Why do memories falter with age? In a preprint posted on bioRxiv on May 22, scientists led by Omer Sharon and sleep researcher Matthew Walker at the University of California, Berkeley, suggested that tau pathology may be responsible. It curtails waves of neuronal activity during sleep that are critical for memory consolidation, the authors conclude.

  • Waves of brain activity during sleep help consolidate memories.
  • Tau tangles in the frontal lobe disrupt these waves, limiting recall.
  • The waves are uncoordinated and do not travel well.

During non-rapid eye movement (NREM) sleep, a slow wave of neural activity traverses from the prefrontal cortex towards the posterior cortex (Massimini et al., 2004). Scientists think these slow waves activate memories in the hippocampus to help make them permanent (Mitra et al., 2016). “This massive slow wave is one of the biggest phenomena in sleep. It is the most synchronized electrical activity you can see in the healthy brain,” Sharon told Alzforum.

A prior study showed that, in people with mild Alzheimer’s disease, slow wave activity faltered as tau pathology worsened (Jan 2019 news) Scientists have wondered why this happens, and if it weakens memory. 

To find out, first author Sharon and colleagues compared sleep data of a group of 61 young, healthy adults, average age around 20, with data from 74 adults around the age of 75. All had had flortaucipir and PiB PET scans to look for neurofibrillary tangles and amyloid plaques, respectively. Within 1.5 years of the scans, both groups participated in two overnight sleep study sessions a week or two apart to measure electrical activity by EEG, and memory by a word association recall test.

Tau Limits Travel. Tangle burden in the frontal cortex predicts the distance traveled by slow oscillations. [Courtesy of Sharon et al., 2024.]

In the young adults, collective slow waves of activity travelled, en masse, from frontal to the parietal cortices during NREM sleep. In older adults, these slow waves were less coordinated, or “lonely,” and less evenly distributed across the cortex. They also traveled shorter distances and were smaller in amplitude and less frequent. 

At first glance, the data seemed to suggest that the slow wave activity wanes with age. However, some older adults had NREM slow waves just like those of the younger adults, indicating age is not the only factor determining slow-wave activities.

What explains the weaker slow-wave activity in most of the older volunteers? When the scientists examined the PET scans, they found that the greater the tangle burden in the frontal cortex, where the slow waves originated, the weaker the waves (image above). Moreover, the less the slow waves traveled, the worse people performed on the overnight memory test.

These results suggest that age-related memory decline might be due, at least in part, to accumulation of neurofibrillary tangles. Further, it hints that memory decline in AD, which correlates much more tightly with tangles than plaques, might not be wholly attributable to tangles affecting local synaptic activity, but to the loss of coordinated slow-wave activity across the cortex region during NREM sleep.

“If slow waves associate with memory in tauopathies and in Alzheimer's disease, then maybe if we improve sleep, we could potentially improve memory in people with very early AD,” said Matthew Pase, Monash University, who was not involved in the study.

Indeed, there is evidence that poor sleep can lead to the accumulation of amyloid and tau, and in turn to dementia. And there's evidence that Alzheimer's disease and dementia can contribute to poor sleep (Dec 2017 conference news; Jan 2018 news; Sep 2021 news). “It's like the chicken and egg, it's quite hard to work out what comes first,” Pase said.

The researchers saw no relationship between slow-wave activity and Aβ in this study.

How slow-wave activity helps solidify memories is unclear. These waves coincide with an influx of calcium in neurons, which fosters neural plasticity (Niethard et al., 2018). Calcium influx in one area of the brain that is followed by a burst in another part of the brain may be creating a kind of sequential plasticity, Sharon said. “This helps the brain learn and store new memories.”

Despite the correlation between tangles and slow-wave activity, Pase noted that other factors, such as ApoE4, could be involved, both driving tau pathology and disrupting NREM sleep (Tranah et al., 2018; Feb 2023 news; Nov 2019 news). To figure out causation, the scientists would need to follow the changes in slow waves over time, he said.

Sharon agreed. “To pinpoint the cause, more longitudinal data is needed,” he said.” The scientist will conduct follow-up measurements to observe any long-term changes in the volunteers’ brains.—Kristel Tjandra

Kristel Tjandra is a freelance writer based in Mount Herman, California.

Comments

  1. I think this is a fascinating and timely research article, demonstrating that the propagation of slow waves across the neocortex during sleep is impaired in older individuals. This impairment is correlated to the degree of tau tangle pathology in the brain, and the authors’ analyses further suggest that the impairment of slow waves is related to deficits in overnight memory consolidation. Slow wave propagation is a highly conserved phenotype, and I am intrigued that this novel human dataset aligns very closely with preclinical papers from our group (Busche et al., 2015; Keskin et al., 2017).

    In our work with APP mouse models, we show that the propagation of slow waves is similarly impaired, and this impairment caused memory deficits in the mice. We found that the slow waves, which originate from deeper layers of the cortex, were more localized and failed to propagate into surrounding tissue in the mice.

    Importantly, this deficit was not dependent on amyloid plaques, consistent with the current human study, but was instead caused by soluble Aβ. We demonstrated that suppressing soluble Aβ rescued the wave deficits, while direct administration of soluble Aβ promoted them. Given that soluble Aβ oligomers cannot yet be directly measured in the human brain, which is a gap, this aspect obviously could not be included in the current analyses. Nonetheless, it is highly likely that the study participants have soluble Aβ in their brains.

    Furthermore, in our studies, we showed that impairments in slow-wave propagation and associated memory deficits in mice could be rescued by enhancing GABAergic inhibition. It would be intriguing to explore if a similar therapeutic approach could be tested in human participants.

    References:

    . Rescue of long-range circuit dysfunction in Alzheimer's disease models. Nat Neurosci. 2015 Nov;18(11):1623-30. Epub 2015 Oct 12 PubMed.

    . BACE inhibition-dependent repair of Alzheimer's pathophysiology. Proc Natl Acad Sci U S A. 2017 Aug 8;114(32):8631-8636. Epub 2017 Jul 24 PubMed.

Make a Comment

To make a comment you must login or register.

References

News Citations

  1. Tau, More than Aβ, Affects Sleep Early in Alzheimer’s
  2. Disturbed Sleep Exerts Toll on Memory and Neurodegeneration
  3. Skimping on Sleep Makes For More Aβ in the Brain
  4. Getting Too Little Sleep? You May Be Accumulating Amyloid.
  5. Secreted by Neurons, ApoE4 Makes Tangles and Degeneration Worse
  6. ApoE4 and Tau in Alzheimer’s: Worse Than We Thought? Especially in Women

Paper Citations

  1. . The sleep slow oscillation as a traveling wave. J Neurosci. 2004 Aug 4;24(31):6862-70. PubMed.
  2. . Human cortical-hippocampal dialogue in wake and slow-wave sleep. Proc Natl Acad Sci U S A. 2016 Nov 1;113(44):E6868-E6876. Epub 2016 Oct 17 PubMed.
  3. . Cortical circuit activity underlying sleep slow oscillations and spindles. Proc Natl Acad Sci U S A. 2018 Sep 25;115(39):E9220-E9229. Epub 2018 Sep 12 PubMed.
  4. . APOEε4 and slow wave sleep in older adults. PLoS One. 2018;13(1):e0191281. Epub 2018 Jan 25 PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Tau pathology leads to lonely non-traveling slow waves that mediate human memory impairment. 2024 May 22 10.1101/2024.05.22.595043 (version 1) bioRxiv.