Magnesium methanolate

Endpoint summary

Key value for chemical safety assessment

Effects on fertility

Additional information

Magnesium methanolate rapidly hydrolyzes in aqueous environments. Toxicity is mediated by its degradation products MeOH and Mg(OH)2 and assessed for these products.

MeOH No impairment of fertility and reproductive performance was found in male and female rats (parent and daughter generations) exposed to methanol (NEDO, 1987).

In a two-generation reproduction study, rats were exposed to methanol by inhalation for 19-20 hours/day (NEDO, 1987). No treatment-related alterations in general observations and reproductive parameters were found. None of the fertility indices, including sexual cycle, days needed for insemination, insemination rate and pregnancy rate showed statistically significant differences. There were no differences for body weight, food consumption and water consumption during gestation and lactation period, either. No abnormalities were observed in findings on delivery and nursing behaviour and necropsy data of F0 animals. No firm conclusions can be drawn about fertility of either sex, as the copulation time of 21 days was comfortably long for successful insemination and gametogenesis was not considered. In the F1 and F2 progeny (both sexes), no histological changes and no effects on testes or ovaries were reported. However, a decrease in brain weights was evident at 1.3 mg/L methanol, but without noticeable histological changes and functional impairments. This phenomenon is believed to represent a change occurring during the prenatal period (Takeda and Katoh, 1988). However, no quantitative data and statistical level were documented for organ weights. The meaning of an apparent shift of testis descent in male offspring in relation to body weight development of the pups in the two following generations is unclear and was not directly addressed by Takeda and Katoh (1988), but detailed by NEDO (1987) and considered a significant difference from untreated controls. Furthermore, it is obvious that this parameter showed considerable variation also between the control groups of both generations.

In a one-generation reproduction study in monkeys (Burbacher et al., 1999), adult female monkeys were exposed to methanol vapour (2.5 hours/day; 0, 200, 600, 1800 ppm) during prebreeding, breeding and pregnancy. No signs of overt maternal toxicity were noted during the study in any of the dose groups. Methanol exposure had no effects on the tested reproductive performance, including menstrual cycles, conception rate, and live-birth delivery rate. However, all methanol-exposed animals had a decrease of about 6 to 8 days in duration of pregnancy compared to control animals. It is not clear whether this decrease in duration of pregnancy was related to methanol exposure, since there was no dose-response and no differences among offspring groups in body weight, size or other physical parameters (head size, crown rump length). Moreover, the duration of pregnancy was within the reported normal range for this species (NTP, 2003). Prenatal exposure to methanol had no effect on infant growth and physical development for the first 9 months. However, results of infant assessments during the first 9 months of life were confounded by the normal variance and the low number of animals. The NOAEC for reproductive effects can be determined to be at the highest concentration tested of 2.39 mg/L (1800 ppm).

In another study, which investigated reproductive effects there was an insignificant increase in morphological anomalies in spermatozoa in male mice at 1000 mg/kg bw/day after oral dosing for five weeks (Ward et al., 1984).

It is also pointed out that in repeated dose toxicity studies, ovarian and testicular tissues have not been adversely affected by methanol administration.

Mg(OH)2

In a combined repeated dose reproductive toxicity screening tests according to OECD No. 422 and GLP no effects on male and female fertility were observed up to the highest dose tested of 1000 mg/kg bw/day. In addition, the absence of effects of magnesium ions on the reproductive organs of males and females was also reported in 90-day dietary studies in rats and mice with magnesium chloride hexahydrate up to the highest dose levels tested (11400 mg/kg bw for males and 13830 mg/kg bw for females in mice, corresponding dose of magnesium hydroxide to 1856 mg/kg bw/day and 2251 mg/kg bw/day; 1600 mg/kg bw of test substance for male rats and 1531 mg/kg test substance for female rats corresponding to a Mg(OH)₂dose of 261 mg/kg bw/day and 249 mg/kg bw/day for males and females respectively), and in a 92-week dietary carcinogenicity study with magnesium chloride hexahydrate up to the highest dose of 2810 mg/kg bw per day in males and 3930 mg/kg bw per day in females, corresponding to 457 and 640 mg magnesium hydroxide/kg bw/day. It can be concluded that the weight of evidence of existing studies indicates that magnesium hydroxide is unlikely to exert effects on male or female fertility at dose levels not causing other systemic effects. Further testing for toxicity to reproduction is therefore not proposed, also taking into consideration animal protection aspects.

Effects on developmental toxicity

Description of key information

Magnesium methanolate rapidly hydrolyzes in aqueous environments. Toxicity is mediated by its degradation products MeOH and Mg(OH)2 and assessed for these products.

MeOH

NOAEC (maternal toxicity) = 1.3 mg/L for rats
NOAEC (teratogenicity = 1.3 mg/L for rats
NOAEC (maternal toxicity) = 2.39 mg/L for monkeys
NOAEC (teratogenicity) = 2.39 mg/L for monkeys
Negative for spermatozoa morphological anomalies: NOAEL (oral) = 1000 mg/kg bw/day

Mg(OH)2

No developmental toxicity was observed in a repeated dose reproduction screening test up to the highest dose of 1000 mg/kg bw/day. A developmental toxicity study with magnesium chloride at dose levels that just did not lead to maternal toxicity (as concluded from a dose range finding study) did not show any developmental toxicity induced by the test substance

. A limited study with magnesium sulphate after i.v. administration to rats corroborates the absence of developmental effects. Taken together, the available evidence suggests that magnesium hydroxide is unlikely to be a developmental toxicant and no further studies are warranted for this endpoint.

Additional information

Magnesium methanolate rapidly hydrolyzes in aqueous environments. Toxicity is mediated by its degradation products MeOH and Mg(OH)2 and assessed for these products.

MeOH

A. Animal data

Developmental toxicity has been observed in many rodent studies, which resulted in a variety of effects in offspring due to prenatal and/or postnatal dosing.

In a developmental study, rats were exposed to 270, 1330 and 6650 mg/m³ methanol by whole-body inhalation from gestation days 7 through 17 for 23 hours/day (NEDO, 1987). In the top dose, maternal toxicity was recorded. In the progeny, there was fetal malformation, increased perinatal mortality and developmental delay. Teratogenic effects occurred only at the maternally toxic exposure concentration. Exposure levels of 1.33 mg/L or less did neither induce toxic symptoms in maternal animals nor structural abnormalities or delay in growth or functional development in the F1-generation. Therefore, the NOAEC for maternal and developmental toxicity is considered to be 1.33 mg/L.

In a second whole-body inhalation developmental study, rats were exposed in chambers from gestation days 1 through 19 at 6650 and 13300 mg/m³ and from gestation days 7 through 15 at 26.6 mg/L for 7 hours/day (Nelson et al., 1985). In the high dose group, significantly reduced food consumption without adverse effect on body weight gain was noted in maternal animals. No signs of maternal toxicity were observed in the lower dose groups. No influence on the number of corpora lutea and implantations was reported. No effects on fetal lethality and resorption were found. There was no evidence of embryotoxic/teratogenic activity of methanol at 6.65 mg/L. At the highest concentration, an increased number of litters with skeletal and visceral malformations was noted. These included in particular rudimentary and extra cervical ribs, exencephaly and encephalocele, and, to minor extent, cardiovascular and urinary-tract defects. In this study a NOAEC for maternal and developmental toxicity of 6.65 mg/L was obtained.

In a developmental whole-body inhalation study, mice were exposed to methanol (1330, 2660, 6650, 9970, 13300, 19940 mg/m³) on gestation days 6 to 15 for 7 hours/day. Additionally, an orally exposed group was included for comparison (Rogers et al., 1991, 1993). There were no signs of maternal toxicity. Developmental effects occurred after inhalation of 2660 mg/m³ (Rogers et al., 1991, 1993). The dose related increase in cervical ribs or ossification sites lateral to the seventh cervical vertebra was significant at 2.66 mg/L. Significant increases in the incidence of exencephaly and cleft palate were observed at 6650 mg/m³. At the highest dose, almost complete resorption of embryos in most litters occurred. Reduced fetal weight was noted at 13300 mg/m³ and above. In this study, NOAECs for maternal and developmental toxicity were 19940 and 1330 mg/m³, respectively.

A study employing a single intraperitoneal injection of mice on gestation day 7 resulted in craniofacial malformations and malformations of the holoprosencephaly spectrum at 1700 mg/kg bw (LOAEL), a NOAEL could not be identified (Fu et al., 1995).

In another study of Fu et al. (1996), pregnant female mice were orally administered 5000 mg/kg methanol on gestation days 6 to 10. The influence of folate in the diet on the incidence of developmental defects in the offspring was investigated. Methanol had no marked influence on maternal folate levels, irrespective of folate supplementation, except in maternal plasma where there was some evidence of a reduction of about 20 %. Methanol treatment was slighty fetotoxic (reduced mean fetal weight and reduced mean crown-rump length), but had no impact on other reproductive parameters. There was some evidence of a teratogenic effect (increased incidences of cleft palate and exencephaly) under folate-adequate supply, but this was hardly statistically significant: cleft palate (2/222 vs. 0/282) and exencephaly (5/222 vs. 1/282 in the respective high-folate control). Likewise, folate deficiency failed to produce significant malformations (in accordance with Heid et al., 1992): cleft palate (5/215 vs. 0/282) and exencephaly (2/215 vs. 1/282 in the respective high-folate control). However, cleft palate, but not exencephaly was significantly increased in the presence of methanol: 39/235 vs. 5/215 and 8/235 vs. 2/215 of folate-poor control, respectively.

In pregnant CD-1 mice given methanol (5000 mg/kg bw/day) from gestation days 6 to 15, one group receiving folate-deficient and the other folate-supplemented diet, there were no pronounced differences in the formate blood levels between both groups: 5.13 ± 0.68 mmol/L (folate-def.) and 3.90 ± 0.94 mmol/L (folate-suppl.) vs. 0.36 ± 0.13 mmol/L (untreated control). Developmental toxicity, however, was significantly higher in folate-deficient dams (Hong et al., 1997). On balance, the results indicate that developmental toxicity of methanol in mice on low dietary folate is not linked to increased formic acid levels (Hong et al., 1997), however, folate deficiency may enhance the teratogenic effects of methanol in mice.

All these rodent studies on the endpoint developmental toxicity, though of scientific interest, are of limited relevance for a classification to humans (see also attachment), since methanol follows a different kinetic pattern in humans (see also chapter 7.1) and causes severe acute toxicity at rather low doses. In contrast, rodents do not respond to methanol in the same sense. Rodents tolerate high methanol doses without signs of toxicity and high doses are also required in order to achieve developmental effects. The experimental species that appears to be much closer to humans in terms of kinetics and acute toxicity is the rabbit. In terms of developmental toxicity to rabbits, so far, only screening experiments are available, which have confirmed that rabbits are apparrently much less sensitive than rats or mice for developmental effects: rabbits received on gestation day 7 or 8 a dose of 4000 mg/kg bw (in two portions; Sweeting et al., 2010; Sweeting and Wells, 2015; DeSesso et al., 2015) via i.p. injection and there was no clear evidence of a developmental effect.

The developmental effects of methanol in rats and mice occur at dose levels at which also other alcohols (e.g. Propanol) are known to cause developmental effects in rats and mice. Methanol has a specifically high toxicity in humans (which is quite exceptional, different from other alcohols and not displayed in rodents) and it is assumed that the classification of methanol as acutely toxic upon all exposure routes would practically preclude an uptake of such doses except for cases of accidents or fraudulent misuse.

On balance, considering the apparent absence of developmental effects of methanol in rabbits and in monkeys (see C., below) and its pronounced acute toxicity in humans, which practically precludes an uptake of doses known to be required for developmental effects in rodents, methanol has not to be classified for developmental toxicity in humans. Overall, the rodent data are not sufficient to presume similar effects in humans.

 

B. Human data

There are no relevant epidemiological studies or case reports which describe an increase in the incidence of malformations in children of mothers exposed to methanol during pregnancy. The limited data available on methanol exposure on reproductive and developmental effects do not show any association (NTP, 2003).

In an epidemiological study, the reproductive effects of various occupations and associated exposures to complex mixtures were examined in women who gave birth to infants with and without cleft lip or cleft palate (Lorente et al., 2000). No association was found between methanol exposure and oral clefts. The small number of subjects exposed to methanol, the lack of individual exposure data, and confounding factors by other chemical exposures did not allow to draw firm conclusions as to the role of methanol on these outcomes.

NTP (2003) reviewed several studies that investigated the association between the periconceptional use of multivitamins containing folic acid and birth defects (e. g. neural tube defects and orafacial clefts). These studies suggest that folate deficiency in humans may lead to greater susceptibility to such effects. However, in all of these reviewed studies, the association between methanol and these effects was not directly investigated. Therefore, no conclusion can be drawn regarding causality between methanol and birth defects based on human data.

NTP (2003) stated that the rodent data on reproductive and developmental toxicity are of relevance for an assessment of the situation in humans even in the light of the known differences in methanol metabolism between rodents and humans. Rodents are adequate models for human exposure as long as formate levels do not accumulate. However, the blood methanol concentrations associated with serious teratogenic effects and reproductive toxicity are in a dose range that is associated with formate accumulation, metabolic acidosis and visual and clinical effects in humans (NTP, 2003) and are far above the European OEL (260 mg/m³).

In humans, transient central nervous system effects generally appear at blood methanol levels higher than 200 mg/L, ocular symptoms appear at blood levels of > 500 mg/L and fatalities haven often occurred in untreated patients with initial blood methanol concentrations in the range of 1500 – 2000 mg/L (see also acute toxicity). Other effects (e.g. marginal, not yet definitive neurological effects observed in primates) may be exhibited at lower inhalation doses and lower methanol blood levels.

C. Overall assessment with special regard to a reproduction toxicity study in nonhuman primates (Burbacher et al., 1999)

C.1 The significance of the study by Burbacher et al. (1999)

In the study by Burbacher et al. (1999; published in a scientific journal in 2004), the health effects of whole body inhalation exposure to methanol for 2.5 hours/day on 7 days/week during prebreeding, breeding and pregnancy (approximately 120, 70, 165 days, respectively) on 11-12 female monkeys (Macaca fascicularis) per dose group and on their offspring (exposed in utero) at 0, 200, 600 and 1800 ppm (0, 0.27, 0.8, 2.39 mg/L) were investigated. The major strenght of this study and its significance for the assessment of the reproductive toxicity of methanol is based on the use of an animal model with high relevance to humans, the inclusion of toxicokinetics evaluations which confirmed established species differences (rodents vs. rabbits, monkeys and humans), the selection of exposure levels known to result in blood concentrations just below that causing nonlinear clearance kinetics in primates, and the conduct of repeated follow-up assessments (developmental/neurobehavioural) of the infants during their first 9 months of life.

C.2 Synopsis of the study by Burbacher et al. (1999)

C.2.1 Blood levels (Toxicokinetics, Maternal health effects)

Exposure to methanol vapours did not affect the health of the adult monkeys prior to or during pregnancy. Daily single 2.5-hour exposures to methanol vapours caused short term elevations in blood methanol concentrations of approx. 1- to 2-fold in the 200 ppm exposure group, 3- to 4-fold in the 600 ppm group and 13- to 16-fold in the 1800 ppm group. The more-than-proportionate increase observed in the 1800 ppm exposure group was accompanied by the appearance of nonlinear elimination kinetics at 1800 ppm, a finding that most likely reflects saturation of methanol metabolism (presumably by hepatic alcohol dehydrogenase). After long-term exposure, peak blood methanol concentrations declined slightly over the first month and remained constant thereafter. The concentration of plasma formate remained at baseline levels during the entire course of the study in all exposure groups. Pregnancy had no effect on methanol disposition. Serum folate concentrations were not affected by pregnancy and methanol exposure. At 1800 ppm, after 5 hour elimination, the residual methanol level was near baseline (max. 2-fold higher). The mean estimated elimination half-lives (for 600 and 1800 ppm) ranged between about 60 to 90 minutes.

 

C.2.2 Toxicity to reproduction (Reproductive (fertility) effects, Developmental effects/Offspring assessments)

Methanol exposure had no effect on most measures of reproductive performance, including menstrual cycles, conception rate, and live-birth delivery rate. However, all methanol-exposed animals had a decrease of about 6 to 8 days in duration of pregnancy compared to control animals. It is not clear whether this decrease was related to methanol exposure, as there was no dose response and no difference among offspring groups in body weight, size or other physical parameters (head size, crown rump length). However, the duration of pregnancy was within the reported normal range for this species (NTP, 2003). Prenatal exposure to methanol had no effect on infant growth and physical development for the first 9 months. The results of behavioural assessments of offspring did not indicate methanol exposure effects on most domains of early behavioural development. No consistent effects due to methanol exposure were observed on early reflex responses, gross motor development, spatial and concept learning and memory, and social behaviour. Methanol exposure was associated with ratings of "low arousal" on the Neonatal Behavioural Scale. The effects were observed when all of the methanol-exposed infants were compared with controls. Further comparisons, however did not indicate that the effect was dose-dependent. In addition, many of the methanol-exposed infants that were the last to receive optimal scores for the arousal items had been delivered via C-section. Thus, this effect may not be directly related to methanol exposure independent of mode of delivery. Methanol exposure was also associated with a delay in early sensorimotor development of male infants only (Visually Directed Reaching Test). The effect was noted after controlling for the shortened gestation length observed for the 3 methanol-exposed groups. The delay was dose-dependent and ranged from 9 days for the 200 ppm exposure group to over 2 weeks for the 600 and 1800 ppm exposure groups. This observation is based on results from group sizes of 8 to 9 infant animals, with 2 to 5 males and 4 to 7 females per group. The statistical significances (linear contrast test based on ANOVA) of p = or < 0.04: This test indicated a significant difference between control infants compared to all methanol-exposed offsprings combined, as well as with the 600 and 1800 ppm groups for males (p < 0.04). The comparison between the control group and the 200 ppm group was nearly significant (p=0.09).

The results of the Fagan-Test of Infant Intelligence indicated a possible effect of methanol exposure on visual recognition memory when testing complex stimuli (social problems). Although there were no mean group differences in the novelty scores across the 4 exposure groups, only the control group exhibited a significant novelty preference for social stimuli.

 

C.2.3 Other observations (Offspring assessments)

Prenatal methanol exposure was associated with the occurrence of a wasting syndrome in 2/7 female offsprings in the 1800 ppm exposure group after approximately 1 year of age, both living in the same cohort. Those females began to show growth retardation with 12 resp. 17 months of age. They became very weak and had to be euthanized at the age of 20 resp. 36 months. Assay results for viral infection, blood chemistry, complete blood count, and liver, kidney, thyroid, and pancreatic function were within the normal range. Necropsies showed signs of severe malnutrition and gastroenteritis. This symptom was not observed in any of the offsprings of the other cohort.

 

C.2.4 No observed adverse effect concentrations (NOAEC)

 Based on the study results, the NOAEC for maternal toxicity/reproductive performance and the NOAEC for fetotoxicity/teratogenicity were considered both to be 1800 ppm (2.39 mg/L). Further data would be required in order to derive a NOAEC for offspring developmental effects (neurobehavioural development). No NOAEC was derived by the study authors.

C.3 Effective assessments of the results of the study by Burbacher et al. (1999)

Considering the assessments below, there is general agreement that no clear robust methanol-related effects on reproduction and/or development were identified up to and including the highest exposure concentration of 1800 ppm in exposed female monkeys and their offspring (exposed in utero).

HEI Statement, in Burbacher et al. (1999)), peer-reviewed conclusion: "The investigators' findings suggest that repeated inhalation exposure to methanol vapours as high as 1800 ppm would not result in accumulation of blood formate above baseline levels. With the exception of an unexpected shortening of gestation, methanol exposure had no effect on reproductive performance. The most significant result to emerge from this study was the wasting observed in two monkeys exposed in utero to 1800 ppm methanol. Although this observation raises concern for prenatal exposures of this magnitude, pregnant women are unlikely to be exposed to such extremely high concentrations of methanol for prolonged periods of time. Overall, the results provide no evidence of a robust effect of prenatal methanol exposure on the neurobehavioural development of nonhuman primate infants during the first nine months of life (...)".

With regard to the study by Burbacher et al. (1999), shorter gestation lengths, higher incidences of Caesarean (C) sections and neurobehavioural findings noted in methanol-exposed monkeys as compared with controls are not considered to represent toxicologically relevant methanol-related effects for the following reasons: Although there was a statistically significant shorter gestation length in methanol-treated groups as compared to the control group (not dose-dependent), gestation lengths for the treatment groups were consistently within the normal range for this species (Chellman et al., 1999). While five C-sections were performed in the treatment groups (2 females at 200 ppm, 2 females at 600 ppm, 1 female at 1800 ppm), there were no C-sections in the control animals. These surgical procedures were performed in response to signs of possible difficulty in maintenance of the pregnancy (e.g. vaginal bleeding) and thus suggest late reproductive dysfunction in the exposed females. However, in the NTP-CERHR Expert Panel review (NTP, 2004), it was noted that there does not appear to be sufficient evidence to support the claim of increased pregnancy complications following methanol exposure, and that the C-sections may not have been necessary. As vaginal bleeding sometimes occurs in macaques 1 to 4 days prior to delivery of a healthy infant, an ultrasound examination should have been performed prior to C-section conduct in order to confirm fetal or placental problems. The reported neurobehavioural findings are not considered as robust (e.g. NTP, 2004). Overall, most neurobehavioural parameters assessed were not affected by the treatment with methanol and there was no strong evidence of a dose-response relationship. Furthermore, there was a lack of proper controls for multiple comparisons in the statistical analyses; pair-wise tests were conducted, even when the overall ANOVA was not significant.

RAC (2014): "An overall assessment of the monkey studies indicated that methanol may have affected the infants, but that the data were not very robust and clearly not sufficient for classification.

Furthermore, there were minimal similarities between the very clear effects noted in rodents and those possibly observed in the monkeys. It is acknowledged that the monkey exposure levels (≤1800 ppm) and exposure time per day (2.5 hours in monkey vs 7 hours in mice), were lower than the LOAEC of 2000 ppm in mice, and the blood methanol concentration was 35 mg/L at the top dose in monkeys when compared to 537 mg/L in mice at the LOAEC. Therefore, developmental toxicity also in monkeys at higher exposure levels cannot be ruled out. The RAC concludes that there is robust evidence of developmental toxicity of methanol in rodents, but very limited indications of developmental toxicity from non-rodent species which have metabolic pathways more similar to humans."

Referring to the assessment by RAC (2014), it has to be recognized that the effects noted in mice at the LOAEC of 2000 ppm (corresponding to a plasma methanol concentration of 537 mg/L; Rogers et al., 1993) were confined to increases in cervical ribs or ossification sites lateral to the seventh cervical vertebra, i.e. to findings (variations, retardations) raising a low level of concern. Only at 5000 ppm (plasma methanol concentration of 1650 mg/L) and above, i.e. at concentrations clearly lethal to humans, an increase in malformations such as exencephaly and cleft palate was noted. Similarly, relevant developmental (teratogenic) effects in rats were only evident at very high exposure levels. In addition, comparing exposure durations, in the study by Burbacher et al. (1999) the actual exposure duration for the monkeys was mostly considerably longer than 2.5 hours, particularly at the high dose, as indicated by toxicokinetics data (continuing exposure longer per day is not appropriate in monkey as it can elicit stress). Comparisons in exposure durations however have to account for interspecies differences in organogenesis, critical periods of development and overall length of gestation. When such considerations are made, then the monkey study can be regarded as comparable to (and far more comprehensive than) the mouse study in terms of exposures relevant for developmental toxicity (teratogenicity) evaluations. Specifically in the monkey study by Burbacher et al. (1999), exposure started prior to conception and continued through all three trimesters of pregnancy.

C.4 The relevance of species differences for the assessment of developmental toxicity

Based on the available evidence, humans are not susceptible to the methanol-related developmental toxicity observed in rats and mice, due to the substantial species differences in metabolism. Different enzymes involved in methanol metabolism provide the basis for the kinetic differences between rodents and humans, which result in the formation of different metabolites at different metabolic rates. Moreover, it is clear from available animal data that, based on differences in metabolism and the formation/accumulation of formic acid in humans, which leads to maternal toxicity at much lower concentrations, the developmental effects observed in rats and mice in the absence of maternal toxicity are not relevant to humans. Methanol is presented in the ECHA Guidance on CLP (2015) as an example for not using rodent toxicity data to classify methanol for acute toxicity and specific organ toxicity on the basis of the non-relevance of rodent toxicity data to humans. This is due to species differences between humans and rodents, rendering the rodent data on methanol irrelevant to humans. The same approach should be applied for developmental toxicity assessment. Methanol has a uniquely high acute toxicity for humans with target organ toxicity for the ophthalmic nerve, which is different to toxicity seen in rodents. There is a large database on the toxicity of methanol and more recent data further support the previous EU decision not to classify methanol for developmental toxicity.

C.5 Comparison of the weight of evidence to the CLP requirements

For a detailed assessment, please refer to the attached document, which provides an argumentation why classification of methanol for developmental toxicity is not required according to the CLP criteria. Of note, the RAC (2014) presented recently an evaluation on the same issue and concluded that, based on the available information, there is not sufficient evidence to classify methanol for developmental toxicity in humans.

C.6 Conclusion

The weight of evidence does not indicate that a classification of methanol for developmental toxicity is required. Even though developmental effects at very high concentrations, exceeding those tested in the Burbacher study (1999), i.e. >1800 ppm, cannot be fully excluded for humans, blood concentrations similar to those observed in rodents at the LOAEC for developmental toxicity would result in severe acute toxicity or lethality in humans. This would require methanol vapour concentrations exceeding the Indicative Occupational Exposure Limit Value (IOELV) of 200 ppm by at least an order of magnitude. Please refer also to the attached document(s).

Mg(OH)2

In a combined repeated dose reproductive toxicity screening tests according to OECD TG 422 and GLP, no effects on embryo or fetal development parameters were observed up to the highest dose tested of 1000 mg/kg bw/day of magnesium hydroxide. No toxicologically relevant changes were noted in any of the parental parameters investigated (i. e. clinical appearance, functional observations, body weight, food consumption, clinical laboratory investigations, macroscopic examination, organ weights, and microscopic examination). Additionally, a developmental toxicity study was reported in the literature with the structural analogue magnesium chloride hexahydrate administered by oral gavage to pregnant Wistar rats. No maternal toxicity and no developmental toxicity was observed in this study. The NOAEL was greater than the highest dose tested of 800 mg/kg bw of test substance per day (corresponding to 130 mg/kg/day of magnesium hydroxide).

In an additional supporting study, the effects of magnesium sulphate on fetal rats after repeated intravenous administrations was investigated. The findings from these study showed that magnesium sulphate did not have any adverse effect on fetal body weight, size and brain development.

Experience with the therapeutic use of magnesium sulphate infusions as tocolytic agent in pregnant women showed no negative effects on newborn infants and their development at dose levels that do not lead to maternal effects and do not lead to a clinically significant imbalance of the Ca²⁺/Mg²⁺ratio.

Justification for classification or non-classification

Magnesium methanolate

Magnesium methanolate rapidly hydrolyzes in aqueous environments. Toxicity is mediated by its degradation products MeOH and Mg(OH)2 and assessed for these products.

Both hydrolysis products should not be classified for toxicity to reproduction. Based on the available information, magnesium methanolate thus does not have to be classified and has no obligatory labelling requirement for this endpoint.

MeOH

Conclusive in rodents, but not employed for classification to humans (see endpoint summary and attached documents).

Based on major species differences between humans and rodents (metabolic pathway/enzymes, mode of action, toxicokinetics), considering the overall weight of evidence, and in line with the evaluation of reproductive toxicity provided by the Committee for Risk Assessment (RAC, 2014), methanol does not appear to be toxic to reproduction. There is no need for classification.

Mg(OH)2

The weight of evidence from a repeated dose reproduction screening study carried out on magnesium hydroxide, which yielded a NOAEL of 1000 mg/kg bw/day, the results of several repeated dose studies of 90-days and longer with magnesium chloride and a developmental toxicity tests with magnesium chloride and sulphate, all indicating no effects on reproductive endpoints up to the highest doses tested plus human experience from medical uses suggests that no classification for any of the hazard categories for reproductive toxicity contained in Regulation (EC) No. 1272/2008 is warranted.

There is no need for classification.