Site-directed mutagenesis of rice serine racemase: Evidence that Glu219 and Asp225 mediate the effects of Mg2+ on the activity

Article abstract of DOI:10.1002/cbdv.200900257

We succeeded in constructing the Glu219Ala/Asp225Ala (i.e., E219A/D225A) serine racemase (SerR) by site-directed mutagenesis, and the effects of Mg ²⁺ on the catalytic efficiency and the structure were compared between the E219A/D225A-SerR and the wild-type protein. This is the first example of a serine racemase whose amino acid residues in the Mg ²⁺-binding site were replaced with other amino acids by site-directed mutagenesis. Neither the serine racemase nor the dehydratase activities of the E219A/D225A-SerR were affected by the addition of Mg²⁺, and Glu219 and Asp225 of the SerR are the essential amino acid residues for Mg²⁺ to affect both kinds of enzyme activities. Therefore, Glu219 and Asp225 mediate the effects of Mg²⁺ on the activity and are important for the SerR to form the Mg²⁺-binding site. Judging from the difference of the K_eq values between the E219A/D225A-SerR and the SerR, Mg²⁺ might affect the equilibrium states in the racemase reaction. The fluorescence quenching analysis of the E219A/D225A-SerR showed that Mg²⁺ bound to Glu219 and Asp225 of the SerR probably causes a conformational change in the ternary structure of the SerR.

Full text of DOI:10.1002/cbdv.200900257

CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
1579
Site-Directed Mutagenesis of Rice Serine Racemase: Evidence That Glu219
and Asp225 Mediate the Effects of Mg2þ on the Activity
by Yoshitaka Gogami, Ai Kobayashi, Toshihiko Ikeuchi, and Tadao Oikawa*
Department of Life Science and Biotechnology, Faculty of Chemistry, Materials, and Bioengineering,
Kansai University, 3-3-35 Yamate-Cho, Suita, Osaka-Fu 564-8680, Japan
(phone: þ81-6-63680812; fax: þ81-6-63888609; e-mail: oikawa@ipcku.kansai-u.ac.jp)
We succeeded in constructing the Glu219Ala/Asp225Ala (i.e., E219A/D225A) serine racemase
(SerR) by site-directed mutagenesis, and the effects of Mg2þ on the catalytic efficiency and the structure
were compared between the E219A/D225A-SerR and the wild-type protein. This is the first example of a
serine racemase whose amino acid residues in the Mg2þ-binding site were replaced with other amino
acids by site-directed mutagenesis. Neither the serine racemase nor the dehydratase activities of the
E219A/D225A-SerR were affected by the addition of Mg2þ, and Glu219 and Asp225 of the SerR are the
essential amino acid residues for Mg2þ to affect both kinds of enzyme activities. Therefore, Glu219 and
Asp225 mediate the effects of Mg2þ on the activity and are important for the SerR to form the Mg2þ
-
binding site. Judging from the difference of the Keq values between the E219A/D225A-SerR and the
SerR, Mg2þ might affect the equilibrium states in the racemase reaction. The fluorescence quenching
analysis of the E219A/D225A-SerR showed that Mg2þ bound to Glu219 and Asp225 of the SerR
probably causes a conformational change in the ternary structure of the SerR.
Introduction. – Serine racemase (EC 5.1.1.18, SerR) is found in various organisms
[1–4] and is a bifunctional enzyme that catalyzes the racemization and the dehydration
(a,b-elimination) of serine. SerR is classified into two groups based on its origin and
function. The first group consists of the SerRs from bacteria, such as Enterococcus
gallinarum [5] and Streptomyces garyphalus [6]. d-Ser produced by the E. gallinarum
SerR is incorporated into the peptideglycan precursor as d-Ala-d-Ser, and this
modification gives a vancomycin resistance to the bacterium. The second group consists
of the SerRs from eukaryotes, such as rice [7], human, mouse [8], fission yeast [9],
cellular slime mold [9], etc. These eukaryotic SerRs are highly specific for Ser and
belong to the fold-type II pyridoxal 5’-phosphate (PLP) enzymes [10]. d-Ser produced
by the human SerR [11] is a co-agonist of NMDA (N-methyl-d-aspartate) receptors,
and, subsequently, d-Ser is expected to be a drug for schizophrenia [12]. d-Ser was also
detected in rice [7], but the physiological function of d-Ser in plants is still unknown.
Recently, we have cloned the rice SerR gene (serr) and enzymologically
characterized the rice SerR [7]. We have found that Mg2þ increases the catalytic
efficiency of the serine racemase activity of the SerR and decreases that of the serine
dehydratase activity. Additionally, the addition of Mg2þ caused a conformational
change of the SerR. We have also observed similar effects of Mg2þ on the catalytic
efficiency and the structure of Arabidopsis thaliana SerR [13], which is a model for
dicotyledonous plants, while Oryza sativa L. (rice) is a model for monocotyledonous
ꢀ 2010 Verlag Helvetica Chimica Acta AG, Zꢁrich

1580
CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
plants. Accordingly, these characteristic effects of Mg2þ on the enzyme activities are
probably common features for plant SerRs, but there is no report of the Mg2þ-binding
site in a SerR studied by a site-directed mutagenesis. The sequence comparison of rice
SerR with the eukaryotic SerRs around the Mg2þ-binding site is shown in Fig. 1,a.
Fig. 1. a) Comparison of eukaryotic serine racemase amino acid sequences around the Mg2þ-binding site.
The amino acid sequences used for the alignment were as follows: Homo sapiens serine racemase
(accession No. NP_068766), Mus musculus serine racemase (accession No. NP_038789), Oryza sativa L.
(rice) serine racemase (accession no. AB425957), Hordeum vulgare (barley) serine racemase (accession
No. BAF_63026), Arabidopsis thaliana serine racemase (accession No. NP_192901), Dictyostelium
discoideum serine racemase (accession No. Q54 HH2), Schizosaccharomyces pombe serine racemase
(accession No. NP_587715), and Saccharomyces cerevisiae serine racemase homolog (accession No.
*
EDN59810). The Glu and Asp residues were conserved between all serine racemases. : Probable amino
acid residues in the magnesium binding site. b) Predicted three-dimensional structure of rice serine
racemase around the Mg2þ-binding site. The Glu219, Ala223, and Asp225 residues were shown as red
sticks and Mg2þ as a blue sphere, respectively.

CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
1581
Based on the coordinates obtained from the crystal structure of the Mg2þ-bound
Schizosaccharomyces pombe SerR [14], we can predict that Mg2þ is hexavalently
coordinated and that the Mg2þ-binding site is formed by two COOꢀ-containing
residues, a main-chain C¼O O-atom, and three well-ordered H2O molecules. In rice
SerR, Glu219, Asp225, and Ala223 are estimated to be involved in the Mg2þ-binding of
the SerR, and Glu219 and Asp225 are two of the candidate amino acids to form the
Mg2þ binding site of the SerR (Fig. 1,b). Identification of these conserved Glu and Asp
residues as ligands of the metal in the eukaryotic SerRs is mainly based on the crystal
structure of Mg2þ-bound SerR from S. pombe [15]. Accordingly, there is no evidence
that these residues mediate the effects of Mg2þ on the activity of SerR.
In this study, we constructed the Glu219Ala/Asp225Ala (i.e., E219A/D225A)-SerR
by site-directed mutagenesis and compared the effects of Mg2þ on its catalytic
efficiency and structure with those of the wild-type protein, to clarify the function of
Mg2þ and identify the Mg2þ-binding site in the SerR.
Results. – Purification and Characterization of the E219A/D225A-SerR. The
E219A/D225A-SerR was expressed in insoluble fractions of a cell-free extract of E. coli
BL21(DE3) harboring the pET-21b-E219A/D225A serr. The purified E219A/D225A-
SerR (ca. 3.3 mg) was obtained at a yield of 0.57% from a cell-free extract prepared
from one liter of culture (Table 1). The purified protein migrated as a single band on
the SDS-PAGE (Fig. 2). The molecular mass of the E219A/D225A-SerR was
determined to be 80.2 kDa by gel filtration and by considering the subunit molecular
mass (39 kDa) estimated by SDS-PAGE (Fig. 2). The quaternary structure was a
homodimer like that of the wild-type protein [7]. The specific activities of the purified
E219A/D225A-SerR for the racemization of l-Ser and d-Ser were 34 and 36 nmol
minꢀ1 mgꢀ1, respectively, while those for the dehydration of l-Ser and d-Ser were 72
and 37 nmol minꢀ1 mgꢀ1, respectively.
Table 1. Purification of the E219A/D225A-Serine Racemase from Oryza sativa L.
Step
Total
activity [U]a)
Total
protein [mg]
Specific
activity [U/mg]a)
Yield [%]
Purification
[fold]
Crude extract
Ni-NTA Agarose
9.60
0.30
570.00
3.30
1.70ꢁ10ꢀ2
100
0.57
1.00
5.60
9.10ꢁ10ꢀ2
a) The activity was expressed as serine dehydratase activity.
As compared to the wild-type protein, the E219A/D225A-SerR showed the lower
catalytic efficiencies (kcat/Km) in both the serine racemase and the serine dehydratase
reactions (Tables 2 and 3). The kcat/Km values of the E219A/D225A-SerR decreased to
16% (for l-Ser) and 23% (for d-Ser) of those of the wild-type protein in the racemase
reaction and to 36% (for l-Ser) and 26% (for d-Ser) in the dehydratase reaction.
These results suggest that Glu219 and Asp225 are important and essential amino acid
residues for both the serine racemase and dehydratase activities of the SerR.
The atomic absorption analysis of the wild-type protein and the E219A/D225A-
SerR revealed the absence of Mg2þ (calculated values in g-atoms Mg per subunit for

1582
CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
Fig. 2. Subunit molecular weight and homogeneity of the E219A/D225A-SerR. The SDS polyacrylamide
gel electrophoresis was conducted with a Ready 108 BR gel cell (Bio-Rad, Japan, Tokyo) equipped with a
Power pac 300 (Bio-Rad). The concentration of acrylamide was 12%T (T¼total % density of mass of
acrylamide and bisacrylamide). Lane 1, protein marker: b-galactosidase from E. coli (116.0 kDa),
bovine serum albumin from bovine plasma (66.2 kDa), ovalbumin from chicken egg white (45.0 kDa),
lactate dehydrogenase from porcine muscle (35.0 kDa), restriction endonuclease Bsp 981 from E. coli
(25.0 kDa), b-galactoglobulin from bovine milk (18.4 kDa), and lysozyme from chicken egg white
(14.4 kDa); Lane 2, crude extract from E. coli BL21(DE3) cells harboring pET-21b serr (5 mg protein);
Lane 3, purified enzyme from E. coli BL21(DE3) cells harboring pET-21b serr (5 mg protein).
Table 2. Kinetic Parameters (kcat, Km, and kcat/Km) of the Serine Racemase Activity of SerR and E219A/
D225A-SerR
Substrate MgCl2 SerR (wild-type protein)
E219A/D225A-SerR
[mm]
kcat [sꢀ1
]
Km [mm] kcat/Km [sꢀ1 mmꢀ1
]
kcat [sꢀ1
]
Km [mm] kcat/Km [sꢀ1 mmꢀ1
]
l-Serine
l-Serine
d-Serine
d-Serine
–
1
–
1
0.36
0.31
0.31
0.20
18.0
10.0
23.0
8.9
0.020
0.031
0.013
0.022
0.42
0.26
0.12
0.14
130.0
78.0
40.0
48.0
0.0032
0.0033
0.0030
0.0030
Table 3. Kinetic Parameters (kcat, Km, and kcat/Km) of the Serine Dehydratase Activity of SerR and E219A/
D225A-SerR
Substrate MgCl2 SerR (wild-type protein)
E219A/D225A-SerR
[mm]
kcat [sꢀ1
]
Km [mm] kcat/Km [sꢀ1 mmꢀ1
]
kcat [sꢀ1
]
Km [mm] kcat/Km [sꢀ1 mmꢀ1
]
l-Serine
l-Serine
d-Serine
d-Serine
–
1
–
1
0.61
0.36
0.37
0.12
28.0
18.0
19.0
6.3
0.022
0.019
0.019
0.018
1.20
0.74
0.27
0.31
150.0
96.0
53.0
58.0
0.0079
0.0077
0.0050
0.0052

CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
1583
the wild-type protein were 1st, 0.01; 2nd, 0.01, and those for E219A/D225A-SerR were
1st, 0.001; 2nd, 0.002).
Effects of Mg2þ on the Enzyme Activities and the Kinetic Parameters. The effects of
Mg2þ on the serine racemase (Fig. 3,a) and the serine dehydratase (Fig. 3,b) activities
were studied under standard assay conditions in the presence of 0.000, 0.001, 0.010,
0.100, or 1.000 mm Mg2þ. Both the serine racemase and dehydratase activities of the
E219A/D225A-SerR were not affected by the addition of Mg2þ (Fig. 3). The effects of
1 mm Mg2þ on the kinetic parameters for the serine racemase and the serine
dehydratase activities were summarized in Tables 2 and 3, respectively. The catalytic
efficiencies (kcat/Km) of both serine racemase and dehydratase reactions catalyzed by
E219A/D225A-SerR were not affected by the addition of 1 mm Mg2þ. In the case of the
wild-type protein, Mg2þ increased the catalytic efficiency of the SerR in the serine
racemase reaction, but slightly decreased it in the serine dehydratase reaction. These
results suggest that Glu219 and Asp225 are the essential amino acid residues for the
Fig. 3. Effects of Mg2þ on a) the serine racemase and b) the serine dehydratase activities of SerR (wild-
type protein) and E219A/D225A-SerR. Black columns, SerR [7]; white columns, E219A/D225A-SerR.
The relative activity was expressed as the percentage of the specific activity relative to the specific
activity measured in absence of Mg2þ with l-serine as substrate, which was 3.4ꢁ10ꢀ2 and 7.2ꢁ10ꢀ2
U/mg for the racemase and the dehydratase activity, respectively. Values are meansꢂSD (n¼3).

1584
CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
SerR to be affected by Mg2þ. As compared with the wild-type protein, the Keq value of
the E219A/D225A-SerR in the serine racemase reaction decreased from 1.5 to 1.1,
which is almost equal to the theoretical value of 1.0 [16].
Fluorescence Quenching Analysis of the Tryptophan (Trp) Residues in the E219A/
D225A-SerR by Acrylamide. The fluorescence quenching analysis in our previous
report [7] revealed that the structure of the SerR was distorted by the addition of Mg2þ
,
and this suggests that the accessibility of acrylamide to the Trp residues in the SerR
decreases and that a structural change occurs. To clarify the effect of Mg2þ on the
structure of SerR, we also studied the effect of Mg2þ on the structure of the E219A/
D225A-SerR by fluorescence quenching analysis of the Trp residues in the E219A/
D225A-SerR by acrylamide in the presence (1 mm) or absence of Mg2þ (Fig. 4). Based
on the data in Fig. 4,a and b, the Stern–Volmer plots in presence and absence of Mg2þ
were described in Fig. 4,c, and the Stern–Volmer constants (Ksv) determined were
summarized in Table 4. The Ksv constants did not change by the addition of Mg2þ in the
presence or absence of d-Ser or l-Ser as a substrate. These results suggest that the
accessibility of acrylamide to the Trp residues in the E219A/D225A-SerR did not
change and that no structural change occurred. Accordingly, the structure of the
E219A/D225A-SerR was not changed by the addition of Mg2þ, and Glu219 and Asp225
of the SerR seem to be essential amino acid residues for the occurrence of the structural
change caused by Mg2þ. The binding of Mg2þ to Glu219 and Asp225 of the SerR causes
the structural change of the SerR, and this might increase the serine racemase activity
and decrease the serine dehydratase activity of the SerR.
Table 4. Stern–Volmer Constants (Ksv) Obtained in the Fluorescence Quenching Analysis of Tryptophan
Residues in the E219A/D225A-SerR by Acrylamide
Substrate
MgCl2 [mm]
Ksv
–
–
–
1
3.03
3.07
l-Serine
l-Serine
–
1
2.98
2.98
d-Serine
d-Serine
–
1
3.27
3.27
Discussion. – Various metal ions are known to affect the serine racemase and
dehydratase activities of SerRs, and the dependence of SerR on divalent cations was
first reported by Cook et al. [17]. l-Serine racemase and dehydratase activities of the
mouse SerR in the presence of 1 mm Mg2þ, Ca2þ, Mn2þ, Fe2þ, Co2þ, Ni2þ, Cu2þ, or
Zn2þ were measured by Strꢀsovsky et al. [18]. The racemase and dehydratase activities
ˇ ˇ
´
were slightly increased by the addition of Mg2þ and Mn2þ, slightly decreased by Ca2þ
,
Fe2þ, and Ni2þ, and significantly inhibited by Co2þ, Cu2þ, and Zn2þ. In the case of the
SerR from Pyrobaculum islandicum [4], both enzyme activities were increased by Cu2þ
and decreased by Fe2þ. In the Schizosaccharomyces pombe enzyme, both activities
were increased by Mg2þ and Ca2þ [9]. Similar properties were also observed in the

CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
1585
Fig. 4. Fluorescence quenching analysis of tryptophan residues in E219A/D225A-SerR with acrylamide.
a) Fluorescence intensities in the presence of Mg2þ (1 mm). b) Fluorescence intensities in the absence of
Mg2þ. The spectrum with the highest fluorescence intensity (F0) was recorded in the absence of
acrylamide and the other spectra (F) in the presence of increasing acrylamide concentrations. c) Stern–
2þ
*
*
Volmer plots in the absence ( ) and presence ( ) of Mg (1 mm).
Hordeum vulgare L. enzyme [19]. Accordingly, Mg2þ has been generally considered to
be an activator of both activities of a SerR. We found that Mg2þ increased the catalytic
efficiency of the serine racemase activity from Oryza sativa L. and slightly decreased
that of the serine dehydratase activity [7]. This could be a way to regulate the activity of
the SerR in rice. But there was no report investigating how metal ions biochemically
increase or decrease both enzyme activities of SerR. To clarify these observations, we

1586
CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
constructed the E219A/D225A-SerR by site-directed mutagenesis and compared the
effects of Mg2þ on its catalytic efficiency and structure with those on the wild-type
protein. This is the first example of a SerR whose amino acid residues in the Mg2þ
binding site were replaced with other amino acids by site-directed mutagenesis.
-
Neither the serine racemase nor the dehydratase activities of the E219A/D225A-
SerR were affected by the addition of Mg2þ (Fig. 3,a and b), and Glu219 and Asp225
mediated the effects of Mg2þ on the activity and were important for the SerR to form
the Mg2þ-binding site. Judging from the difference of the Keq values between the
E219A/D225A-SerR and the SerR, Mg2þ might have an effect on the equilibrium states
in the racemase reaction. The atomic absorption analysis revealed that the E219A/
D225A-SerR and the SerR did not contain Mg2þ. Accordingly, Mg2þ was probably
bound weakly to the SerR via Glu219 and Asp225. The kcat and Km values of both serine
racemase and dehydratase reactions catalyzed by the E219A/D225A-SerR were not
significantly affected by the addition of 1 mm Mg2þ (Tables 2 and 3). These results
suggest that Mg2þ probably also affects the kcat and Km values of the serine racemase
and dehydratase reactions in a different way.
The fluorescence quenching analysis of the E219A/D225A-SerR indicated that
Mg2þ probably binds to Glu219 and Asp225 of the SerR, which causes a conforma-
tional change in the ternary structure of the SerR, and this is one of the reasons for
Mg2þ to increase the serine racemase activity and slightly decrease the serine
dehydratase activity of the SerR. Similar effects of Mg2þ on the enzyme activities and
structures were also observed in the Mg2þ-dependent porphobilinogen synthase
(PBGS) from Pseudomonas aeruginosa [20]. The crystal structure of the enzyme,
solved at 1.67 ꢂ resolution, showed a homooctameric enzyme that consists of four
asymmetric dimers. The monomers in each dimer differed from each other by having a
ꢃclosedꢄ and an ꢃopenꢄ active site pocket. No metal ions were found in the active site of
both monomers, but a single Mg2þ was present at ca. 14 ꢂ distance from the active site
of the enzyme. The Mg2þ-binding site of the enzyme contained Arg181 and Glu245.
The observed differences in the active sites of both monomers were probably induced
by the Mg2þ-binding to this remote site, and a structure-based mechanism for this
allosteric Mg2þ in a rate enhancement was proposed.
The physiological function of a SerR has been extensively studied for the brain
enzyme existing in glial cells, where d-Ser is produced from l-Ser by the enzyme and
subsequently diffuses and co-activates the N-methyl-d-aspartate receptor through the
binding to the so-called ꢃglycine siteꢄ. However, the physiological functions of other
SerRs, especially of a plant SerR, have not been clarified at all, although we found that
germinated, unpolished rice contains a substantial amount of d-Ser, and the ratio of the
d-enantiomer to the l-enantiomer is higher for Ser than for other amino acids [7]. The
Mg2þ concentration found in a rice plant was reported to be 0.16–0.33 mm [21]. This is
enough to activate the SerR in rice. d-Ser and d-Ala are known to inhibit the growth of
A. thaliana, even at low concentrations [22]. Thus, the physiological function of a plant
SerR might be the detoxification of d-Ser taken up from the environment.
We are currently studying the physiological function of a plant SerR with various
biochemical systems.

CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
1587
This work was supported in part by the research grant Strategic Project to Support the Formation of
Research Bases at Private Universities.
Experimental Part
Materials. Escherichia coli BL21(DE3), E. coli NovaBlue, the pET21b vector, and the pT7-blue2
vector were purchased from EMD Biosciences, Inc. (San Diego, CA). The DNA ligation kit (Mighty
Mix) and the restriction enzyme NdeI were acquired from TaKaRa Bio, Inc. (Shiga, Japan), XhoI from
Roche Diagnostics (CH-Basel), the UltraClean15 DNA purification kit from Mo Bio Laboratories, Inc.
(West Carlsbad, CA), and KOD DNA polymerase and Target Clone from TOYOBO Co., Ltd. (Osaka,
Japan). TOYOPEARL HW-55S was purchased from Tosoh Corp. (Tokyo, Japan), Ni-nitrilotriacetic acid
(Ni-NTA) agarose from QIAGEN (D-Hilden), and the DNA ladder from New England Biolabs, Inc.
(Ipswich, MA). A protein molecular weight marker (Fermentas International, Inc., Burlington, ON,
Canada) was used for the sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Gel
filtration calibration kits (GE Healthcare UK Ltd.) were used for the determination of the molecular
mass. All other reagents were the best grade commercially available and were purchased from Kanto
Kagaku Co. (Tokyo, Japan), Kishida Chemical Co. (Osaka, Japan), Sigma-Aldrich Co. (St. Louis, MO),
Tokyo Kasei Kogyo Co. (Tokyo, Japan), or Wako Chemical Co. (Osaka, Japan), unless otherwise stated.
Construction of the E219A/D225A serr by Site-Directed Mutagenesis. The E219A/D225A serr was
constructed by site-directed mutagenesis with an overlap extension PCR method. First, the E219A serr
was prepared, and then the D225A mutation was introduced into it. The experimental strategies consist
of eight PCR experiments. The bacterial strains and primers used in this study are listed in Table 5. To
prepare the E219A serr, the first PCR ran for 30 cycles in a Gene Amp 9700 PCR system (PE Applied
Biosystems, Tokyo, Japan) with a mixture (total volume: 10.0 ml) containing 1 mm of primer P1 (0.1 ml),
1 mm of P4 (0.1 ml), 0.2 mm dNTP (1.0 ml), 1.5 mm MgCl2 (0.6 ml), 10ꢁKOD buffer (1.0 ml), 1 U KOD
DNA polymerase (0.2 ml), 500 mg/ml pET21b serr (0.1 ml), and deionized H2O (7.0 ml). The PCR reaction
conditions used were as follows: pre-heating at 948 for 2 min, denaturation at 988 for 10 s, annealing at
658 for 30 s, and extension at 688 for 30 s. The second PCR was performed under the same conditions as
used for the first PCR, except that P1 and P4 were changed to P2 and P3, resp. The DNAs in each
reaction mixture (10 ml) were separated by a 1.2% agarose gel electrophoresis, and the amplified PCR
product was extracted from the gel and purified with an UltraCleanTM15 DNA purification kit. The
purified DNA fragments obtained from the first and second PCRs were used as megaprimers M1 and M2,
resp. The third PCR was done with the same mixture as used for the first PCR, except that P1 and P4
were changed to M1 and M2, resp. The PCR reaction conditions used were as follows: pre-heating at 948
for 2 min, denaturation at 988 for 10 s, annealing at 558 for 30 s, and extension at 688 for 30 s. The
amplified PCR product was purified by the same procedure as described above and subjected to the forth
PCR to add the restriction enzyme sites (NdeI and XhoI) to the DNA fragment. The forth PCR was done
under the same conditions as used for the first PCR, except that pET21b serr was changed to the third
PCR product. The amplified PCR product was purified by the same procedure as described above and
ligated into a pT7-blue2 vector with a Ligation Mighty mix after a deoxyadenine nucleotide was attached
to both ends of the DNA fragment with Target Clone. E. coli NovaBlue cells were transformed with the
pT7-blue2-E219A serr, and, after confirmation of the DNA sequence by a DNA sequencer SQ5500E
(Hitachi High-Technologies Corp., Tokyo, Japan), the amplified plasmids were used as the template gene
for the fifth PCR to introduce the additional D225A mutation. The fifth PCR was performed under the
same conditions as used for the first PCR, except that P4 and pET21b serr were changed to P5 and pT7-
blue2-E219A serr, resp. The sixth PCR was done under the same conditions as used for the second PCR,
except that P3 and pET21b serr were changed to P6 and pT7-blue2-E219A serr, resp. The purified DNA
fragments obtained from the fifth and sixth PCRs were used as megaprimers M3 and M4, resp. The
seventh PCR was done under the same conditions as used for the third PCR, except that M1 and M2 were
changed to M3 and M4, resp. The amplified PCR product was purified by the same procedure as
described above and subjected to the eighth PCR. The eighth PCR was done under the same conditions
as used for the forth PCR, except that the third PCR product was changed to the seventh one. The

1588
CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
purified DNA fragment obtained by the eighth PCR was subcloned with a pT7-blue2 vector as described
above, and the DNA sequence of the amplified DNA fragment was confirmed by a DNA sequencer as 5’-
NdeI-E219A/D225A serr-XhoI-3’. This DNA fragment was ligated into an NdeI and XhoI-digested pET-
21b vector to form pET-21b-E219A/D225A serr.
Table 5. Bacterial Strains and Primers Used in This Study
Strain or primer
Relevant characterization or primer sequence
Source
Strains
Escherichia coli
BL21(DE3)
Escherichia coli
NovaBlue
Fꢀ ompT hsdSB(rBꢀ mꢀB ) gal dcm (DE3)
EMD Biosciences
EMD Biosciences
endA1, hsdR17, (rꢀK mKþ), supE44, thi-1,
gyrA96, relA1, lac, recA1/Fꢁ, [proABþ,
lac Iq Z DM15, Tn10(tetr)]
Primers used for the construction of the pET21b serr
P1: Oryza serRC-N
(for pET21b)
5’-TCA TAT GGG GAG CAG AGG TG-3’
P2: Oryza serRC-C
(for pET21b)
5’-CTC GAG ACG TTT ATA GAG AGA CTC CC-3’
Primers used for the construction of the pET21b-E219A/D225A
P3: E219A-N
P4: E219A-C
P5: E225A-N
P6: E225A-C
5’-CTG GCA GCA GCA CCA AAG GGT GCT-3’
5’-AGC ACC CTT TGG TGC TGC TGC CAG-3’
5’-TGC TGA TGC TTC TGC CCA GTC C-3’
5’-GGA CTG GGC AGA AGC ATC AGC A-3’
Construction of the Expression Vector of the E219A/D225A serr. After the first and second PCR
experiments, M1 and M2 produced were detected at ca. 650 and 350 bp, resp., by agarose gel
electrophoresis. The overlapped extension with M1 and M2 carried out by the third PCR experiment
succeeded, and the E219A serr (ca. 1,000 bp) was obtained. After ligation of the E219A serr gene with a
pT7-blue2 vector, the mutation of the GAA codon to the GCA codon, corresponding to Glu219 and
Ala219, resp., was confirmed by a DNA sequencer. The mutation of the GAT to the GCT codon in the
pT7-blue2-E219A serr, corresponding to Glu225 and Ala225, resp., was also successfully introduced by
the same procedure with the fifth to seventh PCR experiments. After addition of NdeI and XhoI sites, the
E219A/D225A serr confirmed by a DNA sequencer was ligated into an NdeI and XhoI-digested pET-21b
vector, and the desired pET-21b-E219A/D225A serr vector was constructed successfully.
Expression of the E219A/D225A serr in E. coli BL21(DE3). E. coli BL21(DE3) cells were
transformed with the pET-21b-E219A/D225A serr constructed, and the clone obtained was inoculated
and cultivated in Luria–Bertani medium (1 l) supplemented with ampicillin (100 mg/ml) at 378 with
reciprocal shaking (100 rpm). When the optical density at 600 nm (OD600) had reached a value of ca. 0.6,
the culture was placed at 158 and cultivated again for 24 h.
Purification of the E219A/D225A-SerR. The E. coli BL21(DE3) cells harboring the pET-21b-
E219A/D225A serr were collected by centrifugation (model H-923, Kokusan Co., Ltd., Tokyo, Japan;
26,100g, 15 min, 48). The cells were suspended in a lysis buffer containing a 50 mm sodium phosphate
buffer (pH 8.0), 300 mm NaCl, and 20 mm PLP, and disrupted by ultrasonication (model UD-201, Tomy
Seiko Co., Tokyo, Japan). The disruption conditions used were as follows: output, 6; duty cycle, 30;
operation time, 5 min; operation number, 5. After disruption, the cell suspension was centrifuged
(26,100g, 20 min, 48) and ultracentrifuged (100,000g, 1 h, 48) with a himac CP85b centrifuge (Hitachi
Koki Co., Ltd., Tokyo, Japan). The supernatant obtained was applied to a NiꢀNTA column (1ꢁ2 cm)
equilibrated at a flow rate of 0.3 ml/min with a 50 mm sodium phosphate buffer (pH 8.0) containing
300 mm NaCl (buffer A) plus 10 mm imidazole. The column was washed at a flow rate of 0.5 ml/min

CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
1589
subsequently with buffer A plus 10 mm 1H-imidazole (100 ml), buffer A plus 20 mm imidazole (200 ml),
and buffer A plus 60 mm imidazole (100 ml) to remove the contaminated proteins. The adsorbed protein
was eluted with buffer A plus 150 mm imidazole (50 ml). The fractions containing the SerR were
collected by a Bio-collector (ATTO, Tokyo, Japan), and the protein concentration was measured at
280 nm by spectrophotometry (model V-550, Jasco Co., Ltd., Tokyo, Japan). The purified SerR was
dialyzed against a 20 mm sodium phosphate buffer (pH 8.0) containing 10 mm PLP, and stored at 48.
Enzyme Assay. The serine racemase and dehydratase activities were measured by HPLC and by the
salicylaldehyde method, resp., according to the procedures previously described [7]. To study the
influence of Mg2þ, MgCl2 was added to the reaction mixture at final concentrations of 0.001, 0.010, 0.100,
or 1.000 mm.
Kinetic Parameters. The kinetic parameters of the serine racemase and dehydratase reactions were
determined under standard assay conditions, except for the substrate concentrations. The final substrate
(d- or l-Ser) concentrations used were 2.5, 5.0, 7.5, 12.5, and 15.0 mm. The kinetic parameters (kcat, Km,
and kcat/Km), in the presence (1 mm) or absence of Mg2þ, were determined with Lineweaver–Burk
double-reciprocal plots.
Fluorescence Quenching Analysis of Tryptophan Residues in the E219A/D225A-SerR by Acrylamide.
The conformational change of the E219A/D225A-SerR was estimated by the fluorescence quenching
analysis of Trp residues with acrylamide. The experiments were done in the presence or absence of 1 mm
Mg2þ. The Stern–Volmer quenching constants (Ksv) were calculated based on the Stern–Volmer plots,
according to the method previously reported [7].
Other Methods. Protein concentrations were determined with the method of Bradford, using bovine
serum albumin as standard [23]. Sodium dodecyl polyacrylamide gel electrophoresis (SDS-PAGE) was
carried out with the method of Laemmli [24]. The molecular mass of the purified E219A/D225A-SerR
was measured with a column of TOYOPEARL HW-55S (1.6ꢁ120 cm). The column was equilibrated and
eluted at a flow rate of 0.3 ml/min with a 20 mm Tris·HCl buffer (pH 8.0) containing 0.15m NaCl. The
standard proteins used were thyroglobulin (669 kDa), catalase (232 kDa), albumin (69 kDa), and
chymotrypsinogen A (25 kDa). Mg2þ was measured at r.t. with a polarized Zeeman atomic absorption
spectrophotometer Z-8200 (Hitachi High-Technologies Co., Tokyo). The absorption was monitored with
deuterium-arc background correction. The enzyme solution was used after dialysis against two changes of
300 volumes of ultra-pure H2O at 48 for 12 h. The outer solution of the dialysis tube was used as a blank.
Values are averages of at least two sets of independent determinations.
REFERENCES
[1] H. Wolosker, K. N. Sheth, M. Takahashi, J.-P. Mothet, R. O. Brady Jr., C. D. Ferris, S. H. Snyder,
Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 721.
[2] J. De Miranda, A. Santoro, S. Engelender, H. Wolosker, Gene 2000, 256, 183.
[3] T. Uo, T. Yoshimura, S. Shimizu, N. Esaki, Biochem. Biophys. Res. Commun. 1998, 246, 31.
[4] M. Ohnishi, M. Saito, S. Wakabayashi, M. Ishizuka, K. Nishimura, Y. Nagata, S. Kasai, J. Bacteriol.
2008, 190, 1359.
[5] D. Billot-Klein, D. Blanot, L. Gutmann, J. van Heijenoort, Biochem. J. 1994, 304, 1021.
[6] M.-L. Svensson, S. Gatenbeck, Arch. Microbiol. 1981, 129, 213.
[7] Y. Gogami, K. Ito, Y. Kamitani, Y. Matsushima, T. Oikawa, Phytochemistry 2009, 70, 380.
[8] H. Wolosker, S. Blackshaw, S. H. Snyder, Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 13409.
[9] T. Yoshimura, M. Goto, FEBS J. 2008, 275, 3527.
[10] N. V. Grishin, M. A. Phillips, E. J. Goldsmith, Protein Sci. 1995, 4, 1291.
[11] L. Verrall, M. Walker, N. Rawlings, I. Benzel, J. N. C. Kew, P. J. Harrison, P. W. Burnet, Eur. J.
Neurosci. 2007, 26, 1657.
[12] G. E. Tsai, P. Yang, L. C. Chung, N. Lange, J. T. Coyle, Biol. Psychiatry 1998, 44, 1075.
[13] A. Kobayashi, Y. Gogami, H. Yasuhara, T. Oikawa, Vitamins 2009, 83, 209.
[14] M. Goto, T. Yamauchi, N. Kamiya, I. Miyahara, T. Yoshimura, H. Mihara, T. Kurihara, K. Hirotsu,
N. Esaki, J. Biol. Chem. 2009, 284, 25944.

1590
CHEMISTRY & BIODIVERSITY – Vol. 7 (2010)
[15] F. Baumgart, I. Rodrꢅguez-Crespo, FEBS J. 2008, 275, 3538.
[16] G. E. Briggs, J. B. S. Haldane, Biochem. J. 1925, 19, 338.
[17] S. P. Cook, I. Galve-Roperh, A. Martꢅnez del Pozo, I. Rodrꢅguez-Crespo, J. Biol. Chem. 2002, 277,
27782.
ˇ ˇ
´
´
´
ˇ
[18] K. Strꢅsovsky, J. Jiraskova, C. Barinka, P. Majer, C. Rojas, B. S. Slusher, J. Konvalinka, FEBS Lett.
2003, 535, 44.
[19] Y. Fujitani, T. Horinouchi, K. Ito, M. Sugimoto, Phytochemistry 2007, 68, 1530.
[20] N. Frankenberg, P. T. Erskine, J. B. Cooper, P. M. Shoolingin-Jordan, D. Jahn, D. W. Heinz, J. Mol.
Biol. 1999, 289, 591.
[21] K. Teramoto, Breed. Sci. 1996, 46, 241.
[22] O. Erikson, M. Hertzberg, T. Nꢆsholm, Nat. Biotechnol. 2004, 22, 455.
[23] M. M. Bradford, Anal. Biochem. 1976, 72, 248.
[24] U. K. Laemmli, Nature 1970, 227, 680.