Spacecraft Maximum Allowable Concentrations for Selected Airborne Contaminants: Volume 2 (1996)

King Lit Wong, Ph.D.

Johnson Space Center Toxicology Group

Biomedical Operations and Research Branch

Houston, Texas

Nitromethane is a colorless liquid with a fruity odor (Sax, 1984).

Nitromethane is used as a solvent, an intermediate in the chemical synthesis of pharmaceuticals and pesticides, as a rocket fuel, and a stabilizer for halogenated alkanes (EPA, 1985; ACGIH, 1986). Nitromethane is not known to be used in spacecraft, but it has been predicted to be off-gassed in the Space Station Freedom (Leban and Wagner, 1989).

No pharmacokinetic data have been found in the literature search. Because nitromethane has been found to be toxic in rabbits after oral or inhalation exposures, but not after cutaneous applications on closely clipped abdominal skin (the liquid nitromethane applied was allowed to remain on the skin until totally evaporated) (Machle et al., 1940), it could be inferred that nitromethane is absorbed via the intestinal and respiratory tracts and nitromethane might not be very well absorbed through the skin.

There are no in vivo data on nitromethane metabolism. However, a few studies have been performed with liver microsomal preparations. Incubation of nitromethane with rat liver microsomes under oxidative conditions has been shown to produce nitrite in one study (Marletta and Stuehr, 1983) and to form nitrite and formaldehyde in a stoichiometric fashion in another study (Sakurai et al., 1980). Anaerobic incubation of rat liver microsomes with nitromethane produced only formaldehyde, but not nitrite (Sakurai et al., 1980).

There is no study on the excretory pathways of nitromethane. However, nitromethane has been detected in the urine of mice not exposed to nitromethane (Miyashata and Robinson, 1980).

In in vitro systems, nitromethane is known to interact with certain heme proteins. Incubation of cytochrome P-450, myoglobin, or hemoglobin with nitromethane has been shown to form nitro complexes of these heme proteins (Mansuy, 1977a,b). Nevertheless, it is unknown whether these interactions contribute to any toxicity in the body during nitromethane exposures.

Nitromethane is known to produce certain symptoms and death in laboratory animals. However, whether nitromethane is toxic to the liver is a controversy.

The oral LD₅₀ of nitromethane in mice is 1.44 g/kg (Weatherby, 1955). There are very few inhalation studies with nitromethane. Machle et al. exposed two rabbits and two guinea pigs per group to nitromethane vapor at concentrations ranging from 0.1% to 5% in cumulative exposure durations of 0.25-140 h (the cumulative exposure durations above 6 h consisted of daily 6-h exposures) (Machle et al., 1940). The lethal responses of the two rabbits and two guinea pigs combined varied with concentration x time (C x T) (Machle et al., 1940). Closer scrutiny of the data shows that the lethal response followed Haber's rule only for exposures at concentrations of 0.10% or higher. No mortality was recorded in the group exposed cumulatively to 0.05% nitromethane for 140 h, which is equivalent to a C X T of 7 %-h. The fact that the group exposed at 0.05 % for 140 h was an outlier was apparent after the mortality data were tabulated (see Tables 10-1 and 10-2).

Without reference to specific exposure concentrations, Machle et al. (1940) reported that restlessness, slight mucosal irritation (no occular

discharge, coughing, or sneezing), slight narcosis, salivation, weakness, ataxia, incoordination, circus movements, convulsions, and twitching were produced by the exposures. In addition, the animals appeared uncomfortable or ill. They reported that it took exposures at 3% or 5% nitromethane for more than 1 h or 1% nitromethane for 5 h or more to cause CNS symptoms in rabbits and guinea pigs.

Weatherby (1955) showed that an oral administration of nitromethane at 500 mg/kg in dogs produced marked fatty changes in the periportal and midzonal regions of the liver 32 h after the administration. However, Dayal et al. (1989) reported that nitromethane injected intraperitoneally at 550 mg/kg failed to cause any liver histopathology or any changes in the plasma levels of alanine transaminase, aspartate transaminase, and sorbitol dehydrogenase in mice. Similarly, no evidence of hepatic toxicity has been found with inhalation exposure of nitromethane. Lewis et al. (1979) exposed 10 rats to nitromethane at 98 or 745 ppm, 7 h/d, for 2 d. The level of serum glutamic-pyruvic transaminase (alanine transaminase) did not increase and no histopathology was detected in various tissues, including the liver, in these animals. Therefore, there is no evidence that nitromethane, at least via inhalation exposures, is hepatotoxic.

Subchronic nitromethane exposures might result in liver changes, but

there is no evidence of any liver toxicity when animals inhaled nitromethane repetitively. Subcutaneous injections of nitromethane at 290 mg/kg every other day for 18 d in rats led to an 80% reduction of hepatic histidase activity and two- to three-fold increases in the histidine levels in the plasma and liver (Wang and Lee, 1973). There were no changes in the levels of other amino acids in the plasma and liver, the liver protein content, and the body-weight gain in these rats. Because it is doubtful whether increases in the histidine levels in the plasma and liver have any adverse health impacts, the SMACs are not set according to these biochemical changes.

Weatherby (1955) exposed male rats at 0.1% or 0.25% nitromethane in drinking water for 15 w. Cells with enlarged, prominent nuclei and cytoplasm more granular than in the controls in both the 0.1% and 0.25% groups were observed in the liver. In the rats exposed at 0.25% nitromethane, numerous lymphocytes were observed in the periportal regions of the liver. In both nitromethane groups, the light microscopic morphology of other tissues appeared normal. The National Research Council Subcommittee on SMACs doubted the clinical significance of these minor light microscopic changes in the liver. The subcommittee advised that NASA disregard the hepatic end point in setting nitromethane's SMACs, especially because Lewis et al. (1979) failed to find any increase in alanine transaminase or histopathology in the livers of rats exposed to nitromethane at a concentration of 745 ppm, 7 h/d, 5 d/w, for 10 d or 1, 3, or 6 mo. The hepatic findings of Weatherby (1955) also had doubtful meaning in inhalation exposures of nitromethane because Griffin (1990) also did not detect any increases in the level of alanine transaminase, aspartate transaminase, bilirubin or protein, or any hepatic histopathology in rats exposed to nitromethane at 100 or 200 ppm, 7 h/d, 5 d/w, for 2 y.

Exposures of rats to nitromethane have been shown to result in thyroid toxicity. An exposure of 10 rats at 98 or 745 ppm, 7 h/d, 5 d/w, for 6 mo resulted in an increase in thyroid weight by 19% or 26%, respectively, but no change in the thyroxine level in plasma (Lewis et al., 1979). In contrast, an exposure of five rabbits to nitromethane at 98 ppm, 7 h/d, 5 d/w, for 6 mo caused a 48% reduction in the thyroxine

level in plasma, but no change in the thyroid weight. However, a similar exposure of five rabbits at 745 ppm increased the thyroid weight by 38% and reduced the plasma thyroxine level by 52%. No thyroid weight or thyroxine level changes were detected in 10 rats exposed at 98 or 745 ppm, 7 h/d, 5 d/2, for 2 d, 10 d, 1 mo, or 3 mo. Similarly, no thyroid weight or thyroxine level changes were found in five rabbits exposed to nitromethane at 98 ppm, 7 h/d, 5 d/w, for 1 or 3 mo (Lewis et al., 1979). The thyroid data of Lewis et al. are summarized in Table 10-3.

Unfortunately, in the 2-y study conducted by Griffin (1990), he did not weigh the thyroid or measure the serum levels of thyroid hormones in rats exposed to nitromethane at up to 200 ppm. Despite a close to 50% reduction in thyroxine level and a 20-40% increase in thyroid weight caused by a 6-mo exposure to nitromethane at 98 or 745 ppm in rabbits, the subcommittee on SMACs advised that nitromethane's SMACs need not be set to prevent the thyroid toxicity. The subcommittee's reasoning is that the thyroid changes are not clinically significant in humans and that quite a number of chemicals known to produce thyroid changes of similar magnitude in rodents do not produce any thyroid disease in humans. It should be noted that nitromethane's effect on thyroxine is opposite to that of microgravity. Studies done by NASA showed that the blood concentrations of thyroxine and thyroid-stimulating hormone were increased on the day astronauts returned to earth after several Skylab and space-shuttle missions (Huntoon et al., 1989).

As mentioned earlier, nitromethane is metabolized in vitro to nitrite, so there is a possibility that nitromethane produces methemoglobin. Despite the fact that nitrite was found in the lung, heart, kidney, and spleen of rats exposed to nitromethane at 13,000 ppm for 6 h, no methemoglobin was detected (Dequidt et al., 1973). Similarly, Lewis et al. (1979) did not find methemoglobinemia in rats and rabbits exposed to nitromethane at 745 ppm, 7 h/d, 5 d/w, for 6 mo.

In the study of rats and rabbits exposed to nitromethane for 7 h/d, 5 d/w, the hematocrit and hemoglobin levels were reduced 5-12% in rats exposed at 745 ppm for 10 d or 1, 3, or 6 mo, but not in the rats exposed for 2 d (Lewis et al., 1979). The rabbits appeared to be less sensitive than rats to nitromethane's effects in causing anemia because a 13% reduction in hemoglobin levels was found in the rabbits only at 1 mo and not after 3 or 6 mo of repetitive exposures to nitromethane at 745 ppm. In addition, repetitive exposures of rabbits to nitromethane at 745 ppm for 1, 3, or 6 mo did not decrease the hematocrit.

There is evidence that chronic nitromethane inhalation exposures are devoid of any hematological effects at up to 200 ppm. Griffin (1990) exposed rats to nitromethane at 100 or 200 ppm, 7 h/d, 5 d/w, for 2 y and found no changes in the red-or white-blood-cell counts, hemoglobin concentration, hematocrit, and platelet count.

Lewis et al. (1979) showed that nitromethane caused pulmonary damage in one group of rabbits at one time point. They observed moderate to moderately severe focal lung hemorrhages and congestion, associated with interstitial lung edema, in rabbits exposed to nitromethane at 745 ppm, 7 h/d, 5 d/w, for 1 mo. The interstitial pulmonary edema evidently must not be severe because no increase in the wet-weight-to-dry-weight ratio was found in the lung of these rabbits. These changes in the pulmonary morphology were absent in the rabbits exposed to nitromethane at 98 ppm. No lung histopathology was detected in groups of 10 rats each exposed at 745 ppm, 7 h/d, 5 d/w, for 2 or 10 d or 1, 3, or 6 mo. The apparent pulmonary toxicity discovered in these

rabbits was not relied on in setting the SMACs for three reasons. First, only five rabbits were exposed at 745 ppm for 1 mo. For lung morphological evaluation, something highly susceptible to artifacts, more animals are required to make the result more reliable. Second, the lung changes were seen in the rabbit only at 1 mo but not at 3 or 6 mo. It is highly suspicious that the morphological changes resolved after further repetitive nitromethane exposures. Third, the lung changes have never been reported by other investigators, including Weatherby (1955), who exposed rabbits, rats, and dogs to nitromethane, and Griffin (1990), who exposed rats to nitromethane at up to 200 ppm for 2 y (Machle et al., 1940).

Nitromethane did not produce reproductive toxicity in female rats. Intraperitoneal injections of nitromethane with 0.5 mL (equivalent to 46 mg) every third day, begun 1 w before the rats were bred and continued throughout gestation, failed to produce any changes in the percentage of successful matings, litter size, pup death rate, birth weights, or maternal behavior (Whitman et al., 1977). The negative finding of reproductive toxicity in rats exposed to nitromethane does not prove that nitromethane is devoid of adverse effects on reproduction. According to the subcommittee on SMACs, the rat is not an ideal model to test for some classes of reproductive toxicity, and both male and female effects can be difficult to detect in the absence of histopathological examination. This is particularly germane in an experiment such as this where treatment began only 1 w before mating.

Although an exposure to 2-nitropropane at 207 ppm, 7 h/d, 5 d/w, for only 6 mo produced liver tumors in 10 out of 10 rats, a similar exposure to nitromethane at 745 ppm failed to show any carcinogenicity (Lewis et al., 1979). The lack of carcinogenic response to nitromethane has been proven by Griffin (1990) in rats exposed to nitromethane at 200 or 100 ppm, 7 h/d, 5 d/w, for 2 y.

Nitromethane does not appear to be genotoxic. An exposure of Salmonella typhimurium (strains TA98, TA100, and TA102) to nitromethane at up to 200 µmol per plate did not increase the mutation frequency (Dayal et al., 1989).

No data on nitromethane's developmental toxicity were found in the Toxline and Toxlit databases of the National Library of Medicine.

No data on nitromethane's interaction with other chemicals were found.

The SMACs are set to protect the astronauts against the following toxic end points: CNS symptoms and anemia. For a given exposure duration, the SMACs are set by selecting the lowest acceptable concentration (AC) among the two toxic end points. An acceptable concentration for a toxic end point for a given exposure duration is derived from the no-observed-adverse-effect level (NOAEL) for that end point and that duration. Because all the NOAELs are estimated from animal data, an interspecies factor of 10 is applied on all of them. Other than the AC for anemia, no microgravity safety factor is needed for other toxic end points because they are not known to be affected by microgravity.

Machle et al. (1940) reported that inhalation exposures of rabbits or

guinea pigs led to slight mucosal irritation, but they did not report the exposure concentration at which they detected mucosal irritation in the animals. Similarly, both the ACGIH (1986) and Lewis et al. (1979) commented that nitromethane is a mucosal irritant not supported by any quantitative data. The SMACs are not set based on mucosal irritation because there are no data on the irritating concentrations of nitromethane. Lewis et al., however, did state that the current permissible exposure limit (PEL) of 100 ppm is sufficiently low to prevent respiratory irritation based on animal data and industrial health experience. Because the SMACs are set below 100 ppm, it appears that mucosal irritation will not be a problem if Lewis et al. are correct.

Machle et al. (1940) also stated, without reference to any specific exposure concentration, that nitromethane could produce restlessness, narcosis, ataxia, incoordination, and convulsions in the rabbits and guinea pigs they exposed. Since Lewis et al. (1979) did not mention these symptoms in rats exposed to nitromethane at 98 or 745 ppm in a 90-d study, the NOAEL for CNS symptoms is estimated to be 745 ppm.

The 180-d AC based on CNS symptoms is set equal to the 30-d AC because if CNS symptoms fail to develop in an exposure at 75 ppm for 30 d, it is highly unlikely that the symptoms will occur when the exposure is extended to 180 d. The reason is that the CNS should equilibrate with nitromethane in blood in 30 d.

Exposures of rats to nitromethane at 98 ppm, 7 h/d, 5 d/w, caused no anemia in 2 d to 6 mo, but a similar exposure at 745 ppm produced anemia, as shown in Table 10-7, in 10 d to 6 mo (Lewis et al., 1979).

The 7% increase in the RBC count seen in the rat after 2 d of exposure to nitromethane at 745 ppm (Lewis et al., 1979) is not considered important because it did not repeat itself in all the later time points. Be-

cause no anemia was found in the rat after a 2-d exposure at 745 ppm, 745 ppm is chosen to be the 1-h NOAEL for anemic effects.

To derive the acceptable concentrations based on nitromethane's induction of anemia, a microgravity safety factor of 3 is applied because microgravity is known to reduce the RBC mass by 10-20% (Huntoon et al., 1989). Three is chosen because 5 has been used for microgravity-induced arrhythmia in the derivation of the ACs of other compounds. The anemic effect is not as serious as the arrhythmic effect for the reason that anemia is usually not life threatening, and some arrhythmia could be. So a smaller safety factor is justified for the anemic effect than that for the arrhythmic effect.

With the microgravity safety factor and the traditional interspecies extrapolation factor, the 1-h and 24-h ACs are derived from the fact that a 2-d exposure of rats to nitromethane at 745 ppm at 7 h/d did not produce anemia in the study by Lewis et al. (1979).

Haber's rule is used to derive the 24-h AC.

Lewis et al. (1979) showed that an exposure to nitromethane at 98 ppm, 7 h/d, 5 d/w, for 10 d, 1 mo, or 6 mo failed to produce any changes in the hematocrit, hemoglobin concentration, and RBC count in rats. However, based on Griffin's (1990) finding that an exposure of rats to nitromethane at 200 ppm, 7 h/d, 5 d/w for 2 y, did not produce any changes in the hematocrit, hemoglobin concentration, and RBC count, 200 ppm is selected to be the NOAEL for anemia in a continuous exposure lasting 7, 30, or 180 d.

From the comparison of the various ACs at each time point, the 1-h, 24-h, 7-d, 30-d, and 180-d SMACs are set at 25, 15, 7, 7, and 5 ppm, respectively. As reported in the Toxicity Summary, acute exposures of rabbits and guinea pigs to nitromethane by inhalation have been known to cause death, according to Machle et al (1940). However, those mortality data are not relied on in setting ACs because Machle et al. used only two rabbits and two guinea pigs per combination of exposure concentration and duration. It is of interest, nevertheless, to assess whether exposures at the short-term SMACs are likely to be lethal using the benchmark dose approach as follows (Beck et al., 1993).

The mortality data of rabbits and guinea pigs are combined and grouped together in terms of C x T, which is the product of exposure concentration in percent and exposure duration in hours. But the data of the outlier group (animals exposed at 0.05 % nitromethane, 6 h/d for

more than 23 d) are excluded. The mortality rates are plotted against the logarithms of C x T of the exposure groups. Via probit analysis according to Finney (1971), the lower 95% confidence limit of the LD₅₀ is estimated to be 0.49 %-h.

Because the 24-h and 1-h SMACs are lower than 20 and 490 ppm, respectively, it is highly unlikely that any deaths will result from exposures at these SMACs.

ACGIH. 1986. Nitromethane. P. 439 in Documentation of the Threshold Limit Values and Biological Exposure Indexes. American Conference of Governmental Industrial Hygienists, Akron, Ohio.

Beck, B.D., R.B. Conolly, M.L. Dourson, D. Guth, D. Hattis, C. Kimmel, and S.C. Lewis. 1993. Symposium overview. Improvements in quantitative noncancer risk assessment. Fundam. Appl. Toxicol. 20: 1-14.

Dayal, R., A. Gescher, E.S. Harpur, I. Pratt, and J.K. Chipman. 1989. Comparison of the hepatotoxicity in mice and the mutagenicity of three nitroalkanes. Fundam. Appl. Toxicol. 13:341-348.

Dequidt, J., P. Vasseur, and J. Potencier. 1973. [Experimental toxicological study of some nitroparaffins. 4. Nitromethane.] Bull. Soc. Pharm. Lillie 1973(1):29-35.

EPA. 1985. Health and Environmental Effects Profile for Nitromethane. EPA/600/X-85/116. U.S. Environmental Protection Agency. Cincinnati, Ohio.

Finney, D.J. 1971. Probit Analysis. Cambridge, U.K.: Cambridge University Press.

Griffin, T.B. 1990. Chronic Inhalation of Nitromethane. Lab. Rep. 850902. Coulston International, Inc., White Sands Research Center, Alamogordo, N.M.

Huntoon, C.L., P.C. Johnson, and N.M. Cintron. 1989. Hematology, immunology, endocrinology, and biochemistry. Pp. 222-239 in Space Physiology and Medicine, A.E. Nicogossian, ed. Philadelphia: Lea & Febiger.

Leban, M.I., and P.A. Wagner. 1989. Space Station Freedom Gaseous Trace Contaminant Load Model Development. SAE Technical Paper Series 891513. Warrendale, Pa.: Society of Automotive Engineers.

Lewis, T.R., C.E. Ulrich, and W.M. Busey. 1979. Subchronic inhalation toxicity of nitromethane and 2-nitropropane. J. Environ. Pathol. Toxicol. 2:233-249.

Machle, W., E.W. Scott, and J. Treon. 1940. The physiological response of animals to some simple mononitroparaffins and to certain derivatives of these compounds. J. Ind. Hyg. Toxicol. 22:315-332.

Mansuy, D., P. Gans, J.C. Chottard, and J.F. Bartoli. 1977a. Nitrosoalkanes as iron(II) ligands in the 455-nm-absorbing cyto-

chrome P-450 complexes formed from nitroalkanes in reducing conditions. Eur. J. Biochem. 76:607-615.

Mansuy, D., J.C. Chottard, and G. Chottard. 1977b. Nitrosoalkanes as iron(II) ligands in the hemoglobin and myoglobin complexes formed from nitroalkanes in reducing conditions . Eur. J. Biochem. 76:617-623.

Marletta, M.A., and D.J. Stuehr. 1983. Biologically-derived reactive oxygen and the oxidation of simple nitrogenous compounds. Pp. 56-61 in Oxy Radicals. Their Scavenger System. Vol. 1. Proceedings of the International Conference on Superoxide and Superoxide Dismutase, G. Cohen and R.A. Greenwald, eds. New York: Elsevier Biomedical.

Miyashata, K., and A.B. Robinson. 1980. Identification of compounds in mouse urine vapor by gas chromatography and mass spectrometry. Mech. Age. Dev. 13:177-184.

Sakurai, H., G. Hermann, H.H. Ruf, and V. Ullrich. 1980. The interaction of aliphatic nitro compounds with the liver microsomal monooxygenase system. Biochem. Pharmacol. 29:341-345.

Sax, I. 1984. P. 74 in Dangerous Properties of Industrial Materials. New York: Van Nostrand Reinhold.

Wang, M.-L., and S.-C. Lee. 1973. Alteration of plasma and liver amino acid concentrations in rats treated with nitromethane. J. Chin. Biochem. Soc. 2:25-30.

Weatherby, J.H. 1955. Observations on the toxicity of nitromethane. Arch. Ind. Health 11:102-106.

Whitman, R., B.A. Maher, and R. Abeles. 1977. Deficits in discrimination and maze learning resulting from maternal histidinemia in rats. J. Abnorm. Physchol. 86:659-661.