Environ. Sci. Technol. 2008, 42, 2563–2569
Absorption, Tissue Distribution, And Elimination of Residues after 2,4,6-Trinitro[14C]toluene Administration to Sheep D . J . S M I T H , * ,† A . M . C R A I G , ‡ J. M. DURINGER,§ AND R. L. CHANEY4 Animal Metabolism-Agricultural Chemicals Research Unit, USDA ARS, 1605 Albrecht Blvd. Fargo, North Dakota, 58105-5674, Department of Biomedical Sciences, College of Veterinary Medicine, Oregon State University, Corvallis, Oregon 97331, Department of Environmental and Molecular Toxicology, Oregon State University, Corvallis, Oregon, 97331, and Environmental Management and By-Product Utilization Laboratory, 10300 Baltimore Ave, BARC-West, Beltsville, Maryland, 20705-2350
Received October 17, 2007. Revised manuscript received December 17, 2007. Accepted December 31, 2007.
The compound 2,4,6-trinitrotoluene (TNT) is a persistent contaminant of some industrial and military sites. Biological bioremediation techniques typically rely on the immobilization of TNT reduction products rather than on TNT mineralization. We hypothesized that sheep ruminal microbes would be suitable for TNT destruction after phytoremediation of TNT-contaminated soils by cool-season grasses. Therefore we investigated the fate of [14C]TNT in ruminating sheep to determine the utility of ruminant animals as a portion of the bioremediation process. Three wether sheep were dosed with 35.5 mg each of dietary unlabeled TNT for 21 consecutive days. On day 22 sheep (41.9 ( 3.0 kg) were orally dosed with 35.5 mg of [14C]TNT (129 µCi; 99.1% radiochemical purity). Blood, urine, and feces were collected at regular intervals for 72 h. At slaughter, tissues were quantitatively collected. Tissues and blood were analyzed for total radioactive residues (TRR); excreta were analyzed for TRR, bound residues, and TNT metabolites. Plasma radioactivity peaked within 1 h of dosing and was essentially depleted within 18 h. Approximately 76% of the radiocarbon was excreted in feces, 17% in urine, with 5% being retained in the gastrointestinal tract and 1% retained in tissues. Parent TNT, dinitroamino metabolites, and diaminonitro metabolites were not detected in excreta. Ruminal and fecal radioactivity was essentially nonextractable using ethyl acetate, acetone, and methanol; covalent binding of fecal radioactive residues was evenly distributed among extractable organic molecules (i.e., soluble organic matter, soluble carbohydrate, protein, lipid, and nucleic acid fractions) and undigested fibers (cellulose, hemicellulose, and lignin). This study demonstrated that TNT
reduction within the ruminant gastrointestinal tract leads to substantial immobilization of residues to organic matter, a fate similar to TNT in other strongly reducing environments.
Introduction The compound 2, 4, 6-trinitrotoluene (TNT) has been manufactured for industrial and military purposes since about 1900. In the United States, the production of TNT was stopped in the mid-1980s because of the large amounts of wastes being generated during its manufacture and the economic burden required for waste disposal (1). In spite of the discontinued production of TNT, the U.S. Army has estimated that over 1.2 million tons of soil has been contaminated with high explosives (1). Because TNT and its metabolites are believed to be mutagens (2) and because they are long-lived environmental contaminants (1), substantial efforts have been expended developing remediation tools for nitroaromatic contaminated soils (3, 4). Several bioremediation methods for TNT have been proposed, but the most widely studied for large contamination sites are slurry reactors and composting (1, 3), each based on the microbial degradation of TNT. Phytoremediation has also been proposed because several plant species absorb and transform trinitrotoluene (4). A limitation of aerobic or anaerobic reaction systems has been the relatively slow or nonexistent rates of TNT mineralization that occur in each (5). Recently, Fleischmann et al. (6) have demonstrated that high concentrations of TNT (100 ppm) were rapidly and extensively transformed in anaerobic bovine ruminal fluid. The formation of diaminonitrotoluene isomers and dihydroxyamino-nitrotoluene and other metabolites, however, was only transitory. Each metabolite was further metabolized to polar products which could not be identified because a radiolabeled marker was not used. Nevertheless, the results of Fleischmann et al. (6) suggest that grazing ruminant animals might rapidly and completely metabolize nitroaromatic munitions present in plant materials. Although several studies have investigated the excretion of TNT metabolites in humans after industrial exposures, few studies have examined the absorption, distribution, metabolism, and excretion of TNT in mammals. In rats, significant quantities of TNT metabolites were covalently bound to tissues after intraperitoneal administration of [14C]TNT (7). Later studies have established that TNT is activated to a reactive metabolite (or metabolites) before covalent binding could occur (8). Palmer et al. (9) demonstrated that radioactivity immobilized onto 14C-TNT treated compost was partially available when intratracheally administered to rats, but none of the radioactivity was dosed as parent TNT or its proximal reduction products. The objectives of this study were to determine the absorption, distribution, and elimination of [14C]TNT in sheep after preliminary exposures to unlabeled dietary TNT for a threeweek period.
Experimental Section * Corresponding author phone: (701)-239-1238; fax: (701)-2391430; e-mail:
[email protected]. † Animal Metabolism-Agricultural Chemicals Research Unit. ‡ College of Veterinary Medicine, Oregon State University. § Department of Environmental and Molecular Toxicology, Oregon State University. 4 EnvironmentalManagementandBy-ProductUtilizationLaboratory. 10.1021/es702601n CCC: $40.75
Published on Web 02/13/2008
2008 American Chemical Society
Chemicals. UL-[14C]Toluene was purchased from Sigma Chemical Co., and 2,4,6-trinitro-[14C]toluene was synthesized as previously described (10) and was recrystallized from 95% ethanol to a radiochemical purity of 99.1% (8090 ( 54 dpm/ µg). Unlabeled TNT was synthesized as described by Dennis et al. (11). Chemical purities were assessed by 1H NMR, mass spectral analyses, and by HPLC with radiochemical detection for [14C]TNT or by UV detection (231 nm). VOL. 42, NO. 7, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2563
Test Animals. An animal protocol was approved by the Institutional Animal Care and Use Committee. Four wether sheep (three test animals, one control) were purchased from North Dakota State University and housed within 10 m2 indoor, concrete-floored pens covered with wood shavings. Sheep had free access to a grass-hay, alfalfa mixture and to water. Animals were adapted to the barn facilities, were halter broken, and starting 2 weeks after delivery, animals were trained to eat a daily grain supplement (0.5 kg) through head gates. At the initiation of the study, test animals (n ) 3) received 0.5 kg of ground corn containing 33.1 mg of unlabeled TNT for 21 consecutive days. The level of TNT provided was estimated from TNT plant uptake studies (12) showing a maximal TNT level of 12 µg/g and a maximal estimated forage intake for sheep of about 3.5% of body weight per day. The TNT-fortified corn was prepared by adding sequential aliquots of 2321 mg of TNT dissolved in 50 mL of acetone to 35.0 kg of cracked corn within a stainless steel ribbon mixer. Each acetone aliquot was allowed to evaporate and the grain was mixed for approximately 10 min prior to the addition of the next aliquot of TNT. [14C]TNT Formulation and Administration. A total of 463.5 mg of [14C]TNT (1689 µCi) were dissolved in 5.0 mL of acetone; 0.383 mL of the [14C]TNT acetone solution (129.4 µCi; 35.5 mg) were added to each of three gelatin capsules filled with cracked corn. The acetone was allowed to evaporate and each capsule was capped. At dosing, the three test sheep (41.9 ( 3.0 kg) were placed in metabolism crates (13) and a single [14C]TNT containing capsule was orally administered to each sheep using a balling gun. For sheep nos. 368 and 370, dosing was uneventful, but sheep no 367 fought during the dosing, and the balling gun caused an abrasion in the esophagus near the larynx; thereafter, sheep no. 367 consistently refused feed. Sample Collection and Radiochemical Analyses. Urine, feces, and blood were collected sequentially after dosing until tissues were quantitatively collected at slaughter (see Supporting Information). Expired gases were not collected because mineralization of TNT in soil and anaerobic slurry reactors is minimal (5, 14). Other than collection of urine on ice and rapid freezing of tissues and excreta, precautions were not taken to prevent exposure of samples to oxygen. Quantification of radioactivity in untreated urine and processed feces and tissues (15) was conducted using standard combustion and/or liquid scintillation counting techniques (described in the Supporting Information). Limits of detection and quantification were determined as previously described (16). Characterization of Urinary Radioactivity. Initial efforts to characterize urinary radioactivity were based on literature reports (17). After extraction (see Supporting Information), ethyl acetate layers were removed and dried over sodium sulfate. Aliquots of extracted urine samples were assayed for radioactivity by liquid scintillation counting (LSC) and for parent TNT by GC-MS using a 30 m DB-5 MS column. Aliquots (2 µL) were cool-on-column injected at a column temperature of 100 °C; the temperature was held for 1 min and then ramped to 200 at 3 °C/min. After a 10 min hold at 200 °C, the temperature was increased to 275 °C over the next 3 min. Selected ion monitoring (resolution 5000; EI, 40 eV) was used to detect TNT at m/z 227.13 and ditnitroamino toluene isomers at m/z 197.04. Mass spectral analysis did not include the investigation of TNT dimers that may have formed during excretion or sample preparation. Acid Hydrolysis of Urinary Radiocarbon. Urine samples (1 mL) were hydrolyzed by the addition of 100 µL of 3.3 M HCl and incubation for 30 min. Acidic samples were neutralized with sodium carbonate and then extracted with ethyl acetate. In a second hydrolysis experiment equal volumes of urine and concentrated HCl were mixed and 2564
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 7, 2008
incubated in a shaking water bath at 85 °C for 2 h. Radioactivity was extracted into ethyl acetate as described above. Solid Phase Extraction (SPE) of Urinary Radiocarbon. Control urine samples (1 mL), fortified with ∼5400 (667 ng) or 52 000 dpm (6400 ng) of [14C]TNT, and test samples (1 mL) were loaded onto previously conditioned (5 mL acetone followed by 10 mL nanopure water) C18 SPE tubes (500 mg sorbent; Varian Assoc.). Radioactivity in the aqueous “load” fraction was collected, and radioactivity adsorbed to the sorbent eluted with 100% acetone. Enzyme Hydrolysis of Urinary Radiocarbon. Urine samples (1 mL) were mixed with 2 mL of 1 M sodium acetate buffer (pH 5.0) and 25 µL (5000 U) of glucuronidase/sulfatase from Patella vulgata. Samples were incubated overnight in a shaking water bath at 37 °C. Control incubations containing water in the place of enzyme, and positive control incubations containing 4.7 µg of synthetic ractopamine glucuronide were concurrently run. Hydrolyzed urine samples were loaded onto C18 SPE cartridges, the cartridges rinsed with water, and sequentially eluted with acetone, methanol, and ammonium acetate. Radioactivity in each fraction was determined by LSC. Chromatographic Characterization of Urinary Radiocarbon. Urine extracts were chromatographed using a Waters 600 controller and pump equipped with a Waters 717 autosampler. Radioactivity was eluted from the column (4.6 × 250 mm Symmetry C8; Waters; Milford, MA) using an initial 5 min isocratic solvent consisting of 95/5 (v/v) of water/ acetonitrile followed by a linear gradient to 5% water and 95% acetonitrile over 60 min; the column was flushed with 95% acetonitrile for 5 min prior to returning to starting conditions. Radioactivity was trapped in 2 min fractions for the duration of each chromatographic run. Gel Filtration of Urinary Radiocarbon. A 4.5 × 90 cm column was packed with Sephadex G-75 and equilibrated in 0.05 M phosphate buffer (pH 7.2). To the top of the column a 9 mL aliquot of urine (sheep 368, 0–6 h, 108 056 dpm/mL) was added and allowed to adsorb by gravity flow to the head of the column. Buffer (0.05 M phosphate, pH 7.2) was allowed to flow over the column at 1.25 mL/min via gravity flow and fractions were collected in 10 min increments. Aliquots (0.5 mL) of each fraction were analyzed for radioactivity using LSC. A 2.5 × 60 cm column packed with Sephadex G-10 was loaded with 10 mL of urine (sheep 368, 0–6 h, 108 056 dpm/ mL) and eluted with 1.25 mL/min of phosphate buffer (pH 7.2). Fractions were collected every 5.8 min. Radioactivity in each fraction was determined by LSC counting. Extraction of Radioactivity from Sheep Ruminal Contents. Aliquots (5 g) of ruminal fluid from each sheep were extracted sequentially with 20 mL of ethyl acetate, acetone, acetone/methanol (50/50, v/v), and methanol. Organic solvent was added, samples were vortexed for 30 s, sonicated with a probe sonicator, and then centrifuged at 30 600g for 10 min. Radioactivity in the solvent layer was determined by LSC. Radioactivity Bound to the Neutral Detergent Fiber (NDF) and Acid Detergent (ADF) Cell Wall Fractions. Aliquots (0.25 g wet weight) of feces were weighed in duplicate and transferred to Berzelius beakers for fiber analysis using the method of Van Soest (see refs 18 and 19, and the Supporting Information). Extraction of Fecal Radioactivity into Chemical Fractions. Fecal radioactivity was fractionated into crude aqueous, carbohydrate, lipid, nucleic acid, and protein fractions using the method described by Sutherland and Wilkinson (20) as modified by Carpenter et al. (ref 21; see Supporting Information). Pharmacokinetic Modeling. Pharmacokinetic parameters for serum total radioactive residues (TRR) were estimated
TABLE 1. Disposition of Radiocarbon into Urine, Feces, And Tissues of Sheep Dosed with 35.5 mg of [14C]TNT in a Single Oral Dosea wether fraction
item
urine
367
T0–6 16.8 T6–12 9.5 T12–18 3.2 T18–24 2.2 T24–32 0.5 T32–40 0.0 T40–48 0.7 T48–60 0.3 T60–72 0.4 total: 33.6 feces: T0–6 0.2 T6–12 0.1 T12–18 0.1 T18–24 0.1 T24–32 0.0 T32–40 0.1 T40–48 0.0 T48–60 0.5 T60–72 3.9 total: 5.0 tissues adipose 0.00 kidney 0.03 liver 0.19 skeletal muscle 0.47 bile 0.01 blood 0.00 bone 0.24 brain 0.00 eye 0.02 heart 0.01 large intestine content 8.80 large intestine tissue 0.14 lung 0.03 rumen content 51.18 rumen tissue 0.93 skin 0.38 small intestine content 0.26 small intestine tissue 0.08 spleen 0.01 thyroid 0.00 remainder of carcass 0.31 total: 63.09 cage 0.4 wash total recovery (%): 102.1
368
370
averageb
13.7 2.6 0.8 0.3 0.2 0.2 0.1 0.2 0.1 18.2 0.2 2.7 11.4 6.6 11.7 14.8 9.4 11.4 5.8 74.0 0.00 0.01 0.06 0.00 0.00 0.00 0.00 0.00 0.01 0.00 3.02 0.02 0.00 3.84 0.09 0.06 0.15 0.02 0.00 0.00 0.03 7.31 0.2
11.5 2.6 0.6 0.4 0.2 0.2 0.2 0.1 0.1 15.9 0.2 3.3 12.1 10.7 15.8 16.3 8.6 8.8 3.3 79.1 0.00 0.01 0.08 0.03 0.00 0.00 0.05 0.00 0.00 0.00 1.78 0.01 0.01 1.36 0.03 0.08 0.06 0.01 0.00 0.00 0.02 3.53 0.4
12.6 2.6 0.7 0.4 0.2 0.2 0.2 0.2 0.1 17.1 0.2 3.0 11.8 8.7 13.8 15.6 9.0 10.1 4.6 76.6 0.00 0.01 0.07 0.02 0.00 0.00 0.03 0.00 0.01 0.00 2.40 0.02 0.01 2.60 0.06 0.07 0.11 0.02 0.00 0.00 0.03 5.33 0.3
99.7
98.9
99.3
a
Data are expressed as a percentage of the [14C]TNT dose. b Average of wether 368 and 370.
using WinNonlin version 1.5 (Scientific Consulting, Inc.; Cary, NC). Data were fit to a WinNonlin model 11 which was a two compartment model with no lag time and first order elimination.
Results and Discussion Only about 20% of the radioactivity dosed to sheep was absorbed from the gastrointestinal tracts. Half-lives of absorption were less than 1 h with maximal concentrations of absorbed residues occurring under 90 min. Such rapid appearance of radioactivity in blood serum (see Supporting Information) strongly suggests the ruminal absorption of residues. Elimination of total residues from serum followed a biphasic manner with R- and β-half-lives of 1.9 and 43.8, and 1.0 and 37.3 h, for sheep 368 and 370, respectively. Recovery of dosed radiocarbon was essentially quantitative (Table 1). In wether 367, the majority of the radiolabel
(51%) was recovered in the “rumen contents” fraction. The retention of radioactivity in the rumen, in conjunction with the almost absolute cessation of eating and defecation (335 g of feces vs >3600 g of feces each for wethers 368 and 370) are strong indicators that this animal was not representative of a healthy ruminating animal. For that reason, means presented in Table 1 include only animals 368 and 370. The major route for radiocarbon excretion was the feces, with 76.6% of the administered dose eliminated in fecal matter (Table 1). In contrast, an average of 17.1% of the radiocarbon was excreted in urine, and a majority of that was excreted in the first 6 h of dosing. The major portion of retained radioactivity was present in ruminal and large intestinal contents (2.6 and 2.4%, respectively). Small intestinal contents contained only 0.1% of the dosed radioactivity. Traditionally edible tissues (adipose, kidney, liver, and skeletal muscle) and inedible tissues generally contained less than 0.1% of the dosed radioactivity at slaughter (72 h post dosing). When distribution data were expressed on a concentration basis (µg kg-1 of TNT equivalents; see Supporting Information), TRR ranged from nondetectable in adipose tissue, skeletal muscle, bile, brain, and thyroid to 70, 40, 24, 18, and 11 µg kg-1 wet weight in eye, liver, kidney, rumen tissue, and spleen, respectively (detection limit was ∼4 µg kg-1 wet weight of TNT equivalents). Total tissue radioactive residues may be underestimated because bioaccumulation of nonradioactive residues that could have occurred during the 21 day TNT exposure period were not measured in this study. The quantitative recovery of radiocarbon without the collection of respiratory gases, clearly indicated that mineralization of TNT did not occur within the 72 h study period. Bioremediation through the rapid mineralization of TNT has been elusive using a number of aerobic and anaerobic, microbiologically based systems (5). Our data strongly suggest that the intestinal tracts of ruminants are no different with regards to TNT mineralization than several previously studied anaerobic biomes (23–27). Because it could be argued that ruminal bacteria might adapt to the presence of TNT, a 21 day adaptation period was included in the study in which sheep were provided unlabeled TNT (33.5 mg d-1) as a portion of their daily grain supplement. This exposure to unlabeled TNT would, on a theoretical basis, allow time for microbial adaptations to dietary TNT and would also allow for the induction of biotransformation enzymes that occurs after TNT absorption (28). Urinary excretion data clearly show that the greatest portion of absorbed radioactivity was eliminated during the first 6 h period after dosing (Table 1). Without acid hydrolysis, 4.0-13.5% of the urinary radioactivity excreted during the initial 24 h of the study was extractable into ethyl acetate (see Supporting Information); with acid hydrolysis, extractable radioactivity ranged from 6.5 to 19.1% of the total. Radioactivity fortified into control urine as parent [14C]TNT (0.65 and 6.5 µg L-1) was readily extracted from native (86.9-93.6%) or acid hydrolyzed urine (82.9-85.8%). Radioactivity extracted from urine of dosed sheep was not in the form of parent TNT, or amino-dinitro toluenes or diamino-nitro toluene as determined by GC-MS. In contrast, parent TNT was readily identified in ethyl acetate extracts of fortified urine samples (data not shown). Increasing the severity of acid hydrolysis (i.e., 1:1 dilution with concentrated HCl and heating for 2 h at 85 °C; Supporting Information) did not greatly increase the amount of radiocarbon extracted into ethyl acetate. None of the radioactivity in hydrolyzed samples eluted at the retention time of TNT (data not shown). Figure 1 (top panel) shows the chromatographic distribution of urinary radioactivity after partitioning into water and acetone extracts using C-18 SPE cartridges. The greatest portion of radioactivity eluted as unresolved peaks in the VOL. 42, NO. 7, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2565
FIGURE 1. Top panel: Distribution of urinary radioactivity eluted from C18 solid phase extraction (SPE) tubes using water (solid line; 67.2% of urinary radiocarbon) and acetone (hatched line; 23.5% of urinary radiocarbon). Bottom panel: Distribution of urinary radiocarbon eluted from C18 SPE tubes using acetone (53% of urinary radioactivity), methanol (10% of urinary radioactivity), and a mixture of buffer-methanol (4% of urinary radioactivity) after hydrolysis with β-glucuronidase/sulfatase from Patella vulgata. Chromatography of radioactivity eluted from SPE during sample loading and rinsing with water (26% of the urinary radioactivity) is not shown. very polar portions of the chromatograms, with lesser quantities of radioactivity eluting in fractions 10-15. Hydrolysis with β-glucuronidase from Patella vulgata and subsequent partitioning of radioactivity into water (not shown) acetone, methanol, and buffer (0.05 M ammonium acetate, pH 4.5) fractions (Figure 1, bottom panel) shows that a portion of the radioactivity was shifted away from the chromatographic void volume so that it eluted as broad peaks covering fractions 5-15 for the acetone elution, fractions 7-20 for the methanol fraction, and 7-15 for the buffermethanol fraction. Radioactivity “released” by treatment with β-glucuronidase was not associated with specific UV (230 nm) peaks that could be distinguished from background. Recovery of radiocarbon applied to a Sephadex G-75 column was 98.4%, but the radioactivity eluted as a broad peak from about fraction 90 to fraction 130 (see Supporting Information) that was not associated with measurable protein. For the G-10 column, designed to separate molecules with molecular weights lower than about 700, the cumulative recovery of radioactivity was only 81.5% over 240 fractions, with the radioactivity distributed fairly evenly from fraction 25 (high relative molecular weight) to fraction 150 (lower molecular weight). Broad distribution of the radiocarbon on both reverse phase and size exclusion columns suggests that the radioactivity was associated with a broad range of molecular weights. Two radioactive peaks (corresponding to fractions 67–91 and 92–107, respectively; see Supporting Information) collected from the Sephadex G-10 column were chromatographed using reversed phase chromatography. When chromatographed under hydrophilic conditions (i.e., 95% water, 5% isopropanol), radioactivity was not retained on the column. The anionic ion-paring reagent (22), pentane 2566
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 7, 2008
sulfonic acid (1 mM) did not cause radioactivity to elute at regions other than the void volume, suggesting that the radioactive metabolites were not cationic. Hexadecyltrimethylammonium bromide (1 mM) was then used as a cationic mobile phase additive. For peak 2, about 7% of the radioactivity was retained at a retention time of 39 min as a multiplet of peaks with the remaining radioactivity eluting in the void volume. For peak 1, about 3.5% of the radioactivity was retained at about 38 min with about 43% of the injected radioactivity not retained on the column; 47% of the radioactivity was not recovered (total recovery of 53%). Physical methods (mass spectrometry, NMR) were not used to characterize metabolites because they could not be purified to homogeneity. Thus, the chemical forms in which urinary radioactivity was excreted remain unknown. It was clear, however, that the radiocarbon was not excreted as parent TNT or isomers of aminodinitrotoluene, the metabolic products of TNT most commonly extracted from urine of nonruminant animals (17, 29–31). Aminodinitrotoluene isomers are known to be produced by ruminant microorganisms (6); however, their presence in ruminal fluid was transitory. Relative to the studies of Flieschmann et al. (6), who measured TNT transformation in ruminal fluid at initial concentrations of 100 mg L-1, a low dose (35.5 mg total or 0.85 mg/kg–1 body weight) of [14C]TNT was administered to sheep in this study. Assuming a ruminal content of approximately 4.5 L (32), the initial ruminal concentration of TNT was approximately 8 mg L-1, a level about 12 times lower than the level shown by Fleishmann (6) to be completely reduced in 1 h. Efforts to extract radioactivity from ruminal fluid and feces with solvents commonly used to extract TNT and its free metabolites were not productive (see Supporting Information), with essentially 100% of the radioactivity bound to solids. Significant quantities of nonextractable, covalently bound metabolites have also been measured after [14C]TNT application to soil (33), compost (26; 34), soil-bioslurry reactors (23, 27), sewage sludge (21), sediment (35), cow manure (36), crops or grasses (37–40), trees (41), and pure bacteria (42). The formation of reactive metabolites capable of covalent binding is thought to occur through the sequential reduction of TNT to mono and diamino nitrotoluene isomers through unstable nitroso and hyrdoxylamino intermediates (43). Triaminotoluene, the most reduced metabolite of TNT, is formed only under strictly anaerobic conditions (26, 43). All major products of TNT reduction, namely 4-amino-2,6dinitrotoluene, 2-amino-4,6-dinitrotoluene, 2,4,-diamino-6nitrotoluene, 2,6-diamino-4-nitrotoluene, and triaminotoluene, serve as substrates or precursors capable of covalently binding to carbonyl or quinones (44) present in numerous natural matrices. Hydroxyaminodinitro toluene isomers, nitroso-dinitrotoluene isomers, and triaminotoluene have all been implicated as specific chemical species capable of nucleophilic addition to available carbonyls or other electron acceptors (8, 25, 45, 46). Covalent binding may occur under either aerobic (8) or anaerobic (27) conditions. Although the formation of TNT transformation products was not directly measured in this study, the excretion of radioactivity bound to feces strongly implies their ruminal formation. Ruminal microbes have the capacity to reduce TNT to 2-hydroxyamino-4,6,-dinitro toluene (6), a metabolite implicated in covalent binding (8, 14). Reduction potentials of less than -200 mV are typically required for the complete reduction of TNT to triamino toluene (43, 47); reduction potentials in ruminal fluid of live animals are typically -250 to -450 mV (19). Thus, the complete reduction of monoamino- and diaminonitrotoluene isomers likely occurred in the ruminal fluid of sheep in this study. Ruminal formation of aminodinitrotoluene, diaminonitrotoluene, and(or) triaminotoluene, an unstable molecule under aerobic or anaerobic
TABLE 2. Distribution of fecal radioactivity into chemical fractionsa wether 368
wether 370
b
supernate fractionb
supernate fraction excretion Time (h)
1
2
3
4
5
6
final pellet
1
2
3
4
5
6
final pellet
12–18 18–24 24–32 32–40 40–48
7.1 4.9 5.7 3.8 3.2
3.0 0.9 1.1 1.2 1.9
0.5 1.1 0.8 0.5 0.0
17.2 15.9 16.1 17.3 17.6
12.5 9.2 10.8 9.4 10.0
27.6 4.8 15.6 11.4 22.8
32.1 63.2 49.9 56.4 44.5
3.8 3.7 4.2 3.7 3.2
1.5 0.9 0.7 1.3 1.2
0.5 0.9 0.9 0.8 0.0
16.8 14.8 16.3 16.1 16.8
12.2 10.4 11.0 8.9 8.4
25.0 6.6 10.8 15.6 25.8
40.2 62.7 56.1 53.6 44.6
average std. dev.
4.9 1.5
1.6 0.9
0.6 0.4
16.8 0.8
10.4 1.3
16.4 9.0
49.2 11.9
3.7 0.4
1.1 0.3
0.6 0.4
16.2 0.8
10.2 1.5
16.8 8.5
51.4 9.0
a Data are expressed as the percentage of the TRR present in the indicated fraction. b Fractions 1and 2 represent water soluble organic matter; fraction 3, soluble carbohydrate; fraction 4, lipid; fraction 5, nucleic acid; fraction 6; protein; and the final pellet nonextractable residue presumably made up of indigestible cellulose, hemicellulose, and lignin fractions (see Supporting Information).
TABLE 3. Radioactivity bound to neutral detergent fiber (NDF) and acid detergent fiber (ADF) fractions of fecal mattera wether 368
wether 370
time
NDF
ADF
NDF
ADF
12–18 18–24 24–32 32–40 40–48 average:
51.2 ( 1.5 47.3 ( 1.5 47.9 ( 2.4 48.4 ( 1.6 50.1 ( 0.4 49.0 ( 1.6
22.3 ( 0.7 20.8 ( 0.6 21.4 ( 0.7 22.5 ( 1.1 24.3 ( 0.5 22.3 ( 1.3
53.1 ( 2.1 52.0 ( 0.7 50.8 ( 2.8 53.0 ( 0.7 54.4 ( 0.9 52.7 ( 1.3
23.0 ( 0.3 22.7 ( 1.1 23.4 ( 0.2 23.8 ( 0.2 25.5 ( 0.3 23.7 ( 1.1
a Data are expressed as the means ( std dev. of the percentage of TRR present in fecal material that remained bound to the fiber component after digestion (n ) 3 determinations per time point).
conditions (45), would explain the formation of bound residues in ruminal fluid and fecal matter. The nature of the bound residues excreted in feces was probed using the method of Carpenter et al. (21) in which fecal radioactivity was partitioned into aqueous, soluble carbohydrate, lipid, nucleic acid, and protein fractions (Table 2). Major portions of the fecal radioactivity were associated with the lipids and proteins (about 16% of the total fecal radioactivity associated with each) with approximately 50% of the fecal radioactivity remaining in the pellet. Similar results were obtained using activated sludge (21) with 50% of the starting radioactivity remaining in the pellet, but with approximately 6% of the radioactivity in the protein fraction and 25% in the lipid fraction. In our study, approximately 5% of the fecal radioactivity was extracted into a water rinse (supernate fractions 1 and 2). This radioactivity was very likely associated with water soluble organic compounds and not with free TNT or its free metabolites because previous efforts to extract free metabolites were unsuccessful. Radioactivity in the nucleic acid fraction was high (10% of total radioactivity) relative to that of Carpenter et al. (∼1.5%, ref 21) which may be a function of matrix differences. Nonsoluble carbohydrates are a major component of feces in ruminant animals, therefore the method of Van Soest (18) was used to conduct a crude fractionation of the fecal material (Table 3). Digestion in NDF solution removes soluble proteins, soluble nitrogenous compounds, soluble carbohydrates (sugars), starch, soluble lipids, and in fecal samples, a poorly defined “microbial metabolic component” (18). The radioactivity remaining in NDF insoluble material was associated with cellulose, hemicellulose, and lignin (18). Digestion of fecal material in ADF solution solubilized hemicellulose in addition to soluble materials removed by the NDF procedure.
Combustion analyses of the NDF residue indicated that an average of 50.9% of the radioactivity was collectively bound to cellulose, hemicellulose, and lignin within feces (Table 3), components of feces not associated with the microbial portion of feces. By difference, 49.1% of the fecal radioactivity was associated with NDF-soluble materials. Presumably, the soluble components from the NDF analysis were mostly associated with the microbial mass since the nutritive components of the feed (i.e., digestible carbohydrates and proteins) had been removed by digestion. Further analysis of the fiber component of the feces by digestion in acid detergent indicated that approximately 23% of the fecal radioactivity was bound to the cellulose and lignin portion (ADF fraction) of the plant cell wall (Table 3). By difference, 27.3% of the total radioactivity in the feces was bound to hemicellulose-like material. Our results are consistent with previous work in which [14C]TNT reduction products were covalently associated with lignocellulose, humin, humic acid, and fulvic acid fractions of compost (24, 48). All of the major reduction products of TNT are able to bind to humin, the major chemical fraction of compost (44). In addition, TNT reduction products also covalently bind to purified coniferyl alcohols (a precursor to lignocellulose), humic acid, and to sawdust (44). It was suggested that, in the absence of oxygen, “numerous pathways” exist for the condensation of aromatic amines with carbonyl groups present in organic matter. Evidence for the binding of multiple amine of di and triaminotoluenes to organic matter was established via 15N NMR (44). From a strictly bioremediation perspective, the consumption of TNT by ruminants allows its conversion to forms that are not readily available and might, therefore, represent a viable means of TNT remediation (14). Caveats must be added to this perspective, however. First, use of ruminants would likely be prudent only in pastures with low levels of TNT contamination because of potential TNT toxicity problems. Presumably a threshold exists in which the absorption rate of parent TNT and its reduction products would exceed the rate of ruminal TNT reduction and binding. Once this threshold is met, TNT might cause toxicity, rendering the animal moot as a bioremediation tool. Second, the current study was performed using a low TNT dose. At elevated doses, covalent binding of TNT reduction products might not be as efficient as measured in this study. Absorption of amino- or diaminonitrotoluenes with subsequent excretion of nonbound products into the environment would not be acceptable for bioremediation purposes. In addition, the use of ruminant animals for bioremediation purposes would preclude their use as food animals unless residue depletion studies established appropriate depuration periods. Even with VOL. 42, NO. 7, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2567
these considerations, the apparent large capacity of the rumen to biotransform TNT, the apparent speed at which covalent binding occurs, and the low cost of ruminating animals relative to the costs of existing remediation technologies justifies further research. Such is especially true when one considers possible interactions between bound residues within fecal matter deposited on pastures and the stimulatory effect that vegetation had on the mineralization of bound TNT residues (49).
Acknowledgments We are grateful to Dee Ellig, Michael Giddings, Jason Holthusen, and Chris Thomas for their excellent technical assistance. The use of trade, firm or corporation names in this publication is for the information and convenience of the reader. Such use does not constitute an official endorsement or approval by the United States Department of Agriculture or the Agricultural Research Service of any product or service to the exclusion of others that may be suitable.
Supporting Information Available Details on analytical methods, chromatographic distribution of radioactivity in urine extracts, and kinetics of total radioactive residues in serum of dosed animals. This material is available free of charge via the Internet at http:// pubs.acs.org.
Literature Cited (1) Lewis, T. A.; Newcombe, D. A.; Crawford, R. L. Bioremediation of soils contaminated with explosives. J. Environ. Microbiol. 2004, 70, 291–307. (2) Bolt, H. M.; Degen, G. H.; Dorn, S. B.; Plöttner, S.; Harth, V. Genotoxicity and potentially carcinogenicity of 2,4,6-TNT trinitrotoluene: structural and toxicological considerations. Rev. Environ. Health 2006, 21, 217–228. (3) Rodgers, J. D.; Bunce, N. J. Treatment methods for the remediation of nitroaromatic explosives. Water Res. 2001, 35, 2101–2111. (4) Ramos, J. L.; González-Pérez, M. M.; Caballero, A.; van Dillewijn, P. Bioremediation of polynitrated aromatic compounds: plants and microbes put up a fight. Curr. Opin. Biotechnol. 2005, 16, 275–281. (5) Esteve-Núñez, A.; Caballero, A.; Ramos, J. L. Biological degradation of 2,4,6-trinitrotoluene. Microbiol. Mol. Bio. Rev. 2001, 65, 335–352. (6) Fleischmann, T. J.; Walker, K. C.; Spain, J. C.; Hughes, J. B.; Craig, A. M. Anaerobic transformation of 2,4,6-TNT by bovine ruminal microbes. Biochem. Biophys. Res. Commun. 2004, 314, 957–963. (7) Liu, Y.-Y.; Lu, A. Y. H.; Stearns, R. A.; Chiu, S.-H. L. In vivo covalent binding of [14C]trinitrotoluene to proteins in the rat. Chem.-Biol. Interact. 1992, 82, 1–19. (8) Leung, K. H.; Yao, M.; Stearns, R.; Chiu, S.-H. L. Mechanism of bioactivation and covalent binding of 2,4,6-trinitrotoluene. Chem.-Biol. Interact. 1995, 97, 37–51. (9) Palmer, W. G.; Beaman, J. R.; Walters, D. M.; Creasia, D. A. Bioavailability of TNT residues in composts of TNT-contaminated soil. J. Toxicol. Environ. Health 1997, 51, 97–108. (10) Murray III, A.; Williams, L. D. Organic Syntheses with Isotopes, Part I: Compounds of Isotopic Carbon; Interscience Publishers, Inc.: New York; 1958; pp 893–895. (11) Dennis, W. H.; Rosenblatt, D. H.; Blucher, W. G.; Coon, C. L. Improved synthesis of TNT isomers. J. Chem. Eng. Data 1974, 20, 202–203. (12) Adamia, G.; Ghoghoberidze, M.; Graves, D.; Khatisashvili, G.; Kvesitadze, G.; Lomidze, E.; Ugrekhelidze, D. Zaalishvili. Absorption, distribution, and transformation of TNT in higher plants. Ecotoxicol. Environ. Saf. 2006, 64, 136–145. (13) Robbins, J. D.; Bakke, J. E. Sheep and goat metabolism unit for the collection of excreta and expired 14CO2. J. Anim. Sci. 1967, 26, 424–429. (14) Heiss, G.; Knackmuss, H.-J. Bioelimination of trinitroaromatic compounds: immobilization versus mineralization. Curr. Opin. Microbiol. 2002, 5, 282–287. (15) Benville, P. E.; Tindle, R. C. Dry ice homogenization procedure for fish samples in pesticide residue analysis. J. Agric. Food. Chem. 1970, 18, 948–949. 2568
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 7, 2008
(16) Smith D. J.; Stehly, G. S.; Turberg, M. T. Procedures and Practices for Conducting Residue Studies of Animal Health Compounds in Food Animals. In Handbook of Residue Analytical Methods for Agrochemicals; Lee, P. Ed.; John Wiley & Sons, Ltd.: West Sussex, England, 2003. (17) Yinon, J.; Hwang, D.-G. Metabolic studies of explosives. 5. Detection and analysis of 2,4,6,-trinitrotoluene and its metabolites in urine of munition workers by micro liquid chromatography/mass spectrometry. Biomed. Chromatogr. 1986, 1, 123–125. (18) Van Soest, P. J.; Robertson, J. B.; Lewis, B. A. Carbohydrate methodology, metabolism, and nutritional implications in dairy cattle. J. Dairy Sci. 1991, 74, 3583–3597. (19) Van Soest, P. J. Nutritional Ecology of the Ruminant, 2nd ed.; Cornell University Press: Ithaca, NY, 1994. (20) Sutherland, I. W.; Wilkinson, J. F. In: Methods in Microbiology; Norris, J. R., Ribbons, D. W., Eds.; Academic Press: London, 1971. (21) Carpenter, D. F.; McCormick, N. G.; Cornell, J. H.; Kaplan, A. M. Microbial transformation of 14C-labeled 2,4,6-trinitrotoluene in an activated-sludge system. Appl. Environ. Microbiol. 1978, 35, 949–954. (22) Hawari, J.; Halasz, A.; Paquet, L.; Zhou, E.; Spencer, B.; Ampleman, G.; Thiboutot, S. Characterization of metabolites in the biotransformation of 2,4,6-trinitrotoluene with anaerobic sludge: role of triaminotoluene. Appl. Environ. Mirobiol. 1998, 64, 2200–2206. (23) Funk, S. B.; Roberts, D. J.; Crawford, D. L.; Crawford, R. L. Initialphase optimization for bioremediation of munition compoundcontaminated sites. Appl. Environ. Microbiol. 1993, 59, 2171– 2177. (24) Pennington, J. C.; Hayes, C. A.; Myers, K. F.; Ochman, M.; Gunnison, D.; Felt, D. R.; McCormick, E. F. Fate of 2,4,6trinitrotoluene in a simulated compost system. Chemosphere 1995, 30, 429–438. (25) Achtnich, C.; Sieglen, U.; Knackmuss, H.-J.; Lenke, H. Irreversible binding of biologically reduced 2,4,6-trinitrotoluene to soil. Environ. Toxicol. Chem. 1999, 18, 2416–2423. (26) Kröger, M.; Schumacher, M. E.; Risse, H.; Fels, G. Biological reduction of TNT as part of a combined biological-chemcial procedure for mineralization. Biodegradation 2004, 15, 241– 248. (27) Newcombe, D. A.; Crawford, R. L. Transformation and fate of 2,4,6-trinitrotoluene (TNT) in anaerobic bioslurry reactors under various aeration schemes: implications for the decontamination of soils. Biodegradation [Online early access]. DOI 10.1007/ s10532–007–9103–0. Published Online: February 2, 2007. (28) Reddy, G.; Chandra, S. A. M.; Lish, J. W.; Qualls, C. W., Jr. Toxicity of 2, 4, 6-trinitrotoluene (TNT) in hispid cotton rats (Sigmondon hispidus): Hematological, biochemical, and pathological effects. Int. J. Toxicol. 2000, 19, 169–177. (29) Coombs, M.; Schillack, V. Determination of trinitrotoluene and its metabolites in urine by means of gas-chromatography with mass detection. Int. Arch. Occup. Environ. Health. 1998, 71 (1), S22–S25. (30) Sabbioni, G.; Liu, Y.-Y.; Yan, H.; Sepai, O. Hemoglobin adducts, urinary metabolites and health effects in 2,4,6-trinitrotoluene exposed workers. Carcinogenesis 2005, 26, 1272–1279. (31) Yinon, J.; Hwang, D.-G. Identification of urinary metabolites of 2,4,6,-trinitro-toluene in rats by liquid chromatography-mass spectrometry. Toxicol. Lett. 1985, 26, 205–209. (32) Lyford, S. J. In: The Ruminant Animal, Digestive Physiology and Nutrition; Church, D. C., Ed.; Prentice Hall: Englewood Cliffs, NJ, 1988. (33) Drzyzga, O.; Bruns-Nagel, D.; Gorontzy, T.; Blotevogel, K.; von Löw, E. Anaerobic incorporation of the radiolabeled explosive TNT and metabolites into the organic soil matrix of contaminated soil after different treatment procedures. Chemosphere 1999, 38, 2081–2095. (34) Caton, J. E.; Ho, C.-H.; Williams, R. T.; Griest, W. H. Characterization of insoluble fractions of TNT transformed by composting. J. Environ. Sci. Health 1994, A29, 659–670. (35) Elovitz, M. S.; Weber, E. J. Sediment-mediated reduction of 2,4,6trinitrotoluene and fate of the resulting aromatic (poly)amines. Environ. Sci. Technol. 1999, 33, 2617–2625. (36) Thiele, S.; Fernandes, E.; Bollag, J.-M. Enzymatic transformation and binding of labeled 2,4,6-trinitrotoluene to humic substances during anaerobic/aerobic incubations. J. Environ. Qual. 2002, 31, 437–444.
(37) Vanderford, M.; Shanks, J. V.; Hughes, J. B. Phytotransformation of trinitrotoluene (TNT) and distribution of metabolic products in Myriophyllum aquaticum. Biotechnol. Lett. 1997, 19, 277– 280. (38) Sens, C.; Scheidemann, P.; Klunk, A.; Werner, D. Distribution of 14C-TNT and derivatives in different biochemical compartments of Phaseolus vulgaris. Environ. Sci. Pollut. Res. 1998, 5, 202–208. (39) Bahdra, R.; Wayment, D. G.; Hughes, J. B.; Shanks, J. V. Confirmation of conjugation processes during TNT metabolism by axenic plant roots. Environ. Sci. Technol. 1999, 33, 446–452. (40) Vila, M.; Lorber-Pascal, S.; Laurent, F. Fate of RDX and TNT in agronomic plants. Environ. Pollut. 2007, 148, 148–154. (41) Thompson, P. L.; Ramer, L.; Schnoor, J. L. Uptake and transformation of TNT by hybrid poplar trees. Environ. Sci. Technol. 1998, 32, 975–980. (42) Claus, H.; Bausinger, T.; Lehmler, I.; Perret, N.; Fels, G.; Dehner, U.; Preuss, J.; König, H. Transformation of 2,4,6-trinitrotoluene (TNT) by Raoultella terrigena. Biodegradation 2007, 18, 27–35. (43) Preuβ, A.; Rieger, P.-G. Anaerobic transformation of 2,4,6trinitrotoluene and other aromatic compounds. In: Biodegradation of Nitroaromatic Compounds; Spain, J. C., Ed.; Plenum Press: New York, 1995.
(44) Thorn, K. A.; Kennedy, K. R. 15N NMR investigation of the covalent binding of reduced TNT amines to soil humic acid, model compounds, and lignocellulose. Environ. Sci. Technol. 2002, 36, 3787–3796. (45) Daun, G.; Lenke, H.; Reuss, M.; Knackmuss, H.-J. Biological treatment of TNT-contaminated soil. 1. Anaerobic cometabolic reduction and interaction of TNT and metabolites with soil components. Environ. Sci. Technol. 1998, 32, 1956–1963. (46) Claus, H.; Perret, N.; Bausinger, T.; Fels, G.; Preuβ, J.; König, H. TNT transformation products are affected by the growth conditions of Raoultella terrigena. Biotechnol. Lett. 2007, 29, 411–419. (47) Rieger, P.-G.; Knackmuss, H.-J. Basic Knowledge and Perspectives on Biodegradation of 2,4,6-Trinitrotoluene and Related Nitroaromatic Compounds in Contaminated Soil. In: Biodegradation of Nitroaromatic Compounds; Spain, J. C., Ed.; Plenum Press: New York, 1995. (48) Thorn, K. A.; Pennington, J. C.; Hayes, C. A. 15N NMR investigation of the reduction and binding of TNT in an aerobic bench scale reactor simulating windrow composting. Environ. Sci. Technol. 2002, 36, 3797–3805. (49) Achtnich, C.; Lenke, H. Stability of immobilized 2,4,6-trinitrotoluene metabolites in soil under long-term leaching conditions. Environ. Toxicol. Chem. 2001, 20, 280–283.
ES702601N
VOL. 42, NO. 7, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2569
