Page 10: Guidelines for Canadian Drinking Water Quality: Guideline Technical Document – Toluene, Ethylbenzene and the Xylenes

9.0 Health effects

9.1 Effects in humans

9.1.1 Acute toxicity

Accidental ingestion of toluene was shown to cause severe acute toxicity, including oropharyngeal and gastric irritation with vomiting and hematemesis (Von Burg, 1993). Abdominal pain, hemorrhagic gastritis and central nervous system depression were observed following ingestion of approximately 1 L of paint thinner known to contain toluene (Caravati and Bjerk, 1997). Death was reported to occur within 30 minutes of ingestion of approximately 60 mL (625 mg/kg bw) of toluene in one individual (Ameno et al., 1989). Ingestion of xylene was shown to cause severe gastrointestinal distress (Sandmeyer, 1981). No information was available on human ingestion of ethylbenzene.

Exposure via inhalation to a mixture of VOCs that included toluene, ethylbenzene and xylenes for 2.75 hours at concentrations as low as 5 mg/m3 was associated with irritation of the eyes, nose and throat (Molhave et al., 1986). Additional acute effects of toluene, ethylbenzene and xylene exposure generally target the central nervous system. Accidental exposures to high concentrations of toluene (≥ 10 000 ppm or 37 500 mg/m3) have resulted in central nervous system excitation followed by progressive impairment of consciousness, seizures and coma (IPCS, 1986). Controlled human studies have identified decreased neurological functions at toluene concentrations between 100 and 150 ppm (~ 377 and 566 mg/m3) (Andersen et al., 1983; Baelum et al., 1985; Echeverria et al., 1989). A controlled human co-exposure to toluene and xylene suggested that exposure as low as 200 ppm (~ 754 mg/m 3) toluene (for 7 hours with one break) causes prolongation of reaction time, decrease in pulse and decrease in systolic blood pressure (Ogata et al., 1970). It has been demonstrated in studies of controlled human exposure to m-xylene over several days that concentrations as low as 90 ppm (~ 339 mg/m3) had deleterious effects on reaction time, manual coordination, body balance and equilibrium (Savolainen et al., 1979, 1980). However, Olson et al. (1985) noted that 4-hour exposure of males to toluene at 3.25 mmol/m3, to p-xylene at 2.84 mmol/m3 or to a mixture of both chemicals (toluene at 2.20 mmol/m3 and xylene at 0.94 mmol/m3) had no impact on reaction time, short-term memory or choice reaction either immediately upon exposure or 2 and 4 hours after exposure.

Thus, studies indicate that acute exposure to toluene and xylenes is associated with effects on the central nervous system. Exposure to toluene, ethylbenzene and xylene vapours may also irritate mucous membranes, and exposure via ingestion may cause moderate to severe gastrointestinal effects.

9.1.2 Subchronic and chronic toxicity and carcinogenicity

Studies of exposure in humans are primarily limited to the inhalation route and involve co-exposure to other solvents in the occupational setting. Limited evidence has associated chronic exposure to toluene and xylenes with neurological effects, malignancies and other effects. Very little information was available regarding the chronic toxicity of ethylbenzene due to the lack of working environments with predominant ethylbenzene exposure.

9.1.2.1 Neurological effects

The neurological implications of exposure to toluene and xylenes have been examined in individuals exposed occupationally. However, no studies of ethylbenzene exposure were available, and none of the studies pertained to oral exposure.

Several studies have investigated the effects of toluene in individuals working within the printing and rubber industry, where toluene exposure is prevalent. An effect that has been carefully documented is a reduction in colour vision. Colour vision loss was noted in workers in a printing press at a mean toluene concentration of 120 ppm (452 mg/m3) and in rubber workers that had urinary toluene levels ranging from 60 to 73 µg/L (Zavalic et al., 1998; Cavalleri et al., 2000). In both cases, workers were exposed primarily to toluene and were not exposed to other neurotoxicants. One study at lower toluene levels of approximately 44–48 ppm or 166-181 mg/m3 (and where toluene accounted for at least 90% of the exposure, however, did not show a significant alteration in colour vision (Nakatsuka et al., 1992). Toluene exposure was also associated with increased amplitudes in visually evoked potential of printing press workers exposed to approximately 40–60 ppm (151-226 mg/m3)toluene and employed for an average of 20.6 years (Vrca et al., 1995); a follow-up study, however, did not support these findings (Vrca et al., 1997). Additional studies of toluene exposure pertain to more generalized neurological effects, including decreased concentration and reasoning and self-reporting of subjective symptoms. Several studies have shown that workers exposed to toluene had decreased concentration, memory and reasoning, even at concentrations below 100 ppm or 377 mg/m3 (Foo et al., 1990; Boey et al., 1997; Eller et al., 1999; Neubert et al., 2001; Kang et al., 2005; Nordling Nilson et al., 2010). Other studies of cognitive effects at lower levels report no effect of exposure to toluene at concentrations of up to 50 ppm or 189 mg/m3 (Zupanic et al., 2002; Seeber et al., 2004, 2005). One study demonstrated a clear concentration–response relationship with regards to memory disturbances (Chouaniere et al., 2002). However, the effect was not significant when cumulative exposure was investigated, thus suggesting that the effect may not persist. Furthermore, occupational exposure to toluene caused increases in reporting of subjective symptoms (Ukai et al., 1993; Tanaka et al., 2003), and one study suggested that chronic exposure to toluene may be related to damage to the central autonomic nervous system, as demonstrated by altered nerve conduction in exposed individuals (Murata et al., 1993; Tanaka et al., 2003).

In a repeated measures study by Seeber et al. (2004), the authors investigated the effects of toluene on cognitive function, as measured by attention (symbol–digit substitution, switching attention, simple reaction), memory (digit span, Syndrom-Kurztest) and psychomotor (steadiness, line tracing, aiming, tapping, pegboard) tests in a subsample of 192 subjects who participated in four examinations over 5 years. Exposures and exposure durations were as follows: high current exposure = 26 ppm (98 mg/m3) for 106 subjects; low current exposure = 3 ppm (11 mg/m3) for 86 subjects; long duration = 21 years; and short duration = 6 years. Current exposures were determined by four measures, whereas lifetime-weighted averages were determined using a job exposure matrix. Analyses by repeated measures analysis of variance and stepwise regression adjusting for age, level of education, alcohol consumption and anxiety revealed a significant difference in error time of the steadiness test between exposures and error time in line tracing when exposure level and duration were combined. However, in further analyses, the between-subject effect for exposure level in the steadiness test was not significant, and the results of the line tracing test were not consistent with an exposure-related response. Analyses also showed no differences between exposure groups when they were stratified by duration (short or long). In a later study by Seeber et al. (2005), these results were analyzed using a case-control method in which cases with impaired function were defined as those that were 20% above or below the mean (depending on the measure of effect) of the reference group. The only significant finding (errors with steadiness with the dominant hand) was only marginally significant when errors and error time in a repeated measures approach were included (Seeber et al., 2005).

Non-specific neurological effects of xylene exposure have been observed in an occupational setting and a controlled human study. When exposure to xylenes accounted for more than 70% of total occupational exposure to chemicals (sum of all isomers, geometric mean concentration 14 ppm or 61 mg/m3) in rubber, plastic and printing workers, a significant increase in the prevalence of subjective neurological symptoms was noted. Significant symptoms included dizziness, heavy feeling in the head and headache. However, no dose–response trends were observed, and the potential effects of other chemicals were unclear (Uchida et al., 1993). Chronic symptoms of dizziness, easy fatigability, depressed mood and palpitation were also observed when humans were exposed to approximately to a mixture comprising 50% xylenes along with 24% toluene (corresponding to 18 ppm or 78 mg/m3 xylenes and 10 ppm or 38 mg/m 3 toluene) and traces of other solvents (Wang and Chen, 1993). Savolainen et al. (1985) reported that increases in blood levels of m-xylene correlated with decreased balance following controlled human exposure to a fixed m-xylene concentration of 200 ppm (868 mg/m3) or to fluctuating m-xylene concentrations in the range 135–400 ppm (586–1,736 mg/m3) over 6 weeks (4 hours/day, 6 days/week). The persistence of this effect, however, is not known.

Overall, data indicate that toluene and xylenes may be neurotoxic, although more information was available for toluene due to its widespread use in the printing industry. Toluene exposure may be associated with adverse neurological effects, including colour vision loss and decreased cognitive functions. It is unclear whether ethylbenzene is neurotoxic in humans.

9.1.2.2 Renal, hepatic and other tissue effects

No study has investigated the long-term effects of toluene, ethylbenzene or xylene exposure on renal, hepatic and other tissues via the oral route in humans. However, some studies have focused on occupational exposure to the three chemicals, especially toluene. In the case of toluene, limited data from chronic solvent abusers were also available.

Several studies have investigated impacts on liver toxicity in humans as measured by markers of hepatic toxicity in blood. Most studies that have examined toluene-exposed workers in printing, painting and shoe manufacturing plants have reported negative findings at toluene concentrations of up to 324 ppm or 1 221 mg/m3 (Seiji et al., 1987; Ukai et al., 1993). One study of eight print workers occupationally exposed to low levels of toluene (≤ 200 ppm or 754 mg/m3) reported mild elevation of serum transaminases including aspartate aminotransferase with concomitant pericentral fatty changes in the liver, as measured by liver biopsy (Guzelian et al., 1988). Studies of chronic solvent abusers exposed primarily to toluene have reported kidney injury, including tubular degeneration and necrosis (Kamijima et al., 1994; Kamijo et al., 1998). However, only very mild effects on kidney function, such as alterations in creatinine clearance, were observed in printers (Stengel et al., 1998). This suggests that renal toxicity is likely to increase in a dose-dependent trend in human populations. Studies have also reported acidosis following inhalation exposure to toluene, although this effect most likely does not lead to permanent kidney damage (ATSDR, 2000). Human studies pertaining to tissues other than liver and kidney were not found.

Occupational exposure to ethylbenzene, for the most part, occurs in combination with exposure to other organic solvents. Thus, it is difficult to determine any ethylbenzene-specific effects in human studies. One study evaluated the health impacts of exposure of workers within an ethylbenzene manufacturing facility over 20 years (Bardodej and Cirek, 1988). Although specific air concentrations were not reported, a mean air concentration of 6.4 mg/m3 was estimated from the mean urinary mandelic acid concentrations in the workers (Bardodej and Bardodejova, 1970; ATSDR, 2010). The study did not demonstrate any damage to liver tissue as a result of chronic exposure to ethylbenzene. However, the concentration may have been too low to elicit any health effects.

Only one study has investigated chronic exposure (average of 7 years) to approximately 14 ppm (61 mg/m3) mixed xylenes, representing over 70% of the total exposure to solvents (Uchida et al., 1993). This study did not reveal any changes in serum chemistry that would suggest abnormal liver or kidney effects. Thus, low-level inhalation of xylenes is unlikely to cause any adverse hepatic or nephritic effects.

Thus, toluene appears to cause adverse effects in the kidney and liver upon inhalation exposures in humans. However, there is insufficient information on ethylbenzene and xylenes to draw conclusions with regards to the adverse effects in various tissues.

9.1.2.3 Cancer

Studies investigating the carcinogenic potential of toluene, ethylbenzene and xylenes are limited to occupational exposures. No studies have investigated carcinogenic outcome following oral exposure in humans.

The carcinogenic potential of toluene in humans has been studied in three cohort studies of workers exposed primarily to toluene. Svensson et al. (1990) showed that a minimum of 3 months of exposure of rotogravure printers employed between 1925 and 1985 increased mortality due to gastrointestinal and stomach cancers (standardized mortality ratio [SMR] 2.1, 95% confidence interval [CI] 1.1–3.5; and SMR 2.7, 95% CI 1.1–5.6; for gastrointestinal and stomach cancers, respectively) and increased morbidity due to respiratory tract cancers (standardized incidence ratio [SIR] 1.8, 95% CI 1.0–2.9). However, gastrointestinal cancer was significant only when a 5-year minimum exposure and a 10-year latency period were considered, and exposure to benzene was also noted for those exposed prior to the 1960s. A study of shoe manufacturer workers employed for at least 1 month between 1940 and 1979–1982 showed an increase in lung cancer that was more significant in men (Walker et al., 1993; Lehman and Hein, 2006). However, workers were also exposed to methyl ethyl ketone, acetone and hexane and may have been in contact with benzene. Moreover, smoking was an important confounding factor in the study. In a study by Anttila et al. (1998), Finnish workers were monitored for biological markers of toluene, xylenes and styrene. No significant findings were observed, and possible exposure to benzene was noted. One case-control study determined that professions with the highest exposure to toluene had an increased risk of rectal cancer, but no other forms of cancer (Gérin et al., 1998). Other epidemiological studies have involved individuals exposed to solvent mixtures that included toluene. These studies did not report significant increases in cancer (Wen et al., 1985; Carpenter et al., 1988; Blair et al., 1998; Lundberg and Milatou-Smith, 1998; Costantini et al., 2008), other than an increase in the incidences of Hodgkin's/non-Hodgkin's lymphoma (Olsson and Brandt, 1980; Miligi et al., 2006) and colon cancer (Dumas et al., 2000; Goldberg et al., 2001). However, individuals in these studies were exposed to various other solvents, and thus it is difficult to determine the impact of toluene alone among these findings. Overall, there is little evidence that toluene is associated with cancer, and the significant results that were found may be confounded by exposure to other chemicals, including benzene.

Only one study examined predominant ethylbenzene exposure in workers within an ethylbenzene manufacturing facility. Although the exposure was very low (estimated to be approximately 6.4 mg/m3, based on the mean urinary mandelic acid concentrations in the workers), no excess cases of malignancy were recorded in this facility (Bardodej and Cirek, 1988). Thus, there is insufficient information to determine the carcinogenicity of ethylbenzene in humans.

There is no occupational study in which xylene exposure was predominant, and there is very little evidence of xylene-induced carcinogenicity in humans. One study that examined lymphocytic leukemia among solvent-exposed rubber workers indicated a possible association with exposure to xylenes (Arp et al., 1983). Miligi et al. (2006) reported a potential increased risk of non-Hodgkin's lymphoma for medium to high exposures (according to occupation and exposure control metrics) to xylenes (odds ratio [OR] 1.7, 95% CI 1.0–2.6). However, no increased risk was found by Costantini et al. (2008) for chronic lymphatic leukemia as a result of medium/high xylene exposure. Gérin et al. (1998) determined that professions with the highest exposure to xylenes had an increased risk of rectal cancer, but no excess risk was found for any other cancer. A population-based case-control study that examined rectal cancer determined a slightly elevated risk for those exposed to xylenes (Dumas et al., 2000). However, due to exposure to various other solvents, it is difficult to draw any conclusions from these studies.

Due to the limited studies of toluene, ethylbenzene and xylene exposure in humans and the various confounding factors presented in each study, there is insufficient evidence to establish the carcinogenic effects of the three compounds in humans.

9.1.2.4 Other effects

Studies have investigated the effects of toluene, ethylbenzene and xylenes on hearing and cardiovascular/pulmonary health. None of these studies evaluated oral exposure to the chemicals.

Exposure to solvents in general has been associated with hearing loss beyond what can be expected in work environments with elevated noise. Of the three chemicals of interest, data are limited to toluene. Hearing loss was observed among printing workers exposed primarily to toluene (Morata et al., 1997). Workers exposed on average to 97 ppm (366 mg/m3) toluene had alterations in auditory evoked potentials indicative of hearing impairments (Abbate et al., 1993). Alterations in auditory evoked potentials were further observed in printing press workers exposed to low levels of toluene for an average of 20.3 years (Vrca et al., 1996, 1997). However, investigations in rotogravure printers classified in various groups according to the length and level of exposure found that auditory thresholds were affected by noise, but not by toluene (Schäper et al., 2003). Although it is unclear whether the effects of toluene exposure on hearing in humans are a result of direct ototoxicity or adverse impacts on the brain, data in humans suggest that the effect may be neurological.

Additional studies pertain to adverse cardiovascular, immune and pulmonary effects. Xu et al. (2009) reported significant associations between blood levels of toluene, ethylbenzene, o-xylene and m/p-xylene (0.248, 0.05, 0.058 and 0.210 ng/mL, respectively) and cardiovascular disease prevalence; associations were especially significant for toluene. Yoon et al. (2010) reported that urinary levels of hippuric acid and methylhippuric acid (metabolites of toluene and xylene, respectively) in elderly people are associated with a reduction in forced expiratory volume, suggesting decreased lung function as well as increased urinary markers of oxidative stress. Billionnet et al. (2011) showed that exposure of individuals to ethylbenzene, m/p-xylene and o-xylene is associated with rhinitis. Studies have not reported significant adverse hematological effects of toluene exposure (Matsushita et al., 1975; Ukai et al., 1993); no data were available for ethylbenzene or xylenes.

Thus, occupational exposure to toluene may be associated with hearing loss. Data are not adequate to establish any relationships for risk of adverse cardiovascular or pulmonary effects.

9.1.3 Reproductive and developmental toxicity

Limited data were available on the reproductive and developmental effects of exposure to toluene, ethylbenzene and xylenes in humans. The studies were limited to occupational exposure. Overall, there was very little evidence of reproductive and developmental toxicities for all three chemicals.

9.1.3.1 Reproductive effects

Relevant epidemiological studies include investigations of occupational exposure to organic solvents. Studies were limited to the inhalation route of exposure and primarily investigated spontaneous abortions. No studies examined the risk of reproductive toxicities in humans exposed to ethylbenzene via any route of exposure.

Reproductive studies of occupational exposure to toluene have examined risk of spontaneous abortion and fertility. The only study that investigated an almost exclusive exposure to toluene reported increases in rats of spontaneous abortion. A significant spontaneous abortion rate of 12.4% was reported in women working in an audio speaker factory exposed almost exclusively to toluene (mean concentration 88 ppm or 332 mg/m3), compared with rates of 2.9% and 4.5% in the internal and external control groups, respectively (Ng et al., 1992). An additional study of laboratory workers demonstrated an increased rate of spontaneous abortion in women reporting routine toluene-related work at least 3–5 days/week (OR 4.7, 95% CI 1.4–15.9) (Taskinen et al., 1994). One study of Finnish female shoe workers exposed to high levels of toluene showed an increased risk of spontaneous abortion, although the significance was lost when regression models specific to toluene were applied (Lindbohm et al., 1990). The significant findings may be due to exposure of the shoe workers to additional chemicals, including acetone and hexane. One study reported small, non-significant increases in spontaneous abortion rate upon exposure to toluene in women employed in eight Finnish pharmaceutical companies (Taskinen et al., 1986). Reproductive success was also investigated. Women categorized as highly exposed to toluene according to occupation exhibited a small, non-significant decrease in fertility as measured by time to pregnancy (Sallmén et al., 2008). Although there are few data to support reproductive effects of toluene, exposure may be associated with spontaneous abortions.

Although no studies have investigated the reproductive effects of ethylbenzene, two of the aforementioned studies pertaining to solvent exposure examined the effects of xylenes. Laboratory workers who reported handling xylenes more than three times per week exhibited increased rates of spontaneous abortion (OR 3.1; 95% CI 1.3–7.5) (Taskinen et al., 1994), whereas spontaneous abortion was not affected in another study of occupational exposure (Lindbohm et al., 1990). Like toluene, xylene did not significantly affect fertility in occupationally exposed women (Sallmén et al., 2008).

Although contradictory findings were reported, there is evidence of toluene-induced effects on spontaneous abortion in humans. However, data are insufficient to establish any human reproductive effects associated with exposure to ethylbenzene and xylene.

9.1.3.2 Developmental effects

There are very few epidemiological data on developmental effects associated with exposure to toluene, ethylbenzene and xylenes in humans. Studies of elevated toluene exposure through abuse during pregnancy have reported excess cases of premature birth, reduced birth weight/size, microcephaly and postnatal developmental delays (Arnold et al., 1994; Pearson et al., 1994). However, no association with congenital malformation was found in female laboratory workers reporting frequent exposure to toluene or xylenes (Taskinen et al., 1994). Moreover, a study of neural tube defects in relation to estimated ambient air levels of toluene (0.01–14.3 µg/m3), ethylbenzene (0.01–2.74 µg/m3) and xylenes (0.18–8.84 µg/m3) reported no increased rate of spina bifida or anencephaly (Lupo et al., 2011). Thus, developmental effects of exposure to toluene, ethylbenzene and xylenes are unlikely, other than at very high exposures, as observed in chronic solvent abusers.

9.2 Effects on experimental animals

9.2.1 Acute toxicity

The acute toxicity of toluene, ethylbenzene and xylenes is relatively low. The oral median lethal dose (LD50) for toluene in rats ranges from 5300 to 7400 mg/kg bw, whereas dermal exposure in rabbits resulted in an LD50 of 12 400 mg/kg bw (INRS, 2008). Via inhalation, the 4-hour median lethal concentration (LC50) of toluene is 7500 ppm (2828 mg/m3) in rats and 5308–7440 ppm (20,011-28,048 mg/m3) in mice (INRS, 2008). Oral exposure to ethylbenzene in rats resulted in LD50 values of approximately 2500 mg/kg bw (Wolf et al., 1956), and 4769 mg/kg bw (Smyth et al., 1962). LD50s for exposure of rats to xylene isomers via the oral route range from 3.6 to 5.8 g/kg bw, whereas the 4-hour LC 50s for inhalation exposure are approximately 6500 ppm (28,210 mg/m3) in rats and 4000–5000 ppm (17,360-18,850 mg/m3) in mice (IARC, 1989; WHO, 2004).

9.2.2 Short-term exposure

9.2.2.1 Neurological effects

Multiple studies have investigated the neurological effects of exposure to toluene, ethylbenzene and xylenes in experimental animals. Although most of the data pertain to inhalation exposure, neurotoxic effects following oral exposure were also documented.

Toluene exposure in animals was associated with altered behaviour and central nervous system effects, including modifications in neurotransmission. Only one study of oral exposure through drinking water was found. This study reported that toluene concentrations as low as 17 mg/L over 28 days (corresponding to a daily intake of 5 mg/kg bw) increased norepinephrine, dopamine and serotonin levels in the hypothalamus of male CD-1 mice as well as in other regions of the brain (Hsieh et al., 1990). Another study by oral gavage indicated neuronal necrosis in the dentate gyrus and Ammon's horn of the hippocampus in male and female rats at doses as low as 1250 mg/kg bw per day 5 days per week, after a thirteen-week exposure (Huff, 1990; NTP, 1990). Decreases in brain tissue may be related to breakdown of phospholipids (Kyrklund et al., 1987). Additional studies of exposure by inhalation have also noted alterations in levels of neurotransmitters, including norepinephrine, dopamine and serotonin, within rat brain and in prolactin levels in serum at toluene concentrations as low as 40 ppm or 151 mg/m3 (Ladefoged et al., 1991; Von Euler et al., 1994; Berenguer et al., 2003; Soulage et al., 2004). Studies suggest that these effects can persist considerably after exposure to toluene. For example, an increase in the affinity of dopamine D2 agonist binding in the rat caudate-putamen was observed 29–40 days following exposure to 80 ppm (302 mg/m3) toluene over 4 weeks (6 hours/day, 5 days/week) (Hillefors-Berglund et al., 1995). Six months of exposure to toluene concentrations as low as 500 ppm (1885 mg/m3) for 6 hours/day, 5 days/week, altered levels of norepinephrine, dopamine and serotonin in various regions of rat brain even 4 months post-exposure (Ladefoged et al., 1991). Exposure to toluene by inhalation also altered behaviour in rats at doses as low as 40 ppm (151 mg/m3) over 104 hours/week (Forkman et al., 1991; Berenguer et al., 2003; Beasley et al., 2010, 2012; Bikashvili et al., 2012), whereas others reported no effects on behaviour (Ladefoged et al., 1991; Von Euler et al., 1994). Significant behavioural changes were related to memory, problem solving, sensitization to toluene-induced narcosis and rearing activity. Other studies suggest that toluene inhalation may be related to neuronal death, as demonstrated by reduced weight of the whole brain and cerebral cortex following a 30-day inhalation of 320 ppm (1206 mg/m3) toluene in rats and reduced weight of the subcortical limbic area following a 4-week exposure to toluene in rats at concentrations exceeding 80 ppm or 302 mg/m3 (Kyrklund et al., 1987; Hillefors-Berglund et al., 1995). One study involving intraperitoneal injection of toluene at 300 mg/kg bw in mice demonstrated reduced memory accompanied by transcriptional down-regulation of memory-related genes (i.e., Nr1 and Nr2b) in the hippocampus.

Only two studies have investigated the neurological effects of exposure to ethylbenzene in animals. Li et al. (2010) reported that exposure of up to 500 mg/kg bw per day by oral gavage over 90 days in rats did not induce neurological abnormalities with regards to motor activity, autonomic functions, sensorimotor responses, reaction, gait or any other related clinical observation. Some neurological effects, including a decrease in landing foot splay in male rats and an increase in motor activity in female rats exposed to 750 mg/kg bw per day over 13 weeks (5 days/week), were observed by Mellert et al. (2007). However, the effects were only observed in males. No effects were observed at the lower exposure doses.

Oral exposure of B6C3F1 mice to xylenes by gavage over 13 weeks caused lethargy, rapid and shallow breathing, unsteadiness, tremors and paresis at the high dose of 2000 mg/kg bw, but only for a period of 15–60 minutes, approximately 5–10 minutes post-exposure (NTP, 1986). However, an inhalation concentration of 80 ppm (347 mg/m3) for 6 hours/day, 5 days/week, did not affect dopamine D2 agonist binding in rat caudate-putamen (Hillefors-Berglund et al., 1995). These findings suggest that transient neurological effects may occur shortly after exposure to xylenes. However, studies in rats that measured balance and coordination using the rotarod and other neurobehavioural tests identified some potential longer-term effects of m-xylene exposure. Korsak et al. (1992) demonstrated that exposure of 12 Wistar rats to 100 ppm (434 mg/m3) m-xylene over 6 months and exposure to 1000 ppm (4340 mg/m3) over 3 months significantly decreased rotarod performance and spontaneous activity 24 hours after the final exposure (which would allow most xylene to be eliminated from the animals). In a similar study by the same group, inhalation exposure of 12 male Wistar rats to 0, 50 or 100 ppm (or 0, 217 and 434 mg/m3) resulted in decreased rotarod performance at 100 ppm and decreased latency in the paw-lick response in the hot plate test at 50 ppm (Korsak et al., 1994). Additional studies done in rats indicated that exposure to m-xylene over several weeks affected the ability of the animals to navigate through a maze and caused other neurobehavioural abnormalities (Gralewicz et al., 1995; Gralewicz and Wiaderna, 2001).

Thus, toluene and xylene were associated with altered behaviour, impaired neurotransmission and damage to brain tissue, which may be implicated in persistent neurological dysfunction. It is unclear whether ethylbenzene is associated with adverse neurological effects.

9.2.2.2 Renal, hepatic and other tissue effects

Short-term oral and inhalation exposures to toluene, ethylbenzene and xylenes can affect multiple tissues. The most affected tissues were liver and kidney, although some effects have also been observed in the brain, heart and lung.

Oral and inhalation exposures to toluene are associated with effects in various tissues. Toluene exposure of rats by oral gavage over 13 weeks (5 days/week) and by whole body inhalation over 15 weeks (6.5 hours/day, 5 days/week) increased relative liver and kidney weights in both sexes at doses as low as 625 mg/kg bw and concentrations as low as 1250 ppm or 47125 mg/m3 (NTP, 1990). In rats exposed by inhalation, increases in relative brain, lung and heart weights were observed in both sexes, and increases in relative testis weight were observed in males, at doses exceeding 2500 ppm (9425 mg/m3). An increase in relative liver weight was also apparent in toluene-exposed mice of both sexes at doses as low as 312 mg/kg bw (5 days/week, 13 weeks) and concentrations as low as 625 ppm or 2356 mg/m3 (6.5 hours/day, 5 days/week, 14 weeks) (NTP, 1990). Male mice exposed to more than 1250 mg/kg bw also had increased relative brain and testis weights. Despite liver and kidney weight increases in rats and mice, no significant evidence of histopathological lesions was detected in these tissues. A significant increase in lesions was found only in rats exposed by oral gavage. These rats exhibited an increase in brain necrosis, particularly in the hippocampus, at doses exceeding 1250 and 2500 mg/kg bw in males and females, respectively. Also, an increase in hemorrhaging was observed within the urinary bladder, although rats in this dose group (5000 mg/kg bw) died shortly after exposure (NTP, 1990). Other studies showed that rats exposed to toluene by oral gavage (up to 422 mg/kg bw for up to 193 days) and by intubation (560 mg/kg bw for up to 6 months) displayed no signs of toxicity (Wolf et al., 1956; IPCS, 1986). Thus, toluene exposure was associated with increased weight in some organs, particularly the liver and kidneys, but few signs of histopathological lesions were present, other than in the brain.

It was reported by Mellert et al. (2007) that ethylbenzene administration in male and female rats by oral gavage (up to 750 mg/kg bw per day for 4 and 13 weeks) induced histopathological and serum chemistry changes. At the 13-week exposure mark, serum chemistry changes included increased alanine aminotransferase, total bilirubin, cholesterol, potassium, calcium and magnesium levels at doses exceeding 250 mg/kg bw and histopathological changes including increased liver and kidney weights at doses as low as 75 mg/kg bw. Increased incidences of hepatocyte centrilobular hypertrophy and hyaline droplet nephropathy were observed after 4 and 13 weeks of exposure. However, these increases were not statistically significant.. Another study of oral exposure in rats over 90 days demonstrated enlarged liver and kidney, which were significant at doses exceeding 250 mg/kg bw per day, but only in male rats (Li et al., 2010). Ethylbenzene exposure via inhalation for 13 weeks (6 hours/day, 5 days/week) also resulted in increased liver and kidney weights in male F344/N rats exposed by inhalation at concentrations of 500–750 ppm or 2170-3255 mg/m3 (NTP, 1992). Alkaline phosphatase, a marker of liver dysfunction, was elevated in male and female rats throughout exposure at doses as low as 100 ppm (434 mg/m3). Increased liver weight was also observed in male and female B6C3F1 mice at doses exceeding 750 ppm (3255 mg/m3). However, ethylbenzene did not induce any histopathological changes in any tissues in rats and mice in this study.

Enlarged liver and kidney were also observed in rats that were administered mixed xylenes by oral gavage for 90 consecutive days at doses as low as 750 mg/kg bw per day (Condie et al., 1988). Minimal chronic nephropathy in females was also observed (Condie et al., 1988). Nephrotic effects were not observed in male rats dosed with m-xylene at 0.5 or 2.0 g/kg bw by oral gavage for 5 days/week for 4 weeks (Borriston Laboratories Inc., 1983). Studies of xylene exposure (m- and p-xylene) by oral gavage in rats revealed no abnormal histopathological findings in any tissue or organ at doses as high as 800 mg/kg bw for 90 days (Wolf, 1988a, 1988b). However, survival incidence was decreased, and one study showed evidence of mottled lungs and a failure of lungs to collapse at doses as low as 200 mg/kg bw per day (90-day exposure) (Wolf, 1988a).

Thus, exposure to toluene, ethylbenzene and xylenes is primarily associated with hepatic and renal effects, including increased liver and kidney weights, nephropathy and hepatic hypertrophy. There is also evidence of toluene-mediated effects on the brain.

9.2.2.3 Other effects

(a) Ototoxicity

Ototoxicity was consistently observed for all three chemicals. Although most data were relevant to the inhalation exposure route, one study reported hearing loss following an oral exposure of rats to each of toluene, ethylbenzene and o-, m- and p-xylene individually at 8.47 mmol/kg bw per day (Gagnaire and Langlais, 2005). The study demonstrated ototoxicity, as shown by losses in outer hair cells of the organ of Corti. Ethylbenzene induced severe ototoxicity, as shown by an almost complete loss of three rows of outer hair cells in the medium and apical parts of the cochlea. Moderate ototoxicity was reported for toluene and p-xylene, but not for the other xylene isomers. Exposure to ethylbenzene and to two mixed xylenes by inhalation (6 hours/day, 6 days/week, for 13 weeks) induced hearing loss, as determined by brainstem auditory evoked responses and moderate to severe loss of outer hair cells of the organ of Corti at concentrations ranging from 200 to 800 ppm (or 868 to 3472 mg/m3) for ethylbenzene and from 250 to 2000 ppm (or 1085 to 8680 mg/m3) for mixed xylenes (Gagnaire et al., 2007). Other studies of toluene and ethylbenzene inhalation exposure in experimental animals have demonstrated ototoxicity, but only at relatively high doses of at least 1000 ppm or (3770 mg/m3) and 400 ppm or (1736 mg/m3) for toluene and ethylbenzene, respectively (Johnson and Canlon, 1994; Campo et al., 1997; Lataye and Campo, 1997; Cappaert et al., 1999, 2000). In contrast to humans, experimental animal data regarding ototoxicity suggest that direct damage to the auditory system may be the cause of ototoxicity. The exact role of neurological effects on ototoxicity is unknown.

(b) Immunotoxicity

Several studies have suggested that toluene is immunosuppressive. One study of mice exposed via drinking water showed a decrease in thymus weight, splenocyte lymphoproliferation in response to alloantigens, antibody plaque-forming cell responses and interleukin-2 production, but only at a high dose of 405 mg/L (Hsieh et al., 1989). These findings were supported in another study by the same group using the same doses (Hsieh et al., 1991). An additional study in which the highest dose was 325 mg/L showed no obvious immunotoxic effects (Hsieh et al., 1991).

Exposure of rats to up to 500 ppm (2170 mg/m3) ethylbenzene over 28 days was not immunotoxic, as demonstrated by the lack of plaque-forming cell response to sheep red blood cells (Li et al., 2010). The only evidence of xylene immunotoxicity was a decrease in thymus and spleen weights in rats exposed to p-xylene at 2000 mg/kg bw per day (Condie et al., 1988).

9.2.3 Long-term exposure and carcinogenicity

Long-term animal studies have been carried out for toluene, ethylbenzene and xylenes. Overall, these studies did not support toluene and xylenes as tumour-inducing chemicals via the oral, inhalation or dermal route. However, evidence was found of ethylbenzene-induced tumorigenesis, as well as nephropathy and renal hyperplasia.

The potential carcinogenicity of toluene has been investigated for oral and inhalation exposures. In the oral exposure study, toluene in olive oil at 500 or 800 mg/kg bw was administered by stomach tube (4–5 days/week for 104 weeks) to male and female Sprague-Dawley rats. Following the 104-week exposure, the rats were allowed to die of natural causes. This study suggested an increase in head cancers and leukemia and lymphoma in both sexes and mammary cancers in females (Maltoni et al., 1997). However, dose-related effects were not apparent, and study details were not adequately reported to draw any firm conclusions on carcinogenicity. Tumour incidence was not observed for toluene following a 2-year inhalation study in mice and rats at concentrations of up to 1200 ppm or 4524 mg/m3 (administered 6.5 hours/day, 5 days/week) (NTP, 1990; Huff, 2003). Although a few cases of nasal, kidney and forestomach neoplasms were reported in the female rats, these were determined to be not related to toluene exposure. Almost all rats, including controls, exhibited nephropathy, but the severity was slightly increased in rats of both sexes exposed to 1200 ppm. Thus, there is little evidence of toluene-induced carcinogenicity in experimental animals. However, chronic exposure may be associated with an increased severity of nephropathy.

The carcinogenicity of ethylbenzene has been reported in experimental animals through both the inhalation and oral exposure routes. In the inhalation carcinogenicity bioassays, groups of 50 male and female B6C3F1 mice and F344/N rats were exposed to 0, 75, 250 or 750 ppm (0, 326, 1090 or 3260 mg/m 3) ethylbenzene vapour for 103 and 104 weeks, respectively (Chan et al., 1998; NTP, 1999). A significant, concentration-related increase in the incidence of both alveolar/bronchiolar adenomas and combined alveolar/bronchiolar adenomas and carcinomas of the lung as well as a significant increase in the incidence of alveolar epithelium metaplasia were observed in male mice in the 750 ppm group. Female mice exhibited concentration-related increases in the incidence of both hepatocellular adenomas and combined adenomas and carcinomas, which were significant at the 750 ppm dose compared with concurrent controls, but remained within the historical control ranges. The incidence of eosinophilic foci in the liver was significantly greater in the female mice at 750 ppm, and the eosinophilic foci were considered a precursor to hepatocellular neoplasia. Male rats exhibited a concentration-dependent increase in the incidence of combined renal tubular adenomas and carcinomas, which was significant at 750 ppm. Significant increases in the incidence of renal tubular adenomas in female rats and testicular adenomas in male rats were also observed in the 750 ppm dose group. In rats of both sexes, there was a significant increase in the incidence of focal renal tubular hyperplasia at 750 ppm; this was considered to be a precursor stage of adenoma development by the authors of the study. Dose-dependent increases in the severity of chronic progressive nephropathy were observed in female rats at all exposure levels and in male rats at the highest concentration (Chan et al., 1998; NTP, 1999).

One study reported non-dose-related increases in total malignant tumours and head cancers over controls in male and female Sprague-Dawley rats exposed to ethylbenzene mixed in extra virgin olive oil and administered by stomach tube at 500 and 800 mg/kg bw per day over 104 weeks (Maltoni et al., 1997). However, it is not possible to draw firm conclusions from this study due to the lack of dose–response relationships and inadequate reporting of study details.

Additional studies have investigated long-term effects specific to xylenes. No increased tumour incidence was observed in male and female Fischer 344 rats exposed to mixed xylenes (60% m-xylene, 14% p-xylene, 9% o-xylene and 17% ethylbenzene) at 0–500 mg/kg bw per day or in male and female B6C3F1 mice exposed to mixed xylenes at 0–1000 mg/kg bw per day by oral gavage in corn oil for 103 weeks (5 days/week) (NTP, 1986). Although some cases of testis tumours were reported in the highest exposure group in rats, these effects were considered not to be treatment related. Both male and female mice exposed to the 1000 mg/kg bw per day dose exhibited hyperactivity 5–30 minutes post-exposure as of week 4. Non-dose-related increases in mammary cancers were observed in female rats exposed to xylenes at 500 mg/kg bw per day, and head cancer, lymphomas and leukemias were observed in male and female rats exposed to 500 and 800 mg/kg bw per day over 104 weeks (Maltoni et al., 1997). However, as stated previously, the lack of information provided in this study does not allow for adequate interpretation of the data.

Thus, long-term studies of toluene and xylene toxicity suggest that these chemicals are not likely to induce tumours or other adverse health effects upon chronic exposure. No acceptable study of oral ethylbenzene exposure was available. However, ethylbenzene inhalation was associated with an increased tumour incidence. Long-term exposure to ethylbenzene was also associated with nephropathy and renal hyperplasia.

9.2.4 Genotoxicity

Overall, studies of toluene, ethylbenzene and xylenes using cultured cells and experimental animals provided very little evidence of genotoxic activity.

9.2.4.1 In vitro findings

Overall evidence did not support toluene, ethylbenzene or xylenes as genotoxic agents. Toluene tested negative in the Salmonella typhimurium reverse mutation assay with and without metabolic activation (Nestman et al., 1980; Bos et al., 1981; Connor et al., 1985; Nakamura et al., 1987; NTP, 1990; Huff, 2003). Moreover, toluene did not induce sister chromatid exchanges or chromosomal aberrations in Chinese hamster ovary cells (NTP, 1990) or in human lymphocytes (Gerner-Smidt and Friedrich, 1978), even at concentrations that inhibited cellular growth in human lymphocytes (Richer et al., 1993). Ethylbenzene tested negative in the S. typhimurium reverse mutation assay (Florin et al., 1980; Nestman et al., 1980; Dean et al., 1985; Zeiger et al., 1992), in the Escherichia coli reverse mutation assay (Dean et al., 1985), in the Saccharomyces cerevisiae gene conversion assay and in the mouse lymphoma forward mutation assay (Wollny, 2000; Seidel et al., 2006), both with and without metabolic activation in each assay. One study reported the presence of single, but not double, deoxyribonucleic acid (DNA) strand breaks (Chen et al., 2008).

Ethylbenzene does not induce chromosomal damage in rat liver epithelial cells or in Chinese hamster ovary cells (Dean et al., 1985), although micronuclei were reported in Syrian hamster embryo cells (Gibson et al., 1997). Sister chromatid exchange was generally not observed, although one study reported marginal increases of sister chromatid exchanges in human lymphocytes at a cytotoxic dose of ethylbenzene (Norppa and Vainio, 1983).

Xylene mixtures and individual xylene isomers (o-, m- and p-xylene) tested negative in the S. typhimurium reverse mutation assay (Florin et al., 1980; Bos et al., 1981; Haworth et al., 1983; Connor et al., 1985; Shimizu et al., 1985) and in the E. coli reverse mutation assay (Shimizu et al., 1985; DeMarini et al., 1991), both with and without metabolic activation in each assay. Ethylbenzene did not induce sister chromatid exchanges or chromosomal aberrations in human lymphocytes (Gerner-Smidt and Friedrich, 1978; Richer et al., 1993) or in Chinese hamster ovary cells (Anderson et al., 1990). DNA strand breaks were observed in human lymphocytes, although it was established that they were associated with cytotoxicity (Morozzi et al., 1999). Thus, data do not suggest that toluene, ethylbenzene or xylenes are genotoxic in vitro.

9.2.4.2 In vivo findings

No tangible evidence of toluene, ethylbenzene or xylene genotoxicity was found in experimental animals. Intraperitoneal injection of all three solvents (e.g., toluene, ethylbenzene and individual isomers of xylene [o-, m- and p-xylene]) into mice at doses that ranged from 0.12 to 0.75 mL/kg bw (up to 70% LD50) administered twice did not induce micronuclei in polychromatic bone marrow erythrocytes for any of the solvents except for toluene, for which micronuclei could be detected at concentrations as low as 0.25 mL/kg bw (Mohtashamipur et al., 1985). However, a toluene inhalation exposure of up to 500 ppm (1885 mg/m3) for 6 hours/day over 8 weeks did not induce DNA strand breaks in peripheral blood cells, bone marrow or liver of mice (Plappert et al., 1994). Oral exposure to mixed isomers of xylene of up to 1000 mg/kg bw in mice induced neither chromosomal aberrations nor micronuclei in reticulocytes (ATSDR, 2007). Despite evidence of carcinogenicity in experimental animals, the few studies of ethylbenzene genotoxicity report negative findings. Intraperitoneal exposure to ethylbenzene at doses up to 650 mg/kg bw in both mice and rats revealed no evidence of genotoxic outcomes (Litton Bionetics, 1978; Washington et al., 1983; Mohtashamipur et al., 1985). Thus, toluene, ethylbenzene and xylenes are not likely to be genotoxic.

9.2.5 Reproductive and developmental toxicity

9.2.5.1 Reproductive effects

Convincing evidence of adverse reproductive effects was not found for toluene, ethylbenzene or xylenes. The one study that investigated the reproductive effects of oral toluene exposure revealed no effects on litter viability in mice exposed to a high dose of 2350 mg/kg bw on gestation days 7 through 14 (Smith, 1983). Other studies pertain to the inhalation route of exposure. Decreased sperm counts and weights of the epididymides were reported in male rats exposed to 2000 ppm (7540 mg/m3) for 90 days (6 hours/day) (Ono et al., 1996), whereas female rats exposed to 3000 ppm (11,310 mg/m3) for 7 days produced abundant vacuoles, lytic areas and mitochondrial degeneration in the antral follicles of the ovaries (Tap et al., 1996). However, no treatment-related histopathological lesions were observed in mouse and rat testes and ovaries, even after a 2-year exposure to 1200 ppm (4524 mg/m 3) toluene (NTP, 1990). Ungváry (1985) showed that continuous exposure of pregnant rabbits to 267 ppm (1007 mg/m3) toluene during gestation days 7–20 caused decreased maternal weight gain and abortions. Exposure of rats to doses of up to 3000 ppm toluene revealed no adverse effects on implantation, number or viability of fetuses or sex distribution upon caesarean section on gestation day 20 (Roberts et al., 2007). However, reduced maternal body weight was observed at 1500 ppm. Both two-generation studies for toluene revealed no adverse reproductive effects. Mating, fertility and pregnancy indices in offspring of pregnant rats exposed to up to 1200 ppm (6 hours/day between gestation days 9 and 21) were unaffected (Thiel and Chahoud, 1997). Moreover, impaired reproductive performance was not observed in rats exposed intermittently up to 2000 ppm (6 hours/day for up to 95 days) in comparison to controls (API, 1985).

Only one study examined the direct effects of ethylbenzene exposure in experimental animals via the oral route. Oral administration of 500 mg/kg bw and above in rats resulted in decreased peripheral hormone levels during the diestrous stage (Ungváry, 1986). One two-generation study examined reproductive capability following ethylbenzene inhalation (Faber et al., 2006). Rats were exposed to up to 500 ppm throughout premating (at least 70 days in both sexes), throughout gestation (up to gestation day 20) and on lactation days 5–21 (with oral exposure that would result in the same blood concentration on lactation days 1–4). The study revealed no effect on reproductive parameters, including mating and fertility, gestation time, size and viability of litters and sex distribution. The only significant effect was a reduction in estrous cycle length in F0 (parent) females (Faber et al., 2006). Inhalation exposure to up to 1000 ppm ethylbenzene in Wistar rats for 3 weeks did not affect female fertility (Hardin et al., 1981). An ethylbenzene inhalation exposure in mice and rats ranging from 99 to 975 ppm over 13 weeks caused a decrease in epididymal weight in mice, though no other reproductive effects on sperm or menstrual cycle were observed in mice or rats (NTP, 1992). Thus, there is little evidence of ethylbenzene-induced reproductive effects in experimental animals.

Little information was available for reproductive effects of exposure to xylenes. None was relevant to oral exposure. Testicular alterations were not observed at exposure concentrations as high as 1000 ppm or 4340 mg/m3 (Nylén et al., 1989; Korsak et al., 1994). Thus, xylenes are unlikely to cause adverse reproductive effects.

The information that was available on toluene, ethylbenzene and xylenes does not support adverse reproductive effects of exposure, even at the high doses administered in these studies.

9.2.5.2 Developmental effects

Limited evidence of developmental toxicity was associated with exposure to toluene, ethylbenzene and xylenes.

In a series of experiments in rats, it was determined that oral gavage of toluene at 520 or 650 mg/kg bw caused a reduction in fetal, liver and kidney weights (Gospe et al., 1994, 1996; Gospe and Zhou, 2000). Reductions in heart weight and skeletal ossification were also observed at the higher dose (Gospe et al., 1996; Gospe and Zhou, 2000). Effects on organ weights occurred on gestation day 19 and persisted until postnatal day 10. Although organ weights did not differ from those of controls by postnatal day 21, histological analyses of the brain revealed decreased neuronal packing and alterations in the patterns of staining with bromodeoxyuridine, indicating toluene-induced alterations in neurogenesis and neuronal migration (Gospe and Zhou, 2000). Exposure of rats by inhalation also resulted in reduced fetal weight at toluene concentrations of 1000 mg/m3 (8 hours/day on gestation days 1–21) (IPCS, 1986) and as low as 250 ppm or 943 mg/m3 (6 hours/day on gestation days 6–15) (Roberts et al., 2007). An additional study reported a reduction in sperm count and weight of epididymides in rats exposed to 2000 ppm (7540 mg/m3), 6 hours/day, for 90 days, despite there being no effect on reproductive performance (Ono et al., 1996). A two-generation reproduction study reported no adverse effects in rats exposed to 2000 ppm toluene (6 hours/day) for up to 95 days (API, 1985).

Studies of oral exposure to ethylbenzene were not located. In rats, an increased number of fetuses with extra ribs was observed at inhalation concentrations as low as 100 ppm (434 mg/m3) ethylbenzene (Hardin et al., 1981). Exposure of pregnant rats to ethylbenzene concentrations exceeding 1000 ppm (4340 mg/m3) on gestation days 6–20 caused decreased fetal weight (Saillenfait et al., 2003, 2006, 2007), whereas 2000 ppm (8680 mg/m3) ethylbenzene caused increased incidences of skeletal variations in offspring (Saillenfait et al., 2003). However, no significant findings pertaining to pup survival and weight, neurological functioning or developmental landmarks were found in the F1 and F2 generations of mice exposed to up to 500 ppm or 2170 mg/m3 (for 6 hours/day) for 70 consecutive days (Faber et al., 2006, 2007). Thus, developmental effects may be possible, but only at elevated doses.

Two studies reported adverse developmental effects of xylenes in animals via oral exposure. A high exposure of 2060 mg/kg bw per day on gestation days 6–15 was associated with decreased body weight and an increase in malformations (primarily cleft palate) (Marks et al., 1982). However, no evidence of teratogenic effects was observed in mice exposed to m-xylene at 2000 mg/kg bw per day on gestation days 8–12 (Seidenberg et al., 1986). Inhalation exposure to xylenes has been associated with neurological deficits in offspring at doses as low as 200 ppm or 868 mg/m3 (Hass and Jakobsen, 1993; Hass et al., 1997). Developmental effects, including skeletal abnormalities and decreased fetal weight, have occurred in rats. Taking into consideration the various limitations of the rat studies, these effects were determined to occur only at concentrations above 350 ppm or 1519 mg/m 3 (ATSDR, 2007).

Thus, it is possible that developmental effects may occur as a result of exposure to toluene, ethylbenzene and xylenes. However, the exact role of maternal toxicity in these endpoints is not fully understood.

9.3 Mode of action

Due to their similar chemical properties, toluene, ethylbenzene and xylenes are distributed to lipid-rich tissues, where they exert toxic outcomes through various modes of action.

The lipophilic properties of these solvents are primarily responsible for their acute symptoms of exposure. The narcotic and anesthetic effects of toluene, ethylbenzene and xylenes that occur upon acute exposure may be related to the rapid intercalation of these solvents within the lipid bilayer of nerve membranes, leading to biochemical modifications of membrane-bound proteins and subsequent altered synaptic transmission. One study reported decreases in acetylcholinesterases, total adenosine triphosphate (ATP) and Mg2+-ATPase activities in erythrocytes and synaptosomes of rats exposed to 2000 ppm (7540 mg/m3) toluene for 2 hours (Korpela and Tahti, 1988). Altered synaptic functioning is further supported by global gene expression analysis of various brain regions in the rat following inhalation of toluene over 6 hours (Hester et al., 2011). The analysis revealed pathways associated with altered synaptic plasticity. Similar observations were made for ethylbenzene and xylenes. Ethylbenzene and xylenes decreased activities of Na+/K+-ATPase and Mg2+-ATPase in cultured astrocytes from the cerebella of neonatal Sprague-Dawley rats (Vaalavirta and Tahti, 1995). Effects on the activities of these membrane-bound proteins were more substantive for ethylbenzene and xylenes than for toluene. Overall, these data suggest that toluene, ethylbenzene and xylenes can perturb membrane-bound proteins, leading to adverse central nervous system effects.

Disturbances in brain function may be further related to effects on neurotransmitters, such as persistent alterations in enzymes that regulate their synthesis, degradation and binding. Exposure to toluene by inhalation in experimental animals has resulted in increases in neurotransmitters, including glutamate, taurine, dopamine, norepinephrine, serotonin and acetylcholine, in various brain regions (Rea et al., 1984; Aikawa et al., 1997). Receptor binding of glutamate and γ-aminobutyric acid generally increased in various brain regions following toluene exposure at concentrations exceeding 50 ppm or 189 mg/m3 (Bjornaes and Naalsund, 1988). One study noted that toluene-induced inhibition of N-methyl d-aspartate receptor binding may play a role in the incoordination and memory impairment that occur following toluene exposure (Lo et al., 2009). Xylenes were shown to increase dopamine and catecholamine levels in various brain regions of rats exposed to 2000 ppm (8680 mg/m3), whereas the same concentration of ethylbenzene decreased catecholamine levels (Andersson et al., 1981). An exposure to 750 ppm (3255 mg/m3) ethylbenzene caused a reduction of stratial and tubero-infundibular dopamine in rabbits, whereas exposure to toluene and xylenes had no effect (Mutti et al., 1988). Altered neurotransmitter concentrations resulting from solvent exposure appear to be a transient, reversible effect. Nonetheless, chronic exposure may be associated with persistent alterations in behaviour, depressed mood and loss of memory.

Longer-term neurological effects may be related to cytotoxicity and damage to the central nervous system. Central nervous system cytotoxicity may result from phospholipid degradation, as noted especially for toluene. Toluene exposure has been shown to decrease phospholipid concentrations in the cerebral cortex of rats, leading to a loss of grey matter upon continuous exposure to 320 ppm (1206 mg/m3) toluene over 30 days (Kyrklund et al., 1987). Exposure to 1500 ppm (5655 mg/m3) in rats over 6 months (6 hours/day, 5 days/week) with 4 months of recovery decreased the number of neurons within the regio inferior of the hippocampus (Korbo et al., 1996). Such effects are supported in humans by diagnostic magnetic resonance imaging of chronic solvent abusers (Borne et al., 2005). Chronic abuse of toluene resulted in irreversible damage, including central nervous system atrophy, demyelination and minimal gliosis. Although adverse neurological effects have been documented for ethylbenzene and xylenes, their exact role in tissue damage within the central nervous system is unknown.

The mode of action of toluene, ethylbenzene and xylenes in inducing ototoxicity appears to be attributed to the death of cochlear hair cells in experimental animals (Gagnaire et al., 2001; Gagnaire and Langlais, 2005). This effect appears to be mediated by the presence of a single short side-chain on the benzene ring in aromatic solvents (Gagnaire and Langlais, 2005). However, there is evidence in humans that a neurological component may be involved.

Studies in animals investigating the metabolic interactions between toluene and other chemicals and ototoxicity indicate that toluene-induced hearing loss is caused by toluene itself, not its metabolites (Wallen et al., 1984; Römer et al., 1986; Pryor, 1991; Imbriani and Ghittori, 1997; Campo et al., 1998). Other neurological effects, including central nervous system depression and narcosis, are also believed to involve toluene itself and not its metabolites (ATSDR, 2000).

In the case of xylenes, for neurological effects such as changes in levels of various neurotransmitters and lipid composition observed following exposures to xylene of acute and intermediate durations (Savolainen and Seppalainen, 1979; Andersson et al., 1981; Honma et al., 1983), it is unclear whether the effects are due to xylene itself or to its metabolic intermediates, such as arene oxides or methylbenzaldehyde (Savolainen and Pfäffli, 1980). Methylbenzaldehyde (the product of the action of alcohol dehydrogenase on methylbenzyl alcohol) has been detected in animals; however, its presence has not been confirmed in humans (ATSDR, 2007).

There is strong evidence that tissue damage observed within the central nervous system, liver, kidney and other tissues affected by toluene, ethylbenzene and xylenes would be mediated by oxidative stress. For toluene, substantive evidence of oxidative stress, including increased generation of reactive oxygen species and markers of oxidative damage, has been observed in the brain (Mattia, 1993; Burmistrov et al., 2001; El-Nabi Kamel and Shehata, 2008), liver (Tokunaga et al., 2003) and kidney (Mattia, 1993; Tokunaga et al., 2003) of experimental animals following inhalation exposure to and intraperitoneal injection of toluene. Moreover, the synaptosomes of rats exposed in utero had an increased level of oxidative stress upon reexposure to toluene in vitro, thus indicating long-lasting changes in oxidative status that can affect offspring upon maternal exposure (Edelfors et al., 2002). Markers of oxidative stress have been reported in the brains of rats following inhalation exposure to ethylbenzene concentrations exceeding 433.5 mg/m3 (Wang et al., 2010). This is supported by urinary markers of oxidative stress correlated with ethylbenzene exposure in spray painters (Chang et al., 2011). Although analyses for xylenes were in other tissues, they appeared to cause less oxidative stress, as demonstrated by indicators of oxidative damage in kidney but not in liver (Kum et al., 2007a, 2007b) and the lack of correlation between exposure and excretion of urinary markers (Chang et al., 2011). Such tissue damage may be attributed to induction of apoptosis, inflammation and cellular proliferation, leading to toxicities in affected organs.

Exposure to ethylbenzene has been shown to induce kidney and Leydig cell tumours in rats and lung and liver tumours in mice. Genotoxicity screening studies for ethylbenzene indicate that it is non-genotoxic in vivo and predominantly non-genotoxic in vitro (VCCEP, 2007), and thus other modes of action have been proposed for its propensity towards tumour induction. Ethylbenzene is not likely carcinogenic at doses below a toxic threshold. More information on ethylbenzene's modes of action involved in tumour induction in rodents is presented below.

There is evidence to support the mode of action of kidney tumours in rats resulting from an increased incidence of chronic progressive nephropathy by the primary ethylbenzene metabolite, 1-phenylethanol (VCCEP, 2007). The mode of action may also include a possible weak accentuation of chronic progressive nephropathy by involvement of a2u-globulin in male rats. Because of critical qualitative and, to a certain extent, quantitative species differences, this mode of action is not expected to be relevant in humans. Therefore, rat kidney tumours are not a suitable basis for risk assessment. A study by Hard et al. (2012) thoroughly reevaluated studies of chronic progressive nephropathy in rats. The authors determined there was definitive evidence that advanced stages of chronic progressive nephropathy represent a risk for development of a low incidence of basophilic renal tubule adenomas and their precursor form of hyperplasia in both male and female F344 rats. This work adds to the weight of evidence that chemical exacerbation of chronic progressive nephropathy represents a secondary mode of action for tumour development that is unlikely to have relevance for species extrapolation in risk assessment because there is no counterpart of rat chronic progressive nephropathy (biologically and histopathologically) in humans.

The postulated mode of action for liver tumours in female rats is a phenobarbital-like induction, which was also considered not relevant to humans. Leydig cell tumours were likely induced by alterations of serum testosterone, although such tumours are common in aged rats. Because of many well-documented qualitative and quantitative differences between rats and humans, the Leydig cell tumours observed are not expected to be relevant to human health risk assessment (VCCEP, 2007).

Lung tumours occurring upon inhalation are likely formed as a result of regenerative cellular proliferation following metabolism and exposure to cytotoxic metabolites. The proposed mode of action for ethylbenzene-induced lung tumours is as follows: (1) absorption of ethylbenzene; (2) distribution of ethylbenzene to lung; (3) metabolism of ethylbenzene to active metabolite; (4) detoxification/elimination of active metabolite; (5) possible oxidative stress secondary to high-dose glutathione depletion and/or high-dose-mediated CYP450 ethylbenzene metabolism; (6) arylation of macromolecules, leading to cytotoxicity when detoxification and repair capacities are exceeded; and (7) promotion/progression of lung tumours. The potential role for oxidative metabolites in the mode of action of ethylbenzene-induced lung tumours suggests that these tumours observed in mice are relevant to humans. This mode of action does suggest that there is a threshold below which tumours are not expected to be observed. It should be noted that there are qualitative differences between mouse and human pulmonary metabolism that may lead to increased sensitivity in mice (VCCEP, 2007) and that lung tumours were within the National Toxicology Program historical control range (NTP, 1999). However, considering the mode of action analysis as described above and that other cancer endpoints (liver, Leydig and kidney) have been concluded as being non-relevant to humans, lung tumours were determined to be the most relevant cancer endpoint for humans.

Report a problem or mistake on this page
Please select all that apply:

Thank you for your help!

You will not receive a reply. For enquiries, contact us.

Date modified: