As many as 9% of epileptic seizures occur as a result of drug toxicity.
Identifying compounds with seizurogenic side effects is imperative for assessing
compound safety during drug development, however, multiple marketed drugs
still have clinical associations with seizures. Moreover, current approaches for
assessing seizurogenicity, namely rodent EEG and behavioural studies, are
highly resource intensive. This being the case, alternative approaches have
been postulated for assessing compound seizurogenicity, including in vitro, in
vivo, and in silico methods.
In this thesis, experimental work is presented supporting the use of larval
zebrafish as a candidate model organism for developing new seizure liability
screening approaches. Larval zebrafish are translucent, meaning they are
highly amenable to imaging approaches while offering a more ethical alternative
to mammalian research. Zebrafish are furthermore highly fecund facilitating
capacity for both high replication and high throughput. The primary goal of this
thesis was to identify biomarkers in larval zebrafish, both behavioural and
physiological, of compounds that increase seizure liability.
The efficacy of this model organism for seizure liability testing was assessed
through exposure of larval zebrafish to a mechanistically diverse array of
compounds, selected for their varying degrees of seizurogenicity. Their central
nervous systems were monitored using a variety of different techniques
including light sheet microscopy, local field potential recordings, and
behavioural monitoring. Data acquired from these measurements were then
analysed using a variety of techniques including frequency domain analysis,
clustering, functional connectivity, regression, and graph theory. Much of this
analysis was exploratory in nature and is reflective of the infancy of the field.
Experimental findings suggest that larval zebrafish are indeed sensitive to a
wide range of pharmacological mechanisms of action and that drug actions are
reflected by behavioural and direct measurements of brain activity. For
example, local field potential recordings revealed electrographic responses akin
to pre-ictal, inter-ictal and ictal events identified in humans. Ca2+ imaging using
light sheet microscopy found global increases in fluorescent intensity and
functional connectivity due to seizurogenic drug administration. In addition,
[2]
further functional connectivity and graph analysis revealed macroscale network
changes correlated with drug seizurogenicity and mechanism of action. Finally,
analysis of swimming behaviour revealed a strong correlation between high speed swimming behaviours and administration of convulsant compounds.
In conclusion, presented herein are data demonstrating the power of functional
brain imaging, LFP recordings, and behavioral monitoring in larval zebrafish for
assessing the action of neuroactive drugs in a highly relevant vertebrate model.
These data help us to understand the relevance of the 4 dpf larval zebrafish for
neuropharmacological studies and reveal that even at this early developmental
stage, these animals are highly responsive to a wide range of neuroactive
compounds across multiple primary mechanisms of action. This represents
compelling evidence of the potential utility of larval zebrafish as a model
organism for seizure liability testing.

We assessed similarities and differences in the electrographic signatures of local field potentials (LFPs) evoked by different pharmacological agents in zebrafish larvae. We then compared and contrasted these characteristics with what is known from electrophysiological studies of seizures and epilepsy in mammals, including humans. Ultimately, our aim was to phenotype neurophysiological features of drug-induced seizures in larval zebrafish for expanding knowledge on the translational potential of this valuable alternative to mammalian models. LFPs were recorded from the midbrain of 4-d-old zebrafish larvae exposed to a pharmacologically diverse panel of seizurogenic compounds, and the outputs of these recordings were assessed using frequency domain analysis. This included analysis of changes occurring within various spectral frequency bands of relevance to mammalian CNS circuit pathophysiology. From these analyses, there were clear differences in the frequency spectra of drug-exposed LFPs, relative to controls, many of which shared notable similarities with the signatures exhibited by mammalian CNS circuits. These similarities included the presence of specific frequency components comparable to those observed in mammalian studies of seizures and epilepsy. Collectively, the data presented provide important information to support the value of larval zebrafish as an alternative model for the study of seizures and epilepsy. These data also provide further insight into the electrophysiological characteristics of seizures generated in nonmammalian species by the action of neuroactive drugs.

BACKGROUND AND PURPOSE: Functional brain imaging using genetically encoded Ca2+ sensors in larval zebrafish is being developed for studying seizures and epilepsy as a more ethical alternative to rodent models. Despite this, few data have been generated on pharmacological mechanisms of action other than GABAA antagonism. Assessing larval responsiveness across multiple mechanisms is vital to test the translational power of this approach, as well as assessing its validity for detecting unwanted drug-induced seizures and testing antiepileptic drug efficacy. EXPERIMENTAL APPROACH: Using light-sheet imaging, we systematically analysed the responsiveness of 4 days post fertilisation (dpf; which are not considered protected under European animal experiment legislation) transgenic larval zebrafish to treatment with 57 compounds spanning more than 12 drug classes with a link to seizure generation in mammals, alongside eight compounds with no such link. KEY RESULTS: We show 4dpf zebrafish are responsive to a wide range of mechanisms implicated in seizure generation, with cerebellar circuitry activated regardless of the initiating pharmacology. Analysis of functional connectivity revealed compounds targeting cholinergic and monoaminergic reuptake, in particular, showed phenotypic consistency broadly mapping onto what is known about neurotransmitter-specific circuitry in the larval zebrafish brain. Many seizure-associated compounds also exhibited altered whole brain functional connectivity compared with controls. CONCLUSIONS AND IMPLICATIONS: This work represents a significant step forward in understanding the translational power of 4dpf larval zebrafish for use in neuropharmacological studies and for studying the events driving transition from small-scale pharmacological activation of local circuits, to the large network-wide abnormal synchronous activity associated with seizures.

Aim: Aggression is a behavioural trait characterized by the intention to harm others for offensive or defensive purposes. Neurotransmitters such as serotonin and dopamine are important mediators of aggression. However, the physiological role of the histaminergic system during this behaviour is currently unclear. Here, we aimed to better understand histaminergic signalling during aggression by characterizing the involvement of the histamine H3 receptor (Hrh3). Methods: We have generated a novel zebrafish Hrh3 null mutant line using CRISPR-Cas9 genome engineering and investigated behavioural changes and alterations to neural activity using whole brain Ca2+ imaging in zebrafish larvae and ribosomal protein S6 (rpS6) immunohistochemistry in adults. Results: We show that genetic inactivation of the histamine H3 receptor (Hrh3) reduces aggression in zebrafish, an effect that can be reproduced by pharmacological inhibition. In addition, hrh3−/− zebrafish show behavioural impairments consistent with heightened anxiety. Larval in vivo whole brain Ca2+ imaging reveals higher neuronal activity in the forebrain of mutants, but lower activity in specific hindbrain areas and changes in measures of functional connectivity between subregions. Adult hrh3−/− zebrafish display brain region-specific neural activity changes in response to aggression of both key regions of the social decision-making network, and the areas containing histaminergic neurons in the zebrafish brain. Conclusion: These results highlight the importance of zebrafish Hrh3 signalling for aggression and anxiety and uncover the brain areas involved. Targeting this receptor might be a potential novel therapeutic route for human conditions characterized by heightened aggression.

Abstract
Drug development is a highly resource intensive process that uses large numbers of animals for assessing the safety and efficacy of drugs prior to clinical testing. Improving the efficiency of drug development in terms of financial expenditure and number of animals used is therefore of utmost concern, not only to industry, but also to animal welfare organisations such as the NC3Rs. Poor efficiency in drug development largely stems from drug attrition, particularly attrition in the latter stages of the testing due to the large amount of resources expended at the point of failure. It is therefore imperative that deleterious off-target effects are identified as early as possible. However, typically, identification of seizure as a side-effect of drugs is performed in the later stages of development due to the highly intensive and low-throughput nature of seizure assays. At which point, if a compound fails, a large amount of resources have been squandered. There therefore exists a need for high-throughput and relatively inexpensive seizure liability assays that can be used early in drug development to prevent compounds destined for failure undergoing unnecessary resource intensive testing.
In this thesis we propose a refined approach using non-invasive imaging techniques in non-protected life stage zebrafish as a method for the detection of seizurogenic compounds early in drug development. In addition, we highlight its utility for elucidating the pharmacodynamics of compounds. In this study, a transgenic zebrafish line containing a GCaMP6s calcium sensor under the control of the pan-neuronal promoter elavl3 was used for functional profiling of compounds with varied pharmacologies. Light sheet microscopy was used to record fluorescent activity in three spatial dimensions over time (4-dimensions) from the zebrafish brain after exposure to forty-three different compounds with varied pharmacodynamics and seizure liability profiles. Hierarchical clustering was employed in order to assess if compounds with seizurogenic activity or similar pharmacodynamics elicited specific functional brain activity. It was found that compounds with dopaminergic and serotonergic mechanisms of action elicited highly specific and similar brain activity patterns and that non-seizurogenic drugs also clustered separately from seizurogenic ones. Subsequent analyses, focussed on the utilisation of machine learning techniques, developing a model that could be used to discriminate between compounds with and without potentially seizurogenic effects. It is clear, from the analyses presented here, that drugs do in fact elicit specific brain patterns in zebrafish and that these brain patterns are effectively detected using light sheet microscopy. This system is highly applicable for use within the drug industry and even in its relatively preliminary stages provided an accurate method of discriminating between compounds based on their physiological effects in zebrafish.

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has not been fixed in the paper.

Functional neuroimaging, using genetically-encoded Ca2+ sensors in larval zebrafish, offers a powerful combination of high spatiotemporal resolution and higher vertebrate relevance for quantitative neuropharmacological profiling. Here we use zebrafish larvae with pan-neuronal expression of GCaMP6s, combined with light sheet microscopy and a novel image processing pipeline, for the 4D profiling of chemoconvulsant action in multiple brain regions. In untreated larvae, regions associated with autonomic functionality, sensory processing and stress-responsiveness, consistently exhibited elevated spontaneous activity. The application of drugs targeting different convulsant mechanisms (4-Aminopyridine, Pentylenetetrazole, Pilocarpine and Strychnine) resulted in distinct spatiotemporal patterns of activity. These activity patterns showed some interesting parallels with what is known of the distribution of their respective molecular targets, but crucially also revealed system-wide neural circuit responses to stimulation or suppression. Drug concentration-response curves of neural activity were identified in a number of anatomically-defined zebrafish brain regions, and in vivo larval electrophysiology, also conducted in 4dpf larvae, provided additional measures of neural activity. Our quantification of network-wide chemoconvulsant drug activity in the whole zebrafish brain illustrates the power of this approach for neuropharmacological profiling in applications ranging from accelerating studies of drug safety and efficacy, to identifying pharmacologically-altered networks in zebrafish models of human neurological disorders.