Sleep, brain development, and autism spectrum disorders: Insights from animal models

Abstract

Sleep is an evolutionarily conserved and powerful drive, although its complete functions are still unknown. One possible function of sleep is that it promotes brain development. The amount of sleep is greatest during ages when the brain is rapidly developing, and sleep has been shown to influence critical period plasticity. This supports a role for sleep in brain development and suggests that abnormal sleep in early life may lead to abnormal development. Autism spectrum disorder (ASD) is the most prevalent neurodevelopmental disorder in the United States. It is estimated that insomnia affects 44%-86% of the ASD population, predicting the severity of ASD core symptoms and associated behavioral problems. Sleep problems impact the quality of life of both ASD individuals and their caregivers, thus it is important to understand why they are so prevalent. In this review, we explore the role of sleep in early life as a causal factor in ASD. First, we review fundamental steps in mammalian sleep ontogeny and regulation and how sleep influences brain development. Next, we summarize current knowledge gained from studying sleep in animal models of ASD. Ultimately, our goal is to highlight the importance of understanding the role of sleep in brain development and the use of animal models to provide mechanistic insight into the origin of sleep problems in ASD.

Keywords: ASD; EEG; animal models; autism; development; homeostasis; insomnia; neurodevelopment; rodents; sleep.

Figure 1.

Sleep time as a function of age. Percent time over 24 hours spent in wakefulness (white), rapid eye movement sleep (REMS; red) and non-rapid eye movement sleep (NREMS; blue) on the y-axis. In purple: age ranges in which sleep states are not identifiable using EEG. The x-axis represents age. On the left, sleep time across the lifespan in rats, summarizing studies at the following age ranges: P9 (Seelke and Blumberg, 2008), P12-P60 (Frank and Heller, 1997), 3-12 months (Zepelin et al., 1972). On the right, sleep time across the lifespan in humans, adapted from (Roffwarg et al., 1966).

Figure 2.. Timeline of development of sleep homeostasis in rats.

On top, age from birth: postnatal days 12 (P12), 16 (P16), 20 (P20) and 24 (P24) Below, graphs of changes in non-rapid eye movement sleep (NREMS) delta power (0.5-4.0Hz) after 3 hours of sleep deprivation (SD) corresponding to the different post-weaning ages (P12-P24) depicted above in developing rats. Mean ± SEM NREMS delta power is expressed as percent of the 12 hours mean NREMS delta power value before sleep deprivation at hours 0-3. Hour 0 represents the first hour of light period. Data originally reported by (Frank and Heller, 1997).

Figure 3.. Outline of experimental design for sleep studies in rodent models.

Mice are implanted with electroencephalographic (EEG) and electromyographic (EMG) electrodes under anesthesia. Four EEG electrodes are placed contralaterally over frontal (2) and parietal (2) cortices, and two EMG electrodes are inserted in the nuchal muscles. Mice are allowed 5 days of recovery from surgery prior to habituation to the recording environment. After one-week habituation, mice undergo 24 hours of undisturbed baseline EEG and EMG recording starting at light onset. The next day, mice are sleep deprived (SD) for 5 hours via gentle handling beginning at light onset. Mice are then allowed 19 hours undisturbed recovery sleep. EEG and EMG data are collected via a lightweight, counterbalanced cables, amplified, and digitized at 200 Hz using acquisition software. Wakefulness and sleep states are determined by visual inspection of the EEG waveform, EMG activity, and fast Fourier transform (FFT) analysis by an experimenter blinded to the genotype. Data is scored as wakefulness, NREMS, or REMS with 4 second resolution bins. Sleep parameters are quantified from the scored data and statistically analyzed. Required parameters for sleep characterization are as follows: time in state (wake, NREMS and REMS, baseline and SD day), spectral power for each state (baseline and SD day), latency to sleep following SD, sleep fragmentation (number of bouts in each state per hours, baseline and SD). Standardized protocol adapted from (Ingiosi et al., 2019).