The fish embryo as an alternative model for the assessment of endocrine active environmental chemicals : elucidation of endocrine disruptive mechanisms and identification of relevant effect endpoints using transcriptomics

摘要

Endocrine disruption can impact health and reproduction in wildlife and is therefore highly concerned in the environmental risk assessment. The current approach in the evaluation of endocrine disruption relies on long-term reproductive and animal intensive studies, which implicit time consuming and cost intensive tests. Thus, there is an increasing demand for alternative testing. This thesis intended to examine the suitability of the fish embryo test as such an alternative approach in testing endocrine disruptive chemicals. Although in fish embryos, reproductive organs are not developed and morphological effects cannot be linked to a specific endocrine mode of action, endocrine genes have been previously reported to be responsive to endocrine disrupters already in the embryonic stage. It was therefore hypothesized that gene expression analysis can be a tool for the detection and evaluation of endocrine disruption in fish embryos. To corroborate this hypothesis, endocrine compounds which are known or suspected to cause estrogenic or anti-androgenic effects were tested for their impact on embryonic gene expression. Further, two complementary fish species, zebrafish and medaka, were used to address differences in species-specific responses and to elaborate distinctive advantages for EDC testing. For zebrafish, microarray technology was applied to detect transcriptome response to estrogens and anti-androgens, whereas for medaka, the expression of genes related to hypothalamic-pituitary-gonadal axis and, for comparative purposes, also of selected genes which appeared indicative for estrogenic and anti-androgenic disrupion in the zebrafish microarray results. The microarray results for 48hpf zebrafish showed a clearly discernible transcriptome profile for 17alpha- ethinylestradiol (EE2) and flutamide. The comparison of the further tested endocrine disruptors to the reference compounds allowed a categorization of the corresponding mechanism of action. The gene response to BPA and propanil was predominantly estrogenic and anti-androgenic, respectively. Genistein, linuron and prochloraz showed multiple mechanisms of actions, which comprised estrogenic as well as anti-androgenic effects. Methylparaben showed a contrary response to EE2 and therefore was hypothesized to act anti-estrogenic. Apart from the common regulation of the biomarker cyp19a1b and vtg1, all tested estrogenic compounds altered the expression of growth- as well as isoprenoid- and cholesterol synthesis related genes, suggesting a possible marker function for these genes. Linuron and prochloraz induced neuroendocrine pathways similar to flutamide, suggesting that the altered expression of neuroendocrine genes exerts a possible marker function for anti-androgenic mechanisms. Since only little information is available on the impact of anti-androgens on neurotransmitter activity in fish, these data provide new insights into the mechanisms of anti-androgens in fish. However, no individual marker genes could be identified for the anti-androgenic compounds, which indicated a weak gene response to anti-androgens in 48hpf embryos. Thus, the expression of known endocrine marker and androgen signaling related genes in response to flutamide was examined in 48hpf embryos and 96hpf embryos, in order to test whether a more defined anti-androgenic response can be retrieved in a later developmental stage. Indeed, the regulation of important endocrine genes was detected after 96h exposure to flutamide in zebrafish, favouring the use of 96hpf zebrafish embryos compared to 48hpf embryos for testing anti-androgenic disruption. Next, the zebrafish results were compared to medaka. The response to the tested estrogens in medaka was in line with the corresponding gene response in zebrafish and only the regulation of the steroid receptors was different. This, on the one hand, introduced medaka as a suitable species for testing estrogenic disruption, equally to zebrafish, and on the other hand, confirmed the marker function of the tested genes. Further, the response to EE2 and flutamide was also discernible in medaka embryos, implying that the assessment of different mode of actions is possible in both fish species. The main difference between both species was the response to flutamide: Though the same genes were affected, their regulation was contrary. Further, the downregulation of anti-androgenic marker was unique to medaka. Thus, the use of both fish species elicited inter-species variations in the response to flutamide, implying the complementarity of the use of both fish species. Conclusively, this thesis proves the suitability of medaka and zebrafish embryos for endocrine disruption testing and provides a valuable basis for the development and implementation of a transcriptomics based FET approach, which would allow rapid, animal-test free and cost-effective testing for endocrine disruption.