Here is one more reason teenagers need their zzz's—it appears sleep is prime time for synapse remodeling in adolescence. Researchers led by Giulio Tononi from the University of Wisconsin, Madison, have shown for the first time that in adolescent mice, sleep and waking states affect synaptic modeling. Mice showed a net loss of cortical spines after a full day's sleep, and a net spine gain after a full night of activity. Published online in the October 9 Nature Neuroscience, the result could have implications for the role of sleep in neurological disorders that have their roots in adolescence, such as schizophrenia. "The researchers show very nicely that spine growth and spine density are regulated by sleep in a global downscaling way...[and that] during the adolescent period, sleep contributes to homeostatic regulation," said Jan Born, University of Lübeck, Germany, who was not involved in the study.

Tononi's group has been developing the hypothesis that after a day of brain activity, sleep allows synapse strength to return to a sustainable baseline level where there is still the capacity for potentiation (see Tononi and Cirelli, 2006). In adult fruit flies, synapses become larger, gain AMPA receptors, and fire more strongly at the end of a waking period (see Bushey et al., 2011). "It doesn't take much for those synapses to get to a ceiling level of strength after which you can't potentiate any further," said coauthor Chiara Cirelli. "After this, the synapses are no longer conducive to learning, so there must be a mechanism for them to renormalize."

Knowing that adolescence is marked by a massive pruning of excitatory synapses, and that sleep/wake cycles influence this process in the adult brain, the researchers set out to determine whether sleep favors synaptic pruning in adolescence.

They used a two-photon microscope to observe individual neurons through a thinned portion of the skull in living mice. The researchers manually counted dendritic spines in 23- to 44-day-old mice expressing yellow fluorescent protein (YFP)-H. First author Stephanie Maret and colleagues took two images—one after the mice had been awake for more than eight hours during the night (mice are nocturnal), and then again after they had slept for six to eight hours the next day. In these animals, more spines were lost than gained during sleep, and the overall spine density fell. For a second group of mice, the researchers took images first after sleep, then after a full waking period, at which point there was a net boost in spine density. To make sure sleep—and not just passage of time—drove spine reductions, the researchers imaged a third group of sleep-deprived mice, which were kept awake for an additional six to seven hours after the normal wake period. This group had no net loss of spines, showing gains similar to mice at the end of a normal wake period. Overnight spine loss did not occur in mice that were allowed to sleep for only two to three hours, or in adult mice. The results suggest that a full sleep period is required for net spine loss and that it is unique to adolescents.

Whether sleep/wake cycles drive similar spine dynamics in humans is unclear. "In adults, we know there are serious cumulative consequences of sleep restriction for brain functioning," said Cirelli. "Based on our study, there may be anatomical or structural consequences as well." She said that it is difficult to relate the findings to spine loss and disturbed sleep/wake cycles seen in Alzheimer’s disease, or to aging, since the team neither looked at aging mice, nor saw the net sleep-associated spine change in adults. But Cirelli said that the effects of aging on spine remodeling are worth exploring.

In fact, sleep cycles have previously been shown to affect Aβ production in models of Alzheimers disease. Aβ levels rose during wakefulness and fell during sleep in a study of Tg2576 mice, and the longer those animals were kept awake, the higher the Aβ levels (see ARF related news story on Kang et al., 2009). While in that study, amped-up neuron activity seemed to be associated with the rise in Aβ production, previous research has also shown that when activity is especially high, Aβ levels fall (see ARF related news story on Verges et al., 2011). Tononi intends to explore the spine remodeling process in other brain areas, in other animals, in aging mice, and test how it responds to chronic sleep deprivation.—Gwyneth Dickey Zakaib


  1. Giulio Tononi's findings may or may not be related to the findings in humans of circadian Aβ levels. One hypothesis is that increasing spine and synaptic strength during wakefulness leads to increased Aβ production via increased synaptic activity. Conversely, decreased spine density and synaptic strength during sleep may decrease Aβ levels via decreased synaptic activity.

    View all comments by Randall Bateman

Make a Comment

To make a comment you must login or register.


News Citations

  1. Sleep Deprivation Taxes Neurons, Racks Up Brain Aβ?
  2. Brain Activity and Aβ—The Interstitial Plot Thickens

Paper Citations

  1. . Sleep function and synaptic homeostasis. Sleep Med Rev. 2006 Feb;10(1):49-62. PubMed.
  2. . Sleep and synaptic homeostasis: structural evidence in Drosophila. Science. 2011 Jun 24;332(6037):1576-81. PubMed.
  3. . Amyloid-beta dynamics are regulated by orexin and the sleep-wake cycle. Science. 2009 Nov 13;326(5955):1005-7. PubMed.
  4. . Opposing synaptic regulation of amyloid-β metabolism by NMDA receptors in vivo. J Neurosci. 2011 Aug 3;31(31):11328-37. PubMed.

Other Citations

  1. Tg2576 mice

Further Reading

Primary Papers

  1. . Sleep and waking modulate spine turnover in the adolescent mouse cortex. Nat Neurosci. 2011 Nov;14(11):1418-20. PubMed.