4,4'-isopropylidenediphenol

Long-term toxicity to aquatic invertebrates

Link to relevant study record(s)

Reference

Reference 1

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: Study performed according to OECD 202, Part 2 guidelines (currently OECD 211), under GLP and with analytical confirmation of dose.

Qualifier:

according to guideline

Guideline:

OECD Guideline 211 (Daphnia magna Reproduction Test)

Version / remarks:

Cited as OECD Guideline 202, part 2 (Daphnia sp., Reproduction Test)

Principles of method if other than guideline:

not relevant

GLP compliance:

yes

Specific details on test material used for the study:

CAS no. 80-05-7
molecular formula C15H1602
purity: 99.94%

Analytical monitoring:

yes

Details on sampling:

not specified

Vehicle:

not specified

Details on test solutions:

not specified

Test organisms (species):

Daphnia magna

Details on test organisms:

For more than 10 years, Daphnia magna has been kept in Bayer AG’s ecotoxicology laboratories as a synchronous parthenogenetic strain at 20°C and at a photo period of 16 hr light and 8 hr darkness. The only food source is the single-cell green algae Scenedesmus subspicatus CHODAT from fermenter cultivation (batch operation).

Test type:

semi-static

Water media type:

freshwater

Limit test:

no

Total exposure duration:

21 d

Hardness:

not specified

Test temperature:

not specified

pH:

not specified

Dissolved oxygen:

not specified

Salinity:

not specified

Conductivity:

not specified

Nominal and measured concentrations:

nominal 0.316 and 3.16 mg/L
measured

Details on test conditions:

In order to complement the data for the current EU risk assessment, the chronic Daphnia toxicity (toxicological endpoint: reproduction rate) was determined under GLP conditions in accordance with the OECD Draft Guideline 202 (version dated August 1995) in a semi-static test (medium renewal three times per week) with individual animals in 10 parallels of each concentration step. The test concentrations covered the interval from the lowest 48 hr-EC50 (Stephenson 1983: 3.9 ppm) down to 1/100th of this concentration, following a geometric progression with a factor of 3.16. The results (NOEC/LOEC) are based on the analytically measured values (HPLC: analytical limit of detection: 0.01 ppm) and represent the arithmetic mean of all measured values before and immediately after the renewal of the test media within the semi-static procedure. In a subsequent experiment, the number of pre-adult and adult moults was determined during a 21 d exposure (Bisphenol A: 3.16 or 0.316 ppm; control) .

Reference substance (positive control):

no

Key result

Duration:

21 d

Dose descriptor:

NOEC

Effect conc.:

>= 3.146 mg/L

Nominal / measured:

meas. (arithm. mean)

Conc. based on:

not specified

Basis for effect:

reproduction

Details on results:

The results (NOEC/LOEC) are based on the analytically measured values (HPLC: analytical limit of detection: 0.01 ppm) and represent the arithmetic mean of all measured values before and immediately after the renewal of the test media within the semi-static procedure.

Results with reference substance (positive control):

not relevant

Reported statistics and error estimates:

not specified

In the 21 day reproduction test, the reproduction rate in the control (n=10) was 125.7+/- 8.1 (# of live births/parent) and in the 3.16 ppm exposure group was 124.3 +/- 10.6. The results of the reproduction test showed no adverse effects at the significance level of alpha=0.05. No statistically significant (alpha = 0.05) difference was found in either the number of pre-adult molts or total moults in 21 days at either of the test concentrations. The general moulting behavior was shown to be highly variable under control conditions without any ecotoxicological impact. The results of this study indicate that chronic effects of Bisphenol A exposure do not occur in test organisms exposed to concentrations below the limit for acute toxicity effects.

Validity criteria fulfilled:

yes

Remarks:

Control mortality did not exceed 20 % and the control mean number of live offspring produced per parent was greater than 60.

Conclusions:

Caspers, 1998, conducted a chronic exposure study with D. magna following OECD 202 and applying GLP documentation. Some data are not specified but all in all the presented data is sufficient to consider the study as reliable and appropriate for regulatory purposes.

Executive summary:

Both a standard, chronic Daphnia magna reproduction study as well as a study to assess the molting behavior of Daphnia exposed to Bisphenol A were performed. No statistically significant difference in reproduction was observed in the standard OECD 211 study. The NOEC for reproduction was the highest dose tested of 3.16 ppm. Also, no statistically significant differences in the moulting frequency between the control and the two Bisphenol A concentrations (0.316 ppm and 3.16 ppm) were observed.

Reference 2

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

experimental study

Adequacy of study:

key study

Study period:

August 2005 - November 2005

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

other: US EPA method 100.4

GLP compliance:

yes

Specific details on test material used for the study:

Bisphenol A
Lot No. B0070138
CAS No. 80-05-7
purity of 99.76%

Analytical monitoring:

yes

Details on sampling:

During the definitive study, a composite sample of all twelve replicates of each treatment level and the control solutions was collected and analyzed for Bisphenol A (BPA) at 0-hour (test initiation) and approximately weekly thereafter until test termination. Samples were collected from the approximate midpoint of the test vessel using a disposable glass pipet.

Vehicle:

no

Details on test solutions:

The dilution water (well water) used during this study was from the same source as the water used to culture amphipods and was characterized as having total hardness and total alkalinity ranges (as CaCO3) of 40 to 56 mg/L and 24 to 35 mg/L, respectively, a pH range of 7.4 to 7.8 and a specific conductance range of 150 to180 μmhos/cm. Representative samples of the dilution water source were analyzed periodically for the presence of pesticides, PCBs and toxic metals by GeoLabs, Inc., Braintree, Massachusetts. None of these compounds were detected at concentrations that are considered toxic in any of the water samples analyzed, in agreement with ASTM (2002) standard practices. In addition, representative samples of the dilution water source were analyzed monthly for total organic carbon (TOC) concentration. The TOC concentration of the dilution water source ranged from 0.45 to 1.1 mg/L for July through September 2005. Several species of daphnids (representative freshwater organisms generally recognized to be sensitive to chemical challenges) are cultured in water from the same source as the dilution water utilized in this study and have successfully survived and reproduced over multiple generations. The acceptable performance of the daphnid cultures, in combination with the previously mentioned analyses, confirmed the acceptability of this dilution water for use in bioassays.

Test organisms (species):

other: Hyalella azteca

Details on test organisms:

The freshwater amphipod (Hyalella azteca) was selected as a test organism for several reasons. The amphipods are one of the species recommended in the EPA guidelines for testing, are epibenthic and are widely distributed throughout North and South America. The organism is easily cultured, is frequently used in toxicity tests and reaches reproductive age in a relatively short time period (approximately 35 days at 23 ºC), making it suitable for conducting toxicity tests. The amphipods used during this study were reared at Springborn Smithers. The culture water was laboratory well water and was characterized as soft water. Amphipods for use in the study were obtained by transferring adult amphipods from 20-L glass culture aquaria using wide-bore pipettes and placing them in 9.5-L aquaria with approximately 8 L of water, 8 to 9 days prior to test initiation. Young produced within 24 hours by these isolated adults were then removed and pipetted into holding containers until test initiation. Amphipods were eight days old at test initiation. No amphipod mortality was observed in the test population 48 hours prior to test initiation. During rearing, the amphipods were fed 1.5 mL of a combination of yeast, Cerophyl® and flaked fish food (YCT) suspension (at a concentration of 1.8 g/L) and a unicellular green alga, Ankistrodesmus falcatus. Amphipods were fed once every other day during the culturing and rearing period. During the definitive exposure, amphipods were fed once daily (1.5 mL) until test day 33. After test day 33, they were fed twice daily (1 mL at each feeding). This modification to the feeding regime was made at the request of the Study Sponsor and corresponds to the time the cycle rate and number of vessel turnovers was increased. Representative samples of the food source were analyzed periodically for the presence of pesticides, PCBs and toxic metals by GeoLabs, Inc., Braintree, Massachusetts. None of these compounds have been detected at concentrations considered toxic in any of the samples analyzed. Based on the analysis for pesticides, food sources were considered to be of acceptable quality since analyte concentrations were below levels of concern (ASTM, 2002).

Test type:

flow-through

Water media type:

freshwater

Limit test:

no

Total exposure duration:

42 d

Hardness:

Total hardness as CaCO3: 40 to 56 mg/L
Total alkalinity as CaCO3: 24 to 35 mg/L

Test temperature:

22 to 24 ºC

pH:

7.4 to 7.8

Dissolved oxygen:

6.0 - 9.1 mg/Ls

Salinity:

not specified

Conductivity:

Specific conductivity: 150 to 180 μmhos/cm

Nominal and measured concentrations:

Nomial: 0.19, 0.38, 0.75, 1.5 and 3.0 mg a.i./L
Mean measure (as time-weighted averages): 0.12, 0.22, 0.49, 1.1 and 2.2 mg a.i./L

Details on test conditions:

The toxicity test was conducted using an exposure system consisting of a 2.0-L proportional diluter (Mount and Brungs, 1967) and a set of exposure vessels. The exposure system, constructed entirely of glass and silicone sealant, was designed to provide five concentrations of the test substance, a dilution water control and a pH buffer control. Twelve replicate vessels (A through L) were established for each treatment level and the controls. The exposure vessels were maintained in an area illuminated with fluorescent bulbs at an intensity of 64 to 93 footcandles (690 to 1000 lux). Light intensity was measured with a VWR light meter. The photoperiod was the same as that of the culture area. Sudden transitions from light to dark and vice versa were avoided. The test was conducted in a temperature-controlled water bath which was designed to maintain test solution temperatures at 23 ± 1 ºC.

Reference substance (positive control):

no

Key result

Duration:

42 d

Dose descriptor:

NOEC

Effect conc.:

0.49 mg/L

Nominal / measured:

meas. (TWA)

Conc. based on:

act. ingr.

Basis for effect:

reproduction

Duration:

42 d

Dose descriptor:

LOEC

Effect conc.:

1.1 mg/L

Nominal / measured:

meas. (TWA)

Conc. based on:

act. ingr.

Basis for effect:

reproduction

Duration:

42 d

Dose descriptor:

LC50

Effect conc.:

0.78 mg/L

Nominal / measured:

meas. (TWA)

Conc. based on:

act. ingr.

Basis for effect:

mortality

Details on results:

Preliminary Testing Prior to initiating the definitive study, a preliminary exposure was conducted under flow-through conditions at nominal concentrations of 0.050, 0.13, 0.25, 0.50, 1.0 and 2.0 mg a.i./L, and a dilution water control. Six replicates of ten amphipods each were exposed to each treatment level and the control. Following 42 days of exposure, mean percent survival of 80, 93, 87, 93, 85 and 43% was observed among amphipods exposed to the 0.050, 0.13, 0.25, 0.50, 1.0 and 2.0 mg a.i./L treatment levels, respectively. Mean percent survival of 80% was observed among amphipods exposed to the control. The mean number of offspring per female amphipods was 17, 6.0, 7.6, 7.4, 11 and 0 among amphipods exposed to the 0.050, 0.13, 0.25, 0.50, 1.0 and 2.0 mg a.i./L treatment levels, respectively. The mean number of offspring per female amphipods in the control was 9.4. Based on these results and consultation with the Study Sponsor, nominal concentrations of 0.19, 0.38, 0.75, 1.5 and 3.0 mg a.i./L were selected for the definitive exposure.
Definitive Test
Evaluation of Test Conditions
Results of the water quality parameters measured during the study are presented in Table 1. All water quality parameters measured were unaffected by the concentrations of Bisphenol A (BPA) tested and remained within acceptable ranges for the survival of amphipods. Daily measurement of the temperature in the test solutions and continuous temperature monitoring established that the exposure solution temperature ranged from 22 to 24 ºC during the definitive study.
Analytical Results The diluter system which prepared and delivered the test solutions to the exposure vessels functioned properly during the pretest period and throughout the 42-day definitive study. Analysis of test solutions during the pretest period established that the appropriate concentrations of Bisphenol A (BPA) in the exposure solutions were being delivered. The results of the analysis of the exposure solutions for Bisphenol A (BPA) during the in-life portion of the definitive exposure are presented in Table 2. During the first two weeks of the exposure, measured concentrations of Bisphenol A (BPA) were generally consistent with the nominal concentrations. On test day 14, a power outage resulted in instrumentation failure during the analytical run. Consequently, the samples were compromised and the results were not considered valid. Therefore, the day 14 results were excluded from calculation of the mean measured concentrations. Measured concentrations of Bisphenol A (BPA) dropped during the period between test days 14 and 21. Bisphenol A is known to be readily biodegradable, and it is likely that the microbial communities in the test vessels became acclimated to Bisphenol A as an easily accessible carbon source during the study. In an attempt to prevent further decrease and improve measured concentrations, several modifications were made in-life at test day 21. These modifications were 1) at each weekly observation interval, test vessels were replaced with clean test vessels, 2) vessel turn-over rate was increased from 6 to 12 volume replacements per day and 3) amount of food administered was split between a morning and an evening feeding. These modifications improved recoveries to near nominal levels on test day 22 of the exposure and generally allowed the measured concentrations to remain around or above 50% for the remainder of the exposure (approximately 20 days). Mean measured concentrations (expressed as time-weighted averages) ranged from 58 to 73% of nominal and defined the treatment levels as 0.12, 0.22, 0.49, 1.1 and 2.2 mg a.i./L. The relationship between the nominal treatment levels and the mean measured concentrations established during this study is illustrated in Figure 2. Analysis of the quality control samples resulted in measured concentrations which were consistent with the predetermined recovery range (Appendix 3) and ranged from 91.2% to 103% (N = 27) of the nominal fortified concentrations (0.150, 0.750 and 3.00 mg a.i./L) for Bisphenol A (BPA). Based on these results, it was established that the appropriate precision and quality control was maintained during the analyses of the exposure solutions.
Biological Results Following termination of the exposure, survival of amphipods in the control and pH buffer control was 78 and 75%, respectively. Reproduction (mean cumulative offspring per female) at test termination in the control and pH buffer control was 13 and 10 offspring per female, respectively. Mean total body length for male and female amphipods in the control and pH buffer control was 5.9 and 5.3 mm per amphipod and 5.4 and 4.9 mm per amphipod, respectively. Mean dry weight for male and female amphipods in the control and pH buffer control was 0.90 and 0.77 mg dry weight per amphipod and 0.93 and 0.72 mg dry weight per amphipod, respectively. The EPA test method 100.4 recommends a performance criterion of ≥80% survival in controls after 28 days. This data quality objective was achieved with a pooled control survival at 28 days of 81%. The test method (U.S. EPA, 2000) recommends a reproduction performance objective of > 2 offspring per control female between test days 28 and 42. This data quality objective was also achieved with a mean number of offspring per control female of 12. The recommended reproduction and survival data quality objectives were met (U.S. EPA, 2000), and the results of this study are acceptable for evaluating the long-term exposure of Hyalella azteca to Bisphenol A (BPA). The number of dead amphipods during the chronic exposure are presented in Table 3. The mean percent survival and observations recorded during the chronic test are presented in Table 4. Mean percent survival during the chronic test is illustrated in Figure 3. Following 42 days of exposure, 0% survival was observed among amphipods exposed to the 2.2 mg a.i./L treatment level. Survival of 75, 83, 78 and 72% was observed among amphipods exposed to the 0.12, 0.22, 0.49 and 1.1 mg a.i./L treatment levels, respectively. During the same period, mean percent survival of 78 and 75% was observed among amphipods exposed to the dilution water control and the pH buffer control, respectively. Statistical analysis (t-Test) determined no significant difference between control and pH buffer control for survival, therefore, control and pH buffer control data were pooled (pooled control = 77%). Williams' Test determined a significant difference in survival among amphipods exposed to the 2.2 mg a.i./L treatment level as compared to the pooled control data. This treatment level was excluded from further statistical analyses of reproduction and growth due to this survival effect. Amphipod reproduction during this study is presented in Table 5 and illustrated in Figure 4. At test termination, the cumulative mean number of offspring per female was 9.4, 13, 11 and 8.5 among amphipods exposed to the 0.12, 0.22, 0.49 and 1.1 mg a.i./L treatment levels, respectively. The cumulative number of offspring per female was 13 and 10 in the control and pH buffer control, respectively. Statistical analysis (t-Test) determined no significant difference between control and pH buffer control for reproduction, therefore, control and pH buffer control were pooled (pooled control = 12 offspring per female). Williams' Test determined a significant difference in reproduction among amphipods exposed to the 1.1 mg a.i./L treatment level as compared to the pooled control data. Amphipod growth (length and dry weight) during this study is presented in Table 6 and illustrated in Figure 5 and Figure 6, respectively. At test termination, length among male amphipods exposed to the 0.12, 0.22, 0.49 and 1.1 mg a.i./L treatment levels averaged 5.4, 5.6, 5.7 and 5.6 mm, respectively. Length among male amphipods exposed to the control and pH buffer control averaged 5.9 and 5.4 mm, respectively. Statistical analysis (t-Test) determined a significant difference in length among amphipods exposed to the control compared to the pH buffer control, therefore, pH buffer control data were used for statistical comparisons. Williams' Test determined no significant difference in any of the treatment levels tested when compared to the pH buffer control data. Dry weight among male amphipods exposed to the 0.12, 0.22, 0.49 and 1.1 mg a.i./L treatment levels averaged 0.82, 0.92, 0.94 and 0.91 mg, respectively. Dry weight among male amphipods exposed to the control and pH buffer control averaged 0.90 and 0.93 mg, respectively. Statistical analysis (t-Test) determined no significant difference between control and pH buffer control for dry weight, therefore, control and pH buffer control data were pooled (pooled control = 0.92 mg). Williams' Test determined no significant difference in dry weight among amphipods exposed to any of the treatment levels tested when compared to the pooled control data. At test termination, length among female amphipods exposed to the 0.12, 0.22, 0.49 and 1.1 mg a.i./L treatment levels averaged 4.8, 5.1, 5.1 and 5.1 mm, respectively. Length among female amphipods exposed to the control and pH buffer control averaged 5.3 and 4.9 mm, respectively. Statistical analysis (t-Test) determined a significant difference in length among amphipods exposed to the control compared to the pH buffer control, therefore, pH buffer control data were used for statistical comparisons. Williams' Test determined no significant difference in length for any of the treatment levels tested when compared to the pH buffer control data. Dry weight among female amphipods exposed to the 0.12, 0.22, 0.49 and 1.1 mg a.i./L treatment levels averaged 0.65, 0.73, 0.78 and 0.75 mg, respectively. Dry weight among female amphipods exposed to the control and pH buffer control averaged 0.77 and 0.72 mg, respectively. Statistical analysis (t-Test) determined no significant difference between control and pH buffer control for dry weight, therefore, control and pH buffer control data were pooled (pooled control = 0.74 mg). Williams' Test determined no significant difference in dry weight among amphipods exposed to any of the treatment levels tested when compared to the pooled control data. A summary of the endpoints measured during this study is presented in Table 7. Table 8 presents the 42-day LC50, LOEC and NOEC values. Based on the results of this study, the 42-day LC50 value was determined by moving average analysis to be 0.78 mg a.i./L, with 95% confidence intervals of 0.68 to 0.90 mg a.i./L. Based on mean measured concentrations (expressed at time-weighted averages) and amphipod reproduction, the Lowest-Observed-Effect Concentration (LOEC) was determined to be 1.1 mg a.i./L. The No-Observed-Effect Concentration (NOEC) was determined to be 0.49 mg a.i./L.

Results with reference substance (positive control):

not relevant

Reported statistics and error estimates:

A t-Test was used to compare the pH buffer control data to the control data. If the concentration of the buffer in the pH buffer control caused a statistically significant effect when compared to the control, then treatment data were compared to the pH buffer control. If there was no statistically significant difference between the pH buffer control and the control, both controls were pooled for data analysis. The data were tested for normality and homogeneity of variance using Shapiro-Wilks' Test or Chi-Square Test and Bartlett's Test or Cochran's Test, respectively. If the data passed these qualifying tests, then a parametric method was used to evaluate the results of the test, e.g., Williams' Test or Dunnett's Test. If the data failed the test for normality and homogeneity of variance, then a non-parametric method was used to evaluate the results of the amphipod test, e.g., Dunn's or Steel's One-Many Rank Test. If necessary, mean values were transformed using square root, arcsine square root or log conversion procedures.

Validity criteria fulfilled:

yes

Conclusions:

The study of Cafarella, 2006, is a fully reliable guideline and GLP conform chronic exposure study with Hyallela azteca. The test protocol is providing detailed information on the study design and conduct as well as on the results. This study is appropriate to be used for regulatory purposes.

Executive summary:

A 42-day chronic amphipod study was performed exposing Hyallela azteca to nominal Bisphenol A concentrations of 0.19, 0.38, 0.75, 1.5, and 3.0 mg/L under flow-through conditions.The results of the analysis of exposure solutions established mean measured time-weighted average concentrations ranging from 58-73% of nominal, defined as 0.12, 0.22, 0.49, 1.1 and 2.2 mg/L, respectively. At 42 days, the mean percent survival was 0% in the 2.2 mg/L treatment. Survival of 75, 83, 78, 72, and 77% was observed among amphipods exposed to the 0.12, 0.22, 0.49, 1.1 mg/L and controls, respectively. Survival was only statistically reduced in the 2.2 mg/L dose group. Amphipod reproduction was 12, 9.4, 13, 11 and 8.5 offspring per female in the controls, 0.12, 0.22, 0.49 and 1.1 mg/L dose groups, respectively. No significant difference was observed with respect to growth. Based on the results of this study, the 42-day LC50 value was determined by moving average analysis to be 0.78 mg/L, with 95% confidence intervals of 0.68 to 0.90 mg/L. Based on mean measured concentrations and amphipod reproduction, the LOEC was determined to be 1.1 mg/L. The NOEC was determined to be 0.49 mg/L.

Reference 3

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

experimental study

Adequacy of study:

key study

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: Study performed under GLP guidelines according to US EPA OPPTS Draft Guideline 850.1350. Analytical confirmation of dose was performed.

Qualifier:

according to guideline

Guideline:

EPA OPPTS 850.1350 (Mysid Chronic Toxicity Test)

GLP compliance:

yes

Specific details on test material used for the study:

BPA (CAS 80-05-7)
lot B0070138; purity 99.68%

Analytical monitoring:

yes

Details on sampling:

Samples were removed from one replicate of each treatment level and control prior to the start of the definitive exposure and analyzed for the concentration of bisphenol A present in each vessel. Results of the pretest analysis were used to judge whether sufficient quantities of test substance were being delivered and maintained in the exposure aquaria to initiate the definitive study.

During the in-life phase of the definitive study, samples were removed from alternating replicate solutions of each treatment level and the control on days 0, 7, 14, 16, 17, 21 and 28 for analysis of Bisphenol A concentration. Additional samples were taken at each interval, along with quality control (QC) samples, and stored frozen for future analysis, if necessary. Since the sampling intervals were not equally spaced (e.g., 0, 24 and 72 hours) and exposure concentrations decreased following mysid pairing (see Section 3.4 for explanation), a time weighted average was calculated for each treatment level.

Vehicle:

no

Details on test solutions:

Prior to test initiation and weekly thereafter during the definitive test, until test day 19, a 1.0 mg a.i./mL stock solution was prepared by diluting, for example, 1.003 g (1.000 g as active ingredient) of Bisphenol A in a 1000-mL volumetric flask and bringing to volume with deionized water. The pH of the resultant stock solution was adjusted to approximately 11.3 by the dropwise addition of sodium hydroxide. The stock solution was positioned over a water-driven magnetic stir plate and was partially submerged within an ultrasonic waterbath maintained at 40 °C until the material went into solution. The resultant stock was observed to be clear and colorless. Following test day 19 and weekly thereafter until test termination, a 1.0 mg a.i./mL stock solution was prepared in a 2000-mL volumetric flask using the same procedure as above.

Prior to test initiation, an FMI pump was calibrated to deliver 1.16 mL/cycle of the 1.0 mg a.i./mL stock solution into the diluter's chemical mixing chamber which also received 1.94 L of dilution water per cycle. The mixing chamber was positioned over a water-driven magnetic stir plate and was partially submerged within an ultrasonic waterbath to aid the mixing of the test substance (stock solution) in the dilution water. The solution contained in the mixing chamber constituted the highest nominal test concentration (600 µg a.i./L) and was subsequently diluted (50 %) to provide the remaining nominal exposure concentrations (300, 150, 75, and 38 µg a.i./L). A set of control vessels was also established which contained the same dilution water and was maintained under the same conditions as the treatment level vessels, but contained no Bisphenol A.

Test organisms (species):

Americamysis bahia (previous name: Mysidopsis bahia)

Details on test organisms:

The mysids (≤22 hours post-release) used to initiate the life-cycle test (SSL Lot No. 09A143) were obtained from SSL cultures. The brood stock was originally obtained from the U.S. EPA Atlantic Ecology Division Laboratory located in Narragansett, Rhode Island on 13 May 2009 and allowed to acclimate to exposure conditions for approximately ten months prior to exposure initiation. The culture organisms did not show any sign of sickness, disease, injuries or abnormalities from the day of receipt to the day of exposure initiation. Brood stock and test organisms were determined to be in good health at the start of the exposure phase. Mysids were cultured in six recirculating 76 L glass aquaria containing natural seawater. Standard aquarium undergravel filters were used to provide aeration and a current conducive to feeding. The seawater in the aquaria was characterized as having a salinity range of 21 to 24 ‰, dissolved oxygen ranged from 6.8 to 7.3 mg/L, and a pH range of 7.8 to 8.0 during the 14-day period prior to exposure initiation. Brood stock and test organisms were cultured and tested in seawater from the same source. The area in which the mysids were cultured received a regulated photoperiod of 16 hours of light and 8 hours of darkness at a light intensity range of 76 to 80 footcandles (820 to 860 lux). A commercial aquarium heater was used to regulate culture solution temperature and the temperature ranged from 20 to 27 ºC.

Test type:

flow-through

Water media type:

saltwater

Limit test:

no

Total exposure duration:

28 d

Test temperature:

Daily monitoring for all solutions throughout the test period established that the temperature ranged from 26 to 27 ºC. Continuous temperature monitoring established a temperature range of 25 to 27 ºC during the exposure period.

pH:

Daily monitoring for all solutions throughout the test period established that the pH ranged from 7.3 to 8.1.

Dissolved oxygen:

Daily monitoring for all solutions throughout the test period established that dissolved oxygen concentration was maintained between 5.62 to 7.34 mg/L.

Salinity:

Daily monitoring for all solutions throughout the test period established that the salinity ranged from 20 to 22 ‰.

Nominal and measured concentrations:

Nominal bisphenol A concentrations of 38, 75, 150, 300 and 600 µg a.i./L corresponded to measured time-weighted average concentrations of 18, 41, 74, 170 and 370 µg a.i./L.

Details on test conditions:

Dilute, filtered natural seawater was used as dilution and control water during this study. Seawater was pumped from the Cape Cod Canal, Bourne, Massachusetts from about 1 to 4 meters offshore at a depth of approximately 0.5 meters. The seawater was then transported to the laboratory where the seawater was adjusted to a salinity of 20 ± 3 ‰ with laboratory well water, filtered through 20- and 5-µm polypropylene core filters and intensely aerated for approximately 48-hours within an epoxy-coated fiberglass holding reservoir prior to use. The seawater was pumped under constant pressure through PVC pipes to a head tank, where the water was heated to test temperature and aerated to equilibrate dissolved gases. The seawater used for this study had a salinity range of 20 to 21 ‰ and a pH range of 7.4 to 8.2. Samples of the dilution water were analyzed weekly during the definitive test for total organic carbon (TOC) concentration. The TOC concentration of the dilution water ranged from 0.96 to 0.98 mg/L during the March and April 2010.

The life-cycle toxicity test was conducted using an exposure system consisting of a modified intermittent-flow proportional diluter, a temperature-controlled water bath, and a set of 12 exposure aquaria. The exposure system was designed to provide five concentrations of the test substance and a dilution water control. Two replicates were maintained for all treatments and the controls. Each glass test aquarium measured 39 x 20 x 25 cm. A siphon drain reached a height approximately 5 cm above the bottom of the aquarium allowing the solution volume within the aquarium to fluctuate between approximately 3.9 and 7.0 liters.

During each cycle of the diluter system, approximately 500 mL of exposure solution was delivered to each replicate test vessel with a flow-splitting accuracy of 5 %. For the first 15 days of the study, the diluter provided the exposure solutions to each test vessel at a rate of approximately 7.0 aquarium volume additions per day to provide a 90 % test solution replacement rate of approximately 7 hours (Sprague, 1969). On test day 15, the diluter cycle rate was doubled to help maintain testing concentrations, to approximately 15 aquarium volume additions per day to provide a 90 % test solution replacement rate of approximately 3.5 hours.
For the first 11 days of exposure, each exposure aquarium contained two retention chambers to retain sexually immature mysids. Once all mysid appeared to be sexually mature (test day 11), the mysids in each exposure aquarium were redistributed into one retention chamber and a maximum of ten pairing chambers. The maximum organism loading concentration (based on a maximum mean wet weight of 0.0045 g per mature adult mysid) was 0.0026 g of biomass per liter of flowing test solution per day for the first 15 days of the exposure and 0.0013 g of biomass per liter per day for the remainder of the exposure.

Reference substance (positive control):

no

Key result

Duration:

28 d

Dose descriptor:

NOEC

Effect conc.:

170 µg/L

Nominal / measured:

meas. (TWA)

Conc. based on:

act. ingr.

Basis for effect:

reproduction

Duration:

28 d

Dose descriptor:

LOEC

Effect conc.:

370 µg/L

Nominal / measured:

meas. (TWA)

Conc. based on:

act. ingr.

Basis for effect:

reproduction

Duration:

28 d

Dose descriptor:

NOEC

Effect conc.:

370 µg/L

Nominal / measured:

meas. (TWA)

Conc. based on:

act. ingr.

Basis for effect:

mortality

Duration:

28 d

Dose descriptor:

LOEC

Effect conc.:

> 370 µg/L

Nominal / measured:

meas. (TWA)

Conc. based on:

act. ingr.

Basis for effect:

mortality

Duration:

28 d

Dose descriptor:

NOEC

Effect conc.:

370 µg/L

Nominal / measured:

meas. (TWA)

Conc. based on:

act. ingr.

Basis for effect:

growth

Details on results:

Following 28 days of exposure, survival of 80, 86, 80, 81, 90 and 86 % was observed among mysids exposed to the control, 18, 41, 74, 170 and 370 µg a.i./L concentrations, respectively. Statistical analysis (Williams’ Test) determined no significant difference in survival among organisms exposed to any of the treatment levels tested compared to the control data. Survival by sex was determined once the mysids were paired. Male survival of 85, 82, 80, 76, 92 and 100 % and female survival of 89, 97, 89, 100, 91 and 92 % was observed among mysids exposed to the control, 18, 41, 74, 170 and 370 µg a.i./L concentrations, respectively. Statistical analysis (Williams’ Test) determined no significant difference in male survival among organisms exposed to any of the treatment levels tested compared to the control data. Since no concentration tested resulted in ≥ 50% mortality, the 7-, 14-, 21- and 28 day LC50 values were all empirically estimated to be > 370 µg a.i./L, the highest mean measured concentration tested.

At test termination, the mean number of offspring per female for organisms in the control averaged 25. The mean number of offspring per female averaged 25, 19, 25, 21 and 7.7 among mysids exposed to the 18, 41, 74, 170 and 370 µg a.i./L concentrations, respectively. Statistical analysis (Williams’ Test) determined a significant difference in the mean number of offspring per female among organisms exposed to the 370 µg a.i./L treatment level compared to the control data.

The mean total body length of male mysids exposed to the control was 7.2 mm. The mean total body length of male mysids exposed to the 18, 41, 74, 170 and 370 µg a.i./L treatment levels was 7.3, 7.2, 7.2, 7.2 and 7.1 mm, respectively. Statistical analysis (Williams’ Test) determined no statistically significant difference in the total body length of male mysids exposed any of the treatment levels tested compared to the control data. The mean total body length of female mysids exposed to the control was 7.5 mm. The mean total body length of female mysids exposed to the 18, 41, 74, 170 and 370 µg a.i./L treatment levels was 7.5, 7.5, 7.6, 7.4 and 7.4 mm, respectively. Statistical analysis (Williams’ Test) determined no statistically significant difference in the total body length of female mysids exposed any of the treatment levels tested compared to the control data.

The mean dry body weight of male mysids exposed to the control was 0.89 mg. The mean dry body weight of male mysids exposed to the 18, 41, 74, 170 and 370 µg a.i./L treatment levels was 0.87, 0.95, 0.91, 0.88 and 0.85 mg, respectively. Statistical analysis (Williams’ Test) determined no statistically significant difference in the dry body weight of male mysids exposed any of the treatment levels tested compared to the control data. The mean dry body weight of female mysids exposed to the control was 1.21 mg. The mean dry body weight of female mysids exposed to the 18, 41, 74, 170 and 370 µg a.i./L treatment levels was 1.20, 1.19, 1.31, 1.14 and 1.19 mg, respectively. Statistical analysis (Williams’ Test) determined no statistically significant difference in the dry body weight of female mysids exposed any of the treatment levels tested compared to the control data.

Based on time-weighted average concentrations and the most sensitive endpoint analyzed (reproduction), the Lowest-Observed-Effect Concentration (LOEC) was empirically estimated to be 370 µg a.i./L. The No-Observed-Effect Concentration (NOEC) for bisphenol A and mysids was determined to be 170 µg a.i./L. Therefore, the Maximum-Acceptable-Toxicant Concentration (MATC) was estimated to be 250 µg a.i./L. Since no concentration tested resulted in ≥ 50% mortality, the 28 day LC50 value was empirically estimated to be > 370 µg a.i./L, the highest mean measured concentration tested.

Reported statistics and error estimates:

All statistical conclusions were made at the 95 % level of certainty except in the cases of Shapiro-Wilk’s Test and Bartlett’s Test, in which the 99 % level of certainty was applied. For each endpoint, the performance at each treatment level was compared with the performance of the control organisms. For this study, all data met the assumptions for normal distribution (Shapiro-Wilk's Test) and homogeneity of variance (Bartlett's Test). All data were evaluated using Williams' Test, a parametric procedure, to establish treatment effects. TOXASTAT® Version 3.5 (West, Inc. and Gulley, 1996) was used to perform the statistical computations. The theoretical threshold concentration expected to produce no deleterious effects at the 95 % level of certainty was estimated as the Maximum-Acceptable-Toxicant Concentration (MATC), the geometric mean of the limits set by the lowest mean measured concentration that elicited a statistically significant effect on organism performance (Lowest-Observed-Effect Concentration, LOEC) and the highest mean measured test concentration that showed no statistically significant difference between the exposed organisms and the control (No-Observed-Effect Concentration, NOEC). Determination of these levels is based on the most sensitive of the performance criteria evaluated (e.g., survival, reproductive success, and growth at test termination). The time-weighted average concentrations tested and the corresponding data for mortality derived from the definitive toxicity test were used to estimate the 7-, 14-, 21- and 28-day median lethal concentration (LC50) and the corresponding 95 % confidence intervals. During this study, no concentration tested caused a reduction of 50 % mortality, therefore, the LC50 was empirically estimated to be greater than the highest mean measured concentration tested and no statistical analyses were performed.

Validity criteria fulfilled:

yes

Conclusions:

Lee et al., 2010, conducted a fully valid chronic exposure study with A. bahia according to a standard guideline and applying GLP documentation. The study is in detail reported and the results are complete and plausble. Therefore, the study is appropriate for regulatory purposes.

Executive summary:

A 28-day chronic mysid (Americamysis bahia) life cycle study was performed to determine the effects of Bisphenol A to mysids under flow-through conditions. Mysids (≤22 hours post-release) were used to initiate the life-cycle test. Nominal Bisphenol A concentrations of 38, 75, 150, 300 and 600 µg a.i./L were selected for the definitive exposure and corresponded to time-weighted measured concentrations of 18, 41, 74, 170 and 370 µg a.i./L, respectively. A dilution water control was also employed. First generation (F₀) survival and reproductive success, as well as measurements of growth, as mean total body length and mean dry body weight, for all surviving adult mysids at test termination were assessed.

 

Based on the time-weighted average concentrations and the most sensitive endpoint analysed (reproduction), the Lowest-Observed-Effect Concentration (LOEC) was determined to be 370 µg a.i./L. The No-Observed-Effect Concentration (NOEC) for Bisphenol A and mysids was determined to be 170 µg a.i./L. Therefore, the Maximum-Acceptable-Toxicant Concentration (MATC) was calculated to be 250 µg a.i./L. Since no concentration tested resulted in ≥ 50% mortality, the 28-day LC50 value was empirically estimated to be > 370 µg a.i./L, the highest time weighted average concentration tested.

Reference 4

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

experimental study

Adequacy of study:

key study

Study period:

From 6 September 2005 to 8 September 2005

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

guideline study

Qualifier:

according to guideline

Guideline:

other: ASTM Guideline E1440-91

Deviations:

no

GLP compliance:

yes

Specific details on test material used for the study:

Name: Bisphenol A
Synonym: BPA
Lot No.: B0070138
CAS No.: 80-05-7
Purity: 99.76%

Analytical monitoring:

yes

Details on sampling:

During the in-life phase of the definitive study, samples of test solution were removed from each treatment level and the controls at 0 and 48 hours of exposure for analysis of Bisphenol A (BPA) concentration. Samples analyzed at 0 hour were removed from the mixing vessels prior to division into replicate test vessels. Samples analyzed at test termination were removed from the additional replicates designated for this purpose. Each exposure solution sample was collected from the approximate midpoint of the vessel with a volumetric pipet. Three quality control (QC) samples were prepared at each sampling interval and remained with the exposure solution samples throughout the analytical process. These samples were prepared at Bisphenol A (BPA) concentrations similar to the exposure concentration range. Results of these analyses of the QC samples were used to judge the precision and quality control maintained during the analysis of exposure solution samples.

Vehicle:

no

Details on test solutions:

The dilution water used during this study was from the same source and had the same characteristics as the culture water described above. During holding and prior to use, the dilution water was continuously aerated. Representative samples of the dilution water source were analyzed periodically for the presence of pesticides, PCBs and toxic metals by GeoLabs, Inc., Braintree, Massachusetts. None of these compounds have been detected at concentrations that are considered toxic in any of the water samples analyzed, in agreement with ASTM (2002) standard procedures. In addition, representative samples of the dilution water source were analyzed monthly for total organic carbon (TOC) concentration. The TOC concentration of the dilution water source was 0.43 mg/L for September 2005. Several invertebrate species (e.g., Daphnia magna) are cultured in water from the same source as the dilution water utilized in this study and have successfully survived and reproduced over several generations. The acceptable performance of these invertebrate cultures, in combination with the previously mentioned analyses, confirmed the acceptability of this dilution water for use during the conduct of bioassays.

Test organisms (species):

other: Rotifer (Brachionus calyciflorus)

Details on test organisms:

The rotifer, Brachionus calyciflorus, was selected as the test species and is a commonly used aquatic invertebrate in acute freshwater toxicity testing. The Brachionus calyciflorus used in this toxicity test were obtained as cysts from Florida Aquafarms, a commercial supplier located in Dade City, Florida. The cysts were received at Springborn Smithers on 26 August 2005, assigned SSL Lot No. 05A127, and placed in a refrigerator (approximately 4 ºC) until needed. Approximately 24 hours prior to test initiation, cysts were brought to room temperature, hydrated with moderately hard fortified well water, and placed in 80-mL glass petri dishes containing approximately 50 mL of dilution water at 300 footcandles. Petri dishes were maintained at 24 to 25 ºC. Newly hatched rotifers (≤ 2 hours old) were used to initiate the test. The culture water was prepared by fortifying well water based on the formula for hard water and filtering it through an Amberlite XAD-7 resin column to remove any potential organic contaminants. This water had total hardness and total alkalinity as calcium carbonate (CaCO3) of 84 mg/L and 58 mg/L, respectively, a pH of 8.0, and a specific conductance of 290 μmhos/cm.

Test type:

static

Water media type:

freshwater

Limit test:

no

Total exposure duration:

48 h

Hardness:

Moderately hard fortified well water
Specific conductance: 290 μmhos/cm
Total hardness as CaCO3: 84 mg/L
Total alkalinity as CaCO3: 58 mg/L

Test temperature:

The high temperature reading on this minimum/maximum (22 to 28 ºC) is considered to be an artifact of the placement of the probe near the drum roller. The temperature in the environmental chamber, which housed the drum roller, ranged from 23 to 25 ºC during this same time period,
which was consistent with the temperatures measured in the test solutions during water quality intervals (22 to 24 ºC).

pH:

pH 8.0

Dissolved oxygen:

Dissolved oxygen: 8.9 mg/L

Salinity:

measured but no data reported

Conductivity:

measured but no data reported

Nominal and measured concentrations:

Nominal: 0.47, 0.94, 1.9, 3.8 and 7.5 mg a.i./L
Measured: 0.45, 0.88, 1.8, 3.6 and 7.1 mg a.i./L

Reference substance (positive control):

no

Key result

Duration:

48 h

Dose descriptor:

EC10

Effect conc.:

1.12 mg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

act. ingr.

Basis for effect:

other: total rate of increase

Key result

Duration:

48 h

Dose descriptor:

NOEC

Effect conc.:

1.8 mg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

act. ingr.

Basis for effect:

other: intrinsic rate of population increase.

Duration:

48 h

Dose descriptor:

LOEC

Effect conc.:

3.6 mg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

act. ingr.

Basis for effect:

reproduction

Reported statistics and error estimates:

Statistical analysis (t-Test) determined no significant difference between the dilution water control and buffer control, therefore, the controls were pooled.
Statistical analysis (Wilcoxon's Rank Sum Test) determined a
Springborn Smithers Study No. 13796.6108 Page 18
significant difference for the 3.6 and 7.8 mg a.i./L treatment levels when compared to the pooled control.

The results of the analyses of the exposure solutions for BPA established that the measured concentrations were generally consistent between sampling intervals. Mean measured concentrations ranged from 93 to 95 % of nominal and defined the treatment levels as 0.45, 0.88, 1.8, 3.6, and 7.1 mg/L. At test termination, the average intrinsic rate of increase in the dilution water control and buffer control were statistically similar. The average intrinsic rate of increase for the control, 0.45, 0.88, 1.8, 3.6, and 7.1 mg/L treatment levels was 1.5, 1.45, 1.52, 1, 0.4, and -0.78, respectively. Statistical analysis (Wilcoxon's Rank Sum Test) determined a significant difference for the 3.6 and 7.8 mg/L levels. Based on the mean measured concentrations and the intrinsic rate of increase, the NOEC was determined to be 1.8 mg/L. The LOEC was determined to be 3.6 mg/L. Based on the reported dose-response data, a reliable EC10 could also be calculated using the Trap software. This was determined to be 1.12 mg/l.

Validity criteria fulfilled:

yes

Remarks:

Control performance yielded an intrinsic rate of increase (r value) of greater than 0.65.

Conclusions:

Sayers, 2006, is a fully valid guideline conform and GLP documented study. All relevant informaiton is documented. Reproduction of Brachionus calyciflorus was adequately assessed and the NOEC ov 1.8 mg/L was determined. Based on the reported dose-response data, a reliable EC10 could also be calculated using the Trap software. This was determined to be 1.12 mg/l. EC10 values were derived in accordance with Moermond et al. (2016) and ECHA (2008).The study appears to be appropriate for regulatory purposes.

Executive summary:

The purpose of this study was to determine the chronic toxicity of Bisphenol A to the rotifer (Brachionus calyciflorus) under static test conditions. The intrinsic rate of increase (reproduction) was used to determine the No-Observed-Effect Concentration (NOEC) and the Lowest-Observed-Effect Concentration (LOEC). Based on mean measured concentrations and the intrinsic rate of increase, the No-Observed-Effect Concentration (NOEC) was determined to be 1.8 mg a.i./L. The Lowest-Observed-Effect Concentration (LOEC) was determined to be 3.6 mg a.i./L. Based on the reported dose-response data, a reliable EC10 could also be calculated using the Trap software. This was determined to be 1.12 mg/l. EC10 values were derived in accordance with Moermond et al. (2016) and ECHA (2008).

Reference 5

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

experimental study

Adequacy of study:

key study

Study period:

From 10 February 2006 to 2 December 2006

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: The study was performed meeting the requirements of the protocol according to applicable GLP guidance. Analytical confirmation of dose was performed.

Qualifier:

no guideline available

Principles of method if other than guideline:

No guideline exists for the chronic testing of snails. The test methodology was based on previously conducted research to assess fecundity and juvenile growth of Marisa cornuarietis exposed to Bisphenol A in a flow-through system, utilising standard test methodologies employed in ecotoxicological testing. Testing conducted at 25 °C.

GLP compliance:

yes

Analytical monitoring:

yes

Details on sampling:

BPA concentrations were measured twice during the equilibration period of the adult fecundity and hatchability trials and once prior to the juvenile growth portion of testing. BPA concentrations were also measured at test initiation, weekly during the adult fecundity and juvenile growth trials, twice a week during the hatchability trials, and at test termination. The analytical samples were composites of equal volume sub-samples collected from
each replicate aquarium.

Vehicle:

no

Test organisms (species):

other aquatic mollusc: Marisa cornuarietis (Giant ramshorn snail)

Details on test organisms:

Marisa cornuarietis were obtained as confirmed adults from a population cultured at ABC Laboratories which were descendents of a population of field collected snails received from Dr. Sharon File at the University of Puerto Rico. Wild specimens of M. cornuarietis were collected from Lake Guajataca, Puerto Rico and transported to ABC Laboratories. The snails were cultured under conditions similar to those specified for testing purposes (Don Thomas personal communication, (5). A 12 h light:12 h dark photoperiod with two thirty-minute transition periods was used with illumination provided by fluorescent lights. The cultures were set up on a single-pass flow-through system to maintain adequate water quality. Water temperature of cultures was maintained at 25 ºC. Snails were fed fresh, commercially purchased organically grown romaine lettuce (Lactuca sativa, romaine) and commercial algal wafers (supplied by Hikari, USA and purchased through a distributor).

Test type:

flow-through

Water media type:

freshwater

Limit test:

no

Total exposure duration:

181 d

Remarks on exposure duration:

The adult fecundity study was 181 days. The hatchability trial was 57 days. The juvenile growth study was 90 days post-hatch.

Hardness:

Total Hardness: 230 to 296 mg CaCO3/L
Calcium: 69.8 to 82.6 mg Ca/L

Test temperature:

Temperature: 24.2 to 25.7 °C

pH:

pH: 7.6 to 8.3

Dissolved oxygen:

Dissolved Oxygen: 5.1 to 8.5 mg/L (65 to 107 % sat.)

Nominal and measured concentrations:

Nominal concentrations: 0 (control), 0.10, 1.0, 25, and 640 μg/L
Mean Measured:
Adult Fecundity: Hatchability: Juvenile Growth:

Reference substance (positive control):

no

Key result

Duration:

328 d

Dose descriptor:

NOEC

Effect conc.:

25 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

act. ingr.

Basis for effect:

other: juvenile growth

Duration:

328 d

Dose descriptor:

LOEC

Effect conc.:

640 µg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

act. ingr.

Details on results:

Adult Fecundity Results:
The mean number of eggs per clutch was 92, 95, 95, 98, and 92 eggs/clutch for the control, 0.10, 1.0, 25, and 640 μg BPA/L treatments, respectively. The mean number of eggs perfemale per week was 156, 154, 161, 176, and 162 for the control, 0.10, 1.0, 25, and 640 μg BPA/L treatments, respectively. The mean number of clutches per female per week was 1.7, 1.6, 1.7, 1.8, and 1.8 for the control, 0.10, 1.0, 25, and 640 μg BPA/L treatments, respectively. The mean number of eggs per female per month was 622, 617, 645, 704, and 648 for the control, 0.10, 1.0, 25, and 640 μg BPA/L treatments, respectively. The mean number of clutches per female per month was 6.7, 6.5, 6.8, 7.2, and 7.0 for the control, 0.10, 1.0, 25, and 640 μg BPA/L treatments, respectively. There was no significant effect of BPA on adult egg production at any of the tested concentrations (P = 0.21), however, there was a significant difference among replicates within treatments (P = 0.003). The twosided Dunnett’s test did not detect any significant differences for any BPA-treatment compared to the control.

Hatchability Results:
Percent hatch ranged from 5 to 100 % in the control and from 19 to 100 % in the BPA treatments. Overall mean percent hatch was 90, 97, 95, 93, and 96 % for the control, 0.10, 1.0, 25, and 640 μg BPA/L treatments, respectively. There was no effect of BPA exposure on percent hatch (P = 0.40) and no significant difference in percent hatch among replicate aquaria within treatments (P = 0.14). BPA exposure had a significant effect on time to first hatch (P = 0.008) as did replicate (P = 0.048). A two-sided Dunnett’s test did not show any significant differences between any of the treatments and the control. A Tukey’s test (which compares all pairs against each other) found that the 0.1 μg/L treatment had a significantly shorter hatching time than the 640 μg/L treatment. There was no significant effect of BPA exposure (P = 0.46) or replicate (P = 0.45) on time to 50 % hatch.

Juvenile Growth Results:
After 60 dph the mean blotted wet weight was 1.586, 1.704, 1.701, 1.458, and 1.302 grams for the control, 0.10, 1.0, 25, and 640 μg BPA/L treatments, respectively and the mean blotted wet weight at 90 dph was 3.947, 4.217, 4.252, 3.530, and 3.434 grams for the control, 0.10, 1.0, 25, and 640 μg BPA/L treatments, respectively. The effect of the treatment on male growth rate and wet-weight was not significant. The female growth rate and the female wet-weight were negatively affected at the highest treatment level. This effect was variable-wise significant, and if the F1-growth rate related variables tested in this manuscript would have been the only variables tested, also simultaneously significant.

Analytical:

Mean measured concentrations of BPA from the adult fecundity exposure ranged from 74 to 135 % of the nominal concentrations corresponding to mean measured concentrations of 0.135, 0.913, 18.5, and 554 g BPA/L. No residues of BPA were detected in the control at or above the MQL of 0.0593 g BPA/L. Recoveries of BPA in the QC samples ranged from 69 to 159 % of the nominal concentrations during the adult fecundity exposure period.

Mean measured concentrations of BPA from the hatchability exposure ranged from 69 to 159 % of the nominal concentrations corresponding to 0.159, 0.993, 17.3 and 696 g BPA/L. No residues of BPA were detected in the control at or above the MQL of 0.0593 g BPA/L. Recoveries of BPA in the QC samples ranged from 78 to 128 % of the nominal concentrations during the hatchability exposure period.

Mean measured concentrations of BPA from the juvenile growth exposure ranged from 62 to 67 % of the nominal concentrations corresponding to 0.062, 0.661, 15.5, and 429 g BPA/L. Recoveries of BPA in the QC samples ranged from 84 to 295 % of the nominal concentrations during the juvenile growth exposure period.

Test solution temperatures as measured in the individual test replicates during the adult fecundity exposure ranged from 24.2 to 25.7 °C with an overall replicate mean of 25.0 ± 0.2 °C. Results from the continuous temperature recording indicated that the temperature remained within the 25 ± 1 °C temperature range specified in the protocol. Dissolved oxygen concentrations throughout the adult fecundity exposure ranged from 5.1 to 8.1 mg/L (65 to 103 % saturation) with an overall replicate mean of 7.0 ± 0.5 mg/L. Test solution pH throughout the adult fecundity exposure ranged from 7.6 to 8.2 with an overall replicate mean of 7.9 ± 0.1. Total hardness during the adult fecundity exposure ranged from 230 to 270 mg CaCO3/L with an overall replicate mean of 248 ± 9 mg CaCO3/L. These hardness values should be equated to a calcium concentration greater than or equal to 60 mg/L. Test solution temperatures as measured in the individual test replicates during the hatchability exposure ranged from 24.5 to 25.7 °C with an overall replicate mean of 25.0 ± 0.2 °C. The periodic average temperature as recorded from the continuous temperature recording was 25.5 ± 0.3 ºC. Dissolved oxygen concentrations throughout the hatchability exposure ranged from 7.4 to 8.5 mg/L (93 to 107 % saturation) with an overall replicate mean of 7.9 ± 0.3 mg/L. Test solution pH throughout the hatchability exposure ranged from 7.6 to 8.2 with an overall replicate mean of 7.9 ± 0.1. Total hardness during the hatchability exposure ranged from 242 to 296 mg CaCO3/L with an overall replicate mean of 265 ± 15 mg CaCO3/L. These hardness values should be equated to a calcium concentration greater than or equal to 60 mg/L.

Test solution temperatures as measured in the individual test replicates during the juvenile growth exposure ranged from 24.3 to 25.3 °C with an overall replicate mean of 24.8 ± 0.2 °C. Results from the continuous temperature recording indicated that the temperature remained within the 25 ± 1 °C temperature range specified in the protocol. Dissolved oxygen concentrations throughout the juvenile growth exposure ranged from 7.1 to 8.1 mg/L (90 to 100 % saturation) with an overall replicate mean of 7.8 ± 0.2 mg/L. Test solution pH throughout the juvenile growth exposure ranged from 7.9 to 8.3 with an overall replicate mean of 8.1 ± 0.1. Total hardness during the juvenile growth exposure ranged from 234 to 262 mg CaCO3/L with an overall replicate mean of 246 ± 8 mg CaCO3/L. These hardness values should be equated to a calcium concentration greater than or equal to 60 mg/L.

Biological: During the adult fecundity trial, one male snail was found dead in one of the control chambers after approximately four months. This was the only mortality during the adult fecundity exposure. The female snail in this chamber continued to produce clutches for the remaining two months of the study. There was no significant effect of BPA on adult egg production at any of the tested concentrations (P = 0.21), however there was a significant difference among replicates within treatments (P = 0.003). The two-sided Dunnett's test did not detect any significant differences for any BPA-treatment compared to the control. Snails produced an average of 643 (± 172.8) eggs/female/month (averaged over all treatments and pairs). The contribution of BPA concentration to the total variance in egg production was estimated by treating the effect of BPA levels as a random factor, just as the replicate effect. This analysis indicated that only 1.6 % of the total variance in egg production could be attributed to BPA treatment, whereas 8.2 % was due to differences among replicate tanks and 90.2 % was due to variability among breeding pairs. The within replicate (aquarium) coefficients of variation for this endpoint provide an estimate of the size of among breeding pair variability. These ranged from 10 % to 45 % and averaged 25 %.

Clutch sizes in the hatchability trial ranged from 15 to 280 (overall mean 97.3). Percent hatch varied from 5 % to 100 % (overall mean 93.7). There was no effect of BPA exposure on percent hatch (P = 0.40) and no significant difference in percent hatch among replicate aquaria within treatments (P = 0.14). The Dunnett's test did not detect any significant differences in mean percent hatch between the control and any of the BPA treatments. Analysis of the contribution of BPA concentration to the total variance in hatchability traits indicated that about 1 % of the total variance in percent hatch was due to BPA treatment, 9 % was due to differences among replicate tanks and 90 % was due to differences among breeding pairs. Time to first hatch varied from 8 to 13 days. BPA exposure had a significant effect on time to first hatch (P = 0.008) as did replicate (P = 0.048). A two-sided Dunnett's test did not show any significant differences between any of the treatments and the control. A Tukey's test (which compares all pairs against each other) found that the 0.1 g /L treatment had a significantly shorter hatching time than the 640 g /L treatment. Approximately 29 % of the variance in time to first hatch was due to BPA treatment, 11 % was due to differences among replicate tanks, and 60 % was due to variance among breeding pairs. There was no significant effect of BPA exposure (P = 0.46) or replicate (P = 0.45) on time to 50 % hatch. The amount of variance attributed to BPA treatment and replicate together was less than the amount of variance attributed to the replicate alone, and therefore the amount of variance due to BPA treatment is estimated to be 0 %. About 0.3 % of the variance could be attributed to variance among replicates; the remainder was due to differences among breeding pairs.

Of the 449 juveniles tested in the juvenile growth trial (ca. 25 juveniles per replicate), all survived until the end of the trial, giving a survivorship of 100 % in all treatments. Snails increased rapidly in size in the control and all BPA treatments. Results of the nested ANOVA (with BPA concentration as a fixed effect and the replicates and breeding pairs as random effect factors) found significant effects of BPA on female growth (P = 0.013), female wet weight at 60 dph (P = 0.042), and male growth rate (P = 0.024), and marginally significant effects on male wet weight at 60 dph (P = 0.054). Replicate had a significant effect on male growth (P = 0.014 and wet weight (P = 0.038) and a marginally significant effect on female growth (P = 0.081 and wet weight (P = 0.090). However, by far the most significant effect on juvenile growth rate (for all endpoints) was due to differences among breeding pairs (P < 0.00002 for all endpoints). Two-sided Dunnett's tests, comparing replicate means between each of the BPA concentrations with the control for each of the four growth endpoints individually, found a significant decrease in female growth (P = 0.045), and a marginal effect on female wet weight (P = 0.053) in the 640 g /L treatment compared to the control. In addition, a significant increase in male growth rate (P = 0.045), and a marginal increase in wet weight (P = 0.083), was found in the 1 g /L treatment compared to the control. Between 15-20 % of the variance in the four growth endpoints was due to BPA treatment, 8-13 % was due to differences among replicate tanks, 24-49 % was due to variability among breeding pairs, and 27-42 % was due to variability among siblings from the same breeding pair (i.e., residual or error variance). Thus the variability in growth associated with BPA treatment was less than the variability associated with either breeding pair or inter-juvenile variability.

Validity criteria fulfilled:

not applicable

Conclusions:

A NOEC for juvenile growth rate is estimated to be 25 μg BPA/L based on the observed negative effects on female growth at 640 μg/L. Based on the combined results of the adult fecundity trial, the egg hatchability trial and the juvenile growth trial, a NOEC for Marisa cornuarietis of 25 μg BPA/L is proposed. This value is based on the highest concentration at which no statistically significant impairments in juvenile growth were observed (LOEC=640 μg/L).

Executive summary:

The objective of the study was to evaluate the effects of Bisphenol A on fecundity, hatchability, and growth of Marisa cornuarietis under flow-through conditions at a temperature of 25 degrees Celsius. For the adult fecundity trials, no effects of Bisphenol A were detected at any test concentration. From these results the NOEC for this trait is estimated to be >640 μg BPA/L. For the egg hatchability trial, no significant differences between the control and any Bisphenol A concentration were detected for percent hatch or for time to first- or 50 % hatch. Thus the NOEC for hatchability traits is estimated to be > 640 μg/L. For the juvenile growth trial, significant impairments in female growth were detected at 640 μg/L. Though a statistically significant increase in male (but not female) juvenile growth was observed at 1 μg/L, the contributions of breeding pair and inter-juvenile variability contributed measurably more to the variance in juvenile growth rate than Bisphenol A treatment. Thus a NOEC for juvenile growth rate is estimated to be 25 μg BPA/L based on the observed negative effects on female growth at 640 μg/L. Based on the combined results of the adult fecundity trial, the egg hatchability trial and the juvenile growth trial, a NOEC for Marisa cornuarietis of 25 μg BPA/L is proposed. This value is based on the highest concentration at which no statistically significant impairments in juvenile growth were observed (LOEC=640 μg/L).

Reference 6

Endpoint:

long-term toxicity to aquatic invertebrates

Type of information:

experimental study

Adequacy of study:

key study

Study period:

From 30 August 2006 to 22 December 2006

Reliability:

1 (reliable without restriction)

Rationale for reliability incl. deficiencies:

other: The study was performed according to GLP with analytical confirmation of dose, and met the requirements of the protocol.

Qualifier:

no guideline available

Principles of method if other than guideline:

No guideline exists for the chronic testing of snails. The test methodology was based on previously conducted research to assess fecundity and juvenile growth of Marisa cornuarietis exposed to Bisphenol A in a flow-through system, utilising standard test methodologies employed in ecotoxicological testing. Testing conducted at 22 °C.

GLP compliance:

yes

Analytical monitoring:

yes

Details on sampling:

BPA concentrations were measured twice during the equilibration period, at test initiation, weekly during the adult fecundity trial, and at test termination. The weekly analytical samples were composites of equal volume sub-samples collected from two of the four replicate aquariums. The two replicates analyzed were alternated each week.

Vehicle:

no

Test organisms (species):

other aquatic mollusc: Marisa cornuarietis (Giant ramshorn snail)

Details on test organisms:

Marisa cornuarietis were obtained as confirmed adults from a population cultured at ABC Laboratories which were descendents of a population of field collected snails received from Dr. Sharon File at the University of Puerto Rico. Wild specimens of M. cornuarietis were collected from Lake Guajataca, Puerto Rico and transported to ABC Laboratories. The snails were cultured under conditions similar to those specified for testing purposes (Don Thomas personal communication, (5). A 12 h light:12 h dark photoperiod with two thirty-minute transition periods was used with illumination provided by fluorescent lights. The cultures were set up on a single-pass flow-through system to maintain adequate water quality. Water temperature of cultures was maintained at 25 ºC. Snails were fed fresh, commercially purchased organically grown romaine lettuce (Lactuca sativa, romaine) and commercial algal wafers (supplied by Hikari, USA and purchased through a distributor).

Test type:

flow-through

Water media type:

freshwater

Limit test:

no

Total exposure duration:

84 d

Hardness:

Total Hardness: 224 to 274 mg CaCO3/L, mean 240 ± 12
Calcium: 65.4 to 69.1 mg calcium/L

Test temperature:

Temperature: 21.6 to 23.2 °C, mean 22.4 ± 0.4

pH:

pH: 7.8 to 8.8, mean 8.0 ± 0.1

Dissolved oxygen:

Dissolved Oxygen: 5.2 to 8.7 mg/L (62 to 104 % sat.), mean 7.1 ± 0.7

Nominal and measured concentrations:

Nominal: 0 (control) and 25 μg/L
Mean Measured:

Reference substance (positive control):

no

Key result

Duration:

84 d

Dose descriptor:

NOEC

Effect conc.:

>= 0.64 mg/L

Nominal / measured:

meas. (not specified)

Conc. based on:

test mat.

Basis for effect:

other: survival, reproduction, egg hatchability, egg mass, growth rate

Analytical: Mean measured concentrations of BPA from the 25 µg/L treatment ranged from 11.5 to 24.8 µg BPA/L and from 46 to 99 % of the nominal concentrations. The overall mean and standard deviation was 15 ± 3.2 µg/L. The diluter stock solution averaged 97 % of the nominal concentration. No residues of BPA were detected in the control at or above the MQL of 0.0593 µg BPA/L. Recoveries of BPA in the QC samples ranged from 81 to 120 % of the nominal concentrations during the exposure period.

Test solution temperatures as measured in the individual 25 µg/L replicates during the adult fecundity exposure ranged from 21.6 to 23.3 °C with an overall replicate mean of 22.4 ± 0.4 °C. The mean temperature for the duration of the study as collected by the constant temperature probe was 22.5 ± 0.5 ºC.

Dissolved oxygen concentrations in the individual control and 25 µg/L replicates throughout the adult fecundity exposure ranged from 5.21 to 8.65 mg/L (62 to 103 % saturation) with an overall replicate mean of 7.10 ± 0.7 mg/L. Test solution pH in the individual control and 25 µg/L replicates throughout the adult fecundity exposure ranged from 7.76 to 8.84 with an overall replicate mean of 8.03 ± 0.1. Total hardness in the individual control replicates during the adult fecundity exposure ranged from 224 to 274 mg CaCO3/L with an overall replicate mean of 240 ±12 mg CaCO3/L. These hardness values should be equated to a calcium concentration of 60 mg/L.

Biological: There was 100 % survival of adult snails in all treatments during the course of the 12 week trial. The mean number of eggs per clutch was 105 and 101 eggs/clutch for the control and 25 µg BPA/L treatments, respectively. The mean number of eggs per female per week was 148 ± 15.1 and 158 ± 22.9 for the control and 25 µg BPA/L treatments, respectively. The mean number of eggs per female per month was 592 ± 31and 631 ±61 for the control and 25 µg BPA/L treatments, respectively. The mean number of clutches per female per week was 1.4 ± 0.09 and 1.6 ± 0.20 for the control and 25 µg BPA/L treatments, respectively. The mean number of clutches per female per month was 5.7 ± 0.12 and 6.3 ± 0.23 for the control and 25 µg BPA/L treatments, respectively.

There was no significant effect of exposure to 25 µg/L BPA on adult egg production (P = 0.22), nor was there a significant difference among replicates within treatments (P = 0.22). There was, however a significant amount of variability among breeding pairs. Only 2.9 % of the total variance in egg production could be attributed to BPA treatment, whereas 3.9 % was due to differences among replicate tanks, and 93.2 % was due to variability among breeding pairs. The within-replicate (aquarium) coefficients of variation for this endpoint provide an estimate of the size of the variability among breeding pairs. These ranged from 12 % to 28 % and averaged 16 %.

Validity criteria fulfilled:

not applicable

Conclusions:

There was no significant effect of exposure to 25 μg/L BPA on adult egg production (P = 0.22), nor was there a significant difference among replicates within treatments (P = 0.22).

Executive summary:

This study was performed to examine the potential enhancement of adult fecundity when laboratory-cultured prosobranch snails, Marisa cornuarietis, were exposed to Bisphenol A under flow-through conditions at a temperature of 22 ºC for 12 weeks. There was no significant effect of exposure to 25 μg/L BPA on adult egg production (P = 0.22), nor was there a significant difference among replicates within treatments (P = 0.22). There was, however a significant amount of variability among breeding pairs. Only 2.9 % of the total variance in egg production could be attributed to BPA treatment, whereas 3.9 % was due to differences among replicate tanks, and 93.2 % was due to variability among breeding pairs. The within-replicate (aquarium) coefficients of variation for this endpoint provide an estimate of the size of the variability among breeding pairs. These ranged from 12 % to 28 % and averaged 16 %.

Description of key information

There are six key studies available which are reliable without restriction (Klimisch 1) and which address the long-term toxicity of Bisphenol A to aquatic invertebrates. These include studies on freshwater species: a Daphnia magna reproduction study according to OECD 211 revealing a NOEC of 3.146 mg/L (Caspers, 1998), a rotifer life-cycle study which followed the ASTM Guideline E1440-91 and report a NOEC of 1.8 mg/L (Sayers, 2006a; published in Mihaich et al., 2009), a 328-d reproduction study with Marisa cornuarietis snail with a NOEC of 0.025 mg/L (Warbritton et al., 2007; published as Forbes et al. 2008), and a range-finder which reported a NOEC of 0.025 mg/L (Warbritton et al, 2007b). For the studies with M. cornuarietis there was no guideline available at this point in time. However, these studies were conducted as a follow up of the Oehlmann et al., 2006, study with M. cornuarietis which had severe short-comings. The test setup of Warbritton et al., 2007, in contrast was coordinated with the rapporteur UK of the European Risk Assessment (2003, 2008, 2010) and was performed in accordance with the state of science and technology. Cafarella, 2006 (published in Mihaich et al., 2009), reported a 42-d NOEC of 0.49 mg/L for Hyalella azteca based on reproduction. For marine water species, there is one further key study, namely a life-cycle study according to EPA OPPTS 850.1350 test method with Americamysis bahia which determined a NOEC of 0.17 mg/L (Lee, 2010; published in Mihaich et al. 2018). Thus, in freshwater the key studies by Warbritton et al., 2007, which report a NOEC of 0.025 mg/L and for marine water the study of Lee, 2010, with the NOEC of 0.17 mg/L report of the lowest effect level.

Key value for chemical safety assessment

Fresh water invertebrates

Fresh water invertebrates

Effect concentration:

0.025 mg/L

Marine water invertebrates

Marine water invertebrates

Dose descriptor:

NOEC

Effect concentration:

> 3.146 mg/L

Additional information

Caspers (1998) conducted a chronic Daphnia magna reproduction study (OECD 211) including the assessment on molting of the water flea. Test concentrations were 0, 0.316, and 3.16 mg/L BPA. There was no effect on reproduction or molting behavior and the NOEC was thus equal to the highest test concentration of 3.16 mg/L.

Another key study for assessing the long-term toxicity of Bisphenol A to aquatic invertebrates is a life-cycle rotifer study with Bisphenol A (Sayers, 2006a; published in Mihaich et al., 2009). The purpose of this study was to determine the chronic toxicity of Bisphenol A to the rotifer Brachionus calyciflorus under static test conditions. Based on mean measured concentrations the NOEC for reproduction (intrinsic rate of increase) was determined to be 1.8 mg/L.

Warbritton et al. (2007a) conducted a 328-d study exposing snails (Marisa cornuarietis) to 0, 0.1, 1.0, 25, and 640 μg/L BPA under flow-through conditions. This study aimed to evaluate effects on mortality, fecundity, hatchability, and juvenile growth. There was no effect on fecundity or hatchability and for both endpoints a NOEC of > 640 µg/L was reported. For juvenile growth the NOEC was reported to be 25 µg/L. The precedent study by Warbritton, 2007b, investigated effects on M. cornuarietis over 84 days and reported a NOEC of >= 0.64 µg/L.

Cafarella, 2006 (published in Mihaich et al., 2009), reported a 42-d NOEC of 0.49 mg/L for Hyalella azteca based on reproduction.

In a 28-day life cycle study with the estuarine mysid shrimp (Americamysis bahia) (Lee, 2010; published in Mihaich et al. 2018), endpoints of F0 survival, growth (mean dry body weight and mean total body length) of both male and female mysids and reproduction (number of young released per female) were assessed. Test concentrations were 38, 75, 150, 300 and 600 µg/L BPA but effects were reported based on time-weighted average concentrations (18, 41, 74, 170 and 370 µg/L). Reproduction was the most sensitive endpoint with a NOEC of 170 µg/L BPA. The 28-d LC50 was reported > 370 µg/L as the mortality at the highest concentration was below 50 %.

Further studies were identified which were rated as Klimisch 2. Mansilha et al., 2013, and Jemec et al., 2012, performed their studies on D. magna according to OECD 211 and ISO 10706:2000, respectively, and reported 21-d NOECs based on relevant effects of 3 mg/L and 1.73 mg/L. Brennan et al., 2006, conducted a chronic study with D. magna according to ISO 10706 and reported a NOEC of 1 mg/L. Mu et al., 2005b, performed a D. magna reproduction study according to EPA/660/3-75-009 test method and determined a NOEC of 1.3 mg/L. Sieratowicz et al., 2011, reported a NOEC of 20 µg/L. Pascoe et al., 2002, investigated effects on Hydra vulgaris and reported a NOEC of 42 µg/L. Hill et al., 2002, reported a NOEC of 1.6 mg/L in a non-guideline study with the sponges Heteromyenia sp. and Eunapius fragilis. Finally, Watts et al., 2003, in their study on Chironomus riparius, reported a NOEC of 0.1 mg/L based on the time to larval moult and wet weight of larvae.

Thus, there are several studies which support the key studies and which report NOEC in the range of 0.02-3 mg/L.

In contrast, several other chronic studies which are listed and discussed in this chapter were rated as Klimisch 3 (not reliable) due to major short-comings or Klimisch 4 (not assignable) due to e.g. insufficient documentation and disregarded in the risk assessment. Full justifications for disregard are provided in the endpoint study records and the respective robust study summaries (e.g. Oehlmann et al., 2006, Ladewig et al., 2006, and Schulte-Oehlmann et al., 2001).

Concluding, there are several Klimisch 1 key studies and several Klimisch 2 supporting studies. The key studies by Warbritton et al., 2007, reported the lowest freshwater NOEC of 0.025 mg/L and the key study of Lee, 2010, determined the lowest marine water NOEC of 0.17 mg/L.