Enzymatic and mechanistic studies on the formation of N- phenylglycolohydroxamic acid from nitrosobenzene and pyruvate in spinach leaf homogenate

Article abstract of DOI:10.1021/jf051969f

The biotransformation mechanism of an unknown metabolite formed enzymatically from nitrosobenzene (NOB) and pyruvate in spinach (Spinacea oleracea L.) was investigated using spinach leaf homogenate. The unknown metabolite was identified as N-phenylglycolo

Full text of DOI:10.1021/jf051969f

590 J. Agric. Food Chem. 2006, 54, 590−596
Enzymatic and Mechanistic Studies on the Formation of
N-Phenylglycolohydroxamic Acid from Nitrosobenzene and
Pyruvate in Spinach Leaf Homogenate
RYOSUKE TATSUNAMI AND TADAO YOSHIOKA*
Department of Pharmacy, Hokkaido Pharmaceutical University School of Pharmacy,
7-1 Katsuraoka-cho, Otaru, Hokkaido 047-0264, Japan
The biotransformation mechanism of an unknown metabolite formed enzymatically from nitrosoben-
zene (NOB) and pyruvate in spinach (Spinacea oleracea L.) was investigated using spinach leaf
homogenate. The unknown metabolite was identified as N-phenylglycolohydroxamic acid (PGA). The
activity of PGA formation was decreased by L-alanine, increased by L-serine, and completely inhibited
by aminooxyacetic acid, an inhibitor of transaminases. These results indicate that the transaminase
participates in PGA formation. Indeed, hydroxypyruvate and alanine were produced in the transami-
nation between pyruvate and serine. Hydroxypyruvate served as a direct-acting glycoloyl donor for
PGA formation. A good correlation between the activities of the 200g supernatant of spinach
homogenate and commercial yeast transketolase for PGA formation from several glycoloyl donors
was obtained. These results suggest the following mechanism for PGA formation from NOB and
pyruvate: transamination of L-serine into hydroxypyruvate, which serves as a glycoloyl donor to NOB.
KEYWORDS: N-Phenylglycolohydroxamicacid (PGA); transaminase; transketolase; hydroxypyruvate;
Spinacea oleracea L.
INTRODUCTION
Artificial organic compounds (so-called xenobiotics), includ-
ing agricultural chemicals, pharmaceuticals, and other chemi-
cally engineered products, through their widespread use, make
their way into the environment, where they can be found in the
form of countless metabolites. These xenobiotics, as well as
their metabolites, accumulate in plants, which lack effective
excretion systems (1, 2); as a result, the ingestion of plants that
have been exposed to xenobiotics is believed to represent a
toxicological risk. Therefore, information on the biotransfor-
mation of xenobiotics in plants is needed from the viewpoint
of toxicological effects on animals and humans. There are many
reports on the metabolic pathways and metabolites of xenobi-
otics in plants (3): aromatic compounds taken up by green
leaves (4); factors affecting metabolism and absorption of
xenobiotics by green leaves and roots (5); metabolic pathways
including hydroxylation activity of aromatic compounds; dealky-
lation (6); and acetyl (7), glucose (8), and GSH (9) conjugations.
The metabolites and metabolic pathways for agricultural chemi-
cals such as herbicides have also been investigated (10-12).
N-Substituted aromatic compounds, including nitro and/or
amino aromatics, are organic substances that are widely used
in the environment, and some of these are known to be
carcinogenic or mutagenic. We have studied the biotransfor-
mation and toxicology of these compounds, because their
Figure 1. Structures of PGA and PAA.
bioactivation has been considered as a critical step for their
carcinogenicity and mutagenicity (13, 14). In our previous study,
a nonoxidative pathway for the formation of proximate carci-
nogenic N-arylacetohydroxamic acids from nitroso aromatic
compounds and pyruvate, catalyzed by pyruvate dehydrogenase
complex, was found in animals and also in isolated spinach leaf
cells (15). In the spinach experiment with nitrosobenzene (NOB)
and pyruvate, an unknown metabolite was formed in addition
to N-phenylacetohydroxamic acid (PAA) (Figure 1). The aim
of this study, therefore, was to identify the unknown metabolite
and to reveal the mechanism of its formation from NOB and
pyruvate.
MATERIALS AND METHODS
Apparatus. HPLC was performed with a LC-6A (Shimadzu)
equipped with a SPD-6A spectrophotometric detector (Shimadzu). 1H
NMR spectra were recorded on a JNM-GX270 spectrometer (JEOL,
Japan). Mass spectra were recorded on a M-2000 (HITACHI).
Spectrophotometry was carried out on a UV-200S (Shimadzu).
Materials. NOB, obtained from Tokyo Kasei Kogyo Co. (Tokyo,
Japan), was recrystallized from EtOH before use. Sodium pyruvate and
thiamin diphosphate (TPP) were purchased from Merck (Darmstadt,
* To whom correspondence should be addressed. Tel: +81-134-62-1894.
Fax: +81-134-62-5161. E-mail: yoshioka@hokuyakudai.ac.jp.
10.1021/jf051969f CCC: $33.50
© 2006 American Chemical Society
Published on Web 12/22/2005

N-Phenylglycolohydroxamic Acid in Spinach
J. Agric. Food Chem., Vol. 54, No. 2, 2006 591
Germany). L-Serine, magnesium sulfate, D-fructose 6-phosphate diso-
dium salt, and amino acids standard solution type H were purchased
from Wako Chemical Industries (Osaka, Japan). L-Alanine and amino-
oxyacetic acid were purchased from Kanto Kagaku (Tokyo, Japan) and
Tokyo Kasei Kogyo Co. (Tokyo, Japan), respectively. Sodium D-
xylulose 5-phosphate, glyoxylic acid, lithium â-hydroxypyruvate,
spinach glyoxylate reductase, and yeast transketolase were purchased
from Sigma-Aldrich. PGA was synthesized according to a reported
procedure (16). All other chemicals used were of reagent grade.
Preparation of Spinach Homogenate. Spinach (Spinacea oleracea
L.) was obtained from a local market. Leaves were washed and
homogenized for 20 s in a Waring blender with a medium containing
0.32 M mannitol and 50 mM morpholinopropanesulfonic acid (MOPS)-
KOH (pH 6.9). The homogenates were filtered through two layers of
gauze. After centrifugation of the filtrate at 200g for 5min, an aliquot
of the supernatant was centrifuged at 105 000g for 60 min. A portion
of the 105 000g supernatant was dialyzed using a Spectra/Por (MWCO
25000, SPECTRUM) against 200 volumes of homogenization buffer
to remove low molecular weight material.
Hz, exchangeable with D2O), 7.14 (1H, t, J ) 7.4 Hz), 7.38 (2H, t, J
) 7.9 Hz), 7.65 (2H, d, J ) 7.8 Hz), 10.49 (1H, s, exchangeable with
D2O). MS (EI): m/z 167 (M+), 151, 149, 119, and 109 (base peak).
Assay of Hydroxypyruvate and Alanine in the Transamination.
The incubation mixture (final volume, 7.5 mL) consisted of 50 mM
MOPS-KOH (pH 6.9), 10 mM pyruvate, and 10 mM L-serine. The
incubation was initiated by the addition of the dialyzed 105 000g
supernatant of leaf homogenate (1.5 mL) and carried out at 25 °C.
Samples were removed at various times to determine the amounts of
hydroxypyruvate and alanine. Since the transamination was not affected
by TPP and MgSO4, they were omitted from the incubation mixture
used for the assay of PGA formation activity.
(1) Determination of HydroxypyruVate. To an aliquot of the
incubation mixture (300 µL) was added 5.5 mM aminooxyacetic acid
(30 µL) to stop the reaction. Hydroxypyruvate was quantified with
hydroxypyruvate reductase according to the literature (18) with some
modifications. Briefly, the activity was measured by monitoring the
oxidation of NADH spectrophotometrically at 340 nm in 1 mL reaction
mixture consisting of 50 mM MOPS-KOH (pH 6.9), 0.125 mM
NADH, sample solution (50 µL), and 20 µL of hydroxypyruvate
reductase (Sigma-Aldrich as glyoxylate reductase) at 25 °C. Pyruvate
is not a substrate for this enzyme. Additionally, L-serine and amino-
oxyacetic acid did not affect the reductase activity.
(2) Determination of Alanine. To the second aliquot of the incubation
mixture (300 µL), 25 mg of 5-sulfosalicylic acid was added to
precipitate proteins and stop the reaction. After brief centrifugation,
the supernatant (60 µL) was removed and then 20 µL of 1 mol/L NaOH
(20 µL) was added to it for neutralization. An aliquot (50 µL) was
analyzed by an amino acid analysis system (Shimadzu), according to
the operating instructions. After separation on an ion exchange column
(Shim-pack ISC-07/S1504Na) with a sodium hydroxide gradient in
citrate buffer at pH 3.2-10, amino acids were derivatized with
o-phthalaldehyde and then detected at 450 nm by fluorescence excited
at 348 nm. A standard solution containing 17 amino acids including
L-serine and L-alanine was used for the check of the retention times
and of good separation.
PGA Formation Activity Catalyzed by Transketolase. The
activities of PGA formation were assayed with commercial yeast
transketolase and the 200g supernatant of spinach leaf homogenate.
The incubations were carried out according to the assay of PGA
formation activity using known as glycoloyl donors instead of pyruvate,
hydroxypyruvate, D-fructose 6-phosphate, D-xylulose 5-phosphate, and
fructose at concentrations of 4 times each Km except for D-fructose at
10 mM.
Partial Purification of the Spinach Transketolase. The partial
purification of transketolase from spinach leaf homogenates was
performed according to the reported procedure (17). Fractions obtained
by ion exchange column chromatography were assayed by the standard
transketolase test, and the pooled active fractions were used for the
assay of PGA formation activity.
Assay of PGA Formation Activity. The activity of PGA formation
was essentially assayed according to the previous report (15). The assay
medium (final volume, 1 mL) consisted of 50 mM MOPS-KOH (pH
6.9), 0.32 M mannitol, 0.5 mM TPP, 5 mM MgSO4, 10 mM pyruvate,
and 2 mM NOB [added as 25 µL of a bis(2-methoxyethyl) ether
solution]. Each incubation was initiated by the addition of supernatants
of spinach homogenate (200g and 105 000g), dialyzed 105 000g
supernatant, or partially purified transketolase. Incubation was carried
out at 25 °C with shaking (70 strokes/min) in a 2.5 mL screw-cap vial
equipped with a Teflon-faced seal to prevent NOB volatilization. The
reaction was stopped by heating for 15 s in boiling water, and then the
reaction mixture was cooled immediately and kept on ice until HPLC
analysis. For the HPLC analysis, 800 mg of ammonium sulfate was
added to the heat-treated sample and then the mixture was extracted
with 1 mL of diisopropyl ether (purified by passage through a basic
aluminum oxide column) saturated with the buffer by shaking for 10
min. After brief centrifugation, an aliquot (50 µL) was analyzed by
HPLC.
HPLC Analysis. HPLC analysis was carried out using a LiChrosorb
RP-8 column (Merck, 4 × 250 mm) with detection at 260 nm. CH3-
CN:CH3OH:H2O containing 0.01% desferal mesylate (16) was used
as the mobile phase, at a flow rate of 1 mL/min and at 40 °C. The
mobile phase was a gradient of solvent A [CH3CN:CH3OH:H2O (1.5:
1:18, v/v/v)] and solvent B [CH3OH:H2O (4:1, v/v)] with 0-12 min
100% A, 12-15 min linear gradient to 100% B, 15-20 min 100% B,
and 20-23 min linear gradient to 100% A.
Calculation of Kinetic Constants. Data obtained from the initial
velocity studies on PGA formation were plotted in a double-reciprocal
form to check the fitting with the Michaelis-Menten equation. The
best-fit values of Km and Vmax were obtained by the method of least-
squares with the Taylor expansion (19).
RESULTS
Identification of the Metabolite as PGA. For the isolation of PGA
from a preparative scale of incubation mixture (100 mL), 70 g of
ammonium sulfate was added after 21 h of incubation, and then the
resultant mixture was extracted with 200 mL of diethyl ether. From
the ether solution, PGA was then extracted with 0.1 mol/L NaOH (50
mL) and the aqueous solution was washed with diethyl ether (2 × 10
mL). After the pH of the solution was adjusted to ca. 3 with 5 mol/L
phosphoric acid, the resultant solution was extracted with diethyl ether
(5 × 20 mL). The extract was evaporated under reduced pressure. Final
purification of PGA (Rf ) 0.38) was carried out by a preparative silica
gel TLC (Merck, article 1.13895) with ethyl acetate:benzene (3:1, v/v)
as a developing solvent. Spectral data of the metabolically formed PGA
Structure Confirmation of the Unknown Metabolite. As
in the isolated cell experiment (15), the unknown metabolite
was also formed in an incubation mixture consisting of NOB,
pyruvate, and the 200g supernatant of spinach leaf homogenate.
MS spectral analysis of the isolated metabolite revealed M+ at
m/z 167 and a fragment ion (M - 18)+ at m/z 149. Another ion
equivalent to (M - 16)+ at m/z 151 seems to be the characteristic
one observed for hydroxamic acid compounds (20). On the basis
of the result of high-resolution mass spectrometry, the molecular
1
formula of the metabolite was presumed to be C8H9NO3. H
NMR spectrum of the metabolite showed signals of two protons
(δ 4.76 and 10.49, each 1H) exchangeable with D2O, as well
as aromatic protons (5H). The spectrum also showed a doublet
signal of methylene protons (δ 4.30, 2H), which collapsed to a
singlet signal by the addition of D2O. These results indicate
the metabolite to be PGA (Figure 1).
1
were as follows. H NMR (DMSO-d6): δ 4.30 (2H, d, J ) 6.1 Hz),
4.75 (1H, t, J ) 6.2 Hz, exchangeable with D2O), 7.14 (1H, t, J ) 7.4
Hz), 7.38 (2H, t, J ) 8.1 Hz), 7.65 (2H, d, J ) 7.6 Hz), 10.48 (1H, s,
exchangeable with D2O). MS (EI): m/z 167 (M+), 151, 149, 119, and
109 (base peak). High-resolution MS: calcd for C8H9NO3, 167.0581;
found, 167.0564 (error -1.7 millimass unit).
Therefore, PGA was chemically synthesized according to the
reported procedure (16), and comparison of authentic PGA with
The spectral data of PGA, synthesized chemically, were as follows.
1H NMR (DMSO-d6): δ 4.30 (2H, d, J ) 5.9 Hz), 4.76 (1H, t, J ) 6.2

592 J. Agric. Food Chem., Vol. 54, No. 2, 2006
Tatsunami and Yoshioka
Figure 2. Time courses of PGA and PAA formations in 200g supernatant.
Figure 3. Time courses of PGA formation in the 105 000g supernatant.
PGA formation in the presence of pyruvate (
b
) and in the absence of
) and in the
PGA formation with the 105 000g supernatant in the presence of pyruvate
(b) and in the absence of pyruvate (O); PGA formation with dialyzed
pyruvate ( ); PAA formation in the presence of pyruvate (
O
9
absence of pyruvate (
of 10 mM.
0
). Pyruvate was added at the final concentration
105 000g supernatant in the presence or absence of pyruvate (
Pyruvate was added at the final concentration of 10 mM.
9).
1
Table 1. Effects of
L-Serine and
L-Alanine on PGA Formation
the metabolite proved its identity by producing consistent H
NMR and MS spectra. In addition, the retention times in HPLC
and the Rf values in TLC for these compounds were also
consistent, and both compounds reacted positively to ferric
chloride (hydroxamic acid-iron reaction). On the basis of all
these findings, the unknown metabolite was identified as PGA.
Pyruvate-Dependent Formation of PGA from Spinach
Leaf Homogenate. In comparison with the complete incubation
mixture, PGA formation activity was decreased by approxi-
mately 70% in the absence of pyruvate. PGA formation was
not observed when the 200g supernatant was heat-treated for
15 s in a boiling water bath (data not shown). The time courses
of the formation of PGA and PAA were investigated under
conditions with and without the addition of pyruvate using the
200g supernatant. The formation of PAA was almost completely
dependent on pyruvate, while PGA formation was observed in
the absence of pyruvate, as mentioned above. Furthermore,
although PAA formation increased in proportion to the incuba-
tion time, PGA formation increased at an accelerating rate
(Figure 2). A proportional relationship was observed between
the amount of supernatant added and PAA formation; PGA
formation was not linear but parabolic (data not shown).
The activity of PGA formation in the 200g supernatant was
further fractionated by a differential centrifugation. Approxi-
mately 80% of the activity of the 200g supernatant remained
in the 105 000g supernatant (data not shown). Figure 3 shows
the time courses of PGA formation catalyzed by the 105 000g
supernatant both in the presence and absence of pyruvate. PGA
formation was shown to be pyruvate-dependent, similar to the
results obtained with the 200g supernatant (Figure 2). With the
dialyzed 105 000g supernatant, however, no PGA formation was
observed, even when pyruvate was added.
relative amount
(%) of PGA
formationa
additive
no additive (control)
100
390
60
L
-serineb
L
-alanineb
a Values are expressed as the percentage of the control value for the amount
of PGA formed with pyruvate during the 30-min incubation. b Each amino acid
was added at the final concentration of 10 mM to the incubation mixture containing
10 mM pyruvate and 105000g supernatant.
hydroxypyruvate is formed. As shown in Table 1, the results
revealed a 4-fold increase in the activity when L-serine was
added and a 60% decrease for the addition of L-alanine.
Various amino acids, including L-serine, were detected in the
105 000g supernatant of spinach leaf homogenate by amino acid
analysis (data not shown), as reported by others (22). Since these
results strongly suggest transamination between pyruvate and
L-serine, the formation of hydroxypyruvate and L-alanine was
measured by using the dialyzed 105 000g supernatant. As shown
in Figure 4, formation of hydroxypyruvate with concomitant
formation of alanine was observed. When aminooxyacetic acid
(AOA), an inhibitor of transaminase (23, 24), was added to the
incubation mixture after 20 min, the formation of both hydroxy-
pyruvate and alanine was stopped.
Effects of Amino Acids and r-Oxo Acids in Transami-
nation. Using the dialyzed 105 000g supernatant, PGA forma-
tion activity was measured with some amino acids and R-oxo
acids. The formation of PGA in the presence of pyruvate and
L-serine was inhibited by the addition of D-serine, an inhibitor
of serine:glyoxylate transaminase (25), but no decrease in the
activity of PGA formation was observed with the addition of
L-glutamate or L-aspartate (data not shown). Table 2 shows the
highest PGA formation activity (13.5-fold increase) with gly-
oxylate in the presence of L-serine. The amount of PGA formed
by the addition of other R-oxo acids was less than 30% of that
observed with pyruvate. These results indicate that the spinach
transaminase(s) participating in hydroxypyruvate formation has
high specificity to glyoxylate as the amino acceptor.
Participation of Transamination in PGA Formation. On
the basis of the results obtained and the chemical structure of
PGA, it is difficult to believe that pyruvate serves as the direct-
acting substrate for PGA formation. An intrinsic dialyzable
component, which can react with pyruvate, might participate
in PGA formation. Since pyruvate was proven to serve as the
direct substrate for PAA formation (21), hydroxypyruvate, an
R-oxo acid having a glycoloyl group (-COCH2OH), is con-
sidered to be highly likely to serve as the direct substrate for
PGA formation. Therefore, further experiments were conducted
using the 105 000g supernatant, and a transamination between
pyruvate and L-serine was examined in order to confirm whether
Formation of PGA from Hydroxypyruvate in the 200g
Supernatant. The results obtained suggested that hydroxypyru-
vate formed by transamination between pyruvate and L-serine

N-Phenylglycolohydroxamic Acid in Spinach
J. Agric. Food Chem., Vol. 54, No. 2, 2006 593
Figure 4. Hydroxypyruvate and alanine formation in transamination
between pyruvate and
acid. Hydroxypyruvate formation (
acid ( ); Alanine formation ( ) and in the addition of aminooxyacetic
acid ( ). The incubation was started with 10 mM pyruvate and 10 mM
-serine in the presence of dialyzed 105 000g supernatant. Aminooxyacetic
acid was added at the final concentration of 0.1 mM.
L
-serine and inhibitory effect of aminooxyacetic
Figure 5. Correlation of activities of PGA formation in the spinach 200g
b) and in the addition of aminooxyacetic
supernatant and commercial yeast TK: (A)
6-phosphate, (C) -xylulose 5-phosphate, and (D) hydroxypyruvate. Initial
velocities of PGA formation were measured under the condition described
D-fructose, (B) D-fructose
9
0
O
D
L
in the Materials and Methods. Values are means
±SD (n ) 3).
Table 3. Initial Velocities of PGA Formation
Table 2. Relative Amount of PGA Formation from
Various -Oxo Acids
L-Serine and
R
substratea
V0 (nmol/min)b
relative amount
(%) of PGA
formationb
hydroxypyruvate
pyruvate
9.4
0.089
±
±
0.8
0.004
R
-oxo acida
a Each substrate was added at the final concentration of 0.5 mM (hydroxy-
pyruvate) and 10 mM (pyruvate) to the incubation mixture containing the 200g
supernatant. b Values are obtained by the least-squares method.
pyruvate (control)
glyoxylate
oxalacetate
100
1350
29
mercaptopyruvate
2-ketoglutarate
ketoisovalerate
2-keto-3-methylvalerate
19
17
7
5
Figure 5, there was a positive correlation in the activities of
PGA formation (r ) 0.971, n ) 4, p < 0.05). In experiments
using the partially purified spinach transketolase, PGA formation
with hydroxypyruvate increased in proportion to the incubation
time, and a proportional relationship was observed between the
amount of the transketolase added and PGA formation activity
(data not shown).
a Each
mixture containing 10 mM
R
-oxo acid was added at the final concentration of 1 mM to the incubation
L
-serine and dialyzed 105 000g supernatant. b Values
are expressed as the percentage of the control value for the amount of PGA formed
with pyruvate during the 20-min incubation.
is directly involved in PGA formation. Accordingly, PGA
formation from hydroxypyruvate and NOB was examined using
the 200g supernatant. PGA was shown to be formed with no
time lag (data not shown). Moreover, as shown in Table 3, the
activity level was markedly higher than that observed with
pyruvate. The activity of PGA formation observed with hy-
droxypyruvate was not affected by the addition of AOA (data
not shown).
DISCUSSION
Mechanism of PGA Formation in Spinach Leaf Homo-
genate. In the pyruvate-dependent PGA formation using the
200g supernatant of a spinach leaf homogenate, PGA was
formed at an accelerating rate (Figure 2) and no proportional
relationship between the amount of supernatant added and PGA
formation was observed (data not shown). From the structure
of PGA, pyruvate was considered not to be the direct-acting
substrate for PGA formation. These results, therefore, suggest
a mechanism by which pyruvate is converted into the actual
substrate that reacts with NOB to produce PGA. This hypothesis
is supported by the observation of pyruvate-dependent PGA
formation with 105 000g supernatant but not with the dialyzed
supernatant (Figure 3). An intrinsic low molecular weight
material in the homogenate, which is removed by dialysis, is
thought to play an important role in PGA formation. Further-
more, the activity of PGA formation was enhanced by the
addition of L-serine together with pyruvate (Table 1). On the
other hand, the activity was reduced by the addition of L-alanine
(Table 1). These results suggest that a transaminase catalyzed
the transamination between pyruvate and L-serine participates
in PGA formation. This was also confirmed by the production
of alanine and hydroxypyruvate from pyruvate and L-serine in
the presence of the dialyzed 105 000g supernatant (Figure 4).
Kinetic Analysis of PGA Formation in the 200g Super-
natant. Kinetic analyses of PGA formation with the 200g
supernatant in hydroxypyruvate were performed. PGA formation
with the 200g supernatant exhibits Michaelis-Menten kinetics.
The kinetic parameters of this reaction are as follows: Km
)
2.40 ( 0.01 mM and Vmax ) 186 ( 1 nmol/min/g of spinach
leaf.
Participation of Transketolase in the Formation of PGA.
Hydroxypyruvate is known to be a substrate for transketolase
(26), which catalyzes the transfer of glycoloyl groups of ketoses
into aldoses. Nitroso aromatic compounds have been reported
to serve as glycoloyl acceptors in transketolase-catalyzed
N-arylglycolohydroxamic acids (27). Therefore, a correlation
for PGA formation activities of the 200g supernatant of spinach
leaf homogenate, which is known to present transketolase
activity (28, 29), and commercial yeast transketolase was
examined with various types of glycoloyl donors. As shown in

594 J. Agric. Food Chem., Vol. 54, No. 2, 2006
Tatsunami and Yoshioka
Stage 2: Glycoloyl Group Transfer Reaction. Hydroxy-
pyruvate formed by transamination between pyruvate and
L-serine is likely to serve as the direct-acting substrate for
transketolase, which catalyzes the glycoloyl group transfer from
hydroxypyruvate into NOB. With regard to PGA formation in
nonplant systems, PGA has been reported to be formed from
NOB and D-xylulose 5-phosphate using yeast transketolase (27).
We have also reported PGA formation from NOB and hydroxy-
pyruvate in experiments using rat liver homogenate (36).
Furthermore, it has been reported that hydroxypyruvate serves
as a glycoloyl donor in transketolase reactions (26). As shown
in Figure 5, PGA formation was observed even when D-fructose,
D-fructose 6-phosphate, and D-xylulose 5-phosphate, known
glycoloyl donors in transketolase reactions (26), were used.
Moreover, a positive correlation was found between yeast
transketolase and spinach leaf homogenate in the activities for
PGA formation when various types of glycoloyl donors were
used. Stereochemistry of glycoloyl donors in transketolase has
shown a high specificity for ketoses with C3-L, C4-D config-
uration (37). Structurally, hydroxypyruvate is very different from
these ketoses and has been reported to be an irreversible
glycoloyl donor (26). Furthermore, the formation of PGA from
hydroxypyruvate was also observed in the presence of partially
purified transketolase. Taken together, these findings strongly
suggest that transketolase is involved in the formation of PGA
in spinach leaves.
Scheme 1. Two Stages of the Mechanism for Pyruvate-Dependent
Formation of PGA from NOBa
a AT, transaminase; TK, transketolase.
This transamination reaction was also completely inhibited by
the addition of AOA (Figure 4). As shown in Table 3,
hydroxypyruvate was shown to serve as the direct-acting
substrate for PGA formation from NOB, in which PGA
formation was shown to exhibit Michaelis-Menten kinetics.
On the basis of these results, the enzyme that catalyzed PGA
formation from hydroxypyruvate and NOB is thought to be
transketolase.
It is thus assumed that pyruvate-dependent PGA formation
from NOB proceeds in spinach leaves by way of the two-stage
mechanism shown in Scheme 1. Stage 1 of this mechanism is
a transaminase-catalyzed reaction between pyruvate and L-serine
to give alanine and hydroxypyruvate, and stage 2 is a trans-
ketolase-catalyzed glycoloyl (-COCH2OH) transfer reaction
between hydroxypyruvate and NOB. L-Serine is the abovemen-
tioned intrinsic low molecular weight material in the spinach
homogenate. Stages 1 and 2 of the PGA formation are discussed
separately below.
Stage 1: Transamination. In limited studies (30-32)
relating to spinach leaves, three types of transaminase have been
reported: serine:glyoxylate transaminase (EC 2.6.1.45;
SGT), glutamate:glyoxylate transaminase (EC 2.6.1.4), and
aspartate:2-oxoglutarate transaminase (EC 2.6.1.14). SGT and
glutamate:glyoxylate transaminase are localized in leaf peroxi-
somes, but aspartate:2-oxo-glutarate transaminase isozymes are
found in the chloroplasts, mitochondria, and peroxisomes. As
for SGT, glyoxylate is reportedly more effective than pyruvate
as an amino acceptor (30), and both L-serine and L-alanine have
particularly high capacity to donate amino groups among amino
acids, with the capacity being higher in L-serine than L-alanine
(33). The transamination activity of SGT is also reported to be
inhibited by D-serine and ammonium sulfate (25).
In this study, as shown in Table 1, the activity of PGA
formation was inhibited by the addition of L-alanine. The activity
of L-serine-dependent PGA formation was markedly enhanced
by the addition of glyoxylate compared with that in the case of
addition of pyruvate (Table 2). On the basis of these results
and the inhibition of PGA formation by D-serine and ammonium
sulfate, SGT is thought to be the major enzyme involved in the
transamination reaction between pyruvate and L-serine.
No increase in PGA formation was observed even when PLP
was added to the spinach leaf homogenate. Related findings
showing that the activity of partially purified SGT from spinach
leaves (30) and the activity of asparagine transaminase from
pea plant leaves in the initial stage of purification (34) are not
amplified by the addition of PLP have also been reported.
Moreover, in general, binding to PLP is reportedly stronger with
transaminase derived from plant tissue than with those derived
from animal tissue or microorganisms (35). Consequently, this
is believed to be the reason that the addition of PLP resulted in
no further increase in PGA formation in spinach leaves.
Even without the addition of pyruvate, PGA formation was
observed (Figure 2). This pyruvate-independent PGA formation
might be accounted for by the presence of intrinsic glycoloyl
donors, such as D-fructose 6-phosphate and hydroxypyruvate.
Transketolase is a TPP-dependent enzyme; however, no
enhanced PGA formation activity was observed when TPP was
added (data not shown). This result is believed to be due to the
fact that TPP is strongly embedded in the transketolase active
site (38) and the fact that spinach leaves contain TPP (calculated
as thiamin, approximately 7 µg/g of dried spinach) (39).
Intracellular Localization of Enzymes Involved in PGA
Formation. It was strongly suggested that SGT and transke-
tolase contribute to the formation of PGA. SGT in spinach leaves
has been reported to localize in the peroxisomes (32), while
transketolase has been reported to localize in the chloroplasts
in spinach leaves (28, 29). Consequently, given this intracellular
localization of SGT and transketolase, activity of PGA formation
could not be observed in the 105 000g supernatant. However,
in this study, approximately 80% of the activity of PGA
formation observed with the 200g supernatant of spinach leaf
homogenate remained in the 105 000g supernatant. These results
suggest that during the homogenization process used in this
experiment, destruction of chloroplasts and peroxisomes causes
the elution of transketolase and SGT, respectively.
Toxicological Aspects. The formation of PGA in spinach
using the model compound NOB illustrates possible toxicologi-
cal problems.
First, nitro and/or amino aromatic compounds that are widely
used in the environment are taken up by plants and then
converted into nitroso compounds and subsequently into the
corresponding N-acylhydroxamic acids such as N-acetyl and
N-glycoloyl derivatives through pathways described in this
report and a previous report (15). In plants, which have no
effective systems for excretion (1, 2), these metabolites may
accumulate in structures such as vacuoles or cell walls (1, 2,
40) and be potentially harmful for animals and humans that
consume them. 2-N-(2-Fluorenyl)glycolohydroxamic acid has
been reported to show the same level of mutagenicity as that

N-Phenylglycolohydroxamic Acid in Spinach
J. Agric. Food Chem., Vol. 54, No. 2, 2006 595
(15) Yoshioka, T.; Uematsu, T. Formation of N-arylacylhydroxamic
acids from nitrosoaromatic compounds in isolated spinach leaf
cells. J. Agric. Food Chem. 1998, 46, 606-610.
(16) Corbett, M. D.; Corbett, B. R. Quantitative determination of
N-arylaceto- and N-arylglycolhydroxamic acids in biochemical
reaction mixtures. Anal. Biochem. 1979, 98, 169-177.
(17) Villafranca, J. J.; Axelrod, B. Heptulose synthesis from non-
phosphorylated aldoses and ketoses by spinach transketolase. J.
Biol. Chem. 1971, 246, 3126-3131.
(18) Tolbert, N. E.; Yamazaki, R. K.; Oeser, A. Loclization and
properties of hydroxypyruvate and glyoxylate reductases in
spinach leaf particles. J. Biol. Chem. 1970, 245, 5129-5136.
(19) Sakoda, M.; Hiromi, K. Determination of best-fit values of kinetic
parameters of the Michaelis-Menten equation by the method
of least squares with Taylor expansion. J. Biochem. 1976, 80,
547-555.
of the corresponding N-acetohydroxamic acid derivative (41).
Furthermore, both the hydroxamic acid derivatives have the
ability to form a covalent bond to the DNA of isolated rat cells
(42).
Second, because SGT and transketolase, enzymes for inter-
mediate metabolism, are involved in PGA formation, the
formation of N-arylglycolohydroxamic acids from the corre-
sponding aromatic nitroso compounds in plants could disrupt
both amino acid and sugar metabolisms, which may affect plant
growth and nutritional values.
ACKNOWLEDGMENT
We would like to thank Chie Kijima for technical assistance.
(20) Hinson, J. A.; Mitchell, J. R.; Jollow, D. J. Microsomal
N-hydroxylation of p-chloro-acetanilide. Mol. Pharmacol. 1975,
11, 462-469.
LITERATURE CITED
(21) Yoshioka, T.; Uematsu, T. Formation of N-hydroxy-N-arylac-
etamides from nitrosoaromatic compounds by the mammalian
pyruvate dehydrogenase complex. Biochem. J. 1993, 290, 783-
790.
(22) Riens, B.; Lohaus, G.; Heineke, D.; Heldt, H. W. Amino acid
and sucrose content determined in the cytosolic, chloroplastic,
and vacuolar compartments and in the phloem sap of spinach
leaves. Plant Physiol. 1991, 97, 227-233.
(23) Baskaran, N.; Prakash, V.; Savithri, H. S.; Radhakrishnan, A.
N.; Rao, N. A. Mode of interaction of aminooxy compounds
with sheep liver serine hydroxymethyltransferase. Biochemistry
1989, 28, 9613-9617.
(24) John, R. A.; Charteris, A.; Fowler, L. J. The reaction of amino-
oxyacetate with pyridoxal phosphate-dependent enzymes. Bio-
chem. J. 1978, 171, 771-779.
(25) Rehfeld, D. W.; Tolbert, N. E. Aminotransferases in peroxisomes
from spinach leaves. J. Biol. Chem. 1972, 247, 4803-4811.
(26) Kochetov, G. A. Transketolase from yeast, rat liver, and pig liver.
Methods in Enzymol. 1982, 90, 209-223.
(27) Corbett, M. D.; Chipko, B. R. N-Phenylglycolhydroxamate
production by the action of transketolase on nitrosobenzene.
Biochem. J. 1977, 165, 263-267.
(28) Scnarrenberger, C.; Flechner, A.; Martin, W. Enzymatic evidence
for a complete oxidative pentose phosphate pathway in chloro-
plasts and an incomplete pathway in the cytosol of spinach
leaves. Plant Physiol. 1995, 108, 609-614.
(1) Lamoureux, G. L.; Rusness, D. G. Xenobiotic Conjugation in
Higher Plants. In Xenobiotic Conjugation Chemistry; Paulson,
G. D., Caldwell, J., Hutson, D. H., Menn, J. J., Eds.; American
Chemical Society: Washington, DC, 1986; pp 62-105.
(2) Edwards, V. T.; Hutson, D. H. The Disposition of Plant
Xenobiotic Conjugates in Animals. In Xenobiotic Conjugation
Chemistry; Paulson, G. D., Caldwell, J., Hutson, D. H., Menn,
J. J., Eds.; American Chemical Society: Washington, DC, 1986;
pp 322-340.
(3) Korte, F.; Kvesitadze, G.; Ugrekhelidze, D.; Gordeziani, M.;
Khatisashvili, G.; Buadze, O.; Zaalishvili, G.; Coulston, F.
Organic toxicants and plants. Ecotoxicol. EnViron. Saf. 2000,
47, 1-26.
(4) Ugrekhelidze, D.; Korte, F.; Kvesitadze, G. Uptake and trans-
formation of benzene and toluene by plant leaves. Ecotoxicol.
EnViron. Saf. 1997, 47, 24-29.
(5) Goodman, B. A.; Allison, M. J.; Oparka, K. J.; Hillman, J. R.
Xenobiotics: Their activity and mobility in plants and soils. J.
Sci. Food Agric. 1992, 59, 1-20.
(6) Yang, X.; Kulkarni, A. P. N-dealkylation of aminopyrine
catalyzed by soybean lipoxygenase in the presence of hydrogen
peroxide. J. Biochem. Mol. Toxicol. 1998, 12, 175-183.
(7) Chung, J. G.; Lee, J. H.; Ho, C. C.; Lai, J. M.; Chou, Y. C.;
Teng, H. H.; Hung, C. F.; Huang, S. C. A survey of arylamine
N-acetyltransferase activity in common fruits and vegetables. J.
Food Biochem. 1997, 20, 481-490.
(29) Teige, M.; Melzer, M.; Su¨ss, K.-H. Purification, properties and
in situ localization of the amphibolic enzymes D-ribulose
5-phosphate 3-epimerase and transketolase from spinach
chloroplasts. Eur. J. Biochem. 1998, 252, 237-244.
(30) Nakamura, Y.; Tolbert, N. E. Serine:glyoxylate, alanine:glyoxy-
late, and glutamate:glyoxylate aminotransferase reactions in
peroxisomes from spinach leaves. J. Biol. Chem. 1983, 258,
7631-7638.
(31) Smith, I. K. Purification and characterization of serine:glyoxylate
aminotransferase from kidney bean (phaseolus Vulgaris). Bio-
chim. Biophys. Acta 1973, 321, 156-164.
(32) Yamazaki, R. K.; Tolbert, N. E. Enzymic characterization of
leaf peroxisomes. J. Biol. Chem. 1970, 245, 5137-5144.
(33) Noguchi, T.; Hayashi, S. Peroxisomal localization and properties
of tryptophan aminotransferase in plant leaves. J. Biol. Chem.
1980, 255, 2267-2269.
(8) Loutre, C.; Dixon, D. P.; Brazier, M.; Slater, M.; Cole, D. J.;
Edwards, R. Isolation of a glycosyltransferase from Arabidopsis
thaliana active in the metabolism of the persistent pollutant 3,4-
dichloroaniline. Plant J. 2003, 34, 485-493.
(9) Neuefeind, T.; Reinmer, P.; Bieseler, B. Plant glutathione
S-transferases and herbicide detoxification. Biol. Chem. 1997,
378, 199-205.
(10) Ruhland, M.; Engelhardt, G.; Pawlizki, K. A comparative
investigation of the metabolism of the herbicide glufosinate in
cell cultures of transgenic glufosinate-resistant and non-transgenic
oilseed rape (Brassica napus) and corn (Zea mays). EnViron.
Biosaf. Res. 2002, 1, 29-37.
(11) Hamburg, A.; Puvanesarajah, V.; Burnett, T. J.; Barnekow, D.
E.; Premkumar, N. D.; Smith, G. A. Comparative degradation
of [14C]-2,4-dichlorophenoxyacetic acid in wheat and potato
after Foliar application and in wheat, radish, lettuce, and apple
after soil application. J. Agric. Food Chem. 2001, 49, 146-155.
(12) Hohl, H.-U.; Barz, W. Metabolism of the insecticide Phoxim in
plants and cell suspension cultures of soybean. J. Agric. Food
Chem. 1995, 43, 1052-1056.
(34) Ireland, R. J.; Joy, K. W. Purification and properties of an
asparagine aminotransferase from Pisum sativum leaves. Arch.
Biochim. Biophys. 1983, 223, 291-296.
(35) Givan, C. V. Aminotransferases in higher plants. In The
Biochemistry of Plants, Miflin, B. J., Eds.; Academic Press: New
York, 1980; Vol. 5, pp 329-357.
(13) Purohit, V.; Basu, A. K. Mutagenicity of Nitroaromatic Com-
pounds. Chem. Res. Toxicol. 2000, 13, 673-692.
(14) Lotlikar, P. D. Cytochrome P-450 catalysed activation of
carcinogenic aromatic amines and amides. In N-Oxidation of
Drugs; Hlavica, P., Damani, L. A., Eds.; Chapman & Hall:
London, 1991; pp 315-329.
(36) Yoshioka, T.; Uematsu, T. Biotransformation of nitroso aromatic
compounds and 2-oxo acids to N-hydroxy-N-arylacylamides by
thiamine-dependent enzymes in rat liver. Drug Metab. Dispos.
1998, 26, 705-710.

596 J. Agric. Food Chem., Vol. 54, No. 2, 2006
Tatsunami and Yoshioka
(37) Nilsson, U.; Meshalkina, L.; Lindqvist, Y.; Schneider, G.
Examination of substrate binding in thiamine diphosphate-
dependent transketolase by protein crystallography and site-
directed mutagenesis. J. Biol. Chem. 1997, 272, 1864-1869.
(38) Schneider, G.; Lindqvist, Y. Crystallography and mutagenesis
of transketolase: Mechanistic implications for enzymatic thiamin
catalysis. Biochim. Biophys. Acta 1998, 1385, 387-398.
(39) Hebrero, E.; Santos-Buelga, C.; Garcia-Moreno, C. Changes in
thiamin content during the storage of spinach. J. Agric. Food
Chem. 1988, 36, 144-147.
amide 2-(glycolylamino)fluorene to give the glycolylhydroxamic
acid. A new xenobiotic reaction. Chem. Res. Toxicol. 1990, 3,
296-300.
(42) Corbett, M. D.; Lim, L. O.; Corbett, B. R.; Johnston, J. J.
Wiebkin, P. Covalent binding of N-hydroxyl-N-glycolyl-2-
aminofluorene to rat hepatocyte DNA: In vitro and cell-
suspension studies. Chem. Res. Toxicol. 1988, 1, 41-46.
Received for review August 11, 2005. Revised manuscript received
November 14, 2005. Accepted November 20, 2005.
(40) Sandermann, H., Jr. Plant metabolism of xenobiotics. Trends
Biochem. Sci. 1992, 17, 82-84.
(41) Corbett, M. D.; Corbett, B. R.; Quintana, S. J.; Hannothiaux,
M.-H.; Wei, C.-I. Microsomal N-hydroxylation of the glycol-
JF051969F