Adolescent changes in the homeostatic and circadian regulation of sleep

Abstract

Sleep deprivation among adolescents is epidemic. We argue that this sleep deprivation is due in part to pubertal changes in the homeostatic and circadian regulation of sleep. These changes promote a delayed sleep phase that is exacerbated by evening light exposure and incompatible with aspects of modern society, notably early school start times. In this review of human and animal literature, we demonstrate that delayed sleep phase during puberty is likely a common phenomenon in mammals, not specific to human adolescents, and we provide insight into the mechanisms underlying this phenomenon.

Fig. 1

Two potential mechanisms underlying delayed circadian phase during puberty. These mechanisms are illustrated using a PRC depicting the circadian phase response of male degus to a light pulse presented at different times of the day. Phase shift magnitude is graphed in hours, with positive values indicating phase advance, and negative values indicating phase delay. Time of day is graphed in reference to the former light cycle (zeitgeber time, lights on during ZT 0–12). Therefore, if pubertal animals have a delay in the phase of the circadian pacemaker, then the phase of the PRC would also be delayed in these models. Model a: An elongation of τ during puberty would cause the circadian pacemaker and its rhythmic output to phase delay relative to the light cycle. The delay would provide more light exposure at a phase when the pacemaker is sensitive to advancing phase shift and thus allow entrainment (because τ −24 h = Φ, with Φ representing the magnitude of necessary daily phase resetting). Model b: An increase in the circadian pacemaker's relative sensitivity to the phase-delaying effects of light would also cause the pacemaker to delay relative to the light cycle. This delay would result because the pacemaker would need more light exposure at a phase when it is sensitive to advancing phase shift.

Fig. 2

Pubertal mice exhibit an exaggerated delay in circadian phase in response to evening light. Female mice were placed into constant darkness (DD) for two weeks and then exposed to a 15-min light pulse (150 lx). Circadian phase shift in response to the pulse was calculated in reference to sham (no pulse) conditions, and circadian time was defined in reference to the activity rhythms of the individual mice (activity onset = CT12). Therefore, unlike figure 1, if pubertal animals have a delay in the phase of the circadian pacemaker, the phase of the PRC would not be delayed in this figure. The dotted line represents the phase response of mice that were likely to be pubertal at the time of the light-pulse (P49, n = 34) and the dark line represents the phase response of adults (P140, n = 34). Each point represents the average phase shift produced by light presented during a 1.5-hour bin. The sample size for each bin is represented by the size of the data point (n = 1–8). Adapted from Weinert and Kompauerova [52].