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

Chiu-Wing Lam, Ph.D., and John T. James, Ph.D.

Johnson Space Center Toxicology Group

Biomedical Operations and Research Branch

Houston, Texas

Indole is a colorless crystalline solid. It has an intense fecal odor at moderate concentrations; however, the odor at very low concentrations is pleasant (Merck Index, 1989).

Indole is a naturally occurring compound, constituting about 2.5% of jasmine oil and 1% of orange-blossom oil, and in both cases, it contributes to their fragrances (Kirk-Othmer Concise Encyclopedia of Chemical Technology, 1985). Indole is a component of perfumes (Kirk-Othmer Concise Encyclopedia of Chemical Technology, 1985; Merck Index,

1989). Paradoxically, indole has an intense fecal odor (Merck Index, 1989), presumably at higher concentrations. It is also a bacterial decomposition product of tryptophan in the gut (Hammond et al., 1984; Eisele, 1986). Indole is one of the odorous components found in sewage and animal wastes, including human feces (Veber, 1967, as cited in Sgibnew and Orlova, 1971; Karlin et al., 1985). The compound is also found in animal tissues where putrefactive processes have occurred, presumably by the decomposition of tryptophan (Eisele, 1986). Trace levels of indole are expected to be present in manned spacecraft.

Indole, an anaerobic metabolite of tryptophan, can be condensed with serine by microorganisms to recover tryptophan (Meister, 1965). However, this biosynthetic pathway for tryptophan has not been observed in humans and rats. Absorbed indole from the gut is hydroxylated to form indoxyl, which conjugates with sulfate to produce indican (indoxylsulfuric acid) in the liver (Meister, 1965). Indoxyl and indican are found in human plasma and urine. A mean plasma indican concentration of 3 mg/L was reported in 56 males (range 1.2-4.8 mg/L) and 44 females (range 0.6-5.4 mg/L) (Geigy Pharmaceuticals, 1974). The daily urinary excretion of indoxylsulfate in normal adults was reported to average 200 mg (range 140-250 mg) (Haddox and Saslaw, 1963).

Sgibnew and Orlova (1971) reported that indole was not detected in the blood of rabbits exposed at 10 mg/m³ for 3 h. When 10 mg of indole was injected intravenously into each of the five rabbits, an average plasma indole concentration of 0.3 mg% was detected only in the blood samples taken 15 min after the injection. Because of the inability to detect indole in the blood thereafter, Sgibnew and Orlova concluded that indole was quickly removed and rendered harmless. However, it should be pointed out that 0.3 mg% was close to the detection limit of the calormetric method employed by these authors. Indole concentrations in the blood were investigated by Hammond et al. (1980) in cows dosed orally with the compound at 50, 100, or 200 mg/kg. Plasma indole concentrations reached peak concentrations of 4.5, 8, and 20 mg/mL, respectively, 3 h after dosing. The plasma concentrations decreased to 4.4%, 7.4%, and 38% of the corresponding peak concentra-

tions after 12 h and to 2%, 0.6%, and 1.4% after 24 h. Indole was not detected (<0.02 mg/mL) in plasma of these cows 72 h after injection. These results indicate that indole was rapidly eliminated from the blood of the cows.

Indole at a few parts per million has an unpleasant odor and can elicit toxic symptoms, such as nausea. The consistent toxicological property of indole, an aromatic amine, observed in animal studies is its ability to cause the formation of Heinz bodies, which are known to be produced by other aromatic amine compounds, such as aniline (Smith, 1986). Chronic studies by the subcutaneous route have shown that indole might have a weak leukemogenic activity in mice, but not in hamsters. The toxicity of indole is summarized in the Table 6-1.

The human odor threshold of indole was reported to be 0.45 mg/m³ (»0.1 ppm) (Sgibnew and Orlova, 1971). Very unfavorable odor was perceived in concentrations approaching 9.0 mg/m³; 2 of the 12 test subjects complained of nausea. According to the authors, brief exposures at this concentration did not produce electrocardiographic and electroencephalographic changes (exposure length not specified). Exposing 20 mice and 15 rats to indole at a concentration of 9-10 mg/m³ for 2 h produced no toxic signs except some unrest during the first 15 min; no deaths occurred during the 14-day post-exposure observation period (Sgibnew and Orlova, 1971).

The hematological effects and subsequent renal lesions induced by indole were observed in four cows given the compound at a single oral dose of 100 mg/kg and then 200 mg/kg 2 w later (Hammond et al., 1980). Hemolysis was observed 24 h after each chemical exposure. At the high dose, all four cows had blood-colored urine. Necropsy 1 w later revealed renal tubular epithelial degeneration attributable to hemoglobinuric nephrosis. There was a grayish-brown discoloration of the endothelial surfaces of the aorta and other elastic arteries, and a mild

cloudy swelling in hepatocytes. No lesions were found in the lung. The dose of indole that produced no observable clinical signs of toxicity, including hemolysis, in these cows was 50 mg/kg given in a single dose 1 mo before the 200-mg/kg dose.

Inhalation studies were conducted by Sandage (1961) in 10 rhesus monkeys, 50 rats, and 100 mice exposed continuously to indole at a concentration of 10.5 ppm (50 mg/m³) for 90 d. Hematological examination of the exposed rodents revealed that numerous Heinz bodies were present in the blood. Heinz-body occurrence was most prominent in mice and was observed after 3-4 d of exposure. It was slightly less prominent in the rats and was observed after 7-10 d of indole exposure. No Heinz bodies were observed in monkeys until 70 d of exposure. In the mice, the appearance of Heinz bodies was accompanied by anisocytosis (presence in blood of erythrocytes showing excessive variation in size) and polychromatophilia, with many erythrocytes exhibiting diffuse basophilia. There was a general leukocytosis and marked reticulocytosis with about 80% of the reticulated cells appearing morphologically atypical. Instead of the fine thread-like reticulum of normal reticulocytes, there was a profuse, densely staining nuclear material that filled the cell, which appeared to consist of coarse granules. Pathological studies on 25 % of the exposed mice revealed 95 % of the animals had pigment in the renal tubular cells. However, renal abnormalities were not found in any exposed monkeys or rats examined (about 25% of those exposed).

Significant elevation of serum sodium, cholinesterase, and amylase levels in monkey blood serum was also noted by Sandage (1961), who suggested that such elevated levels provided a clue to neurological effects. However, pathological examination showed no brain damage. Histopathological studies of the heart, lung, liver, and kidney from the exposed monkeys revealed no statistical difference from that of control monkeys. Two monkeys from the exposed group and none from the controls died. Sandage concluded that the death rate (2 of 10) of the monkeys exposed to indole could not be considered ''significant.'' The cause of death was not given.

There are three studies in mice that provide some evidence that indole might be leukemogenic. Eckhart and Stich (1957) reported that three of the 50 RFH-bred white mice given indole at 0.5 mg (in 50% propylene glycol) subcutaneously every 3-4 d for 140 d (weekly dose approximately 40 mg/kg) developed leukemia and three developed aleukemic myelosis; 10 of the initial 50 mice died in the first 4 w of indole administration (4 mg per mouse). Leukemia was not observed among the 100 control mice; however, no indication was given that the control mice were given sham injections of the vehicle. According to these authors, the spontaneous rate of leukemia observed in 1000 RFH mice was 1:500.

A similar study was conducted by Dzioev (1974), who injected indole subcutaneously in mice at a weekly dose of 1 mg for 9 mo. The author reported that of 60 surviving mice, 13 showed leukosis and 21 had pulmonary adenomas. Exposure to tryptophan (presumably serving as controls) resulted in tumors in 8 of the 28 surviving animals. Subcutaneous injection of indole in mice (C-57) was also conducted by Rauschenbach et al. (1963), who administered a dose of 2.5 mg per mouse once a week for 5-10 mo. No leukemia was observed in 17 mice that survived longer than 1 y; however, one mouse developed adenocarcinoma of the breast and seven were leukemoid. Cancer or hematopoietic changes were not observed in the 30 mice given tryptophan (presumably serving as controls) and surviving longer than 1 y (10 died before 1 y). These carcinogenicity results of indole reported in the latter two studies (Rauschenbach et al., 1963; Dzioev, 1974) were classified by NIOSH as equivocal (NIOSH, 1985-86). In addition, the leukemogenic effect of indole was not observed in hamsters chronically exposed to indole. In a study to investigate the role of indole in the carcinogenicity of 2-acetylaminofluorene, 23 male and 30 female Syrian golden hamsters were given 1.6% indole alone in the diet as controls for 10 mo (Oyasu et al., 1972). These animals showed no tumors of the bladder or liver.

The SMACs for indole depend not only on the toxicological effects induced, but also on the interactions with spaceflight-induced changes and the normal turnover of indole in man. The toxicological effects induced by indole include nausea, hematological changes, mortality, and possibly leukemogenic effects. Because spaceflight was known to induce approximately a 10% reduction in red-cell mass (Huntoon et al., 1989), this finding was considered when the hematological effects of indole were used to set an acceptable concentration (AC) for effects on red cells. Inhaled indole is not known to induce effects on the respiratory system; therefore, systemically mediated effects, produced by an absorbed dose from inhalation or other route of administration, might be considered equivalent. Moderate amounts of indole are normally absorbed from the gastrointestinal tract, so there is a significant systemic body burden. If the amount of indole entering the body via respiration

is small compared with the normal load, then there will be no toxic effects from the inhaled indole.

Brief exposures at 2 ppm indole induced nausea in 2 of 12 test subjects (Sgibnew and Orlova, 1971). Slight nausea is tolerable for brief periods as long as crew-member performance is not affected; however, interpretation of the duration of exposure is difficult from the original study. It was estimated that reducing the exposure concentration by a factor of 2 would have caused no more than slight nausea in a few people even if the exposure were for 1 h. Hence, the 1-h AC (nausea) was set at 1 ppm. Since nausea induced by odorous compounds is a threshold effect, and nausea would be less tolerable for a 24-h period, the 24-h AC (nausea) was set at 0.5 ppm to minimize the chance of nausea. For longer exposures, nausea is not acceptable, so the 7-d, 30-d, and 180-d ACs were calculated from the 2-ppm lowest-observed-adverse-effect level (LOAEL) as follows:

The factor of 10 was applied to estimate a no-observed-adverse-effect level (NOAEL) from the LOAEL, and the factor of the square root of 12 divided by 10 was applied to account for a small number of test subjects (12 subjects). This value is about one-third the concentration that gave a sweet odor when inhaled by test subjects for the first time, but not during the second exposure (Sgibnew and Orlova, 1971).

The key study for this toxic end point was the continuous 90-d inhalation study reported by Sandage (1961). Many toxic end points were assessed in the exposures of mice, rats, and monkeys to indole at a concentration of 10.5 ppm. For human risk assessment, the study has several shortcomings, but represents the best available data. The control

animals were not sham exposed; they were kept in a separate room that was not environmentally the same as the exposure chambers; and the findings were reported statistically as increases or decreases compared with control groups rather than numerically (except mortality). Favorable points of the study were that several species were exposed, the concentrations were measured analytically (the actual measurements were not reported), and the data were apparently subjected to statistical analysis.

In this 90-d continuous inhalation study (Sandage, 1961), exposures at 10.5 ppm, which was the only concentration tested, produced Heinz bodies in mice, rats, and monkeys after 3, 7, and 70 d of exposure, respectively. Because only a single concentration was used, the conventional derivation of a LOAEL from a dose-response curve could not be applied. However, employing an unconventional species-based approach, it was noted that 3 x 10.5, 7 x 10.5, and 70 x 10.5 d-ppm induced Heinz bodies in one, two, and three animal species, respectively. Hence, 3 x 10.5 d-ppm was estimated to be the LOAEL. Using the NRC guideline, the LOAEL was divided by 10 to get a NOAEL of 1 ppm. The 7-d NOAEL of 0.4 ppm, which is equal to 1 ppm x 3/7, was then divided by 10 (species factor) and 3 (spaceflight factor, see below) to obtain the 7-d AC of 0.015 ppm. The hematological effect was not observed 24 h after the exposure; therefore, 10.5 ppm was the NOAEL for the 24-h exposure. By applying the same safety factors (species and spaceflight), the 24-h AC would be 0.3 ppm. Because only one exposure concentration was tested in the study, the LOAEL for 30-d and 180-d exposures could not be obtained or extrapolated, and 30-d and 180-d ACs could not be established based on these data.

The factor of 3 for spaceflight-induced effects was appropriate because the magnitude of the red-cell mass changes observed in astronauts has been about 10% (Huntoon et al., 1989). This change has caused no clinically detectable effects and could be considered adaptive in nature. The mechanism of this change might be due to microgravity effects on the kidney resulting from fluid shifts. Kidneys, which produce erythro-poietin, play an important role in red-cell production. However, the mechanism of the red-cell mass changes has not been clearly elucidated at this time. The hematology factor should be smaller than the factor of 5 used for cardiac effects because cardiac effects have actually caused a cosmonaut to be returned early from a mission (Gazenko et al., 1990).

The same study used to set the ACs for hematology was used to set the ACs for mortality because it is the only long-term inhalation study available (Sandage, 1961). The author reported that there was no statistically significant increase in mortality in the exposed groups compared with the control groups; however, there were more deaths in the exposed groups by the end of the 90-d study. A comparison is given below for mortality incidences.

Because the study did not provide information on the cause of death, and it appears that more deaths occurred in the exposed groups, mortality was considered in setting the SMACs. The 30- and 180-d ACs for mortality were set by dividing the exposure at a concentration of 10.5 ppm by factors of 10 to get to a NOAEL for mortality and 10 for interspecies extrapolation. Haber's rule (90 d/180 d) was used to decrease the 180-d exposure concentration. The ACs for 30 and 180 d were 0.1 and 0.05 ppm, respectively.

As described in the Toxicity Summary, there are several reports that indole is leukemogenic when given by injection. NIOSH (1985-1986) has classified two of the studies (Rauschenbach et al., 1963; Dzioev, 1974) as equivocal, and the remaining study (Eckhart and Stich, 1957) appears positive for leukemogenic effects. However, the study control groups might not have been given injections of the vehicle used when indole was administered, and the results of the study are difficult to extrapolate to inhalation exposures. The injection dosages were highly

toxic bolus quantities, whereas inhalation dosages of an equivalent amount would be through long-term, low-level administrations. Furthermore, it is difficult to estimate the risk of carcinogenesis from a study in which only one exposure group was used. It also appears that mortality is a more important end point because in the study in which 6 of 40 mice showed leukemogenic changes, 10 of 50 died before any animals showed these changes. Because the inhalation data have been used to estimate ACs for mortality, it was not necessary to establish ACs for cancer based on injection data, which were obtained for purposes other than estimating an inhalation risk.

Daily urinary excretion of indican averages 200 mg in normal adults, and this is equivalent to 110 mg of indole absorbed from the gastrointestinal tract. It is reasonable to assume that a 5% increase in this indole input from an inhalation source would be toxicologically insignificant. Hence, 5 mg/d could enter through the respiratory system without causing an effect. For a 70-kg man breathing 20 m³/d, an indole concentration of 0.25 mg/m³ (0.05 ppm) would cause an additional uptake of 5 mg/d, assuming 100% of the inhaled dose is absorbed. Thus, 0.05 ppm should be a lower bound on any SMAC value selected. The 7-, 30-, 180-d SMACs were set at the metabolic load threshold of 0.05 ppm based on these considerations.

Dzioev, F.K. 1974. [Study of carcinogenic action of tryptophan and some of its metabolites]. Voprosy Onkologh 20:75-81.

Eckhart, H., and W. Stich. 1957. Untersuchungen uber experimentelle leukamien. II. Mitteilung die indole-leukamie bei der weiben maus. Klinische Wochenschrift 35:504-511.

Eisele, G.R. 1986. Distribution of indole in tissues of dairy cattle, swine and laying pullets. Bull. Environ. Contam. Toxicol. 37:246-262.

Gazenko, O.G., A.I. Grigor'yev, S.A. Bugrov, V.V. Yegorov, V.V. Bogomolov, I.B. Kozlovskaya, and I.K. Tarasov. 1990. [Review of the major results of medical research during the flight of the second prime crew of the Mir Space Station.] Kosmicheskaya Biologiya i Aviakosmicheskaya Meditsina 23(4):3-11; translated and abstracted in USSR Space Life Sciences Digest, L.R. Stone and R. Tetter, eds., NASA Contractor Report 3922(34):119-120.

Geigy Pharmaceuticals. 1974. P. 575 in Documenta Geigy Scientific Tables. Geigy Pharmaceuticals, Ardsley, N.Y.

Haddox, C.H., and M.S. Saslaw. 1963. Urinary 5-methoxytryptamine in patients with rheumatic fever. J. Clin. Invest. 4:435-441.

Hammond, A.C., J.R. Carlson, and R.G. Breeze. 1980. Indole toxicity in cattle. Vet. Record 107:344-346.

Hammond, A.C., B.P. Glenn, G.B. Huntington, and R.G. Breeze. 1984. Site of 3-methylindole and indole absorption in steers after ruminal administration of L-tryptophan. Am. J. Vet. Res. 45:171-174.

Huntoon, C.L., P.C. Johnson, and N.M. Cintron. 1989. Hematology, immunology, endocrinology, and biochemistry. P. 222 in Space Physiology and Medicine, 2nd Ed., A. Nicogossian, C. Huntoon, and S. Pool, eds. Philadelphia: Lea & Febiger.

Karlin, D.A., A.J. Mastromarino, R.D. Jones, J.R. Stroehlein, and O. Lorentz. 1985. Fecal skatole and indole and breath methane and hydrogen in patients with large bowel polyps or cancer. J. Cancer Res. Clin. Oncol. 109:135-141.

Kirk-Othmer Concise Encyclopedia of Chemical Technology. 1985. P. 639 in Kirk-Othmer Concise Encyclopedia of Chemical Technology. New York: Wiley-Interscience.

Meister, A. 1965. Pp. 841-883 in Biochemistry of the Amino Acids, 2nd Ed., Vol 2. New York: Academic Press.

Merck Index. 1989. P. 786 in Merck Index, 11th Ed. Rahway, N.J.: Merck & Co.

NIOSH. 1985-86. Indole. In Registry of Toxic Effects of Chemical Substances. National Institute of Occupational Safety and Health, Cincinnati, Ohio.

Oyasu, R., T. Kitajima, M.L. Hopp, and H. Sumie. 1972. Enhancement of urinary bladder tumorigenesis in hamsters by co-administration of 2-acetylaminofluorene and indole. Cancer Res. 32:2027-2033.

Rauschenbach, M.O., E.I. Jarova, and T.O. Protasova. 1963. Blastomogenic properties of certain metabolites of tryptophane. Acta Unio Int. Cancrum 19:660-662.

Sandage, C. 1961. Tolerance Criteria for Continuous Inhalation Exposure to Toxic Material. II. Effects on Animals of 90-Day Exposure to H ₂S, Methyl Mercaptan, Indole, and a Mixture of H₂S, Methyl Mercaptan, Indole and Skatole. ASD Tech. Rep. 61-519 (II). Biomedical Laboratory, Aerospace Medical Laboratory, Wright-Patterson Air Force Base, Dayton, Ohio.

Sgibnew, A.K., and T.A. Orlova. 1971. K voprosu izucheniia toksichnostic. Pp. 190-195 in Problemy Kosmicheskoi. Biologii, Vol. 16 [Translation: Problem of studying the toxicity of indole. Pp. 233-239 in Problems of Space Biology, Vol. 16], V.N. Chernigovskiy, ed. Academy of Sciences of the USSR, Department of Physiology. Moscow: Nauka Press.

Smith, R. 1986. Toxic responses of the blood. P. 239 in Casarett and Doull's Toxicology: The Basic Science of Poisons, 3rd Ed., C.D. Klaassen, M.O. Amdur, and J. Doull, eds. New York: Macmillan.

Veber, T.V. 1967. P. 276 in Chelovek Pod Vodoy i v Kosmose. Moscow: Voyenizdat Press.