Sleep is a rose, as the Persians say.
Vladimir Nabokov, Lolita
NO ONE KNOWS why we sleep. The obvious answer is that we do it because we’re tired. But the brain is 95 per cent as active during sleep as during our waking hours. And that, given our evolutionary legacy as prey for larger predators, not to mention the amount of time we lose for reproduction and food gathering, makes the reason for sleep all the more mysterious.
For decades there have been theories to better explain it, ranging from wound healing and heat regulation, to memory consolidation and dream-induced creative thinking. But recent work, published in leading scientific journals, suggests that sleep may also exist to shield the brain from Alzheimer’s.
It began in 2005, when a group of psychiatrists at St James Hospital in Dublin demonstrated a link between sleep disturbances, like increased insomnia and daytime sleeping, and the severity of dementia in their Alzheimer’s patients.1 This didn’t come as a surprise–many brain disorders correlate with sleep disturbances–and so it was mainly viewed as practical knowledge to help doctors and caregivers choose the right medicine to help their patients sleep.
But others believed the link was more far-reaching; that something deeper was at play. So scientists set about replicating the paradigm in Alzheimer’s mice.
Among them was a group at Washington University, St Louis, led by a man named David Holtzman. His team showed that beta-amyloid levels fluctuate with sleep–wake cycles: depriving mice of sleep raises beta-amyloid levels, while chemically stimulating sleep lowers them.2 That was 2009, and the discovery was soon supported by the group’s converse observation, in 2012, that vaccinating the mice against beta-amyloid restored normal sleeping patterns.3
One year later, Maiken Nedergaard and her team at the University of Rochester, New York, found evidence that the brain actually cleans itself during sleep,4 removing beta-amyloid using a network of microscopic channels filled with spinal fluid called the glymphatic pathway: a kind of plumbing system composed of glial cells that clears the brain of waste products. She compared the sleeping brain to a ‘dishwasher’ clearing out molecular ‘dirt’. The findings were impressive. The only nagging caveat was that mice (besides being mice) are nocturnal; they have different sleeping behaviours to us. A human study was desperately needed.
This came from researchers at the University of California, Berkeley, led by a neuroscientist called Matthew Walker. For Walker, a young, fresh-faced, jocular professor originally from England, sleep is more intimately tied to memory than people realise. He likes to preface his lectures by assuring people that they are allowed to fall asleep during his talk. ‘Knowing what I know about the relationship between sleep and memory,’ he recently told one audience, ‘it’s actually the greatest form of flattery, for me, to see people like you not being able to resist the urge to strengthen what I’m telling you by falling asleep.’5
In July 2015 Walker and his team recruited twenty-six healthy people with an average age of seventy-five, and set out to uncover the relationship between sleep, beta-amyloid and memory.6 To begin, Walker performed PiB-PET scans on the recruits to gauge their brain’s amyloid quota. Then he made the participants memorise sets of word pairs before asking them to spend the night in a sleep laboratory, where their sleeping patterns could be professionally monitored.
Sleep moves in roughly ninety-minute cycles, made up of rapid eye movement (REM) and non-rapid eye movement (NREM) phases. REM sleep only lasts around ten minutes and happens while dreaming (though why the eyes move is unknown). Deep and dreamless NREM sleep dominates each cycle, but becomes less dominant later in the night as REM sleep encroaches slightly more into its time. Memory consolidation is thought to occur during an NREM phase called slow-wave sleep–a period of synchronised, low-frequency pulses of electrical activity. This is what Walker was particularly interested in for the human study.
The next morning, Walker’s recruits retook the word-pair test during a functional MRI brain scan. As it turned out, the participants with the highest beta-amyloid levels had the lowest slow-wave activity and scored the worst on memory recall. The reduced slow-wave activity was most prominent in the prefrontal cortex–the brain region where beta-amyloid accumulation was at its highest. This was seen after correcting for age, sex and brain size. What’s more, the participants were told to abstain from stimulants such as coffee and alcohol two days before testing. Mathematically, there is a linear relationship by which beta-amyloid affects sleep, which then affects memory. This makes sleep itself a possible candidate for therapeutic intervention in Alzheimer’s. Walker published the study in Nature Neuroscience and, predictably, received full-bore sensationalism from the press.
But there are two problems with the study. First, the findings are correlative: they don’t prove cause and effect. For that, people’s sleeping habits would have to be followed over several years. Second, it’s possible the results were swayed by the recruits’ having to sleep in a new environment. Walker had asked them to make home sleep logs and told them they could sleep in the laboratory in the same way, but without somehow measuring the difference between the two settings, the findings remain open to interpretation.
Which is precisely what Holtzman and fellow neurologist Brendan Lucey have done. ‘Despite these issues,’ they wrote in an opinion article accompanying the study, ‘the study by [Walker’s team] provides important new insights into the changes in sleep and memory in preclinical Alzheimer’s disease, as well as indicating potential new avenues for investigation.’7 They’ve argued for alternative ways that the trifecta of beta-amyloid, sleep and memory are connected. According to them, it may be that beta-amyloid affects sleep and memory simultaneously, or that sleep problems associated with ageing affect memory and beta-amyloid, which in turn affects sleep, creating a self-perpetuating loop of havoc. How tangles fit into this is unknown. But hurdles aside, few would dispute the importance of a good night’s sleep, and this blossoming new area of research wholly and irrefutably strengthens that mandate.
What are we to make of so much ambivalence? If my grandfather were alive today, he’d probably tell you to live like a rock star; the austere life he led certainly didn’t protect him. But still, there’s no denying that the evidence for non-pharmacological countermeasures against Alzheimer’s, albeit conflicting, exists. Of course, as a scientist, I’d be the first to say that evidence alone isn’t enough–it has to be good evidence: large samples, widely replicated, and so forth–but since we know that these lifestyle measures are good for us anyway, the most sensible approach is to play it safe. So follow a Mediterranean diet. Exercise. Avoid stress. Stimulate your mind. Sleep. You’ve got nothing to lose and everything to gain.