Abstract
Military activities associated with training, munitions manufacturing, and demilitarization has resulted in soil residues of munition compounds and their breakdown products. Two isomers of dinitrotoluene (2,4- and 2,6-) are often found in soil associated with those activities at considerable concentrations. Consequently, issues regarding the effects of exposure to birds that visit these habitats require evaluation. To provide data useful to a risk assessment approach, we followed a controlled dosing regime (gavage) using 2,4-dinitrotoluene (2,4-DNT) in the Northern Bobwhite (Colinus virginianus) for 60 days following a 14-day range-finding study and the determination of a LD50 using the up/down method. The LD50 was determined to be 55 mg/kg using corn oil as a vehicle. Individuals dosed exceeding this level were moribund or died within 60 h of exposure. Morbidity and death occurred during the 14-day range-finding study at dosing regimens of 35 and 55, but not at 15, 5, and 0.5 mg/kg-day. Compound-related morbidity/mortality occurred in the 60-day study during the first week of exposure at 25 and 15, but not at 5, 1, and 0 mg/kg-day. Overt signs of toxicity occurred with both sexes at the onset of exposure. Signs included weight loss, diarrhea, and lethargy. Dose-related changes in egg production, ovary, kidney, and brain mass, and body weight, but not feed consumption, were found. Changes in kidney mass and histological observations suggest accumulation of nitrogenous waste may be the cause of morbidity. These data suggest that oral 2,4-DNT exposures are more acutely toxic and has a different etiology than 2,4,6-trinitrotoluene in birds.
Keywords
Dinitrotoluenes (DNTs) are used in the production of polyurethane, manufacturing of dyes, and in specific explosive mixtures (Ellis et al. 1985; Brower et al. 1994; ATSDR 1998). Environmental sources of DNTs have been predominantly associated with trinitrotoluene (TNT) manufacturing and use (Brower et al. 1994) and have been identified in various media in over 122 hazardous waste sites in the United States (ATSDR 1998). Most of these are sites used and maintained by the Department of Defense, and environmental medium concentrations are attributed to operations associated with munitions manufacturing, use, and disposal.
Of the six isomers, 2,4-DNT is one of the most prevalent (Ellis et al. 1985). Soil residues of 2,4-DNT have been reported at numerous U.S. Army installations, with reported concentrations as high as 237 mg/kg (Walsh and Jenkins 1992; Jenkins et al. 2001). Because of restricted human access, large tracts of undisturbed land have become habitat for a diverse collection of fauna, especially birds. Many avian species intentionally ingest soil to assist in food digestion and therefore may be exposed to DNT (McCann 1961; Giofriddo and Wink et al. 1993; Best et al. 1992). To date, no data exist that have evaluated the effects of oral 2,4-DNT exposure in birds.
Effects of oral 2,4-DNT exposures have been investigated in mammals. Primary effects include the production of methemoglobin, anemia, peripheral neuropathy, jaundice (hepatotoxicity), tremors, and sensitization (Lee et al. 1975, 1978; Ellis et al. 1978; Hong et al. 1985; Lane et al. 1985). Reproductive effects (i.e., testicular and ovarian toxicity) have also been reported in rats (Bloch et al. 1988). Oral LD50values for 2,4-DNT range from 240 to 650 mg/kg in rats and 1340 to 1954 mg/kg in mice. Toxicity observed in both species included ataxia and cyanosis, with death occurring within the first 24 hours (Lee et al. 1975; Ellis et al. 1978; Lane et al. 1985).
Because 2,4-DNT contaminated soils exist at habitats used by birds, it is important that effects from exposure to these soils are evaluated. This work evaluated the toxicity of 2,4-DNT under controlled laboratory conditions to the Northern Bobwhite (Colinus virginianus). Northern Bobwhite were selected because they are widely distributed throughout eastern North America, exist across a wide variety of habitats, and because of their life histories, have the potential for significant exposure among native avian species (Brennan 1999).
MATERIALS AND METHODS
Animal Husbandry
Northern Bobwhite, approximately 12 weeks of age, were hatched from breeding stock at the Virginia-Maryland Regional College of Veterinary Medicine, Virginia Tech, Blacksburg, VA, USA. A subset of individuals were necropsied and evaluated to determine the health of the population. Following positive confirmation of health criteria, birds were shipped by commercial carrier to the U.S. Army Center for Health Promotion and Preventive Medicine for the conduct of the study. Upon arrival, birds were held individually in 24-cage units, each cage 11 inch wide × 12 inch tall × 15 inch deep constructed of ½ inch by 1 inch polyvinylchloride (PVC)-coated wire. Each unit contained two adjustable-height, automatic nipple drinkers and individual 18-gauge stainless steel feeding troughs. Each enclosure had a rubber kong for environmental enrichment. Birds were kept in constant photoperiod durations of 16-light: 8-dark with humidity maintained between 30% and 70% and the temperature between 64°F and 79°F.
Birds were acclimated/quarantined for greater than 14 days and determined to be in good health by the attending veterinarian before being released for use in the study. Birds were uniquely identified by cage card and subcutaneous transponder. Tap water was provided ad libitum by an automated watering system that was checked daily. Birds were provided with certified starter crumbs (Nutrena Premium Gamebird Feed; Cargill, Minneapolis, MN, USA) weekly, which was weighed along with any spillage.
Throughout the study, birds were monitored daily for changes in health disposition, and the onset of morbidity. Animals exhibiting signs of distress were humanely euthanized according to protocol.
Test Article
The test compound (2,4-DNT) was obtained from Sigma-Aldrich (St. Louis, MO, USA) and was determined to be 96% pure. The sample was dissolved in ethyl acetate and analyzed by gas chromatography using a flame ionization detector. The chromatography was done using a nonpolar column. A blind control yielded a recovery of 98%. DNT was mixed in a corn oil vehicle for the 14-day range finding and the 60-day subchronic studies at target concentrations of 7 mg/ml and 5 mg/ml, respectively. Concentrations of 2,4-DNT were confirmed analytically at target concentrations following a 2-week stability study. The samples were diluted with ethyl acetate and analyzed using a gas chromatograph fitted with one of three detectors, a mass spectrometer, an electron capture detector, or a flame ionization detector. The mass spectrometer and the electron capture detectors are very sensitive and were used so that the samples could be diluted sufficiently to minimize the effect of the corn oil on the instruments. Blind controls were analyzed with the samples and had an average recovery of 102%. Compound and vehicle solutions were prepared every two weeks for the 60-day exposures and administered volumetrically based upon the most recent body weight for each individual bird.
Acute Toxicity
Determination of the LD50 was conducted using the up/down (UDP) method (OECD 2001; ASTM 2002; USEPA 2002). Briefly, a single female was orally administered 2,4-DNT in corn oil via gavage at the expected LD50 (175 mg/kg). Following a 60-h observation period, another female was dosed either 0.5 log higher (1.3× dose) or lower depending on outcome, i.e., mortality or morbidity. This process was repeated until more than three reversals occurred and the specified likelihood ratios exceeded the critical value (LR criterion; USEPA 2002). The LD50 estimate is based on whether recovery occurs following a 14-day observation period. Compound was weighed specific to each target dose and mixed in 1 ml of corn oil immediately prior to administration.
Subacute Toxicity
Six treatment groups consisting of seven bobwhites of mixed sex were administered either 0, 0.5, 5, 15, 35, or 55 mg DNT/kg body weight (bw) daily for 14 days. Oral doses were determined volumetrically based on a 7 mg/ml solution. Controls received an equivalent volume of vehicle without DNT as the birds in the high-dose group. Birds were weighed on days −1, 0, 3, 7, and 14. Mean body weight of birds between treatment groups were not different (194.2 ± 2.2 g and 189.1 ± 3.6 g for males and females, respectively; p > .67). Following 14 days of exposure, birds were phlebotomized via the jugular vein, anesthetized/euthanized through CO2 asphyxiation, and evaluated through gross necropsy. Brain, liver, kidney, and gonads were weighed to the nearest 0.001 g and presented as a ratio to body weight.
Subchronic Toxicity
The 60-day oral study was conducted using male and female C. virginianus applying the same general methods described above. A total of 120 birds were randomly sorted to one of five groups containing 12 male and 12 female birds. Mean weights of birds within each group at the initiation of the study were not significantly different (3.6 g; 202.7 ± 1.7 g and 203.3 ± 2.0 g for males and females, respectively; p > .994). Birds in these groups were orally dosed with either corn oil or a 5 mg/ml corn oil solution of 2,4-DNT to either 1, 5, 15, or 25 mg/kg-day based on the most recent body mass measurement. All dosing occurred between the hours of 0800 and 1200 daily for 60 days. Birds were weighed on days −1, 0, 7, 14, and weekly thereafter. Daily dosing procedures including measuring feed and animal weights and recording any overt observations using the Labcat Software System (Innovative Programming Associates, Princeton, NJ, USA). Following 60 days of exposure, birds were bled and humanely euthanized as described previously. No plasma chemistry or blood parameters were collected from birds that died or were euthanized for humane considerations before scheduled study end.
Hematology
Whole blood was evaluated for enumeration of erythrocytes and leucocytes (red blood cells [RBCs] and white blood cells [WBCs], respectively), packed cell volume (PCV), total protein solids, and five-part leukocyte differential conducted using the same standard methods described in Gogal et al. (2002). Hemoglobin content (Hb) was determined from whole blood using a HemoCue Hemoglobin photometer (HemoCue AB, Angelholm, Sweden). Lithium heparin collection tubes were used for all blood samples.
Plasma Chemistry Analyses
Following collection of aliquots for hematology parameters, the remaining whole blood from each heparinized tube was centrifuged to obtain plasma for clinical chemistry analyses. A VetTest chemistry analyzer (IDEXX Laboratories, Westbrook, ME, USA) was then used to evaluate the following plasma parameters: Aspartate aminotransferase (AST), glucose (GLU), creatine kinase (CK), triglycerides (TRIG), uric acid (URIC), phosphorus (PHOS), albumin (ALB), globulin (GLOB), lactate dehyrogenase (LDH), calcium (Ca), and alkaline phosphatase (ALKP). Plasma electrolytes (e.g., sodium, chloride, potassium) were also analyzed using a VetLyte Electrolyte Analyzer (IDEXX Laboratories).
Histopathology
Tissues were collected, trimmed, fixed in formalin, and embedded in paraffin. These tissues were then sectioned at 6 microns, stained with hematoxylin and eosin, and examined via light microscopy. The pathologist did not know treatment group of each tissue until after the histopathologic examination was completed.
Statistical Analysis
Tests of the data for normality and equal variances for each group were conducted to satisfy the assumptions associated with parametric tests. If the data failed to fit a normal distribution or were of unequal variance distributions, they were either log transformed and reevaluated or ranked and analyzed using a two-way analysis of variance (ANOVA) on the ranks. Normally distributed data were analyzed using a two-way ANOVA using sex and treatment as variables. Change in body weights were evaluated separately for each sex using a one-way ANOVA. If significant, a pairwise multiple comparison procedure was conducted using the Holm-Sidak method. Deviations in these procedures are noted where applicable. Statistical significance was defined at the p < .05 level. To allow for a consistent characterization of the results, means, and standard errors of the mean are presented for all data. Tests were conducted using SigmaStat Ver. 3.0 (SPSS, Chicago, IL, USA).
This study was conducted consistent with the standards found in Title 40 Code of Federal Regulations (CFR), Part 792, Good Laboratory Practices. The investigators and technicians adhered to the following guidelines; the Public Health Service Policy on Humane Care and Use of Laboratory Animals, “U.S. Government Principles for the Utilization and Care of Vertebrate Animals Used in Testing, Research, and Training,” and the Animal Welfare Act.
RESULTS
Acute Toxicity
Both birds dosed at 175 mg/kg died, as did three out of four dosed at 55 mg/kg. All three dosed at 17.5 mg/kg survived. All deaths were largely unremarkable and occurred at < 72 h post dose. Initial observations included watery stools, lethargy, and a single case of excessive drinking. The partial response at 55 mg/kg and complete responses at the other exposures resulted in an estimated LD50 of 55 mg/kg with 95% confidence intervals of 19.95 to 78.6 mg/kg.
Subacute Toxicity
All birds in 25 and 35 mg/kg/d groups died within 72 h accompanied by marked weight loss (data not shown). Given the small sample size, data were pooled across gender and analyzed using an ANOVA. Kidney/bw ratios were higher for the 35 mg/kg-day groups than the corn oil controls (Figure 1). Liver/bw ratios were increased for birds in the 35 mg/kg-day groups. No differences were found in spleen, brain, testes, ovary/bw ratios between treatments (data not shown for ovary and testes).
Given the chronology of unscheduled deaths in the 35 and 55 mg/kg-day groups, no hematology data were collected from those individuals. No differences were found in the clinical chemistry data between treatments; however, trends in electrolyte and triglycerides levels corroborated with the mass/bw changes are suggestive of adverse kidney and liver effects.
Subchronic Toxicity
All females and 9/12 males died or were moribund in the high dose group, most within 1 week (Table 1). Three males and four females died or were moribund in the 15 mg/kg-day groups. Symptoms associated with the 15 and 25 mg/kg-day groups included lethargy, diarrhea, and weight loss. During the 5th week of exposure, two, one, and one females in the control, 1, and 5 mg/kg-day groups, respectively, developed signs consistent with a respiratory infection (dyspnea, mucus discharge around the nares) and were humanely euthanized. Given the quick onset and distribution of individuals between groups, this condition was not considered to be treatment related.
Mean change in body weights were lower for birds in the two high-dose treatments compared with the others (Figure 2). Females were gravid and gained weight over time in most treatments. Birds of both sexes in the 25 mg/kg-day group lost body mass, and it was reduced for males in the15 mg/kg-day. Females in the 15 mg/kg-day group gained less mass than those in the other treatments. Mean feed consumption/day was not different across treatments (p > .49; data not shown).
Females in the 25 mg/kg-day group laid fewer eggs/day than all other groups (Figure 3). This was consistent with ovary/bw relationships (Table 2). Testes/bw ratios were not affected by treatment (p > .47).
Brain, liver, and kidney/bw ratios were affected by treatment (Table 2). Brain/bw ratios were increased in the 15 and 25 mg/kg-day groups for both sexes. Female kidney/bw ratios were increased in the 5 mg/kg-day group, and for both sexes in the 15 and 25 mg/kg-day groups. Only male liver/bw was increased in the 15 and 25 mg/kg-day groups. There was a significant interaction between sex and treatment.
Triglyceride levels were higher in males from the 5 and 15 mg/kg-day groups compared with the control and 1 mg/kg-day groups (Table 3), consistent with increasing liver/bw ratios. There were no other differences in the plasma chemistry data that were attributable to treatment; however, sufficient variation in these parameters contributed to this finding. This was particularly evident in females, where no treatment related effects were found. Moreover, differences between males and females were evident for most parameters (Table 3). There was a general increasing tendency in plasma uric acid concentrations with dose, which was consistent with the kidney/bw data, suggestive of adverse kidney effects.
PCV, Hb, and RBC counts were all affected by treatment and gender (Table 4). Only females showed a difference in PCV yet lacked a dose response and were within published ranges for this species (Ritchie, Harrison, and Harrison 1994). RBC values were lower according to treatment for females, but not males of birds in the 15 mg/kg-day versus the control group. Hb concentrations were lower for females, but not males according to treatment (Table 4).
Acute inflammation of the kidneys, consisting of multifocal infiltrations of heterophils and lymphocytes, was present in 5/6 birds from the 25 mg/kg-day group and 1/1 male from the 15 mg/kg-day group that died or were sacrificed for humane considerations. Chronic inflammation, consisting of multifocal lymphocytic infiltration, was present in the kidneys of 1/3 of the unscheduled death females of the 15 mg/kg-day group. The kidneys of that female also had gout tophi. Based on the histologic severity of this observation, none of the lesions would have been considered to be life threatening. The kidneys of a small number of birds had unusual deposition of basophilic granules in the glomeruli. This material was histologically consistent with mineral deposition, though this was not confirmed with special staining procedures. This mineral deposition only affected a few number of glomeruli, thus would not be expected to have an adverse effect on renal function. The precise etiology of this condition was not apparent from the examination. However, the combination of histologic observations in the kidneys suggests a treatment-related effect, perhaps due to accentuation of a spontaneous disease process such as visceral nephritic accumulation of urates (gout).
Splenic hemosiderosis consisted of an increased amount of hemosiderin contained within the macrophages of the spleen. Changes of this type can indicate intravascular hemolysis or an increased rate of erythrocyte sequestration by the spleen. The incidence pattern was not directly attributable to treatment. The liver from a single male had mild focal necrosis. Given this observation was found in only one individual, it could not be attributable to treatment.
DISCUSSION
Physiologic differences between species in vertebrate classes can account for observed differences in toxicity. Effects reported in mammals from repetitive oral 2,4-DNT exposures include RBC lysis, reduction in Hb and PCV, methemoglobinemia, splenic hemosiderosis, testicular degeneration, and neuropathy (Ellis et al. 1985; Hong et al. 1985; Lee et al. 1985). Rats exposed to high levels of DNT in feed (>34 mg/kg-day) showed a reduction in feed consumption, were lethargic, lost weight, and died after continued exposure (Lee et al. 1985). Anemia was associated with dose in mice, rats, and dogs, along with a yellowish-orange coloration of the urine (chromaturia; Elllis et al. 1985; Hong et al. 1985; Lee et al. 1985). Hemosiderin-like pigment was reported in macrophages found in the spleen, liver, lung, kidney, and elsewhere in mammalian species receiving relatively significant exposures (Hong et al. 1985, Ellis et al. 1985). Neurotoxicity, manifested through demyelination of peripheral nerves and lesions of the cerebellum, was noted as the most prominent cause of deaths in dogs (Ellis et al. 1985). Overt symptoms consisted of hindquarter paralysis and ataxia.
We observed no evidence of ataxia or other neuromuscular effects in the present study. Although some hematological parameters were affected by dose, they were not outside normal ranges for this species. No evidence of chromaturia was evident in this study; however, incidence of diarrhea was positively associated with exposure.
Unscheduled mortality from repetitive exposures occurred earlier in C. virginianus than in other mammalian species (Ellis et al. 1985; Hong et al. 1985; Lee et al. 1985) and was associated with weight loss but not feed consumption. A feeding study that evaluated a similar compound, 2,4-dinitrophenol (DNP), found an increase in water consumption, decrease in body mass, and an increase in oxygen consumption associated with exposure in the same species (Dominguez et al. 1993). Although we recognize that there may be important toxicokinetics differences between DNP and DNT, similarities in these effects may help to describe the mechanistic basis for these observed effects.
Studies investigating the metabolism of 2,4-DNT in mammals have described 2,4-dinitrobenzyl alcohol as the primary metabolite (Bond and Rickert 1981; Medinsky and Dent 1983). Other studies have provided evidence supporting the importance of anaerobic bacterial transformation and enterohepatic circulation to toxicity and metabolism (Medinsky and Dent 1983). Proton donation coefficients (pKa) between these compounds are likely different, therefore it is unknown if toxic mechanisms are similar. However, the etiology of 2,4-DNT toxicity is markedly different from that of 2,4,6-TNT (Gogal et al. 2002; Johnson et al. 2005), suggesting that pathways and metabolites different from those of TNT are responsible for toxicity. Incidences of diarrhea, reduction of body mass, and anecdotal observations of increased water consumption were consistent with those described in the DNP study (Dominguez et al. 1993). Changes in kidney/bw ratios occurred in females at 5 and males in the 15 mg/kg-day groups, respectively. Treatment-related histopathologic changes in the kidney and variable increases in plasma uric acid concentrations are consistent with the finding of nephritic accumulation of urates (visceral gout). Currently, there are inadequate data to determine whether this condition is a direct result of increased systemic nitrogen concentrations from compound exposure or a consequence of inefficient urate elimination resulting from impaired kidney function. This condition, however, is not uncommon in avian species and can produce symptoms of emaciation, lethargy, depression, and death (Shivaprasad 1998).
Dose-related changes in liver mass and plasma triglyceride levels are suggestive of adverse hepatic effects. However, these results were not consistently corroborated with adverse histopathologic observations. Triglyceride levels in males of the 5 mg/kg-day group were higher than controls, though 10-fold less than egg-laying females. Though statistically significant, the lack of other corroborating evidence at this exposure level suggests the biological importance of this finding is uncertain; however, it is suggestive of early adverse hepatic effects. Changes is brain mass occurred in both sexes at levels where mortality occurred (15 mg/kg-day). Because no specific pathological changes were found upon histological examination, the importance of this measure requires further investigation.
In summary, DNT caused dose-related changes in bw, changes in brain and kidney/bw ratios, histological nephritic changes, and mortality in males and females exposed to repetitive oral exposures at and greater than 15 mg/kg-day. Females had reduced ovary/bw ratios and egg production at 25 mg/kg-day. Males experienced a mild but statistically significant increase in plasma triglyceride levels at exposures of and greater than 5 mg/kg-day, and liver/bw changes at exposures at and exceeding 15 mg/kg-day. Females exposed to ≥5 mg/kg-day had a greater kidney/bw ratio. Treatment-related mortality occurred in both sexes of the 15 and 25 mg/kg-day groups. Therefore, the lowest adverse effect level (LOAEL) is 5 mg/kg-day based on early indicators of kidney effects in females and liver effects in males. The no observed adverse effect level (NOAEL) was determined to be 1 mg/kg-day based on the absence of indications suggestive of adverse effects.
Footnotes
Figures and Tables
Acknowledgements
The authors thank Robert McKenzie for analytical support, Patricia Beall for laboratory coordination, George Parker for the histopathology, Maj. Joseph Novak for necropsy support, and Maj. Stephen Dalal for veterinary care. This work was funded by the U.S. Army Environmental Quality Technology Program through the U.S. Army Corps of Engineers, Engineering Research and Development Center (ERDC).
Disclaimer: The views expressed in this paper are those of the authors and do not necessarily reflect the views and policies of the U.S. Army. Mention of trade names or commercial products does not constitute endorsement or recommendation for use.