Effects of UV filters (benzophenones and octocrylene) and mifepristone on different life stages of zebrafish (Danio rerio)

This thesis addresses effects of personal care products (PPCPs) and pharmaceuticals to zebrafish (Danio rerio). Personal care products including cosmetics and sunscreens are used for human welfare and contain organic UV absorbing chemicals, so-called UV filters to protect the human skin from UV irradiation. Such compounds are also used to prevent degradation of consumable products by absorption of harmful UV irradiation. Some of the PPCPs may be of environmental concern, because of their endocrine activity, persistence and adverse effects to biota. They enter the aquatic environment either directly from industrial and urban discharges and recreational activities, or indirectly via discharges of wastewater resulting from incomplete removal in wastewater treatment plants. Depending on their physico-chemical properties (usually small organic molecules, logKOW > 3) these chemicals bear the potential for bioaccumulation.
This thesis addresses the effects of the UV filters, benzophenone-3 (BP-3), benzophenone-4 (BP-4) and octocrylene (OC), as well as the anti-progestin mifepristone (MIF, RU486) and progesterone (P4) on different developmental stages of zebrafish. We focused on transcriptomics, using microarrays and RT-qPCR, to elucidate the modes of action of these chemicals and tried to link them to physiological changes in fish. Thereby we investigated alterations of mRNA expression of genes involved in the endocrine system (steroid hormone receptors) and steroidogenesis (steroidogenic enzymes) in multiple tissues.
In adult zebrafish males and embryos the UV filter BP-3 indicated an antiestrogenic and antiandrogenic activity at 2 - 312 ug/L based on transcriptional down-regulation of steroid hormone receptors (estrogen receptor (ers1) and androgen receptor (ar)) and several steroidogenic enzymes (P450 aromatase B (cyp19b), 3beta-hydroxysteroid dehydrogenase (hsd3b), 17beta-hydroxysteroid dehydrogenase type 3 (hsd17b3), 11beta-hydroxylase (cyp11b) and 11beta-hydroxysteroid dehydrogenase type 2 (hsd11b2)). Microarray analysis in liver of adult fish did not reveal novel modes of action of BP-3 at 312 ug/L. Vitellogenin, the biomarker for estrogenicity, was not induced at transcriptional (vtg1) and protein level (VTG), and no histological changes were observed in testes. We found that BP-3 is partly metabolized to the estrogenic benzophenone-1 (BP-1), similar to humans, and that the metabolite may contribute to the observed effects of BP-3.
A similar targeted gene approach was applied in multiple tissues of adult zebrafish males and embryos exposed to BP-4. Based on transcriptional down-regulation of steroid hormone receptors (esr1, esr2b, ar), vitellogenin (vtg1, vtg3) and some steroidogenic enzymes (hsd17b3, P450 aromatase A (cyp19a)), we hypothesized an antiestrogenic and antiandrogenic activity of BP-4 at 28 - 2782 ug/L, similarly to previously reported in vitro activities. Potential effects on the thyroid hormone synthesis were hypothesized due to transcriptional alterations of some marker genes (hematopoietically-expressed homeobox protein (hhex) and paired box 8 (pax8)) in the developing thyroid.
In contrast, the frequently used UV filter OC did not show any hormonal activities in adult zebrafish males and embryos at 22 - 383 ug/L, based on gene transcription analysis and plasma sex steroid hormone levels. The global transcriptomics analysis by microarrays indicated other unexpected modes of action of OC, such as interference with blood cell/vessel formation, blood circulation, haematopoeisis and organ development. The expressional changes of transcripts involved in these pathways at 22 - 383 ug/L were similar in adult males and embryos. They varied in time and indicate a similar mode of action at different life stages. Further experiments are necessary to show as to what extent and how these molecular mechanisms of OC result in physiological changes in zebrafish.
Reproductive effects of MIF and P4 were addressed in a reproduction assay, where breeding groups of zebrafish were exposed to 5 - 77 ng/L MIF and 25 ng/L P4, respectively. In addition, embryos (F1 generation) were exposed to the same concentrations for 96 - 120 h. In this study physiological endpoints including egg reproduction, plasma sex hormone levels and gonad histology were compared to molecular (transcriptomics) and in vitro effects of MIF and P4 to get first insights into their potential environmental risks. Reproductive performance (egg production) was increased after a 21 days exposure of adult zebrafish to 5 and 77 ng/L MIF, but was not affected by 39 ng/L MIF and 25 ng/L P4. Reproductive effects were reflected in histopathological observations in ovaries of females, showing post-ovulatory follicles, atretic follicles and proteinaceous fluid. The levels of plasma sex steroid hormone levels (11-ketotestosterone and 17beta-estradiol) were not affected in adult fish. Furthermore, the fertility of laid eggs, hatching and survival of embryos (F1 generation) was not affected by MIF and P4. Reproductive effects could only partly be linked to transcriptional alterations of target genes involved in the endocrine system (esr1, ar, progesterone receptor (pgr), glucocorticoid receptor (gr) and vtg1) and steroidogenesis (hsd17b3, hsd3b, cyp19a and cyp11b). The alterations were similar in adult (F0 generation, after 21 days) and embryonic (F1 generation, after 120 h) zebrafish. The observed expressional alterations of MIF and P4 in vivo were reflected partially by in vitro activities, but the antiprogestogenic activity of MIF was lacking in the yeast-based system.