There are increasing concerns about the occurrence of pharmaceuticals in our aquatic environment. Several reports have already demonstrated that behaviors of aquatic life were altered by exposure to environmental levels of neuroactive pharmaceutical contaminants. Many such neuroactive pharmaceuticals excreted by human cannot be removed effectively by conventional wastewater treatment processes and are discharged to the aquatic environment.

Neurobehavioural test is one of the major approaches for the assessment of integrated functioning (e.g., motor, sensory, cognitive and autonomic functions) of the nervous system, supplemented by in vitro neurocytological assessment for the characterization of molecular and cellular mechanisms. Impairment or alteration in neurobehaviour indicates changes in neurological functioning. However, associating in vitro neurocytological endpoints with observable in vivo neurodevelopmental and neurobehavioural effects is usually complicated by the lack of systemic biological processes and inter-cellular/ inter-tissue networking in in vitro models.

Metabolomics is an emerging research strategy for toxicology research, in which quantitative changes to a set of small molecule metabolites in biological specimens in response to in vivo or in vitro exposure to the toxicant of interest are assessed. Previous studies have already revealed that the profile of neurotransmitters and their metabolites in the CNS of a model fish was sensitive to in vivo neurotoxicological impacts. It is reasonable to infer that the manifestation of neurobehavioural effects induced by neurotoxic substances should be accompanied by the perturbation of neurometabolic profile of the CNS. Zebrafish (Danio rerio) is an increasingly recognized vertebrate model for neurobehavioural studies and is also an ideal model for translational neuroscience research. Fluoxetine is a commonly prescribed antidepressant, which acts as a selective serotonin reuptake inhibitor (SSRI) or norepinephrine-dopamine disinhibitor (NDDI) in the CNS. The occurrence of fluoxetine in the aquatic environment has been reported worldwide especially at the downstream of sewage treatment plants.

In this work, we aim to evaluate the feasibility of using targeted neurometabolomic profiling of the adult zebrafish whole brain to assess neurobehavioural changes (determined by novel tank diving test) induced by subacute exposure to fluoxetine. The first chapter of this thesis introduces essential concepts involved in this study. In the second chapter, the development of a metabolomics study workflow combined with novel tank diving test is reported. Procedures of novel tank diving test was optimized to accommodate the facilities and the metabolomics study workflow. An analytical protocol for the LC-MS/MS-based quantitative neurometabolomic profiling via dansylation labeling was adopted. Zebrafish was exposed to 5 concentrations of fluoxetine (ranging from environmental realistic concentration to common laboratory study concentration) for a period of 14 days. Afterwards, novel tank test was conducted to evaluate behavioural effects of the drug to the fish. Immediate after behavioural test, the whole brain tissue of the fish was dissected and was subjected to the neurometabolic profiling procedures. Swimming behavior of the fish in novel tank diving test was recorded, and the videos were analysed with a tracking software and neurobehavioural endpoints were quantitated. In the third chapter, a multivariate statistical technique, orthogonal-partial least squares (OPLS), was utilized to analyse the perturbation of the targeted neurometabolic profiles of the CNS of fish and to correlate with the behavioural changes induced by the drug. levels of fluoxetine and norfluoxetine in the brain tissue of fish were also measured. The final chapter includes concluding remarks of the study and an evaluation on the utilisation of targeted neurometabolomics platform in neurotoxicological study. Results of our study provide metabolomics information to link the neurological effects of neuroactive chemicals with their in vivo neurobehavioural impacts. With further verification studies, this profiling approach can also become a new tool for ecotoxicological risk assessment of neuroactive contaminants.