As people go through phases of deep sleep, when neuronal activity patterns form slow waves across the brain, pulses of cerebrospinal fluid bathe the brain, possibly as a means of taking out molecular trash. Critical to the cleaning process is the water channel aquaporin-4 (AQP4), which is present in the membranes of astrocytes, according to experiments in rodents. AQP4 is necessary for the flow of CSF through the brain as it travels via a series of lymph-like vessels that make up the glymphatic system and clears waste from the brain.

In a paper published May 5 in PLOS Biology, researchers connect the dots between AQP4 and sleep in people. They showed that individuals with a set of eight single nucleotide polymorphisms (SNPs) in AQP4 have more slow wave energy as they sleep, which is characteristic of deep states of slumber, than those...

“This paper’s exciting because it gives the first hint that maybe in humans the way we sleep and how deeply we sleep might actually be coupled to how efficient our brains are at carrying out this fluid flow exchange process,” says Laura Lewis, a biomedical engineer at Boston University. She did not participate in the study, but led the team that demonstrated the connection between slow wave neuronal activity during deep sleep and CSF influx in people last year.

See “Waves of Fluid Bathe the Sleeping Brain, Perhaps to Clear Waste

Although previous studies have shown that people have some aspects of the glymphatic system, it’s not clear whether the factors responsible for fluid flow are also present in humans. The transgenic and radio-label tracing methods researchers have used to explore the glymphatic system in rodents are too invasive to be used in people, so when Sebastian Holst, a postdoc in the Neurobiology Research Unit at Copenhagen University Hospital, and his colleagues wanted to look for more evidence of the glymphatic system in the human brain, they needed another way in.

Because AQP4 plays a central role in glymphatic function in rodents, they started by investigating what kind of variations in AQP4 exist in people and whether those variants might affect fluid flow and therefore sleep. In the dbSNP database, the researchers found a group of eight AQP4 SNPs that are almost always inherited together and account for about one-fifth of the copies of the gene present in people in Europe, the US, and East or South Asia. In the current study, the authors refer to these eight SNPs as the minor haplotype and to the more common sequence as the major haplotype.

A 2017 study showed that “there’s actually a functional link between one of these variants . . . and how much aquaporin-4 seems to be expressed,” says Holst. Cultured cells carrying the minor haplotype had lower AQP4 expression, which—if it holds true in the human brain—could change glymphatic flow.

In the latest study, the research team turned to previously collected blood samples from 123 people who had been part of a sleep deprivation study. They genotyped the samples and found that 52 subjects carried at least one copy of the minor haplotype, while 71 had only the major haplotype. Those with the minor haplotype, who likely have lower AQP4 expression, showed more slow wave electrical activity during sleep, especially at the beginning of the night, than those with the major haplotype. They also self-reported being less sleepy and performed better on a task that measured their alertness every three hours over 40 hours of prolonged wakefulness.

“Although we can’t be sure yet, it really seems that the deeper we sleep, the more slow waves the brain produces [and] the more clearance there is,” Holst tells The Scientist. “The fact that we can see this association between this channel that allows that [glymphatic] flow in the rodent brain and how deeply someone sleeps . . . is an indication that this is also present in the human brain. We still have a lot of interesting study ahead of us to really show that this is the case.”

The authors hypothesize that people with the major haplotype, who express more AQP4, need less slow wave electrical activity to stimulate glymphatic flow. In contrast, individuals who express less AQP4 sleep more deeply, perhaps to generate more slow wave electrical activity, which then initiates glymphatic flow. These interactions establish a feedback loop between sleep, electrical activity, and glymphatic fluid flow, they write.

A study like this one “really validates the preclinical work that’s ongoing and can give great insights into neurodegenerative diseases,” says Philip Haydon, a neuroscientist at Tufts University who did not participate in the study. In addition to the demonstrated necessity of AQP4 for glymphatic clearance, prior research has also drawn associations between AQP4 variants and neurodegenerative disorders, such as Alzheimer’s disease.

“In patients with Alzheimer’s disease, it’s known that you can have reduced . . . slow wave activity during non-REM sleep, and there’s also the idea that in patients with Alzheimer’s disease you get less glymphatic clearance,” he explains. “Now according to their paper, you might predict less glymphatic clearance would stimulate more slow wave activity, but you’re seeing in patients there’s a reduction in slow wave activity.” This raises questions about whether the feedback loop the authors predict has broken down during aging and disease, he says.

“I don’t have any answers. I just have questions. And what I love is you have a real observation that is very clearly documented in people,” Haydon adds. “And this now allows us to go back to animal models and start asking the questions that hopefully then will allow us to come back to people.” 

S.M. Ulv Larsen et al., “Haplotype of the astrocytic water channel AQP4 is associated with slow wave energy regulation in human NREM sleep,” PLOS Biologydoi:10.1371/journal.pbio.3000623, 2020.

Correction (May 19): The article has been updated to reflect the incidence of the major and minor haplotypes in people. The Scientist regrets the error.

Interested in reading more?

The Scientist ARCHIVES

Become a Member of

Receive full access to more than 35 years of archives, as well as TS Digest, digital editions of The Scientist, feature stories, and much more!
Already a member?