How damaged sleep circuits may let Alzheimer’s proteins build up

People with Alzheimer’s disease can sleep poorly at night and still become unusually sleepy during the day. This combination may look like a simple disturbance of rest, but the biology is more complicated. Sleep is coordinated by several interacting brain circuits, and those circuits can themselves become vulnerable to Alzheimer’s-related protein pathology.

A review published on April 21, 2023, proposes that this relationship may run in both directions. Alzheimer’s disease can disrupt the systems that regulate sleep and wakefulness. Poor or fragmented sleep may then interfere with the brain’s ability to maintain and clear proteins. The resulting strain could add to the problems already affecting neurons and memory. The review presents this as a promising explanation to investigate, not as a complete causal sequence already demonstrated in people.

The brain’s sleep controls are part of the disease story

A useful place to begin is the locus coeruleus, a small brainstem region involved in arousal, wakefulness and memory. Its noradrenaline-producing neurons help keep the brain alert and influence the balance between waking and sleeping. The review describes the locus coeruleus as one of the earliest regions affected by abnormal tau, with tau pathology appearing there before substantial accumulation in some other regions associated with Alzheimer’s disease.

Because the locus coeruleus connects with the cortex, hippocampus, thalamus, basal forebrain and hypothalamus, damage there could affect more than alertness. It may alter the timing and quality of sleep while also disturbing circuits involved in memory. The review links these early changes with the appearance of sleep problems during Alzheimer’s progression.

The hypothalamus adds another layer. It contains populations that promote wakefulness, including orexin-producing and histaminergic neurons, as well as neurons involved in promoting sleep and rapid eye movement sleep. The review describes tau pathology and loss among some wake-promoting populations, while another sleep-related population appears more resistant. In people with Alzheimer’s disease, orexin-producing neurons may be reduced even as cerebrospinal-fluid orexin is sometimes elevated. That combination suggests that the sleep-wake system is not simply switched off; its signals may become poorly balanced, with nighttime sleep loss and increased daytime napping emerging from a more complex disruption.

Protein maintenance is more than one cleanup system

Proteostasis is the brain’s ongoing maintenance of functioning proteins. It includes making, folding, repairing and removing proteins so that they do not accumulate in damaging forms. In Alzheimer’s disease, this maintenance is already under strain. Undigested cellular material, amyloid-beta and tau can build up in vulnerable regions, contributing to neuronal injury.

Several internal systems participate in protein handling. The ubiquitin-proteasome system breaks down selected proteins, while autophagy and the lysosomal system help process larger or damaged cellular material. The unfolded protein response helps cells respond when proteins are misfolded or accumulating. These are intracellular processes: they operate within cells and help determine whether a neuron can keep its internal protein environment under control.

The review also discusses fluid-associated clearance, including the movement of substances through the brain’s glymphatic system. This is a different route from intracellular degradation. It concerns how material can move out of brain tissue through fluid pathways, with slow-wave sleep associated with conditions that support this clearance. Losing slow-wave sleep could therefore affect both the internal handling of proteins and the movement of material through the surrounding fluid environment, although the precise human sequence remains unresolved.

Neuronal activity matters as well. The review summarizes evidence that increased activity can promote the spread or release of amyloid-beta and tau, while sleep and circadian disruption can affect the proteasome, autophagy, the unfolded protein response and glymphatic clearance. In this account, sleep is not merely a period when the brain becomes inactive. It is a changing physiological state that may influence several protein-maintenance processes at once.

Amyloid-beta does not affect sleep in one uniform way

Animal experiments summarized by the review help show why the relationship cannot be reduced to “more Alzheimer’s protein means less sleep.” In zebrafish, injected amyloid-beta assemblies of different lengths produced different effects. Short assemblies increased activity in hypothalamic neurons and promoted wakefulness through particular receptor signals. Longer assemblies reduced neuronal activity and promoted sleep through a pathway involving prion protein. Very long assemblies had no observed effect in that experiment.

In mice, injected amyloid-beta oligomers produced fragmented sleep, with less non-rapid eye movement and rapid eye movement sleep and more transitions between stages. That effect was not observed in mice lacking prion protein, pointing to a context-dependent mechanism rather than a universal response to amyloid-beta itself.

These findings come from controlled animal experiments using injected assemblies, not from ordinary sleep patterns in patients. They show that the physical form of a protein assembly, the species being studied and the experimental setting can all shape the outcome. They also help explain why people with Alzheimer’s disease may experience a mixture of shortened nighttime sleep, frequent awakenings and daytime sleepiness rather than one predictable sleep pattern.

Human studies measure different pieces of the problem

Human observations support an association between sleep and Alzheimer’s-related pathology, but the measurement method changes what can reasonably be concluded. In older adults, weaker electrical activity in the 1–2 Hz slow-wave range during non-rapid eye movement sleep has been associated with greater tau signal on brain imaging. In people with mild-to-moderate Alzheimer’s disease, sleep-wake disturbances have been associated with cerebrospinal-fluid levels of phosphorylated tau.

Other studies capture different aspects of sleep. Movement-based monitoring over several days has linked objectively measured sleep impairment with greater tau imaging signal in early disease-related regions. Self-reported sleep problems have been associated with both amyloid-beta and tau measures, even when people’s reported sleep did not match their objectively measured sleep as closely. Subjective experience and movement-based estimates therefore tell related but nonidentical stories.

The biological measurements also need to remain separate. Brain imaging can show where a tracer signal is distributed. Blood tests measure material circulating in plasma. Cerebrospinal-fluid tests sample a different compartment, closer to the brain. These methods cannot be treated as interchangeable indicators of one underlying quantity.

Short sleep-deprivation experiments add another piece. After one night of restricted or absent sleep, studies summarized in the review reported changes in cerebrospinal-fluid or blood measures of amyloid-beta and tau, with the size and pattern differing across experiments and compartments. One-night changes in a biomarker can demonstrate that sleep loss is associated with a measurable biological shift. They do not show that one poor night causes Alzheimer’s disease, nor do they quantify one person’s long-term dementia risk or predict what will happen to that person’s memory.

Better sleep is a possible intervention question

The review reports different results for two orexin-related sleep medicines. A trial of suvorexant in people with insomnia and Alzheimer’s disease reported a significant improvement in total sleep time compared with placebo. A lemborexant trial involving irregular sleep-wake patterns in mild-to-moderate Alzheimer’s disease produced a more limited picture: some reduction in restlessness was reported, but there was no strong overall benefit across sleep outcomes in the review’s account.

Those findings concern sleep-related outcomes. They do not, on the evidence described here, establish that either treatment improved protein clearance, protected neurons or slowed cognitive decline. A drug that changes sleep may affect protein maintenance, but that proposed connection requires its own measurements and longer-term testing.

Behavioral approaches raise a related but distinct question. Sleep education and cognitive behavioral therapy for insomnia do not directly manipulate the molecular machinery that handles proteins. If they improve sleep, however, they could help researchers examine whether changes in sleep are accompanied by changes in protein-maintenance measures or cognitive outcomes. That would be a way to test the relationship without assuming that every sleep improvement has a direct molecular effect.

Translation is difficult because the relevant evidence spans several levels at once: molecular pathways inside cells, vulnerability in particular neuronal populations, electrical and behavioral sleep patterns, and measures of cognition. Sleep-deprivation experiments can also involve stress, and stress-related changes may affect cellular protein handling independently of sleep loss. Future work would need to separate these influences while following people across disease stages, rather than treating a short-term biomarker movement as proof of a long-term clinical effect.

What this could change

For people in the early stages of Alzheimer’s disease, the longer-term possibility is a treatment strategy that improves sleep while helping preserve cognitive function. That could involve combining a sleep-directed approach with a therapy aimed at amyloid-beta, tau or another part of protein maintenance. The practical benefit would depend on showing that changing sleep and protein handling together produces durable effects on disease progression, not merely a better night’s sleep or a temporary biomarker shift.

The 2023 review’s contribution is to connect sleep regulation, vulnerable neuronal circuits and protein maintenance into one testable clinical question. If later research can identify which patients, disease stages and combinations of interventions respond, sleep could become more than a symptom to manage. It could become one part of a broader strategy for protecting the brain’s ability to maintain its proteins and preserve cognition.

Key Sources

• Primary source: Proteostasis failure exacerbates neuronal circuit dysfunction and sleep impairments in Alzheimer's disease.
  DOI or Journal Page · Review or Meta-Analysis · 2023-04-21