Characterization of oxidation products of TNT metabolism in aquatic phytoremediation systems of Myriophyllum aquaticum

Article abstract of DOI:10.1021/es990436i

TNT transformation processes in sediment-free, 'natural', aquatic phytoremediation systems of Myriophyllum aquaticum were investigated with specific interest in oxidation products. Extraction procedures combining liquid-liquid extractions and solid-phase extractions were developed for the isolation of the mostly acidic, oxidized TNT metabolites. Six compounds unique from the reduction products of TNT were isolated and characterized by UV-vis, ¹H, and ¹³C NMR spectroscopy, by mass spectroscopy, and by chemical synthesis where feasible. These compounds include 2-amino-4,6- dinitrobenzoic acid, 2,4-dinitro-6-hydroxy-benzyl alcohol, 2-N-acetoxyamino- 4,6-dinitrobenzaldehyde, 2,4-dinitro-6-hydroxytoluene, and two binuclear metabolites unique from the customary azoxytetranitrotoluenes. The monoaryl compounds show clear evidence of oxidative transformations, methyl oxidation and/or aromatic hydroxylation. It is possible that oxidative transformation(s) preceded nitro reduction since studies on exposure of M. aquaticum to either 2-amino-4,6-dinitrotoluene or 4-amino-2,6-dinitrotoluene did not yield any of the oxidation products identified here. The accumulation of oxidation products was significant: 2-amino-4,6-dinitrobenzoic acid, 4.4%; 2,4-dinitro-6-hydroxy-benzyl alcohol, 8.1%; 2-N-acetoxyamino-4,6- dinitrobenzaldehyde, 7.8%; and, 2,4-dinitro-6-hydroxytoluene, 15.6%. The binuclear metabolites accounted for an estimated 5.6%. This study is the first direct evidence for oxidative transformations in aquatic phytoremediation systems. TNT transformation processes in sediment-free, `natural', aquatic phytoremediation systems of Myriophyllum aquaticum were investigated with specific interest in oxidation products. Extraction procedures combining liquid-liquid extractions and solid-phase extractions were developed for the isolation of the mostly acidic, oxidized TNT metabolites. Six compounds unique from the reduction products of TNT were isolated and characterized by UV-vis, ¹H, and ¹³C NMR spectroscopy, by mass spectroscopy, and by chemical synthesis where feasible. These compounds include 2-amino-4,6-dinitrobenzoic acid, 2,4-dinitro-6-hydroxy-benzyl alcohol, 2-N-acetoxyamino-4,6-dinitrobenzaldehyde, 2,4-dinitro-6-hydroxytoluene, and two binuclear metabolites unique from the customary azoxytetranitrotoluenes. The monoaryl compounds show clear evidence of oxidative transformations, methyl oxidation and/or aromatic hydroxylation. It is possible that oxidative transformation(s) preceded nitro reduction since studies on exposure of M. aquaticum to either 2-amino-4,6-dinitrotoluene or 4-amino-2,6-dinitrotoluene did not yield any of the oxidation products identified here. The accumulation of oxidation products was significant: 2-amino-4,6-dinitrobenzoic acid, 4.4%; 2,4-dinitro-6-hydroxy-benzyl alcohol, 8.1%; 2-N-acetoxyamino-4,6-dinitrobenzaldehyde, 7.8%; and, 2,4-dinitro-6-hydroxytoluene, 15.6%. The binuclear metabolites accounted for an estimated 5.6%. This study is the first direct evidence for oxidative transformations in aquatic phytoremediation systems.

Full text of DOI:10.1021/es990436i

Environ. Sci. Technol. 1999, 33, 3354-3361
2
,4,6-trinitrotoluene (TNT) and related synthesis byproducts.
Characterization of Oxidation
Products of TNT Metabolism in
Aquatic Phytoremediation Systems
of Myriophyllum aquaticum
Lately, phytoremediation or plant-driven bioremediation has
been viewed as a viable, low-cost option for the amelioration
of TNT-contaminated media (2). In light of such interest,
elucidating the metabolic fate of TNT in plants is relevant
to determining environmentally acceptable end point(s) of
such processes (3). Investigations into phytoremediation of
TNT include a diverse array of plant systems: yellow nutsedge
(
4), bush beans (5), aquatic and wetland species (6-9), axenic
root cultures (6, 10), garden vegetables (11), hybrid poplar
12), and other terrestrial species (13). In all of these studies,
R . B H A D R A , † , ‡ R . J . S P A N G G O R D ,
§
†
, |
J . B . H U G H E S , † A N D
D . G . W A YM E N T ,
(
J . V . S H A N K S * ,
†
the activity of the reductive pathway has been reported,
customarily leading to the formation of 2-amino-4,6-dini-
trotoluene and 4-amino-2,6-dinitrotoluene. Further, the
formation of conjugates of 2-amino-4,6-dinitrotoluene and
George R. Brown School of Engineering, Rice University,
Houston, TX, and Biopharmaceutical Development Division,
SRI International, Menlo Park, CA
4
-amino-2,6-dinitrotoluene has been recently elucidated as
a significant process of plant-catalyzed metabolic fate of TNT
beyond early reduction products (10).
TNT transformation processes in sediment-free, “natural”,
aquatic phytoremediation systems of Myriophyllum
aquaticum were investigated with specific interest in
oxidation products. Extraction procedures combining liquid-
liquid extractions and solid-phase extractions were
developed for the isolation of the mostly acidic, oxidized
TNT metabolites. Six compounds unique from the reduction
products of TNT were isolated and characterized by UV-
vis, H, and C NMR spectroscopy, by mass spectroscopy,
and by chemical synthesis where feasible. These compounds
include 2-amino-4,6-dinitrobenzoic acid, 2,4-dinitro-6-
hydroxy-benzyl alcohol, 2-N-acetoxyamino-4,6-dinitroben-
zaldehyde, 2,4-dinitro-6-hydroxytoluene, and two binuclear
metabolites unique from the customary azoxytetranitro-
toluenes. The monoaryl compounds show clear evidence
of oxidative transformations, methyl oxidation and/or aromatic
hydroxylation. It is possible that oxidative transformation(s)
preceded nitro reduction since studies on exposure of
M. aquaticum to either 2-amino-4,6-dinitrotoluene or 4-amino-
In contrast, the activity of oxidative pathway(s) of TNT
metabolism in plants has remained largely uninvestigated.
This could include transformation of the ring-substituted
methyl group and the hydroxylation of the aromatic ring.
Oxidative metabolism of herbicides is particularly well-
documented as an initial reaction leading to activation or
deactivation of the parent herbicide (14, 15). A majority of
these reactions in plants are dependent on mixed-function
oxygenases (mfo) (16, 17). Evidence of such transformations
during plant metabolism of TNT could facilitate the elucida-
tion of additional routes to conjugate formation and se-
questration of xenobiotic explosives in plants and a greater
understanding of metabolic and toxicology profiles during
phytoremediation operations. The studies presented here
were directed at the objective of isolating and characterizing
products of oxidative metabolism of TNT in plants. This was
accomplished by examining the product profile in the
extracellular sediment-free, aqueous matrix of the aquatic
plant Myriophyllum aquaticum exposed to TNT. Myriophyl-
lum is capable of metabolizing TNT in “natural” systems (6,
1
13
9
), and the present analysis provides evidence for oxidative
2,6-dinitrotoluene did not yield any of the oxidation
metabolism of TNT in its presence.
products identified here. The accumulation of oxidation
products was significant: 2-amino-4,6-dinitrobenzoic acid,
Materials and Methods
Chem icals and Materials. 2,4,6 Trinitrotoluene (TNT),
4
.4%; 2,4-dinitro-6-hydroxy-benzyl alcohol, 8.1%; 2-N-
acetoxyamino-4,6-dinitrobenzaldehyde, 7.8%; and, 2,4-dinitro-
-hydroxytoluene, 15.6%. The binuclear metabolites
4
-amino-2,6-dinitrotoluene, and 2-amino-4,6-dinitrotoluene
6
from Chemservice (Westchester, PA) were used for amend-
ment of M. aquaticum experimental systems. Reference
standards employed for analytical quantification and/ or
product identification include the following: TNT, 2-amino-
accounted for an estimated 5.6%. This study is the first
direct evidence for oxidative transformations in aquatic
phytoremediation systems.
4
,6-dinitrotoluene, and 4-amino-2,6-dinitrotoluene from
Chemservice; 2,4-diamino-6-nitrotoluene, 2,6-diamino-4-
nitrotoluene, 4-hydroxylamino-2,6-dinitrotoluene, 2-hydrox-
ylamino-4,6-dinitrotoluene, 2,2′,6,6′-tetranitro-4,4′-azoxy-
toluene (4,4′-azoxy), 4,4′,6,6′-tetranitro-2,2′-azoxytoluene
(2,2′-azoxy), and 2-amino-4,6-dinitrobenzoic acid gifted by
Dr. R. J. Spanggord (SRI International, Menlo Park, CA); 2,4-
dihydroxylamino-6-nitrotoluene by catalytic biosynthesis
according to Hughes et al. (18); 2,4,6-triaminotoluene from
Sigma-Aldrich (St. Louis, MO); and plant metabolites of TNT,
namely, the 2-amino-4,6-dinitrotoluene-derived conjugates
TNT-5.5 and 2A-5.6 and the 4-amino-2,6-dinitrotoluene-
derived conjugates TNT-6.7 and 4A-6.6 purified from ni-
troaromatic-amended C. roseus axenic roots as in Bhadra et
Introduction
The cleanup of contaminated soil and water at facilities
involved in the manufacturing, handling, packaging, and
testing of explosives continues to be a serious challenge due
to the extent and recalcitrance of explosives in the environ-
ment (1). The primary culprit is the nitroaromatic compound
*
Corresponding author present address: Department of Chemical
Engineering, Iowa State University, Ames, IA 50011-2230; phone:
(
515) 294-4828; fax: (515) 294-2689; e-mail: jshanks@iastate.edu.
†
Rice University.
‡
Present address: Department of Chemical and Bioresource
14
al. (10). For product analysis, [ring-U- C]-2,4,6-trinitrotolu-
ene (Chemsyn, Lenexa, KS), with specific activity of 21.58
mCi/ mmol and purity of 98%, was used. Solvents for NMR
Engineering Colorado State University, Fort Collins, CO 80523.
§
SRI International.
|
Present address: Department of Chemistry, Morningside College,
Sioux City, IA 51106.
3 3
analysis, CD OD (99.8 atom %) and CD CN (99.8 atom %),
3
3 5 4
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 19, 1999
10.1021/es990436i CCC: $18.00

1999 Am erican Chem ical Society
Published on Web 08/13/1999

were obtained from Aldrich (Milwaukee, WI). HPLC grade
-propanol, acetonitrile, methanol, hexane, and ethyl acetate
stages, aliquots treated to further LLE followed by solid-phase
extraction (SPE) as described below.
2
supplied by EM Science (Gibbstown, NJ) were used for
liquid-liquid extraction (LLE), solid-phase extraction (SPE),
and HPLC. Supelclean LC-8 from Supelco (Bellefonte, PA)
and silica were employed for solid-phase extraction. â-Glu-
curonidase from Helix pomatia (EC 3.2.1.31) and â-glucosi-
dase from almonds (EC 3.2.1.21) supplied by Sigma Chemicals
In the next stage, approximately half the volume of the
crude extract was diluted to 500 mL by adding ethyl acetate,
and extracted thrice with 350-mL batches of 1% (w/ v) aqueous
NaHCO . The objective of this LLE stage was to separate
3
acidic metabolites from neutral analytes (TNT and its
customary monoaminodinitrotoluene derivatives) that re-
mained in the organic phase. The aqueous phase from the
extractions was pooled, its pH adjusted to ∼2, and extracted
with ethyl acetate to obtain protonated acidic TNT metabo-
lites in organic solution. The organic phase from the final
LLE was collected and concentrated to ∼100 mL for finer
separation by several stages of SPE.
The final organic extract from the LLE separations
described above was further fractionated by a series of C8
solid-phase extractions (SPE). Visual observation of bands
and HPLC analysis were employed to monitor eluant
(
St. Louis, MO) were used for enzymatic hydrolysis of purified
analytes. Stock solutions were prepared in HPLC-grade
1
4
methanol: TNT, 25 mg/ mL, and C-TNT, 3130 dpm/ µL.
Plants. M. aquaticum (parrot feather) plants were main-
tained outdoors, rooted in a bottom layer of compost and
gravel, in 20-gal containers that were recharged periodically
with water. Prior to the start of each study, plants were
removed from outdoor storage and acclimated to indoor
conditions for 24 h.
8
Analytical Methods. C reverse-phase HPLC with PDA
detection as described in Hughes et al. (6) was employed for
the quantification of the following: TNT; and the reduction
derivatives, 2-amino-4,6-dinitrotoluene, 4-amino-2,6-dini-
trotoluene, 4-hydroxylamino-2,6-dinitrotoluene, 2-hydroxy-
lamino-4,6-dinitrotoluene, 2,4-diamino-6-nitrotoluene, 2,6-
diamino-4-nitrotoluene, and 2,4-dihydroxylamino-6-nitro-
toluene. A simple modification of the same method was
employed for the HPLC analysis of oxidized TNT metabo-
fractions. The first SPE stage was elutions on a C
8
stationary
phase with mixtures of H PO (10 mM) and 2-propanol. The
3
4
column was hand-packed with ∼40 g solids in a glass tube
of dimensions, 250 (L) × 90 mm (φ). The elution gradient
3 4 3 4
was from H PO to 79/ 21 (% v/ v) H PO / 2-propanol; final
washes include MeOH followed by ethyl acetate. Six previ-
ously unreported TNT metabolites were separated by this
procedure, either pure or in mixtures with TNT, monoamines,
and/ or each other. One TNT-metabolite eluted as a pure
lites isolated in this study, ion-suppression HPLC on C
stationary phase using a mobile phase of 82/ 18 (% v/ v) 10
mM H PO / 2-propanol. The azoxy-tetranitrotoluenes, 4,4′-
8
fraction with 95/ 5 H
as a pure fraction with 90/ 10 H
metabolite,which eluted in 85/ 15 H
3
PO
4
/ 2-propanol, and a second eluted
PO / 2-propanol. A third TNT
PO / 2-propanol with
3
4
3
4
azoxy and 2,2′-azoxy, were analyzed by C18 reverse-phase
HPLC with PDA detection as described in Bhadra et al. (10).
The TNT conjugate metabolites, TNT-5.5 and 2A-5.6, and
TNT-6.7 and 4A-6.6 were analyzed by C18 reverse-phase HPLC
with PDA detection as described in Bhadra et al. (10). Peak
identification in all cases was based on comparisons to
retention times and UV-vis spectra of authentic standards
as well as spiking with authentic standards as necessary.
3
4
TNT as the major impurity, was purified by silica-SPE as
outlined in Bhadra et al. (10). The remainder three metabolites
coeluted as mixtures in elutions of 83/ 17 to 81/ 19 H PO /
3 4
2-propanol. TNT and monoamines were removed from these
fractions with silica-SPE as previously performed. Beyond
that, an additional re-elution on C
8
3 4
-SPE with H PO (10 mM)/
2-propanol was sufficient to purify one of these TNT
metabolites. Another C -SPE method, which used gradient
elutions of H PO (10 mM)/ CH CN from 80/ 20 to 75/ 25 (%
v/ v), was formulated for the separation of the remainder two
analytes. Except the first C -SPE of the crude extract,
1
H NMR spectra were acquired on a Bruker AC-250 MHz
8
1
spectrometer relative to either CD
3
OD ( H: δ, 3.31 and 4.87
3
4
3
1
13
ppm) or CD
3
CN ( H: δ, 1.94 ppm). C NMR spectra were
OD
CN ( C: δ, 118.69 and 1.39). Chemical
shifts are reported in ppm and coupling constants in hertz
Hz). Mass spectroscopy was performed by direct probe
chemical ionization mass spectroscopy (MS-CI) on a Finnigan
MAT 95 spectrometer with CH as the ionizing gas. Radio-
activity was detected and quantified using a Beckman LS
acquired with the same spectrometer relative to CD
3
8
1
3
13
(
C: δ, 49.15) or CD
3
subsequent SPE fractionations were executed in 30-mL glass
syringes with 20-mL bed volume. All pure fractions/ TNT
metabolites were dried without heating under vacuum as
previously described and redissolved in MeOH for chemical
analysis.
Hydrolysis of TNT Metabolites. To evaluate the formation
of hydrolyzable metabolites of TNT, such as conjugates and
sugar adducts, metabolites that were isolated were treated
by enzymatic hydrolysis as described in Bhadra et al. (10).
The enzymes used include â-glucuronidase and â-glucosi-
dase.
(
4
6
500 scintillation counter as described in Hughes et al. (6).
Extraction and Separation of Transform ation Products.
Intracellular unbound metabolites of nitroaromatic trans-
formation were determined as described in Bhadra et al.
(
10), starting with cold extraction of freeze-dried biomass
aliquots with methanol. Extracellular products of TNT
transformation by M. aquaticum were extracted from the
residual aqueous milieu of exposed plants in the large-scale
aquarium study (described below in Materials and Methods).
The aqueous mixture was extracted by LLE (liquid-liquid
extraction) in 1-L batches with equal volumes of ethyl acetate
Transform ation Studies. Studies on nitroaromatic me-
tabolism in natural aquatic systems of M. aquaticum were
conducted at three scales: TNT-exposure studies in a reaction
volume of 250 mL in 500-mL wide-mouth Erlenmeyer flasks;
TNT-exposure studies in a reaction volume of 24 L in a 30-L
aquarium; and aminodinitrotoluene-exposure studies in a
reaction volume of 6 Lin 10-L aquariums. At the start of each
study, M. aquaticum plants, previously acclimated to room
temperature for 1 day, were washed thoroughly to remove
adherent soil, gravel, and microflora; patted dry; weighed;
and introduced into the aqueous nitroaromatic solution in
each reactor. Initial biomass density in these studies were as
follows: 250 mL scale, 100 g/ L fresh weight (FW); 24 L scale,
∼40 g/ L FW; and 6 L scale, ∼30 g/ L FW. All reaction systems
were either maintained in darkness or cloaked adequately
from light to prevent photodegradation reactions. Aquarium
volumes were calibrated and marked at 1-L intervals, and
each aquarium was mounted on wooden blocks so that the
(
HPLC-grade) following an initial pH adjustment to ∼2 by
the addition of H PO (1 M). LLEs were conducted in 2-L
3
4
separatory funnels; for each batch, the aqueous phase was
discarded and the organic phase collected. After the entire
aqueous solution was extracted in this manner, the organic
extracts were pooled and concentrated (without heating) to
6
1
00 mL under strong vacuum (10- mmHg) using a rotary
evaporator (Buchler model RE-121). This initial LLE stage
enabled the preferential separation of neutral analytes
(
including TNT and reduction derivatives) and acidic TNT
metabolites from the aqueous milieu. The concentrated crude
extract was stored at 4 °C, and in subsequent separation
VOL. 33, NO. 19, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 3 5 5

aqueous solution could be mixed using Teflon-covered stir
bars driven by several magnetic stirrers placed beneath the
tank.
The initial concentrations of nitroaromatic reactants in
these studies were as follows: 250 mL scale, 75 mg/ L TNT,
1
4
and 18 900 dpm/ mL C-TNT; 24 L scale, 64 mg/ L TNT and
1
4
1
4
8.5 dpm/ mL C-TNT; and 6 L scale, 23 mg/ L 2-amino-
,6-dinitrotoluene or 22 mg/ L 4-amino-2,6-dinitrotoluene.
For the small-scale study (in 500-mL shake flasks), stock
solutions were added to tap water in shake flasks to attain
1
4
initial target levels of TNT and C-TNT. A different approach
was adopted for the larger-scale tank studies since methanol
that was used as solvent in stock solutions, supported
microbial growth causing progressive cloudiness. For these,
the appropriate mass of nitroaromatic compound (TNT,
2
-amino-4,6-dinitrotoluene or 4-amino-2,6-dinitrotoluene)s
calculated from the product of the reaction volume (24 or
L) and the initial target concentrationswas dissolved in 10
1
4
FIGURE 1. Levels of TNT-derived C-label and nitroaromatic
compounds (TNT, 2-A-4,6-DNT and 4-A-2,6-DNT) in the aqueous
phase of sediment-free M. aquaticum exposed to TNT, and the
6
mL of HPLC-grade ethyl acetate and applied across the
bottom of the tank. The solution was then air-dried, resulting
in the nitroaromatic being plated across the bottom of the
aquarium. The appropriate volume of tap water was then
poured into the tank and stirred for 3-4 days. Dissolution
was monitored visually (absence of flakes) as well as by
measuring aqueous concentration with HPLC; 2-3 days were
sufficient for complete dissolution and to reach the initial
target concentration.
14
extractability of C-label into ethyl acetate. Extractability analysis
was preceded either by no pH adjustment or adjustment to pH 1,
, 8, or 13. Scale of study, 250 mL of reaction volume in a 500-mL
Erlenmeyer flask. Initial concentrations: TNT, 75 mg/L; C-TNT,
8,900 dpm/mL.
4
14
1
phase, pre-extraction pH adjustment appeared to have little
effect (Figure 1). The initial data are also an estimate of
extraction efficiencies. Over time, the levels of nitroaromatics
The purpose of the small-scale study (in 500-mL shake
flasks) was to examine the pH dependence of the extractability
of plant metabolites of TNT into ethyl acetate. Further, this
study segment was aimed at formulating an optimal approach
for separating TNT metabolites from the dilute aqueous phase
of TNT-exposed experimental plant systems. TNT-derived
(
b) and 14C-label (0) in the extracellular aqueous phase
decreased, with the former being lower than the latter as
reported previously (6). This provided strong evidence for
M. aquaticum-mediated transformation to products other
than the reduction derivatives of TNT. A profile of trans-
1
4
formants with pK -dependent hydrophobicities was also
C-label was employed to evaluate the extractability of TNT
a
evident from the pH dependence (at pH of 1, 4, 8, or 13) of
the extractability of aqueous-phase TNT metabolites into
ethyl acetate. Beyond the initial time point, 14C extractabilities
at most pH were higher than the level of TNT and monoamine
derivatives and were favored by pre-extraction adjustment
to acidic pH. The extractability outcome at pH 1 was
significant from that with no pH adjustment (R ) 0.02, paired
t-test) and also from the extractability outcome at pH 13 (R
metabolites, since the possibility of unidentified ones being
formed appeared significant (6). One TNT-amended flask
was sacrificed at each sampling event, and the extractability
1
4
of aqueous-phase C-label into ethyl acetate measured
following pH adjustment of 25-mL aliquots to 1, 4, 8, or 13.
The control in this situation was no pH adjustment. The
purpose of the aquarium-scale nitroaromatic-exposure stud-
ies was to monitor nitroaromatic metabolism by plants and
in addition, to harvest the aqueous phase milieu for TNT
metabolites. For these studies, nitroaromatic-containing
aqueous controls devoid of plants were maintained to
monitor abiotic degradation, evaporation, and transpiration.
The controls were performed in a reaction volume of 6 L in
)
0.01, paired t-test) over the duration of the study.
Myriophyllum-catalyzed Biosynthesis and Separation
of TNT Metabolites. A 24-L sediment-free, aquatic system
of M. aquaticum was employed to generate TNT metabolites
for subsequent separation. The levels of TNT, its reduction
products 2-amino-4,6-dinitrotoluene and 4-amino-2,6-dini-
trotoluene, and 14C-label in the extracellular aqueous phase
of the TNT-exposed plants are depicted in Figure 2. TNT (2)
and its aminated derivatives (0, 9) combined for a total
residual level of 18% initial moles following exposure for 12
days. At the same time, the level of TNT-derived 14C-label (b)
in the aqueous phase was 64% of initial, the difference
presumably representing unidentified products.
1
0-L aquariums for monoamine exposure studies and in a
reaction volume of 12.5 L in 20-L aquariums for TNT-exposure
studies. Samples were removed periodically to monitor
extracellular concentrations of known nitroaromatic com-
1
4
pounds, detectable products, and C-label. Prior to each
sampling event, the aqueous volume in the control tank was
measured, and the appropriate volume of water added to
each tank to compensate for evaporation losses. At the end
of each aquarium-scale experiment, the plants were separated
from the aqueous phase, rinsed with freshwater, and frozen
at -20 °C. The aqueous phase volume was measured,
combined with the rinsate, and the combined rinsate was
stored at 4 °C. Subsequently, it was utilized for separation
and analysis of TNT metabolites.
TNT-derived plant metabolites were separated from the
extracellular aqueous phase of the large-scale aquarium
reactor study as described in Materials and Methods following
the exposure of M. aquaticum to TNT for 12 days. Ninety
14
percent of the aqueous-phase TNT-derived C-label or 58%
14
of initial C-label was separated by liquid-liquid extraction
with ethyl acetate after initial pH adjustment to ∼2. The final
LLE-stage extract (base extracted) contained several UV-
active, TNT-derived analytes of interest. As shown in Figure
Results
pH Dependence of Extractability of TNT Metabolites. The
initial characterization of TNT metabolites in aquatic systems
of M. aquaticum was performed in 500-mL shake flasks as
described in the previous section. The pH dependence of
the extractability of aqueous-phase TN metabolites into ethyl
acetate was examined. At the start of the exposure study,
when only unmetabolized TNT was present in the aqueous
3, these were eluted by ion-suppression C
times of 5.1, 7.3, 8.4, 10.2, 11.6, and 14.5 min. These analytes
were separated and purified by C column chromatography
8
-HPLC at retention
8
as described in Materials and Methods and subjected to
further analysis. Som e 2-am ino-4,6-dinitrotoluene and
4-amino-2,6-dinitrotoluene was also found in the extract at
3
3 5 6
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 19, 1999

FIGURE 2. Levels of TNT-derived 14C-label, TNT, 2-A-4,6-DNT, and
-A-2,6-DNT in the extracellular aqueous phase of sediment-free
4
FIGURE 4. Levels of 2-A-4,6-DNT or 4-A-2,6-DNT and products in
aquatic systems of M. aquaticum exposed to either monoamine:
(A) levels in the extracellular aqueous phase over a 32-day exposure
period; (B) total levels in the system (aqueous and unbound biomass
phase) after exposure for 32 days. Scale of study, 6-L reaction volume
in a 10-L aquarium. Initial concentrations: 2-amino-4,6-dinitro-
toluene, 23 mg/L; 4-amino-2,6-dinitrotoluene, 22 mg/L.
M. aquaticum and biomass-free aqueous control over a 12-day period.
Live plants were exposed to initial levels of 64 mg/L TNT and 18.5
14
dpm/ml C-TNT in a 24-L reaction volume in a 30-L aquarium. The
biomass free aqueous control was conducted with initial levels of
14
2
5 mg/L TNT and 33 dpm/mL C-TNT in a 12.5-L reaction volume in
a 20-L aquarium.
zaldehyde (HPLC retention time, 8.4 min); and 2,4-dinitro-
6
-hydroxytoluene (HPLC retention time, 10.2 min). Further,
the list of TNT metabolites characterized also includes two
binuclear (or diaryl) metabolites unique from the customarily
detected azoxy-tetranitrotoluene derivatives of TNT; these
have HPLC retention times of 11.6 and 14.5 min. The
1
13
assignments are supported by H and C NMR spectroscopy
Tables 1-3), mass spectroscopy (Table 4), and wherever
(
feasible, such as in the case of 2-amino-4,6-dinitro benzoic
acid, independently by chemical synthesis.
1
On the basis of H NMR, each monomer derivative of
FIGURE 3. Chromatogram (at 230 nm) of ion-suppression C -HPLC
8
TNT isolated in this study has two magnetically nonequivalent
aromatic protons (Table 1). Similar to the nonequivalent
aromatic protons of 2-amino-4,6-dinitrotoluene, the non-
equivalent aromatic protons of monomers isolated in this
study exhibited long-range coupling of the order 2-2.25 Hz.
With respect to nonaromatic protons, two benzylic protons
(at δ ) 4.97 ppm) were detected in the metabolite identified
as 2,4-dinitro-6-hydroxy-benzyl alcohol, close to Schoolery’s
predictions (δ ) 4.64 ppm) for an internal methylene group
attached to terminal phenyl and hydroxy residues (19).
Another proton resonance detected in the same metabolite
at 9.6 ppm is that of the ring-substituted hydroxyl proton,
close to the reported value (δ ) 9.25 ppm) of the hy-
droxyl proton in phenol. Alkyl protons corresponding to
methyl groups were detected: for ring-substituted methyl in
the metabolite identified as dinitrohydroxytoluene (δ )
2.61 ppm) and for the acetoxymethyl (δ ) 2.6 ppm) in the
metabolite identified as N-acetoxyaminodinitrobenzalde-
hyde.
analysis of the extracted aqueous phase of M. aquaticum exposed
to 64 mg/L TNT for 12 days in a 30-L aquarium. Extraction includes
two LLE steps: an initial extraction into ethyl acetate after pH
adjustment to ∼2 and a second step of base extraction with 1%
3
(w/v) NaHCO . The HPLC method employed a Novapak-HR column
(
H
7.8 × 300 mm, Waters) and a mobile phase of 82/18 (% v/v) 10 mM
3
PO /2-propanol at 5 mL/min. Peak labels are: A, 2-amino-4,6-
4
dinitrobenzoic acid; B, 2,4-dinitro-6-hydroxybenzyl alcohol; C, 2-N-
acetoxyamino-4,6-dinitrobenzaldehyde; D, TNT; E, 2,4-dinitro-6-
monohydroxytoluene; F, binuclear metabolite 1; G, binuclear
metabolite 2; H, 2-A-4,6-DNT; I, 4-A-2,6-DNT.
retention times of 21.8 and 24 min, respectively, due to the
slight solubility of ethyl acetate in water.
Characterization and/or Identification of TNT Metabo-
lites. The characterization of TNT metabolites detected
initially by HPLC-UV, resulted in the identification of several
mono- and diaryl compounds. The monomers include (refer
to Figure 3) 2-amino-4,6-dinitrobenzoic acid (HPLC retention
time, 5.1 min.); 2,4-dinitro-6-hydroxybenzyl alcohol (HPLC
retention time, 7.3 min); 2-N-acetoxyamino-4,6-dinitroben-
The intact aromatic ring of the metabolites identified as
2-amino-4,6-dinitrobenzoic acid and 2,4-dinitro-6-hydroxy-
VOL. 33, NO. 19, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 3 5 7

1
TABLE 1. Summary of H NMR Resonances of Mono-aryl TNT-Metabolites Isolated from the Extracellular Aqueous Phase of M.
aquaticum Exposed to TNT, Compared to Nitroaromatic Standards
other
H
H-3b
H-5b
H-7b
1 b
analyte
2
4
,4,6-trinitrotoluenea (RA ) CH3; RB ) RC ) RD ) NO2)
8.95(s)
7.21 (s)
c
c
2.64 (s)
2.21 (s)
a
-am ino-2,6-dinitrotoluene (RA ) CH3; RB ) RD )
NO2; RC ) NH2)
a
2
2
2
2
-am ino-4,6-dinitrotoluene (RA ) CH3; RB ) NH2;
7.7 (d) J ) 2.25 Hz
7.6 (d) J ) 2.13 Hz
8.13 (d) J ) 2.25 Hz
9.45 (d) J ) 2.18 Hz
7.8 (d) J ) 2.35 Hz
7.8 (d) J ) 2.13 Hz
7.85 (d) J ) 2.25 Hz
8.93 (d) J ) 2.22 Hz
2.22 (s)
RC ) RD ) NO2)
a
-am ino-4,6-dinitrobenzoic acid (RA ) COOH; RB )
NH2; RC ) RD ) NO2)
a
,4-dinitro-6-hydroxy-benzyl alcohol (RA ) CH2OH;
4.97 (s)
9.6 (s)
RB ) RC ) NO2; RD ) OH)
a
-N-acetoxyam ino-4,6-dinitrobenzaldehyde (RA ) CHO;
2
2
.62 (s)
.61 (s)
RB ) NHOCOCH3; RC ) RD ) NO2;)
a
4
2
-N-acetoxyam ino- 2,6-dinitrobenzaldehyde
8.85 (s)
9.64 (d) J ) 2.1 Hz
c
a
,4-dinitro-6-hydroxytoluene (RA ) CH3; RB )
RC ) NO2; RD ) OH)
8.97 (d) J ) 2.1 Hz
a
2
,6-dinitro-4-hydroxytoluene
8.93 (s)
c
a
1
CN ( H: δ, 1.94). b s, singlet; d, doublet. c Chem ical shift equivalent to H-3
1
Chem ical shifts relative to CD
due to arom atic sym m etry.
3
OD ( H: δ, 3.31 and 4.87) and/or CD
3
13
TABLE 2. Summary of C NMR Resonances of Mono-aryl TNT-Metabolites Isolated from the Extracellular Aqueous Phase of M.
aquaticum Exposed to TNT, Compared to Nitroaromatic Standards
analyte
C-1
C-2
C-3
C-4
C-5
C-6
C-7
other C
2
4
2
2
2
2
4
2
2
,4,6-trinitrotoluenea
-am ino-2,6-dinitrotoluene
-am ino-4,6-dinitrotoluene
134.9
112.8
126.2
114.8
129.5
153.0
153.8
150.1
150.4
148.9
123.6
113.1
117.2
106.0
114.6
147.5
149.6
147.4
153.0
152.4
c
c
d
d
15.70
13.76
13.6
a
b
108.7
106.4
111.0
150.9
151.7
159.3
a
b
-am ino-4,6-dinitrobenzoic acid
,4-dinitro-6-hydroxy-benzyl alcohol
∼179
a
55.8
03.3
-N-acetoxyam ino- 4,6-dinitrobenzaldehyde 130.4b 141.2b 122.6b 155.2b 114.9b 150.4b
2
165.96, 15.3
-N-acetoxyam ino- 2,6-dinitrobenzaldehyde 122.7b 150.4b 122.6b 146.0b
c
d
,4-dinitro-6-hydroxytoluene
,6-dinitro-4-hydroxytoluene
125.8b 150.9b 110.8b 147.5b 116.8b 157.8b
119.8b 150.9b 116.8b 154.3b
1
5.6
c
d
a
OD (13C: δ, 49.15) and/or CD
CN (13C: δ, 118.69 and 1.39). b Chem ical shift estim ated from Ewing (24). c Chem ical
Chem ical shifts relative to CD
3
3
d
shift equivalent to C-3 due to arom atic sym m etry. Chem ical shift equivalent to C-2 due to arom atic sym m etry
1
TABLE 3. Summary of H NMR Resonances of Binuclear TNT-Metabolites Isolated from the Extracellular Aqueous Phase of M.
aquaticum Exposed to TNT, Compared to Azoxy Standards
aromatic
protons
non-aryl
protons
analyte
4
2
,4′,6,6′-tetranitro-2,2′-azoxytoluene
9.32(d), 9.19 (d)
2.65 (s)
2.76 (s)
2.65 (s)
2.68 (s)
2.18 (s)
1.29 (s)
9
.02 (s), 8.8 (d)
8.99 (s)
.15 (s)
,2′,6,6′-tetranitro-4,4′-azoxytoluene
9
binuclear m etabolite 1
rt in ion-suppression C8-HPLC, 11.6 m in)
8.75 (d), J ) 2.5 Hz;
8.44 (d), J ) 2.5 Hz
(
7
7
.58 (d), J ) 2 Hz
.44 (d), J ) 1.75 Hz
binuclear m etabolite 2
rt in ion-suppression C8-HPLC, 14.5 m in)
8.26 (s)
2.58 (s)
1.27 (s)
(
7.78 (d), J ) 1.75 Hz;
7
.46 (d), J ) 1.75 Hz
benzyl alcohol were confirmed by the six aromatic carbons
proposed assignments: 2-amino-4,6-dinitrobenzoic acid,
227.0; and 2,4-dinitro-6-hydroxybenzyl alcohol, 214.1. In
addition, characteristic fragments, less a molecule of water,
were observed for both. In the case of 2-amino-4,6-dini-
trobenzoic acid, the identity was also confirmed by inde-
pendently synthesizing the analyte.
1
3
in the C NMR spectra of both isolates (Table 2) in the
1
3
proximity of estimated values. A C-resonance detected in
the latter at δ ) 55.8 ppm was that of the benzylic carbon.
The molecular ions observed by desorption chemical ioniza-
tion MS (Table 4) verified the molecular masses of both
3
3 5 8
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 19, 1999

TABLE 4. Summary of Mass Spectral Fragments of TNT-Metabolites Isolated from the Extracellular Aqueous Phase of M.
aquaticum Exposed to TNT
observed molecular
primary fragment
ion observed
observed
reagent ions
a
analyte
ion (M + 1)
2
2
-am ino-4,6-dinitrobenzoic acid
,4-dinitro-6-hydroxy-benzyl alcohol
228.0
215.1
270.1
199.2
210.0
197.1
228.1
181.1
364.1, 336.1, 292.1,
248.1, 198.1
332.1, 320.1, 292.1
256.1, 268.1
243.1
298.1, 310.1
227.2, 239.2
N-acetoxyam ino dinitrobenzaldehyde
dinitro m onohydroxy toluene
binuclear m etabolite 1
(rt in ion-suppression C8-HPLC, 11.6 m in)
binuclear m etabolite 2
(RT in ion-suppression C8-HPLC, 14.5 m in)
a
+
+.
Corresponds to M + C
2
H
5
3 5
or M + C H
Desorption chemical ionization MS of TNT-derived
2-amino-4,6-dinitrobenzoic acid, 4.4%; 2,4-dinitro-6-hy-
droxy-benzyl alcohol, 8.1%; 2-N-acetoxyamino-4,6-dini-
trobenzaldehyde, 7.8%; 2,4-dinitro-6-hydroxytoluene, 15.6%.
Exposure of M. a qua ticum to 2-A-4,6-DNT and 4-A-2,6-
DNT. To investigate the sequence of transformations, M.
aquaticum plants were exposed separately to 2-amino-4,6-
dinitrotoluene and 4-amino-2,6-dinitrotoluene in 6-L aquatic
systems of roughly the same biomass density as with TNT-
exposure studies. The depletion of both monoamines fol-
lowed each other closely with apparent pseudo-first-order
kinetic rates (normalized to biomass) of 0.00195 and 0.0025
L gFW day , respectively. The depletion of either mono-
amine in biomass-free, aqueous controls was marginal. De-
tectable transformation products that emerged from a total
analysis (extracellular and intracellular unbound) of each
aquatic plant system after exposure for 32 days, include low
levels (<1% of initial) of 2,4-diamino-6-nitrotoluene and 2,4-
dihydroxylamino-6-nitrotoluene. These were accumulated
in the extracellular aqueous phase only; none was detected
in the biomass phase. Only in the case of exposure to 2-amino-
4,6-dinitrotoluene, the TNT conjugate metablite 2A-5.5 (10)
accumulated extracellularly to a maximum level of 0.05%.
The transformation of 4-amino-2,6-dinitrotoluene also ap-
peared more complete than that of 2-amino-4,6-dinitro-
toluene, 26 and 35%, respectively, remained untransformed.
A larger residual fraction accumulated intracellularly in both
cases, about 60-70% of the total. Related compounds that
were identified in TNT-exposure studies, namely, 2-amino-
4,6-dinitrobenzoic acid, 2-N-acetoxyamino-4,6-dinitro ben-
zaldehyde, and the binuclear metabolites, were not de-
tected when the starting material was either amino-dini-
trotoluene.
metabolites identified as 2-N-acetoxyamino dinitrobenz-
aldehyde and 2,4-dinitro-6-hydroxytoluene, yielded molec-
ular masses of 269.1 and 198.2, respectively (Table 4);
characteristic fragment ions were obtained, less an acetyl
1
3
residue and a molecule of water, respectively. The
C
NMR spectra (Table 2) supported the following in these
isolates: a ring-substituted methyl carbon in the dinitro-
hydroxytoluene (δ ) 15.6 ppm); acetoxy carbon (δ ) 165.9
ppm), acetoxy methyl carbon (δ ) 15.3 ppm), and ring-
substituted carbonyl carbon (δ ) 203.3 ppm) in the N-
acetoxyamino dinitrobenzaldehyde. In addition, isomers of
both compounds were possibly coeluted during separation
as observed from the appearance of 12 resonances in the
-
1
-1
1
3
aromatic region of C NMR spectra, and of two symmetric
1
aromatic protons as well in H NMR spectra. The spectral
evidence supports the proposed structures. Positive iden-
tifications will need comparative data from authentic stan-
dards.
The two remainder TNT-derived analytes were designated
as diaryl metabolites based on a combination of evidence.
Their molecular masses are considerably greater than that
of TNT and monoamine derivatives, and monomer analytes
isolated in this study (Table 4). Both analytes have resonances
1
for four aryl protons in their H NMR spectra as well as for
six nonaryl protons (Table 3). The six nonaromatic protons
occur as a pair of singlet resonances, one corresponding to
an intact ring-substituted methyl-group (δ ) 2.2-2.6 ppm).
Both are composed of dissimilar aromatic units: binuclear
metabolite 1 is composed of two nonsymmetric aryl units
inferred from the two pairs of nonequivalent protons with
long-range coupling and binuclear metabolite 2 is composed
of one symmetric and one nonsymmetric aromatic unit. It
is unlikely that these compounds are conjugates with sugar
adducts since characteristic resonances were neither ob-
Discussion
The objective of this study was to investigate oxidative
transformation as a possible biocatalytic route for the
metabolism of TNT by plants. A growing body of investiga-
tions have reported the reductive transformation of TNT by
plants (4-9, 11-13); the resultant list of TNT metabolites
includes 2-amino-4,6-dinitrotoluene; 4-amino-2,6-dinitro-
toluene; 2,4-diamino-6-nitrotoluene; 4-hydroxylamino-2,6-
dinitrotoluene; and azoxytetranitrotoluenes. Oxidative trans-
formations of TNT, however, remain comparatively un-
reported but are of interest from the perspectives of diversity
of plant processes, of environmental fate in natural systems,
and of acceptable “designed” fate in phytoremediation
systems. The results of studies reported herein, conducted
in sediment-free, natural aquatic systems of M. aquaticum,
provide evidence for an array of metabolic products of TNT
arising from plant-catalyzed oxidative metabolism, including
oxidation of the ring-substituted methyl group and aromatic
hydroxylation.
1
served in the H NMR spectra, nor did enzymatic hydroly-
sis with â-glucuronidase and â-glucosidase result in their
decomposition. Comparison of proton resonances (Table
) with those of commonly reported azoxytetranitrotolu-
enes also establishes their difference from those com-
pounds.
3
Contribution of New Metabolites to Mole Balance. An
additional analysis was performed to quantify total levels of
the freshly identified TNT-derived metabolites in the 24-L
1
4
aquatic system of M. aquaticum by employing C-based
response factors for HPLC areas at 230 nm. For monoaryl
1
4
TNT-metabolites, the molar C activity per mole of isolated
metabolite was assumed to be the same as that of the 14C-
TNT and for binuclear (or diaryl) TNT-metabolites, double
1
4
that of the C-TNT. After exposure to TNT for 12 days, the
extracellular aqueous milieu contained significant levels of
these metabolites, 35.9% of initial TNT moles as the
characterized monomers and 5.6% as the binuclear me-
tabolites. The breakdown of the former group comprises
In this study, a number of unique plant-based products
were isolated from the extracellular, aqueous milieu of TNT-
VOL. 33, NO. 19, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 3 5 9

exposed M. aquaticums2-amino-4,6-dinitrobenzoic acid;
,4-dinitro-6-hydroxybenzyl alcohol; 2-N-acetoxyamino-4,6-
metabolism, such as ring-hydroxylated and alkyl-hydroxy-
lated derivatives often occurred as the phenolic â-glucoside,
â-O-glucoside, glycosyl ester, or amino acid conjugate (14).
It is similarly possible that the oxidized metabolites of TNT
could occur intact within the biomass, or as conjugates that
would require hydrolysis prior to the analysis performed in
this study. Studies are presently ongoing that are examining
this aspect of nitroaromatic fate in M. aquaticum.
2
dinitrobenzaldehyde and isomer; 2,4-dinitro-6-hydroxytolu-
ene and isomer; and two binuclear metabolites. This analysis
is supported by a suit of analytical evidence: UV-vis (Figure
1
13
3
), H and C NMR spectroscopy (Tables 1-3), mass spec-
troscopy (Table 4), and where feasible, such as in the case
of 2-amino-4,6-dinitrobenzoic acid, independent chemical
synthesis. The formation of 2-amino-4,6-dinitrobenzoic acid
likely involves multistep alkyl hydroxylation, formation of
the benzyl group followed by that of the benzoyl group, as
in the metabolism of the herbicide oxadiazon in rice (20)
and peanut (21) plants. The metabolites 2-N-acetoxyamino-
The oxidized metabolites appeared in the aquatic system
of TNT-exposed M. aquaticum together with the customarily
reported reduction products: 2-amino-4,6-dinitrotoluene
and 4-amino-2,6-dinitrotoluene. After exposure to TNT for
1
2
2 days, the residual aqueous phase contained 10% TNT,
.3% 2-amino-4,6-dinitrotoluene, 5.3% 4-amino-2,6-dini-
4
,6-dinitrobenzaldehyde and 2,4-dinitro-6-hydroxybenzyl
alcohol may represent modified intermediates/ byproducts
of this pathway. As isolated, these compounds also exhibit
additional modifications, such as acetoxy addition at the aryl
amino group and aromatic hydroxylation, respectively. From
the monoamine exposure studies conducted here, it appears
that methyl oxidation most likely precedes nitro reduction,
since neither amino dinitrobenzoic acids nor acetoxyami-
nodinitrobenzaldehydes were detected when either 2-amino-
trotoluene, and an estimated total of 36% monomeric
oxidized metabolites. It demonstrates that aquatic plants
exposed to TNT, in this case M. aquaticum, are capable of
both types of metabolism, reductive and oxidative. From
our current state of understanding of herbicide metabolism
in plants, the most plausible enzyme candidates for the
oxidative metabolism of TNT are mixed function oxygenases
(
mfo). In most general terms, these are cytochrome P-450
4
,6-dinitrotoluene or 4-amino-2,6-dinitrotoluene were the
groups of enzymes that are localized primarily in the
microsomes (endoplasmic reticulum) of plant cells, requiring
molecular oxygen as a second substrate, and NADPH or
NADH as cofactors (17, 23). Our present findings may be
evidence for the activity of such enzymatic processes during
TNT metabolism in plants. To further improve our under-
standing of the activity of oxidative metabolism relative to
nitro reduction, temporal analyses of TNT-exposed plant
systems, where both reductive and oxidative metabolites are
monitored over time, will be necessary. Clearly, our inves-
tigation of oxidative metabolites in TNT-exposed plant
systems is important to the total analysis of metabolic fate
of TNT and related nitroaromatic xenobiotics in nature and
in phytoremediation processes.
starting material for plant exposure. A cautionary note in
this matter is the possible noninduction of the relevant
enzyme system, direct N-conjugation of the amine, or partial
to complete methyl hydroxylation followed by rapid con-
jugation in the monoamine-feed studies. The product profile
for exposure of M. aquaticum to 2-amino-4,6-dinitrotoluene
suggests the second possibility since the corresponding
conjugate, 2A-5.6 was detected; 2A-5.6 is a conjugate
metabolite that is formed by the addition of a six-carbon
sugar to 2-amino-4,6-dinitrotoluene and is a hypothesized
precursor of bound TNT metabolites in plants (10). Another
interesting aspect of oxidative metabolism in M. aquaticum
is the ring hydroxylation of TNT evidenced by the formation
of 2,4-dinitro-6-hydroxybenzyl alcohol and 2,4-dinitro-6-
hydroxytoluene. In both cases, aryl hydroxylation was
accompanied by the elimination of a nitro group. Aromatic
hydroxylation of the herbicide 2,4-D is known to occur
Acknowledgments
We would like to express our appreciation to Dr. Terry
Marriot, Director, Mass Spectrometry Center at Rice Uni-
versity and Ms. Anne Bellatore of University of Texas, Health
Science Center, for their assistance with mass spectrometric
analysis of samples. This material is based in part upon work
supported by the Texas Advanced Technology Program under
Grant 003604-045, by NSF Young Investigator Award BES-
(
reviewed by Hatzios (16)), but the elimination of ring-
substituted chlorine is a minor reaction; for this herbicide,
the “NIH shift” mechanism was determined to be the
dominant enzyme catalyzed pathway in which ring hydroxy-
lation is accompanied by chlorine migration and the forma-
tion of 4-OH-2,3-D and/ or 4-OH-2,5-D. The difference in
aryl hydroxylation outcomes between TNT and 2,4-D may
lie among other possibilities, in the degree of ring substitu-
tion of the parent compound. With respect to the binu-
9
257938 to J.V.S. and by DSWA Project 01-97-1-0020.
Literature Cited
clear metabolites, it is evident that their resolution in C
HPLC was pK dependent, achieved only by ion-suppres-
sion using 10 mM H PO . This leads to the possibility of
constituent aryl units bearing aromatic hydroxyl group(s)
that exhibit low pK in the presence of electronegative ring
8
-
(
(
(
1) Wang, L. K.; Hao Wang, M.; Ciccone, V. J.; Ross, R. G. Effluent
Water Treat. J. 1982, 22, 222-225.
2) Medina, V. F.; Rivera, R.; Larson, S. L.; McCutcheon, S. C. Soil
Groundwater Cleanup 1998, February/ March, 19-24.
a
3
4
a
3) Schnoor, J. L.; Licht, L. A.; McCutcheon, S. C.; Wolfe, N. L.;
substituents, in this case one or more nitro-group of the
parent compound (TNT). At this point, however, with the
analytical evidence at hand, the aryl hydroxylation of
binuclear metabolites, isolated in this study and, conse-
quently, their relationship to oxidative metabolism of TNT,
can only be hypothesized.
Carreira, L. H. Environ. Sci. Technol. 1995, 29, 9, 318-323.
4) Palazzo, A. J.; Leggett, D. C. J. Environ. Qual. 1986, 15, 49-52.
5) Harvey, S.; Fellows, R. J.; Cataldo, D. A.; Bean, R. M. J. Chromatogr.
(
(
1
990, 518, 361-374.
(
6) Hughes, J. B.; Shanks, J. V.; Vanderford, M.; Lauritzen, J.; Bhadra,
R. Environ. Sci. Technol, 1997, 31, 266-271.
In our study, the products were isolated exclusively from
the aqueous phase. This is in agreement with the concep-
tualized “green liver” metabolism of organic xenobiotic
compounds in plants; products of the initial phase (phase
I and/ or phase II) of xenobiotic detoxification are more
hydrophilic and less bioavailable than the parent molecule
(7) Vanderford, M.; Shanks, J. V.; Hughes, J. B. Biotechnol. Lett.
997, 19, 277-280.
1
(
8) Best, E. P. H.; Zappi, M. E.; Fredrickson, H. L.; Sprecher, S. L.;
Larson, S. L.; Ochman, M. In Bioremediation of Surface and
Subsurface Contamination; Bajpai, R. K., Zappi, M. E., Eds.;
Annals of New York Academy of Science: New York, 1997; Vol.
8
29, pp 179-194.
(
17, 22). Our isolation of oxidized metabolites of TNT from
(
9) Pavlostathis, S. G.; Comstock, K. G.; Jacobson, M.; Saunders:
the aqueous phase of TNT-exposed plants does not, however,
preclude their occurrence within TNT-exposed plant bio-
mass. In studies with herbicides, products of oxidative
F. M. Environ. Toxicol. Chem. 1998, 17, 2266-2273.
(10) Bhadra, R.; Wayment, D. G.; Hughes, J. B.; Shanks, J. V. Environ.
Sci. Technol. 1999, 33, 446-452.
3
3 6 0
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 19, 1999

(
11) Larson, S. L. In Bioremediation of Surface and Subsurface
Contamination; Bajpai, R. K., Zappi, M. E., Eds.; Annals New
York Academy of Science: New York, 1997; Vol. 829, pp 195-
(19) Silverstein, R. M.; Bassler, G. C.; Morrill, G. C. Spectrometric
Identification of Organic Compounds, 5 ed.; John Wiley & Sons:
New York, 1991; p 419.
(20) Ishizuka, K.; Hirata, H.; Fukunaga, K. Agric. Biol. Chem. 1975,
39, 1431-1446.
2
01.
12) Thompson, P. L.; Ramer, L.; Schnoor, J. L. Environ. Sci. Technol.
998, 32, 975-980.
13) Scheidemann, P.; Klunk, A.; Sens, C.; Werner, D. J. Plant Physiol.
998, 152, 242-247.
(
(
(
1
(
(
(
(
21) Bingham, S. W.; Shaver, R. L.; Guyton, C. L. J. Agric. Food Chem.
980, 28, 735-740.
22) Coleman, J. O. D.; Blake-Kalff, M. M. A.; Davies, T. G. E. Trends
Plant Sci. 1997, 2, 1144-1151.
23) Hatzios, K. K.; Penner, D. Metabolism of herbicides in higher
plants; CRC Press: Boca Raton, 1982; p 15-74.
24) Ewing, D. F. Org. Mag. Reson. 1979, 12, 499-523.
1
1
14) Cole, D. In Progress in Pesticide Biochemistry and Toxicology;
Hutson, D. H., Roberts, T. R., Eds.; John Wiley & Sons: New
York, 1983; Vol. 3, pp 199-254.
15) Cole, D. J. Pest. Sci. 1994, 42, 209-222.
16) Hatzios, K. K. In Environmental Chemistry of Herbicides; Grover,
R., Cessna, A. J., Eds.; CRC Press: Boca Raton, 1991; Vol. 2, pp
(
(
1
41-185.
Received for review April 19, 1999. Revised manuscript re-
ceived June 28, 1999. Accepted June 30, 1999.
(
(
17) Sandermann, J. H. Pharmacogenetics 1994, 4, 225-241.
18) Hughes, J. B.; Wang, C. Y.; Bhadra, R.; Richardson, A. Environ.
Toxicol. Chem. 1998, 17, 343-348.
ES990436I
VOL. 33, NO. 19, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 3 6 1