2.2.1.1. Death

Historically, the greatest number of deaths among munitions workers were caused by adverse effects of 2,4,6-trinitrotoluene on the liver. Initial clinical symptoms included nausea, vomiting, pain in the abdomen, fatigue, dizziness, petechiae, and jaundice. Exposure to 2,4,6-trinitrotoluene eventually led to 475 deaths in the United States during World War I (McConnell and Flinn 1946). It is important to note, however, that the route of exposure was probably not exclusively inhalation, but also dermal.

No studies were located regarding death in animals after inhalation exposure to 2,4,6-trinitrotoluene.

2.2.1.2. Systemic Effects

No studies were located regarding cardiovascular, gastrointestinal, musculoskeletal, or renal effects in humans or animals after inhalation exposure to 2,4,6-trinitrotoluene.

Respiratory Effects. Extremely limited information is available regarding respiratory effects in humans after inhalation exposure to 2,4,6-trinitrotoluene. One study of occupational exposure (Morton et al. 1976) reported several cases of respiratory difficulties in ammunition plant workers who were exposed to 2,4,6-trinitrotoluene in the air at a level that was well above the current TLV of 0.5 mg/m³ (ACGIH 1993). However, there are several serious limitations to this study. The report does not state the exact air concentration of 2,4,6-trinitrotoluene, the duration of exposure, or the number of exposed workers with respiratory difficulties. It also does not specify the nature of those difficulties. The air concentration of 2,4,6-trinitrotoluene in the same plant was brought down to 0.3 mg/m³ within a month, but there is no information on whether the respiratory difficulties disappeared or persisted in the affected workers (Morton et al. 1976).

No studies were located regarding respiratory effects in animals after inhalation exposure to 2,4,6-trinitrotoluene.

Hematological Effects. In England during World War I, there were numerous cases of anemia and some reports of fatal aplastic anemia among workers using 2,4,6-trinitrotoluene in the production of explosives (Hathaway 1985). Similar experiences were seen in other countries involved in the war. However, with the improvement of protective measures used during the manufacturing process, the number of cases of 2,4,6-trinitrotoluene toxicity has decreased dramatically. The results of 2,4,6-trinitrotoluene exposure on hemoglobin, hematocrit, and reticulocyte numbers are well documented. A dose-response relationship between 2,4,6-trinitrotoluene exposure and effects on the hematologic system was found in 626 workers exposed to 2,4,6-trinitrotoluene when they were compared with 865 nonexposed controls. Tests were taken over a 6-week period. However, the actual duration of the workers’ exposures was not specified (Army 1976). The 2,4,6-trinitrotoluene estimated mean exposure levels ranged from <0.01 to 1.49 mg/m³; this range includes exposures higher than the present TLV of 0.5 mg/m³ (ACGIH 1993). Dose-related reductions in hemoglobin (9.9% lower than control) and hematocrit (11.6% lower that controls) and 50% higher reticulocyte counts were noted in exposed workers (Army 1976).

No abnormal values for hemoglobin were found in 43 workers employed in the manufacture of 2,4,6-trinitrotoluene who were monitored over a period of 5 months (Morton et al. 1976). During that time the air levels of 2,4,6-trinitrotoluene rose from 0.3 to 0.8 mg/m³. This finding is indirectly supported by another occupational exposure study. Activities of two mitochondrial enzymes, δaminolevulinic acid synthase and heme synthase, were measured in reticulocytes in a chronic occupational exposure study of workers who developed cataracts (Savolainen et al. 1985). Mean 2,4,6-trinitrotoluene concentrations were 0-35 mg/m³. Although the levels of the two enzymes were lower in the exposed workers, none of them had clinical anemia. This finding indicates that 2,4,6-trinitrotoluene may not affect hemoglobin synthesis in the bone marrow and that possible effects on reticulocytes may occur in the circulation after the oxygenation of cells in the lungs has taken place.

Hyperplasia of the bone marrow is the first reaction of the hematopoietic system to 2,4,6-trinitrotoluene poisoning. The shaft of the femur and the ribs are usually filled with active red marrow. If the exposure to 2,4,6-trinitrotoluene continues, the bone marrow becomes hypocellular (Army 1978a).

No studies were located regarding hematological effects in animals after inhalation exposure to 2,4,6-trinitrotoluene.

Hepatic Effects. Toxic hepatitis has been the principal manifestation of 2,4,6-trinitrotoluene toxicity in humans (Army 1978a). During World War I, 2,4,6-trinitrotoluene production increased and many cases of toxic hepatitis were fatal (Army 1978a). Industrial hygiene techniques improved by World War II; consequently, both the number of toxic hepatitis cases and the number of fatalities due to 2,4,6-trinitrotoluene exposure decreased dramatically (Army 1978a).

A statistically significant increase in hepatic enzymes (serum glutamic-oxaloacetic transaminase [SGOT] and lactic dehydrogenase [LDH]) was noted in ammunition plant workers when the 2,4,6-trinitrotoluene level in the air increased from 0.3 to 0.8 mg/m³ (during a 4-month period) at the same time that 2,4,6-trinitrotoluene production increased from 80% to 100% (Morton et al. 1976). Both these 2,4,6-trinitrotoluene air levels were close to the TLV of 0.5 mg/m³ (ACGIH 1993). The SGOT increase was 20% above the maximal normal value (Morton et al. 1976). In the same group of workers, the LDH values increased from about 51 units to over 106, but since isoenzyme studies were not performed, it is difficult to say if this increase was due to liver toxicity or to hemolysis (hemoglobin levels decreased only slightly-see Hematological Effects above and Morten et al. 1976).

No significant differences in liver function were noted in a cross-sectional epidemiology study of 626 munitions workers from four plants exposed to 2,4,6-ttinitrotoluene when compared to 865 nonexposed controls (Army 1976). The majority of the workers were exposed to 0.5 mg/m³ or less of 2,4,6-trinitrotoluene, and a few were exposed to 1.5 mg/m³. No changes were noted in the other liver parameters evaluated in the study: LDH, bilirubin (total and direct), alkaline phosphatase, SGOT, and serum glutamic-pyruvic transaminase (SGPT) (Army 1976). One possible explanation for these findings is that exposure to 2,4,6-trinitrotoluene causes more liver toxicity in potentially susceptible workers, and that in some cases of long-term exposure, liver cells may adapt to moderate exposure levels (Hathaway 1985). A case-control study of Chinese workers exposed to 2,4,6-trinitrotoluene indicates that the likelihood of liver injury is increased among those workers who are heavy drinkers as compared to workers who are not heavy drinkers (Li et al. 1991). However, the parameters measured to arrive at the diagnosis of liver damage in these cases are not discussed.

No studies were located regarding hepatic effects in animals after inhalation exposure to 2,4,6-trinitrotoluene.

Dermal Effects. In an occupational exposure study, several of the workers handling 2,4,6-trinitrotoluene in an ammunition plant complained of dermatitis (Morton et al. 1976). Inhalation seemed to be the major route of exposure, although dermal exposure was possible. Therefore, the dermatitis cannot be attributed to either a local or a systemic effect with certainty. In addition, the study did not report the precise number of affected workers or the duration and level of exposure.

No studies were located regarding dermal effects in animals after inhalation exposure to 2,4,6-trinitrotoluene.

Ocular Effects. The appearance of cataracts is believed to be an effect of 2,4,6-trinitrotoluene exposure (Hathaway 1985; Savolainen et al. 1985) and is often associated with chronic exposures. Irreversible equatorial lens opacities/cataracts were reported in 6 out of 12 Finnish workers exposed to 2,4,6-trinitrotoluene for an average of 6.8 years (2.1-11.5 years of exposure) (Harkonen et al. 1983). The principal routes of exposure were probably inhalation and dermal, although this is not clearly indicated in the report. Therefore, it is uncertain whether cataracts are a local or a systemic effect of 2,4,6-trinitrotoluene exposure. The opacities were detectable only on the periphery of the lens and appeared either continuous or discontinuous. The opacities of the lens were bilateral and symmetrical and did not affect visual fields or visual acuity. The workroom 2,4,6-trinitrotoluene air concentration was about 0.3 mg/m³ with a range of 0.14-0.58 mg/m³ (Harkonen et al. 1983). The progression of the cataract stops if the exposure to 2,4,6-trinitrotoluene stops. The mechanism of 2,4,6-trinitrotoluene cataract formation is not understood, but the possibility was raised that free radicals may play a role (Harkonen et al. 1983).

No studies were located regarding ocular effects in animals after inhalation exposure to 2,4,6-trinitrotoluene.

2.2.1.3. Immunological and Lymphoreticular Effects

No studies were located regarding immunological or lymphoreticular effects in humans or animals after inhalation exposure to 2,4,6-trinitrotoluene.

2.2.1.4. Neurological Effects

Very limited information is available regarding neurological effects in humans following inhalation exposure to 2,4,6-trinitrotoluene. Several workers who handled 2,4,6-trinitrotoluene in an ammunition plant reported altered taste, but no quantitative data are provided in the study (Morton et al. 1976). The concentration of 2,4,6-trinitrotoluene in the air was 0.3 mg/m³, which was below the TLV of 0.5 mg/m³ (ACGIH 1993). However, no details were provided on the exposure time or symptoms, so it is difficult to estimate the extent of the effect.

No studies were located regarding neurological effects in animals following inhalation exposure to 2,4,6-trinitrotoluene.

2.2.1.5. Reproductive Effects

A case-control study in two 2,4,6-trinitrotoluene plants in China indicated that 50 of the 104 workers that were examined for possible effects on semen had significantly lower semen volumes and a smaller percentage of motile spermatozoa, as well as a significantly higher incidence of sperm malformation, than the 33 controls (Li et al. 1993). Controls were clerks matched to the workers by income and by the city where they lived. However, exposure to 2,4,6-trinitrotoluene was not estimated. The only data available were annual measurements of air concentrations, so dose and effects cannot be correlated. Furthermore, confounding variables (beyond smoking and drinking) were not discussed. Possible important variables would include simultaneous exposures to other chemicals and heat in the workplace.

No studies were located regarding reproductive effects in animals after inhalation exposure to 2,4,6-trinitrotoluene.

2.2.1.6. Developmental Effects

No studies were located regarding developmental effects in humans or animals after inhalation exposure to 2,4,6-trinitrotoluene.

2.2.1.7. Genotoxic Effects

The presence of mutagenic compounds in the urine of workers exposed to 2,4,6-trinitrotoluene was confirmed in two investigations (Ahlborg et al. 1985, 1988a). The initial study involved the screening of urine samples from 97 workers in a chemical plant producing pharmaceuticals and explosives (Ahlborg et al. 1985). Included in the study was a group of 14 individuals exposed to a maximum air concentration of 0.29 mg/m³ 2,4,6-trinitrotoluene. Urine samples were collected at the conclusion of a work shift, concentrated on XAD-2 resin, and assessed for mutagenic activity using Salmonella typhimurium TA98 and Escherichia coli WP2 uvrA in the presence or absence of exogenous metabolic activation. Baseline data for each participant were established from samples collected following a 4-week vacation. The excretion of genotoxic agents was indicated by a significant increase (p<0.01) in mutant colonies of strain TA98 without metabolic activation in the urine of workers exposed to 2,4,6-trinitrotoluene; baseline samples for this group were uniformly negative. The exclusion of smokers from the 2,4,6-trinitrotoluene exposure group did not alter these findings. Unmetabolized 2,4,6-trinitrotoluene was detected in the urine of workers with the highest level of mutagenic activity. The findings of this study are consistent with the demonstrated mutagenic activity of 2,4,6-trinitrotoluene in S. typhimurium TA98 without S9 activation (see Section 2.4).

In the follow-up study, urine samples from 50 individuals exposed to varying concentrations of 2,4,6-trinitrotoluene in the workplace were evaluated (Ahlborg et al. 1988a). Subjects were divided into three groups: no exposure (2,4,6-trinitrotoluene air concentrations were too low to be detected), mid-range exposure (0.1-0.3 mg/m³), and high-range exposure (0.2-0.5 mg/m³). For each individual, pre- and postexposure samples were collected, and health status data relative to smoking habits, alcohol consumption, diet, and medication were obtained. Samples were concentrated and assessed for mutagenic activity in S. typhimurium strain TA98 and a derivative of TA98 deficient in nitroreductase activity (TA98NR); the assays were conducted without exogenous metabolic activation. The concentrations of 2,4,6-trinitrotoluene and two major metabolites (4-aminodinitrotoluene [4-ADNT] and 2-aminodinitrotoluene [2-ADNT]) were also determined. In agreement with the earlier findings, evidence of mutagenic activity was present in the urine of groups exposed to 2,4,6-trinitrotoluene, but significant genotoxicity was confined to urine from individuals in the high exposure group. Strain TA98, rather than TA98NR, is the most sensitive indicator of induced gene mutations because of endogenous nitroreductase activity. The finding suggests that bacterial nitroreductase activity is the most probable primary cause of 2,4,6-trinitrotoluene-induced gene mutations. Although the relevancy of this finding to humans is not known, comparable nitroreductase activity may be present via intestinal microflora or mammalian cells. In contrast to the results of the earlier study (Ahlborg et al. 1985) in which detectable levels of 2,4,6-trinitrotoluene were found in the urine samples that exhibited the highest level of mutagenicity, no correlation between 2,4,6-trinitrotoluene concentration and mutagenesis was seen. However, a weak correlation was seen between the concentration of the major metabolite (4-ADNT) and mutagenesis. The study authors concluded that the wide variation in individual urine sample mutagenicity data in conjunction with toxicokinetics and individual rates of 2,4,6-trinitrotoluene uptake probably accounts for the lack of a correlation. Similarly, calculation of the uptake of 2,4,6-trinitrotoluene through inhalation based on air concentration provided much lower estimates than expected from the urine concentration of the metabolites. This finding indicates that dermal absorption may contribute significantly to total uptake.

No studies were located regarding genotoxic effects in animals after inhalation exposure to 2,4,6-trinitrotoluene.

Other genotoxicity studies are discussed in Section 2.4.

2.2.1.8. Cancer

In a case of chronic occupational exposure to 2,4,6-trinitrotoluene, a 61-year-old male died of hepatocellular carcinoma (Garfinkel et al. 1988). Although he was exposed to 2,4,6-trinitrotoluene for 39 years as an ammunition plant worker, it is not known if 2,4,6-trinitrotoluene had a promoting role or any role in the development of the primary liver carcinoma. No discussion of exposure routes was included. In general, inhalation is assumed to be the primary pathway for worker exposure, but dermal contact and incidental ingestion due to hand-mouth contact cannot be ruled out.

No studies were located regarding cancer in animals after inhalation exposure to 2,4,6-trinitrotoluene.

2.2.2.1. Death

No studies were located regarding death in humans after oral exposure to 2,4,6-trinitrotoluene. However, during World Wars I and II, many fatal cases of toxic jaundice and aplastic anemia occurred. The fatalities were attributed to 2,4,6-trinitrotoluene exposure during the manufacturing of munitions (Army 1978a; McConnell and Flinn 1946).

The concentrations at which 2,4,6-trinitrotoluene is acutely lethal in animals vary depending on the species and sex. Reported oral LD₅₀ values for 2,4,6-trinitrotoluene are 1,010 and 1,320 mg/kg/day for male rats, and 795 and 820 mg/kg/day for female rats (Army 1978b; Dilley et al. 1982b). Acute oral LD₅₀ values in male and female mice are 1,012 and 660 mg/kg/day, respectively. Doses were administered by gavage with oil as a vehicle (Army 1978b; Dilley et al. 1982b). The animals developed tremors, followed by mild convulsions, l-2 hours after exposure. In some animals, death occurred within 4 hours following the exposure. The animals that survived the convulsions were still alive 14 days after exposure (Dilley et al. 1982b).

2,4,6-Trinitrotoluene was found to be lethal to beagle dogs receiving 32 mg/kg/day orally (by capsule) for 26 weeks. One female dog died during week 16 after exhibiting considerable weight loss, diarrhea, and ataxia. A second female dog was sacrificed in a moribund state during week 14 of the study. The second female dog was observed to be dehydrated and emaciated, with low body temperature and signs of an advanced icteric state. No deaths occurred in male dogs (Levine et al. 1990b).

No deaths were observed in Fisher-344 rats fed 125 mg/kg/day of 2,4,6-trinitrotoluene for 13 weeks (Levine et al. 1990a). Similar observations were made in the chronic exposure studies. No changes in survival rates were seen in the same breed of rats that were fed 50 mg/kg/day of 2,4,6-trinitrotoluene for 24 months (Army 1984a).

The LD₅₀ values and all reliable LOAEL values for death in each species and duration category are recorded in Table 2-1 and plotted in Figure 2-1.

2.2.2.2. Systemic Effects

No studies were located regarding respiratory, musculoskeletal, dermal, or ocular effects in humans or animals after oral exposure to 2,4,6-trinitrotoluene.

The highest NOAEL values and all reliable LOAEL values for each study and for each end point are recorded in Table 2-1 and plotted in Figure 2-1.

TABLE 2-1

Levels of Significant Exposure to 2,4,6-Trinitrotoluene - Oral.

Figure 2-1

Levels of Significant Exposure to 2,4,6-Trinitotoluene – Oral.

Cardiovascular Effects. No studies were located regarding cardiovascular effects in humans after oral exposure to 2,4,6-trinitrotoluene.

Very limi.ted information is available regarding cardiovascular effects in animals after oral exposure to 2,4,6-trinitrotoluene. Intermediate exposure of beagle dogs to a 2,4,6-trinitrotoluene dose as high as 32 mg/kg/day for 26 weeks did not cause any changes in electrocardiogram results or heart rates (Levine et al. 1990b).

Gastrointestinal Effects. No studies were located regarding gastrointestinal effects in, humans after oral exposure to 2,4,6-trinitrotoluene.

Adverse gastrointestinal effects were reported in dogs after intermediate oral exposure to 2,4,6-trinitrotoluene. Dogs receiving 20 mg/kg/day of 2,4,6-trinitrotoluene for 13 weeks had mucoid stools and diarrhea (Dilley et al. 1982b). In another longer intermediate exposure study, histopathology revealed inflammation of a part of the small intestine in beagle dogs fed 0.5, 2, 8, or 32 mg/kg/day 2,4,6-trinitrotoluene for 6 months. Although not dose related, the observed enteritis was more frequent in dogs treated with the highest dose of 2,4,6-trinitrotoluene. None of the control animals had enteritis (Levine et al. 1990b).

Hematological Effects. No studies were located regarding hematological effects in humans after oral exposure to 2,4,6-trinitrotoluene.

Anemia is one of the frequent signs of 2,4,6-trinitrotoluene toxicity. Adverse effects on standard hematologic parameters were observed in rats (Dilley et al. 1982b; Jiang et al. 1991; Levine et al. 1984, 1990a), mice (Dilley et al. 1982b), and dogs (Dilley et al. 1982b; Levine et al. 1990b) after intermediate oral exposures to 2,4,6-trinitrotoluene. Compensatory responses occurring as a result of anemia (including reticulocytosis, macrocytosis, and increased levels of nucleated erythrocytes) were observed in Fischer-344 rats fed 125 mg/kg/day of 2,4,6-trinitrotoluene for 13 weeks (Levine et al. 1990a). Dose-related anemia was also observed in Fischer-344 rats fed 10 or 50 mg/kg/day 2,4,6-trinitrotoluene for 24 months (Army 1984a). In this chronic exposure study, both male and female rats had reduced hematocrit, hemoglobin, and red blood cells. These hematological effects were observed throughout the entire duration of the study in male rats, but only for the first year in female rats. Methemoglobin was observed in male rats at doses of 10 and 50 mg/kg/day. Howell-Jolley and Heinz Bodies occurred at 50 mg/kg/day in male rats. Therefore, male rats seemed somewhat more sensitive than female rats. Reticulocytosis, but not macrocytosis, was present as a compensatory response to the anemic state in all animals. Histopathology revealed splenic lesions consisting of sinusoidal congestion, extramedullary hematopoiesis, and increased amounts of hemosiderin-like pigment (Army 1984a). The findings are consistent with the hypothesis that 2,4,6-trinitrotoluene induces anemia by causing hemolysis through oxidative damage which is mediated by 2,4,6-trinitrotoluene and/or its metabolites. This conclusion is further supported by the presence of methemoglobinemia, produced by the oxidation of the heme iron, observed when 2,4,6-trinitrotoluene was fed to rats at 300 mg/kg/day for 13 weeks (Levine et al. 1984), to rats at 10 or 50 mg/kg/day for 24 months (Army 1984a), and to dogs at 32 mg/kg/day for 6 months (Levine et al. 1990b). Mild anemia was also noted in B6C3F₁ mice fed 70 mg/kg/day 2,4,6-trinitrotoluene for 24 months (Army 1984b). Anemia, as indicated by dose-dependent decreases in hematocrit and hemoglobin levels and decreased erythrocyte counts, was observed in dogs administered 2,4,6-trinitrotoluene (via capsules) for 6 months (Levine et al. 1990b). The anemia was compensated by reticulocytosis, macrocytosis, and an increased number of nucleated erythrocytes. There was an elevated level of methemoglobin in all dogs treated with 32 mg/kg/day of 2,4,6-trinitrotoluene. No effects on methemoglobin levels, or blood cells and Heinz body counts were found in monkeys following gavage administration of 2,4,6-trinitrotoluene at dose levels as high as 1.0 mg/kg/day (Martin and Hart 1974).

Bone marrow fibrosis was present in a significant number of female rats fed 50 mg/kg/day of 2,4,6-trinitrotoluene for 24 months (Army 1984a).

Dogs treated daily with 8 mg/kg/day of 2,4,6-trinitrotoluene for 6 months had approximately a 68% and 22% increase in platelet levels over the control animals for males and females respectively (Levine et al. 1990b). A similar increase was noted in Fischer-344 rats treated with 50 mg/kg/day of 2,4,6-trinitrotoluene for 24 months (Army 1984a). This increase occurred during the 2nd year of treatment and was not present at the end of the study period, week 104 (Army 1984a). Although the increase in the number of platelets appeared to be related to 2,4,6-trinitrotoluene treatment, its significance is not clear.

Hepatic Effects. No studies were located regarding hepatic effects in humans after oral exposure to 2,4,6-trinitrotoluene. However, toxic jaundice was often fatal in workers exposed to 2,4,6-trinitrotoluene in ammunition plants during World War I (Army 1978a). Between 1916 and 1941, 475 cases of toxic jaundice were recorded in a British ammunition plant; 125 of these were fatal (Army 1978a). During World War II, only eight fatal cases of toxic hepatitis were recorded in the United States because industrial hygiene techniques had improved since the first world war (Army 1978a).

Limited information is available for adverse hepatic effects in animals after acute oral exposure to 2,4,6-trinitrotoluene. The most common adaptive change observed in mice, rats, and dogs during intermediate exposure was an increase in liver weight and/or size. A significant increase in liver weight was noted in rats receiving 200 mg/kg/day of 2,4,6-trinitrotoluene for 6 weeks (Jiang et al. 1991) or 125 mg/kg/day for 13 weeks (Levine et al. 1984). A similar observation was made in male mice treated with 193 mg/kg/day for 13 weeks (Dilley et al. 1982b). Dogs treated with 20 mg/kg/day of 2,4,6-trinitrotoluene for 13 weeks also had increased liver weight (Dilley et al. 1982b). Another adaptive response to 2,4,6-trinitrotoluene-induced hepatotoxicity was a reduction in serum glutamicpyruvic transaminase (SGPT) in rats treated with 160 mg/kg/day and dogs treated with 20 mg/kg/day for 13 weeks (Dilley et al. 1982b). No change in serum glutamic-oxaloacetic transaminase (SGOT) was noted in those same animals (Dilley et al. 1982b). Dose-related changes such as hepatocytomegaly and cloudy swelling were present in dogs after exposure to doses of 0.5 mg/kg/day or greater of 2,4,6-trinitrotoluene for 6 months (Levine et al. 1990b). In addition, a reduction in SGPT activity was noted in dogs administered 8 or 32 mg/kg/day. Necrotic lesions in the liver were found in mice treated with 193 mg/kg/day for 13 weeks (Dilley et al. 1982b), while hepatic cirrhosis was seen in dogs treated with 8 mg/kg/day of 2,4,6-trinitrotoluene for 6 months (Levine et al. 1990b). Monkeys administered 1 mg/kg/day 2,4,6-trinitrotoluene (by gavage) for 90 days displayed ironpositive material in the liver (Martin and Hart 1974). However, results of the bromosulfophthalein (BSP) dye test revealed no effects on liver function. This study was limited because only three monkeys per sex per were used, precluding statistical analysis of the data. In addition, there was a high frequency of emesis in treated animals. These results indicate that there is a species difference regarding 2,4,6-trinitrotoluene hepatotoxicity. Dogs seem to respond to lower concentrations of 2,4,6-trinitrotoluene than do mice or rats.

The hepatotoxicity of 2,4,6-trinitrotoluene was reflected in elevated levels of cholesterol present in the serum after intermediate and chronic exposures. Increased serum cholesterol levels were present in rats treated with 25 (Levine et al, 1984), 125 (Levine et al. 1990a), and 160 mg/kg/day (Dilley et al. 1982b) of 2,4,6-trinitrotoluene for 13 weeks. A similar finding was seen in dogs fed 20 mg/kg/day of 2,4,6-trinitrotoluene for the same time period (Dilley et al. 1982b). Increased cholesterol levels were also found in male and female rats treated with 50 mg/kg/day for 24 months (Army 1984a).

Evidence for adverse hepatic effects has also been revealed in studies of chronic exposure. Doserelated hepatomegaly resulting from hepatocellular hyperplasia was observed in Fischer-344 rats given 10 or 50 mg/kg/day of 2,4,6-trinitrotoluene for 24 months (Army 1984a).

Serum lipids were affected by chronic administration of 2,4,6-trinitrotoluene to rats (Army 1984a). The levels of serum triglycerides were affected differently in male and female rats exposed to 2, 10, or 50 mg/kg/day of 2,4,6-trinitrotoluene (Army 1984a). Treatment-related hypotriglyceridemia was observed in females at 2 mg/kg/day; this change was statistically significant at 10 and 50 mg/kg/day during week 104 of treatment. In male rats treated with 50 mg/kg/day, a significant reduction in triglyceride levels was seen after 104 weeks of treatment (Army 1984a). The results indicate that female rats were more susceptible to 2,4,6-trinitrotoluene-induced reduction in serum triglyceride levels than male rats that showed reduced triglycerides only at the highest treatment doses. Hypoglyceridemia was also observed in mice treated with 70 mg/kg/day 2,4,6-trinitrotoluene for 24 months (Army 1984b), and a decrease in glucose levels was observed in dogs treated with 8 or 32 mg/kg/day 2,4,6-trinitrotoluene (Levine et al. 1990b).

Jaundice (icterus) was observed in beagle dogs treated with 32 mg/kg/day for 26 weeks (Levine et al. 1990b). The presence of jaundice was evidenced by elevated bilirubin levels in both serum and urine and increased urobilinogen values (Levine et al. 1990b). Histopathological analysis of these animals revealed hemosiderosis in Kupffer’s cells in all dogs receiving 8 and 32 mg/kg/day and in one female receiving 2 mg/kg/day for 26 weeks (Levine et al. 1990b).

One of the proposed mechanisms of 2,4,6-trinitrotoluene-induced toxicity is an increase in free radical levels which occurs after exposure. Superoxide radicals and hydrogen peroxide were measured in mitochondria and microsomes from livers of monkeys treated with 0, 60, or 120 mg/kg/day of 2,4,6-trinitrotoluene for 12 weeks (Kong et al. 1989). The amount of superoxide radicals was indirectly measured by the formation of adrenochrome from adrenalin, and hydrogen peroxide production was evaluated by the conversion of methanol to formaldehyde. There was a dose-dependent increase in superoxide radicals and hydrogen peroxide production in liver mitochondria and microsomes (Kong et al. 1989). These findings were confirmed when mitochondria and microsomes obtained from various organs were treated in vitro with 0, 0.04, 0.2, or 1 mmol of 2,4,6-trinitrotoluene and then tested for adrenochrome and formaldehyde production. Different amounts of hydrogen peroxide were produced in the mitochondria of the various organs. The highest amount of hydrogen peroxide was present in the liver followed by brain, testicle, kidney cortex, and kidney medulla (Kong et al. 1989).

Renal Effects. Discoloration of the urine is among the first indications of 2,4,6-trinitrotoluene intoxication in humans. The color of urine ranges from abnormal amber to a deep red, and in most cases the results are positive for Webster’s test (a qualitative urine test for 2,4,6-trinitrotoluene based on the formulation of purple color in acidified urine samples following extraction with ether and treatment with potassium hydroxide) (Army 1978a).

Accumulation of yellowish-brown pigment in the renal cortex of rats treated with 125 mg/kg/day of 2,4,6-trinitrotoluene for 13 weeks was seen during histopathological analysis. The authors suggest that this pigment may have represented known photolytic decomposition products of 2,4,6-trinitrotoluene (Levine et al. 1984, 1990b). The same observation was made in female rats treated with 10 or 50 mg/kg/day for 24 months (Army 1984a). A dose-related increase in granular pigment within the cytoplasm of epithelial cells of proximal convoluted tubules was observed at the end of the treatment period in male rats receiving either 10 or 50 mg/kg/day for 24 months (Army 1984a). Increased filtration rate was also present in these chronically exposed animals (Army 1984a).

2.2.2.3. Immunological and Lymphoreticular Effects

Limited information was located regarding immunological effects in humans after oral exposure to 2,4,6-trinitrotoluene. However, an early reaction to 2,4,6-trinitrotoluene intoxication was an increase in the number of mononuclear leukocytes found in the blood counts of 105 exposed individuals (Army 1978a). This increase seems to precede any other symptom and remains positive for 2-3 months; therefore, it would be helpful in the differential diagnosis of 2,4,6trinitrotoluene poisoning, especially when Webster’s test is negative (Army 1978a). The route of exposure and precise dose were not defined in this report.

An increase in lymphocyte numbers was also seen in nine fatal cases of 2,4,6-trinitrotoluene toxicity in humans. The normal range for lymphocytes is 20-40% of the total white blood cell count. In the affected patients, the average was 78% of the total white blood count (range, 61-92%) (Army 1978a). Patients who recovered after 2,4,6-trinitrotoluene intoxication had lymphocyte counts which were approximately 46% of the total white cell count (Army 1978a).

No adverse lymphoreticular effects as assessed by changes in lymphocyte levels and histological changes in the spleen were seen in dogs treated with 2.0 mg/kg/day, rats treated with 34.7 mg/kg/day, or mice treated with 35.7 mg/kg/day of 2,4,6-trinitrotoluene for 13 weeks (Dilley et al. 1982b).

Increased spleen weight was seen in mice, rats, and dogs after an exposure of 13 weeks to 2,4,6-trinitrotoluene. The splenomegaly is possibly related to an increased clearance of hemolysed cells (Dilley et al. 1982b; Levine et al. 1984, 1990a). Male and female dogs exposed to 20 mg/kg/day, mice exposed to 193 mg/kg/day (Dilley et al. 1982b), and rats exposed to 125 mg/kg/day (Levine et al. 1984, 1990a) and 160 mg/kg/day (Dilley et al. 1982b) all had increased spleen weights after 13 weeks of exposure. Spleen enlargement related to 2,4,6-trinitrotoluene administration was also noted in dogs treated with 8 or 32 mg/kg/day for 26 weeks. The ratios of myeloid to erythroid cells were also significantly lower in these dogs, and they had varying degrees of splenic congestion (Levine et al. 1990b). Splenic hemosiderosis related to 2,4,6-trinitrotoluene treatment was also present in rats receiving 160 or 300 mg/kg/day (Dilley et al. 1982b; Levine et al. 1984), mice receiving 193 mg/kg/day (Dilley et al. 1982b), and dogs receiving 20 mg/kg/day (Dilley et al. 1982b) of 2,4,6-trinitrotoluene for 13 weeks. Lymphopenia was present in mice treated with 193 mg/kg/day for 13 weeks (Dilley et al. 1982b). Increased globulin levels and leukocytosis were noted in the dogs treated with 20 mg/kg/day after a 4 week recovery period. Leukocytosis was observed in rats fed 160 mg/kg/day of 2,4,6-trinitrotoluene for 13 weeks (Dilley et al. 1982b). Sinusoidal congestion, extramedullary hematopoiesis, and hemosiderin-like pigment in the spleen were observed in male and female rats fed 10 or 50 mg/kg/day of 2,4,6-trinitrotoluene for 24 months (Army 1984a). Enlargement of the spleen and lymph nodes was noted in female mice treated with 70 mg/kg/day of 2,4,6-trinitrotoluene for 24 months (Army 1984b).

2.2.2.4. Neurological Effects

No studies were located regarding neurological effects in humans after oral exposure to 2,4,6-trinitrotoluene.

In acute exposure studies, rats fed 182 mg/kg/day of 2,4,6-trinitrotoluene for 4 days showed no signs of neurotoxicity as measured by changes in zoxazolamine paralysis time and hexobarbital sleeping time (Short and Lee 1980). However, in a single-dose oral LD₅₀ study in rodents, rats and mice showed signs of inactivity, were tremulous, developed convulsions, and died (Dilley et al. 1982b).

Similar observations were made in intermediate-duration studies. No signs of neurotoxicity were seen after 13 weeks of 2,4,6-trinitrotoluene treatment in dogs receiving 0.2 mg/kg/day (Dilley et al. 1982b), monkeys receiving 1 mg/kg/day (Martin and Hart 1974), or rats receiving 1.42 mg/kg/day (Dilley et al. 1982b). Dogs treated with 32 mg/kg/day for 6 months were ataxic (Levine et al. 1990b), while inactivity was observed in dogs after treatment with 20 mg/kg/day for 13 weeks (Dilley et al. 1982b). Dose-related changes in behavior such as lethargy and/or ataxia were seen in rats treated with 34.7 or 125 mg/kg/day for 13 weeks (Dilley et al. 1982b; Levine et al. 1984). Brain lesions with focal vacuolation were seen in rats receiving 300 mg/kg/day of 2,4,6-trinitrotoluene for 13 weeks (Levine et al. 1984).

No significant signs of neurotoxicity were seen in Fischer-344 rats treated with up to 50 mg/kg/day of 2,4,6-trinitrotoluene for 24 months (Army 1984a). The combined results of acute and intermediate exposure studies indicate species differences in 2,4,6-trinitrotoluene-induced neurotoxicity, with dogs being more sensitive than rats or mice.

2.2.2.5. Reproductive Effects

No studies were located regarding reproductive effects in humans after oral exposure to 2,4,6-trinitrotoluene.

Significantly decreased testes weights and testes zinc and copper concentrations were observed in male rats exposed to 200 mg/kg/day 2,4,6-trinitrotoluene for 6 weeks (Jiang et al. 1991). In addition, the serum ceruloplasmin concentration was significantly lower in the male rats. Zinc metabolism seems to be affected more than copper metabolism by 2,4,6-trinitrotoluene treatment. Although there was a close correlation between testicular weight and testicular zinc concentration, the role of zinc in decreasing testicular weight is not clear (Jiang et al. 1991). However, zinc is known to be essential for maintenance of normal testicular function (Jiang et al. 1991). No adverse reproductive effects were noted in rats exposed to 1.42 mg/kg/day of 2,4,6-trinitrotoluene for 13 weeks (Dilley et al. 1982b). However, male rats treated with 125, 160, or 300 mg/kg/day for the same period of time had serious reproductive effects such as degenerated germinal epithelium, testicular atrophy, and atrophic seminiferous tubules (Dilley et al. 1982b; Levine et al. 1984, 1990a). Testicular atrophy was not reversible in rats allowed 4 weeks of recovery (Dilley et al. 1982b).

2.2.2.6. Developmental Effects

No studies were located regarding developmental effects in humans or animals after oral exposure to 2,4,6-trinitrotoluene.

2.2.2.7. Genotoxic Effects

No studies were located regarding genotoxic effects in humans after oral exposure to 2,4,6-trinitrotoluene Results were negative from in viva studies employing the oral route of exposure to ascertain whether 2,4,6-trinitrotoluene has the potential to induce clastogenic effects in somatic cells or increase the frequency of unscheduled deoxyribonucleic acid (DNA) synthesis (UDS) in liver cells. However, the findings from somatic cell cytogenetic assays with rats were compromised and, therefore, do not fully support a negative conclusion.

In the bone marrow test, groups of five male Sprague-Dawley rats were administered dietary concentrations of 0.002% or 0.25% 2,4,6-trinitrotoluene for 28 days; two additional groups of five rats each were similarly treated and allowed a 28-day recovery period (Army 1978c). At the conclusion of the treatment or recovery period, animals were sacrificed; bone marrow cells were harvested and examined for abnormal chromosome morphology. No animals died prior to the scheduled sacrifice. The study authors attributed the reduced body weight observed in the high-dose group to the palatability of the test material rather than to a toxic effect. The slight depression in the mitotic indices for high-dose animals at the conclusion of treatment was not considered indicative of a cytotoxic effect on the target organ (bone marrow cells). Although no chromosome aberrations were scored in the exposure groups immediately after treatment or following the 28-day recovery period, the failure to demonstrate overt toxicity in the test animals or cytotoxic effects on the target organ renders the study insufficient to fully support the conclusion that 2,4,6-trinitrotoluene was negative in this in vivo cytogenetics assay.

2,4,6-Trinitrotoluene was administered by oral gavage at doses of 100, 200, 500, or 1,000 mg/kg to male Alderley Park rats and at doses of 200, 500, and 1,000 mg/kg to male Fischer-344 rats (up to three rats/group/strain) to investigate UDS in liver cells (Ashby et al. 1985). There was no evidence of a cytotoxic or genotoxic effect on the hepatocytes of either strain 12 hours after 2,4,6-trinitrotoluene exposure.

Other genotoxicity studies are discussed in Section 2.4.

2.2.2.8. Cancer

No studies were located regarding cancer in humans after oral exposure to 2,4,6-trinitrotoluene. However, a preliminary study of a German population living near the sites of two World War II munitions plants indicates an association between increased rates of some types of leukemia and living in a town near 2,4,6-trinitrotoluene waste from these plants (Kolb et al. 1993). The study shows increased relative risk of acute myelogenous leukemia (AML) for adult males and females living near the former explosives plants when compared with adults in a neighboring county. The relative risk is particularly high for individuals over 65 years of age. However, study case numbers are very small. The relative risk for chronic myelogenous leukemia (CML) is also increased for males but there was only one case among females so comparisons could not be made. The relative proximity of the cases of leukemia to the sites of 2,4,6-trinitrotoluene manufacture or disposal is not known, nor are any 2,4,6-trinitrotoluene concentrations in the environment reported. No investigation of confounding variables (i.e., benzene exposure or occupational exposure to carcinogens) has been done. The study concludes that a causal relationship is suggested, but further investigation of the living and working conditions of the populations is required (Kolb et al. 1993).

In a chronic study, groups of 150 (75 males and 75 female) Fischer-344 rats were exposed to 0, 0.4, 2.0, 10.0, and 50.0 mg/kg/day of 2,4,6-trinitrotoluene in their food for 24 months (Army 1984a). A statistically significant number of female rats (12/55 or 21.8%) exposed to 50-mg/kg/day doses developed urinary bladder carcinomas. Urinary bladder papillomas were present in 1/55 and 5/55 female rats fed 10 and 50 mg/kg/day, respectively. No metastases were observed in any of the animals. Histopathologic lesions included increased incidence of hyperplastic, preneoplastic, and neoplastic changes of the mucosal epithelium of the urinary bladder. The cancer incidence observed in this chronic exposure study is further supported by renal and urinary bladder hyperplasia observed in treated animals. None of the control animals developed lesions of the urinary bladder. In a similar study conducted in groups of 150 B6C3F₁ mice (75 males and 75 females), a statistically significant incidence (p<0.01) of leukemia and/or malignant lymphoma of the spleen was present in female mice receiving 70 mg/kg/day of 2,4,6-trinitrotoluene for 24 months (Army 1984b). Leukemia and/or malignant lymphoma of the spleen was noted in 28% and 32% of the female mice administered 1.5 and 10 mg/kg/day, respectively. The increase in the cancer incidence in the female mice receiving 1.5 or 10 mg/kg/day of 2,4,6-trinitrotoluene was not statistically significant. Histopathology revealed that leukemia was of granulocytic or lymphocytic type, while lymphoma was histiocytic, lymphocytic, or of a mixed type. All the lesions were treatment related and systemic in nature. The neoplasias involved other organs and tissues such as adrenals, bone marrow, brain, gastrointestinal tract, eyes, kidneys, liver, lungs, and lymph nodes. The occurrence of combined leukemia/malignant lymphoma seemed to be dose related but was not statistically significant. Based on the information from these two chronic animal studies, EPA has classified 2,4,6-trinitrotoluene as a possible human carcinogen (Group C) (EPA 1989b).

2.2.3.1. Death

Adverse effects of 2,4,6-trinitrotoluene on the liver and hematopoietic system have caused the greatest number of deaths among munitions workers. There were 475 deaths out of about 17,000 cases of 2,4,6-trinitrotoluene poisoning in the United States within 7.5 months during World War I (McConnell and Flinn 1946). In the same report, 22 cases of death due to occupational exposure to 2,4,6-trinitrotoluene during World War II were described (McConnell and Flinn 1946). In this series of fatal cases, 8 died from toxic hepatitis, 13 died from aplastic anemia, and 1 died probably from the combination of both these conditions (McConnell and Flinn 1946). The authors indicated that exposure occurred by dermal contact and inhalation.

No studies were located regarding death in animals after dermal exposure to 2,4,6-trinitrotoluene.

2.2.3.2. Systemic Effects

No studies were located regarding respiratory, cardiovascular, gastrointestinal, or musculoskeletal effects in humans or animals after dermal exposure to 2,4,6-trinitrotoluene.

Hematological Effects. Acute hemolytic disease was described in three ammunition plant workers who filled shells with 2,4,6-trinitrotoluene (Djerassi and Vitany 1975). All three were deficient in glucose-6-phosphate dehydrogenase (G6PD), an enzyme that catalyzes the oxidation of glucose 6-phosphate to 6-phosphoglucono-lactone. All three cases developed acute severe hemolysis 2-3 days after being exposed to 2,4,6-trinitrotoluene and had very similar symptoms: paleness, weakness, and vertigo. They also had decreased hemoglobin levels, decreased hematocrit, and increased reticulocyte numbers (Djerassi and Vitany 1975). All three recovered and had no further complications when examined 5 and 10 years later. Limitations of this report are that the routes of exposure are not specified, although it seems that there was dermal and inhalation exposure, and that the level of exposure is not known. However, the authors note that the air levels of 2,4,6-trinitrotoluene were higher than the allowed daily exposure limit, which was 1.5 mg/m³ at that time.

No studies were located regarding hematological effects in animals after dermal exposure to 2,4,6-trinitrotoluene.

Hepatic Effects. A modified MacLagen test (thymol turbidity test) was used to give evidence of cirrhosis or hepatitis resulting from 2,4,6-trinitrotoluene exposure in ammunition workers (Goodwin 1972). In a retrospective study spanning 20 years, the data showed that 40 out of 4,641 workers had >5 MacLagen units (2.9 MacLagen units is considered to be normal); the length of exposure and dose of 2,4,6-trinitrotoluene were not specified. However, the hepatotoxicity was reversible. All the workers with a MacLagen test result of >5 units were transferred to other jobs, and their readings returned to normal within 3 weeks (Goodwin 1972).

No studies were located regarding hepatic effects in animals after dermal exposure to 2,4,6-trinitrotoluene.

Renal Effects. No adverse renal effects were reported in humans after acute exposure to 2,4,6-trinitrotoluene as measured by Webster’s reaction and the aminodinitrotoluene (ADNT) test (Hassman and Hassmanova 1976). In this study, the exposure route was not clearly specified, but it seems that dermal contact and inhalation were the major routes of exposure.

No studies were located regarding renal effects in animals after dermal exposure to 2,4,6-trinitrotoluene.

Dermal Effects. Allergic contact dermatitis with erythema (Goh 1988) and erythematous papillar rush with edema (Goh and Rajan 1983) were reported in two ammunition workers after intermediate exposures to 2,4,6-trinitrotoluene. In both cases, patch tests were used to determine the allergen. The results showed that 5% is the most suitable concentration for patch testing in order to avoid false positive or negative results (Goh and Rajan 1983). In both affected workers, dermal reactions developed on parts of the body that were exposed and in direct contact with 2,4,6-trinitrotoluene, such as hands and forearms. These reactions subsided and disappeared when the workers were transferred to different jobs and were no longer in contact with 2,4,6-trinitrotoluene.

No studies were located regarding dermal effects in animals after dermal exposure to 2,4,6-trinitrotoluene.

Ocular Effects. The development of cataracts in humans is believed to be specific to 2,4,6-trinitrotoluene exposure (Hathaway 1985) and is often associated with chronic exposures. Equatorial lens opacities/cataracts were reported in 6 out of 12 Finnish workers (mean age, 39.5±8.9 years) exposed to 2,4,6-trinitrotoluene for an average of 6.8±4.7 years (Harkonen et al. 1983). The principal routes of exposure were dermal and inhalation, although this fact is not clearly indicated in the report. Therefore, it is not known whether cataracts were a systemic or a local effect. The opacities were detectable only on the periphery of the lens and appeared continuous or discontinuous. The opacities of the lens were bilateral and symmetrical and did not affect visual fields or visual acuity. The workroom 2,4,6-trinitrotoluene air concentration was about 0.3 mg/m³ with a range of 0.14-0.58 mg/m³ (Harkonen et al. 1983). There was no control population in the study. The formation of cataracts did not progress further after exposure to 2,4,6-trinitrotoluene was terminated; however, the cataracts were not reversible. The mechanism of 2,4,6-trinitrotoluene cataract formation is not understood, but the authors raised the possibility that oxidative damage may play a role since cellular defense mechanisms protective against oxidative damage (i.e., glucose-6-phosphate dehydrogenase and glutathione concentration) may be deficient after exposure to 2,4,6-trinitrotoluene (Harkonen et al. 1983).

No studies were located regarding ocular effects in animals after dermal exposure to 2,4,6-trinitrotoluene.

2.2.3.3. Immunological and Lymphoreticular Effects

Two ammunition plant workers developed an allergic contact dermatitis with erythema (Goh 1988) and erythematous papillar rush with edema (Goh and Rajan 1983) after intermediate exposures to 2,4,6-trinitrotoluene. Patch tests were used to identify the allergen using 5% as the most suitable concentration (Goh and Rajan 1983). Dermal reactions were observed to occur on the exposed parts of the body such as the hands and forearms. Once removed from environments containing 2,4,6-trinitrotoluene, the workers sensitivity subsided and the dermatitis disappeared.

No studies were located regarding immunological or lymphoreticular effects in animals after dermal exposure to 2,4,6-trinitrotoluene.

2.2.3.4. Neurological Effects

No studies were located regarding neurological effects in humans or animals after dermal exposure to 2,4,6-trinitrotoluene.

2.2.3.5. Reproductive Effects

No studies were located regarding reproductive effects in humans or animals after dermal exposure to 2,4,6-trinitrotoluene.

2.2.3.6. Developmental Effects

No studies were located regarding developmental effects in humans or animals after dermal exposure to 2,4,6-trinitrotoluene.

2.2.3.7. Genotoxic Effects

Two studies involving genotoxic effects in individuals occupationally exposed to 2,4,6-trinitrotoluene suggested that in addition to inhalation exposure, dermal exposure may have occurred (Ahlborg et al. 1985, 1988a). These studies are discussed in detail in Section 2.2.1.7. No other studies were located regarding genotoxic effects in humans or animals after dermal exposure to 2,4,6-trinitrotoluene.

Other genotoxicity studies are discussed in Section 2.4.

2.2.3.8. Cancer

In a case of chronic occupational exposure to 2,4,6-trinitrotoluene, a 61-year-old male died of hepatocellular carcinoma (Garfinkel et al. 1988). The routes of exposure were not specified in the study, and it is possible that the worker was exposed via the inhalation and/or dermal route.

No studies were located regarding cancer in animals after dermal exposure to 2,4,6-trinitrotoluene.

2.3.1.1. Inhalation Exposure

Studies that directly measure the absorption of 2,4,6-trinitrotoluene in humans following inhalation exposure of known amounts of this chemical were not located. The amount of 2,4,6-trinitrotoluene in the urine as measured by Webster’s reaction and the amount of 2-ADNT in the urine were compared in 88 factory workers working with 2,4,6-trinitrotoluene (Hassman and Hassmanova 1976). The concentration of 2,4,6-trinitrotoluene in the air varied from 0.045 to 0.93 mg/m³ in different parts of the plant. The results showed that there was a good correlation between the amounts of 2,4,6-trinitrotoluene and its main metabolite ADNT in the urine of exposed workers. The levels of ADNT increased rapidly during the workday and then declined within 24 hours to close to the levels at the beginning of the workday. These results suggest that 2,4,6-trinitrotoluene is rapidly absorbed and eliminated in the course of an acute inhalation exposure. The study is limited in that there was no information on the route of exposure, the length of exposure in the course of a working day, or the exposure dose.

In an attempt to simulate inhalation exposure, 50 mg/kg of radiolabelled 2,4,6-trinitrotoluene suspended in methyl cellulose was instilled into the trachea of anesthetized, tracheotomized Sprague-Dawley rats (Army 1981d). At the same time, another group of rats was treated orally with the same dose of radiolabelled 2,4,6-trinitrotoluene. Both groups were sacrificed 4 hours later, and tissue and urine samples were collected for radioactivity analysis. The rate of absorption was faster after intratracheal instillation than after oral administration of 2,4,6-trinitrotoluene. Urinary excretion averaged 19.3% of the dose after intratracheal administration and 14.6% of the dose after oral administration. These results indicate that there are differences in the absorption rate of 2,4,6-trinitrotoluene depending on the administration route.

2.3.1.2. Oral Exposure

Discoloration of the urine is among the first indications that metabolism has occurred after 2,4,6-trinitrotoluene absorption in humans. The color of urine ranges from abnormal amber to deep red (Army 1978a).

Similar observations were made in rats and mice. Sixty minutes after a single exposure to 10,000 mg/kg/day of 2,4,6-trinitrotoluene, the urine of mice and rats becomes red in color (Dilley et al. 1982b). This is an indirect indication of 2,4,6-trinitrotoluene absorption. A more direct estimate of absorption of 2,4,6-trinitrotoluene was done in rats, mice, rabbits, and dogs after a single oral dose of 50 mg/kg of radiolabelled 2,4,6-trinitrotoluene (Army 1981d). Twenty-four hours later, the recovery of radiolabel was measured in rats, mice, dogs, and rabbits. The largest percentage of radioactivity was recovered from urine: 59.5%, 59%, and 61% for rats, mice, and dogs, respectively. Rabbits had a slightly higher recovery (74.3%) of radiolabelled 2,4,6-trinitrotoluene in their urine, with a proportional decrease in the radioactivity recovered from the gastrointestinal tract and feces (Army 1981d). Total recovery of the radioactivity (including feces, gastrointestinal tract, and urine) was 92.2%, 94.4%, 94.2%, and 103.6% in rats, mice, dogs, and rabbits, respectively. The red pigment in the urine was not detected in rabbits or dogs. The results of this study indicate that 2,4,6-trinitrotoluene is relatively quickly absorbed after oral administration and that a majority of the ingested compound is excreted within 24 hours.

Findings in rats, mice, and dogs support the observations made in humans. Discoloration of urine was noted in both rats and mice exposed to 34.7 and 35.7 mg/kg/day, respectively, for 13 weeks (Dilley et al. 1982b). Dogs treated with 20 mg/kg/day for 13 weeks had urine that was orange in color (Dilley et al. 1982b), while dogs receiving 8 or 32 mg/kg/day for 26 weeks had light to dark brown urine throughout the treatment period (Levine et al. 1990b). No adverse hepatic effects and no change in urine color were observed in monkeys treated with 1 mg/kg/day for 90 days (Martin and Hart 1974). The Martin and Hart (1974) study was limited in that only three monkeys per sex per group were utilized, thereby precluding statistical analyses of the data. Also, a high frequency of emesis was observed in the treated monkeys. It is believed that the species differences in urine color are due to the presence of unidentified metabolites of 2,4,6-trinitrotoluene. Species differences in 2,4,6-trinitrotoluene toxicity may be attributed to the different metabolic pathways of 2,4,6-trinitrotoluene and its metabolites. Once identified, these urine metabolites may be used as markers of 2,4,6-trinitrotoluene exposure.

2.3.1.3. Dermal Exposure

Although data are limited regarding absorption of 2,4,6-trinitrotoluene following dermal exposure in humans, it appears that it occurs rapidly (Woollen et al. 1986). 2,4,6-Trinitrotoluene absorption was assessed by measuring the urinary concentration of one of its metabolites, ADNT, in 25 exposed workers. There were wide variations between individual workers in the rate of clearance of ADNT from the body. Furthermore, when urine samples were collected from a subgroup of workers from the original group of 25, eight out of nine subjects had detectable ADNT levels in their urine even though these workers had been away from the workplace for 17 days. This is an indication that a portion of absorbed 2,4,6-trinitrotoluene or its metabolites is slowly excreted (Woollen et al. 1986). Additionally, when five workers from the total group of 25 exposed workers were monitored more closely during two workshifts, it was shown that 2,4,6-trinitrotoluene was absorbed rapidly during the exposure period. The limitations of this study are that the dermal exposure dose was not measured and workers’ were also exposed to 2,4,6-trinitrotoluene via the inhalation route.

The differences in absorption and excretion of radiolabelled 2,4,6-trinitrotoluene after dermal and oral exposures were investigated in mice, rabbits, rats, and beagle dogs (Army 1981d). Rats and mice were exposed dermally and orally (by gavage) to 50 mg/kg, and dogs and rabbits to 5 or 50 mg/kg. Twenty-four hours after exposure, animals were sacrificed and urine, gastrointestinal tract, feces, and various tissues were analyzed for radioactivity. Total recovered radioactivity was significantly lower in all species after dermal exposure as compared to after oral exposure (Army 1981d). The highest degree of dermal absorption was observed in rabbits and mice, in which 68.3% and 41.7% of the administered radiolabel was recovered, respectively. The total recovery of radiolabel in dogs and rats was much lower, 17% and 24%, respectively. These results suggest that absorption of radiolabelled 2,4,6-trinitrotoluene is species-dependent and is significantly lower after dermal rather than oral exposure.

2.3.2.1. Inhalation Exposure

No studies were located regarding distribution following inhalation exposure to 2,4,6-trinitrotoluene in humans.

In rats that received 50 mg/kg of radiolabelled 2,4,6-trinitrotoluene by intratracheal administration, the highest tissue concentrations of radiolabel after 4 hours were found in both male and female animals in the fat (82 and 155 µg eq/g, respectively) and gastrointestinal tract (82 and 40 µg eq/g, respectively) (Army 1981d). Since the gastrointestinal tract contained considerable amounts of radioactivity, some of the rats in this experiment were bile-duct cannulated in order to collect bile and estimate the amount of radiolabel. The results were given as a percent of the administered dose. The radioactivities recovered in bile-duct cannulated male and female rats were 20% and 15%, respectively, from the bile, and 18% and 13%, respectively, from the urine (Army 1981d). When these results were compared to excretion results following oral administration, the percent of radioactivity recovered in the urine and bile was significantly higher after intratracheal exposure in both male and female rats. The opposite was true for the amount of radioactivity recovered in the gastrointestinal tract; it was significantly higher in the orally treated rats regardless of cannulation (Army 1981d). These results indicate that the route of administration contributes to the differences in 2,4,6-trinitrotoluene distribution.

2.3.2.2. Oral Exposure

No studies were located regarding distribution following oral exposure to 2,4,6-trinitrotoluene in humans.

Twenty-four hours after administration of a single oral dose of radiolabelled 2,4,6-trinitrotoluene to rats and mice (100 mg/kg), and rabbits and dogs (5 mg/kg), the blood and different tissues were analyzed for radioactivity. The blood and liver, kidney, spleen, lungs, brain, and skeletal muscle of dogs contained a higher percentage of radioactivity than the blood and tissues of rats, mice, and rabbits (Army 1981d). Recovery of radioactivity was greatest in the liver, skeletal muscle, and blood in all four species. However, the amount of radioactivity recovered from tissues was small, ranging from <0.1% to 5.4% of the dose, because the majority of the label is excreted in urine (an average of 60% of the dose) and feces (an average of 11% of the dose) (Army 1981d). This indicates both rapid absorption and rapid distribution in different species after oral exposure to 2,4,6-trinitrotoluene. The limitations of this study are that a small number of animals was analyzed for distribution of radiolabel and that the amount of unchanged 2,4,6-trinitrotoluene was not discussed.

2.3.2.3. Dermal Exposure

No studies were located regarding distribution following dermal exposure to 2,4,6-trinitrotoluene in humans or animals. However, the recovery of radiolabel after a single dermal application of 50 mg/kg of 2,4,6-trinitrotoluene was significantly lower in rats, mice, dogs, and rabbits than in those exposed orally (Army 1981d). Skin and fat around the site of dermal application were not included in the final radiolabel recovery estimates. This may account for the lower radiolabel recovery obtained after the dermal exposure (Army 1981d). The recovery of radiolabel from urine, feces, gastrointestinal tract, blood, and tissue differed among the four species examined: rabbits (56.9%) > mice (41.7%) > rats (22.8) > dogs (15.9%).

2.3.3.1. Inhalation Exposure

No studies were located regarding metabolism following inhalation exposure to 2,4,6-trinitrotoluene in humans or animals. However, in a retrospective study (covering a 5-year period) of 2,4,6-trinitrotoluene workers, no correlation was found between the presence in urine of one of the main 2,4,6-trinitrotoluene metabolites, ADNT, and the results of Webster’s reaction, which measures urine 2,4,6-trinitrotoluene levels (Hassman and Hassmanova 1976). This result provides indirect evidence that 2,4,6-trinitrotoluene is metabolized completely and that no detectable amounts of unchanged compound are present in the urine. This study is limited in that it does not clearly define the route of exposure and does not specify the dose or the length of exposure.

2.3.3.2. Oral Exposure

No studies were located specifically addressing metabolism following oral exposure to 2,4,6-trinitrotoluene in humans.

The 2,4,6-trinitrotoluene molecule may undergo various metabolic transformations, such as oxidation of the methyl group, oxidation of the benzene ring, reduction of the three nitro groups, and conjugation (EPA 1989b). A metabolic pathway illustrating some possible transformation products of 2,4,6Mnitrotoluene is shown in Figure 2-2 (Army 1981d). The failure to detect unmetabolized 2,4,6-trinitrotoluene in the urine of humans (Hassman and Hassmanova 1976) provides indirect evidence that 2,4,6-trinitrotoluene is extensively metabolized. Trace amounts of unmetabolized 2,4,6-trinitrotoluene were found in the urine of rats, mice, rabbits, and dogs (Army 1981d). Several metabolites have been identified in human urine: 4-ADNT, 2-ADNT, 2,4-diatnino-6-nitrotoluene, 4-hydroxylamino-2,6-dinitrotoluene, and amino-nitrocresol (Army 1986c; Channon et al. 1944; Lemberg and Callaghan 1945).

Figure 2-2

Possible Biotransformation Products of 2,4,6-TNT.

2,4,6-Trinitrotoluene was extensively metabolized in rats, mice, dogs, and rabbits after a single oral dose of 50 mg/kg of radiolabelled 2,4,6-trinitrotoluene (Army 1981d). Only minute amounts of the unmetabolized 2,4,6-trinitrotoluene were found in urine. The majority of urinary metabolic products have high polarity and very low extractability in organic solvents. It was therefore difficult to identify them. The metabolic profiles of urine from the four species differed only quantitatively (Army 1981d). 4,6-Diamine, 2,6-diamine, and monoamines of 2,4,6-trinitrotoluene were the predominant metabolites detected in the urine of rats. Smaller quantities of 2- and 4-hydroxylamines and azoxytoluene were present. In contrast to rat urine, greater amounts of the monoamines and hydroxylamines and smaller quantities of polar metabolites and diamines were found in the urine of mice. The urine of dogs contained appreciable amounts of diamines and monoarnines and small amounts of the 4-hydroxylamine and 2-hydroxylamine. Substantial amounts of monoarnines, hydroxylamines, and diamines were noted in rabbit urine (Army 1981d). Treatment of urine of all species with β-glucuronidase increased the amount of extractable radioactivity, indicating that conjugation of 2,4,6-trinitrotoluene metabolites with UDP-glucuronic acid is an important route of metabolism. Urine from treated mice contained the least amount of glucuronide conjugates (Army 1981d).

2.3.3.3. Dermal Exposure

No studies were located regarding metabolism following dermal exposure to 2,4,6-trinitrotoluene in humans.

The differences in metabolic profiles from the urine of mice, rats, dogs, and rabbits after a single dermal dose of 50 mg/kg of radiolabelled 2,4,6-trinitrotoluene were only quantitative (Army 1981d). There was an increased amount of unchanged 2,4,6-trinitrotoluene in urine after a single dermal exposure which was not found after a single oral exposure (Army 1981d). This suggests that the exposure route plays a role in the extent of 2,4,6-trinitrotoluene biotransformation.

2.3.4.1. Inhalation Exposure

4-ADNT, which is considered to be a major metabolite of 2,4,6-trinitrotoluene, was shown to be present in the urine of munitions workers exposed to 0.045-0.93 mg/m³ of 2,4,6-trinitrotoluene by both the Webster reaction and by a polarographic technique (Hassman and Hassmanova 1976). Similar findings were made using a sensitive gas chromatographic method (Almog et al. 1983). No unchanged 2,4,6trinitrotoluene was detected in this study. However, no detail was provided regarding the exposure dose or route.

Recovery of radiolabel from urine of Sprague-Dawley rats 4 hours after intratracheal instillation of 50 mg/kg of radiolabelled 2,4,6-trinitrotoluene was similar to the recovery after oral exposure. The amount of radiolabel in the urine, expressed as a percentage of the dose, was 19% and 13% in male and female animals, respectively (Army 1981d).

2.3.4.2. Oral Exposure

No studies were located regarding excretion following oral exposure to 2,4,6-trinitrotoluene in humans.

The results from animal studies indicate that urine is the major excretion route after a single oral dose of radioactive 2,4,6-trinitrotoluene. The excretion of radioactive label was studied in Sprague-Dawley rats after a single oral dose of 100 mg/kg (Army 1981d). In the course of 24 hours after exposure, 53-65% of the radioactivity was recovered from the urine; 2-8% from the feces; and 30-34% from the gastrointestinal tract and its contents. Similar results on the recovery of radiolabel in urine were obtained after oral exposure of albino CD1 mice (100 mg/kg), rabbits (5 mg/kg), and dogs (50 mg/kg) to radioactive 2,4,6-trinitrotoluene (Army 1981d). These results were confirmed when a 24-hour recovery of radiolabelled 2,4,6-trinitrotoluene was evaluated in rats, mice, dogs, and rabbits after a single oral exposure of 50 mg/kg (Army 1981d). The percentages of radiolabel in urine were 59.5%, 59%, and 61% for rats, mice, and dogs, respectively. The highest amount of radiolabel was recovered from the urine of rabbits, 74.3%. The amounts of label from feces were 11%, 24%, 5%, and 22% for rats, mice, rabbits, and dogs, respectively. Although the number of animals in these studies was relatively small, the results indicate that there are species differences regarding excretion after acute oral exposure to 2,4,6-trinitrotoluene. The urine of rats and mice in these studies was bright red in color, indicating formation of some species-specific, unknown metabolite(s); no such color was observed in dogs and rabbits (Army 1981d).

2.3.4.3. Dermal Exposure

No studies were located regarding excretion following dermal exposure to 2,4,6-trinitrotoluene in humans.

Total recovery of radioactive label (from blood, liver, kidneys, lungs, spleen, brain, muscle, gastrointestinal tract, feces, and urine) after a single dermal exposure of 50 mg/kg of radiolabel 2,4,6-trinitrotoluene was evaluated in rats, mice, dogs, and rabbits and compared to total recovery after oral exposure (Army 198 Id). In all species, the total recovery of the radiolabel was significantly lower after dermal exposure as compared to oral exposure. Total radioactivity recovered after dermal exposure was 16%, 23%, 42%, and 57% in dogs, rats, mice, and rabbits, respectively. For comparison, the recovery after oral exposure was 92%, 94%, 94%, and 104% in rats, dogs, mice, and rabbits (Army 1981d). These results indicate that the total recovery of the label after a single dermal exposure to 2,4,6-trinitrotoluene varies depending on the species. Recovery is also affected by the exposure route; recovery after oral exposure is significantly higher than after dermal exposure (Army 1981d).

Systemic Effects

Respiratory Effects. Extremely limited information was located regarding respiratory effects in humans after exposure to 2,4,6-trinitrotoluene. One study of occupational exposure (Morton et al. 1976) reported several cases of respiratory difficulties in workers exposed to 2,4,6-trinitrotoluene levels in the air that were above the TLV of 0.5 mg/m³ (ACGIH 1993). However, there are several limitations to this study. The report does not state the exact air concentration of 2,4,6-trinitrotoluene, the exposure duration, or the number of exposed workers reporting difficulties. No details are given about the nature of the reported difficulties. No studies in animals were located that described potential respiratory effects of 2,4,6-trinitrotoluene. Therefore, insufficient evidence exists to assess the relevance of these findings to public health.

Cardiovasclilar Effects. No studies were located regarding cardiovascular effects in humans. Intermediate oral exposure to doses as high as 32 mg/kg/day of 2,4,6-trinitrotoluene for 26 weeks did not cause any changes in electrocardiogram or heart rates in beagle dogs (Levine et al. 1990b). The available information is not sufficient to evaluate the effects of 2,4,6-trinitrotoluene on populations living close to ammunition plants.

Gastrointestinal Effects. There are no studies on gastrointestinal effects in humans after exposure to 2,4,6-trinitrotoluene. However, two studies in dogs reported adverse gastrointestinal effects following intermediate oral exposure to 2,4,6-trinitrotoluene. Dogs receiving 20 mg/kg/day for 13 weeks had mucoid stools and diarrhea (Dilley et al. 1982b), while inflammation of a part of the small intestine was observed in beagle dogs fed 0.5, 2, 8, and 32 mg/kg/day of 2,4,6-trinitrotoluene for 25 weeks (Levine et al. 1990b). The inflammation was dose-dependent and was more pronounced in dogs receiving the highest dose. Based on the available information, it is possible, although unlikely, that oral exposure to 2,4,6-trinitrotofuene may cause some adverse gastrointestinal effects, but it is not known if such effects would occur after dermal exposure in the vicinity of an ammunition plant or a demilitarization facility. Although dermal exposure has been indicated as the most likely route of exposure in occupational situations, limited exposure to 2,4,6-trinitrotoluene may result from ingestion of produce contaminated by deposition from fugitive particles or resuspension of contaminated soil, or from ingestion of animal products from animals that graze in the vicinity of an ammunition or demilitarization facility (Army 1986d).

Hematological Effects. Fatal cases of aplastic anemia among workers engaged in the production of explosives have not been reported in the recent literature, although they occurred in England (Hathaway 1985) and other countries involved in World War I. The incidence of adverse health effects of 2,4,6-trinitrotoluene including aplastic anemia have decreased dramatically because of improvements in protective measures in munitions factories.

Dose-related reductions in hemoglobin and hematocrit and a 50% increase in reticulocyte counts were noted in 626 workers exposed to 2,4,6-trinitrotoluene air levels ranging from <0.1 to 1.49 mg/m³ (Army 1976). The duration of exposure was not specified. In another study, no abnormal values for hemoglobin were found in 43 workers employed in the manufacture of 2,4,6-trinitrotoluene who were monitored for 5 months (Morton et al. 1976).

Acute hemolytic disease was described in three ammunition plant workers (Djerassi and Vitany 1975) who were also glucose-6-phosphate dehydrogenase (GGPD) deficient. This study is limited in that the exposure route is not specified, so it is not clear how relevant the finding is for general public health.

Anemia (consisting of reduced number of red blood cells and reduced hemoglobin and hematocrit) is one of the major signs of 2,4,6-trinitrotoluene toxicity. These adverse effects were observed in rats (Dilley et al. 1982b; Jiang et al. 1991; Levine et al. 1984, 1990a), mice (Dilley et al. 1982b), and dogs (Dilley et al. 1982b; Levine et al. 1990b) after intermediate oral exposure to 2,4,6-trinitrotoluene. Similar observations were made in Fischer-344 rats fed 10 or 50 mg/kg/day for 24 months (Army 1984a). In this chronic exposure study, male rats were somewhat more sensitive than female rats to the toxic effects of 2,4,6-trinitrotoluene. Reticulocytosis was present as a compensatory response to the anemic state in all animals. Methemoglobinemia was noted in rats fed 300 mg/kg/day for 13 weeks (Levine et al. 1984), in rats fed 10 or 50 mg/kg/day for 24 months (Army 1984a), and in dogs fed 32 mg/kg/day for 6 months (Levine et al. 1990b).

Bone marrow fibrosis and leukocytosis were present in rats orally exposed to 2,4,6-trinitrotoluene for 24 months or 13 weeks; these animals were also anemic (Army 1984a; Dilley et al. 1982b).

Dogs and rats had increased platelet levels after exposure for 6 and 24 months, respectively (Army 1984a; Levine 1990b). The significance of this finding was not discussed.

Based on this information, it seems unlikely that sufficient amounts of 2,4,6-trinitrotoluene would be present near ammunition plants to cause adverse hematological effects in the population living in the vicinity. Individuals who are G6PD deficient may need to be evaluated as a potentially susceptible population.

Hepatic Effects. Toxic hepatitis has been the principal manifestation of 2,4,6-trinitrotoluene toxicity in humans, and many cases recorded during World War I were fatal (Army 1987a).

Reports on adverse hepatic effects of 2,4,6-trinitrotoluene in humans have been located. Increases in hepatic enzymes (SGOT and LDH) were noted in ammunition plant workers exposed when air levels of 2,4,6-trinitrotoluene rose from 0.3 to 0.8 mg/m³ (Morton et al. 1976). The duration of the study was 5 months. In another study, no significant differences in liver function (LDH, bilirubin, alkaline phosphatase, SGOT and SGPT) were noted in a cross-sectional study on 626 munitions workers exposed to 0.5 mg/m³ of 2,4,6-trinitrotoluene (Army 1976).

In a retrospective study spanning 20 years, liver cell irritation (measured with the MacLagen thymol turbidity test) was present in 40 munitions workers (Goodwin 1972). The report did not specify the exposure dose or route. Exposure of animals to moderate-to-high levels (0.5-200 mg/kg/day) of 2,4,6-trinitrotoluene over intermediate-to-chronic periods has been reported to cause adverse effects such as jaundice, elevated serum and urine bilirubin levels, hyperplasia, cloudy swelling, focal necrosis and cirrhosis of the liver, changes in the levels of serum triglycerides, and increased serum cholesterol levels (Army 1984a; Dilley et al. 1982b; Levine et al. 1984, 1990b). It is not known if chronic exposure to the levels of 2,4,6-trinitrotoluene described above would cause similar adverse hepatic effects in exposed humans.

These degenerative effects are distinct from the adaptive changes observed in livers of a number of nonhuman species in response to exposure to 2,4,6-trinitrotoluene. The most common adaptive change observed in several animal species during intermediate exposure to 2,4,6-trinitrotoluene was an increase in liver weight and/or size (Dilley et al. 1982b; Jiang et al. 1991; Levine et al. 1984). It is not known if these effects would occur in humans.

Renal Effects. No studies were located regarding renal effects in humans after exposure to 2,4,6-trinitrotoluene. However, studies have shown that discoloration of the urine is among the first indications of 2,4,6-trinitrotoluene exposure in humans and is due to the presence of 2,4,6-trinitrotoluene metabolites. The color of urine ranges from abnormal amber to deep red (Army 1978a).

Discoloration of the urine from the presence of 2,4,6-trinitrotoluene metabolites also occurs in rats, mice, and dogs. Sixty minutes after acute exposure to 10,000 mg/kg/day of 2,4,6-trinitrotoluene, the urine in mice and rats became red in color because of the presence of 2,4,6-trinitrotoluene metabolites (Dilley et al. 1982b). The same effect was observed in rats and dogs treated with higher doses and for a longer period (Dilley et al. 1982b). Increased filtration rate was present in rats chronically exposed to 10 and 50 mg/kg/day of 2,4,6-trinitrotoluene (Army 1984a).

In rats treated with higher doses of 2,4,6-trinitrotoluene for 13 weeks (Levine et al. 1984; 1990b) or 24 months (Army 1984a), histopathological analysis revealed the accumulation of yellowish-brown pigment in the renal cortex and in the epithelial cells of proximal convoluted tubules. It is therefore possible that persons exposed to extremely high levels of 2,4,6-trinitrotoluene may be at increased risk of renal toxicity.

Dermal Effects. Exposure to 2,4,6-trinitrotoluene can cause dermatitis in workers handling the compound (Morton et al. 1976). Two incidences of allergic contact dermatitis were reported in two ammunition plant workers after intermediate-duration exposure to 2,4,6-trinitrotoluene (Goh 1988; Goh and Rajan 1983). These findings indicate that prolonged exposure to relatively low levels of 2,4,6-trinitrotoluene may cause an allergic reaction manifested by dermatitis appearing in the areas of contact with the chemical.

Ocular Effects. The appearance of irreversible cataracts is believed to be specific to 2,4,6-trinitrotoluene exposure. It is often associated with chronic exposures to relatively low levels of 2,4,6-trinitrotoluene (Hathaway 1985). Equatorial lens opacities/cataracts were reported in 6 out of 12 workers exposed to 2,4,6-trinitrotoluene for an average of 6.8 years (Harkonen et al. 1983). The average concentration of 2,4,6-trinitrotoluene in the air was 0.3 mg/m³, which is well below the threshold limit of 0.5 mg/m³ (ACGIH 1993). It is therefore possible that chronic exposure to low levels of 2,4,6-trinitrotoluene in the vicinity of ammunition plants may have adverse ocular effects.

Immunological and Lymphoreticular Effects. An increase in the number of mononuclear leukocytes was found in reviewing blood cell counts of 105 individuals exposed to 2,4,6-trinitrotoluene (Army 1978a). This increase precedes any other symptom, remains positive for 2-3 months, and could be helpful in differential diagnosis. Also increased were lymphocyte numbers in nine cases of fatal 2,4,6-trinitrotoluene poisoning (Army 1974). Since the doses necessary to produce these effects were not established, the possibility that susceptible persons living in the vicinity of ammunition plants may be exposed to sufficient amounts of 2,4,6-trinitrotoluene to trigger such immunological effects cannot be excluded.

Exposure to 2,4,6-trinitrotoluene can cause dermatitis in workers handling the compound (Morton et al. 1976). Two incidences of allergic contact dermatitis were reported in two ammunition plant workers after intermediate exposure to 2,4,6-trinitrotoluene (Goh 1988; Goh and Rajan 1983). These findings indicate that prolonged exposure to relatively low levels of 2,4,6-trinitrotoluene may cause an allergic reaction manifested by dermatitis appearing in the areas of contact with the chemical.

Increased spleen weight was the most common effect seen in several nonhuman species after intermediate exposure to medium-to-high doses of 2,4,6-trinitrotoluene (Dilley et al. 1982b; Levine et al. 1984, 1990a). Other changes, such as splenic congestion and hemosiderosis, reduced lymphocyte counts, increased lymphocyte counts, and increased globulin levels, were also noted. Thus, persons exposed to sufficiently high levels of 2,4,6-trinitrotoluene near ammunition plants may be at risk of developing immune system or lymphoreticular effects including splenomegaly with splenic congestion and hemosiderosis, lymphocytosis due to reduced lymphocyte counts, and higher globulin levels.

Neurological Effects. Only minor neurological effects, such as altered taste, were noted in humans after an inhalation exposure of 0.3 mg/m³ of 2,4,6-trinitrotoluene (Morton et al. 1976). On the basis of this limited information, it is difficult to speculate on possible adverse neurological effects that may occur in 2,4,6-trinitrotoluene-exposed people living in the vicinity of ammunition plants.

Rats showed no signs of neurotoxicity after acute exposure to 182 mgtkglday (Short and Lee 1980). However, when fed an extremely high dose of 10,000 mg/kg/day, both rats and mice showed signs of inactivity; some developed convulsions and died (Dilley et al. 1982b). Similar observations were made in the intermediate-duration studies in dogs, rats, and monkeys fed low doses of 2,4,6-trinitrotoluene (Dilley et al. 1982b; Martin and Hart 1974). When higher doses were used (32 mg/kg/day for 26 weeks), dogs became ataxic (Levine et al. 1990b). In rats exposed to 300 mg/kg/day of 2,4,6-trinitrotoluene for 13 weeks, brain lesions (consisting of focal vacuolation and malacia of the white tracts of the cerebellar folia) were seen in histopathological analysis (Levine et al. 1984). In contrast, no significant signs of neurotoxicity were seen in rats treated with up to 50 mg/kg/day for 24 months (Army 1984a). Since these results indicate species differences after acute, intermediate, and chronic exposures, it is difficult to estimate potential neurotoxic effects for humans living close to ammunition plants.

Reproductive Effects. A preliminary case control study of workers in two 2,4,6-trinitrotoluene plants in China indicates exposure to 2,4,6-trinitrotoluene may have adverse effects on several indicators of male reproductive status (Li et al. 1993). Workers exposed to 2,4,6-trinitrotoluene had significantly lower semen volumes and a smaller percentage of motile spermatozoa as well as a significantly higher incidence of sperm malformation than the control group. However, exposure concentrations and route of exposure are not known. Possible important variables which are not discussed include exposure to other chemicals and heat in the workplace.

Serious reproductive effects, such as testicular atrophy and atrophic seminiferous tubules, were observed in rats treated with high doses of 2,4,6-trinitrotoluene for 13 weeks (Dilley et al. 1982b; Levine et al. 1984, 1990a). These changes were not reversible after a 4-week recovery period. Based on this limited information, adverse reproductive effects after exposure of males to sufficiently high concentrations of 2,4,6-trinitrotoluene cannot be excluded.

Developmental Effects. No studies were located regarding developmental effects in humans or animals following exposure to 2,4,6-trinitrotoluene by any exposure route. It is therefore not possible to predict the potential developmental toxicity of 2,4,6-trinitrotoluene at hazardous waste sites or near ammunition plants.

Genotoxic Effects. No studies were found that directly assess the potential of 2,4,6-trinitrotoluene to induce genotoxic effects in humans. However, there is convincing evidence that the urine of individuals occupationally exposed to 2,4,6-trinitrotoluene contains mutagenic components (Ahlborg et al. 1985, 1988a). The primary metabolite of 2,4,6-trinitrotoluene appears to be 4-aminodinitrotoluene (6ADNT); major intermediate forms, including 4-ADNT, are weakly mutagenic in bacteria (Spanggord et a1.1982b). By contrast, 2,4,6-trinitrotoluene is a confirmed mutagen in bacterial and mammalian cells in vitro and;2,4,6-trinitrotoluene-induced mutagenesis is either markedly diminished or abolished by the inclusion of exogenous metabolic activation into these test systems (see discussion of in vitro results below).

The detection of unmetabolized 2,4,6-trinitrotoluene in the urine of exposed workers exhibiting a high level of mutagenic activity (Ahlborg et al. 1985) tends to support the assumption that the parent compound rather than its derivatives was responsible for the observed response. However, a more detailed follow-up study found no correlation between mutagenicity and 2,4,6-trinitrotoluene concentration in urine (Ahlborg et al. 1988a). It is nevertheless possible that these conflicting results could be resolved if more appropriate concentration procedures improved the detection of 2,4,6-trinitrotoluene. This would elucidate the possible connection between mutagenesis and the concentrations of 2,4,6-trinitrotoluene and/or its metabolites in the urine of exposed workers.

Only one in vitro study employing a human cell line (WI-38 human fibroblasts) was found in the existing literature (Army 1978~). In this study, target cells were exposed to 2,4,6-trinitrotoluene doses ranging from 2 to 2,000 µg/mL without S9 activation and from 6 to 6,000 µg/mL in the presence of an uninduced mouse liver homogenate. Precipitation of the test material occurred at ≥200 µg/mL (without S9) and at 375 µg/mL (in the presence of S9). At nonactivated levels of 500 and 1,000 µg/mL, a significant (p<0.05) increase in UDS was obscured by discoloration of the samples; however, significant (p<0.05) effects were also obtained at 250 µg/mL. No discoloration of the test samples occurred in the S9-activated phase of testing, and no evidence of a genotoxic response was uncovered. Although a definitive conclusion could not be reached because of compound interference with the nonactivated assay results, the data do not suggest that 2,4,6-trinitrotoluene was genotoxic in this human cell line. Similarly, the lack of an effect in the presence of auxiliary metabolic activation is consistent with other in vitro assay results.

The single in vivo animal study assessing potential adverse effect on the chromosome structure of somatic cells following oral administration of 2,4,6-trinitrotoluene was negative but compromised because neither a toxic effect in the rats nor a cytotoxic effect on the target organ (i.e., bone marrow cells) was demonstrated (Army 1978c). However, the results of a well-conducted mouse micronucleus assay, which evaluated 2,4,6-trinitrotoluene at a level (80 mg/kg) that approximated 80% of the maximum tolerated dose, provided no indication of a clastogenesis (Ashby et al. 1985). For this study, groups of five male CBA x Balb C mice received single intraperitoneal injections of 40 or 80 mg/kg 2,4,6-trinitrotoluene; animals were sacrificed 24, 48, and 72 hours post-treatment, and bone marrow cells were examined for the presence of micronucleated polychromatic erythrocytes (MPEs). Results indicated that there were no significant increases in the frequency of MPEs in bone marrow cells sampled over the entire hematopoietic cycle. Similarly, the in vivo / in vitro rat liver UDS assay performed by the same investigators was negative.

As the above discussion indicates, deleterious genetic events resulting from exposure to 2,4,6-trinitrotoluene have not been extensively investigated in either humans or animals. Since the only available studies in humans were from occupational settings, inhalation must be considered an important pathway of 2,4,6-trinitrotoluene exposure. However, quantifiable dermal absorption indicates that 2,4,6-trinitrotoluene is absorbed through the skin. It also suggests that dermal absorption plays an important role in 2,4,6-trinitrotoluene uptake, which may be more important than uptake through inhalation (Ahlborg et al. 1988a). The identification of the agent(s) responsible for the mutagenic activity observed in the urine of workers exposed to 2,4,6-trinitrotoluene is of considerable importance, because human metabolism of 2,4,6-trinitrotoluene has not been fully characterized. It is possible that the parent compound, a known mutagen, and its metabolites, which are also mutagens in S. typhimurium TA98, are contributors to 2,4,6-trinitrotoluene-induced mutagenesis (Spanggord et al. 1982b).

Although only three in vivo studies were found, the overall results provided no evidence that 2,4,6-trinitrotoluene is genotoxic in whole animals. This assumption is supported by the results of the in vitro UDS assay with human cells indicating that 2,4,6-trinitrotoluene-induced UDS was abolished by the inclusion of exogenous metabolic activation (Army 1978c). Similar results, as discussed below, were obtained in other test systems using both cultured bacterial and mammalian cells.

The implications of both the whole animal and in vitro human cell assay findings are highly relevant to human health. If 2,4,6-trinitrotoluene can be reduced to nonmutagenic metabolic products, the potential health hazard to humans would be greatly reduced. Refer to Tables 2-2 and 2-3 for a further summary of these studies.

In contrast to the absence of genotoxicity in animal studies, numerous investigators (Ahlborg et al. 1985, 1988a; Army 1978a, 1978c, 1979b, 1980d; Kaplan and Kaplan 1982c; Pearson et al. 1979; Spanggord et al. 1982b; Whong and Edwards 1984; Won et al. 1976) have demonstrated that 2,4,6-trinitrotoluene is a microbial mutagen. There is good agreement that 2,4,6-trinitrotoluene primarily causes frameshift mutations in S. typhimurium TA1537, TA1538, and TA98 and that the mutagenic response is not dependent on auxiliary metabolic activation nor substantially influenced by microbial nitroreductase activity. Data also exist showing that 2,4,6-trinitrotoluene is mutagenic in S. typhimurium strains TA1535 and TA100, which detect agents that cause base-pair substitution mutations (Army 1978c, 1979b, 1980d; Whong and Edwards 1984). The weight of evidence, however, is consistent with a frameshift mutagen; appreciably higher levels of 2,4,6-trinitrotoluene (≥30 µg/plate) were required to achieve positive responses in TA1535 and TA100 as compared to the reactivity (≥2 µg/plate) of 2,4,6-trinitrotoluene with strains TA1538 and/or TA98. Metabolites of 2,4,6-trinitrotoluene have not been shown as consistently to be mutagens in S. typhimurium. While the parent compound induced a dose-related increase in mutant colonies of S. typhimurium TA98 over a concentration range of 2-10 µg/plate, the seven investigated metabolites were negative (Won et al. 1976). In a study using the same assay system, one of the tested metabolites (2-amino-4,6-dinitrotoluene) was positive, while the second one (4-amino-2,6-dinitrotoluene) was only slightly positive; both needed nitroreductase to induce mutagenicity (Spanggord et al. 1982b). Similarly, in a third study the four possible mono- and diamino metabolites of 2,4,6-trinitrotoluene were all less mutagenic than the parent compound in TA98 or TA100 (Tan et al. 1992). Mutagenicity in Salmonella tester strains seems dependent on endogenous nitroreductase activity. Strains deficient in nitroreductase show decreased sensitivity to 2,4,6-trinitrotoluene while strains constructed with increased nitroreductase activity show increased sensitivity (Einisto et al. 1991).

TABLE 2-2

Genotoxicity of 2,4,6-Trinitrotoluene In Vitro.

TABLE 2-3

Genotoxicity of 2,4,6-Trinitrotoluene In Vivo.

2,4,6-Trinitrotoluene is also capable of causing gene mutations in mammalian cells (Styles and Cross 1983). In a well-conducted study, 2,4,6-trinitrotoluene (8-l,000 µg/mL) caused dose-dependent cytotoxicity and significant increases in mutation at the TK^+/− locus in mouse lymphoma cells. In agreement with other in vitro assay findings, S9 activation was not required to demonstrate the response. 2,4,6-Trinitrotoluene was negative under conditions of exogenous metabolic activation. No studies investigating potential clastogenic effects in vitro were found.

Although the database for in vitro genetic toxicology testing with 2,4,6-trinitrotoluene is limited, a high degree of concordance exists among different assay systems. Based on the existing information, there is sufficient valid in vitro data to conclude that 2,4,6-trinitrotoluene is a direct-acting mutagen in bacterial and mammalian cells. There is also suggestive evidence that 2,4,6-trinitrotoluene is a directacting genotoxic agent in cultured human cells. Refer to Table 2-2 for a further summary of these results.

Cancer. One preliminary epidemiological study of German populations living in the proximity of former munitions plants suggests 2,4,6-trinitrotoluene may increase leukemia rates in exposed adult human populations (Kolb et al. 1993). However, the case numbers in this study are very small. Also the proximity of the cases of leukemia to sites where 2,4,6-trinitrotoluene was manufactured during World War II or to disposal sites is not reported nor are any environmental 2,4,6-trinitrotoluene concentrations. Therefore, there is only slight circumstantial evidence linking the observed leukemia cases to possible 2,4,6-trinitrotoluene exposure. Furthermore, no investigation of confounding variables (e.g., benzene exposure or occupational exposure to carcinogens) has been done. Several female Fisher-344 rats developed urinary bladder carcinoma after exposure to 10 or 50 mg/kg/day of 2,4,6-trinitrotoluene for 24 months (Army 1984a). A similar study conducted in B6C3F₁ mice showed that a statistically significant (p<0.01) incidence of leukemia and/or malignant lymphoma of the spleen was present in female mice receiving 70 mg/kg/day for 24 months (Army 1984b). On the basis of this result, EPA has classified 2,4,6-trinitrotoluene as a possible human carcinogen (Group C) (EPA 1989b).