Treatment of trinitrotoluene by crude plant extracts

Article abstract of DOI:10.1016/j.chemosphere.2003.12.014

Crude plant extract solutions (spinach and parrotfeather) were prepared and spiked with 2,4,6-trinitrotoluene (TNT) (20 mgl^-1). 90-h TNT removal by these solutions was compared to controls. Spinach and parrotfeather extract solutions removed 99

Full text of DOI:10.1016/j.chemosphere.2003.12.014

Chemosphere 55 (2004) 725–732
www.elsevier.com/locate/chemosphere
Treatment of trinitrotoluene by crude plant extracts
a,
a
b
Victor F. Medina *, Steven L. Larson , Lovell Agwaramgbo ,
c
Waleska Perez , Lynn Escalon
d
a
Environmental Engineering Branch, EP-E, Environmental Laboratory, ERDC,
US Army Corps of Engineers (USACE), 3909 Halls Ferry Road, Vicksburg, MS 39180, USA
b
Department of Chemistry, Dillard University, New Orleans, LA 70122, USA
c
Department of Chemistry, Purdue University, West Lafayette, IN 47907, USA
d
Analytical Services, Inc., Vicksburg, MS 39180, USA
Received 10 February 2003; received in revised form 18 August 2003; accepted 29 December 2003
Abstract
Crude plant extract solutions (spinach and parrotfeather) were prepared and spiked with 2,4,6-trinitrotoluene (TNT)
(20 mg lꢀ1). 90-h TNT removal by these solutions was compared to controls. Spinach and parrotfeather extract solu-
tions removed 99% and 50% of the initial TNT, respectively; TNT was not eliminated in the controls or in extract
solutions where removal activity was deactivated by boiling. A first-order removal constant of 0.052 hꢀ1 was estimated
for spinach extract solutions treating 20 mg lꢀ1 TNT concentrations, which compared favorably to intact plant removal.
Concentration variation was described by Michaelis–Menton kinetics. Detectable TNT degradation products repre-
sented only a fraction of the total TNT transformed, and the transformation favored the formation of 4-aminodini-
trotoluene. The results indicated that crude plant extracts transform TNT, without the presence of the live plant.
Ó 2004 Published by Elsevier Ltd.
Keywords: Phytoremediation; Trinitrotoluene; Plant extracts
1. Introduction
cannot work in areas where the contaminant concen-
tration is toxic to the plant; and (3) the treatment rate is
limited by the rate of contaminant uptake by the plant.
If contaminants could react with extracted plant chem-
icals, in the absence of the plant itself, applications of
phytoremediation could be expanded.
Phytoremediation of organic compounds is promis-
ing for the treatment of a wide range of environmental
contaminants in soil and water matrices (Dushenkov
et al., 1995; Newman et al., 1997; Hughes et al., 1998;
Nzengung et al., 1999). The majority of applications
described in the literature involve the growth of plants in
the zone of contamination, and having the plant uptake
and accumulate or uptake and transform the contami-
nant. Although effective in many cases, these approaches
suffer from the several inherent drawbacks, including:
(1) the treatment is limited to areas where the plant roots
can absorb and uptake the contaminant; (2) the system
Because it has been well studied and is transformed
and degraded by plants, 2,4,6-trinitrotoluene (TNT) is a
model organic compound for phytoremediation studies.
Previous studies have focused on in vivo transformation
processes (Hughes et al., 1998; Thompson et al., 1998;
Larson et al., 1999); however, there is much evidence that
transformations occur outside the plant (in vitro) as well.
For example, studies with Myriophyllum sp. reported the
recovery of transformation products (aminodinitrotolu-
enes (ADNT), diaminonitrotoluene, and hydroxylamin-
odinitrotoluenes) in culture medium (Hughes et al., 1998;
Pavlostathis et al., 1998; Rivera et al., 1998; Medina
et al., 2000a,b, 2002). The plant either exuded these
* Corresponding author. Tel.: +1-601-634-4283.
E-mail address: victor.f.medina@erdc.usace.army.mil (V.F.
Medina).
0045-6535/$ - see front matter Ó 2004 Published by Elsevier Ltd.
doi:10.1016/j.chemosphere.2003.12.014

726
V.F. Medina et al. / Chemosphere 55 (2004)725–732
metabolites after uptake and transformation, or they
were formed from parent TNT in solution. Root-asso-
ciated bacteria may have also formed the metabolites,
but removal of periphyton from Myriophyllum cultures
did not eliminate the appearance of the metabolites, and
transformation products were not detected in the non-
planted controls. Photolytic reactions could be ruled out,
as the transformation products were not found in the
controls.
tography. The methods have been previously described
(Rivera et al., 1998; Medina et al., 2000a,b). The method
detection limit for TNT was determined to be 0.003
mg lꢀ1 and the laboratory reporting limit was 0.020
mg lꢀ1
.
Controls were prepared using the same jars as the
experimental reactors and consisted of DI water with no
plants added. They underwent identical filtering and
shaking as the experimental reactors.
The goal of this study was to determine if plants
could transform TNT in aqueous solution external to
the plant. The study focused on crude plant extracts
released from the plant during submersion and agitation
in aqueous solution.
Several experiments were conducted using this ap-
proach: screening study, boiled extract study, and kinet-
ics study. In addition, an intact plant experiment was
conducted for comparison. A concentration variation
study was conducted to develop additional understand-
ing of the extract reaction kinetics.
2.2. Plants
2. Experimental methods
Two plant species were tested: parrotfeather (Myr-
iophyllum aquaticum) and spinach (Spinicia oleracea).
Parrotfeather was chosen because it has been widely
tested for the treatment of munitions (Hughes et al.,
1998; Rivera et al., 1998; Larson et al., 1999; Medina
et al., 2000a,b, 2002). Parrotfeather was purchased at a
local nursery and was cultivated in hydroponic tanks in
tap water. Spinach was chosen because it is commonly
used in botanical studies, is widely available, and is easy
to grow. Some of it was purchased from a local super-
market and was used within a day of purchase. Spinach
was also grown in the laboratory, particularly for the
intact plant uptake study. The plants were thoroughly
washed, first with tap water and then in deionized water.
2.1. Basic experimental approach
Fig. 1 is a schematic showing the basic experimental
approach. Plants were placed in autoclaved 200-ml
screw-top jars, and deionized (DI) water was added to
create a biomass concentration of 50 g lꢀ1. The jars were
placed on a shaker table for 24 h. The plants were then
removed from the jars, and the remaining solutions were
filtered through a 0.22-lm filter to remove all particulate
matter and microorganisms. There are several studies
that indicate that 0.22-lm membrane filters are effective
at removing microorganisms from aqueous solution
(Logan et al., 1993; Madaeni, 1999; Hijnen et al., 2000;
Ohtani et al., 2000). Then, a 100-mg lꢀ1 aqueous TNT
solution was added to the filtered solutions in the jars to
form the target concentration (usually 20 mg lꢀ1). The
jars were incubated in the dark at 20 °C, sampled, and
analyzed for TNT and its breakdown products. We as-
sumed that, for a given plant, the extract concentration
would be about the same for a given extraction period.
However, we did not take measurements to confirm this.
Concentrations of TNT, ADNT, and other break-
down products were determined directly from the
aqueous solutions using high-pressure liquid chroma-
2.3. Plate counting for heterotrophic bacteria
Standard Method 9215D, which is a filtration-based
plate counting method, was used to enumerate hetero-
trophic bacteria from solution (Greenberg et al., 1992).
Methods for extracting microorganisms from the plant
surfaces were discussed in Medina et al. (2000b).
A bleach treatment method (described in Medina
et al. (2000b)) was used to remove microorganisms living
on the plant surfaces. The method has been documented
to remove 96–100% of bacteria from parrotfeather and
spinach, respectively (Medina et al., 2000b).
Bleach-
Treated Plant
Place in DI
Water
Shake
Overnight
2.4. Screening experiment
Untreated
Plant
Parrotfeather and spinach were separated into two
batches: bleach-treated (see above) and untreated. Un-
treated plants did not undergo the bleach-treatment
process. The experiment involved the following: one con-
trol, one bleach-treated spinach, one untreated spinach,
one bleach-treated parrotfeather, and one untreated
parrotfeather. The control and the untreated parrot-
feather were triplicated. The jars were incubated in the
Remove
Plant
Filter
(0.22 µm)
Add Saturated TNT
to make a 20 mg/l
Solution (or desired
concentration)
Sample & Analyze for
TNT & Transformation
Products
Fig. 1. Schematic of the experimental design.

V.F. Medina et al. / Chemosphere 55 (2004)725–732
2.8. Concentration variation comparison study
727
dark at 20 °C for 90 h, sampled, and analyzed for TNT
and its breakdown products.
To investigate the effect of initial contaminant con-
centration, the experimental procedure was repeated
using crude spinach extract solutions on TNT concen-
2.5. Boiled extracts experiment
trations of 0.05, 2.0, 5.5, 10.5, and 21.0 mg lꢀ1
.
Heat is an effective means for removing enzymatic
activity. Medina et al. (2000a) found that intact plants
did not remove TNT at a temperature of 54 °C and
higher. Budge and Parrish (1999) established boiling
as an effective means for removing lipolytic enzymatic
activity, and van Beelen and Burris (1995) used boil-
ing to inactivate nitroreductase enzymes isolated from
soil. Therefore, boiling was used to deactivate enzy-
matic activity in this study. Extract solutions were
heated on a hot plate and boiled for 5 min. A water-
cooled condenser was used to recover evaporated liquid.
The solution was then allowed to cool to room tem-
perature.
Michaelis–Menton kinetics was then used to relate
removal kinetics to concentration. The Michaelis–
Menton equation can be rearranged to yield the Eadie–
Hofstee equation:
v
S0
v0 ¼ Vm
ꢀ
0 Km
By graphing v0 versus v0=S0, the y-intercept gives Vm and
the slope of the line gives Km (Zubay, 1988). Because the
same term (v0) is in both the x and y term, it is expected
that a high correlation coefficient will result. However,
the method allows for more even spacing of data points
compared to Lineweaver–Burke plots and has been
commonly used in recent published literature (Iraburu
et al., 1994; Chalot et al., 1996; Kajikawa et al., 1997;
Javelle et al., 1999).
A portion of spinach extract solution underwent the
boiling procedure to deactivate enzymatic activity. Both
untreated and boiled extract solutions were then spiked
with TNT to form 20-mg lꢀ1 concentrations. In addition,
controls without extracts were prepared. Triplicate reac-
tors were prepared for each condition. Removal of TNT
and generation of amino-transformation products were
compared at 90 h.
3. Results
3.1. Screening study
2.6. Kinetic experiment
Concentrations of TNT were reduced during 90 h of
exposure to crude plant extracts from both spinach and
parrotfeather but remained unchanged in the control
solution (Fig. 2). Spinach-based extract solutions were
more effective, reducing over 95% of the initial TNT
concentration compared to about 50% removal for
parrotfeather extract solution. Plate counts indicated
that the bleach treatment removed 100% of heterotro-
phic bacteria from spinach and 97% of the bacteria from
the parrotfeather. In addition, plate counts indicated
that filtration using a 0.22-lm membrane removed 100%
of heterotrophic bacteria.
Small amounts of ADNT (1–3 mg lꢀ1) were detected
in both the spinach and parrotfeather extract solutions
following 90 h but not in the control. Although the
ADNT concentrations were relatively low, the concen-
trations were well above the method detection limit.
Spinach extracts were prepared and used to treat a
23-mg lꢀ1 TNT solution, and their performance were
compared to control solutions. Triplicate reactors were
prepared for each condition (extract and control).
Measurements of TNT and aminonitro-transformation
products were taken at 0, 3, 4, 5, 18, 40, 75, 97, and 147
h, enabling calculation of kinetic rates. These rates were
compared to intact plants.
A first-order rate equation was used to analyze ki-
netic rates of TNT removal, and the pseudo-first-order
rate constant, k1, was then determined graphically. In
actuality, the removal of TNT was expected to be sec-
ond-order, in relation to enzyme (or extract) concen-
tration (Pavlostathis et al., 1998; Medina et al., 2000a).
However, for this study, we assumed this to be in excess
and therefore could be considered a constant.
2.7. Removal of TNT by intact plants comparison study
3.2. Boiled extract solution study
For comparison, removal of TNT by intact spinach
and parrotfeather plants was conducted using methods
previously described (Medina et al., 2000a). A plant
density of 50 g lꢀ1 was used. Triplicate reactors were
prepared for each plant. This removal was compared to
that of triplicated extract solutions generated by identi-
cal plant densities.
Using unboiled extract solutions, TNT was removed;
however, the chemical was not removed from the con-
trols (Fig. 3). Only a small amount of TNT (1.6 mg lꢀ1
)
was removed from boiled extract solutions (boiled be-
fore introduction of TNT), compared to 24.85 mg lꢀ1
removed from the unboiled extract solution. The error
bars, depicting the variation associated between different

728
V.F. Medina et al. / Chemosphere 55 (2004)725–732
tions were more than two orders of magnitude lower
35
30
25
20
15
10
5
than what was found in the unboiled extract solution
reactors.
Analysis of variance (ANOVA) was conducted on the
triplicate reactor data collected during this experimental
set, using an error allowed, a, of 0.01. Comparing the
initial concentration with the ending concentrations of
the three test conditions (boiled, unboiled extract, and
control), each had a statistically significant difference.
The ANOVA comparison of the initial concentration,
the boiled extract solution, and the control also showed
a statistically significant difference. Comparisons be-
tween the initial concentration and the boiled extract,
between the initial concentration and the control, and
between the control and the boiled extract all had no
statistical significant differences. All comparisons with
the unboiled extract solution had statistically significant
differences.
0
Control
SU
SB
PFU
PFB
TNT (0 hr) TNT (90 hr) ADNT
Fig. 2. TNT removal after 90 h of exposure to plant extracts
released in water (SU ¼ untreated spinach extract solution,
SB ¼ bleach-treated spinach extract solution, PFU ¼ un-
treated parrotfeather extract solution, PFB ¼ bleach-treated
parrotfeather extract solution). Error bars represent standard
deviation of measurements on triplicate reactors.
replicates, were relatively small, indicating the process is
repeatable.
3.3. Kinetic study
In each case, a small amount of 2ADNT was formed
at the end of the 90-h period. The 2ADNT concentra-
tions were about equal for each reactor. In addition,
4ADNT was detected at a level of about 5 mg lꢀ1 in the
extract reactors. While 4ADNT was found in both the
boiled extract and the control reactor, the concentra-
Kinetic studies revealed that the removal of TNT by
spinach extracts was first-order with a rate constant of
0.052 hꢀ1 ðr2 ¼ 0:96Þ (Fig. 4). The extracts degraded
TNT to 2ADNT and 4ADNT, with 4ADNT being
formed at levels 10 times higher than that of 2ADNT.
Whole plant studies indicated that spinach and parrot-
30
25
20
15
10
5
TNT
4ADNT
2ADNT
Error bars
show two
std. dev. of
triplicate
reactors
0
Initial (0 hr)
Boiled (90 hr)
Unboiled (90 hr)
Control (90 hr)
Fig. 3. Comparison of TNT removal after 90 h of exposure to spinach extracts in water, with comparison to control (no exudates) and
boiled extracts. Initial is the starting TNT concentration. Error bars represent standard deviation of measurements on triplicate
reactors.

V.F. Medina et al. / Chemosphere 55 (2004)725–732
729
Table 1
Comparison of measured removal rates versus predicted values
25
20
15
10
5
TNT
4ADNT
2ADNT
using Michaelis–Menton parameters (Vm ¼ 59:75 mg lꢀ1 hꢀ1
and Km ¼ 5375 mg lꢀ1
)
Initial TNT
concentration TNT removal rate based on Michaelis–
Measured
Calculated TNT removal
(mg lꢀ1
)
rate
(mg lꢀ1 hꢀ1
Menton kinetics
(mg lꢀ1 hꢀ1
)
)
0.05
2.0
0.0006
0.022
0.061
0.117
0.232
0.0006
0.022
0.061
0.117
0.233
5.5
10.5
21.0
0
0
20
40
60
80
100 120 140 160
Time (hr)
Fig. 4. Kinetic study results for spinach exudate removal of
TNT. Error bars (representing standard deviation of triplicate
reactors) are given for TNT but are quite small.
enzymes involved in the reaction. Michaelis–Menton
kinetics (discussed in ‘‘Experimental Section’’) is a com-
monly used theory for enzyme saturation kinetics.
Fig. 6 is an Eadie–Hofstee plot of the data giving the
Michaelis–Menton constants for the reaction (Vm
¼
feather had first-order rate constants of 0.141 and 0.036
hꢀ1 respectively. Therefore, the TNT removal rate con-
stant by crude spinach extracts was about 1/2 that of the
intact spinach plant and was greater than that of intact
parrotfeather. This indicates that removal rates by the
crude extracts are reasonably fast and could be com-
petitive with that of intact plants.
59:75 mg lꢀ1 hꢀ1 and Km ¼ 5375 mg lꢀ1). Comparing the
measured reaction rates (or reaction velocities) with
those predicted using the Michaelis–Menton constants
indicate that the values are extremely close (Table 1).
Table 2 gives the concentrations of 4ADNT and
2ADNT found in the concentration variation study. The
0.30
3.4. Concentration variation study and relationship to
Michaelis–Menton kinetics
0.25
y = -5375 x + 59.75
R2 = 0.97
0.20
A concentration variation experiment was conducted
in which crude spinach extract solutions were used to
treat initial concentrations of TNT varying from 0.5 to
21.5 mg lꢀ1. The measured pseudo-first-order rate con-
stants decreased with increasing TNT concentrations
(Fig. 5), a similar pattern to what was found with whole
plant studies (Medina et al., 2000a). Although the pseu-
do-first-order rate constant decreased with increasing
concentration, the reaction velocity increased (Table 1).
These patterns might be explained by saturation of the
0.15
0.10
0.05
0.00
0.01106
0.01108
v/S (hr-1)
0.01110
0.01112
Fig. 6. Eadie–Hofstee plot for concentration variation study.
Vm is the y-intercept (59.75 mg lꢀ1 hꢀ1), and Km is the absolute
value of the slope (5375 mg lꢀ1).
0.14
0.12
Table 2
0.10
0.08
0.06
0.04
0.02
0.00
Final concentrations of 4-amino-2,6-dinitrotoluene (4ADNT)
and 2-amino-4,6-dinitrotoluene in comparison to initial TNT
concentrations after 90 h of reaction with spinach extract
solutions
Initial TNT
concentration
Final 4ADNT
concentration
Final 2ADNT
concentration
(mg lꢀ1
)
(mg lꢀ1
)
(mg lꢀ1
)
0
5
10
15
20
25
0.05
2.0
0.023
0.74
1.49
2.21
3.15
0.000
0.00
0.05
0.11
0.37
Initial TNT Conc. (mg l-1)
5.5
Fig. 5. Change of first-order rate constants with respect to
initial TNT concentration for a series of spinach extract
experiments.
10.5
21.0

730
V.F. Medina et al. / Chemosphere 55 (2004)725–732
concentrations of both transformation products in-
creased with increasing initial TNT concentration. The
concentration of 4ADNT was considerably higher in
each case than the concentration of 2ADNT.
from spinach (Baysdorfer and Robinson, 1985; Nakag-
awa et al., 1985; Shah and Spain, 1996; Shah and
Campbell, 1997), and these have been demonstrated to
react with tetryl (Shah and Spain, 1996) and nitroben-
zene (Shah and Campbell, 1997). Presumably, spinach
extracts removed TNT more completely than parrot-
feather extracts because the spinach extracts had more
nitroreductase activity than the parrotfeather extracts.
However, plant protein or enzymes were not directly
measured in this study.
4. Discussion/conclusions
The method developed for the experiment was de-
signed to isolate and eliminate competing variables for
TNT removal (e.g., photo-reactions, adsorption, micro-
bial reactions), leaving released extracts as the most
likely agent responsible for TNT transformation in
solutions. Table 3 summarizes the isolation of variables.
The literature indicates that filtration by a 0.22-lm filter
should remove most bacteria from the extract solutions
(Logan et al., 1993; Madaeni, 1999; Hijnen et al., 2000;
Ohtani et al., 2000), and our own measurements indi-
cated that filtration removed 100% of heterotrophic
bacteria. Therefore, we conclude that microbial uptake
and metabolism played a negligible role in the trans-
formation. It might be possible that microbial enzymes
passed through our filtration step. However, the dissipa-
tion of TNT was approximately equal for bleach-treated
(which have their surface microorganisms removed)
plants when compared to untreated plants, further sup-
porting that microbial degradation was minimal in these
experiments.
Some plant chemical production and release is in-
duced by the presence of the contaminant or by envi-
ronmental conditions. For example, induced production
of metal-interacting phytochelating agents has been
observed in several studies, usually in nutrient limited
€
conditions (Romheld, 1991; Salt et al., 1997), and nitrate
availability induced the synthesis and regulation of
NAD(P)H nitrate reductase from the bryophyte Sphag-
num magellanicum (Deising and Rudolph, 1987). How-
ever, this does not appear to be the case in our studies.
The living plant was removed before the TNT was added
to the solution, indicating that the plant must have re-
leased the factor responsible for the transformation in
the absence of the contaminant. Therefore, the presence
of the released factor does not appear to induced by
TNT. This is consistent with intact plant studies, where
previous exposure to munitions does not seem to in-
crease removal rates (Comstock, 1996; Pavlostathis
et al., 1998; Medina et al., 2000b). In fact, some intact
plant studies suggest the rate is even decreased after
previous TNT exposure (Medina et al., 2000b).
The released factor responsible for transformation
may have been a nitroreductase enzyme (Schnoor et al.,
1995; van Beelen and Burris, 1995) or a combination of
nitrate- and nitrite-reductases that plants use to process
nutrients (Larcher, 1995; Taiz and Zeiger, 1998). Ni-
trate-, nitrite-, and nitroreductases have been isolated
The fact that the extracts removed TNT without the
physical presence of the plant suggests that the activity is
relatively stable. This is consistent with the literature in
Table 3
Summary of isolation of competing variables for TNT removal
Variable
Means of isolation
Volatilization
1. TNT is relatively non-volatile (vapor pressure at 80 °C ¼ 0.05 mm Hg)
2. Experiments conducted in screw-top jars
3. Accounted for by control
Adsorption on glass surfaces
Uptake by plant
Accounted for by control
Plant removed before TNT was added, eliminating this possibility
Reactions with light
1. Jars were incubated in the dark
2. Accounted for by control
Adsorption with solid plant particles
Particles were removed by filtration of liquid before TNT addition
Microbial degradation associated with the
solutions and glassware
1. All glassware and solutions were autoclaved before use
2. Accounted for by control
Microbial degradation associated with the
plants themselves
1. Use of bleach treatment to kill microbes before the plant was placed in the
hydroponic system (verified by plate counting)
2. Microbes removed by filtration (0.22 lm) before TNT spiking (verified by
plate counting)

V.F. Medina et al. / Chemosphere 55 (2004)725–732
731
which stable enzymatic activity has been described
(Skujins, 1967). The sources of these enzymes can be
difficult to determine, and generally, they have been
assumed to have microbial origins. However, plant ori-
gins are possible (Skujins, 1967). In fact, stable enzymes
capable of transforming TNT were isolated from
aquatic sediments (van Beelen and Burris, 1995). Al-
though the source of these enzymes was not determined,
the authors noted that the sediments contained plant
roots. Furthermore, the enzymes seemed ubiquitous;
active enzymes were found in 10 of 11 freshwater sites
sampled. In a similar study, stable peroxidase enzymes
were extracted and purified from soils (Bollag et al.,
1987). Once again, the source was not isolated, but the
enzyme had similar properties to those isolated from
horseradish.
required to test the efficacy of these ideas, treatment
systems based on enhancing the treatment potential of
plant extracts have the potential to greatly enhance the
application of phytoremediation.
Acknowledgements
The authors would like to acknowledge funding
provided by the Environmental Chemistry Branch of the
US Army Corp of Engineers––Engineer Research and
Development Center, Vicksburg, MS.
References
Baysdorfer, C., Robinson, J.M., 1985. Metabolic interactions
between spinach leaf nitrite reductase and ferrodoxin-
NADP reductase. Plant Phys. 77, 318–320.
The kinetic rates of removal by the crude extract
solutions were relatively fast, comparing well to the re-
moval rates of whole plant studies. A significant portion
of this activity may be from released compounds asso-
ciated with the plant’s life, or even released during the
death of plant tissue. If so, then a significant amount of
treatment of TNT phytosystems may actually occur
outside the plant.
Bollag, J.-M., Chen, C.M., Sarkar, J.M., Loll, M.J., 1987.
Extraction and purification of a peroxidase from a soil. Soil
Biol. Chem. 19, 61–67.
Budge, S.M., Parrish, C.C., 1999. Lipid class and fatty acid
composition of Pseudo-nitzshia multiseries and Pseudo-
nitzschia pungens and effects of lipolytic enzyme deactiva-
tion. Phytochemistry 52, 561–566.
The amount of transformation products formed was
minimal, and they were largely degraded over time. It is
not clear if some buildup of breakdown products would
occur, as found in some continuous flow-through reac-
tor studies (Rivera et al., 1998). These studies also
indicate that the extracts preferentially form 4ADNT as
compared to 2ADNT. This could be used as a tool to
understand if these types of reactions are occurring, at
least compared to adsorption reactions.
Chalot, M., Brun, A., Botton, B., Soderstrom, B., 1996.
Kinetics, energetics, and specificity of a general amino acid
transporter from the ectomycorrohizal fungus Paxillus
involutus. Microbiology 142, 1749–1756.
Comstock, K.K., 1996. Transformation of 2,4,6-trinitrotoluene
(TNT) by the aquatic plant Myriophyllum spicatum. M.S.
thesis, Georgia Tech., Atlanta, GA.
Deising, H., Rudolph, H., 1987. Nitrate-induced de novo
synthesis and regulation of NAD(P)H nitrate reductase
from Sphagnum. Physiol. Plant. 71, 477–482.
Removal also followed Michaelis–Menton kinetics.
The maximum velocity (Vm) calculated for the crude
spinach extracts was actually about a factor of ten
higher than that calculated for intact parrotfeather
studies (Medina et al., 2000a). Also, the half saturation
constant (Km) was also much higher for the spinach
extract solutions compared to intact parrotfeather (94.2
mg lꢀ1, from Medina et al. (2000a)). An important
consequence of the high Km value is that it is much
higher than the actual solubility of TNT in water (125
mg lꢀ1). This indicates that the reaction would be
essentially pseudo-first-order for all possible concentra-
tions. As such, this result is analogous to that found in
intact plant studies (Medina et al., 2000a).
Dushenkov, V., Nanda Kumar, P.B.A., Motto, H., Raskin, I.,
1995. Rhizofiltration: the use of plants to remove heavy
metals from aqueous streams. Environ. Sci. Technol. 29,
1239–1245.
Greenberg, A.E., Clesceri, L.E., Eaton, A.D., 1992. Stan-
dard Methods for the Examination of Water and Waste-
water. American Public Health Association, Washington,
DC.
Hijnen, W.A.M., van Veenendaal, D.A., van der Speld,
W.M.H., Visser, A., van der Kooij, D., 2000. Enumeration
of faecal indicator bacteria in large water volumes using on
site membrane filtration to assess water treatment efficiency.
Water Res. 34, 1659–1665.
Hughes, J.S., Shanks, J., Vanderford, M., Lauritzen, J.,
Bhadra, R., 1998. Transformation of TNT by aquatic
plants and plant tissue cultures. Environ. Sci. Technol. 32,
266–271.
Stable enzymatic activity may allow for the devel-
opment of extract-based remediation technologies. Such
treatment systems would have several advantages over
the traditional approach of growing the plants in the
zone of contamination. Plants would not uptake the
contaminant, and the potential for exposure of herbi-
vores to contaminants is eliminated. Further, crude
plant extracts can be applied to highly contaminated
soils where the plants cannot live. Although research is
Iraburu, M.J., Lopez-Zabalza, M.J., Saenz, M., Santiago, E.,
1994. Molecular aspects of activation mechanisms of CF1.
Phytochemistry 36, 559–563.
Javelle, A., Chalot, M., Soderstrom, B., Botton, B., 1999.
Ammonium and methylamine transport by the ectomycor-
rhizal fungus Paxillus involtus and ectomycorhizas. FEMS
Microbiol. Ecol. 30, 355–366.

732
V.F. Medina et al. / Chemosphere 55 (2004)725–732
Kajikawa, H., Amari, M., Masaki, S., 1997. Glucose transport
by mixed ruminal bacteria from a cow. Appl. Environ.
Microbiol. 63, 1847–1851.
Pavlostathis, S.G., Comstock, K.L., Jacobson, M.E., Saunders,
F.M., 1998. Transformation of 2,4,6-trinitrotoluene (TNT)
by the aquatic plant Myriophyllum spicatum. Environ.
Toxicol. Chem. 17, 2266–2273.
Larcher, W., 1995. Physiological Plant Ecology. Springer-
Verlag, New York.
Rivera, R., Medina, V.F., Larson, S.L., McCutcheon, S.C.,
1998. Phytoremediation of TNT contaminated groundwa-
ter. J. Soil Contam. 7, 511–529.
Larson, S.L., Jones, R.P., Escalon, L., Parker, D., 1999.
Classification of explosives transformation products in plant
tissue. Environ. Toxicol. Chem. 18, 1270–1276.
€
Romheld, V., 1991. The role of phytosiderophores in acquisi-
Logan, B.E., Hilbert, T.A., Arnold, R.G., 1993. Removal of
bacteria in laboratory filters: models and experiments.
Water Res. 27, 955–962.
tion of iron and other micronutrients in Graminaceous
species: An ecological approach. Plant Soil 130, 127–134.
Salt, D.E., Pickering, I.J., Prince, R.C., Gleba, D., Dushenkov,
S., Smith, R.D., Raskin, I., 1997. Metal accumulated by
aquacultured seedlings of Indian Mustard. Environ. Sci.
Technol. 31, 1636–1644.
Madaeni, S.S., 1999. The application of membrane technology
for water disinfection. Water Res. 33, 301–308.
Medina, V.F., Larson, S.L., Bergstedt, A.E., McCutcheon,
S.C., 2000a. Phyto-removal of trinitrotoluene from water
using batch kinetics studies. Water Res. 34, 2713–2722.
Medina, V.F., Jeffers, P., Larson, S.L., Perez, W., 2000b.
Sterilization of plants for phytoremediation studies by
bleach treatment. Int. J. Phytorem. 2, 287–295.
Schnoor, J.L., Licht, L.A., McCutcheon, S.C., Wolfe, N.L.,
Carreira, L.H., 1995. Phytoremediation of organic and
nutrient contaminants. Environ. Sci. Technol. 29, 318A–
323A.
Shah, M.M., Campbell, J.A., 1997. Transformation of nitro-
benzene by ferredoxin NADP oxidoreductase from spinach
leaves. Biochem. Biophys. Res. Commun. 241, 794–796.
Shah, M.M., Spain, J.C., 1996. Elimination of nitrite from the
explosive 2,4,6-trinitrophenylmethylnitramina (tetryl) cata-
lyzed by ferredoxin NADP oxidoreductase from spinach.
Biochem. Biophys. Res. Commun. 220, 563–568.
Skujins, J.J., 1967. Enzymes in soil. In: McLaren, A.D.,
Peterson, G.H. (Eds.), Soil Biochemistry. Marcel Dekker,
Inc., New York, pp. 371–414.
Medina, V.F., Larson, S.L., McCutcheon, S.C., 2002. Evalua-
tion of continuous flow-through phytoreactors for the treat-
ment of TNT-contaminated water. Environ. Prog. 21, 29–36.
Nakagawa, H., Yonemura, Y., Yamamoto, H., Sato, T.,
Ogura, N., Sato, R., 1985. Spinach nitrate reductase:
purification, molecular weight, and subunit composition.
Plant Phys. 77, 124–128.
Newman, L.A., Strand, S.E., Choe, N., Duffy, J., Ekuan, G.,
Ruszaj, M., Shurtleff, B.B., Wilmoth, J., Heilman, P.,
Gordon, M., 1997. Uptake and biotransformation of
trichloroethylene by hybrid poplars. Environ. Sci. Technol.
31, 1062–1067.
Taiz, L., Zeiger, E., 1998. Plant Physiology. Sinauer Associates,
Inc., Publishers, Sunderland, MA.
Thompson, P.L., Ramer, L.A., Schnoor, J.L., 1998. Uptake
and transformation of TNT by hybrid poplar. Environ. Sci.
Technol. 32, 279–284.
Nzengung, V.A., Wang, C., Harvey, G., 1999. Plant-mediated
transformation of perchlorate into chloride. Environ. Sci.
Technol. 33, 1470–1478.
van Beelen, P., Burris, D.R., 1995. Reduction of the explosive
2,4,6-trinitrotoluene by enzymes from aquatic sediments.
Environ. Toxicol. Chem. 14, 2115–2123.
Ohtani, T., Kaneko, A., Fukuda, N., Hagiwara, S., Sadanori,
S., 2000. Development of a membrane disinfection system
for closed hydroponics in a greenhouse. J. Ag. Eng. Res. 77,
227–232.
Zubay, G., 1988. Biochemistry. McMillan Publishing Com-
pany, New York.