Reductive transformation of bound trinitrophenyl residues and free TNT during a bioremediation process analyzed by immunoassay

Article abstract of DOI:10.1021/es9901765

To follow the fate of bound metabolites of TNT in soil, a synthetic trinitrophenyl residue covalently linked to humic acids was used as model compound. A selective monoclonal antibody was able to detect chemical changes of the nitro groups of the bound re

Full text of DOI:10.1021/es9901765

Environ. Sci. Technol. 1999, 33, 3421-3426
genic potential of TNT and particularly its reduced metabo-
lites (3) require remediation of these sites.
Reductive Transformation of Bound
Trinitrophenyl Residues and Free
TNT during a Bioremediation
As an alternative to thermal processes of decontamination
of soil, biological remediation techniques were considered.
Since all attempts failed to establish a complete mineraliza-
tion of TNT in soil, most biological treatment processes were
directed toward an irreversible binding of reduced metabo-
lites of TNT to soil organic matter. The difficulties to achieve
mineralization of TNT can be explained by the high xenobiotic
character of nitroaromatic and particularly polynitroaromatic
compounds and the resistance of the highly oxidized trinitro-
substituted aromatic ring against oxidative attack (4). In
contrast, initial cometabolic reduction of the nitro groups of
TNT has often been observed in cultures of bacteria (5-14)
and fungi (15-18). The successive reduction of the nitro
groups leads to an activation of the molecule toward
electrophiles. While under aerobic conditions partially
reduced metabolites such as ADNT and DANT were formed,
complete reduction of all nitro groups can only be ac-
complished under strict anaerobic conditions (19). Triami-
notoluene (TAT) or highly reactive intermediates such as
hydroxylaminodinitrotoluenes strongly interact with soil
components (19). If these partially reduced metabolites bind
to soil, the resulting structures carry intact nitro groups. To
minimize the chance that nitro- or aminoaromatic com-
pounds were released from the soil, all three nitro groups
should be reduced which is the precondition of multivalent
binding. A reduction of nitro groups of covalently bound
metabolites, however, cannot be affected by microbial cells
but may be possible through soluble reduction equivalents
that act as redox mediators (20). A general evidence that
these nitro groups can be further transformed is still missing.
Process Analyzed by Immunoassay
†
C H R I S T O F A C H T N I C H ,
P E T E R P F O R T N E R ,
M I C H A E L G . W E L L E R , ‡
‡
‡
R E I N H A R D N I E S S N E R ,
H I L T R U D L E N K E , * , A N D
†
†
, §
H A N S - J O A C H I M K N A C K M U S S
Fraunhofer-Institut f u¨ r Grenzfl a¨ chen- und
Bioverfahrenstechnik, Nobelstrasse 12,
D-70569 Stuttgart, Germany, Institut f u¨ r Wasserchemie,
Technische Universit a¨ t M u¨ nchen, Marchioninistrasse 17,
D-81377 M u¨ nchen, Germany, and Institut f u¨ r Mikrobiologie,
Universit a¨ t Stuttgart, Allmandring 31,
D-70569 Stuttgart, Germany
To follow the fate of bound metabolites of TNT in soil, a
synthetic trinitrophenyl residue covalently linked to humic
acids was used as model compound. A selective
monoclonal antibody was able to detect chemical changes
of the nitro groups of the bound residues. The general
possibility of reductive transformations of nitro groups of
bound molecules and the reductions rates should be
determined. In comparison to the reduction of free TNT
and its metabolites, the reductive transformation of the bound
trinitrophenyl residue was delayed, and the transformation
rate was considerably slower. Trinitrophenyl residues
also could be detected by the immunoassay in humic acids
extracted from TNT contaminated soil. The reductive
transformation of these trinitrophenyl residues started after
the reduction of free TNT. At the end of the treatment,
small amounts of these residues were still detectable
indicating that some of these structures were not completely
reduced during the process. From present results one
can conclude that the further reduction of nitro groups of
bound metabolites requires a prolonged anaerobic
15
Besides using NMR spectroscopic methods of [ N]TNT (21),
immunological techniques may also be used to recognize
such transformations.
In this paper we investigated whether nitro groups of
covalently bound and partially reduced metabolites are
further transformed by a previously described anaerobic/
aerobic treatment process (22). For this purpose, a synthetic
conjugate of a 2,4,6-trinitrotoluene derivative with humic
acids and real contaminated soil was treated. To detect
reductive transformations of the nitro groups of covalently
bound molecules during the bioremediation process, an
optimized immunoassay which had been developed by
Pfortner et al. (23-25) was used. The immunoassay was
designed to detect only covalently bound trinitrophenyl
residues. This system should simulate the transformation of
TNT covalently bound to humic acids. Therefore, this new
approach should allow to elucidate whether nitro groups of
bound TNT in soil organic matter can be reduced.
treatment. Not only the monitoring of free nitroaromatic
compounds is recommended during the bioremediation
process but also the measurement of bound residues to
determine the optimal conditions and duration of the treatment.
Experimental Section
Chem icals. Highly pure 2,4,6-trinitrotoluene (TNT) was
supplied by MBB Deutsche Aerospace (Schrobenhausen,
Germany). 2-Amino-4,6-dinitrotoluene (2-ADNT), 4-amino-
Introduction
2
,4,6-Trinitrotoluene (TNT) was the major explosive produced
during World War II: In Germany the production amounted
to approximately 800 000 tons of TNT (1). Due to the
biological persistence TNT and related components are still
present at high concentrations at former production sites
and can be found in the water supplies of the neighboring
communities (2). The toxicity, the mutagenic, and carcino-
2
(
2
(
,6-dinitrotoluene (4-ADNT), 2,4-diamino-6-nitrotoluene
2,4-DANT), 2,6-diamino-4-nitrotoluene (2,6-DANT), and
,4,6-triaminotoluene (TAT) were obtained from Promochem
Wesel, Germany). A mixture of 2-hydroxylamino-4,6-dini-
trotoluene and 4-hydroxylam ino-2,6-dinitrotoluene (2-
HADNT/ 4-HADNT) was generated by enzymatic reduction
of TNT with xanthine oxidase (Merck, Darmstadt, Germany)
and NADH as described by Michels and Gottschalk (18).
N-(2,4,6-trinitrophenyl)-4-aminobutyric acid (TNP-C4) was
obtained from Research Organics (Cleveland, U.S.A.); per-
oxidase-labeled goat anti-mouse immunoglobulin class G
*
Corresponding author phone: +49-711-970-4216; fax: +49-711-
9
70-4200; e-mail: len@igb.fhg.de.
†
Fraunhofer-Institut f u¨ r Grenzfl a¨ chen- und Bioverfahrenstechnik.
Technische Universit a¨ t M u¨ nchen.
Universit a¨ t Stuttgart.
‡
§
1
0.1021/es9901765 CCC: $18.00

1999 Am erican Chem ical Society
VOL. 33, NO. 19, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 4 2 1
Published on Web 08/28/1999

Im m unoassay Procedure. Polystyrene microtiter plates
were coated with 200 µL of a 0.05% solution of methylated
bovine serum albumin in phosphate buffered saline (pH 7.6)
overnight. The coated plates were washed three times, and
2
00 µL of a 10 mg/ L humic acid sodium salt solution of
samples or blank in 12 replicates was incubated for 1 h.
Another washing step was performed, and 50 µL of water
was added in six wells of each sample, while the remaining
six wells of each sample were filled with 50 µL of a TNT
solution (1 mg/ L in water). Two hundred microliters of the
monoclonal TNT antibody (0.74 mg/ L) in phosphate buffered
saline, containing 1% w/ v poly(acrylic acid) and 0.2% w/ v
gelatin, was added to all wells and incubated for 15 min.
After a further washing step, 200 µL of a 1:10 000 dilution of
peroxidase-labeled goat anti-mouse IgG was added for 30
min. A final washing step was performed, and 200 µL of freshly
prepared substrate solution was added. The enzymatic
reaction was stopped after 25-30 min by adding 100 µL of
FIGURE 1. Structure of the synthetic 2,4,6-trinitrophenyl-4-ami-
nobutyric acid/humic acids (TNP-C4-HA) conjugate.
(
(
IgG) was from Sigma (Steinheim, Germany); gelatin, poly-
acrylic acid), and methylated bovine serum albumin were
obtained from Fluka (Neu-Ulm, Germany). The monoclonal
TNT antibody was kindly provided by Strategic Diagnostics
Inc. (Newark, DE). Buffers and the substrate solution were
prepared as previously described (24). All other chemicals
were obtained from commercial sources.
Synthesis of the TNP-C4-HA Conjugate. The synthetic
N-(2,4,6-trinitrophenyl)-4-aminobutyric acid/ humic acids
conjugate (designated as TNP-C4-HA conjugate, Figure 1)
was synthesized according to an earlier published procedure
5
2 4
% v/ v H SO , and the absorbance at 450 nm was measured
by a TR 400 MTP reader from SLT (Gr o¨ dingen, Austria). The
difference of the signals without and with competitive
inhibition of the antibody with TNT resulted from highly
specific antibody/ antigen interactions and was used as
parameter for the amount of covalently bound trinitrophenyl
residue in the sample.
Calculation of the TNT Equivalents. A TNP-C4-HA
standard conjugate, which was earlier characterized (24) in
detail (coupling density: 14 µmol TNP-C4/ g humic acids),
was used for calibration of the immunoassay for the detection
of trinitrophenyl residues which were bound to humic acids.
Solutions of 10 mg/ L TNP-C4 standard conjugate and HA,
respectively, were prepared. These solutions were mixed in
different ratios so that samples with different amounts of
bound TNP-C4 but constant HA content resulted. After
measurement in the described immunoassay, the determined
signal differences showed a linear correlation with the TNP-
C4 content of the samples. This calibration graph was used
to calculate the molar content of bound TNT equivalents in
the unknown samples.
Setup for Reduction Experim ents. Com etabolic Reduc-
tion of TNT and TNP-C4-HA Conjugate in Solution. The
experiments in the aqueous system were carried out in bottles
(500 mL), which were closed with rubber stoppers. After
autoclaving 180 mL of 50 mM sodium phosphate (pH ) 7.3),
crystals of TNT were added and dissolved by boiling for 3-5
h. After cooling, the gas phase was exchanged 4 times with
argon. Before inoculation, a 2 mL mineral salts solution and
a 2.2 mL glucose solution (2 M) were added anaerobically
with a syringe (19). The experiments were started by adding
2.0 mL of a mixed culture of TNT reducing bacteria. All
suspensions were incubated on a rotary shaker in horizontal
position at 30 °C and 100 rpm. Samples were centrifuged for
10 min at 14 000 rpm (approximately 13000g) before the
supernatants were analyzed by HPLC to measure the
concentration of TNT and its reduction products. Three
different experiments were performed: reduction of the
TNP-C4-HA conjugate with a small amount of TNT (27
µM) added,sas a controlsreduction of TNT in the presence
of the same humic acids without TNP-C4, and reduction of
TNT without humic acids and conjugate. Detailed incubation
conditions are given in Table 1.
(
23). N-(2,4,6-Trinitrophenyl)-4-aminobutyric acid (TNP-
C4) was converted to the corresponding TNP-C4-N-
hydroxysuccinimide-ester with dicyclohexylcarbodiimide
and coupled to primary amino functions of humic acids
previously isolated from a standard soil (see below) through
extraction with diluted NaOH. Purification of the coupling
product was achieved through gel chromatographic separa-
tion eliminating components with low molecular weight. The
density of coupling groups in the humic acids was the same
as reported in ref 23.
Soils and Isolation of Hum ic Acids. The contaminated
soil used in the slurry experiment was obtained from a former
TNT production site at Stadtallendorf near Marburg (Ger-
many). The soil had the following characteristics: total
organic carbon, 2.5%; N(total), 0.1%; P(total), 0.09; density,
1
.23 g/ mL; clay-fraction (diameter < 20 µm), 20.8%. These
were determined according to Lenke et al. (22). The soil was
air-dried and sieved to a mesh of 2 mm. For detection of
bound trinitrophenyl residues humic acids were extracted
from the soil at different times during the anaerobic/ aerobic
treatment. Prior to humic acids extraction, the extractable
nitroaromatic compounds were removed from the soil
samples by exhaustive Soxhlet-extraction with methanol.
Subsequently, humic acids were isolated with 0.1 M tetra-
sodium pyrophosphate as described previously (23).
A standard soil (type 2.2) purchased from the Land-
wirtschaftliche Untersuchungs- und Forschungsanstalt Spey-
er (Speyer, Germany) was used for the isolation of humic
acids for synthesis of the TNP-C4-HA conjugate and soil-
free reduction experiments. Samples from the experiment
with the synthetic TNP-C4-HA conjugate were diluted to
a humic acid concentration of 10 mg/ L, and the im-
munological response was measured without any further
treatment.
Humic acids for sorption experiments with HADNT were
extracted from TNT contaminated soil (Hessisch-Lichtenau,
Germany, 22). The extraction procedure was described
elsewhere (26).
Analytical Methods. HPLC Analysis of Nitroarom atic
Com pounds. The concentration of TNT, 2-ADNT, 4-ADNT,
2
,4-DANT, and 2,6-DANT were determined by reverse-phase
Reduction of TNT in Contam inated Soil. For the soil
experiments, a slurry of 850 g of contaminated soil and 850
mL of sterile sodium phosphate buffer (19) was filled in a 1.3
l bioreactor. The soil slurry was mixed with a high-grade
steel stirrer (150 rpm). The temperature of the bioreactor
was kept at 30 °C, and the pH value was maintained at 7.0
by adding 10 M NaOH solution. The bioreactor was closed,
and the gas exchange with the environment was minimized
to maintain the established anaerobic conditions. The
high performance liquid chromatography (HPLC) as de-
scribed recently (22). Concentrations of TAT were determined
by reversed-phase HPLC according to Daun et al. (19).
Isomers of hydroxylaminodinitrotoluene were measured by
reversed-phase HPLC with Lichrocart 125-4 column, filled
with 5 µm particles of Lichrospher 100 RP-8 (Merck,
Germany). A mobile phase of 40/ 60 (v/ v) acetonitrile/
phosphate buffer (15 mM, pH 7.3) was used.
3
4 2 2
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 19, 1999

TABLE 1: Conditions of Cometabolic Reduction of TNT in an
Aqueous System in Presence/Absence of Humic Acids
experiment
TNT with
TNT without TNT with
humic acid
unit
m L
humic acids humic acids conjugatea
vol (50 m M PP,b
pH ) 7.3)
concnd TNT
concnd hum ic
acids
193
193
189
m M
% (w/v)
0.32
0.35
0.98
0.027
concnd conjugatea % (w/v)
source of hum ic
acids
0.66
soil
soil
type 2.2
type 2.2a
concnd glucose
duration of the
expt
m M
h
22
200
22
200
22
850
a
TNP-C4-HA conjugate. b Sodium phosphate buffer.
experiment was started by the addition of glucose (initial
concentration 10 mM) and a mixed culture (19). Glucose
was added at an average rate of 1.0 mmol glucose/ d. Samples
of 3-4 mL soil-slurry were taken from the top of the bioreactor
2
by flushing with N . The samples were centrifuged for 10
min at 4000 rpm (approximately 2500g) to separate the
supernatant from the soil. The remaining soil pellet of each
sample was extracted two times for 1 h with methanol. Both
methanol extracts were collected and vaporized at 30 °C.
The concentration of TNT and its metabolites were measured
in the supernatant and in the methanol extract by HPLC.
The soil pellet was dried at room temperature to refer the
amount of extractable TNT and reduction products to the
dry mass.
FIGURE 2. Cometabolic reduction of TNT in the presence of the
TNP-C4-HA conjugate. (A) TNT and its metabolites analyzed by
HPLC (starting concentration 0.027 mM) and (B) immunoassay of
extracted TNP-C4-HA conjugate. The signal difference correlates
with the amount of bound trinitrophenyl residues measured by
immunoassay. The signal difference of the immunoassay of humic
acids without TNP-C4 conjugate is also given (blank).
Sorption and Desorption of Hydroxylam inodinitrotolu-
ene to Hum ic Acids. The experiments were carried out in
closed bottles (125 mL) filled with 20 mL of 50 mM sodium
phosphate buffer (pH ) 7.3) and 0.75% (w/ v) humic acids.
Humic acids were extracted from TNT contaminated soil
laminodinitrotoluene and TAT strongly interact with the
humic acid part of the TNP-C4-HA conjugate. Similar
observations were also made during reduction experiments
with the same humic acids without TNP-C4 (data not
shown). Obviously, the modification of humic acids with
TNP-C4 did not change their ability to interact with certain
reduced metabolites of TNT.
(
Hessisch-Lichtenau). To obtain anaerobic conditions the
gas phase was replaced by argon. A solution of the isomers
of HADNT in methanol (0.5 mL; final concentration: 36 µM)
was added anaerobically. Incubation was carried out on a
rotary shaker in horizontal position at 30 °C and 100 rpm.
2
While avoiding an excess of O , aliquots were taken and
immediately analyzed by HPLC. For desorption experiments
humic acids were precipitated with HCl (10 M) and centri-
fuged. The precipitate was extracted with 5 mL of each
different solvent (methanol; NaOH, 0.5 M; sodium phosphate
buffer, 50 mM, pH ) 7.3; acetonitrile) 1 h at 30 °C on a rotary
shaker. After centrifugation the extracts were analyzed by
HPLC for HADNT and other reduced TNT metabolites. The
desorption experiments were carried out in the presence
The binding of partially reduced metabolites such as
HADNT to humic acids leads to molecules which still carry
two unreduced nitro groups. Further and finally complete
reduction to amino groups is the precondition of additional
linking to soil organic matter. For safety and maximum
stability of the immobilized pollutants covalent bindings
should if possible involve all three N-functions of the original
nitro groups of TNT. These cross linkages may increase the
stability of the bound residues so that no low molecular weight
nitro- or aminoaromatic compounds would be released from
the soil even during long-term incubation or natural turnover
of organic matter. As a precondition of this multiple cross-
linking between reduced TNT metabolites and the humic
substances, it is necessary that the reduction of nitro groups
continues even if nitroaromatic metabolites are already
covalently bound to humic material (Figure 5, reaction 3).
The trinitrophenyl moiety of the synthetic TNP-C4-HA
conjugate (Figure 1) can serve as a model of a bound residue
and be detected by an immunoassay with a highly specific
monoclonal antibody. Chemical changes of the bound
trinitrophenyl moiety (i.e. further reduction of nitro groups
and covalent binding to humic acids) will lead to structures,
which are not recognized by the monoclonal antibody and
the signal of the immunoassay should thus decrease (25, 27).
To correlate the decrease of the immunoassay signal with
2
and absence of O .
Results and Discussion
Reduction of Free TNT and TNP-C4-HA Conjugate. The
cometabolic reduction of TNT during a fed batch fermenta-
tion of glucose in the presence of the TNP-C4-HA conjugate
is shown in Figure 2 A. The conditions of the fermentation
experiment are listed in Table 1. TNT was reduced to DANT
with isomeric HADNT and ADNT as intermediates. TAT was
only detected in the control experiment without the TNP-
C4-HA conjugate or humic acids (data not shown). The
presence of the TNP-C4-HA conjugate caused a significant
decrease of the total amount of TNT and its metabolites in
solution (Figure 2A), particularly during the initial reduction
step (TNT to HADNT) and the last reduction step (DANT to
TAT). According to previous sorption experiments with TNT
and its metabolites (19), it can be concluded that hydroxy-
VOL. 33, NO. 19, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 4 2 3

humic acids. However, it cannot completely be ruled out
that other chemical alterations might have occurred in the
nitroaromatic compound which have lowered the residual
cross-reactivity. Due to the stability of the amid bond of the
TNP-C4-HA conjugate, the hydrolysis and release of TNP
residues can be excluded under the conditions used in the
experiment (23).
Remarkably, a significant decrease of the immunoassay
signal occurred only after most of the soluble TNT is converted
to the DANT isomers. As expected, the tranformation of nitro
groups of molecules which are bound to organic matter
proceeds much slower than the reduction of free nitroaro-
matic compounds. Obviously, the bound molecules were
less accessible to microorganisms or reducing species in the
medium. Moreover, substructures of the humic matrix in
the surrounding of the bound molecules could decrease the
potential of reduction of the nitro groups which also lowers
the reaction rate of the nitro groups of the covalently bound
molecules, when compared with those of free nitroaromatic
molecules. Nevertheless, the data from this experiments may
indicate that covalently bound nitroaromatic residues were
subject to further reductive transformation. The humic acids
without TNP-C4 conjugate which was used as control
experiment showed no immunological signal during the
entire treatment process. This clearly showed thatseven with
high concentrations of TNTsthe binding of reduced TNT to
humic acid did not generate structures which are similar to
the TNP conjugate. This observation supports the assumption
that the N-functions are mainly responsible for the binding.
To check whether these results are transferable to samples
of humic acids from TNT contaminated soil, a biological
anaerobic/ aerobic treatment of soil samples from a con-
taminated site was carried out.
Bound Trinitrophenyl Residues during Reductive Treat-
m ent of Contam inated Soil. The soil used for this experiment
was from a former TNT production plant in Stadtallendorf
near Marburg (Germany), and earlier examinations had
shown a strong signal in the immunoassay indicating bound
trinitrophenyl residues (24). These residues could originate
from reactions of partially oxidized methyl groups with soil
organic matter. The formation of such nitrobenzoic acids
has been described earlier by several authors during incuba-
tion of TNT with bacterial cultures (28) and in contaminated
soils (29-34). As intermediates of the oxidation of the methyl
group highly reactive aldehyde groups could be formed and
bind covalently to soil organic matter. 2,4,6-Trinitrobenzal-
dehyde was identified as an intermediate of a photocatalytic
degradation of TNT (35) and was found in soil of contami-
nated sites (30).
FIGURE 3. Reduction of TNT and its metabolites during the
anaerobic/aerobic treatment of TNT contaminated soil in a soil
slurry. (A) Amounts of TNT and its metabolites which were desorbed
from the soil with methanol. (B) Amounts of trinitrophenyl residues
measured by immunoassay from extracted humic acids of the treated
soil expressed as nmol TNP equivalents per kg dry soil.
During the anaerobic/ aerobic treatment of soil from
Stadtallendorf the concentrations of TNT, ADNT, and DANT
were measured in the supernatant of the slurry (data not
shown) and in methanol soil extracts (Figure 3A). The data
were very similar to those of experiments with contaminated
soil from Hessisch-Lichtenau during an anaerobic/ aerobic
treatment (22). Figure 3A shows the succession of the
reduction of TNT to the isomers of ADNT and DANT. Isomers
of hydroxylaminodinitrotoluene or TAT could not be detected
in the supernatant or in the methanol soil extracts. This
verifies again the reactivity of these intermediates toward
humic components of the soil.
The immunoassay signals of the covalently bound trini-
trophenyl residues (Figure 3B), measured after extraction of
humic acids, declined only slowly during the first reaction
period (0-7 days of incubation) and decreased significantly
after the ADNTs concentrations had reached their maxima.
During further anaerobic and aerobic treatment (after 42
days) the immunoassay signal decreased very slowly. After
the end of the treatment process, small amounts of bound
trinitrophenyl residues were still detectable.
FIGURE 4. Kinetic of interaction of isomers of hydroxylaminod-
initrotoluene with humic acids under anaerobic conditions. A control
experiment is run without humic acids under the same conditions.
the reduction of the nitro groups of TNP-C4-HA conjugate
a small amount of free TNT was added to the solution (Table
1
). In contrast to the fast reduction of free TNT and its
metabolites (Figure 2A), the immunoassay signal was nearly
constant during the initial 35 h (Figure 2B). Therefore, it can
be concluded that only minor or no changes of the bound
trinitrophenyl moiety took place during this time period.
The decrease of the signal differences after 35 h indicates
that the covalently bound trinitrophenyl residue was chemi-
cally converted to structures which were not bound any more
by the monoclonal antibody. Under the experimental condi-
tions, it is likely that further reduction of nitro groups
increases the reactivity of the groups toward organic soil
components. This may lead to an extensive cross-linking to
3
4 2 4
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 19, 1999

FIGURE 5. Potential reactions of reduced metabolites of TNT with soil. Reduction of nitro groups of free molecules (reaction 1), binding
of partially reduced (reaction 2a) and completely reduced metabolites (reaction 2b) to soil organic matter, and further reduction of nitro
groups of bound molecules (reaction 3). For multivalent binding at least two nitro groups should be reduced.
Immunoassay signal data obtained with the bioreme-
diation experiment of a TNT contaminated soil may indicate
that covalently bound trinitrophenyl residues were present.
They were further transformed during the biologically
mediated reduction. In contrast to the reduction of the
corresponding free TNT and metabolites, such transforma-
tion of covalently bound trinitrophenyl residues was con-
siderably slower. In addition, it started later than the reduction
of the corresponding molecules in solution which were
preferentially reduced.
at present all biological treatment processes aim at an
irreversible binding of transformed metabolites of TNT to
soil components. The fate of the contaminants during
biological treatment and the analysis of the remediated soil
is mostly monitored by the disappearance of free TNT and
1
4
its metabolites and by using radiolabeled [ C]TNT. These
methods, however, do not allow for following the fate of
partially reduced metabolites which are covalently and
irreversibly bound to the soil. Especially the extent of
reduction of nitro groups and the mode of incorporation of
the residues into soil organic matter were not analyzed. At
Coupling of HADNT to Hum ic Acids. To show the
significance of binding of partially reduced metabolites of
TNT to humic acids sorption experiments were carried out.
Figure 4 shows the kinetics of sorption of isomeric HADNT
to humic acids which were extracted from TNT contaminated
soil. The concentration of HADNT rapidly decreased, and
after 120 min all HADNT was no longer detectable. These
results are similar to observations of sorption of HADNT to
commercial humic acids from brown coal (19). The rate of
sorption of HADNT to humic acids is much higher than that
observed with TNT, 4-ADNT, DANT, and TAT (19). These
observations suggest that the reaction of HADNT with organic
compounds is an important mechanism for initial binding
and immobilization of reduced metabolites of TNT (Figure
1
3
15
present only NMR spectroscopy ( C or N) or immunological
tests can detect bound molecules and determine the degree
of reduction and the status of immobilization at the end of
a treatment process. The present results show that in addition
to the fast disappearance of extractable TNT and its
metabolites their complete and irreversible incorporation
into the soil matrix must be demonstrated.
As shown here by the immunoassay, the detected trini-
trophenyl residues are present in contaminated soils only in
minor amounts. Although bound residues generated by
chemisorption of HADNT and probably other reduced
metabolites to humic acids are not detected with the
monoclonal antibody used in the present immunoassay (27),
it is likely that the rate and the extent of reduction of any
covalently bound structure (e.g. TNT bound by an oxidized
methyl group or reduced nitro group) is slow.
5
, reaction 2a). Extractions with different solvents, both under
anaerobic and aerobic conditions, did not release HADNTs
or related compounds (ADNT, DANT). This confirmed the
irreversible character of the binding of HADNT to humic
acids. These HADNT humic acids coupling products did not
interact with the trinitrophenyl monoclonal antibody which
confirmed its highly specific character.
In Figure 5, a hypothetical reaction scheme for reduction
and binding of TNT and its metabolites during the anaerobic/
aerobic treatment process is shown. Three major reactions
must be considered. (i) Reduction of nitro groups of free
molecules (reaction 1): This reaction is well documented in
the literature (4, 19). (ii) Binding of reactive reduced
intermediates to soil organic matter (reaction 2a + 2b): Both
Significance for Biorem ediation of TNT Contam inated
Soils. Since extensive mineralization of TNT in soils cannot
be accomplished by the available bioremediation techniques,
VOL. 33, NO. 19, 1999 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 4 2 5

partially reduced metabolites (HADNT; reaction 2a) and
completely reduced metabolites (TAT; reaction 2b) were
shown to be highly reactive in the present of humic acids
(6) Naumova, R. P.; Selivanovskaya, S. Y.; Cherepneva, I. E. Appl.
Biochem. Microbiol. 1989, 24, 409-413.
(
7) Preuss, A.; Fimpel, J.; Diekert, G. Arch. Microbiol. 1993, 159,
45-353.
8) Funk, S. B.; Roberts, D. J.; Crawford, D. L.; Crawford, R. L. Appl.
Environ. Microbiol. 1993, 59, 2171-2177.
3
(
(
19). (iii) Further slow reduction of the bound residues
reaction 3) which was pointed out in the present work.
The rates of the three reactions may be very different.
(
(9) Boopathy, R.; Kulpa, C. F.; Wilson, M. Appl. Microbiol. Biotech.
1993, 39, 270-275.
10) Boopathy, R. B.; Manning, J.; Montemagno; C.; Kulpa, C. Curr.
Microbiol. 1994, 28, 131-137.
11) Boopathy, R.; Kulpa, C. F. Can. J. Microbiol. 1994, 40, 273-278.
12) Fiorella, P. D.; Spain, J. C. Appl. Environ. Microbiol. 1997, 63,
Obviously, the reactive HADNT intermediates interact very
fast with humic acids and generate nonextractable residues.
The slow and delayed transformation of the bound trini-
trophenyl residues, as observed by the immunoassay, indicate
that the bound HADNT is reduced much slower than the
free molecules (reaction 1 versus reaction 3). It must be
assumed that multivalent binding of metabolites of TNT is
limited by the extent of reduction of nitro groups of the bound
molecules. Therefore, the anaerobic phase of the biological
treatment process and the reduction of the nitro groups may
be crucial. For progressive cross-linking of the contaminants
the anaerobic phase has to be continued even though free
components are no longer detectable in organic soil extracts.
(
(
(
2
007-2015.
(13) Khan, T. A.; Bhadra, R.; Hughes, J. J. Ind. Microbiol. Biotechnol.
1997, 18, 198-203.
(
14) Vorbeck, C.; Lenke, H.; Fischer, P.; Spain, J. C.; Knackmuss,
H.-J. Appl. Environ. Microbiol. 1998, 64, 246-252.
15) Parrish, F. W. Appl. Environ. Microbiol. 1977, 34, 232-233.
16) Stahl, J. D.; Aust, S. D. Biochem. Biophys. Res. Commun. 1993,
(
(
1
92, 477-482.
(17) Bumpus, J. A.; Tatarko, M. Curr. Microbiol. 1994, 28, 185-190.
(18) Michels, J.; Gottschalk, G. Appl. Environ. Microbiol. 1994, 60,
187-194.
1
5
This is in accord with N NMR measurements which show
that unreduced nitro groups of bound molecules were still
found at the end of the anaerobic treatment process (21).
Thus, the immunoassay for nitroaromatic residues bound
to humic acids can be a suitable tool for monitoring the
progress of bioremediation of TNT contaminated soil.
Moreover, this test system allows for following structural
changes of bound residues during the treatment. Through
combination of this newly developed technique with estab-
(
19) Daun, G.; Lenke, H.; Reuss, M.; Knackmuss, H.-J. Environ. Sci.
Technol. 1998, 32, 1956-1963.
20) Dunnivant, F. M.; Schwarzenbach, R. P.; Macalady, D. L. Environ.
Sci. Technol. 1992, 26, 2133-2141.
(21) Knicker, H.; Bruns-Nagel, D.; Drzyzga, O.; L o¨ w, E.; Steinbach,
K. Environ. Sci. Technol. 1999, 33, 343-349.
22) Lenke, H.; Warrelmann, J.; Daun, G.; Hund, K.; Sieglen, U.; Walter,
(
(
(
U.; Knackmuss, H.-J. Environ. Sci. Technol. 1998, 32, 1964-
1
971.
23) Pfortner, P.; Weller, M. G.; Niessner, R. Fresenius J. Anal. Chem.
1998, 360, 192-198.
1
4
lished methods for the detection of bound residues (i.e.
C
1
5
labeling, N NMR spectroscopy) additional information
about the formation and long-term stability of bound residues
become available. If an appropriate antibody for the detection
of bound dinitrotolyl residues becomes available, it should
be possible to follow the formation and fate of bound ADNT
residues which are suspected to be key metabolites during
bioremediation under composting (e.g. aerobic rather than
anaerobic) conditions.
(24) Pfortner, P.; Weller, M. G.; Niessner, R. Fresenius J. Anal. Chem.
998, 360, 781-783.
1
(
25) Pfortner, P. Entwicklung von Enzymimmunoassays (ELISAs)
zur Bestimmung von huminstoffgebundenen Nitroaromaten.
Ph.D. Thesis, Technische Universit a¨ t M u¨ nchen, 1998.
26) Achtnich, C.; Sieglen, U.; Knackmuss, H.-J.; Lenke, H. Environ
Toxicol Chem. In press
(
(27) Zeck, A.; Weller, M. G.; Niessner, R. Fresenius J. Anal. Chem.
999, 364, 113-120.
1
(
28) Vanderberg, L. A.; Perry, J. J.; Unkefer, P. J. Appl. Microbiol.
Biotechnol. 1995, 43, 937-945.
29) Kaplan, D. L. Curr. Opin. Biotechnol. 1992, 3, 253-260.
(30) Jenkins, T. F.; Walsh, M. E.; Schumacher, P. W.; Miyares, P. H.
J. Assoc. Off. Anal. Chem. 1989, 72, 890-899.
31) Preiss, A.; Levsen, K.; Humpfer, E.; Spraul, M. Fresenius J. Anal.
Chem. 1996, 356, 445-451.
32) Preiss, A.; Lewin, U.; Wennrich, L.; Findeisen, M.; Efer, J. Fresenius
J. Anal. Chem. 1997, 357, 676-683.
33) Breitung, J.; Bruns-Nagel, D.; Steinbach, K.; Blotevogel, K.-H.;
Gorontzy, T.; Dillert, R.; Winterberg, R.; Stoffers, H.; Haas, R.;
M u¨ ller, M.; Asbach, P.; Kaminski, L.; L o¨ w, E.; Gemsa, D. Z.
Umweltchem. O¨ kotox. 1996, 8, 249-254.
Acknowledgments
(
We gratefully acknowledge Albert Niedermeier for performing
some of the experiments. We would like to thank Strategic
Diagnostics Inc., Newark, DE, for providing us with mono-
clonal TNT antibodies. The partial financial support by the
Deutsche Forschungsgemeinschaft and the Fonds der Che-
mischen Industrie is gratefully acknowledged.
(
(
(
Literature Cited
(
(
(
1) Preuss, J.; Haas, R.; Koss, G. Geogr. Rundschau 1988, 40, 31-38.
2) Haas, R.; v. L o¨ w, E. Forum St a¨ dte-Hygiene 1986, 37, 33-43.
3) Toxicity of Nitroaromatic Compounds; Rickert, D. E., Ed.;
Hemisphere: Washington, DC, 1985.
(
(
34) Schmidt, T. C.; Petersmann, M.; Kaminski, L.; L o¨ w, E.; Stork, G.
Fresenius J. Anal. Chem. 1997, 357, 121-126.
35) Spanngord, R. J.; Mabey, W. R.; Chou, T. W.; Smith, J. H. In
Toxicity of Nitroaromatic Compounds; Rickert, D. E., Ed.;
Hemisphere: Washington, DC, 1985; pp 15-33.
(
(
4) Rieger, P.-G.; Knackmuss, H. J. Basic knowledge and perspectives
on biodegradation of 2,4,6-trinitrotoluene and related com-
pounds in contaminated soil. In Biodegradation of nitroaromatic
compounds; Spain, J., Ed.; Plenum Press: New York, 1995; pp
Received for review February 15, 1999. Revised manuscript
received June 29, 1999. Accepted July 12, 1999.
1
-18.
5) McCormick, N. G.; Feeherry, F. E.; Levinson, H. S. Appl. Environ.
Microbiol. 1976, 31, 949-958.
ES9901765
3
4 2 6
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 33, NO. 19, 1999