The AP-2 transcription factors are crucial for regulating sleep in both vertebrate and invertebrate animals. In mice, loss of function of the transcription factor AP-2β (TFAP2B) reduces non-rapid eye movement (NREM) sleep. When and where TFAP2B functions, however, is unclear. Here, we used the Cre-loxP system to generate mice in which Tfap2b was specifically deleted in the nervous system during development and mice in which neuronal Tfap2b was specifically deleted postnatally. Both types of mice exhibited reduced NREM sleep, but the nervous system-specific deletion of Tfap2b resulted in more severe sleep phenotypes accompanied by defective light entrainment of the circadian clock and stereotypic jumping behavior. These findings indicate that TFAP2B in postnatal neurons functions at least partly in sleep regulation and imply that TFAP2B also functions either at earlier stages or in additional cell types within the nervous system.

We examined the associations between electroencephalogram (EEG)-based sleep characteristics and physical health parameters in general adults via a cross-sectional study recruiting 100 volunteers aged 30-59 years. Sleep characteristics were measured at home using a portable multichannel electroencephalography recorder. Using the k-means +  +  clustering method, according to 10 EEG-based parameters, participants were grouped into better (n = 39), middle (n = 46), and worse (n = 15) sleep groups. Comparing 50 physical health parameters among the groups, we identified four signals of difference (P < 0.05), including systolic (sBP) and diastolic blood pressure (dBP), γ-glutamyl transpeptidase (γ-GTP), and serum creatinine, where sBP reached a Bonferroni-corrected threshold (P < 0.001). The sBP was higher by 7.9 (95% confidence interval 1.9-13.9) and 15.7 (7.3-24.0) mmHg before adjustment and 5.4 (- 0.1-10.9) and 8.7 (1.1-16.3) mmHg after adjustment for age, sex, body mass index, smoking, drinking habits, and 3% oxygen desaturation index in the middle and worse sleep groups, respectively, than in the better group. As another approach, among 500 combinations of EEG-based and physical health parameters, there were 45 signals of correlation, of which 4 (N1% and sBP, dBP, γ-GTP, and triglycerides) reached a Bonferroni-corrected threshold (P < 0.0001). Thus, EEG-based sleep characteristics are associated with several physical health parameters, particularly sBP.

Automated sleep stage scoring for mice is in high demand for sleep research, since manual scoring requires considerable human expertise and efforts. The existing automated scoring methods do not provide the scoring accuracy required for practical use. In addition, the performance of such methods has generally been evaluated using rather small-scale datasets, and their robustness against individual differences and noise has not been adequately verified. This research proposes a novel automated scoring method named "MC-SleepNet", which combines two types of deep neural networks. Then, we evaluate its performance using a large-scale dataset that contains 4,200 biological signal records of mice. The experimental results show that MC-SleepNet can automatically score sleep stages with an accuracy of 96.6% and kappa statistic of 0.94. In addition, we confirm that the scoring accuracy does not significantly decrease even if the target biological signals are noisy. These results suggest that MC-SleepNet is very robust against individual differences and noise. To the best of our knowledge, evaluations using such a large-scale dataset (containing 4,200 records) and high scoring accuracy (96.6%) have not been reported in previous related studies.

Sleep-stage classification is essential for sleep research. Various automatic judgment programs, including deep learning algorithms using artificial intelligence (AI), have been developed, but have limitations with regard to data format compatibility, human interpretability, cost, and technical requirements. We developed a novel program called GI-SleepNet, generative adversarial network (GAN)-assisted image-based sleep staging for mice that is accurate, versatile, compact, and easy to use. In this program, electroencephalogram and electromyography data are first visualized as images, and then classified into three stages (wake, NREM, and REM) by a supervised image learning algorithm. To increase its accuracy, we adopted GAN and artificially generated fake REM sleep data to equalize the number of stages. This resulted in improved accuracy, and as little as one mouse's data yielded significant accuracy. Due to its image-based nature, the program is easy to apply to data of different formats, different species of animals, and even outside sleep research. Image data can be easily understood; thus, confirmation by experts is easily obtained, even when there are prediction anomalies. As deep learning in image processing is one of the leading fields in AI, numerous algorithms are also available.

Sleep is affected by the environment. In rodents, changes in the amount of rapid eye movement sleep (REMS) can precede those of other sleep/wake stages. The molecular mechanism underlying the dynamic regulation of REMS remains poorly understood. Here, we focused on the sublaterodorsal nucleus (SLD), located in the pontine tegmental area, which plays a crucial role in the regulation of REMS. We searched for genes selectively expressed in the SLD and identified copine-7 (Cpne7), whose involvement in sleep was totally unknown. We generated Cpne7-Cre knock-in mice, which enabled both the knockout (KO) of Cpne7 and the genetic labeling of Cpne7-expressing cells. While Cpne7-KO mice exhibited normal sleep under basal conditions, the amount of REMS in Cpne7-KO mice was larger compared to wildtype mice following cage change or water immersion and restraint stress, both of which are conditions that acutely reduce REMS. Thus, it was suggested that copine-7 is involved in negatively regulating REMS under certain conditions. In addition, chemogenetically activating Cpne7-expressing neurons in the SLD reduced the amount of REMS, suggesting that these neurons negatively regulate REMS. These results identify copine-7 and Cpne7-expressing neurons in the SLD as candidate molecular or neuronal components of the regulatory system that controls REMS.

The amount, quality, and diurnal pattern of sleep change greatly during development. Developmental changes of sleep/wake architecture are in a close relationship to brain development. The fragmentation of wake episodes is one of the salient features in the neonatal period, which is also observed in mature animals and human individuals lacking neuropeptide orexin/hypocretin signaling. This raises the possibility that developmental changes of lateral hypothalamic orexin neurons are relevant to the development of sleep/wake architecture. However, little information is available on morphological and physiological features of developing orexin neurons. To address the cellular basis for maturation of the sleep/wake regulatory system, we investigated the functional development of orexin neurons in the lateral hypothalamus. The anatomical development as well as the changes in the electrophysiological characteristics of orexin neurons was examined from embryonic to postnatal stages in orexin-EGFP mice. Prepro-orexin promoter activity was detectable at embryonic day (E) 12.0, followed by expression of orexin A after E14.0. The number of orexin neurons and their membrane capacitance reached similar levels to adults by postnatal day (P) 7, while their membrane potentials, firing rates, and action potential waveforms were developed by P21. The hyperpolarizing effect of serotonin, which is a major inhibitory signal for adult orexin neurons, was detected after E18.0 and matured at P1. These results suggest that the expression of orexin peptides precedes the maturation of electrophysiological activity of orexin neurons. The function of orexin neurons gradually matures by 3 weeks after birth, coinciding with maturation of sleep/wake architecture.