Sleep and synaptic homeostasis: a hypothesis

Abstract

During much of sleep, the cerebral cortex is rippled by slow waves, which appear in the electroencephalogram as oscillations between 0.5 and 4.5 Hz. Slow waves are regulated as a function of previous wakefulness, being maximal at the beginning of sleep and then progressively returning to a baseline level. This paper discusses a hypothesis about the significance of slow-wave activity and its homeostatic regulation. The hypothesis is as follows:1. Wakefulness is associated with synaptic potentiation in several cortical circuits; 2. Synaptic potentiation is tied to the homeostatic regulation of slow-wave activity;3. Slow-wave activity is associated with synaptic downscaling; 4. Synaptic downscaling is tied to the beneficial effects of sleep on performance.The hypothesized link between sleep and synaptic homeostasis is supported by several lines of evidence and leads to testable predictions.

Overlapping memory replay during sleep builds cognitive schemata

“…If left unchecked, potentiation during sleep-dependent consolidation, as well as during wakeful encoding, would eventually lead to synaptic saturation. The synaptic homeostasis hypothesis [29,30] proposes that all synapses belonging to neurons that were potentiated during the day are downscaled during SWS. Because each synapse is downscaled by a percentage of its current strength, the ratio between synapses is maintained.…”

“…8 will be more strongly activated and their interconnections will therefore become stronger than connections relating to a single memory (Figure 2) [31]. Under the synaptic homeostasis hypothesis [29,30], recently potentiated synapses will be downscaled during SWS but the strongly potentiated interconnections between areas of overlap will survive this process. Conversely, Weaker connections between nonoverlapping portions of these memories will be lost during downscaling (Figure 2c).…”

“…This suggests that these structures might interact during schema formation, and therefore this finding supports a role for the vmPFC in schema formation and/or in the representation of partially consolidated schemata. explain how, in combination with the synaptic downscaling that occurs during SWS [29,30], such overlap could lead to the gradual formation of conceptual schemata. Overlap in memory reactivation can occur either simultaneously or sequentially ( Figure 1).…”

Rapid alterations in cortical protein profiles underlie spontaneous sleep and wake bouts

“…Proposed roles for sleep include the maintenance of body temperature [McGinty and Szymusiak, 1990;Wehr, 1992], energy homeostasis [Adam, 1980;Walker and Berger, 1980;Benington and Heller, 1995], immune function [Majde and Krueger, 2005;Opp, 2005], synaptic plasticity [Tononi and Cirelli, 2006] and memory consolidation/reconsolidation [Stickgold and Walker, 2005;Born et al, 2006]. The amount of time devoted to sleep, the deleterious effects that accompany SD, its' proposed roles in the maintenance of a variety of biological functions and the apparent conservation in animals are consistent with the idea that sleep-wakefulness is a fundamental behavior, similar to eating and drinking.Though it has long been hypothesized that the impact of sleep on the intracellular milieu affects neural function [Rechtschaffen, 1998;Steriade and Timofeev, 2003;Tononi and Cirelli, 2003], relatively little is known of the intracellular consequences of sleep due to its' complex underpinnings and the lack of appropriately sensitive high throughput technologies. Early studies showed increased regional protein and RNA contents during sleep [Giuditta et al, 1980;Ramm and Smith, 1990;Nakanishi et al, 1997] transcriptional activation.…”

“…Though it has long been hypothesized that the impact of sleep on the intracellular milieu affects neural function [Rechtschaffen, 1998;Steriade and Timofeev, 2003;Tononi and Cirelli, 2003], relatively little is known of the intracellular consequences of sleep due to its' complex underpinnings and the lack of appropriately sensitive high throughput technologies. Early studies showed increased regional protein and RNA contents during sleep [Giuditta et al, 1980;Ramm and Smith, 1990;Nakanishi et al, 1997], consistent with the stimulation of protein biosynthesis by transcriptional activation.…”