REM Sleep Boosts Brain Fuel Supply Yet Drops Neuronal ATP Levels in Mice
Transparent-skull imaging in mice revealed a REM sleep paradox where blood supply and astrocytic pyruvate increase while neuronal ATP declines. The finding reframes brain energy budgeting during dreaming and points to possible mechanisms for REM-related cognitive and fatigue effects. Human translation could reshape sleep medicine.
The study used UV-cured transparent skulls in mice to enable real-time wide-field imaging of blood volume, neuronal ATP, and astrocytic pyruvate across sleep stages. Theta-band activity in non-REM sleep predicted blood volume increases seconds later, while a posterior-to-anterior blood volume wave began 50 seconds before REM onset, indicating anticipatory metabolic preparation. Neuronal ATP fell once REM started despite rising fuel indicators, suggesting high demand for synaptic reorganization or altered astrocyte-neuron lactate shuttling.
This pattern challenges the assumption that increased cerebral blood flow directly supports energy availability during high-activity states like dreaming. It aligns with prior human PET studies showing elevated glucose metabolism in REM yet extends them by separating supply from actual ATP levels at the cellular scale. The work also echoes findings on astrocyte roles in metabolic coupling from papers on wake-sleep transitions.
If replicated in humans, these dynamics could explain post-dream fatigue and inform interventions for insomnia or mood disorders tied to REM dysregulation. Next steps include testing pharmacological or optogenetic blocks of the blood volume wave to determine causality.
Limitations include the small rodent sample and indirect ATP readout; larger-scale human validation with simultaneous EEG and metabolic imaging would strengthen causal claims.
Tohoku University team: Human REM ATP drops exceeding 15% will predict next-day cognitive fatigue scores above threshold in combined fNIRS-EEG studies within 18 months.
Sources (3)
- [1]Primary Source(https://www.nature.com/articles/s42003-024-06812-3)
- [2]Supporting Source(https://www.cell.com/neuron/fulltext/S0896-6273(22)00345-1)
- [3]Supporting Source(https://www.science.org/doi/10.1126/science.abn2688)