Glycosidation of chlorophenols by Lemna minor

Article abstract of DOI:10.1897/02-649

Metabolic fate of xenobiotics in plant tissues has an important role in the ultimate fate of these compounds in natural and engineered systems. Chlorophenols are an important class of xenobiotics used in a variety of biocides and have been shown to be res

Full text of DOI:10.1897/02-649

Environmental Toxicology and Chemistry, Vol. 23, No. 3, pp. 613–620, 2004
2004 SETAC
Printed in the USA
0730-7268/04 $12.00 .00
᭧
ϩ
GLYCOSIDATION OF CHLOROPHENOLS BY LEMNA MINOR
J
AMES A. DAY and F. MICHAEL
SAUNDERS*
School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332-0512 USA
(
Received 2 January 2003; Accepted 13 June 2003)
Abstract—Metabolic fate of xenobiotics in plant tissues has an important role in the ultimate fate of these compounds in natural
and engineered systems. Chlorophenols are an important class of xenobiotics used in a variety of biocides and have been shown
to be resistant to microbial degradation. Three chlorophenyl glycosides were extracted from tissues of Lemna minor exposed to
2,4-dichlorophenol (DCP). The products were identified as 2,4-dichlorophenyl-
␤
-D-glucopyranoside (DCPG), 2,4-dichlorophenyl-
1)- -D-apiofuranoside(DCPAG).
␤
-D-(6-O-malonyl)-glucopyranoside (DCPMG) and 2,4-dichlorophenyl- -D-glucopyranosyl-(6
␤
→
␤
Identification was based on reverse phase retention (C18), electrospray mass spectra collected in negative and positive mode (ESI-
NEG and ESI-POS, respectively), and nuclear magnetic resonance (NMR) spectra comparisons to reference materials synthesized
in the laboratory. Liquid chromatography-mass spectrometry (LC-MS) analysis of plants exposed to 2,4,5-trichlorophenol (TCP)
formed analogous compounds: 2,4,5-trichlorophenyl-
copyranoside (TCPMG) and 2,4,5-trichlorophenyl- -D-glucopyranosyl-(6
drolysis with -glucosidase was ineffective in releasing the -glucosides with chemical modifications at C6. Presence of these
glucoconjugates confirmed that L. minor was capable of xenobiotic uptake and transformation. Identification of these products
suggested that chlorophenols were incorporated into vacuoles and cell walls of L. minor
␤
-D-glucopyranoside (TCPG), 2,4,5-trichlorophenyl-
␤-D-(6-O-malonyl)-glu-
␤
→
1)- -D-apiofuranoside (TCPAG). Enzyme catalyzed hy-
␤
␤
␤
.
Keywords—Glycosides
Phytoremediation
Plant metabolism
Xenobiotics
INTRODUCTION
been reported using this technique. However, enzymatic hy-
drolysis has potential shortcomings associated with substrate
Plants play an important role in the environment by pro-
viding habitat and food sources for a variety of ecosystems
[1]. Xenobiotic accumulation in plant tissues may have sig-
nificant impacts on the health of ecosystems and associated
food webs. In addition, successful implementation of phyto-
remediation and natural attenuation strategies with aquatic
plants requires positive identification of metabolites of plant
origin to satisfy scientific and regulatory concerns.
Previous studies on plant metabolism of xenobiotics have
described three types of cometabolic reactions, or phases, anal-
ogous to liver detoxification [2]. Phase 1 processes are char-
acterized by chemical modification of xenobiotics, e.g., oxi-
dation, reduction, or hydrolysis. Phase 1 products are usually
hydrophilic and have reactive functional groups amenable to
further modification. Phase 2 processes are condensations of
xenobiotics or phase 1 products with common biomolecules
such as carbohydrates, amino acids, or polypeptides. Phase 2
products have increased molecular weight and generally in-
creased hydrophilicity [3]. Phase 3 processes result in excre-
tion or sequestration of xenobiotics or metabolites into tissues
that minimize toxic effects. Phase 3 processes in plants are
dominated by sequestration [2].
specificity of enzyme preparations. In contrast, intact ␤-gly-
cosides of phenol, 2,4-dichlorophenol, and 2,4,5-trichlorophe-
nol have been extracted from axenic Lemna gibba [11,12].
Formation of malonate conjugates and malonate esterifi-
cation of conjugates is widespread in plant studies with xe-
nobiotics [5]. Malonyl addition to glucose is proposed to occur
by esterification of malonyl-coenzyme A at C6. Malonate ad-
dition is hypothesized to block further metabolism by signaling
sequestration of xenobiotics in vacuoles [13,14] or preventing
incorporation of multiple carbohydrate units onto the glycoside
[15]. Malonyl glycosides of pentachlorophenol [16] and DCP
[17] previously have been reported with axenic plants and cell
cultures.
The objectives of this study were to demonstrate that plant
metabolism plays an important role in active xenobiotic re-
moval in systems closely resembling a natural aquatic eco-
system with naturally occurring microbial populations and pro-
vide positive identification of metabolites in plant tissues. To-
ward this end, an experimental system was developed using
Lemna minor because it is easily cultured and has a recognized
role in toxicological investigations [18–20]. Both DCP and
TCP were chosen as representative xenobiotics due to use of
these compounds and their derivatives in a variety of biocides,
e.g., 2,4-dichlorophenoxyacetic acid, 2,4,5-trichlorophenoxy-
acetic acid, and triclosan.
Glycosidation, a typical phase 2 process, occurs in a variety
of plant species. In one study, 28 of 31 plant species formed
␤
-glycosides of 1,4-dihydroxybenzene or 1,3-dihydroxyben-
zene [4]. Also, a variety of herbicides have been shown to
form -O- and -N-glycosides [5,6]. Glycosides of hydrox-
␤
␤
MATERIALS AND METHODS
Materials
ylated 2,4-dichlrophenoxyacetic acid (a phase 1 product) from
soybean and wheat were identified by enzyme-catalyzed hy-
drolysis [7]. Similarly,
␤-glycosides of pentachlorophenol
All chemical reagents were reagent grade or better. DCP,
[7,8], trinitrotoluene [9], and trichloroethene [10] have also
TCP,
mond meal, deuterium oxide, malonic acid, sodium methoxide,
benzoyl chloride, 1,2:3,5-di-O-isopropylidene- -D-apiose, D-
␣-bromo-(2,3,4,6-tetra-O-acetyl)-D-glucopyranose, al-
* To whom correspondence may be addressed
(michael.saunders@ce.gatech.edu).
␣
613

614
Environ. Toxicol. Chem. 23, 2004
J.A. Day and F.M. Saunders
arabinose, -bromo-(2,3,4-tri-O-acetyl)-L-arabinopyranose,
␤
High-performance liquid chromatography-MS
and tert-butyl isocyanide were obtained from Sigma-Aldrich
(St. Louis, MO, USA). Acetonitrile (ACN), acetic acid, sodium
hydroxide, phosphoric acid, ethyl acetate, dichloromethane,
and methanol were obtained from Fisher Scientific (Pittsburgh,
PA, USA).
Lemna minor was obtained from stock sources maintained
at an outdoor, pilot-scale artificial wetland. A portion of the
culture was acclimated to laboratory conditions and main-
tained in 60-L containers over several months. Each 60-L unit
contained approximately 20 L of water, a 5 to 10 cm layer of
commercial potting soil [21], and complete surface coverage
by L. minor. Water was replenished on a weekly basis with
house service deionized water, and potting soil was added
every two to three months.
Liquid chromatography-MS was performed with an Agilent
(Palo Alto, CA, USA) Model 1100 equipped with a XDB C-
8 guard column (2.1
analytical column (2.1
ϫ
12.5 mm, 5
␮
m, Agilent) and SB-C18
m, Agilent). Reverse-
ϫ
150 mm, 5
␮
phase separation of native glycosides was accomplished with
linear gradients. Eluent flow rate was 0.3 mL/min with a 10-
␮
L injection volume. Solvent A contained 0.02% glacial acetic
acid and 5% ACN in water. Solvent B was ACN. Eluent was
held at 5% B for 4 min, raised from 5 to 38% B at 25 min,
raised from 38 to 100% B at 33 min, and held at 100% B until
50 min. All separations were conducted at 35
and phenols were detected by ultraviolet absorbance (␭ ϭ 210
nm, bandwidth 16 nm) with a diode array detector and a
ЊC. Glycosides
ϭ
single quadropole mass selective detector via an electrospray
interface (ESI) in positive (ESI-POS) or negative (ESI-NEG)
Media for exposure assessments was prepared by filtering
(Whatman GF/D, Clifton, NJ, USA) stock culture media. A
mode. The drying gas, N2, was delivered 10 L/min at 350
ЊC
and a nebulizer pressure of 2.07
ϫ
105 Pa (30 psig). Capillary
typical media preparation contained 213
Ϯ 35 mg/L suspended
and fragmentor voltages were maintained at 4000 V and 80
V, respectively. The mass spectrometer was operated in scan
solids and 30 6 mg/L chemical oxygen demand. Desired
Ϯ
initial concentrations of chlorophenols were prepared in single
batches and divided among experimental units. Experimental
units consisted of 250-mL Erlenmeyer flasks containing 100
mode from
m/z 50–1000. Aqueous and extracted samples were
amended with 0.1-mL of 1 M acetate buffer (pH 5.0) per mL
sample and filtered through a 0.2
filter prior to analysis.
␮m perfluorotetraethylene
mL of media and 3.5 to 4.0 g fresh wt (gFW) of L. minor
.
Experimental units were closed to the atmosphere with butyl-
rubber stoppers. Reported values are averages and standard
deviations of at least three experimental units unless otherwise
noted.
Plant tissue samples were prepared by separating intact
plants from media with a course screen (1–2 mm). Plants were
air-dried for 30 min and blotted dry with a paper towel. Plant
mass (gFW) obtained from an experimental unit was recorded
Perbenzoylated carbohydrates were also separated on the
Agilent 1100 system with eluents identical to chlorophenol
analyses. The gradient began at 65% B, increased to 88% B
in 30 min, and immediately stepped to 100% B for 10 min.
Separations were carried out at 35ЊC. Perbenzoylated carbo-
hydrates were monitored at 230 nm diode array detector and
a mass selective detector equipped with an atmospheric pres-
sure chemical ionization interface (APCI). The APCI data were
prior to freezing (-80
Њ
C). Frozen plants were ground to powder
collected in positive mode with corona current of 5
illary voltage of 2500 V, N2 flow rate of 4 L/min at 350
vaporizer temperature at 325 C, and nebulizer pressure of 4.14
105 Pa (60 psig). Retention was recorded relative to excess
␮
A, cap-
with precooled (-80
Њ
C) mortars and pestles. Frozen tissues
Њ
C,
were extracted three times with 5-mL aliquots of 80% ACN
per gFW. Solids were separated from each aliquot by centri-
fugation. Centrifugate was collected after each extraction and
combined (crude extract). Analytical extracts were prepared
Њ
ϫ
benzoic acid in final product mixture.
by eluting 10-mL crude extract over a preconditioned 1 g
ϫ
Preparative chromatography
6-mL solid phase extraction (SPE) tube (Supelco LC-18, Bel-
lefonte, PA, USA). Tubes were preconditioned by eluting 5-
mL ACN and, subsequently, 5-mL 5%-ACN. Tubes were
flushed with an additional 3-mL ACN. Under these conditions,
interfering compounds such as chlorophyll were retained on
the SPE tube and target analytes were not retained. The 10-
mL sample load and 3-mL flush were combined for analysis.
Final volume of eluent was determined by recording mass and
density of eluent. Resulting solutions were diluted 1:1 with
water and analyzed by LC-MS. In some cases, extracts were
further concentrated by evaporation under a stream of N2. In
these cases, phenolic concentrations were determined before
evaporation of sample. Large-scale extracts for 1H-NMR anal-
Preparative liquid chromatography was performed on an
Agilent 1050 system equipped with a SB-C18 guard column
(4.5
(4.5
ϫ
ϫ
12 mm, 5
250 mm, 5
␮
␮
m, Agilent) and SB-C18 analytical column
m, Agilent). Eluents were identical to the
analytical system. Preparative separations were performed
with an injection volume of 0.1 mL. Preparative samples were
eluted with a linear gradient, 0 to 65% B over 40 min. Eluent
composition was immediately switched to 100% B for an ad-
ditional 10 min. Separations were performed at 35ЊC. Fractions
were collected from LC effluent with a Foxy Jr. fraction col-
lector (ISCO, Lincoln, NE, USA).
Metabolite synthesis
ysis were prepared in a similar manner with 10 g
ϫ 60-mL
Both 2,4-dichlorophenyl-
␤-D-glucopyranoside (DCPG)
SPE tubes. Larger tubes were preconditioned with 50 mL ACN
and 50 mL 10% ACN. Analytes were flushed with an addi-
tional 50-mL ACN. Large-scale extracts were concentrated by
evaporation under a gentle stream of N2. For experiments with
inactivated plants, living plants were treated with two freeze-
and 2,4,5-trichlorophenyl- -D-glucopyranoside (TCPG) were
␤
synthesized by adaptation of the Koenigs-Knorr method [12].
Approximately 9 mmol of parent phenol (DCP or TCP) and
5 to 6 mmol of
␣-bromo-(2,3,4,6-tetra-O-acetyl)-D-glucopy-
ranose were dissolved in 10 mL 1 M NaOH and 15 mL of
acetone. Reaction mixtures were stirred overnight in closed
flasks. Acetone was removed by evaporation under a stream
of N2 leaving slurries of water and precipitate. Slurries were
diluted to 75 mL with water and extracted three times with
thaw cycles (-80–50ЊC) over 30 min, pretreated with 2 g/L
NaN3 for 24 to 48 h or macerated with a laboratory blender.
Inactivated tissues were then treated in the same manner as
active plants.

Glycosidation of chlorophenols by Lemna minor
Environ. Toxicol. Chem. 23, 2004
615
30-mL aliquots of dichloromethane. Aqueous layers were dis-
carded. Organic layers were extracted three times with 30-mL
aliquots of 1 M NaOH. Organic fractions were evaporated to
dryness under a stream of N2. Resulting solids were redissolved
for 3 hr in 2 M trifluoroacetic acid at 100 C. Partially meth-
Њ
ylated residues were reduced to alditols with deuterated sodium
borohydride and acetylated with pyridine/acetic anhydride.
Partially methylated alditol acetates were analyzed by GC-MS
(Hewlett-Packard 5970, Palo Alto, CA, USA) with electron
impact ionization. Products were separated on a Supelco 2330
fused-silica capillary column (30 m) with a thermal gradient:
in 30 mL warm methanol containing 16
␮M sodium methoxide
(
ϳ
35ЊC). Mixtures were heated until solutions became ho-
mogenous. After 1.5 h, 0.4 mL 2 M HCl was added to quench
the reaction. Solutions were evaporated to dryness under N2.
Further purification of the material was accomplished using
the preparative chromatographic method. Typical yields of pu-
rified products were approximately 45%.
80
Њ
C for 2 min, raised to 170
ЊC at 30ЊC/min, raised to 240ЊC
at 40
Њ
C/min and held at 240 C for an additional 5 min [27].
Њ
1H-NMR
Malonyl glycosides were synthesized by adapting the meth-
od of Roscher et al. [22]. One hundred mg of parent glycoside
(DCPG or TCPG) were combined with three molar equivalents
of malonic acid. Contents were dissolved in 10 mL of pre-
viously dried ACN. Acetonitrile was dried by saturation with
anhydrous sodium sulfate and allowed to stand at least 1 h
prior to use. Reactions were initiated by adding 1.5 molar
equivalents of tert-butyl-isocyanide to solutions. Mixtures
All samples were prepared from the purest solutions avail-
able. Synthetic and extracted samples (2–5 mg of dry product)
were dissolved in deuterium oxide (HOD, 99.9%). Correlation
spectroscopy (COSY) and proton, H, spectra were obtained
on a Bruker DRX 500 (Billerica, MA, USA). All shifts were
1
normalized to the residual deuterated water (HOD) signal,
4.5.
␦
ϭ
Hydrolysis
Enzymatic hydrolysis was accomplished with
were immediately closed and heated to 50ЊC for 1 h. Mixtures
were removed from heat and stirred overnight. Solvent was
removed by evaporation under a stream of N2. Reaction re-
sulted in approximately 50% conversion to several isomers.
Desired products were separated from reaction mixtures by
preparative chromatography. Typical yields of desired prod-
ucts were approximately 12%.
␤
-glucosi-
dase from almond meal. Enzymatic hydrolysis was carried out
in 0.1 M acetate buffer (pH 5.0). -Glucosidase was dissolved
at 1 to 3 mg/mL almond meal in pH 5.0 acetate buffer and
filtered (0.2 m) to remove particulate matter. Reactions were
␤
␮
initiated by adding 0.1 mL of almond meal solution to 1.0 mL
of aqueous test solutions. Dilute acid hydrolysis was carried
out in 0.02 M HCl at room temperature. Acid hydrolysis was
Perbenzoylation
Carbohydrate-containing samples were hydrolyzed in 1.2
carried out in 1.2 M HCl at 55ЊC.
M HCl at 100
benzoylation of hydrolysates was carried out by a modification
of the method of Oehlke et al. [24]. A 100- L sample of
hydrosylate was combined with 240 L 8 M NaOH and vor-
texed. Perbenzoylation was initiated by adding 40- L benzoyl
chloride. Reaction mixtures were vortexed 5 min. Immediately
reaction mixtures were acidified with 40 L 1.4 M H3PO4 and
extracted with a 400- L aliquot of ethyl acetate (vortexed 1
min). Three hundred L of ethyl acetate extract were pipetted
ЊC for 30 min prior to derivatization [23]. Per-
RESULTS AND DISCUSSION
␮
To examine xenobiotic fate in plant controls, the extraction
procedure was tested for DCP with inactivated L. minor with
initial concentrations of 5–100 ␮M DCP. Lemna minor inac-
tivated by freeze-thaw cycles rapidly sorbed approximately
20% of DCP from media reaching equilibrium in less than 6
h. Average recovery of DCP from inactivated systems was
␮
␮
␮
␮
␮
98.9
Ϯ 11.9% (5 levels, 3 replicates each). The TCP also
off the mixture and evaporated to dryness under N2, redis-
solved in 60% ACN, and analyzed by LC-APCI-MS. Refer-
ence chromatograms were prepared from hydrolysates of D-
arabinose, D-galactose, D-xylose, D-glucose, 1,2:3,5-di-O-is-
rapidly reached equilibrium with inactivated tissue under sim-
ilar experimental conditions. Plants pre-exposed to 2 g/L NaN3
exhibited similar sorptive capacity to thermally inactivated
tissues. In neither case were any transformation products ob-
served in media or plant tissues. In a further effort to separate
plant metabolism from microbial processes, active plants were
macerated and immediately exposed to chlorophenols for pe-
riods up to 48 h. Again, these systems exhibited no evidence
of the metabolites found in tissues of active plants.
opropylidene-␣-D-apiose (yielded free apiose), and ␤-bromo-
(2,3,4-tri-O-acetyl)-L-arabinopyranose (yielded L-arabinose).
Identifications were based on relative retention to benzoic acid,
product distribution, and APCI mass spectra.
Benedict’s test
Like systems containing inactive plants, no metabolites
were detected in media of systems containing active plants.
However, active L. minor removed chlorophenols from media
at rapid rates. For example, pseudofirst-order removal rate con-
Benedict’s solution was prepared by addition of 1.7 g cupric
sulfate, 17.3 g sodium citrate, and 11.7 g sodium carbonate to
70 mL water. Reactions were carried out at 50
Benedict’s reagent was added to 2 mL of sample (
Њ
C. One mL of
10 mM
Ϫ
stants of 0.01 to 0.17 h 1 were measured in systems with initial
ϳ
in analyte). After 1 h, reducing sugars produced a red precip-
itant, Cu2O [25]. Glucose was used as a positive control and
sucrose was used as a negative control.
concentrations of 3–100 ␮M DCP and nominal plant mass of
3.5 to 4.0 gFW L. minor in 100 mL media. Exclusion or
removal of active L. minor from the system halted aqueous
removal of chlorophenols and eliminated suspended microbial
activities as possible causative agents for observed removals.
Extracts prepared from the tissues of active plants indicated
that transformation products were present in plant tissues. A
chromatogram of a typical active plant extract from an ex-
periment with DCP is shown in Figure 1. Despite numerous
peaks in ultraviolet and MS traces, chlorophenol derivatives
were identified from extracted ion chromatograms (EIC) for
Linkage analysis
A sample of DCPA was submitted to the Complex Car-
bohydrate Research Center (CCRC), University of Georgia
(Athens, GA, USA) for glycoside linkage analysis. Samples
were dissolved in dry dimethyl sulfoxide saturated with NaOH.
Two sequential additions of methyl iodide were added to meth-
ylate free hydroxyl groups [26]. Samples were then hydrolyzed

616
Environ. Toxicol. Chem. 23, 2004
J.A. Day and F.M. Saunders
and TCP [11,12]. Likewise, nominal mass of DCPC was iden-
tical to that of 2,4-dichlorophenyl-␤-D-(6-O-malonyl)-gluco-
pyranoside, which was observed in transgenic cotton exposed
to DCP and 2,4-dichlorophenoxyacetic acid [17]. By analogy,
TCPC was hypothesized to be 2,4,5-trichlorophenyl-␤-D-(6-
O-malonyl)-glucopyranoside. Because no reference materials
were commercially available, DCPG, DCPMG, TCPG, and
TCPMG (structures shown in Fig. 2 for DCP conjugates; TCP
conjugates are similar) were prepared in the laboratory. All
synthesized compounds co-eluted and produced identical ESI-
NEG mass spectra to extracted compounds. Positive identifi-
cation was provided by comparison of the 1H-NMR spectra of
extracted and synthetic compounds, which were identical (Ta-
1
1
ble 2). The H signal assignments were confirmed with H-
COSY spectra (data not shown). The number of active protons
expected from each structure (four for DCPG and six for
DCPMG) was confirmed by direct injection of deuterated sam-
ples into ESI-interface.
No previous reports of analogs to DCPA and TCPA could
be found in the literature. Dilute acid hydrolysis of DCPA and
TCPA yielded DCPG and TCPG as transient intermediates,
and DCP and TCP as terminal products, respectively. There-
Fig. 1. Typical chromatographic trace of active plant extracts exposed
to 2,4-dichlorophenol. Trace A collected by extracting 161 from trace
B, the total ion chromatogram. Trace C was collected by ultraviolet
absorbance at 210 nm.
fore, it was concluded that DCPA/TCPA contained the ␤-glu-
copyranoside core structure of the other products. An isotope
calculator (Mass Spec Pro Calculator, ChemSW, Fairfield, CA,
USA) was used to predict possible molecular formulas based
mass to charge ratios (
phenols, i.e., 161 for DCP and
traces were also examined for potential phase 1 metabolites:
Dichlorodihydroxybenzenes ( 177), chlorophenols (
127), chlorohydroquinones ( 142), and chlorodihydroxy-
benzenes ( 144). No conclusive evidence was found for
m
/
z
) of ESI-NEG base peaks of parent
on the
␤-glycoside structure with CHO additions. Potential
m
/
z
m/z
195 for TCP. The EIC
matches included tartaryl esters or pentosyl-substituted gly-
cosides. Tartaryl esters of DCPG were prepared in the same
manner as DCPMG by substituting an equimolar amount of
tartaric acid for malonic acid. None of the tartaryl esters
formed co-eluted with DCPA, nor shared similarity with the
mass spectrum of DCPA. The 1H-NMR spectra of DCPA (Ta-
ble 2) contained a second anomeric proton (␦ ϭ 4.79) and an
increased number of signals in the ␦ ϭ 3.4 to 4.0 region con-
sistent with an additional pentose residue. These data sug-
gested that DCPA contained a pentosyl moiety glycosidically
m
/
z
m/z
m/z
m/z
any phase 1 metabolites in media or plant tissues.
Examination of plant extract ESI-NEG mass spectra of the
peaks in Figure 1 showed six chromatographic peaks with
recognizable dichlorophenyl signatures. The peak at 29 min
was identified as DCP by comparison of retention time (tr) to
standard material. Peaks at 19.5 min and 16.8 min provided
limited structural information and were not observed consis-
bound to the ␤-glucopyranoside core.
Acid hydrolysates of DCPA and TCPA were perbenzoy-
lated and compared to reference chromatograms of perben-
zoylated D-arabinose, L-arabinose, D-xylose, and apiose (Ta-
ble 3). Based on these data, the pentosyl unit of DCPA and
TCPA was determined to be apiose. The J value of the apiosyl
anomeric carbon, 3.2 Hz (Table 2), was consistent with the (3-
tently. Remaining peaks, designated DCPA ( r
DCPB (tr 17.8 min), and DCPC (tr 21.0 min) were con-
sistently observed. Degree of chlorination for m 2 ion clus-
t ϭ 17.0 min),
ϭ
ϭ
ϩ
ters in each mass spectrum was determined by comparison to
hypothetical values based on relative abundance of 35Cl/37Cl
(
ϳ
3:1). It was observed that ESI-NEG spectra for DCPA,
DCPB, and DCPC contained high values with double the
number of chlorine atoms from parent compounds. For ex-
ample, the mass spectrum of DCPC had an m 2 ion cluster
819 with four Cl atoms, and ESI-POS spectra of DCPC
contained a single ion cluster at 433 409 24 with
hydroxymethyl)-
yl- -D-apioside (3.7 Hz) reported by Ishii and Yanagisawa
[28]. The shifts in C6HaЈ and C6HbЈ were analogous to ob-
servations with DCPMG and supported assignment of a (1 6)
␤-D-erythrofuranose conformation of meth-
m/z
␤
ϩ
→
m/z ϭ
linkage. Furthermore, DCPA/TCPA were not reactive with
Benedict’s reagent, which indicated both anomeric carbons
participated in glycosidic bonds.
m
/
z
ϭ
ϩ
two Cl atoms. Therefore, it was deduced that the compounds
ϩ
formed dimers in negative mode and Na adducts in positive
A linkage analysis was performed by CCRC on a sample
of DCPA. Apiose was the only pentose residue identified in
the sample. Analysis of the mass spectrum showed that the
ion distribution of the apiose residue derived from DCPA was
similar to a terminal apioside residue (Table 4) reported by
Brillouet et al. [29]. The chromatogram of derivatized car-
bohydrate residues was dominated (44% of integrated area)
by 1,6-linked glucose residues further supporting designation
mode. With this information, nominal molecular weights of
the compounds represented by these peaks were established
and used to generate mass assignments in Table 1. Chromato-
grams of extracts from plants exposed to TCP contained anal-
ogous peaks, TCPA, TCPB, and TCPC, with nominal molec-
ular weights consistent with addition of a single chlorine atom,
relative to DCP.
Nominal molecular weights assigned to DCPB and TCPB
of a glucopyranosyl-(6
and TCPA were determined to be 2,4-dichlorophenyl-
copyranosyl-(6 1)- -D-apiofuranoside (DCPAG) and 2,4,5-
→
1)-apiofuranoside. Therefore, DCPA
were identical to nominal masses of 2,4-dichlorophenyl-
glucopyranoside and 2,4,5-trichlorophenyl- -D-glucopyrano-
side, which were observed in axenic L. gibba exposed to DCP
␤-D-
␤
-D-glu-
␤
→
␤

Glycosidation of chlorophenols by Lemna minor
Environ. Toxicol. Chem. 23, 2004
617
Table 1. Negative mode electrospray mass spectral data for chlorophenol metabolites
tra
m
/
zb
Clx
Assignment
DCPAGd
tra
m
/
zb
Clx
Assignment
TCPAGe
c
Ϫ
17.0
125
161
293
455
491
1
2
0
2
3
[M
Ϫ
H
Ϫ
162
Ϫ
HCl]
20.4
195
293
489
525
3
0
2
3
[M
[M
Ϫ
H
[M
H
Ϫ
Ϫ
ϩ
Ϫ 162]
Ϫ
[M
[M
Ϫ
H
[M
H
Ϫ
162]
Ϫ
C6H5Cl2]
Ϫ
Ϫ
Ϫ
Ϫ
Ϫ
ϩ
ϩ
C6H5Cl2]
H]
Cl]
Na]
H]
Ϫ
Ϫ
[M
Cl]
Ϫ
[M
[M
ϩ
ϩ
(479)
(513)
[M
ϩ
Na]
DCPGf
TCPGg
Ϫ
Ϫ
17.8
125
161
323
359
383
485
647
(347)
1
2
2
3
2
[M
Ϫ
[M
H
Ϫ
162
Ϫ
HCl]
21.9
195
357
383
417
(381)
3
3
4
3
[M
Ϫ
[M
[M
H Ϫ 162]
Ϫ
Ϫ
Ϫ
H
Ϫ
162]
Ϫ
H]
Ϫ
Ϫ
[M
[M
Ϫ
ϩ
H]
Cl]
ϩ
Cl]
Ϫ
Ϫ c
[M
[M
ϩ
ϩ
Ac]
ϩ
Ϫ
h
[M
ϩ
Ac]
Na]
Ϫ
4
[M
[M
ϩ
C6H5Cl2]
NAi
ϩ
[M
M
ϩ
Ϫ H]
ϩ
Ϫ
Na]
DCPMGj
TCPMGk
Ϫ
Ϫ
Ϫ
20.0
161
365
409
2
2
2
4
[M
[M
Ϫ
Ϫ
[M
H
H
Ϫ
Ϫ
Ϫ
H ]
162]
23.4
195
399
443
887
(467)
3
3
3
6
[M
[M
Ϫ
Ϫ
H
H
Ϫ
Ϫ
162]
Ϫ
CO2]
CO2]
Ϫ
Ϫ
[M Ϫ H]
Ϫ
Ϫ
819
[M
ϩ
[M
M
ϩ
Ϫ
Na]
H]
[M
ϩ
M
Ϫ H]
ϩ
ϩ
(433)
[M
ϩ
Na]
a Values reported in min.
b Mass to charge ratios. Values in parentheses collected in positive mode.
c Degree of chlorination determined by comparison of percent relative abundance of m
to theoretical values based on the natural abundance of 35Cl/37Cl.
ϩ
2 intensities
d 2,4-dichlorophenyl-
e 2,4,5-trichlorophenyl-
f 2,4-dichlorophenyl-
-D-glucopyranoside.
g 2,4,5-trichlorophenyl-
-D-glucopyranoside.
h Ac
CH3CO.
i Relative abundance values too small for comparison.
j 2,4-dichlorophenyl-
-D-(6-O-malonyl)-glucopyranoside.
k 2,4,5-trichlorophenyl-
-D-(6-O-malonyl)-glucopyranoside.
␤
-D-glucopyranosyl-(6
→
→
1)-
1)-
␤
-D-apiofuranoside.
-D-apiofuranoside.
␤
-D-glucopyranosyl-(6
␤
␤
␤
ϭ
␤
␤
trichlorophenyl-
side (TCPAG), respectively (Fig. 2).
Extracted and synthetic glycosides were subjected to en-
zyme-catalyzed hydrolysis with -glucosidase from almond
meal. Samples containing DCPG and TCPG were rapidly hy-
drolyzed with complete aglycan release (up to 213 M) in less
than 1 h. However, derivatized glycosides such as DCPMG
and DCPAG were resistant to hydrolysis under the same con-
ditions. For example, a statistically insignificant fraction of
␤
-D-glucopyranosyl-(6
→
1)-
␤
-D-apiofurano-
tegrated area). Only a small fraction of DCPAG (10.2
Ϯ
6.3%)
was released, even after extending the reaction time to 18 h.
Therefore, -glucosidase was found ineffective at accurately
␤
␤
representing the extent and identity of glycosides formed in
plant tissues.
␮
Identification of metabolites in conjunction with the reac-
tion sequence observed in Figure 3 suggested that chlorophe-
nols were metabolized by L. minor via the pathway shown in
Figure 4. This is consistent with the three-phase detoxification
scheme previously discussed. Malonated glycosides (DCPMG
and TCPMG) suggested that sequestration occurred in vacu-
oles of L. minor [13,14].
DCPMG was hydrolyzed to DCP (0.7
Ϯ 0.7% of initial in-
Apiosyl residues in conjunction with xenobiotic glycosides
of L. minor represented a novel and important discovery. Api-
ose is a rare carbohydrate with unusual structural properties.
Apiose was first discovered as a component of a flavanoid
glycoside extracted from Petroelinum sativum [30]. Apiose,
a branched-chain pentose, is exclusively associated with plant
metabolism [31] and is known to be a unique component of
cell walls in L. minor [32–34]. Detection of this residue strong-
ly supported that L. minor, and not a microorganism, was
responsible for metabolites observed in this research, and close
association of apiose and cell walls of L. minor also suggested
that apiosyl glycosides of chlorophenols were indicative of a
phase 3 biotransformation. This conclusion implied that other
plant species would likely contain analogous glycosides with
substitution of more common pentosyl units of hemicellulose,
Fig. 2. Chemical structures of chlorophenol glycosides extracted from
Lemna minor. DCPG 2,4-dichlorophenyl- -D-glucopyranoside;
2,4-dichlorophenyl- -D-(6-O-malonyl)-glucopyranoside;
2,4-dichlorophenyl- -D-glucopyranosyl-(6 1)- -D-api-
ofuranoside; and 2,4,5-trichlorophenol glycosides follow a similar ter-
minology.
ϭ
␤
DCPMG
DCPAG
ϭ
ϭ
␤
␤
→
␤
e.g., arabinose or xylose. For example,
␤-D-xylopyranosyl-

618
Environ. Toxicol. Chem. 23, 2004
J.A. Day and F.M. Saunders
Table 2. Proton nuclear magnetic resonance data for chlorophenyl glycosidesa. (The 2,4-dichlorophenyl
[DCP]-glycoside abbreviations are in Fig. 2 caption)
Atomb
DCPGc
DCPMGc
DCPAGc
HC-3
HC-5
HC-6
HC-1
HC-2
HC-4
HC-5
7.24 (2.5) d
7.02 (8.9, 2.5) dd
6.92 (9.2) d
7.10 (2.5) d
6.93 (9.0, 2.5) dd
6.79 (9.2) d
7.28 (2.5) d
7.07 (9.5, 2.5) dd
6.96 (9.1) d
4.86 (7.6) d
3.30–3.40 m
3.24 (9.1, 8.8) td
3.30–3.40 m
3.72 d
Ј
Ј
Ј
Ј
4.83 (7.4) d
4.73 (7.6)
, HC-3Јb
3.26–3.35m
3.27–3.37 m
3.21 (9.7, 8.9) td
3.30–3.40 m
3.20 (9.2, 9.4) td
3.46 m
HC-6a
HC-6b
3.61 (12.2, 2.29) dd
3.47 (12.4, 5.6) dd
4.18 (12.4, 2.0) dd
4.04 (12.0, 5.5) dd
3.58 d
HC-1
HC-2
ٞ
ٞ
ٞ
ٞ
4.79 (3.2) d
3.68 (3.2) d
3.76 m
3.47 m
6
HC-4a
HC-5a
, 4b
, 5b
ٞ
ٞ
Number of active H’se
4
6
a Recorded in deuterium oxide, 500 MHz. Chemical shift (
␦
) in ppm relative to residual water at 4.5
ppm, coupling constants (J) in Hz. d, doublet; dd, doublet-doublet; t, triplet; m, multiplet.
b Atom labels refer to Figure 2.
c Interpreted as a degenerate dd.
d Complex splitting makes these signals impossible to distinguish.
e Active protons determined by increase in mass of deuterated samples directly infused into electrospray-
interface.
(6
→
1)-D-glucose glycosides (primeverosides) of anthraqui-
results were observed in experiments with TCP (data not
shown).
Biosynthesis of glycosidic bonds usually occurs by se-
quential reaction of an aglycan with a nucleotide diphosphate.
Because nucleotide diphosphate formation requires at least one
equivalent of ATP [36], a minimal energetic cost may be as-
none were detected in cell cultures of Morinda citrifolia [35].
Progression of observed reactions were followed by sac-
rificial sampling of experimental units (Fig. 3). Uptake of DCP
was characterized by a short, rapid phase followed by a slower,
sustained phase consistent with a diffusion-controlled parti-
tioning process. Both DCPG and DCPMG were detected in
the first sample at approximately 1 h. The DCPAG was not
detected until the second sampling point at 8 h. These results
suggested that a constitutive enzyme catalyzed formation of
DCPG and DCPMG, while DCPAG required induction or an
undetected intermediate before the reaction sequence could
continue. Both DCP and DCPG quickly reached pseudo-steady
state concentrations and never accumulated to an appreciable
extent. In contrast, DCPMG and DCPAG continued to accu-
mulate over the course of the experiment. It should be noted
that no metabolites were ever observed in the aqueous phase,
even after volumetric concentration of 20 to 100 times. Similar
signed to the formation of
␤-glucosides, DCPG or TCPG, of
approximately 30.5 kJ/mol product formed. This value is the
Ϫ
approximate free energy released by hydrolysis of adenosine
triphosphate (ATP) to adenosine diphosphate (ADP) under
standard biological conditions: 25ЊC, pH 7.0 and unit activity
[37]. Similarly, an additional ATP equivalent must be used to
form malonyl or apiosyl derivatives. Thus, metabolites de-
scribed in this research represent an energetic sink to plants
storing chlorophenols in these forms. Therefore, it may be
concluded that conjugation of chlorophenols reduced free en-
ergy available for primary metabolic processes associated with
growth and maintenance. In addition, free-energy losses are
also likely to be associated with dissipation of electrochemical
potentials by free chlorophenols in plant tissues. However,
these losses are mitigated by glycosylation, which effectively
Table 3. Monosaccharide indentification by high-performance liquid
chromatography.a (Chlorophenyl-glycosides abbreviations are
presented in Fig. 2 caption)
Table 4. Comparison of ion distribution of 1-linked apiose from
Brillouet et al. [29] and pentosyl residue reported by the Complex
Carbohydrate Research Center, University of Georgia (USA);
dimethyl tetrachloroterephthalate (DCPA)
Compound
Peak 1
Peak 2
Peak 3
DL-arabinoseb
D-xylosec
D-apiose
1.41 (0.96)
1.71 (1.00)c
1.66 (0.26)
2.70 (1.00)
2.68 (1.00)
2.72 (1.00)
1.69 (0.26)
2.72 (1.00)
1.67 (0.25)
2.72 (1.00)
1.53 (0.34)
1.77 (1.00)
3.19 (0.36)
1.79 (1.00)
1.79 (1.00)
2 (1.00)
Brillouet et al. [29]
Sample derived from DCPA
D-glucose
Mass to
Relative
Mass to
Relative
D-galactose
DCPGd (glucose)
DCPAGd (apiose)
(glucose)
charge ratio abundance (%)
charge ratio
abundance (%)
87
101
117
131
146
161
173
201
205
233
26.6
100.0
88.3
87
101
118
132
146
161
174
202
205
234
21.7
100.0
76.7
30.0
5.0
40.8
14.2
3.3
TCPAGd (apiose)
(glucose)
30.9
Ͻ
5.0
a Retention reported relative to benzoic acid, ( r
t -tbenzoic Acid)/tbenzoic acid.
36.2
8.5
Reported values are the average of two determinations. Values in
parentheses are fraction of largest peak in set.
Ͻ
Ͻ
Ͻ
5.0
5.0
5.0
b D-arabinose and L-arabinose were indistinguishable by this method.
c The peak for D-xylose contained two unresolved compounds.
d Abbreviations defined in Table 1.
Ͻ
Ͻ
5.0
5.0

Glycosidation of chlorophenols by Lemna minor
Environ. Toxicol. Chem. 23, 2004
619
particular, glycosidation of xenobiotics is attributed to plants
because this type of detoxification would represent a large
expenditure of potential energy, reducing power and carbon
sources to heterotrophic microbial populations operating under
natural nutrient-limiting conditions [39]. These losses are mit-
igated in plant species due to the high levels of carbohydrates
formed from photosynthetic activity. Location of all observed
metabolites within plant tissues also supports the conclusion
that all observed transformations were mediated by plant bio-
chemistry. The strongest evidence of plant-mediated transfor-
mation was the detection of apiosides, which are exclusive to
the plant kingdom [31].
The nature of the two glycoside derivatives, e.g., DCPMG
and DCPAG, indicated the fate of DCP in tissues of L. minor
.
As stated previously, malonated glycosides have been asso-
ciated with sequestration in plant-cell vacuoles [13,14]. Ob-
served malonated glycosides, DCPMG and TCPMG, in plant
tissues were consistent with these reports, and localization
within plant tissues was counterindicative of microbial me-
tabolism in media or on plant surfaces.
Fig. 3. Concentration profile of batch plant system inoculated with
2,4-dichlorophenol (see Fig. 2 caption for glycoside descriptions).
a Total mass in 100-mL aqueous phase. b Total mass extracted from
plant tissues. c Relative mass units were derived by normalizing peak
area to the g fresh weight of plant in the reactor. 2,4-dichlorophenol
(DCP); refer to Fig. 2 legend for acronym definitions.
With L. minor, DCP and TCP transport into tissues was
directly observed with media and tissue measurements. Sub-
sequent formation of
parent chlorophenols in tissues. In turn,
␤
-glycosides required accumulation of
-glycosides of DCP
blocks phenols from scavenging electrons from energy pro-
duction.
␤
or TCP served as precursors for glycosyl derivatives, i.e., mal-
onyl and apiosyl glycosides of the parent compound. The ul-
timate fate of DCP and TCP was sequestration into cell walls
as apiosyl glycosides and vacuoles as malonyl glycosides. In
the sequestered form, the chlorophenols were less toxic to
plants. No dechlorination was observed for either substrate.
In natural systems, this work implied that considerable atten-
tion should be directed toward assessing contaminant uptake
and sequestration in exposed flora. In addition to metabolites
found in this research, it is quite likely that a large number of
similar glycosides of contaminants probably await discovery
in other species. Also, evidence was presented that reliance
on enzyme catalyzed hydrolysis provided misleading infor-
mation on the extent of glycoside conjugation in plant tissues.
In retrospect, strong structural similarities of xenobiotic
conjugates reported here and natural products suggested an
improved strategy for finding phytotransformation products of
xenobiotics. Natural products chemically related to xenobiotics
can provide new insight into the variety and extent of products.
For example, chlorophenolic glycosides reported herein, have
natural product analogs such as flavanoids, flavanols, and tri-
terpenoids, which have all been found to form glycosides in
multiple species, e.g., Dirca palustris [40], Bidensandicola
andicola [41], and Proteaceae species [42]. Therefore, if nat-
ural product analogs can be identified, it is reasonable to as-
sume that xenobiotic glycosides can be formed in similar sys-
tems. Furthermore, the breadth of these types of reactions
across species further suggested that plant roles in xenobiotic
fate may be sorely underestimated.
CONCLUSIONS
The results reported here were consistent with prior reports
of glycosidation as the initial step in xenobiotic assimilation
by plants, e.g., [11,12,38]. Although these previous reports
were based on studies of axenic plants or cell cultures, this
work demonstrated that the same processes were prominent in
natural plant systems. Over the time-scales investigated here,
the rapid assimilation of chlorophenols did not allow selection
for chlorophenol-degrading microorganisms as no microbial
transformation products were detected in the media. Attribut-
ing metabolism to L. minor is also consistent with observed
differences in nature of plant and microbial detoxification. In
Acknowledgement—We thank Les Gelbaum, Georgia Tech NMR Cen-
1
ter and Julia Kubanek, for assistance collecting and interpreting H-
NMR spectra, as well as Parastoo Azadi, for performing the linkage
analysis. This work was supported in part by the U.S. Department of
Energy at the Savannah River Site through a collaborative project
with the Education, Research and Development Association (Contract
CE FC09-97SR18911) and a National Science Foundation Traineeship
in Environmental Science and Engineering awarded to James A. Day
III through Georgia Institute of Technology. Use of commercial brand
names is included for information only and does not imply acceptance
or endorsement by the authors.
Fig. 4. Assimilation of 2,4-dichlorophenol (DCP) and 2,4,5-trichlo-
rophenol (TCP) observed in Lemna minor; adenosine triphosphate
(ATP)/ adenosine diphosphate (ADP); uridine diphosphoglucuronic
acid (UDP).

620
Environ. Toxicol. Chem. 23, 2004
J.A. Day and F.M. Saunders
REFERENCES
21. Taraldsen JE, Norberg-King TJ. 1990. New method for deter-
mining effluent toxicity using duckweed (Lemna minor). Environ
Toxicol Chem 9:761–767.
1. Wang W, Gorsuch JW, Lower WR. 1990. Overview. In Wang, W,
Gorsuch, JW, Lower WR, eds, Plants for Toxicity Assessment.
American Society for Testing and Materials, Philadelphia, PA, pp
1–2.
22. Roscher R, Steffen J-P, Herdrich M, Schwab W, Schreier P. 1996.
Synthesis of phenyl-6Ј-O-malonyl-␤ϪD-glucopyranoside. Facile
preparation of malonylated glycoconjugates. J Agric Food Chem
44:1626–1629.
2. Sandermann H Jr. 1992. Plant metabolism of xenobiotics. Trends
Biochem Sci 17:82–84.
23. Pazur JH. 1994. Neutral polysaccharides. In Chaplin MF, Kennedy
JF, eds, Carbohydrate Analysis: A Practical Approach. Oxford
University, Oxford, UK, pp 73–124.
24. Oehlke J, Brudel M, Blasig IE. 1994. Benzoylation of sugars,
polyols and amino acids in biological fluids for high-performance
liquid chromatographic analysis. Journal of Chromatography B
655:105–111.
3. Landis WG, Yu M-H. 1995. Introduction to Environmental Tox-
icology: Impacts of Chemicals upon Ecological Systems. Lewis,
Boca Raton, FL, USA.
4. Pridham JB. 1964. The phenol glucosylation reaction in the plant
kingdom. Phytochemistry 3:493–497.
5. Sandermann H. 1982. Metabolism of environmental chemicals:
A comparison of plant and liver enzyme systems. In Klekowski
EJ, ed, Environmental Mutagensis, Carcinogenesis, and Plant
Biology, Vol 1, Praeger, New York, NY, USA, pp. 3–70.
6. Shimabukuro RH, Lamoureux GL, Frear DS. 1982. Pesticide me-
tabolism in plants: Reactions and mechanisms. In Matsumura F,
Krishna Murti CR, eds, Biodegradation of Pesticides. Plenum,
New York, NY, USA, pp 21–65.
7. Harms H, Langebartels C. 1986. Standardized plant cell suspen-
sion test systems for an ecotoxicologic evaluation of the metabolic
fate of xenobiotics. Plant Sci. 45:157–165.
8. Scheel D, Schaefer W, Sandermann H. 1984. Metabolism of pen-
tachlorphenol in cell suspension cultures of soybean (Glycine max
L.) and wheat (Triticum aestivum L.). General results and iso-
lation of lignin metabolites. J Agric Food Chem 32:1237–1241.
9. Bhadra R, Wayement DG, Hughes JB, Shanks JV. 1999. Confir-
mation of conjugation processes during TNT metabolism by axe-
nic plant roots. Environ Sci Technol 33:446–452.
25. Brown WH. 1995. Organic Chemistry. Saunders College, Fort
Worth, TX, USA.
26. Ciucanu J, Kerek F. 1984. A simple and rapid method for the
permethylation of carbohydrates. Carbohydr Res 131:209–217.
27. York WS, Darvill AG, McNeill M, Stevenson TT, Albsersheim
P. 1986. Isolation and characterization of cell walls and cell wall
components. In Weissbach A, Weissbach H, eds, Methods in En-
zymology, Vol 118. Academic, Orlando, FL, USA, pp 3–40.
28. Ishii T, Yanagisawa M. 1998. Synthesis, separation and NMR
spectral analysis of methyl apiofuranosides. Carbohydr Res 313:
189–192.
29. Brillouet J-M, Gunata Z, Bitteur S, Cordonnier R, Bosso C. 1989.
Terminal apiose: A new sugar constituent of grape juice glyco-
sides. J Agric Food Chem 37:910–912.
30. Bell DJ. 1962. Apiose. In Whistler RL, Wolfrom ML, eds, Meth-
ods in Carbohydrate Chemistry, Vol I. Academic, New York,
NY, USA, pp 260–263.
10. Shang TQ, Doty SL, Wilson AM, Howald WN, Gordon MP. 2001.
Trichloroethylene oxidative metabolism in plants: The trichlo-
roethanol pathway. Phytochemistry 58:1055–1065.
31. Watson RR, Orenstein NS. 1975. Chemistry and biochemistry of
apiose. In Tipson RS, Horton D, eds, Advances in Carbohydrate
Chemistry and Biochemistry, Vol 31. Academic, New York, NY,
USA, pp 135–184.
11. Ensley HE, Barber JT, Polito MA, Oliver AI. 1994. Toxicity and
metabolism of 2,4-dichlorophenol by the aquatic angiosperm
Lemna gibba. Environ Toxicol Chem 13:325–331.
32. Hart DA, Kindel PK. 1970. Isolation and partial characterization
of apiogalacturonans from the cell wall of Lemna minor
116:569–579.
33. Roberts RM, Shah RH, Loewus F. 1967. Inositol metabolism in
. Biochem
12. Ensley HE, Sharma HA, Barber JT, Polito MA. 1997. Metabolism
of chlorinated phenols by Lemna gibba, duckweed. In Kruger
EL, Anderson TA, Coats JR, eds, Phytoremediation of Soil and
Water Contaminants Symposium. American Chemical Society,
Washington, DC, pp 238–253.
13. Komossa D, Langebartels C, Sandermann H. 1995. Metabolic
processes for organic chemicals in plants. In Trapp S, McFarlane
JC, eds, Plant Contamination. Lewis, Boca Raton, FL, USA, pp,
69–103.
J
plants, IV. Biosynthesis of apiose in Lemna and Petroselinum
.
Plant Physiol 42:659–666.
34. Golovchenko VV, Ovodova RG, Shashkov AS, Ovodov YS. 2002.
Structural studies of the pectic polysaccharide from duckweed,
Lemna minor L. Phytochemistry (Oxford) 60:89–97.
35. Van Der Plas LH, Hagendoorn MJM, Jamar DCL. 1998. An-
thraquinone glycosylation and hydrolysis. In Morinda citrofolia
cell suspensions: Regulation and function. J Plant Physiol 152:
235–241.
14. Schmitt R, Sandermann H. 1982. Specific localization of ␤-D-
glucoside conjugates of 2,4-dichlorophenoxyacetic acid in soy-
bean vacuoles. Z Naturforsch 37:772–777.
36. Brett CT, Waldron KW. 1996. Physiology and Biochemistry of
15. Edwards R, Gatehouse JA. 1999. Secondary metabolism. In Lea
PJ, Leegood RC, eds, Plant Biochemistry and Molecular Biology.
John Wiley, New York, NY, USA, pp 193–218.
Plant Cell Walls. Chapman & Hall, London, UK.
37. Voet D, Voet JG. 1995. Biochemistry. John Wiley, New York,
NY, USA.
16. 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 chlo-
rinated phenols and anilines. Arch Biochem Biophys 287:341–
350.
38. Sandermann H, Scheel D, Trenck TVD. 1984. Use of plant cell
cultures to study the metabolism of environmental chemicals.
Ecotoxicol Environ Saf 8:167–182.
39. Hall JC, Wickendon JS, Kerrin YFY. 2001. Biochemical conju-
gation of pesticides in plants and microorganisms: An overview
of similarities and divergences. In Hall JC, Hoagland RE, Za-
blotowicz RM, eds, Pesticide Biotransformation in Plants and
Animals, Vol 777. American Chemical Society, Washington, DC,
pp 89–118.
17. Laurent F, Debrauwer L, Rathahao E, Scalla R. 2000. 2,4-Dichlo-
rophenoxyacetic acid metabolism in transgenic tolerant cotton
(
Gossypium hirstum). J Agric Food Chem 48:5307–5311.
18. American Public Health Association, American Water Works As-
sociation, and Water Pollution Control Federation. 1998. Standard
Methods for the Examination of Water and Wastewater, 20th ed.
American Public Health Association, Washington, DC.
19. American Society for Testing and Materials. 1997. Standard guide
for conducting static toxicity tests with Lemna gibba G3. E 1415–
91, Philadelphia, PA.
40. Ramsewak RS, Nair MG, DeWitt DL, Mattson WG, Zasada J.
1999. Phenolic glycosides form Dirca palustris
ural Products 62:1558–1561.
. Journal of Nat-
41. Tommasi ND, Piacente S, Pizza C. 1998. Flavanol and chalcone
ester glycosides from Bidens andicola
ucts 61:973–977.
. Journal of Natural Prod-
20. U.S. Environmental Protection Agency. 1996. Ecological effects
test guidelines: Aquatic plant toxicity using Lemna spp., Tiers I
and II. EPA 712-C-96-156. Washington, DC.
42. Verotta L, Orsini F, Pelizzoni F, Torri G, Rogers CB. 1999. Poly-
phenolic glycosides from African Proteaceae
Products 62:1526–1531.
. Journal of Natural