Do sedatives engage natural sleep pathways? It is usually assumed that anesthetic-induced sedation and loss-of-righting-reflex (LORR) arise by influencing the same circuitry to lesser or greater extents. For the α2 adrenergic receptor agonist dexmedetomidine, we find that sedation and LORR are in fact distinct states, requiring different brain areas, the preoptic hypothalamic area and locus coeruleus (LC) respectively. Selective knockdown of α2A adrenergic receptors from the LC abolished dexmedetomidine-induced LORR, but not sedation. Instead, we found that dexmedetomidine-induced sedation resembles the deep recovery sleep that follows sleep deprivation. We used TetTag-pharmacogenetics in mice to functionally mark neurons activated in the preoptic hypothalamus during dexmedetomidine-induced sedation or recovery sleep. The neuronal ensembles could then be selectively reactivated. In both cases NREM sleep, with the accompanying drop in body temperature, was recapitulated. Thus α2 adrenergic receptor-induced sedation and recovery sleep share hypothalamic circuitry sufficient for producing these behavioral states.

In humans and other mammalian species, lesions in the preoptic area (POA) of the hypothalamus cause profound sleep impairment1–5, indicating a crucial role of the POA in sleep generation. However, the underlying circuit mechanism remains poorly understood. Electrophysiological recordings and c-Fos immunohistochemistry showed the existence of sleep-active neurons in the POA, especially in the ventrolateral preoptic area (VLPO) and median preoptic nucleus (MnPO)6–9. Pharmacogenetic activation of c-Fos-labeled sleep-active neurons has been shown to induce sleep10. However, the sleep-active neurons are spatially intermingled with wake-active neurons6,7, making it difficult to target the sleep neurons specifically for circuit analysis. Here, we have identified a population of POA sleep neurons based on their projection target and discovered their molecular markers. Using a lentivirus expressing channelrhodopsin-2 (ChR2) or a light-activated chloride channel (iC++) for retrograde labeling, bidirectional optogenetic manipulation, and optrode recording, we showed that the POA GABAergic neurons projecting to the tuberomammillary nucleus (TMN) are both sleep active and sleep promoting. Furthermore, translating ribosome affinity purification (TRAP) and single-cell RNA-seq identified candidate markers for these neurons, and optogenetic and pharmacogenetic manipulations demonstrated that several peptide markers (cholecystokinin, corticotropin releasing hormone, and tachykinin 1) label sleep-promoting neurons. Together, these findings provide easy genetic access to sleep-promoting POA neurons and a valuable entry point for dissecting the sleep control circuit.

SummaryHistaminergic neurons in the tuberomammilary nucleus (TMN) of the hypothalamus form a widely projecting, wake-active network that sustains arousal. Yet most histaminergic neurons contain GABA. Selective siRNA knockdown of the vesicular GABA transporter (vgat, SLC32A1) in histaminergic neurons produced hyperactive mice with an exceptional amount of sustained wakefulness. Ablation of the vgat gene throughout the TMN further sharpened this phenotype. Optogenetic stimulation in the caudate-putamen and neocortex of “histaminergic” axonal projections from the TMN evoked tonic (extrasynaptic) GABAA receptor Cl− currents onto medium spiny neurons and pyramidal neurons. These currents were abolished following vgat gene removal from the TMN area. Thus wake-active histaminergic neurons generate a paracrine GABAergic signal that serves to provide a brake on overactivation from histamine, but could also increase the precision of neocortical processing. The long range of histamine-GABA axonal projections suggests that extrasynaptic inhibition will be coordinated over large neocortical and striatal areas.

SummaryCircadian clocks allow anticipation of daily environmental changes [1]. The suprachiasmatic nucleus (SCN) houses the master clock, but clocks are also widely expressed elsewhere in the body [1]. Although some peripheral clocks have established roles [1], it is unclear what local brain clocks do [2, 3]. We tested the contribution of one putative local clock in mouse histaminergic neurons in the tuberomamillary nucleus to the regulation of the sleep-wake cycle. Histaminergic neurons are silent during sleep, and start firing after wake onset [4–6]; the released histamine, made by the enzyme histidine decarboxylase (HDC), enhances wakefulness [7–11]. We found that hdc gene expression varies with time of day. Selectively deleting the Bmal1 (also known as Arntl or Mop3 [12]) clock gene from histaminergic cells removes this variation, producing higher HDC expression and brain histamine levels during the day. The consequences include more fragmented sleep, prolonged wake at night, shallower sleep depth (lower nonrapid eye movement [NREM] δ power), increased NREM-to-REM transitions, hindered recovery sleep after sleep deprivation, and impaired memory. Removing BMAL1 from histaminergic neurons does not, however, affect circadian rhythms. We propose that for mammals with polyphasic/nonwake consolidating sleep, the local BMAL1-dependent clock directs appropriately timed declines and increases in histamine biosynthesis to produce an appropriate balance of wake and sleep within the overall daily cycle of rest and activity specified by the SCN.

Rapid eye movement (REM) and non-REM (NREM) sleep are controlled by specific neuronal circuits. Here we show that galanin-expressing GABAergic neurons in the dorsomedial hypothalamus (DMH) comprise separate subpopulations with opposing effects on REM versus NREM sleep. Microendoscopic calcium imaging revealed diverse sleep-wake activity of DMH GABAergic neurons, but the galanin-expressing subset falls into two distinct groups, either selectively activated (REM-on) or suppressed (REM-off) during REM sleep. Retrogradely labeled, preoptic area (POA)-projecting galaninergic neurons are REM-off, whereas the raphe pallidus (RPA)-projecting neurons are primarily REM-on. Bidirectional optogenetic manipulations showed that the POA-projecting neurons promote NREM sleep and suppress REM sleep, while the RPA-projecting neurons have the opposite effects. Thus, REM/NREM switch is regulated antagonistically by DMH galaninergic neurons with intermingled cell bodies but distinct axon projections.