Registration Dossier

Ecotoxicological information

Endocrine disrupter testing in aquatic vertebrates – in vivo

Administrative data

Endpoint:
fish: other
Type of information:
experimental study
Adequacy of study:
key study
Study period:
biological test Jan 29-Apr 11, 2019
Reliability:
1 (reliable without restriction)
Rationale for reliability incl. deficiencies:
guideline study
Justification for type of information:
ECHA decision on substance evaluation: Request of Fish Sexual Development Test, OECD 234

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)
GLP compliance:
yes (incl. certificate)

Test material

Reference
Name:
Unnamed
Test material form:
solid: particulate/powder
Details on test material:
CAS number: 94-13-3
Specific details on test material used for the study:
CAS number: 94-13-3
Molecular mass: 180.2 g/mol
Purity (according to CoA): 99.7 %
Date of analysis: April 05, 2018
Physical state: Powder
Colour: White
Aqueous solubility: 500 mg/L (20 °C)
Log Pow (25°C): 3.04

Sampling and analysis

Analytical monitoring:
yes
Details on sampling:
see attched report

Test solutions

Details on test solutions:
see attached full report

Test organisms

Aquatic vertebrate type:
fish
Test organisms (species):
Danio rerio (previous name: Brachydanio rerio)

Study design

Test type:
flow-through
Water media type:
freshwater
Limit test:
no
Total exposure duration:
70 d

Test conditions

Hardness:
5.5 to 6.0 mmol/L
Test temperature:
The mean water temperature was in the range of 26.8 °C to 27.5 °C in all test vessels
pH:
Mean 7.52- 7.66 Min 7.2 Max 7.83
Dissolved oxygen:
Mean 89 98 %, Min 78%
Conductivity:
229-254 µS/cm
Nominal and measured concentrations:
The concentrations of propyl 4-hydroxybenzoate were assessed by chemical analysis using LC-MS/MS. The LOQ was set to 0.75 µg propyl 4-hydroxybenzoate/L. At test start, after 7 days and after 14 days, samples were taken from all test vessels in order to confirm correct dosing. During the remaining course of the study, samples were taken on a weekly basis, alternating between the vessels that were served by the same dosage pump. At test end, samples of all vessels were analysed.
If samples were found outside the desired concentration range of 80 – 120 % of nominal concentrations, retain samples were analysed alongside two QCs to confirm the correct concentration in the tank. For calculations of the mean, the mean of the original and the retain sample was calculated. This mean value was used for calculation of the vessel and concentration mean.
Mean concentrations per treatment of the propyl 4-hydroxybenzoate during the course of the study were between 80.5 % (nominal concentration of 640.0 µg propyl 4-hydroxybenzoate/L) and 127.4 % (nominal concentration of 6.4 µg propyl 4-hydroxybenzoate/L) of the nominal concentration of the test item.
Details on test conditions:
see attached full report
Reference substance (positive control):
no

Results and discussion

Effect concentrationsopen allclose all
Duration:
70 d
Dose descriptor:
NOEC
Effect conc.:
>= 518 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
mortality
Remarks on result:
other: Post hatch survival at 70dpf
Duration:
70 d
Dose descriptor:
LOEC
Effect conc.:
> 518 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
mortality
Remarks on result:
other: Post hatch survival at 70 dpf
Duration:
70 d
Dose descriptor:
NOEC
Effect conc.:
165 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
sex ratio
Duration:
70 d
Dose descriptor:
LOEC
Effect conc.:
518 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
sex ratio
Duration:
70 d
Dose descriptor:
NOEC
Effect conc.:
165 µg/L
Nominal / measured:
meas. (initial)
Conc. based on:
test mat.
Basis for effect:
other:
Remarks:
number of female animals
Duration:
70 d
Dose descriptor:
LOEC
Effect conc.:
518 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
other:
Remarks:
number of female animals
Duration:
70 d
Dose descriptor:
NOEC
Effect conc.:
>= 518 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
other:
Remarks:
number of male animals
Duration:
70 d
Dose descriptor:
LOEC
Effect conc.:
> 518 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
other:
Remarks:
number of male animals
Duration:
70 d
Dose descriptor:
NOEC
Effect conc.:
>= 518 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
vitellogenin level
Remarks:
females
Duration:
70 d
Dose descriptor:
LOEC
Effect conc.:
> 518 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
vitellogenin level
Remarks:
females
Duration:
70 d
Dose descriptor:
NOEC
Effect conc.:
>= 518 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
vitellogenin level
Remarks:
males
Duration:
70 d
Dose descriptor:
LOEC
Effect conc.:
> 518 µg/L
Nominal / measured:
meas. (geom. mean)
Conc. based on:
test mat.
Basis for effect:
vitellogenin level
Remarks:
males
Details on results:
see any other information on results, executive summary and attached full report

Any other information on results incl. tables

        Summary of NOEC/LOEC determination during the course of the study

Parameter

NOEC / LOEC

Nominal concentration

propyl 4-hydroxybenzoate [µg/L]

NOEC / LOEC

Mean measured concentration

propyl 4-hydroxybenzoate [µg/L]

Post-hatch survival at 35 dpf

≥640 / > 640***

≥ 518 / > 518***

Total length at 35 dpf

202 / 640**

165 / 518**a

Post-hatch survival at 70 dpf

≥640 / > 640***

≥ 518 / > 518***

Total length at 70 dpf (all fish)

202 / 640*

165 / 518*a

Wet weight at 70 dpf (all fish)

64.0 / 202**

59.9 / 165**a

Length at 70 dpf (males)

≥ 640 / > 640a

≥ 518 / > 518b

Length at 70 dpf (females)

64.0 / 202**

59.9 / 165**a

Weight at 70 dpf (males)

202 / 640**

165 / 518**a

Weight at 70 dpf (females)

64.0 / 202**

59.9 / 165**a

Female to male ratio

202 / 640*

165 / 518*

Sex [% females]

202 / 640*

165 / 518*

Sex [% males]

≥640 / > 640***

≥ 518 / > 518***

Sex [% undifferentiated]

202 / 640*

165 / 518*

VTG content [ng/µg protein]; females

≥ 640 / > 640***

≥ 518 / > 518***

VTG content [ng/µg protein];

males

≥640 / > 640a

≥ 518 / > 518b

VTG content [ng/µg protein];

undifferentiated

≥640 / > 640a

≥ 518 / > 518b

* Statistical evaluations were performed with the Williams T-Test (α=0.05)

** Statistical evaluations were performed with the Jonckheere-Terpstra Test (α=0.05)

*** Statistical evaluations were performed with the Dunnetts T-Test (α=0.05)

aIncreased size; not considered as adverse effect

bNo valid statistical analysis possible due to a small number of values.

Applicant's summary and conclusion

Validity criteria fulfilled:
yes
Conclusions:
Assessment of endocrine disruption properties of propyl 4-hydroxybenzoate
In the Community strategy for endocrine disruptors' an ED is defined as follows: An endocrine disruptor is an exogenous substance or mixture that alters function(s) of the endocrine system and consequently causes adverse health effects in an intact organism, or its progeny, or (sub)populations". From this it can be derived that for the verification of an endocrine disrupting effect, an adverse effect must arise and must be related to a specific endocrine mode of action.
For the test substance an estrogenic mode of action was assumed since a certain ability to bind to estrogen receptors was reported. Using in vivo testing, it should be proofed whether a specific mode of action and in consequence, an adverse health effect will occur.
Generally, in ecotoxicology the population and not the individuum of a species is the protection objective when the potential ecotoxicological risk of as substance is assessed. This is also true for the assessment of endocrine disruption. Therefor to identify an adverse effect, it must have consequences for the species on population level.
The fish sexual development test was carried out because “The combination of the two core endocrine endpoints, vitellogenin (VTG) concentration and phenotypic sex ratio enable the test to indicate the mode of action of the test chemical (OECD 234)”.
The following parameters, which may be related to an endocrine disrupting mode of action were investigated in the study.
Effect on sex ratio
A significant effect on the alteration of the number of female fish was only found in the highest concentration tested. The differences for male fish did not follow a concentration-dependent trend and were not statistically significant.
Since the number of male fish does not decrease significantly, it cannot be concluded that a drift from female to male animals has manifested.
But due to the increase of female fish an effect on the sex ratio was calculated for the highest test concentration.
Strength of effect
One criterion for the validity of the test is the sex ratio of fish in the controls. It must be in the range of 70/30, or 30/70 (males /females). The validity on this parameter is given for the test as the sex ratio based on mature individuals in controls was at 61.0 % females to 39.0 % males. This ratio is well within historical data observed at Fraunhofer IME, with a mean value of 64.4 % females to 35.6 % males. Historical data show that the strain used, tends to produce more female than male progeny.
Only in the highest concentration the share of females increased significantly to a level of 81.3 % This is an increase of 20 % compared to the controls in the experiment and 17% compared to the historical control data.
The acceptance of a very wide range of sex ratio in the validity criteria is caused by the naturally high plasticity in the sex ratio of the offspring. In consequence changes in sex ratio cannot be assumed a critical parameter for species survival and therefor a shift from one side to the other is of no biological relevance for the maintenance of the population.
The number of males was low in the controls and all treatments but showed no statistically significant decrease. From population ecology it is known that the number of females is more relevant than the number of males as long as a genetically diverse fertilization is possible. So, it can be concluded that the low amount of male fish has no relevance for species survival.
Since the increase of female fish in the experiment is not biologically relevant for species survival it cannot be seen as an adverse effect.

Vitellogenin
No statistically significant differences between controls and treatments were determined for the VTG content in female fish. For males, very low VTG values were determined. There was no statistically significant difference determined between controls and treatment samples. An estrogen modulated receptor interaction would result in a prominent VTG increase in male blood plasma.
So, no feminization of male fish was caused by the test substance.

Histopathology
Histopathological examination of the exposed fish showed no relevant lesions compared to control fish. In male fish no hints for feminization (e.g. occurrence of testis -ova) were found.

Conclusion
To verify an endocrine disrupting effect, an adverse effect must arise and must be related to a specific endocrine mode of action.
Based on the data presented, there is no evidence that propyl 4-hydroxybenzoate has endocrine disrupting properties.
Since there is no increase of vitellogenin in male fish, which could be read as a step in a feminization process, neither the amount of male fish decreases significantly, nor the histopathology of male fish showed any feminization, it is concluded that there is no hint for an estrogenic effect of propyl 4-hydroxybenzoate. An estrogen modulated receptor interaction would result in a prominent VTG increase in male blood plasma, but this was not observed at any treatment level.
As no effect on biomarker response (vitellogenin, histopathology) was detected, there is no evidence for a primary endocrine impact of the test item to fish.
The increase of female fish in the highest test concentration is not assumed to be an adverse. And, due to the missing evidence for an estrogenic mode of action, no connection to adverse effects can be proved which is an absolute precondition for the identification of endocrine disrupting chemicals.
In summary it is not justified to state that propyl 4-hydroxybenzoate has endocrine disrupting properties




Executive summary:

Summary

A fish sexual development test (FSDT) with zebrafish (Danio rerio), sponsored by Clariant, was performed at the Fraunhofer Institute for Molecular Biology and Applied Ecology (IME). The objective of this study was the assessment of effects of continuous exposure to the test item on the early life stages and sexual differentiation of zebrafish (Danio rerio), following the OECD test guideline 234. Deviating from the OECD TG 234, the study was run until 70 dpf, as it turned out that the size of the fish at 63 dpf would have not guaranteed a proper sex determination. The study was conducted with nominal concentrations of 6.40, 20.2, 64.0, 202, 640 µg propyl 4-hydroxybenzoate/L in four replicates each under flow through conditions. An untreated control was run in parallel. Exposure was started with 30 fertilized eggs per test vessel and replicate. Endpoints that were determined included hatching success and rates, and mortalities during the early life stage and the juvenile growth. At day 35 pf and when groups were terminated (day 70 pf), fish were digitally photographed. Fish lengths were determined by evaluating photographs using electronically supported analysis. Single wet weights (blotted dry) were determined on day 70 pf (test end).

Sex ratios were determined macroscopically by inspection of the gonads and by histopathological verification. Blood samples of all fish were taken and measured for the vitellogenin (VTG) concentration. Furthermore, a histopathological examination of the fish gonads was performed.

 

Chemical analysis

Mean concentrations per treatment of the propyl 4-hydroxybenzoate during the course of the study were between 80.5 % (nominal concentration of 640 µg propyl 4-hydroxybenzoate/L) and 127.4 % (nominal concentration of 20.2 µg propyl 4-hydroxybenzoate/L) of the nominal concentration of the test item. As the mean measured concentrations thus differed from the desired 80 – 120 % of target concentrations, it was decided to base the biological effects on mean measured concentrations (7.66 µg/L; 25.4 µg/L; 59.9 µg/L; 165 µg/L; 518 µg/ L).

 

Biological effects

Early life stage

Hatching rate

First larvae started to hatch at 3 dpf across all treatment levels. Hatch was completed at 5 dpf in all replicates, with no difference between treatments. Complete hatch was observed in all replicates.

 

Post-hatch survival at 35 dpf

At 21 dpf, the post-hatch survival rate in controls reached a mean value of 84.2 %. At 35 dpf, post-hatch survival in control vessels was also at 83.3 % and thus met the validity criterion for survival in controls of ≥ 70 %. Survival rates under treatment conditions at 35 dpf varied between 82.5 % (mean measured concentration of 7.66 µg propyl 4-hydroxybenzoate/L) and 89.2 % (mean measured concentration of 518 µg propyl 4-hydroxybenzoate/L). Statistical analyses of post-hatch survival at 35 dpf revealed no significant differences between control and treatments (Williams Test,α=0.05; one-sided smaller).

 

Total length at 35 dpf

In this study, mean total length of 2.17 cm was observed in controls. The mean length of larvae in the treatment conditions were found in the range of 2.20 cm (59.9 µg/L) and 2.35 cm (518 µg/L). The larvae were thus larger in terms of length under treatment conditions. Statistical analyses revealed a significant increase in size compared to control values for the highest treatment level (Jonckheere-Terpstra Test, p<0.05; one-sided greater).

Thus, the NOEC based on the endpoint size in terms of length after the ELS was defined as 165 µg propyl 4-hydroxybenzoate/L (mean measured concentration).

 

Test termination

Post-hatch survival at 70 dpf

Only few fish were found dead during the juvenile growth phase. No other signs of disease were observed. Statistical analyses of post-hatch survival at 70 dpf revealed no significant differences between control and treatments (Williams Test,α=0.05; one-sided smaller). The NOEC for the endpoint post-hatch survival at test end was defined at ≥ 518 µg propyl 4-hydroxybenzoate/L (mean measured concentration).

 

Total length and wet weight at 70 dpf

At 70 dpf, growth in terms of length was determined. Juvenile fish in the control displayed a mean measured length of 3.3 cm. The mean length of fish under treatment conditions varied between 3.3 cm and 3.6 cm. The mean measured weight of fish in controls was measured at 0.381 g, the mean weight of fish under treatment conditions varied between 0.379 g and 0.503 g. Thus, the size of the juvenile fish was increasing with increasing test item concentrations.

Statistical analyses revealed a significant increase in size in terms of total length at the highest test concentration (Williams Test, p<0.05; one-sided greater) and wet weight at the two highest test concentrations (Jonckheere-Terpstra Test, p<0.05; one-sided smaller) at test end compared to control values.

Length in females increased concentration-dependent from a mean value of 3.4 cm at the lowest test concentration to a mean value of 3.6 cm at the highest test concentration. The difference at the two highest test concentrations was statistically significantly different compared to controls (Jonckheere-Terpstra Test, p<0.05, one-sided greater).

The mean values in controls for the parameter weight were determined to be 0.334 g for males and 0.446 g for females. Weight in males increased concentration-dependent with a statistically significant difference at the highest treatment level (Jonckheere-Terpstra Test, p<0.05, one-sided greater). Also for weight, the increase is due to the two single males in replicate 3 and 4 of the highest treatment level, with a weight of 0.853 g and 0.600 g, respectively.

Weight in females increased concentration-dependent from a mean value of 0.424 g at the lowest test concentration to a mean value of 0.510 g at the highest test concentration. The difference at the two highest test concentrations was statistically significantly different compared to controls (Jonckheere-Terpstra Test, p<0.05, one-sided greater).

Thus, the overall NOEC for the parameter growth was at 59.9 µg propyl 4-hydroxybenzoate/L (mean measured concentration), and determined for length and weight of females.

However, an increased size is not considered as negative and is likely the result of a more progressed maturation of the fish with increasing concentrations (see following paragraph).

 

Histological determination of fish sex at test termination

The obtained histopathological data confirmed a trend to an increased number of females, while the number of undifferentiated fish decreased in a concentration-dependent manner. In controls, the mean value for females was at 45.9 % of all fish,while it increased to a level of 81.3 % at the highest test concentration. Correspondingly, the number of undifferentiated fish decreased from a mean value of 26.3 % in controls to a level of 3.6 % at the highest test concentration. Also the number of males decreased, from a mean value of 27.8 % in controls to a mean value of 15.1% at the highest test concentration. The decrease in the relative value of males however did not follow a linear trend.

Statistical analyses revealed significant differences in the number of females and the number of undifferentiated fish at the highest concentration level (females: Williams T-Test; p<0.05; one-sided greater; undifferentiated fish: Williams T-Test; p<0.05; one-sided smaller). No statistically significant difference was observed for the number of males (Dunnett’s Multiple T-Test; α=0.05; one-sided smaller).

 

Histopathology

Examination of the sex ratio of exposed fish showed that the chemical treatment induced a maturation of females. Animals at higher concentration were further developed compared to controls or lower treatment concentrations. Proportion of males was always low and further decreased with increasing treatment concentration. Furthermore, in higher concentrations, pathology in the ovaries increased. The high amount of lesions in the ovaries at the highest concentration is most likely due to the incapability to lay eggs. Ovaries in this concentration were mostly in stage 3 or 4. As substrate to facilitate the egg deposition was not provided, egg-bound related changes are probable.

Vitellogenin content in blood plasma

No statistically significant differences between controls and treatments were determined for the VTG content female fish. For controls, mean values of 490.8 ng VTG/µg protein were measured, while treatments displayed VTG concentrations between 313.7 ng VTG/µg protein (59.9 µg propyl 4-hydroxybenzoate /L) and 654.1 ng VTG/µg protein (165 µg propyl 4-hydroxybenzoate /L), with no concentration-response relationship. For males and undifferentiated fish, low VTG values were determined. There was no statistically significant difference determined between controls and treatment samples.

 

Conclusion

Statistical evaluation revealed no effect of the test item propyl 4-hydroxybenzoate hatch and post-hatch survival. However, larvae under treatment conditions displayed a concentration-dependent increase in length at 35 dpf. At 70 dpf, size in terms of length was statistically significantly different compared to controls at the highest test concentration, size in terms of wet weight was statistically significantly different to controls at the two highest test concentrations. The differences in size were especially evident in females, with a statistically significant difference in the length and wet weight at a concentration of 165 µg propyl 4-hydrobenzoate/L. One explanation for an increased size is a bacterial growth, which was seen in the highest concentration and which could serve as additional feed for fish exposed to this test level.

However, an increased size is not considered as negative and is likely the result of a more progressed maturation of the fish with increasing concentrations.

 

The evaluation of statistically significant differences in the sex ratio between controls and exposed fish revealed a concentration-dependent increase of females, which was statistically significant at the highest test concentration. At the same time, the number of undifferentiated fish decreased concentration-dependent, with a statistically significant difference at the highest test concentration (518 µg propyl 4- hydroxybenzoate/L (mean measured concentration)). The differences in the ratio of males did not follow a concentration-dependent trend and was not statistically significant. Data from histological evaluation of the maturation stage indicated that with increasing concentrations, females displayed more mature gonads, with the majority of ovaries between maturity stage 1 and maturity stage 2 in controls and the majority of ovaries between 3 and 4 at the highest test concentration. The number of females in stage 0 (undifferentiated fish, several of them with degenerating oocytes suggesting the entry in the transition phase) decreased with increasing concentrations.

 

No statistically significant differences between controls and treatments were determined for the VTG content in female fish. For males and undifferentiated fish, very low VTG values were determined. There was no statistically significant difference determined between controls and treatment samples.

 

Based on the endpoint sex ratio (shift of the sex ratio towards females, decreased number of undifferentiated fish), the following NOEC was determined:

 

NOEC = 165 μg propyl 4-hydroxybenzoate/L (mean measured concentration),

corresponding to

NOEC = 202 μg propyl 4-hydroxybenzoate/L (nominal concentration).

 

 

Assessment of endocrine disruption properties ofpropyl 4-hydroxybenzoate

In theCommunity strategy for endocrine disruptors' an ED is defined as follows: An endocrine disruptor is an exogenous substance or mixture that alters function(s) of the endocrine system and consequently causes adverse health effects in an intact organism, or its progeny, or (sub)populations". From this it can be derived that for the verification of an endocrine disrupting effect, an adverse effect must arise and must be related to a specific endocrine mode of action.

For the test substance an estrogenic mode of action was assumed since a certain ability to bind to estrogen receptors was reported. Using in vivo testing, it should be proofed whether a specific mode of action and in consequence, an adverse health effect will occur.

Generally, in ecotoxicology the population and not the individuum of a species is the protection objective when the potential ecotoxicological risk of as substance is assessed. This is also true for the assessment of endocrine disruption. Therefor to identify an adverse effect, it must have consequences for the species on population level.

The fish sexual development test was carried out because“The combination of the two core endocrine endpoints, vitellogenin (VTG) concentration and phenotypic sex ratio enable the test to indicate the mode of action of the test chemical (OECD 234)”.

The following parameters, which may be related to an endocrine disrupting mode of action were investigated in the study.

Effect on sex ratio

A significant effect on the alteration of the number of female fish was only found in the highest concentration tested. The differences for male fish did not follow a concentration-dependent trend and were not statistically significant.

Since the number of male fish does not decrease significantly, it cannot be concluded that a drift from female to male animals has manifested.

But due to the increase of female fish an effect on the sex ratio was calculated for the highest test concentration.

Strength of effect

One criterion for the validity of the test is the sex ratio of fish in the controls. It must be in the range of 70/30, or 30/70 (males /females). The validity on this parameter is given for the test as the sex ratio based on mature individuals in controls was at 61.0 % females to 39.0 % males. This ratio is well within historical data observed at Fraunhofer IME, with a mean value of 64.4 % females to 35.6 % males. Historical data show that the strain used, tends to produce more female than male progeny.

Only in the highest concentration the share of females increased significantly to a level of 81.3 % This is an increase of 20 % compared to the controls in the experiment and 17% compared to the historical control data.

The acceptance of a very wide range of sex ratio in the validity criteria is caused by the naturally high plasticity in the sex ratio of the offspring. In consequence changes in sexratio cannot be assumed a critical parameter for species survival and therefor a shift from one side to the other is of no biological relevance for the maintenance of the population.

The number of males was low in the controls and all treatments but showed no statistically significant decrease.From population ecology it is known that the number of females is more relevant than the number of males as long as a genetically diverse fertilization is possible. So, it can be concluded that the low amount of male fish has no relevance for species survival.

Since the increase of female fish in the experiment is not biologically relevant for species survival it cannot be seen as an adverse effect.

 

Vitellogenin

No statistically significant differences between controls and treatments were determined for the VTG content in female fish. For males, very low VTG values were determined. There was no statistically significant difference determined between controls and treatment samples. An estrogen modulated receptor interaction would result in a prominent VTG increase in male blood plasma.
So, no feminization of male fish was caused by the test substance.

 

Histopathology

Histopathological examination of the exposed fish showed no relevant lesions compared to control fish. In male fish no hints for feminization (e.g. occurrence of testis -ova) were found.

 

Conclusion

To verify an endocrine disrupting effect, an adverse effect must arise and must be related to a specific endocrine mode of action.

Based on the data presented, there is no evidence thatpropyl 4-hydroxybenzoatehas endocrine disrupting properties.

Since there is no increase of vitellogenin in male fish, which could be read as a step in afeminizationprocess, neither the amount of male fish decreases significantly, nor the histopathology of male fish showed any feminization, it is concluded that there is no hint for an estrogenic effect of propyl 4-hydroxybenzoate. An estrogen modulated receptor interaction would result in a prominent VTG increase in male blood plasma, but this was not observed at any treatment level.

As no effect on biomarker response (vitellogenin, histopathology) was detected, there is no evidence for a primary endocrine impact of the test item to fish.

The increase of female fish in the highest test concentration is not assumed to be an adverse. And, due to the missing evidence for an estrogenic mode of action, no connection to adverse effects can be proved which is anabsolute preconditionfor the identification of endocrine disrupting chemicals.

In summary it is not justified to state thatpropyl 4-hydroxybenzoatehas endocrine disrupting properties.

 

Table1:        Summary of NOEC/LOEC determination during the course of the study

Parameter

NOEC / LOEC

Nominal concentration

propyl 4-hydroxybenzoate[µg/L]

NOEC / LOEC

Mean measured concentration

propyl 4-hydroxybenzoate [µg/L]

Post-hatch survival at 35 dpf

≥640 / > 640***

≥ 518 / > 518***

Total length at 35 dpf

202 / 640**

165 / 518**a

Post-hatch survival at 70 dpf

≥640 / > 640***

≥ 518 / > 518***

Total length at 70 dpf (all fish)

202 / 640*

165 / 518*a

Wet weight at 70 dpf (all fish)

64.0 / 202**

59.9 / 165**a

Length at 70 dpf (males)

≥ 640 / > 640a

≥ 518 / > 518b

Length at 70 dpf (females)

64.0 / 202**

59.9 / 165**a

Weight at 70 dpf (males)

202 / 640**

165 / 518**a

Weight at 70 dpf (females)

64.0 / 202**

59.9 / 165**a

Female to male ratio

202 / 640*

165 / 518*

Sex [% females]

202 / 640*

165 / 518*

Sex [% males]

≥640 / > 640***

≥ 518 / > 518***

Sex [% undifferentiated]

202 / 640*

165 / 518*

VTG content [ng/µg protein]; females

≥ 640 / > 640***

≥ 518 / > 518***

VTG content [ng/µg protein];

males

≥640 / > 640a

≥ 518 / > 518b

VTG content [ng/µg protein];

undifferentiated

≥640 / > 640a

≥ 518 / > 518b

* Statistical evaluations were performed with the WilliamsT-Test (α=0.05)

** Statistical evaluations were performed with the Jonckheere-Terpstra Test (α=0.05)

*** Statistical evaluations were performed with the Dunnetts T-Test (α=0.05)

aIncreased size; not considered as adverse effect

bNo valid statistical analysis possible due to a small number of values.