2,2'-oxydiethanol

Endpoint summary

Key value for chemical safety assessment

Genetic toxicity in vitro

Description of key information

Data of a whole battery of robust high quality in vitro studies as well as an in vivo test show that the test item does not possess any mutagenic or genotoxic properties.
Diethylene glycol was not mutagenic in the bacterial reverse mutation test in the absence and the presence of metabolic activation. In addition, a CA and HGPRT assay were negative with and without metabolic activation. In addition, an in vivo micronucleus test with mouse according to the OECD TG 474 under GLP conditions was also negative.

Link to relevant study records

Reference

Reference 1

Endpoint:

in vitro gene mutation study in bacteria

Type of information:

experimental study

Adequacy of study:

key study

Study period:

27 Nov - 17 Dec 2012

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 471 (Bacterial Reverse Mutation Assay)

Version / remarks:

1997

Deviations:

no

Qualifier:

according to guideline

Guideline:

EU Method B.13/14 (Mutagenicity - Reverse Mutation Test Using Bacteria)

Version / remarks:

2008

Deviations:

no

Qualifier:

according to guideline

Guideline:

EPA OPPTS 870.5100 - Bacterial Reverse Mutation Test (August 1998)

Deviations:

no

GLP compliance:

yes

Type of assay:

bacterial reverse mutation assay

Specific details on test material used for the study:

- Name of test material (as cited in study report): Diethylene glycol
- Physical state: colorless clear liquid
- Analytical purity: 99.8 % (BASF SE analytical report 12L00357)

Target gene:

- His operon for S. typhimurium strains
- Trp operon for the E. coli strain

Species / strain / cell type:

S. typhimurium TA 1535, TA 1537, TA 98 and TA 100

Additional strain / cell type characteristics:

other: TA 98: rfa-, uvrB-, R-factor; TA 100: rfa-, uvrB-, R-factor; TA 1535: rfa-, uvrB-; TA 1537: rfa-, uvrB

Species / strain / cell type:

E. coli WP2 uvr A

Additional strain / cell type characteristics:

other: WP2: trp-; uvr A-

Metabolic activation:

with and without

Metabolic activation system:

S9 liver mix prepared from Wistar rats treated with 80 mg/kg bw phenobarbital i.p. and β-naphthoflavone orally, each on three consecutive days.

Test concentrations with justification for top dose:

First experiment (standard plate test, with and without metabolic activation, 3 plates/dose or control): 0, 33, 100, 333, 1000, 2500 and 5000 µg/plate
Second experiment (preincubation test with and without metabolic activation, 3 plates/dose or control): 0, 33, 100, 333, 1000, 2500 and 5000 µg/plate

Vehicle / solvent:

- Vehicle(s)/solvent(s) used: ultrapure water
- Justification for choice of solvent/vehicle: good solubility of the test item in the vehicle

Untreated negative controls:

yes

Remarks:

sterility control

Negative solvent / vehicle controls:

yes

Remarks:

ultrapure water

True negative controls:

no

Positive controls:

yes

Remarks:

with S9-mix

Positive control substance:

4-nitroquinoline-N-oxide

9-aminoacridine

other: 2-aminoanthracene (2-AA); N-methyl-N'-nitro-N-nitrosoguanidine (MNNG); 4-nitro-o-phenylendiamine (NOPD)

Details on test system and experimental conditions:

STANDARD PLATE TEST (SPT)
According to Ames et al., Mut Res 31: 347-364 (1975) and Maron & Ames, Mut Res 113: 173-215 (1983)

In the standard plate test, tubes were filled with 2mL portions of soft agar and kept in a water bath at 42 to 45°C. This soft agar consisted of 100 mL agar and 10 mL amino acid solution. As amino acid solution for the soft agar was used 0.5 mM histidine and 0.5 mM biotin for TA strains and 0.5 mM tryptophan for the E. coli strain.
Then following components are added:
0.1 mL test solution or vehicle
0.1 mL fresh bacterial culture
0.5 mL S9 -mix or phosphate buffer
After mixing samples were poured onto Vogel-Bonner (minimal glucose agar plates) plate and incubated for 48 - 72 hrs in the dark at 37°C.

PREINCUBATION TEST (PIT)
According to Yahagi et al. Mut Res 48: 121-129 (1977) and Matsushima et al., In: Norpoth, K.H. and R.C. Garner, Short-Term Test Systems for Detecting Carcinogens, Springer Verlag Berlin, Heidelberg, New York (1980)

For the preincubation test 0.1 mL test solution or vehicle, 0.1 mL bacterial suspension and 0.5 mL of either S9 mix or phosphate buffer were incubated at 37°C for 20 minutes. After addition of 2 mL soft agar, samples were poured onto agar plates and incubated again at 37°C for 48 to 72 hrs.
For the E. coli strain, plate test differed again in mixture of amino acid solution of the soft agar, the histidine component used for the TA strains being replaced by tryptophan.

Evaluation criteria:

An assay is accepted when the following criteria are met:
1.) number of colonies in the negative control is in the historical control range
2.) no indication of bacterial contamination (checked by sterility control)
3.) number of colonies in the positive controls are in the range of historical control data
4.) titer of viable bacteria is ≥ 10 E+8/mL

Toxicity is detected by:
1.) decrease in the number of revertants
2.) titer reduction
3.) clearing or diminution of the background lawn

Precipitation:
As long as no interference between precipitation and colony counting occurs is 5 mg/plate set as maximum dose even for relatively insoluble compounds.

A test chemical is to be considered as mutagenic when:
1.) increase of number of revertant colonies is reproducible and dose-related.
2.) in at least 1 tester strain doubling of colony counts with or without S-9 mix or after adding a metabolizing system is seen.

A test chemical is to be considered as non-mutagenic when:
1.) the number of revertants is inside the range of historical negative control data in 2 experiments performed independently from each other.

Species / strain:

S. typhimurium, other: TA 1535, TA 1537, TA 98, TA 100

Remarks:

SPT

Metabolic activation:

with and without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

no cytotoxicity

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Species / strain:

E. coli WP2 uvr A

Remarks:

SPT

Metabolic activation:

with and without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

no cytotoxicity

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Species / strain:

S. typhimurium, other: TA 1535, TA 1537, TA 98, TA 100

Remarks:

PIT

Metabolic activation:

with and without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

no cytotoxicity

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Species / strain:

E. coli WP2 uvr A

Remarks:

PIT

Metabolic activation:

with and without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

no cytotoxicity

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Additional information on results:

TEST-SPECIFIC CONFOUNDING FACTORS
No precipitation was detected.

ADDITIONAL INFORMATION ON CYTOTOXICITY:
No cytotoxic effects were seen in either the standard plate test (SPT) or the preincubation test (PIT) with and without metabolic activation.

CONTROLS
Negative and positive controls were as expected and confirmed the validity and sensitivity of the test method and system.

Experiment 1: Standard plate-incorporation test

SPT without S9-Mix  [mean no. of mutations/ plate]
Dosage [µg/plate]	TA 1535	TA 100	TA 1537	TA 98	WP2 uvrA
Solvent control	14	48	8	21	103
33	10	51	8	15	100
100	12	49	9	19	103
333	12	53	7	20	106
1000	13	47	7	18	100
2500	15	52	9	21	104
5000	12	55	6	18	104
Respective positive control	1354	949	331	723	1048
SPT with S9-Mix  [mean no. of mutations/ plate]
Dosage [µg/ plate]	TA 1535	TA 100	TA 1537	TA 98	WP2 uvrA
Solvent control	17	57	9	28	111
33	16	60	8	27	104
100	21	57	10	27	105
333	16	63	10	28	108
1000	16	62	11	32	112
2500	18	66	8	26	104
5000	17	61	9	29	113
Respective positive control	461	1225	664	776	196
Experiment 2: Preincubation test PIT without S9-Mix  [mean no. of mutations/ plate]
Dosage [µg/ plate]	TA 1535	TA 100	TA 1537	TA 98	WP2 uvrA
Solvent control	10	47	6	34	98
33	12	46	9	31	97
100	12	42	8	34	101
333	10	48	9	32	107
1000	10	52	6	31	103
2500	10	52	7	31	99
5000	13	45	7	37	109
Respective positive control	1639	1345	868	890	986
PIT with S9-Mix  [mean no. of mutations/ plate]
Dosage [µg/ plate]	TA 1535	TA 100	TA 1537	TA 98	WP2 uvrA
Solvent control	13	51	11	49	108
33	12	53	10	47	107
100	11	60	9	44	106
333	12	56	11	47	107
1000	11	48	13	47	108
2500	12	51	10	48	103
5000	12	58	13	46	105
Respective positive control	512	1440	313	1224	301

Reference 2

Endpoint:

in vitro gene mutation study in mammalian cells

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

study well documented, meets generally accepted scientific principles, acceptable for assessment

Qualifier:

equivalent or similar to guideline

Guideline:

OECD Guideline 479 (Genetic Toxicology: In Vitro Sister Chromatid Exchange Assay in Mammalian Cells)

Qualifier:

equivalent or similar to guideline

Guideline:

OECD Guideline 476 (In Vitro Mammalian Cell Gene Mutation Test)

GLP compliance:

yes (incl. QA statement)

Remarks:

testing lab.

Type of assay:

in vitro mammalian cell gene mutation test using the Hprt and xprt genes

Specific details on test material used for the study:

Analyses supplied by the sponsor indicated that the sample contained (by weight) 99.3% diethylene glycol, 0.09% monoethylene glycol, 0.39% triethylene glycol, 0.03% water, and 0.19% unknown components. The specific gravity was 1.1184. For the definitive tests, the test chemical was added directly into the cell culture medium of the test systems. Diethylene glycol is infinitely soluble in water with no solvent incompatibility as indicated in the information supplied by the sponsor. No other analyses of stability or concentration in the test system were performed.

Target gene:

CHO cells

Species / strain / cell type:

Chinese hamster Ovary (CHO)

Metabolic activation:

with and without

Metabolic activation system:

rat liver S-9 mix

S9 liver homogenate, prepared from Aroclor l254-induced, Sprague-Dawley male rats, as purchased from Microbiological Associates (MBA), Bethesda, MD. The S9 preparation was screened for metabolic activity by the supplier and at BRRC prior to use in our general testing program. Data from MBA showed that the S9 preparation was active with three different activation dependent mutagens in Salmonella bacterial strains TA98 and TA100.

Test concentrations with justification for top dose:

<= 50 mg/mL

Vehicle / solvent:

Test chemical was added directly into the cell culture medium of the test systems.

Untreated negative controls:

yes

Negative solvent / vehicle controls:

yes

True negative controls:

yes

Positive controls:

yes

Positive control substance:

other: dimethylnitrosamine and ethylmethanesulfonate

Details on test system and experimental conditions:

Selection of Test Concentrations:
Preliminary experiments were performed with CHO cells to determine an appropriate range of test concentrations in which the highest concentration would not kill over 90% of the treated cells. Test results wíth diethylene glycol indicated that concentrations up to 50 mg/ml, tested respectively with and without S9 metabolic activation, produced no excessive cytotoxic effects upon cell culture growth or mitotic indices. The 50 mg/ml (5.°G w/v) dose is the maximum dose used in the standard BRRC protocol and it was the highest dose tested with and without S9 activation in this study.


Test Procedures (HPGRT):
At 2 to 3 day intervals after treatment with the test agent, approximately 3 to 5 x 10^5 cells are subcultured in 100 mm tissue culture dishes in F12-D5 medium and incubated at 37°C in a 5 to 6% C02 atmosphere . After a period of at least 7 days to allow "expression" the mutant phenotype, cells of are dissociated with 0.05% to 0.075% trypsin, counted and plated at a concentration of 2 x 10^5 cells/dish in five 100 mm culture dishes (1 x 10^6 total cells) which each contain 10 ml of F12-D5 (TG) selective medium. At this same time interval, cells are diluted and 100 cells/dish are added to four culture plates containing F12-D5 medium (without TG) to assess viability (plating efficiency) of the treated cell population; the value obtained for the viable fraction for cells exposed to various doses and to control agents is used to correct the mutant frequency for cell populations with variable growth ability. All cultures are then incubated for an additional 6 to 8 days to allow growth of cells; medium is then discarded and colonies are fixed and stained for counting. The number of colonies in selection plates and in the viability test are counted by electronic methods, and data are recorded both as mutants/10^6 total cells and mutants/10^6 viable cells

Test Procedures (SCE Assay):
For testing without without metabolic activation, between 2 to 3 x 10^5 cells are plated into 75 cm2 culture flasks in F12-5 medium approximately 40-48 hrs prior to treatment and then incubated at 37°C in a 5 to 6% C02 atmosphere. Appropriate concentrations of the test agent or control chemicals are added to the cells and 3 µg/ml BrdU is added to all flasks. Cells are treated with test agents for 5 hrs, media is then removed by suction, cells are rinsed with buffered, physiological salt solution and fresh medium containing 3 µg/ml BrdU is added for at least 24 hrs of additional incubation at 37°C to allow two rounds of cell division. Cells are then harvested and chromosomes are prepared for SCE staining.
Treatment of cells for testing of chemicals which require metabolic activation for mutagenic effectiveness is performed similarly as for treatments without activation, except for three modifications :
1. Approximately 40 to 48 hrs prior to exposure to the test agents, between 2 to 3 x 10^5 cells are added to F12-5 medium in 75 cm2 culture flasks.
2. Before treatment with the test agents, F12-5 medium is removed and F12
medium without serum is added.
3. S9 metabolic activation mixture is added to each flask (including solvent and positive controls) before addition of test agents.
4. Cells are treated for a total of 2 hrs (rather than 5 hrs as in tests without S9) and then incubated for at least 28 additional hours before harvest for chromosome preparation. The exact harvest time can vary depending upon the cytotoxicity of the chemical.

Control Agents:
Positive and solvent control materials were tested concurrently with the test sample to assure the sensitivity of the test system. For the CHO and SCE assays, dimethylnítrosamíne (DMN)-CAS 62-75-9 and ethylmethanesulfonate
(EMS)-CAS 62-50-0 were used as the positive control agents to assure the reliability and sensitivity of the test system for detecting metabolic activation dependent and independent mutagens, respectively. Cell culture medium was used as the vehicle and solvent control agent for this test chemical.

Evaluation criteria:

HPGRT:
The criteria for interpretation of the test results as a positive or negative response depend upon both the level of statistical significance from the concurrent control and the evidence of a dose-response effect following treatment. When a definite dose-response relationship is not evident but one oi more marginally significant values are obtained, a careful examination of the data from the concurrent positive and negative controls and comparisons to historical control data are used to evaluate the possible biological significance of the responses. Historical negative control data indicate that an average spontaneous mutation frequency in CHO cells of approximately 3 to 4 mutants/10^6 viable cells with a 95 percentile range of 0 to 18 mutants/10^6 viable cells. Statistical comparisons against unusually high or low
spontaneous controls are subjectively scrutinized in respect to the above
expected variability of this test system. Tests are usually repeated to clarify
the biological significance of the data when questionable results are obtained.

For evaluation criteria SCE, see "Any other information on materials and methods incl. tables"

Statistics:

The data were analyzed after transformation of the mutation frequencies (MF) and SCE values according to the conversion method of Box and Cox (1964). For CHO mutation studies with a concurrent control frequency of zero mutants, the variance of recent historical controls was used for the statistical analyses. For SCE data, statistical analyses of historical data at BRRC indicate that an exponent of 0.15 is the appropriate value for transformation of SCE values. Positive controls for the CHO mutation test were run concurrently to assess the sensitivity of the assays in comparison to historical experience with the test system. Data for positive control agents were not compared statistically whenever differences were at least 5 times the concurrent negative control value and results were within the historical positive control range.

Species / strain:

Chinese hamster Ovary (CHO)

Metabolic activation:

with and without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

other: > 50 mg/mL

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Additional information on results:

The relative cytotoxicity of the various concentrations, tested both in the presence and absence of an S9 metabolic activation system, was determined by measuring the relative growth of treated and control cells incubated overnight following removal of the test chemical. The authors observed that diethylene glycol was not highly cytotoxic when tested either with or without S9 metabolic activation. A concentration of 50 mg/ml produced 16% inhibition of growth without S9 and 27% inhibition with S9. For the definitive tests, a concentration range between 30 to 50 mg/ml was tested in the mutagenicity tests with and without S9. The 50 mg/ml (5% w/v) dose is the usual maximum dose for non-cytotoxic chemicals tested by the BRRC Standard Protocol to avoid possible artefacts produced by non-physiological cell-culture conditions at higher doses.

Executive summary:

The test substance was evaluated for potential genotoxic activity using the Chinese Hamster Ovary (CHO) Mutation test and the Sister Chromatid Exchange (SCE) test. The results indicated that the test substance did not produce a dose-related or repeatable, significant mutagenic effect in either of the in vitro screening tests employed. No detectable cytotoxicity to CHO cells was produced by the test substance even at the maximum concentration of 50 mg/mL (5 % w/v) typically evaluated as the highest dose in these tests following the standard BRRC protocol. The test substance was neither genotoxic nor cytotoxic to CHO cells under the conditions of these two in vitro test systems.

Reference 3

Endpoint:

in vitro cytogenicity / chromosome aberration study in mammalian cells

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

study well documented, meets generally accepted scientific principles, acceptable for assessment

Qualifier:

equivalent or similar to guideline

Guideline:

OECD Guideline 473 (In Vitro Mammalian Chromosome Aberration Test)

GLP compliance:

yes (incl. QA statement)

Remarks:

testing lab.

Type of assay:

in vitro mammalian cell transformation assay

Specific details on test material used for the study:

Briefly, the test chemical was analyzed by the sponsor and found to consist of 99.3% diethylene glycol, 0.09% monoethylene glycol, 0.39% triethylene glycol, 0.19% unknown impurities and 0.03% water. The test chemical was added directly into the cell culture medium of the test system without dilution . No analyses of stability or concentration in the medium were performed. Diethylene glycol is infinitely soluble in water and no solvent incompatibilities were listed by the sponsor.

Target gene:

CHO cells

Species / strain / cell type:

Chinese hamster Ovary (CHO)

Metabolic activation:

with and without

Metabolic activation system:

rat liver S-9 mix

Test concentrations with justification for top dose:

30 - 50 mg/ml

Vehicle / solvent:

Cell-culture medium was used as the solvent as the test material was added directly into the cell-culture test system.

Untreated negative controls:

yes

Negative solvent / vehicle controls:

yes

True negative controls:

not specified

Positive controls:

yes

Positive control substance:

other: cyclophosphamide and triethylenemelamíne

Details on test system and experimental conditions:

Selection of Test Concentrations: Preliminary experiments were performed with CHO cells to determine an appropriate range of test concentrations in which the highest concentration would kill, no more than (approximately) 90% of the treated cells. Test concentrations for the SCE test were chosen so that sufficient numbers of cells in the second division (M2) would be available for determination of SCEs. Test results with diethylene glycol indicated that concentrations up to 50 mg/ml, the maximum concentration used for evaluation of test chemicals following the BRRC standard test protocol, were not cytotoxic to CHO cells. For the definitive tests, 50 mg/ml was also selected as the maximum dose and additional concentrations below this dose were evaluated. In this tests, doses between 30 and 50 mg/ml were evaluated for genotoxic potential.
Control Agents: Positive and solvent control materials were tested concurrently with the test sample to assure both the sensitivity of the test systems and the concurrence of the results to historical test performance at BRRC. For this assays, dimethylnitrosamine (DMN)-CAS 62-75-9 and ethylmethanesulfonate (EMS)-CAS 62-50-0 were used as positive control agents to assure the sensitivity and reliability of the test system for detecting metabolic activation dependent and independent mutagens, respectively. Cell culture medium was used as the negative control.
Metabolic Activation: S9 liver homogenate, prepared from Aroclor 1254-induced, Sprague-Dawley male rats, was purchased- from Microbiological Associates, Bethesda, MD. The S9 preparation used for the CHO test was screened for appropriate metabolic activity by the supplier with three activation-dependent mutagens and Salmonella strains TA98 and TA100. A volume of 50 ul of S9 homogenate was used for each 1.0 ml of the S9 activation mixture.

Evaluation criteria:

For evaluations of direct clastogenic potential, CHO cells were exposed to diethylene glycol and appropriate controls for the complete 12 hour period without S9 activation. Determination of indirect genotoxic potential, requiring metabolic activation by liver S9-homogenate, was studied with a 2-hour exposure period to the test chemical and S9 activation system.

Statistics:

Statistical analyses of the test data employed the Fisher's Exact Test (one-tailed) to determine significance of differences between the test and control populations. This statistical test was considered appropriate for the analysis of the data because it is a distribution independent test and cytogenic data often vary from a normal distribution required for parametric analyses. A positive test result was interpreted by the attainment of differences from the control at the p < 0.05 level of significance for at least one test concentration, together with an indication of a concentration-related effect of exposures or
reproducibility between the duplicate cultures. Rounding of data to either two decimal places or to the appropriate number significant figures was performed for presentation on tables. Although statistically significant decreases in genotoxicity indices can occur because of cytotoxic responses, only statistically significant increases ín responses above control values are indicated on tables for simplicity. All statistical tests were performed to determine whether the treatment with the test agent produced a response statistically different from the value(s) obtained with the concurrent solvent control cultures.

Species / strain:

Chinese hamster Ovary (CHO)

Metabolic activation:

with and without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

other: > 50 mg/mL

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Additional information on results:

Diethylene glycol did not produce dose-related, statistically significant increases in the incidence of chromosome aberrations in CHO cells in tests conducted with and without the addition of a rat-liver homogenate, S9 metabolic activation system. A single statistical indication of an increase above control values was obtained for the lowest dose sampled at 12 hr after test initiation in the test with S9 activation. This result was not repeated at higher doses and the level of the increase was essentially the same as the variation in the concurrent negative control values. For these reasons, the statistical indication was not considered biologically relevant. The positive control agents, cyclophosphamide and triethylenemelamine, both produced significant increases in chromosome aberrations indicative of the appropriate, reliability of the test system. The control cultures employing only the culture growth medium had low and acceptable levels of chromosome aberrations typical for these cultured cells. Diethylene glycol was concluded to lack significant genotoxic activity under the conditions employed for this in vitro test system.

No statistically significant increase in the number of SCEs was observed with any of the test cultures treated with the test agent. The incidence of SCEs in all of the cultures exposed to the test agent was within the historical negative control range for this test at BRRC (Methods, Section G1). Diethylene glycol was considered inactive under the non-activation conditions of this sensitive in vitro assay.

Table 1: Preliminary cytotoxicity test for determination of appropriate dose range for subsequent genotoxicity tests

Test chemical [mg/mL]	Initial cell concentration* [cells/plate x 10^5]	Final cell concentration** [cells/plate x 10^5]		Percentage relative to solvent control [%]
		Without S9 mix	With S9 mix	Without S9 mix	With S9 mix
0.0 (solvent-medium) 30 35 40 45 50	5.0 5.0 5.0 5.0 5.0 5.0	19.76 20.42 20.42 18.79 16.15 19.84	21.57 18.86 20.79 20.50 21.24 22.43	100.0 103.3 103.3 95.1 81.7 100.4	100.0 87.4 96.4 95.0 98.5 104.0

*5 x 10⁵CHO cells were inoculated uniformly into all culture flasks approximately 24 h prior to chemical treatments

** determined approximately 20 to 24 h after removal of the test agents

Table 2: Cytotoxicity testing summary - mitotic indices (percentage of dividing cells) after an 8 h treatment

Test chemical [mg/mL]	No. of cells counted	No. of cells in metaphase	% cells in metaphase	% of control
Test without S9 activation system
0.0 30 35 40 45 50	500 500 500 500 500 500	10 10 9 10 10 11	2.0 2.0 1.9 2.0 2.0 2.2	100.0 100.0 90.0 100.0 100.0 110.0
Test with S9 activation system
0.0 30 35 40 45 50	500 500 500 500 500 500	9 9 10 8 13 11	1.9 1.9 2.0 1.6 2.5 2.2	100.0 100.0 111.1 88.9 144.4 122.2

Table 3: Summary of combined test data and statistical analyses

Test chemical [mg/mL]	Total cells with aberrations	Total cells scored	Percentage aberrant cells (± SD)	Level of statistical* difference above solvent control value	Total cells with aberrations	Total cells scored	Percentage aberrants cells (± SD)	Level of statistical* difference above solvent control value
Test without S9 activation system
	8 h sampling time				12 h sampling time
0.0 30 40 50 TEM 1.0 µg/mL	2 3 7 3 18	100 100 100 100 50	2.0 (0.0) 3.0 (4.2) 7.0 (4.2) 3.0 (1.4) 36.0	- NS NS NS c	2 2 4 2 -	100 100 100 100 -	2.2 (2.8) 2.0 (2.8) 4.0 (0.0) 2.2 (0.0) -	- NS NS NS -
Test with S9 activation system
	8 h sampling time				12 h sampling time
0.0 30 40 50 Cyclophosphamide 15 µg/mL	6 4 1 6 9	100 100 100 100 50	6.0 (0.0) 4.0 (0.0) 1.0 (1.4) 6.0 (0.0) 18	- NS NS NS a	1 7 4 5 -	100 100 100 100 -	1.0 (1.4) 7.0 (1.4) 4.0 (2.8) 5.0 (1.4) -	- a NS NS -

*significance based on analyses using Fisher’s Exact Test, 1-tailed: a = 0.05 > p > 0.01; c = p < 0.001; NS = p > 0.05

Conclusions:

The test substance did not produce significant, dose-related effects in the CHO in vitro chromosome aberration assay when tested with and without addition of an S9 metabolic activation. The results indicate that the test substance does not possess significant clastogenic or cytotoxic potential under the conditions of this in vitro test system.

Executive summary:

The test substance was evaluated for potential genotoxic activity using the Chinese Hamster Ovary (CHO) in vitro chromosome aberration test system. Test concentrations of diethylene glycol ranged from 10 mg/mL to 50 mg/mL in tests both with and without a rat-liver S9 metabolic activation system. Test concentration ranges were chosen on the basis of a preliminary cytotoxicity test. The highest three doses which did not produce excessive cytotoxic inhibition of mitotic cells were scored for incidences of chromosome aberrations.

Results obtained in tests with and without a rat-liver homogenate, S9 metabolic activation system, indicated that the test substance did not produce significant, dose-related increases in chromosome aberrations in comparison to values of control cultures. No evidence of clastogenic or cytotoxic effects were observed under the conditions of this test system up to the maximum 50 mg/mL dose used in this test following the standard BRRC protocol.

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed (negative)

Genetic toxicity in vivo

Link to relevant study records

Reference

Endpoint:

in vivo mammalian somatic cell study: cytogenicity / erythrocyte micronucleus

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

OECD Guideline 474 (Mammalian Erythrocyte Micronucleus Test)

GLP compliance:

yes

Type of assay:

micronucleus assay

Specific details on test material used for the study:

Purity: 99.932 area %

Species:

mouse

Strain:

NMRI

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Weight at study initiation: 26.9 g
- Housing: individual housing in appropriately labeled cages, Makrolon cages, type M I
- Bedding: Type Lignocel FS 14 fibres, dustfree bedding, supplied by SSNIFF, Soest, Germany
- Diet: standardized pelleted feed (Maus/Ratte Haltung "GLP", Provimi Kliba SA, Kaiseraugst, Switzerland), ad libitum
- Water: drinking water from bottles, ad libitum
- Acclimation period: at least 5 days

ENVIRONMENTAL CONDITIONS
- Temperature (°C): 20 - 24°C
- Humidity (%): 30 - 70%
- Air changes (per hr): fully air-conditioned rooms with central air conditioning
- Photoperiod (hrs dark / hrs light): 12 / 12

Route of administration:

intraperitoneal

Vehicle:

purified water

Details on exposure:

The substance to be administered per kg body weight was dissolved in purified water. To achieve a solution of the test substance in the vehicle, the test substance preparation was shaken thoroughly. All test substance formulations were prepared immediately before administration.

Duration of treatment / exposure:

once intraperitoneally

Frequency of treatment:

once intraperitoneally

Dose / conc.:

500 mg/kg bw/day (nominal)

Dose / conc.:

1 000 mg/kg bw/day (nominal)

Dose / conc.:

2 000 mg/kg bw/day (nominal)

No. of animals per sex per dose:

5

Control animals:

yes, concurrent vehicle

Positive control(s):

vincristine sulfate

Tissues and cell types examined:

Bone marrow cells

Details of tissue and slide preparation:

The animals were sacrificed and the bone marrow of the two femora was prepared 24 and 48 hours after administration from the animals of the highest dose group of 2000 mg/kg bw and those of the vehicle control group. In the test groups of 1000 mg/kg and 500 mg/kg bw and in the positive control groups, the 24-hour sacrifice interval was investigated only. After staining of the preparations, 2000 polychromatic erythrocytes were evaluated per animal and investigated for micronuclei. The normocytes with and without micronuclei occurring per 2000 polychromatic erythrocytes were also recorded.
As vehicle control, male mice were administered merely the vehicle purified water by the same route and in the same volume as the animals of the dose groups, which gave frequencies of micronucleated polychromatic erythrocytes within the historical vehicle control data range.

Evaluation criteria:

A finding is considered positive if the following criteria are met:
• Statistically significant and dose-related increase in the number of PCEs containing micronuclei.
• The number of PCEs containing micronuclei has to exceed both the concurrent vehicle control value and the range of the historical vehicle control data.
A test substance is considered negative if the following criteria are met:
• The number of cells containing micronuclei in the dose groups is not statistically significant increased above the concurrent vehicle control value and is within the range of the historical vehicle control data.

Statistics:

The statistical evaluation of the data was carried out using the program system MUKERN (BASF SE). The asymptotic U test according to MANN-WHITNEY (modified rank test according to WILCOXON) was carried out to clarify the question whether there are statistically significant differences between the untreated control group and the treated dose groups with regard to the micronucleus rate in polychromatic erythrocytes. The relative frequencies of cells containing micronuclei of each animal were used as a criterion for the rank determination for the U test. Statistical significances were identified as follows: * p ≤ 0.05 ** p ≤ 0.01. However, both biological relevance and statistical significance were considered together.

Sex:

male

Genotoxicity:

negative

Toxicity:

no effects

Vehicle controls validity:

valid

Negative controls validity:

valid

Positive controls validity:

valid

Additional information on results:

Both positive control substances, cyclophosphamide for clastogenicity and vincristine sulfate for spindle poison effects, led to the expected increase in the rate of polychromatic erythrocytes containing small or large micronuclei.
A slight inhibition of erythropoiesis determined from the ratio of polychromatic to normochromatic erythrocytes was detected at 2 000 mg/kg body weight at 48-hour sacrifice interval.
According to the results of the present study, the single intraperitoneal administration of Diethylene glycol did not lead to a relevant increase in the number of polychromatic erythrocytes containing either small or large micronuclei. The rate of micronuclei was close to the range of the concurrent vehicle control in all dose groups and at all sacrifice intervals and within the range of the historical vehicle control data.

Summary table - Induction of Micronuclei in bone marrow cells - single intraperitoneal administration

Test group	Sacrifice	Animal	Micronuclei in PCE		Number
	interval [hrs]	No.	total^a [‰]	large MN^b [‰]	of NCE^c
Vehicle control purified water	24	5	1.3	0.0	3 820
Test substance 500 mg/kg bw.	24	5	0.6	0.0	4 389
Test substance 1 000 mg/kg bw.	24	5	0.6	0.0	4 227
Test substance 2 000 mg/kg bw.	24	5	0.6	0.0	3 941
Positive control cyclophosphamide 20 mg/kg bw.	24	5	20.8**	0.1	6 045
Positive control vincristine sulfate 0.15 mg/kg bw.	24	5	54.3**	17.1**	5 979
Vehicle control purified water	48	5	0.9	0.1	3 138
Test substance 2 000 mg/kg bw.	48	5	1.9	0.0	5 331

PCE	= polychromatic erythrocytes
NCE	= normochromatic erythrocytes
bw.	= body weight
^a	= sum of small and large micronuclei
^b	= large micronuclei (indication for spindle poison effect)
^c	= number of NCEs observed when scoring 10 000 PCEs    (indication for target organ toxicity)
*	= p ≤ 0.05
**	= p ≤ 0.01

Conclusions:

Under the experimental conditions chosen here, the test substance has no chromosome-damaging (clastogenic) effect nor does it lead to any impairment of chromosome distribution in the course of mitosis (aneugenic activity) in bone marrow cells of NMRI mice in vivo.

Executive summary:

The test substance was assessed for its potential to induce chromosomal damage (clastogenicity) or spindle poison effects (aneugenic activitiy) in NMRI mice using the micronucleus test method. For this purpose, the test substance, dissolved in purified water, was administered once intraperitoneally to male animals at dose levels of 500 mg/kg, 1000 mg/kg and 2000 mg/kg body weight in a volume of 10 mL/kg body weight in each case.

The animals were sacrificed and the bone marrow of the two fermora was prepared 24 and 48 hours after administration in the highest dose group of 2000 mg/kg body weight and in the vehicle controls. In the test groups of 1000 mg/kg and 500 mg/kg body weight and in the positive control groups, the 24 -hour sacrifice interval was investigated only. After staining of the preparations, 2000 polychromatic erythrocytes were evaluated per animal and investigated for micronuclei. The normocytes with and without micronuclei occuring per 2000 polychromatic erythrocytes were also recorded.

As vehicle control, male mice were administered merely the vehicle purified water by the same route and in the same volume as the animals of the dose groups, which gave frequencies of micronucleated polychromatic erythrocytes within the historical vehicle control data range.

Both positive control substances, cyclophosphamide for clastogenicity and vincristine sulfate for spindle poison effects, led to the expected increase in the rate of polychromatic erythrocytes containing small or large micronuclei.

A slight inhibition of erythropoiesis determined from the ratio of polychromatic to normochromatic erythrocytes was detected at 2000 mg/kg body weight at 48 -hour sacrifice interval.

According to the results of the present study, the single intraperitoneal administration of the test substance did not lead to a relevant increase in the number of polychromatic erythrocytes containing either small or large micronuclei.

The rate of micronuclei was close to the range of the concurrent vehicle control in all dose groups and at all sacrifice intervals and within the range of the historical vehicle control data.

Thus, under the experimental conditions of this study, the test substance does not induce cytogenetic damage in bone marrow cells of NMRI mice in vivo.

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed (negative)

Additional information

Genetic toxicity in vitro

Bacterial reverse mutation tests

A test for bacterial gene mutagenicity was conducted with diethylene glycol according to the OECD TG 471 under GLP conditions with the following bacterial strains: Salmonella typhimurium TA 98, TA 100, TA 1535 and TA 1537 and E. coli WP2 uvrA (BASF, 2013). The test concentrations were 0, 33, 100, 333, 1000, 2500, and 5000 µg/plate for the standard plate test with and without S9-mix, and for the preincubation test with and without S9-mix, respectively. Negative (sterility and solvent) and positive controls were considered. Under the experimental conditions chosen, the test item diethylene glycol was not mutagenic in the bacterial reverse mutation test in the absence and the presence of metabolic activation. Neither precipitation nor cytotoxicity were noticed. All negative and positive controls were as expected and confirmed the validity, suitability and sensitivity of the test method and system used.

This result is supported by a an older study provided by Union Carbide (UCC, 1984), which also reported a negative result for diethylene glycol tested according to the EU Method B10, which is similar to the OECD TG 471, using the following strains of Salmonella typhimurium: TA 98, TA 100, TA 1535, TA 1537 and TA 1538. In this study, the tested concentrations ranged between 1000 and 11800 µg/plate and testing was done with and without S9 mix.

In a less reliable and poor reported study of Krug et al.(1986) the mutagenic potential of diethylene glycol in Salmonella tester strains TA 98, TA 100, TA 102, and TA 104 in the presence and absence of liver homogenate was investigated. The S9-mix was prepared from Aroclor 1254 pretreated female rats. A weak mutagenic effect was detected in strain TA 104 in the presence of "S9-Mix" (maximum: 2,2 fold increase over the spontaneous reversion frequency at 315 µmol DEG/plate). Due the limited available documentation of these results they were not taken into account for classification and labeling.  

 

Chromosome aberration assay  

A chromosome aberration assay with CHO cells tested in absence and presence of S9 mix was conducted with the test item equivalent or similar to OECD TG 473 (Union Carbide, 1984). Diethylene glycol at concentrations of 30 - 50 mg/mL did not produce dose-related, statistically significant increases in the incidence of chromosome aberrations in CHO cells in tests conducted with and without the addition of a rat-liver homogenate, S9 metabolic activation system. A single statistical indication of an increase above control values was obtained for the lowest dose sampled at 12 h after test initiation in the test with S9 activation. This result was not repeated at higher doses and the level of the increase was essentially the same as the variation in the concurrent negative control values. For these reasons, the statistical indication was not considered biologically relevant. The positive control agents, cyclophosphamide and triethylenemelamine, both produced significant increases in chromosome aberrations indicative of the appropriate, reliability of the test system. The control cultures employing only the culture growth medium had low and acceptable levels of chromosome aberrations typical for these cultured cells. Diethylene glycol was concluded to lack significant genotoxic activity under the conditions employed for this in vitro test system.  

 

HGPRT

A HGPRT assay with CHO cells tested in absence and presence of S9 mix was conducted with the test item equivalent or similar to OECD TG 476 (Union Carbide, 1984). The relative cytotoxicity of the various concentrations, tested both in the presence and absence of an S9 metabolic activation system, was determined by measuring the relative growth of treated and control cells incubated overnight following removal of the test chemical. It was observed that diethylene glycol was not highly cytotoxic when tested either with or without S9 metabolic activation. A concentration of 50 mg/mL produced 16% inhibition of growth without S9 and 27% inhibition with S9. For the definitive tests, a concentration range between 30 to 50 mg/mL was tested in the mutagenicity tests with and without S9. In the main test, diethylene glycol was neither genotoxic nor cytotoxic to CHO cells under the conditions of this in vitro test system.

 

Genetic toxicity in vivo

An in vivo micronucleus test with mouse was conducted with diethylene glycol according to the OECD TG 474, under GLP conditions (BASF, 2009). The male animals received a single intraperitoneal administration of the test item at following dose levels: 0, 500, 1000 and 2000 mg/kg bw. The single intraperitoneal administration of the test item in mice did not lead to a relevant increase in the number of polychromatic erythrocytes containing either small or large micronuclei. The rate of micronuclei was close to the range of the concurrent vehicle control in all dose groups and at all sacrifice intervals. Also, it was within the range of the historical vehicle control data. Both positive control substances, cyclophosphamide for clastogenicity and vincristine sulfate for spindle poison effects, led to the expected increase in the rate of polychromatic erythrocytes containing small or large micronuclei. Thus no indication for mutagenic potential was noted even after administration of very high doses and for a non-relevant worst case administration route (i.p.).

In less reliable studies (Barilyak, 1985 and Krug, 1986) DEG was further tested for in vivo genotoxicity:

An increase in chromosome damage in the bone marrow cells was reported after administration of 1/5 of the LD50 of DEG by gavage in hamster (Barilyak, 1985). In rats, DEG administration (dose not reported) caused dominant lethal mutations (Barilyak, 1985). In a micronucleus test (species not reported), a single intraperitoneal injection of 60% of LD50 of DEG caused chromosomal fragments. This induction was suppressed when the animals were pretreated during 7 days with a low daily dose of DEG (4 % of the LD50) (Krug et al., 1986). Due the limited available documentation of these results they were not taken into account for classification and labeling.  

 

SUMMARY  

In summary, the data of a whole battery of robust high-quality in vitro studies as well as results from an in vivo micronucleus test show that the test item does not possess any mutagenic or genotoxic properties. In addition, DEG and other ethylene glycols have no structural alerts that are known to lead to DNA reactivity and the overall evidence points to negative genotoxicity potential. Even though some old and very poorly documented abstracts show some positive results, new state of the art studies are available clearly documenting the absence of any genotoxic effects. The questionable results were thus judged as disregarded studies.

In a recently conducted literature search, no further relevant information on genetic toxicity were identified (please refer to IUCLID section 12).

Justification for classification or non-classification

The available experimental test data are reliable and suitable for classification purposes under Regulation (EC) No 1272/2008. As a result the substance is not considered to be classified for mutagenicity under Regulation (EC) No 1272/2008.