Sleep and Memory: The Impact of Sleep Deprivation on Transcription, Translational Control, and Protein Synthesis in the Brain

Abstract

In countries around the world, sleep deprivation represents a widespread problem affecting school age children, teenagers and adults. Acute sleep deprivation and more chronic sleep restriction adversely affect individual health, impairing memory and cognitive performance as well as increasing the risk and progression of numerous diseases. In mammals, the hippocampus and hippocampus-dependent memory is vulnerable to the effects of acute sleep deprivation. Sleep deprivation induces changes in molecular signaling, gene expression and may cause changes in dendritic structure in neurons. Genome wide studies have shown that acute sleep deprivation alters gene transcription, although the pool of genes affected varies between brain regions. More recently, advances in research have drawn attention to differences in gene regulation between the level of the transcriptome compared with the pool of mRNA associated with ribosomes for protein translation following sleep deprivation. Thus, in addition to transcriptional changes, sleep deprivation also affects downstream processes to alter protein translation. In this review, we focus on the multiple levels through which acute sleep deprivation impacts gene regulation, highlighting potential post-transcriptional and translational processes that may be affected by sleep deprivation. Understanding the multiple levels of gene regulation impacted by sleep deprivation is essential for future development of therapeutics that may mitigate the effects of sleep loss.

1. Introduction

Sleep deprivation represents a widespread public health problem in the United States as well as other countries worldwide (Hafner et al. 2017). In the U.S., more than half of school age children get insufficient sleep on a nightly basis (Tsao et al. 2021). In adolescents and teenagers, the problems of sleep deprivation affect approximately 70% of individuals (Center for Disease Control and Prevention, 2017 (Centers for Disease & Prevention 2009; Wheaton et al. 2018). In population models, approximately 27% of adults experience mild sleep restriction sleeping less than 7 hours nightly with an additional 18% sleeping less than 6 hours (Hafner et al. 2017). In addition to problems with chronic sleep restriction, approximately 70% of U.S. adults experience acute sleep deprivation at least one night per month with more than 30% of adults experiencing sleep deprivation on a regular basis (Centers for Disease & Prevention 2009). The pervasiveness of individuals experiencing regular sleep loss is seen worldwide. In the Netherlands, 43.2% of adults reported insufficient sleep in a national survey (Kerkhof 2017), while in Italy, 29.5% of adults reported insufficient sleep (Varghese et al. 2020). In the United Kingdom, approximately 16% of adults sleep less than six hours per night with another 19% receiving six to seven hours sleep per night (Hafner et al. 2017). In Australia, more than one-third of adults and half of teenagers experience inadequate sleep (Adams 2016; AIHW 2021). Many countries in Asia also have high proportions of individuals experiencing sleep deprivation with Japan often considered the world’s most sleep deprived country with 56% of adults in Japan estimated as receiving inadequate sleep (Hafner et al. 2017; Kakamu et al. 2021). Although studies of sleep from African countries are more limited, recent studies suggest that insufficient sleep and poor sleep quality is also a problem across age groups with studies highlighting sleep deprivation in older adults and in university students (Lankrew Ayalew 2022; Nsengimana et al. 2023; Stranges et al. 2012; Wang et al. 2019). Despite the growing number of population studies on sleep and sleep disorders across multiple countries, the question of whether the incidence of sleep deprivation among individuals is increasing can be difficult to answer as many large-scale studies rely upon self-reported or subjective sleep tests and the definitions of insufficient sleep can vary. However, the pervasiveness of sleep deprivation across age groups and societies necessitates increased understanding of the impacts of sleep loss.

Sleep deprivation results in significant decrements in cognition and adversely impacts individual health. Sleep disorders increase the incidence of metabolic diseases, cardiac diseases, cancer, as well as neuropsychiatric and mood disorders. Sleep and circadian rhythm disorders have also been linked to the increased incidence and accelerated progression of neurodegenerative diseases including Alzheimer’s disease (Sabia et al. 2021; Shi et al. 2018; Wang & Holtzman 2020; Wu et al. 2019). At a societal level, economic losses attributed to sleep loss due to lower workplace productivity, increased disease incidence and higher mortality risks, result in approximate annual costs in the U.S. between $280 to $411 billion with this range expected to increase to $318 to $456 billion by 2030 (Hafner et al. 2017). The cumulative costs for the economic impacts of sleep loss are predicted to be between 4 to 7 trillion dollars for the U.S. alone by 2030 (Hafner et al. 2017). In Japan, the economic impacts of sleep deprivation are estimated at greater than $138 billion annually. In Germany and the United Kingdom, economic losses due to sleep deprivation are between $50 - $60 billion annually (Hafner et al. 2017). In Australia, the annual economic costs for sleep deprivation are considered between $35 to 45 billion (Hillman et al. 2018; Streatfeild et al. 2021). Given the serious consequences of sleep loss to individuals and society, it is important to understand the behavioral, cellular and molecular consequences of sleep deprivation, particularly those processes affecting memory and performance.

2. Consequences of Sleep Deprivation

Sleep deprivation disrupts numerous physiological processes causing stress on many tissues and organs. In this section, we focus on the impacts of sleep deprivation on memory and the underlying molecular and neural changes induced by sleep deprivation (Figure 1).

Figure 1. Overview of the impacts of sleep deprivation on hippocampus dependent memory and examples of molecular pathways that are affected.

2.1. Memory

The adverse impacts of acute sleep deprivation on memory and performance are phylogenetically conserved with associative memory affected from invertebrates to humans. Table 1 provides examples of the impacts of sleep deprivation on memory across species. In Drosophila and honeybees, sleep deprivation results in deficits in memory consolidation ((Hussaini et al. 2009), particularly spatial memory in bees (Beyaert et al. 2012). Moreover, extended wake (sleep deprivation) in Drosophila decreases basal and induced neuronal activity (Bushey et al. 2015). In the marine mollusk, Aplysia californica, sleep deprivation prior to training inhibits both short and long-term associative memory (Krishnan et al. 2016a; Krishnan et al. 2016b). Sleep deprivation in zebrafish results in deficits in object discrimination (Pinheiro-da-Silva et al. 2017a), spatial learning (Pinheiro-da-Silva et al. 2017b), and conditioned avoidance (Pinheiro-da-Silva et al. 2018). In mammals, the effects of acute sleep deprivation are particularly apparent for hippocampus dependent memory (reviewed in (Havekes et al. 2015; Kreutzmann et al. 2015). Acute sleep deprivation inhibits the consolidation of hippocampus-dependent long-term memory for spatial object recognition, contextual fear conditioning, and spatial memory mice and rats (Graves et al. 2003; Guan et al. 2004; Hagewoud et al. 2010; Xu et al. 2010). In fact, sleep deprivation induces the use of alternate learning strategies to bypass hippocampus-dependent memory such as seen with spatial learning for food reward, although memory impairments are still observed in more complex reversal learning (Hagewoud et al. 2010). Furthermore, acute sleep deprivation inhibits the subsequent induction of hippocampal long-term potentiation (LTP) and heterosynaptic plasticity through synaptic tagging as seen in rodent studies (Vecsey et al. 2009; Vecsey et al. 2018; Wong et al. 2019). REM sleep deprivation also induces deficits in hippocampal LTP (Davis et al. 2003; McDermott et al. 2003). Three days of REM sleep deprivation in rats impairs hippocampus dependent memory with contextual fear conditioning, but had no effect on amygdala dependent cued fear conditioning (McDermott et al. 2003).

Table 1.

Examples of the impacts of sleep deprivation on memory from invertebrates to humans

Findings related to memory	Method and length of sleep deprivation	Organism	References
Deficits in memory consolidation and spatial memory	Bees were intermittently sleep deprived (5 min sleep dep, 5 min rest) during the entire light and dark phase	Foraging honeybees (Apis mellifera carnica)	Hussaini et al. 2009 Beyaert et al. 2012
Decreases basal and induced neuronal activity	6, 12, and 24 hours of sleep deprivation in rotating box	Drosophila	Bushey et al. 2015
Inhibits short and long-term associative memory	9 hours of sleep deprivation using context changes and tactile stimulation	Marine mollusk (Aplysia californica)	Krishnan et al. 2016a Krishnan et al. 2016b
Deficits in object discrimination, spatial learning and conditioned avoidance	6 hours of sleep deprivation by extending the light phase or brief light pulses during the dark phase	Zebrafish (Danio rerio)	Pinheiro-da-Silva et al. 2017a Pinheiro-da-Silva et al. 2017b Pinheiro-da-Silva et al. 2018
Impairs memory consolidation for contextual fear conditioning	5 hours of sleep deprivation by gentle stroking	Mus musculus	Graves et al. 2003
Impairs spatial working memory	6 or 12 hours of sleep deprivation		Hagewoud et al. 2010
Impairs spatial learning and memory	3 hours of sleep deprivation for 30 consecutive days by gentle touch with a brush		Xu et al. 2010
Inhibits the induction of hippocampal long-term potentiation and heterosynaptic plasticity	5 hours of sleep deprivation by gentle handling		Vecsey et al. 2009 Vecsey et al. 2018 Wong et al. 2019
Inhibits the consolidation of hippocampus-dependent long-term memory for spatial memory	6 hours of sleep deprivation by placing novel objects in the animal’s cage.	Rattus norvegicus	Guan et al. 2004
Impairs hippocampus dependent memory contextual fear conditioning	REM sleep deprivation on single small platform in the middle of a water tank for 72 h.		McDermott et al. 2003
Decreases hippocampal activity during episodic memory training using visual stimuli, resulting in deficits in memory encoding	One night of sleep deprivation by waking activity (Internet, e-mail, short walks, reading and board games)	Homo sapiens	Yoo et al. 2007
Alteration in the connectivity of hippocampal networks			Kaufmann et al. 2016
Deficits in declarative memory (word-pair task)			Kuhn et al. 2016
Neural network remodeling			Pesoli et al. 2022
Decreases activation of prefrontal cortex, posterior parietal cortex and supplementary motor area	Extended sleep deprivation		Mu et al. 2005

In humans, fMRI studies have been used to evaluate the effects of sleep deprivation on brain activity. One night of sleep deprivation significantly decreases hippocampal activity during episodic memory training using visual stimuli, resulting in deficits in memory encoding (Yoo et al. 2007). Interestingly, while hippocampal activity necessary for memory was impaired, alternate networks were strengthened including coupling between the hippocampus and basic vigilance networks in the upper brainstem and thalamus (Yoo et al. 2007). More recent fMRI imaging assessments of hippocampal impairments following sleep deprivation found that sleep deprivation altered the connectivity of hippocampal networks (Kaufmann et al. 2016). These results are reminiscent of the bypass of hippocampal learning networks seen in rodent studies. One night of sleep deprivation also causes significant deficits in declarative memory using a word-pair task often used to assess hippocampus-cortical network function (Kuhn et al. 2016). Thus, in humans as well as in animal models, the hippocampus appears particularly vulnerable to the effects of sleep deprivation. Sleep deprivation also affects the cortex with decreases in LTP-like plasticity observed; however, baseline cortical excitability and synaptic strength were increased after sleep deprivation (Kuhn et al. 2016). fMRI studies have also shown that extended sleep deprivation decreases activation of parts of the prefrontal cortex, the posterior parietal cortex and the supplementary motor area following a working memory task (Mu et al. 2005). Studies using magnetoencephalography to evaluate functional connectivity found that brain networks underwent widespread remodeling after sleep deprivation, potentially as a mechanism to redistribute cognitive load (Pesoli et al. 2022). Sleep deprivation also reduces connectivity and induces alterations within the dorsal attention network, the visual network and the default mode network (Gao et al. 2015; Kaufmann et al. 2016).

2.2. Molecular Signaling

At the molecular level, probably the most well-studied pathway affected by sleep deprivation is the cAMP/PKA pathway. Acute sleep deprivation inhibits cAMP/PKA signaling in the hippocampus decreasing cAMP levels (Vecsey et al. 2009; Wong et al. 2019). Increasing cAMP levels through pharmacological inhibition of phosphodiesterases counters the effects of sleep deprivation restoring synaptic plasticity and enabling the molecular consolidation of long-term memory after sleep deprivation (Vecsey et al. 2009). Moreover, if cAMP is specifically increased only in excitatory neurons during sleep deprivation, the adverse impact of sleep deprivation on long-term spatial memory is ameliorated (Havekes et al. 2014). Downstream the cAMP/PKA signaling results in the phosphorylation of the transcription factor CREB, a transcription factor necessary for many forms of long-term memory. Acute sleep deprivation decreases basal levels of pCREB in the hippocampus as well as learning induced phosphorylation of CREB (Vecsey et al. 2009; Wong et al. 2019; Xu et al. 2010). The cAMP/PKA pathway is also targeted during chronic REM sleep deprivation. Multiple days of REM sleep deprivation depresses cAMP levels in the hippocampus and impairs long-term spatial memory in the Morris water maze (Maher et al. 2021). As with acute sleep deprivation, administration of phosphodiesterase inhibitors mitigates the impact of chronic sleep deprivation on spatial memory (Maher et al. 2021). The impact of sleep deprivation on cAMP/PKA signaling is phylogenetically conserved. In zebrafish, the expression of PKA and CREB are reduced after sleep deprivation (Lee et al. 2020). Moreover, after a fear conditioning avoidance task, sleep deprivation reduced the training induced nuclear increases in PKA and CREB and decreased CREB DNA binding (Lee et al. 2020).

Interestingly, cAMP/PKA signaling also functions in the regulation of wakefulness and sleep homeostasis. Reduction in PKA activity results in increased sleep fragmentation and decreased sleep rebound after sleep deprivation (Hellman et al. 2010). CREB also appears to function in wakefulness as the deletion of Creb1 in forebrain excitatory neurons increases sleep, yet significantly decreases rebound sleep following sleep deprivation (Wimmer et al. 2021). In Drosophila, the cAMP/PKA has been shown to play a key role in regulating sleep homeostasis (Hendricks et al. 2001). Thus, the cAMP/PKA pathway and CREB dependent transcription have evolutionarily conserved roles in the regulation of sleep/wake activity and homeostatic sleep needs.

Sleep deprivation also affects other kinase signaling pathways including the extracellular signal-regulated kinase (ERK) pathway. Six hours of total sleep deprivation has been shown to reduce ERK activity in the hippocampus of rats (Guan et al. 2004) and mice (Wong et al. 2019). Furthermore, chronic REM sleep deprivation also decreases hippocampal levels of phospho-ERK1/2 (Ravassard et al. 2009). Changes in ERK signaling are also seen after sleep deprivation in Drosophila (Vanderheyden et al. 2013). Thus, a significant impact of sleep deprivation appears to be through changes in MAPK signaling.

Brain derived neurotrophic factor (BDNF) is a key regulator of both presynaptic and post-synaptic plasticity, and is highly expressed in the hippocampus (Edelmann et al. 2014; Lin et al. 2018; Wang et al. 2022). Consequently, the effects of sleep deprivation on BDNF have also been well-studied in animal models and in humans (reviewed in (Rahmani et al. 2020). Early research found that acute REM sleep deprivation in rats decreased BDNF levels in the cerebellum and the brainstem, but not the hippocampus (Sei et al. 2000). However, total sleep deprivation reduces BDNF in hippocampus (Alkadhi & Alhaider 2016). Five hours of acute sleep deprivation in mice reduces both the pro-BDNF form that occurs prior to cleavage and mature BDNF in the hippocampus (Wong et al. 2019). Longer periods of sleep restriction also reduce BDNF RNA and protein levels (Guzman-Marin et al. 2006). When sleep deprivation occurs post-learning after training for hippocampus dependent learning in the Morris Water Maze, BDNF mRNA expression is decreased (Karabulut et al. 2019). Interestingly, the effects of sleep deprivation on BDNF signaling appear to occur through its low affinity receptor, the p75 neurotrophin receptor (p75NTR). The effects of acute sleep deprivation are alleviated with restoration seen in the levels of ERK phosphorylation, pCREB and BDNF in p75NTR knockout mice (Wong et al. 2019). In humans, one night of sleep deprivation decreases BDNF levels in plasma (Kuhn et al. 2016). The effects of acute and chronic sleep deprivation on cellular signaling mechanisms appear widespread, potentially affecting numerous downstream targets and multiple cellular processes (Table 2).

Table 2.

Sleep deprivation induces changes in molecular signaling

Molecular changes	Method and length of sleep deprivation	Organism	References
Inhibits cAMP/PKA signaling in the hippocampus (decrease cAMP and pCREB levels). Reduces ERK activity and both pro-BDNF and mature BDNF in the hippocampus.	5 hours of sleep deprivation by gentle handling	Mus musculus	Vecsey et al. 2009 Xu et al. 2010 Wong et al. 2019
Reduces cAMP levels in the hippocampus	REM sleep for 12 hours a day for six consecutive days		Maher et al. 2021
Reduction in PKA activity results in increased sleep fragmentation and decreased sleep	6 hours of sleep deprivation by gentle handling		Hellman et al. 2010
Deletion of Creb in forebrain excitatory neurons increases sleep, but decreases rebound sleep			Wimmer et al. 2021
BDNF mRNA expression is decreased	3 hours of REM sleep deprivation post-learning (Morris Water Maze)		Karabulut et al. 2019
Reduction of PKA and CREB levels	One night sleep deprivation	Zebrafish (Danio rerio)	Lee et al. 2020
cAMP/PKA regulates sleep homeostasis	6 hours of sleep deprivation by mechanical stimuli	Drosophila	Hendricks et al. 2001
Increased ERK signaling increases sleep; sleep deprivation increases ERK signaling potentially to induce rebound sleep	12 hours of sleep deprivation		Vanderheyden et al. 2013
Reduction of ERK activity in the hippocampus	6 hours of sleep deprivation	Sprague Dawley rat	Guan et al. 2004
Decreases hippocampal levels of phospho-ERK1/2	Chronic REM sleep deprivation for 72 hours on multiple platforms in a standard container filled with water	Sprague Dawley rat	Ravassard et al. 2009
Reduction of BDNF RNA and protein levels	Treadmill method (3s on/12s off) for 48 hours		Guzman-Marin et al. 2006
Decreases BDNF levels in the cerebellum and the brainstem	6 hours of REM sleep deprivation	Wistar rat	Sei et al. 2000
Reduction of BDNF in the hippocampus	24 hours of sleep deprivation in the “Modified Multiple Platforms” paradigm		Alkadhi & Alhaider 2016
Decreases BDNF levels in plasma	One night of total sleep deprivation with standardized activities	Homo sapiens	Kuhn et al. 2016

2.3. Morphological Changes

Although the impacts of sleep deprivation on behavior, memory, network connectivity and cellular signaling appear striking, the evidence is less clear regarding morphological changes in neurons (Table 3). Long-term memory and synaptic plasticity are associated with dynamic regulation of dendritic spine number and synapses. Sleep/wake cycles and the circadian clock regulate synapse number in invertebrates and vertebrate model systems resulting in changes in synaptic strength (Appelbaum et al. 2010; Bushey et al. 2011; Fernandez et al. 2008; Maret et al. 2011; Mehnert et al. 2007). In Drosophila, rhythmic variations are observed for the synapses of circadian pacemaker neurons with peaks occurring during the day and troughs at night for axon diameter, dendritic branching, synapse size and synapse number (Fernandez et al. 2008; Mehnert et al. 2007; Mehnert & Cantera 2011). Sleep/wake cycles also regulate synaptic transmission with changes in synapse number increased when flies are awake and decreased during sleep (Bushey et al. 2011). Similarly, in zebrafish, overall synapse number is reduced during the night, but interestingly, inhibitory synapse number is increased (Appelbaum et al. 2010; Elbaz et al. 2017). Dendritic spine number also appears to be regulated by sleep/wake cycles in adolescent mice with decreased spine number seen during sleep (Maret et al. 2011). Thus, synaptic transmission is regulated across species by sleep/wake cycles. More recent studies have also shown that some dendritic spines are reduced in size during sleep in mice, along with the elimination of some spines (de Vivo et al. 2017; Li et al. 2017). Thus, sleep regulates dendritic and spine morphology across species.

Table 3.

Examples of the impacts of sleep deprivation on neuronal structure

Morphological findings	Method and length of sleep deprivation	Organism	References
Increases overall synapse number	6 hours of sleep deprivation by vibration	Zebrafish	Appelbaum et al. 2010
Decreases the number of inhibitory synapses	24 hours of sleep deprivation with constant light		Elbaz et al. 2017
Golgi staining: Decreased spine number in the CA1 and dentate gyrus	5 hours of sleep deprivation by gentle handling	Mus musculus	Havekes et al. 2016 Raven et al. 2019 Wong et al. 2019
Genetic fluorescent labeling: increased total spine number (thin spines) and spine volume.			Gisabella et al. 2020
Branch-specific effects on spine density in the CA1			Bolsius et al. 2022
No effect on spine number or the shape of the dendritic tree in CA1 neurons – Animals group housed with enrichment prior to sleep deprivation			Brodin et al. 2022
Branch specific reduction in dendritic spines in cortex	7 hours of sleep deprivation by gentle handling after motor learning task		Yang et al. 2014
Reduction in training-induced new spines in adolescent mice	7 hours of sleep deprivation by gentle handling after motor learning task		Adler et al. 2021
Decreased spine density in the dorsal hippocampus	6 hours of sleep deprivation for 21 consecutive days		Huang et al. 2022
Synapses are strengthened when awake and weakened during sleep	Wake/sleep cycle		de Vivo et al. 2017
REM sleep prunes new dendritic spines in the motor cortex	Sleep		Li et al. 2017
Decreased dendritic spine number in the CA1 of the hippocampus	24 hours of total sleep deprivation	Rattus norvegicus	Acosta-Pena et al. 2015

It has been more difficult to discern the effects of sleep deprivation on neuron structure and synaptic morphology, as considerable variation occurs depending upon the type of neuron and the brain region studied. In zebrafish, six hours of sleep deprivation, but not shorter periods, has been shown to increase overall synapse number (Appelbaum et al. 2010). However, more recent research found that sleep deprivation decreases the number of inhibitory synapses (Elbaz et al. 2017). Thus, inhibitory synapse number is increased during sleep, but in sleep deprivation, inhibitory neurotransmission is weakened. In mammalian research, differences between studies, may arise due to the techniques used to assess synapse number, the neurons examined or even the relative proximity of the dendrites to neuronal cell bodies. In mice, studies using Golgi staining to visualize dendritic branches and spines found that five hours of acute sleep deprivation decreases spine number in the CA1 and dentate gyrus (Havekes et al. 2016; Raven et al. 2019; Wong et al. 2019). In rats, twenty-four hours of sleep deprivation also decreased dendritic spine number in the CA1 of the hippocampus (Acosta-Pena et al. 2015). In contrast, recent research using genetic fluorescent labeling to visualize dendritic spines in excitatory CA1 neurons determined that five hours of sleep deprivation increased total spine number, primarily through an increase in thin spines, as well as an increase in spine volume (Gisabella et al. 2020). Although methodological differences in spine visualization may account for differences in spine measurements between these studies, it is more likely that differences in assessment of dendritic spines after sleep deprivation between these studies arise from positional variance in spine dynamics across dendritic branches. In-depth analysis of the effect of sleep deprivation on spine density found that sleep deprivation has branch specific effects in the CA1 (Bolsius et al. 2022). A recent study in which mice were group housed with enrichment prior to sleep deprivation found that a single session of acute sleep deprivation had no effect on spine number or the shape of the dendritic tree in CA1 neurons (Brodin et al. 2022). This study perhaps suggests that group housing and enrichment can aid in minimizing the effects of acute sleep deprivation. On the other hand, chronic sleep restriction of young mice for three weeks results in decreased spine density in the dorsal hippocampus (Huang et al. 2022). Branch specificity in spine dynamics and a reduction in spine number after sleep deprivation were previously observed in the primary motor cortex after learning (Yang et al. 2014). Animals trained for a motor learning task followed by seven hours of sleep deprivation showed a branch specific reduction in dendritic spines compared with animals that slept after training (Yang et al. 2014). Similarly, in adolescent mice, sleep after learning promotes spine formation and the development of new synapses in the primary motor cortex, while sleep deprivation reduces the number of learning-induced new spines (Adler et al. 2021).

3. Sleep Deprivation Targets Gene Regulation at Multiple Levels

In addition to impacts on cellular signaling, sleep deprivation also affects gene regulation at multiple levels including changes in chromatin accessibility, transcription, post-transcriptional stability of mRNA and translation. In the next sections, we survey some of the potential mechanisms through which sleep deprivation targets gene regulation. Although research using invertebrate models and non-mammalian vertebrates have substantially contributed to our understanding of molecular signaling, learning and memory, and the functions of sleep; for clarity, we will focus on studies from mammalian models in our discussion of the processes impacted by sleep deprivation in gene regulation.

3.1. Transcription

3.1.1. Tissue Specific Effects of Sleep Deprivation on Transcription

Early research investigating molecular differences between sleep and wake focused on transcriptional approaches and changes in gene expression using microarrays or RNA sequencing to examine differential gene expression in various brain regions with sleep/wake cycles and sleep deprivation. Gene transcription varies with sleep/wake cycles and the subset of regulated genes varies depending upon the brain region or cell type examined (Cirelli et al. 2004; Cirelli & Tononi 2000; Mackiewicz et al. 2007). Estimates suggest that up to 10% of cortical transcripts are regulated with sleep/wake cycles (Scarpa et al. 2018). In the cortex, multiple time points of gene expression data and modeling of transcriptome dynamics suggests that sleep/wake cycles, in particular the length of time awake, drive changes in gene expression (Hor et al. 2019). These studies suggest that changes in gene expression with sleep/wake cycles are intertwined with homeostatic sleep regulation.

Sleep deprivation appears to impact gene transcription with a high degree of regional and cellular specificity within the brain as evidenced by the considerable differences observed between the hippocampus and the cortex. For more in-depth information on the specific genes affected by acute sleep deprivation in the cortex and hippocampus, we refer the reader to the following studies (Cirelli & Tononi 1998; Gaine et al. 2021; Terao et al. 2003; Thompson et al. 2010; Vecsey et al. 2012). While studies of gene changes with sleep deprivation are interesting, on a practical note, there is considerable interest in understanding how these changes in gene expression can be reversed. Most of the studies on recovery sleep have examined gene expression after acute sleep deprivation. In the hippocampus, three hours of recovery sleep is sufficient to return many genes to baseline expression levels following acute sleep deprivation (Gaine et al. 2021; Vecsey et al. 2012). However, there are some genes which maintain aberrant expression even after recovery sleep including the mRNA splicing factor Srsf7 , the G protein coupled serotonin receptor Htr1a, a potassium voltage gated channel subfamily member Kcnv1 and the transcription factor Elk1 (Gaine et al. 2021; Vecsey et al. 2012). The functions of these genes affecting transcription, mRNA splicing and neurotransmission could potentially mediate the persistent effects of acute sleep deprivation. In the cortex, acute sleep deprivation induces differential expression of hundreds of genes, including core circadian clock genes (Gerstner et al. 2016; Hor et al. 2019; Ingiosi et al. 2019). Interestingly, as in the hippocampus, the amount of recovery sleep necessary for gene expression to return to baseline levels varies depending upon the gene examined (Gerstner et al. 2016; Ingiosi et al. 2019). Sleep deprivation also induces differences in gene regulation in non-brain tissues. In humans, two nights of total sleep deprivation induces gene dysregulation in the blood with 212 genes significantly affected, with most genes downregulated (Uyhelji et al. 2018).

3.1.2. Sleep deprivation Induces Epigenetic Changes

Gene transcription depends on the accessibility of the chromatin allowing transcription factors and the transcriptional machinery to bind to the DNA. Epigenetic modifications affecting chromatin structure and DNA accessibility can be affected by sleep deprivation, thus influencing gene transcription (reviewed in (Gaine et al. 2018). DNA methylation in upstream gene regulatory sequences represses gene transcription. In rodents, acute sleep deprivation was found to induce methylation changes in the promoters of more than 136 genes in the cerebral cortex, with both increased and decreased methylation observed (Massart et al. 2014). In humans, one night of sleep deprivation increases DNA methylation in the promoter regions of core circadian clock genes CRY1 and PER1 as seen in adipose tissue (Cedernaes et al. 2015). Acute sleep deprivation has also been shown to change DNA methylation for genes involved in fatty acid metabolism in humans (Skuladottir et al. 2016). A broader study of acute sleep deprivation and DNA methylation in humans using whole blood samples found that sleep deprivation changed methylation in 269 genes (Nilsson et al. 2016). A derived modification of methylation, DNA hydroxymethylation, is also seen in the mammalian brain, with hydroxymethylation occurring more downstream in the body of the gene. Hydroxymethylation has been associated with bidirectional gene regulation, with enhanced expression seen for some genes (Kriaucionis & Heintz 2009; Li & Liu 2011; Meng et al. 2015). Sleep deprivation affects hydroxymethylation in the cortex to a greater degree than methylation, with changes in DNA hydroxymethylation observed for 4697 genes (Massart et al. 2014). Interestingly, in cell cultures, cAMP signaling has been shown to enhance DNA hydroxymethylation providing another potential link through which cAMP/PKA may influence gene transcription (Camarena et al. 2017). Thus, in animal models and humans, acute sleep deprivation alters patterns of DNA methylation affecting gene transcription.

Histone acetylation is another epigenetic modification that has been examined after sleep deprivation, given its importance in memory formation. The acetylation of histone proteins to increase chromatin accessibility constitutes an important regulatory step necessary for the increased gene expression seen with the molecular consolidation of long-term memory and synaptic plasticity (Levenson et al. 2004; McQuown et al. 2011). Histone deacetylase inhibitors, such as Trichostatin A, facilitate hippocampus dependent long-term memory when weak training paradigms are used or even to reverse memory impairments in disease and stress models (Hawk et al. 2011; Pandey et al. 2015; Vargas-Lopez et al. 2016). Given the effects of acute and chronic sleep deprivation on hippocampus dependent long-term memory, researchers investigating histone acetylation found that five hours acute sleep deprivation in mice significantly decreases two prominent histone acetylation marks, H3K9 and H4K12 and significantly increases the abundance of the histone deacetylase, HDAC2 (Wong et al. 2020). Histone deacetylation and HDAC2 activity are associated with memory impairments (reviewed in (Pao & Tsai 2022). Treatment with an FDA approved histone deacetylase inhibitor restored histone acetylation levels and blocked the impairments in LTP and long-term memory induced by sleep deprivation (Wong et al. 2020). Moreover, inhibition of histone deacetylase inhibitors during chronic sleep deprivation also facilitates the consolidation of long-term memory (Duan et al. 2016). In the cerebral cortex, the effects of acute sleep deprivation on gene expression were correlated with chromatin accessibility for 24% of the genes examined (Hor et al. 2019). Changes in chromatin accessibility are detected quickly with sleep deprivation and for the most part, returned to baseline levels by 12 hours following sleep deprivation (Hor et al. 2019). Thus, it appears that one mechanism through which sleep deprivation may impact the molecular consolidation of memory is through chromosome accessibility and histone acetylation regulation of gene transcription.

3.2. Sleep Deprivation Alters mRNA Processing

Following mRNA transcription, newly formed mRNA transcripts undergo maturation and processing. In the nucleus, newly formed mRNA transcripts are subject to splicing and polyadenylation prior to nuclear export. Within the cytoplasm, mRNA stability and lifespan may also be regulated as well as the propensity for translation (reviewed in (Radhakrishnan & Green 2016). In the following sections, we discuss potential mechanisms in post-transcriptional processing that may be affected by sleep deprivation between transcription and translation and the evidence that suggests these processes are targeted (Figure 2).

Figure 2. Post-transcriptional processes impacted by sleep deprivation.

Sleep deprivation affects mRNA splicing, polyadenylation, and miRNA mediated degradation. Sleep deprivation also decreases mTOR signaling resulting in 4EBP mediated inhibition of translation.

3.2.1. Alternative Splicing and Polyadenylation

The human genome is estimated to contain 24,000 protein coding genes resulting in more than 100,000 proteins due to 95% of genes giving rise to multiple transcripts (Keren et al. 2010; Kim et al. 2018; Pan et al. 2008). Alternative splicing facilitates neuronal function by increasing the number of mRNA transcripts arising from a single gene, consequently increasing protein diversity. Additionally, alternative splicing may alter mRNA translation through changes in the stability of mRNA transcripts or decay rates. In neurons, mRNA splicing patterns are dynamically regulated to promote neural plasticity and synaptic function (Vuong et al. 2016). Disease related changes in pre-mRNA splicing and splicing pathologies have been associated with metabolic disorders, cancers, and neurodegenerative diseases including Alzheimer’s disease (reviewed in (Daguenet et al. 2015; Kim et al. 2018; Li et al. 2021a). Circadian regulation of alternative splicing has been shown for multiple tissues (Marcheva et al. 2020; McGlincy et al. 2012), suggesting that sleep/wake cycles may also affect alternative splicing. While alternative splicing of mRNA transcripts following sleep deprivation has not been well studied, technological advances and increased resolution in genome wide RNA sequencing will undoubtedly further our understanding of alternative splicing and sleep in the next few years.

The regulation of mRNA stability occurs, at least in part, via polyadenylation, the addition of adenosine monophosphates to the 3’ terminal end of the RNA transcript that augments RNA stability and increases translation efficiency (reviewed in (Charlesworth et al. 2013). As our understanding of the mechanisms and functions of polyadenylation have grown, disease associated changes in gene expression due to mis-steps in polyadenylation and deadenylation have been identified, underscoring the importance of polyadenylation in gene regulation (Dharmalingam et al. 2022; Ivshina et al. 2014). Circadian rhythms in polyadenylation and deadenylation have been shown in neurons and astrocytes (Gendreau et al. 2018; Gerstner et al. 2012; Kojima et al. 2012; Parnell et al. 2021). Changes in polyadenylation have also been associated with long-term potentiation in the hippocampus (Fontes et al. 2017). Potentially, if sleep deprivation disrupts the circadian clock or core clock genes, then subsequent changes in polyadenylation may affect RNA stability and gene transcription. Interestingly, two genes that are downregulated by sleep deprivation in the hippocampus, Cirbp and one Rbm3 isoform (Gaine et al. 2021) have been implicated in circadian gene expression through the regulation of polyadenylation (Liu et al. 2013). Decreased expression of Cirpb or Rbm3 resulted in the shortening of the 3’UTR for more than 100 genes and subsequent decreased circadian rhythms for many genes (Liu et al. 2013). Additionally, sleep deprivation increases the expression of Srsf7 in the hippocampus, a gene recently reported to have expression inversely correlated with polyA tail length in the 3’ UTR (Gaine et al. 2021; Schwich et al. 2021). However, the regulation of RNA processing and polyadenylation by sleep/wake cycles appears complex and may vary with brain regions. In the cortex and hypothalamus, the expression of polyadenylation factor 2 (Cpsf2) decreases during sleep, suggesting that polyadenylation may decrease with sleep (Mackiewicz et al. 2007). In humans, two nights of sleep deprivation has been shown to decrease the mRNA levels in the blood of cytoplasmic polyadenylation element binding protein CPEBP4, a gene involved in mediating polyA elongation and translation (Uyhelji et al. 2018). Although additional research needs to be done, it is clear that regulation of polyadenylation to affect mRNA stability represents a potential mechanism through which sleep deprivation could post-transcriptionally impact gene regulation.

3.2.2. mRNA stability: Degradation through miRNA

An important mechanism through which gene expression may be regulated is through microRNA (miRNA) mediated degradation of mRNA. miRNA sequences are comprised of 18–25 nucleotides that bind to the 3’ untranslated region (3’ UTR) of a gene resulting in a region of double stranded RNA that is then subject to cleavage and degradation or translational repression (Iwakawa & Tomari 2015; Kim & Pak 2020). There are approximately 2500 miRNAs in the human genome (Alles et al. 2019; Huang et al. 2020); miRTarBase, release 9.0), with studies estimating that more than 60% of protein coding mRNA transcripts are targeted by miRNAs (Friedman et al. 2009). miRNAs are dynamically regulated through transcription and degradation. The half-life of miRNAs can vary, although half-life is usually measured in hours (reviewed in (Kim & Pak 2020). Functionally, miRNAs can be targeted to dendrites and play a prominent role in synaptic plasticity (Kindler & Kreienkamp 2012; Schratt 2009; Song 2020; Steward & Worley 2001).

The regulation of sleep also appears dependent in part on miRNA regulation of gene expression. Pharmacological studies in rats using targeted miRNA inhibitors found that inhibition of miRNA-138 in the cortex suppressed total sleep duration and slow wave sleep during NREM, in particular during the animal’s normal rest phase (Davis et al. 2012). However, the effect of miRNAs on sleep depends upon the miRNA examined and presumably the mRNA transcripts that it targets. Injection of a pre form miR-132 (mimetic), a miRNA known to be involved in memory, decreased NREM sleep and lowered sleep intensity, although REM sleep was increased (Davis et al. 2011). Circadian and/or regulation by sleep/wake cycles occurs for many miRNAs including miR-132 (Aten et al. 2018; Cheng et al. 2007; Davis et al. 2012; Davis et al. 2011). miRNAs have also been shown to play a prominent role in regulating entrainment of the circadian clock in mammals as well as functioning in translational control affecting the period length of the circadian clock (Alvarez-Saavedra et al. 2011; Cheng et al. 2007; Park et al. 2020; Zhou et al. 2021). Circadian regulation of miRNA expression has been observed in plants, fungi, invertebrates and vertebrate animals (reviewed in (Parnell et al. 2021).

Changes in miRNA expression with sleep deprivation could mediate mRNA degradation, thereby affecting gene specific translation. In rats, eight hours of sleep deprivation resulted in changes in 50 miRNAs (Davis et al. 2007). Interestingly, as with gene expression, brain region specific differences were seen in the direction of miRNA regulation (Davis et al. 2007). In the prefrontal and somatosensory cortices, more than 90% of the differentially regulated miRNAs were significantly decreased. However, in the hippocampus the directionality was reversed with the significant upregulation of 49 miRNAs, of which 37 were decreased in the cortex such as miR-132 (Davis et al. 2007). This suggests that miRNAs in the hippocampus play a contributing role in the decreased gene expression seen after sleep deprivation.

During memory consolidation, miRNA regulation is bidirectional and depends upon the specific miRNA and its targets. Downregulation of some miRNAs permits upregulated gene-specific protein translation; whereas, the upregulation of other miRNAs facilitates the degradation of mRNA transcripts for genes that act as limits in memory formation (reviewed in (Aksoy-Aksel et al. 2014; Fiorenza & Barco 2016; Wei et al. 2017). The impairments in long-term memory seen in REM sleep deprivation following training in the Morris water maze are associated with changes in the hippocampal expression of miRNAs known to be related to memory including miR-132, miR-182 and miR-124 (Karabulut et al. 2019). With chronic REM sleep deprivation after Morris water maze training, miRNA-191a and miRNA-155 were found to have increased expression in the hippocampus correlated with impairments in long-term memory (Mohammadipoor-Ghasemabad et al. 2019). Similar to the dysregulation of miRNAs seen with sleep deprivation in rodent studies, one night total sleep deprivation in humans significantly increases expression of at least two miRNAs measured in blood samples (Weigend et al. 2019). Furthermore, in a multi-national European study of children and adolescents, miRNA profiles from blood plasma samples differed significantly for short sleepers compared with normal sleepers (Iacomino et al. 2020). This lends further evidence to the hypothesis that sleep deprivation post-transcriptionally affects gene expression through changes in miRNA regulation of mRNA stability. Interestingly, sleep disorders that increase sleep including narcolepsy induce miRNA dysregulation (Holm et al. 2014)). Thus, in humans and other mammals, miRNA regulation of mRNA transcripts affects sleep, while sleep loss induces region specific changes in miRNA expression, potentially providing a mechanism for sleep deprivation induced cognitive impairments.

To better understand how miRNA regulation may be involved in gene expression changes following sleep deprivation, we analyzed recent acute sleep deprivation studies for the transcriptome and translatome, the subset of mRNA associated with ribosomes, of the hippocampus (Gaine et al. 2021; Lyons et al. 2020). To identify conserved binding motifs for miRNAs, we analyzed mRNA transcripts for genes that were significantly downregulated in the translatome following sleep deprivation, but were not transcriptionally regulated. We used the ENSEMBL 3’UTR sequences obtained with the BioMart utility (www.ensembl.org/biomart/martview) to search for conserved miRNA binding motifs in the 3’UTR of transcripts downregulated by sleep deprivation in the translatome of hippocampal excitatory neurons ((Lyons et al. 2020), GEO series accession GSE156925). Conserved motifs were identified using unbiased motif discovery with the MEME suite with motifs subsequently compared with the miRNA binding site database using Tomtom as detailed (https://github.com/YannVRB/Review-Sleep-Deprivation-Journal-of-Neurochemistry.git). We found two miRNAs, miR-7045–3p and miR-6971–5p, with a high predicted probability for downregulating multiple genes that were downregulated with sleep deprivation in the translatome (Figure 3). There were 5 genes that can be downregulated by both miRNAs suggesting that miRNA regulation is a strong possibility for downregulating these genes. Interestingly, miR-6971 abundance exhibits circadian cycles in the liver with peak expression in the early subjective night (Wang et al. 2016). However, no research has been done on these miRNAs with sleep deprivation or circadian regulation in the brain.

Figure 3. Prediction analysis of miRNAs associated with downregulation of gene expression after sleep deprivation.

Analysis of genes down regulated in the translatome, but not the transcriptome, revealed predicted 3’ UTR binding sites for two miRNAs, miR-7045–3p and miR-6971–5p.

4. Impact of Sleep Deprivation on Protein Synthesis

Although the above processes contribute to altering the pool of mRNA transcripts available for translation after sleep deprivation, recent studies have shown considerable differences between the effects of acute sleep deprivation on the transcriptome compared to the translatome, particularly in the hippocampus (Delorme et al. 2021a; Delorme et al. 2021b; Gaine et al. 2021; Lyons et al. 2020; Puentes-Mestril et al. 2021). The disparity between gene regulation evident at the level of the hippocampal transcriptome and regulation in the translatome after acute sleep deprivation (Gaine et al. 2021; Lyons et al. 2020) suggests that additional translational mechanisms may also be affected by sleep deprivation.

4.1. Sleep Deprivation Alters Protein Abundance

During deep sleep, protein synthesis has been shown to increase in non-human primates in many regions of the brain, although variation occurs between brain regions (Nakanishi et al. 1997). In rodent models, rates of protein synthesis have also been shown to increase across the brain during slow wave sleep, although the same correlation was not observed during REM sleep (Ramm & Smith 1990). These studies suggest that areas particularly vulnerable to sleep deprivation may show altered rates of protein synthesis. Indeed, acute sleep deprivation has been found to decrease protein synthesis as well as altering mRNA stability (Lamon et al. 2021; Tudor et al. 2016). As with gene transcription, there are regional specific effects of sleep deprivation on the proteome. In mice, six hours of acute sleep deprivation decreases polyribosomal mRNA complexes in cortical tissue, presumably decreasing protein synthesis (Naidoo et al. 2005). More direct assessments of in vivo protein synthesis in brain tissue have been done using the incorporation of puromycin into growing polypeptide chains and subsequent quantification through western blotting, a technique known as ribopuromycylation or the SUnSET assay (Schmidt et al. 2009). Five hours of sleep deprivation was found to decrease in vivo protein synthesis by approximately 50% in the hippocampus (Tudor et al. 2016). In eight hour sleep deprivation experiments with subsequent analysis of synaptic proteins in the cerebral cortex of Wistar rats, the protein abundance of 78 proteins was altered after sleep deprivation (Gulyassy et al. 2022). In contrast to what has been shown for the hippocampus, almost 70% of cortical proteins were upregulated by sleep deprivation including upregulation of metabotropic GABA receptors. Interestingly this proteomic data suggested suppression of vesicle exocytosis and impaired endocytosis (Gulyassy et al. 2022). In the mouse forebrain as a whole, six hours of sleep deprivation altered the synaptic proteome and protein abundance (Bruning et al. 2019; Noya et al. 2019).

In humans, the effect of sleep deprivation on protein synthesis has been primarily investigated through studies of skeletal muscle mass, which is primarily dependent on protein synthesis (de Boer et al. 2007). In young adults, male and female, acute sleep deprivation decreases protein synthesis in muscle by 18% as measured in pre- and post- sleep deprivation muscle biopsies using ¹³C labeled phenylalanine (Lamon et al. 2021). However, in the same study, markers of protein degradation were not affected by acute sleep deprivation (Lamon et al. 2021). This research suggests that sleep deprivation targets protein synthesis, rather than protein degradation. Five days of partial sleep restriction in humans has also been shown to reduce myofibrillar protein synthesis (Saner et al. 2020). Similarly, four days of REM sleep deprivation in rats also caused decreased muscle mass (Dattilo et al. 2012; Monico-Neto et al. 2015). These studies raise the question of whether protein synthesis overall is affected or whether specific subsets of genes are targeted.

4.2. Sleep deprivation targets the initiation of translation

To better understand how sleep deprivation affects protein abundance and protein synthesis, it is necessary to look at the molecular mechanisms that underlie translation. Prior to the association of mRNA transcripts with ribosomes, upstream signaling through the target of rapamycin (mTOR) pathway provides a crucial level of regulation for cap-dependent translation. The mTOR complex phosphorylates two downstream targets, S6 kinase and eukaryotic initiation factor 4e binding protein (4EBP). Phosphorylation induces the activation of the serine/threonine S6 kinase resulting in the phosphorylation of downstream targets including ribosomal protein S6 (RPS6), a component of the 40S small ribosome subunit (reviewed in (Magnuson et al. 2012; Yi et al. 2021). RPS6 phosphorylation is dynamically regulated and can occur through multiple pathways which either activate S6 kinase or induce direct phosphorylation of RPS6 (Hutchinson et al. 2011; Yi et al. 2021). Phosphorylation of RPS6 affects mRNA binding to the ribosome, thus regulating translation. In a second pathway affecting translation, S6 kinase phosphorylates BMAL1 facilitating the interaction of BMAL1 with the translation initiation complex (Lipton et al. 2015; Marti et al. 2017). In the hippocampus, no change has been observed in the phosphorylation of S6 kinase after 5 hours of sleep deprivation (Tudor et al. 2016), suggesting that this is not a major pathway through which sleep deprivation impacts protein synthesis.

The second branch of the mTOR pathway is the phosphorylation of 4EBP2, the most prominent 4EBP in the brain (Banko et al. 2005). 4EBP2 regulates the initiation of cap-dependent translation by regulating the association of the eukaryotic initiation factors needed for the cap-binding protein complex. In its hypophosphorylated form, 4EBP2 binds to the initiation factor eIF4E preventing eIF4G from binding (Figure 2). mTOR phosphorylation of 4EBP causes the dissociation of 4EBP from the translation initiation factor eIF4E, thus allowing the subsequent association of eIF4E and eIF4G (reviewed in (Thoreen et al. 2012). eIF4E and eIF4G bind to eIF4A forming the complex that binds to the 5’cap of the mRNA transcript, allowing the initiation of translation (Ma & Blenis 2009). Acute sleep deprivation decreases activation of the mTORC complex resulting in decreased levels of phosphorylated 4EBP2 in the hippocampus (Tudor et al. 2016). Consequently, a decreased association of eIF4E and eIF4G is also seen in the hippocampus after sleep deprivation (Tudor et al. 2016). Thus, the mTOR-4EBP2 pathway appears to be a major target of sleep deprivation to decrease protein synthesis in the hippocampus. Interestingly, chronic sleep deprivation experiments in rats also found that sleep deprivation increased levels of hypophosphorylated (active) 4EBP and muscle atrophy (de Sa Souza et al. 2016), suggesting that the impact of sleep deprivation on protein synthesis through the mTOR-4EBP pathway affects multiple tissues.

eIF4E phosphorylation also promotes the initiation of translation (reviewed in (Amorim et al. 2018). Acute sleep deprivation in rats decreases levels of eIF4E phosphorylation in the dentate gyrus (Gronli et al. 2012). Similarly, two nights of total sleep deprivation decreases eiF4E expression, supporting the hypothesis that sleep deprivation decreases protein synthesis through suppression of translation initiation (Uyhelji et al. 2018). The initiation of translation is complex involving numerous initiation factors (reviewed in (Komar & Merrick 2020). The initiation of translation may also be regulated through an eIF2α dependent mechanism. eIF2α dependent translation has been shown important for local mRNA translation in dendrites and synaptic plasticity (reviewed in (Oliveira & Klann 2022). Dysregulation and accumulation of eIF2α phosphorylation has been observed in models of Alzheimer’s disease pathology and neurodegeneration (Chang et al. 2002a; Chang et al. 2002b). Six hours of acute sleep deprivation was found to double the level of phosphorylated eIF2α in the mouse cortex (Naidoo et al. 2005). However, in the hippocampus, acute sleep deprivation does not appear to affect eIF2α phosphorylation levels (Tudor et al. 2016). Thus, as with transcription, the mechanism through which sleep deprivation affects the initiation of translation also appears to vary depending upon region of the brain. Although much of the research to date has focused on the initiation of translation, sleep deprivation may also interfere with peptide elongation by impacting translation elongation factors. Phosphorylation of the eukaryotic elongation factor eEF2 inhibits the elongation of translation (reviewed in (Ballard et al. 2021). Acute sleep deprivation has been shown to increase eEF2 phosphorylation in the mouse hippocampus and prefrontal cortex (Gronli et al. 2012). Thus, in addition to the impacts on translation mediated by 4EBP, sleep deprivation also affects the abundance and modification of translation initiation factors.

4.3. Analysis of the translatome

In order to more specifically understand the genes and pathways through which sleep deprivation affects neural function and memory, it is important to analyze gene regulation at the level of the translatome. Development of new techniques for the isolation of mRNA transcripts associated with ribosomes has facilitated this type of analysis. Translating ribosome affinity purification and RNA sequencing (TRAP-Seq) with transgenic mouse lines or viral delivery of transgenes has enabled cell type specific isolation of the translatome (Dougherty 2017; Heiman et al. 2008; Nectow et al. 2017; Sanz et al. 2019; Sanz et al. 2009). Combination of TRAP-Seq approaches with specificity for ribosomes containing phosphorylated ribosomal protein S6 also been used for translatome profiling of activated neurons (Jiang et al. 2015; Knight et al. 2012). Research using TRAP-Seq techniques has been used to investigate gene regulation in hippocampus excitatory neurons following acute sleep deprivation (Delorme et al. 2021a; Lyons et al. 2020). These studies employed different periods and methods for sleep deprivation with Delorme and colleagues using 3 hours of sleep deprivation that included direct disturbance of the animal, while Lyons and colleagues used 5 hours of sleep deprivation that included only cage tapping or shaking (Delorme et al. 2021a; Lyons et al. 2020). Given these different approaches, comparative analysis of the resulting data sets may provide insight into the temporal profile of the changes in the translatome that occur during acute sleep deprivation. We compared the differential gene expression in hippocampal excitatory neurons from studies using 3 hours sleep deprivation (Delorme et al. 2021a) or 5 hours of sleep deprivation (Lyons et al. 2020) finding a common set of 51 differentially expressed genes (Figure 4a). The overlapping data set included transcription factors, members of the AMPK signaling pathway, and genes involved in RNA methylation, which suggests that additional genes may be regulated at later time points. Interestingly, there were a large number of genes that were exclusively regulated in one or the other data set. Whether these differences arose due to temporal regulation of the translatome during sleep deprivation or due to the different methods of sleep deprivation and prior housing remains an open question that warrants further investigation.

Figure 4. Comparative analysis of the translatome in hippocampal excitatory neurons after sleep deprivation and long-term potentiation.

a) Comparison of differentially expressed genes after 3 hours of sleep deprivation and 5 hours of sleep deprivation. b) Comparison of differentially expressed genes in the translatome after 5 hours of sleep deprivation with changes observed after LTP in the hippocampus. Genes highlighted in yellow have opposite directions of regulation between sleep deprivation and LTP.

Comparison of differentially regulated genes in the translatomes after sleep deprivation and synaptic plasticity (long-term potentiation) can potentially provide insight into specific mechanisms underlying the inhibition of memory consolidation by acute sleep deprivation. We compared differentially expressed genes in translatome of the excitatory neurons of the hippocampus after acute sleep deprivation (Lyons et al. 2020) with previously identified differentially expressed genes in the excitatory neurons of the CA1 and dentate gyrus of the hippocampus after the induction of long-term potentiation (Chen et al. 2017) in Figure 4b. As may be expected given the different effects of sleep deprivation and LTP on neural plasticity, the bulk of gene regulation in the translatome was exclusive to one data set or the other, although there were a few genes commonly regulated in both data sets that exhibited different directions of regulation. These results suggest that the changes in gene regulation in excitatory neurons induced by sleep deprivation constitute more than just opposite regulation of genes involved in synaptic plasticity. Five hours of acute sleep deprivation has also been shown to activate somatostatin expressing inhibitory neurons in the hippocampus, with the activation of these neurons appearing to be a mechanism through which sleep deprivation inhibits long-term memory (Delorme et al. 2021b). Thus, sleep deprivation affects gene regulation in both excitatory and inhibitory neurons of the hippocampus to inhibit synaptic plasticity and long-term memory consolidation.

5. Future Directions

The past twenty years has seen an explosion of research examining the impacts of sleep deprivation on cognitive performance, health and lifespan. Despite the evidence of adverse health impacts, acute and chronic sleep deprivation continue to be a significant public health problem in the United States and worldwide affecting children, teenagers and adults. The complexity of regional and cell-type specific effects in the brain following sleep deprivation make it clear that more research needs to be done to fully understand the molecular and cellular impacts of sleep deprivation on the brain. Although this is not an all-inclusive list, we identified several areas of research warranting further research for which increased understanding has the potential to enable advances for human health.

5.1. Molecular Changes during Recovery Sleep

The homeostatic need for sleep results in compensatory rebound sleep following sleep deprivation. The time needed for recovery following acute or prolonged sleep restriction can vary depending upon the physiological process or genes examined. Rebound sleep combined with a full night’s sleep following acute sleep deprivation is often insufficient to mitigate the adverse impacts of sleep deprivation as seen from studies across phyla. For example, in the marine mollusk Aplysia, the effects of acute sleep deprivation on short and long-term learning and memory persist for more than 24 hours (Krishnan et al. 2016a). In Drosophila, researchers using a model of Parkinson’s disease found that one night of acute sleep deprivation induced persistent deficits in short-term memory that lasted at least three days (Seugnet et al. 2009). In humans, some studies have reported that one long night of recovery sleep is sufficient to return performance scores to baseline following one night of acute sleep deprivation (Lamond et al. 2007). However, individuals with shorter periods of recovery sleep (6 or 8 hours) following acute sleep deprivation still exhibit performance deficits in neurobehavioral tasks (Lamond et al. 2007; Sallinen et al. 2004). For more challenging neurobehavioral tasks, two to three nights of recovery sleep were needed to restore performance following sleep deprivation (Ikegami et al. 2009). Recent research analyzing hippocampal connectivity and episodic memory found that two nights of recovery sleep following total sleep deprivation were sufficient to restore hippocampal connections, but not normal episodic memory (Chai et al. 2020). A few studies have investigated cellular changes during recovery following chronic sleep deprivation in rodent models and in humans. In mice, changes in neuroinflammation, microglial activation and neuronal degeneration in the hippocampus have been observed three weeks after seven days of REM sleep deprivation (Yin et al. 2017). In rats, three weeks of recovery sleep is insufficient to completely restore changes in the dendritic arborization of hippocampal CA3 that occur with three weeks of chronic sleep deprivation (Konakanchi et al. 2022). Thus, recovery from either acute sleep deprivation or chronic sleep restriction appears to be a lengthier process than previously envisioned.

The human brain displays some ability to adapt to mild chronic sleep loss to stabilize performance and attention at a reduced level; however, adaptation has not been observed during a seven day paradigm of severe sleep restriction with participants showing continuing declines in performance (Belenky et al. 2003). With mild sleep restriction (5 to 7 hours in bed), three days of recovery sleep was insufficient to return performance in psychomotor vigilance tasks to baseline levels restriction (Belenky et al. 2003). More recent research has shown that 7 days of recovery sleep following a 30% sleep reduction paradigm for ten days was insufficient to restore baseline performance on the Stroop test (Ochab et al. 2021). In individuals subjected to one week of sleep restriction followed by two night of recovery sleep, performance deficits in psychomotor vigilance tests persisted (Pejovic et al. 2013). Thus, for the millions of teenagers and adults in which chronic sleep loss occurs during the week, increased sleep on the weekends appears to be insufficient to restore cognitive performance to normal levels. Furthermore, researchers found that weekend recovery sleep was insufficient to account for the metabolic disturbances associated with sleep deprivation including changes in insulin sensitivity in healthy young adults (Depner et al. 2019). Interestingly, two days of recovery sleep following one week of mild sleep restriction does appear to normalize cortisol levels and cytokines (Pejovic et al. 2013). As a caveat, a recent study of one night total sleep deprivation followed by two nights of recovery sleep determined that although cortisol levels changed with sleep deprivation, they were not a marker of resilience or vulnerability to sleep deprivation based on multiple neurobehavioral tests (Yamazaki et al. 2021).

Molecular studies of rebound sleep have focused primarily on the identification of transcripts and proteins that changed following acute sleep deprivation and then returned to baseline within 2 – 3 hours of recovery sleep. For example, following 6 h sleep deprivation in mice, the phosphorylated glial protein NDRG2, a marker of sleep need, returns to baseline phosphorylation levels within 2 hours of recovery sleep (Suzuki et al. 2013). In transcriptome studies of the hippocampus following acute sleep deprivation, most genes, including immediate early genes and transcription factors examined return to baseline expression levels with three hours of recovery sleep with some notable exceptions (Gaine et al. 2021; Vecsey et al. 2012). In contrast, the splicing factor Srsf7 was found to reverse directions during recovery sleep supporting the hypothesis that sleep deprivation changes mRNA splicing patterns and recovery sleep may reduce alternate splicing (Gaine et al. 2021). However, studies for longer periods of recovery sleep have not been performed in the hippocampus. In the cortex, 2.5 h of recovery sleep also normalizes many of sleep deprivation associated changes in gene expression; however, some genes required up to 6 hours to return to basal levels (Gerstner et al. 2016). At the proteomic level, 78 synaptic proteins in the cerebral cortex were changed after 8 hours of sleep deprivation with 56 upregulated, while 39 synaptic proteins were altered following 16 hours of recovery sleep with 15 upregulated (Gulyassy et al. 2022). However, no intermediate time points in recovery sleep were examined. With the limited number of studies done examining the time course of gene regulation and protein synthesis during recovery from sleep deprivation, higher temporal resolution molecular studies of multiple brain regions need to be done to fully understand the recovery patterns in gene regulation after sleep deprivation.

5.2. Long-term Impact of Chronic Sleep Disturbances

In a recent study commissioned by the American Academy of Sleep Medicine, sleep disturbances in U.S. adults have increased during the Covid-19 pandemic with 56% of individuals surveyed reporting chronic sleep disturbances (AASM Sleep Prioritization Study March 2021). Future studies on the impacts of sleep loss need to include additional paradigm complexity including repeated bouts of acute sleep deprivation, the persistence of sleep fragmentation arising from sleep deprivation, chronic sleep loss, and the combinatorial effects of sleep deprivation and circadian dysfunction on memory and performance. Importantly, researchers should also consider the long-term effects of chronic sleep disturbances across age groups.

Models of chronic sleep restriction in rodents have shown that sleep restriction in adolescent and young adult animals results in long-lasting impairments in learning and memory. Chronic sleep restriction in pregnant dams results in impairments in hippocampus-dependent memory in the offspring including reductions in spatial memory in the Morris water maze, long-term potentiation and excitatory synaptic activity (Peng et al. 2016). Ten days of chronic sleep loss in adolescent rats leads to hippocampus-dependent memory impairments that are still evident after four weeks of recovery, although hippocampus independent memory was not similarly impacted (Howard & Hunter 2019). Chronic sleep restriction in adolescent mice also results in changes in synaptic structure impairing synaptic reorganization (Nagai et al. 2021). Recent research found that 12 weeks of chronic sleep restriction in young adult wild type mice induced persistent adverse impacts including a decrease in CA1 neurons and a reduction in hippocampal volume as well as increased tau phosphorylation twelve months later (Owen et al. 2021). Given the potential life-long consequences of chronic sleep loss in adolescents and young adults and the increasing problem of sleep deprivation in children and teenagers, more research needs to done at the molecular and cellular level to understand chronic sleep loss.

5.3. Sleep Disturbances with Aging

To date, most of the studies on sleep deprivation have been done using young animals or in young adults. However, changes in sleep patterns with aging or stress may affect memory and cognitive function. More research needs to be done on the molecular impacts of sleep disturbances with aging, particularly given the links between sleep disturbances and the increased progression of age-related diseases, including neurodegenerative diseases such as Alzheimer’s disease. Potentially, behavioral therapies or lifestyle changes may assist in ameliorating the cognitive impacts of sleep changes that occur with age. For example, exercise has been shown to improve memory and psychomotor performance in middle aged and elderly individuals, even increasing hippocampal volume (De la Rosa et al. 2019; Erickson et al. 2011). In rodent models, the intensity and duration of exercise appears to correlate with increased BDNF activation in the hippocampus and memory performance, although higher intensity exercise is needed for cortical activation (Aguiar et al. 2008; Cefis et al. 2019).

Disrupted sleep patterns and circadian rhythm disorders have been associated with an increased risk and incidence of Alzheimer’s disease and neurodegeneration in animal models and in humans. In humans, longitudinal studies of sleep in individuals starting in middle-age have shown that decreased sleep increases the risk of dementia by 30%, particularly Alzheimer’s disease (Sabia et al. 2021; Shi et al. 2018). At the molecular level, 12 weeks of chronic sleep restriction in a mouse tauopathy model accelerated the progression of pathologies including motor impairments and decreased neural volume in multiple brain regions (Zhu et al. 2018). Chronic sleep fragmentation has also been found to increase the deposition and accumulation of amyloid β with the severity of sleep fragmentation correlated to amyloid β deposition (Minakawa et al. 2017). Similarly, studies in humans have shown increases in Aβ plaque formation in individuals with poor sleep (Spira et al. 2014; Spira et al. 2013). Moreover, increased Aβ aggregation also increases sleep disturbances exacerbating disease progression (reviewed in (Chen et al. 2021; Sun et al. 2022). Sleep deprivation may, in part, increase the risk of neurological diseases through decreased clearing of metabolic waste via the glymphatic system (reviewed in (Achariyar et al. 2016; Bishir et al. 2020; Rasmussen et al. 2018). There have been many recent reviews discussing the interactions between sleep and circadian disorders and neurodegenerative diseases, particularly Alzheimer’s disease. For more information, (Vanderheyden et al. 2018; Wang & Holtzman 2020; Wu et al. 2019).

Although Alzheimer’s disease is the most studied neurodegenerative disease in regard to sleep issues, sleep disorders also are associated with other dementias and neurodegenerative diseases. Multiple studies have shown that individuals with Lewy body dementia experience even a greater percentage of sleep disruptions than Alzheimer’s patients, up to twice as much (Cagnin et al. 2017; Koren et al. 2023). Systematic reviews have found that approximately 90% of Lewy body dementia patients have at least one sleep disorder (Elder et al. 2022). Focusing on specific aspects of sleep disorders in dementia has allowed researchers to identify specific sleep disturbances associated with these conditions. For example, researchers have found that non-REM sleep with hypertonia may serve as a diagnostic biomarker for Lewy body dementia and Parkinson’s disease with dementia (Levendowski et al. 2022). Sleep deprivation and sleep disturbances have also been associated with Parkinson’s disease, with more than 60% of patients experiencing sleep issues (Louter et al. 2012; Maggi et al. 2023). In mice carrying a mutation in leucine-rich repeat kinase (Lrrk2), an autosomal dominant mutation associated with Parkinson’s disease, chronic sleep restriction increased the progression and severity of Parkinson pathologies (Liu et al. 2022b). Although the impact of specific gene changes that occur with sleep deprivation and disturbances in neurodegenerative diseases remains unknown, it is clear that more research needs done to understand the molecular mechanisms through which sleep disturbances increase the incidence or progression of neurodegenerative diseases. Given the increasing aging population in many countries including the U.S., more molecular and cellular research needs to be done to understand the effect of sleep disruption on age-related diseases and to identify potential mechanisms to alleviate these conditions.

5.4. Functions of Sleep and Systems Interactions

In addition to the connections between sleep and neuronal function, the functions of sleep are widespread interacting with multiple physiological systems including development, metabolism, the cardiovascular system, the immune system and muscle growth (reviewed in (Mullington et al. 2009; Tobaldini et al. 2017; Zielinski et al. 2016). Sleep deprivation damages multiple organs through oxidative stress and inflammation, including damage to the thyroid gland, liver, heart, and kidneys (Li et al. 2021b; Periasamy et al. 2015). In humans, sleep deprivation also affects immune system responses increasing inflammation (reviewed in (Mullington et al. 2009; Simpson & Dinges 2007). Systemic inflammatory markers are also increased following sleep deprivation in animal models as well as the activation of microglia (Liu et al. 2022a; Periasamy et al. 2015; Poroyko et al. 2016; Zhang et al. 2021). Thus, it is possible that indirect effects of sleep deprivation on other systems may affect neural function. Undoubtedly, future research will focus on system wide interactions of sleep.

5.5. Conclusion

Although sleep is generally thought of as a rhythmic daily behavior, it is clear that both acute and chronic sleep deprivation can induce changes in molecular signaling and gene expression that extend past the duration of the sleep loss. It is critical to understand the detailed molecular impacts of sleep deprivation as changes in gene expression underlie long-term memory and contribute to the increased incidence and exacerbation of neurodegenerative diseases. As the individual and societal factors contributing to sleep deprivation and sleep disturbances appear unlikely to abate, more research needs to be done to identify pathways and gene targets through which the impacts of sleep deprivation may be mitigated. The benefits of research on gene regulation and protein synthesis associated with sleep disturbances will also provide vital insight into gene dysregulation and identify ways in which to these molecular changes might be reversed. Increasing resilience to the adverse effects of sleep loss will have significant impacts on the human, societal and economic costs of sleep deprivation and sleep disturbances.

List of Abbreviations:

4EBP

Eukaryotic initiation factor 4e binding protein

BDNF

Brain derived neurotrophic factor

cAMP

Cyclic AMP, adenosine 3’,5’-cyclic monophosphate

Cpebp4

Cytoplasmic polyadenylation element binding protein

Cpsf2

Cleavage and polyadenylation factor 2

CREB

cAMP response binding element

CRY

Cryptochrome

eIF

Eukaryotic initiation factor

Elk1

ETS like Transcription Factor 1

ERK

Extracellular signal-regulated kinase

fMRI

Functional magnetic resonance imaging

Hrt1a

5-hydroxytryptamine receptor 1a

Kcnv1

Potassium voltage-gated channel subfamily V member 1

LTP

Long-term potentiation

miRNA

Micro RNA

mRNA

Messenger RNA

mTOR

Mammalian Target of Rapamycin

NREM

Non rapid eye movement sleep

pCREB

phosphorylated cyclic AMP response binding element

PER1

Period circadian protein homolog 1

PKA

Protein kinase A

REM

Rapid eye movement sleep

RNA

Ribonucleic acid

RPS6

Ribosomal protein S6

Srsf7

Serine and arginine rich splicing factor 7

TRAP-Seq

Translating ribosome affinity purification and RNA sequencing

UTR

Untranslated Region

Data Availability Statement

Data analyzed in the current study is available at GEO series accession GSE156925 with the analysis pipeline at https://github.com/YannVRB/Review-Sleep-Deprivation-Journal-of-Neurochemistry.git.