The connection between sleep and brain development
Question #1: Why do animals sleep?
Question #2: What happens to sleep-deprived animals?
Answer: They die, but the reason is unknown.
Yu Hayashi decided to devote himself to sleep research mainly because he was surprised to find that so much about sleep was still unknown even in the 21st century.
“An American research group discovered in 1983 that rats die within three weeks if you completely deprive them of sleep. Forty years later, we still do not precisely know why that happens. REM sleep was discovered in the 1950s. Decades later, we still do not precisely know why REM sleep is necessary in the first place.”
We spend one-third of our days asleep, an indispensable state not only for humans but also for other animals. Yet, sleep is at the very frontier of science.
Hayashi originally studied brain development at the Graduate School of Science at the University of Tokyo. He was trying to clarify how the brain developed after birth at the molecular and cellular level through interactions with the environment. After completing his doctoral studies, Hayashi was contemplating what he should do as a postdoctoral fellow when he stumbled across a certain research paper.
“It was an old paper from the 1960s reporting measurements and records of sleep states throughout the life cycle from two-week-old babies to 76-year-olds. According to the study, although newborn babies spent much time asleep and demonstrated both REM and non-REM sleep, they had a particularly high amount of REM sleep. Reading that paper gave me the idea that REM sleep might be related to rapid brain development in newborns. That was my way into sleep research.”
Hayashi was surprised to learn that much about sleep was still unknown.
“My first aim was to create a strain of mice in which I could control REM and non-REM sleep cycles. I thought that a lot of research had already been done, much was already known and I would just have to jump in… but it turned out that quite the opposite was true. In the end, I decided that I had to research the fundamental mechanisms of sleep first."
That was in around 2008.
“Although my supervisor, Dr. Shigeyoshi Itohara, was very supportive, I had just become a postdoctoral fellow, which meant that my knowledge about sleep was limited and I was a beginner at doing research on mice, so I was completely at a loss. Around that time, a Japanese professor working in France who had published an important paper suggesting that the REM sleep center might be located in a certain part of the brainstem in cats was back in Japan. I got to meet him, Dr. Kazuya Sakai, by chance at a conference I participated in. When I told him I wanted to elucidate the function of REM sleep by using genetically engineered mice, he was very sympathetic towards the idea and offered some advice. His words greatly encouraged me at that time and he has supported me ever since, for which I am very grateful.”
And so, sleep research has taken Hayashi to a world full of unexpected possibilities.
Why has REM sleep evolved?
Non-REM sleep is characterized by a slow heartbeat, calm breathing, and the dominance of delta waves. REM sleep, on the other hand, is characterized by an irregular heartbeat and breathing, rapid eye movements, and the dominance of alpha and beta waves. It is also the sleep stage when dreams occur. Although the differences might make it seem like non-REM sleep is deep and REM sleep is shallow, if anything, REM sleep is deeper than average non-REM sleep.
“REM sleep occurs in rodents and some bird species, including chickens. Birds have descended from dinosaurs, but if we go back long enough in time, we find that the human lineage has a reptilian ancestor as well. Considering that both birds and mammals have REM sleep, we could speculate that lizards and other reptiles have it, too. Recent studies have shown that there is a stage of increased brain activity and simultaneous eye movements during sleep in reptiles. Based on these findings, we could further speculate that the common ancestor of reptiles, birds, and mammals already had REM sleep. So, what about fish? Although we cannot speak of exactly REM and non-REM sleep, it seems that fish also have two types of sleep, which might be the prototype of REM sleep.”
Hayashi believes that REM sleep must have been a unique adaptation in animals with complex brains. In other words, non-REM sleep is the more evolutionarily ancient. His sleep research on genetically engineered nematodes leads him to the following conclusion.
By the way, nematodes, worms about 1 mm long, also “sleep.” Naturally, scientists cannot measure brain waves in nematodes, so they determine whether nematodes are asleep using a definition of sleep based on resting behavior.
“Some researchers argue that REM sleep is evolutionarily older, and non-REM sleep is evolutionarily more recent. However, we have found that the same gene that promotes sleep in nematodes is also present and promotes non-REM sleep in mice. Therefore, I am rather convinced that non-REM sleep has older roots.”
Why, then, has REM sleep evolved in certain animals?
“That is one of the big questions I hope to answer. It is challenging to study REM sleep in humans because a short nap is not enough to enter REM sleep. Moreover, because the amount of REM sleep is so short, it is difficult to observe what is happening in the brain. So, we developed a method to directly observe the changes in brain activity during sleep and found that blood flow in the cerebral cortex increased twofold during REM sleep. Moreover, we know that the cerebral cortex has been growing larger over the course of vertebrate evolution. The human cerebral cortex also fits the pattern as well. From this, we could hypothesize that REM sleep has evolved to keep a sizable cortex at a consistent level of health.”
If the hypothesis is true, that would mean that REM sleep is essential for higher brain functions. Research papers suggesting just that have been published recently.
“Comparing the sleep characteristics of healthy people over 60 showed that those who developed dementia within the next 12 years had less REM sleep. In other words, decreased REM sleep is possibly a sign of developing dementia.”
So now we think that REM sleep plays a crucial role in human health. However, the full extent of the function of REM sleep remains unknown, fueling Hayashi’s passion.
Controlling sleep in mice
Based on their detailed observations of the neurons that play a role in sleep in mice, Hayashi and his colleagues succeeded in controlling these neurons and created “artificial” mice with different sleep patterns. By using this method unique to Hayashi’s lab, REM sleep in these mice can be turned on and off at will, so to speak.
“We have achieved this by using genetic engineering which is a time-consuming and fastidious process. Every cell has the same DNA, but not every cell uses the same genes. Therefore, depending on the cell, different genes will be turned on or off even though the complete genetic base is the same which gives each cell individual characteristics. Even cells close to each other might be using different sets of genes, so one becomes a cell that controls respiration and another that controls REM sleep. We can tell which genes are turned on by looking at the RNA in the cell. Let us say we want to figure out the function of a cell with a specific active gene. To do that, we can use promoter genes to either activate, inhibit, or destroy the cells with the target gene, a technique first used in neuroscience by Nobel laureate Susumu Tonegawa. By observing which function is affected, we can determine the role of the cells in which the target gene is active.”
That is how Hayashi discovered the switch in the brain that controls REM sleep in mice. Of course, we do not have similar control over REM sleep in humans yet, but at least we do know that there must be cells in the brain stem that can turn REM sleep on and off.
“As the function of the cells that control REM sleep gradually declines due to aging or stress, the amount of REM sleep decreases. As a result, blood flow to the cerebral cortex might be compromised, which might lead to cognitive impairment. That is how I see the process. In my lab, we managed to increase the amount of REM sleep in mice by inserting genes that increase neural activity into the cells that control REM sleep. Although we cannot apply this to humans, if we could develop an alternative method to increase REM sleep in humans, I believe we may be able to delay or prevent the progression of dementia.”
Can you stay alive without sleep?
Besides understanding REM sleep, Hayashi focuses on engineering nematodes that stay alive even without sleep.
As mentioned in the pop quiz, animals die if they are completely deprived of sleep. What precisely kills them, however, is still unknown. Hayashi is determined to find it out.
“For a while, it was believed that an increase of commensal bacteria in the bloodstream might overpower the immune system, leading to sepsis and death. However, it later turned out that even when the growth of bacteria is suppressed with antibiotics and other drugs, sleeplessness still leads to death. There are many other hypotheses, such as that the accumulation of free radicals during a sleepless period leads to death. However, none has solid evidence, and each has shortcomings. So, I wonder if the reason is beyond the limit of human imagination.”
If the ordinary method of using our imagination to come up with a hypothesis and then test it experimentally fails… what do we do then? Instead of formulating a hypothesis in advance, a more suitable approach might be to start by experimenting, randomly manipulating genes, and then choosing intriguing phenomena for further investigation. Hayashi determined that nematodes would be the perfect subject for such a process.
“It was already known that sleep-deprived nematodes tended to die, so I thought it would be possible to clarify the role of sleep using even nematodes. We experimented with random mutations in various genes and discovered a mutant that slept more than twice as well as a normal nematode. We then investigated why the mutant nematodes slept better. We found that mutant nematodes tended to accumulate a lot of defective proteins - proteins that are not folded properly - throughout their bodies.”
Hayashi then moved on to mice. He found that mice that could not remove defective proteins slept longer, which could have meant that one major role of sleep was to remove defective proteins that accumulated while the organism was awake and active.
“We still do not know why sleep removes defective proteins. However, we have found that sleep temporarily suppresses protein synthesis, which might facilitate cleaning up defective proteins. I imagine there are other factors as well, but for the time being, at least we can be reasonably certain about this one.”
The paper reporting the discovery that animals sleep when defective proteins accumulate in the peripheral tissues was published recently, in 2023. The accumulation of defective proteins is truly one of the hot topics in current sleep research.
Why do animals dream?
Hayashi loved animals since his childhood.
“I especially loved fish. My grandfather taught me how to cast nets and we would often go fishing in the river. I would take the fish home, keep them as pets, and observe them for a few days until the next time we went fishing when I would release them back into the wild.”
He also spent five years, from the first to the fifth grade, in Texas and Rhode Island, where his computer scientist father was working. There, Hayashi had many opportunities to experience the richness of nature, making him even more interested in the living environment. It was a natural progression for him to choose biology when he entered the University of Tokyo.
“I was originally interested in animal behavior and through that, I got interested in the brain sciences. I think most neuroscience researchers start with an interest in human rather than animal behavior, but in my case, it was the opposite.”
While he was working at Kyoto University, Hayashi would go to Lake Biwa with his children to catch fish whenever his time permitted. He smiles as he remembers that he often wondered about the fish caught in their nets, how they slept, or whether they dreamed.
“Why do we dream? I am fascinated by this question, too. The difficult thing about dream research is that it relies on self-reporting. So, dream research can only be done on humans. But there is a possibility that we might be able to do dream research on mice.”
When people dream during REM sleep, their bodies do not move because REM sleep induces atonia, otherwise known as sleep paralysis. This makes Hayashi wonder if REM sleep has evolved for dreaming. He has successfully engineered mice that can stand up, jump around, and move their bodies even during REM sleep.
“There are people who suffer from something called REM sleep behavior disorder which means that they “act out” their dreams, including speaking loudly and moving their bodies. We are now investigating whether the state of REM sleep-related cells in these people is the same as the state of cells in the engineered mice that can move during REM sleep. If we find that they are indeed similar, we could infer that moving mice are “acting out” their dreams as well, so in a sense we can “read” what they are dreaming. Then, we could study brain activity while dreaming in animals which could speed up the process of understanding the role of dreams.”
That is how Hayashi works: he tackles the many unknowns of sleep, one question at a time.
“Being a biologist is a lot of fun. Even though other fields tend towards large-scale research, in biology, many discoveries can be made doing small-scale research. In that sense, it is biology that has a place for individuals to make a big difference. Natural science is about finding new underlying principles. I hope many young people will come and join us at the School of Science."
As for his sleep, Hayashi says he might snore a little.
“My wife says I snore too much.”
※Year of interview: 2023
Interview/Text: Minoru Ota
Photography: Junichi Kaizuka