-directed mutagenesis.
Mutation of the PLP-binding lysine (K56) to alanine abol-
ished both serine racemase and dehydrase activities. Incu-
bation of D- and L-serine with the resultant mutant enzyme,
K56A, resulted in the accumulation of PLP-serine external
aldimine, while less amounts of pyruvate, a-aminoacrylate,
antipodal serine and quinonoid intermediate were formed.
An alanine mutation of Ser81 (S81) located on the opposite
side of K56 against the PLP plane converted the enzyme
from serine racemase to L-serine dehydrase; S81A showed no
racemase activity and had significantly reduced D-serine
dehydrase activity, but it completely retained its L-serine
dehydrase activity. Water molecule(s) at the active site of the
S81A mutant enzyme probably drove D-serine dehydration
by abstracting the a-hydrogen in D-serine. Our data suggest
that the abstraction and addition of a-hydrogen to L- and
D-serine are conducted by K56 and S81 at the si- and re-sides,
respectively, of PLP.
Keywords D-serine ꢀ Serine racemase ꢀ Dictyostelium
discoideum ꢀ Pyridoxal 50-phosphate
Abbreviations
SR
Serine racemase
DdSR Serine racemase of Dictyostelium discoideum
SpSR Serine racemase of Schizosaccharomyces pombe
PLP
Pyridoxal 50-phosphate
Introduction
Eukaryotic serine racemase (SR) is a pyridoxal 50-phos-
phate (PLP)-dependent enzyme that catalyzes the racemi-
zation of D- and L-serine (Wolosker et al. 1999). In addition
to serine racemization, SR catalyzes the dehydration of
D- and L-serine to yield pyruvate and ammonia (De Miranda
et al. 2002). SR is widely distributed among eukaryotic
organisms, including, mammals, plants and fungi. The
primary structure of SR shows sequence homology with
enzymes in the serine/threonine dehydratase family, and
SR is classified into the fold-type II group of PLP-depen-
dent enzymes (Yoshimura and Goto 2008). In the mam-
malian brain, SR localizes to astrocytic glia cells and
neurons, and supplies D-serine, which serves as a physio-
logical co-agonist of the N-methyl-D-aspartate (NMDA)
receptor, a key component in glutamatergic neurotrans-
mission (Mothet et al. 2000; Kartvelishvily et al. 2006).
The SR deletion in mouse shows an approximate 80–90 %
decrease in the forebrain D-serine content and showed
impaired learning and memory (Inoue et al. 2008; Mori and
Inoue 2010). In the mammalian brain, SR is thought to serve
as a sole endogenous source of D-serine and as a modulator
of brain functions. Dysfunction of the NMDA receptor
is reportedly associated with various neurodegenerative
T. Ito amd M. Maekawa contributed equally to this work.
T. Ito ꢀ M. Maekawa ꢀ S. Hayashi ꢀ H. Hemmi ꢀ
T. Yoshimura (&)
Department of Applied Molecular Biosciences, Graduate School
of Bioagricultural Sciences, Nagoya University, Frou-chou,
Chikusa, Nagoya, Aichi 464-8601, Japan
e-mail: yosimura@nuagr1.agr.nagoya-u.ac.jp
M. Goto
Department of Biomolecular Science, Faculty of Science, Toho
University, Miyama 2-2-1, Funabashi, Chiba 274-8510, Japan
123

1074
T. Ito et al.
disorders, such as stroke, schizophrenia, Alzheimer’s dis-
ease, and amyotrophic lateral sclerosis (Nishikawa 2011).
SR is regarded as a pharmacological target for these dis-
eases. Elucidation of the structure–function relationship of
SR would be valuable for developing mechanism-based
inactivators or activators of the enzyme.
We previously cloned, purified, and examined the
properties of an SR from the cellular slime mold Dictyos-
telium discoideum (DdSR), which exhibits 38 and 46 %
sequence homology with SpSR and human SR. DdSR is a
dimer in solution and is activated by Na? as well as other
activators such as Mg2? and Mg2?ATP (Ito et al. 2012a).
This unique stimulation by Na? is probably caused by its
binding to the divalent cation-binding site because muta-
tion of the divalent metal ion coordinating residues E207
and D213 abolished both Na? and Mg2? stimulation.
In the present study, we examined the reaction mecha-
nism and roles of active site residues of DdSR by substi-
tution with alanine, serine, and/or cysteine by site-directed
mutagenesis. The K56 residue of DdSR, which corresponds
to K57 of SpSR, anchors PLP and is involved in a-proton
shuttling on the si-face of the PLP plane. The S81 mutation
located on the re-face of PLP (the opposite side of K56
with respect to the PLP plane) completely lost racemase
activity, but its L-serine dehydrase activity was not affec-
ted. These data suggest that the serine racemization cata-
lyzed by DdSR proceeds through a two-base mechanism
that involves K56 and S81.
To date, crystal structures of a few SRs have been
solved. Goto et al. first solved the crystal structures of four
kinds of Schizosaccharomyces pombe SR (SpSR)—wild-
type enzyme (WT), WT bound with an ATP analog,
modified enzyme, and the modified enzyme with serine
(Goto et al. 2009; Yamauchi et al. 2009). More recently,
Smith et al. (2010) published the structures of human and
rat enzymes in the presence or absence of malonate, which
is an orthosteric inhibitor. These structures revealed that
SRs have an overall fold typical of the b-family of PLP
enzymes that consists of two domains—a large and a small
flexible domain. The cofactor PLP lies in a groove at the
domain interface and is linked covalently to a lysine resi-
due (K57 in SpSR and K56 in the human enzyme) to form
an internal Schiff base.
The reaction mechanism of PLP-dependent amino acid
racemase has been intensively studied with the alanine
racemase of a thermopile, Geobacillus stearothermophilus.
The alanine racemase belongs to the fold-type III PLP
enzyme family, and its primary and three-dimensional
structures are distinctly different from those of eukaryotic
SRs (Yoshimura and Goto 2008; Shaw et al. 1997). The
alanine racemase reaction proceeds through a 2-base
mechanism, with Lys39 bound to PLP and Tyr2650 lying
on the side opposite to Lys39, against the PLP plane. Lys39
abstracts the a-hydrogen of D-alanine resulting in the pro-
duction of an anionic intermediate that is stabilized by
PLP. Re-protonation of the a-carbon occurs via Tyr2650
on the side of the PLP plane opposite to the side where the
a-hydrogen abstraction takes place, and L-alanine is pro-
duced. The reaction from L-alanine to D-alanine proceeds
through the reverse process—the a-hydrogen of L-alanine
is abstracted by Tyr2650, and Lys39 donates a proton on the
opposite side of PLP to produce D-alanine (Watanabe et al.
2002). Although the eukaryotic SR is structurally different
from the alanine racemase, Goto et al. postulated a similar
two-base mechanism for the enzyme on the basis of wild-
type SpSR models complexed with the substrates L-serine
and D-serine. The side chains of the PLP-binding K57 and
S82 are located at the si- and re-face in the PLP plane
Materials and methods
Materials
PLP, D- and L-serine, nicotinamide adenine dinucleotide
(NADH), isopropyl-b-D-thiogalactoside (IPTG), methanol,
and lactic dehydrogenase from pig heart were obtained
from Wako Pure Chemicals (Osaka, Japan). ATP was
purchased from Sigma-Aldrich Japan (Tokyo, Japan).
Synthetic oligonucleotide primers were obtained from
Fasmac Inc. (Tsukuba, Japan). All other chemicals were of
the highest grade commercially available.
Construction and purification of DdSR mutants
The point mutation of DdSR was introduced by PCR with
KOD-plus DNA polymerase (Toyobo, Osaka, Japan) using
complementary primers pair containing the desired muta-
tion. The K56, S80, S81, H84, K111, R132, and P150 were
mutated and obtained the K56A, S80A, S80C, S81A,
H84A, K111A, R132A or P150S mutant enzymes. The
following primers and their complementary primers were
used: 50-caggaagttttgcaatgcgtggtgc-30 for K56A, 50-ggtgttg
taactcatgctagtggtaatcatggtcaagc-30 for S80A, 50-ggtgttgtaa
ctcattgtagtggtaatcatggtcaagc-30 for S80C, 50-gttgtaactcatag
tgctggtaatcatggtcaagc-30 for S81A, 50-gtagtggtaatgccggtca
agcattatc-30 for H84A, 50-gcaccatcagttgcattgaatgctatttgtgg-30
for K111A, 50-gtaaagcaacattggaagcagcagaatcaaatactaaac-30 for
R132A and 50-cattcttgtaagttaattcattcattcgataatttacaagttatcgc-30
˚
within a distance of 3.3 and 3.0 A, respectively, from the
a-carbon of the substrate (Goto et al. 2009). In the con-
version from L-serine to D-serine, a proton is probably
abstracted from the a-carbon of the L-serine moiety by the
e-amino group of K57, which is located on the si-face of
the PLP-L-serine intermediate. A proton is then donated by
S82 to the a-carbon of the resultant anionic intermediate.
123

Catalytic mechanism of serine racemase from D. discoideum
1075
for P150S. The previously constructed DdSR expression
plasmid pDdSR was used as a template (Ito et al. 2012a).
Mutation of DNA inserts was verified by DNA sequencing.
Each wild-type and mutant plasmid was introduced into the
E. coli Rosetta 2 (DE3) cells (Novagen, Madison, WI,
USA). The wild-type and the mutant enzymes were over-
expressed with 0.1 mM IPTG, and purified according to the
same method for the wild-type enzyme as described pre-
viously (Ito et al. 2012a).
activity catalyzed by WT, the pyruvate formed was deter-
mined with lactic dehydrogenase and NADH as described
above. The D-serine dehydrase activity catalyzed by S81A
were calculated by the decrease in the amount of the
D-serine, which was determined by the above-mentioned
HPLC method.
Spectroscopic analysis
UV–visible spectra were obtained with a Shimadzu
UV-2450 spectrophotometer (Kyoto, Japan). The spectra
were recorded in 50 mM Tris–HCl buffer (pH 8.5) con-
taining 0.5 mM MgCl2, 0.5 mM ATP and 25 mM D- or
L-serine in the wavelength range from 250 to 550 nm using
1-cm light pass cell at 25 °C.
Fluorescence spectra of the wild-type and mutant
DdSRs (0.22 mg/ml) were measured with a Shimadzu
RF5300PC spectrofluorophotometer in 50 mM Tris–HCl
buffer (pH 8.5) containing 0.5 mM ATP, 0.5 mM MgCl2
and 25 mM D- or L-serine at 25 °C. Excitation was per-
formed at 330 or 420 nm, and the excitation and emission
slit widths were 5 nm. Emission spectra were recorded in
the wavelength ranging from 340 to 540 nm (when
excited at 340 nm) and from 430 to 630 nm (excited at
420 nm). The data were obtained by the subtraction of the
blank (spectra of the sample without enzyme) from
the obtained spectra. In the presence of D- or L-serine, the
spectra were recorded immediately after mixing of the
amino acid.
Assay of enzymatic activity
Serine dehydration catalyzed by the wild type and mutant
enzymes was assayed by monitoring the pyruvate for-
mation with lactate dehydrogenase and NADH by fol-
lowing the decrease in absorbance at 340 nm according
to the previously described protocols (Ito et al. 2008,
2012b). If otherwise the reaction (500 ll) was performed
in a mixture containing 50 mM Tris–HCl buffer (pH 8.5),
20 lM PLP, 1 mM ATP, 1 mM MgCl2, 50 mM substrate,
and DdSR (0.9–7 lM) at 30 °C for 5 min. The reaction
was stopped by boiling for 5 min and the mixture was
centrifuged at 21,0009g for 5 min. Then, 200 ll of the
supernatant was collected and transferred to the 800 ll of
reaction solution consisting of 50 mM Tris–HCl buffer
(pH 8.5), 3 mM NADH and 10 U lactic dehydrogenase,
and incubated at 30 °C for 5 min. Pyruvate formed was
calculated from the decrease in the absorbance at
340 nm.
Serine racemase activity was assayed by measuring
the amount of each antipodal serine formed from D- or
L-serine. The DdSR reaction was performed in a 500 ll
mixture containing 50 mM Tris–HCl (pH 8.5), 20 lM
PLP, 1 mM ATP, 1 mM MgCl2, 1 mM L-homocysteic
acid (L-HCA) as an internal standard, 50 mM substrate,
and DdSR at 30 °C for 5 min. The reaction was stopped
by boiling for 5 min and the mixture was centrifuged at
21,0009g for 5 min. Amino acids were derivatized with
BOC-L-Cys-OPA reagent (t-butoxycarbonyl-L-cysteine
and O-phatalaldehide mixture) to produce diastereomeric
derivatives which were separated by HPLC using a Cos-
mosil 5C18-AR-II column (4.6 9 150 mm, Nacalai tes-
que, Kyoto, Japan) as described previously (Ito et al.
2012a).
Circular dichroism studies
Circular dichroism (CD) spectra were collected with a
JASCO J720WI CD spectrophotometer with a path length
of 0.2 cm. The proteins were dissolved in 50 mM borate–
NaOH buffer (pH 8.5) at a final concentration 5.7 lM. The
scans were performed from 200 to 260 nm at 50 nm per
min at 25 °C. The raw data were converted to molar
ellipticity values based on the molecular weight and the
concentration of DdSRs.
Results
Effect of methanol on the DdSR activity was assayed as
follows. The reactions were performed in a mixture con-
taining 50 mM Tris–HCl (pH 8.5), 20 lM PLP, 1 mM
ATP, 1 mM MgCl2, 1 mM L-HCA, 10 mM substrate and
increasing concentrations (0–40 %) of methanol at 30 °C.
The reactions were stopped by boiling for 5 min. The
racemase activities catalyzed by WT were determined with
HPLC. For the determination of the D-serine dehydrase
Spectroscopic properties of DdSR
We first analyzed the spectroscopic properties of the wild-
type enzyme (WT). Based on our previous finding that
DdSR, similar to other SRs, is stimulated by MgATP (Ito
et al. 2012a), we obtained all data in the presence of
MgATP in Tris–HCl buffer at pH 8.5. In the presence of
MgATP, the wild-type DdSR (WT) gave an absorption
123

1076
T. Ito et al.
Fig. 1 Spectral properties of
DdSRs in the presence of D- or
L-serine. UV–visible spectra of
WT (a, b 0.5 mg/mL), K56A (c,
d 0.77 mg/mL) and S81A (e,
f 0.77 mg/mL). In each panel,
spectrum in the absence of
amino acid is shown as a dotted
line. Each spectrum was
recorded at 25 °C with 50 mM
Tris–HCl (pH 8.5) containing
0.5 mM MgCl2 and 0.5 mM
ATP. Time-dependent spectral
changes caused by the addition
of 25 mM L-serine (a, c, and
e) and 25 mM D-serine (b, d,
and f) are also shown. The 3
spectra in each panel,
(A)
(B)
0.15
0.15
0.1
0.05
0
0.1
0.05
0
270
320
370
420
470
520
270
320
370
420
470
520
Wavelength (nm)
Wavelength (nm)
(C)
(D)
0.2
0.15
0.1
0.2
0.15
0.1
represented as solid lines, are
the spectra at 1, 3, and 5 min
(from bottom to top at 320 nm)
after the addition of substrate
0.05
0
0.05
0
270
320
370
420
470
520
270
320
370
420
470
520
Wavelength (nm)
Wavelength (nm)
(E)
(F)
0.3
0.2
0.1
0
0.3
0.2
0.1
0
270
320
370
420
470
520
270
320
370
420
470
520
Wavelength (nm)
Wavelength (nm)
spectrum with wavelength maxima at 315 and 416 nm
(Fig. 1a, broken line). The absorbance values at 315 and
416 nm are probably attributed to the two forms of the
Lys56-PLP internal aldimine—enolimine (unprotonated
internal aldimine) and ketoenamine (protonated internal
aldimine) tautomers, respectively. Incubation of D- and
L-serine with WT resulted in a rapid decrease in the absor-
bance at 416 nm and a gradual increase in the absorbance
at approximately 310 nm (Fig. 1a, b, solid line). The
absorbance change at 416 nm indicated a decrease in the
levels of the ketoenamine tautomer and/or the external
aldimine. On the other hand, the increase in absorbance at
310 nm was partially due to the accumulation of pyruvate
formed from D- and L-serine. When excited at 420 nm,
WT exhibited an emission fluorescence centered at
495 nm. The peak originated from the ketoenamine tau-
tomer of the internal Schiff base. The addition of 25 mM
D- or L-serine caused a rapid peak shift from 495 to
485 nm, which increased the emission intensity (Fig. 2a).
An increase in the emission peak indicated the formation
of external aldimine (PLP–serine complex) (Vaccari et al.
1996). Each emission peak was decreased in a time-
dependent manner (data not shown), reflecting a decrease
in the ketoenamine form and the external aldimine tau-
tomers, as observed by UV–visible spectrophotometry
(Fig. 1a, b).
123

Catalytic mechanism of serine racemase from D. discoideum
1077
Fig. 2 Emission spectra of WT
(a), untreated K56A (b), S81A
(A) 500
(C)300
(c) and reconstituted K56A
(d) in the absence and presence
of D- or L-serine. Emission
spectra upon excitation at
420 nm of WT (0.1 mg/mL),
S81A (0.1 mg/mL), and K56A
(0.03 mg/mL) were recorded at
25 °C in a 50 mM Tris–HCl
(pH 8.5) containing 0.5 mM
MgCl2 and 0.5 mM ATP. The
spectra were recorded in the
absence of serine (bold line),
presence of 25 mM L-serine
(thin line), and presence of
25 mM D-serine (dotted line).
The spectra of WT and S81A
were recorded immediately after
amino acid mixing. The spectra
of untreated and reconstituted-
K56 were taken 6 min after the
addition of amino acid
400
200
100
0
300
200
100
0
430
480
530
580
630
430
480
530
580
630
Wavelength (nm)
Wavelength (nm)
(B) 600
(D) 180
160
500
400
300
200
100
0
140
120
100
80
60
40
20
0
430
480
530
580
630
430
480
530
580
630
Wavelength (nm)
Wavelength (nm)
Mutation of the active-site residues of DdSR
absorbs at 325 and 388 nm at neutral pH. The 416-nm
absorption peak is possibly derived from the Schiff base
between PLP and a free amino acid, or PLP and some
residue of the enzyme other than K56.
To study the reaction mechanism of DdSR, we mutated one
of each of the seven active-site residues—K56, S80, S81
H84, K111, R132, and P150—that are highly conserved
among SRs and thus obtained K56A, S80A, S80C, S81A,
H84A, K111A, R132A and P150S mutant DdSRs. Each
mutant residue corresponds to the following residues in
SpSR, whose three-dimensional structure has been deter-
mined: K57, S81, S82, H85, K112, R133, and P151. WT
and mutant DdSRs were overexpressed in Escherichia coli
cells and purified according to a previously described
protocol (Ito et al. 2012a).
Extensive dialysis of K56A with dialysis buffer (50 mM
borate–NaOH pH 8.5, 200 mM NaCl, 1 mM hydroxyl-
amine and 1 % glycerol) for 16 h resulted in the enzyme
exhibiting a single peak at 280 nm (Fig. 4, line 2), indi-
cating the apo-enzyme formation. Further dialysis of the
apo-K56A with the buffer containing PLP produced the
holo-K56A (reconstituted K56A). The reconstituted K56A
absorbed at 280 nm and showed a broad absorption band at
approximately 320 and 400 nm (Fig. 4, line 3).
Incubation with D- and L-serine resulted in no absorption
spectral changes with K56A (Fig. 1c, d). An increase in the
absorption at approximately 320 nm would have been
observed if the enzyme had catalyzed the dehydration of
substrate. The absence of the absorption maxima at the 470
and 500 nm indicated the absence of the a-aminoacrylate
and quinonoid intermediates, respectively. When excited at
420 nm, K56A exhibited a strong emission peak at 481 nm
(Fig. 2b). When estimated at 490 nm, the fluorescence
intensity (per lg protein) of K56A was approximately
eightfold higher than that of the WT. Unlike WT, the
addition of D- or L-serine to K56A did not cause immediate
changes in the fluorescence spectra (data not shown).
However, prolonged incubation of D- or L-serine with
K56A led to a gradual increase in the emission intensity
without changing the kmax (Fig. 2b). These observations
Effects of the K56 mutation
Based on the structure of SpSR bound to L-serine, Goto
et al. suggested that the side chain amine nitrogen of the
PLP-binding K57 corresponding to K56 of DdSR is located
˚
within 3.3 A of the substrate Ca, and it probably catalyzes
the abstraction and addition of the a-hydrogen of the
external Schiff base formed from L-serine and PLP (Fig. 3)
(Goto et al. 2009).
Although we expected that the K56A mutant enzyme
would lose its ability to bind to PLP, its absorption spec-
trum exhibited two peaks at approximately 280 and
416 nm (Fig. 1c, broken line and Fig. 4, line 1). The
absorption maximum at 416 nm is characteristic of the
Schiff base, but not of the free PLP aldehyde, which
123

1078
T. Ito et al.
Fig. 3 Active site of modified SpSR with substrate serine (SpSRm).
The active site of modified SpSR (Goto et al. 2009) containing 4
water molecules is depicted as stick model. The Schiff base lysine
residue (K57) is converted to lysino-D-alanyl residue. Distances
between S82-OH and a-carbon of substrate, water molecule and the
a-carbon, and S81-OH and carboxy oxygen atoms of substrate are
indicated. The K57, S81, S82, H85, K112, R133, and P151 of SpSR
correspond to the K56, S80, S81, H84, K111, R132, and P150 of
DdSR, respectively. Structural figure was generated using the
Molscript (Kraulis 1991)
not shown), suggesting that the mutation did not cause any
gross conformational alternations.
0.4
0.3
0.2
In some PLP enzymes, addition of a small primary
amine such as methylamine or ethylamine recovers the
activity of the mutant enzyme following a mutation of the
lysine residue at the active site (Watanabe et al. 1999). We
added 1, 25, or 1,000 mM of methylamine or ethylamine to
the reaction mixture with K56A, however, we did not
detect any racemase or dehydrase activities with D- or
L-serine (data not shown).
0.1
1
3
Role of S81 for abstraction/donation of the a-hydrogen
of D-serine
2
Wavelength (nm)
0
250
350
450
550
The active site structure of SpDR strongly suggests that
S81 of DdSR is located on the re-face of PLP, which is the
opposite side of the catalytic lysine (K56) across the PLP
plane (Fig. 3). Furthermore, its hydroxyl group is oriented
toward the a-proton of the substrate moiety in the PLP-D-
serine Schiff base. The structure also suggests that S81 is
a catalytic base for the abstraction and donation of the
a-hydrogen of D-serine.
The CD spectrum of S81A was identical to that of WT
in the far-UV region (between 200 and 260 nm), suggest-
ing that the mutation did not result in gross changes in the
structure of the enzyme (Fig. 5). The S81A mutant enzyme
showed absorption maxima at 280 and 412 nm (Fig. 1e,
dotted line). The addition of 25 mM L-serine to S81A
resulted in a decrease in the absorbance at 412 nm, with a
concomitant increase in the absorbance at 320 nm
(Fig. 1e). In contrast, addition of D-serine resulted in
almost no spectral change in the S81A mutant enzyme
Fig. 4 UV–visible spectra of K56A. UV–vis spectra of untreated-
K56A (line 1), apo-K56A (line 2) and reconstituted-K56A (line 3) in
50 mM Tris–HCl pH 8.5 were shown. Each spectrum was taken for a
solution containing 0.9 mg/ml protein
probably reflect the accumulation of PLP-serine Schiff
base. Accumulation of the external aldimine with K56A
suggests that K56 functions as a residue that abstracts the
a-hydrogen from the Schiff base intermediate.
We also examined the catalytic activities of the
untreated and reconstituted K56A. As shown in Table 1,
neither untreated nor reconstituted K56A exhibited race-
mase activity with D- or L-serine. The mutant enzymes did
not show serine dehydrase activity with either D- or
L-serine. K56A showed no apparent difference from WT in
the circular dichroism (CD) spectrum (Fig. 5) and the elution
profile generated using gel-filtration chromatography (data
123

Catalytic mechanism of serine racemase from D. discoideum
Table 1 Relative activities of mutant enzymes
1079
values of S81A for L-serine dehydration were 47.5 mM
and 68.9 min-1, respectively. The kcat/Km value was
1.45 min-1 mM-1, which is approximately one-fourth that
of WT (5.86 min-1 mM-1, Ito et al. 2012a). Conversion of
SR to L-serine dehydratase by the S81 mutation is com-
patible with the expected role of S81; i.e., S81 serves as
a catalytic residue for the abstraction and addition of the
a-hydrogen of D-serine. However, unexpectedly, the S81A
mutant catalyzed D-serine dehydration very slowly (at
approximately 3.3 % of the rate of the reaction catalyzed
by WT), with Km and kcat values of 97.5 mM and
1.22 min-1 (Table 1), respectively.
Racemase
Dehydrase
(L- to D-Ser) (D- to L-Ser) (L-Ser to Pyr) (D-Ser to Pyr)
WT
100
ND
ND
83.9
ND
35.3
100
ND
ND
133
ND
9.1
100
ND
100
ND
ND
131
3.3
K56A
S80A
S80C
S81A
H84A
ND
54.5
95.9
61.1
5.4
16.4
2.3
K111A 2.6
R132A 58.2
P150S 396
2.1
54.2
250
40.0
29.4
49.3
73.6
Attribution of water to D-serine dehydration catalyzed
by S81A
The serine dehydrase activities were obtained by lactic dehydroge-
nase coupling assay method. The racemase activities were determined
by HPLC. The data are represented as percentages of activity, with
the activity of WT being assumed to be 100 %
The weak D-serine dehydrase activity of S81A raised the
question of whether S81 really catalyzes abstraction (and
donation) of the Ca-H of D-serine. One possible explana-
tion is that DdSR possesses another mechanism for a-
proton abstraction of D-serine in addition to S81.
ND not determined
5
0
Aspartate aminotransferase from chicken has been
found to very slowly racemize some amino acids, such as
L-Glu, L-Asp, L-Phe, L-Tyr, and L-Ala in the presence of the
cognate oxo-acid despite the absence of ionizable catalytic
bases. The reaction rate decreased by the diffusion of water
molecules of the active site when methanol was added to
the reaction mixture, whereas the canonical aminotrans-
ferase activity of the enzyme remained unchanged (Koch-
har and Christen 1992). These phenomena can be explained
by assuming that the water molecule in the active site is
responsible for the racemization. The water molecule
probably donates a proton to Ca of the anionic intermediate
of PLP–amino acid complex on the opposite side of the
PLP plane where a-hydrogen abstraction occurs. Like the
aspartate aminotransferase, the SpSR and DdSR lack
the residues capable of hydrogen transfer on the re-side of
PLP other than S81. However, in the active site of SpSR, 3
-5
-10
-15
WT
K56A
-20
S80A
S81A
-25
200
210
220
230
240
250
260
Wavelength (nm)
Fig. 5 CD spectra of the wild type and mutant DdSRs. The far-UV
CD spectra of WT DdSR (bold line), K56A (thin line), S80A (bold
broken line), and S81A (thin dotted line) were recorded as described
in ‘‘Materials and methods’’
˚
water molecules exist within a distance of 3.6 A from the
(Fig. 1f). When excited at 420 nm, the S81A mutant
enzyme displayed an emission spectrum with a major peak
at kmax of 495 nm (Fig. 2c, bold line). The fluorescence
emission spectrum, including its intensity, was almost
identical to that of WT. The emission band was blue-
shifted by 5–490 nm, and the intensity increased by the
addition of L-serine as observed with WT (Fig. 2c). In
addition, D-serine induced a blue shift of the spectrum
(kmax 490 nm) and increased the emission intensity
(Fig. 2c). These results suggest that S81A forms an
external Schiff base upon addition of both D- and L-serine.
The mutation of S81 to the alanyl residue completely
abolished the racemase activity, but its L-serine dehydrase
activity was fully retained (Table 1). The Km and kcat
substrate Ca atom (Fig. 3). Therefore, we cannot rule out
the possibility that a water molecule is responsible for
D-serine dehydration catalyzed by S81A DdSR.
We then examined the contribution of water mole-
cule(s) to D-serine dehydrase activity catalyzed by S81A.
Dehydration of L-serine was slightly increased rather than
decreased by the addition of methanol to the reaction
mixture (up to 40 % v/v), confirming that S81A is active
under these conditions. In contrast, the rate of D-serine
dehydration catalyzed by S81A decreased with the
increasing concentration of methanol. The catalytic activity
was suppressed to 38 % of that under purely aqueous
conditions, when 40 % of solvent water was replaced
by methanol (Fig. 6). In contrast, almost no effects on
123

1080
T. Ito et al.
WT (dehydrase)
WT (racemase)
S81A (dehydrase)
420 nm, S80A showed an emission spectral peak at
approximately 492 nm. The fluorescence emission spec-
trum was almost identical to that of WT. However, no
emission spectral changes were observed upon the addition
of D- or L-serine, suggesting that S80A lacks the ability to
form a PLP-serine Schiff base (data not shown). These
observations suggest that S80 is essential for substrate
binding. In contrast, substitution of S80 by cysteine (the
resultant enzyme was S80C) caused little difference in the
associated rates of serine racemase and dehydrase reactions
compared with the WT. As shown in Table 1, the rates of
L-serine racemization, L-serine dehydration, D-serine race-
mization, and D-serine dehydration catalyzed by S80C were
83.9, 54.5, 133, and 131 % of those of WT, respectively.
The SH-group can be fully substituted for an OH-group at
position 80.
120
100
80
60
40
20
0
0
10
20
30
40
50
% Methanol, v/v
Fig. 6 Effect of methanol on D-serine racemization and dehydration
catalyzed by WT and S81A. Serine racemization and dehydration
catalyzed by WT (1.9 lM) and S81A (7.4 lM) were performed
at 30 °C in the 500 lL of reaction mixture consisting of 50 mM
Tris–HCl (pH 8.5), 20 lM PLP, 1 mM ATP, 1 mM MgCl2, 10 mM
D-serine, 1 mM L-homocysteic acid (L-HCA) as an internal standard,
and indicated concentrations of methanol. The reactions with WT and
S81A enzymes were terminated by boiling for 5 min after incubation
for 5 min and 4 h, respectively. The racemase (filled diamonds for
WT) and dehydrase (open diamonds for WT, and open circles for
S81A) activity were determined by analyzing the rates of L-serine
formation and D-serine consumption using HPLC. The rates in the
absence of methanol were regarded as 100 %
We also examined the effects of the mutation of H84,
K111, R132, and P150, which are thought to be located
around PLP (see Fig. 3). Relative activities of the mutant
enzymes were compared with those of WT (Tables 1, 2).
No significant differences in the rates of racemization and
dehydration reactions were observed with these mutant
enzymes. The rates of L-serine racemization (L- to D-ser-
ine), D-serine racemization (D- to L-serine), L-serine dehy-
dration, and D-serine dehydration catalyzed by H84A were
35.3, 9.1, 61.1, and 16.4 % of WT, respectively. The
replacement of R132 with alanine (R132 corresponds to
R133 of SpSR and R135 of human SR) caused a slight
decrease in the enzyme’s catalytic activities. The rates of
L-serine racemization, D-serine racemization, L-serine
dehydration, and D-serine dehydration catalyzed by R132A
were 58.2, 54.2, 54.0, and 49.3 % of WT. Although R133
is regarded as a key residue for recognition of the substrate
in modified spSR (Goto et al. 2009), and R135 of human
SR is suggested to stabilize the transition state to enhance
the reaction rate (Smith et al. 2010), no significant changes
were observed in the activities as a result of the mutation of
the corresponding residue (R132) of DdSR. Mutation of
K111 brought about a more marked decrease in the cata-
lytic activities. The catalytic efficiencies of L-serine
dehydration and racemization were observed when meth-
anol was added to the WT reaction (Fig. 6), indicating that
the water molecule is unlikely to participant in the WT
reaction. These results, taken together with those from the
crystallographic data of SRs, provide evidence for the
assignment of S81 as the sole catalytic base for a-hydrogen
transfer of PLP-D-serine external aldimine. The small
activity of D-serine dehydration catalyzed by S81A is
probably attributed to the water molecule(s).
Effect of mutating S80, H84, K111, R132, and P150
As shown in Table 1, substitution of alanine for S80
resulted in complete loss of all enzyme activity. In the
closed form of SpSR structure, a loop consisting of
81Ser-82Ser-83Gly-84Asn-85His recognizes the substrate
serine, and the hydroxyl group of S81, which corresponds
to S80 of DdSR, coordinates with the carboxyl group of the
Table 2 Kinetic parameters of D- and L-serine dehydration catalyzed
by WT and S81A
Reaction
Km (mM)
kcat (min-1
)
k
cat/Km (min-1/mM)
˚
substrate at a distance of 2.5 and 3.0 A, respectively
L-Ser dehydrase
(Fig. 3). S80 of DdSR probably plays a role in substrate
binding by interacting with the carboxyl group of the
substrate. To confirm this, the S80 mutant enzyme was
incubated with or without the substrate and subjected to
spectral analysis. S80A exhibited absorbance maxima at
412 nm, but no change in the absorbance was observed
with D- and L-serine (data not shown). When excited at
(WT)
6.7 1.10
47.5 1.81
39.3 1.50
68.9 1.17
5.86(1)
1.45
(S81A)
D-Ser dehydrase
(WT)
22.4 3.82
97.5 11.3
4.31 0.25
1.22 0.08
0.19
0.01
(S81A)
(1)
The values are from Ito et al. (2012a)
123

Catalytic mechanism of serine racemase from D. discoideum
1081
racemization, D-serine racemization, L-serine dehydration,
and D-serine dehydration catalyzed by K111A were 2.6,
2.1, 5.4, and 2.3 % of those of WT, respectively. The rates
of L- and D-serine racemization increased following the
mutation of P150 to serine. The P150S mutation altered the
partition between racemization and dehydration. This
mutant exhibited a 250 and 396 % increase in the rates of
L- and D-serine racemization, respectively, whereas the
rates of L- and D-serine dehydration were 29.4 and 73.6 %
relative to those of WT. A similar effect was observed in
P153S mouse SR (Foltyn et al. 2005). Because these
mutants showed significant catalytic activities, H84, K111,
R132, and P150 were not essential for either racemization
or dehydration of D- and L-serine.
the S82A SpSR catalyzed L-serine dehydration at a similar
rate to that of the wild-type SpSR (Km = 45 mM,
Vmax = 450 nmol min-1 mg for S82A and Km = 36 mM,
Vmax = 870 nmol min-1 mg for WT), it abolished both
racemase and D-serine dehydrase activity (Goto et al.
2009). Unlike SpSR, S81A of DdSR shows a little D-serine
dehydration activity, which probably occurs via a mecha-
nism different from WT. With S81A, the a-hydrogen of
D-serine is probably abstracted by a water molecule because
addition of a high concentration of methanol decreased the
rate of D-serine dehydration without any loss in L-serine
dehydration. Neither the serine racemase nor D-serine
dehydrase activity of WT were affected by the addition of
methanol, indicating that the water molecule-mediated
catalysis hardly occurs with WT. These observations con-
firm that the Ser81 residue is responsible for the abstrac-
tion/donation of the a-hydrogen of D-serine-PLP complex
at the re-side of PLP.
Discussion
As determined from the structures of SpSR and human SR,
K56 is the PLP-binding lysine residue of DdSR. However,
purified K56A exhibited an absorption peak at 416 nm,
which suggests the presence of an Schiff base of PLP.
Similar Schiff base formation has been reported with some
PLP-dependent enzymes lacking the PLP-binding lysine
residue. The K87T a2b2-tryotophan synthase (Miles et al.
1989) and K42A O-acetylserine sulfhydrylase (Rege et al.
1996) contained PLP, thereby forming an Schiff base with
the free serine, and methionine or leucine, respectively.
Our data suggest that the K56A DdSR contains an Schiff
base of PLP with a free amino acid or some residue of
DdSR. Upon addition of L-serine, the untreated- and
reconstituted-K56A gradually formed the external aldim-
ine, but D-serine and pyruvate were not produced.
Absorption spectra of the enzyme indicated the absence of
the formation of a-aminoacrylate or quinonoid intermedi-
ate with L-serine. These results suggest that K56 is dis-
pensable for the external aldimine formation, but is
involved in the a-proton abstraction from the L-serine
moiety of the PLP-L-serine external aldimine. This finding
is compatible with the expected role of the active-site
lysine, as suggested by crystallography studies of SpSR
and human SR: its e-amino group is situated on the si-face
of PLP and implicated in abstracting and donating the
a-hydrogen of L-serine.
The implication of Ser81 in D-serine dehydration is
reminiscent of L-serine dehydratase. The L-serine dehy-
dratase from rat liver (LSDH) exhibits 27 % sequence
identity with the primary structure and a similar quaternary
structure with SRs (Yamada et al. 2003). In spite of the
similarity, LSDH catalyzes neither serine racemization nor
D-serine dehydration. Comparison of the active site
between SR and LSDH showed a critical difference at
position 81. Ser81 of DdSR is located in the 80Ser-81-
Ser-82Gly-83Asn-84His segment (the so-called asparagine
loop), which is highly conserved among SRs. The segment
is also conserved among LSDHs, but the residue corre-
sponding to S81 of DdSR is replaced by Ala (A65 in the rat
enzyme). A similar segment, 167Ser-168Thr-169Gly-170Asn,
is also found in the E. coli D-serine dehydratase belonging
to the same PLP-dependent enzyme family of SRs, fold-
type II. According to the active site model of the E. coli
D-serine dehydratases, the threonine residue (Thr168,
underlined) is located in the corresponding position of
DdSR S81 and is suggested to abstract the a-proton of
D-serine (Urusova et al. 2012). Presence of the residue with
a hydroxyl group at this position is probably critical for its
reactivity toward D-serine.
Site-directed mutagenesis studies suggest that the serine
racemase reaction catalyzed by DdSR proceeds through a
two-base mechanism with K56 and S81 in a manner similar
to the bacterial alanine racemase (Fig. 7). In contrast to
racemization, L-serine dehydration can be achieved using
one catalytic residue, as exemplified by the observation
that S81A completely catalyzes L-serine dehydration.
Interestingly, we found that K56A did not catalyze D-serine
dehydration, suggesting that K56 is also essential for
D-serine dehydration. K56 is most probably involved in the
release of an a-aminoacrylate from the a-aminoacrylate
intermediate (Fig. 7, IV) through transaldimination.
According to the active-site structures of SpSR and
human SR, the hydroxyl side chain of S81 of DdSR is
probably located close to the a-carbon of substrate on
the re-face of PLP. S81 is the candidate catalytic base for
a-hydrogen transfer of D-serine. S81A displayed impaired
serine racemase and D-serine dehydrase activities, whereas
it exhibited a functional L-serine dehydrase activity. S81 is
not required for L-serine dehydration. Similar results were
recently reported with the S82A mutant of SpSR. Although
123

1082
T. Ito et al.
Fig. 7 Proposed reaction mechanism of DdSR. The cofactor PLP
forms Schiff base with Lys56 (Internal aldimine, I). DdSR is expected
to maintain the N1 pyridine nitrogen in the unprotonated state through
the interaction with S308. K56 and S81 are likely to be the acid–base
catalyst that shuttle proton to the L-Ser (II) and D-Ser external
aldimine (V), respectively, in a two-base mechanism for serine
racemization. L-Serine dehydration (II ? III ? IV) can be achieved
using one catalytic residue (K56). D-Serine dehydration
(V ? III ? IV) requires two residues, S81 and K56. It is possible
that a-hydrogen abstraction and hydroxyl group elimination proceed
in a concerted fashion (II ? IV or V ? IV) in the serine dehydra-
tion. Racemization of serine without formation of the distinct
carbanionic intermediate can be also possible, if the a-hydrogen
abstraction and addition proceed concertedly (Watanabe et al. 2002)
However, lack of the transaldimination is not the reason for
the inability of the D-serine dehydration by K56A, because
K56A did not accumulate the detectable amount of a-am-
inoacrylate intermediate (IV) [or carbanionic intermediate
(III)] with D-serine, but accumulated a PLP-D-serine
external aldimine (V). These observations suggest that
formation of the a-aminoacrylate intermediate (IV) from
D-serine external aldimine (V) requires K56. To examine
whether K56A catalyzes a-hydrogen abstraction from
D-serine, we incubated K56A with [a-1H] D-serine in D2O
that the a-hydrogen abstraction occurs concertedly with the
elimination of the b-hydroxyl group. These speculations
are compatible with that of Griswold and Toney (2011):
Although racemization and dehydration (b-elimination)
reaction start with the proton abstraction at Ca, unlike
racemization, b-elimination can occur through a concerted
mechanism (E2-type mechanism) without forming of a
distinct carbanionic intermediate and such an E2-type
mechanism of the b-elimination reaction has been pro-
posed in some PLP-dependent enzymes such as rat L-serine
dehydratase (Yamada et al. 2003), Salmonella typhimurium
O-acetylserine sulfhydrylase (Cook 2003), Saccharomyces
cerevisiae D-serine dehydratase (Ito et al. 2012b) and G.
stearothermophilus alanine racemase (in a,b-elimination
of b-Cl-D-alanine, Griswold and Toney (2011). One of
the possible mechanism for the concerted reaction of
a-hydrogen abstraction and hydroxyl group elimination
catalyzed by DdSR is as follows: The abstracted
a-hydrogen is transferred to the b-hydroxyl group, which is
1
and monitored the a-hydrogen–deuterium exchange by H
NMR. As a result, we found no a-hydrogen–deuterium
exchange with D-serine (data not shown). This suggests that
K56A catalyzed no a-hydrogen abstraction from D-serine
or that the abstracted hydrogen fully returned to Ca without
exchanging with the solvent deuterium. If the former
possibility holds true, the absence of the a-hydrogen
abstraction denies the formation of the carbanionic inter-
mediate (Fig. 7, III). This can be explained by assuming
123

Catalytic mechanism of serine racemase from D. discoideum
1083
cerevisiae. FEBS J 279:612–624. doi:10.1111/j.1742-4658.2011.
08451.x
Kartvelishvily E, Shleper M, Balan L, Dumin E, Wolosker H (2006)
Neuron-derived D-serine release provides a novel means to
activate N-methyl-D-aspartate receptors. J Biol Chem 281:
14151–14162. doi:10.1074/jbc.M512927200
Kochhar S, Christen P (1992) Mechanism of racemization of amino
acids by aspartate aminotransferase. Eur J Biochem 203:563–
569. doi:10.1111/j.1432-1033.1992.tb16584.x
Kraulis PJ (1991) MOLSCRIPT: a program to produce both detailed
and schematic plots of protein structures. J Appl Cryst 24:946–
950. doi:10.1107/S0021889891004399
Miles EW, Kawasaki H, Ahmed SA, Morita H, Morita H, Nagata S
(1989) The beta subunit of tryptophan synthase. Clarification of
the roles of histidine 86, lysine 87, arginine 148, cysteine 170,
and cysteine 230. J Biol Chem 264:6280–6287
Mori H, Inoue R (2010) Serine racemase knockout mice. Chem
Biodivers 7:1573–1578. doi:10.1002/cbdv.200900293
Mothet JP, Parent AT, Wolosker H, Brady RO Jr, Linden DJ, Ferris
CD, Rogawski MA, Snyder SH (2000) D-serine is an endogenous
ligand for the glycine site of the N-methyl-D-aspartate receptor.
Proc Natl Acad Sci USA 97:4926–4931. doi:10.1073/pnas.97.9.
4926
Nishikawa T (2011) Analysis of free D-serine in mammals and its
biological relevance. J Chromatogr B Analyt Technol Biomed
Life Sci 879:3169–3183. doi:10.1016/j.jchromb.2011.08.030
Rege VD, Kredich NM, Tai CH, Karsten WE, Schnackerz KD, Cook
PF (1996) A change in the internal aldimine lysine (K42) in O-
acetylserine sulfhydrylase to alanine indicates its importance in
transimination and as a general base catalyst. Biochemistry
35:13485–13493. doi:10.1021/bi961517j
then eliminated as a H2O, a more efficient leaving group
than OH-. An amino acid racemase should posses some
hydrogen relay system between the catalytic residues such
as K56 and S81 of DdSR, if not, the racemization reaction
would stop after single turnover. It is thus possible that the
a-hydrogen is abstracted from D-serine by S81 and relayed
to the b-hydroxyl group of D-serine via K56. We also
cannot rule out the possibility that dehydration of D-serine
occurs after its conversion to L-serine through the
racemization.
An understanding the reason behind the inability of K56
to catalyze D-serine dehydration will help the elucidation of
the mechanism of bifunctionality of SRs catalyzing both
racemization and dehydration.
Acknowledgments This work was supported in part by the Program
for Promotion of Basic and Applied Researches for Innovation in Bio-
oriented Industry (BRAIN) (to T. Y.), Grant-in-Aid for Research
22380058 (to T. Y.) from the Ministry of Education, Culture, Sports,
Science, and Technology, Grant-in-Aid for Research Activity Start-up
22880014 (to T. I) and Grant-in-Aid for Young Scientists (B) 24780098
(to T. I.).
References
Shaw JP, Petsko GA, Ringe D (1997) Determination of the structure
˚
of alanine racemase from Bacillus stearothermophilus at 1.9-A
Cook PF (2003) Alpha, beta-elimination reaction of O-acetylserine
sulfhydrylase. Is the pyridine ring required? Biochim Biophys
Acta 1647:66–69. doi:10.1016/S1570-9639(03)00052-9
De Miranda J, Panizzutti R, Foltyn VN, Wolosker H (2002) Cofactors
ofserineracemasethatphysiologicallystimulatethesynthesisofthe
N-methyl-D-aspartate (NMDA) receptor coagonist D-serine. Proc
NatlAcadSci USA 99:14542–14547. doi:10.1073/pnas.222421299
Foltyn VN, Bendikov I, De Miranda J, Panizzutti R, Dumin E,
Shleper M, Li P, Toney MD, Kartvelishvily E, Wolosker H
(2005) Serine racemase modulates intracellular D-serine levels
resolution. Biochemistry 36:1329–1342. doi:10.1021/bi961856c
Smith MA, Mack V, Ebneth A, Moraes I, Felicetti B, Wood M,
Schonfeld D, Mather O, Cesura A, Barker J (2010) The structure
of mammalian serine racemase: evidence for conformational
changes upon inhibitor binding. J Biol Chem 285:12873–12881.
doi:10.1074/jbc.M109.050062
Urusova DV, Isupov MN, Antonyuk S, Kachalova GS, Obmolova G,
Vagin AA, Lebedev AA, Burenkov GP, Dauter Z, Bartunik HD,
Lamzin VS, Melik-Adamyan WR, Mueller TD, Schnackerz KD
(2012) Crystal structure of D-serine dehydratase from Esche-
richia coli. Biochim Biophys Acta 1824:422–432. doi:10.1016/
j.bbapap.2011.10.017
through an alpha, beta-elimination activity.
280:1754–1763. doi:10.1074/jbc.M405726200
J Biol Chem
Goto M, Yamauchi T, Kamiya N, Miyahara I, Yoshimura T, Mihara
H, Kurihara T, Hirotsu K, Esaki N (2009) Crystal structure of a
homolog of mammalian serine racemase from Schizosaccharo-
myces pombe. J Biol Chem 284:25944–25952. doi:10.1074/jbc.
M109.010470
Vaccari S, Benci S, Peracchi A, Mozzarelli A (1996) Time-resolved
fluorescence of tryptophan synthase. Biophys Chem 61:9–22.
doi:10.1016/0301-4622(96)00020-8
Watanabe A, Kurokawa Y, Yoshimura T, Kurihara T, Soda K, Esaki
N (1999) Role of lysine 39 of alanine racemase from Bacillus
stearothermophilus that binds pyridoxal 50-phosphate. Chemical
rescue studies of Lys39– [ Ala mutant. J Biol Chem 274:
4189–4194. doi:10.1074/jbc.274.7.4189
Watanabe A, Yoshimura T, Mikami B, Hayashi H, Kagamiyama H,
Esaki N (2002) Reaction mechanism of alanine racemase from
Bacillus stearothermophilus: X-ray crystallographic studies of
the enzyme bound with N-(50-phosphopyridoxyl)alanine. J Biol
Chem 277:19166–19172. doi:10.1074/jbc.M201615200
Wolosker H, Sheth KN, Takahashi M, Mothet JP, Brady RO Jr, Ferris
CD, Snyder SH (1999) Purification of serine racemase: biosyn-
thesis of the neuromodulator D-serine. Proc Natl Acad Sci USA
96:721–725. doi:10.1073/pnas.96.2.721
Griswold WR, Toney MD (2011) Role of the pyridine nitrogen in
pyridoxal 50-phosphate catalysis: activity of three classes of PLP
enzymes reconstituted with deazapyridoxal 50-phosphate. J Am
Chem Soc 133:14823–14830. doi:10.1021/ja2061006
Inoue R, Hashimoto K, Harai T, Mori H (2008) NMDA- and beta-
amyloid1-42-induced neurotoxicity is attenuated in serine rac-
emase knock-out mice. J Neurosci 28:14486–14491. doi:
10.1523/JNEUROSCI.5034-08.2008
Ito T, Hemmi H, Kataoka K, Mukai Y, Yoshimura T (2008) A novel
zinc-dependent D-serine dehydratase from Saccharomyces cere-
visiae. Biochem J. 409:399–406. doi:10.1042/BJ20070642
Ito T, Murase H, Maekawa M, Goto M, Hayashi S, Saito H, Maki M,
Hemmi H, Yoshimura T (2012a) Metal ion dependency of serine
racemase from Dictyostelium discoideum. Amino Acids
43:1567–1576. doi:10.1007/s00726-012-1232-z
Yamada T, Komoto J, Takata Y, Ogawa H, Pitot HC, Takusagawa F
(2003) Crystal structure of serine dehydratase from rat liver.
Biochemistry 42:12854–12865. doi:10.1021/bi035324p
Ito T, Koga K, Hemmi H, Yoshimura T (2012b) Role of zinc ion for
catalytic activity in D-serine dehydratase from Saccharomyces
123

1084
T. Ito et al.
Yamauchi T, Goto M, Wu HY, Uo T, Yoshimura T, Mihara H,
Kurihara T, Miyahara I, Hirotsu K, Esaki N (2009) Serine
racemase with catalytically active lysinoalanyl residue. J Bio-
chem 145:421–424. doi:10.1093/jb/mvp010
Yoshimura T, Goto M (2008) D-amino acids in the brain: structure
and function of pyridoxal phosphate-dependent amino acid
racemases. FEBS J 275:3527–3537. doi:10.1111/j.1742-4658.
2008.06516.x
123