Registration Dossier

Administrative data

Description of key information

In animals, after inhalation and oral repeated exposure, the principal affected organs are liver and kidneys. Mostly these changes appeared after inhalation exposure to concentrations of 3000 ppm (10710 mg/m3) and above or at oral doses greater than 209 mg/kg bw/day (exposure via drinking water) or 300 mg/kg bw/day (exposure by gavage).
The SCOEL has derived an 8-h TWA of 50 ppm (178.5 mg/m3) for inhalation exposure based on the available toxicity data.
No dermal repeated dose toxicity studies are available.
Human studies provided limited information on MTBE long-term-effects.

Key value for chemical safety assessment

Repeated dose toxicity: via oral route - systemic effects

Endpoint conclusion
Endpoint conclusion:
adverse effect observed
Dose descriptor:
209 mg/kg bw/day
Study duration:
Quality of whole database:
Two key studies - one showing lowest NOEL shown above

Repeated dose toxicity: inhalation - systemic effects

Endpoint conclusion
Endpoint conclusion:
adverse effect observed
Dose descriptor:
2 856 mg/m³
Study duration:

Repeated dose toxicity: inhalation - local effects

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed

Repeated dose toxicity: dermal - systemic effects

Endpoint conclusion
Endpoint conclusion:
no study available

Repeated dose toxicity: dermal - local effects

Endpoint conclusion
Endpoint conclusion:
no study available

Additional information

Across repeated dose toxicity studies, signs of clinical toxicity are commonly reported at high exposure concentrations (typically ≥1000 mg/kg bw/day) such as ataxia and hypoactivity.

After inhalation and oral repeated exposure, the principal affected organs are liver and kidneys. Mostly these changes appeared at inhalation concentrations of 3000 ppm (10710 mg/m3) and above or at oral doses greater than 209 mg/kg bw/day (exposure via drinking water) or 300 mg/kg bw/day (exposure by gavage).

Degenerative changes and protein droplet nephropathy of the male rat kidney proximal convoluted tubules are common findings. In inhalation studies, only slight indications of kidney toxicity are seen up to 1000 ppm. At 4000 ppm, the male rats show clear protein droplet nephropathy. The results from the oral studies show slightly different effect levels: Williams et al. (2000) noted protein droplet nephropathy visible in light microscopy in most male rats at 250 mg/kg bw/day. The 90-day study by Robinson et al. (1990) reported this effect only at 1200 mg/kg bw/day. The study by Zhou et al. (1999) mentioned no signs of nephropathy up to 1000 mg/kg bw/day in the same rat strain. In the oral drinking water study (CIIT, 2007), α2u-globulin related nephropathy was observed at all tested doses.

Based on the available data it can be concluded that these effects (protein droplet accumulation in the proximal tubules leading to proliferation and eventually to nephropathy) are caused by the accumulation of α2u-globulin, which is specific to male rats. Thus, these effects are considered to have no relevance for the human toxicological risk assessment (see also review of Hard (2006) in the section on carcinogenicity).

Regarding inhalation exposure, the overall NOAEC for MTBE is 800 ppm (2856 mg/m3) based on absolute and relative organ weights (mainly liver) in a 90 day rat study at 4000 ppm (14280 mg/m3) (Bushy Run Research Center, 1989a) and in the two-year study at 3000 ppm (10710 mg/m3) (Bushy Run Research Center, 1992a, see section 5.8.3 regarding carcinogenicity). The relative liver weight increase in the 90-day study at 4000 ppm was 20% for males and 13% for females.

In the two-year study (Bushy Run Research Center, 1992a discussed in section 5.8.3 on carcinogenicity), a relative liver weight increase of 20% was seen at 3000 ppm. No data were available for males due to early termination; chronic progressive nephropathy (CPN) was the major cause of death in males from the 3000 ppm and 8000 ppm groups in that study. The absence of these effects at 500 and 400 ppm in the other 13-week (Inveresk Research International, 1980) and the 28-day study (Bushy Run Research Center, 1993a) rat studies, respectively, supports the selection of NOAEC value of 800 ppm. In the study of Inveresk Research International (1980), a NOAEC of 500 ppm was chosen due to reduced lung weight seen at 1000 ppm. However, this finding was not seen in any other study except transiently at 14 days in the 90-day oral study. Apart from slightly increased severity and incidence of CPN, there were no significant signs of adverse microscopic changes at 400 ppm in the two-year carcinogenicity study. CPN is a common pathological event in ageing rats that can be enhanced under conditions of prolonged exposure to a xenobiotic chemical. As explained in the EU Risk Assessment Report for MTBE, the use of the two-year study for other toxicological endpoints is somewhat limited since haematological analysis and urinalysis was performed only for the control and high concentration animals (European Commission, 2002).

With regard to oral exposure, liver abnormalities have been found in the 28 and 90 day oral exposure studies. In the 90-day oral study (Robinson et al. 1990), at 300 mg/kg bw/day only the female rats had a statistically significant increase in kidney weight. However, this was not accompanied by degenerative microscopic findings, which only appeared in the males of the 1200 mg/kg bw/day group. In addition, clinical chemistry analyses for the females showed no adverse signs that would support kidney toxicity at that level. Therefore, the critical effect is considered the statistical significant weight increase in the male liver at 900 mg/kg bw/day and the overall oral NOAEL is considered to be 300 mg/kg bw/day based on the available gavage studies.

A 13-week drinking water study with Wistar rats is available for assessment (CIIT, 2007), where the only organ exhibiting treatment-related effects in males and females was the kidney with histopathological effects observed in males at all dose levels. These were shown to be α2u-globulin related and therefore not considered relevant for human health risk assessment. Based on increased relative kidney weights in males and females and depressed body weights in males, NOAELs of 209 mg/kg bw/day (males) and 272 mg/kg bw/day (females) are derived. Relative kidney weights were elevated in males and females at the higher tested dose levels (21% for males at 514 mg/kg bw/day and females at 650 mg/kg bw/day, 19% for males at 972 mg/kg bw/day and 30% for females at 1153 mg/kg bw/day) at the end of 13 weeks of exposure.

Effects on kidneys were also evident in a recent two-year drinking water study with Wistar rats discussed in section 5.8.3 on carcinogenicity (Dodd et al., 2010). As part of the two-year drinking water study, results of a 13-week screening study are available and the results of toxicity evaluation after one-year of exposure, which have been published separately (Bermudez et al., 2012).

Weights of male kidneys were increased at the end of two years of exposure to 7.5 mg/ml MTBE and this was already observed during the 13-week screening study (Dodd et al., 2010). Kidney weights of females had a statistically significant increase at 15 mg/mL after a 13-week exposure, but only the left kidney after a one-year exposure and there was no statistically significant difference after two-year exposure (Dodd et al., 2010).

Evidence of CPN was observed in males and females as low as 25 and 49 mg/kg bw/day, respectively. The CPN was more severe in males and was exacerbated in the high MTBE exposure groups of 330 mg/kg bw/day for males and 1042 mg/kg bw/day for females. As discussed above with respect to the two-year carcinogenicity study by inhalation, CPN is a common pathological event in ageing rats that can be enhanced under conditions of prolonged exposure to a xenobiotic chemical and therefore has a limited relevance to human health risk assessment.

It should be noted that Williams et al. (2000) found increased relative kidney weight and protein droplet nephropathy with significantly increased severity and incidence in the same male rat species (Sprague-Dawley) already at 250 mg/kg bw/day by gavage. The effects seen in liver, namely weight increase, hypertrophy and slight morphological changes, were mostly recorded at doses of 500 mg/kg bw/day and higher. These effects may be adaptive responses, which is corroborated by the fact that there are few obvious signs of liver toxicity in the rat even in the two-year carcinogenicity studies (see section regarding carcinogenicity). In any case, the EU Risk Assessment Report for MTBE (European Commission, 2002) put less weight on the study by Williams et al. (2000) since it was conducted mainly to observe changes in endocrine homeostasis with limited weight put on statistical analysis of other toxicological endpoints.

In a repeated dose study with exposures of either two or four weeks in Sprague-Dawley rats at doses of 0, 400, 800, 1600mg/kg bw/day by gavage (Dong-mei et al., 2009). A large number of variations in relative organ weights was observed, however none of the findings appear to have been reproduced between the experiments, even though conducted at the same exposures and experimental conditions, and there is no evidence of dose-response in any of the organ weights (Dong-met et al., 2009). An unexplained high (30%) mortality in the low dose group gives further doubts on the reliability of the study. Therefore this study has been disregarded for risk assessment purposes. 

In the Williams et al. study (2000) with Sprague-Dawley rats investigating endocrine effects, the highest exposure group of 1500 mg/kg bw/day had a decreased dihydrotestosterone (DHT) after 28 days.Triiodothyronine (T3) decreased significantly (19% of control) in rats dosed with 1000 and 1500 mg/kg bw/day for 28 days, as well as a statistical decrease in luteinizing hormone (LH) at 1500 mg/kg bw/day, but without a change in serum thyroid-stimulating hormone (TSH) or thyroxine (T4). However, in a separate experiment at a single dose of 1500 mg/kg for 15 days, T3 and LH were not seen as decreased although testosterone in both serum and testicular interstitial fluid were significantly decreased compared to controls.


Williams et al. (2000) stress that caution must be used when interpreting the high-dose responses due to clear signs of clinical toxicity being apparent in these animals throughout the study such as hypoactivity and dehydration. Furthermore histopathology found lesions in the liver at oral dose levels 500, 1000 and 1500 mg/kg bw/day, as well as kidney lesions at all doses, indicating a potential change in liver and kidney function as a result of specific MTBE toxicity in the rat. Williams et al. (2000) highlight that when excluding the high dose group, the only weak, if any, observed effect on the endocrine system of the male rats was the notable change in T3 serum concentrations. In particular, there were no effects on hormone levels at 250 and 500 mg/kg bw/day in the studies.

When considered together with the clear clinical toxicity evident at 1500 mg/kg bw/day and the lack of dose response for hormone levels measured, there is no evidence that the effects seen in Williams et al. (2000) at 1500 mg/kg are a result of endocrine disruption. Effects on endocrine homeostasis, such as levels of T3, testosterone and LH, can potentially be caused by stress (e. g. Everds et al., 2013) or may be secondary effects of MTBE toxicity.

As MTBE may have particular effects on the liver, a parallel study by Williams & Borghoff (2000) examined biotransformation reactions in the rat and found that the MTBE exposures at or above 1000 mg/kg bw/day by gavage can result in increased activity UDP-glucuronyltransferase and P450-related enzymes. UDPGT is the major enzyme for elimination of T3 and T4 in rodents, but an explanation of a decrease in T3 without a change T4 is not provided by the researchers (Williams & Borghoff, 2000). With regards to potential effects on testosterone levels, Williams & Burghoff (2000) deem the degree of induction of specific CYP enzymes by high exposures to MTBE as particularly low compared to known inducers and question whether the induction is sufficient to decreases in serum testosterone via enhanced clearance.

With regards to potential for stress impacting the endocrine homeostasis, Williams et al. (2000) report “difficulty administering MTBE, a noxious irritating chemical”. This difficulty may be partly from the volumes administered and potential aspiration hazard of MTBE, as well as any odour/taste aversion (although gavage is used, at high test concentration amounts exhaled are much higher than odour threshold).

For instance, effects of stress on corticosterone levels from gavage and aspiration of test articles during gavage has been reported by Brown et al., 2000. Stress effect on costcosterone levels and the functioning of endocrine relevant organs (adrenals, testes) from immobilization stress are reported in the rat by Pellegrini et al., 1998.  This is important to consider because elevated corticosterone is known to reduce testosterone production (Monder et al., 1994) and this effect has been seen with other ethers (Tohei et al., 1997).  Stress is also known to result in T4 conversion to reverse T3 instead of T3 although the database on measured levels of reverse T3 in animals is limited, it does include an observation in rats under immobilization stress (Bianco et al., 1987).

Unfortunately, corticosterone levels were not measured by Williams et al. (2000), although statistically significant increased absolute and relative adrenal weights were observed at the 1500 mg/kg bw/day exposure. Effects on corticosterone levels are however reported in other studies with male rats following gavage administration of MTBE, potentially with doses as low as 40 mg/kg bw/day (e.g. de Peyster et al., 2003). Changes in corticosterone levels have also been seen in MTBE inhalation studies at dose concentrations above NOAEC and metabolic saturation in a 90 day repeat dose study in Fischer rats (Lington et al., 1997), as well as the 2-year carcinogenicity studies in rats and mice (Bird et al., 1997). In these studies, corticosterone was generally seen as increased at high exposure groups, although not always significantly, with the exception of the 2-year carcinogenicity study where it was seen to be decreased in male rats at 8000 ppm compared to controls.

Increased adrenal weight is an indicator of stress, which was evident in the higher exposure groups after 90-day exposure in rats (Lington et al. 1997), but only in male rats and mice in the highest exposure group (Bird et al., 1997). Increased adrenal weight is seen at high dose levels in other repeated dose studies reported in the IUCLID, although there is not a consistent pattern evident (Robinson et al., 1990; Bushy Run Research Center, 1993a; de Peyster & Mihaich, 2013). The variability on corticosterone levels and effects on adrenal glands across studies supports potential effects of stress, rather than a specific mode of action or direct toxicity to the adrenal gland, as stress is a variable that is difficult to control and may be affected by multiple compounding factors.

In the two-year drinking water carcinogenicity study in Wistar rats (Dodd et al. 2010),discussed in section 5.8.3 on carcinogenicity, there were statistically significant elevated levels in T3 in males after a 6-month exposure at the mid-dose groups of 0.5 mg/mL and 3 mg/mL. However, the blood chemistry was normal at the end of a 12-month exposure period and there were also no other changes in thyroid parameters (TSH or T4) or thyroid histopathology. The increase in T3 observed after the 6-month exposure period was not considered to be related to MTBE-exposure (Dodd et al., 2010).

Given the lack of potential for MTBE to cause reproductive or developmental toxicity, as discussed in section 5.9.3 on reproductive toxicity, there is no supporting evidence in the reproductive and developmental toxicity database on developmental toxicity to suggest that MTBE has an effect on T3.

Overall, the observed significant difference in T3 reported by Williams et al. (2000) is questionably related to MTBE  (significance was only at p< 0.05 with some measurements of T3 in the 1000 mg/kg bw/day exposure group within the range of controls) and also of questionable biological relevance. 

For consideration of other endocrine-related endpoints examined during repeated-dose exposures, the high exposure inhalation study at 8000 ppm by Moser et al. (1998) in female B6C3F1 mice has shown that MTBE can cause changes in estrogen sensitive tissues without affecting serum estrogen levels: MTBE exposure significantly decreased body weight gain and ovary and pituitary weight at 4 and 8 months and uterine weight at all time points; after 8 months of exposure, there was a significantly increased the length of the estrous cycle by increasing the mean number of days in both the estrus and the nonestrus stages; histological evaluation of H&E-stained tissues showed a decrease in the number of uterine glands after subchronic MTBE exposure.


The female reproductive system is more susceptible to stress than the male system and is particularly sensitive to the stress associated with decreased feed intake and decreased body weight gain, with the most sensitive reproductive parameter being disturbance of the estrous cycle, with stress generally leading to an extended duration of the estrous cycle (Everds et al., 2013). Moser et al. (1998) did not measure feed intake, but noted that decreases in body weight gain were mild compared to that typically seen in food restriction studies. Moser et al. (1998) did not measure corticosterone levels, although the carcinogenicity study in mice did see increased corticosterone in female mice at 8000 ppm that were not statistically significant (Bird et al., 1997). 


Because Moser et al. (1998) did not observe changes in levels of serum estrogen, this indicates that there is not an effect from hepatic catabolism of endogenous estrogen. .Moser et al. (1998) suggest that the effects observed may through an anti-estrogenic mode of action. However, a weight of evidence analysis on endocrine interactions presented in section 5.10.3 on specific investigations concludes that there are no clear supportive evidence of MTBE having a direct effect on the estrogen pathway. Therefore the effects observed by Moser et al. (1998) may likely be a result of general or specific toxicity. Although it is not possible to derive a NOAEL from the Moser et al. (1998) because the study exposure was at such a high concentration, the effects on organs weights have also been observed in other studies (pituitary, uterine, ovary), NOAELs from other toxicological studies on MTBE can adequately cover the effects from general or specific toxicity.


In addition, if changes in liver induction of P450 enzymes are resulting in enhanced metabolism of steroid hormones, this type of induction is unlikely to be significant at low doses of MTBE as only minimal levels of enhanced activity are reported at high doses and impact at high doses is unlikely to be reflective of low dose. Furthermore, is this P450 activity is unlikely to impact homeostasis at even high MTBE exposure levels, as discussed in section on specific investigations.

The potential for endocrine disruption of MTBE based on the findings of Williams et al. (2000) and Moser et al. (1998) were reviewed by the EU Scientific Committee on Toxicity, Ecotoxicity and the Environment (CSTEE) which concluded that the “effects are judged not to be critical” (CSTEE, 2001). This is in line with recent discussions on endocrine disruption classifications, where potential endocrine effects are not deemed as relevant when observed with other toxic effects and/or may arise from secondary non-specific toxicity.

Other findings on the potential for MTBE to interact with the endocrine system are discussed in section 5.9.3 on reproductive toxicity. The results of an evaluation on the potential for MTBE to interact with the endocrine system are presented in section on specific investigations.


No dermal repeated dose toxicity studies are available. In the EU Risk Assessment Report (European Commission, 2002) it is noted that due to the lipid extraction properties of MTBE, it can be presumed that repeated skin exposure may result in skin fatigue (and consequent risk of toxic eczema), an effect common to a variety of organic solvents. No quantitative data on this effect are available.

Human studies provided limited information on MTBE long-term-effects. Most of the studies have been unsuccessful in controlling the variables, have mixed exposure or are biased.


The SCOEL (2006) has also evaluated the repeated dose toxicity database of MTBE with the focus on inhalation exposure and concluded that there are no effects of toxicological significance for human health below 1000 ppm, with findings reported only at 3000 ppm and above. Based on that, the SCOEL derived an 8-h TWA of 50 ppm (178.5 mg/m3). This IOELV will be used as the starting point for the development of inhalation and dermal DNELs for workers and the general population.

Regarding oral exposure, the NOAEL of 209 mg/kg bw/day will be used as the starting point for DNEL derivation.


Repeated dose toxicity: via oral route - systemic effects (target organ)digestive: liver; urogenital: kidneys  


Repeated dose toxicity: inhalation - systemic effects (target organ)digestive: liver; urogenital: kidneys

Justification for classification or non-classification

In accordance with EU Classification, Labelling and Packaging of Substances and Mixtures (CLP) Regulation (EC) No. 1272/2008, classification is not necessary for repeated dose toxicity based on the available data. The NOAEC of 800 ppm (2856 mg/m3) / 6 hr/day from the inhalation key study (and the NOAEC of 400 ppm (1428 mg/m3) / 6 hr/day from the chronic inhalation study), the oral NOAEL of 209 mg/kg bw/day (exposure via drinking water) and the oral NOAEL of 300 mg/kg bw/day (exposure by gavage) are all above the cut-off values that trigger STOT-repeated exposure classification.