The Role Of Cholinergic Basal Forebrain Neurons In Sleep-Wake Regulation

Background and Objective:Sleep-wake cycle is an advanced brain function. Sleep disorder or sleep deprivation cause negative emotions, and even serious neuropsychological diseases. Rodents mostly sleep in daytime and are active at night. According to electroencephalographic (EEG) and electromyographic (EMG) recording, the normal sleep state is divided into slow-wave sleep (SWS) and rapid eye movement (REM) sleep. The basal forebrain (BF) consists of a group of structures, including substantia innominata (SI), nucleus of the horizontal limb of the diagonal band (HDB), magnocellular preoptic nucleus (MCPO), which provide the major cholinergic innervation of the cerebral cortex. In the BF, the cholinergic neurons comprise only-5%of the total cell population, while GABAergic neurons account for～35%and glutamatergic neurons account for～55%. Previous studies reported that the BF plays a crucial role in increasing cortical activition during REM sleep and wakefulness.Previous studies usually used electrical stimulation and pharmacological techniques to activate or damage cholinergic BF neurons. The specificity of the electrical stimulation is poor so it is difficult to clarify the role of the cholinergic neurons in the BF in sleep-wake circuits by electrical stimulation of the BF. In rodents, sleep or wakefulness episodes last for only a couple of minutes due to immediate sleep-wake transitions, so pharmacological method could not effectively unravel the immediate sleep-wake transitions in rat or mouse. Furthermore, it is not possible to determine the role of cholinergic BF neuronal activity in sleep-wake transitions with precise temporal relationship using these traditional approaches. So, new tools are necessary to selectively manipulate the neuronal activity of the cholinergic BF neurons relevant to natural sleep-wake events in freely-moving mice.Optogenetics is a technology that combines optical control and genetic targeting using cell-type specific promoters and optical-sensitive proteins e.g., channelrhodopsin-2(ChR2) for excitation or archearhodopsin (Arch) for inhibition for the precise manipulation of neuronal functions. Light-sensitive ion channels respectively have cation or anion selectivity to specific wavelengths of light stimuli, which could cause changes of the membrane potential of the cell membrane and selectively excite or inhibit cells. Optogenetics method has cell-type selectivity with a high spatial and temporal resolution, thus operates neurons for non-invasive precise manipulation.To identify a conclusive, causal role for the cholinergic neurons in the BF in natural sleep-wake events, we selectively activated these neurons with ChR2in the ChAT-ChR2-EYFP transgenic mice (ChAT-ChR2mice) or selectively silenced these neurons with Arch in the ChAT-Arch-GFP mice (ChAT-Arch mice).Methods:ChAT-ChR2mice and ChAT-Arch mice weighing25-30g were used in the present study. We got the ChAT-Arch mice by hybridizing ROSA-Arch-GFP mice and ChAT-Cre mice, and then identified genotype via PCR. In in vitro experiment,4-weeks old ChAT-Arch mice were used. After deeply anesthetized with pentobarbital sodium (100mg/kg, i.p.) and decapitated. The brain was cut coronally into300-μm slices followed by whole cell recording. Action potentials were recorded under current-clamp mode and light-induced current were recorded under voltage-clamp mode. For in vivo experiments, animals were anesthetized and mounted on a small-animal stereotaxic frame. The cannula was placed above the BF [anteroposterior (AP):-0.7mm, mediolateral (ML):1.6mm, dorsoventral (DV):4.5mm]. Four skull screw holes were drilled, into which screws fit tightly and were driven through the skull to the surface of the dura. The animals were also implanted with a custom-made EEG/EMG unit placed on the rear of the skull, posterior to the cannula implantation. EEG signals were recorded from electrodes placed on the frontal cortex (AP:+2.0mm, ML:1.0mm). Two stainless steel wires were inserted into neck muscles as EMG electrodes. After the surgical procedures, animals were allowed to recover in individual chambers for at least7days. Each animal was transferred to a recording chamber and connected to an EEG/EMG headstage and the optical fiber. For stimulating ChR2or Arch, a473-nm or589-nm solid-state laser diode was used. The power of the laser at the tip of the optical fiber was measured by a optical power meter (PM10, Coherent) before experiments. For in vitro experiment, we used continuous yellow light or pulses with20-ms width and20-Hz frequency for30or60s. For in vivo acute experiment30-ms blue light stimulation at20Hz for15s and continuous yellow light stimulation for30s were used. For in vivo chronic experiment6-h yellow light inactivation (continuous for60s, once two minutes) was delivered. After experiments, the brains of the recorded animals were taken out to exam the location of optical fiber. The brain slices (30μm) including the BF were used for immunohistochemistry to exam the specificity and efficiency of expression of ChR2or Arch in the cholinergic BF neurons.Results:1. Immunohistochemistry reaction of ChAT revealed that97.6±0.9%(n=4mice) ChR2-EYFP and92.1±1.2%(n=5mice) Arch-GFP positive cells were ChAT positive.2. The20Hz/20ms or continuous yellow light illumination in vitro for30s or60s inhibited the activity of the Arch-expressing cholinergic BF neurons.3. There was a significantly increased probability of SWS-to-waking transition during15-s photostimulation (20Hz/30ms) in inactive period of the ChAT-ChR2mice (66.1±6.2%), while in wild type (WT) mice, the probability of SWS-to-waking transition during15-s photostimulation was17.9±0.3%(P＜0.01, n=60stimulations in3mice). The results in active period were similar to those observed in inactive period.4. There was also a significantly increased probability of SWS-to-REM transition during15-s photostimulation in inactive period of the ChAT-ChR2mice (19.5±3.3%), while in WT mice, the probability of SWS-to-REM transition during15-s photostimulation was2.9±2.9%(P＜0.05, n=60stimulations in3mice). The results in active period were similar to those observed in inactive period.5. During the inactive period there was a significantly reduced probability of SWS-to-wake transition during30-s bilateral inactivation in the ChAT-Arch mice (11.7±2.1%), while in control mice and unilateral inactivation of ChAT-Arch mice, the probability was33.0±2.5%and27.5±3.1%(P＜0.01, n=60stimulations in3mice), but there was no difference of the transition from waking-to-SWS in the ChAT-Arch mice compared with control after30-s bilateral inactivation. The results in active period were similar to those observed in inactive period.6. During the inactive period continuous inactivation of cholinergic BF neurons during SWS extended the duration of SWS from the baseline (73.5±5.3s) to105±10.5s in the ChAT-Arch mice with bilateral inactivation (P＜0.05, n=75stimulations in3mice). Meanwhile, we also found the REM sleep duration was not affected in the continuous unilateral or bilateral inactivation of cholinergic BF neurons during REM sleep in inactive period and active period.7. Six hours of inactivation (13:00to19:00) during the inactive period and inactivation (19:00to1:00) during the active period in the ChAT-Arch mice both increased the amount of SWS (P＜0.05, n=3mice) and decreased the amount of wakefulness during the6hours (P＜0.05, n=3mice), and the amount of REM sleep did not changed. Meanwhile, there was a compensatory increase of wakefulness and reduction of SWS in the following18h after6-h inactivation. But the total time of wakefulness, SWS and REM sleep in24h did not change after inactivation compared with baseline.Conclusion:Activation of cholinergic BF neurons alone is sufficient to promote but not necessary to maintain wakefulness or REM sleep episodes. In addition, as an important arousal system in the brain, the changes of functional status in BF do not affect the sleep rhythm, sleep homeostasis as well as sleep-wake architecture.