Why Do We Need to Sleep?

TSUKUBA, Japan—Outside the International Institute for Integrative Sleep Medicine, the heavy fragrance of sweet Osmanthus trees fills the air, and big golden spiders string their webs among the bushes. Two men in hard hats next to the main doors mutter quietly as they measure a space and apply adhesive to the slate-colored wall. The building is so new that they are still putting up the signs.

The institute is five years old, its building still younger, but already it has attracted some 120 researchers from fields as diverse as pulmonology and chemistry and countries ranging from Switzerland to China. An hour north of Tokyo at the University of Tsukuba, with funding from the Japanese government and other sources, the institute’s director, Masashi Yanagisawa, has created a place to study the basic biology of sleep, rather than, as is more common, the causes and treatment of sleep problems in people. Full of rooms of gleaming equipment, quiet chambers where mice slumber, and a series of airy work spaces united by a spiraling staircase, it’s a place where tremendous resources are focused on the question of why, exactly, living things sleep.

Ask researchers this question, and listen as, like clockwork, a sense of awe and frustration creeps into their voices. In a way, it’s startling how universal sleep is: In the midst of the hurried scramble for survival, across eons of bloodshed and death and flight, uncountable millions of living things have laid themselves down for a nice, long bout of unconsciousness. This hardly seems conducive to living to fight another day. “It’s crazy, but there you are,” says Tarja Porkka-Heiskanen of the University of Helsinki, a leading sleep biologist. That such a risky habit is so common, and so persistent, suggests that whatever is happening is of the utmost importance. Whatever sleep gives to the sleeper is worth tempting death over and over again, for a lifetime.

The precise benefits of sleep are still mysterious, and for many biologists, the unknowns are transfixing. One rainy evening in Tsukuba, a group of institute scientists gathered at an izakaya bar manage to hold off only half an hour before sleep is once again the focus of their conversation. Even simple jellyfish have to rest longer after being forced to stay up, one researcher marvels, referring to a new paper where the little creatures were nudged repeatedly with jets of water to keep them from drifting off. And the work on pigeons—have you read the work on pigeons? another asks. There is something fascinating going on there, the researchers agree. On the table, dishes of vegetable and seafood tempura sit cooling, forgotten in the face of these enigmas.

In particular, this need to make up lost sleep, which has been seen not just in jellyfish and humans but all across the animal kingdom, is one of the handholds researchers are using to try to get a grip on the bigger problem of sleep. Why we feel the need for sleep is seen by many as key to understanding what it gives us.

Biologists call this need “sleep pressure”: Stay up too late, build up sleep pressure. Feeling drowsy in the evenings? Of course you are—by being awake all day, you’ve been generating sleep pressure! But like “dark matter,” this is a name for something whose nature we do not yet understand. The more time you spend thinking about sleep pressure, the more it seems like a riddle game out of Tolkien: What builds up over the course of wakefulness, and disperses during sleep? Is it a timer? A molecule that accrues every day and needs to be flushed away? What is this metaphorical tally of hours, locked in some chamber of the brain, waiting to be wiped clean every night?

In other words, asks Yanagisawa, as he reflects in his spare, sunlit office at the institute, “What is the physical substrate of sleepiness?”

Biological research into sleep pressure began more than a century ago. In some of the most famous experiments, a French scientist kept dogs awake for more than 10 days. Then, he siphoned fluid from the animals’ brains, and injected it into the brains of normal, well-rested canines, which promptly fell asleep. There was something in the fluid, accumulating during sleep deprivation, that made the dogs go under. The hunt was on for this ingredient—Morpheus’s little helper, the finger on the light switch. Surely, the identity of this hypnotoxin, as the French researcher called it, would reveal why animals grow drowsy.

In the first half of the 20th century, other researchers began to tape electrodes to the scalps of human subjects, trying to peer within the skull at the sleeping brain. Using electroencephalographs, or EEGs, they discovered that, far from being turned off, the brain has a clear routine during the night’s sleep. As the eyes close and breathing deepens, the tense, furious scribble of the waking mind on the EEG shifts, morphing into the curiously long, loping waves of early sleep. About 35 to 40 minutes in, the metabolism has slowed, the breathing is even, and the sleeper is no longer easy to wake. Then, after a certain amount of time has passed, the brain seems to flip a switch and the waves grow small and tight again: This is rapid eye movement, or REM, sleep, when we dream. (One of the first researchers to study REM found that by watching the movements of the eyes beneath the lids, he could predict when infants would wake—a party trick that fascinated their mothers.) Humans repeat this cycle over and over, finally waking at the end of a bout of REM, minds full of fish with wings and songs whose tunes they can’t remember.

Sleep pressure changes these brain waves. The more sleep-deprived the subject, the bigger the waves during slow-wave sleep, before REM. This phenomenon has been observed in about as many creatures as have been fitted with electrodes and kept awake past their bedtimes, including birds, seals, cats, hamsters, and dolphins.

If you needed more proof that sleep, with its peculiar many-staged structure and tendency to fill your mind with nonsense, isn’t some passive, energy-saving state, consider that golden hamsters have been observed waking up from bouts of hibernation—in order to nap. Whatever they’re getting from sleep, it’s not available to them while they’re hibernating. Even though they have slowed down nearly every process in their body, sleep pressure still builds up. “What I want to know is, what about this brain activity is so important?” says Kasper Vogt, one of the researchers gathered at the new institute at Tsukuba. He gestures at his screen, showing data on the firing of neurons in sleeping mice. “What is so important that you risk being eaten, not eating yourself, procreation ... you give all that up, for this?”

The search for the hypnotoxin was not unsuccessful. There are a handful of substances clearly demonstrated to cause sleep—including a molecule called adenosine, which appears to build up in certain parts of the brains of waking rats, then drain away during slumber. Adenosine is particularly interesting because it is adenosine receptors that caffeine seems to work on. When caffeine binds to them, adenosine can’t, which contributes to coffee’s anti-drowsiness powers. But work on hypnotoxins has not fully explained how the body keeps track of sleep pressure.

For instance, if adenosine puts us under at the moment of transition from wakefulness to sleep, where does it come from? “Nobody knows,” remarks Michael Lazarus, a researcher at the institute who studies adenosine. Some people say it’s coming from neurons, some say it’s another class of brain cells. But there isn’t a consensus. At any rate, “this isn’t about storage,” says Yanagisawa. In other words, these substances themselves don’t seem to store information about sleep pressure. They are just a response to it.

Sleep-inducing substances may come from the process of making new connections between neurons. Chiara Cirelli and Giulio Tononi, sleep researchers at the University of Wisconsin, suggest that since making these connections, or synapses, is what our brains do when we are awake, maybe what they do during sleep is scale back the unimportant ones, removing the memories or images that don’t fit with the others, or don’t need to be used to make sense of the world. “Sleep is a way of getting rid of the memories in a way that is good for the brain,” Tononi speculates. Another group has discovered a protein that enters little-used synapses to cause their destruction, and one of the times it can do this is when adenosine levels are high. Maybe sleep is when this cleanup happens.

There are still many unknowns about how this would work, and researchers are working many other angles in the quest to get to the bottom of sleep pressure and sleep. One group at the Tsukuba institute, led by Yu Hayashi, is destroying a select group of brain cells in mice, a procedure that can have surprising effects. Depriving mice specifically of REM sleep by shaking them awake repeatedly just as they’re about to enter it (a bit like what happens to the parents of crying babies) causes serious REM sleep pressure, which mice have to make up for in their next bout of slumber. But without this specific set of cells, mice can miss REM sleep without needing to sleep more later. Whether the mice get away totally unscathed is another question—the team is testing how REM sleep affects their performance on cognitive tests—but this experiment suggests that where dreaming sleep is concerned, these cells, or some circuit they are part of, may keep the records of sleep pressure.

Yanagisawa himself has always had a taste for epic projects, like screening thousands of proteins and cellular receptors to see what they do. In fact, one such project brought him into sleep science about 20 years ago. He and his collaborators, after discovering a neurotransmitter they named orexin, realized that the reason the mice without it kept collapsing all the time was that they were falling asleep. That neurotransmitter turned out to be missing in people with narcolepsy, who are incapable of making it, an insight that helped trigger an explosion of research into the condition’s underpinnings. In fact, a group of chemists at the institute at Tsukuba is collaborating with a drug company in an investigation of the potential of orexin mimics for treatment.

These days, Yanagisawa and collaborators are working on a vast screening project aimed at identifying the genes related to sleep. Each mouse in the project, exposed to a substance that causes mutations and fitted with its own EEG sensors, curls up in a nest of wood chips and gives in to sleep pressure while machines record its brain waves. More than 8,000 mice so far have slumbered under observation.

When a mouse sleeps oddly—when it wakes up a lot, or sleeps too long—the researchers dig into its genome. If there is a mutation that might be the cause, they try to engineer mice that carry it, and then study why it is the mutation disrupts sleep. Many very accomplished researchers have been doing this for years in organisms like fruit flies, making great progress. But the benefit to doing it in mice, which are extremely expensive to maintain compared to flies, is that they can be hooked up to an EEG, just like a person.

A few years ago, the group discovered a mouse that just could not seem to get rid of its sleep pressure. Its EEGs suggested it lived a life of snoozy exhaustion, and mice that had been engineered to carry its mutation showed the same symptoms. “This mutant has more high-amplitude sleep waves than normal. It’s always sleep-deprived,” says Yanagisawa. The mutation was in a gene called SIK3. The longer the mutants stay awake, the more chemical tags the SIK3 protein accumulates. The researchers published their discovery of the SIK3 mutants, as well as another sleep mutant, in Nature in 2016.

While it isn’t exactly clear yet how SIK3 relates to sleepiness, the fact that tags build up on the enzyme, like grains of sand pouring to the bottom of an hourglass, has the researchers excited. “We are convinced, for ourselves, that SIK3 is one of the central players,” says Yanagisawa.

As researchers probe outward into the mysterious darkness of sleepiness, these discoveries shine ahead of them like flashlight beams, lighting the way. How they all connect, how they may come together into a bigger picture, is still unclear.

The researchers hold out hope that clarity will come, maybe not next year or the next, but sometime, sooner than you might think. On an upper story at the International Institute for Integrative Sleep, mice go about their business, waking and dreaming, in row after row of plastic bins. In their brains, as in all of ours, is locked a secret.