3,4-Dichloroaniline is detoxified and exported via different pathways in Arabidopsis and soybean

Article abstract of DOI:10.1016/S0031-9422(03)00289-9

The metabolic fate of [UL-¹⁴C]-3,4-dichloroaniline (DCA) was investigated in Arabidopsis root cultures and soybean plants over a 48 h period following treatment via the root media. DCA was rapidly taken up by both species and metabolised, predominantly to N-malonyl-DCA in soybean and N-glucosyl-DCA in Arabidopsis. Synthesis occurred in the roots and the respective conjugates were largely exported into the culture medium, a smaller proportion being retained within the plant tissue. Once conjugated, the DCA metabolites in the medium were not then readily taken up by roots of either species. The difference in the routes of DCA detoxification in the two plants could be explained partly by the relative activities of the respective conjugating enzymes, soybean containing high DCA-N-malonyltransferase activity, while in Arabidopsis DCA-N-glucosyltransferase activity predominated. A pre-treatment of plants with DCA increased DCA-N-malonyltransferase activity in soybean but not in Arabidopsis, indicating differential regulation of this enzyme in the two plant species. This study demonstrates that DCA can undergo two distinct detoxification mechanisms which both lead to the export of conjugated metabolites from roots into the surrounding medium in contrast to the vacuolar deposition more commonly associated with the metabolism of xenobiotics in plants.

Full text of DOI:10.1016/S0031-9422(03)00289-9

Phytochemistry 63 (2003) 653–661
www.elsevier.com/locate/phytochem
3
,4-Dichloroaniline is detoxified and exported via different
pathways in Arabidopsis and soybean
a
b
b
c
Si-Houy Lao , Caroline Loutre , Melissa Brazier , Julian O.D. Coleman ,
a,
d
b
David J. Cole , Robert Edwards , Frederica L. Theodoulou *
aCrop Performance and Improvement Division, Rothamsted Research, Harpenden AL5 2JQ, UK
bSchool of Biological and Biomedical Sciences, University of Durham, Durham DH1 3LE, UK
cSchool of Biological and Molecular Sciences, Oxford Brookes University, Gypsy Lane, Oxford OX3 0BP, UK
dFormerly of Aventis Crop Science, Fyfield Road, Ongar, Essex CM5 0HW, UK
Received 4 February 2003; received in revised form 28 April 2003
Abstract
1
4
The metabolic fate of [UL- C]-3,4-dichloroaniline (DCA) was investigated in Arabidopsis root cultures and soybean plants over
a 48 h period following treatment via the root media. DCA was rapidly taken up by both species and metabolised, predominantly
to N-malonyl-DCA in soybean and N-glucosyl-DCA in Arabidopsis. Synthesis occurred in the roots and the respective conjugates
were largely exported into the culture medium, a smaller proportion being retained within the plant tissue. Once conjugated, the
DCA metabolites in the medium were not then readily taken up by roots of either species. The difference in the routes of DCA
detoxification in the two plants could be explained partly by the relative activities of the respective conjugating enzymes, soybean
containing high DCA-N-malonyltransferase activity, while in Arabidopsis DCA-N-glucosyltransferase activity predominated. A
pre-treatment of plants with DCA increased DCA-N-malonyltransferase activity in soybean but not in Arabidopsis, indicating dif-
ferential regulation of this enzyme in the two plant species. This study demonstrates that DCA can undergo two distinct detox-
ification mechanisms which both lead to the export of conjugated metabolites from roots into the surrounding medium in contrast
to the vacuolar deposition more commonly associated with the metabolism of xenobiotics in plants.
#
2003 Elsevier Ltd. All rights reserved.
Keywords: Glycine max (Fabaceae) soybean; Arabidopsis thaliana (Brassicaceae); Herbicide metabolism; Conjugation; Export; Detoxification;
Dichloroaniline; Malonyltransferase; Glucosyltransferase; Roots
1
. Introduction
interest in its metabolic fate in plants (Bartha et al.,
983; Sandermann et al., 1983; Harvey et al., 2002).
1
The ability of plants to take up and chemically trans-
form xenobiotics has important implications for envir-
onmental and food science. Chlorinated anilines such as
DCA metabolism has been investigated in a small
number of plants, with differences observed in the pro-
portions and identities of biotransformation products.
In soybean, N-malonylation was the major route of
DCA metabolism in cell cultures (Winkler and Sander-
mann, 1989; Harms and Langbartels, 1986; Gareis et
al., 1992; Schmidt et al., 1995), excised leaves (Gareis et
al., 1992) and whole plants (Bockers et al., 1994). In
wheat, both plants and suspension-cultured cells meta-
bolised DCA mainly by N-glucosylation (Schmitt et al.,
1985, 1995; Winkler and Sandermann, 1989, 1995;
Bockers et al., 1994), while in carrot cultures, DCA
underwent both malonylation and glucosylation
3
,4-dichloroaniline (DCA) are intermediates in chemical
synthesis as well as being metabolites of acylanilide,
phenylurea, and carbamate pesticides (Parris, 1980; Van
der Trenck et al., 1981). DCA can persist in the envir-
onment as insoluble residues in soil and plants and can
also photodimerise to form carcinogens (Harvey et al.,
2002). Consequently, DCA has been classified as a
compound of environmental concern, with substantial
*
Corresponding author. Fax: +44-1582-763010.
E-mail address: freddie.theodoulou@bbsrc.ac.uk
F.L. Theodoulou).
(Schmidt et al., 1994). The enzymes responsible for cat-
alysing the conjugation of DCA to glucose and malonic
(
0031-9422/03/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0031-9422(03)00289-9

654
S.-H. Lao et al. / Phytochemistry 63 (2003) 653–661
acid respectively have been described in plants (Frear,
968; Sandermann et al., 1991; Matern et al., 1984). In a
dues in Arabidopsis roots. The percentage of the dosed
radiolabel in the Arabidopsis growth medium continued
to decline during the course of the experiment, to a final
value of 20% after 48 h treatment.
1
number of studies, species-dependent differences in
DCA metabolism have been attributed to differences in
the activity of the respective DCA-N-glucosyltransfer-
ase and DCA-N-malonyltransferase enzymes (Sander-
mann et al., 1991; Schmidt et al., 1995).
2.2. Metabolic fate of DCA in soybean and Arabidopsis
In this study, we have compared DCA metabolism in
soybean, a species where the detoxification of this
xenobiotic is relatively well characterised, with its
metabolism in the model species, Arabidopsis thaliana.
Ultimately, the intention here is to characterise the
enzymes and transporters involved in DCA metabolism
in these two species using a combination of biochemical
and molecular approaches, with the potential aim of
manipulating xenobiotic metabolism in plants.
The radioactive metabolites present in the plant tis-
sues and respective media were analysed by thin layer
chromatography (TLC) and quantified, following
Phosphorimaging. For reference, the profiles of meta-
bolites seen in both soybean and Arabidopsis after a 48
1
4
h incubation with [ C]-DCA are shown in Fig. 3. The
identification of DCA metabolites was based on their
co-chromatography in two TLC solvent systems and by
HPLC (data not shown) with prepared standards of
1
4
DCA-N-malonate (M-DCA) and [ C]-DCA-N-b-d-
glucoside (G-DCA), which are both well characterised
metabolites of DCA in plants (Winkler and Sander-
mann, 1989, 1992; Gareis et al., 1992).
2
. Results
2
.1. Time course of 14C DCA uptake
Hydroponically-grown soybean seedlings and
Arabidopsis root cultures were treated with identical,
1
4
sub-toxic doses of [UL- C]-3,4-dichloroaniline (45.2
nmol, 37 kBq) for 48 h to investigate the uptake and
distribution of radioactivity over time. Plants were
extracted with methanol to determine incorporation
into soluble metabolites, with any non-extractable
radioactivity defined as being a bound residue which
was quantified after digestion with alkali. Radioactivity
in the medium was quantified directly, with the aqueous
medium partitioned with ethyl acetate to recover DCA
and its metabolites (average recovery 86% in soybean
and 96% in Arabidopsis). In each case, the radioactivity
present was quantified by liquid scintillation counting
(
LSC; Fig. 1). In soybean plants, radioactivity in the
medium declined rapidly over 0–4 h (Fig. 1A). A major
proportion of the loss in radioactivity was due to the
1
4
uptake of [ C]-DCA into the soybean plants, where the
radiolabel accumulated as both extractable and bound
residues, predominantly in the roots. After 4 h, the dis-
tribution of radiolabel between medium and root
remained constant. There was a moderate net loss of
radioactivity from the system after 48 h, which was
1
4
attributed to the volatilisation of [ C]-DCA from the
1
4
medium due to aeration. Fluorographs of the [ C]-
DCA treated soybean plants showed that, at the end of
the time course, the majority of the label had accumu-
lated in the roots, with a small proportion in the mer-
istem (Fig. 2).
In Arabidopsis, the rate of net uptake of radioactivity
into cultured roots was also rapid, with soluble residues
in the roots comprising 65% of total label after 24 h of
treatment (Fig. 1B). Unlike soybean, only 3–4% of the
administered label was incorporated into bound resi-
Fig. 1. Distribution of radioactivity following addition of [14C]-DCA
to the root medium of soybean and Arabidopsis. Hydroponically-
grown soybean seedlings (A), and Arabidopsis root cultures (B) were
treated with 45.2 nmol (37 kBq) [UL-14C]-3,4-dichloroaniline for 48 h.
At various time points, the radioactive metabolites were extracted
from the plants and medium and quantified by LSC as: soluble resi-
dues in the plant (closed circles), bound residues in the plant (open
circles), radioactivity in the growth medium (closed triangles). Data
points represent the means of 2–4 replicates, with error bars showing
standard deviations. Representative of four experiments.

S.-H. Lao et al. / Phytochemistry 63 (2003) 653–661
655
Fig. 3. Identification of DCA metabolites by TLC. Radioactive DCA
metabolites were extracted from plants and culture medium and ana-
lysed by TLC using chloroform/methanol/water (60:35:8; v/v/v).
Lanes: (1) [14C]-3,4-dichloroaniline, (2) N-14C-d-glucopyranosyl)-3,4-
dichloroaniline, (3) N-14C-malonyl-3,4-dichloroaniline, (4) Arabidopsis
plant soluble extract, (5) Arabidopsis growth medium, (6) soybean
plant soluble extract, (7) soybean growth medium.
other 8% ran as a diffuse spot near the solvent front.
1
4
When the ‘‘authentic’’ [ C]-DCA was scraped off and
re-run in the same chromatography system, the radio-
activity again partitioned into the two moieties in a
similar ratio (9:1). It was concluded that the minor,
more hydrophobic radioactive entity was an artefact of
chromatography, rather than an impurity and in all
subsequent analyses, these two radioactive bands were
1
4
collectively quantified as parent [ C]-DCA. When ana-
1
4
lysed by HPLC, [ C]-DCA ran as a single radioactive
entity (data not shown).
Previous studies have demonstrated that DCA can
spontaneously form the respective N-b-d-glucopyr-
anosyl conjugates when incubated with d-glucose
(Winkler and Sandermann, 1992). It was important,
therefore, to demonstrate that DCA metabolism in the
root cultures was due to reactions mediated in planta,
rather than due to non-enzymic reactions. When [ C]-
DCA was incubated with Arabidopsis growth medium
Fig. 2. Distribution of radiolabel in soybean plant fed [14C]-DCA
through the root medium. Hydroponically-grown, 10 day-old soybean
seedlings were treated with 45.2 nmol (37 kBq) [UL-14C]-3,4-
dichloroaniline. After 3.5 days, plants were analysed by phosphor-
imaging (A). Radioactivity in the dissected plant parts was quantified
by LSC and the results expressed as % of the original dose (A) and
nmol of [14C]-DCA equivalents g FW (B), with results showing the
means of duplicated experimentsꢁS.E.
14
containing 1% (w/v) d-glucose in place of sucrose, G-
DCA was seen to accumulate at a rate of 0.15 nmol h
ꢀ
1
ꢀ1
under the dosing conditions used in the plant feeding
studies. When sucrose, the sugar used routinely in the
1
4
plant studies was tested, [ C]-DCA was not converted
to polar conjugates. Similarly, the rate of DCA con-
jugate synthesis in conditioned media taken from the
Arabidopsis root cultures or soybean plants was also
below the limit of detection. These experiments con-
firmed that any metabolism of DCA observed in the
feeding studies was due to metabolism in planta.
1
4
In all chromatograms, the parent [ C]-DCA ran as
two radioactive entities, with the proportion of radio-
activity in each band determined by the solvent system.
In the chloroform:methanol:water system, 92% of the
1
4
radioactivity in the [ C]-DCA dosing solution co-chro-
matographed with authentic unlabelled DCA, while the

656
S.-H. Lao et al. / Phytochemistry 63 (2003) 653–661
1
4
After a 48 h incubation with [ C]-DCA, distinct pro-
Table 1
Quantification of major DCA metabolites formed over 48 h in soy-
bean plants and Arabidopsis root cultures dosed with 45.2 nmol [14C]-
DCAa
files of radioactive metabolites were seen in soybean and
Arabidopsis (Fig. 3). Extracts from soybean plants con-
tained three major metabolites of DCA which were
designated S1, S2 and S3, respectively. Based on co-
chromatography, S1 was identified as G-DCA and S2 as
M-DCA. Compound S3 did not co-chromatograph with
any available standards. In the soybean root medium,
Time (h)
Extracted metabolites (nmol)
DCA
A1
G-DCA
M-DCA
A4/S3
A5
Soybean root extract
1
4
0.5
2
1.5
4.4
4.2
5.5
5.0
4.2
3.8
ND
0.4
0.6
2.2
1.5
0.6
0.6
0.9
1.3
0.1
0.4
0.4
0.5
0.2
0.2
0.2
only S2 (M-DCA) was observed in addition to [ C]-
DCA. Extracts from Arabidopsis plants contained five
major metabolites of DCA, designated A1–A5. Com-
pound A1 was not identified but migrated very closely
to A2, which in turn co-chromatographed with G-DCA.
Metabolite A3 appeared to be M-DCA while A4 was
identical to the unknown soybean DCA metabolite S3.
The polar compound A5 was also not characterised.
Although metabolites A1, A4 and A5 were not identi-
fied, their lability to acid hydrolysis and associated
release of DCA suggested that they were all N-con-
jugates of the parent compound. In the Arabidopsis root
medium, only A1 and G-DCA were observed as DCA
metabolites.
4
0.3
8
12
ND
ND
ND
ND
2
4
4
8
Soybean medium
0
2
4
8
1
2
4
.5
29.5
15.8
7.8
2.0
4.7
7.0
4.1
4.4
11.3
11.7
15.1
12.0
2
4
8
3.3
3.2
Arabidopsis root extract
4
8
1
2
4
2.7
2.2
3.4
2.2
1.3
3.1
2.5
2.5
2.0
0.9
18.2
9.7
0.5
0.3
0.5
0.6
0.8
0.9
0.9
1.4
1.7
2.5
0.5
0.5
0.6
1.4
2.0
1
4
The metabolism of [ C]-DCA in soybean plants and
Arabidopsis root cultures was studied over a 48 h feed-
ing period and the major recovered metabolites (Fig. 3)
quantified in both media and plant tissue (Table 1).
2
4
8
10.9
6.5
3.0
Arabidopsis medium
1
4
With soybean, parent [ C]-DCA steadily declined from
the medium, to be replaced with M-DCA. In contrast,
the M-DCA content in the soybean plant remained vir-
tually constant throughout this period. By comparison
with the proportion of DCA metabolised by malonyla-
tion, the amounts of G-DCA and metabolite S3 formed
were modest, with G-DCA accumulating transiently at
4
8
1
2
4
2.7
3.0
5.6
2.2
2.8
0.3
2.2
3.0
5.8
6.7
6.1
9.5
ND
ND
1.1
2
4
8
12.8
22.0
17.5
0.5
0.6
a
Results represent the means of duplicated studies with the varia-
tion in the plants being <10% of the mean. ND=not detected.
2
h and then declining.
In Arabidopsis, G-DCA was the dominant metabolite
GT) were determined in crude root extracts from the two
species. In both cases, product formation was monitored
by quantifying the amount of radiolabelled M-DCA or
G-DCA formed which partitioned into ethyl acetate
(Brazier et al., 2002). With both assays, the activities
reported significantly underestimate the true catalytic
efficiencies of the respective enzymes, as in both cases
they were incubated with low concentrations of the radi-
at all time points in both the roots and the media. In
contrast, M-DCA remained a minor metabolite, con-
firming that glucosylation was the major route of DCA
conjugation in Arabidopsis. In the roots, the accumula-
tion of G-DCA was particularly marked over the first 4
h, after which time its relative abundance steadily
declined. During this loss of G-DCA, the amounts of
DCA and M-DCA remained constant, A4 and A5
slowly accumulated and A1 slowly declined. The selec-
tive disappearance of G-DCA from the roots was asso-
ciated with an increase in the glucoside in the medium,
suggesting that the conjugate was being exported. Simi-
larly, the slower decline in A1 in the roots was mirrored
by its steady accumulation in the medium.
1
4
14
olabelled co-substrate, either UDP-[ C-glucose] or [ C-
malonyl]-CoA. This was done to enhance the sensitivity
of the assays by keeping the radioactive specific activities
of the substrates high. However, with both DCA-N-MT
and DCA-N-GT activities, it was demonstrated that
product formation was strictly dependent on protein
content for all comparative assays performed. In soy-
ꢀ
1
bean roots, DCA-N-MT activity was high (12 nkat g
protein), whereas DCA-N-GT was barely detectable (0.5
2
.3. Transferase activities
ꢀ
1
pkat g protein). In Arabidopsis root cultures, DCA-N-
GT and DCA-N-MT activities were readily detectable
although DCA-N-GT activity was two orders of magni-
In order to account for the differences in DCA con-
jugation in soybean and Arabidopsis, the activities of 3,4-
dichloroaniline-N-malonyltransferase (DCA-N-MT)
and 3,4-dichloroaniline-N-glucosyltransferase (DCA-N-
ꢀ
1
tude greater than DCA-N-MT (6.3 nkat g protein and
ꢀ
1
52 pkat g protein, respectively). No DCA-N-MT or

S.-H. Lao et al. / Phytochemistry 63 (2003) 653–661
657
DCA-N-GT activity could be determined in the media
immersing either the soybean or Arabidopsis roots, con-
firming that the synthesis of the respective conjugates
occurred within the plant tissues.
Since many xenobiotic-metabolising enzymes have
been shown to be inducible (Marrs, 1996; Tommasini et
al., 1997; Robineau et al., 1998), the effects of DCA pre-
treatment on DCA-N-MT and DCA-N-GT activity
Arabidopsis, in agreement with the DCA feeding studies
(Fig. 1). After 48 h of DCA treatment, about 5–10% of
the 14C in the medium is unchanged DCA (see Table 1)
and it is possible that the small amount of uptake and
incorporation in the cross-feeding study was due to the
presence of this parent compound. In both soybean and
Arabidopsis, the majority of the radioactivity (76–93%)
remained in the medium during the cross-feeding exper-
iment, irrespective of the source of the labelled metabo-
lites, suggesting that the export of M-DCA or G-DCA
constitutes an effective mechanism of exclusion which
extends across the species.
(
assayed at saturating DCA) were determined. Pre-
treatment of Arabidopsis root cultures with 100 mM
DCA for up to 48 h had no effect on DCA-N-MT or
DCA-N-GT activity (data not shown). In soybean roots
however, DCA-N-MT activity was increased by 2- to 6-
fold depending on the length of treatment with 100 mM
DCA (Fig. 4A). Increasing the DCA concentration to
2
00 mM gave no further enhancement in activity com-
pared to 100 mM pre-treatment. Kinetic analysis of the
effect of DCA pre-treatment indicated that although
VmaxDCA was somewhat variable between experiments
ꢀ
(
control: 248.83ꢁ272.40 nkat
g
1; DCA-treated:
ꢀ
1
1
increased by DCA pre-treatment (fig. 4B). In contrast,
376ꢁ474 nkat g ), this parameter was always
KmDCA did not change substantially (control:
2
3.09ꢁ3.17 mM; DCA-treated: 30.75ꢁ5.15 mM).
2
.4. Cross-feeding study
Since metabolic fate studies indicated that DCA was
exported from roots in a species-specific manner, pre-
dominantly as M-DCA in soybean and G-DCA in
Arabidopsis, a cross-feeding experiment was conducted
to determine whether or not these metabolites could be
reabsorbed by either plant. Soybean seedlings and Ara-
bidopsis root cultures were grown in the presence of 37
1
4
kBq [UL- C]-DCA (45.2 nmol). After 48 h, the meta-
bolites released into the medium by each plant species
were extracted with ethyl acetate, concentrated and
quantified by LSC. Untreated plants from both species
were then incubated for 24 h in 20 ml fresh medium
containing 6.67 kBq 14C-labelled metabolites from each
species and plants and media analysed by LSC to
determine the partitioning of 14C between the growth
medium and plant material.
Soybean seedlings took up approximately equal
amounts of 14C-labelled compounds whether they were
derived from soybean or Arabidopsis growth medium. In
each case, 12% of the supplied radioactivity was recov-
ered in the extractable fraction from the plants while 5%
became incorporated into bound residues, regardless of
source of the metabolites. Arabidopsis root cultures
showed a slight preference for uptake of radiolabelled
compounds derived from Arabidopsis culture medium:
the soluble fraction recovered from plants corresponded
to 13% of the dosed radioactivity when Arabidopsis
metabolites were supplied, compared with only 5% for
soybean. Almost no bound residues were formed in
Fig. 4. Effect of DCA treatment on DCA-N-MT activity. (A) Hydro-
ponically-grown soybean plants were treated with 100 mM DCA (light
grey bars) or 200 mM DCA (black bars) for 12–48 h. Roots were har-
vested at 10 days and DCA-N-MT assayed. Open bars represent con-
trols (methanol treatment). (B). Soybean roots were treated with 100
mM DCA (circles) or an equal vol. of methanol (triangles) for 24 h,
harvested at 10 days and DCA-N-MT activities were then determined
for a range of concentrations of DCA with each replicated four times
(
meansꢁS.E.). KmDCA and VmaxDCA were determined by calculating
the least-squares fit to a Michaelis–Menten function. In this experi-
ment, KmDCA values were 20.84ꢁ4.04 mM and 34.39ꢁ3.91 mM in
control and DCA-treated plants, respectively. VmaxDCA values were
441.45ꢁ21.73 nkat g and 1040.75ꢁ35.1 nkat gꢀ1 in control and
ꢀ1
DCA-treated plants, respectively.

658
S.-H. Lao et al. / Phytochemistry 63 (2003) 653–661
3
. Conclusions
metabolism in both suspension cultured cells (Winkler
and Sandermann, 1989) and the roots of whole plants
(Bockers et al., 1994). Significantly, we have also
demonstrated that once exported, the M-DCA was not
readily re-imported into the roots of soybean or Arabi-
dopsis. Since M-DCA is chemically stable, its export
would therefore represent an effective route and long-
term mechanism for detoxification (Winkler and
Sandermann, 1989, 1992). Extracellular export of M-
DCA may result from the inability of soybean to
deacylate this metabolite (Matern et al., 1984). The
insecticide, phoxim [O,O-diethyl a-cyanobenzylide-
Our results demonstrate that N-malonylation of DCA
is the major route of detoxification in soybean, while in
Arabidopsis N-glucosylation predominates. In both
cases, the respective conjugates were rapidly exported
from their site of synthesis in the roots into the sur-
rounding medium. The N-malonylation of DCA in
soybean and carrot has been reported previously
(
Winkler and Sandermann, 1989; Gareis et al., 1992;
Bockers et al., 1994; Schmidt et al., 1994, 1995). In
contrast, N-glucosylation is a minor route of metabo-
lism of DCA in soybean, but a major route in monocots
such as wheat and rice (Still, 1968; Schmitt et al., 1985;
Winkler and Sandermann, 1989; Schmidt et al., 1995;
Bockers et al., 1994). Interestingly the N-malonylation
of DCA in soybean was linked with insoluble residue
formation (Gareis et al., 1992; Bockers et al., 1994). The
N-malonylation of DCA in soybean can be readily
explained by the presence of a highly active DCA-N-
MT in the roots with the small proportion of G-DCA
formed reflecting the very low activities of DCA-N-GT.
neamino-oxyphosphononthioate;
sphinothioloxyimino)-2-phenylacetonitrile]
exported from soybean suspension culture cells as its N-
malonyl conjugate (Hohl and Barz, 1995), suggesting
2-(diethoxypho-
is also
¨
that plasma membrane transport of N-malonylated
xenobiotics is a general detoxification mechanism for
synthetic amines. This is in contrast to the N-malonates
of natural products such as the N-malonyl-conjugate of
the ethylene precursor 1-aminocyclopropane-1-car-
boxylic acid (ACC), which is selectively transported into
the vacuole (Yang and Hoffman, 1984). In Arabidopsis,
the small amount of M-DCA formed was also exported,
though this was negligible compared to the efflux of G-
DCA (Table 1). This was in contrast to wheat, with
neither G-DCA nor other related glycosides being
exported from the root (Harms and Langbartels, 1986;
Bockers et al., 1994). Neither soybean nor Arabidopsis
roots supported an appreciable net uptake of the
N-glucosyl conjugate when it was supplied to hydro-
ponic growth medium. However, unlike the N-mal-
onate, G-DCA is labile to both non-enzymic and
microbial hydrolysis (Winkler and Sandermann, 1992),
suggesting that under field conditions the DCA would
be released in the rhizosphere and become available for
uptake by plants.
Malonylation and glucosylation of natural products
are well known to influence the transport and storage of
natural products. Thus O-malonylated glucosides of
natural products such as anthocyanins and the iso-
flavone phytoalexin medicarpin, as well as O-malonates
of chlorophyll degradation products, are compartmen-
talised within the vacuole (Harborne and Self, 1987;
Mackenbrock et al., 1993; Barz and Mackenbrock,
1994; Hinder et al., 1996; Tommasini et al., 1998; Lu et
al., 1998). It has been proposed that the malonyl group
not only protects the b-glucosyl group from attack by
b-glucosidases but also acts as a signal for vacuolar
00
In earlier studies, the 6 -O-malonate of G-DCA was
reported to be a significant metabolite of DCA in soy-
bean (Gareis et al., 1992; Bockers et al., 1994). Attempts
to demonstrate the presence of this metabolite proved
inconclusive, although its chromatographic properties
resemble those of metabolite S3 (=A4).
In Arabidopsis, the major metabolite was identified as
G-DCA. In addition to G-DCA and M-DCA, three
other polar, radiolabelled metabolites could be detected,
all of which appear to be N-conjugates. In terms of
abundance, compound A1 was the second most sig-
nificant metabolite, accumulating along with G-DCA in
the medium over time. The presence of a small amount
of M-DCA in Arabidopsis was presumably due to the
low, but measurable DCA-N-MT activity present,
which was 100-fold less active than the DCA-N-GT.
Induction experiments revealed that DCA-N-MT is
differentially regulated in Arabidopsis and soybean.
Arabidopsis root cultures exhibited a low, constitutive
activity which was not elevated by DCA pre-treatment,
whereas DCA-N-MT was increased in soybean roots
following pre-incubation of roots in 100 mM DCA. The
DCA pre-treatment of soybean roots resulted in an
increase in VmaxDCA for the DCA-N-MT, but only a
modest change in KmDCA, suggesting that pre-treatment
increases the abundance of the enzyme present either by
increasing synthesis or preventing protein turnover.
This could provide a potential route to the molecular
identification of DCA-N-MT.
¨
sequestration (Matern et al., 1983; Koster et al., 1984;
Schmitt et al., 1985; Sandermann, 1994). In all these
examples, O-malonylation can be considered to be a
freely reversible modification through the action of
malonylesterases located in the cytoplasm or the tono-
plast (Matern, 1983; Mackenbrock et al., 1992). In
contrast, N-malonates of xenobiotics are recalcitrant to
In both soybean and Arabidopsis, a major sink of
DCA metabolism involved export of the M-DCA and
G-DCA conjugates into the respective media. In soy-
bean, the export of M-DCA into the media has pre-
viously been reported to be a major route of DCA

S.-H. Lao et al. / Phytochemistry 63 (2003) 653–661
659
hydrolysis and appear to be more effectively detoxified
by extracellular deposition. It also appears that N-gly-
cosides of DCA are readily exported from the roots of
Arabidopsis plants, but not from other species such as
wheat. These observations raise some interesting ques-
tions about the specificity and regulation of N-conjugate
transporters in plants and understanding these processes
could be an important tool in manipulating xenobiotic
metabolism in plants in the future. To that end, future
work will concentrate on identifying the genes encoding
the proteins responsible for the conjugation and trans-
port of xenobiotics in Arabidopsis and other plants.
stem. Samples were weighed, frozen in liquid nitrogen
ꢂ
and stored at ꢀ80 C.
4.3. Extraction of radioactive metabolites
Frozen plant tissue was ground in a pestle and mortar
with acid-washed sand in the presence of ice-cold
methanol (10 v/w). After centrifugation (3000 g, 5 min),
the supernatant was decanted and the pellet frozen at
ꢂ
ꢀ20 C. The volume of the supernatant was noted and a
sample assayed by liquid scintillation counting (LSC).
The extract was then concentrated under reduced pres-
ꢂ
sure at 50 C and the residue resuspended in methanol
(1 ml) and re-assayed by LSC. Non-extractable radio-
4
. Experimental
activity in the plant material was determined after
extensively washing the pellet from the initial extraction
with methanol. The pellet was then incubated in 2 ml 2
4
.1. Chemicals
ꢂ
M NaOH, at 37 C for 20 h prior to the addition of 320
3
de-Haen (Germany) and unlabelled malonic acid from
Fluka. [UL- C]-3,4-Dichloroaniline (817.7 MBq/
mmol) was purchased from Sigma-Aldrich (UK) and
,4-Dichloroaniline (DCA) was obtained from Riedel-
¨
ml concentrated HCl. Samples (100 ml) were then ana-
lysed by LSC. Growth medium was made up to a
known volume with water and 200 ml analysed by
mLSC. The medium was then partitioned two times
against one v/v ethyl acetate and the organic phases
combined and concentrated to 1 ml under vacuum, at
1
4
1
4
[
2- C-Malonyl]-Coenzyme A (2.035 GBq/mmol) and
1
4
Uridine diphospho-D-[UL- C]-glucose (11.84 GBq/
mmol) from Amersham Pharmacia Biotech (UK).
N-malonyl-3,4-dichloroaniline (M-DCA) was prepared
by chemical synthesis (Matern et al., 1984).
ꢂ
40 C. The residue was redissolved in 1 ml methanol and
re-assayed by LSC.
4.4. Cross feeding study
1
4
4
.2. Feeding studies with [ C]-DCA
Four soybean seedlings and four A. thaliana root cul-
tures were transferred into 50 ml Hoagland’s medium
and Gamborg B5 basal salt medium respectively, each
containing 37 kBq [UL- C]-DCA. After 48 h, plants
were removed and the radiolabelled metabolites present
in the growth media were assayed by LSC and then
extracted by partitioning with 1 v/v ethyl acetate. The
organic phase was evaporated under vacuum and the
residue redissolved in 1 ml of methanol and assayed by
LSC. After concentrating the methanolic extract to 50
A. thaliana, ecotype Col 0 seeds (Lehle Seeds, TX
USA) were sterilised in 70% (v/v) ethanol for 1 min
followed by sodium hypochlorite (20% v/v) for 10 min.
After rinsing with sterile distilled water, 10 seeds were
transferred to 50 ml Gamborg B5 liquid medium
1
4
(
(
Sigma-Aldrich, UK), pH 5.8, supplemented with 1%
w/v) sucrose. After growing for 17 days in the dark
1
4
with constant shaking, 37 kBq (45.2 nmol) [ C]-DCA
was added in 50 ml methanol to each flask. At timed
intervals, the root cultures and medium were separated
by vacuum filtration and frozen in liquid nitrogen prior
ml under a stream of N , the equivalent of 6.67 kBq of
2
1
4
C-labelled metabolites was added to 20 ml of fresh
ꢂ
to storage at ꢀ80 C.
media in the following combinations: (1) fresh soybean
plants treated with DCA metabolites from soybean (2)
fresh Arabidopsis root cultures treated with DCA meta-
bolites from soybean (3) fresh soybean plants treated
with DCA metabolites from Arabidopsis (4) fresh Ara-
bidopsis root cultures treated with DCA metabolites
from Arabidopsis. After 24 h, the plant material and
growth media were radioassayed as described above.
Soybean seeds (Glycine max L. var. Chapman) were
ꢂ
sown in vermiculite and grown at 24 C, with 110 mE
ꢀ1
ꢀ1
m
s light intensity (16 h photoperiod), for 10 days.
For DCA metabolism studies, the seedlings were trans-
ferred to Erlenmeyer flasks, each containing 50 ml
Hoagland’s No. 2 basal salt mixture (Sigma-Aldrich,
UK), pH 6.0 and incubated for 24 h with constant
1
4
aeration. [UL- C]-3,4-dichloroaniline (37 kBq; 45.2
nmol) was added to the media (50 ml), and the seedlings
incubated with constant aeration and illumination.
Duplicate batches, each of three seedlings, were har-
vested at timed intervals, the roots were rinsed with
distilled water, blotted dry and plants separated into
roots, cotyledons, primary leaves, secondary leaves, and
4.5. Identification of metabolites
Samples were applied onto silica gel TLC plates pre-
coated with fluorescent indicator (UV254, Sigma-
Aldrich, UK). The compounds were separated in
chloroform/methanol/water (60:35:8; v/v/v) or ethyl

660
S.-H. Lao et al. / Phytochemistry 63 (2003) 653–661
ꢂ
acetate/propan-2-ol/water (63:23:11; v/v/v). UV-
absorbing spots were detected under UV light at 254 nm
and 366 nm, respectively. Radioactive metabolites were
detected by autoradiography using BioMax MR Film in
combination with BioMax TranScreen LE intensifying
screens (Kodak Scientific Imaging System, Cambridge)
and individual metabolites quantified using a Bio-Rad
GS-525 Phosphor-imager which had been calibrated
with known amounts of 14C metabolites. The identity of
metabolites was confirmed by their co-chromatography
in both TLC systems with authentic unlabelled or 14C-
labelled standards. In addition, radioactive metabolites
were co-chromatographed with authentic standards
using a reverse-phase HPLC column (Symmetry C18,
a final vol. of 75 ml. After incubation at 30 C for 20
min, 125 ml Tris–HCl, pH 8.0 was added and the gluco-
sylated conjugates were partitioned into 200 ml ethyl
acetate and 100 ml quantified by LSC. Controls con-
sisted of incubating the protein extract under identical
conditions except that DCA was substituted with
methanol.
To prepare radioactive reference metabolites of M-
DCA and G-DCA, the respective organic phases from
the reactions were pooled and then concentrated under
a stream of N prior to analysis by TLC.
2
Acknowledgements
3
.5 mm, 4.6ꢃ30 mm). The initial solvent system of ace-
tonitrile/1% (v/v) phosphoric acid (5:95; v/v) was
changed after 2 ml to give a linear gradient to 100%
acetonitrile at 7.6 ml. The flow rate was 0.8 ml.min 1
We gratefully acknowledge Aventis CropScience UK
and BBSRC for CASE PhD studentships awarded to S-
h L, CL and MB. Rothamsted Research receives grant-
aided support from the BBSRC of the UK.
ꢀ
.
4
.6. Enzyme extraction and assay
Frozen soybean tissue was ground to a fine powder
References
using a pestle and mortar and suspended in 1.5 vol.
homogenising buffer [200 mM Tris–HCl, pH 7.5, 2 mM
Bartha, R., You, I.-S., Saxena, A., 1983. Humus-bound residues of
phenylamide herbicides: their nature, persistence and monitoring,
Vol. 3. In: Miyamoto, J., Kearney, P.C. (Eds.), Pesticide Chemistry.
Pergamon, Oxford, pp. 345–350.
MgCl , 5% (w/v) PVPP, 5 mM DTT, 1 mg/ml leupeptin,
2
1
mg/ml pepstatin A, 1 mM PMSF]. The slurry was
strained through four layers of muslin and the filtrate
ꢂ
Barz, W., Mackenbrock, U., 1994. Constitutive and elicitation induced
metabolism of isoflavones and pterocarpans in chickpea (Cicer
arietinum) cell suspension cultures. Plant Cell Tiss. Org. Cult. 38,
centrifuged (20,000g, 10 min at 4 C). The resulting
supernatant was employed for enzyme assays.
For Arabidopsis, the frozen tissue was ground under
liquid nitrogen using a pestle and mortar and the
resulting powder was resuspended in 3 vol. extraction
buffer [0.2 M Tris–HCl pH 8, 5% (w/v) PVPP, 1 mM
DTT]. The preparation was filtered and centrifuged as
above and proteins in the supernatant precipitated after
adjusting to 70% (w/v) with respect to ammonium sul-
phate and collected by centrifugation (10,000g, 40 min).
Protein pellets were resuspended in extraction buffer
and desalted on a G25 column. In all cases, the protein
concentration was determined using Coomassie dye
Reagent (BioRad) and g-globulin as reference protein.
1
99–211.
Bockers, M., Rivero, C., Thiede, B., Jankowski, J., Schmidt, B., 1994.
Uptake, translocation, and metabolism of 3,4-dichloroaniline in
soybean and wheat plants. Z. Naturforsch. 49c, 719–726.
Brazier, M., Cole, D.J., Edwards, R., 2002. O-Glucosyltransferase
activities toward phenolic natural products and xenobiotics in wheat
and herbicide-resistant and herbicide-susceptible black-grass (Alo-
pecurus myosuroides). Phytochemistry 59, 149–156.
Frear, D.S., 1968. Herbicide metabolism in plants. I. Purification and
properties of UDP-glucoside: arylamine N-glucosyl-transferase
from soybean. Phytochemistry 7, 381–390.
Gareis, C., Rivero, C., Schuphan, I., Schmidt, B., 1992. Plant metabolism
of xenobiotics. Comparison of the metabolism of 3,4-dichloroani-
line in soybean excised leaves and soybean cell suspension cultures.
Z. Naturforsch. 47c, 829–832.
Harborne, J.B., Self, R., 1987. Malonylated cyanidin 3-glucosides in
Zea mays and other grasses. Phytochemistry 26, 2417–2418.
Harms, H., Langbartels, C., 1986. Standardised plant cell suspension
test systems for an ecotoxicologic evaluation of the metabolic fate of
xenobiotics. Plant Sci. 45, 157–165.
4
.7. Enzyme assays
For DCA-N-MT assays, the mixture consisted of 100
mM Bis-Tris Propane–HCl, pH 6.5, 300 mM DCA, 9
mM [2- C]-malonyl-CoA (3.7 kBq), 22 mM cold mal-
onyl-CoA, 0.1% (w/v) BSA and 20 mg crude protein, in
a final vol. of 100 ml. After incubation at 35 C for 5
min, the reaction was stopped with of 3 ml glacial acetic
acid and the malonylated conjugates partitioned into
200 ml ethyl acetate and 100 ml of the organic phase
quantified by LSC.
The DCA-N-GT assay mixture consisted of 200 mM
Tris–HCl, pH 8.0, 5 ml of 1 mM DCA, 10 ml of
1
4
Harvey, P.J., Campanella, B.F., Castro, P.M.L., Harms, H., Licht-
¨
fouse, E., Schaffner, A.R., Smrcek, S., Werck-Reichhart, D., 2002.
Phytoremediation of polyaromatic hydrocarbons, anilines and phe-
nols. Environ. Sci. Pollut. Res. 9, 29–47.
Hinder, B., Schellenberg, M., Rodoni, S., Ginsburg, S., Vogt, E.,
ꢂ
¨
Martinoia, E., Matile, P., Hortensteiner, S., 1996. How plants dis-
pose of chlorophyll catabolites. Directly energized uptake of tetra-
pyrrolic breakdown products into isolated vacuoles. J. Biol. Chem.
2
71, 27233–27236.
hl, H.-U., Barz, W., 1995. Metabolism of the insecticide Phoxim in
Ho
¨
plants and suspension cultures of soybean. J. Agric. Food Chem. 43,
1052–1056.
1
4
[
UL- C]-UDP-glucose (962 Bq) and 100 mg protein, in

S.-H. Lao et al. / Phytochemistry 63 (2003) 653–661
661
Ko
¨
ster, J., Bussmann, R., Barz, W., 1984. Malonyl-Coenzyme A: iso-
0
Sandermann, H., 1994. Higher plant metabolism of xenobiotics: the
‘‘green liver’’ concept. Pharmacogenetics 4, 225–241.
0
flavone 7-O-glucoside-6 -O-malonyltransferase from roots of chick
pea (Cicer arietinum L.). Arch. Bichem. Biophys. 234, 513–521.
¨
Lu, Y.P., Li, Z.S., Drozdowicz, Y.M., Hortensteiner, S., Martinoia,
Sandermann, H., Schmitt, R., Eckey, H., Bauknecht, T., 1991.
Plant biochemistry of xenobiotics: isolation and properties of
soybean O- and N-glucosyl and O- and N-malonyltransferases
for chlorinated phenols and anilines. Arch. Biochem. Biophys.
287, 341–350.
E., Rea, P.A., 1998. AtMRP2, an Arabidopsis ATP binding cassette
transporter able to transport glutathione S-conjugates and chloro-
phyll catabolites: functional comparisons with AtMRP1. Plant Cell.
1
0, 267–282.
Schmidt, B., Rivero, C., Thiede, B., 1995. 3,4-dichloroaniline N-glu-
cosyl and N-malonyltransferase activities in cell cultures and plants
of soybean and wheat. Phytochemistry 39, 81–94.
Mackenbrock, U., Vogelsang, R., Barz, W., 1992. Isoflavone and
pterocarpan malonylglucosides and b-1,3-glucan- and chitin-hydro-
lases are vacuolar constituents in chickpea (Cicer arietinum L.). Z.
Naturforsch. 47c, 815–822.
Schmidt, B., Thiede, B., Rivero, C., 1994. Metabolism of the pesticide
metabolites 4-nitrophenol and 3,4-dichloroaniline in carrot (Daucus
carota) cell suspension cultures. Pest. Sci. 40, 231–238.
Schmitt, R., Kaul, J., Trenck, T.V.D., Schaller, E., Sandermann, H.,
1985. b-d-glucosyl and O-malonyl-b-D-glucosyl conjugates of pen-
tachlorophenol in soybean and wheat: identification and enzymatic
synthesis. Pest. Biochem. Physiol. 24, 77–85.
Mackenbrock, U., Gunia, W., Barz, W., 1993. Accumulation and
metabolism of medicarpin and maakiain malonylglucosides in eli-
cited chickpea (Cicer arietinum L.) cell suspension cultures. J. Plant
Physiol. 142, 385–391.
Marrs, K.A., 1996. The functions and regulation of glutathione S-
transferases in plants. Annu. Rev. Plant Physiol. Plant Mol. Biol.
Still, G.G., 1968. Metabolism of 3,4-dichloroporionanilide in plants:
the metabolic fate of the 3,4-dichloroaniline moiety. Science 159,
992–993.
4
7, 127–158.
Matern, U., 1983. Acylhydrolases from parsley (Petroselinum hor-
tense). Relative distribution and properties of four esterases hydro-
lyzing malonic acid hemiesters of flavonoid glucosides. Arch.
Biochem. Biophys. 224, 261–271.
Tommasini, R., Vogt, E., Schmid, J., Fromentau, M., Amrhein, N.,
Martinoia, E., 1997. Differential expression of genes coding for
ABC transporters after treatment of Arabidopsis thaliana with
xenobiotics. FEBS Lett. 411, 206–210.
Matern, U., Heller, W., Himmelspach, K., 1983. Conformational
changes of apigenin 7-O-(6-O-malonylglucoside), a vacuolar pig-
ment from parsley, with solvent composition and proton concen-
tration. Eur. J. Biochem. 133, 439–448.
¨
Tommasini, R., Vogt, E., Fromenteau, M., Hortensteiner, S., Matile,
P., Amrhein, N., Martinoia, E., 1998. An ABC-transporter of Ara-
bidopsis thaliana has both glutathione-conjugate and chlorophyll
catabolite transport activity. Plant J. 13, 773–780.
Matern, U., Feser, C., Heller, W., 1984. N-malonyltransferases from
peanut. Arch. Biochem. Biophys. 235, 218–227.
Van der Trenck, K.T., Hunkler, D., Sandermann, H., 1981. Incor-
poration of chlorinated anilines into lignin. Z. Naturforsch. C 36,
714–720.
Parris, G.E., 1980. Environmental and metabolic transformations of pri-
mary aromatic amines and related compounds. Residue Rev. 76, 1–30.
Robineau, T., Batard, Y., Nedelkina, S., Cabello-Hurtado, F., LeRet,
M., Sorokine, O., Didierjean, L., Werck-Reichhart, D., 1998. The
chemically inducible plant cytochrome P450 CYP76B1 actively
metabolizes phenylureas and other xenobiotics. Plant Physiol. 118,
Winkler, R., Sandermann, H., 1989. Plant metabolism of chlorinated
anilines: isolation and identification of N-glucosyl and N-malonyl
conjugates. Pest. Biochem. Physiol. 33, 239–248.
Winkler, R., Sandermann, H., 1992. N-glucosyl conjugates of chlori-
nated anilines—spontaneous formation and cleavage. J. Agric.
Food Chem. 40, 2008–2012.
1
049–1056.
Sandermann, H., Scheel, D., v.d. Trenck, T., 1983. Metabolism of
environmental chemicals by plants—copolymerization into lignin.
Appl. Polym. Sci.: Appl. Polym. Symp. 37, 407–420.
Yang, S.F., Hoffman, N.E., 1984. Ethylene biosynthesis and its reg-
ulation in higher plants. Annu. Rev. Plant Physiol. 35, 155–189.