― 257 ―EEG/EMG recording and sleep scoringMice were kept at a constant light/dark cycle (12:12). The implanted EEG and EMG wires were connected to an EEG/EMG acquisition system (Intan) through extension cables and a commutator that could rotate to relieve the torque on the extension cables that is generated as the mouse moved. The EEG and EMG signals were high-pass filtered at 0.5 Hz and recorded at a sampling rate of 1 kHz. For sleep scoring, custom MATLAB scripts were used. Briefly, the EEG/EMG data were chopped into 4-s epochs that were subjected to FFT. The vigilance state in each epoch was manually classified as wakefulness, NREM sleep, and REM sleep based on EEG band powers and the integral of EMG signals. The state with the most extended duration was assigned when a single epoch is composed of multiple vigilance states. The amount, episode duration, and power of sleep spindles (8-15 Hz) were then analyzed for each vigilance state and compared between the groups.Immunostaining and image analysisCoronal sections including the VM were collected from the VMCalb1-ablated mice. The sections were stained with a primary antibody, mouse anti-Calb1 antibody (CB300, Swant), followed by a secondary antibody, Alexa 647 anti-mouse IgG (A-21236, Thermo Fisher). After being mounted on a glass slide, the sections were imaged under a microscope. For image analysis, the percentage of pixels occupied by Calb1 signals in the VM was analyzed using ImageJ. Pre-pulse inhibition (PPI)Mice were subjected to a home-made PPI device to measure pre-pulse inhibition 1). After EEG/EMG recording, the mice were single-housed for at least one week. They were then acclimatized in a box on the PPI device for 10 minutes, followed by 10 trials (without pre-pulse) and 10 trials (with pre-pulse). The data on sound intensity and acoustic startle response were collected. The pulse was 20 milliseconds long with a sound intensity of 117 dB, and the pre-pulse was 4 milliseconds long with a sound intensity of 76 dB. The interval between pre-pulse and pulse is 30 milliseconds. The interval between trials is randomized within a range of 10 to 30 seconds.Results and DiscussionDecreased sleep spindles in the parietal EEG in the VM-ablated miceI aimed to examine whether the ventromedial nucleus of the thalamus (VM) is involved in the reduced activity of sleep spindles that have been observed in patients suffering from schizophrenia and their first-degree relatives 2). The VM was ablated by virally expressing Cre recombinase and Cre-dependent diphtheria toxin A, which blocks protein synthesis and thus kills all the cells nonspecifically around the VM in WT mice. As a result, the amount of NREM sleep was slightly increased, and the episode duration of NREM sleep remained unchanged in the mice ablated with VM (VM-ablated mice) during the light phase (Fig. 1A, B). Sleep spindles (cortical oscillations in the range of 8-16 Hz) were slightly reduced in the frontal cortex and reduced mainly in the parietal cortex in the VM-ablated mice, indicating the VM may be necessary for generating normal sleep spindles broadly across the cortex (Fig. 1C, D).
元のページ ../index.html#259