The developmental emergence of tonic and phasic REM sleep in rats

Abstract

REM sleep is composed of two substates--phasic and tonic--that differ in their behavioral, sensory, and electrophysiological features. Although these substates are well characterized in adults, their developmental trajectory remains unclear. Here, we examined the development of tonic and phasic REM in rats from postnatal day (P) 12-24, spanning a period of rapid corticothalamic development. We recorded local field potentials and single units from primary motor cortex (M1), together with high-speed video and electromyographic recordings of the nuchal muscle. Periods of behavioral quiescence along with high delta power indicated NREM sleep, whereas periods of sustained muscle atonia and low delta power indicated REM sleep. At P16, M1 theta oscillations first appeared, and the delay to the first twitch increased, revealing the start of a distinct twitch-free portion of REM sleep. Motivated by this, we divided REM sleep into phasic and tonic periods, with and without twitching, respectively. Spiking activity and gamma power were consistently higher during phasic REM. At P20, phasic REM also showed faster theta oscillations than tonic REM. At P24, tonic REM was accompanied by a distinct alpha oscillation. These results show that the features distinguishing the two REM substates appear sequentially across development, revealing a progressive differentiation of REM sleep into tonic and phasic periods, a developmental refinement that may support increasingly complex forms of sleep-dependent plasticity. Significance StatementInfancy is marked by rapid circuit formation and by the dominance of REM sleep, a state thought to support early neural development. Yet sleep itself undergoes massive changes throughout this period, and the developmental timeline of REM sleep remains poorly understood. Using neural recordings and high-speed video from developing infant rats, we show that REM sleep gradually divides into two substates--tonic and phasic--across infancy. These substates exhibit age-dependent differences in movement, neural firing, and cortical oscillations, revealing increasing microstructural complexity in REM sleep. Our findings identify when these substates first emerge and how their defining features unfold over time, providing a framework for understanding how REM sleep supports developmental plasticity throughout early life.