Registration Dossier

Ecotoxicological information

Endocrine disrupter testing in aquatic vertebrates – in vivo

Administrative data

Endpoint:
fish early-life stage: reproduction, (sub)lethal effects
Type of information:
experimental study
Adequacy of study:
key study
Study period:
May 23rd to December 4th, 2019
Reliability:
1 (reliable without restriction)
Rationale for reliability incl. deficiencies:
guideline study

Data source

Reference
Reference Type:
study report
Title:
Unnamed
Year:
2020
Report Date:
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:
deviations from the study plan
GLP compliance:
yes (incl. certificate)

Test material

Reference
Name:
Unnamed
Type:
Constituent

Sampling and analysis

Analytical monitoring:
yes
Details on sampling:
The test item quantification was performed according to a validated analytical method to measure the concentration of test item. All replicates were analyzed individually during the test at least twice a week.

Test solutions

Details on test solutions:
The test solutions were prepared by dilution of these stock solutions with a 10 000 factor to obtain a DMSO concentration approximately of 0.1 mL/L. Since the test item was poorly soluble under the test conditions, stock solutions were prepared in solvent dimethylsulfoxide. The test solutions were prepared by dilution of this stock solution with vigorous stirring to aid dispersion. Fresh stock solutions were prepared when the previous solutions were almost finished (approximately every 5 days). The test item stock solution was stored at room temperature. From 28/06/2019, the stock solutions were prepared in 250 mL in order to compensate for the increase in flow rate.

Test organisms

Aquatic vertebrate type:
fish
Test organisms (species):
Danio rerio (previous name: Brachydanio rerio)
Details on test organisms:
TEST ORGANISM
- Common name: Zebra fish
- Source: Fertilised eggs of Danio rerio produced from the laboratory breeders were used. Zebrafish genitors were supplied from PRECI, SFR Biosciences US8/UMS3444, LYON, France
- Life stage: fertilised eggs
- Method of breeding: At the laboratory, male and female breeders were separated in aquariums with a continuous renewal of the water. The day before the experiment, 5 males and 3 females were introduced in 3 nest boxes at approximately 27°C ± 2. The "nest boxes" used were homemade systems made of buckets, the bottom of which is provided with a mesh allowing the passage of eggs. These buckets were arranged in glass aquariums.

REARING CONDITIONS FOR GENITORS
- Food type: once per day with small pellets (e.g. Small Granular, Special Diets Services – Protein 52%, Lipid 12%, Ash 10%, Fibre 2%) and once with frozen food (chironomus, artemia, from e.g. HAEGEL)
- Amount: ad libitum, food is supplied in excess and uneaten food is removed approximately 1 hour after feeding
- Frequency: twice a day
- Medium: water from the water system (mix of tap water and deionized water). The water system consists in a biological filter (noodles), UV treatment and particle filter with continuous circulation.
- Temperature: 21°C ± 2
- Light/dark cycle: 16 h/8 h
- Dissolved oxygen concentration: ≥ 80 % of the air saturation value

METHOD FOR PREPARATION AND COLLECTION OF FERTILIZED EGGS
- Numbers of parental fish (i.e. of females used to provide required number of eggs): 3 females
- Method of collection of fertilised eggs: Eggs laying and fertilisation occurred at « dawn », simulated by an increase of light intensity. Eggs were harvested.
- Subsequent handling of eggs: a first selection was done in order to eliminate unfertilised eggs. To avoid genetic bias, eggs were collected from three breeding groups, mixed and randomly selected to initiate the test. Eggs were checked individually before their introduction in the test aquaria.

POST-HATCH FEEDING
- Start date: 27/05/2019
- Type/source of feed: adapted food. Surplus food and faeces were removed, as necessary to avoid accumulation of waste.
- Frequency of feeding: depending on the stage of development, fish were fed with adapted food several times a day

Study design

Test type:
flow-through
Water media type:
freshwater
Limit test:
no
Total exposure duration:
64 d
Remarks on exposure duration:
The exposure started as soon as possible after fertilization and before cleavage of the blastodisc commences and no later than 12 h post fertilization. The test duration of the study was until 60 days post-hatch.

Test conditions

Hardness:
9.53 - 12.8 °dH
Test temperature:
25.1-28.8 °C
pH:
8.0-8.3
Dissolved oxygen:
89-99.6 %
Conductivity:
301.0 - 364.5 μS/cm
Nominal and measured concentrations:
0 -10 – 32 – 100 – 320 and 1000 µg/l (nominal)
0 - 10.4 - 28.2 - 66.6 - 229.4 and 588.2 µg/L (measured, arithmetic mean)
Details on test conditions:
TEST SYSTEM
- Test vessel: glass aquarium of approximately 7L, filled to 5 L.
- Type: open
- Aeration: no
- Type of flow-through: The different test item solutions were prepared with peristaltic pumps from the stock solution in mixing cells. The flow in each aquarium was controlled with aquaria individual taps
- Renewal rate of test solution: The flow rate during the test was measured once a week. flow rate (control) 17.9-53.7
- No. of organisms per vessel: 30 eggs per replicate. On day 32 of the study period the number of fish per replicate was checked and for C06 concentration (nominal concentration at 320 µg.L-1) all alive fish were pooled in the same replicate so that fish density between treatment levels was kept as equal as possible
- No. of vessels per concentration (replicates): four
- No. of vessels per control (replicates): four
- Vehicle control performed: yes
- No. of vessels per vehicle control (replicates): four

TEST MEDIUM / WATER PARAMETERS
- Source/preparation of dilution water: the water used for testing was freshwater obtained from a mixture of deionized water and tap water issued from the laboratory's water system
- Total organic carbon: 0.52 mg/l
- Particulate matter: < 2 mg/l
- Total organophosphorus pesticides: <50 ng/l
- Total organochlorine pesticides plus polychlorinated biphenyls: <5 ng/l
- Unionised ammonia: <5 µg/l
- Residual chlorine: 0 µg/l
- Total organic chlorine: <25 ng/l
- Ca/mg ratio: 3.2
- Culture medium different from test medium: water from the water system (mix of tap water and deionized water). The water system consists in a biological filter, UV treatment and particle filter with continuous circulation
- Intervals of water quality measurement: Temperature was automatically recorded every two hours. Dissolved oxygen, pH and conductivity were measured once a week. Hardness was measured at the beginning and at the end of the test

OTHER TEST CONDITIONS
- Adjustment of pH: not reported
- Photoperiod: light/dark cycle of 14/10 hrs
- Light intensity: 540-1080 lux

EFFECT PARAMETERS MEASURED: Every day, the number of surviving and hatched organisms were recorded. Eggs, embryo and larvae or dead juveniles were removed as soon as they were recorded. At the end of the test: the wet weights per replicate of surviving fish was determined (blotted dry) and the standard length of surviving fish was measured. Fish were measured using a numeric calliper.
- Fish Sampling: fish sampling was performed at termination of the test. All surviving fish were euthanized in order to carry out the measurements of weight and length. For this, fish were placed for about 10 minutes in a crystallizer containing sodium methanesulfonate or MS222 (CAS number 886-86-2) at a lethal concentration of 300 mg.L-1 buffered with 300 mg NaHCO3 per L. After euthanasia, the length of fish was measured individually with numeric calliper and fish were transferred in previously weighed aluminium platforms and weighed to obtain the wet weight. The weighing scale used had a precision of 0.1 mg.
- Sampling for VTG analysis and sex determination via histological evaluation: all fish were sampled and prepared for analysis of sex and VTG. Head and tail of each fish were separated from the body of the fish by cuts made right behind the pectoral fins, and right behind the dorsal fin, using a scalpel. The head and tail part from each fish were pooled, weighed and individually numbered, frozen in liquid nitrogen and stored at -70ºC or less for VTG analysis. For histological evaluation, the body parts of the fish were numbered and fixed in Bouin liquid for 24 hours. Subsequently, samples were rinsed with 70% reagent grade ethanol and stored in 10% neutral buffered formalin until further processed.
All fish were analyzed histologically to determine sex. For the VTG measurements, at first, a sub-sampling of 16 fish from each replicate was done and then all fish were used for the VTG measurement to increase the robustness of this endpoint.

VTG analysis
The head and tail were homogenised in 1.5 mL Eppendorf with 1 mL of homogenisation buffer. Stainless steel beads were added and Eppendorf were placed in Retsch Mixer Mill (MM-400) to grind sample. The samples were placed on ice until centrifugation at 4°C (15 000 g for 30 min). 50 µL of supernatant without fat or pellet fractions were dispensed into two tubes stored at -80°C until use.
The VTG measurements were performed using zebrafish Vitellogenin ELISA kit from Biosense (Product Number: V01008402, Batch: 38220009; expiry date: 06/08/2020). This ELISA kit utilises specific binding between antibodies and vitellogenin (VTG) to quantify VTG in samples from zebrafish. The microplates wells were pre-coated with a specific Capture antibody that binds to VTG in standard and sample added to the wells. A different Vtg-specific Detecting antibody is added to create a sandwich of Vtg and antibody, which is detected with an enzyme-labelled Secondary antibody. The enzyme activity is determined by adding a substrate that gives a coloured product, and the colour intensity is directly proportional to the amount of Vtg present.
The day of measurement, samples were thawed at room temperature and diluted in PBS buffered (Phosphate buffered saline) at two concentrations. Measurements were performed on two replicates of each concentration at 492 nm with microplate reader (Varioskan). Homogenisation buffer: 12 mL Tris-HCl (SIGMA; Ref T2194; Batch: SLCB5415 expiry date: 31/01/2020) pH 7,4 (+ 120 µl Protease inhibitor cocktail (SIGMA; Ref: P8340; Batch: 118M4154V; expiry date: 22/12/2022)

Histopathological evaluation
The euthanized, dissected and fixed fish bodies were sent to the laboratory for histology processing and pathology evaluation. For evaluation, no fish was available from group 7. From 6, there were only fish from one replicate. A total of 559 fish were evaluated.
The fish were trimmed, processed, embedded in paraffin wax and cut at step sections at an approximate thickness of 2-4 µm, and stained by hematoxylin and eosin (HE).
For trimming up to cutting, the gonads were embedded together for transversal sectioning in the same paraffin block, and each block was cut until the largest cross-sectional area of the gonads was apparent in unstained sections. Six step sections (each 2-4 microns thick) were made at approximately 50-micron intervals, and all six sections were mounted on glass slides. From all samples, the gonads, liver and kidney of all treatment groups (kidneys and livers were present on the same sections as gonads) were evaluated.
The animal data were transferred manually into EXCEL sheets. The microscopic findings were recorded during histopathologic examination by the pathologist and directly entered into the data sheets.
The microscopic findings were recorded during histopathology examination by the pathologist and directly entered into the PathData System
- Diagnostic Terminology: Histologic changes including stages were described. Severity scores were assigned on a scale of grade 1 (minimal) to 4 (severe). The sex, and all changes and stages are described in the single animal data tables
- Other Findings: Furthermore, histopathological diagnoses were made according to the list of primary and secondary diagnoses established by the ‘OECD Guidance Document for the Diagnosis of Endocrine-Related Histopathology of Fish Gonads’ (Johnson R, Wolf J, Braunbeck T, January 7, 2009)

TEST CONCENTRATIONS
- Spacing factor for test concentrations: 3.2
- The spacing factor 3.2 was chosen to include the high levels from previous tests conducted on the substance (10 and 32 μg/L).

RANGE FINDING STUDY
- No DRF was performed. The test conentrations were chosen according to the ECHA decision on substance evaluation: the highest concentration should be the water solubility of DCHP (ca. 1 mg/L).

STATISTICAL ANALYSIS - EXPRESSION OF RESULTS
The hatching, mortality, abnormality and percentage were determined for each concentration. EC10 (Effect Concentration for 10% effect), EC50 and confidence limits were obtained by calculation according to Regtox® Macro (“Vindimian”), using the Hill model.
The fitting is based on the algorithm of Marquardt (1963) which is robust and fast. It needs initial estimates of the parameters and optimize them by successive iterations.
The confidence intervals on the parameters are estimated by a bootstrap simulation which is entirely non parametric and is considered to be well adapted to non-linear models (Efron 1990).
The NOEC (No Observed Effect Concentration) were determined by statistical test from replicate data on hatching, mortality and on length and weight. Statistical calculations were made using R statistical software.
The proportions of sex were determined as concentration proportions ± Standard Deviation of males, females, intersex and undifferentiated.
The VTG was analysed as concentration means ± Standard Deviation of males, females, intersex and undifferentiated.
The NOEC (No Observed Effect Concentrations) were determined by statistical test from replicate data on proportions of sex and VTG concentration. Statistical calculations were made using R statistical software. Statistical analysis was performed only for proportion of female and proportion of male, the number of data concerning intersex and undifferentiated was low (only three intersex and two in the control, only one undifferentiated)
Results of EC50, and EC10 estimations were based on measured concentrations
Reference substance (positive control):
no

Results and discussion

Effect concentrationsopen allclose all
Duration:
64 d
Dose descriptor:
EC10
Effect conc.:
319.1 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Basis for effect:
number hatched
Remarks on result:
other: (95% confidence limits: 251.2 – 399.4 µg.L-1)
Duration:
64 d
Dose descriptor:
EC50
Effect conc.:
821.8 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Basis for effect:
number hatched
Remarks on result:
other: (95% confidence limits: 712.3 – 965.3 µg.L-1)
Duration:
64 d
Dose descriptor:
EC10
Remarks:
survival
Effect conc.:
85.5 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Basis for effect:
mortality
Remarks on result:
other: (95% confidence limits: 62.2 – 161.8µg.L-1)
Duration:
64 d
Dose descriptor:
EC50
Remarks:
survival
Effect conc.:
153.8 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Basis for effect:
mortality
Remarks on result:
other: (95% confidence limits: 128.8 – 201.2 µg.L-1)
Duration:
64 d
Dose descriptor:
NOEC
Effect conc.:
66.6 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Basis for effect:
number hatched
Duration:
64 d
Dose descriptor:
NOEC
Remarks:
survival
Effect conc.:
66.6 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Basis for effect:
other: survival
Duration:
64 d
Dose descriptor:
NOEC
Effect conc.:
10.4 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Basis for effect:
length
Duration:
64 d
Dose descriptor:
NOEC
Effect conc.:
10.4 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Basis for effect:
wet body weight
Duration:
64 d
Dose descriptor:
NOEC
Effect conc.:
66.6 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Basis for effect:
sex ratio
Duration:
64 d
Dose descriptor:
NOEC
Effect conc.:
< 10.4 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Basis for effect:
vitellogenin level
Remarks:
females
Duration:
64 d
Dose descriptor:
NOEC
Effect conc.:
66.6 µg/L
Nominal / measured:
meas. (arithm. mean)
Conc. based on:
test mat.
Basis for effect:
vitellogenin level
Remarks:
males
Details on results:
- Overall mortality/survival: given in the tables and the study report attached
- Days to hatch or time to release of young: The hatching occurred between days 3 and Days 5 (median was approximately day 4)
- Numbers hatched, Numbers of offspring produced: given in the study report attached
- Type and number of developmental / reproductive effect: change of VTG, sex ratio, mortality and growth (weight and length)
- Effect concentrations exceeding solubility of substance in test medium: an acute toxicity of test item on hatching and survival was highlighted from 229.4 μg.L-1 (equivalent to 320 μg.L-1, nominal). At this concentration, the test item is probably under its limit of solubility in test medium. Indeed, a precipitate appeared in the mixing cell at this concentration. A chronic effect on fish weight was highlighted from 28.2 μg.L-1 (equivalent to 32 μg.L-1, nominal).
- Observed effects at each concentration for each observation time: given in the tables and the study report attached
- Fish weights (individual and mean values): given in the tables and the study report attached

Abnormality: Malformations (oedema and spinal deformation) and a state of inactivity or a low activity appeared during the first days post hatching for high concentration. Deceased fish of the 320 μg.L-1 concentration showed a colour appearance darker than that of the control population and a low activity before dying. This coloration is a sign of stress on fish and usually occurs before dying. However, in some cases, this coloring does not lead to death and the fish after a period of stress can regain a normal appearance and behavior. The black coloration was observed in addition to a weak activity but, the fish were not marked individually and therefore it is difficult to conclude whether all the fish having presented a black coloration are dead.

Histopathological: The following histopathological changes were observed during the evaluation: Oocyte atresia, Ovary inflammation, Liver vacuolation, Liver basophilia, Intestinal inflammation, Granulomas, Tubular basophilia, Mononuclear foci.
- Gross Lesions: There were no gross lesions submitted, and no gross lesion was noted during trimming
- Microscopic Findings- Reproductive System: there were no differences in staging in ovaries and testes comparing Negative and DMSO control groups with test item-treated groups, except in Group 6 (320 µg/L). For Groups 1 to 5, the distribution of sexes was at a random manner, i.e., lowest relative male numbers were noted at 10 µg/L, however, at 100 µg/L, the relative male number was comparable to solvent controls. At 320 µg/L, there were only 20 animals available for evaluation, Therefore, the high dose group values need to be excluded from the evaluation of sex distribution. In the Negative Control Group 1, there were two hermaphrodites. In one case, there was one cortical alveolar oocyte found adjacent to a small testicular remnant at Stage 2 (Ova-testis). In the second animal of negative control group, there were two follicles embedded into testicular tissue at Stage 3 (Testis-Ova). In one animal from Group 5 (100 µg/L), there was testicular tissue at Stage 3. However, in one location, there was ovarian tissue with perinucleolar oocytes not embedded into testicular tissue. In general, the relative portion of males was low throughout all study groups compared to females (ranging from 21.3 to 30.0% for males, including solvent control and test item-treated groups), except for the control group with 39.0%. The accepted sex ratio according to the OECD 234 guideline for zebrafish (% males or females) is 30-70%. However, the lowest portion of males was found in the low dose group. There was no dose-relationship that could be considered to be an indicator for an abnormal sex-reversal. Therefore, no test item-related effect could be established based on the sex ratio. Other findings were rarely encountered, and consisted of follicular atresia, proteinaceous fluid in ovaries, or focal granuloma or focal inflammation within the ovaries
- Microscopic Findings - Liver: There were no lesions induced by the test item. The findings consisted of the physiological basophilia in females, but also in one single male from the Negative and DMSO control, as well as from group 3 (10 µg/L). The findings in animals are not deemed to be test item-related due to their random distribution. In addition, there was hepatocellular vacuolation. Although there were single animals at 100 µg/L that showed a higher degree of vacuolation, these alterations were considered to be within the range of normal background alterations. There was one animal with a mononuclear cell focus in Group 5 (100 µg/L)
- Microscopic Findings- Kidney: there was only one lesion consisting of a focal tubular basophilia in the kidneys of female no. F253 (Group 3: 10 µg/L, Replicate R2)
- Microscopic Findings - other: in two animals, there was a minimal submucosal inflammation in the intestine. There were no further findings recorded in this study
Under the conditions of this study, Dicyclohexyl phthalate (DCHP), did not induce any test item-related lesion in the reproductive organs, liver, and kidneys. There were no test item-related differences for the ratio of males to females. Therefore, based on the histological evaluation, there were no morphological findings induced by the test item Dicyclohexyl phthalate (DCHP)

VTG measurements:The standard deviation for the VTG measurement in females is important because at the end of the test, the females had different gonadic stage (between 0 and 4) and the VTG production depends on this evolution. In normal conditions, zebrafish gonad differentiation occurred around 25 dpf (days post fertilization) (Wei Ge, 2013). They have highlighted that the onset of puberty is directly regulated by the reproductive axis, the functionality of this axis seems to be influenced by the growth axis. Their data showed that the puberty onset of female zebrafish was closely related to body size, but not age. No matter how old the fish was (60 dpf vs. 90 dpf), puberty did not initiate if the total body length did not reach 1.8 cm or the body weight was below 100 mg. For each concentration, some fish did not attain these threshold level and therefore this could be the reason of the increase the standard deviation.

Concentrations: As most of the measured concentrations (red value) were not within ±20% of nominal or initial concentrations, all results and subsequent calculations were based on arithmetic mean concentrations. The variation of measured concentrations was probably due to a low solubility of DCHP in test media despite using DMSO. Indeed, the stability of concentration in test solution was higher in the low concentration (10 and 32 μg.L-1) and a precipitate was generated in the mixing cell for the concentration at 320 μg.L-1 and in aquaria for the concentration at 1000 μg.L-1 (a non-GLP analysis was performed on this precipitate to confirm the DCHP identity). It seems that as the test item concentration increases (C03 to C07), the difference between the analytical measurements and the nominal value increases. For DEHP phthalate, a tendency to form colloidal dispersions was highlighted above 3 μg.L-1 (Oehlmann and al., 2009). Perhaps DCHP has a similar behaviour and this could explain the decrease of measured concentration and also the increase in measured concentration if colloids or particles were sampled.

When a statistical analysis of VTG measurements in females is carried out by not taking into account the C06 concentration, a decrease in VTG production was highlighted for 10.4 μg.L-1 (equivalent to 10 μg.L-1, nominal) and 28.2 μg.L-1 (equivalent to 32 μg.L-1, nominal) concentrations. A decrease in the production of vitellogenin in females was highlighted for the concentrations 10.4 μg.L-1 (10 μg.L-1, nominal) and 28.2 μg.L-1 (32 μg.L-1, nominal). A statistical effect on growth (weight) was also highlighted for the concentration 28.2 μg.L-1 (32 μg.L-1, nominal). No statistical effect was highlighted for fish whole length but the data showed a decrease of length for the concentrations 10.4 μg.L-1 (10 μg.L-1, nominal) and 28.2 μg.L-1 (32 μg.L-1, nominal). Moreover, even there is no statistical effect at these concentrations, it was observed that the proportion of female obtained for these two concentrations (78.9 % for 10.4 μg/l and 76.4 % for 28.2 μg/l) were higher than the control. As the sex-ratio is skewed in all conditions, it was really difficult to show potential substance-induced changes, especially a feminization. In this study, the parameter that indicated a disturbance of the endocrine is a decrease in VTG in the female. However, these effects (decrease of weight, decrease of length, decrease of female VTG production) are not monotonous and have not been observed for the concentration 66.6 μg.L-1 (100 μg.L-1, nominal) but intra-variability is higher in this group for measured parameters and inter-variability is high especially for weight and length (Standard Deviation is higher than for the others concentration). This phenomenon could be explained by problems of solubility of DCHP in concentrations higher than 32 μg.L-1 with possible formation of colloids as indicated in the literature for certain phthalates (Oehlmann and al., 2009), making DCHP less bioavailable at high concentrations. In fact, the analytical monitoring of the test solutions showed that the lowest test concentrations were more stable. It seems that as the test item concentration increases (C03 to C07), the difference between measured values and nominal values (measured values lower than nominal values) also increases probably by precipitation of DCHP.
Reported statistics and error estimates:
under "any other information on results incl. tables"

Any other information on results incl. tables

Reported statistics

-Hatching: A statistical analysis has been performed to compare both controls. As normality of the data and variance homogeneity are fulfilled a t-test can be achieved and highlighted no siginificant difference between both control conditions (p>0.05). In this context, both control conditions can be gathered for further analysis. A statistical analysis using non-parametric test (Normality not met; data not shown) highlighted a significant difference compared to the control condition from 229.4 μg.L-1 (320 μg.L-1, nominal)(C06; p<0.05).

- Survival: A statistical analysis has been performed to compare both controls. As normality of the data and variance homogeneity are fulfilled a t-test can be achieved and highlighted no siginificant difference between both control conditions (p>0.05). In this context, both control conditions can be gathered for further analysis.A statistical analysis using non-parametric test (Homogeneity of the variance not met; data not shown) highlighted a significant difference in comparison with the control condition from 229.4 μg.L-1 (320 μg.L-1, nominal) (C06; p<0.05).

- Fish length: A statistical analysis has been performed to compare both controls. As normality of the data and variance homogeneity are fulfilled a t-test can be achieved and highlighted no siginificant difference between both control conditions (p>0.05). Both control conditions can be gathered for further analysis. As normality of the data and variance homogeneity are fulfilled a parametric test (ANOVA) can be achieved and highlighted siginificant difference between conditions (p<0.05). A pairwise comparisons using Dunnett’s test for multiple comparisons with control highlighted a significant difference compared to the control condition from 28.2 μg.L-1 (32 μg.L-1, nominal) (C04; p<0.05). In C05 condition (66.6 μg.L-1, 100μg.L-1, nominal), a high intra-variability of the data did not permit to show any significant difference compared to the control.

- Fish body weight: A statistical analysis has been performed to compare both controls. As normality of the data and variance homogeneity are fulfilled a t-test can be achieved and highlighted no siginificant difference between both control conditions (p>0.05). Both control conditions can be gathered for further analysis.As normality of the data and variance homogeneity are fulfilled a parametric test (ANOVA) can be achieved and highlighted siginificant difference between conditions (p<0.05). A pairwise comparisons using Dunnett’s test for multiple comparisons with control highlighted a significant difference compared to the control condition from 28.2 μg.L-1 (32 μg.L-1, nominal) (C04; p<0.05). In C05 condition (66.6 μg.L-1, 100μg.L-1, nominal), a high intra-variability of the data did not permit to show any significant difference compared to the control.

- Sex ratio: A statistical analysis has been performed to compare both controls. As normality of the data and variance homogeneity are fulfilled, a t-test can be achieved and highlighted a siginificant difference between both control conditions (p>0.05). Solvent control condition was used for further analysis. A statistical analysis using non-parametric test (Normality not met; data shown) highlighted no significant difference compared to the control condition from each treatment (p<0.05). Normality and variance homogeneity for C06 were not performed because C06 contains only one replicate. With a single C06 replicate, nonparametric statistical tests do not allow the different treatments to be discriminated. Thus, analytical test was performed below without C06 results. A statistical analysis using non-parametric test (Normality not met; data shown) highlighted no significant difference to the control condition from each treatment (p<0.05).A statistical analysis has been performed to compare both controls. As normality of the data and variance homogeneity are fulfilled a t-test can be achieved and highlighted a siginificant difference between both control conditions (p>0.05). Solvent control conditions was used for further analysis.

- VTG measurement - females: A statistical analysis has been performed to compare both controls. As normality of the data and variance homogeneity are fulfilled a t-test can be achieved and highlighted no siginificant difference between both control conditions (p>0.05). Both control conditions can be gathered.A statistical analysis using non-parametric test (Normality not met; data not shown) highlighted no significant difference to the control condition. Normality and variance homogeneity for C06 were not performed because C06 contains only one replicate. With a single C06 replicate, nonparametric statistical tests do not allow the different treatments to be discriminated. Thus, analytical test was performed below without C06 results

- VTG measurement - males: A statistical analysis has been performed to compare both controls. As normality of the data and variance homogeneity are not met a Wilcoxon test can be achieved and highlighted no siginificant difference between both control conditions (p>0.05). Both control conditions can be gathered. A statistical analysis using non-parametric test (Normality not met; data not shown) highlighted no significant difference to the control condition. With a single C06 replicate, nonparametric statistical tests do not allow the different treatments to be discriminated. Thus, the analytical test was performed below without C06 results.

Applicant's summary and conclusion

Validity criteria fulfilled:
yes
Conclusions:
The decrease in VTG production in females as well as growth parameters (length and weight) at concentrations of 10 and 32 μg.L-1 could be a sign of a potential endocrine disruptor effect of DCHP.
Executive summary:

The objective of this study was to assess the potential endocrine disrupters of the test item on Danio rerio according to the OECD guideline 234 (28 July 2011). The Fish Sexual Development Test (FSDT) assesses early life-stage effects and potential adverse consequences of putative endocrine disrupting chemicals (e.g. oestrogens, androgens and steroidogenesis inhibitors) on sexual development. The combination of the two core endocrine endpoints, vitellogenin (VTG) concentration and phenotypic sex ratio enables the test to indicate the mode of action of the test chemical.

Male and female genitors were separated in aquariums with a continuous renewal of the water. The day before the experiment, males and females were added in the nest box. Egg laying and fertilization occurred at dawn, simulated with lighting. Eggs were harvested and a first selection was done in order to eliminate unfertilized eggs. Fertilized eggs of Danio rerio produced from the laboratory genitors were used as soon as possible after fertilization. The exposure started as soon as possible after fertilization and before cleavage of the blastodisc commences and no later than 12 h post fertilization to ensure exposure during early embryonic development. 28 days post fertilization the number of fish per replicate were redistributed, so that each replicate contained as equal a number of fish as possible. If exposure related mortality occurs, the number of replicates was reduced appropriately so that fish density between treatment levels was kept as equal as possible. The test duration of the study was until 60 days post-hatch. The fish were distributed into 7 groups each containing 4 replicates. The fish were treated at 10, 32, 100, 320, or 1000 μg/L (groups 3, 4, 5, 6, and 7). The test concentrations were chosen considering that the highest concentration should be the water solubility of DCHP (ca. 1 mg/L). Group 1 served as negative control. Group 2 served as Solvent control. The euthanized, dissected and fixed fish bodies were processed for histolopathological evaluation. VTG measurement, sex ratio, hatching and survival rate, fish body weight and length were the key parameters evaluated. As some of the measured concentrations were not within the range of 80% - 120% of nominal or initial concentrations, all results were based on arithmetic means of measured concentrations 10.4, 28.2, 66.6, 229.4 and 588.2 μg/L.

Under the conditions of this study, no findings were noted in reproductive organs that were related to DCHP. There were no differences in the sex-distribution. However, at 320 μg/L, there were only 20 animals available for evaluation from which two animals only were males. Therefore, the high dose group values need to be excluded from the evaluation of sex distribution. Furthermore, there were no differences in ovarian and testicular stages. There were three cases of hermaphrodites randomly distributed throughout the groups, including the controls. It might be considered that at day 60, juvenile hermaphroditism should not be present any longer. No test item-related findings were noted in kidneys and livers and no morphological findings induced by the test item.

An acute toxicity of test item on hatching and survival was highlighted from 229.4 μg.L-1 (equivalent to 320 μg.L-1, nominal). At this concentration, the test item is probably under its limit of solubility in test medium. Indeed, a precipitate appeared in the mixing cell at this concentration. A chronic effect on fish weight was highlighted from 28.2 μg.L-1 (equivalent to 32 μg.L-1, nominal). For the evaluation of length and weight, the sex ratio and the production of VTG, the values of the high dose group, 229.4 μg.L-1 (equivalent to 320 μg.L-1, nominal), must be excluded from the evaluation.

A decrease in the production of vitellogenin in females was highlighted for the concentrations 10.4 μg.L-1 (10 μg.L-1, nominal) and 28.2 μg.L-1 (32 μg.L-1, nominal). A statistical effect on growth (weight) was also highlighted for the concentration 28.2 μg.L-1 (32 μg.L-1, nominal). No statistical effect was highlighted for fish whole length but the data showed a decrease of length for the concentrations 10.4 μg.L-1 (10 μg.L-1, nominal) and 28.2 μg.L-1 (32 μg.L-1, nominal). Moreover, even there is no statistical effect at these concentrations, it was observed that the proportion of female obtained for these two concentrations (78.9 % for 10.4 μg/l and 76.4 % for 28.2 μg/l) were higher than the control. As the sex-ratio is skewed in all conditions, it was really difficult to show potential substance-induced changes, especially a feminization. In this study, the parameter that indicated a disturbance of the endocrine is a decrease in VTG in the female. However, these effects (decrease of weight, decrease of length, decrease of female VTG production) are not monotonous and have not been observed for the concentration 66.6 μg.L-1 (100 μg.L-1, nominal) but intra-variability is higher in this group for measured parameters and inter-variability is high especially for weight and length (Standard Deviation is higher than for the others concentration). This phenomenon could be explained by problems of solubility of DCHP in concentrations higher than 32 μg.L-1 with possible formation of colloids as indicated in the literature for certain phthalates, making DCHP less bioavailable at high concentrations. In fact, the analytical monitoring of the test solutions showed that the lowest test concentrations were more stable. It seems that as the test item concentration increases (C03 to C07), the difference between measured values and nominal values (measured values lower than nominal values) also increases probably by precipitation of DCHP.

The decrease in VTG production in females as well as growth parameters (length and weight) at concentrations of 10 and 32 μg.L-1 could be a sign of a potential endocrine disruptor effect of DCHP.