Wirkmuster hormonaktiver Substanzen im Zebrabärbling (Danio rerio) : neue methodische Bewertungsansätze unter Berücksichtigung unterschiedlicher Belastungsszenarien

The examination and identification of hormone active substances has great economical and also political relevance. Chemicals, plant protection products and biocides identified as so-called endocrinedisruptors (ED) are produced in great amounts and released to the environment. Following EU authorization processes they are handled under hazard based assessment. Moreover, pharmaceutical hormones for human or veterinary use can interact directly with the hormone system at very low concentrations. By wastewater and surface run-off the ED substances can enter the aquatic environment and thus can be a threat for fish populations. The aim of this work was to examine different biological parameters characterizing an endocrine active impact on the sexual endocrine system of zebrafish following an exposure to ED substances. The first part of this thesis aimed to analyse the sensitivity of indicative parameters, i.e. biomarkers, with focus on receptor mediated and enzyme modulated mechanisms. Effect concentrations of both indicative and population relevant endpoints derived from short term and long term studies, respectively, were compared. The data obtained was used to close data gaps for less studied endocrine modes of action. In a next step, the use of these results within a tiered testing approach to identify hormone active substances in fish was discussed. The results were considered within a proposal for a new testing guideline within the OECD framework to analyse the endocrine impacts in an extended one generation reproduction test using zebrafish(Danio rerio) as test species. The proposed test design considers an existing test protocol with Japanese medaka (Oryzias latipes) and is intended to be placed in the OECD guideline developing program. The aim of the second part of this thesis was to assess the effect of a pulsed exposure to an EDsubstance. An important aspect of endocrine modes of action can be the distance between exposure to an ED substance and the final manifestation of an effect. Two fish life cycle tests were performed. A first study under continuous exposure aimed to get information on the intrinsic toxicity while in a second study a static approach was performed to deliver information on effects after a very short (pulsed) exposure. By comparison of continuous and pulsed application, exposure related effect thresholds and sensitive windows of exposure were identified. Based on the studies described in this thesis, but furthermore using historical data from Fraunhofer IME and literature data, the most relevant and sensitive windows of exposure and effect manifestation were identified and ranked according to their population relevance. The effect concentrations obtained for the lipoprotein vitellogenin and the sex steroide 11-ketotestosterone in blood plasma were in good compliance with effect levels derived for population relevant parameters. Within a tiered testing approach to identify an impact of endocrine active compounds in fish, the significance and value of biomarker responses could be confirmed. Apical endpoints were identified considering their relevance for fish populations. These are life stage specific survival rates, generation time, reproduction capacity and finally sex ratio. Considering the time until an effect becomes evident after a pulsed exposure to an ED compound, the parameters could moreover be ranked. Life stage specific mortalities can occur already shortly after a single exposure pulse. Sub-lethal effects can become evident in a delay of growth. If this delay is lasting it will finally prolong the time until maturation and can delay the total generation time. More sustainable are impacts on reproduction capacity as survival of population essentially depends on it. If the exposure results in an alteration and significant shift of sex ratio in a fish group, the final impact is irreversible and means most and sustainable adverse consequences for the affected fish population. To find an appropriate test strategy for identification of an ED substance affective to fish populations, it has to be considered that depending on the mode of action there can be a gap between the sensitive time period where exposure occurs and the time point an effect becomes evident. For most of the modes of action assessed, an exposure of the sexual development induced effects with relevance for fish population. This refers to Estrogen receptor (ER) –agonists and –antagonists, to Androgen receptor (AR) –agonists and finally to Aromatase inhibitors. For AR –antagonists and inhibitors of the basic steroid synthesis, not enough data was available to identify the sensitive exposure window. Effects with population relevance were found in terms of a reduction of reproduction capability (ER -agonists and –antagonists, AR-antagonists, Inhibitors of basic steroid synthesis), or became evidentby a shift in sex ratio (ER –antagonists, AR –agonists and aromatase inhibitors). A delay in development in terms of reduced growth and a delay of the time to first spawning were observed for ER –agonists. The indicative parameters vitellogenin and 11-keto testosterone showed corresponding sensitivity compared to the affected sensitive population relevant endpoints for most of the mode of actions assessed. However, for AR –agonists the vitellogenin response was less sensitive by several concentration steps, i.e. one order of magnitude, compared to the most sensitive apical endpoint. For the AR -Antagonist, the steroide 11-keto testosterone was more sensitive than vitellogenin. To find an appropriate definite test to identify an ED substance with impact on fish populations, the sensitivity of the different parameters has to be considered. Fish life cycle tests (FLCT) are able to record all population relevant and indicative parameters. However, these studies are cost and time consuming. If the available data pool already allows to derive information on endpoint sensitivity for a test substance, the test strategy can be adapted. If it can be expected that e.g. sex ratio will be affected, a more focussed test can be sufficient, e.g. a fish sexual development test (FSDT).For the future assessment of hazard or risk of ED substances to the environment it will be necessary to derive mechanistic data to find explanation for the effect observed. Beside the well known and described ED modes of action (i.e. enzyme and receptor interactions), the challenge will be to interpret the recorded effects for new substance classes as there are often related to different steps of steroid synthestis. Tailored test methods have to be developed and has to be integrated in the current assessment strategies for EDs