Priority Substances List Assessment Report for Hexachlorobutadiene

2.0 Summary of Information Critical to Assessment of "Toxic" Under CEPA 1999

2.1 Identity and physical/chemical properties

Hexachlorobutadiene (CAS registry number 87-68-3), referred to hereafter as HCBD, has the empirical molecular formula C4Cl6, the structural formula shown in Figure 1 and a molecular weight of 260.76 g/mol. HCBD is a colourless liquid with a water solubility of 3.20 mg/L at 25°C (Gradiski et al., 1975), a log K_ow of 4.90 (Chiou, 1985), a vapour pressure of 20 Pa at 20°C (Pearson and McConnell, 1975) and a Henry's law constant of 1044 Pa·m³/mol (Shen, 1982). Synonyms for HCBD include 1,1,2,3,4,4-hexachloro-1,3-butadiene, hexachloro-1,3-butadiene, perchlorobutadiene and perchloro-1,3-butadiene. Additional information on physical and chemical properties of HCBD is presented in Environment Canada (1999).

Figure 1 Structure of HCBD

2.2 Entry characterization

2.2.1 Production, importation and use

HCBD has never been commercially produced in Canada. It is produced as a by-product during the production of certain chlorinated chemicals, such as tetrachloroethylene, trichloroethylene, vinyl chloride, allyl chloride, epichlorohydrin and carbon tetrachloride (U.S. EPA, 1980; Kusz et al., 1984; Choudhary, 1995).

In the past, HCBD was imported into Canada for use as a solvent (Environment Canada, 1979), but it is no longer imported or used (Environment Canada, 1997c). In addition, HCBD was not included on the National Pollutant Release Inventory (NPRI, 1994).

HCBD was used as a solvent for C4 and higher hydrocarbons and elastomers, as a hydraulic fluid, as a heat transfer liquid in transformers and as a chemical intermediate in the production of chlorofluorocarbons and lubricants (U.S. EPA, 1980; Manahan, 1992). It was also used to recover chlorine-containing gas in chlorine plants, in gyroscopes and in insulating fluids, and it had widespread application as a fumigant for treating grapes against Phylloxera in the former Soviet Union, France, Italy, Greece, Spain and Argentina (IARC, 1979; IPCS, 1994). Recent information on the use of HCBD is not available (IPCS, 1994).

2.2.2 Sources and releases

2.2.2.1 Natural sources

There are no natural sources of HCBD in the environment.

2.2.2.2 Anthropogenic sources

In the 1970s, formation of HCBD as a waste by-product was estimated to be 1.5% of total tetrachloroethylene production (Brown et al., 1975). Some of this waste was emitted to the aquatic environment in industrial effluents and to air from stacks. Since the closing of the two tetrachloroethylene plants in Canada in 1985 and 1992, there have been no major point sources of HCBD. Current Canadian sources are minor but potentially numerous. They include possible releases in landfill leachates, releases during refuse combustion and releases as a by-product in the production of other chlorinated chemicals, such as vinyl chloride, allyl chloride and epichlorohydrin.

Based on 12-month average concentrations, HCBD was detected (detection limit 10 ng/L) in 4 of 26 effluent streams from organic chemical manufacturing plants in Ontario and in 9 of 74 final discharge streams monitored between 1989 and 1991. Estimated loadings at these sites ranged from <1 to 9 g/day; the total loading from this sector was estimated to be 20 g/day (OME, 1992). Until recently, the most significant point source of HCBD in Canada appeared to be the Cole Drain, which discharges into the St. Clair River at Sarnia, Ontario, and includes outfalls from an industrial landfill and a few industrial companies. Loadings from the Cole Drain appear to have decreased from 140 g/day in 1985 (OME, 1991) to 30 g/day in 1995 (Kauss, 1996). In a survey of the Cole Drain final mixing chamber discharge in 1995, a maximum concentration of 0.9 µg HCBD/L was detected (Kauss, 1996). Since 1998, the discharge from the Cole Drain has been practically eliminated as a result of remediation activities. The industrial landfill that was the primary source of HCBD in the Cole Drain was completely remediated and decommissioned, and the bed of the Cole Drain itself was remediated and restored in 1998 (Sarnia_Lambton Environmental Association, 2000; Scott Munroe 2000).

The inadvertent production and use of HCBD in the United States are other potential sources of HCBD to the Canadian environment through atmospheric long-range transport or transboundary movement in shared water systems. Evidence for long-range transport of HCBD was provided by Mudroch et al. (1992), who found that HCBD was present at concentrations ranging from 0.01 to 0.23 ng/g at various sediment depths in samples taken from Great Slave Lake in 1987. According to the United States Toxic Release Inventory, 2 tonnes of HCBD were released to the environment in the United States in 1995; 75% of this total was to the air, 15% to water and 10% to underground injection (Toxic Release Inventory, 1997). The load to the atmosphere, however, does not include all possible releases from every type of industrial facility (ATSDR, 1994).

2.3 Exposure characterization

2.3.1 Environmental fate

2.3.1.1 Air

In air, HCBD persists until it is either degraded photochemically or adsorbed to particulate matter and deposited to water or soil. Estimates of its half-life in air based on photochemical degradation through reactions with hydroxyl radicals and ozone range from 60 days (ATSDR, 1994) to three years (Howard et al., 1991).

Class and Ballschmiter (1987) calculated that HCBD would have a tropospheric half-life of 840 days in the northern hemisphere and 290 days in the southern hemisphere, based on a hydroxyl radical rate constant of 2 x 10^-14 cm³/molecule per second and a hydroxyl radical concentration of 7 x 10⁵ molecules/cm³ in the north and 17  x 10⁵ molecules/cm³ in the south.

These data indicate that HCBD meets the criteria for persistence in air (half-life ≥2 days) in accordance with the Persistence and Bioaccumulation Regulations of CEPA 1999.

2.3.1.2 Water

HCBD was completely degraded by wastewater microbiota within seven days of exposure under aerobic conditions (Tabak et al., 1981). Degradation of HCBD is very slow under anaerobic conditions (Johnson and Young, 1983; Govind et al., 1991; Howard, 1991). The half-life of HCBD in water is proportional to the amount of organic matter in the aqueous media; in natural waters, the half-life is estimated to be 4-52 weeks (Howard et al., 1991).

2.3.1.3 Sediments

Sediments are a sink for HCBD released to water. In sediments with high organic content, the compound is not expected to persist; however, measured values for the half-life in sediment are not available. HCBD will eventually be biodegraded in aerobic sediments.

2.3.1.4 Soils

The half-life of HCBD in soil depends upon the chemical, physical and biological heterogeneity of the soil and climatic conditions. Howard et al. (1991) estimated the half-life to be 4-26 weeks, based on aerobic biodegradation rates; these authors suggested that HCBD may not biodegrade in anaerobic zones of soil and that evaporation would be a significant transport mechanism from soil surfaces. In a dune infiltration study in the Netherlands, HCBD was found to be mobile in sandy soils, with an average residence time of 100 days and little biodegradation (Howard, 1991).

Fragiadakis et al. (1979) examined residues of radio-labelled HCBD in soil-plant systems and observed that 4% of the original radioactivity was bound in non-extractable residues in the top 50 cm of soil after two years, suggestive of potential long-term accumulation. The remaining 96% of the original radioactivity was unaccounted for and was believed to have volatilized.

2.3.1.5 Biota

HCBD partitions preferentially into lipid phases. Although HCBD accumulates in the tissues of freshwater aquatic invertebrates and fish, it does not biomagnify through food chains because of its fast depuration rate (Environment Canada, 1983). HCBD tends to be preferentially accumulated in the livers of fish. The bioconcentration factors (BCFs) in muscle and liver were 700 and 10 000, respectively, in dab, Limanda limanda (Pearson and McConnell, 1975). HCBD was eliminated from the tissues of goldfish (Carassius auratus) with a half-life of 6.3 days (Leeuwangh et al., 1975).

BCFs ranging from 1 to 19 000 on a whole-body basis have been reported for HCBD in the literature. The highest BCF reported was determined in rainbow trout (Oncorhynchus mykiss) in a field study (Oliver and Niimi, 1983). This wide range can be explained in part by species differences in metabolism or differences in exposure concentrations (ATSDR, 1994). It takes longer for equilibrium to be reached in fish at lower exposure concentrations than at higher levels (69 days at 0.1 ng/L versus 7 days at 3.4 ng/L) (Oliver and Niimi, 1983). BCFs were more than two-fold greater at the higher exposure levels than at the lower concentrations, indicating that rates of detoxification and elimination by fish are concentration dependent.

HCBD also bioconcentrates in aquatic invertebrates, but to a somewhat lesser degree than in fish, with a maximum reported BCF of 2000 for the mussel, Mytilus edulis (Pearson and McConnell, 1975). Contamination of water by HCBD led to uptake of the substance by caged mussels in the St. Clair River (Kauss and Hamdy, 1985; OME/MDNR, 1991).

HCBD does not appear to bioaccumulate in plants. In a field study with radio-labelled HCBD, no significant degree of accumulation occurred in roots, leaves or stems of potato or carrot plants (Fragiadakis et al., 1979).

The available data for fish indicate that HCBD meets the criteria for bioaccumulation (BCF ≥5000) in accordance with the Persistence and Bioaccumulation Regulations of CEPA 1999.

2.3.1.6 Environmental partitioning

The distribution of HCBD in the environment was estimated using EQC Level III, a steady-state, non-equilibrium fugacity model (DMER and AEL, 1996). The results of the modelling show that HCBD tends to remain in the environmental compartment into which it is released. If HCBD is emitted into air, more than 98% would be found in the air, about 1% in soil and less than 1% in water and sediments. If released to soil, about 99% would be found in the soil and about 1% in air. If released to water, about 70% would be found in the water, about 15% in each of air and sediments and less than 1% in soil. Values for input parameters were as follows: molecular weight, 260.76 g/mol; vapour pressure, 20 Pa; water solubility, 3.20 mg/L; log K_ow, 4.90; Henry's law constant, 1044 Pa·m³/mol; half-life in air, 1700 hours; half-life in water, 550 hours; half-life in soil, 550 hours; and half-life in sediment, 550 hours. Justification for the selection of these input parameters is presented in DMER and AEL (1996). Modelling was based on an assumed default emission rate of 1000 kg/hour into a region of 100 000 km², which includes a surface water area (20 m deep) of 10 000 km². The height of the atmosphere was assumed to be 1000 m. Sediments and soils were assumed to have an organic carbon content of 4% and 2% and a depth of 1 cm and 10 cm, respectively. The estimated percent distribution predicted by this model is not affected by the assumed emission rate.

The predicted distributions suggest that little intermedia transport will occur when HCBD is discharged to air or soil. By comparison, disposal to water has the potential for significant transport of HCBD to the air and sediment compartments.

2.3.2 Environmental concentrations

The closure of tetrachloroethylene production plants, changes in industrial processes and improvements in waste treatment processes, including improvements in containment facilities and spill prevention, have resulted in greatly reduced loadings of HCBD in the Canadian environment since the early 1980s; HCBD has only rarely been detected in recent monitoring programs in areas removed from former sources.

2.3.2.1 Air

HCBD was detected (detection limit 0.1 µg/m³) in only 153 of 9231 samples (i.e., less than 2%) of outdoor air from 46 sites across Canada surveyed from 1989 to early 1997. It has not been detected at any of these sites since 1994. The maximum concentration measured was about 4 µg/m³ in Windsor in 1992. Mean concentrations at each site, calculated by assuming a concentration of one-half the detection limit of 0.1 µg/m³ in those samples that did not contain detectable levels of HCBD, ranged from 0.05 to 0.07 µg/m³ (Dann, 1997).

No data on levels of HCBD in indoor air in Canada or in "uncontaminated" areas in other countries were identified.

2.3.2.2 Drinking water

HCBD has not been detected in drinking water (detection limits ranging from 0.7 pg/L to 5 µg/L) in most provincial monitoring programs in Canada (Environment Ontario, 1987; Kendall, 1990; Jobb et al., 1993; Alberta Environmental Protection, 1996; Riopel, 1996; Zanette, 1996). It was detected (detection limit 1 ng/L) in only 5 of 2994 samples of treated drinking water from 143 sites across Ontario surveyed in 1991-1995; the maximum concentration measured was 6 ng/L in Port Dover (OMEE, 1996).

2.3.2.3 Surface water

The highest reported concentration of HCBD in Canadian surface waters was 1.3 µg/L, which was measured in the St. Clair River in 1984 (OME/MDNR, 1991); levels have decreased substantially (i.e., 500-fold) since 1984, based on a measurement of 0.0027 µg/L downstream from the Cole Drain in 1994, the highest concentration reported that year (Kauss, 1996). Since 1990, concentrations of HCBD in surface water from southern Ontario have generally been less than 0.001 µg/L (Environment Canada et al., 1995; L'Italien, 1996). A maximum concentration of 24 µg/g dry weight was measured in suspended sediments from the St. Clair River in 1985 (Oliver and Kaiser, 1986); in 1989, the highest level detected was 0.01 µg/g dry weight (Chan, 1993).

2.3.2.4 Sediments

The maximum level of HCBD in sediment in the St. Clair River, near Sarnia, Ontario, where the greatest contamination by HCBD in Canada has reportedly occurred, prior to 1986 was 430 µg/g dry weight (lowest reported concentration of 0.0001 µg/g dry weight); it was detected (detection limit not specified) in 59 of 65 sampling sites in 1985 (Oliver and Pugsley, 1986). The highest concentration measured in recent years was 310 µg/g dry weight, downstream from the Cole Drain at a depth of 5-15 cm in 1994; in this survey, HCBD was detected (detection limit 0.001 µg/g dry weight) in 148 of 153 samples (Farara and Burt, 1997; Kauss, 1997). In the top 5 cm of sediment in a 2-km stretch of the St. Clair River in an industrialized zone in 1994, concentrations of HCBD ranged from <0.001 to 243 µg/g dry weight (detectable in 37 of 39 samples; detection limit 0.001 µg/g dry weight), with a geometric mean of 0.64 µg/g dry weight (Bedard and Petro, 1997). In these samples, the 99th-, 95th- and 90th-percentile values were 194, 60.9 and 18.7 µg/g dry weight, respectively, while the median was 0.9 µg/g dry weight.

2.3.2.5 Soils

In the only identified relevant survey in Canada, HCBD was not detected (detection limit 0.05 µg/g dry weight) in 24 samples of agricultural soils from across the country or in 6 samples from areas that had repeatedly received heavy applications of pesticides (Webber and Wang, 1995).

2.3.2.6 Biota

No recent data on HCBD concentrations in biota have been identified. Levels in rainbow troutcollected from Lake Ontario in 1981 ranged from 0.06 to 0.3 ng/g (mean 0.2 ng/g) (Oliver and Niimi, 1983). Levels of up to 10 ng/g have been detected in composite samples of coho salmon (Oncorhynchus kisutch) collected from the Great Lakes in 1980 (Clark et al., 1984). The maximum concentration of HCBD in caged mussels, Elliptio complanata following three weeks of exposure on the sediment surface near three industrial areas of the St. Clair River was 36 ng/g wet weight (Kauss and Hamdy, 1 985; OME/MDNR, 1991; Kauss, 1997).

2.3.2.7 Food

Data on levels of HCBD in foodstuffs (in addition to those discussed in Section 2.3.2.6) are limited primarily to earlier studies conducted in other countries. Concentrations of HCBD in beverages, bread, butter, cheese, eggs, fruits, meats, milk, oils and potatoes ranging from non-detectable to 3.7 µg/kg (grapes) were reported in the United Kingdom (McConnell et al., 1975), while in Germany, concentrations of HCBD in chicken, eggs, fish, margarine, meat and milk ranged from non-detectable to 42 µg/kg (egg yolk) (Kotzias et al., 1975) (detection limits were not specified in either report). HCBD was not detected in samples of eggs or vegetables and was detected in only 1 of 20 samples of milk produced or grown in the vicinity of organic chemical manufacturing plants in the United States (detection limits of 5 or 40 µg/kg) (Yip, 1976). In a survey of breast milk of women from five regions of Canada, HCBD was not detected in any of 210 samples analysed (detection limit 1.2 µg/L) (Mes et al., 1986).

2.3.2.8 Multimedia exposure study

In a recent pilot multimedia exposure study, samples of personal air, tap water, beverages and food from 44 households in the Toronto area were analysed for HCBD. None of the samples contained detectable amounts of HCBD, although the detection limits in this study were generally higher than those reported in other studies discussed above (i.e., 0.64 µg/m³ for air, 2 µg/L for water and 0.09-0.9 µg/kg for food and beverages), and the analytical recovery of HCBD was not determined (Zhu, 1997).

2.4 Effects characterization

2.4.1 Ecotoxicology

2.4.1.1 Pelagic organisms

HCBD preferentially accumulates in the livers of fish (Pearson and McConnell, 1975). Once in the liver, it can be biotransformed into polar metabolites that will reach the kidneys via the bile and could become nephrotoxic in fish (Anders and Jakobson, 1985; Yang, 1988; IPCS, 1994).

The available data on toxicity for sensitive receptors indicate that chronic effects occur at concentrations an order of magnitude below those causing acute effects. In most cases, freshwater fish and marine crustacea are more sensitive than their marine and freshwater counterparts, respectively.

The lowest available chronic value was a 28-day Lowest-Observed-Effect Concentration (LOEC) of 13 µg/L reported for the fathead minnow (Pimephales promelas ), based on survival and growth (Benoit et al., 1982). No chronic data on toxicity were identified for aquatic invertebrates. The lowest identified acute value was a 96-hour LC₅₀ of 32 µg/L for the marine mysid shrimp, Mysidopsis bahia (U.S. EPA, 1980). For fish, the lowest identified acute value was a 96-hour LC₅₀ of 90 µg/L for the goldfish (Leeuwangh et al., 1975). In other studies, acute toxicity was reported only at concentrations of HCBD above 100 µg/L (Pearson and McConnell, 1975; Laseter et al., 1976; Dow Chemical Co., 1978; Juhnke and Lüdemann, 1978; Laska et al., 1978; Slooff, 1979; U.S. EPA, 1980; Walbridge et al., 1983; Geiger et al., 1985; Mayer and Ellersieck, 1986; Roederer et al., 1989). The most sensitive freshwater invertebrate identified was the aquatic sowbug, Asellus aquaticus , with a 96-hour LC₅₀ of 130 µg/L (Leeuwangh et al., 1975). Bacteria and plants are less sensitive to HCBD than fish or invertebrates (Knie et al., 1983).

2.4.1.2 Benthic organisms

There were no acute or chronic toxicity studies using benthic organisms identified for HCBD. In the absence of such data, the water-sediment Equilibrium Partitioning approach can be used to estimate a Critical Toxicity Value (CTV) for HCBD for benthic organisms. The principle behind this approach is that sediment organic carbon is the main factor influencing partitioning of non-polar organic compounds into sediments (Di Toro et al., 1991). For HCBD, the CTV for the most sensitive freshwater pelagic invertebrate multiplied by the organic carbon/water partition coefficient (K_oc) and the organic content of the sediment (f_oc) can be used to estimate a CTV for benthic organisms using the equation:

CTV_benthic = f_oc x K_oc x CTV_pelagic

where:

• f_oc is 0.02, based on the mean organic carbon
  content for all surficial sediment samples from the St. Clair River in 1994, expressed on a dry-weight basis (Kauss, 1997),
• K_oc is 80 000 L/kg, based on the log K_oc of
  4.90 (Oliver and Kaiser, 1986) for HCBD, and
• CTV_pelagic is 13 µg/L, the 28-day LOEC for
  fathead minnow (Benoit et al., 1982).

Therefore:

The CTV for HCBD for benthic organisms is therefore estimated to be 20.8 µg/g dry weight.

2.4.2 Abiotic atmospheric effects

Class and Ballschmiter (1987) calculated that HCBD would have a tropospheric half-life of 840 days in the northern hemisphere and 290 days in the southern hemisphere. These half-lives are sufficiently long to allow HCBD to reach the stratosphere and react with the ozone present there (Bunce, 1996).

Worst-case calculations were made to determine if HCBD has the potential to contribute to depletion of stratospheric ozone, ground-level ozone formation or climate change (Bunce, 1996).

The Ozone Depletion Potential (ODP) was calculated to be 0.07 (relative to the reference compound CFC-11, which has an ODP of 1), based on the following formula:

where:

• t_HCBD is the atmospheric lifetime of
  HCBD (4.2 years),
• t_CFC-11 is the atmospheric lifetime
  of CFC 11 (60 years),
• M_CFC-11 is the molecular weight of CFC-11
  (137.5 g/mol),
• M_HCBD is the molecular weight of HCBD
  (260.8 g/mol),
• n_Cl is the number of chlorine atoms in the
  HCBD molecule (6),
• n_Br is the number of bromine atoms in the
  HCBD molecule (0), and
• α is a measure of the effectiveness of bromine in ozone depletion with respect to chlorine.

The Photochemical Ozone Creation Potential (POCP) was estimated to be 0.01 (relative to the value of an equal mass of the reference compound ethene, which has a POCP of 100), based on the following formula:

POCP = (k_HCBD /k_ethene) x (M_ethene /M_HCBD) x 100

where:

• k_HCBD is the rate constant for the reaction of
  HCBD with OH radicals (9.5 x 10-15 cm³/mol per second),
• k_ethene is the rate constant for the reaction of
  ethene with OH radicals (8.5 x 10-12 cm³/mol per second),
• M_ethene is the molecular weight of ethene
  (28.1 g/mol), and
• M_HCBD is the molecular weight of HCBD
  (260.8 g/mol).

The Global Warming Potential (GWP) was calculated to be 0.037 (relative to the reference compound CFC-11, which has a GWP of 1), based on the following formula:

GWP = (t_HCBD /t_CFC-11) x (M_CFC-11 /M_HCBD) x (S_HCBD /S_CFC-11)

where:

• t_HCBD is the atmospheric lifetime of HCBD
  (4.2 years),
• t_CFC-11 is the atmospheric lifetime of CFC-11
  (60 years),
• M_CFC-11 is the molecular weight of CFC-11
  (137.5 g/mol),
• M_HCBD is the molecular weight of HCBD
  (260.8 g/mol),
• S_HCBD is the infrared absorption strength of
  HCBD (2389/cm² per atmosphere, default), and
• S_CFC-11 is the infrared absorption strength of
  CFC-11 (2389/cm² per atmosphere).

These figures imply that HCBD is not likely to contribute significantly to ground-level ozone formation, but it does have the potential to contribute somewhat to depletion of stratospheric ozone and to climate change.

2.4.3 Experimental animals and in vitro

2.4.3.1 Acute toxicity

HCBD is moderately acutely toxic, with LD₅₀s of 65-116 mg/kg-bw in mice, 200-580 mg/kg-bw in rats and 90 mg/kg-bw in guinea pigs (Murzakaev, 1963; Gulko et al., 1964; Gradiski et al., 1975; K_ociba et al., 1977a, 1977b). Birner et al. (1995) observed necrosis of the pars recta of the proximal renal tubules in Wistar rats administered a single dose of 200 mg/kg-bw; renal tubular necrosis was als o in duce d in laboratory animals exposed to single doses of several metabolites of HCBD (Lock and Ishmael, 1979; Jaffe et al., 1983; Lock et al., 1984; Nash et al., 1984).

2.4.3.2 Short-term and subchronic toxicity

Although the database is limited, in available short-term and subchronic studies in rats and mice, the renal proximal tubules appear to be the principal site of injury at the lowest doses that cause effects following oral or inhalation exposure. Although decreases in body weight gain were sometimes also observed at the lowest exposure levels at which effects were observed, these decreases were generally associated with reduced food consumption.

Increased relative kidney weight and histopathological changes, including degeneration of the proximal tubular epithelial cells, necrosis and regeneration, and alterations in biochemical parameters in the blood and urine (consistent with renal damage) were reported in short-term studies in Wistar or Sprague-Dawley rats exposed to HCBD in the diet or by gavage for 2-4 weeks at doses as low as 2.5 mg/kg-bw per day (K_ociba et al., 1971; Harleman and Seinen, 1979; Stott et al., 1981; Jonker et al., 1993). Jonker et al. (1993) observed female rats to be more sensitive to the nephrotoxic effects than male rats, as histopathological changes in the kidney occurred in females at 100 and 400 ppm in the diet (approximately equivalent to doses of 5 and 20 mg/kg-bw per day, respectively) and in males only at 400 ppm, although effects on kidney weight and biochemical parameters were noted in both sexes at 100 ppm and above. In the only identified short-term study in mice, there was a dose-related increase in severity of renal toxicity, characterized by pale kidney cortices and necrosis of the cortex and/or outer medulla, in male and female B6C3F₁ mice administered concentrations of HCBD equivalent to doses as low as 3 mg/kg-bw per day in the diet for two weeks (Yang et al., 1989; NTP, 1991).

In a subchronic study in which groups of 10 male or female Wistar-derived rats were administered doses of 0, 0.4, 1.0, 2.5, 6.3 or 15.6 mg HCBD/kg-bw per day in arachid oil by gavage for 13 weeks, there was a dose-related increase in relative kidney weight, which was significant in females at the two highest doses and at all doses in males. Histopathological changes in the kidney, consisting of large, prominent hyperchromatic nuclei and f_ocal necrosis of epithelial cells and nuclear detritus, were observed in the renal proximal tubules in females at 2.5 mg/kg-bw per day and above and in males at 6.3 mg/kg-bw per day and above. There were also dose-related decreases in urine osmolarity (indicative of compromised urine-concentrating ability of the kidneys), which were significant in females at 2.5 mg/kg-bw per day and above and in males at the highest dose only (Harleman and Seinen, 1979). The Lowest-Observed-Adverse-Effect Levels (LOAELs), based on renal effects, are considered to be 2.5 and 6.3 mg/kg-bw per day in females and males, respectively, with No-Observed-Adverse-Effect Levels (NOAELs) of 1.0 and 2.5 mg/kg-bw per day in females and males, respectively (the authors presented these latter values as "no effect levels").

Effects on the kidney were also observed in groups of 10-34 male or female Sprague-Dawley rats administered doses of 0, 0.2, 2.0 or 20 mg HCBD/kg-bw per day in the diet for approximately 148 days. The kidneys of only five animals per group were examined histopathologically. The relative weight of the kidney was significantly increased in both sexes at 20 mg/kg-bw per day, whereas the kidneys in males administered the two highest doses were "roughened" and mottled in appearance. There was minimal or moderate renal tubular dilation and hypertrophy with f_oci of renal tubular epithelial degeneration and regeneration in four of five male or female rats in the high-dose group; these lesions also occurred in one female at 2.0 mg/kg-bw per day. Renal changes that are characteristic of this strain of rats occurred in all dose groups, but with greater severity at 2.0 and 20 mg/kg-bw per day (Schwetz et al., 1977). The No-Observed-Effect Level (NOEL) and Lowest-Observed-Effect Level (LOEL) for effects on the kidney are considered to be 0.2 and 2.0 mg/kg-bw per day, respectively. (Note: The latter value was not considered a LOAEL because of the lack of statistical significance of the observed effects.)

In the only subchronic study in mice, diets containing 0, 1, 3, 10, 30 or 100 ppm HCBD (which the authors calculated to be equivalent to doses of 0, 0.1, 0.4, 1.5, 4.9 and 16.8 mg/kg-bw per day for males and 0, 0.2, 0.5, 1.8, 4.5 and 19.2 mg/kg-bw per day for females) were administered to groups of 10 B6C3F₁ mice of each sex for 13 weeks. Dose-related reductions in relative and/or absolute kidney weights were reported; these reductions were significant in males in the three highest dose groups and in females in the two highest dose groups. The incidence and severity of renal tubular epithelial regeneration, characterized by increased basophilia of the tubular cell cytoplasm, occasional mitosis and an increased number of nuclei, increased in an exposure-related manner (0/10, 1/10, 9/10, 10/10, 10/10 and 10/10 [females] and 0/10, 0/10, 0/10, 0/9, 10/10 and 10/10 [males] at 0, 1, 3, 10, 30 and 100 ppm, respectively). Females appeared to be more sensitive than males, as the incidence of this lesion was significantly increased at 3 ppm and above in females and at 30 ppm and above in males; renal tubular regeneration was also observed in 1 of 10 female mice exposed to 1 ppm. (The lesion in this mouse was assigned a severity score of 2; Elwell, 1993.) Unlike the observation of renal necrosis in the short-term study, only regenerative changes were observed in this study, which the authors suggested was indicative of adaptation and compensation by the kidney tubular epithelium for cell loss. Based on the histopathological effects in the kidney, the authors considered the NOAEL in male mice to be 1.5 mg/kg-bw per day; a no-effect level for female mice was not presented by the authors, as renal tubular regeneration was observed in all dose groups (Yang et al., 1989; NTP, 1991). Therefore, because of the lack of statistical significance of the response in the female mice in the lowest dose group (for which data on the incidence of this lesion in historical controls at the National Toxicology Program were not available for comparison) and the severity of the renal tubular regeneration in the one mouse in this dose group, as well as the lack of data on food consumption for individual animals (i.e., it is unclear whether this effect may have been a function of increased food consumption), 0.2 mg/kg-bw per day is considered to be the LOEL for renal toxicity in females in this study.

In the only short-term or subchronic study identified in which animals were exposed to HCBD by inhalation, renal proximal tubular degenerat ion an d adrenal cortical degeneration were noted in groups of four male or female Alderley Park SPF rats exposed to 25 ppm (267 mg/m³) HCBD and above for up to 15 days. Renal toxicity was not observed at lower concentrations (5 ppm [53 mg/m³] or 10 ppm [107 mg/m³]) (Gage, 1970).

2.4.3.3 Chronic toxicity and carcinogenicity

The identified information on the chronic toxicity and carcinogenicity of HCBD is extremely limited. In the only long-term study identified, groups of 39 or 40 (90 in controls) male and female Sprague-Dawley rats were administered doses of 0, 0.2, 2.0 or 20 mg HCBD/kg-bw per day in the diet for two years. Mortality was significantly increased in males in the 20 mg/kg-bw per day group during the last two months of the study. Body weight gain was significantly decreased and absolute and relative kidney weights were significantly increased in both sexes at this dose. There were significant increases in urinary coproporphyrin in males and females at 20 mg/kg-bw per day and in females at 2.0 mg/kg-bw per day; however, other urinary biochemical parameters were not altered. Histopathological changes, including multif_ocal or disseminated hyperplasia and f_ocal adenomatous proliferation of the renal tubular epithelium, were observed in rats exposed to the highest dose and "possibly" at 2.0 mg/kg-bw per day, with females being more sensitive than males (incidence and statistical significance not specified). The incidence of renal tumours (adenomas, adenocarcinomas and carcinomas, combined) was significantly increased in rats of both sexes administered 20 mg/kg-bw per day (males: 1/90 [1.1%], 0/40 [0%], 0/40 [0%] and 9/39 [23.1%] at 0, 0.2, 2.0 and 20 mg/kg-bw per day, respectively; females: 0/90 [0%], 0/40 [0%], 0/40 [0%] and 6/40 [15.0%] at 0, 0.2, 2.0 an d 20 mg/kg-bw per day, respectively). There were no significant increases in the incidence of tumours at other sites. The authors concluded that HCBD induced renal tumours only at a dose level greater than that which caused observable non-neoplastic injury (K_ociba et al., 1977a). The NOEL for non-neoplastic kidney damage was considered to be 0.2 mg/kg-bw per day, with a LO(A)EL of 2.0 mg/kg-bw per day. (It is not possible to determine whether the effects at this dose were adverse on the basis of information presented in the published account of the study.)

Additional limited screening bioassays contribute little to the assessment of the potential carcinogenicity of HCBD. HCBD did not induce local or distant tumours following chronic dermal application or short-term intraperitoneal administration in sensitive strains of mice (Theiss et al., 1977; Van Duuren et al., 1979), although the extent of histopathological examination was limited in these studies; nor did HCBD initiate the induction of skin papillomas in mice in a long-term initiation-promotion assay (Van Duuren et al., 1979).

2.4.3.4 Genotoxicity

Although the results of available studies are not completely consistent, there is some limited evidence that HCBD is genotoxic under certain conditions. The results of early standard Ames tests were negative in both the presence and absence of liver S-9 metabolic activation (De Meester et al., 1980; Stott et al., 1981; Haworth et al., 1983; Reichert et al., 1983). However, HCBD induced gene mutations in Salmonella typhimurium in the presence of liver S-9 mix with enhanced protein content (Reichert et al., 1984) and in the presence of liver microsomes and glutathione (GSH), with a greater response with both liver and kidney microsomes and GSH (Vamvakas et al., 1988). Positive results were also obtained for the Ara test in Salmonella, only in the absence of liver S-9 metabolic activation (Roldán-Arjona et al., 1991). HCBD induced sister chromatid exchanges in Chinese hamster ovary cells (with and without S-9), but not chromosomal aberrations (Galloway et al., 1987); chromosomal aberrations were also not induced in peripheral human lymphocytes, although the exposure levels tested were much lower (German, 1988).

One author reported the induction of chromosomal aberrations in the bone marrow of mice exposed to HCBD orally (≥2 mg/kg-bw) or by inhalation (10 mg/m³) (German, 1988), whereas negative results have been reported in other studies in rats exposed to greater concentrations or doses (Schwetz et al., 1977; NIOSH, 1981). Increased DNA synthesis and minor amounts of DNA alkylation were observed in the kidney of rats administered single or repeated oral doses of 20 mg HCBD/kg-bw (Stott et al., 1981). In addition, there was significant covalent binding to mitochondrial DNA in the kidney of mice orally exposed to 30 mg HCBD/kg-bw (Schrenk and Dekant, 1989).

Several of the metabolites of HCBD have been mutagenic in Salmonella. The cysteine conjugate, which appears to be the most potent of the metabolites tested, is likely cleaved by bacterial β-lyase to mutagenic intermediates (Dekant et al., 1986). The mutagenic activity of the S-conjugate is enhanced by the presence of rat renal microsomes and mitochondria, which exhibit high γ-glutamyl transpeptidase activity (Vamvakas et al., 1988). Similarly, the mercapturic acid metabolite was mutagenic only in the presence of metabolic activation, which would provide N-deacetylase (Wild et al., 1986), whereas the bis-conjugates were not active under any conditions (Vamvakas et al., 1988).

2.4.3.5 Reproductive and developmental toxicity

Subchronic or chronic oral administration of up to 20 mg HCBD/kg-bw per day did not induce histopathological changes in the testes or ovaries or effects on estrous cycle or sperm parameters in B6C3F₁ mice or Sprague-Dawley rats (K_ociba et al., 1977a; NTP, 1991). In developmental studies, effects on body weight and histopathological changes in the kidney were observed in fetuses of rats (Sprague-Dawley, Wistar and CD strains) exposed to oral doses or airborne concentrations of HCBD that also induced decreased body weight gain and/or renal effects in the dams (Schwetz et al., 1977; Harleman and Seinen, 1979; Hardin et al., 1981; Saillenfait et al., 1989; NTP, 1990).

2.4.3.6 Neurological effects and effects on the immune system

Although data are limited, results of available short-term, subchronic and chronic studies in rodents do not indicate that neurological effects or effects on the immune system are critical endpoints associated with exposure to HCBD; that is, such effects were not observed at doses lower than those that induced effects on the kidney (Kociba et al., 1977a; Harleman and Seinen, 1979; Yang et al., 1989; NTP, 1991). However, no studies on the effects of HCBD on the function of the immune system were identified.

2.4.3.7 Toxicokinetics and mechanism of action

The site-specific renal toxicity of HCBD is closely correlated with the accumulation of active metabolites in the pars recta of the proximal tubule. HCBD is initially conjugated with GSH in the liver to form sulphur conjugates, which are hydrolysed in the bile duct, intestine and kidney. These S-cysteine conjugates and their mercapturic acid derivatives (formed by N-acetylation)
are concentrated in the kidney, where the pentachloro-sulphur conjugate is subsequently cleaved by renal b-lyase (which is localized in the pars recta) to reactive thiol metabolites, which may covalently bind to cellular acromolecules
(causing cytotoxicity) and/or bind to DNA to induce mutation. (Note: Although metabolism of HCBD may be qualitatively similar in experimental animals and in humans, some very limited data indicate that the activity of b-lyase in the kidney of humans may be several-fold less than that in the kidney of rats [McCarthy et al., 1992; Lock, 1994].) In addition, sulphoxidation of one of the mercapturic acid derivatives to electrophilic metabolites has been recently demonstrated in rats exposed to HCBD in vivo and in human liver microsomes (Birner et al., 1995).

Although it is known that these electrophilic metabolites induce damage in renal tubular epithelial cells and mutations in Salmonella and bind to DNA, it has not been firmly established whether the initial step in kidney tumour formation is a result of genetic damage or epigenetic events (possibly in the mitochondria) (Stott et al., 1981; Schrenk and Dekant, 1989; Dekant et al., 1990; Henschler and Dekant, 1990). Unlike the mechanism of action associated with other halogenated hydrocarbons, accumulation of a2µ-globulin and hyaline droplet formation are not involved in the formation of renal tumours induced by HCBD.

2.4.4 Humans

The limited identified studies in humans, which include a cross-sectional study on liver function and a survey of the frequency of chromosomal aberrations in exposed workers (German, 1986; Driscoll et al., 1992), are inadequate to contribute meaningfully to evaluation of the toxicity of HCBD.