Registration Dossier

Ecotoxicological information

Endocrine disrupter testing in aquatic vertebrates – in vivo

Currently viewing:

Administrative data

Endpoint:
fish juvenile: (sub)lethal effects
Type of information:
experimental study
Adequacy of study:
key study
Study period:
from 2019-07-05 to 2020-04
Reliability:
1 (reliable without restriction)
Rationale for reliability incl. deficiencies:
guideline study

Data source

Reference
Reference Type:
study report
Title:
Unnamed
Year:
2020

Materials and methods

Test guideline
Qualifier:
according to
Guideline:
OECD TG 234 (Fish Sexual Development Test)
Version / remarks:
28 July 2011
Deviations:
yes
Remarks:
As per request of the evaluating authority, additional histopathological measurements of the liver and kidneys was performed.
GLP compliance:
yes

Test material

Reference
Name:
Unnamed
Type:
Constituent
Test material form:
liquid
Details on test material:
- Name of test material (as cited in study report): OMC
- Physical state: Liquid

Sampling and analysis

Analytical monitoring:
yes
Remarks:
LC-MS/MS
Details on sampling:
Test solution from a representative replicate tank at the control and the test concentration was sampled for chemical analysis during the equilibration phase at least once. Test solution from each replicate tank in the limit concentration and control was sampled for chemical analysis on study day (SD) 0 and on in-life study conclusion (end of the exposure period). Thus, during the remainder of the in-life study samples were collected from one replicate tank of the control and limit concentration. The replicate selected for weekly sampling rotated (week 1 – A, week 2 – B, week 3 – C, week 4 –D, and repeat) throughout the in-life phase until the in-life phase was completed. Test solutions (20 mL) were collected in duplicate, centrifuged (2795g or 5000 rpm for 30 min) and subsequently diluted with 20 mL acetonitrile (ACN). Samples were stored refrigerated (ca. 4 °C) in 40 mL glass vials. The duplicate sample was retained and stored at 4 °C (1 - 9 °C) to be used as necessary in the event of the need for re-analysis.

Test solutions

Vehicle:
no
Details on test solutions:
Since the range finding study did not indicate substance-related effects at the water solubility limit under test conditions, a limit test design was selected. Therefore, the test concentration series was a limit test consisting of 1x and 0.00x (control), where x was the water solubility limit value. Based on a target MTC of 50.0 μg/L (centrifuged samples), the target test concentrations were 50.0 and 0.00 (control) μg/L.

- Preparation of saturator columns:
Eight saturator columns were prepared, two for control groups and six for preparation of test item stock solutions. In general, columns were prepared by 125 g/arm pyrolytically cleaned glass wool into each arm of u-shaped glass columns (diameter = 1.54 cm, length = 152.4 cm). To further clean the glass wool, the glass wool was subsequently rinsed with acetone and the acetone purged from the system prior to test substance loading. For the columns used to prepare test substance stock solution, approximately 10 g of the test item was dissolved in 30 mL acetone and distributed equally to both arms of the column. A vacuum pump was used to pull the solution through the glass wool to provide an even coat. The acetone was then evacuated from the columns leaving the test item. Once the saturator columns were dried and the acetone evacuated, the columns were plumbed, and dechlorinated tap water pumped through them at a rate of 4 mL/min for 10 days to rinse the columns and remove any glass particles from the glass wool.

- Determination of the water solubility:
Prior to in-life testing, water solubility was determined at test conditions. On three consecutive days, three samples were collected, centrifuged (2795 x g or 5000 rpm for 30 min.), and preserved. After collection, samples were stored at 4 °C (1 - 9 °C). Once all samples were collected, centrifuged, and preserved, samples were analysed. Three primary factors were used to determine if saturation was achieved: 1) in-life test conditions were maintained throughout assessment, 2) differences between three consecutive samples was ≤20 %, and 3) differences between centrifuged and uncentrifuged samples was ≤20 %. When each of the three criteria were met, the mean concentration of the three consecutive samples was considered saturation or the practical water soluble concentration, and the in-life study commenced.

Test organisms

Aquatic vertebrate type:
fish
Test organisms (species):
Danio rerio (previous name: Brachydanio rerio)
Details on test organisms:
TEST ORGANISM
- Common name: zebrafish
- Source: Embryos were from breeding of adults from an in-house culture. Adult fish were originally obtained from Aquatic Research Organisms (Hampton, NH, USA)
- Brood stock selection:
The exposure phase was initiated with newly fertilized embryos produced from adult fish from a laboratory supply of reproductively mature animals cultured at 26 ± 1°C. The fish were actively spawning. Test fish were selected from a single laboratory stock acclimated for at least two weeks prior to the test under conditions of water quality and illumination similar to those used in the test. Brood stock fish were obtained from an in-house culture, as shipping of adult fish is stressful and will interfere with reliable spawning. Apparently healthy looking parental fish were used for the breeding. Parental fish did not receive treatment for disease in the two-week period preceding breeding. Moribund fish or fish with clinical signs of disease were not used to produce the embryos used in the study. Since fertilized eggs needed to be obtained from at least 3 clutches, 6 pair of zebrafish were bred concurrently so that egg laying and fertilization occurred as synchronously as possible. One false bottom egg crate was placed in each tank to separate breeding fish from eggs. During the egg collection phases, eggs were cultured in a separate brood chamber.

METHOD FOR PREPARATION AND COLLECTION OF FERTILIZED EGGS
The addition of larvae to each treatment tank at test initiation used a random assignment process. Fertilization, stage, and normalcy of development were ascertained using a dissecting microscope. Since the random assignment process spanned a period of time, the stages at the beginning of assignment and end of assignment were monitored. Since exposure was required as early as possible after fertilization before cleavage of the blastodisc commenced the following procedure was used:
1. Determination of egg clutches that were of sufficient quantity and quality. Fertilization (≥90 %), development stage (ca. 4 hpf), and rate of normal development (≥90 %) was determined using a dissecting microscope. Approximately, 240 normal-appearing embryos from 3 clutches were used to initiate the study.
2. Once the clutches to be used were identified, they were pooled.
3. Pooled embryos were subsequently sorted to remove unfertilized or abnormal embryos. The number of pooled embryos exceeded 240.
4. Random assignment of embryos to tanks commenced and finished when 30 embryos were assigned to each tank (3 rounds of 10 embryos/round). The developmental stage at the time of first allocation and final allocation for each tank was noted. A representative sample (voucher) was collected at the beginning and conclusion of random assignment. The sub-sample of voucher embryos was preserved in 10 % neutral buffered formalin (NBF) to verify the stage of development at test initiation. The stage of development at the initiation of embryo assignment was 6 and 7 at the conclusion of assignment to the exposure tanks.

POST-HATCH FEEDING
- Type/source of feed: brine shrimp nauplii
- Amount given: ad libidum
- Frequency of feeding: twice per day
- Uneaten food and faecal material were removed from the test vessels daily 30 minutes after the last feeding by carefully cleaning the bottom of each tank using suction. Feeding was stopped 24 h before end of the test.

Study design

Test type:
flow-through
Water media type:
freshwater
Remarks:
Dechlorinated (DeCl2) tap water
Limit test:
yes
Total exposure duration:
63 d
Remarks on exposure duration:
ca. 63 days (until day 60 after end of hatch)

Test conditions

Hardness:
84 - 120 mg/L as CaCO3
Test temperature:
25.8 - 27.1 °C
pH:
6.7 - 7.6
Dissolved oxygen:
5.9 - 8.1 mg/L
Conductivity:
376 - 567 µS/cm
Nominal and measured concentrations:
Nominal concentrations: 0 (control) and 50 µg/L
Mean measured concentrations:
Details on test conditions:
TEST SYSTEM
- Test vessel: glass tanks in a flow-through system
- Material, size, headspace, fill volume: The system had water-contact components of glass (aquaria), stainless steel (diluter housing and water bath), and Teflon (tubing responsible for test material delivery). Exposure tanks were glass aquaria (with approximate measurements of 22.5 x 14.0 x 16.5 cm deep) equipped with standpipes that result in an actual tank volume of 10.0 L and minimum water depth of 10 to 15 cm.
- Aeration: none
- Type of flow-through: flow-through diluter system - Benoit Mini-Diluter
- Renewal rate of test solution (frequency/flow rate): 2.7 L/min, ca. 6.5 tank volumes/day
- No. of organisms per vessel: 30 freshly fertilised eggs (embryos)
- No. of vessels per concentration (replicates): 4
- No. of vessels per control (replicates): 4
- Vehicle control performed: no
- Biomass loading rate: At 28 dpf (SD 28), the number of fish per replicate was assessed to determine if each replicate contains as equal a number of fish as possible. The loading rate in the control ranged from 0.84 g/L to 1.19 g/L and 0.32 g/L to 0.55 g/L in the test item treatment at the end of exposure. Since the difference between the number of fish in each replicate of the control or the test item treatment at 28 dpf was ≤ 4 test organisms, no re-distribution was performed.

TEST MEDIUM / WATER PARAMETERS
- Source/preparation of dilution water: Dechlorinated (charcoal-filtered) tap water prepared at the laboratory using 4-filter system - a multimedia filter to remove suspended solids in the feed water; a 10” pre-treatment filter (5 µm) to remove any additional solids; a 0.1 m³ activated virgin carbon treatment filter to remove chlorine, ammonia, and higher molecular weight organics; and a 5 µm polishing filter to remove any carbon particles from the carbon treatment phase.
- Total organic carbon: <0.3 - 1.3 mg/L
- Alkalinity: 36 - 64 mg/L as CaCO3
- Iodide: 8.1 - 8.3 µg/L
- Ammonia: <0.06 µg/L
- Chlorine: <0.05 µg/L
- Culture medium different from test medium: no
- Intervals of water quality measurement: twice a month

OTHER TEST CONDITIONS
- Adjustment of pH: none
- Photoperiod: 16 h light : 8 h dark
- Light intensity: 570 - 946 lux

EFFECT PARAMETERS MEASURED:
- survival,
- behaviour of hatchlings and juveniles,
- body weights and length,
- VTG concentration,
- sex ratios,
- histopathology.

TEST CONCENTRATIONS
- Justification for using less concentrations than requested by guideline: Limit test is generally accepted. No effects were observed in the range-finding study, therefore, a limit test at the water solubility of the test substance under the test conditions was performed.

RANGE FINDING STUDY
- Test concentrations: 0 (control), 2.0 and 20.0 µg/L
- Results used to determine the conditions for the definitive study: In the range-finding study, survival above 90 % in all replicates and 100 % hatching success was observed throughout the exposure period. thus, a limit test at the water solubility limit was chosen for the definitive test.
Reference substance (positive control):
no

Results and discussion

Effect concentrationsopen allclose all
Key result
Duration:
63 d
Dose descriptor:
NOEC
Effect conc.:
< 46.9 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
gonadal histology
Remarks:
and hepatotoxicity
Key result
Duration:
63 d
Dose descriptor:
LOEC
Effect conc.:
46.9 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
gonadal histology
Remarks:
and hepatotoxicity
Key result
Duration:
63 d
Dose descriptor:
NOEC
Effect conc.:
< 46.9 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
length
Remarks:
and weight
Key result
Duration:
63 d
Dose descriptor:
LOEC
Effect conc.:
46.9 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
length
Remarks:
and weight
Duration:
63 d
Dose descriptor:
NOEC
Effect conc.:
>= 46.9 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
mortality
Key result
Duration:
63 d
Dose descriptor:
NOEC
Effect conc.:
>= 46.9 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
sex ratio
Duration:
63 d
Dose descriptor:
NOEC
Effect conc.:
>= 46.9 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
number hatched
Key result
Duration:
63 d
Dose descriptor:
NOEC
Effect conc.:
>= 46.9 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
vitellogenin level
Key result
Duration:
63 d
Dose descriptor:
NOEC
Effect conc.:
>= 46.9 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
genetic/phenotypic sex ratio
Remarks:
proportion of undifferentiated gonads

Any other information on results incl. tables

Determination of practical water solubility

Three consecutive samples met the three primary factors used to determine if saturation had been achieved: 1) in-life test conditions maintained throughout assessment, 2) difference in concentration ≤20 %, and 3) differences between centrifuged and uncentrifuged ≤20 %. The difference between each of the three consecutive samples relative to the mean ranged from 1.9 - 5.7 %. The difference between centrifuged and uncentrifuged samples for the three consecutive samples ranged from 7.3 - 11.6 %. The mean concentration based on the three consecutive centrifuged samples was 48.6 μg/L test item. Based on these results, the target concentration was set at 50.0 μg/L test item.

 

Water quality measurements and test system performance

Iodide (I-) levels in the dilution water were measured on SD 0 and 63 (in-life conclusion) of the study and contained 8.1 ± 0.08 and 8.3 ± 0.03 μg/L I-, respectively. All physicochemical water quality parameters in study were within acceptable ranges.

 

Confirmation of test concentrations

Target OMC concentrations selected for the FSDT definitive study were 0.0 (control) and 50.0 μg/L. The dilution water control solutions showed no detectable levels of the test item (<3.04 μg/L limit of quantitation, LOQ). Dilution water control samples collected during the study contained < LOD (LOD = 0.912 μg/L). Since time interval between sample collection events was consistent throughout study, the mean measured concentration represented a reasonable estimation of exposure concentration. The mean measured concentration represented the average of each data point from SD 0, 7, 14, 21, 28, 35, 42, 49, 56, and 63 for each replicate on SD 0 and 63; and sequential sampling of each replicate (A-D) of the control and each treatment per facility SOP. Interquartile ranges (IQRs) determined for the 0.0 (control), and 50.0 μg/L treatment were 0.46 - 0.46 and 24.15 - 61.35 μg/L test item, respectively. Based on IQR analysis of the control and each treatment group, five outliers were identified, original sample 50.0 μg/L replicates A-D SD 0 (8.26, 7.91, 7.30, and 7.75 μg/L test item), and replicate B SD 7 (22.8 μg/L test item). In this case, results of the duplicate samples of 50.0 μg/L replicates A-D (40.5, 40.3, 44.4, and 44.8 μg/L test item) on SD 0 and replicate B (47.4 μg/L test item) on SD 7 were used in the analysis. The corresponding mean measured concentrations in the definitive study were <LOD and 46.9 μg/L test item, respectively. The coefficient of variation (CV) [(Standard deviation/mean)100] was based on the standard deviation of the four replicate means (n = 4) for the control and the four replicate means (n = 4) for each treatment per facility SOP. The CVs of the intra-replicate means of the measured test concentrations for the control and 50.0 μg/L test item treatment were 0.0 % and 5.2 %, respectively. Each sample was within 20 % (35.5 - 56.3 μg/L test item) of the mean measured OMC concentration which were acceptable based on the criteria established in the test guidance and the study plan.

 

Mortality

No mortality was observed until hatching was completed on SD 4. Mean survival in the control and 46.9 μg/L treatment (mean measured concentration) were 82.5 % and 81.7 %, respectively. Survival in the 46.9 μg/L treatment was not significantly different from the control (t-test, p = 0.654).

 

Hatching success

Hatching was observed from SD 2-4 (end of hatch) in both the control and the test item treatment. The hatching success in the control and the treatment was 100 %. Therefore, no statistical analysis of this endpoint was performed.

 

Sex ratio

Sex ratios of control fish was 51.5 : 44.4 : 4.0 (%Female [F] : %Male [M] : %Undetermined [U]). Phenotypic sex ratios of fish exposed to 46.9 μg/L test item (mean measured concentration) was 55.1 : 34.7 : 10.2 (%F : M : U). Phenotypic sex ratio in the treatment was not significantly different from the control (t-test, p = 0.683 female and p = 0.174 male). In addition, the incidence of undetermined phenotypic sex (undifferentiated gonad) was not significantly different from the control (Mann-Whitney Rank Sum test, p = 0.057) indicating that the occurrence of primary sexual differentiation was not altered by the test item exposure in the present study.

 

Growth

Standard Length

Mean standard lengths were 22.1 mm in control female zebrafish and 15.1 mm in female zebrafish exposed to 46.9 μg/L test item (mean measured concentration). Mean standard lengths were 19.4 mm in control male zebrafish and 15.7 mm in male zebrafish exposed to the test item. Mean standard lengths were 15.1 mm in control undifferentiated zebrafish and 14.4 mm in undifferentiated zebrafish exposed to 46.9 μg/L test item. Standard lengths in both female and male fish exposed to the test item were significantly less than the control (t-test, p < 0.001 and p < 0.006, respectively). Standard length in undifferentiated fish exposed to the test item was not significantly different from the control (Mann-Whitney Rank Sum test, p = 0.800).

 

Body Weight

Mean body weights were 194.6 mg in control female zebrafish and 79.9 mg in female zebrafish exposed to the test item. Mean body weights were 132.0 mg in control male zebrafish and 83.3 mg in male zebrafish from the treatment group. Mean body weights were 169.8 mg in control undifferentiated zebrafish and 70.3 mg in undifferentiated treated zebrafish. Body weights in female and male fish exposed to the test item were significantly less than the control (Mann-Whitney Rank Sum test, p = 0.029 and t-test, p = 0.002). Body weight in undifferentiated fish exposed to the test item was not significantly different than the control (Mann-Whitney Rank Sum test, p = 0.133). Overall, the effects on growth occurred after the onset of gonadal differentiation.

 

Plasma Vitellogenin (VTG)

Mean plasma VTG levels were 571485.3 ng/mL in control female zebrafish and 564631.1 ng/mL in female zebrafish exposed to 46.9 μg/L test item (mean measured concentration). Mean plasma VTG levels were 255.4 ng/mL in control male zebrafish and 252.3 ng/mL in male zebrafish exposed to the test item. Plasma VTG LOQs were 4800.0 and 4.8 ng/mL for female and male zebrafish, respectively. Plasma VTG levels in female and male fish exposed to 46.9 μg/L test item were not significantly different from the control (t-test, p = 0.880 and Mann-Whitney Rank Sum test, p = 0.114, respectively). Plasma VTG levels in undifferentiated fish exposed to the test item were not significantly different from the control (Mann-Whitney Rank Sum test, p = 0.533).

 

Histopathology

Kidney

There were no exposure-related effects in the kidney.

 

Liver

Exposure-related effects in the liver included hepatocyte karyomegaly (minimal to moderate), oval cell proliferation (mild to moderate), single cell necrosis (minimal to mild), and increased mitotic figures (minimal). These findings were not distinct in undifferentiated or hermaphroditic fish.

Hepatocyte karyomegaly was characterised by nuclear enlargement with only modest cytoplasmic enlargement. Karyomegalic nuclei, which were oval or irregular in appearance, were frequently observed in paired or binucleate cells, and were typically hyperchromatic with prominent nucleoli. Because slight anisokaryosis noted as a variation in nuclear size was evident in the livers of occasional control fish, diagnoses of karyomegaly were not recorded unless affected nuclei were at least twice the diameter of average hepatocyte nuclei in the same liver section. The prevalence and severity of karyomegaly in the 46.9 μg/L test item-exposed males and females were comparable, and this finding was not observed in control fish. Oval cell proliferation occurred in three test item-exposed females. Oval cell proliferation was characterised by the presence of small cells with ovoid hyperchromatic nuclei and scant amounts of faintly basophilic fibrillar cytoplasm, which coursed through the liver in short chains or small clusters. Single cell necrosis was observed occasionally in the livers of control fish of both sexes, but the prevalence of this finding increased substantially in the test item-exposed males and females relative to control. Necrotic cells were small, round or irregular, with eosinophilic cytoplasm and pyknotic or karyorrhectic nuclei, consistent with an apoptotic form of cell death. Necrotic cells were typically surrounded by a clear halo and most often occurred within the cytoplasm of sinusoidal macrophages or other hepatocytes (phagocytosis). Increased mitotic figures occurred in only a single test item-exposed female; however, as that particular female additionally had mild hepatic karyomegaly, increased mitotic division was interpreted as further evidence of a regenerative response to hepatic injury.

 

Gonads

Exposure-related effects in the gonads included decreased mean ovarian stage scores in test item-exposed fish relative to controls. While the majority of control females (32 of 52 specimens) had ovarian stage scores of 1.0, characterised as immature ovaries characterised by the presence of both cortical alveolar and perinucleolar phase oocytes; the majority of females exposed to the test item (36 of 54 specimens) had Stage 0.0 ovaries, in which the most developed oocytes were perinucleolar phase. Undifferentiated gonads occurred in 3 of the 99 control fish and 8 of 98 fish exposed to 46.9 μg/L test item, respectively. Two of the 98 fish exposed to the test item were hermaphroditic. Seven instances of “intersex” specimens were identified in the present study, four of which occurred in control fish, and three in test item-exposed fish. Specific findings consisted of testicular oocytes (minimal) in three control males; ovarian spermatogenesis (minimal) in one control and one test item-exposed female; and hermaphroditism (not graded for severity) in two exposed fish. The finding of testicular oocytes was recorded for gonads in which one or more perinucleolar oocyte was embedded within the germinal epithelium of tissue that was otherwise clearly testicular. Conversely, in gonads with ovarian spermatogenesis, the majority of examined gonads was specifically ovarian, with the exception of a small portion of spermatogenic tissue in a single section. In fish diagnosed as hermaphroditic, different segments of the same gonad appeared definitively testicular or ovarian. Hermaphroditic gonads were not staged, and because these fish could not be definitively sexed, they were tabulated as a separate category resulting in either male, female, hermaphrodite, and undifferentiated designations. An additional intersex-like finding recorded in two control males was the presence of a gonadal duct with a female phenotype. This developmental anomaly was characterised by attachment of the testis to the dorsal coelomic mesothelium at two sites creating an intervening space. It should be noted that immature testes in male fish typically have a single attachment site with no space. All of these intersex changes were specifically and clearly interpreted as different manifestations of juvenile hermaphroditism, which is a well-documented phenomenon in normal laboratory zebrafish. Consequently, none of these findings were considered to be related to the test item treatment.

 

Summary of Histopathological Findings

The liver findings of hepatocyte karyomegaly, oval cell proliferation, single cell necrosis (increased prevalence compared to controls), and increased mitotic figures in test item-exposed fish were interpreted as indications of hepatic toxicity. Oval cells are progenitors of hepatocytes and bile duct epithelial cells, and oval cell proliferation in particular is typically associated with substantial hepatocyte loss. However, delayed ovarian development observed in test item-exposed females, and increased prevalence of undifferentiated gonads, was most likely related to the treatment-induced decrease in somatic growth. Although test item exposure was associated with reduced growth (mean body weight and standard length) relative to controls in both sexes, female fish appeared to be affected to a greater degree than males. As might be anticipated, fish with undifferentiated gonads were generally among the smallest fish in the study.

The seven instances of “intersex” gonads identified in present study, of which four were identified in control fish, were not considered to be treatment-related. During early development (2-3 wpf), the majority of zebrafish have gonads that exhibit the female (ovarian) phenotype, which begin gradually transitioning to the male (testicular) phenotype at approximately 5 weeks post-fertilization (wpf). Thus, at ca. 8 wpf at which point the exposure concluded and fish were sacrificed in the present study, it is not surprising that remnants of this transitional phase were evident as male-dominated (testicular oocyte), female-dominated (ovarian spermatogenesis), or neither-dominant (hermaphroditism) forms of intersex. Most importantly, no treatment-related differences in sex ratio or the frequency of undifferentiated gonads were noted. The subtle, but statistically insignificant, increase in the ratio of females to males in the test item-exposed fish compared to controls was most likely a function of delayed transition from the male to female phenotype as opposed to treatment-induced gonadal feminization.

 

Clinical signs of toxicity

Clinical signs of toxicity were not observed during the conduct of the present study.

Applicant's summary and conclusion

Validity criteria fulfilled:
yes
Conclusions:
In conclusion, the present study indicated that exposure to 46.9 μg/L test item (ca. limit of water solubility) did not reduce survival, alter sex ratio, increase the proportion of undifferentiated gonads, or alter plasma vitellogenin levels relative to control zebrafish. However, both standard length and body weight were reduced in female and male fish, but not in undifferentiated fish exposed to 46.9 μg/L mean measured test item concentration. Early life stage exposure of zebrafish to 50 μg/L (target concentration) test item was associated with morphologic changes in the liver consistent with hepatotoxicity. It is reasonable to surmise that hepatotoxicity was responsible for the decreased somatic growth experienced by test item-treated fish, which in turn was likely the cause of delayed ovarian differentiation in females as determined by gonadal stage scoring. The low incidence of intersex in both control and test item-exposed fish, which manifested as either testicular oocytes, ovarian spermatogenesis, or hermaphroditism, was not considered to be related to the test item treatment, but was instead a function of gonadal immaturity in a species that is known to exhibit juvenile hermaphroditism. Based on these results, the NOEC was <46.9 μg/L and the LOEC was 46.9 μg/L. The highest tested concentration represented the limit of solubility under test conditions. However, it should be noted that the response observed was not related with an endocrine-mediated mode of action.
Executive summary:

Freshly fertilized zebrafish eggs were exposed to the limit of solubility of the test substance under test conditions and dilution water control in accordance with OECD 234. Prior to in-life testing, practical water solubility of the test item was determined at test conditions.

To prepare stock solution required for dilution to the test solution, eight saturator columns were prepared, two for control groups and six for preparation of test item stock solutions.

A flow-through diluter system (Benoit Mini-Diluter) was used in the performance of the FSDT exposure. The Benoit mini-diluter system had water-contact components of glass (aquaria), stainless steel (diluter housing and water bath), and Teflon® (tubing responsible for test material delivery). Exposure tanks were glass aquaria equipped with standpipes with an actual tank volume of 10 L. The flow rate to each tank was 2.7 L/h (6.5 volume exchanges/d). Fluorescent lighting was used to provide a photoperiod of 16 h light and 8 h dark at an intensity that ranged from 540 to 1000 lux (lumens/m2) at the water surface. Water temperature was maintained at 27 ± 2 °C and did not differ by more than ±1.5 °C between test chambers at any one time. In addition, pH was maintained between 6.5 to 9.0, and the dissolved oxygen (DO) concentration was maintained at ≥ 4.9 mg/L (> 60 % of the air saturation) in each test tank.

Temperature was measured daily; and pH, DO, and light intensity (lux) were measured two times per week in all control and treatment tanks. Total hardness, alkalinity, and conductivity were measured in all chambers at in-life initiation and conclusion. Iodine as I- was measured in the dilution water on life initiation and conclusion. TOC was measured in the dilution water weekly throughout the in-life phase.

Test solution from each replicate tank in the limit concentration and control was sampled for chemical analysis on SD 0 and on in-life study conclusion. Thus, during the remainder of the in-life study samples were collected from one replicate tank of the control and limit concentration. The replicate selected for weekly sampling rotated (week 1 – A, week 2 – B, week 3 – C, week 4 –D, and repeat) throughout the in-life phase until the in-life phase was completed.

Thirty embryos per replicate tank, and 4 replicates per treatment (120 embryos/treatment) during exposure phases were used. Embryos from three clutches were double-selected using a dissecting microscope to ensure fertilization and normal appearing development and mixed prior to addition to replicate tanks. Exposure was started within 4 h after fertilization. The test chemical was delivered to the exposure chambers using a continuous-flow diluter. The exposure phase was conducted for 60 days post-hatch (dph). All exposure condition parameters were evaluated at pre-exposure phase d 0, in-life exposure phase d 0, and in-life test termination. Samples for analytical monitoring were collected on SD 0, 7, 14, 21, 28, 35, 42, 49, 56, and 63 (in-life termination) and were analysed after centrifugation.

The following parameters were determined throughout the exposure period - survival (daily), body weights and length, behaviour of hatchlings and juveniles (daily), vitellogenin (VTG), sex ratios, histopathology of the kidney, liver and gonads.

Results of the practical water solubility analyses of centrifuged samples indicated that the mean concentration based on the three consecutive centrifuged samples was 48.6 μg/L test item. Therefore, the target test item concentrations selected for the FSDT study were 0.0 (control) and 50 μg/L. The corresponding mean measured concentration in the definitive study was 46.9 μg/L for the treatment and <LOD for the control. Mean survival in the control and the treatment was 82.5 % and 81.7 %, respectively. Mortality in the treatment was not significantly different from the control. The hatching success in the control and the treatment was 100 % and was complete by 4 days post-fertilization (dpf). Phenotypic sex ratio in the treatment was not significantly different from the control for both males and females. In addition, the incidence of undetermined (undifferentiated) phenotypic sex (undifferentiated gonad) was not significantly different from the control. Standard lengths in both female and male fish exposed to 46.9 μg/L test item were significantly less than those of the control group. Standard length in undifferentiated fish exposed to the test item was not significantly different from the control. Body weights in female and male fish exposed to the test item were significantly less than those of the control. Body weight in undifferentiated test item-exposed fish was not significantly different from the control. Plasma VTG levels in female and male fish exposed to the test item concentration were not significantly different from the control. Plasma VTG levels in undifferentiated fish exposed the test item were not significantly different from the control.

Exposure-related effects observed in the liver included hepatocyte karyomegaly (minimal to moderate), oval cell proliferation (mild to moderate), single cell necrosis (minimal to mild), and increased mitotic figures (minimal). These findings were not distinct in undifferentiated or hermaphroditic fish. The liver findings of hepatocyte karyomegaly, oval cell proliferation, single cell necrosis (increased prevalence compared to controls) and increased mitotic figures in 46.9 μg/L test item-exposed fish were interpreted as indications of hepatic toxicity. Exposure-related effects in the gonads included decreased mean ovarian stage scores in 46.9 μg/L test item-exposed fish relative to controls. However, the delayed ovarian differentiation observed in the treatment-exposed females was related to the treatment-induced decrease in somatic growth. The seven instances of “intersex” gonads identified (four of which were identified in control fish) were not considered to be treatment-related. The subtle, but statistically insignificant, increase in the ratio of females to males in the test item-exposed fish compared to controls was most likely a function of delayed transition from the male to female phenotype as opposed to treatment-induced gonadal feminization. No significant effects on behaviour or signs of overt toxicity were noted.

In conclusion, the present study indicated that exposure to 46.9 μg/L test item (ca. limit of water solubility) did not reduce survival, alter sex ratio, increase the proportion of undifferentiated gonads, or alter plasma vitellogenin levels relative to control zebrafish. However, both standard length and body weight were reduced in female and male fish, but not in undifferentiated fish exposed to 46.9 μg/L mean measured test item concentration. Early life stage exposure of zebrafish to 50 μg/L (target concentration) test item was associated with morphologic changes in the liver consistent with hepatotoxicity. It is reasonable to surmise that hepatotoxicity was responsible for the decreased somatic growth experienced by test item-treated fish, which in turn was likely the cause of delayed ovarian differentiation in females as determined by gonadal stage scoring. The low incidence of intersex in both control and test item-exposed fish, which manifested as either testicular oocytes, ovarian spermatogenesis, or hermaphroditism, was not considered to be related to the test item treatment, but was instead a function of gonadal immaturity in a species that is known to exhibit juvenile hermaphroditism. Based on these results, the NOEC was <46.9 μg/L and the LOEC was 46.9 μg/L. The highest tested concentration represented the limit of solubility under test conditions. However, it should be noted that the response observed was not related with an endocrine-mediated mode of action.