Registration Dossier

Administrative data

Key value for chemical safety assessment

Additional information

Mutagenicity in bacterial strains Data on 1,2,3 - propanetriol, glycidyl ethers (GE-100): 1,2,3-propanetriole, glycidyl ethers was investigated using the Salmonella/microsome test for point mutagenic effects in doses up to 5000 µg per plate on five Salmonella typhimurium LT2 mutants (TA98, TA 100, TA 1535, TA 1537 and TA 1538; Banduhn, 1986). The study was performed according to the OECD Guideline 471and EU Method B13/14 with deviations (five Salmonella strains tested) and considered to be of the highest quality (reliability Klimisch 1). The strains have the following genotypes: S. typhimurium TA 1535 his G46 rfa- uvrB-, S. typhimurium TA 1537 his C3076 rfa- uvrB-, S. typhimurium TA 1533 his D3052 rfa- uvrB-, S. typhimurium TA 98 his D3052 rfa- uvrB- R + and S. typhimurium TA 100 his G46 rfa- uvrB- R +. The bacteria were treated with the test material using the Ames plate incorporation method at up to eight dose levels (1.58, 5, 15.8, 50, 158, 500, 1580 and 5000 µg per plate), in triplicate, both with and without the addition of a rat liver homogenate metabolising system (cofactor-supplemented post-mitochondrial fraction prepared from the livers of rats treated with the enzyme-inducing agent Aroclor 1254.). Results for the negative controls (spontaneous mutation rates) were considered to be acceptable. The vehicle (water) control plates gave counts of revertant colonies within the normal range. The positive controls  methyl methanesulfonate  (MMS), 9-Aminoacridine (9AMA), 2-nitrofluorene  (2 -NF), N-ethyl-N'-nitro-N-nitrosoguanidine(ENNG), and benzo(a)pyrene (BaP) had a marked mutagenic effect, as was seen by a biologically relevant increase of induced revertant colonies compared to the corresponding negative controls. So all of the positive control chemicals used in the test induced marked increases in the frequency of revertant colonies, both with and without metabolic activation. Thus, the sensitivity of the assay and the efficacy of the S9-mix were validated. No cytotoxic effect of the test article was observed. In the experiments, a dose-dependent elevation in the number of revertant colonies was observed with the strains TA 98, TA 100 and TA 1535 in the absence as well as in the presence of S-9 mix. The revertant rates were enhanced up to a 2-3-fold increase versus concurrent controls in TA 93, up to 14-fold in TA 100 and up to about 30-fold in TA 1535. No clear-cut enhancement of revertant rates was observed in TA 1537 and in TA 1533, except in the second experiment with TA 1538 in the absence of S-9 mix, where a slight increase up to a doubling was observed at the highest concentration. A mutagenic activity of the test article was ascertained, starting at approx. 153 µg/µL in the experiments without metabolic activation and at approx. 50 µg/µL with activation. These findings were confirmed in the second, independent experiment. In conclusion, it can be stated that during the described mutagenicity test and under the experimental conditions reported, the test article, GE 100, showed a reproducible dose-dependent mutagenic activity. So during the described mutagenicity test and under the experimental conditions reported, the test article induced point mutations by base-pair changes and frameshifts in the genome of the strains used. Therefore, 1,2,3-propanetriol, glycidyl ethers is considered to be mutagenic in this Salmonella typhimurium reverse mutation assay.

Mutagenic activity of the target substance(1,2,3-propanetriol, glycidyl ethers) was examined with the DNA repair test (solid and liquid) and the reversion test (spot and soft agar methods) using the bacterial strains of E. coli (WP2, WP2uvrA, CM571 and WP100, these strains are nearly isogenic and have a tryptophan-deficiency that is suppressible by ochre suppressor mutation) and S. typhimurium (TA98 and TA100), (Ohtani and Nishioka, 1980). The study was performed similar to the OECD Guideline with deviations (different strains and no metabolic activation system was used) and considered to be of high quality (reliability Klimisch 2). For the DNA repair tests, solid method and liquid method, four strains of E. coli and two strains (WP2 and WP100) were used, respectively. For the reversion test, E. coli WP2uvrA and S. typhimurium TA98 and TA100 were used. It has been elucidated that TA98 is mutated by frameshift type mutagens and the other two strains by base-change type mutagens. The results of the DNA repair test of solid method showed that GE-100 (10 % in DMSO) caused a positive effect. The inhibiting zones in the DNA repair test of solid method for the positive samples were always greater in CM571 and WP100 which are deficient in recombination repair than in WP2 and WP2uvrA. This suggests that the epoxide resins caused DNA damage which can be repaired by the process of recombination. The results of the DNA repair test of liquid method (strains tested: E.coli WP100 and E.coli WP2) showed that GE-100 had a remarkable value of DIG50 (differential inhibition for 50% growth = log (chemical concentration resulting in 50% growth in WP2/chemical concentration resulting in 50% growth in WP100): 1.37 and it was almost comparable to positive controls, AF-2 and MMS.

From the results, it is possible to estimate that DNA damaging capacity of these epoxide resins could be similar to the positive controls 2-acetylaminofluorene (AF-2) and methylmethanesulfonate (MMS). The results of the reversion test with the strains of E. coli and S. typyimurium for the samples of epoxide compounds show that all those (data not shown here) which were positive in the DNA repair test induce revertants in E. coli WP2uvrA and in S. typhimurium TA100 but not in TA98. This suggests that these epoxide compounds induce the mutation of base-pair substitution type without metabolic activation. Those which showed no killing effect and no DNA damaging capacity were negative in the reversion test. In the experiment with the commercial adhesive agents, some of them (data not shown) were also mutagenic in TA100 with or without a hardening agent. These results obtained suggest that epoxide resins produce DNA damage which can be repaired by the process of recombination and induce mutation of base-pair substitution type without metabolic activation. Epoxide compounds which have a higher molecular weight and lower solubility showed neither any killing effect nor mutagenic effect in the DNA repair test as well as the reversion test. The reason seems to be that their molecular sizes and solubilities are not small and high enough to pass through the cellular membrane and to reach DNA.

In summary,the results indicate that some epoxide compounds which have relatively lower molecular weight induce mutation. It applies to the target substance GE-100. It is suggested that the mutagenicity of epoxide resins may be influenced by their solubilities and / or transportation through cellular membrane.

Data on Polyglycidyl Ether of Substituted Glycerin (EPON 562)

The "Patty´s Industial Hygiene and Toxicology" contains information about Polyglycidyl ether of substituted glycerine (EPON 562) (Hine et al., 1981). It is mentioned, that the substance, tested in S. typhimurium TA 98, Ta 100, TA 1531 and TA 1533, at concentrations of 50 to 1000 µg per plate, caused a positive results in TA 98 and 100 but negative at 50 µg; and negative results in TA 1531 and 1533.

Mutagenicity and Chromosome Aberration in mammalian cells

Data on read-across substances

The target chemical was profiled as "Epoxides" by the "US EPA New Chemical Categories" (OECD QSAR Toolbox, v3.1, 2013). Common properties of epoxides are high reactivity, cytotoxicity, and high probability of mutagenic potential and/or carcinogenicity. Therefore chemicals with the same profiling result have been retrieved from the database. The chemicals containing other chemical elements in their structure and/or other organic functional groups were considered dissimilar to the target chemical and have been removed from the domain. The target chemical is obtained by the reaction of epichlorohydrin with glycerol. Therefore, epichlorohydrin is considered to be a suitable candidate for read-across. Glycidol is the simplest representative of glycidyl ethers category (HPV, Epoxy Resin Systems Task Group (ERSTG), 2001). The other category members possess epoxy moieties in their structures and their profiling results regarding the ability to bind to proteins and to DNA (property which is likely responsible for genetic toxicity) are similar to those of the target chemical. Therefore, they considered to be suitable for read-across. Three chemicals have been removed from the domain as they are too lipophilic compared to the target chemical. The target chemical is predicted to be positive in Mouse lymphoma cells, in Chinese hamster Lung (CHL) cells as well as in in vivo Micronucleus Test.

In a mammalian gene mutation assay (Mouse Lymphoma Assay, similar to OECD 476), L5178Y cell cultures were exposed to the read-across substances glycidol and butyl glycidyl ether at concentrations of 8, 15, 23, 30, 45, 60, 75, 94, 125, 187, 250 µg/mL (glycidol) or 84, 100, 130, 164, 200, 256, 300, 320, 400, 500, 640, 800 µg/mL (butyl glycidyl ether) with and without metabolic activation (Thompson et al., 1981). The metabolic activation system was either liver homogenates prepared from Aroclor-1254-induced Sprague-Dawley rats or rats injected with corn oil (non-induced). Glycidol and butyl glycidyl ether were both tested up to cytotoxic concentrations, i.e. the highest dose was set at twice the level that killed 50% of the organisms in the toxicity assay. Positive controls (ethyl methanesulfonate, 620 µg/mL, –S9; 2-acetylaminofluorene, 100 µg/mL, + induced S9; dimethylnitrosamine, 74 µg/mL, + uninduced S9) induced the appropriate responses. For both glycidol and butyl glycidyl ether there was a concentration-related positive response as well as the stipulated at least three-fold increase of the mutation frequency over background without or with both available metabolic activation systems. Thus, it can be concluded that 1,2,3-propanetriol, glycidyl ethers possesses mutagenic activity in the Mouse Lymphoma Assay, too. This study is classified as acceptable, reliable with restrictions and satisfies the requirements for OECD Guideline 476 for in vitro mammalian cytogenicity data.

Other genotoxicity data in mammalian cells

In an unscheduled DNA synthesis assay similar to OECD guideline 482, WI38 cells were exposed to the read-across substances glycidol and butyl glycidyl ether at concentrations of 0, 0.375, 0.75, 1.5, 3.0, 6.0 µg/mL (glycidol, without S9); 0, 0.037, 0.111, 0.333, 1.0, 3.0 µg/mL (glycidol, with S9); 0, 0.24, 0.36, 0.53, 0.8, 1.2 µg/mL (butyl glycidyl ether, without S9); 0, 0.5, 1.0, 2.0, 4.0, 8.0 µg/mL (butyl glycidyl ether, with S9) (Thompson et al., 1981). Glycidol and butyl glycidyl ether were tested just below the level which produced cytotoxicity. The positive controls (4-nitroquinoline-N-oxide (4NQO), dimethylnitrosamine (DMN)), induced the appropriate response.

There was no evidence or a dose related positive response for both compounds without metabolic activation that unscheduled DNA synthesis, as determined by radioactive tracer procedures, was induced. Glycidol induced a dose-related positive response with metabolic activation, butyl glycidyl ether induced demonstrable, although not considered positive responses with metabolic activation. The study was classified as reliable with restrictions (Klimisch 2) and satisfies the requirements for OECD guideline 482 for other genotoxicity data.


Justification for selection of genetic toxicity endpoint
No study is selected since all available studies are positive.

Short description of key information:
1) Banduhn, 1986 - Salmonella typhirium - Ames test - TA98, 100, 1535 - positive, 1537 and 1538 - unclear results ;
2) Pattys, 1981 - Salmonella typhimurium - TA 98 and Ta 100 - positive, TA 1531 and 1533 - negative;
3) Ohtani and Nishioka, 1980 - Salmonella typhimurium and E. coli - positive;
4) Thompson et al., 1981. - Mouse Lymphoma - positive; UDS Assay -positive (with metabolic activation); negative (without metabolic activation);
5) Toolbox predictions - are all positive (in vitro and in vivo).

Endpoint Conclusion: Adverse effect observed (positive)

Justification for classification or non-classification

The target substance showed a reproducible dose-dependent mutagenic activity in Salmonella typhimurium reverse mutation assay (Banduhn, 1986). The chemical was predicted positive in Mouse lymphoma cells, in Chinese hamster Lung (CHL) cells as well as in in vivo Micronucleus Test (the OECD QSAR Toolbox v3.1). By this modelling tool, the epoxy ring was identified as a structural alert reacting with nucleophilic sites of DNA by SN2 mechanism and causing mutations.

There are numerous studies available publically for a variety of structurally similar epoxy compounds. The chemicals containing the same glycidyloxy moieties in their structures are almost all positive in a variety of genetic toxicity studies in vitro and in vivo. For instance, the read-across substances glycidol and butyl glycidyl ether showed a concentration-related positive response as well as the stipulated at least three-fold increase of the mutation frequency over background without or with metabolic activation in the Mouse Lymphoma Assay (Thompson et al., 1981). Ohtani et al. (1981) pointed to a trend in the genetic toxicity potency of epoxy compounds depending on their molecular weights and solubilities. The epoxy resins with large molecular size and low water solubility showed neither killing effect nor mutagenic effect in the DNA repair test as well as the reversion test in bacterial test system (Ohtani et al., 1981). Nevertheless, low molecular epoxides including GE-100 (the target substance) were positive (Ohtani et al., 1981). This assumption was confirmed in a lot of studies conducted with alkyl glycidyl ethers containing hydrocarbon chains of different lengths. The ethers with C-4 hydrocarbon side chain showed a definite response while the C-8 or higher ethers showed very weak or no responses in the Ames Test, Mouse Lymphoma Assay or UDS Test (Thompson et al., 1981).

Based on the extrapolation from glycidol, epichlorohydrin, allyl-, butyl-, isopropyl-, phenyl glycidyl ethers as well as diglycidyl ether and alkyl glycidyl ethers (category created for the read-across, see statement attached to this file) and taking into account molecule size, side chain length(s) and lipophilicity, it can be concluded that mutagenic activity of 1,2,3-propanetriol, glycidyl ethers cannot be ruled out. Therefore classification and labelling are warranted according to the criteria of the European regulation (EC) No. 1272/2008:

Mutagenicity, Cat 2, H 341 (suspected of causing genetic defects)