Functional analysis of the zebrafish visual system

The retina communicates with the brain via upwards of 20 different subtypes of retinal ganglion cell (RGC). Each RGC subtype is thought to convey information about different features of the visual world to the brain. But clearly we do not see the world as a disjointed set of attributes. These parallel streams of visual information leaving the retina must be integrated in the brain to generate a coherent neural representation of the visual world. A necessary first step towards understanding how the brain does this is to describe, as fully as possible, the nature of the information that leaves the retina and to map where in the brain this information is sent. Our approach is to generate transgenic zebrafish that express genetically-encoded reporters of neural activity (GCaMPs) specifically in RGCs. We then perform high-speed volumetric functional imaging of RGC axons within different visual areas of the brain during presentation of visual stimuli. By clustering and categorizing RGC responses to different visual stimuli we are building functional maps which describe the functional diversity and organization of retinal inputs to the brain.

To address this question we are performing functional imaging of the postsynaptic targets of RGC axons within the brain. We are focussing on the zebrafish optic tectum, an evolutionarily conserved area of the brain that integrates visual with somatosensory and auditory inputs to direct reorienting movements of the eyes, head and body that are associated with behaviours such as catching prey and avoiding predators. By comparing and contrasting how RGCs and their postsynaptic targets respond to the same visual stimuli we hope to gain insight into how the brain integrates its inputs from the retina and then selects and executes an appropriate behavioural response. We are also developing a genetic toolkit for the selective labelling and manipulation of tectal cell types in order to describe the individual components of tectal circuitry and the interactions between them that convert visual input into motor commands.

No area of the brain works in isolation. Instead, behaviour is an emergent property of the dynamic interactions between multiple areas of the brain. Because we are able to rapidly image the whole zebrafish brain we are able to observe these dynamic interactions in real time, and by doing so provide insight into the brain-wide circuits that control behaviour. We are particularly interested in identifying the sensory and motor components associated with approach vs avoidance behaviours and describing how the strength of sensorimotor coupling is modulated according to the type of sensory stimulus and prior experience.

A unique strength of the zebrafish is that larvae are accessible to imaging at all stages of development and that the visual system first becomes functional very early during brain development. We are using the imaging approaches described above to examine the functional properties of nascent visual circuits, the process of circuit maturation and the influence of genes and sensory experience on the functional development of visual circuits. We are also combining the ability image the larval zebrafish brain with genome editing techniques (TALENS and CRISPR/Cas9) to study how normal patterns of circuit assembly and function are perturbed in zebrafish models of human neurodevelopmental disorders such as autism, schizophrenia and epilepsy. Our studies will reveal whether sensory perception, behavioural choice, sensorimotor gating and functional connectivity are altered in ways that might explain the some of the behavioural manifestations of these disorders.