Báo cáo khoa học: Mechanism of the reaction catalyzed by dehydroascorbate reductase from spinach chloroplasts doc

Mechanism of the reaction catalyzed by dehydroascorbate reductase
from spinach chloroplasts
Taise Shimaoka
1,2
, Chikahiro Miyake
2
and Akiho Yokota
1,2
1
Graduate School of Biological Sciences, Nara Institute of Science and Technology, Nara, Japan;
2
Research Institute of Innovative
Technology for the Earth, Kyoto, Japan
Dehydroascorbate reductase (DHAR) reduces dehydro-
ascorbate (DHA) to ascorbate with glutathione (GSH) as
the electron donor. We analyzed the reaction mechanism of
spinach chloroplast DHAR, which had a much higher
reaction specificity for DHA than animal enzymes, using a
recombinant enzyme expressed in Escherichia coli. Kinetic
analysis suggested that the reaction proceeded by a
bi-uni-uni-uni-ping-pong mechanism, in which binding of
DHA to the free, reduced form of the enzyme was followed
by binding of GSH. The K
m
value for DHA and the summed
K
m
value for GSH were determined to be 53 ± 12 l
M
and
2.2 ± 1.0 m

M
, respectively, with a turnover rate of
490 ± 40 s
)1
. Incubation of 10 l
M
DHAR with 1 m
M
DHA and 10 l
M
GSH resulted in stable binding of GSH to
the enzyme. Bound GSH was released upon reduction of the
GSH–enzyme adduct by 2-mercaptoethanol, suggesting that
the adduct is a reaction intermediate. Site-directed muta-
genesis indicated that C23 in DHAR is indispensable for the
reduction of DHA. The mechanism of catalysis of spinach
chloroplast DHAR is proposed.
Keywords: dehydroascorbate reductase; catalytic mechan-
ism; ping-pong mechanism; oxidative stress; ascorbate.
Ascorbate functions not only as an antioxidant but also as a
substrate for ascorbate peroxidase (APX) and violaxanthin
de-epoxidase in chloroplasts [1,2]. APX catalyzes the
decomposition of hydrogen peroxide in the active oxygen-
scavenging system and the reaction catalyzed by violaxan-
thin de-epoxidase in the xanthophyll cycle is involved in the
down-regulation of the activity of photosystem II. These
enzymes are involved in the dissipation of excess light energy
and protect plants from oxidative stress. In reactions
catalyzed by APX and violaxanthin de-epoxidase, ascorbate
is oxidized to monodehydroascorbate (MDA) and then

dehydroascorbate (DHA) is produced via the spontaneous
disproportionation of MDA. The regeneration of ascorbate
is essential for the maintenance of the activity of the active
oxygen-scavenging system and the xanthophyll cycle. MDA
and DHA are reduced to ascorbate by MDA reductase and
ferredoxin, and DHA reductase (DHAR) in chloroplasts,
respectively [3–6].
The reduction of DHA to ascorbate by DHAR
(EC 1.8.5.1) involves GSH as the electron donor. Enzymes
that reduce DHA are distributed not only in plant cells but
also in mammalian cells [7–14]. However, the enzymatic
properties of spinach chloroplast DHAR are different from
those of other DHA-reducing enzymes. The specific activity
of spinach chloroplast DHAR was found to be seven times
higher than that of DHAR from rice bran [8,15]. Moreover,
spinach chloroplast DHAR has a 100-fold lower K
m
value
for DHA and several-fold higher specific activity
than porcine DHAR and other DHA-reducing enzymes
[7,11–13,15,16].
DHA-reducing enzymes commonly include a C-X-X-C
motif. It has been demonstrated by site-directed muta-
genesis that the C22 residue in pig liver thioltransferase is
essential for the reduction of DHA [17]. Spinach chloroplast
DHAR also has this motif [15] but the highly efficient
reduction of DHA by spinach chloroplast DHAR cannot
be explained by this motif alone. The difference in k
cat
between spinach chloroplast DHAR and other DHA-

reducing enzymes might be due to differences in mecha-
nisms of catalysis.
Models for catalysis by DHAR were proposed for pig
liver thioltransferase and trypanothione:glutathione disul-
fide thioltransferase from Trypanosoma cruzi [18,19].
However, the validity of these models has not been
confirmed by kinetics. In the present study, we examined
the mechanism of catalysis by spinach chloroplast DHAR.
Our kinetic studies showed that catalysis by spinach
chloroplast DHAR proceeds by a bi-uni-uni-uni-ping-pong
Correspondence to Akiho Yokota, Graduate School of Biological
Sciences, Nara Institute of Science and Technology,
8916-5 Takayama, Ikoma, Nara 630-0101, Japan.
Fax: + 81 743 72 5569, Tel.: + 81 743 72 5560,
E-mail:
Abbreviations: APX, ascorbate peroxidase; AsA, ascorbate; DHA,
dehydroascorbate; DHAR, dehydroascorbate reductase; GSH,
glutathione; GSSG, oxidized glutathione; K
eq
, equilibrium constant;
K
DHA
m
, K
m
for DHA; K
GSH1
m
, K
m

for the first-binding molecule of
GSH; K
GSH2
m
, K
m
for the second-binding molecule of GSH;
K
DHA
i
, inhibition constant for DHA; K
GSH1
i
, inhibition constant for
the first-binding molecule of GSH; K
GSH2
i
, inhibition constant for the
second-binding GSH; K
AsA
m
, K
m
for AsA; K
GSSG
m
, K
m
for GSSG;
K

AsA
i
, inhibition constant for AsA; K
GSSG
i
, inhibition constant for
GSSG; MDA, monodehydroascorbate; V
max
, maximum reaction rate;
V
maxf
, maximum rate of the forward reaction; V
maxr
, maximum rate of
the reverse reaction; V
f
, forward reaction rate; V
r
, reverse reaction rate.
Enzyme: Dehydroascorbate reductase (EC 1.8.5.1).
(Received 11 September 2002, revised 26 December 2002,
accepted 7 January 2003)
Eur. J. Biochem. 270, 921–928 (2003) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03452.x
mechanism. A disulfide bond was formed between the
enzyme reacted with DHA and GSH, and site-directed
mutagenesis revealed that the C23 was essential for DHAR
activity. The role of the C residue in the reaction catalyzed
by DHAR is discussed.
Experimental procedures
Materials

DHA was purchased from Sigma (St Louis, MO, USA).
GSH, GSSG and 2-mercaptoethanol were obtained from
Wako Pure Chemical Industries (Osaka, Japan). 4-Fluoro-
7-sulfamoylbenzofurazan was obtained from Dojin (Kuma-
moto, Japan). Other chemicals and reagents were of the
highest purity commercially available.
Assay of the activity of DHAR
The reaction rate of DHAR was determined by monitoring
the glutathione-dependent production of ascorbate at
265 nm, as described in a previous report [15]. The reaction
mixture contained 50 m
M
potassium phosphate (pH 7.0),
1m
M
EDTA, DHA and GSH at the indicated concentra-
tions, and purified enzyme.
Analysis of data
All data were fitted to theoretical lines or curves by the least-
squares method with the computer program
KALEIDAGRAPH
3.08d (Synergy Software, PA, USA).
Purification of recombinant DHAR from
Escherichia coli
E. coli BL21 (DE3) harboring pET3a-DHAR, in which the
cDNA for the mature form of spinach chloroplast DHAR
had been ligated [15], was grown in 200 mL of LB medium
supplemented with 50 lgÆmL
)1
ampicillin at 37 °Cfor

12–16 h.
All procedures for purification were performed at 0–4 °C.
The cultured cells were collected by centrifugation at 4620 g
for 10 min and resuspended in 50 mL of 50 m
M
potassium
phosphate (pH 7.8) that contained 1 m
M
EDTA, 40 m
M
2-mercaptoethanol, 1 m
M
phenylmethylsulfonyl fluoride
and 20% (v/v) glycerol. The cells were disrupted with a
French Pressure Cell Press (AFPS-20KM; Amicon, MA,
USA) at 6895 MPa. The cell extract was centrifuged at
10 000 g for 10 min and the supernatant was used for
purification of recombinant DHAR.
The supernatant was brought to 40% saturation by
addition of (NH
4
)
2
SO
4
and allowed to stand for 30 min with
gentle stirring. After centrifugation at 10 000 g for 10 min,
the supernatant was applied to a column (1.6 · 10 cm) of
butyl-Toyopearl (Tosoh, Tokyo, Japan), which had been
equilibrated with buffer A, which was 50 m

M
potassium
phosphate (pH 7.8) containing 1 m
M
EDTA, 10 m
M
2-mercaptoethanol and 20% (v/v) glycerol. The adsorbed
enzyme was eluted with a linear gradient of (NH
4
)
2
SO
4
(40–0% saturation) in buffer A. The active fractions were
appliedtoacolumn(2.6 · 60 cm) of Superdex 75 prep grade
(Amersham Pharmacia, Uppsala, Sweden), which had been
equilibrated with buffer A containing 0.15
M
KCl. The
column was developed with the same buffer and the active
fractions were pooled and stored at )80 °C.
The concentration of purified enzyme was quantitated
with a Protein Assay kit from Bio-Rad (Hercules, CA,
USA) with bovine serum albumin as the standard.
Quantitation of DHAR-bound GSH
DHAR in buffer A containing 0.15
M
KCl was passed
through a column of PD-10 (Amersham Pharmacia), which
had been equilibrated with N

2
-purged 50 m
M
potassium
phosphate (pH 7.0) that contained 1 m
M
EDTA, to remove
2-mercaptoethanol under anaerobic conditions. The
reduced-form enzyme (10 l
M
)wasallowedtoreactwith
1m
M
DHA and 10 l
M
GSH at room temperature for 30 s
in the presence of 50 m
M
potassium phosphate (pH 7.0)
that contained 1 m
M
EDTA under N
2
. DHA, GSH and
phosphate were removed from the mixture under anaerobic
conditions by passage through a column of PD-10 (Amer-
sham Pharmacia), which had been equilibrated with 50 m
M
O
2

-free borate buffer (pH 8.0) that contained 1 m
M
EDTA.
The protein fraction was collected in the small vial, which
waspurgedwithN
2
gas through the rubber cap during the
collection. Then 100 lL of the enzyme solution were
incubated with 50 m
M
2-mercaptoethanol at 25 °Cfor
30 min, and evaporated to dryness. The residue was
dissolved in 50 lL of distilled water. The 2-mercaptoetha-
nol-free solution was mixed with 50 lLof1m
M
4-fluoro-
7-sulfamoylbenzofurazan in 0.1
M
borate buffer (pH 8.0)
and incubated at 50 °C for 5 min. After incubation, the
mixture was cooled on ice and acidified by addition of
30 lLof0.1
M
HCl. The acidified solution was applied to a
column (2.3 · 250 mm) of Wacosil-II 5C18 HG (Wako,
Osaka, Japan), which was part of an HPLC system and
which had been equilibrated with a mixture of 50 m
M
potassium hydrogen phthalate (pH 4.0) and acetonitrile
(92 : 8, v/v). The column was developed with the same

solution at a flow rate of 1.0 mLÆmin
)1
. The GSH that had
been modified by 4-fluoro-7-sulfamoylbenzofurazan was
detected fluorometrically with an excitation at 380 nm and
an emission at 510 nm.
Site-directed mutagenesis of DHAR
Three C residues (C9, C23, C26) in mature DHAR from
spinach chloroplasts were individually mutated at one or
two positions. The plasmid pET3a-DHAR [15] was digested
with SacIandXbaI. The excised DNA fragment containing
the cDNA for part of the mature form of spinach
chloroplast DHAR was inserted at the SacI–XbaIsitein
pUC18. Amplification by PCR was carried out with the
pUC18 vector that contained the fragment as template and
the following primers: for C9S, P2 and P3; for C23S, P1 and
P4; for C26S, P1 and P5; and for C9S/C26S, P1, P2, P3 and
P5. The oligonucleotide primers sequences used in PCR
were as follows: P1, 5¢-AGCTTGTTGGGGGTGGT
GAC-3¢;P2,5¢-AATCTGTCACCACCCCCAAC-3¢;P3,
5¢-CCTTGACG
GATATTTGGAGTG-3¢;P4,5¢-TGGCG
ATT
CTCCATTTTGCCAAAGAGTG-3¢;andP5,5¢-TG
GCGATTGTCCATTTT
CCCAAAGAGTG-3¢.Mutated
bases are underlined. The PCR products were phosphoryl-
ated and self-ligated. After mutation of DHAR, the
922 T. Shimaoka et al.(Eur. J. Biochem. 270) Ó FEBS 2003
sequences of the DNA fragments were confirmed by

nucleotide sequencing with vector primers, M13 reverse
and forward primers, and a Thermo Sequenase II dye
terminator cycle-sequencing Premix kit (Amersham Phar-
macia) with an automated DNA sequencer (model 373;
Applied Biosystems, CA, USA). The mutated DNA
fragments were used individually to replace the SacI-XbaI
fragment of pET3a-DHAR. Proteins encoded by the
vectors that included cDNAs for mutant DHARs,
namely pDHAR-C9S, pDHAR-C23S, pDHAR-C26S and
pDHAR-C9S/C26S, were expressed in E. coli BL21 (DE3).
Recombinant enzymes were purified as described above.
SDS/PAGE
SDS/PAGE was performed with 12.5% polyacrylamide
gels as described by Laemmli [20]. Proteins on the gels were
stained with Coomassie Brilliant Blue R-250 (Nacalai
tesque, Kyoto, Japan).
Western blotting
After separation by SDS/PAGE of each DHAR, the
protein on the gel was transferred to a poly(vinylidene
difluoride) membrane by a semidry blotting method. A
specific antibody, raised in rabbit against recombinant
spinach chloroplast DHAR, was used at a dilution of
1/2000 in 30 m
M
Tris/HCl (pH 7.5) that contained 200 m
M
NaCl and 5% (w/v) skim milk. Immunoreactive proteins,
which bound with antibodies against rabbit IgG that had
been conjugated to horseradish peroxidase, on the
poly(vinylidene difluoride) membrane were revealed with

an Immunostaining HRP-1000 kit (Konica, Tokyo, Japan)
according to the manufacturer’s instructions.
Results and discussion
Initial velocity and product inhibition of the reaction
catalyzed by spinach chloroplast DHAR
We purified recombinant DHAR to homogeneity from
E. coli that expressed a cDNA for mature DHAR from
spinach chloroplasts (Fig. 1). Because the K
m
values for
DHA and GSH, and k
cat
of the recombinant DHAR were
the same as those of the DHAR purified from fresh leaves
of spinach [15], we used the recombinant DHAR in this
study.
DHAR catalyzes the reduction of DHA to ascorbate
with GSH as the electron donor [21,22], as follows:
DHA þ 2GSH ! Ascorbate þ GSSG
Thus, the reduction of DHA by DHAR is a ter-bi reaction.
We measured the initial velocity of the reaction catalyzed
by DHAR with the concentration of GSH fixed at 0.2, 0.3,
0.5, 0.8, 1.0, 2.0 or 4.0 m
M
, varying the concentration of
DHA from 0.02 to 0.5 m
M
. We also measured the activity
when we varied the concentration of GSH from 0.2 to
4.0 m

M
with the concentration of DHA fixed at 0.02, 0.03,
0.05, 0.07, 0.1, 0.2 or 0.5 m
M
. Double-reciprocal plots for
activity vs. various concentrations of one substrate yielded
straight lines at the various fixed concentrations of the other
substrate (Fig. 2A,B). The lines crossed in the second
quadrant. The velocity of the reaction catalyzed by DHAR
in the absence of reaction products can be expressed as Eqn
(1) because the DHAR reaction is a ter bi reaction:
v ¼ V
max
½DHA½GSH
2
=fð½DHAÀ½GSHÞ ð1Þ
The numerator of Eqn (1) is represented by the product of
the maximum velocity (V
max
) and the concentrations of
substrates, and the denominator by the function of the
concentrations of substrates. The straight lines of the
double-reciprocal plots in Fig. 2 indicate that the denomi-
nator of Eqn (1) for the DHAR-catalyzed reaction does not
include a constant, and show that the DHAR-catalyzed
reaction proceeds via a ping-pong mechanism. Two mech-
anisms have been proposed for ping-pong-ter-bi reactions:
bi-uni-uni-uni-ping-pong and uni-uni-bi-uni-ping-pong [23].
By replotting the slopes and the intercepts with the y-axis
of the lines in Fig. 2 against the reciprocals of the

concentrations of the substrates (Fig. 2, insets a–d), straight
lines are obtained. In the case of ping-pong mechanisms,
Eqn (1) is transformed as follows:
v ¼ V
max
 f
1
ð½GSHÞ Â ½DHA=f½DHAþf
2
ð½GSHÞg
ð2AÞ
¼ V
max
 f
1
ð½DHAÞ Â ½GSH=f½GSHþf
2
ð½DHAÞg
ð2BÞ
Fig. 1. SDS/PAGE and Western blotting analysis of purified wild type
and mutant forms of DHAR. Recombinant wild type and mutant forms
of DHAR were produced in E. coli and purified as described in
ÔExperimental proceduresÕ. The purified wild type and mutant enzymes
were subjected to SDS/PAGE on a 12.5% polyacrylamide gel. The
proteins on the gel were stained with Coomassie Brilliant Blue R-250
(lanes 1–5) or transferred to a poly(vinylidene difluoride) membrane
for Western blotting analysis (lanes 6–10). Each lane was loaded with
2 lg of enzyme for staining with Coomassie Brilliant Blue and 10 ng of
protein for Western blotting. M, Molecular mass markers; lanes 1 and
6, wild type DHAR; lanes 2 and 7, C9S DHAR; lanes 3 and 8, C23S

DHAR; lanes 4 and 9, C26S DHAR; and lanes 5 and 10, C9S/C26S
DHAR. The following proteins were used as molecular mass markers:
phosphorylase b (94 kDa), bovine serum albumin (67 kDa), ovalbu-
min (43 kDa), carbonic anhydrase b (30 kDa), trypsin inhibitor
(20.1 kDa), and a-lactalbumin (14.4 kDa).
Ó FEBS 2003 Catalytic mechanism of chloroplast DHAR reaction (Eur. J. Biochem. 270) 923
The straight lines in insets a–d of Fig. 2 indicate that both
1/f
1
([GSH]) and f
2
([GSH])/f
1
([GSH]) are represented by the
linear function of the reciprocals of the concentrations of
GSH and, moreover, that both 1/f
1
([DHA]) and f
2
([DHA])/
f
1
([DHA]) are represented by the linear function of the
reciprocals of the concentrations of DHA. Therefore, it
appears that the reaction catalyzed by DHAR proceeds via
a bi-uni-uni-uni-ping-pong mechanism and that the sub-
strate that binds last is GSH. The last-binding substrate of
GSH means that the first product is ascorbate and that
GSSG is the final product in the catalytic cycle of the
DHAR reaction, as GSSG is produced from two molecules

of GSH.
We performed product-inhibition studies to confirm the
order of binding of DHA and GSH to DHAR. Double-
reciprocal plots of reaction velocity against the concentra-
tion of DHA in the presence of 4.0 m
M
GSH and 0, 2.0 or
5.0 m
M
GSSG gave straight lines that pivoted counter-
clockwise on the point at which the lines intersected
(Fig. 3A). Double-reciprocal plots of velocity vs. the
concentration of GSH in the presence of 0.5 m
M
DHA
and 0, 2.0 or 5.0 m
M
GSSG yielded parabolic curves
(Fig. 3B). These results indicate that GSSG acts as a
competitive inhibitor with respect to DHA and as a mixed-
type inhibitor with respect to GSH. The competitive
inhibition by GSSG with respect to DHA is consistent with
a mechanism in which DHA binds first to the enzyme
(Fig. 4).
The activity of spinach chloroplast DHAR was inhibited
by incubation with iodoacetic acid and such inhibition was
suppressed by the addition of DHA [15], suggesting that a C
residue of DHAR might interact with DHA. In the
chemical reaction between DHA and GSH, glutathionyl
Fig. 3. Inhibition of the DHAR-catalyzed reaction by GSSG at various

concentrations of DHA (A) and GSH (B). The concentrations of GSH
and DHA were 4.0 m
M
in (A) and 0.5 m
M
in (B). The concentrations
of GSSG were 0 m
M
(d), 2.0 m
M
(m)and4.0m
M
(j). Each reaction
mixture contained 20 ng of enzyme. For other details, see Experi-
mental procedures.
Fig. 2. Double-reciprocal plots of the initial velocity vs. the concentra-
tion of one substrate at various fixed concentrations of the other sub-
strate. (A) Reciprocals of initial rates of reduction of DHA are plotted
against the reciprocals of concentrations of DHA at several fixed
concentrations of GSH. The concentrations of GSH were 0.3 m
M
(d),
0.5 m
M
(j), 0.8 m
M
(m), 1.0 m
M
(s), 2.0 m
M

(h), 3.0 m
M
(n)and
4.0 m
M
(e). In insets (a) and (b), the slope and the intercept on the
y-axis, respectively, are replotted against the reciprocals of the con-
centrations of GSH. (B) Reciprocals of initial rates of reduction of
DHA were plotted against the reciprocals of concentrations of GSH at
fixed concentrations of DHA. The concentrations of DHA were
0.02 m
M
(d), 0.03 m
M
(j), 0.05 m
M
(m), 0.07 m
M
(s), 0.1 m
M
(h),
0.2 m
M
(n)and0.5m
M
(e). In insets (c) and (d), the slope and the
intercept on the y-axis, respectively, are replotted against the recipro-
cals of the concentrations of DHA. Each reaction mixture contained
20 ng of enzyme. For other details, see Experimental procedures.
924 T. Shimaoka et al.(Eur. J. Biochem. 270) Ó FEBS 2003

hemiketal is formed first as a reaction intermediate [24].
Therefore, it is likely that a cysteinyl-thiohemiketal complex
is formed between DHA and the sulfhydryl group of a C
residue of DHAR in its reduced form.
The most plausible reaction mechanism that incorporates
our steady-state kinetic studies seems to be the bi-uni-uni-
uni-ping-pong mechanism (Fig. 4). The sulfhydryl group of
a C residue in the reduced enzyme, E-S
–
, reacts with DHA.
The reducing equivalents of the sulfhydryl groups of the C
residue reduce DHA to ascorbate, and generate the oxidized
form of the enzyme, E-S-SG. A second molecule of GSH
then reduces E-S-SG to generate E-S
–
and GSSG.
We calculated the kinetic parameters, as described in the
Appendix, and summarized in Table 1. The calculated K
GSH
m
was the sum of the values of K
m
for the first-binding molecule
of GSH and the second-binding molecule of GSH since
we were unable to calculate separately K
GSH1
m
and K
GSH2
m

.
The mechanism of catalysis suggests that the differences
in specific activity between spinach chloroplast DHAR and
other DHA-reducing proteins, such as the thioltransferase
from pig liver and trypanothione:glutathione disulfide
thioltransferase from T. cruzi, might be due to differences
in the mechanisms of catalysis. The mechanism proposed
for T. cruzi enzyme [19] includes the formation of gluta-
thionyl-thiohemiketal by DHA and GSH [24] on the
enzyme. Our steady-state kinetic studies for spinach
chloroplast DHAR do not support such a mechanism
(Figs 2 and 3). In contrast, the reduction of DHA to
ascorbate by C23S DHAR had the DHAR activity similar
to that of the T. cruzi enzyme, and may proceed via the
reaction mechanism of the T. cruzi enzyme.
Two mechanisms were proposed for the pig liver enzyme
[18]. One is the same as that for spinach chloroplast DHAR.
However, it is not clear, because neither the steady-state
kinetics nor the structure of the reaction intermediate has
been examined with the pig liver enzyme. In the other
mechanism, an intramolecular disulfide bond was proposed
to be formed in the enzyme during the DHA-reducing
reaction. However, mutation of C9 and C26 to S residues
results in the appearance of all DHAR activity in the present
study (Table 1).
Detection of the oxidized form enzyme, E-S-SG
To detect E-S-SG, we performed the following experiment.
We reacted E-S
–
with an excess of DHA to generate a

cysteinyl-thiohemiketal complex, E-S-DHA. Then, we
incubated E-S-DHA in 2-mercaptoethanol-free medium
with equimolar GSH to that of the enzyme. E-S-SG was
freed of residual DHA and GSH by gel filtration and then
excess 2-mercaptoethanol was added to E-S-SG to reduce
the disulfide bond that had formed between the enzyme and
GSH. The GSH released from E-S-SG was detected by
HPLC as described in Experimental procedures. When E-S
–
was reacted with excess DHA, we detected GSH with a
retention time of 7.4 min (Fig. 5). The detected GSH was
5.8% of the reacted enzyme when we quantified them by the
standard addition method. No GSH was detected at this
retentiontimewhentheenzymewasreactedwithGSHonly
(Fig. 5). The results indicate that the reduced form of
DHAR reacted first with DHA and then E-S-DHA reacted
with GSH to generate E-S-SG. The low yield of E-S-SG
might be due to the higher rate of the reaction between
E-S-SG and the second GSH than that of the reaction
between E-S-DHA and the first GSH.
Identification of the C residue involved in the reaction
catalyzed by chloroplast DHAR
Spinach chloroplast DHAR contains three C residues,
namely C9, C23 and C26. C9 and C23 are conserved in
plant DHARs. C26 is conserved in spinach chloroplast and
in Arabidopsis DHARs but is replaced to the S residue in
rice bran DHAR (Fig. 6). We purified the mutated DHARs
that had been expressed in E. coli.SDS/PAGErevealedthat
each enzyme had been purified to homogeneity, and the
purified enzymes were confirmed to be forms of DHAR by

Western blotting with antibodies specific for spinach
chloroplast DHAR (Fig. 1). The kinetic parameters of
wild type and mutant DHARs were
1
described as above.
C23S DHAR had almost no activity (Table 1). The k
cat
of
C26Swashalfthatofthewild-typeDHAR,whilethek
cat
of
Fig. 4. The most plausible mechanism of the reaction catalyzed by
dehydroascorbate reductase from spinach chloroplasts. E-S
–
and E-S-S-G
mean the reduced form and the oxidized form enzymes, respectively
[19].
Table 1. Kinetic parameters of wild-type and mutant forms of DHAR from spinach chloroplasts. Values of k
cat
were calculated using a molecular mass
of 24 kDa. Values of k
cat
and K
m
are given as means ±SD (n ¼ 3–5).
k
cat
(s
)1
) K

m
for DHA (l
M
) K
m
for GSH (m
M
) k
cat
=K
DHA
m
(
M
)1
Æs
)1
) k
cat
=K
GSH
m
(
M
)1
Æs
)1
)
Wild-type 490 ± 40 53 ± 12 1.1 ± 0.5 (9.2 ± 2.1)Æ10
6

(5.2 ± 1.9)Æ10
5
C9S 420 ± 30 19 ± 14 0.95 ± 0.07 (3.1 ± 2.4)Æ10
7
(4.5 ± 0.6)Æ10
5
C23S <1 – – – –
C26S 280 ± 30 26 ± 6 0.69 ± 0.2 (1.1 ± 0.3)Æ10
7
(4.2 ± 0.9)Æ10
5
C9/26S 210 ± 10 58 ± 7 1.1 ± 0.3 (3.7 ± 0.3)Æ10
6
(2.0 ± 0.5)Æ10
5
Ó FEBS 2003 Catalytic mechanism of chloroplast DHAR reaction (Eur. J. Biochem. 270) 925
C9S was slightly lower than that of wild-type DHAR. The
k
cat
of C9S/C26S was 210 ± 10 s
)1
. The respective K
m
values for the two substrates of wild type C9S, C26S and
C9S/C26S DHARs, were almost identical. These results
indicate that C23 is essential for spinach chloroplast DHAR
to have the high specific activity and suggest that this
residue may be involved in the formation of the disulfide
bond with GSH in the cysteinyl-thiohemiketal. Other C
residues might also be involved in the reaction of the

spinach enzyme, but their contributions were not significant
(Table 1).
Influence of reaction products on the activity of DHAR
In a previous paper [15], we discussed the ability of spinach
chloroplast DHAR to function as an ascorbate-regener-
ating enzyme in vivo. In the present study, we analyzed the
effects of product-inhibition on the activity of DHAR to
clarify the possible effects of reaction products on the
activity of DHAR in vivo. Spinach chloroplasts contain
12–25 m
M
ascorbate [25,26] and 3–4 m
M
glutathione
[26,27]. More than 90% of the ascorbate and a similar
percentage of glutathione are found in the reduced forms
under nonstress conditions. Thus, the concentrations of
DHA and GSSG might be 2.5 m
M
and 0.4 m
M
at
maximum, respectively. When we assayed the activity of
spinach chloroplast DHAR in the presence of 4.0 m
M
GSH
and 0.5 m
M
DHA, the activity decreased by 75% upon
addition of 2.0 m

M
GSSG, a concentration of GSSG that is
much higher than that in chloroplasts (Fig. 3). Considering
this result, we can speculate that the inhibition of DHAR
activity by GSSG might not affect the activity of spinach
chloroplast DHAR in vivo, under conditions where the
concentrations of GSH and GSSG are 4 m
M
and 0.4 m
M
,
respectively. By contrast, when we assayed the activity of
chloroplast DHAR in the presence of 1.0 m
M
GSH and
0.1 m
M
DHA, the activity decreased by 40% upon addition
of 20 m
M
ascorbate (T. Shimaoka, C. Miyake & A. Yokota,
unpublished results). This finding suggests that ascorbate
lowers the activity of DHAR in spinach chloroplasts, in
which the concentrations of ascorbate and DHA are 25 m
M
and 2.5 m
M
, respectively. In our earlier estimate of the rate
of formation of superoxide at photosystem I [28], we
proposed that MDA would be formed at a rate of

300 lmolÆmg chlorophyll
)1
Æh
)1
in the reaction catalyzed by
APX for decomposition of hydrogen peroxide at a light
intensity of 1400 lmol photonsÆm
)2
Æs
)1
in air. Most of the
MDA formed in the water-water cycle is directly reduced to
ascorbate by ferredoxin [1]. If 10% of the MDA were
disproportionated to ascorbate and DHA, the rate of
formationofDHAwouldbe15 lmolÆmg chlorophyll
)1
Æh
)1
.
This rate corresponds closely to 20% of the maximum
activity of chloroplast DHAR that we measured in our
previous study [15]. Therefore, it appears that DHAR can
reduce all available DHA to ascorbate under nonstress
conditions, even if the maximum activity of DHAR is
inhibited by 40% by ascorbate, the concentration of which
might range from 12 to 25 m
M
. However, we cannot ignore
the possibility that the activity of DHAR, in terms of the
regeneration of ascorbate, might be limited under stress

conditions, where the rate of production of DHA is elevated.
Acknowledgements
This study was partly supported by the Petroleum Energy Center and
the Research Association for Biotechnology subsidized by the Ministry
of Economy, Trade and Industry of Japan.
References
1. Asada, K. (1999) The water-water cycle in chloroplasts: scaven-
ging of active oxygens and dissipation of excess photons. Annu.
Rev. Plant Physiol. Plant Mol. Biol. 50, 601–639.
2. Niyogi, K.K. (1999) Photoprotection revisited: genetic and
molecular approaches. Annu. Rev. Plant Physiol. Plant Mol. Biol.
50, 333–359.
3. Jablonski, P.P. & Anderson, J.W. (1981) Light-dependent reduc-
tion of dehydroascorbate by ruptured pea chloroplasts. Plant
Physiol. 67, 1239–1244.
4. Nakano, Y. & Asada, K. (1981) Hydrogen peroxide is scavenged
by ascorbate-specific peroxidase in spinach chloroplasts. Plant Cell
Physiol. 22, 867–880.
5. Hossain, M.A., Nakano, Y. & Asada, K. (1984) Mono-
dehydroascorbate reductase in spinach chloroplasts and its parti-
cipation in regeneration of ascorbate for scavenging hydrogen
peroxide. Plant Cell Physiol. 25, 385–395.
Fig. 6. The amino acid sequences around the C-X-X-C motif in spinach
chloroplast DHAR and DHA-reducing enzymes from other organisms.
White letters in black boxes represent C residues of the following
sequences: spinach chloroplast DHAR [15], Arabidopsis chloroplast
DHAR [15], rice bran DHAR [30], bovine protein disulphide iso-
merase (PDI) [31], T. cruzi trypanothione-glutathione thioltransferase
(p52) [32], rat liver DHAR [33], rice glutaredoxin (Grx) [12], and pig
liver thioltransferase [34].

Fig. 5. Detection of DHAR-bound GSH by HPLC. The 2-mercapto-
ethanol-reduced enzyme (10 l
M
), from which 2-mercaptoethanol had
been removed under anaerobic conditions was reacted with 1 m
M
DHA and 10 l
M
GSH (solid line) or 10 l
M
GSH (dashed line) at room
temperature for 30 s in 50 m
M
potassium phosphate (pH 7.0) that
contained 1 m
M
EDTA under aerobic conditions. For other details
2
,
see Experimental procedures.
926 T. Shimaoka et al.(Eur. J. Biochem. 270) Ó FEBS 2003
6. Miyake, C. & Asada, K. (1994) Ferredoxin-dependent photo-
reduction of the monodehydroascorbate radical in spinach
thylakoids. Plant Cell Physiol. 35, 539–549.
7. Wells, W.W., Xu, D.P., Yang, Y. & Rocque, P.A. (1990) Mam-
malian thioltransferase (glutaredoxin) and protein disulfide iso-
merase have dehydroascorbate reductase activity. J. Biol. Chem.
265, 15351–15364.
8. Kato, Y., Urano, J., Maki, Y. & Ushimaru, T. (1997) Purification
and characterization of dehydroascorbate reductase from rice.

Plant Cell Physiol. 38, 173–178.
9. Hossain, M.A. & Asada, K. (1984) Purification of dehydro-
ascorbate reductase from spinach and its characterization as a
thiol enzyme. Plant Cell Physiol. 25, 85–92.
10. Dipierro, S. & Borraccino, G. (1991) Dehydroascorbate reductase
from potato tubers. Phytochemistry 30, 427–429.
11. Xu, D.P., Washburn, M.P., Sun, G.P. & Wells, W.W. (1996)
Purification and characterization of a glutathione dependent
dehydroascorbate reductase from human erythrocytes. Biochem.
Biophys. Res. Commun. 221, 117–121.
12. Sha, S., Minakuchi, K., Higaki, N., Sato, K., Ohtsuki, K., Kurata,
A., Yoshikawa, H., Kotaru, M., Masumura, T., Ichihara, K. &
Tanaka, K. (1997) Purification and characterization of gluta-
redoxin (thioltransferase) from rice (Oryza sativa L.). J. Biochem.
(Tokyo) 121, 842–848.
13. Washburn, M.P. & Wells, W.W. (1999) Identification of the
dehydroascorbic acid reductase and thioltransferase (Gluta-
redoxin) activities of bovine erythrocyte glutathione peroxidase.
Biochem. Biophys. Res. Commun. 257, 567–571.
14. Vethanayagam, J.G., Green, E.H., Rose, R.C. & Bode, A.M.
(1999) Glutathione-dependent ascorbate recycling activity of rat
serum albumin. Free Rad. Biol. Med. 26, 1591–1598.
15. Shimaoka, T., Yokota, A. & Miyake, C. (2000) Purification and
characterization of chloroplast dehydroascorbate reductase from
spinach leaves. Plant Cell Physiol. 41, 1110–1118.
16. Maellaro, E., Del Bello, B., Sugherini, L., Comporti, M. & Casini,
A.F. (1997) Purification and characterization of glutathione-
dependent dehydroascorbate reductase from rat liver. Methods
Enzymol. 279, 30–35.
17. Yang, Y. & Wells, W.W. (1991) Identification and characteriza-

tion of the functional amino-acids at the active center of pig liver
thioltrasferase by site-directed mutagenesis. J. Biol. Chem. 266,
12759–12765.
18. Washburn, M.P. & Wells, W.W. (1999) The catalytic mechanism
of the glutathione-dependent dehydroascorbate reductase activity
of thioltransferase (glutaredoxin). Biochemistry 38, 268–274.
19. Moutiez, M., Quemeneur, E., Sergheraert, C., Lucas, V.,
Tartar, A. & Davioud-Charvet, E. (1997) Glutathione-dependent
activities of Trypanosoma cruzi p52makesitanewmemberof
the thiol:disulphide oxidoreductase family. Biochem. J. 322,
43–48.
20. Laemmli, U.K. (1970) Cleavage of structural proteins during the
assembly of the head of bacterophage T4. Nature 227, 680–685.
21. Mapson, L.W. (1958) Metabolism of ascorbic acid in plants:
function. Annu. Rev. Plant Physiol. 9, 119–150.
22. Foyer, C.H. & Halliwell, B. (1977) Purification and properties of
dehydroascorbate reductase from spinach leaves. Phytochemistry
16, 1347–1350.
23. Segel, I.H. (1975) Enzymatic Kinetics: Behavior and Analysis of
Rapid Equilibrium and Steady-State Enzyme Systems. Wiley, New
York, USA.
24. Winkler, B.S., Orselli, S.M. & Rex, T.S. (1994) The redox couple
between glutathione and ascorbic acid: a chemical and physiolo-
gical perspective. Free Rad. Biol. Med. 17, 333–349.
25. Foyer, C., Rowell, J. & Walker, D. (1983) Measurement of the
ascorbate content of spinach leaf protoplasts and chloroplasts
during illumination. Planta 157, 239–244.
26. Law, M.Y., Charles, S.A. & Halliwell, B. (1983) Glutathione and
ascorbic acid in spinach (Spinacia oleracea) chloroplasts. Biochem.
J. 210, 899–903.

27. Foyer, C.H. & Halliwell, B. (1976) The presence of glutathione
and glutathione reductase in chloroplasts: a proposed role in
ascorbic acid metabolism. Planta 133, 21–25.
28. Miyake, C. & Yokota, A. (2000) Determination of the rate of
photoreduction of O
2
in the water-water cycle in watermelon
leaves and enhancement of the rate by limitation of photosynth-
esis. Plant Cell Physiol. 41, 335–343.
29. King, E.L. & Altman, C. (1956) A schematic method of deriving
the rate laws for enzyme-catalyzed reactions. J. Phys. Chem. 60,
1375–1378.
30. Urano, J., Nakagawa, T., Maki, Y., Masumura, T., Tanaka, K.,
Murata, N. & Ushimaru, T. (2000) Molecular cloning and char-
acterization of a rice dehydroascorbate reductase. FEBS Lett. 466,
107–111.
31. Yamauchi, K., Yamamoto, T., Hayashi, H., Koya, S., Takikawa,
H., Toyoshima, K. & Horiuchi, R. (1987) Sequence of membrane-
associated thyroid hormone binding protein from bovine liver: Its
identity with protein disulphide isomerase. Biochem. Biophys. Res.
Commun. 146, 1485–1492.
32. Schoneck, R., Plumas-Marty, B., Taibi, A., Billaut-Mulot, O.,
Loyens,M.,Gras-Masse,H.,Capron,A.&Ouaissi,A.(1994)
Trypanosoma cruzi cDNA encodes a tandemly repeated domain
structure characteristic of small stress proteins and glutathione
S-transferase. Biol. Cell 80, 1–10.
33. Ishikawa, T., Casini, A.F. & Nishikimi, M. (1998) Molecular
cloning and functional expression of rat liver glutathione-depen-
dent dehydroascorbate reductase. J. Biol. Chem. 273, 28708–
28712.

34. Yang, Y., Gan, Z R. & Wells, W.W. (1989) Cloning and
sequencing the cDNA encoding pig liver thioltransferase. Gene 83,
339–346.
Appendix
Mathematical representations of the kinetic model
According to the King–Altman method [29], the reaction
scheme can be drawn as Fig. 4. The complete rate
equation for the bi uni uni uni ping pong is obtained from
the above five, four-sided King–Altman interconversion
patterns is:
in the absence of reaction products,
m
V
max
¼
½DHA½GSH
2
K
DHA
i
K
GSH1
m
½GSHþðK
GSH1
m
þ K
GSH2
m
Þ½DHA½GSHþK

DHA
m
½GSH
2
þ½DHA½GSH
2
ðA1Þ
Ó FEBS 2003 Catalytic mechanism of chloroplast DHAR reaction (Eur. J. Biochem. 270) 927
and in the presence of reaction products,
m ¼
V
maxf
V
maxr
½DHA½GSH
2
À
AsA
½
GSSG
½
K
eq

V
r
K
DHA
i
K

GSH1
m
½GSHþV
r
K
GSH2
m
½DHA½GSHþV
r
K
GSH1
m
½DHA½GSH
þV
r
K
DHA
m
½GSH
2
þ V
r
½GSH
2
þ
V
f
K
GSSG
m

½DHA½AsA
K
DHA
i
K
eq
þ
V
f
K
GSSG
m
½DHA½GSH½AsA
K
DHA
i
K
GSH1
i
K
eq
þ
V
f
K
GSSG
m
½AsA
K
eq

þ
V
f
K
AsA
m
½GSSG
K
eq
þ
V
f
½AsA½GSSG
K
eq
þ V
r
K
DHA
m
K
GSH2
i
½GSH½GSSG
þ
V
r
K
DHA
i

K
GSH1
i
½GSH½GSSG
K
GSSG
i
þ
V
r
K
DHA
m
K
GSH2
i
½GSH½AsA½GSSG
K
AsA
i
K
GSSG
i
þ
V
r
K
DHA
m
½GSH

2
½GSSG
K
GSSG
i
ðA2Þ
Determination of kinetic parameters
We determined Michaelis constants and k
cat
from our
initial-velocity experiments using Eqn (A2). We can trans-
form Eqn (A2) as follows:
v ¼
V
max
½GSH
½GSHþK
GSH1
m
þ K
GSH2
m
½DHA
½DHAþ
K
DHA
i
K
GSH1
m

þ K
DHA
m
½GSH
K
GSH1
m
þ K
GSH2
m
½GSH
ðA3Þ
The plots of initial velocity at various fixed concentrations
of GSH and varying concentrations of DHA were fitted
to the Michaelis–Menten equation
3
v ¼ m
1
· [DHA]/
([DHA] + m
2
). Calculations for m
1
and m
2
were made
by application of the computer program
KALEIDAGRAPH
3.08d, where m
1

and m
2
represent V
max
½GSH=ð½GSHþ
K
GSH1
m
þ K
GSH2
m
Þ and ðK
DHA
i
K
GSH1
m
þ K
DHA
m
½GSHÞ=
ðK
GSH1
m
þ K
GSH2
m
þ½GSHÞ, respectively. To determine
V
max

, K
GSH1
m
þ K
GSH2
m
, K
DHA
i
and K
DHA
m
, we generated
double-reciprocal plots between m
1
and the concentration
of GSH, and double-reciprocal plots between m
1
/m
2
and the
concentration of GSH. The slope and intercept of the former
plot gave V
max
and K
GSH1
m
þ K
GSH2
m

while the slope and
intercept of the latter plot gave K
DHA
i
and K
DHA
m
.
2
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6
4
3
7
7
7

7
7
7
7
7
7
7
7
7
7
7
7
7
5
928 T. Shimaoka et al.(Eur. J. Biochem. 270) Ó FEBS 2003