Redox Control of Sleep

Sleep disturbances are among the most common medical problems, with an estimated prevalence of 10-15% in the general and 30-60% in the older population. They accompany many medical, psychiatric, and neurological conditions.

Fatigue and insomnia are common symptoms of endocrine and metabolic disorders, whereas inadequate sleep, be it a consequence of lifestyle choices, work or personal pressures, chronic pain, or respiratory dysfunction, is an established risk factor for diabetes and obesity.

Similarly complex cause-and-effect relationships also characterize the sleeping difficulties in psychiatric illnesses. Insomnia and hypersomnia are core symptoms of major depressive disorder, while reduced sleep need is a defining feature of manic episodes. Inadequate sleep often triggers episodes, contributes to relapse, and increases the risk of substance abuse comorbidity.

The sleep fragmentation that occurs during normal ageing is accelerated or augmented in many neurodegenerative diseases, including Alzheimer's and Parkinson's. Injury to sleep-promoting neurons may account, in part, for the characteristically poor sleep quality of Alzheimer's patients. In Parkinson's disease, sleep disruptions are among the most diagnostic biomarkers during the prodromal stage and among the most common non-motor signs in symptomatic disease.

Despite the prevalance of sleep disturbances and the proven benefit of treating them for improving many comorbid conditions, therapeutic options remain limited. They include behavioural interventions to improve sleep hygiene and the use of benzodiazepine and antihistamine sedatives. These medications are associated with a wide range of adverse effects, such as morning sedation, rebound insomnia, anterograde amnesia, confusion and injury, and addiction.

The development of new therapeutic concepts requires a deeper understanding of the neuronal control of sleep. The present programme will test the generality of a sleep-regulatory mechanism we have discovered in Drosophila and examine its suitability as a target for pharmacological intervention. The centrepiece of this mechanism is a potassium channel beta-subunit that senses changes in cellular redox chemistry with the help of a stably bound nicotinamide (NADPH) cofactor. Sleep loss elevates mitochondrial reactive oxygen species in sleep-inducing neurons, which register this rise by converting the beta-subunit to the NADP+-bound form. The oxidation of the cofactor boosts the frequency of action potentials by accelerating membrane repolarization and thereby promotes sleep. Energy metabolism, oxidative stress, and sleep, three processes implicated independently in lifespan, ageing, and degenerative disease, are thus mechanistically connected.

The proposed programme will pursue three goals:

Project 1 will examine whether potassium channel beta-subunits regulate sleep in mammals. We will quantify sleep in mice carrying mutations in the three KCNAB genes, singly or in combination, and localize the sleep-relevant sites of action by re-introducing wild-type beta-subunits into confined brain regions.

Project 2 will test whether small-molecule oxidoreductase substrates, such as breakdown products of peroxidized lipids, can stably alter the redox potential of the bound cofactor. If the NADP+:NADPH ratio of the cofactor encodes the brain's record of sleep debt or waking time, such molecules represent prototypes of sleep-regulatory drugs.

Project 3 will search for chemicals that can stimulate (or prevent) sleep by acting as beta-subunit substrates but possess more favourable pharmacological properties than peroxidized lipids.

Together, this programme promises to lay the foundation for a rational new approach to the treatment of sleep disorders, with potentially significant health and economic benefits.

Our work in Drosophila showed that rising sleep pressure activates two dozen sleep-inducing neurons with projections to the dorsal fan-shaped body (dFB). Sleep need is encoded in the electrical excitability of these neurons, which fluctuates because two potassium conductances, voltage-gated Shaker (Kv1) and the leak channel Sandman, are modulated antagonistically. As a consequence, dFB neurons are electrically silent during waking and persistently active during sleep.

The critical dFB-intrinsic transducer of sleep pressure is Shaker's beta-subunit, an aldo-keto reductase with a stably bound NADPH cofactor. During waking, the mitochondria of metabolically idle dFB neurons release reactive oxygen species and/or aldehydic breakdown products of peroxidized lipids, whose reduction by the beta-subunit flips the cofactor to the oxidized form. This slows the inactivation of the associated potassium conductance, boosts the repolarizing force that restores the resting membrane potential after a spike, and so enables tonically active neurons to fire at higher rates-precisely the changes required to induce sleep.

Exogenous substrates for reduction or oxidation by beta-subunits thus offer a potential new mode of intervention in sleep disorders.

We will explore this mechanistic concept in three objectives: 1) We will investigate the role of potassium channel beta-subunits-and, in particular, their oxidoreductase activity-in the regulation of mammalian sleep. We will analyze sleep in Kcnab knockout mice and localize sleep-control regions by focal viral transduction of wild-type beta-subunits. 2) We will test if lipid peroxidation products serve as endogenous substrates for beta-subunits and study the stoichiometry and regulation of the oxidoreductase reaction. 3) We will develop high-throughput target- and cell-based screens to explore the wider chemical space of beta-subunit substrates, as an initial step toward characterizing the sleep-modifying potential of hits.