Spacecraft Water Exposure Guidelines for Selected Contaminants: Volume 2 (2007)

Raghupathy Ramanathan, Ph.D. NASA-Johnson Space Center Houston, Texas

Caprolactam is a cyclic amide, derived from epsilon-aminocaproic acid, from which nylon 6 is polymerized (see Table 6-1). Caprolactam is a monomer primarily used in the manufacture of the synthetic polymer nylon 6, fibers and resins, synthetic leather, and as a polyurethane cross linker. Nylon 6 (polycaprolactam) is used in the production of tire cords, carpeting, plastics, and food-packaging materials. Caprolactam has been

approved by the U.S. Food and Drug Administration (FDA) for use in food contact films. Occupational exposure to caprolactam occurs primarily from the manufacture of nylon 6 fibers and resins. Highly soluble in water, caprolactam leaches from clothing made from polyamide fibers when the clothing is soaked in simulated perspiration (Statsek and Ivanova 1978). It has been found in groundwater, surface waters, and finished water (IARC 1986; EPA 1988). Water produced in the Shuttle from fuel cells is iodinated, collected in large containers called contingency water containers (CWCs), and transferred to the International Space Station (ISS) for use by the crew. These CWC bags are used to store drinking water containing silver used as a biocide, and for collection and storage of humidity condensate that will be processed to potable water. The bags are lined with a material called Combitherm. When the Crew and Thermal Systems Division of the National Aeronautics and Space Administration (NASA) was performing material compatibility testing to see if CWC-stored water undergoes quality degradation, it was found that the total organic carbon (TOC) concentration increased. It was determined that caprolactam was the only contributor to this increased TOC. It was also found that the Combitherm material leaches caprolactam during storage, and irrespective of the biocide used (iodine or silver), the leaching continued. A concentration of 16 milligrams per liter (mg/L) of caprolactam was found at the end of 24 weeks (wk). Thus, NASA was prompted to evaluate caprolactam for potential health hazards at this concentration and to recommend a spacecraft water exposure guideline (SWEG).

No human data are available on the absorption of caprolactam from an oral dose. However, from an animal study of disposition pharmacokinetics described below (Unger et al. 1981), it appears that caprolactam is almost completely absorbed from the gastrointestinal (GI) tract, because within 24 hours (h), more than 80% of the administered dose (¹⁴C-caprolactam) was recovered in urine, feces, and expired air.

Two sets of studies were carried out by Unger et al. (1981) on the disposition kinetics of caprolactam when dosed via the oral route. In the first set of experiments, a single oral bolus of ¹⁴C-caprolactam in water

was given to male Fischer rats at a concentration of 0.18 mg per kilogram (kg). Groups of animals were killed after 0.5, 1, 2, 3, 4, 6, 15, or 24 h following dosing; urine was also collected. In addition, exhaled ¹⁴C (as carbon dioxide [CO₂]) was quantified. Twenty-four hours after a single oral bolus dose of ¹⁴C-caprolactam at 0.18 mg/kg, about 77% of radioactivity was excreted in the urine, 3.5% in the feces, and 1.5% in the expired air. Elimination of radioactivity was most rapid in the urine during the initial 6 h following the dose. Analysis of urine indicated that 24 h after dosing, only 2.3% of the radioactivity was in the form of the parent compound. There were two major unidentified metabolites, one comprising 79% and the other 17.7%. After dosing, the peak concentrations of radioactivity in the tissues (nanogram [ng] equivalents of caprolactam per gram of tissue) were similar to that found in the blood (for example, it was about 128 ng/g in blood, 151 ng/g in liver, and about 140 ng/g in spleen), except for stomach (1,907 ± 286 ng/g, including contents), kidney (247 ± 19 ng/g), and bladder (1,240 ± 223 ng/g)—the tissues associated with ingestion and excretion (Unger and Friedman 1980; Unger et al. 1981).

In the second experiment, rats were orally treated with caprolactam at 1.5 g/kg for 7 days (d) and 24 h after the last dose; radioactive caprolactam was administered at the same dose. Animals were killed 6 h after receiving the radioactive dose, and their blood, tissues, urine, feces, and expired air were collected. Another set of animals was administered caprolactam containing ¹⁴C at 1.5 g/kg, and 24-h urine samples were collected and analyzed for metabolites.

When a single dose of ¹⁴C-caprolactam at 1.5 g/kg was administered and studied after 6 h, the pattern of distribution was the same as that observed at the low dose except that 40% of the radioactivity was still in the stomach,, whereas in the previous experiment, only 6% of the radioactivity was in the stomach after 6 h. Additionally, at 6 h, 14% was excreted in the urine in the high-dose group, and 39% was excreted in the urine in the low-dose group. The authors did not specify how much was in the stomach tissue or in the contents.

In the 7-d study, when radioactive caprolactam was given after 7 d of pretreatment, the tissue distribution was similar to that from a single bolus dose. However, at 6 h, there was a fivefold increase in the excretion of radioactive CO₂ in the expired air (about 0.25% of the administered dose). The results of this study indicated that caprolactam is very

rapidly and almost completely absorbed at low doses and also rapidly eliminated. From the distribution of radioactivity in the GI tract, it appears that caprolactam is predominantly and directly absorbed from the stomach rather than from the intestine. The fact that liver does not show a very high percentage of the radioactivity may suggest that caprolactam either undergoes extensive biotransformation or is rapidly eliminated from the liver and finally excreted in the urine. In the high-dose group, an appreciable quantity of parent compound was excreted in the urine,, which suggests that biotransformation in liver may not be required for the elimination of caprolactam in urine. This also suggests that the metabolic pathway in liver probably becomes saturated at high doses.

Kirk et al. (1987) reported that rats fed a diet containing 3% caprolactam for 2 or 3 wk excreted approximately 16% of the caprolactam ingested as the 4-hydroxy metabolite and a small amount as the nonhydroxylated acid, 6-aminohexanoic acid. The 4-hydroxy metabolite rearranges spontaneously in acidic aqueous medium to an equilibrium mixture in which 6-amino-gamma-caprolactone is the major component and 6-amino-4-hydroxyhexanoic acid is a minor component.

The distribution of ¹⁴C-caprolactam was also studied by wholebody autoradiography in male, female, and pregnant mice on day 14.5 of gestation which were given an average of ¹⁴C-caprolactam at 6.6 mg/kg by oral intubation in water (Waddell et al. 1984). Pregnant mice were frozen 20 minutes (min) and 1, 3, 9, and 24 h after oral administration of the compound. From the pattern of intensity of radioactivity as measured by autoradiography, one can understand the distribution and elimination. The nonpregnant mouse was frozen 3 h after oral dosing, and two male mice were frozen 20 min and 9 h after intravenous (iv) administration. Autoradiography data indicated that radioactivity was rapidly absorbed from the stomach and distributed throughout the entire animal, including the fetuses. There was efficient elimination by the kidney and liver, as evidenced by the shift in density in autoradiograph from the stomach to these organs. Material secreted by the liver into bile and intestinal contents appeared not to be reabsorbed via the enterohepatic circulation. The kinetics of distribution and elimination appeared to be the same in male, female, and pregnant animals. The distribution into and removal from the fetuses was typical of molecules that diffuse freely across the placenta. There was no retention of radioactivity in any fetal tissue, and there was no localized concentration of caprolactam in any specific tissue.

There are several Russian reports indicating that industrial exposures of factory workers to caprolactam resulted in various serious adverse effects on neurologic (headaches to seizures), gynecologic, obstetric, GI (nausea and vomiting), cardiovascular, dermatologic, and hepatic systems. Because these factory workers were also exposed to several organic solvents and other compounds, it is very difficult to associate these effects with caprolactam alone. Because of difficulties involved in getting full translations of these Russian articles for evaluation, these studies have not been included in this document. Caprolactam is suspected to cause contact dermatitis (allergic skin reaction) in an occupational environmental setting.

The Crew and Thermal Systems Division of NASA conducted a fluid compatibility study on water stored in CWCs, the bladders of which were made of a material called Combitherm-140. Earlier results had indicated that a chemical material was leaching out of the bags, increasing the TOC in water and was traced entirely to the increased concentrations of caprolactam, one of the ingredients of the water bags. The water was of the exact composition as that of water available to crew on the ISS, water with silver biocides and minerals. The water contained silver, fluoride, calcium, and magnesium at concentrations of 0.5, 0.55, 29.4, and 4.91 mg/L, respectively. Taste tests were done at 12, 48, and 64 wk after storing ISS-quality water in CWCs using 10-12 panelists who participated in the sensory panel. The caprolactam concentrations were 4.7, 11.5, and 12.6 mg/L, respectively. The sensory panel consisted of odor, flavor, and overall acceptability. The scores were compared with the concentrations of caprolactam in the water. Water from the Johnson Space Center (JSC) drinking water and Ozarka commercial bottled water served as controls for this blinded study. Based on the results, it was concluded that at the maximum concentration of 13 mg/L of caprolactam, the water was declared acceptable for drinking. It must be stressed that the subjects tasted the water only one time and not every day over duration.

Target organs for acute and chronic toxicities from inhalation exposures were suspected to be eye, skin, the respiratory system, neurologic system, and the GI tract (nausea and vomiting). Most of the reports on toxicity are from Russian studies using inhalation exposures on animals. In animals injected with caprolactam via iv or intraperitoneally (ip), observed effects have included reproductive/teratogenic effects, cardiovascular system effects such as hypertension and hypotension depending on the dose, anemia, and degeneration of the renal tubular epithelium (see Gross 1984).

The National Toxicology Program (NTP) conducted a carcinogenesis bioassay of caprolactam in F344 rats and B6C3F1 mice in 1982. In the survival determination studies, rats and mice were given single gavage doses of caprolactam at 681, 1,000, 1,470, 2,150, and 3,160 mg/kg for male and female rats and 1,000, 1,470, 2,150, 3,160 and 4,640 mg/kg for male and female mice. The doses were delivered in corn oil. Deaths occurred in rats receiving more than 1,470 mg/kg (males) and 1,000 mg/kg (females). Estimated LD₅₀ (the dose causing death in 50% of test subjects) values are shown in Table 6-2.

In a comparative oral toxicity study in different species of animals, a single dose of caprolactam at 1,000 mg/kg resulted in 70% lethality in mice, 60% in rabbits, 30% in guinea pigs, and 30% in rats. In all animal species, death resulted from violent epileptic convulsions, salivation, bleeding from the nostrils, respiratory arrest, tremors, and low body temperature (Savelova 1960, cited in Gross 1984).

Six hours after oral administration of a single bolus of caprolactam at 1.5 g/kg to male F344 rats, there were marked inductions of tyrosine amino transferase (TAT, L-tyrosine:2-oxoglutarate aminotransferase) and tryptophan oxygenase (TPO) (Friedman and Salerno 1980). The two enzymes are involved in the first steps of tyrosine and tryptophan catabolism, respectively. Induced activities were seen at 3 h after dosing with a maximum effect occurring at 6 h. Even 24 h postdosing, the induction of enzymes was significantly higher than in untreated controls. A dose-response study of the induction of these two enzymes after single

oral doses of 300, 600, 900, 1,200 and 1,500 mg/kg was conducted. Both enzymes, when measured 5 h after dosing, increased linearly with the dose up to the maximum dose tested, but TAT was induced only at doses higher than 300 mg/kg and increased by 100% at 600 mg/kg. In the case of TPO, a 100% increase was observed even at 300 mg/kg, and the activity continued to increase with doses up to 1,500 mg/kg, the maximum dose used in the studies.

In a biochemical study, 90-d-old adult female Sprague-Dawley rats (CD strain) were treated by gavage with two doses of caprolactam at 425 mg/kg (Kitchin and Brown 1989) to give a total dose in 24 h of 850 mg—one dose within 21 h of sacrifice and one 4 h before sacrifice (n = 13). Caprolactam was administered as a saline solution. Serum ornithine decarboxylase, alanine amino transferase (ALT, also called serum glutamic pyruvic transaminase, SGPT), hepatic glutathione content, hepatic cytochrome P-450 concentrations, and DNA damage (as measured by the alkaline elution of DNA) were measured. There was a statistically significant increase in the activity of SGPT (marker enzyme for abnormal liver function). Single- and double-stranded breaks in liver DNA were also measured in the study and were not different from controls. No other parameters were found to be altered.

In another study (Friedman and Salerno 1980), caprolactam was fed to rats for 7 d as 1% and 5% of their diet. The animals were pair fed, and

the authors estimated the doses as 1.1 g/kg/d and 1.8 g/kg/d, respectively. Liver protein synthesis was studied using incorporation of ³H-leucine. Although food restriction (pair feeding) alone increased this activity by 60%, the inclusion of caprolactam blocked this increase in protein synthesis. Also, the authors reported that food-conversion efficiency (gain in weight per gram of diet consumed), was reduced by 1.8 g/kg/d. While induction of TAT was also observed in this 7-d study compared to pairfed controls, TAT activity was lower than corresponding pair-fed controls (although, the difference was not statistically significant) (Friedman and Salerno 1980). This induction did not appear to require RNA synthesis, because RNA synthesis inhibitors did not prevent the observed induction of TAT and TPO. It must be pointed out that with a single oral bolus dose, the increased activities of TPO and TAT were several-fold higher than were observed in the week-long pair-feeding study in which the caprolactam was administered via the diet. Thus, the data will be used for the 10-d acceptable concentration (AC), which is based on the effect of caprolactam on increased amino acid catabolism, which is also reflected on the effect of protein synthesis.

In a three-generation reproduction study by Serota et al. (1988), F344 albino rats (10 males and 20 females) were fed diets containing caprolactam at 0, 50, 250, or 500 mg/kg/d for 10 wk. Body weights of the parental generations and their offspring (in the 250 and 500 mg/kg/d groups) were significantly reduced. In this study, the authors noted that on microscopic evaluation of the kidney sections and gross lesions, there was also a slight increase in the severity of spontaneous nephropathy accompanied by the presence of granular casts in some animals of the 500mg/kg/d group. According to the authors, this was related to the administration of caprolactam. A lowest-observed-adverse-effect level (LOAEL) of 500 mg/kg/d and a no-observed-adverse-effect-level (NOAEL) of 250 mg/kg/d were identified.

Powers et al. (1984) administered caprolactam in the feed to F344, Sprague-Dawley, and Wistar rats for 90 d at 0, 0.01, 0.05, 0.1, and 0.5% (or 0, 5, 25, 50, and 250 mg/kg/d). The authors measured glomerular filtration rates (by ³H-inulin clearance) and other renal parameters (urine volume, sodium, potassium, chloride, osmolality, protein, creatinine, glucose [a proximal tubular damage marker], and alkaline phosphatase [a

marker for kidney brush border membrane damage]). Blood samples were analyzed for blood urea nitrogen (BUN), protein, creatinine, hematocrit, sodium, potassium, and chloride. There were no significant differences in the urine parameters in any of the three strains of rats. However, a dose-related increase in BUN was observed in male F344 and Sprague-Dawley rats in the 0.1% (not statistically significant) and 0.5% caprolactam groups. If the compound is a suspect nephrotoxin, BUN is used as a marker for glomerular filtration rate (GFR); BUN rises when GFR slows. A LOAEL and a NOAEL for the increased BUN are identified as 250 mg and 50 mg/kg/d, respectively. Although there was an increase of about 15% in the ratio of kidney weight to body weight in 0.1%- and 0.5%-dose groups of male rats, this effect was not seen in female rats. Renal histopathology data showed that although eosinophilic hyaline droplets were present in the tubules of all groups including controls, they were found at a higher concentration in higher-dosage groups. In the 0.5% group, there was also an increase in the number of tubules with basophilic and hyperplasic epithelial cells. The authors suggested that a slight nephrosis might be present because of 0.5% caprolactam. Furthermore, in the 0.5% group, the frequency and degree of chronic inflammation and interstitial lymphoid cells were the highest. This occurred without any change in chronic inflammation or nephropathy. These effects were seen only in male rats of all strains, and any effect seen in females was not consistent with the dose.

Rabbits given caprolactam at 500 mg/kg/d for 6 months (mo) exhibited cellular changes in the gastric and intestinal mucosa along with lower hemoglobin values. The details of how the caprolactam was administered are not available (Savelova 1960, cited by Gross 1984).

NTP (1982) conducted a 14-d repeated-dose study in which groups of five F344 male and female rats and groups of five B6C3F1 male and female mice were fed a diet containing caprolactam at 0, 5,000, 10,000, 15,000, 20,000, or 30,000 ppm (estimated dose range of 0-4,500 mg/kg/d for rats and mice). There was no mortality in any of the species or sexes of these animals. However, NTP reported pale, mottled kidneys in all treated groups of dosed male rats in incidences of 60-100%. No histopathology was carried out to identify any specific lesions that coincide with the mottled kidneys. No such changes were seen in mice. In both the acute and 14-d studies, NTP did not look at any serum or urine clinical chemistry. Several studies have reported decreased weight gain and thus reduced body weight in animals administered caprolactam by various

routes. When caprolactam was given in drinking water to rats (670 mg/kg/d, duration not known), weight loss was observed, and it was determined that it was due in part to decreased water intake (Goldblatt et al. 1954). Because of the lack of detailed data, this study cannot be used for AC derivations.

NTP also conducted a 13-wk subchronic feeding study (NTP 1982) in which groups of male and female F244 rats were fed diets containing caprolactam at 0, 625, 1,250, 2,500, 5,000, or 7,500 ppm (estimated dose range of 0-1,125 mg/kg for male and female rats). Male and female mice were fed diets containing caprolactam at 0, 5,000, 10,000, 15,000, 20,000, or 30,000 ppm (estimated dose range of 0-6,000 mg/kg). There were weight gain depressions and a reduction in food intake in all caprolactam-treated groups of rats. In the highest-dose group of rats, food consumption was decreased by 23% in males and by 19% in females. No compound-related adverse histopathology was noted. In mice, there were two deaths at the high dose, and as in rats, depressions in weight gain and reduced food consumption were noted.

In the early 1970s, the Central Institute of Nutrition and Food Research conducted dose-range finding and subchronic toxicity studies on caprolactam in the feed using two strains of males and females of Wistarderived (CIVO strain) and Sprague-Dawley rats at two different concentration ranges. In the 28-d study, with caprolactam constituting 5% of the diet (about 3.8 g/kg), renal damage consisting of hyaline droplet degeneration in the epithelium of the proximal convoluted tubules was reported in both sexes of CIVO Wistar-derived rats, with minimal changes (only in males) at 1% (850 mg/kg). In the 90-d study, females of both strains of rats seemed to be less sensitive to these effects. The above mentioned effects were seen at doses greater than 0.05% in Sprague-Dawley male rats and greater than 0.3% in the Wistar-derived male rats (Wijnands and Fern 1969; de Knecht-van Eekelen and van der Meulen 1970, cited in Gross 1984). Increased kidney weights were seen only in male rats of both strains. No proteinuria could be demonstrated. The nephrotoxicity of caprolactam could be supported by the degeneration in the epithelium of the convoluted tubules of rats receiving caprolactam by ip at 50 mg/kg or 100 mg/kg for 6 mo (see Gross 1984). Because these data were presented only as an abstract and details of the data were not available for review, this data cannot be used for AC derivation. However, this study indicated that caprolactam might be a nephrotoxin.

NTP (1982) conducted a chronic feed study in which male and female F344 rats and male and female B6C3F1 mice were given caprolactam in their diet at 0, 3,750, or 7,500 ppm (estimated doses for rats of 0, 560, and 1,120 mg/kg/d) or 0, 7,500, and 15,000 ppm (estimated doses for mice of 0, 1,500, or 3,000 mg/kg/d) for a period of 103 wk. Evaluations were made of grossly visible lesions and histopathology of all organs of animals surviving at the end of 105 wk (103 wk plus 2 wk on a normal diet). Necropsies were performed. Only periodic data on body weight and food consumption were collected, and no serum or urine clinical chemistry measurements were done. From the dose-related decrement in mean body weight gain and feed consumption, it appears that the LOAEL is 1,120 mg/kg/d for the rat, and 3,000 mg/kg/d for mice (NTP 1982).

The various types of neoplasms occurring in dosed animals did not appear to be related to caprolactam ingestion. The degenerative and inflammatory lesions of the type and frequency seen in dosed rats are usually observed as a function of age in Fischer rats.

Epsilon-aminocaproic acid, from which caprolactam is derived, has been known to cause allergic contact dermatitis (Shono 1989). Tanaka et al. (1993) presented a case report from Japan in which a 43-year (y)-old woman with a history of wearing nylon body stockings developed scaly erythema on her trunk. The lesions were resolved after she stopped wearing the body stocking. She also developed the same lesions around her waist from wearing nylon panty hose. When a patch test was done using the monomer, epsilon-aminocaproic acid 3% and 5% in petrolatum, positive reactions were obtained for both concentrations. Other cases of dermatitis have been reported in Russian factory workers involved in nylon synthesis (Kelman 1986). In one case history, a 62-y-old man who had worked in a textile factory (in the nylon drawing factory) for 29 y, presented with an 18-mo history of itchy erythematous scaly patches of eczema involving the neck, chest, and extensor limbs, although the back and flexor limbs were also involved to a lesser degree (Aguirre et al. 1995). This individual had positive reaction in the open test and patch test when caprolactam 5% aqueous patches were used. Fifteen healthy

controls tested negative to caprolactam 5% aq. When the subject left work for 2 mo, his lesions completely resolved. These results indicate that there are only isolated case reports of allergic contact dermatitis in spite of the very wide use of nylon products. No surveys are available.

Under the International Program on Chemical Safety (IPCS), caprolactam was tested extensively in short-term mutagenicity tests, and the results were published in Progress in Mutation Research, Vol. 5 (IPCS 1985). In addition, a special issue of Mutation Research (Ashby and Shelby 1989a) was devoted to studies on in vitro and in vivo genotoxicity of caprolactam (Ashby and Shelby 1989a, b). A summary of the genetic toxicology of caprolactam was published by Brady et al. (1989). A summary of caprolactam genotoxicity studies that had been present in the International Agency for Research on Cancer (IARC) monographs, Vol. 71 (1999) has been modified and included as Table 6-3.

Negative results were found in several in vitro and in vivo short-term genetic tests of caprolactam-treated bacterial, yeast, or fungal systems with or without exogenous metabolic activation as measured by gene mutations, gene conversions, or aneuploidity. In Drosophila melanogaster, caprolactam induced somatic cell mutations in four studies and a marginal increase in sex-linked recessive mutations in one study. No gene mutations were seen in mammalian cells exposed in vitro to caprolactam. In human lymphocyte cultures in vitro, caprolactam failed to induce sister chromatid exchanges, but it increased chromosomal aberrations.

Caprolactam treatment in vivo did not increase DNA single-stranded breaks in hepatocytes, did not induce sister chromatid exchanges, micronuclei, or chromosomal aberrations in mouse bone marrow, and did not induce morphologic abnormalities in mouse sperm. Salamone (1989) orally treated mice with various doses of caprolactam in corn oil (0, 222, 333, 500, 750, and 1,125 mg/kg) for 5 d. Sperm samples were taken 35 d after the last treatment. Abnormal sperms per 1,000 sperm were counted. There was no difference between the percentage of abnormal sperms in controls and treated groups (Henderson and Grigliatti 1992). In another study, unscheduled DNA synthesis (UDS) in F344 rat spermatocytes exposed to caprolactam was studied (Working 1989). F344 rats were gavaged with a single bolus dose of caprolactam

in water at 750 mg/kg, spermatogenic cells were isolated at 12, 24, or 48 h after treatment, and induction of UDS was measured by quantitative autoradiography. Caprolactam did not induce UDS (see Table 6-3). In one study, when cultured human lymphocytes from one male donor and one female donor were treated at 7.5 mg/mL, slightly increased frequencies of chromosomal aberrations were seen in the cells from the male donor but not the female (Kristiansen and Scott 1989). However, in another study, Sheldon (1989b) tested caprolactam in the in vitro human lymphocyte cytogenetic assay in the presence and absence of S9 mix at dose concentrations up to 5.5 mg/mL using lymphocytes obtained from a male and a female donor. Statistically significant increases in chromosomal damage were observed in cells from both donors. Sheldon (1989b) concluded that caprolactam induces chromosomal damage in human lymphocytes in vitro at high-dose concentrations. Thus, the results are inconsistent.

Although in a few systems mutagenic responses have been documented, they are weak and occur only at very high concentrations. The results, taken as a whole, indicate that there is not strong evidence to conclude that caprolactam is genotoxic (IARC 1999).

No epidemiologic data relevant to the carcinogenicity of caprolactam were available.

IARC has designated caprolactam as a Group 4. This is based on the studies described below. IARC states that a Group 4 “agent (mixture, exposure circumstance) is probably not carcinogenic to humans.”

A carcinogenesis bioassay of caprolactam in F344 rats and B6C3F1 mice (feed study) conducted by Litton Bionetics Inc. for NTP (Toxicology Report 214) reported that diets containing caprolactam at 3,750-7,500 parts per million (ppm) given to groups of 50 male or female F344 rats (estimated doses of 560 and 1,120 mg/kg, respectively) and 7,500 or 15,000 ppm given to groups of 50 male and female mice for 103 wk (with proper controls) did not show any carcinogenic activity (NTP 1982). Only the mean body weights of dosed rats and mice decreased compared to those of control groups in a dose-related fashion. The incidences of caprolactam-treated animals with specific-site tumors did not differ significantly from those seen in controls. For example, pituitary

carcinomas in male rats showed a positive trend with dose: 0/46 in controls, 0/49 at low dose, and 3/43 at high dose. In female rats, the trend was not dose specific: 2/49 in control, 1/49 at medium dose, and 1/47 in high dose. Although increased proportions of interstitial-cell tumors of the testis were seen in the highest-dose groups, this occurred at concentrations of historical controls. This response was not considered to be associated with the administration of caprolactam.

The tumor-promoting potential of caprolactam by oral feeding was also evaluated in two multistage tumor promotion studies in rats. The studies indicated that caprolactam is not a tumor promoter. Fukushima et al. (1991) conducted a complex multistage initiation-promotion study on a group of male F344/DuCrj rats. Rats received ip injections of N-nitrosodiethylamine followed by four ip injections of N-methyl-N-nitrosourea followed by 0.1% N-bis (2-hydroxy-propyl) nitrosamine in their drinking water for 2 wk. The rats were then given caprolactam at 10,000 mg/kg diet for 16 wk. In addition, five rats received vehicle without carcinogens during the first step of the treatment period and were then given 10,000 mg/kg diet for 16 wk. Histologic examination of most organs and any gross lesions and quantification of the placental form of glutathione S-transferase positive (GST-P) foci of the liver were performed after 20 wk of caprolactam treatment. Caprolactam showed no modifying effect on carcinogenesis in any organ after a 16-wk treatment.

In another study by Hasegawa and Ito (1992), male F344 rats were administered a single ip injection of N-diethyl nitrosamine in 9% weight per volume (w/v) of saline at 200 mg/kg. After a 2-wk recovery period, rats were given either 10,000 mg/kg diet or basal diet for 6 wk. A two-thirds partial hepatectomy was done at week 3, and animals were killed at week 8. GST-P foci of the liver were quantified. There were no significant differences in either the numbers or areas of GST-P foci between the caprolactam-treated group and the caprolactam-untreated group, indicating that there was no modifying effect by caprolactam.

Recently, caprolactam was tested in a transgenic p53-deficient mouse bioassay (p53 +/ heterozygous model) because this mouse has been found to be highly susceptible to the induction of tumors when challenged with certain carcinogens. Wild-type strain C57BLp53 (+/) and C57BL/6 transgenic p53-deficient mice were exposed to caprolactam at 15,000 ppm in their diet for up to 26 wk. The cell-replicating fraction (RF) in the liver was determined, and no consistent and pertinent changes were present in the treated mice. Caprolactam did not produce neoplasms of any type in p53 (+/) mice (Iatropoulos et al. 2001). The data cannot

be considered conclusive, as diethylnitrosamine, a classical liver carcinogen, did not show any tumors in this mouse strain.

Russian animal studies have reported that caprolactam administration affected both male and female reproductive performance in rats. Effects included shortening of the estrus cycle, a reduction in the number of follicles as well as an increase in corpora lutea, and damage to gonadal function. Because all papers are in Russian, with no English abstracts, we were unable to critically evaluate them and provide descriptions; therefore they could not be included in this document.

Serota et al. (1988) conducted a three-generation reproductive toxicity study in F344 albino rats on the effects of caprolactam in the diet at 0, 1,000, 5,000 and 10,000 ppm (estimated doses of 0, 50, 250, and 500 mg/kg). In this study, rats were given caprolactam in the diet, and in each generation, the rats were mated 10 wk after a dietary exposure to caprolactam. No treatment-related deaths, clinical signs, changes in reproductive performance, number of pups, or gross pathology findings was observed in the parental animals. However, decreases in body weight were observed in the P2 and P3 generations of rats, which showed that in utero and postnatal exposures of offspring resulted in body weight change at doses that did not cause substantial body weight reductions in adult P1 generations. It should further be noted that although caprolactam might be maternally toxic, fetotoxic, and developmentally toxic, it was not teratogenic. The offspring data revealed no treatment-related effect with respect to gross appearance, gross pathology, survival, number of pups, or percentage of male pups.

Unrelated to reproductive toxicity, histopathology findings indicated that in the high-dose P1 males, a slight increase in the severity of spontaneous nephropathies occasionally accompanied by granular casts was observed.

In a three-generation reproduction study discussed earlier (Serota et al. 1988), the body weights (on days 1, 7, and 21 of lactation) of offspring of rats administered caprolactam at 1,000, 5,000, 10,000 ppm in their diet for 10 wk revealed significantly lower mean values in the high-

dose male and female animals of all filial generations. The mean body weights of both sexes in the mid-dose group were generally lower than those of the controls. These effects seem to be related to the ingestion of caprolactam.

In a developmental study designed for examining the teratogenic effects of caprolactam ingestion, Gad et al. (1987) reported that in rats gavaged orally with caprolactam during days 6-11 of gestation at doses of 100, 500, or 1,000 mg/kg/d, the incidence of resorption was nearly 10-fold higher in rats administered 1,000 mg/kg and mean implantation efficiency at the this dose was about 65% compared to 82% in controls. In this study, clinical signs were observed and noted during days 0, 6, 11, 15, and 20 of gestation. On day 20 of gestation, all surviving females were killed to examine the uterus, ovaries, and fetuses.

Gad et al. (1987) evaluated the developmental toxicity potential of caprolactam in rats and rabbits after an oral dose of various concentrations of caprolactam. In the first study, mature 8-wk-old female F344 rats (20/dose group) were intubated with caprolactam in distilled water at 0, 100, 500, or 1,000 mg/kg/d on gestation days 6-15, the period of organogenesis in rats. In the 1,000 mg/kg group, the maternal mortality was greater than 50%. On days 6-11, the mean body weight changes (and food consumption) in the two highest-dose groups were less than those of controls and the lowest-dose group. The mean incidence of resorption in the highest-dose group was nearly 10-fold higher than in the controls and all other dose groups. Skeletal variants (including incomplete ossification of the skull or vertebral column and the presence of extra ribs) were markedly increased among offspring from animals exposed to the highest dose (Gad et al. 1987). No other dose-related malformations or anomalies were noted among the offspring of any exposure group. For this rat oral gavage study in which doses of 0, 100, 500, and 1,000 mg/kg/d on gestation days 6-15 were used, 1,000 and 500 mg/kg/d appear to be the LOAEL and NOAEL, respectively, for fetal resorption (Gad et al. 1987).

In the rabbit study, pregnant New Zealand white rabbits were intubated with caprolactam intragastrically with water at 50, 150, or 250 mg/kg/d on gestation days 6-28. Four out of 25 rabbits dosed with 250 mg/kg had convulsions immediately. On day 29 of gestation, dead fetuses, empty implantation sites, and resorptions were recorded. Mortality was observed in the highest-dose group, and maternal body weight gain was significantly depressed. Significantly lower mean fetal weights were seen in the 150 and 250 mg/kg groups. An increased incidence of thirteenth ribs among fetuses whose mothers had been exposed to the high-

est concentration of caprolactam was also reported (Gad et al. 1987). A NOAEL of 50 mg/kg is identified in this species. Pregnant astronauts are barred from spaceflight, and consideration of developmental effects in this section is included only for completeness of analyses (NRC 2000).

A large number of sedative and convulsing agents are lactam derivatives, and thus there has been considerable interest in whether caprolactam can induce such activity. When administered parenterally, caprolactam caused convulsions and death. It has been shown that at high doses, caprolactam has convulsant properties (Elison et al. 1971), and the convulsion-inducing caprolactams may be acting as gamma-aminobutyric acid antagonists (Kerr et al. 1976; Skerritt et al. 1985; Kerr et al. 1986). Some of the LD₅₀ studies have reported that death was caused by convulsions. Caprolactam itself is a weak agent compared to the alkylated derivatives of this lactam. Lack of effect of caprolactam when injected intraventricularly is probably because of its poor penetration of the blood brain barrier. The metabolite of caprolactam, the epsilon-aminocaproic acid, is weaker than caprolactam as a convulsant. In a Russian study, it was reported that rats administered caprolactam at a concentration of 15 mg/kg/d (probably in feed) for 2 mo developed loss of reflexes and other nonspecific effects probably related to neurologic systems (Savelova 1960, cited in Gross 1984). Because of a lack of available details on these experiments, data on neurologic end points could not be used for AC derivation.

A list of studies on the toxicity from oral intake of caprolactam is summarized in Table 6-4.

The Office of the Safe Drinking Water of the U.S. Environmental Protection Agency (EPA) has not proposed a maximum contaminant level (MCL), a secondary maximum contaminant level (SMCL), nor a maximum contaminant level goal (MCLG). The Agency for Toxic Substances and Disease Registry (ATSDR) has not developed a toxicologic profile for caprolactam, and thus no minimal risk levels (MRLs) are known for oral exposure through drinking water or diet. In 1977, the Safe Drinking Water Committee of the National Research Council (NAS

1977) concluded that in view of the paucity of data on the long-term oral toxicity of caprolactam, estimates of the toxicity of low-dose oral exposures cannot be made with any confidence. However, EPA has derived an RfD (oral reference dose) for caprolactam (EPA 1988) (see Table 6-5) using the reductions in body weights of offspring in the three-generation reproduction study with rats (Serota et al. 1988).

The following discussion provides a rationale for proposing SWEG values for caprolactam in drinking water for 1 d, 10 d, 100 d, and 1,000 d (see Table 6-6). The values listed were based on ACs for each duration according to Methods for Developing Spacecraft Water Exposure Guidelines (NRC 2000). An intraspecies factor is not usually used, because astronauts come from a homogenous healthy population and there is no evidence of a group of healthy persons having excess susceptibility to caprolactam. However, if there are known variations in the metabolism of the compound or susceptibility to the compound regardless of health status, then an individual factor could be applied. Although developmental data are included in the document under the toxicity section, it is only for comprehensiveness, and the data will not be used in setting SWEGs, because pregnant astronauts are barred from spaceflight.

A search of the literature indicates there are few case reports and no survey reports of known allergic reactions to caprolactam in an occupational setting. There are no reports of such allergic reactions from ingestion. The long-term oral studies on animals have not evaluated this end point. A review of the caprolactam ingestion studies indicates that there are very few toxicity end points. Most of the reported studies indicate reduced food intake and decreased body weight gain at high doses. It is not clear if the reduced food intake noted in many studies included here is because of taste aversion (palatability problem) or because of neurologic effects.

EPA derived the RfD for caprolactam from the results of the three-generation reproduction study by Serota et al. (1988). This study has been described in detail under “Reproductive Toxicity.” In this study, male and female F344 rats were fed diets containing caprolactam at 0,

1,000, 5,000, or 10,000 ppm (0, 50, 250, or 500 mg/kg/d as calculated by EPA) for three generations. Mean body weight and food consumption were reduced in both parental generations of the 250- and 500-mg/kg/d groups. Body weight of offspring was reduced at these dietary concentrations. Histopathology indicated a slight increase in the severity of nephropathy in males in the high-dose group of the first parental generation. No adverse effects were noted at 1,000 ppm (50 mg/kg/d); therefore, 1,000 ppm was chosen as the NOAEL to serve as the basis for the RfD. Thus, EPA did not use any specific end point for obtaining this NOAEL. A modifying factor of 100 for uncertainty was used that included 10 for the species factor and 10 for uncertainty in the threshold for sensitive humans.

Thus, an RfD of 0.5 mg/kg/d or 35 mg/d for a 70 kg human was derived. If we were to extend the calculations for use by NASA, an AC for more than 1,000 d would be as follows:

It should be noted that EPA did not use any time factor even though the three-generation exposure in this study consisted of a total of 30 wk, or 210 d, and not a lifetime for rats or mice.

The data on sensory threshold values measured for caprolactam cannot be used. The olfactory threshold was found to be 0.3 mg per cubic meter (m³). Caprolactam is not expected to vaporize out of water because of its low vapor pressure at room temperature. Therefore, an odor threshold value is not calculated. NASA taste tests indicated that at least up to a concentration of 13 mg/L, water containing caprolactam will not be objectionable to drink. It may be overly conservative to use this value as a 1-d AC because such data for higher concentrations are not available.

TAT and TPO are two enzymes that are involved in the first steps of tyrosine and tryptophan metabolism. After oral administration of a bolus of caprolactam at 1.5 g/kg to male F344 rats, there were marked inductions of TAT and TPO activities at 3 h with maximum inductions at 6 h (Friedman and Salerno 1980). Although these data pertain to the induction of two important amino acid metabolizing enzymes by caprolactam, whether it can be considered an adverse effect in this acute exposure is questionable. Thus, the data were not used for deriving a 1-d AC.

A second study that was considered for the 1-d AC was that of Kitchin and Brown (1989). Oral ingestion of caprolactam at 425 mg/kg (twice) by gavage resulted in a significant increase in the activity of ALT (30%) over controls. Clinically, an increase in this enzyme indicates hepatocyte necrosis or damage. While 850 mg/kg appears to be a LOAEL, a NOAEL was not identified, because there were no dose-response data. However, with the changes in the liver protein metabolism enzymes, TAT and TPO, reported by Friedman and Salerno (1980), in conjunction with these changes reported by Kitchin and Brown (1989), it can be concluded that caprolactam resulted in adverse effects on liver. Because the investigators administered one dose at 21 h and one more dose 4 h before sacrifice, the effect may have been from the combined dose of 850 mg/kg. Therefore, this dose will be used in deriving an AC.

The 1-d AC was calculated based on hepatotoxicity as follows:

where

850 mg/kg = LOAEL;

70 kg = nominal body weight;

10 = LOAEL to NOAEL;

10 = species extrapolation factor; and

2.8 L/d = nominal water consumption.

Data from Friedman and Salerno (1980), in which rats fed caprolactam in their diet for 7 d at doses of 1.1 and 1.8 g/kg (calculated by the authors), showed inhibition of liver protein synthesis and also induction of liver TPO measured at the end of the treatment period. Diet restriction increased protein synthesis by at least 63%, but a dose of 1.8 g/kg blocked this increase. Also, TPO activity was found to be increased at the 1.1 g/kg dose. Therefore, 1.1 g/kg is used as the LOAEL for the induction of this enzyme. Although we considered that the change in activity of the enzyme may be an adaptive response to a single dose, the change in conjunction with an inhibition of protein synthesis allows us to use these data to derive a 10-day AC after applying a time extrapolation factor.

10-d AC for adverse effect on amino acid metabolism is calculated as follows:

where

1,100 mg /kg = LOAEL;

70 kg = nominal body weight;

10 = LOAEL to NOAEL;

10 = species extrapolation factor;

2.8 L/d = nominal water consumption; and

10 d/7 d = time extrapolation factor.

Powers et al. (1984) studied three strains of male rats fed diets containing caprolactam at doses of 0, 0.01, 0.05, 0.1, 0.5, and 2.5% (or 0, 1, 10, 50, and 250 mg/kg/d) for 90 d. A dose-related increase in BUN was observed in male F344 and Sprague-Dawley rats at 0.1% and 0.5% dose groups but was statistically significant only in the 0.5% dose group. If the compound is a suspected nephrotoxin, BUN is used as a marker for GFR; BUN rises when GFR slows. A LOAEL and NOAEL for increased BUN are identified as 250 and 50 mg/kg/d, respectively. Although there was an increase of about 15% in the kidney weight to body weight ratio in the 0.1% and 0.5% dose groups of male rats, this effect was not seen in female rats. The increase in the ratio is because of an absolute increase in kidney weight, probably indicating hypertrophy of the kidney. Renal histopathology data revealed that, although eosinophilic hyaline droplets were present in the tubules of all groups including controls, this was found at a higher concentration in higher-dosage groups. Although the numbers were minimal, the number of rats affected increased with the dose of caprolactam. These effects were seen only in male rats of all strains, and any effect seen in females was not consistent with the dose.

Thus, a 100-d AC for renal toxicity can be derived as follows:

where

50 mg/kg/d = NOAEL for BUN;

70 kg = nominal body weight;

10 = species extrapolation factor;

2.8 L/d = nominal water consumption; and

100 d/90 d = time extrapolation factor.

Another study that was considered for calculating a 100-d AC was Serota et al. (1988), a three-generation reproductive study in which rats were fed diets containing caprolactam at 0, 50, 250, or 500 mg/kg/d. In this study, the authors noted that there was a slight increase in the severity of nephropathy on histologic examination of the male rats of the 500 mg/kg/d group. The duration of each generation was 10 wk (70 d). Body weights of the parental generations and their offspring (of the 250 and 500 mg/kg/d dose groups) were significantly reduced. The reported slight increase in severity in spontaneous nephropathy in exposed groups, ac-

companied by the presence of granular casts in some rats, was used as a toxicologic end point. A NOAEL of 250 mg/kg/d for severity of nephropathy was identified. The developmental effects observed in the study are not the basis for the 100-d AC.

Thus, a 100-day AC for nephropathy can be derived as follows:

where

250 mg/kg = NOAEL;

70 kg = nominal body weight;

10 = species extrapolation factor;

2.8 L/d = nominal water consumption; and

100 d/70 d = time extrapolation factor.

The SWEG for the 100-d duration will be accepted as the SWEG for 10-d duration, instead of the calculated value of 200 mg/L as 10-d AC. The reasons are outlined below.

The 1-d AC was based on hepatotoxic effects derived from a LOAEL of 850 mg/kg. A factor of 10 was used for LOAEL to NOAEL. The 10-d AC was based on adverse effects on amino acid catabolism and protein synthesis (Friedman and Salermo 1980). This data was used to calculate a 10-d AC of caprolactam at 200 mg/L. In this study, hepatotoxic indices were not measured. It is quite possible that 10 d of continuous ingestion of water containing caprolactam at 200 mg/L will lead to hepatoxicity, because it approaches the LOAEL for hepatotoxicity observed in 1 d and will not provide any margin of safety. The rationale for using 100 mg/L as the 10-d AC and for adopting the 100-d AC obtained using the Powers et al. (1984) study is as follows: NTP (1982) conducted a carcinogenesis bioassay of caprolactam in the feed for 103 wk in F344 rats and B6C3F1 mice. The study also included a 1-d gavage and a 14-d feed protocol. In these protocols, the focus was on food and water consumption, body weight gain, and survival rates in addition to general clinical observations and necropsy. NTP reported that at the end of 2 wk, there were no deaths; however, pale, mottled kidneys occurred in all groups of dosed rats in incidences of 60-100%. NTP did not discuss the significance of this observation. The calculated dose rates for male and female rats for the lowest-dose group that showed the pale, mottled kid-

neys are about 860 mg/kg/d, and 750 mg/kg/d respectively. Although the biologic significance of mottled kidney is not clear, it appears that the kidney is somewhat affected. This is somewhat in concordant with the observations from the 90-d study of Powers et al. (1984) that showed that the kidney is a target organ. NTP did not conduct any clinical chemistry on blood or urine. Therefore, it was decided to adopt the 100-d AC derived from the Powers et al. (1984) study for 10 d, because it would offer enough margin of safety for kidney effects and for hepatotoxicity. In addition, the 28-d and 90-d rat subchronic caprolactam-feed study conducted by the Central Institute of Nutrition in the Netherlands reported changes in the kidney resulting from caprolactam feeding. However, as details of the study were unavailable as a full report, the data could not be critically evaluated for NASA in the determination of 10-d or 100-d AC.

The 2-y carcinogenesis bioassay study sponsored by NTP (1982) is the only study that can be considered for 1,000-d AC derivation. In this study, rats and mice were exposed to caprolactam in their diet for 2 y (male and female F344 rats to 0, 3,750, or 7,500 ppm, and male and female B6C3F1 mice to 0, 7,500, or 15,000 ppm). These mean concentrations for males and females corresponded to estimated doses of 0, 560, and 1,120 mg/kg/d for rats and 0, 1,500, and 3,000 mg/kg/d for mice. The feed consumption in the highest-dose groups was only 70-80% of that of controls, and thus, body weights were lower. The NRC committee had recommended in the past that body weight changes should not be used to set ACs when it is known that changes occurred because of reduced food consumption. According to the NTP report (NTP 1982), the frequency of the large number of degenerative, proliferative, and inflammatory lesions encountered in caprolactam-treated rats was not different from that in control rats. However, there were some observations of toxicologic concern. One is that in rats, testicular interstitial cell tumors were observed in increasing proportions as a function of dose, even though they occur historically at 80%. The Cochran-Armitage linear trend analysis was statistically significant. Also, carcinomas of the pituitary were observed in increased proportions in high-dose male rats. The Cochran-Armitage linear trend analysis was statistically significant in the positive direction. However, the committee recommended that NTP data could not be used to set an AC based on non-neoplastic lesions, because

NTP concluded that it did not find any evidence of neoplastic or non-neoplastic lesions related to dietary caprolactam.

The 100-d AC was adopted as the 1,000-d AC in the absence of sufficient data for long- term ingestion. The use of the 100-d AC without any time factor for 1,000 d was justified by the fact that caprolactam does not accumulate in the body and is excreted efficiently. Furthermore, if one were to use the lowest dose (560 mg/kg/d) used in the NTP 2-y carcinogenicity study for rats, which did not produce any compound-related long-term adverse tissue pathology, and calculate an AC using only the species extrapolation factor, a concentration of 1,400 mg/L would be obtained as the AC. Thus, the use of 100 mg/L as the 1,000-d AC is justified and conservative. See Table 6-7 for a summary of all ACs and final SWEGs for all durations.

ACGIH (American Conference of Governmental Industrial Hygienists). 1991. Documentation of Threshold Limit Values and Biological Exposure Indices, 6th Ed. American Conference of Governmental Industrial Hygienists, Cincinnati, OH.

Aguirre, A., R. Gonzalez Perez, J. Zubizarreta, N. Landa, C. Sanz de Galdeano, and J.L. Diaz Perez. 1995. Allergic contact dermatitis from epsilon-caprolactam. Contact Dermatitis 32:174-175.

Ashby, J., and M.D. Shelby, eds. 1989a. Assessment of Genotoxicity In Vivo of the Rodent Non-Carcinogens Caprolactam (CAP) and Benzoin (ZOIN). Mutat. Res.

Ashby, J., and M.D. Shelby. 1989b. Overview of the genetic toxicity of caprolactam and benzoin. Mutat. Res. 224:321-324.

Bermudez, E., T. Smith-Oliver, and L.L. Delehanty. 1989. The induction of DNA-strand breaks and unscheduled DNA synthesis in F-344 rat hepatocytes following in vivo administration of caprolactam or benzoin. Mutat. Res. 224(3):361-364.

Brady, A.L., H.F. Stack, and M.D. Waters. 1989. The genetic toxicology of benzoin and caprolactam. Mutat. Res. 224:391-403.

Carls, N., and R.H. Schiestl. 1994. Evaluation of the yeast DEL assay with 10 compounds selected by the International Program on Chemical Safety for the evaluation of short-term tests for carcinogens. Mutat. Res. 320(4):293-303.

de Knecht-van Eekelen, A., and H.C. van der Meulen. 1970. Subchronic (90 day) toxicity study with caprolactam in Sprague-Dawley albino rats. Central Institute for Nutrition and Food Research, Netherlands, page 13.

Elison, C., E.J. Lien, A.P. Zinger, M. Hussain, G.L. Tong, and M. Golden. 1971. CNS activities of lactam derivatives. J. Pharm. Sci. 60:1058-1062.

EPA (U.S. Environmental Protection Agency). 1988. Health and environmental effects profile for caprolactam. ECAO-CIN-G018. Office of Health and Environmental Assessment, U.S. Environmental Protection Agency, Cincinnati, OH.

Fahrig, R. 1989. Possible recombinogenic effect of caprolactam in the mammalian spot test. Mutat. Res. 224:373-375.

Foureman, P., J.M. Mason, R. Valencia, and S. Zimmering. 1994a. Chemical mutagenesis testing in drosophila. IX. Results of 50 coded compounds tested for the National Toxicology Program. Environ. Mol. Mutagen. 23:51-63.

Foureman, P., J.M. Mason, R. Valencia, and S. Zimmering. 1994b. Chemical mutagenesis testing in Drosophila. X. Results of 70 coded chemicals tested for the National Toxicology Program. Environ. Mol. Mutagen. 23:208-227.

Friedman, M.A. and A.J. Salerno (1980). Influence of caprolactam on rat-liver tyrosine aminotransferase and tryptophan oxygenase. Food Cosmet Toxicol. 18:39-45.

Fukushima, S., A. Hagiwara, M. Hirose, S. Yamaguchi, D. Tiwawech, and N. Ito. 1991. Modifying effects of various chemicals on preneoplastic and neoplastic lesion development in a wide-spectrum organ carcinogenesis model using F344 rats. Jpn. J. Cancer Res. 82:642-649.

Gad, S.C., K. Robinson, D.G. Serota, and B.R. Colpean. 1987. Developmental toxicity studies of caprolactam in the rat and rabbit. J. Appl. Toxicol. 7:317-326.

Goldblatt, M.W., M.E. Farquharson, G. Bennett, and B.M. Askew. 1954. Epsilon-Caprolactam. Br. J. Ind. Med. 11(1):1-10.

Gross, P. 1984. Biologic activity of epsilon-caprolactam. Crit. Rev. Toxicol. 13:205-216.

Hasegawa, R, and N. Ito. 1992. Liver medium-term bioassay in rats for screening of carcinogens and modifying factors in hepatocarcinogenesis. Food Chem. Toxicol. 30(11):979-992

Henderson, D.S., and T.A. Grigliatti. 1992. A rapid somatic genotoxicity assay in Drosophila melanogaster using multiple mutant mutagen-sensitive (mus) strains. Mutagenesis 7:399-405.

Hohenesee, F. 1951. On the pharmacological and physiological effects of epsilon-caprolactam [in German]. Faserforsch. Textiltech. 2:299-303.

Howard, C.A., T. Sheldon, and C.R. Richardson. 1985. Tests for induction of chromosomal aberrations in human peripheral lymphocytes in culture. Pp 457-276 in Progress in Mutation Research, Vol 5, Evaluation of Short-Term Tests for Carcinogens, J. Ashby, and F.J. deSerres eds. Amsterdam: Elsevier Science Publishers.

HSDB (Hazardous Substances Data Bank). 2006. Caprolactam. U.S. National Library of Medicine. [Online]. Available at: http://toxnet.nlm.nih.gov/cgibin/sis/search/f?./temp/~SCop00:1 [access September 5, 2006].

IARC (International Agency for Research on Cancer). 1986. IARC Monographs on the Evaluation of Carcinogenic Risk of Chemicals to Humans, Vol 39: Some chemicals used in plastics and elastomers. Lyon, France: International Agency for Research on Cancer.

IARC (International Agency for Research on Cancer). 1999. IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 71: Reevaluation of some organic chemicals, hydrazine and hydrogen peroxide. Lyon, France: International Agency for Research on Cancer.

Iatropoulos, M.J., A.M. Jeffrey, G. Schluter, H.G. Enzmann, and G.M. Williams. 2001. Bioassay of manifold and caprolactam and assessment of response to diethylnitrosamine in heterozygous p53-deficient (+/) and wild type (+/+) mice. Arch. Toxicol. 75:52-58.

IPCS (International Program on Chemical Safety). 1985. Overview and conclusions of the IPCS collaborative study on in vitro assay systems. Pp. 117-174 in Progress in Mutation Research, Vol. 5, Evaluation of short-term

tests for carcinogens, J. Ashby, and F.J. de Serres, eds. Amsterdam: Elsevier Science Publishers.

IRIS (Integrated Risk Information System). 1998. Caprolactam (CASRN: 105-60-2) Oral RfD Assessment. Integrated Risk Information System, U.S. Enviromental Protection Agency, Washington, DC [online]. Available: http://www.epa.gov/iris/subst/0357.htm [accessed May 2, 2006].

Kelman, G.R. 1986. Effects of human exposure to atmospheric epsilon-caprolactam. Hum. Toxicol. 5:57-59.

Kerr, D.I., B.J. Dennis, E.L. Breuker, R.H. Prager, A.D. Ward, and T. Duong. 1976. Antagonism of GABA-mediated inhibition in the central nervous system by caprolactam derivatives. Brain Res. 110:413-416.

Kerr, D.I., J. Ong, R.H. Prager, and D.A. Ward. 1986. Caprolactam-barbiturate interaction at the GABAA receptor complex in the guinea-pig intestine. Eur. J. Pharmacol. 124:203-206.

Kirk, L.K., B.A. Lewis, D.A. Ross, and M.A. Morrison. 1987. Identification of ninhydrin-positive caprolactam metabolites in the rat. Food Chem. Toxicol. 25:233-239.

Kitchin K.T., and J.L. Brown. 1989. Biochemical studies of promoters of carcinogenesis in rat liver. Teratog. Carcinog. Mutagen. 9:273-285

Kristiansen, E., and D. Scott. 1989. Chromosomal analyses of human lymphocytes exposed in vitro to caprolactam. Mutat. Res. 224:329-332.

Merck Index. 1989. An Encyclopedia of Chemicals, Drugs, and Biologicals, 11th Ed. S. Budavari, M.J. O’Neil, and A. Smith, eds. Whitehouse Station, NJ: Merck & Co.

Neuhauser-Klaus, A., and W. Lehmacher. 1989. The mutagenic effect of caprolactam in the spot test with (T x HT) F1 mouse embryos. Mutat. Res. 224:369-371.

Norppa, H., and H. Jarventaus. 1989. Induction of chromosome aberrations and sister-chromatid exchanges by caprolactam in vitro. Mutat. Res. 224:333-337.

NRC (National Research Council). 1977. Drinking Water and Health. Washington DC: National Academy Press.

NRC (National Research Council). 2000. Methods for Developing Spacecraft Water Exposure Guidelines. Washington, DC: National Academy Press

NTP (National Toxicology Program). 1982. Carcinogenesis bioassay of caprolactam. F344 Rats and B6C3F1 Mice (Feed Study). Technical Report Series #214 (CAS no: 105-60-2). U.S. Department of Health and Human Services, National Toxicology Program, Research Triangle Park, NC.

Powers, W.J., J.C. Peckham, K.M. Siino, and S.C. Gad. 1984. Effects of subchronic dietary caprolactam on renal function. Pp. 77-96 in Proceedings of a Symposium on an Industry Approach to Chemical Risk Assessment: Caprolactam and related compounds as a case study. Pittsburgh: Industrial Health Foundation.

Salamone, M.F. 1989. Abnormal sperm assay tests on benzoin and caprolactam. Mutat. Res. 224:385-389.

Savelova, V.A. 1960. Maximum allowable concentration for caprolactam in reservoirs. Sanit. Okhrana. Vodoemov Zagryazeniya Prom Stochnymi Vodanis 4:156, as cited in Crit. Rev. Toxicol. 1984, 13:205-216.

Serota, D.G., A.M. Hoberman, M.A. Friedman, and S.C. Gad. 1988. Three-generation reproduction study with caprolactam in rats. J. Appl. Toxicol. 8:285-293.

Sheldon, T. 1989a. An evaluation of caprolactam and benzoin in the mouse micronucleus test. Mutat. Res. 224:351-355.

Sheldon, T. 1989b. Chromosomal damage induced by caprolactam in human lymphocytes. Mutat. Res. 224:325-327.

Shono, M. 1989. Allergic contact dermatitis from epsilon-amino-caproic acid. Contact Dermatitis 21:106-107

Skerritt, J.H., G.A. Johnston, S.C. Chow, R.L. MacDonald, R.H. Prager, and A.D. Ward. 1985. Differential modulation of gamma-aminobutyric acid receptors by caprolactam derivatives with central nervous system depressant or convulsant activity. Brain Res. 331(2):225-233

Statsek, N., and T. Ivanova. 1978. Hygienic investigations of synthetic polyamide fabrics and clothing made thereof [in Russian]. Gig. Sanit. 10:38-41.

Tanaka, M., S. Kobayashi, and S. Miyakawa. 1993. Contact dermatitis from nylon 6 in Japan, Contact Dermatitis 28:250.

Unger, P.D., and M.A. Friedman. 1980. High-pressure liquid chromatography of caprolactam and its metabolites in urine and plasma. J. Chromatogr. 187:429-435.

Unger, P.D., A.J. Salerno, and M.A. Friedman. 1981. Disposition of [14C] caprolactam in the rat. Food Cosmet. Toxicol. 19:457-462.

Vogel, E.W. 1989. Caprolactam induces genetic alterations in early germ cell stages and in somatic tissue of D. melanogaster. Mutat. Res. 224:339-342.

Waddell, W.J., C. Marlowe, and M.A. Friedman. 1984. The distribution of [14C]caprolactam in male, female and pregnant mice. Food Chem. Toxicol. 22:293-303.

Wijnands, W.C., and V.J. Feron. 1969. Range finding (28 day) toxic study with caprolactam in rats. Central Institute for Nutrition and Food Research, the Netherlands.

Working, P.K. 1989. Assessment of unscheduled DNA synthesis in Fischer 344 rat pachytene spermatocytes exposed to caprolactam or benzoin in vivo. Mutat. Res. 224:365-368.

Wurgler, F.E., U. Graf, and H. Frei. 1985. Somatic mutation and recombination test in wings of Drosophila melanogaster. Pp. 325-340 in Progress in Mutation Research, Vol 5, Evaluation of Short-Term Tests for Carcinogens, J. Ashby, and F.J. deSerres, eds. Amsterdam: Elsevier Science Publishers.