ABSTRACT
The zebrafish has emerged as a useful model system to discover and characterize genetic and neuronal circuits
that regulate vertebrate sleep. However, a limitation of this model is that sleep is determined using behavioral
criteria and not the electroencephalogram (EEG) and electromyogram (EMG) measures that are used to define
mammalian sleep and wake states. Using EEG-like electrophysiological recordings, we have observed that
zebrafish exhibit large amplitude slow waves of brain activity during periods of behavioral quiescence, and small
amplitude fast waves of brain activity during periods of behavioral activity, similar to those observed during
mammalian NREM sleep and wakefulness, respectively. In Aim 1, we will perform a series of experiments to test
the hypothesis that these patterns of brain activity reflect zebrafish sleep and wake states. Validation of this
hypothesis will increase the usefulness of zebrafish as a sleep model since EEG recordings enable more precise
categorization of sleep and wake states, and the classification of EEG rhythms will facilitate comparisons
between zebrafish and mammalian sleep phenotypes. Sleep EEG rhythms may also reveal distinct sleep states
that cannot be captured using behavioral analyses. In Aim 2, we will identify the neurons that generate the slow
waves that we observe during zebrafish behavioral quiescence by performing simultaneous electrophysiological
recordings and whole-brain GCaMP6f neuronal activity imaging. The slow waves that are observed during
mammalian NREM sleep are thought to be largely generated by synchronous firing of cortical and thalamic
neurons. This hypothesis is based on a combination of intracellular recordings of single neurons in the cortex
and thalamus, and EEG recordings that capture the collective activity pattern of cortical neurons near the surface
of the brain. However, it has not been possible to comprehensively determine the contribution of individual
neurons to the slow waves at whole-brain scale. As a result, it is unclear whether the slow waves observed
during mammalian NREM sleep are generated by relatively small or large populations of neurons, and whether
similar patterns of neuronal activity are present in brain regions in addition to the cortex and thalamus. We will
address these questions for the slow waves that we observe during zebrafish behavioral quiescence by
performing simultaneous whole-brain GCaMP6f imaging, with single neuron resolution, and electrophysiological
recordings, in order to directly visualize the neurons in the entire brain whose activity oscillates at 2-4 Hz. Results
from these experiments may generate new hypotheses regarding the neuronal basis of slow waves during NREM
sleep that can be tested using targeted recordings in mammals. This project has the potential to reveal a new
layer of similarity between mammalian and zebrafish sleep, and to predate the known emergence of mammalian-
like patterns of brain activity during sleep by >450 million years. This project will also establish and validate a
method that is essential for a subsequent Targeted Brain Circuits Project R01 application that will aim to
determine the function of the slow waves that are observed during sleep.