4,4'-[2,2,2-trifluoro-1-(trifluoromethyl)ethylidene]diphenol; bisphenol AF

Short-term toxicity to fish

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

fish embryo acute toxicity (FET)

Type of information:

experimental study

Adequacy of study:

key study

Study period:

Not reported

Reliability:

2 (reliable with restrictions)

Rationale for reliability incl. deficiencies:

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

Qualifier:

according to guideline

Guideline:

other: ISO 15088: Water Quality - Determination of the acute toxicity of waste water to zebrafish eggs (Danio rerio)

Version / remarks:

2007

Deviations:

yes

Remarks:

test extended to 96 h

GLP compliance:

no

Remarks:

Published study not conducted to GLP

Specific details on test material used for the study:

STABILITY AND STORAGE CONDITIONS OF TEST MATERIAL
- Storage condition of test material: not reported
- Stability under test conditions: not reported
- Solubility and stability of the test substance in the solvent/vehicle: not reported
- Reactivity of the test substance with the solvent/vehicle of the cell culture medium: not reported

TREATMENT OF TEST MATERIAL PRIOR TO TESTING
- Treatment of test material prior to testing: Stock solution prepared at 15 mg/L in ultrapure water by mixing for 24 h.
- Preliminary purification step (if any): no
- Final dilution of a dissolved solid, stock liquid or gel: stock solution = 15 mg/L
- Final preparation of a solid: no

FORM AS APPLIED IN THE TEST (if different from that of starting material): liquid

OTHER SPECIFICS: no

Analytical monitoring:

yes

Details on sampling:

- Concentrations: 15 mg/L stock solution measured, only.
- Sampling method: Subsample of stock taken for HPLC analysis
- Sample storage conditions before analysis: Not reported

Vehicle:

no

Details on test solutions:

PREPARATION AND APPLICATION OF TEST SOLUTION (especially for difficult test substances)
- Method: Treatment solutions prepared from a 15 mg/L stock solution. Test concentrations were 0.50, 0.80, 1.0, 2.0, 4.0, 5.0 and 6.0 mg/L along with a dilution water control.
- Eluate: Standard dilution water as specified in ISO 7346-1 and ISO 7346-2.
- Differential loading: 1 embryo per 1 mL of treatment solution.
- Controls: Dilution water
- Chemical name of vehicle (organic solvent, emulsifier or dispersant): n/a
- Concentration of vehicle in test medium (stock solution and final test solution(s) or suspension(s) including control(s)): n/a
- Evidence of undissolved material (e.g. precipitate, surface film, etc.): not reported

Test organisms (species):

Danio rerio (previous name: Brachydanio rerio)

Details on test organisms:

TEST ORGANISM
- Common name: Zebrafish
- Strain: Danio rerio (Hamilton Buchanan)
- Source: ECT Oekotoxikologie, Flörsheim, Germany
- Age at study initiation: 3 h post-fertilisation (embryos)
- Length at study initiation: n/a
- Weight at study initiation: n/a
- Method of breeding: Bred in aquarium (60 x 30 x 30 cm) containing 45 L of tap water at constant temperature (26 ± 1 ºC) and 12 h light : 12 h dark photoperiod. Fish were fed three times per day with commercially available dried fish food (TetraMin, Tetra, Germany) and three times per week with young daphnids.
- Maintenance of the brood fish: see above.

ACCLIMATION
- Acclimation period: n/a
- Acclimation conditions (same as test or not): n/a
- Type and amount of food during acclimation: n/a
- Feeding frequency during acclimation: n/a
- Health during acclimation (any mortality observed): not reported

QUARANTINE (wild caught)
- Duration: n/a
- Health/mortality: n/a

FEEDING DURING TEST
- Food type: n/a
- Amount: n/a
- Frequency: n/a

Test type:

static

Water media type:

freshwater

Limit test:

no

Total exposure duration:

96 h

Remarks on exposure duration:

Extended from 48 h as prescribed in ISO 15088

Hardness:

Not reported

Test temperature:

Not reported

pH:

Not reported

Dissolved oxygen:

Not reported

Salinity:

Not reported

Conductivity:

Not reported

Nominal and measured concentrations:

Nominal

Details on test conditions:

TEST SYSTEM
- Test vessel: 24-well plates
- Type (delete if not applicable): closed
- Material, size, headspace, fill volume: not reported
- Aeration: no
- Type of flow-through (e.g. peristaltic or proportional diluter): n/a
- Renewal rate of test solution (frequency/flow rate): n/a
- No. of organisms per vessel: 1 embryo per well
- No. of vessels per concentration (replicates): not reported
- No. of vessels per control (replicates): not reported
- No. of vessels per vehicle control (replicates): not reported
- Biomass loading rate: 1 embryo per well

TEST MEDIUM / WATER PARAMETERS
- Source/preparation of dilution water: as per ISO 7346-1 and ISO 7346-2 (as prescribed in ISO 15088) i.e. 294 mg/L CaCl2.2H2O, 123.3 mg/L MgSO4.7H2O, 63.0 mg/L NaHCO3, 5.5 mg/L KCl
- Total organic carbon: not reported
- Particulate matter: not reported
- Metals: not reported
- Pesticides: not reported
- Chlorine: not reported
- Alkalinity: not reported
- Ca/mg ratio: not reported
- Culture medium different from test medium: no
- Intervals of water quality measurement: not reported

OTHER TEST CONDITIONS
- Adjustment of pH: not reported
- Photoperiod: not reported
- Light intensity: not reported

EFFECT PARAMETERS MEASURED (with observation intervals if applicable) : lethal effects; egg coagulation, missing heratbeat, missing somites, missing tail detachment from yolk sac, sublethal malformations; no eye, body pimentation, missing blood fow, pericardial edema were inspected after 24 and 48 h exposure.

TEST CONCENTRATIONS
- Spacing factor for test concentrations: n/a
- Justification for using less concentrations than requested by guideline: not reported
- Range finding study: no
- Test concentrations: n/a
- Results used to determine the conditions for the definitive study: n/a

Reference substance (positive control):

no

Key result

Duration:

48 h

Dose descriptor:

EC50

Effect conc.:

4.8 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

mortality (fish)

Key result

Duration:

48 h

Dose descriptor:

EC50

Effect conc.:

3.3 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: missing body pigmentation

Key result

Duration:

48 h

Dose descriptor:

EC50

Effect conc.:

3 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: missing blood flow

Key result

Duration:

48 h

Dose descriptor:

EC50

Effect conc.:

2.5 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: pericardial edema

Key result

Duration:

72 h

Dose descriptor:

EC50

Effect conc.:

2.2 mg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: hatching inhibition

Details on results:

- Behavioural abnormalities: not reported
- Observations on body length and weight: not reported
- Other biological observations: not reported
- Mortality of control: not reported
- Other adverse effects control: not reported
- Abnormal responses: not reported
- Any observations (e.g. precipitation) that might cause a difference between measured and nominal values: not reported
- Effect concentrations exceeding solubility of substance in test medium: not reported

Results with reference substance (positive control):

n/a

Reported statistics and error estimates:

EC20 EC50 and EC 80 values calculated by non-linear regression model using GraphPad Prism (version 6.0, USA).

Sublethal observations / clinical signs:

Table 1:       EC Values of Bisphenol AF for Zebrafish Embryos

Basis for effect	Dose descriptor	Value (mg/L)
Mortality (48 h)	EC20	3.3
	EC50	4.2 (95 % C.L.: 3.6 – 4.7)
	EC80	5.2
Missing body pigmentation (48 h)	EC20	2.9
	EC50	3.3 (95 % C.L.: 2.4 – 4.5)
	EC80	3.7
Missing blood flow (48 h)	EC20	2.2
	EC50	3.0 (95 % C.L.: 0.67 – 13.6)
	EC80	4.2
Pericardial edema (48 h)	EC20	2.0
	EC50	2.8 (95 % C.L.: 1.4 – 5.4)
	EC80	3.9
Hatching inhibition (72 h)	EC20	1.9
	EC50	2.2 (95 % C.L.: 1.8 – 2.8)
	EC80	2.6

Validity criteria fulfilled:

not specified

Conclusions:

48 h EC50 for mortality was 4.2 (95 % C.L.: 3.6 - 4.7) mg/L for Zebrafish embryos.

Study conducted in accordance with ISO 15088 however it is not clear from the data reported if the guideline validity criteria were met.

Executive summary:

The 4 day acute toxicity of Bisphenol AF to zebrafish (Danio rerio) fertilished embryos (3-h post fertilisation) was studied under static conditions.  Fertilized eggs/embryos (10 embryos per test concentration/ control) of zebrafish were exposed to a control and test chemical (nominal) concentrations of 0.50, 0.80, 1.0, 2.0, 3.0, 4.0, 5.0 and 6.0 mg/L.  The test system was maintained under conditions as prescribed in ISO 15088, with the exception that the test was prolonged to 96 -h. The 48 h LC50 values, based on mortality, was 4.2 (95 % C.L.: 3.6 - 4.7) mg/L.   The sublethal effects observed included missing body pigmentation (48 h EC50 = 3.3 mg/L), missing blood flow (48 h EC50 = 3.0 mg/L), pericardial edema (48 h EC50 = 2.8 mg/L) and hatching inhibition (72 h EC50 = 2.2 mg/L).   The most sensitive end point was hatching success.

 

This toxicity study is classified as acceptable and satisfies the guideline requirement for a fish embryo acute toxicity test in acordance with ISO 15088 (2007).

 

Results Synopsis;

 

Test Type: Static

The most sensitive endpoint was hatching success with an 72 h EC50 of 2.2 (95 % C.L.: 1.8 - 2.8) mg/L.

The most relevant endpoint for acute lethal toxicity to fish was mortality at 48 hours. The 48 hour LC50 was 4.2 (95 % C.L.: 3.6 - 4.7) mg/L.

 

Applicant's interpretation:

For the fish test the ISO 15088 (2007) guideline was followed (acute toxicity to fish the acute toxicity of waste water to zebrafish eggs (Danio rerio)).  The test is not a traditional method however, it is considered acceptable with expert judgement. For the test it has been shown in a study using an early version of the acute embryo toxicity test with zebrafish (Danio rerio) that is was a more sensitive alternative to cell line tests (Lange et al., 1995). Nagel (2002) documented that the zebrafish embryo assay is a very promising tool to replace the acute fish test. The ISO state: This International Standard has been elaborated as a substitute for the acute fish toxicity test. Applied to waste water, it gives the same or similar results as achieved from the acute fish toxicity tests (e.g. ISO 7346-1 or ISO 7346-2). If used for single substances, different sensitivities from both test systems are possible.  Mortality is used as the surrogate endpoint for fish acute toxicity endpoints, this is indicated by 1) lack of tail detachment from egg, 2) egg coagulation and 3) heartbeat is not detectable.  

The fish were cultured in the laboratory in an aquarium containing 45 L of tap water at 26 ± 1 °C, with a 12 hour photoperiod per day.  The fish were therefore naive to chemical pollution.  Fish were fed 3 times daily with fish food and 3 times per week with young daphnids. The concentrations used testing the toxicity of BPAF were 0.50, 0.80, 1.0, 2.0, 3.0, 4.0, 5.0, 6.0 mg/L and a control (dilution water only).  These were place into 24-well plates 3 hours post fertilisation. Ten eggs per concentration and the control were used, each was placed separately in its on individual well on the plate.  The authors state, in principle, the test was compliant with the guideline used, but the exposure of embryos was extended to 4 days. However, 48 hour endpoints were also noted in accordance with the guideline.  After 24 and 48 h of exposure, lethal (egg coagulation, missing heartbeat, missing somites, missing tail detachment from the yolk sac) and sublethal (no eye and body pigmentation, missing blood flow, pericardial oedema) malformations were inspected using a stereoscopic microscope (Nikon SMZ 1000 with a DS-Fi1 digital camera) and a NIS-Elements Documentation 2.2 imaging software. Embryos were checked daily for mortality, developmental malformations and body length (a distance from the most anterior part of the head to the tip of the tail) after 96 h of exposure. For each chemical at least three independent experiments were performed. Concentration response curves for each tested species exposed to BPAF were plotted using Origin (version 8.1) data analysis and graphing software (OriginLab). The EC50 and LC50 with 95% confidence limits and EC20/LC20, EC80/LC80 values were calculated by non-linear regression model using GraphPad Prism (version 6.0, GraphPad Software Inc., San Diego). All stock solution concentrations were determined using HPLC.  Although, the test did not monitor test concentrations throughout the duration of the this test, it was verified during the chronic Daphnia magna test performed by the same authors and reported in the same paper that BPAF was stable (within 90 %) of nominal dosed concentrations after 3 days in Daphnia medium (Elendt M4).  The medium is much more complex (more salts are added) and the data from this study confirm that 1) the dosing performed by the laboratory is accurate and 2) that the test item is likely to be stable in less complex medium such as that used for the acute embryo toxicity fish test.  

It can be seen from the results of the report, in the sigmoidal plot of data that control mortality was < 10 %.  However a positive control was used during the experiment.  The data only fulfils one of the 2 validity criteria.  However, due to the concordance of data from the other tested materials (bisphenol A (BPA) and bisphenol F (BPF)) with existing data it is expected that the test system used and the laboratory practices were conducive of a reliable and high integrity study.  The other chemicals tested and there results fell within the known ranges seen from other experiments data using similar or the same tests. For instance, Tisler et al. (2016) showed BPA 48 hour LC50 for was 15.9 mg/L, in a study by Song et al. (2014) the 144 hour LC50 was 7.5 mg/L, although shown to be more toxic the duration was longer.  Further in concordance with Song et al. (2014) there was no effect on hatchability below <=1.9 mg/L. In this case BPA may be used as surrogate positive controls, outwith those traditionally used or advised.  Based on the information one validity criteria is met, but the other information shows that the study is valid and reliable.

 BPAF at the higher concentrations tested showed malformations in development (i.e. > 1 mg/L), however if this was statistically significant was not assessed.  The purpose of the study, however, was not to determine the subs-lethal effects, but the LC50 of BPAF.  Which is well documented and in line with the definition of mortality as defined by the guideline.  The 48-hour LC50 was 4.2 (95 % confidence limit: 3.6-4.7) mg/L.  Based on the quality of the study and relevance of the endpoint this will be used as a key value for chemical safety assessment.

Reference 2

Endpoint:

fish embryo acute toxicity (FET)

Type of information:

experimental study

Adequacy of study:

key study

Study period:

2015

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 236 (Fish embryo acute toxicity (FET) test)

Deviations:

yes

Remarks:

Some of the guideline validity criteria were not reported or were outwith acceptable Guideline limits

GLP compliance:

no

Remarks:

Published study not conducted to GLP

Specific details on test material used for the study:

STABILITY AND STORAGE CONDITIONS OF TEST MATERIAL
- Storage condition of test material: Not reported
- Stability under test conditions: Not reported
- Solubility and stability of the test substance in the solvent/vehicle: Not reported
- Reactivity of the test substance with the solvent/vehicle of the cell culture medium: Not reported

TREATMENT OF TEST MATERIAL PRIOR TO TESTING
- Treatment of test material prior to testing: BPAF stock solutions (50 mg/L, 5 mg/L and 0.5 mg/L) were prepared by dissolving BPAF in dimethyl sulphoxide (DMSO) (100 mL), followed by dilution to 1 L with pure water containing 60 mg Instant Ocean sea salt, respectively
- Preliminary purification step (if any): n/a
- Final dilution of a dissolved solid, stock liquid or gel: 50 mg/L, 5 mg/L and 0.5 mg/L
- Final preparation of a solid: n/a

FORM AS APPLIED IN THE TEST (if different from that of starting material): Liquid

OTHER SPECIFICS: none

Analytical monitoring:

yes

Remarks:

HPLC analysis of stock solutions conducted.

Details on sampling:

- Concentrations: 50 mg/L, 5 mg/L and 0.5 mg/L
- Sampling method: Not reported
- Sample storage conditions before analysis: Not reported

Vehicle:

yes

Remarks:

Dimethylsulfoxide (DMSO)

Details on test solutions:

PREPARATION AND APPLICATION OF TEST SOLUTION (especially for difficult test substances)
- Method: BPAF stock solutions (50 mg/L, 5 mg/L and 0.5 mg/L) were prepared by dissolving BPAF in dimethyl sulphoxide (DMSO) (100 mL), followed by dilution to 1 L with pure water containing 60 mg Instant Ocean sea salt, respectively. These solutions were further diluted to 5, 50 and 500 μg/L and named as the second stock solutions. Test water was prepared by dissolving 60 mg of Instant Ocean sea salt in 1 L of pure water to prevent interference by impurities in tap water. DMSO solvent (0.1%, treated as control) was freshly prepared by adding 1 mL of DMSO in 1 L of test water
-Eluate: Test water was prepared by dissolving 60 mg of Instant Ocean sea salt in 1 L of pure water.
- Differential loading:
- Controls: DMSO solvent (0.1%, treated as control) was freshly prepared by adding 1 mL of DMSO in 1 L of test water
- Chemical name of vehicle (organic solvent, emulsifier or dispersant): DMSO
- Concentration of vehicle in test medium (stock solution and final test solution(s) or suspension(s) including control(s)): 0.1 % v/v
- Evidence of undissolved material (e.g. precipitate, surface film, etc.): not reported

Test organisms (species):

Danio rerio (previous name: Brachydanio rerio)

Details on test organisms:

TEST ORGANISM
- Common name: Zebrafish (Danio rerio)
- Strain: AB strain
- Source: Not reported
- Age at study initiation (mean and range, SD): 3 months
- Length at study initiation (length definition, mean, range and SD): Not reported
- Weight at study initiation (mean and range, SD): Not reported
- Method of breeding: Not reported
- Maintenance of the brood fish: Throughout the study, adult fish were maintained under an artificial light/dark period of 14/10 h and a constant temperature (28 ± 1 °C) in an aerated aquarium system. Adult fish were fed twice daily with a commercial flake food (Charoen Pokphand Group, Bangkok, Thailand) complemented with freshly hatched Artemia nauplii (Charoen Pokphand Group) without any solvent or BPAF.

ACCLIMATION
- Acclimation period: Not reported
- Acclimation conditions (same as test or not): Not reported
- Type and amount of food during acclimation: Adult fish were fed twice daily with a commercial flake food (Charoen Pokphand Group, Bangkok, Thailand) complemented with freshly hatched Artemia nauplii (Charoen Pokphand Group) without any solvent or BPAF.
- Feeding frequency during acclimation: As above
- Health during acclimation (any mortality observed): Not reported

QUARANTINE (wild caught)
- Duration: n/a
- Health/mortality: n/a

FEEDING DURING TEST
- Food type: Not fed
- Amount: n/a
- Frequency: n/a

Test type:

semi-static

Water media type:

freshwater

Limit test:

no

Total exposure duration:

168 h

Remarks on exposure duration:

168 hours post fertilisation

Hardness:

Not reported

Test temperature:

28.5 °C

pH:

Not reported

Dissolved oxygen:

Not reported

Salinity:

Not reported

Conductivity:

Not reported

Nominal and measured concentrations:

Nominal concentrations: 5, 50, 500 µg/L

Details on test conditions:

TEST SYSTEM
- Test vessel: 500 mL beaker
- Type (delete if not applicable): not reported
- Material, size, headspace, fill volume: 500 mL beaker filled with 300 mL test solution
- Aeration: not reported
- Type of flow-through (e.g. peristaltic or proportional diluter): n/a
- Renewal rate of test solution (frequency/flow rate): Solutions replaced avery 12 h
- No. of organisms per vessel: 300 embryos per replicate
- No. of vessels per concentration (replicates): 3
- No. of vessels per control (replicates): n/a
- No. of vessels per vehicle control (replicates): 3
- Biomass loading rate: 300 embryos per 300 mL test solution

TEST MEDIUM / WATER PARAMETERS
- Source/preparation of dilution water: Test water was prepared by dissolving 60 mg of Instant Ocean sea salt in 1 L of pure water
- Total organic carbon: not reported
- Particulate matter: not reported
- Metals: not reported
- Pesticides: not reported
- Chlorine: not reported
- Alkalinity: not reported
- Ca/mg ratio: not reported
- Culture medium different from test medium: not reported
- Intervals of water quality measurement: not reported

OTHER TEST CONDITIONS
- Adjustment of pH: not reported
- Photoperiod: 14:10 h (light: dark)
- Light intensity: not reported

EFFECT PARAMETERS MEASURED (with observation intervals if applicable) : At 168 hpf (hours post fertilisation) the following parameters were measured; body weight length determination, thyroid hormones measurement, and gene expression analysis

TEST CONCENTRATIONS
- Spacing factor for test concentrations: x10
- Justification for using less concentrations than requested by guideline: The highest concentration was based on 25% of the concentration for 50% of maximal effect (EC50) (based on the malformation rate, 24 h EC50 of BPAF was 2.00 mg/L for the embryos). The lowest concentration was based on 1/3 of an environmental investigation concentration.
- Range finding study: No
- Test concentrations: n/a
- Results used to determine the conditions for the definitive study: The highest concentration was based on 25% of the concentration for 50% of maximal effect (EC50) (based on the malformation rate, 24 h EC50 of BPAF was 2.00 mg/L for the embryos). The lowest concentration was based on 1/3 of an environmental investigation concentration.

Reference substance (positive control):

no

Key result

Duration:

168 h

Dose descriptor:

NOEC

Effect conc.:

>= 500 µg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

mortality (fish)

Key result

Duration:

168 h

Dose descriptor:

NOEC

Effect conc.:

>= 500 µg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: Length

Key result

Duration:

168 h

Dose descriptor:

NOEC

Effect conc.:

>= 500 µg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

weight

Key result

Duration:

72 h

Dose descriptor:

NOEC

Effect conc.:

5 µg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: hatchability

Remarks:

No. of individuals hatched at 72 hpf

Key result

Duration:

96 h

Dose descriptor:

NOEC

Effect conc.:

>= 500 µg/L

Nominal / measured:

nominal

Conc. based on:

test mat.

Basis for effect:

other: hatchability

Remarks:

No. of individuals hatched at 96 hpf

Details on results:

- Behavioural abnormalities:
- Observations on body length and weight: No effect observed at any concentration.
- Other biological observations: BPAF exposure caused significant alterations in zebrafish whole-body T4 and T3 contents at 168 hpf. The concentrations of TT4 and TT3 were significantly affected by different treatment (Kruskal-Wallis test, p = 0.0032 and p = 0.0108, respectively). The TT4 contents significantly decreased by 30.08% and 34.44% at 50 and 500 μg/L exposure groups, respectively, in comparison with the SC group. The concentrations of whole-body TT3 in tissue homogenate were significantly reduced by 19.42% and 24.71% at 50 and 500 μg/L exposure groups, respectively, in comparison with the SC group. However, no significant differences were observed at the 5 μg/L exposure group. The concentrations of FT4 and FT3 were significantly affected by different treatment (Kruskal-Wallis test, p = 0.0205 and p = 0.002, respectively). Exposure to high concentrations (50 and 500 μg/L BPAF) significantly reduced whole-body FT4 contents (by 22.46% and 28.55%, respectively). In comparison with SC, the FT4 levels were unchanged at 5 μg/L BPAF exposure group. Nevertheless, a reduction of FT3 levels was observed at all exposure groups, which was extremely significant in every exposure group. The basal FT3 contents in the control zebrafish was 10.637 ± 0.527 pg/g, and were significantly reduced at different treatments: 42.23% reduction,
6.145 ± 0.286 pg/g at 5 μg/L group; 61.56% reduction, 4.089 ± 0.490 pg/g at 50 μg/L group; 60.53% reduction, 4.198 ± 0.556 pg/g at 500 μg/L group (** p < 0.01). The transcription of tsh-β gene was significantly up-regulated by 72.4% upon exposure to 50 μg/L of BPAF (0.01 < p < 0.05) whereas a highly significant decrease was observed at 500 μg/L. The slc5a5 gene transcription was significantly down-regulated 44.3% and 56.1% at 50 and 500 μg/L BPAF exposure groups, respectively. The mRNA expression of gene encoding thyroglobulin (tg) was up-regulated transcriptionally by 68.8% whereas the transthyretin (ttr) gene was significantly induced by 112.8%, 104.5%, and 118.1% at a concentration-dependent manner after exposure to 5, 50, and 500 μg/L BPAF. The expression of dio1 significantly increased at 50 and 500 μg/L BPAF exposure in comparison with SC (p < 0.05). In addition, the transcription of dio2 gene was significantly up-regulated by 40.4% after exposure to 50 μg/L BPAF. Two thyroid hormone receptors (tr-α and tr-β) were examined. The transcription of tr-α gene significantly decreased by 43.6% and 64.4% at 50 and 500 μg/L BPAF groups, respectively, whereas significant down-regulation of tr-β mRNA levels were observed at 50 μg/L in comparison with the control.
- Mortality of control: < 10 % at 96 hpf
- Other adverse effects control: not reported
- Abnormal responses: not reported
- Any observations (e.g. precipitation) that might cause a difference between measured and nominal values: not reported
- Effect concentrations exceeding solubility of substance in test medium: not reported

Results with reference substance (positive control):

N/A

Reported statistics and error estimates:

The data of the developmental parameters and hormone assay are presented as the mean ± standard error of the mean (SEM). The differences between the control and each exposure group were evaluated by Kruskal-Wallis test (nonparametric test) followed by Nemenyi multiple comparison test. These statistical tests were conducted using SPSS for Windows 13.0 Software (SPSS, Chicago, IL, USA). Homogeneity of variances and normality of the data were analyzed by Levene’s test and Kolmogorov-Smirnov test. Statistical analyses were performed by using GraphPad Prism 6.01 Int. J. Environ. Res. Public Health 2015, 12 13074 computer program (GraphPad Software, Inc., La Jolla, CA, USA) and the rest of the data were also presented as mean ± standard error of the mean (SEM). The differences in each observation were evaluated by one-way ANOVA followed by Dunett’s test for identifying the differences between exposure groups and the control group. The differences were considered significant at two different levels (* p < 0.05; ** p < 0.01) relative to the controls

Sublethal observations / clinical signs:

No significant effects on fish survivability, fish length or fish weight were noted in any of the treatment concentrations versus the control fish. A significant delay in hatching was observed at the top two concentrations (50 and 500 µg/L) at 72 hpf, however no significant delays were noted at 96 hpf, versus the control fish.

Effects on gene expression are detailed in the section above.

Validity criteria fulfilled:

no

Remarks:

a. Cannot be determined if fertilisation rate was greater than or equal to 70 %. b. Individiuals were maintained at 28 ºC (TG states 26 ± 1 ºC as acceptable limits). c. Positive control not used. d. % DO at 96 hpf not reported.

Conclusions:

The NOEC for mortality, fish length and weight was greater than or equal to 500 µg/L.

The NOEC for hatchability was 5 µg/L, where a significant delay in hatching at 72 hpf was observed in the 50 and 500 µg/L groups, versus the control.

Total T4 (TT4), Total T3 (TT3) and Free T4 (FT4) hormone concentrations were significantly reduced at 50 and 500 µg/L. Free T3 (FT3) concentrations were significantly (p<0.01) reduced at 5, 50 and 500 µg/L.

The modification of hypothalamic-pituitary-thyroid axis (HPT) gene mRNA levels in zebrafish were significantly altered to some extent (increase or decrease of induction) at all tested concentrations.

Executive summary:

In a 168 hour acute toxicity study, zebrafish (Danio rerio) embryos (2 h post fertilisation) were exposed to Bisphenol AF at nominal concentrations of 5, 50 and 500 µg/L under semi-static conditions (solutions were changed every 12 h).  The 168 h NOEC for survivability, fish length and fish weight was greater than or equal to the highest tested concentration of 500 µg/L.  A significant delay in hatching was observed at 50 and 500 µg/L with no significant delays observed at 5 µg/L. Additionally, hormone concentrations Total T4 (TT4), Total T3 (TT3) and Free T4 (FT4) were significantly reduced at 50 and 500 µg/L.  Free T3 (FT3) concentrations were significantly (p<0.01) reduced at 5, 50 and  500 µg/L. The modification of hypothalamic-pituitary-thyroid axis (HPT) gene mRNA levels in zebrafish larvae were significantly altered to some extent (increase or decrease of induction) at all tested concentrations.

 

This toxicity study is classified as supplementary as not all of the validity criteria to fulfill the guideline requirements were met/ reported [OECD 236, Danio rerio].

 

Results Synopsis;

 

Test Type: Semi static and renewed every 12 h.

 

LC50: > 500 µg/L

NOEC for mortality, fish length and fish weight: Greater than or equal to 500 µg/L

NOEC for hatch time: 5 µg/L

NOEC for hormone/ gene expression alteration: Could not be determined (less than lowest tested concentration)

 

Applicant's interpretation:

In the key study by Tang et al., 2015, using methods similar or equivalent to the OECD 236 (2013) survival (168 hpf), length (168 hpf), weight (168 hpf) and hatchability (72 and 96 hpf) were measured after exposure to 5, 50 or 500 µg/L of BPAF or a solvent control (DMSO).  The chemical and its source was well defined.  Stocks were made using DMSO (100 mL) and then adding the solvent-test time mix to 1 L of pure water containing 60 mg instant ocean sea salt.  DMSO control was set to 0.1 % in line with the concentration in the test item exposures.  The authors noted that BPAF has a half-life in water of 180 days, and therefore they did not expect the concentration in the test system to significantly decrease over 168 hours, however, there was no analytical data to verify this. Throughout the study, adult fish were maintained under an artificial light/dark period of 14/10 h and a constant temperature (28 ± 1 °C) in an aerated aquarium system. Adult fish were fed twice daily with a commercial flake food (Charoen Pokphand Group, Bangkok, Thailand) complemented with freshly hatched Artemia nauplii (Charoen Pokphand Group) without any solvent or BPAF.

The test was conducted at 28 ± 1°C, this is above the temperature advised by the OECD 236 guideline of 26 ± 1°C, but it is not expected this significantly impacted on the test or health of the organism.  The test was initiated with fertilized eggs of zebrafish (D. rerio) obtained from non-exposed adults (AB strain, aged 3 months). Humidity was 70 % with a 14 h photoperiod.

In support of Tisler et al. (2016), up to and including 500 µg/L, no effects on survival were seen.  Furthermore, although at 72 hpf hatchability appeared decreased compared to the solvent control at 96 hpf there were no significant differences in hatchability.  No other endpoints for development showed significant effects.

A study by Song et al. (2014) (Section 6.1.2 - Long term toxicity to fish) was conducted using similar methods to those described in the OECD guidelines 234 (2011) and 236 (2013).  The study further supports that the LC50 of the test item is above 2.0 mg/L as seen in the study by Tisler et al. (2016).  

Zebrafish embryos were collected and rinsed with mineral medium. Only healthy embryos were selected for the test. It is unclear when the embryos were fertilised, but all measures are referred to as hours post fertilisation. A vehicle and non-vehicle control were included to confirm that the used level of DMSO was not detrimental to fish health.  In a preliminary study following similar guidelines as the OECD 236 fish embryos were exposed in 4 mL glass vials to  a range of concentrations between 0.5 and 10 mg/L for 144 hours. Studies were conducted with 30 embryos per dose and all studies were repeated at least three times. Developmental lesions, including oedema and hemorrhage, as well as death and date of hatching, were recorded daily. The LC50 was determined to 1.75 mg/L and the NOEC was 1.0 mg/L.  Although 10 % mortality was observed within the 1 mg/L group this was not seen as biologically significant and still falls within the possible expected mortality of healthy unexposed populations which is further highlighted by the validity criteria which allow such mortality rates within the controls.  Hatching rate was not significantly effected in all surviving fish.  Developmental anomalies were detected in fish exposed to the test item at 1.5 mg/L but the NOEC for abnormalities (seen as oedema) was 1.0 mg/L.  At 96 hours, there were no mortalities at any concentration and no effects on hatching rates.

The study only reports nominal concentrations, however concentrations were checked via LC/MS, but not results were presented. It was noted that the test concentrations (including DMSO) were homogenous in the test replicates during the study.  There is sufficient evidence within the literature to suggest that 1) the test item is stable in environmental matrices, 2) the chemical is sufficiently soluble at the tested level, and 3) test item will not shift pH or have a high oxygen demand.  For example Tisler et al. 2016 verified during a chronic Daphnia magna test that the test item was stable (i.e. within 90 % of nominal dosed concentrations) after 3 days in Daphnia medium (Elendt M4).  The medium used here is much less complex and similar stability would be expected. Further, DMSO appears to be an appropriate solvent in all fish testing with high rates of hatching success and survival, and minimal noting of malformations or ill health in Song et al. (2014), Tang et al. (2015) and Yang et al. (2014), which all used DMSO as a solvent.  Song et al. (2014) also included a DMSO and medium control to show that no significant effect was caused by the use of the solvent.  An algae study conducted by DuPont (2007) and the chronic Daphnia magna study conducted by Tisler et al. (2016) provide further evidence that water quality parameters will not be greatly affected by these concentrations of the test item.

For validity criteria according to OECD 236 (2013) although temperature was 28 ± 0.5 °C, it was seen as acceptable for this species, it was also noted that the embryos of control group developed normally in embryo medium, and almost all embryos were hatched between 48 and 72 h likely complying with hatchability requirements of => 80 %. However, the survival in both controls, fertilisation success and positive control results on the embryos is unknown. Sufficient embryos were used per concentration and the test was repeated 3 times, giving some confidence in the presented results.

 

Description of key information

48h LC50 = 4.2 mg/L, ISO 15088 (2007), Tišler et al., 2016

96h LC50 = >= 0.5 mg/L; OECD 236; Tang et al., 2015

Key value for chemical safety assessment

Fresh water fish

Fresh water fish

Dose descriptor:

LC50

Effect concentration:

4.2 mg/L

Additional information

There are 2 key studies on short-term toxicity to fish.  The studies were equivalent or similar to the determination of the acute toxicity of waste water to zebrafish eggs (Danio rerio) (ISO 15088, 2007) and the fish embryo acute toxicity test (FET) (OECD 236, 2013).

 

In the Key Study by Tisler et al., 2016, they sought to run a suite of aquatic ecotoxicity tests to identify hazards to the aquatic environment and conduct a risk assessment using measured concentrations of test item from the environment to determine a risk characterisation ratio (i.e. predicted exposure concentration (PEC)/predicted no effect concentration (PNEC). Firstly, a suite of low tier aquatic toxicity tests were conducted on V. fischeri (bacteria), Desmodesmus subspicatus (green algae), Daphnia magna (invertebrate) and Danio rerio (fish; common name zebrafish).  

 

For the fish test the ISO 15088 (2007) guideline was followed (acute toxicity to fish the acute toxicity of waste water to zebrafish eggs (Danio rerio)).  The test is not a traditional method however, it is considered acceptable with expert judgement. For the test it has been shown in a study using an early version of the acute embryo toxicity test with zebrafish (Danio rerio) that is was a more sensitive alternative to cell line tests (Lange et al., 1995). Nagel (2002) documented that the zebrafish embryo assay is a very promising tool to replace the acute fish test. The ISO state: This International Standard has been elaborated as a substitute for the acute fish toxicity test. Applied to waste water, it gives the same or similar results as achieved from the acute fish toxicity tests (e.g. ISO 7346-1 or ISO 7346-2). If used for single substances, different sensitivities from both test systems are possible.  Mortality is used as the surrogate endpoint for fish acute toxicity endpoints, this is indicated by 1) lack of tail detachment from egg, 2) egg coagulation and 3) heartbeat is not detectable.  

 

The fish were cultured in the laboratory in an aquarium containing 45 L of tap water at 26 ± 1 °C, with a 12 hour photoperiod per day.  The fish were therefore naive to chemical pollution.  Fish were fed 3 times daily with fish food and 3 times per week with young daphnids. The concentrations used testing the toxicity of BPAF were 0.50, 0.80, 1.0, 2.0, 3.0, 4.0, 5.0, 6.0 mg/L and a control (dilution water only).  These were place into 24-well plates 3 hours post fertilisation. Ten eggs per concentration and the control were used, each was placed separately in its on individual well on the plate.  The authors state, in principle, the test was compliant with the guideline used, but the exposure of embryos was extended to 4 days. However, 48 hour endpoints were also noted in accordance with the guideline.  After 24 and 48 h of exposure, lethal (egg coagulation, missing heartbeat, missing somites, missing tail detachment from the yolk sac) and sublethal (no eye and body pigmentation, missing blood flow, pericardial oedema) malformations were inspected using a stereoscopic microscope (Nikon SMZ 1000 with a DS-Fi1 digital camera) and a NIS-Elements Documentation 2.2 imaging software. Embryos were checked daily for mortality, developmental malformations and body length (a distance from the most anterior part of the head to the tip of the tail) after 96 h of exposure. For each chemical at least three independent experiments were performed. Concentration response curves for each tested species exposed to BPAF were plotted using Origin (version 8.1) data analysis and graphing software (OriginLab). The EC50 and LC50 with 95% confidence limits and EC20/LC20, EC80/LC80 values were calculated by non-linear regression model using GraphPad Prism (version 6.0, GraphPad Software Inc., San Diego). All stock solution concentrations were determined using HPLC.  Although, the test did not monitor test concentrations throughout the duration of the this test, it was verified during the chronic Daphnia magna test performed by the same authors and reported in the same paper that BPAF was stable (within 90 %) of nominal dosed concentrations after 3 days in Daphnia medium (Elendt M4).  The medium is much more complex (more salts are added) and the data from this study confirm that 1) the dosing performed by the laboratory is accurate and 2) that the test item is likely to be stable in less complex medium such as that used for the acute embryo toxicity fish test.  

 

It can be seen from the results of the report, in the sigmoidal plot of data that control mortality was < 10 %.  However a positive control was used during the experiment.  The data only fulfils one of the 2 validity criteria.  However, due to the concordance of data from the other tested materials (bisphenol A (BPA) and bisphenol F (BPF)) with existing data it is expected that the test system used and the laboratory practices were conducive of a reliable and high integrity study.  The other chemicals tested and there results fell within the known ranges seen from other experiments data using similar or the same tests. For instance, Tisler et al. (2016) showed BPA 48 hour LC50 for was 15.9 mg/L, in a study by Song et al. (2014) the 144 hour LC50 was 7.5 mg/L, although shown to be more toxic the duration was longer.  Further in concordance with Song et al. (2014) there was no effect on hatchability below <=1.9 mg/L. In this case BPA may be used as surrogate positive controls, outwith those traditionally used or advised.  Based on the information one validity criteria is met, but the other information shows that the study is valid and reliable.

 

BPAF at the higher concentrations tested showed malformations in development (i.e. > 1 mg/L), however if this was statistically significant was not assessed.  The purpose of the study, however, was not to determine the subs-lethal effects, but the LC50 of BPAF.  Which is well documented and in line with the definition of mortality as defined by the guideline.  The 48-hour LC50 was 4.2 (95 % confidence limit: 3.6-4.7) mg/L.  Based on the quality of the study and relevance of the endpoint this will be used as a key value for chemical safety assessment.

 

In the key study by Tang et al., 2015, using methods similar or equivalent to the OECD 236 (2013) survival (168 hpf), length (168 hpf), weight (168 hpf) and hatchability (72 and 96 hpf) were measured after exposure to 5, 50 or 500 µg/L of BPAF or a solvent control (DMSO).  The chemical and its source was well defined.  Stocks were made using DMSO (100 mL) and then adding the solvent-test time mix to 1 L of pure water containing 60 mg instant ocean sea salt.  DMSO control was set to 0.1 % in line with the concentration in the test item exposures.  The authors noted that BPAF has a half-life in water of 180 days, and therefore they did not expect the concentration in the test system to significantly decrease over 168 hours, however, there was no analytical data to verify this. Throughout the study, adult fish were maintained under an artificial light/dark period of 14/10 h and a constant temperature (28 ± 1 °C) in an aerated aquarium system. Adult fish were fed twice daily with a commercial flake food (Charoen Pokphand Group, Bangkok, Thailand) complemented with freshly hatched Artemia nauplii (Charoen Pokphand Group) without any solvent or BPAF.

 

The test was conducted at 28 ± 1, this is above the temperature advised by the OECD 236 guideline of 26 ± 1, but it is not expected this significantly impacted on the test or health of the organism.  The test was initiated with fertilized eggs of zebrafish (D. rerio) obtained from non-exposed adults (AB strain, aged 3 months). Humidity was 70 % with a 14 h photoperiod.

 

In support of Tisler et al. (2016), up to and including 500 µg/L, no effects on survival were seen.  Furthermore, although at 72 hpf hatchability appeared decreased compared to the solvent control at 96 hpf there were no significant differences in hatchability.  No other endpoints for development showed significant effects.

 

A study by Song et al. (2014) (Section 6.1.2 - Long term toxicity to fish) was conducted using similar methods to those described in the OECD guidelines 234 (2011) and 236 (2013).  The study further supports that the LC50 of the test item is above 2.0 mg/L as seen in the study by Tisler et al. (2016).  

 

Zebrafish embryos were collected and rinsed with mineral medium.  Only healthy embryos were selected for the test. It is unclear when the embryos were fertilised, but all measures are referred to as hours post fertilisation.  A vehicle and non-vehicle control were included to confirm that the used level of DMSO was not detrimental to fish health.  In a preliminary study following similar guidelines as the OECD 236 fish embryos were exposed in 4 mL glass vials to  a range of concentrations between 0.5 and 10 mg/L for 144 hours. Studies were conducted with 30 embryos per dose and all studies were repeated at least three times. Developmental lesions, including oedema and hemorrhage, as well as death and date of hatching, were recorded daily. The LC50 was determined to 1.75 mg/L and the NOEC was 1.0 mg/L.  Although 10 % mortality was observed within the 1 mg/L group this was not seen as biologically significant and still falls within the possible expected mortality of healthy unexposed populations which is further highlighted by the validity criteria which allow such mortality rates within the controls.  Hatching rate was not significantly effected in all surviving fish.  Developmental anomalies were detected in fish exposed to the test item at 1.5 mg/L but the NOEC for abnormalities (seen as oedema) was 1.0 mg/L.  At 96 hours, there were no mortalities at any concentration and no effects on hatching rates.

 

The study only reports nominal concentrations, however concentrations were checked via LC/MS, but not results were presented. It was noted that the test concentrations (including DMSO) were homogenous in the test replicates during the study.  There is sufficient evidence within the literature to suggest that 1) the test item is stable in environmental matrices, 2) the chemical is sufficiently soluble at the tested level, and 3) test item will not shift pH or have a high oxygen demand.  For example Tisler et al. 2016 verified during a chronic Daphnia magna test that the test item was stable (i.e. within 90 % of nominal dosed concentrations) after 3 days in Daphnia medium (Elendt M4).  The medium used here is much less complex and similar stability would be expected. Further, DMSO appears to be an appropriate solvent in all fish testing with high rates of hatching success and survival, and minimal noting of malformations or ill health in Song et al. (2014), Tang et al. (2015) and Yang et al. (2014), which all used DMSO as a solvent.  Song et al. (2014) also included a DMSO and medium control to show that no significant effect was caused by the use of the solvent.  An algae study conducted by DuPont (2007) and the chronic Daphnia magna study conducted by Tisler et al. (2016) provide further evidence that water quality parameters will not be greatly affected by these concentrations of the test item.

 

For validity criteria according to OECD 236 (2013) although temperature was 28 ± 0.5 °C, it was seen as acceptable for this species, it was also noted that the embryos of control group developed normally in embryo medium, and almost all embryos were hatched between 48 and 72 h likely complying with hatchability requirements of => 80 %. However, the survival in both controls, fertilisation success and positive control results on the embryos is unknown. Sufficient embryos were used per concentration and the test was repeated 3 times, giving some confidence in the presented results.

 

The 144 hour post fertilisation LC50 of 1.75 mg/L and a 144 hpf NOEC of 1.5 mg/L. In accordance with the guideline durations of 96-hours there were no observed significant mortalities and no effects on hatching rates at any exposure concentration.  The key endpoint from the study is the 96-hr LC50 > 2 mg/L and supports observations of Tisler et al., 2016.

 

In conclusion, the key value for chemical safety is an LC50 of 4.2 mg/L (Tisler, 2016).