Globally, antidepressant drugs are one of the most commonly prescribed classes of pharmaceuticals used for the treatment of psychiatric conditions, such as anxiety and depression. As a result, antidepressants have been widely detected in the aquatic environment, albeit at relatively low concentrations. Antidepressants act via modulation of brain monoaminergic signalling systems (predominantly serotonergic, adrenergic, dopaminergic), which show a high degree of structural conservation across diverse animal phyla. As a consequence, non-target organisms in the aquatic environment may be at risk from the effects of exposure. In this thesis, using the zebrafish (Danio rerio), l investigated the bioconcentration potential of a range of antidepressant drugs in fish tissues (spanning three major therapeutic classes), and assessed their subsequent effects on features of physiology, behaviour and neuronal activity. A wide range of exposure concentrations were employed, to elucidate the potential risk to fish in the wild, including those of environmental relevance.
In the first instance, acute 5-day exposures of zebrafish embryo-larvae to 9 antidepressant drugs (in isolation) were carried out, before subsequently assessing their effects on locomotor behaviours, including thigmotaxis, which is used as a measure of anxiogenesis, or anxiolysis. As part of this assessment, uptake was measured in whole body tissues and the data were compared with internal concentrations predicted using the Fish Plasma Model (FPM). All compounds tested were found to bioconcentrate in fish tissues to varying degrees. The reliability of the FPM to accurately predict uptake based upon compound lipophilicity was found to vary for the different test compounds. The behavioural assessments revealed that exposure to antidepressants at concentrations several orders of magnitude above those of environmental relevance, induced hypolocomotion, most notably during periods of darkness. There was little evidence of antidepressant-induced anxiolytic behaviours at any of the concentrations tested.
In the next phase of work, a chronic sublethal exposure to the tricyclic antidepressant amitriptyline was undertaken to assess the effects on zebrafish physiology and behaviour at early life stages. The phenotypic effects assessed were those linked to established therapeutic effects in humans i.e. changes in behaviour and specific monoamine pathways. Amitriptyline was found to bioconcentrate in both the whole body and brain tissues of 28-day old zebrafish, but was readily transformed to its major active metabolite nortriptyline. Comparisons of the water and internal tissue concentrations (bioconcentration factors) revealed that in the aquatic environment, amitriptyline is unlikely to reach levels in fish tissues that would be expected to induce therapeutic effects (based upon the effective doses reported in humans). Despite this, at these environmentally relevant concentrations, the relative expression of the serotonin transporter slc6a4a was found to be downregulated, suggesting pharmacological activity. Drug exposure at concentrations above those of environmental relevance were found to accelerate hatch rate and induce hypolocomotive behaviours. Following a period of depuration in clean water, drug-induced behavioural phenotypes were recovered, supporting a drug-specific effect.
Using a transgenic zebrafish with a genetically-encoded pan-neuronal Ca2+ indicator in combination with light sheet microscopy, the effects of aqueous antidepressant exposure on zebrafish larvae brain activity was investigated. All drugs tested were found to alter the neural activity and functional connectivity between distinct anatomical regions in the larval zebrafish brain, further supporting pharmacological activity for acute exposures, albeit at concentrations higher than those detected in the environment. The patterns of activity (i.e. which region exhibited increased or decreased activity versus the control) were distinct for each compound, although some commonalities in the brain regions being (de)activated by drug treatment were evident within therapeutic classes. Furthermore, most of the antidepressants tested were found to modify the neural response of larvae to the introduction of a ‘stressor’ stimulus (abrupt light flashes). By gaining a greater understanding of which neural circuits are influenced by exposure to CNS-active drugs, this may help in directing the development of more targeted behavioural tests.
The findings in this thesis collectively suggest that antidepressant-induced effects on zebrafish physiology, behaviour and neuronal activity were clearly evident only at concentrations above those detected in the aquatic environment supporting the notion that they present a low-level risk to fish populations. Despite this, given factors including their pseudopersistance, the influence of mixtures, food-chain transfer, and the potential for transgenerational inheritance, combined with the continual global rise in prescription rates, the potential risk level may rise in the future.

Antidepressants are one of the most commonly prescribed pharmaceutical classes for the treatment of psychiatric conditions. They act via modulation of brain monoaminergic signaling systems (predominantly serotonergic, adrenergic, dopaminergic) that show a high degree of structural conservation across diverse animal phyla. A reasonable assumption, therefore, is that exposed fish and other aquatic wildlife may be affected by antidepressants released into the natural environment. Indeed, there are substantial data reported for exposure effects in fish, albeit most are reported for exposure concentrations exceeding those occurring in natural environments. From a critical analysis of the available evidence for effects in fish, risk quotients (RQs) were derived from laboratory-based studies for a selection of antidepressants most commonly detected in the aquatic environment. We conclude that the likelihood for effects in fish on standard measured end points used in risk assessment (i.e. excluding effects on behavior) is low for levels of exposure occurring in the natural environment. Nevertheless, some effects on behavior have been reported for environmentally relevant exposures, and antidepressants can bioaccumulate in fish tissues. Limitations in the datasets used to calculate RQs revealed important gaps in which future research should be directed to more accurately assess the risks posed by antidepressants to fish. Developing greater certainty surrounding risk of antidepressants to fish requires more attention directed toward effects on behaviors relating to individual fitness, the employment of environmentally realistic exposure levels, on chronic exposure scenarios, and on mixtures analyses, especially given the wide range of similarly acting compounds released into the environment.