2,2'-[(1-methylethylidene)bis(4,1-phenyleneoxymethylene)]bisoxirane

Endpoint summary

Key value for chemical safety assessment

Genetic toxicity in vitro

Description of key information

BADGE was positive in a number of in vitro assays but negative in in vivo assays

Link to relevant study records

Reference

Reference 1

Endpoint:

in vitro gene mutation study in bacteria

Remarks:

Type of genotoxicity: gene mutation

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: Guideline study conducted under GLP

Qualifier:

according to guideline

Guideline:

OECD Guideline 472 (Genetic Toxicology: Escherichia coli, Reverse Mutation Assay)

Deviations:

no

Qualifier:

according to guideline

Guideline:

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

Deviations:

no

Qualifier:

according to guideline

Guideline:

EPA OPPTS 870.5265 (The Salmonella typhimurium Bacterial Reverse Mutation Test)

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

Target gene:

TA98 hisD3052
TA100 hisG46
TA1535 hisG46
TA1537 hisC3076
WP2uvrA trp

Species / strain / cell type:

E. coli WP2 uvr A pKM 101

Species / strain / cell type:

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

Test concentrations with justification for top dose:

Up to 25 micrograms/plate

Vehicle / solvent:

DMSO

Untreated negative controls:

yes

Negative solvent / vehicle controls:

yes

Positive controls:

yes

Positive control substance:

benzo(a)pyrene

Remarks:

Migrated to IUCLID6: 2-nitrofluorene, 2-aminoantracene, sodum azide, ICR-191, 4-nitroquinoline-N-oxide

Details on test system and experimental conditions:

In addition to a mutation in either the histidine or tryptophan opérons, the tester strains contain two additional mutations which enhance their sensitivity to some mutagenic compounds. A mutation of either the uvrA gene (Escherichia coli) or the uvrB gene (Salmonella typhimurium), results in a deficient DNA excision repair system which greatly enhances the sensitivity of these strains to some mutagens. Since the uvrB deletion extends through the bio gene, the Salmonella typhimurium tester strains containing this deletion also require the vitamin biotin for growth.

The Salmonella typhimurium tester strains also contain the rfa wall mutation which results in the loss of one of the enzymes responsible for the synthesis of part of the lipopolysaccharide (LPS) barrier that forms the surface of the bacterial cell wall. The resulting cell wall deficiency increases permeability to certain classes of chemicals such, as those containing large ring systems (i.e., benzo[a]pyrene) that would otherwise be excluded by a normal intact cell wall.

Strains TA98 and TA100 also contain the pKMlOl plasmid, which further increases the sensitivity of these strains to some mutagens. The mechanism by which this plasmid increases sensitivity to mutagens has been suggested to be by modifying an existing bacterial DNA repair polymerase complex involved with the mismatch-repair process.

Tester strains TA98 and TA 1537 are reverted from histidine dependence (auxotrophy) to histidine independence (prototrophy) by frameshift mutagens. Tester strains TA100, TA1535, and WP2«vrA are reverted from auxotrophy to prototrophy by base substitution mutagens.

Source of Tester Strains. The Salmonella typhimurium tester strains in use at Covance were received directly from Dr. Bruce Ames, Department of Biochemistry, University of California, Berkeley. The Escherichia coli tester strain, WP2«vrA, was received from The National Collection of Industrial Bacteria, Torrey Research Station, Scotland (United Kingdom).

Frozen Permanent Stocks. Frozen permanent stocks were prepared by growing fresh overnight cultures, adding DMSO (0.09 mL/mL of culture) and freezing away appropriately vialed aliquots. Frozen permanent stocks of the tester strains were stored at <-70°C.

Master Plates. Master plates of the tester strains were prepared by streaking each tester strain from a frozen permanent stock onto minimal agar appropriately supplemented with either histidine and biotin or tryptophan, and for strains containing the pKMlOl plasmid, ampicillin. Tester strain master plates were stored at 5 ± 3°C.

Preparation of Overnight Cultures
Inoculation. Overnight cultures for use in all testing procedures were inoculated by transferring a colony from the appropriate master plate to a flask containing culture medium. Inoculated flasks were placed in a shaker/incubator which was programmed to begin operation (shaking, 125 ± 25 rpm; incubation, 37 ± 2°C) so that the overnight cultures were in log phase or late log phase when turbidity monitoring began.

Harvest. To ensure that cultures were harvested in late log phase, the length of incubation was determined by spectrophotometric monitoring of culture density. Cultures were harvested once a predetermined density was reached which ensures that cultures had reached a density of at least 0.5 X 109 cells per mL and that the cultures have not overgrown. Overgrown (stationary) cultures may exhibit decreased sensitivity to some mutagens. Cultures were removed from incubation when the target density was reached and were held at 5 ± 3°C until used in the assay.

Confirmation of Tester Strain Genotype. Tester strain cultures were checked for the following genetic markers on the day of their use in the mutagenicity assay:
rfa Wall Mutation. For the Salmonella tester strains, the presence of the rfa wall mutation was confirmed by demonstration of the sensitivity of the culture to crystal violet. An aliquot of an overnight culture of each strain was overlaid onto plates containing selective media and an antibiotic sensitivity disk containing 10 u,g of crystal violet was added. Sensitivity was demonstrated by inhibition of bacterial growth in a zone immediately surrounding the disk.
pKMlOl Plasmid. The presence of the pKMlOl plasmid was confirmed for cultures of tester strains TA98 and TA100 by demonstration of resistance to ampicillin. An aliquot of an overnight culture of each strain was overlaid onto plates containing selective media and an antibiotic sensitivity disk containing 10 jxg of ampicillin was added. Resistance was demonstrated by growth in the zone immediately surrounding the disk.

Characteristic Number of Spontaneous Revertants. The mean number of spontaneous revertants per plate in the vehicle controls that is characteristic of the respective strains was demonstrated by plating 100 jxL aliquots of each culture along with the appropriate vehicle on
selective media.

Culturing Broth. The broth used to grow overnight cultures of the tester strains was Vogel-Bonner salt solution (Vogel and Bonner, 1956) supplemented with 2.5% (w/v) Oxoid Nutrient Broth No. 2 (dry powder).

Minimal Bottom Agar Plates. Bottom agar (25 mL per 15 x 100 mm petri dish) was Vogel-Bonner minimal medium E (Vogel and Bonner, 1956), supplemented with 1.5% (w/v) agar and 0.2% (w/v) glucose.

Top Agar for Selection of Revertants. Top (overlay) agar was prepared with 0.7% agar (w/v) and 0.5% NaCl (w/v) and was supplemented with 10 mLof 1) 0.5 mM histidine/biotin solution per 100 mL agar for selection of histidine revertants, or 2) 0.5 mM tryptophan solution per 100 mL of agar for selection of tryptophan revertants. For the agar overlay, 2.0 mL of the supplemented top agar was used.

Control Articles
Vehicle Controls. Vehicle controls were plated for all tester strains in the presence and absence of S9 mix. The vehicle control was plated, using a 50 uJL aliquot of dimethylsulfoxide (equal to the maximum aliquot of test article dilution plated), along with a 100 uL aliquot of the appropriate tester strain and a 500 uJL aliquot of S9 mix (when necessary), on selective agar.

Sterility Controls. The most concentrated test article dilution was checked for sterility by plating a 50 uL aliquot (the same volume used in the assay) on selective agar. The S9 mix was checked for sterility by plating 0.5 mL on selective agar.

S9 Metabolie Activation System
S9 Homogenate. Liver microsomal enzymes (S9 homogenate) were purchased from Molecular Toxicology, Inc., Batch 1204 (40.0 mg of protein per mL). The homogenate was prepared from male Sprague-Dawley rats that had been injected (i.p.) with Aroclor™ 1254 (200 mg per mLin corn oil) at 500 mg/kg as described by Ames et al., (1975).
S9 Mix. The S9 mix was prepared immediately prior to its use in any experimental procedure.

Rangefinding assay
The growth inhibitory effect (cytotoxicity) of the test article to the test system was determined in order to allow the selection of appropriate concentrations to be tested in the mutagenicity assay.
Design. The rangefinding study was performed using tester strains TA 100 and WP2«vrA in both the presence and absence of S9 mix. Ten concentrations of test article were tested at one plate per concentration. The test article was checked for cytotoxicity up to a maximum concentration
of 5 mg per plate.
Rationale. The cytotoxicity of the test article observed on tester strain TA100 is generally representative of that observed on the other Salmonella typhimurium tester strains and because of the comparatively high number of spontaneous revertants per plate observed with this strain, gradations of cytotoxicity can be readily discerned from routine experimental variation. The Escherichia coli tester strain WP2«vrA does not possess the rfa wall mutation that the Salmonella typhimurium strains have and thus, a different range of cytotoxicity may be observed. Also, the cytotoxicity induced by a test article in the presence of S9 mix may vary greatly from that observed in the absence of S9 mix. Therefore, this would require that different test article concentration ranges be tested in the mutagenicity assay based on the presence or absence of the microsomal enzymes.
Evaluation of the Rangefinding assay. Cytotoxicity is detectable as a decrease in the number of revertant colonies per plate and/or by a thinning or disappearance of the bacterial background lawn.
Selection of the Maximum Concentration for the Mutagenicity Assay. Cytotoxicity was observed in the rangefinding study and the highest concentration level of test article used in the subsequent mutagenicity assay was a concentration which gave a reduction of revertants per plate and/or a thinning or disappearance of the bacterial background lawn.
Mutagenicity Assay
Design. The assay was performed using tester strains TA98, TA100, TA1535, TA1537, and WP2uvrA both in the presence and absence of S9 mix along with the appropriate vehicle and positive controls. The concentrations of test article were selected based on the results of the rangefinding assay. The results of the initial mutagenicity assay were confirmed in an independent experiment.
Frequency and Route of Administration. The tester strains were exposed to the test article via the preincubation modification of the Ames Test originally described by Yahagi et al. (1975) and Maron and Ames (1983). This methodology has been shown to detect a wide range of classes of chemical mutagens. In the preincubation methodology, S9 mix (or phosphate buffer, where appropriate), the tester strain, and the test article were preincubated prior to the addition of molten agar. The agar and the preincubation reaction mixture were mixed and then overlaid onto a minimal agar plate. Following incubation, revertant colonies were counted. All concentrations of the test article, the vehicle controls and the positive controls were plated in triplicate.
Plating Procedures
These procedures were used in both the rangefinding study and the mutagenicity assay. Each plate was labeled with a code which identified the test article, test phase, tester strain, activation condition and concentration level. The S9 mix and dilutions of the test article were
prepared immediately prior to their use.
When S9 mix was required, 500 uL of S9 mix was added to 13 X 100 mm glass culture tubes, which had been pre-heated to 37 ± 2°C. To these tubes was added 100 uL of tester strain and 50 uL of vehicle or test article concentration. When S9 mix was not required, 500 uL of 0.1M
phosphate buffer was substituted for the S9 mix. After the required components had been added, the mixture was vortexed and allowed to incubate for 20 ± 2 minutes at 37 ± 2°C. Two mL of molten selective top agar was then added to each tube, and the mixture was vortexed and overlaid onto the surface of 25 mL of minimal bottom agar contained in a 15 x 100 mm petri dish. After the overlay solidified, the plates were inverted and incubated for 52 ± 4 hours at 37 ± 2°C. Positive control articles were plated using a 50 uL plating aliquot.

Evaluation criteria:

Scoring the Plates
Plates which were not evaluated immediately following the incubation period were held at 5 ± 3°C until such time that colony counting and bacterial background lawn evaluation could take place.
Bacterial Background Lawn Evaluation. The condition of the bacterial background lawn was evaluated both macroscopically and microscopically (using a dissecting microscope) for indications of cytotoxicity and test article precipitate. Evidence of cytotoxicity was scored relative to the vehicle control plate and was recorded along with the revertant counts for all plates at that concentration level. Lawns were scored as 1) normal, 2) slightly reduced, 3) moderately reduced, 4) extremely reduced, 5) absent, or 6) obscured by precipitate. If present on the plates, macroscopic precipitate was scored as slight, moderate or heavy.
Counting Revertant Colonies. Revertant colonies were counted either by automated colony counter or by hand. If there was sufficient test article precipitate on the plates at any concentration that interferes with automated colony counting, then the plates at that concentration
were counted manually.

Statistics:

For all replicate platings, the mean revertants per plate and the standard deviation were calculated.

Species / strain:

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

Metabolic activation:

with and without

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

cytotoxicity

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Remarks on result:

other: all strains/cell types tested

Remarks:

Migrated from field 'Test system'.

Conclusions:

The results of the Salmonella-Escherichia co/i/Mammalian-Microsome Reverse Mutation Assay Preincubation Method with a Confirmatory Assay indicate that under the conditions of this study, the test article, BADGE-2HCL, did not cause a positive increase in the mean number of revenants per plate with any of the tester strains either in the presence or absence of microsomal enzymes prepared from Aroclor™-induced rat liver (S9). Hence, BADGE-2HCL was considered to be non mutagenic in this assay.

Executive summary:

The objective of this study was to evaluate the test article, Bisphenol A-bis-(3-chlor-2-hydroxypropyl)-ether (BADGE-2HCL), for the ability to induce reverse mutations either in the presence or absence of mammalian microsomal enzymes at 1) the histidine locus in the genome of several strains of Salmonella typhimurium and at 2) the tryptophan locus of Escherichia coli strain WP2«vrA. This assay satisfied the following guidelines: U.S. EPA (1998), EEC (2000), and OECD (1997).

The concentrations tested in the mutagenicity assay were selected based on the results of a , rangefinding assay using tester strains TA 100 and WP2wvrA and ten concentrations of test article ranging from 6.67 to 5000 u,g per plate, one plate per concentration, both in the presence and absence of S9 mix.

The tester strains used in the mutagenicity assay were Salmonella typhimurium tester strains TA98, TA100, TA1535, and TA1537 and Escherichia coli tester strain WP2uvrA. The assay was conducted with a minimum of six concentration levels of test article in both the presence and absence of S9 mix along with concurrent vehicle and positive controls using three plates per concentration. The concentrations tested in the mutagenicity assay with the Salmonella tester strains ranged from 1.00 to 1000 \ig per plate in both the presence and absence of S9 mix (0.333 to 1000 ug per plate with TA 1537 in the absence of S9 mix). The concentrations tested in the

mutagenicity assay with Escherichia coli tester strain WP2uvrA ranged from 10.0 to 5000 ug per plate in both the presence and absence of S9 mix. The results of the initial mutagenicity assay were confirmed in an independent experiment.

The results of the Salmonella-Escherichia co/i/Mammalian-Microsome Reverse Mutation Assay Preincubation Method with a Confirmatory Assay indicate that under the conditions of this study, the test article, BADGE-2HCL, did not cause a positive increase in the mean number of revertants per plate with any of the tester strains either in the presence or absence of microsomal enzymes prepared from Aroclor™-induced rat liver (S9). Hence, BADGE-2HCL was considered to be non mutagenic in this assay.

Reference 2

Endpoint:

in vitro gene mutation study in mammalian cells

Remarks:

Type of genotoxicity: gene mutation

Type of information:

experimental study

Adequacy of study:

key study

Study period:

not specified

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: The GLP study was conducted according to guideline/s

Reason / purpose for cross-reference:

reference to same study

Reason / purpose for cross-reference:

reference to other study

Qualifier:

no guideline followed

Principles of method if other than guideline:

mouse lymphoma assay using L5178Y cell line as described by Clive et al., 1979.

Clive, D., Johnson, K.O., Spector, J.F.S., Batson, A.G. and Brown, M.M.M. (1979). Validation and characterization of the L5178Y/TK+/- mouse lymphoma mutagen assay system. Mutat. Res 59:61-108

GLP compliance:

yes

Type of assay:

mammalian cell gene mutation assay

Target gene:

thymidine kinase gene (TK+/-)

Species / strain / cell type:

mouse lymphoma L5178Y cells

Metabolic activation:

with and without

Metabolic activation system:

The S-9 fraction was obtained from the livers of male Sprague-Dawley rats that were dosed with 500 mg/kg body weight of Aroclor 1254 and killed five days later.

Test concentrations with justification for top dose:

0.0018-0.013 mg/mL nonactivated; 0.032-2.4 mg/L activated

Untreated negative controls:

yes

Negative solvent / vehicle controls:

yes

True negative controls:

not specified

Positive controls:

yes

Remarks:

Ethylmethane sulfonate without S9 and Benzo(a)pyrene with S9

Positive control substance:

ethylmethanesulphonate

Details on test system and experimental conditions:

Cells were treated with various test material concentrations, a negative control (vehicle) and a positive control (Ethylmethane sulfonate) concurrently for toxicity and mutagenicity.

Cells were counted on test day 2 for colony density. Cultures were selected for cloning for treated and both control groups, plated, and cultured for 10 days to determine viability. Mutant counts were recorded, and tutant frequency was calculated.

Species / strain:

mouse lymphoma L5178Y cells

Metabolic activation:

with

Genotoxicity:

negative

Cytotoxicity / choice of top concentrations:

cytotoxicity

Remarks:

toxicity was noted in suspension growth at 0.18 mg/ml and higher

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Species / strain:

mouse lymphoma L5178Y cells

Metabolic activation:

without

Genotoxicity:

positive

Cytotoxicity / choice of top concentrations:

cytotoxicity

Remarks:

toxicity was observed on suspension growth at 0.003 mg/ml and higher

Vehicle controls validity:

valid

Untreated negative controls validity:

valid

Positive controls validity:

valid

Remarks on result:

other: all strains/cell types tested

Remarks:

Migrated from field 'Test system'.

Chemical analyses were performed on samples of EPON 828 used to dose the cells. The nominal concentration of both solutions was 100 mg/ml and the actual amount, as determined by analysis, was 104 +/- 1 mg/ml for Assay II (without S9) and 103 +/- mg/ml for Assay III (with S9). No anal;ysis was done on samples from Assay I as this was used as a preliminary assay to determine toxicity and establish dose levels for the mutagenicity assays (II and III).

In the absence of S-9 and EPON 828, there was a toxic effect on both suspension growth and soft agar growth and an increase in the mutant frequency at concentrations of 0.003 mg/ml and higher. The assay was repeated (Assay II) with 0.0075 mg/ml as the highest concentration and an additional five concentrations in decreasing 1/8 log10 dilutions. The results were similar to those obtained in the preliminary assay. There was evidence of a toxic effect on suspension growth and on soft agar growth commencing at 0.0024 mg/ml and a dose-related increase in mutants at all concentrations in the absense of S-9.

When EPON 828 was exposed to the cells in the presence of metabolic activation, no toxicity was noted on suspension growth or soft agar growth in Assay I at concentrations up to 0.1 mg/ml and there was no increase in the mutant frequency at any concentrations. Therefore in the repeat assay (III) a higher concentration, 0.24 mg/ml, was used and the other doses were set at 1/8 log 10 dilutions therefrom. Toxicity was apparent in suspension growth at 0.18 mg/ml and higher and in soft agar growth at 0.24 mg/ml. No increase in mutant frequency which exceeded 2-fold the background was noted at any concentration tested in the presence of S-9 metabolic activation.

Conclusions:

Interpretation of results (migrated information):
positive

EPON 828 was a direct acting mutagen in the mouse lymphoma gene mutation assay and such activity was eliminated by the addition of a metabolizing enzyme fraction derived from rat liver.

Executive summary:

EPON 828 was tested in the mouse lymphoma cell mutagenicity assay for its ability to induce gene mutation at the thymidine kinae locus in the absence and presence of activation by a rat liver microsome (S-9) fraction. Cells were exposed to concentrations of EPON 828 up to and including those which resulted in toxicity to the cells.

In the absence of activation, EPON 828 in DMSO produced a dose-related cytotoxic effect at concentrations of 0.0024 mg/ml and higher and a dose-related increase in the mutant frequency at concentrations from 0.0018 to 0.0056 mg/ml. The positive control (EMS, 0.8 mg/ml) produced a mutant frequency approximately 20 times that of the untreated control. In the presence of activation EPON 828 produced cytotoxic effects at concentrations of 0.18 mg/ml and higher; however, there was no increase in mutant frequency at any concentration tested. The positive control (P[a]P, 0.005 mg/ml) produced mutant frequency values greater than twofold that of the untreated control.

It can be concluded that APON 828 is a direct acting mutagen in the mouse lymphoma gene mutation assay and that this activity is removed by the addition of a rat liver metabolizing enzyme fraction.

Endpoint conclusion

Endpoint conclusion:

adverse effect observed (positive)

Genetic toxicity in vivo

Description of key information

BADGE was positive in a number of in vitro assays but negative in in vivo assays

Link to relevant study records

Reference

Reference 1

Endpoint:

in vivo mammalian germ cell study: cytogenicity / chromosome aberration

Remarks:

Type of genotoxicity: chromosome aberration

Type of information:

experimental study

Adequacy of study:

key study

Study period:

not stated

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

other: The study was not conducted according to guideline/s and GLP but the report contains sufficient data for interpretation of study results

Reason / purpose for cross-reference:

reference to same study

Reason / purpose for cross-reference:

reference to other study

Qualifier:

no guideline followed

Principles of method if other than guideline:

Dominant lethal study in which male mice were orally gavaged with test material and subsequently mated with untreated female mice over a period of six weeks. The number of females pregnant and number of offspring in each litter was determined.

GLP compliance:

no

Type of assay:

rodent dominant lethal assay

Species:

mouse

Strain:

other: Tif: MAG f (SPF)

Sex:

male

Details on test animals or test system and environmental conditions:

Animals were 3-4 months of age at the time of test, were fed a standard rodent diet and water ad libitum, and were kept in environmentally-adequate housing facilities.

Route of administration:

oral: gavage

Vehicle:

polyethylene glycol (PEG 400)

Details on exposure:

The test material was administered orally in single doses to 20 male albino mice per group, which were then mated to untreated females from the same strain over a period of 6 weeks. At the end of each week, the 2 females per male were replaced by new ones, repeated for 6 weeks to cover the stages of the maturation of the germ cell from the A-spermatogonia to the mature spermatozoon. Doses of 3333 mg/kg and 10,000 mg/kg were given in polyethylene glycol (PEG 400). A control group was given only the vehicle.

Duration of treatment / exposure:

One dose. Test material was dissolved in polyethylene glycol 400.

Frequency of treatment:

once

Post exposure period:

Each male mouse was allowed to mate with two untreated females beginning six hours after receiving a single oral dose of test material. Each group of two untreated females remained with the treated male mouse for a week. After one week the females were removed and replace by another group of two untreated females.

Remarks:

Doses / Concentrations:
Doses of 3333 mg/kg and 10,000 mg/kg
Basis:
nominal conc.

No. of animals per sex per dose:

20 male mice/group

Control animals:

yes, concurrent vehicle

Tissues and cell types examined:

Females were necropsied on the 14 day of gestation. The number of live embryos and embryonic deaths were listed. In addition, the uteri were placed in a solution of ammonium sulphide in order to detect sites of early embryonic resorptions.

Statistics:

To compare the total number of implantations indicating pre-implantation loss, the t-test or Mann-Whitney's U-test was used. The total numbers of mated and pregnant dams or embryonic deaths were compared with the aid of the X2-test or Fisher's exact test.

Sex:

male

Genotoxicity:

negative

Toxicity:

yes

Remarks:

In-life observations included diarrhea in males of the high dose group two days after treatment.

Vehicle controls validity:

valid

Additional information on results:

In-life observations included diarrhea in males of the high dose group two days after treatment. There were no adverse effects on females associated with any of the groups.

The data on mating ratio, numbers of implantations, and embryonic deaths are comparable for all groups.

The data on mating ratio, on the numbers of implantations and embryonic deaths were comparable for all groups.

The diarrhea observed two days after dosing male mice was most likely the result of the vehicle rather than the test material.

Conclusions:

Interpretation of results (migrated information): negative
The test material was negative in the dominant lethal assay.

Executive summary:

A mouse dominant lethal study was conducted with DGEBPA. There was no evidence of an effect on male fertility or number of offspring/litter.

Reference 2

Endpoint:

in vivo mammalian somatic cell study: gene mutation

Type of information:

experimental study

Adequacy of study:

key study

Study period:

05 December 2017 - 11 December 2018

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Remarks:

We confirm that the laboratory utilized standardized approaches to dose formulation and analysis in accordance with their SOPs and that the analytical data clearly support that the objective to administer BADGE at the targeted dose levels was achieved.

Qualifier:

according to guideline

Guideline:

OECD Guideline 488 (Transgenic Rodent Somatic and Germ Cell Gene Mutation Assays)

Version / remarks:

2013

Deviations:

no

GLP compliance:

yes (incl. QA statement)

Type of assay:

transgenic rodent mutagenicity assay

Species:

rat

Strain:

other: F344 Big Blue

Details on species / strain selection:

Rats have been used historically in safety evaluation and genotoxicity studies and are recommended by regulatory agencies. Because this study was conducted in accordance with regulatory guidelines, alternatives could not be considered.
The Big Blue® in vivo mutation assay is a Transgenic Rodent (TGR) Mutation assay described in OECD Test Guideline 488 (OECD, 2013). TGR assays in general, and the Big Blue® assay in particular, have been reviewed by OECD (OECD, 2009 and 2011a) and are identified in OECD Test Guideline 488 (OECD, 2011b and OECD, 2013) as being appropriate to investigate in vivo mutagenicity in any tissue of interest. In addition, the TGR assays are recommended to investigate a potential mutagenic mode of action in the etiology of rodent tumors.

Sex:

male

Details on test animals or test system and environmental conditions:

TEST ANIMALS
- Source: BioReliance colony housed at Taconic Biosciences, Inc., Germantown, NY
- Age at study initiation: 9-10 weeks (initial cohort), 13-14 weeks (extended cohort)
- Weight at study initiation: 213.1 to 249.2 grams (initial cohort), 236.3 to 333.6 grams (extended cohort)
- Assigned to test groups randomly: yes, under following basis: by body weight
- Housing: multiple-housed during acclimation and following randomization in polycarbonate cages
- Diet (e.g. ad libitum): TEKLAD Global Diet #2018C (Certified 18% Protein Rodent Diet, Envigo, Madison, WI) in pellet form, in stainless steel rodent feeders, ad libitum
- Water (e.g. ad libitum): ad libitum
- Acclimation period: 13 or 40 days prior to the first dose administration, for the initial or extended cohorts, respectively

ENVIRONMENTAL CONDITIONS
- Temperature (°C): 20.5 to 24.0ºC
- Humidity (%): 30 to 70%
- Air changes (per hr): at least 10
- Photoperiod (hrs dark / hrs light): 12/12

Route of administration:

oral: gavage

Vehicle:

- Vehicle(s)/solvent(s) used: 0.5% Methocel A4M methyl cellulose ethers with 0.1% Tween 80 in deionized water
- Amount of vehicle (if gavage or dermal): 10 mL/kg bw
- Type and concentration of dispersant aid (if powder):
- Lot/batch no. (if required): Methocel A4M: BCBR9701V; Tween 80: MKBQ9736V

Justification for vehicle:
Justification for the use of 0.5% Methocel A4M methyl cellulose ethers with 0.1% Tween 80 in deionized water as the vehicle for formulation of suspensions of BADGE includes the following considerations:
- Oral gavage in a vehicle was specified in the ECHA decision and is consistent with previous repeated-dose animal testing
- 0.5% Methocel A4M methyl cellulose ethers with 0.1% Tween 80 in deionized water as vehicle for dose formulation was used for the following repeated-dose animal testing: 28- and 90-day oral toxicity, 24-month chronic toxicity and carcinogenicity testing, 1- and 2-Generation Reproductive Toxicity, and OECD 414 Developmental toxicity testing.
- The ability, as demonstrated in the prior toxicity studies and as demonstrated in this study, to prepare stable dose formulations at the targeted dose concentrations (Text Table 1 summarizes the analytical results for concentration verification, homogeneity, and stability conducted on the current study). Additional details on dose formulation and analysis including chromatograms are provided in other sections of this dossier and in attachments.

Details on exposure:

PREPARATION OF DOSING SOLUTIONS:
An appropriate amount of test substance (no correction factor was used) along with approximately 70% of the required amount of vehicle were weighed and added to the beaker; the contents were stirred until homogeneous in appearance.
The remaining amount of vehicle was added and the contents were again stirred until homogeneous in appearance. The contents were homogenized using a polytron (if needed); the mix was further homogenized by sonication. After sonication, the formulations were heated between 35-45°C, while stirring. BADGE formulations were stored at 37°C, protected from light, prior to delivering to the animal facility for dosing, or when not in use in the animal facility. Dose formulations were stirred for at least 30 minutes prior to use for dosing as well as during dose administration. This process is standard methodology to ensure resuspension and homogeneity.

Justification for the 3-day dose formulation preparation and use schedule.
BADGE formulations were prepared at least once every 3 days.

Concentration verification for each dose level was performed for the first and last dose formulations used in each of the two study phases.

Dose formulation homogeneity was evaluated for each dose level for the first dose formulation of each study phase and the high and low dose level for the last dose formulation of the first study phase. Homogeneity evaluations were performed by analyzing two samples each from the top, middle, and bottom levels of the dose formulation. Thus, there were 6 samples in total, 2 from each location (top, middle, bottom). The Relative Standard Deviation was calculated for each level as well as for the average of all 3 levels.

The range, as indicated in the Text Table 1 (below), represents the highest and lowest average value seen among the top, middle, and bottom levels. There was a wider than anticipated degree of variability between sampling levels for the high-dose group (1000 mg/kg) in Phase 1. Homogeneity at this level improved in Phase 2 as demonstrated by the results of the homogeneity analysis in that phase. The High-dose group from Phase 1 was not used in the assessment of mutagenicity.


Stability testing was conducted on the first and last dose formulations prepared in “Phase 1” and then once (first dose formulation) during Phase 2. Unfortunately, the first stability test in Phase 1 was inadvertently performed at room temperature and thus not relevant to stability of the dose formulations which were maintained at 37C. The second stability test was conducted at 37C however the results were not reported in the study report because the T=0 homogeneity criteria were not met. We have added the T=3-day results from the second stability analysis to Text Table 1 (below) for completeness. The stability test performed in Phase 2 was also done at 37C and demonstrated stability over the dose range.

Justification for utilization of prepared dose formulations over the 3-day period is derived from the stability that was demonstrated on this study as well as from the known chemical/physical properties of BADGE and the stability of such dose formulations utilized in several other repeated dose toxicity studies.
As reported in the study report (page 159) and presented herein (Text Table 2) and in the attachment (Supplemental information Regarding Dose Formulation and Analysis for the study titled: In Vivo Mutation Assay with BADGE (ERC#1) at the cII Locus in Big Blue® Transgenic F344 Rats ) the stability of formulations prepared at the high (100 mg/ml) and low (5 mg/ml) dose concentrations were analyzed directly after preparation and then again after a 3-day holding period at 37C. The concentration of BADGE after the 3-day period was essentially unchanged at both concentration levels. The concentration of BADGE in the low dose level was 4.66 mg/ml at T = 0 and 4.71 mg/ml after 3 days. The concentration of BADGE in the high dose level was 121 mg/ml at T = 0 and 120 mg/ml after 3 days. The stability demonstrated by the analysis was further supported by the lack of any emergence of additional peaks in the chromatography as would be expected if BADGE was undergoing hydrolysis during the 3-day maintenance period. Example chromatograms from the evaluation of the high dose level are presented in the attached supplemental document.
Loss of BADGE due to hydrolysis is possible however the extent of such loss in the dose formulations is severely limited by low solubility and modest hydrolysis rate. The water solubility of BADGE has been determined experimentally, 6.9 mg/L at 20⁰C. Though the vehicle for dose formulations include a low concentration of Methocel (0.5%) and Tween 80 (0.1%) the vehicle was principally distilled-deionized water. As shown in the table below the concentration of BADGE in each dose formulation was in far excess of the water solubility (833 to 16,666-fold excess). Under these conditions it is understandable that a suspension of BADGE would result and that the bulk of the test material would have limited contact with the aqueous phase.
The epoxy functionality of BADGE is susceptible to hydrolysis and the half-life of BADGE in water at pH 7 was determined at 30⁰C (73.5 hrs) and 40⁰C (28 hrs). From these data one can estimate that the hydrolysis half-life at 37⁰C, the temperature at which dose formulations were maintained, to be approximately 42 hrs. Thus, BADGE dissolved in the aqueous phase of the dose solutions would be expected to undergo hydrolysis and with a half-life of approximately 42 hrs. However, as hydrolysis would have been limited to BADGE dissolved in the aqueous vehicle, which is limited by the low water solubility, the total mass susceptible to hydrolysis over the 3 day holding period represents only a small fraction of the total mass of BADGE in any of the dose formulations (dissolved (ug/ml) + suspension (mg/ml)). As a conservative estimation, assuming a dose formulation solubility for BADGE of 10 ug/ml, and two half-lives over the 3-day dosing period, the total loss of BADGE to hydrolysis for any of the dosing solutions would be no more than 20 ug/ml. This would amount to a loss of only 0.4% of the total mass of BADGE in the 5 mg/ml dose solution.

The sum total of the evidence that includes the practical demonstration of 3-day dose formulation stability under the actual use conditions within the study provide justification for the dose formulation schedule and use. The use of this 3-day dose formulation schedule as compared with daily dose preparation does not impact on the validity of the study.

Additional Notes:
Results for stability testing at room and refrigerated temperatures were reported in the study report (Study Report Appendix C). These results were included in the report because they were conducted, however, stability at room and refrigerated temperatures were not relevant to the dose formulations used to dose animals on study and were not pursued further. Additionally, the study report misstated the results of the 3-day at 37C stability evaluation. The report will be amended to correctly state that stability assessments for formulations held at room temperature and at cold temperatures did not meet acceptance criteria. The stability assessment for dose formulations held at 37C were stable and met the acceptance criteria. Acceptance criteria for formulation analysis of concentration, stability and homogeneity are defined by the laboratory's SOP. Acceptance criteria are specific to the character of the formulation. Solutions have narrower acceptance limits than suspensions. Acceptance criteria initially stated in the study protocol was set with the anticipation that dose formulations would be solutions. The protocol was amended with acceptance criteria deemed appropriate by the laboratory for suspensions.

BADGE is known to solidify at low ambient temperatures as was apparent during the days before the initial dosing of animals. As such the Sponsor recommended that the bulk test article be warmed to between 40 – 60C to acquire liquified samples for formulation. This realization occurred before preparation of dose formulations used on day 1. The laboratory observed improved ease with formulation preparation and use by incorporating warming to 37C into the process of dose formulation and maintenance.

Duration of treatment / exposure:

28 d (except 1000 mg/kg bw/d dose group: 25 d; positive control: days 1, 2, 3, 12, 19, and 26)

Frequency of treatment:

daily (except positive control: days 1, 2, 3, 12, 19, and 26)

Dose / conc.:

50 mg/kg bw/day (actual dose received)

Dose / conc.:

250 mg/kg bw/day (actual dose received)

Dose / conc.:

1 000 mg/kg bw/day (actual dose received)

Dose / conc.:

500 mg/kg bw/day (actual dose received)

Remarks:

(extended phase)

Dose / conc.:

1 000 mg/kg bw/day (actual dose received)

Remarks:

(extended phase)

No. of animals per sex per dose:

6

Control animals:

yes, concurrent vehicle

Positive control(s):

ethylnitrosurea
- Route of administration: oral(gavage)
- Doses / concentrations: 20 mg/kg/day

Tissues and cell types examined:

liver, duodenum, and glandular stomach were collected for cII mutant analysis; testes and cauda were also collected but not analyzed for mutants
In addition, for Groups 6-8 (extended cohort), the median lobe of the liver (with the associated mass), approximately one third of the glandular stomach, and one of the three 1-inch sections of the duodenum, were saved in 10% neutral buffered formalin (10% NBF) for possible future staining and microscopic evaluation.

Details of tissue and slide preparation:

CRITERIA FOR DOSE SELECTION:
Dose levels were also selected based on available toxicity data in rats from a 28-day study, where 40 Ralf (SPF) rats, 5 males and 5 females per dose group, were administered BADGE daily by gavage for 28 days at doses of 0, 50, 200, or 1000 mg/kg bw per day. A no observed effect level (NOEL) of 1000 mg/kg bw was identified after 28 days of repeated, once daily oral gavage administration.

DETAILS OF DNA PREPARATION:
Isolated DNA was processed using Packaging Reaction Mix (PRM), purchased from New York University, New York, NY. This product is similar to Transpack manufactured by Agilent, Santa Clara, CA. PRM or Transpack were used to isolate the recoverable lambda shuttle DNA vectors from the genomic DNA and to package the lambda shuttle vector DNA using phage proteins and cofactors to create infectious lambda phage particles. Methods followed BioReliance SOP’s, based on Agilent instruction manual titled “λ Select-cII Mutation Detection System for Big Blue® Rodents” (Agilent, 2015) and Agilent instruction manual titled “Transpack Packaging Extract for Lambda Transgenic Shuttle Vector Recovery” (Agilent, 2009b).

METHOD OF ANALYSIS:
Isolated DNA was processed using Packaging Reaction Mix (PRM), which is used to isolate the recoverable lambda shuttle DNA vectors from the genomic DNA. Phage head, tail, and tail fibers from the packaging mix are then assembled around lambda shuttle vector DNA creating infectious lambda phage particles.
Packaged phage were incubated overnight at 37 ± 1.0°C, and then scored for plaque formation and titer determination; cII mutant selection plates were incubated for two days (nominally, 40-48 hours) at 24 ± 0.5°C, and then scored for mutant plaques. At least 125,000 phage were evaluated from at least 2 packagings for each dose and tissue.

The individual animal is considered the experimental unit. The mutant frequency (MF) was calculated (number of mutant phage / number of total phage screened) for each tissue analyzed from each animal. Since this ratio is extremely small and may not be normally distributed, a log10 transformation of the MF data was performed.
The statistical analysis of MF was conducted as follows: the positive control (Group 5) was independently compared to either of the vehicle controls (Group 1 and Group 6). In the second part of the analysis, test substance-treated groups were compared to their concurrent vehicle controls (i.e., Groups 2-3 vs. Group 1 and Groups 7-8 vs. Group 6). Lastly, the extended phase vehicle control (Group 6) was analyzed against the initial vehicle control (Group1), in order to assess the impact of potential differences in background mutation rates between phases.
In all instances, log10-transformed MF data from the vehicle control and treated groups were evaluated using a One-Way Analysis of Variance (ANOVA). The suitability of using the parametric ANOVA was confirmed by testing parameters of the log10-transformed MF data for normality and equal variance. If the data were normally distributed and exhibit equal variance, the parametric ANOVA analysis would be used; if either test failed, a
non-parametric method would be used.

Evaluation criteria:

Validity criteria:
Vehicle control values: The average mutant frequency of the vehicle controls should be within reasonable limits of the laboratory historical controls and literature values.
Positive control values: the positive control must induce a statistically significant increase in mutant frequency as compared with the concurrent vehicle control (P<0.05 will be considered significant) .

Criteria for a positive response: The test item will be considered to have produced a positive response if it induces a statistically significant increase in the frequency of cll mutants in any dose level outside the 95% control limits of the historical background mutant frequency range. Biological significance will be an important consideration ion the final determination of a positive response.

Criteria for a negative response: A test item will be considered to have produces a negative response if no significant increase in cll mutant frequency is observed.

Criteria for an equivocal response: equivocal responses will be evaluated by the study director on a case-by-case basis considering both statistical significance and biological relevance.

Statistics:

The incidence of all effects was analyzed separately by dose level. Dunnett’s test was conducted on body weight, body weight changes, and organ weight data. All statistics compared treated groups versus their concurrent control (i.e., Groups 2-5 vs. Group 1 and Groups 7-8 vs. Group 6), and were based on a significance value of p < 0.05.

Sex:

male

Genotoxicity:

negative

Toxicity:

yes

Vehicle controls validity:

valid

Positive controls validity:

valid

Big Blue® Assay Results

Summary of Mutant Frequency Data

Dose Level (mg/kg/day)	Liver Mean [Median] ± SD (x 10-6)	Duodenum Mean [Median] ± SD (x 10-6)	Glandular Stomach Mean ± SD (x 10-6)
0 (Vehicle control, Initial phase)	25.1 [26.5] ± 7.7	43.3 [40.0] ± 18.8	24.9 ± 4.5
50	24.7 ± 7.2	40.2 ± 20.1	19.1 ± 3.5
250	24.0 ± 6.3	29.3 ± 9.6	23.4 ± 6.9
20 mg/kg ENU (a)	239.1**### ± 65.2	840.9** [826.0##] ± 85.8	512.7**### ± 77.6
0 (Vehicle control, Extended phase)	36.2 [37.6*] ± 5.2	34.6 [34.2] ± 12.4	31.2 ± 9.3
500	25.0# ± 3.8	26.4 ± 8.9	19.7 ± 0.9
1000	34.9 ± 10.0	46.5 ± 12.0	28.5 ± 10.9

a = Days 1, 2, 3, 12, 19, and 26 only; SD = Standard Deviation.

* = Statistically significant (Kruskal-Wallis test, p < 0.05) compared to Group 1.

** = Statistically significant (One-way ANOVA, p < 0.001) compared to Group 1.

# = Statistically significantly lower (One-way ANOVA, p < 0.05) compared to Group 6.

## = Statistically significant (Kruskal-Wallis test, p < 0.01) compared to Group 6.

### = Statistically significant One-way ANOVA, p < 0.001) compared to Group 6.

 

 

Mutant frequencies (MF) for liver, duodenum, and glandular stomach in BADGE-treated animals were not statistically elevated over the correspondent controls at any dose level. MF in ENU-treated animals was statistically elevated over both vehicle controls for liver, duodenum, and glandular stomach, demonstrating the responsiveness of the test system to ENU, a directacting mutagen.

 

 

Formulation stability

BADGE formulated in 0.5% Methocel A4M methyl cellulose ethers with 0.1% Tween 80 in DI water, at concentrations of 4.66 and 121 mg/mL, was stable at 37°C for at least 3 days.

 

Mortality

Group 4 (1000 mg/kg/day) animals (initial phase) were terminated on Day 25, as directed by protocol amendment #2; all other animals survived until their scheduled terminal sacrifice on Day 31.

 

Clinical Signs

There were no remarkable clinical observations associated with BADGE treatment at doses up to and including 500 mg/kg/day in either the initial or extended phases.

Text Table 1 shows the comparison of post-dose cage-side observations between the 1000 mg/kg/day groups in the initial phase and the extended phase.

During the initial phase, severe signs of toxicity were noted post-dose (cage-side or unscheduled observations) at 1000 mg/kg/day (Group 4), beginning with Day 6, and these included: decreased motor activity, ruffled fur, hunched posture, squinty eyes, labored breathing, and diarrhea. After Day 9 however, most animals appeared normal. Diarrhea was also noted in two animals during detailed hands-on observations at 1000 mg/kg/day, on Day 8.

During the extended phase, signs of toxicity noted during cage-side or unscheduled observations (at 1000 mg/kg/day) included decreased motor activity, ruffled fur, hunched posture, and squinty eyes, starting with Day 9. All animals appeared normal after Day 12.

Thin appearance was also noted during hands-on observations at 1000 mg/kg/day (Group 8), on Day 8. No labored breathing or diarrhea were noted in the extended phase.

All other observations noted were unremarkable, and likely unrelated to the test substance administration as they also occurred in the control group. Red discharge from the left eye noted in a single animal at detailed hands-on observations on Days 22, 29, and 31, and as an unscheduled observation on Day 27, was considered an accidental injury and thus incidental to BADGE treatment. Overall, the occurrence of the signs of toxicity during cage-side or unscheduled observations seemed reduced in incidence and frequency in Group 8 compared to Group 4 with no diarrhea (evidence of severe toxicity) reported in Group 8 compared to Group 4.

 

Body Weights and Body Weight Gains

There were no statistically significant or otherwise remarkable differences in mean body weight between the concurrent control, and the BADGE-treated groups up to and including 500 mg/kg/day in both the initial and extended phase. In the extended phase, the mean body weight of the 1000 mg/kg/day animals (Group 8) was also not statistically significantly different from the concurrent control.

During the initial phase, toxicologically and statistically significant lower mean body weights were noted for the 1000 mg/kg/day dose group, starting with Day 8 (14% lower than control and a weight loss of 10.1% of the mean Day 1 body weight) and continuing through the last body weight on Day 22 (13% lower than the vehicle control). The severe, acute body weight losses coupled with the clinical observations occurring in the same time frame, were considered in the decision to terminate Group 4 and start the extended phase.

Although there were few statistically significant differences in comparisons of mean body weight of the BADGE-treated groups, there were notable, statistically significant differences in body weight gains that were related to BADGE treatment at doses of 250 mg/kg/day and higher.

In the initial phase, the mean body weight gain of the 250 mg/kg/day group was statistically significantly lower (25% of control) than the concurrent control in the interval from Day 22 to 29, but the overall weight gain (Day 1 to 31), was similar.

During the extended phase, the body weight changes at both 500 and 1000 mg/kg/day tended to be significantly lower with weight loss (2.4% and 5.7%, respectively) relative to the initial body weight, more notably during the first week of treatment. The overall (Day 1-31) mean body weight gains were 42% and 66% lower than control at 500 and 1000 mg/kg/day, respectively, and both of these were statistically significant differences.

A dose of 50 mg/kg/day was a clear no effect level for BADGE-related body weight changes.

For ENU-treated animals, the overall (Day 1-31) mean body weight gains were statistically significantly different, and 25% lower as compared to the vehicle control group.

 

Gross Necropsy Findings

A few gross observations were made in the livers of two control and two 1000 mg/kg/day animals in the extended phase; among these observations, a firm mass was noted in the median lobe of the liver for one 1000 mg/kg/day male. None of the gross necropsy findings were considered treatment-related.

 

Organ Weight Analysis

Organ weights were collected mainly for predicting the number of possible DNA extractions from a tissue, not for toxicity evaluation. Any statistically significant differences between BADGE-treated and control organ weights are considered incidental findings.

Conclusions:

There was no treatment-related mortality and no evidence of an increase in mutant frequency at the cII gene in liver, duodenum, or glandular stomach of F344 Big Blue® male rats after 28 days of once daily oral gavage treatment with BADGE, at doses ranging from 50 mg/kg/day up to the limit dose of 1000 mg/kg/day.

Executive summary:

This study investigated the effect of BADGE (Bisphenol A Diglycidyl Ether; CAS 1675-54-3) on mutant frequency at the cII gene in liver, glandular stomach, and duodenum from male transgenic Fischer 344 (F344) Big Blue® rats. The Big Blue® Assay is a Transgenic Rodent (TGR) Mutation assay described in OECD Test Guideline 488 (OECD, 2013).

The study consisted of 47 transgenic F344 Big Blue® male rats assigned to eight groups, as detailed below. The initial study design used 5 groups of 6 male rats each: one vehicle control (Group 1), 3 BADGE-treated groups (2, 3 and 4) and one positive control (Group 5) (initial phase). Due to signs of excessive toxicity coupled with concerns regarding formulation homogeneity, the high dose (Group 4) animals were terminated early, and a second cohort (extended phase) of 17 male rats was added, drawn from the same breeding group to include: one vehicle control (Group 6, 5 animals),

and two BADGE-treated groups (Groups 7 and 8, 6 animals each) (extended phase).

Groups 1 and 6 received the vehicle (0.5% Methocel A4M methyl cellulose ethers with 0.1% Tween 80 in deionized water). Test substance-treated animals received BADGE, formulated in the vehicle, at the dose levels presented in the table below. Animals in Groups 1-3 and 6-8 were dosed once daily via oral gavage for 28 consecutive days. Animals in Group 4 were dosed once daily by oral gavage for 25 consecutive days, except on Day 8 (as directed by the Study Director). Positive control animals (Group 5) receivedN-ethyl-N-nitrosourea (ENU) in buffer solution, pH 6.00 by oral gavage at 20 mg/kg/day, on Days 1, 2, 3, 12, 19, and 26. All doses were administered based upon body weight at a volume of 10 mL/kg body weight (bw).

 

Group	Dose Levels (mg/kg/day)	Number of Males
Group 1 (Vehicle control)	0	6
Group 2	50	6
Group 3	250	6
Group 4	1000	6
Group 5 (Positive control, ENU)	20 (Days 1, 2, 3, 12, 19, and 26)	6
Group 6 (Vehicle control, extended phase)	0	5
Group 7 (extended phase)	500	6
Group 8 (extended phase)	1000	6
Total animals		47

 

The Group 4 animals were terminated by carbon dioxide (CO2) overdose on Day 25, and discarded without necropsy. The remaining animals were terminated by CO2 overdose on Day 31. A partial necropsy was performed for animals in Groups 1-3 and 5-8; the liver, duodenum, glandular stomach, testes, and cauda were collected, weighed, flash frozen, and stored at or below -60°C.

Liver, duodenum, and glandular stomach from the first five surviving animals/group were processed for DNA isolation and analysis of cII mutants, following BioReliance SOPs. Samples of liver, duodenum, and glandular stomach from the extended study were retained in 10% neutral buffered formalin (10% NBF) for possible histopathology.

All animals survived to their scheduled termination. There were no remarkable, BADGE-related clinical observations at dose levels up to and including 500 mg/kg/day. At 1000 mg/kg/day, BADGE-related clinical observations included transient observation of decreased motor activity, ruffled fur, hunched posture, and squinty eyes that began on Day 6 or Day 9 for the initial and extended dosing phases, respectively. Slight to moderate diarrhea and labored breathing were noted on Days 6, 7, and 8 (and once on Day 19) only at 1000 mg/kg/day during the initial phase, but these effects were not observed during the extended phase.

There were no statistically significant or otherwise remarkable differences in mean body weight between the concurrent control, and the BADGE-treated groups up to and including 500 mg/kg/day in both the initial and extended phases. In the extended phase, the mean body weight of the 1000 mg/kg/day animals was also not statistically significantly different from the concurrent control. Although there were several statistically significant differences of mean body weights and/or body weight gains (bwg) for the BADGE-treated group that was terminated early on Day 25 (Group 4, 1000 mg/kg/day), there were also a few intervals of statistically significantly lower absolute bwg compared to the concurrent control, that were related to BADGE treatment, among the other groups: a single interval (Days 22-29) at 250 mg/kg/day in the initial phase, and several intervals, including Days 1-31 for both 500 and 1000 mg/kg/day, in the extended phase. The bwg data also included single intervals for each dose level with significantly higher bwg values compared to controls.

Repeated treatment with BADGE, up to a limit dose of 1000 mg/kg/day did not result in elevated mutant frequencies (MF) at thecIIgene in liver, duodenum, or glandular stomach of F344 Big Blue® male rats. The lack of mutation induction in these portals of entry and systemic tissues obviated the need to analyze the testes or cauda for mutations. The treatment with ENU produced statistically significant increases in MF for all tissues evaluated, demonstrating the utility of the test system to detect and quantify induced mutants following exposure to a known direct-acting mutagen. The study design and results obtained met protocol-specified assay acceptance criteria and were consistent with the study requirements of OECD TG 488 for transgenic rodent mutation assays.

In conclusion, there was no treatment-related mortality and no evidence of an increase in mutant frequency at the cII gene in liver, duodenum, or glandular stomach of F344 Big Blue® male rats after 28 days of once daily oral gavage treatment with BADGE, at doses ranging from 50 mg/kg/day up to the limit dose of 1000 mg/kg/day.

Endpoint conclusion

Endpoint conclusion:

no adverse effect observed (negative)

Additional information

In the key studies, BADGE produced a positive response in strains TA100 and TA1535 with and without metabolic activation. It was also positive in strain TA1537 with metabolic activation. It was positive in yeast with and without metabolic activation and was also positive in a chromosomal aberration assay using rat liver. A number of in vivo assays were conducted and all were negative. These included, mouse micronucleus, dominant lethal, chromosome aberration, mouse spermatocytes and DNA damage/repair.

In vitro tests

Bacterial reverse mutation assay

The results of the Salmonella-Escherichia coli/Mammalian-Microsome Reverse Mutation Assay Preincubation Method with a Confirmatory Assay indicate that under the conditions of this study, the test article, BADGE-2HCL, did not cause a positive increase in the mean number of revenants per plate with any of the tester strains either in the presence or absence of microsomal enzymes prepared from Aroclor™-induced rat liver (S9). Hence, BADGE-2HCL was considered to be non mutagenic in this assay.

 

Several other bacterial reverse mutation assays indicate that the compound does not induce frameshift mutations, but has the ability to induce base pair substitutions in the presence and absence of metabolic activation.

 

in vitro gene mutation study in mammalian cells

Positive and negative results were obtained in the mouse lymphoma gene mutation assay with and without metabolic activation.

 

In vitro chromosome aberration assays

The effect of Diglycidyl ether of bisphenol A was studied in rat liver cells without metabolic activation. Cells in metaphase were analyzed for structural chromosome damage. BADGE is considered to lead to the induction of chromosome damage in vitro.

 

The effect of DGEBPA on the formation of chromosomal aberrations in human peripheral lymphocytes was examined. The test material did not cause a significant increase over control levels of chromatid aberrations at dose levels as high as 50 µg/ml. At 100 µg/ml cytotoxicity was noted as only a few mitotic cells were scored.

 

BADGE is reported to be very stable in the absence of metabolizing enzymes (S9 mix) but is rapidly hydrolyzed to a diglycol (BADGE*2H2O) by S9 mix. BADGE but not BADGE*2H2O, induced micronuclei and gene mutations at the HPRT locus in cultured Chinese hamster V79 cells. The induced micronuclei consisted of acentric chromosomal fragments and did not contain whole chromosomes/chromatids, as was shown by staining with CREST antikinetochore antibodies. It is concluded that BADGE exhibits clastogenic and mutagenic potential, which is lost after hydrolysis of the epoxide rings and converted to aneuploidogenic potential after cleavage to bisphenol A.

 

in vitro DNA damage and/or repair study

In the in vitro rat primary hepatocyte unscheduled DNA synthesis (UDS) assay, the test material, diglycidyl ether of bisphenol A (DGEBPA), did not induce significant increases in UDS. In the assay described in this report, freshly prepared rat hepatocytes were exposed to DGEBPA at concentrations ranging from 255 ug/ml to 0.005 ug/ml in the presence of 5 uCi/ml 3HTdr (20 Ci/mmole). The test material was insoluble in media from 255 ug/ml to 10.2 ug/ml but appeared soluble at lower concentrations. Treatments from 51.1 to 255 ug/ml were not analyzed for nuclear labeling due to high toxicity. Cultures treated with concentrations of test material ranging from 0.511 to 25.5 ug/ml covered a good range of toxicity (79.5 to 100% survival) and were selected for analysis of nuclear labeling. None of the treatments with the test material caused increases that were considered different from the control cultures. DGEBPA was therefore evaluated as inactive in the rat primary hepatocytes UDS assay.

 

in vitro mammalian cell transformation assay

The mutagenic activity of epoxy resins to induce neoplastic transformation in cultured cells was investigated in a clone of baby hamster kidney (BHK) cells using the soft agar cloning technique. Diglycidyl ether of bisphenol A (DGEBPA) and EPIKOTE 828 induced neoplastic transformation in BHK cells.

 

In vivo tests

Transgenic rodent mutagenicity assay

This study investigated the effect of BADGE on mutant frequency at the cII gene in liver, glandular stomach, and duodenum from male transgenic Fischer 344 (F344) Big Blue® rats in accordance with OECD Test Guideline 488.

The initial study design used 5 groups of 6 male rats each: vehicle control, 50, 250, and 1000 mg/kg bw/d, positive control (20 mg/kg bw/d ENU). Due to signs of excessive toxicity coupled with concerns regarding formulation homogeneity, the high dose animals were terminated early, and a second cohort (extended phase) of 17 male rats was added, drawn from the same breeding group to include: one vehicle control (5 animals), and two BADGE-treated groups (500, and 1000 mg/kg bw/d, respectively) (extended phase).

Animals were dosed once daily via oral gavage for 28 consecutive days, positive control animals (Group 5) receivedN-ethyl-N-nitrosourea (ENU) in buffer solution, pH 6.00 by oral gavage at 20 mg/kg/day, on Days 1, 2, 3, 12, 19, and 26. Liver, duodenum, and glandular stomach from the first five surviving animals/group were processed for DNA isolation and analysis of cII mutants.

All animals survived to their scheduled termination. There were no remarkable, BADGE-related clinical observations at dose levels up to and including 500 mg/kg/day. At 1000 mg/kg/day, BADGE-related clinical observations included transient observation of decreased motor activity, ruffled fur, hunched posture, and squinty eyes that began on Day 6 or Day 9 for the initial and extended dosing phases, respectively. Slight to moderate diarrhea and labored breathing were noted on Days 6, 7, and 8 (and once on Day 19) only at 1000 mg/kg/day during the initial phase, but these effects were not observed during the extended phase.

There were no statistically significant or otherwise remarkable differences in mean body weight between the concurrent control, and the BADGE-treated groups up to and including 500 mg/kg/day in both the initial and extended phases. In the extended phase, the mean body weight of the 1000 mg/kg/day animals was also not statistically significantly different from the concurrent control. Although there were several statistically significant differences of mean body weights and/or body weight gains (bwg) for the BADGE-treated group that was terminated early on Day 25 (Group 4, 1000 mg/kg/day), there were also a few intervals of statistically significantly lower absolute bwg compared to the concurrent control, that were related to BADGE treatment, among the other groups: a single interval (Days 22-29) at 250 mg/kg/day in the initial phase, and several intervals, including Days 1-31 for both 500 and 1000 mg/kg/day, in the extended phase. The bwg data also included single intervals for each dose level with significantly higher bwg values compared to controls.

Repeated treatment with BADGE, up to a limit dose of 1000 mg/kg/day did not result in elevated mutant frequencies (MF) at the cII gene in liver, duodenum, or glandular stomach of F344 Big Blue® male rats. The lack of mutation induction in these portals of entry and systemic tissues obviated the need to analyze the testes or cauda for mutations. The treatment with ENU produced statistically significant increases in MF for all tissues evaluated, demonstrating the utility of the test system to detect and quantify induced mutants following exposure to a known direct-acting mutagen. The study design and results obtained met protocol-specified assay acceptance criteria and were consistent with the study requirements of OECD TG 488 for transgenic rodent mutation assays.

In conclusion, there was no treatment-related mortality and no evidence of an increase in mutant frequency at the cII gene in liver, duodenum, or glandular stomach of F344 Big Blue® male rats after 28 days of once daily oral gavage treatment with BADGE, at doses ranging from 50 mg/kg/day up to the limit dose of 1000 mg/kg/day.

 

Rodent dominant lethal assay

A mouse dominant lethal study was conducted with DGEBPA. The test material was administered orally in single doses to 20 male albino mice per group, which were then mated to untreated females from the same strain over a period of 6 weeks. At the end of each week, the 2 females per male were replaced by new ones, repeated for 6 weeks to cover the stages of the maturation of the germ cell from the A-spermatogonia to the mature spermatozoon. Doses of 3333 mg/kg and 10,000 mg/kg were given in polyethylene glycol (PEG 400). A control group was given only the vehicle.

The data on mating ratio, on the numbers of implantations and embryonic deaths were comparable for all groups. There was no evidence of an effect on male fertility or number of offspring/litter.

 

In vivo micronucleus assay

The objective of this in vivo assay was to evaluate the ability of the test article, Diglycidyl ether of bisphenol A (DGEBPA), to induce micronuclei in bone marrow polychromatic erythrocytes of ICR mice. The test article was suspended in corn oil and dosed by oral gavage at 500, 2500 and 5000 mg/kg as specified by the sponsor. The animals were dosed with the test article and were killed 24, 48 and 72 hours after dosing for extraction of the bone marrow. Ten animals (five males and five females) were randomly assigned to each dose/kill time group. Negative and positive control groups killed 24 hours after dosing were included in the assay. DGEBPA did not induce a significant increase in micronuclei in bone marrow polychromatic erythrocytes under the conditions of this assay and is considered negative in the mouse bone marrow micronucleus test.

 

Chinese hamsters were orally gavaged on two consecutive days with 0, 825, 1650 and 3300 BADGE/kg in 20 ml/kg polyethylene glycol 400. Cyclophosphamide, 128 mg/kg in 20 ml/kg PEG400, was used as the positive control. 24 hours after the last gavage, the animals were sacrificed and bone marrow was harvested from the shafts of both femurs. Cells from the bone marrow were fixed on slides and 1000 bone marrow cells each were scored per animal and the following anomalies were registered: a) single Jolly bodies, b) fragments of nuclei in erythrocytes, c) micronuclei in erythroblasts, d) micronuclei in leucopoietic cells, e) polyploid cells.

The experiment was performed to evaluate any mutagenic effect on somatic interphase cells in vivo. Mutagenic effects present themselves in interphase cells in form of nucleus anomalies of bone marrow cells. These anomalies occur in interphase cells as a consequence of damage during the mitotic process. The increase in anomalies shows a clear dose dependency, comparable to the occurrence of chromosome aberrations in metaphase preparations.

The bone marrow smears from animals treated with various doses of the test item showed no significant differences from the control. The incidence of bone marrow cells with anomalies of nuclei corresponds to the frequency observed in the control group.

By contrast, a positive control experiment with cyclophosphamide (128 mg/kg) yielded 7.73% cells with anomalies of nuclei. This is significantly different from the controls treated with the vehicle (PEG 400) alone.

It is concluded that under the conditions of this experiment, no evidence of mutagenic effects was obtained in Chinese hamsters treated with BADGE.

 

Several supporting studies were also negative.

 

Comet assay

The effect of Diglycidyl ether of Bisphenol A on the integrity of rat liver DNA was investigated using the alkaline elution assay and was compared with that produced by methyl methanesulphonate.

No effect on the integrity of liver DNA was demonstrated 6 h after a single oral exposure to 500 mg DGEBPA/kg.

Justification for classification or non-classification

The weight of evidence supports the conclusion that although BADGE appears to be genotoxic in in vitro test systems, in vivo it is not clastogenic or genotoxic. As such, there is no requirement for classification, nor is there a need to perform further studies to assess potential Germ cell mutagenicity.