Alzheimer’s Sleep Problems May Be Reversible by Calming Brain Immune Cells

Scientists have discovered an unexpected reason why Alzheimer’s disease disrupts sleep—and their findings suggest the problem may be reversible. In experiments involving mice with Alzheimer’s-like brain changes, researchers restored more than two hours of sleep per day without removing the amyloid plaques associated with the disease.

The overlooked culprit may be the brain’s own immune system. Specialized immune cells called microglia can respond to amyloid plaques by triggering inflammation that interferes with healthy sleep.

Researchers at the University of Kentucky found that temporarily eliminating most microglia allowed mice to sleep longer and spend more time in deep, restorative sleep. The findings suggest a potential new approach to treating one of Alzheimer’s disease’s most disruptive symptoms.

The study was published in Alzheimer’s & Dementia. The research team was led by Shannon L. Macauley, Ph.D., an associate professor of physiology at the University of Kentucky College of Medicine, and Nicholas J. Constantino, Ph.D., a recent doctoral graduate and the study’s first author.

Why Alzheimer’s Disease Can Disrupt Sleep

Macauley described the discovery as “paradigm-shifting” because it challenges the traditional explanation for severe sleep problems in Alzheimer’s disease.

Amyloid plaques are deposits of sticky protein fragments that accumulate between nerve cells. They are a characteristic feature of Alzheimer’s disease and have long been investigated for their role in the condition.

The brain also contains microglia, immune cells that remove debris, respond to injury and help maintain a healthy environment for neurons.

When microglia detect amyloid plaques, however, their defensive response may become excessive. Instead of simply addressing the threat, the cells release inflammatory signals that can interfere with normal brain activity and sleep.

Scientists have traditionally linked sleep problems in Alzheimer’s disease primarily to damaged neurons or amyloid plaque accumulation. The new findings suggest that the brain’s inflammatory response to those plaques may play a major role.

“Basically, we showed that it is not the plaques themselves, or solely dysfunctional neurons, that cause sleep loss but actually microglia,” Macauley said. She compared the process to a defense system responding so broadly that it creates disruption throughout the entire house.

How Researchers Studied Alzheimer’s and Sleep

The scientists studied two groups of mice. One group was genetically predisposed to develop amyloid plaques, while the other consisted of normal “wild-type” mice that experienced typical aging.

The mice were examined at two ages: six months, when plaques first appeared in the genetically modified animals, and 18 months, when the disease model showed more advanced pathology. Comparing the groups helped researchers separate changes associated with Alzheimer’s-like pathology from those linked to normal aging.

Each mouse was fitted with a small device that recorded electroencephalography (EEG) and electromyography (EMG).

• EEG measures the brain’s electrical activity and the rhythms associated with wakefulness and different stages of sleep.
• EMG measures electrical activity in muscles.

Using both measurements allowed the researchers to identify sleep stages and determine how long the animals spent in each one.

The team also used light-sheet microscopy to examine the relationship between amyloid plaques and microglia throughout the brain. This imaging technique makes biological tissue transparent and illuminates it with a thin layer of laser light, creating detailed three-dimensional maps of structures that are otherwise difficult to observe.

Drug Temporarily Eliminated 87% of Microglia

To determine whether microglia were directly responsible for the sleep disruption, the researchers conducted an experiment to temporarily eliminate most of the cells.

The mice received Pexidartinib, also known as PLX3397, in their food for 14 days. The drug, originally developed for cancer research, blocks a cellular signal that microglia need to survive.

After treatment, approximately 87% of the mice’s microglia had been depleted. This allowed the scientists to test whether sleep improved when the cells driving the inflammatory response were largely absent.

The researchers also analyzed the animals’ brain activity using a mathematical method called Fitting Oscillations and One Over Frequency, or FOOOF. This approach separates electrical signals into:

• Periodic activity: rhythmic electrical waves commonly associated with brain activity.
• Aperiodic activity: the background electrical activity or “noise” in the brain signal.

Examining these components helped the team identify changes in brain function, including whether the brain remained unusually active during periods when it should have been resting.

Sleep Problems Did Not Worsen as Plaques Increased

One of the study’s most unexpected findings was that sleep disruption did not continue worsening as amyloid plaques accumulated.

The researchers initially expected increasing plaque levels to produce progressively more severe sleep problems. Instead, the mice developed substantial sleep disturbances relatively early, and those problems remained largely unchanged even as plaque levels increased dramatically.

“I expected that as plaque burden became more severe, sleep disruption would also worsen,” Constantino said. “The disruptions in sleep and cortical EEG activity that occur at 6 months, when plaques first emerge, did not worsen by 18 months, despite more than double the amount of plaque burden.”

The researchers described this pattern as a ceiling effect. Sleep disruption appeared to reach a certain level early in the disease process rather than increasing in direct proportion to amyloid plaque accumulation.

The finding suggests that the initial inflammatory response triggered by plaques may be particularly important. Once microglia become activated and disrupt sleep, additional plaque accumulation may not cause an equivalent increase in sleep loss.

Alzheimer’s Disrupts Deep Sleep Differently Than Normal Aging

The study also identified a difference between sleep changes caused by normal aging and those associated with Alzheimer’s-like pathology.

Sleep consists of several stages with different functions. Rapid eye movement, or REM, sleep is associated with vivid dreaming and plays an important role in memory processing.

In the study, normal aging primarily reduced REM sleep. Alzheimer’s-like pathology had a different effect: it selectively disrupted non-rapid eye movement, or NREM, sleep.

NREM sleep includes the deepest and most restorative stages of sleep. During this time, the brain performs maintenance functions that support learning, memory and recovery. Deep sleep is also associated with processes that help clear metabolic waste from the brain.

“That restorative sleep is super important for physical repair, learning and memory and washing out the toxins of the day,” Macauley said.

Losing deep sleep could therefore have consequences beyond daytime fatigue. Researchers suspect that poor sleep may interfere with processes that maintain the brain’s internal environment, potentially contributing to a cycle in which Alzheimer’s pathology disrupts sleep and inadequate sleep further aggravates the disease.

Restoring More Than Two Hours of Sleep Without Removing Plaques

The most striking result occurred after the researchers temporarily depleted most of the microglia in mice with Alzheimer’s-like pathology.

The animals gained more than two hours of sleep per day. They also spent longer periods in restorative NREM sleep, giving them more opportunities to transition naturally into REM sleep, which is important for memory formation.

These improvements occurred without reducing amyloid plaque levels. The plaques remained in the brain, but the mice slept substantially better.

The results provide evidence that microglia-driven inflammation contributes directly to sleep disruption and that suppressing this inflammatory response can reverse at least some of the associated sleep problems.

The findings also suggest that improving sleep may not require removing amyloid plaques first.

However, the research was conducted in mice, and the same benefits have not been demonstrated in people. Temporarily eliminating most of the brain’s immune cells is not an established or practical treatment for patients with Alzheimer’s disease.

The researchers are now investigating whether the benefits can be reproduced by adjusting microglial activity rather than eliminating the cells.

Could Medications Calm Overactive Microglia?

Because microglia perform essential protective functions, eliminating them is unlikely to be an ideal long-term treatment strategy. A more practical approach may be to modify how the cells respond to amyloid plaques and regulate inflammation.

The research team is investigating existing medications, including the diabetes drug Metformin and the anti-seizure medication Stiripentol. The drugs are being studied for their potential to change how microglia produce and use energy, a process known as cellular metabolism.

Altering microglial metabolism could potentially reduce excessive immune activation and its harmful effects on sleep. These drug strategies remain experimental for this purpose, and additional research is needed to determine whether they can safely produce similar benefits in humans.

Researchers are particularly interested in whether treating microglial dysfunction during the early stages of Alzheimer’s disease could improve sleep and quality of life before substantial memory loss develops.

“If we can target that process, it might help with quality of life, attention, cognition and confusion,” Macauley said.

EEG May Help Detect Alzheimer’s-Related Brain Changes

Beyond identifying a potential treatment target, the study found electrical activity patterns that may help distinguish Alzheimer’s-related pathology from normal brain aging.

The researchers believe these patterns could eventually make EEG a useful tool for detecting and monitoring changes associated with Alzheimer’s disease.

EEG records electrical signals through electrodes placed on the scalp. Compared with more expensive imaging procedures, it is relatively affordable, does not require surgery and can be repeated to track changes over time.

The team envisions EEG becoming a “readily accessible, affordable and longitudinal biomarker of Alzheimer’s disease.” A biomarker is a measurable biological sign that provides information about a person’s health or a disease. A longitudinal biomarker can be measured repeatedly to monitor how a condition develops or changes.

Portable EEG systems could eventually allow researchers and clinicians to monitor people in their homes and screen for electrical patterns associated with Alzheimer’s disease. People with concerning results could then be referred for additional testing.

This approach could be particularly valuable in rural communities across Kentucky, where patients may need to travel long distances to reach specialized medical centers. However, the potential use of EEG for Alzheimer’s screening still requires validation in human studies.

What the Findings Could Mean for Alzheimer’s Treatment

The study suggests that treating Alzheimer’s disease may require looking beyond amyloid plaques and examining how the brain responds to them.

By identifying microglia as a major contributor to sleep disruption in their mouse models, the University of Kentucky researchers have uncovered a potential way to restore an essential brain function without first removing amyloid deposits.

The next challenge is determining whether calming overactive microglia can deliver similar benefits in people. If future research confirms the findings, targeting inflammation could eventually help improve sleep, cognition and quality of life for people affected by Alzheimer’s disease.

Research reported in this publication was supported by the National Institute on Aging of the National Institutes of Health under Award Numbers R01AG068330, R01AG093847 and P30AG072946, and by the National Institute of General Medical Sciences of the National Institutes of Health under Award Numbers P30GM127211 and P20GM148326. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

This work was supported by a $287,236 award from the Cure Alzheimer’s Fund.

This work was supported by a $250,000 award from The CART Fund (Coins for Alzheimer’s Research Trust).