Spacecraft Maximum Allowable Concentrations for Selected Airborne Contaminants: Volume 3 (1996)

John T. James, Ph.D.

Johnson Space Center Toxicology Group

Biomedical Operations and Research Branch

National Aeronautics and Space Administration

Houston, Texas

tert-Butanol is a crystalline solid below 25°C and a colorless, volatile liquid with a camphorlike odor above 25°C. The odor threshold is 47 ppm (Amoore and Hautala, 1983).

tert-Butanol is used in perfumes, cosmetics, aerosol sprays, paint removers, and defoaming agents; industrially, it is used in production processes, separations, and cleaning and as a gasoline additive and dehydrating agent (WHO, 1987; Lington and Bevan, 1994; NTP, 1994). Analysis of air from recent space-shuttle flights showed that this alcohol is found in about 15% of the air samples at concentrations that sometimes exceed 1 mg/m³ (James et al., 1994).

The rate of absorption of tert-butanol has not been investigated directly; however, studies with other goals suggest that the alcohol is rapidly absorbed by inhalation, intraperitoneal injection, or orally. For example, Bechtel and Cornish (1975) reported that rats given 500 mg/kg orally reached their maximum serum concentration of 450 ppm (36 mg/dL¹) in 45 to 50 min.

The tissue distribution of tert-butanol appears to be according to the water content of the tissue (Bechtel and Cornish, 1975).

Because the metabolic pattern for tert-butanol is qualitatively and quantitatively much different from primary and secondary butanols, its means of elimination differs from that of other butanols. The ''major portion'' of a 500-mg/kg oral dose given to rats was exhaled as the

unmetabolized alcohol (Bechtel and Cornish, 1975). Metabolites of tert-butanol could not be found in the blood of exposed rats, whereas metabolites of other butanols were readily detected. In rabbits given an oral dose of tert-butanol at 4 mmol/kg, 24% of the dose was excreted in the urine as a glucuronide, as compared with 14% for sec-butanol and 1.8% for n-butanol (Kamil et al., 1953).

Because of very slow metabolism (see below), the rate of elimination of tert-butanol from blood is much slower than that of most alcohols. The peak blood concentration in rabbits orally administered 2 mL/kg (1.6 g/kg) was 200 mg/dL, but detectable amounts (60 mg/dL) were present 70 h after the dose was given (Saito, 1975). Similarly, Beaugé et al. (1981) reported that female rats given 25 mmol/kg (1.8 g/kg) had blood concentrations of 13.2 mM (100 mg/dL) after 2 h and 11.3 mM (80 mg/dL) after 20 h. Gaillard and Derache (1965) reported that rats given 2 g/kg orally had a blood concentration peak of 124 mg% after 2 h, but after 8 h, the blood concentration had only decreased to 110 mg%.

The rate of tert-butanol elimination can be increased by repeated exposure to the alcohol; however, that does not occur quickly in some animal models. Thurman et al. (1980) showed that rats given oral doses of the alcohol every 8 h for 1 d required 26 h after the last dose to eliminate it from the blood to the detection limit, whereas after 2-1/2 d of exposure, the elimination time was only 18 h. McComb and Goldstein (1979a) found that to maintain a blood concentration of tert-butanol of about 6 mM (40 mg/dL) in male Swiss-Webster mice over 9 d of continuous inhalation exposure, the exposure concentration had to be increased from 50 µmol/L during day 1 to 120 µmol/L during day 9. In contrast to these findings, female C57BL mice pre-exposed to tert-butanol (five doses of 10.5 mmol/kg every 12 h) showed the same rate of elimination up to 12 h after exposure as mice that were not pre-exposed (Faulkner et al., 1989). Because Aarstad et al. (1985) showed that the cytochrome P-450 concentration in rat-liver microsomes was induced by inhalation of 2000 ppm (6 h/d, 3 d), it is reasonable to speculate that the increased rate of elimination in the rat is due in part to enzyme induction.

As would be expected, the slow metabolism of tert-butanol leads to prolonged signs of intoxication in animals. The recovery from intoxication (measured by using the tilted plane test) was slowest for tert-bu-

tanol, when compared with other butanols administered to rats, at a dose of 16 mmol/kg (1.2 g/kg) (Wallgren, 1960). Seven hours after administration of tert-butanol, the rats continued to have a performance decrement of 30%, which was the maximum decrement seen for tert-butanol.

Limited data are available on the metabolism of tert-butanol from in vivo models and in vitro systems; however, there are no data available on its metabolism in humans. Some animal models indicate that much of the alcohol is eliminated unchanged by respiration, and a lesser amount is conjugated to glucuronic acid and excreted in the kidney (Bechtel and Cornish, 1975; Kamil et al., 1953). In rats given 1 g/kg, the elimination appeared to be first order with a half-life of 9.1 h, and acetone was identified in the blood (Baker et al., 1982). Using radio-labeled doses between 0.75 and 2 g/kg, the investigators found that they could recover 0 to 9.5% (molar basis) of the dose as acetone in the urine and expired air. A small amount of labeled carbon dioxide was also eliminated from rats given a dose of 1.75 g/kg. In mice, data suggest a pseudo-zero-order elimination process after a dose of 5, 10, or 20 mmol/kg (0.4, 0.7, or 1.5 g/kg, respectively) was given by intraperitoneal injection (Faulkner and Hussain, 1989).

In vitro studies of the metabolism of tert-butanol have suggested an additional important metabolite of the alcohol. Using rat-liver microsomes, a source of cytochrome P-450-linked electron transport, to generate ·OH from H₂O₂, Cederbaum and Cohen (1980) showed that tert-butanol could be oxidatively demethylated to form formaldehyde. These investigators suggested that hydroxyl radicals are involved as follows:

That reaction would explain the production of formaldehyde as well as acetone reported by others. Later studies demonstrated the ability of tert-butanol to act as a hydroxyl radical scavenger in vitro (Cederbaum et al., 1983).

Many studies are designed to understand how tert-butanol induces physical dependency; however, they are of little value in demonstrating the toxic effects caused by the alcohol. In short-term experiments, the primary effect is CNS depression, whereas prolonged exposures target the urinary bladder and kidney. Data were found on animal exposures by various routes; however, studies using human test subjects have not been reported.

Short-term inhalation studies consist of a report that is not available in detail (Battelle, 1988a) and incidental observations during a developmental toxicity study (Nelson et al., 1989). Unsteadiness was reported in pregnant rats at the end of 7-h inhalation exposures to 2000 ppm, and ataxia was reported at the end of 3500 or 5000 ppm exposures (Nelson et al., 1989). Although the exposures were for 19 d, it is reasonable to assume that the signs of narcosis were observed during or after each acute exposure. Ataxia was noted in 10 out of 10 rats exposed to 900 ppm 6 h/d for 12 d, but mice did not exhibit ataxia until the exposures were at or above 1750 ppm (Battelle, 1988a). A single 7-h exposure to 7000 ppm caused coma in 10 of 10 rats and 10 of 10 mice (Battelle, 1988a). Target concentrations have been reported here for the 12-d Battelle study; however, because analytical concentrations were 20% above target concentrations during a 13-w study reported by the same laboratory (Battelle, 1988b), it is possible that the actual exposures during the 12-d study were 20% above target concentrations.

Acute effects have been reported in animals exposed by noninhalation routes. Ataxia (in 1 of 10) and hypoactivity (in 2 of 10) were reported in male rats given oral doses as low as 1.0% (wt/vol) in drinking water (Lindamood et al., 1992). The ND₅₀ dose (causing stupor and loss of voluntary movement in 50% of the animals) was reported to be 19 mmol/kg (1.4 g/kg) in rabbits dosed by oral gavage and the ND₅₀ was 48 mmol/kg (3.5 g/kg) (Munch, 1972). Thus, tert-butanol was found to be comparable in toxicity to other butanols, which had an ND₅₀ range of 11 to 19 mmol/kg and an LD₅₀ range of 41 to 66 mmol/kg. Other LD₅₀ values include the following: 0.9 g/kg given intraperi-

toneally in mice; 1.5 g/kg given intravenously in mice; and 3.5 g/kg given orally in rats (WHO, 1987). Microencephaly (16% decrease in brain weight) was induced in neonatal rats given tert-butanol in a milk formula on postnatal days 4 to 7 (dose, 0.6 to 2.7 g/kg); however, there were no reductions in liver or heart weights (Grant and Samson, 1982).

Even though the liver is not thought to be a target of tert-butanol, a single oral dose of 25 mmol/kg (1.8 g/kg) given to female Wistar rats induced a 3-fold accumulation of triacylglycerols in the liver 20 h after dosing (Beaugé et al., 1981). The authors seem to attribute the result to a nonspecific stress effect related to the hypothermia action of this alcohol rather than a specific action of the alcohol.

Long-term oral and inhalation studies of tert-butanol have shown that the urinary bladder and kidney are targets for injury. Considerable differences were found in the responses of different sexes and in the two species of rodents evaluated.

An abstract, data tables, and a pathologist's summary of observations were made available on a subchronic inhalation study contracted by the National Toxicology Program (NTP) and conducted in 1988 by Battelle (1988b). Male and female rats (F344) and mice (B6C3F₁) were exposed to tert-butanol at 0, 134, 270, 540, 1080, and 2100 ppm for 6 h/d for 65 d with breaks for weekends and holidays. Those average analytical concentrations appear to be about 20% above the target concentrations for tert-butanol. Measurements were taken and observations made in the following categories to detect toxic effects: weekly body weights and clinical observations, hematology, clinical chemistry (rats only), urinalysis (rats only), gross pathology, organ weights, and histopathology. The findings on rats were as follows: Exposures did not affect survival or body-weight gains; the hemoglobin, hematocrit, and red-blood-cell (RBC) counts were decreased (amount not given) in males exposed at 1080 and 2100 ppm; there were other scattered changes in clinical laboratory results and increased kidney weights were found in conjunction with increased severity of nephropathy in males (Table 3-1). The Battelle (1988b) investigation concluded that kidneys in the two highest exposure groups of male rats were affected by the

exposure. The investigators felt that the slight anemia in male rats was of little biological importance. Male rats exposed by inhalation did not show urinary-bladder lesions when subjected to exposures above 800 mg/kg/d, as reported in the 13-w drinking-water study (Lindamood et al., 1992). Nine of 10 female rats exposed at 2100 ppm showed kidney mineralization, but no such lesions were reported in the inhalation controls (Battelle, 1988b). That might have been an oversight because female control rats in the 13-w drinking-water study all showed kidney mineralization (Lindamood et al., 1992). Shifts in the relative counts of white blood cells were reported in female rats exposed by inhalation at 2100 ppm (Battelle, 1988b).

In mice the findings were as follows: the 1080- and 2100-ppm groups showed depressed gains in body weight, few if any clinical signs, no consistent changes in organ weights, and no dose-related gross or microscopic lesions. The high-dose male mice did show a leukocyte shift from lymphocytes to neutrophils and possibly an increased incidence of renal tubule regeneration (4 of 10 in the high-dose group compared with 1 of 10 in the control group) (Battelle, 1988b).

In the 13-w drinking-water study, B6C3F₁ mice and F344 rats of both sexes were given 0, 0.25, 0.5, 1.0, 2.0, and 4.0% (wt/vol) tert-butanol in their drinking water (Lindamood et al., 1992). The amount of tert -butanol consumed by the rats of both sexes averaged about 250 mg/kg/d (low dose) to 3500 mg/kg/d (high dose). In mice, the low-dose group consumed an average of 320 (males) or 570 mg/kg/d (females) and the high-dose group consumed 6200 (males) or 7500

mg/kg/d (females). All male rats and 6 of 10 female rats in the high-dose group died before the end of the study. In mice, 6 of 10 males and 4 of 10 females in the high-dose group died. Male and female rats had reduced urinary volumes in association with crystaluria (probably uric acid). Lesions consisted of urinary-tract calculi; renal, pelvic, and uretral dilatation; and thickening of the urinary-bladder mucosa. Male rats exhibited hyaline droplet accumulation in renal tubules. That is a lesion characteristic of α2u-globulin nephropathy; however, complexes with that protein were not specifically found in the kidney (Takahashi et al., 1993). The highest no-effect level for urinary-tract lesions was 800 mg/kg/d in male rats (Lindamood et al., 1992).

A 2-y chronic-exposure drinking-water study was conducted in rats and mice as a followup to the subchronic study (NTP, 1994). The dose was adjusted depending on the susceptibility of the specific species and sex of the rodents. Male F344 rats were administered the alcohol at 0, 1.25, 2.5, or 5.0 mg/mL of their drinking water and female rats were given twice those concentrations. Male and female B6C3F ₁ mice were given concentrations fourfold higher than the male rats. The average daily doses delivered are listed in Table 3-2.

The survival rates of male rats exposed at 5 mg/mL, female rats at 10 mg/mL, and male mice at 20 mg/mL were significantly below controls. Kidney mineralization and renal tubule hyperplasia were statistically increased in the high-dose group of male rats, but were not increased in any group of female rats. The incidence of transitional epithelial hyperplasia in rat kidney was statistically increased in males exposed at 2.5 and 5.0 mg/mL and in females at 10 mg/mL. The average severity of nephropathy in female rats increased from 1.6 in controls to 2.9 in the high-dose group (one, minimally; two, mildly, and three, moderately). Although the low-dose group of female rats (2.5 mg/mL or 0.175 g/kg/d), with an average severity of nephropathy at 1.9, was statistically above the controls, the slight increase over controls, in the subcommittee's opinion, does not represent an adverse effect of biological significance. The no-observed-adverse-effect level (NOAEL) for nephropathy in female rats was 0.175 g/kg/d. The incidences of thyroid-gland follicular hyperplasia were increased in all male mice and in all mid-and high-dose groups of mice. The incidences of chronic inflammation and transitional-cell hyperplasia of the urinary bladder were increased in the high-dose groups of male and female mice. In the 2-y study, male mice were the most sensitive species and sex for urinary-bladder injury. The NOAEL appeared in these mice when drinking an average of 1035 mg/kg/d. The NOAEL in the 13-w study for the same sex and species was 1570 mg/kg/d; therefore, the incidence and severity of these lesions do not increase much with prolonged exposure (i.e., from 90 to 720 d). Possible neoplastic effects of tert-butanol are reported below.

The 2-y drinking-water study (NTP, 1994) demonstrated some potential for tert-butanol to induce tumors in some of the test groups. "Some evidence of carcinogenic activity" was reported for the induction of renal tubule adenomas and carcinomas in male rats exposed at average doses of 0, 85, 195, or 420 mg/kg/d. The incidences of adenoma and carcinoma combined were as follows in the four groups (controls to high dose): 8 of 50, 13 of 50, 19 of 50, and 13 of 50, respectively. The decreased incidence in the high-dose group was presumably due to the shorter survival time, which limited the time for tumors to develop.

No evidence of carcinogenic activity was found in female rats. Equivocal evidence of carcinogenicity was reported in male mice because of marginal increases in thyroid adenomas and carcinomas combined. "Some evidence of carcinogenicity" was reported in female mice because of increased incidences of follicular-cell adenomas in the thyroid gland. The incidences reported for controls and the three exposed groups of female mice were 2 of 58, 3 of 60, 2 of 59, and 9 of 59, respectively.

Without exception, tert-butanol has been negative in tests of its genotoxic potential. It was negative in Salmonella typhimurium strains TA98, TA100, TA1535, and TA1537 for gene mutation with and without the S9 mix (Zeigler et al., 1987). tert-Butanol was negative in a mouse lymphoma mutagenicity test, with and without the S9 mix (McGregor et al., 1988). Galloway et al. (1987) has reported an overall negative result for induction of sister chromatid exchanges in Chinese hamster ovary (CHO) cells with and without S9 activation. One of the four trials (-S9) was weakly positive; the others were totally negative. Similarly, the same study reported a negative result for chromosomal aberrations in CHO cells, with and without S9 activation; however, one trial (+S9) gave an equivocal result. The percent of micronucleated peripheral blood erythrocytes was not increased in male and female mice exposed for 13 w at concentrations up to 40,000 ppm in drinking water (MacGregor et al., 1990).

Studies were not found that examined the in vivo functional capacity of the reproductive system after exposure to tert-butanol; however, histological evidence from the 2-y drinking-water study gave no indication of reproductive effects (NTP, 1994). No increased incidence of lesions was found in male rats and mice after histopathological examination of the epididymis, penis, preputial gland, prostate, seminal vesicle, and testes. Similarly negative results were found in female rats and mice after histopathological examination of the clitoral gland, ovary, uterus,

and vagina. Evaluation of sperm morphology and vaginal cytology in rats and mice from the 13-w portion of this study revealed that the only exposure-related effect was a prolonged or unclear estrus cycle in the high-dose female mice (NTP, 1994). It has been reported that, unlike ethanol, tert-butanol did not affect the in vitro fertilizing capacity of mouse spermatozoa at aqueous concentrations up to 4000 mg/L (Anderson et al., 1982).

Several studies have been published on the developmental toxicity of tert-butanol administered by various routes, including inhalation. Some of the studies were designed to answer questions about the role of acetaldehyde in the developmental toxicity induced by ethanol. The rationale was that if tert-butanol, which is poorly metabolized, were found to cause developmental toxicity, then that would be evidence that ethanol itself, rather than its metabolite (acetaldehyde), causes the developmental toxicity. The results are mixed. For example, Daniel and Evans (1982) report evidence that tert-butanol, given to mice at concentrations of 0.5, 0.75, and 1.0% (wt/vol) during gestational days 6 to 20, was 5 times more potent than ethanol in inducing delay in post-parturition physiological and psychomotor performance scores. That result was taken as evidence that ethanol, not acetaldehyde, was responsible for the developmental effects caused by excess ethanol consumption. In contrast, Faulkner et al. (1989) gave tert-butanol at a concentration of 10.5 mmol/kg to CBA/J or C57BL mice on gestational days 6 to 18 and found more resorptions per litter but no decreased body weights or malformations in the fetuses. That result was taken as indirect evidence that acetaldehyde, not ethanol, is the cause of developmental toxicity when ethanol is consumed in excess.

Inhalation data suggest that tert-butanol is not selective for developmental toxicity (Nelson et al., 1989). Dams exposed at 2000, 3500, or 5000 ppm for 7 h/d during gestational days 1 to 19 produced fetuses with subnormal weight gain (all three exposures) and increased skeletal variations (top two exposures). Dams in all groups exhibited an unsteady gait toward the end of exposure each day, and the highest exposure group failed to gain weight at the rate of controls (a difference of approximately 70 g after 19 d). Nelson et al. (1989) pointed out that

the developmental effects reported by Daniel and Evans (1982) were elicited after much higher doses that would be difficult to achieve by inhalation. Nelson et al. (1989) concluded that developmental effects would be likely to occur only in the presence of maternal toxicity.

The potential for tert-butanol to affect the toxicity of other chemicals has not been studied extensively. Cornish and Adefuin (1967) showed that carbon tetrachloride-induced liver injury in rats, as measured by increases in serum SGOT, could be increased by oral administration of the alcohol at 1.4 g/kg 16 to 18 h before inhalation of the chlorocarbon at 2000 ppm for 2 h. The activity of the enzyme in the serum of tert -butanol-exposed rats was increased approximately 50-fold over that in control rats. The effect could not be demonstrated when the alcohol was given only 2 h before inhalation of the chlorocarbon.

In experiments designed to understand the development of physical dependence on alcohols, McComb and Goldstein (1979b) showed that tert-butanol can substitute for ethanol in the induction of the dependent state in mice. The larger alcohol was 4 to 5 times more effective in producing the dependent state than the smaller alcohol (McComb and Goldstein, 1979a).

Because no data are available on human exposures to tert-butanol, the rationale for acceptable concentrations must depend entirely on data obtained in animal models. In 1987, WHO considered the toxicity data base inadequate for setting occupational exposure guidelines, and in 1989, the Cosmetics Ingredient Review Expert Panel concluded that the available data were insufficient to support the safety of tert-butanol as used in cosmetics (Brandt, 1989). Since that time, excellent data have become available on the long-term effects of exposure to this alcohol via the oral route of administration (Lindamood et al., 1992; NTP,

1994). In addition, we were able to use results from an inhalation study (not yet peer reviewed) sponsored by NTP as part of the data base for setting acceptable concentrations (Battelle, 1988a, b). In general, the guidelines promulgated by the NRC (1992) were used to derive acceptable concentrations (ACs) for tert-butanol.

Without human data on the CNS effects of tert-butanol, the AC is based on animal data. Nelson et al. (1989) reported ''unsteadiness'' in pregnant rats at the end of 7-h exposures at concentrations as low as 2000 ppm (6200 mg/m³); hence, that concentration was taken as a lowest-observed-adverse-effect level (LOAEL). The steepness of the dose-response curve (see Table 3-3) suggests that a factor of only 3, rather than the usual 10, would be adequate to estimate a NOAEL; therefore, the NOAEL was estimated to be 2000 mg/m³. From the Battelle results (1988a), a NOAEL for ataxia in rats was 450 ppm (1400 mg/m³) and in mice it was 900 ppm (2800 mg/m³). Thus, in the most susceptible species (rats), the NOAEL for ataxia appears to be about 1500 mg/m³. Applying the default species factor of 10, gives an AC for humans of 150 mg/m³ (50 ppm). This limit is independent of exposure time since blood concentrations stabilize a few hours after the start of an exposure and prolonged exposure leads to reduced blood concentrations for a given airborne concentration (McComb and Goldstein, 1979a).

A dose-dependent increase in the severity of renal lesions was found in exposed male rats in the drinking-water and inhalation studies (Table 3-1), but was not found not in females or other species (NTP, 1994; Battelle, 1988b). Such lesions are difficult to apply to human risk assessment, because male rats produce α2u-globulin in large amounts, and this protein binds hydrocarbons or their metabolites to yield a complex that is difficult to degrade (Swenberg et al., 1989). The formation of such a complex has not been demonstrated specifically for tert-buta-

nol; however, a complex has been demonstrated for structurally similar compounds, such as 2, 4, 4-trimethyl-2-pentanol (Borghoff et al., 1995). Since humans do not produce this globulin, they should have a markedly reduced risk for hydrocarbon-induced nephropathy compared with male rats (Borghoff et al., 1990). Eventually, the nephropathy (as observed in the 13-w inhalation study) can lead to cell death and carcinogenesis through restorative cell replication (Borghoff et al., 1990), as was evident in the 2-y drinking-water study (NTP, 1994).

The risk analysis for renal injury must consider the fact that nephropathy was reported only in female rats in the 13-w drinking water study, and only in male rats in the 13-w inhalation study. In female rats, the incidence of nephropathy in the drinking-water study showed a NOAEL at an average dose of 0.50 g/kg/d (0.5%) for 13 w (NTP, 1994). The highest inhalation dose to female rats was 2100 ppm (measured average concentration), and that was a NOAEL (Battelle, 1988b). These inhalation exposures were for 6 h, so a 0.3-kg rat inhaling air at 0.15 m³/d with tert-butanol at 6.5 g/m³ (2100 ppm) with an uptake of 50% would receive a daily dose as follows:

That dose is less than the oral NOAEL; therefore, it is not surprising that nephropathy was not reported in female rats in the inhalation study. Using the default species factor of 10 and conservatively assuming Haber's rule, an AC (renal injury) is calculated as follows:

To extend this NOAEL to shorter or longer exposure times, the highest NOAELs in female rats were compared in the drinking-water studies (Table 3-6). The LOAELs were included in Table 3-6 to indicate that the NOAELs were within a factor of 2 of the lowest-effect level. An eightfold increase in the exposure time (from 13 w to 2 y) resulted in a decrease in the NOAEL of only threefold (from 0.50 g/kg/d to 0.175 g/kg/d).

Hence, an increase in astronaut exposure time from 30 d to 180 d

(sixfold) should require only a reduction of threefold in the exposure concentration to be assured of a NOAEL. The 180-d AC was determined to be 120 mg/m³ (about one third the 30-d AC of 350 mg/m³). The 7-d (168 h) AC can be set from the female rat data as recommended by the NRC (1992) for extrapolation to shorter exposure times, that is, by not increasing the exposure concentration. Using the default species factor of 10, the 7-d AC was calculated to be 650 mg/m³ for nephropathy.

The findings on urinary-bladder injury were analogous to those for nephropathy; lesions were found in the 13-w drinking-water study, but not in the inhalation study. The 13-w drinking water study suggested a NOAEL for urinary-bladder injury of 0.8 g/kg/d (1%) in the male rat, the most sensitive species and sex. This dose is well above that calculated for the highest inhalation dose (0.4 g/kg/d). The highest dose in the 2-y study (0.42 g/kg/d) did not cause urinary-bladder injury in male rats, suggesting that this lesion is more dependent on dose than length of exposure. This conclusion is also reached by inspection of the data on male mice. The NOAEL for 13-w exposures in the drinking-water study was 1.6 g/kg/d, whereas the NOAEL in the 2-y study was 1.0 g/kg/d (a LOAEL was found at 2.1 g/kg/d). These data suggest that using Haber's rule to extrapolate the 13-w inhalation NOAEL of 6.5 g/m³ to other exposure times would be too conservative. The data suggest that the 13-w inhalation NOAEL would be suitable for 7 and 30 d of exposure, and half that concentration would easily be a NOAEL for 180 d of exposure. Hence, applying the default species factor, the ACs were set as follows:

7-d and 30-d ACs = 650 mg/m³

180-d AC = 300 mg/m³.

After review of the possible mechanisms leading to the thyroid adenomas in mice, consideration of the differences between rodent and human thyroid function, and the lack of any lesions in the inhalation studies, it was concluded that the data were inappropriate for human health risk estimates. The observations and rationale are summarized below.

Thyroid lesions were not reported in rats exposed to tert-butanol by any route; however, the incidence of follicular-cell hyperplasia statistically increased in male mice given the alcohol in drinking water for 2 y at concentrations of 5, 10, or 20 mg/mL and in female mice given 10 or 20 mg/mL. The incidence of follicular-cell adenomas was increased only in the 20-mg/mL group of female mice. Such adenomas are thought to result from progression of follicular-cell hyperplasia and can progress to carcinomas over time (one male in the 20-mg/mL exposure group had a carcinoma). Such proliferative lesions in mice might not be appropriate as models for human thyroid lesions, depending on the mechanism of injury. These lesions could be due to direct action of tert-butanol or a metabolite on follicular-cell DNA, but genotoxicity has not been demonstrated for tert-butanol despite extensive testing. Even if direct mechanisms were involved, McClain (1992) has pointed out that the strong promoting effect of thyroid-stimulating hormone (TSH) in rodents is likely to lead to over estimation of the risk in other species with lower TSH levels.

The weight of evidence suggests a nongenotoxic mechanism that might involve the stimulation of excess production of TSH, but humans appear to be less sensitive than experimental animals to TSH-stimulating chemicals (Hill et al., 1989). There are marked differences in thyroid function between humans and rodents, and these differences make the "rodent an inappropriate model for the extrapolation of cancer risk to man for chemicals that operate secondary to hormone imbalance" (McClain, 1992). The serum TSH is many times higher in rodents than in humans, and rodents lack the thyroid-binding globulin found in humans and some other species. These differences suggest a much higher

thyroid-gland activity in rodents when compared with humans (Dohler et al., 1979). These differences in thyroid activity are reflected in the interspecies differences in baseline thyroid tumor incidences. For example, the expected lifetime incidence of thyroid cancer in adult humans is 0.2% (males) and 0.5% (females), whereas the estimated neoplasm incidence at the end of the 2-y drinking-water study in control mice was 3.6% (males) and 5.6% (females) (NCRP, 1989; NTP, 1994).

A final argument against using the thyroid data from the drinking-water study for a human inhalation risk estimate is that no effects on the thyroid were observed in any groups involved in the 13-w inhalation study (Battelle, 1988b). Because the highest concentration in the rodent inhalation study was 2100 ppm, it is very unlikely that a less susceptible species, such as humans, would have a significant risk of thyroid tumors at a 40-fold lower concentration (50 ppm vs. 2100 ppm) than the one that was a NOAEL in rodents.

The adverse effects resulting from exposure to tert-butanol would not be expected to be potentiated by the physiological and biochemical changes caused by spaceflight.

The most important weakness in the data base is the lack of information about the effects of short-term exposures in humans. No data were found on the irritation thresholds or potential performance decrements in humans exposed for several hours to tert-butanol. Other alcohols show significant irritant properties and cause readily measured performance decrements. The data base for long-term exposures appears to be relatively complete, particularly in view of the recent completion of a 13-w inhalation study. The study has not been peer reviewed by NTP.

Aarstad, K., K. Zahlsen, and O. G. Nilsen. 1985. Inhalation of butanols: Changes in the cytochrome P-450 enzyme system. Arch. Toxicol. Suppl. 8:418-421.

ACGIH. 1995. 1995-1996 Threshold Limit Values and Biological Exposure Indices. American Conference of Governmental Industrial Hygienists, Cincinnati, Ohio.

Amoore, J. E., and E. Hautala. 1983. Odor as an aid to chemical safety: Odor thresholds compared with threshold limit values and volatilities for 214 industrial chemicals in air and water dilution. J. Appl. Toxicol. 3(6):272-290.

Anderson, R. A., Jr., J. M. Reddy, C. Joyce, B. R. Willis, H. Van der Ven, and L. J. D. Zaneveld. 1982. Inhibition of mouse sperm capacitation by ethanol. Biol. Reprod. 27:833-840.

Baker, R. C., S. M. Sorensen, and R. A. Deitrich. 1982. The in vivo metabolism of tertiary butanol by adult rats. Alcohol Clin. Exp. Res. 6(2):247-251.

Battelle. 1988a. Repeated-dose Inhalation Study of t-Butanol in Rats and Mice. Abstract and data tables. Prepared by Battelle, Columbus, Ohio, for the National Institute of Environmental Health Sciences, Research Triangle, N.C.

Battelle. 1988b. Subchronic Inhalation Study of t-Butanol in Rats and Mice. Abstract and data tables. Prepared by Battelle, Columbus,

Ohio, for the National Institute of Environmental Health Sciences, Research Triangle, N.C.

Beaugé, F., M. Clément, J. Nordmann, and R. Nordmann. 1981. Liver lipid disposal following t-butanol administration to rats. Chem.-Biol. Interact. 38:45-51.

Bechtel, D., and H. Cornish. 1975. Metabolism and biological disposition of butyl alcohols in the rat. Toxicol. Appl. Pharmacol. 33:175.

Borghoff, S. J., B. G. Short, and J. A. Swenberg. 1990. Biochemical mechanisms and pathobiology of alpha-2-microglobulin nephropathy. Annu. Rev. Pharmacol. Toxicol. 30:349-367.

Borghoff, S. J., M. L. Gargas, M. E. Andersen, and R. B. Conolly. 1995. Development of a mechanism-based dosimetry model for 2, 4, 4-trimethyl-2-pentanol-induced alpha-2u-globulin nephropathy in male Fischer 344 rats. Fundam. Appl. Toxicol. 25:124-137.

Brandt, K. R. 1989. Final report on the assessment of t-butyl alcohol. J. Am. College Toxicol. 8:627-641.

Cederbaum, A. I., and G. Cohen. 1980. Oxidative demethylation of t-butyl alcohol by rat liver microsomes. Biochem. Biophys. Res. Commun. 97(2):730-736.

Cederbaum, A. I., A. Qureshi, and G. Cohen. 1983. Production of formaldehyde and acetone by hydroxyl-radical generating systems during the metabolism of tertiary butyl alcohol. Biochem. Pharmacol. 32(23):3517-3524.

Cornish, H. H., and J. Adefuin. 1967. Potentiation of carbon tetrachloride toxicity by aliphatic alcohols. Arch. Environ. Health. 14:447-449.

Daniel, M. A., and M. A. Evans. 1982. Quantitative comparison of maternal ethanol and maternal tertiary butanol diet on postnatal development. J. Pharmacol. Exp. Ther. 222:294-300.

Dohler, K. D., C. C. Wong, and A. Von Zur Muhlen. 1979. The rat as a model for the study of drug effects on thyroid function: Consideration of methodological problems. Pharmacol. Ther. 5:305-318

Faulkner, T. P., and A. S. Hussain. 1989. The pharmacokinetics of tertiary butanol. Res. Commun. Chem. Pathol. Pharmacol. 64(1):31-39.

Faulkner, T. P., J. D. Wiechart, D. M. Hartman, and A. S. Hussain.

1989. The effects of prenatal tertiary butanol administration in CBA/J and C57BL/6J mice. Life Sci. 45:1989-1995.

Gaillard, D., and R. Derache. 1965. Métabolisation de différents alcools. Présents dans les boissons alcooliques chez le rat. Pharmacie Montpellier 25:51-62.

Galloway, S. M., M. J. Armstrong, C. Reuben, S. Colman, B. Brown, C. Cannon, A. D. Bloom, F. Nakamura, M. Ahmed, S. Duk, J. Rimpo, B. H. Margolin, M. A. Resnick, B. Anderson, and E. Zeiger. 1987. Chromosome aberrations and sister chromatid exchanges in Chinese hamster ovary cells: Evaluations of 108 chemicals. Environ. Mol. Mutagen. 10(Suppl. 10):1-175.

Grant, K. A., and H. H. Samson. 1982. Ethanol and tertiary butanol induced microcephaly in the neonatal rat: Comparison of brain growth parameters. Neurobehav. Toxicol. Teratol. 4:315-321.

Hill, R. N., L. S. Erdreich, O. E. Paynter, P. A. Roberts, S. L. Rosenthal , and C. F. Wilkinson. 1989. Thyroid follicular cell carcinogenesis. Fundam. Appl. Toxicol. 12:629-697

James, J. T., T. F. Limero, H. J. Leano, J. F. Boyd, and P.A. Covington. 1994. Volatile organic contaminants found in the habitable environment of the space shuttle: STS-26 to STS-55. Aviat. Space Environ. Med. 65:851-857.

Kamil, I. A., J. N. Smith, and R. T. Williams. 1953. Studies in detoxication. Biochem. J. 53:129-136.

Lindamood, C., III, D. R. Farnell, H. D. Giles, J. D. Prejean, J. J. Collins, K. Takahashi, and R. R. Maronpot. 1992. Subchronic toxicity studies of t-butyl alcohol in rats and mice. Fundam. Appl. Toxicol. 19:91-100.

Lington, A. W., and C. Bevan. 1994. Alcohols. Pp. 2585-2760 in Pattys Industrial Hygiene and Toxicology, 4th Ed., G. D. Clayton and F. E. Clayton, eds. New York: Wiley-Interscience.

MacGregor, J. T., C. M. Wehr, P. R. Henika, and M. D. Shelby. 1990. The in vivo erythrocyte micronucleus test: Measurement at steady state increases assay efficiency and permits integration with toxicity studies. Fundam. Appl. Toxicol. 14:513-522.

McClain, R. M. 1992. Thyroid gland neoplasia: non-genotoxic mechanisms. Toxicol. Lett. 64/65:397-408.

McComb, J. A., and D. B. Goldstein. 1979a. Quantitative comparison of physical dependence on tertiary butanol and ethanol in mice: Cor-

relation with lipid solubility. J. Pharmacol. Exp. Ther. 208(1):113-117.

McComb, J. A., and D. B. Goldstein. 1979b. Additive physical dependence: Evidence for a common mechanism in alcohol dependence. J. Pharmacol. Exp. Ther. 210(1):87-90.

McGregor, D. B., A. Brown, P. Cattanach, I. Edwards, D. McBride, and W. J. Caspary. 1988. Responses of the L5178Y tk⁺/tk^- mouse lymphoma cell forward mutation assay. II. 18 coded chemicals. Environ. Mol. Mutagen. 11:91-118.

Munch, J. C. 1972. Aliphatic alcohols and alkyl esters: Narcotic and lethal potencies to tadpoles and to rabbits. Ind. Med. 41(4):31-33.

Nelson, B. K., W. S. Brightwell, A. Khan, J. R. Burg, and P. T. Goad. 1989. Lack of selective developmental toxicity of three butanol isomers administered by inhalation to rats. Fundam. Appl. Toxicol. 12:469-479.

NCRP. 1989. Guidance on Radiation Received in Space Activities. NCRP Rep. No. 98. National Council on Radiation Protection and Measurements, Bethesda, Md.

NRC. 1992. Guidelines for Developing Spacecraft Maximum Allowable Concentrations for Space Station Contaminants. Washington, D.C.: National Academy Press.

NTP. 1994. Toxicology and Carcinogenesis Studies of t-Butyl Alcohol in F344/N Rats and B6C3F₁ Mice (Drinking-Water Studies). Tech. Rep. No. 436. NIH Publ. No. 95-3167. National Institutes of Health, National Toxicology Program, Research Triangle Park, N.C.

Saito, M. 1975. Studies on the metabolism of lower alcohols. Nichidai Igaku Zasashi 34:569-585.

Swenberg, J. A., B. Short, S. Borghoff, J. Strasser, and M. Charbonneau. 1989. The comparative pathobiology of α2u-globulin nephropathy. Toxicol. Appl. Pharmacol. 97:35-46.

Takahashi, K., C. Lindamood, III, and R. R. Maronpot. 1993. Retrospective study of possible alpha-2u-globulin nepropathy and associated cell proliferation in male Fischer 344 rats dosed with t-butyl alcohol. Environ. Health Perspect. 101(Suppl 5):281-286.

Thurman, R. G., K. Winn, and B. Urquhart. 1980. Rat brain cyclic amp levels and withdrawal behavior following treatment with t-butanol. Adv. Exp. Med. Biol. 126:271-281.

U.S. Department of Labor. 1995. Air Contaminants—Permissible Ex-

posure Limits. Title 29, Code of Federal Regulations, Part 1910, Section 1910.1000. Washington, D.C.: U.S. Government Printing Office.

Wallgren, H. 1960. Relative intoxicating effects on rats of ethyl, propyl and butyl alcohols. Acta Pharmacol. Toxicol. 16:217-222.

WHO. 1987. Butanols-four isomers: 1-butanol, 2-butanol, tert-butanol, isobutanol. Environmental Health Criteria 65. Geneva: World Heath Organization.

Zeiger, E., B. Anderson, S. Haworth, T. Lawlor, K. Mortelmans, and W. Speck. 1987. Salmonella mutagenicity tests. III. Results from the testing of 255 chemicals. Environ. Mutagen. 9(Suppl. 9): 1-110.