Why Does Sleep Slow-Wave Activity Increase After Extended Wake? Assessing the Effects of Increased Cortical Firing During Wake and Sleep

Abstract

During non-rapid eye movement (NREM) sleep, cortical neurons alternate between ON periods of firing and OFF periods of silence. This bi-stability, which is largely synchronous across neurons, is reflected in the EEG as slow waves. Slow-wave activity (SWA) increases with wake duration and declines homeostatically during sleep, but the underlying mechanisms remain unclear. One possibility is neuronal "fatigue": high, sustained firing in wake would force neurons to recover with more frequent and longer OFF periods during sleep. Another possibility is net synaptic potentiation during wake: stronger coupling among neurons would lead to greater synchrony and therefore higher SWA. Here, we obtained a comparable increase in sustained firing (6 h) in cortex by: (1) keeping mice awake by exposure to novel objects to promote plasticity and (2) optogenetically activating a local population of cortical neurons at wake-like levels during sleep. Sleep after extended wake led to increased SWA, higher synchrony, and more time spent OFF, with a positive correlation between SWA, synchrony, and OFF periods. Moreover, time spent OFF was correlated with cortical firing during prior wake. After local optogenetic stimulation, SWA and cortical synchrony decreased locally, time spent OFF did not change, and local SWA was not correlated with either measure. Moreover, laser-induced cortical firing was not correlated with time spent OFF afterward. Overall, these results suggest that high sustained firing per se may not be the primary determinant of SWA increases observed after extended wake.

Evidence That Homeostatic Sleep Regulation Depends on Ambient Lighting Conditions during Wakefulness

“…Along these lines, we speculate that the experienced ambient light intensity modulates "synaptic load" during wakefulness, being higher in a brighter than dim environment, and eventually leading to more EEG SWA in the subsequent night. This is in line with data in mice which demonstrated that sustained neuronal activation induced by active exploration in awake animals leads to an increase in SWA during subsequent sleep [29]. If this was true, also a more local response in EEG SWA should be expected, as demonstrated in several studies on use-dependent aspects of sleep regulation [7,8,30,31].…”

Sleep, synaptic homeostasis and neuronal firing rates

“…In another study the same group of mice underwent 6 hours of sleep deprivation with exposure to novel objects and, a few days apart and in counterbalanced order, 6 hours of optogenetic stimulation during NREM sleep [46]. In the latter condition laser stimulation was restricted to a local population of cortical neurons, with the goal of forcing these cells to maintain high levels of activity during sleep, as high as those seen during sleep deprivation.…”

“…The conclusion was that a sustained increase in population firing alone is unlikely to be the main reason why sleep pressure builds up during prolonged wake. A more likely candidate is synaptic potentiation, which can also account for the increase in neuronal coupling and cortical synchrony observed after prolonged wake [46]. A long-standing hypothesis is that neurons fire less during NREM sleep to recover from the “fatigue” accrued during wake, when overall synaptic activity is higher than in sleep [47,48].…”

Subcortical theta oscillations alternate with high amplitude neocortical population within synchronized states

“…Our results show that BL5 UP states have relatively high amplitude, while SUB UP states have lower amplitude. Amplitude during sleep has previously been characterized by slow-wave activity (SWA), which is high at sleep onset and slowly subsides throughout the night (Aeschbach & Borbély, 1993; Dijk, Brunner, & Borbély, 1990; Rodriguez et al, 2016; Vassalli & Dijk, 2009). Because of this change, we hypothesize that BL5-dominated states occur earlier during sleep, while SUB-states occur later, consistent with LOW states (Miyawaki et al, 2017).…”