By mapping the brain during prolonged wakefulness, researchers pinpointed neuronal populations whose activity rises with time awake and tested how strongly they shape the pressure to sleep.

Study: Wake-activated neuronal populations that regulate sleep drive. Image Credit: Radiological imaging / Shutterstock

In a recent study published in the journal Nature, researchers conducted comprehensive whole-brain activity mapping and targeted circuit manipulations in mice to elucidate the poorly understood neural mechanisms governing homeostatic sleep drive.

The study specifically investigated whole-brain responses across prolonged sleep deprivation, recovery sleep, and unperturbed circadian cycles to identify specific neuronal populations that track wake duration and regulate sleep drive.

Study findings identified the median raphe (MR) and the anterior medial preoptic area (aMPO) as candidate subcortical regions encoding sleep deficit. Furthermore, the study demonstrated that genetically targeted activation of deprivation-responsive neurons in these regions triggered longer, more intense non-rapid eye movement (NREM) sleep resembling recovery sleep, while chronic co-inhibition of MR GABAergic and serotonergic neurons reduced NREM sleep by nearly 70%; most surviving mice maintained high-arousal wakefulness without the marked behavioral deficits typically associated with severe sleep deprivation.

The authors concluded that sleep drive is an active, cell-type-specific neural process regulated by wake-activated circuits. These findings represent a significant step forward in science's understanding of mammalian sleep homeostasis.

Background

Human and animal studies have established that prolonged wakefulness inevitably escalates sleep drive, creating an irresistible physiological pressure that is normally compensated for in mammalian systems by increased sleep duration and intensity.

This homeostatic sleep pressure is reflected in elevated NREM slow-wave activity in the delta frequency range (0.5–4 Hz). Until now, the circuit-level mechanisms that generate this homeostatic urge to sleep have remained elusive.

The circuit-level mechanisms generating sleep drive remain poorly understood, including which neuronal populations track sleep deficit and how these circuits influence downstream sleep- and wake-related regions.

About the study

The present study aimed to address these unanswered questions and map the neural circuitry underlying sleep drive by systematically mapping brain-wide responses to prolonged wakefulness. The brain-mapping dataset comprised 162 cleared mouse brains, which were examined across 26 experimental conditions using whole-brain immediate-early gene (FOS) immunostaining, tissue clearing, and light-sheet fluorescence microscopy.

The study’s main experimental comparisons included: 1. Six-hour sleep deprivation in the light phase using novel objects or water-mist-induced grooming, 2. Post-deprivation recovery sleep, and 3. Undisturbed 24-hour circadian cycles. Dark-phase sleep deprivation was also used to help distinguish deprivation-related responses from circadian effects.

These datasets were acquired using automated voxel-wise clustering, which was further used to categorize brain-wide FOS dynamics into three distinct response profiles: Type 1 (early-peaking stimulus-associated responses), Type 2 (sleep-correlated recovery peaks), and Type 3 (wake-correlated, deprivation-responsive accumulation).

Researchers then tested the causal roles of these neuronal populations using TRAP2 (Fos-2A-iCreERT2) transgenic mice to selectively capture deprivation-activated neurons. Subsequent evaluations utilized Cre-dependent chemogenetic activation (hM3Dq), chronic inhibition via inward-rectifying potassium channels (Kir2.1), whole-cell patch-clamp electrophysiology, anterograde tracing, and contextual fear conditioning assays.

Study findings

The study’s whole-brain analyses revealed that Type 3 activation patterns were predominantly restricted to subcortical regions, specifically the MR and aMPO, where cellular activation scaled with wake duration.

Notably, chemogenetic activation of deprivation-TRAP cells in either the MR or aMPO during the dark phase was associated with a two- to threefold increase in NREM sleep duration and a significant amplification of NREM delta power (P < 0.01).

Anatomical tracing subsequently revealed that MR deprivation-TRAP neurons project broadly across subcortical targets, including the sleep-promoting lateral preoptic area (LPO) and wake-associated lateral hypothalamic area (LHA). Chemogenetic activation increased FOS activity in the LPO while suppressing it in the LHA, and selective activation of LPO-projecting MR neurons increased NREM sleep.

Electrophysiological slice recordings revealed that 6 hours of sleep deprivation specifically depolarized the resting membrane potential and lowered the action-potential firing threshold of MR GABAergic cells (P < 0.05), increasing their spontaneous firing rates. Serotonergic neurons did not show significant changes in resting membrane potential or firing threshold.

Histological classification demonstrated that MR deprivation-responsive cells were predominantly GABAergic (Vgat+) or serotonergic neurons. Approximately 20% of all MR Vgat+ cells and 40% of all MR serotonergic cells were deprivation-TRAP labeled.

Chemogenetic co-activation of MR Vgat+ and Sert+ neurons was shown to synergistically induce prolonged, high-intensity NREM sleep. Conversely, chronic co-inhibition via Kir2.1 was found to suppress NREM sleep by nearly 70% and to blunt delta power accumulation (β = -0.120, P = 0.013). Notably, while chronic co-inhibition caused 16.7% lethality, surviving mice maintained sustained, high-arousal wakefulness without significantly increased anxiety-like behavior in the assays used and were able to encode persistent contextual memories, although recent memory strength was modestly weaker than in controls.

Importantly, the chronically inhibited mice also showed reduced delta-power accumulation, fewer sleep attempts, and little rebound sleep despite extended wakefulness, suggesting that they accumulated less sleep pressure rather than simply tolerating conventional sleep debt.

Conclusions

The analysis concludes that homeostatic sleep drive is actively encoded and regulated by wake-activated neuronal populations within the MR and aMPO, with MR GABAergic and serotonergic neurons making particularly important contributions. Its findings show that sleep deprivation increases the intrinsic excitability of MR GABAergic neurons, suggesting that these physiological shifts may contribute to the accumulation or expression of sleep pressure while leaving the upstream mechanisms unresolved.

While further research is needed to determine whether these mouse circuits have comparable roles in humans, this study provides a mechanistic framework for investigating abnormal sleep drive and could ultimately inform research into sleep disorders.

Journal reference:

• Joo, W., Diester, C., Bitsikas, V., Panopoulou, M., Hidalgo, A., Ntemos, K., ... & Schier, A. F. (2026). Wake-activated neuronal populations that regulate sleep drive. Nature. DOI:10.1038/s41586-026-10928-3. https://www.nature.com/articles/s41586-026-10928-3