Genetic dissection of hypothalamic sleep circuitry

Abstract

Do sedative drugs engage the natural sleep circuitry to produce their effects? In particular, the sedated state induced by hypnotic drugs acting at adrenergic type 2A receptors (adra2a) is difficult to distinguish from natural deep non-REM sleep, but most researchers believe these are distinct vigilance states. Here I use pharmacogenetics to demonstrate that 2 adrenergic receptor agonist (dexmedetomidine)-induced sedation, and the recovery sleep induced following sleep deprivation, do in fact share a common neuronal circuitry such that recovery sleep and sedation appear to be the same. To do this I used the TetTaging method. I expressed c-fos-promoter inducible hM3Dq receptor genes selectively in the preoptic area (PO). I then gave mice sleep deprivation and allowed them recovery sleep or sedative doses of dexmedetomidine. Only neurons previously active during rebound sleep following sleep-deprivation or those disinhibited by dexmedetomidine expressed hM3Dq receptors. Remarkably, NREM sleep with the accompanying drop in body temperature could be induced by selectively re-activating these “neural engrams” using systemic CNO a few days later. After sleep deprivation and recovery sleep, selective reactivation of neuronal ensembles in either MnPO or the LPO areas was sufficient to induce powerful NREM sleep; after dexmedetomidine-induced sedation, only selective reactivation of the LPO area could induce NREM sleep. Selective ablation of the vesicular GABA transporter gene in the LPO neurons hindered dexmedetomidine’s ability to induce sedation, suggesting that the neurons disinhibited by dexmedetomidine were GABAergic.