2,4-Dinitrotoluene

Chem. Abstr. Serv. Reg, No.: 121-14-2

Chem. Abstr. Name: 1-Methyl-2,4-dinitrobenzene

IUPAC Systematic Name: 2,4-Dinitrotoluene

Synonyms: Dinitrotoluene; 2,4-dinitrotoluol; DNT; 2,4-DNT; 4-methyl-1,3-dinitrobenzene

2,6-Dinitrotoluene

Chem. Abstr. Serv. Reg. No.: 606-20-2

Chem. Abstr. Name: 2-Methyl-1,3-dinitrobenzene

IUPAC Systematic Name: 2,6-Dinitrotoluene

Synonyms: 2,6-DNT; 1-methyl-2,6-dinitrobenzene

3,5-Dinitrotoluene

Chem. Abstr. Serv. Reg. No.: 618-85-9

Chem. Abstr. Name: 1-Methyl-3,5-dinitrobenzene

IUPAC Systematic Name: 3,5-Dinitrotoluene

Synonyms: 3,5-DNT

2,4-Dinitrotoluene

1. Description: Yellow needles from ethanol or carbon disulfide (Booth, 1991; Lide, 1993)
2. Boiling-point: 300 °C (decomposes) (Lide, 1993)
3. Melting-point: 71 °C (Lide, 1993)
4. Spectroscopy data: Infrared (prism [175, 258], grating [8017]), ultraviolet (UV) [2550], nuclear magnetic resonance (proton [3229], C-13 [4627]) and mass spectral data have been reported (Sadtler Research Laboratories, 1980)
5. Solubility: Slightly soluble in water (270 mg/L at 22 °C); soluble in acetone, benzene, diethyl ether and ethanol (Mabey et al., 1982; Lide, 1993)
6. Volatility: Vapour pressure, 0.00011 mm Hg [0.015 Pa] at 20 °C; relative vapour density (air = 1), 6.27 (Howard, 1989; Booth, 1991)
7. Stability: Combustible when exposed to heat or flame; can react with oxidizing materials. Decomposes when heated at ≥ 250 °C. Mixture with nitric acid is a high explosive. Mixture with sodium carbonate (or other alkalies) can decompose with significant increase in pressure at 210 °C (Sax & Lewis, 1989).
8. Octanol/water partition coefficient (P): log P, L98 (Hansch et al., 1995)
9. Conversion factor: mg/m³ = 7.45 × ppm¹

2,6-Dinitrotoluene

1. Description: Rhombic needles from ethanol (Lide, 1993)
2. Boiling-point: 285 °C (Howard, 1989)
3. Melting-point: 66 °C (Lide, 1993)
4. Spectroscopy data: Infrared (prism [17378], grating [676]), UV [5514], nuclear magnetic resonance (proton [895]) and mass spectral data have been reported (Sadtler Research Laboratories, 1980)
5. Solubility: Slightly soluble in water (180 mg/L at 20 °C); soluble in ethanol (Mabey et al., 1982; Lide, 1993)
6. Volatility: Vapour pressure, 0.018 mm Hg [2.4 Pa] at 20 °C (Mabey et al., 1982)
7. Octanol/water partition coefficient (P): log P, 2.10 (Hansch et al., 1995)
8. Conversion factor: mg/m³ = 7.45 × ppm¹

3,5-Dinitrotoluene

1. Description: Yellow rhombic needles from acetic acid (Lide, 1993)
2. Boiling-point: Sublimes (Lide, 1993)
3. Melting-point: 93 °C (Lide, 1993)
4. Density: 1.2772 at 11 °C/4 °C (Lide, 1993)
5. Solubility: Soluble in benzene, chloroform, diethyl ether and ethanol (Lide, 1993)
6. Stability: Combustible when exposed to heat or flame; can react with oxidizing materials. A moderate explosion hazard when exposed to heat (Sax & Lewis, 1989)
7. Octanol/water partition coefficient (P): log P, 2.28 (United States National Library of Medicine, 1995)
8. Conversion factor: mg/m³ = 7.45 × ppm¹

(a) Water

2,4-Dinitrotoluene has been detected in seawater, river water and in wastewater from 2,4,6-trinitrotoluene production. 2,4-Dinitrotoluene was found in Dokai Bay, Japan, at concentrations of up to 206 µg/L (Hashimoto et al., 1982), in water from the River Rhine in the Netherlands at 0.3 µg/L (Zoeteman et al., 1980), in water samples obtained from the River Potomac near Quantico, VA, United States, at a concentration of < 10 µg/L (Hall et al., 1987) and in Waconda Bay, Lake Chickamauga, TN, United States (range of means, < 0.10–22.1 µg/L) (Putnam et al., 1981). It has also been detected in groundwater at levels ranging from 2 to 90 500 µg/L (0.002 to 90.5 ppm) near a nitroaromatic manufacturing facility in Pasadena, TX, United States (Matson, 1986), and was detected but not quantified, in groundwater (one sample) at the Hawthorne Naval Ammunition Depot, NV, United States (Pereira et al., 1979).

2,4-Dinitrotoluene has also been found in condensate wastewater from 2,4,6-trinitrotoluene manufacture at an unspecified concentration (Liu et al., 1984) and in wastewater from 2,4,6-trinitrotoluene production at an average concentration of 9700 µg/L in 29/54 samples (Spanggord & Suta, 1982). It has been detected in effluents from coal mining (18 µg/L in 1/49 samples), iron and steel manufacture (530 µg/L in 1/5 samples), aluminium forming (77 µg/L in 1/2 samples), foundries (mean, 26 µg/L; range, 7–50 µg/L in 4/4 samples) and organic chemical manufacture (mean, 14 000 µg/L in 4 samples) (United States Environmental Protection Agency, 1980; Howard, 1989).

2,4-Dinitrotoluene concentrations of 3.1 and 13.0 µg/L were detected in surface-water samples collected from two brooks near Hischagen/Waldhof, Germany, which were close to sites used for Second World War munitions manufacture; the river into which the brooks fed (River Losse) was found to have a concentration of 0.5 µg/L. Two ponds in the Clausthal-Zellerfeld region of Germany, again near sites of previous munitions manufacture, had levels of 1.2 and 0.8 µg/L; the ponds feed into the River Oder, which was found to have a level of 0.02 µg/L. Concentrations at three locations (Brunsbüttel, Brokdorf, Lauenburg) of the River Elbe ranged from 0.1 to 1.3 µg/L (Feltes et al., 1990).

Concentrations of 2,4-dinitrotoluene ranged from 700 to 1180 µg/L in groundwater samples collected near a former explosives factory in Elsnig, Germany (Steuckart et al., 1994).

In wastewaters generated in the manufacture of 2,4,6-trinitrotoluene, 2,4-dinitrotoluene was found in all 54 samples at concentrations ranging from 40 to 48 600 µg/L over a 12-month sampling period (Spanggord et al., 1982a).

In 1993, estimated quantities of 2,4-dinitrotoluene released into the environment by industrial facilities in the United States were 850 kg into air and 150 kg into water (United States National Library of Medicine, 1995).

2,6-Dinitrotoluene has also been detected in seawater, in raw wastewater from a textile plant and in wastewater from 2,4,6-trinitrotoluene production. It was found in Dokai Bay, Japan, at concentrations ranging from “not detected” to 14.8 µg/L (Hashimoto et al., 1982). It was found at concentrations ranging from 1.3 to 38.7 µg/L (mean, 19.4 µg/L) at Waconda Bay, Lake Chichamauga, TN, United States (Putnam et al., 1981) and was detected at concentrations ranging from not detected to 76 800 µg/L (mean, 16 763 µg/L) in groundwater near a nitroaromatic plant in Pasadena, TX, United States (Matson et al., 1986). The concentration of 2,6-dinitrotoluene in raw wastewater from a textile plant was 50 µg/L in one sample (Rawlings & Samfield, 1979) and averaged 4300 µg/L in 12/54 samples in wastewater from 2,4,6-trinitrotoluene production (Spanggord & Suta, 1982). It was also detected (5 µg/L) in wastewater from a nitrobenzene plant (Shafer, 1982). It was detected in effluents from coal mining (30 µg/L in 1/49 samples), iron and steel manufacture (range, 47–140 µg/L in 2/8 samples), nonferrous metals manufacture (max., 16 µg/L), foundries (mean, 20 µg/L in 6/6 samples; range 4–50 g/L), organic chemical manufacture (mean, 3800 µg/L in 4 samples), paint and ink formulations (max., 10 µg/L) and textile mills (54 µg/L in 1 sample) (United States Environmental Protection Agency, 1980; Howard, 1989).

2,6-Dinitrotoluene concentrations of 4.1 and 7.6 µg/L were detected in surface-water samples collected from two brooks near Hischagen/Waldhof, Germany, in the vicinity of munitions manufacture during the Second World War; the river into which the brooks fed (River Losse) had a concentration of 0.1 µg/L. Two ponds in the Clausthal-Zellerfeld region of Germany, again near previous munitions manufacture, had levels of 0.07 and 0.3 µg/L; the ponds feed into the River Oder which had a level of 0.02 µg/L. Concentrations at three locations (Brunsbüttel, Brokforf, Lauenburg) of the River Elbe ranged from 0.04 to 0.5 µg/L.

2,6-Dinitrotoluene has also been detected in seawater, in raw wastewater from a textile plant and in wastewater from 2,4,6-trinitrotoluene production. It was found in Dokai Bay, Japan, at concentrations ranging from “not detected” to 14.8 µg/L (Hashimoto et al., 1982). It was found at concentrations ranging from 1.3 to 38.7 µg/L (mean, 19.4 µg/L) at Waconda Bay, Lake Chichamauga, TN, United States (Putnam et al., 1981) and was detected at concentrations ranging from not detected to 76 800 µg/L (mean, 16 763 µg/L) in groundwater near a nitroaromatic plant in Pasadena, TX, United States (Matson et al., 1986). The concentration of 2,6-dinitrotoluene in raw wastewater from a textile plant was 50 µg/L in one sample (Rawlings & Samfield, 1979) and averaged 4300 µg/L in 12/54 samples in wastewater from 2,4,6-trinitrotoluene production (Spanggord & Suta, 1982). It was also detected (5 µg/L) in wastewater from a nitrobenzene plant (Shafer, 1982). It was detected in effluents from coal mining (30 µg/L in 1/49 samples), iron and steel manufacture (range, 47–140 µg/L in 2/8 samples), nonferrous metals manufacture (max., 16 µg/L), foundries (mean, 20 µg/L in 6/6 samples; range 4–50 g/L), organic chemical manufacture (mean, 3800 µg/L in 4 samples), paint and ink formulations (max., 10 µg/L) and textile mills (54 µg/L in 1 sample) (United States Environmental Protection Agency, 1980; Howard, 1989).

2,6-Dinitrotoluene was found in all 54 samples taken from wastewaters generated in the manufacture of 2,4,6-trinitrotoluene at a concentration ranging from 60 to 14 900 µg/L over a 12-month sampling period (Spanggord et al., 1982a).

The concentration of 2,6-dinitrotoluene ranged from not detected to 510 µg/L in groundwater samples collected near a former explosives factory in Elsnig, Germany (Steuckart et al., 1994).

In 1993, estimated quantities of 2,6-dinitrotoluene released into the environment by industrial facilities in the United States were 210 kg into air and 100 kg into water (United States National Library of Médecine, 1995).

3,5-Dinitrotoluene has been detected at concentrations ranging from 162 to 339 µg/L in condensate effluent resulting from the production of 2,4,6-trinitrotoluene (Spanggord & Suta, 1982).

In wastewaters generated in the manufacture of 2,4,6-trinitrotoluene, 3,5-dinitrotoluene was found in 51/54 samples with a range of 140µ6480 µg/L over a 12-month sampling period (Spanggord et al., 1982a).

(b) Soil and sediments

In the United States, both 2,4- and 2,6-dinitrotoluenes were detected in one of two soil samples taken near the Buffalo River at the former site of a dye manufacturing plant (Nelson & Hites, 1980). 2,4- and 2,6-Dinitrotoluenes were detected in sediment from Waconda Bay, Lake Chickamauga, TN, at concentrations ranging from < 2.5 to ≤ 7.9 µg/kg and < 13–17 µg/kg, respectively (Putnam et al., 1981).

Concentrations of 2,4- and 2,6-dinitrotoluenes ranged from 3.4 to 226 mg/kg and 0.2 to 12.1 mg/g, respectively, in soil samples collected from two explosive ordnance sites in Mississippi and Alaska, United States (Jenkins & Walsh, 1992).

3.1.1. Mouse

Groups of 50 male and 50 female B6C3F1 mice, approximately six weeks of age, were fed diets containing 0 (control), 0.008 or 0.04% [0, 80 or 400 mg/kg diet (ppm)] ‘practical-grade’ 2,4-dinitrotoluene [purity stated to be > 95% by the supplier] for 78 weeks. This was followed by 13 weeks of control diet for all groups, after which time the mice were killed and subjected to complete histopathological examination. Studies with the low and high doses of 2,4-dinitrotoluene were run at different times, and, therefore, each had its own control group. Mean body weights of treated mice were reduced compared to controls during the study; final body-weight reductions were 9% and 18% in low- and high-dose males and 11% and 24% in low- and high-dose females, respectively. Survival was 78% in high-dose males compared to 74% in concurrent controls and 90% in low-dose males compared to 82% in concurrent controls. The respective figures in females were 72% at the high dose (concurrent controls, 70%) and 84% at the low dose (concurrent controls, 78%). No increase in tumour incidence was reported for any site in treated mice (United States National Cancer Institute, 1978).

In a screening assay based on increased multiplicity and incidence of lung tumours in a strain of mice highly susceptible to the development of this neoplasm, groups of 26 male and 26 female strain A mice, six to eight weeks old, were given 2,4-dinitrotoluene (at a purity of 92–95%, with the major impurity being 2,6-dinitrotoluene) in tricaprylin by gavage twice a week for 12 weeks (total doses, 0, 1200, 3000 and 6000 (maximal tolerated dose, MTD) mg/kg bw). Surviving mice were killed 18 weeks after the last treatment and examined for the gross appearance of lung tumours. Survival was 45/50 in controls, 47/52 at the low dose, 48/52 at the mid dose and 44/52 at the high dose. There was no increase in lung tumour incidence or in the number of lung tumours per mouse when compared to controls. The incidence of lung tumours in survivors was 27% in controls, 28% at the low dose, 31% at the mid dose and 23% at the high dose (Stoner et al., 1984).

After two weeks of acclimatization, groups of 38 male and 38 female weanling Charles River CD-1 mice received diets containing 0, 0.01%, 0.07% and 0.5% [100, 700 and 5000 mg/kg diet (ppm)] 2,4-dinitrotoluene (98.5–99% 2,4-dinitrotoluene, 1–1.5% 2,6-dinitrotoluene; Lee et al., 1985) for 24 months. After 12 months, eight males and eight females from each group were killed and necropsied. The remaining mice were killed at 24 months. Estimates of 2,4-dinitrotoluene intake for controls and low-, mid- and high-dose mice were 0, 14, 95 and 898 mg/kg bw per day, respectively. Body weights of low- and mid-dose females did not differ from those of controls. After three months, weight gains of mid-dose males were lower than those of controls [approximately 10%]; high-dose males and females were also lighter than controls [by approximately 22–35%]. Survival of high-dose mice began to decrease after month 8; median survival was reached in 21, 21, 19 and 9 months for control, low-, mid- and high-dose males and 20.5, 19, 20.5 and 10 months for control, low-, mid- and high-dose females, respectively. Tumours, described as cystic adenoma, cystic papillary adenoma, cystic papillary carcinoma and solid carcinoma, were observed in the renal tubular epithelium. The incidence of these tumours in males was 0/24 in controls, 6/22 at the low dose, 16/19 at the mid dose and 10/29 at the high dose; 2/8 high-dose males examined at one year also had a renal tumour. One of 23 high-dose females developed a renal tumour at 23 months (Hong et al., 1985).

3.1.2. Rat

Groups of 50 male and 50 female Fischer 344 rats, approximately six weeks of age, were fed diets containing 0, 0.008 (0.0075% for the first 19 weeks) or 0.02% [0, 80 (75) or 200 mg/kg diet (ppm)] ‘practical-grade’ 2,4-dinitrotoluene [purity stated to be > 95% by the supplier] for 78 weeks followed by control diet for 26 weeks. After this time they were killed and subjected to complete histopathological examination. The studies with low- and high-dose rats were run at different times and, therefore, each had its own control group. Mean body weights of treated rats were reduced compared to controls during the study; compared to their concurrent controls, body-weight reductions were very slight in low-dose males, 20–25% in high-dose males, variable in low-dose females and up to 18% in high-dose females. Survival was 58% in high-dose males compared to 52% in concurrent controls and 58% in low-dose males compared to 64% in concurrent controls. The respective figures in females were 52% at the high dose (concurrent controls, 48%) and 62% at the low dose (concurrent controls, 62%). No integumentary tumours were seen in controls, but the incidences of these tumours were increased in low- and high-dose males — fibromas of the skin/subcutaneous tissue occurred in 7/49 low-dose and 13/49 high-dose males; lipomas occurred in 3/49 high-dose males; fibrosarcomas occurred in 1/49 low-dose and 2/49 high-dose males, and squamous-cell papillomas and basal-cell carcinomas each occurred in single low-dose males. In treated females, the incidence of fibroadenomas of the mammary gland was increased (23/50 at the high dose versus 4/23 in controls; p < 0.016, Fisher's exact test) (United States National Cancer Institute, 1978).

After two weeks acclimatization, groups of 38 male and 38 female weanling Charles River CD rats received diets containing 0%, 0.0015%, 0.01% and 0.07% [15, 100 and 700mg/kg diet (ppm)] 2,4-dinitrotoluene (98.5–99% 2,4-dinitrotoluene, 1–1.5% 2,6-dinitrotoluene) for up to 24 months. After 12 months, eight males and eight females from each group were killed and necropsied. The remaining rats were killed at 24 months. Estimates of 2,4-dinitrotoluene intake for control, low-, mid- and high-dose males were 0, 0.57 ± 0.02, 3.92 ± 0.15 and 34.5 ± 0.8 mg/kg bw per day, respectively, and those for females were 0, 0.71 ± 0.02, 5.14 ± 0.18 and 45.3 ± 1.4 mg/kg bw per day, respectively. Body weights of low-dose rats did not differ from those of controls. Weight gains of mid-dose rats were lower after month 9 [by as much as approximately 12% in males and females]. [High-dose rats were up to approximately 33–38% lighter than controls.] Survival of low- and mid-dose rats did not differ from that of controls, whereas all high-dose males and all but one high-dose females died before the end of the study, with 50% mortality being reached between 19 and 20 months for each sex. [Survival was approximately 40–45% for the controls.] At 12 months, neoplastic nodules of the liver were diagnosed in 1/8 low-dose males, 6/7 high-dose males and 7/8 high-dose females with hepatocellular carcinoma in 1/8 high-dose females. None was found in the other groups. In rats surviving longer than 12 months, the incidences of hepatocellular neoplastic nodules and hepatocellular carcinomas were as follows: males — hepatocellular neoplastic nodules: 1/25 in controls, 2/27 at the low dose, 1/19 at the mid dose, 2/27 at the high dose; hepatocellular carcinomas: 1/25, 0/27, 1/19, 6/27, respectively [p = 0.06; Fisher's exact test]; females — hepatocellular neoplastic nodules: 0/23 in controls, 2/28 at the low dose, 2/26 at the mid dose, 5/25 at the high dose; hepatocellular carcinomas: 0/23, 0/28, 1/26, 10/25, respectively [p = 0.03; Fisher's exact test]. The incidences of mammary gland tumours (predominantly fibroadenomas) in females were 11/23 in controls, 1/28 at the low dose, 16/26 at the mid dose and 21/25 at the high dose [p < 0.05; Dunnett's test]; the incidences of skin tumours (mostly fibromas) in males were 2/25 in controls, 4/27 at the low dose, 3/19 at the mid dose and 15/27 at the high dose [p < 0.05; Dunnett's test] (Lee et al., 1985).

Groups of 28 male CDF (Fischer 344)/CrlBR rats, weighing 130–150 g and having been in quarantine for four weeks, received a control diet or a diet containing a sufficient quantity of 2,4-dinitrotoluene [‘purified’, but purity unspecified] to provide a dose of 27 mg/kg bw per day for 52 weeks. The rats were then killed and the livers and lungs evaluated histopathologically. Body-weight gain of the treated group was about 25% less than that of the controls at the end of the study. No early death was indicated. Hepatocyte degeneration and vacuolization were apparent in the majority of treated animals; acidophilic and basophilic foci were reported in 70% and 10% of the livers of treated rats, respectively. One treated rat had a hepatic neoplastic nodule and no liver tumour was noted in the control group (Leonard et al., 1987). [The Working Group noted the small numbers of animals used, the short duration of the study and the incomplete histopathology.]

3.1.1. Mouse

In a screening assay based on increased multiplicity and incidence of lung tumours in a strain of mice highly susceptible to the development of this neoplasm, groups of 26 male and 26 female strain A mice, six to eight weeks old, were given 2,6-dinitrotoluene (at a purity of 98%) in tricaprylin twice a week by gavage for 12 weeks (total doses, 0, 1200, 3000 and 6000 (MTD) mg/kg bw). Surviving mice were killed 18 weeks after the last dose and examined for the gross appearance of lung tumours. Survival was 45/50 in controls, 49/52 at the low dose, 50/52 at the mid dose and 38/52 at the high dose. There was no increase in lung tumour incidence or in the number of lung tumours per animal when compared to controls. The incidence of lung tumours in surviving mice was 27% in controls, 1.8% at the low dose, 22% at the mid dose and 34% at the high dose (Stoner et al. 1984).

3.1.2. Rat

In a study designed to evaluate the influence of pectin-induced changes in gut microflora on the hepatocarcinogenicity of 2,6-dinitrotoluene, groups of 30 male CDF (Fischer 344)/CrlBR rats, weighing 130–150 g and having been in quarantine for two weeks, were placed on one of three diets containing sufficient quantities of 2,6-dinitrotoluene (at a purity of 99.9%) to produce daily doses of 0, 0.6–0.7 or 3–3.5 mg/kg bw. Ten animals from each treatment group were killed at three, six and 12 months and livers were evaluated histopathologically. The diets used were open formula, cereal-based NIH-07, purified AIN-76A and AIN-76A with 5% added pectin. Groups receiving the high dose of 2,6-dinitrotoluene gained about 10% less weight than the respective control groups. No early death was reported. The number and size of γ-glutamyl transpeptidase (γ-GT)-staining foci increased in a time- and dose-dependent manner in animals given 2,6-dinitrotoluene in the NIH-07 diet. The group on the NIH-07 diet receiving the high dose of 2,6-dinitrotoluene exhibited a 100% incidence of liver tumours (6/10 with hepatocellular carcinomas and/or 6/10 with neoplastic nodules) at 12 months. Three of 10 rats receiving the low dose in NIH-07 diet had neoplastic nodules at 12 months. No tumour was observed in rats receiving the control diets or 2,6-dinitrotoluene in the AIN-76A diet with or without added pectin (Goldsworthy et al., 1986). [The Working Group noted that pectin did not influence the tumour outcome in this experiment].

Groups of 28 male CDF (Fischer 344)/CrlBR rats, weighing 130γ150 g and having been in quarantine for four weeks, received a control diet or a diet containing a sufficient quantity of 2,6-dinitrotoluene [‘purified’, but purity unspecified] to provide doses of 7 or 14 mg/kg bw per day for 52 weeks. After this time, the rats were killed and the livers and lungs evaluated histopathologically. Body-weight gains of the low-dose and high-dose groups were about 18% and 32% less than those of the controls at the end of the study. No early death was indicated. Hepatocyte degeneration and vacuolization were apparent in the majority of treated animals; acidophilic and basophilic foci were reported in over 90% of treated rats. No liver tumour was noted in the control group. Neoplastic nodules were found in 18/20 rats at the low dose and 15/19 at the high dose. Hepatocellular carcinomas, described as trabecular, occurred in 17/20 at the low dose and 19/19 at the high dose, and one tumour described as an adenocarcinoma was found in a low-dose rat. Cholangiocarcinomas occurred in 2/20 low-dose rats. Liver tumours metastasized to the lung in 3/20 rats at the low dose and 11/19 at the high dose (Leonard et al., 1987).

3.1.1. Mouse

In a screening assay based on increased multiplicity and incidence of lung tumours in a strain of mice highly susceptible to the development of this neoplasm, groups of 26 male and 26 female strain A mice, six to eight weeks old, were given 2: 1 mixtures of 2,4-dinitrotoluene (at a purity of 92–95%, with the major impurity being 2,6-dinitrotoluene) and 2,6-dinitrotoluene (at a purity of 98%) in tricaprylin twice a week by gavage for 12 weeks (total doses, 0, 1200, 3000 and 6000 (MTD) mg/kg bw). Surviving mice were killed 18 weeks after the last dose and were examined for the gross appeareance of lung tumours. Survival was 45/50 in controls, 48/52 at the low dose, 48/52 at the mid dose and 48/52 at the high dose. There was no increase in lung tumour incidence or in the number of lung tumours per mouse when compared to controls. The incidence of lung tumours in survivors was 27% in controls, 35% at the low dose, 35% at the mid dose and 33% at the high dose (Stoner et al., 1984).

3.1.2. Rat

Popp and Leonard (1982), Rickert et al. (1984) and Leonard et al. (1987) reported various aspects of a study on the carcinogenicity of technical-grade dinitrotoluene (76.5% 2,4-dinitrotoluene, 18.8% 2,6-dinitrotoluene, 2.43% 3,4-dinitrotoluene, 1.54% 2,3- dinitrotoluene, 0.69% 2,5-dinitrotoluene and 0.04% 3,5-dinitrotoluene) in Fischer 344 rats. In Rickert et al. (1984), data that outlined the incidence of hepatic neoplastic lesions in Fischer 344 rats at all sample intervals were presented in tabular form. The compound was given in the diet at doses of 0, 3.5, 14 and 35 mg/kg bw per day. Due to the mortality in animals at the high-dose level, the final kill of these animals was carried out at 55 weeks. At the high-dose level, the incidences of hepatocellular carcinomas in males were 2/10 at 26 weeks, 10/10 at 52 weeks and 20/20 at 55 weeks, and in females were 0/10, 4/10 and 11/20, respectively; the incidences of neoplastic nodules in males were 0/10 at 26 weeks, 3/10 at 52 weeks and 5/20 at 55 weeks, and in females were 0/10, 8/10 and 12/20, respectively. In the mid-dose group sampled at 26, 52, 78 and 104 weeks, the incidences of hepatocellular carcinomas in males were 0/10, 3/10, 19/20 and 22/23, and in females were 0/10, 0/10, 10/20 and 41/63, respectively; similarly, the incidences of neoplastic nodules were 0/10, 4/10, 11/20 and 16/23 in males, and 0/10, 0/10, 0/20 and 53/69 in females. In the low-dose group, no liver tumour was observed in any animal at 26 or 52 weeks. At 104 weeks, the incidences of hepatocellular carcinomas were 9/70 and 12/61 in males and females, respectively (neoplastic nodules, 11/70 and 12/61, respectively). In control animals, a single hepatocellular carcinoma was observed in males (1/61) at 104 weeks, at which time neoplastic nodules were also seen in 9/61 males and 5/57 females.

Groups of 28 male CDF (Fischer 344)/CrlBR rats, weighing 130–150 g and having been in quarantine for four weeks, received a control diet or a diet containing a sufficient quantity of technical-grade dinitrotoluene (76.5% 2,4-dinitrotoluene, 18.8% 2,6-dinitrotoluene, 2.43% 3,4-dinitrotoluene, 1.54% 2,3-dinitrotoluene, 0.69% 2,5- dinitrotoluene and 0.04% 3,5-dinitrotoluene) to provide a dose of 35 mg/kg bw per day for 52 weeks. After this time the rats were killed and the livers and lungs were evaluated histopatho-logically. Body-weight gain of the treated rats was about 26% less than that of the controls at the end of the study. Hepatocyte degeneration and vacuolization were apparent in the majority of treated animals; acidophilic and basophilic foci were reported in over 90% of treated animals. No liver tumour was noted in the control group. Neoplastic nodules were found in 10/19 treated rats, hepatocellular carcinomas (trabecular) in 9/19 and cholangiocarcinomas in 2/19 (Leonard et al., 1987).

(a) Whole animals

A scheme for the metabolism of 2,6-dinitrotoluene is presented in Figure 1.

Figure 1.

Metabolism of 2,6-dinitrotoluene

Male Wistar rats (200 g) were given an oral dose of 22.2 mg/kg bw 2,4-dinitro-[³H]-toluene in salad oil [not identified] (Mori et al., 1978). Radioactivity in blood and liver reached a peak at 6 h after administration with an elimination half-life from blood of 22 h [liver elimination half-life unspecified]. Approximately 10% of the radioactivity administered was excreted in the bile within 24 h. After seven days, about 46% of the radioactivity was excreted in urine or faeces (Mori et al., 1977).

The distribution of uniformly [¹⁴C]-ring-labelled 2,4-dinitrotoluene after oral gavage in corn oil was investigated in male Fischer 344 rats (80–90 days old) at three doses (10, 35 and 100 mg/kg bw) and in females at a dose of 100 mg/kg bw (Rickert & Long, 1980). In male rats, terminal elimination half-lives of total radioactivity ranged from 61 (for the two lowest doses) to 27 h (for the high dose) in plasma and from 51 to 36 h in livers, respectively. Terminal elimination half-lives of radioactivity were similar in male and female rats; however, in female livers, concentrations of radioactivity were only one-half of those found in males. The tissues of both sexes contained unmetabolized 2,4-dinitrotoluene, 2,4-dinitrobenzoic acid and 2-amino-4-nitrobenzoic acid. No evidence of the presence of 2,4-diaminotoluene was found in any tissue examined.

Rickert and Long (1981) also investigated the metabolism and excretion of 2,4-dinitrotoluene in a separate study of male and female Fischer 344 rats given 10, 35 or 100 mg/kg bw uniformly [¹⁴C]-ring-labelled 2,4-dinitrotoluene by oral gavage in corn oil. In both males and females, urine was the major route of excretion of 2,4-dinitrotoluene metabolites at all doses: 4-(N-acetyl)amino-2-nitrobenzoic acid, 2,4-dinitrobenzoic acid, 2-amino-4-nitrobenzoic acid and 2,4-dinitrobenzyl alcohol glucuronide were the major metabolites, comprising > 85% of the radioactivity excreted in rat urine. Both male and female rats showed dose-dependent changes in urinary excretion of 2,4-dinitrotoluene metabolites, with a smaller percentage of the dose being excreted in urine at the higher concentrations. The most significant sex-dependent difference was the greater percentage of 2,4-dinitrotoluene excreted as 2,4-dinitrobenzyl alcohol glucuronide by female rats after the 10- or 35-mg/kg dose. The finding that urine was the major route for elimination of ¹⁴C is in agreement with a report by Lee et al. (1975); however, Mori et al. (1978) found that the faeces were the major route of 2,4-dinitrotoluene excretion. This difference among studies may be due to strain differences or to the use of 2,4-dinitro-³H-toluene by Mori et al. (1978).

Rickert et al. (1981) explored the role of gut metabolism in the activation of 2,4-dinitrotoluene in Fischer 344 rats (80–90 days old) weighing 120–150 g (females) or 200–250 g (males). Both conventional and axenic rats received a single oral dose of 35 mg/kg bw uniformly [¹⁴C]-ring-labelled 2,4-dinitrotoluene by gavage. Throughout the study, axenic rats, which are lacking in intestinal microflora, were housed in sterile isolation units. Conventional rats were housed in typical temperature- and humidity- controlled exposure rooms. Axenic males and females excreted a smaller fraction of the 2,4-dinitrotoluene dose in the urine than did conventional animals. Most notably, amounts of 4-(N-acetyl)amino-2-nitrobenzoic acid and 2-amino-4-nitrobenzoic acid excreted by axenic animals were one-tenth to one-fifth of those excreted by conventional animals. Additionally, hepatic covalent binding was decreased by one-half in axenic animals. These data suggested to the authors that intestinal microflora play a major role in the appearance of reduced urinary metabolites and of covalently bound material after 2,4-dinitrotoluene administration.

The biliary excretion and enterohepatic circulation of 2,4-dinitrotoluene was investigated in male Fischer 344 rats (200 g) given oral doses of 35, 63 or 100 mg/kg bw uniformly [¹⁴C]-ring-labelled 2,4-dinitrotoluene in corn oil or female rats (160 g) given 35 mg/kg bw uniformly [¹⁴C]-ring-labelled 2,4-dinitrotoluene (Medinsky & Dent, 1983). The excretion of ¹⁴C in bile of male rats was related linearly to dose, with biliary elimination of ¹⁴C accounting for approximately 25% of the dose. After a dose of 35 mg/kg bw 2,4-dinitrotoluene, females excreted less ¹⁴C in the bile than males (18% of the dose). Over 90% of the radioactivity in the bile was identified as the glucuronide conjugate of 2,4-dinitrobenzyl alcohol. Biliary elimination half-lives for ¹⁴C ranged from 3.3 to 5.3 h.

Sayama et al. (1989a) also investigated the biliary excretion of 2,4-dinitrotoluene, 2,4-dinitrobenzyl alcohol and 2,4-dinitrobenzaldehyde in male Wistar rats (180–220 g). 2-Acetylamino-4-nitrotoluene, 2,4-dinitrobenzyl alcohol, 2,4-dinitrobenzaldehyde and unchanged 2,4-dinitrotoluene were detected in the nonhydrolysed neutral basic fraction of bile from rats dosed orally with 2,4-dinitrotoluene (40 mg/kg bw in salad oil [mixed vegetable oil]). The major biliary metabolite of 2,4-dinitrotoluene in male Wistar rats was 2,4-dinitrobenzyl alcohol glucuronide (11.8% of the dose). A variety of minor metabolites, most notably 2,4-dinitrobenzaldehyde (0.27%) and 4-amino-2-nitro(2-amino-4-nitro)benzyl alcohol sulfate (1.5%), were also formed. Similar metabolites were eliminated in the bile after oral administration of 2,4-dinitrobenzyl alcohol and 2,4-dinitrobenzaldehyde.

The metabolism and excretion of the isomer 2,6-dinitrotoluene was investigated by Long and Rickert (1982) in male and female Fischer 344 rats (80–90 days old, 200 and 150 g, respectively). Uniformly [¹⁴C]-ring-labelled 2,6-dinitrotoluene (> 99% radio-chemically pure) was dissolved in corn oil and administered by oral gavage (10 mg/kg bw). The major route of excretion of ¹⁴C after a single dose was via the urine (males, 53.6 ± 2.6%; females, 54.0 ± 4.8%). Faecal excretion accounted for 17.9% (males) and 19.8% (females) of the dose. Analysis of the urine by HPLC revealed three major metabolites that accounted for 95% of the urinary ¹⁴C. These metabolites were identified as 2,6-dinitrobenzoic acid, 2,6-dinitrobenzyl alcohol glucuronide and 2-amino-6-nitrobenzoic acid. No sex-dependent difference in the total amounts of the individual metabolites was noted. These results were analogous to those found after administration of 10 mg/kg bw 2,4-dinitrotoluene to male and female Fischer 344 rats. The only major difference in the disposition of the two isomers is that no N-acetylaminonitrobenzoic acid was found after administration of 2,6-dinitrotoluene in vitro. This may reflect steric hindrance to N-acetylation of an amino group adjacent to a methyl group.

Mori et al. (1989a) also reported on the metabolism of 2,6-dinitrotoluene in male Wistar rats (180–200 g). Rats were dosed orally with a solution of 2,6-dinitrotoluene (75 mg/kg bw) in 1 ml of salad oil [type unspecified]. The bile ducts of some rats were cannulated prior to dosing. The major urinary metabolite identified by chromatography with standards was 2,6-dinitrobenzyl alcohol, which represented 1.5% of the dose excreted in 24 h. The glucuronide conjugate of 2,6-dinitrobenzyl alcohol was eliminated in the bile and accounted for 30% of the dose excreted in 24 h. Small percentages (< 0.1%) of other metabolites were also detected in bile. In these studies, no 2,6-dinitrobenzoic acid was detected in the urine. [This discrepancy between the findings of Mori et al. (1989a) and those of Long and Rickert (1982) may be explained by methodological differences, or may indicate strain differences in the metabolism of 2,6-dinitrotoluene.]

Chadwick et al. (1990) examined the gastrointestinal enzyme activity and activation of 2,6-dinitrotoluene in male CD-1 mice (38.2 g) and male Fischer 344 rats (204 g) dosed orally with 75 mg/kg bw in dimethyl sulfoxide (DMSO). Mice metabolized significantly more 2,6-dinitrotoluene to mutagenic urinary metabolites than did Fischer 344 rats. Mutagenicity was evaluated by adding β-glucuronidase-hydrolysed urine samples to cultures of Salmonella typhimurium strain TA98 and quantitating revertants detected in urine. The investigators noted that native intestinal nitroreductase activity was markedly higher in the CD-1 mouse than in the Fischer 344 rat and that this correlated with the increase in revertants. In contrast to these observations, a comparison of the metabolism and binding of 2,6-dinitrotoluene in Fischer 344 rats and A/J mice indicated a higher hepatic DNA binding of 2,6-dinitrotoluene in the rats (Dixit et al., 1986). In addition, the metabolism of 2,6-dinitrotoluene was very slow in the small intestines of A/J mice (Schut et al., 1983).

The hepatic macromolecular covalent binding and intestinal disposition of 2,6-dinitrotoluene was compared with that of 2,4-dinitrotoluene (Rickert et al., 1983). Male Fischer 344 rats, 80–90 days old, received 10 or 35 mg/kg bw of each uniformly [¹⁴C]-ring-labelled isomer orally by gavage in corn oil Covalent binding of radioactivity after administration of 2,6-dinitrotoluene was always two- to five-fold higher than binding after administration of 2,4-dinitrotoluene.

Further studies comparing the hepatic macromolecular covalent binding of 2,6- and 2,4-dinitrotoluenes were conducted by Kedderis et al. (1984) in male Fischer 344 rats (180–260 g). Animals were administered intraperitoneal injections of the sulfotransferase inhibitors 2,6-dichloro-4-nitrophenol or pentachlorophenol (40 µmol/kg bw) in propane-1,2-diol or vehicle alone. Subsequently, animals were dosed orally with 2,6-dinitro[3-³H]toluene or uniformly [¹⁴C]-ring-labelled 2,4-dinitrotoluene dissolved in corn oil. At 12 h after dinitrotoluene administration, the livers were removed and processed for isolation of DNA, and urine was analysed by HPLC. Prior administration of the sulfotransferase inhibitors resulted in a significant decrease in the hepatic macro-molecular covalent binding of both isomers with the decrease being more pronounced for 2,6-dinitrotoluene. Covalent binding to hepatic DNA was reduced by > 95% (2,6-dinitrotoluene) or > 84% (2,4-dinitrotoluene). The authors suggested that metabolites formed via a sulfotransferase-dependent pathway were responsible for the majority of the covalent binding of 2,6-dinitrotoluene to hepatic DNA.

(b) In-vitro studies

(a) Single-dose studies

Data on acute toxicity following a single oral dose of five dinitrotoluene isomers in rats and mice were reported by Vernot et al. (1977). LD₅₀ values and 95% CIs were estimated using the moving-average technique. Values obtained for 2,4- and 2,6-dinitrotoluenes are presented below (Table 3). The animals used in these studies were male Sprague-Dawley rats (200–300 g) and CF-1 mice (22–28 g). The vehicle in which the materials was administered was not specified.

Table 3

Acute toxicity in rats and mice following single oral dose of dinitrotoluenes.

Male Fischer 344 rats (180–200 g) were injected intraperitoneally with 2,6-dinitrotoluene and its metabolites 2,6-diaminotoluene and 2-amino-6-nitrotoluene dissolved in DMSO at doses of 1.2 mmol/kg bw (219 mg/kg) or 0.3 mmol/kg (55 mg/kg) (La & Froines, 1993). Intraperitoneal administration of 0.3 mmol/kg 2,6-dinitrotoluene to rats resulted in 50% lethality within two days. 2,6-Diaminotoluene and 2-amino-6-nitrotoluene did not produce the death of any animal even at 1.2 mmol/kg. Histopathological examination of livers showed that 2,6-dinitrotoluene induced extensive centrilobular haemorrhagic necrosis, whereas no evidence of necrosis was observed in rat livers at either dose level of 2,6-diaminotoluene or 2-amino-6-nitrotoluene.

Studies by La and Froines (1992a) investigated the toxicity of 2,4- and 2,6-dinitrotoluenes. All rats administered 150 mg/kg bw 2,6-dinitrotoluene either intraperitoneally or by gavage died within 24 h. None of the rats administered 375 mg/kg bw 2,4-dinitrotoluene died. Only the 2,6-isomer exhibited hepatotoxicity producing extensive centrilobular haemorrhagic necrosis. According to the authors, their findings were consistent with previous studies that compared the toxicities of 2,4- and 2,6-dinitrotoluenes. For example, Leonard et al. (1987) demonstrated increased activities of γ-glutamyltransferase and alanine aminotransferase only in rats treated with 2,6-dinitrotoluene. Increased activities of these serum enzymes are indicative of liver injury. Additionally, Mirsalis and Butterworth (1982) demonstrated differences in toxicity between 2,4- and 2,6-dinitrotoluenes in vitro. Hepatocytes isolated from rats pretreated with 2,6-dinitrotoluene (100 mg/kg) did not survive in culture because of its toxicity.

(b) Repeated-dose studies

Kozuka et al. (1979) investigated the subchronic toxicity of 2,4-dinitrotoluene incorporated into a standard commercial diet at a concentration of 0.5% and fed ad libitum for six months to male Wistar rats, seven weeks old at the start of the study. The average consumption of 2,4-dinitrotoluene calculated from the amount of diet consumed was approximately 66 mg per day during the first three months and 75 mg per day in the subsequent three months. [The estimated daily dose of 2,4-dinitrotoluene ranged from 330 mg/kg bw at the beginning of the study to 500 mg/kg bw at the end.] A variety of adverse clinical signs were noted in the test group, including piloerection, whitened skin colour, humpback, incoordination, decreases in spontaneous movements and general weakness. The mortality of the treated rats was about 71% after a 26-week period. There were significant reductions in body-weight gain. The relative weights of liver, spleen and kidney were increased significantly (approximately two-fold compared with controls), while testes weights were significantly decreased. The formation of puruloid matter was noted in the livers of treated rats. The percentage of methaemoglobin was significantly increased in treated rats. At the end of the study, methaemoglobin levels of 7% were noted in the treated rats compared with 1.13% in the controls. In rats fed 0.5% 2,4-dinitrotoluene for one month, triglycerides and glucose levels in serum were elevated compared with controls. Serum enzymes (GOT, lactate dehydrogenase (LDH) and alkaline and acid phosphatase) were also increased, suggesting evidence of hepatotoxicity. No histopathology was reported.

Weanling CD rats were fed 2,4-dinitrotoluene at concentrations of 0, 0.07, 0.2 or 0.7% of the diet for 13 weeks (Lee et al., 1985). The low dose, estimated to be 34 mg/kg bw per day in males and 38 mg/kg bw per day in females, caused depressed weight gain. The mid dose, estimated to be 93 or 108 mg/kg bw per day day, was toxic and caused reticulocytosis, splenic haemosiderosis and decreased spermatogenesis. The high dose, 266 or 145 mg/kg bw per day, was progressively more toxic, causing death in all of the animals by the end of the 13-week period. Anaemia, reticulocytosis, excessive pigment deposits in the spleen and decreased spermatogenesis were the most marked findings.

Weanling CD-1 mice were fed 0, 0.07, 0.2 or 0.7% 2,4-dinitrotoluene in the diet for 13 weeks (Hong et al., 1985). The highest dose caused mild anaemia with concurrent reticulocytosis, mild degeneration of the seminiferous tubules, mild hepatocellular dysplasia and pigmentation in the Kupffer cells of the liver in males and females. Decreased erythrocyte counts, haematocrit and haemoglobin concentration were also noted.

The chronic toxicity of 2,4-dinitrotoluene was investigated in beagle dogs (Ellis et al., 1985), five to six months old at the time of study, that were given 2,4-dinitrotoluene in a hard gelatin capsule. The major adverse effect of 2,4-dinitrotoluene in dogs was a neuropathy characterized by incoordination and paralysis. Vacuolation, endothelial proliferation and gliosis of the cerebella of some of the affected dogs was observed. These effects were seen within two years in one dog given 1.5 mg/kg bw per day, within six months in all dogs given 10 mg/kg bw per day and within two months in all dogs given 25 mg/kg bw per day. Some dogs progressed to complete paralysis leading to death. Methaemoglobinaemia was also noted and the presence of Heinz bodies was common. Biliary tract hyperplasia was also noted.

In studies conducted for up to two years in which 2,4-dinitrotoluene was added to the feed of CD rats (weanlings at start of study) at doses of 0.0015, 0.01 and 0.07%, the highest dose resulted in a shortened lifespan (Lee et al., 1985). Toxic anaemia, atrophy of the testes and depression of spermatogenesis were observed. After feeding for 12 months, lesions occurred in the liver. The initial lesions were small foci or areas of altered hepatocytes referred to by the authors ‘hyperplastic foci’. In these animals, the liver architecture was preserved. The lesions progressed to ‘hyperplastic nodules’.

Weanling CD-1 mice were fed 0.01%, 0.07% or 0.5% 2,4-dinitrotoluene in the diet for up to two years (Hong et al., 1985). The authors estimated that these concentrations provided doses equal to 14, 95 or 898 mg/kg bw per day. Toxic anaemia was observed in high-dose males and females after feeding of 2,4-dinitrotoluene for 12 months. Methae-moglobin formation and the presence of numerous Heinz bodies were noted. By 21 months, all high-dose mice had died. In this group, generalized abnormal pigmentation was noted in many tissues, which was dose-dependent and increased with the length of treatment. Liver and kidney were the organs most affected. Hepatocellular dysplasia was observed. The authors use this term to characterize metabolic, degenerative and hyperplastic alterations of the cells. In males, the incidence of this lesion was increased in all three dose groups. In females, the incidence was only increased in the high-dose group.

(c) In-vitro toxicity

A reduction in the incorporation of tritiated thymidine was observed in primary cultures of aortic smooth muscle cells from dinitrotoluene-exposed animals relative to vehicle controls (Ramos et al., 1991a^,b). Male Sprague-Dawley rats (150–180 g) were injected intraperitoneally on five days a week for eight weeks with 2,4- or 2,6-dinitrotoluene (0.5, 5 or 10 mg/kg bw) or vehicle oil (control). Histopathological evaluation of aortae from animals exposed to either isomer showed dysplasia and rearrangement of aortic smooth muscle cells at all doses tested. Exposure of smooth muscle cells from naïve animals to dinitrotoluene in vitro (1, 10 or 100 µM) did not alter the extent of thymidine incorporation into DNA (Ramos et al., 1991b).

The relation of various structural parameters to hepatotoxic potential was investigated by exposing isolated rat hepatocytes from young adult male Sprague-Dawley rats (320–410 g) to six dinitrotoluene isomers (Spanggord et al., 1990). Dinitrotoluene-induced hepatotoxicity correlated with inhibition of protein synthesis and increases in lactate dehydrogenase release but not with lipid peroxidation, ortho- and para-Substituted isomers were more hepatotoxic at the same concentration than meta-substituted isomers. One group, 2,3-, 3,4- and 2,5-dinitrotoluenes, in which the nitro substituents are oriented either ortho or para to each other, was toxic at appreciably lower concentrations than the other group, which included 2,6-, 2,4- and 3,5-dinitrotoluenes in which the nitro groups are oriented meta to each other. It was concluded that the reducibility of the nitro groups of the parent chemicals is the principal factor for determining their hepatotoxic potential. The results indicated that dinitrotoluene isomers and/or their metabolites produced by the hepatocytes are directly cytotoxic to the cells without the need for metabolic activation by gut microflora. Comparison of the effects of 2,6-dinitrotoluene and 2,4-dinitrotoluene indicated that the inhibition of protein synthesis was similar between the two chemicals. However, at all concentrations tested, the release of LDH from the hepatocytes was higher after 2,6-dinitrotoluene administration compared to 2,4-nitrotoluene administration in vitro. Thus, the LDH release correlates with differences in hepatotoxicity between 2,6- and 2,4-dinitrotoluenes observed in in-vivo studies.

(a) Macromolecular adducts

2,4-Dinitrotoluene was reported to bind covalently to hepatic DNA in Fischer 344 rats after oral administration. Upon intraperitoneal injection, 2,4-dinitrotoluene showed DNA binding in liver, lung, small intestine and large intestine of mice and rats (Dixit et al., 1986). Using ³²P-postlabelling after intraperitoneal administration of 2,4-dinitrotoluene to Fischer 344 rats, three distinct adducts were detected in liver, kidney, lung and mammary gland. DNA binding was highest in the liver (La & Froines, 1992a^,b).

2,6-Dinitrotoluene was reported to bind covalently to hepatic macromolecules, including DNA, in Fischer 344 rats after oral administration and intraperitoneal injection. Twice as much ¹⁴C was found to be bound covalently to hepatic macromolecules in male than in female Fischer 344 rats after oral dosing of 10 mg/kg bw 2,6-dinitro[¹⁴C]toluene. The covalent binding to rat liver macromolecules is increased in the presence of gut microflora (Long & Rickert, 1982). Hepatic macromolecular binding was increased by feeding pectin (DeBethizy et al., 1983), Aroclor 1254 (Chadwick et al., 1993) or coal-tar creosote (Chadwick et al., 1995). After intraperitoneal injection into A/J mice, 2,6-dinitrotoluene showed DNA binding in liver but not in extrahepatic tissues such as lung, small intestine and large intestine; in rats, DNA binding was found in liver, lung and large intestine but not in small intestine (Dixit et al., 1986). Using ³²P-postlabelling after intraperitoneal administration of 2,6-dinitrotoluene to Fischer 344 rats, four distinct adducts were detected in liver, kidney, lung and mammary gland; DNA binding was highest in the liver, and the 2,6-isomer produced a greater adduct yield than the 2,4-isomer (La & Froines, 1992a, 1993).

(b) Mutation and allied effects

2,4-Dinitrotoluene (technical grade) induced both forward and reverse mutations in S. typhimurium in the presence and absence of an exogenous metabolic system. 2,4-Dinitrotoluene (technical grade) did not induce gene mutation to 6-thioguanine resistance in Chinese hamster ovary cells with or without rat liver preparations for metabolic activation. Technical-grade 2,4-dinitrotoluene, both in the presence and absence of S9-mix, caused a dose-related decrease in survival but no induction of mutation in the mouse lymphoma test. Technical-grade 2,4-dinitrotoluene did not induce morphological transformation of Syrian hamster embryo (SHE) cells.

When administered in vivo to rats, 2,4-dinitrotoluene (technical grade) induced a dose-related increase in unscheduled DNA synthesis in hepatocytes. Treated female rats showed a much lower level than males. Unscheduled DNA synthesis was observed in rats with a normal complement of gut flora, but not in rats lacking a gut flora, which indicates that metabolism by gut flora is a necessary step in the genotoxicity of this nitroaromatic compound. After in-vivo treatment with 2,4-dinitrotoluene (technical grade) by gavage, cultured rat lymphocytes showed a positive sister chromatid exchange response but no cell cycle inhibition or mitotic depression (Kligerman et al., 1982).

Technical-grade 2,4-dinitrotoluene inhibited intercellular communication at toxic concentrations in the Syrian hamster embryo cell line BPNi, but not in the V79 Chinese hamster lung cell assay.

Technical-grade 2,4-dinitrotoluene was negative in the mouse bone-marrow micro-nucleus test, the mouse dominant lethal test and the mouse spot test.

High-purity 2,4-dinitrotoluene was genotoxic in the S. typhimurium umu test. In the S. typhimurium reverse mutation assay, a number of different studies gave a heterogeneous picture. 2,4-Dinitrotoluene was negative in mutagenicity tests with the E. WP2 uvrA strain. It had little or no mutagenic activity in S. typhimurium TA98 using the standard plate test with or without rat liver S9, but was found to have a clear flavin mononucleotide-dependent mutagenic activity in a modified preincubation assay with hamster S9 (Dellarco & Privai, 1989).

In Drosophila melanogaster, high-purity 2,4-dinitrotoluene induced sex-linked recessive lethal mutations after injection, but failed to induce lethal mutations after feeding and translocations after injection.

High-purity 2,4-dinitrotoluene, at the highest dose tested, induced a small number of DNA strand breaks in rat hepatocytes in vitro.

High-purity 2,4-dinitrotoluene did not induce gene mutation to 6-thioguanine resistance in Chinese hamster ovary cells with or without rat liver preparations for metabolic activation. The induction of gene mutations was reported for mouse lymphoma cells in the absence of a metabolic activation system. 2,4-Dinitrotoluene produced a small but reproducible increase in sister chromatid exchange in Chinese hamster cells in vitro with S9, but not without S9; chromosomal aberrations were not induced with or without S9.

When high-purity 2,4-dinitrotoluene was incubated anaerobically with whole cells from rabbit intestine, culture extracts were mutagenic to S. typhimurium strain TA98NR (nitroreductase-deficient) (Combes & Walters, 1986).

High-purity 2,4-dinitrotoluene has been reported to be negative in a number of other mammalian genotoxicity assays: unscheduled DNA synthesis in primary rat hepatocytes and in human hepatocytes in vitro, the dominant lethal assay and the sperm morphology test in mice.

When administered to rats in vivo, high-purity 2,4-dinitrotoluene induced a weak response in unscheduled DNA synthesis in hepatocytes.

High-purity 2,4-dinitrotoluene inhibited intercellular communication at toxic concentrations in the Syrian hamster embryo cell line BPNi, but not in the V79 Chinese hamster lung cell assay. 2,4-Dinitrotoluene did not induce morphological transformation of Syrian hamster embryo (SHE) cells.

2,6-Dinitrotoluene is weakly mutagenic in the S. typhimurium reverse mutation test without metabolic activation and in TA1535 and TA1537 with rat liver S9. 2,6-Dinitrotoluene is metabolized by S. typhimurium strains TA98, TA98/1,8-DNP₆ and TA98NR. Results indicate that the low mutagenic activity of 2,6-dinitrotoluene is not due to low reductive metabolism by bacteria, but due to the lack of mutagenic activity of the bacterial reductive products (Sayama et al., 1992). 2,6-Dinitrotoluene had little or no mutagenic activity in S. typhimurium TA98 using the standard plate test with or without rat liver S9, but was found to have a clear flavin mononucleotide-dependent mutagenic activity in a modified preincubation assay with hamster S9 (Dellarco & Prival, 1989).

2,6-Dinitrotoluene did not induce morphological transformation of Syrian hamster embryo (SHE) cells.

2,6-Dinitrotoluene induced DNA strand breaks in rat hepatocytes in vitro.

2,6-Dinitrotoluene did not induce gene mutation to 6-thioguanine resistance in Chinese hamster ovary cells with or without rat liver preparations for metabolic activation.

Oral administration of 2,6-dinitrotoluene by gavage led to excretion of mutagenic 2,6-dinitrotoluene-derived metabolites detectable in hydrolysed urines in a microsuspension bioassay (DeMarini et al., 1989) using S. typhimurium strain TA98 without rat liver S9 activation. In a similar urinary assay, a positive response was observed with Fischer 344 rats as well as with CD-1 mice.

When administered in vivo to rats, 2,6-dinitrotoluene induced a strong response in unscheduled DNA synthesis in hepatocytes.

2,6-Dinitrotoluene inhibited intercellular communication at toxic concentrations in the Syrian hamster embryo cell line BPNi, but not in the V79 Chinese hamster lung cell assay.

3,5-Dinitrotoluene was mutagenic in the S. typhimurium reverse mutation test without metabolic activation and also in some tester strains with rat liver S9.

3,5-Dinitrotoluene did not induce gene mutation to 6-thioguanine resistance in Chinese hamster ovary cells with or without rat liver preparations for metabolic activation. It did not induce unscheduled DNA synthesis in primary rat hepatocytes in vitro.