Slipping into deep sleep frees the brain to clear waste, including Aβ, in Alzheimer’s disease. Does the same happen to α-synuclein in Parkinson’s disease? Yes, according to researchers led by Daniela Noain and Christian Baumann at the University of Zurich in the December 8 Science Translational Medicine. Synucleinopathy mice that got deeper slow-wave sleep cleared α-synuclein aggregates, possibly through better glymphatic function and upregulated proteostasis. In contrast, sleep-deprived mice accumulated more α-synuclein and slowed down homeostatic processes. This study provides the first evidence of how sleep quality influences protein pathology in PD.

  • The narcolepsy drug sodium oxybate deepened slow waves in snoozing mice.
  • Well-rested mice had more fluid-shuttling aquaporin-4 in their perivascular spaces.
  • Deeper slow-wave sleep upregulated proteins involved in proteostasis.

“While we cannot yet say for sure that [optimizing sleep] will slow progression of synuclein pathology, there is now a rationale for believing that it might,” Andrew Lim, University of Toronto, wrote to Alzforum (full comment below). The findings excited Simon Schreiner at U Zurich, who was not a part of the current study. “They perfectly fit the bidirectional relationship of sleep with neurodegenerative diseases,” Schreiner told Alzforum.

People with AD often snooze intermittently, missing out on the deeper and more restorative slow-wave sleep during which waste removal ramps up in the brain (Nov 2019 news; Dec 2017 conference news). Disrupting SWS allows amyloid plaques and tau tangles to aggregate, which further fragments sleep (Jun 2015 news; Jan 2019 news; Jan 2019 news). 

Regarding sleep disturbances in PD, rapid eye movement (REM) sleep behavior disorder has been known for years (Sep 2010 news). More recently, scientists have begun to appreciate that SWS abnormalities in non-REM sleep also plague people with Parkinson’s (Schreiner et al., 2021). Might their SWS disturbance be similarly connected, at a molecular level, to PD's main protein aggregates and their clearance?

To find out, first author Marta Morawska turned to mice deficient in the dopamine transport protein vesicular monoamine transporter 2. Without VMAT2, dopamine builds up and damages neurons, triggering α-synuclein aggregation, loss of motor function, and sleep disturbances (Taylor et al., 2011). The scientists implanted an electroencephalography/electromyography device into the skulls of 5-month-old mice to track their sleep over 24 hours. Indeed, VMAT2-deficient animals spent more time awake, with less REM and non-REM sleep, than their wild-type siblings.

How about old mice? Aged VMAT2-deficient mice did not tolerate the EEG/EMG implantation procedure, so the scientists were unable to analyze their sleep. Instead, they did so in 14-month-old, wild-type mice. The scientists either sedated them with the narcolepsy drug sodium oxybate or kept them awake by placing them on a small platform over water for 16 hours. During 24 hours of EEG/EMG recording under each condition, mice that took sodium oxybate had more slow waves during non-REM sleep, whereas sleep-deprived animals had shallower waves and more fragmented non-REM sleep.

Could getting more restful SWS over an extended period affect synuclein aggregation? The scientists fed 14-month-old VMAT2-deficient and wild-type mice sodium oxybate, or placed them on the over-water platform every day for four months, then labeled prefrontal cortex slices with antibodies against α-synuclein and phosphorylated α-synuclein. Well-rested mice had less phosphorylated synuclein and fewer aggregates than controls, while the opposite was true in the sleepless mice (see image below).

Likewise, researchers fed sodium oxybate to 5.5-month-old A53T mice, which carry mutant human α-synuclein and develop Lewy-body-like synuclein aggregates. The drug boosted their clearance so well that western blots of their midbrain tissue looked almost like those from wild-type. “I thought it was fascinating that sleep altered pathology so dramatically in mice genetically destined to accumulate synuclein,” Schreiner told Alzforum.

Bye-bye Synuclein? Compared to control VMAT2-deficient mice (left), mice that caught more slow-wave sleep (middle) had somewhat less phosphorylated synuclein (top) and fewer synuclein aggregates (bottom) in their prefrontal cortices. Sleep-deprived mice (right) had more of both. [Courtesy of Morawska et al., Science Translational Medicine, 2021.]

To get a sense of how the brain might be clearing aggregated α-synuclein, the scientists searched immunostained cortical slices for phosphorylated synuclein co-localized with the proteasome marker ubiquitin or the endosome/lysosome marker LC3B, and for astrocytes co-localized with the glymphatic protein aquaporin-4. To their eyes, only the latter differed with sleep. A bit more aquaporin-4 overlapped with astrocytes in sodium oxybate-treated mice than in controls, while the protein remained unchanged in sleep-deprived animals. A part of the glymphatic waste removal system, aquaporin-4 water channels, when located at astrocytic end-feet surrounding blood vessels, aid interstitial fluid movement (Aug 2012 news; Dec 2016 news). 

Commentators hesitated to credit glymphatics for the cleanup. Erik Musiek, Washington University in St. Louis, captured their sentiment. “The [aquaporin-4 staining] seemed quite modest and not enough to conclude anything about glymphatic function,” he wrote to Alzforum (full comment below).

To get an unbiased sense of protein changes driven by sleep, Morawska ran mass spec on cortical samples from VMAT2-deficient mice. Compared to normal slumber, sleeping better or worse changed the mice's expression of more than 900 proteins in 150 processes. Broadly speaking, sound sleep upregulated homeostasis, while sleep deprivation downregulated it (see image below).

Snoozing to Homeostasis. Compared to mice that slept normally (top), mice that got more SWS (middle) had enhanced protein folding (1), breakdown (2), dephosphorylation (3), and release (4), as well as more aquaporin-4 (parallel blue lines) at perivascular astrocyte end-feet (5). The authors believe that, together, all this reduced synuclein aggregation and promoted its clearance. The opposite was true in sleep-deprived mice (bottom). [Courtesy of Morawska et al., Science Translational Medicine, 2021.]

Altering sleep in mice may not directly translate to people, as humans and rodents have different stages of sleep (Matsumoto and Tsunematsu, 2021). The scientists also do not know if sodium oxybate affects the neuropathology of people with PD.

Even so, some clinicians have reported that the drug improves sleep in individual cases and are testing this in three small trials (Liebenthal et al., 2016). In an open-label study of 30 people with PD who took sodium oxybate for six weeks, average time spent in SWS reportedly doubled and excessive daytime sleepiness waned (Ondo et al., 2008). A randomized, cross-over Phase 2 trial led by Baumann showed much the same in 12 people with PD (Büchele et al., 2018). An ongoing study at Stanford is testing an eight-week course of sodium oxybate treatment in 24 people with REM sleep behavior disorder, often a prodrome of Parkinson's or another synucleinopathy.

Schreiner cautioned that sodium oxybate is a tricky drug to take, especially long-term. “There is a complicated titration to find the right initial dose for each person, and they have to take it twice daily—before bed and in the middle of the night—for optimal effectiveness,” he said. Lim agreed, noting the drug’s propensity to cause adverse events in older adults.

Indeed, the authors themselves are eschewing pharmacological sleep interventions in their next study. Rather, Noain said they will use auditory stimulation by playing certain tones during slow-wave sleep to try to specifically enhance or diminish those brain waves in mouse models of AD and PD (Moreira et al., 2021).—Chelsea Weidman Burke


  1. Morawska et al. have expanded upon previous studies showing that sleep enhancement/deprivation can influence Aβ and tau deposition to now show that the same holds true for α-synuclein. They have taken care to characterize the effects of sodium oxybate on slow-wave sleep in mice, and use this intervention to enhance SWS in two different models of α-synucleinopathy, both of which are major strengths.

    They also added a sleep-deprivation arm using the platform-over-water method. In general, sleep deprivation increases synuclein aggregation, while SWS enhancement mitigates it. It is difficult to compare the sleep-deprivation and -enhancement methods directly, as one is pharmacological (oxybate) and the other behavioral, and potentially stressful. However, this relationship is similar to that seen with other pathogenic proteins.

    Mechanistically, there are some interesting leads but no very specific conclusions at this point. There is some change in Aqp4 staining, though the effect seemed quite modest and not really enough in my opinion to conclude anything about glymphatic function. Proteomics revealed a signature of increased lysosomal function, ubiquitin function, protein folding/breakdown with sleep enhancement and a reciprocal effect in deprivation. These findings are consistent with other studies, suggesting that sleep regulates proteostasis broadly.

    The findings also dovetail nicely with previous work from Brendan Lucey, David Holtzman, and colleagues showing increased CSF α-synuclein in humans after sleep deprivation (see Holth et al., 2019). Morawska et al. have presented an important observation for the Parkinson's disease field, as well as for dementia with Lewy bodies research. This should open the door for further study of sleep in these conditions and possible therapeutic use of sleep-modifying agents.


    . The sleep-wake cycle regulates brain interstitial fluid tau in mice and CSF tau in humans. Science. 2019 Feb 22;363(6429):880-884. Epub 2019 Jan 24 PubMed.

  2. This is a really exciting paper. It provides evidence that, in at least some model systems for synucleinopathy, sleep manipulations can in principle affect the accumulation of synuclein neuropathology. As such, it does for synucleinopathies what previous papers by Kang and colleagues (Kang et al., 2009) and Holth and colleagues (Holth et al., 2019) did for amyloid and tau, respectively.

    From a mechanistic perspective, I think important questions remain, including to what degree these effects are driven by neuronal activity/production versus extracellular clearance versus modulation of proteostatic mechanisms. 

    From a translational perspective, the extent to which these mechanisms are important in human synucleinopathies like Parkinson’s disease remains a completely open question. There will be several challenges in answering this question. 

    First, we will need appropriate sleep therapeutic interventions in older adults that will have a similar effect on slow-wave sleep as sodium oxybate, without the safety issues that surround use of this drug in older patients with and without PD. It is likely that for any intervention to be effective in patients, it will need to be given long-term, and possibly to neurologically asymptomatic patients, and it is not clear that sodium oxybate fits the bill given its propensity to cause adverse effects in older adults. 

    Second, we lack great biomarker readouts for the burden of synuclein pathology in humans. While all this work is being done, I do think this study provides a mechanistic rationale for using existing tools to optimize sleep in patients with, or at risk for, synucleinopathies. While we cannot yet say that this will for sure slow progression of synuclein pathology, there is at least now a rationale for believing that it might.


    . Amyloid-beta dynamics are regulated by orexin and the sleep-wake cycle. Science. 2009 Nov 13;326(5955):1005-7. PubMed.

    . The sleep-wake cycle regulates brain interstitial fluid tau in mice and CSF tau in humans. Science. 2019 Feb 22;363(6429):880-884. Epub 2019 Jan 24 PubMed.

  3. The main finding of this compelling study is that modulating sleep slow waves influences neuropathological outcome in two different mouse models of α-synucleinopathy. Specifically, the study showed less α-synuclein accumulation after enhancing slow waves with sodium oxybate compared to placebo, whereas sleep deprivation had an opposite effect. Intriguingly, the authors identified changes in the AQP4-mediated/glymphatic pathway and several processes related to protein homeostasis as potential mechanisms through which sleep slow waves could influence accumulation of α-synuclein.

    These findings fit with previous studies on the link between slow-wave sleep and pathological protein accumulation in Alzheimer's disease and imply that similar mechanisms could be present in synucleinopathies such as Parkinson's disease.

    As such, the findings resonate with our clinical studies in people with Parkinson's, suggesting that deeper slow-wave sleep is associated with better cognitive performance cross-sectionally and slower motor progression over time, although prospective confirmation of the latter observation is pending.

    The present study is exciting, as it provides more rationale to further explore the role and therapeutic potential of sleep, especially slow-wave sleep, in clinical populations with neurodegenerative disorders, including synucleinopathies. This is of interest because there are pharmacological and emerging, highly specific non-pharmacological methods for enhancing slow-wave sleep in humans.

    There are several additional findings, including (1) sex differences in the therapeutic response, (2) important observations on the trajectories of sleep, behavioral impairment, and neuropathology in synuclein mice, and (3) discordance between neuropathological and clinical outcome, which warrant further exploration.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Deep Sleep Makes Waves for CSF
  2. Disturbed Sleep Exerts Toll on Memory and Neurodegeneration
  3. Does Amyloid Disturb the Slow Waves of Slumber—and Memory?
  4. Another Reason to Catch Some Zzzs: Sleep Regulates Tau Release
  5. Tau, More than Aβ, Affects Sleep Early in Alzheimer’s
  6. Bad Dream—Sleep Disorder Plus Neuroimaging Markers Spell Trouble
  7. Brain Drain—“Glymphatic” Pathway Clears Aβ, Requires Water Channel
  8. Dearth of Water Channels a Sign of ‘Glymphatic’ Breakdown in Alzheimer’s?

Research Models Citations

  1. α-synuclein A53T Mouse (Tg)

Paper Citations

  1. . Reduced Regional NREM Sleep Slow-Wave Activity Is Associated With Cognitive Impairment in Parkinson Disease. Front Neurol. 2021;12:618101. Epub 2021 Feb 19 PubMed.
  2. . VMAT2-Deficient Mice Display Nigral and Extranigral Pathology and Motor and Nonmotor Symptoms of Parkinson's Disease. Parkinsons Dis. 2011;2011:124165. PubMed.
  3. . Association between Sleep, Alzheimer's, and Parkinson's Disease. Biology (Basel). 2021 Nov 3;10(11) PubMed.
  4. . A Case of Rapid Eye Movement Sleep Behavior Disorder in Parkinson Disease Treated With Sodium Oxybate. JAMA Neurol. 2016 Jan;73(1):126-7. PubMed.
  5. . Sodium oxybate for excessive daytime sleepiness in Parkinson disease: an open-label polysomnographic study. Arch Neurol. 2008 Oct;65(10):1337-40. PubMed.
  6. . Sodium Oxybate for Excessive Daytime Sleepiness and Sleep Disturbance in Parkinson Disease: A Randomized Clinical Trial. JAMA Neurol. 2018 Jan 1;75(1):114-118. PubMed.
  7. . Closed-loop auditory stimulation method to modulate sleep slow waves and motor learning performance in rats. Elife. 2021 Oct 6;10 PubMed.

External Citations

  1. open-label study
  2. Phase 2 trial
  3. ongoing study

Further Reading


  1. . Slow-wave sleep and motor progression in Parkinson disease. Ann Neurol. 2019 May;85(5):765-770. Epub 2019 Mar 27 PubMed.

Primary Papers

  1. . Slow-wave sleep affects synucleinopathy and regulates proteostatic processes in mouse models of Parkinson's disease. Sci Transl Med. 2021 Dec 8;13(623):eabe7099. PubMed.