Testing the role of sleep in homeostatic plasticity

We spend 1/3 of our lives asleep, unconscious and disconnected from the world. It seems like a waste of time, yet all animals sleep, so it seems sleep must fulfill some important biological purpose. What might that purpose be?

Sleep improves memory and our minds work less well when we're sleep-deprived, suggesting that sleep has a neurological function. Indeed, one hypothesis posits that sleep is required for the brain to maintain stable levels of activity. This is important because neurons in the brain are all connected: when a neuron fires an electrical impulse, it sends chemical signals to other neurons that either excite them (make them fire) or inhibit them (stop them from firing). If excitation and inhibition become imbalanced, a neural network can spiral out of control into a seizure (too much excitation) or silence (too much inhibition). Yet our brains are constantly changing as we learn based on sensory experience. To stop these changes from unbalancing excitation and inhibition, the brain readjusts neurons and their connections to compensate for the changes and restore stable activity levels, a process called "homeostatic plasticity". It's thought that this process might be best carried out during sleep, a time of inactivity with little sensory input - much as shops do inventory checks after hours.

Although this idea has much supporting evidence, it's not clear exactly how sleep is involved in homeostatic plasticity. First, is sleep specifically required for particular *kinds* of homeostatic plasticity? One influential hypothesis posits that connections between neurons are mainly strengthened when we're awake, and weakened when we're asleep. This idea is supported by much, but not all, evidence. Could sleep's role in homeostatic plasticity be governed by a different logic? For example, perhaps sleep is important for adjusting the strength of connections between neurons but not neurons' own intrinsic ability to be excited by other neurons ('excitability'), or for adjusting the activity of excitatory but not inhibitory neurons.

Second, by what molecular mechanisms does sleep influence homeostatic plasticity? Two interesting candidates are "reactive oxygen species" (byproducts of metabolism that can be dangerous yet also play important signaling roles) and levels of certain synaptic proteins (molecules that help neurons signal to each other). Each one is regulated by sleep and plays a role in homeostatic plasticity. Could one or both be a common nexus by which sleep influences homeostatic plasticity?

We will address these questions using the olfactory system of the fruit fly Drosophila. Like humans, flies sleep, and we have developed a new model system for studying homeostatic plasticity in the intact brain in flies. Here, neurons called "Kenyon cells" excite, and are inhibited by, a neuron called "APL". If we artificially activate APL for 4 days (producing excess inhibition), the circuit compensates for the perturbation, which is revealed as higher activity in Kenyon cells when we lift the artificially imposed excess inhibition. This effect arises both because APL becomes less active and because Kenyon cells get more excitation, and it requires sleep: it's reduced when we stop flies from sleeping, and it's enhanced when we use a genetic trick to force them to sleep extra.

We will test what kinds of homeostatic plasticity sleep is required for, by testing whether sleep is required for (1) a variety of forms of homeostatic plasticity (e.g., excess excitation from Kenyon cells, excess exposure to natural odours) and (2) different possible underlying cellular mechanisms (e.g., changing connection strength between neurons or intrinsic excitability). We will test *how* sleep modulates homeostatic plasticity by measuring and manipulating reactive oxygen species and synaptic protein levels in normal and sleep-deprived flies and testing how this affects homeostatic plasticity.

Sleep is widely believed to be important for neuronal homeostatic plasticity, but it remains unclear what kinds of plasticity sleep is required for and through what molecular mechanisms sleep modulates plasticity. We will address these unknowns in the Drosophila mushroom body, which exhibits sleep-dependent homeostatic plasticity. The mushroom body's principal neurons, Kenyon cells (KCs), receive feedback inhibition from the APL neuron. After we artificially activate APL (=excess inhibition) for 4 d and then remove the perturbation, KCs show higher-than-normal odour responses, evidence of homeostatic compensation for excess inhibition. This effect occurs via both reduced activity in APL and increased excitation of KCs, and it requires sleep: it's reduced in sleep-deprived flies and it's enhanced in flies genetically induced to sleep too much.

To reveal what kinds of homeostatic plasticity sleep is required for (e.g., increasing vs. decreasing activity? only certain types of circuits or manipulations?), we will extend our preliminary results by testing whether sleep is required for other forms of adaptation (e.g., to excess excitation from KCs or prolonged odour exposure), as well as their reversal after the perturbation is gone. We will dissect the circuit mechanisms underlying these forms of adaptation (e.g., modifying pre-synaptic release, post-synaptic receptors, or intrinsic excitability) and test if sleep differentially modulates different mechanisms (e.g., synaptic vs. intrinsic plasticity? plasticity of excitatory vs. inhibitory neurons?). Finally, we will test two candidate molecular mechanisms linking sleep and homeostatic plasticity: reactive oxygen species (ROS) and synaptic protein levels. We will use genetic reporters to measure ROS and synaptic proteins with/without sleep deprivation, with/without activity perturbations, and test how manipulating ROS or synaptic protein levels in particular cell types affects homeostatic plasticity.