Molecular biomarkers of endocrine disruption in small model fish

Abstract

A wide range of environmental contaminants can interfere with hormonal regulation in vertebrates. These endocrine disrupting chemicals (EDCs) are of high relevance for human and wildlife health, since endocrine signalling controls many essential physiological processes which impact on the individual's health, such as growth and development, stress response, and ultimately reproduction and population development. Small fish represent a cost-effective model for testing potential EDCs allowing the possibility to integrate from molecular to phenotypic and functional effects. We have comprehensively reviewed exposure-effect data from four different small model fish: zebrafish, medaka, fathead minnow, and the three-spined stickleback. The majority of available data refer to EDCs interfering with reproductive hormones. However, we have also included interactions with other hormone systems, particularly the thyroid hormones. We demonstrate that the available data clearly indicates the predictive potential of molecular biomarkers, supporting the development and regulatory application of simple molecular-based screening assays using small model fish for EDC testing.

Introduction

There is an increasing public concern over the adverse effects of anthropogenic contaminants on the health and well-being of both humans and wildlife (WHO, 2002). To date, a wide variety of both developmental and reproductive disorders observed in wildlife species have clearly been linked to the exposure to environmental contaminants which act as endocrine disrupting chemicals (EDCs, Edwards et al., 2006, Tyler et al., 1998, WHO, 2002), that is “exogenous agents that interfere with the production, release, transport, metabolism, binding, action or elimination of natural hormones in the body responsible for the maintenance of homeostasis and the regulation of developmental processes” (Kavlock et al., 1996). EDCs are of high environmental relevance, since many essential physiological processes which impact on the individual's health, such as growth and development, stress response, and ultimately reproduction and population development are controlled by hormones. Regardless of the initial type of interaction, the effects of EDCs with different modes of action are often very similar and resemble changes in endogenous hormone levels.

In vertebrates, different types of hormone systems could principally be targets of environmental chemicals: reproductive hormones such as estrogens and androgens, thyroidal hormones, corticosteroids, growth hormone and their associated hypothalamus/pituitary releasing and stimulating hormones. However, research has to date focused nearly entirely on the interference with the steroidal reproductive hormones for obvious reasons. Firstly, reproductive hormones control one of the most important endpoints in the risk assessment of environmental chemicals, that is all aspects of reproductive development from sex differentiation to puberty, aspects of which are critical for population viability. Secondly, reproductive hormones are of vital importance during the early critical stages of embryonic development and sex differentiation. As a result, even though the binding affinity of most EDCs to the ER/AR is low compared to endogenous steroid hormones, exposure to even low concentrations of xenoestrogens/xenoandrogens can disrupt normal developmental/reproductive processes. This is evident for exposure to micropollutants, that is, anthropogenic chemicals continuously detected at low concentrations in the environment. A well-known example is ethinylestradiol (EE₂), which is discharged into the aquatic environment via municipal effluents due to its use in contraceptive pills. EE₂ is found in the lower ng/L-range in aquatic surface water, a concentration that leads to sustainable impacts on reproduction and population development in fish (Larsson et al., 1999).

Fish are considered as one of the primary risk organisms for EDCs, especially those interacting with the reproductive hormones. Firstly, sex determination in teleost fishes is characterised as being very labile, and can be disturbed or even functionally reversed by the external application of hormones at critical developmental stages (Francis, 1992). Secondly, as the aquatic environment represents the ultimate sink for most anthropogenic contaminants, fish populations are directly exposed to a wide variety of EDCs, originating from industrial, agricultural or municipal effluents. There is now unequivocal evidence showing that EDCs can have long-term effects on reproduction and subsequent population development in natural fish populations (Kidd et al., 2007). However, investigations based on the evaluation of reproductive endpoints are costly, time-consuming and laborious. This has generally resulted in these endpoints not being routinely integrated into most chemical safety assessments. However, since endocrine disruption can be clearly linked to molecular interactions, expression of appropriate biomarkers could be used as predictors for reproductive or other effects.

Small fish represent a cost-effective model for large scale chemical testing. They are easy to maintain in small laboratory aquaria, have a short life-cycle, and readily produce many young. In contrast to targeted in vitro cellular models such as the yeast screen assay for estrogenic compounds or primary cell cultures, small fish models allow the possibility to integrate alternative modes of action, such as compounds that modify endogenous hormone levels. Furthermore, uptake, metabolism and excretion as important constraints of bioavailability are considered. The attractiveness of using small fish models in chemical testing programmes is further enhanced by recent advances in molecular approaches. DNA markers for molecular sex determination are now available for some small fish models (medaka and stickleback, Griffiths et al., 2000, Matsuda et al., 2002). Moreover, the genome has now been sequenced for three of the species highlighted in this review (medaka, zebrafish and stickleback), resulting in the potential development of new molecular tools.

For this review we have selected four different model fish that are ideally suited to establish links between molecular biomarkers and phenotypic and functional effects of EDCs: zebrafish, medaka, fathead minnow, and the three-spined stickleback. Each of these organisms is well established for the testing of environmental chemicals providing different advantages and possibilities such as detailed genetic characterisation, sex specific markers, and/or use as an experimental and biomonitoring model. Due to their short life-cycle, the selected species allow for the analysis of reproductive performance, which is one of the most important endpoints for long-term effects with implications on population development. A brief description of the model small fish species is given in Table 1 (detailed information on the different models can be obtained from the review of Ankley and Johnson, 2004 and Katsiadaki et al., 2007).

In this review, we summarise the available data on endocrine disruption in fish highlighting the links between molecular, phenotypic and physiological effects (in particular reproduction). We focus nearly exclusively on small fish models, since they provide the experimental basis to establish appropriate correlations between the different effect levels including reproductive and transgenerational effects. For the first time, data are presented in a comprehensive manner, providing an overview of the effect levels of different molecular and phenotypic endpoints in the selected model species. The majority of available data refer to EDCs interfering with reproductive hormones. However, we have also reviewed interactions with other hormone systems, indicating further potential applications for small fish models. We show that the available data clearly indicate the predictive potential of molecular biomarkers supporting the development of screening assays for endocrine disruption such as initiated by the OECD (OECD, 2006).