Why do we need sleep? In their long search for answers, scientists have often uncovered only more thought-provoking mysteries about what sleep is, how it evolved and the benefits that it provides. In this episode, Steven Strogatz — the noted mathematician, author and host of The Joy of Why — speaks with Dragana Rogulja, an assistant professor of neurobiology at Harvard Medical School who recently discovered how sleep deprivation causes death in fruit flies. Then he continues the conversation with Alex Keene, a neurogeneticist at Texas A&M University who studies cave fish to understand more about the evolutionary history of sleep.
Steven Strogatz (00:03): I’m Steve Strogatz, and this is The Joy of Why, a podcast from Quanta Magazine that takes you into some of the biggest unanswered questions in science and mathematics today. Today, we’re going to be talking all about sleep.
Why do we sleep anyway? We spend about a third of our lives asleep, so it seems like it must be pretty important. But there’s still so much about it that we don’t understand. One thing that sleep researchers are pretty sure of is that every system in our body seems to be impacted by sleep. When we miss out on sleep, it impairs our circulation, our digestion, immune system, metabolism, and of course, brain function. And sleep deprivation doesn’t need to be long term to do damage. In fact, if you go without sleep long enough, you will die. But why, exactly?
Dragana Rogulja knows something about that. We’ll be talking with her in a minute. She’s an associate professor of neurobiology at Harvard Medical School. She studies why we need to sleep, and how the brain switches between being asleep and being awake. She also looks at the lethal effects of sleep deficiencies.
Later, we’ll be hearing from Alex Keene of Texas A&M University, who studies the neural regulation of sleep, and the role it plays in the bigger picture of how sleep has evolved. He does that by looking in an unexpected place — in fish that live in caves in Mexico. But first, Dragana Rogulja, thank you so much for joining us today.
Dragana Rogulja (01:35): Thank you so much for inviting me. This is incredible.
Strogatz (01:38): Yeah, I’m super excited to talk to you about your work. But first, I was hoping we could talk broadly about sleep in general. Like, we all understand that, you know, like, I remember asking my mother, “Why do I have to go to sleep?” when I was a little kid. And she said, well, because you’re tired, it’s going to help you get rest. But sleep seems like something really different than just mere rest. Because in sleep, we have this whole altered state of consciousness. How is sleep really different from rest?
Rogulja (02:07): Well, you’re probably sitting right now and kind of resting in some way. But you’re definitely not sleeping, right? So, yeah, what is it that’s so different? And I would say that, for me, what is the kind of most defining characteristic of sleep is that kind of loss of awareness of the external environment and of your internal state, in many ways.
Usually, when we study sleep in humans or other mammals, we do these recordings where we look at electrical activity of the brain, right, and you can see these waves change, and you can’t do that in simple animals that sleep. Yet we know that they really do enter these states where they disconnect. They stop moving, but you can stop moving anyway, right? So, you stop moving, but this is coupled with that loss of awareness, relaxation of the body. And I think it’s a tricky thing, asking why. I think we want to get at the why. But the way that we want to get to the why is by asking how. What are the most primitive things about sleep that we can understand?
Strogatz (03:09): I notice you mention other animals besides people. I mean, we have this very, naturally, human-centric view of what sleep is about, we think about our dreams. But as you say, maybe to get at the how questions we should be looking, possibly, at other animals? Who sleeps in the animal world?
Rogulja (03:24): What we think today is really that sleep is as old as animals themselves. So there are these animals that we refer to as the living fossils, because supposedly, they haven’t changed much throughout animal evolution. And as we look at the simple animals, like jellyfish, and Hydra now, so animals that have very, very primitive nervous systems. It is very clear that they engage in these forms of behavior that I would say, for all practical purposes are really like our sleep. They disconnect, they stop paying attention to what’s going on around them, they can’t respond to external stimulation, unless that stimulation is very strong. So we see that in basically the simplest animals.
Strogatz (03:25): That’s amazing. I hadn’t really heard about this
Rogulja (03:26): The biggest problem to me is that there’s too much focus on the relationship between the brain and sleep because we tend to think of it from our human-centric perspective, like you said, because we dream, et cetera. But if we accept the fact that these simple animals sleep, then we really have to think beyond the brain because we did not appear with these big, beautiful brains, you know, just out of nowhere.
Strogatz (04:33): Yeah, it’s a really interesting point. So, okay, so let’s try to think a little less about brains. Allan Hobson had this remark that sleep is “of the brain, for the brain and by the brain.”
Rogulja (04:44): Yeah, I don’t think that’s right. I don’t think that’s right. I don’t think that’s right. I mean, it is for the brain for sure. But it is also for many other — for all the other, you know, pieces of our body, I think, really. So I don’t think that that’s right. Even saying it’s by the brain or of the brain, I mean animals that don’t have a brain, that have very simple nervous systems, very simple nervous systems, really rudimentary nervous system, they do sleep. So I guess that could be a sort of semantics issue, you know, is it a brain or not a brain, but I don’t even think it’s just for the nervous system.
And actually, we have some evidence now that, you know, it’s not even just of the brain, other places in the body can regulate sleep. I mean, we and others have evidence for that, that, you know, signals to regulate sleep can come from other places in the body. And it’s not just for the brain, we also have evidence for that.
Strogatz (05:35): So much evidence has accumulated now, thanks to work by you and your colleagues. You have made use of a wonderful model organism, we call it from our, again, from our human-centric perspective. I mean, I don’t know if they think of themselves as model organisms: the fruit fly, we know so much about them genetically, developmentally. And now you’re using them to teach us about what sleep might be for.
Rogulja (05:59): Yeah. So it was shown several decades ago that sleep in flies checks off all those boxes. You know, there’s certain criteria that you have to pass to be considered sleeping. And it was shown that this is really the case in flies. They do enter these states where they stay immobile for many hours, and of course, being immobile does not mean you’re sleeping. But again, they do enter this state where they disconnect from their external environment to a large degree. Same thing as happens to us, right? When you’re sleeping, you just — you don’t respond to stimulation.
They’re amenable to all kinds of genetic manipulations. And it became clear that they need sleep, in a sense that if you prevent them from sleeping, bad things can happen and they can die. We’re using flies, primarily, in the lab. And then when we find something that we think is kind of a, you know, an important discovery, then we test those findings in mice, and that does give us more confidence. And so far, things that we have been cross-checking, it’s pretty much the same, you know, and I’ve really come to kind of think that flies are just like us. Plus, they can fly. So it’s very, very cool.
Strogatz (07:08): That’s amazing. I mean, because it’s — not many — other than the movie The Fly, I mean — not many people would think of themselves as connected to flies, but it’s, it’s fantastic the unity of life on Earth, how we can learn so much about ourselves from flies and mice. I mean, we are kind of relatives in a deep way.
Rogulja (07:26): Oh, my god, absolutely. I’ve come to think that we’re all — essentially we’re all the same. The more biology you know, the less can you think, I think, of ourselves as separate from everything else.
Strogatz (07:39): Suppose you do the kinds of experiments that you have done and that people before you have been doing for decades, where you deprive an animal, in this case a fly or maybe a mouse, of sleep, and then ask, if you do that enough, if you make them go without sleep long enough, and they die from it, what exactly killed them? And you have a clue, a very important clue
Rogulja (08:04): Yeah, that’s exactly where we started. So, when I started my lab, this is a question I’ve been interested in for a long time. Why, right? Why do you need to sleep? And “why” is an interpretation, right? We can say, like, okay, this is why it happens. But what we can experimentally, really, show is like, what happens, right? Like how things go.
When I started the lab, a postdoc came to my lab, Alex Vaccaro, who was just ideal for this. And we talked about, kind of, how to approach this question. And we decided to take a new approach where we would be agnostic about the reason for why animals would die without sleep. Exactly trying to stay away from that thinking of sleep is of, for, by the brain. So we just thought, okay, let’s see, if we really deprive flies of sleep, and we try to do it in different ways. So, to have different methodologies, non-overlapping methodologies, and then try to look at their lifespan and see if there’s a certain time that they die. And then, can we find what happens preceding that?
(09:04) We jumped into this with some faith, but I had very little hope, honestly, that we would find something like what we ended up finding, because it just seemed so — I had a feeling, you know, like, all kinds of things could be falling apart, all over the body, right? Even if it’s not the brain, there could be many, many different things happening in the body, and it might be really difficult to pinpoint the exact cause of death.
The first thing that really surprised me was that when you deprive animals of sleep, they crash, they die prematurely. This is in flies, and it was very reproducible when they die. So it really depended on how much sleep they lost. So the more sleep you lose, the faster you die. But if you have different methodologies, which all produce the same loss of sleep, you ended up dying with the same kind of kinetics. So that really surprised me. That happens at a specific time.
The reason why that was important is because it suggested to us that there really might be some specific events, something that we could dig up somewhere in the body. That was the first thing that we were like, okay, if we can find a real correlation between survival and loss of sleep, then maybe we’ll be able to find what’s going on.
Strogatz (10:17): Can I ask you to pause right there? Because you use the word kinetics, and I have a guess what you mean, tell me if this is the right picture. Like, suppose I had 100 flies, and then I start depriving them of sleep, that maybe a certain amount survive one day, and then a certain amount survived to the second day, and so on. And you’re making a graph like that.
Rogulja (10:35): Exactly. That’s right. And then, so what happens is that in the beginning, they all look — everybody’s 100% alive. And then the controls keep on living. And at some point pretty early on, depending on how much sleep you lose. The more sleep you lose, the sooner you crash, these survival curves start going down. So it’s like 80% of sleep-deprived animals are alive, 60%. And so, if we remove all sleep, lifespan can be caught in you know, you live a quarter, your lifespan is a quarter of the control, you know, so it’s a really, yes, it’s a very strong effect.
Strogatz (11:11): That’s brutal. So when you make a graph of this thing that you’re calling the survival curve, the number surviving as a function of the amount of time that you’ve sleep-deprived them, what does the curve look like?
Rogulja (11:22): Yeah, that’s a great question. And something that was really critical in this whole journey. So it seems like that there is a certain point, and that point depends, seemingly, solely on how much sleep you lose, where all of a sudden, these sleep-deprived animals massively start dying. So, under this condition that we looked at first, for example, where the controls live up to 40 days, around day 10, it’s about 90% sleep loss. So around day 10, they start crashing, and by day 20, they’re all dead. So the last of the surviving of these sleep-deprived animals is dead by day 20. And then the controls live to 40 days. And so at day 10 is that inflection point where they start crashing, and that really gave us a window, when to look for bad things that were happening in the body. Yeah, so that was a really critical point.
Strogatz (12:16): And so did you start, like, looking at different organs?
Rogulja (12:18): Yeah, so it was actually very simple. I mean, the idea was very, very simple. What we started from Alex, the postdoc, who started this project, and then Yosef Kaplan Dor, who joined her, another postdoc. The idea was to, now, take all organs that we could take out from the fly, just dissect the whole animal, do pathology on it, so to speak, and look at anything that we could think of, what are some markers of bad things happening? Markers of cell death, markers of DNA being damaged. And so we just looked at everything all over the place. Okay, all over the body. And that was really the critical thing is that we did not limit ourselves to the brain, we just thought, okay, bad things could be happening anywhere, let’s just take all the organs.
And then when we did that, it was actually really, really quick that we got to this surprising answer, which was the bad things were happening in the gut, specifically.
Strogatz (13:12): The gut. That is not obvious. I mean, right? Losing sleep somehow messes something up in the fly’s gut. So, whoa, that’s, that’s really, really interesting.
Rogulja (13:23): Yeah, it was shocking. You know, it’s one of those things that, now, I’ve gotten used to it. It’s been a few years. And it’s just like, yeah, of course, it’s that, you know? But when we first got these results, yeah, it was really weird, you know?
So what Alex did was, one of the things that she looked at was levels of reactive oxygen species. We can talk about that a little bit, these are molecular derivatives of oxygen that are extremely chemically active, they’re very labile. And she saw that exactly around the time when these animals start massively dying, sleep-deprived animals, so there’s a huge, huge, huge increase in reactive oxygen species, specifically in the gut. That was immediately preceding that.
Strogatz (14:04): I’ve heard people talk about free radicals. Is that a different thing?
Rogulja (14:07): So, reactive oxygen species, like the name says, it’s derived from oxygen, very reactive. Free radicals are the most reactive forms of these reactive oxygen species, okay. They’re very, very damaging, but free radicals can be derived, also, not from oxygen, you know, from some other thing. But the critical thing there is that you have an unpaired electron in their outer orbital, in their valence orbital. So that’s the orbital that engages in chemical reactions. And you need electrons to be paired for stability. So these molecules are kind of wobbly. And they attack cellular molecules. They steal electrons, so to speak, from DNA from protein from, from fats, they oxidize them. So this is very similar to rusting, right? Or, like, when you cut an apple, you expose it to air, it gets oxidized. That’s the brown stuff, okay? So you turn these cellular molecules also into dangerous molecules, free radicals, which then attack other things.
Strogatz (15:03): Oh boy.
Rogulja (15:04): Yeah, it was crazy, and so, what happens, what we saw is that you have these reactive oxygen species accumulating, and then you track oxidation of the gut, or what is called oxidative stress. And that’s the thing that I refer to, the fact that you steal electrons from cellular molecules, and you destroy them, and so you — eventually, you see cells dying, massively dying in the gut. And this all happens, and then they die.
Strogatz (15:29): Just to make sure, since it’s been a while since I did chemistry or biochemistry, maybe some of our listeners the same thing. The basic point is, if you, a fly, go without sleep, or are forced to go without sleep for too long, you’re going to build up very abnormally high levels of these reactive oxygen species in your gut. I see them abbreviated as ROS; do you pronounce it as “ross”?
Rogulja (15:49): “Ross”, or reactive oxygen species. ROS, yeah.
Strogatz (15:52): So these reactive oxygen species are kind of like an internal rust or some kind of poison. That’s the point, right?
Rogulja (15:58): Yeah, that’s exactly the point. I think of it as, like, rusting, you know, rusting of a pipe, expose apple to air, it turns brown, that’s oxidation. Oxidation is essentially, you oxidize other molecules means you steal electrons from them. Okay.
So, that’s what happens. That’s what oxidative stress is. And what’s really dangerous about it is that it can propagate, because one molecule attacks another, turns it bad, and that one attacks another. And so this is what we saw. And we saw it in flies, and we saw it with every method of sleep loss that we could think of. And then we checked in mice, and you see the same thing. But the most interesting thing came when we tried to show causation between that and the death that follows. Just because something precedes death, it doesn’t mean that it’s causal, right? It could be correlation.
And so what Alex and Yossi did, and others that, others on the team, was to try to neutralize these molecules. So to get rid of reactive oxygen species in the gut specifically, and then see if this could allow survival, normal life without sleep. And we tried this thinking, okay, it’s a logical thing to do. But did I have any kind of faith that that would work? I would say no, I mean, it really seemed like a fantasy, you know.
(17:17) It was shocking, it was absolutely shocking. We would all gather around and look at these flies every day. I mean, they were just simply fed certain antioxidants, you could neutralize their reactive oxygen species, and they could survive. But when we did this by feeding antioxidants, we were thinking, okay, you — you know, when you eat something you don’t know where that could go and act anywhere in the body, right, just because we’re not seeing ROS in other places, it doesn’t mean that they’re not there. So, then what we wanted to do was to, basically, through genetic manipulations, only do this in the gut.
So you can imagine, if you express an antioxidant, and if you put an antioxidant enzyme — we have these tricks, right? So, antioxidant is something that neutralizes oxidants, like ROS, you only put it in the gut. And then you ask, can this rescue survival? And it does, as long as we get rid of these things in the gut, animals can survive.
Strogatz (18:15): It’s so, it’s so brilliant. I mean, these experiments are so ingenious and sort of so simple that it must make all your colleagues feel like, ah, why didn’t I do that? This field has been — you know, right? But of course, you have the power of genetic engineering now, that’s really been helpful.
Rogulja (18:30): Yeah. I don’t know if they feel like that. But I will say this, yeah, it was very simple. That’s the thing, though, I will tell you. I’ve never been a part of something that went smoother. Like, I’ve never worked on a tougher question, I think, and never been a part of something that went smoother.
Strogatz (18:46): So let me get it. It’s sort of like saying, again, it’s oversimplifying, but it’s sort of like if I did rust remover — not rust remover — but these antioxidants. Just in the gut, that’s enough to save the flies that would have otherwise been dead.
Rogulja (18:58): That’s absolutely right. That’s exactly what happened. And another question that we wanted to understand in the lab is exactly how do you do this sensory disconnect? Like how do you enter this state of sensory disconnect? Why is it that you do not process information the same way when you’re asleep as when you are awake?
And so Iris Titos, another postdoc in the lab, she did this screen, and screen is something where you just try to get rid of a bunch of genes, like, one by one, right? So, we’re looking for manipulations where it would — it would make animals extremely responsive, now, during sleep, or extremely unresponsive, you know, someone who can sleep through an earthquake. And what she found is a signal that originates in the gut. And so this was a completely separate study that also led us to the gut. She found this molecule that’s secreted from the gut, in response to high-protein diet, and that signals to the brain to put you in deeper sleep, to put you in the state of greater sensory disconnect. So that’s an example of where the gut is really dictating the quality of sleep.
Strogatz (20:05): This is wild. This, this reminds you of people saying, after a turkey dinner on Thanksgiving, you’ll have too much tryptophan and you’re going to want to go to sleep.
Rogulja (20:13): In our case, we showed it’s not tryptophan. At least, you know, this molecule is made not in response to specific amino acids, but it’s just sensing how proteinaceous food is. And so for me, the way that I interpret it is, you know, I used to say this thing, which we all say, working in sleep, it’s like, “It’s dangerous. Why would you be in that state, you can’t do anything?” But I actually think you’re better off if you had a good meal, you’re better off hiding somewhere, and not moving, and sleeping, right? So you can essentially afford to disconnect and sleep deeply. And you don’t have to run across the field and look for food and expose yourself to danger.
Strogatz (20:52): Huh, very interesting. So sleep is really tied to digestion, or something.
Rogulja (20:58): Yeah, I think so. I think so.
Strogatz (21:00): Well, that would explain why every animal up and down the, whatever, the tree of life, is going to need to do this, we all have to eat. Everyone knows eating is important. Metabolism is important.
Rogulja (21:11): Well, there’s also something else, you know, like the gut is one of the first organs that, that appeared in animal evolution. And I used to think of it as — before this study, I would not have been fascinated by the gut because you just think, oh, you eat to survive. But think about, like, everything that you’re made of essentially, it has to come through your gut. You have to eat it, extract some molecules, turn it into yourself.
This is like a place close to the middle of your body where wild things happen. You have to break down tissues of other animals, of plants, of different things without harming yourself. It’s nothing else that you can think of in your body is exposed to anything approaching what your gut is exposed to.
Strogatz (21:52): That is a fascinating note to end on. Thank you so much for this really enlightening and delightful conversation, Dragana.
Rogulja (21:59): Thank you, Steve.
Announcer (22:05): The Joy of Why is a podcast from Quanta Magazine, an editorially independent publication supported by the Simons Foundation. Funding decisions by the Simons Foundation have no influence on the selection of topics, guests or other editorial decisions in this podcast, or in Quanta.
Strogatz (22:26): My next guest is Alex Keene. He’s a neurogeneticist who studies sleep. He also heads up the biology department at Texas A&M University. Alex has studied the molecular basis for sleep, and memory formation, in fruit flies. And he’s looked at Mexican cave fish to gain a better understanding of the neural mechanisms of sleep, and how they’ve evolved. Alex, thank you so much for joining us today.
Alex Keene (22:51): Thanks so much for having me.
Strogatz (22:52): This is great. I’m dying to hear about the cave fish. But maybe first, we should begin with a sort of broader question about the role of model genetic organisms in the study of sleep. What is that all about?
Keene (23:05): There is, I think, a revolution that happened about 25 years ago. Before that sleep was largely studied in mammalian systems. But circadian rhythms, the 24-hour patterns were known to be widely conserved. And a lot of the work was done in very simple models. And so that community that really studied circadian rhythms, led by Paul Shaw at the Neuroscience Institute, and Ahmita Sehgal and Allan Pack at Penn had the idea to bring those experiments into fruit flies, which were this simple model with powerful genetic tools, where you could do gene discovery. And since then, it’s taken off in a number of other systems, zebrafish, C. elegans, these simple worms, and they’ve used them to find out some fundamental features of sleep that we now know are very conserved with human sleep.
Strogatz (23:51): I think going back to the 1970s, people were trying to look for the genetic basis of circadian clocks in flies. And you’re saying, because that’s a, kind of an adjacent field to sleep, it’s sort of the same strategy except, what, we’re now looking for genes that might regulate sleep or underlie sleep in some way?
Keene (24:09): Yeah, so circadian rhythms are kind of a clean, straightforward thing to study, because you know almost every animal on earth has this 24-hour clock in their head. Whereas sleep was seen as, I think, much more complex and harder to define. But what the field had gotten really good at is recording animals’ behavior. And when these groups started to look really closely at fly sleep, they could find all these behavioral characteristics that we use to identify sleep in mammalian systems, like, an increased arousal threshold or a species-specific posture that animals go into in their sleep, were taking place in flies too. And then they could transfer a lot of these approaches that had been used to find the genes and neurons involved in circadian rhythms into sleep. And that, I think, really started the field and it’s just been rocketing forward ever since.
Strogatz (25:00): Just, sort of, full disclosure, my Ph.D. was about sleep. Although I’m a mathematician, I worked on the math of human circadian rhythms and sleep cycles. But it was about 35 years ago. And so I’m a little bit of a Rip Van Winkle here. I don’t work on this anymore. So it’s almost like you’re waking me up from a big sleep and telling me how the field has progressed. If a person really believes in evolution, you know, and that life has evolved here on Earth, presumably, from a last universal common ancestor, all of us around today. From an evolutionary perspective, you might expect, of course, lots of organisms have something like sleep.
Keene (25:37): Yeah. And I think, I mean, sleep might be unique in terms of why people thought this, and because the sleep was always thought to be an emergent property of many different neurons, right, and you needed a complex ensemble of neurons to generate sleep, so, you know, the mammalian brain has millions or billions of neurons. And so it’s not something that anyone ever thought you could localize to an individual cell. But now, I mean, I think with the finding that, you know, an animal with 300 neurons can sleep is broadly accepted, I think we need to rethink kind of what sleep is. What is the minimum number of neurons that you need to enter a sleep-like state is a big question.
Strogatz (26:18): Oh, I was just gonna ask you that. Is C. elegans as far down as we’ve gone?
Keene (26:22): Yeah, I mean, I think C. elegans have a very primitive nervous system. There’s also Hydra and jellyfish that have been identified or characterized for sleep. The really remarkable thing about the work in C. elegans and Hydra is, it’s not just that they have these behavioral characteristics of sleep, but a lot of the molecular machinery involved in sleep appears to be conserved. So the same genes, from this little microscopic worm all the way up to humans, are involved. For example, if you give Hydra melatonin, they enter a sleep-like state.
Strogatz (26:53): Hm. Well, maybe we should now shift gears to go deep into caves in Mexico. You have done a lot of work, you and your group, on Mexican cave fish. Tell us a little about them, and why they’re so interesting to study in connection with sleep.
Keene (27:08): This is a really remarkable, I think, biological system. If you go into northeastern Mexico, there are these surface fish that live in rivers, and they look like normal fish. They have eyes, they’re pigmented. And then, there are 30 different caves around — basically within a county of each other, 50-kilometer area, and they have cave fish that have been trapped there. And so, they’re still the same species, they can still interbreed if we take them out and put them in the lab. But these cave fish look completely different. And the really cool thing is, they’ve lost their eyes, and they’ve lost their pigmentation, but it happened independently in each cave. And so scientists have studied these fish for almost 100 years to figure out how evolution happens repeatedly.
What’s happened recently is we’ve grown to appreciate the differences in their behavior and their physiology. So for example, when I was a postdoc at NYU, we looked at sleep differences in these cave fish. And the impetus for that is there’s lots of speculation that the reason there’s variable sleep throughout the animal kingdom is because the animals have different foraging needs. So for example, if you’re a large mammal like an elephant, you need to eat most of the day. And for that reason, elephants only sleep a few hours a night. And so we knew that food was poor in the caves. And so we tested sleep. And we found that these cave fish had repeatedly evolved to sleep much less. And that was kind of the start of using this organism as a genetic model to look at sleep.
Strogatz (28:31): This idea of being trapped in a cave, I want to understand that a little better. It’s something like what, hundreds of thousands of years ago, there was some kind of geological phenomenon? What, flooding or something that caused them to get trapped?
Keene (28:44): Yeah, so we think that these fish got trapped a few hundred thousand years ago, in all these different caves. The idea is the fish got stuck in there. And then over time, you know, what was needed to survive in a cave is very different than what was needed to survive in a river. And one of the things that we were particularly interested in is sleep. And it looks like what we know now, they sleep less, and our hypothesis is that’s because there’s limited food and if there’s limited food, when you’re asleep, that’s time away from foraging.
Strogatz (29:12): So what, more broadly, are these cavefish telling us about sleep?
Keene (29:16): One of the fundamental questions I think we are trying to address is, do they need less sleep? Or are they chronically sleep depriving themselves? Because one idea you could think of is that they’ve become more efficient sleepers. And it’d be fascinating to know what a brain is like that sleeps so efficiently that it only needs a few hours’ sleep.
More importantly, we can look at the differences in sleep between each population of cave fish. And understand how genetic variation leads to sleep differences. And that’s really important because sleep in humans is incredibly variable. Some people need five hours of sleep. Other people need eight hours of sleep. You can have two people who are insomniacs, but for completely different biological reasons. And that’s really important because the way you might treat that insomnia is different based on the underlying biological problem. If we can go beyond looking at sleep in a few model systems that have been highly inbred in the laboratory, in leveraging the power of this variation that’s out there, that evolution’s created, it’ll tell us a lot more about the variation between individual people, and I think that’s ultimately necessary to treat sleep better, and to understand how it’s working.
Strogatz (30:29): Let’s see if I get this, then, about the independent evolution. I mean, that aspect is very interesting, that you say there’s these 30 caves, they don’t have, to first approximation, maybe almost no interaction with each other. So that when you speak of the fish evolving in this cave, starting a few hundred thousand years ago, or the other cave, they’re like separate experiments. They don’t have any particular contact, and yet, they’re all evolving so that they don’t have eyes, they sleep less, they forage more. Is that what you’re saying?
Keene (30:59): Exactly. And what’s so cool about this system is that we can take them into the lab, and we can breed them and study them. So one of my favorite experiments, and this was done in the 70s, is you take two populations of cave fish from different caves, neither of them have eyes. But if you cross them together, the progeny have eyes. Because it’s — it’s different genetic pathways that lead to the eye loss in each one. But then when you cross them together, each of the offspring get one functional copy from each parent, and then their eyes reappear.
Strogatz (31:33): This is like classic textbook stuff out of high school biology. Little a, big A.
Keene (31:38): Yeah, and it’s amazing when, when you can see it in the lab. And we have some evidence that the way that these fish lost sleep is very different. So for example, we can give one drug that will restore sleep in a population of fish, but have no effect in a different population. And so it’s not a giant leap to translate that to people. Why is Ambien more effective in some people, but not in others, right? And there are probably differences that are leading to sleep loss. And if we can understand why this is happening, we have much more targeted sleep medicine, which is probably the ultimate goal.
Strogatz (32:13): Hmm. Is there such a thing as an insomnia gene? I mean, are you trying to identify genes, maybe even single genes, associated with different kinds of insomnia?
Keene (32:23): Yeah, so, it’s funny, there’s a gene called insomnia that was identified in flies, and the mutants have very little sleep. But I think, more importantly, there are lots of genes that impact sleep. And there lots of different neurons that impact sleep. And that’s what makes this such a complicated system to work with. With circadian rhythms, if we want to have a comparison, there’s a clock in your brain. And we know where that central clock is located. It’s located in the hypothalamus. But I think with sleep, there are sleep circuits that are diffusely located all over your brain. And so you can imagine if there are all these different inputs, and you strengthen one and you get sleep and you strengthen another, and you get wakefulness, there are lots of different ways to impact the system. And so I think there are probably hundreds of genes that impact sleep, and hundreds of different sets of neurons, if we look at the human brain, and so figuring that out is going to be really complex.
One of the things we found in cave fish is that the gene hypocretin, also called orexin, which is thought to be a main regulator of sleep in mammalian systems — narcoleptics have reduced hypocretin signaling, for example — and this gene varies by cave and cave fish. And so cave fish with really short sleep have higher levels of this wake-promoting gene. And so what we think is, it’s possible that variability in hypocretin levels between individual people is contributing to our sleep differences.
Strogatz (33:50): Let me underline that last point, I want to make sure I got that. You’re saying there’s a gene that makes — hypocretin or orexin, you said — are those proteins?
Keene (33:59): So it’s called hypocretin and orexin. It’s a neuropeptide, so it’s released from neurons and binds to a receptor. And it was first identified in these really classic studies in dogs from Emmanuel Mignot’s group at Stanford. And they had these narcoleptic dogs, and he was hunting for the gene that caused narcolepsy. And he found a mutation in the receptor for hypocretin. And it caused the dogs, when they would get excited, to essentially fall asleep with cataplexy, so they wouldn’t move, they’d be paralyzed.
And then later it was found that humans with narcolepsy often had a loss of hypocretin neurons in their hypothalamus. Since then, there’s been a lot of focus on this gene. And so, we just found in cave fish that variable levels of this gene track with how much the fish sleeps. And so, it’s possible that this translates to humans as well. And, you know, really good sleepers might have lower levels of hypocretin. And poor sleepers or insomniacs might have higher levels of this gene. But like I said, it’s probably a complex process, but that’s just one idea of how you could take variable levels of a gene and use it to understand human biology.
Strogatz (35:09): It’s just astonishing, the unity of biology here, that there’s this gene that we have, that dogs have, that your cavefish have.
Keene (35:18): I think what’s amazing to me is that this acceptance that sleep has really old evolutionary origins is recent. In the history of the field of sleep studies, 20 years is not a long time, or 25 years, for people to have made this shift. But I think now that there’s this acceptance, I think there’s just going to be so much knowledge that’s gained when we can take these simple organisms, relatively simple organisms, and apply what we learn to more complex brains.
Strogatz (35:46): One fairly recent discovery that got lots of buzz, I think deservedly, was that zebrafish have two patterns of neuronal activities. One that seems kind of like REM sleep, the sleep that we associate with the dreaming state, in people and in mammals. Why was that discovery about these different patterns, roughly like sleep stages, so important in terms of understanding sleep and evolution of sleep?
Keene (36:13): Yeah, I think this is a really remarkable study from Philippe Mourrain’s lab at Stanford. And the reason is, it’s always thought that, or at this point, I think it’s accepted that these simpler non-mammalian models sleep. But structurally, their sleep can’t be like human sleep, that we have sleep made up of stages, light and deep sleep, REM and non-REM. And what Philippe Mourrain’s lab did is they took advantage of the fact that you can use genetic tools to record the activity of the brain in these zebrafish. They’re transparent when they’re really small, and so you put them under a microscope. And he recorded their activity. And he found that the activity pattern in this area, the dorsal pallium, which is the evolutionary precursor to the mammalian neocortex, had these REM- and non-REM-like sleep patterns. He then used further drug treatments to find that the same drugs that would induce REM-like sleep would induce REM-like patterns in zebrafish. That really showed that there’s kind of a direct line. Not only in the behavior of sleeping, but in the structure of the brain’s activity during sleep, from non-mammalian models all the way up to humans.
Strogatz (37:23): So also, you’ve been looking at how sleep evolves in response to challenges from the environment. So what kinds of environmental challenges are you looking at, and what are you finding from that?
Keene (37:35): We’ve been looking at this, I think, in flies and fish, and I think this is one of the more underexplored areas of sleep. We tend to talk about sleep — I mean, we, the scientific community in general — as a unified process, you sleep or you don’t, but sleep changes based on our internal and external state. So, for example, animals ranging from flies, all the way up to humans, will change their sleep when they’re food-deprived, right, because if you’re really hungry, you probably want to go find food instead of taking the time to sleep. Stress impacts sleep, aging impacts sleep. And I think what we’ll find is the genes involved in sleep differ in each one of those conditions.
And one of the ways I think about this a lot, that might be hard to address directly scientifically, given the tools we currently have, is, I think, sleep loss — the impact of it might vary greatly, depending on the context. If you stay up all night and don’t sleep because you’re really stressed out, that might affect the brain very differently than if you stay up all night because you’re having fun watching movies on TV or hanging out. And so, what is the biological basis of that? I think it’s really important to understand that.
Strogatz (38:42): That’s an interesting thought.
Keene (38:43): And there’s evidence for this, for example — so, our lab and others have found that if you starve a fruit fly, and it loses sleep, it doesn’t need to rebound from that. So it doesn’t need to make up that lost sleep. But if you were to sleep deprive it by mechanically shaking it, it would need to make up that reduced sleep. And the way I think most of us interpret that is, it’s evolutionarily built-in to find ways to compensate for sleep, when it’s part of the brain’s wiring. But when it’s not, when it’s something like mechanical shaking, the brain’s unprepared to do that. And so, there’s a consequence, which is you need to sleep more. So, I think that just highlights the need to look at sleep in all these different contexts, not just a comfortable animal that’s sitting there in its home arena.
Strogatz (39:28): What do you feel like we know about how sleep evolved? And what is it that remains to be learned?
Keene (39:33): We know now that genetically sleep is very conserved, and even at the neural level, from the simplest models, all the way up to humans. I think the two big things we don’t know are how variation in genetics leads to sleep differences. And I say this especially in people, but I think we can learn about this through other models.
So for example, there have been huge, they call them GWAS studies, where they look at the DNA of individuals across huge populations, and most people are probably familiar with this through, from companies like 23andMe. These have also been done for sleep, and what they come up with is lots of hits, lots of genes that are associated with sleep. But it’s hard to verify what those do. And if we could understand how that variation between us impacts sleep duration and quality, I think that’s really important. And then I think that from the evolutionary perspective, the really big thing we don’t know is, what is it about different environments that drive sleep differences?
So I can tell you, the elephant sleeps only a few hours a night and the armadillo sleeps 18 hours a night, but what is it about their behavior, about their evolution that caused those dramatic differences, that made our cave fish lose sleep? And it’s really tricky to study, because it’s hard to go back in time, we can’t go back in time, and trace how their sleep changed over time. It’s not like the fossil record where you can look at changes in bone structure, let’s say, like we’ve done for early hominids. And so what we end up doing is comparing a lot of existing species. And it’s a challenge, but I think it’s, it’s achievable.
Strogatz (41:09): Thinking about sleep evolution research, is there one question that you could identify that you’re most excited about?
Keene (41:16): I think I’m most excited about this idea of linking the animal’s natural ecology to sleep itself, because I think that’s a tangible one. More and more animals are being used to study sleep now, and a lot of that’s being driven by the development of CRISPR. That gives us the ability now to manipulate genes in neurons in just about any model we look at, when before that was limited to a small subset of biological models.
So I’ll give you an example. We started looking at sleep in fish, these cichlid fish that you commonly see in pet stores, from Lake Malawi, in Africa. And there are hundreds of species of fish that all share the same lake and live in close proximity to each other. But they have really different sleep patterns, and sleep durations. And so, we think we can use that, just like we’re using our cavefish, to understand what is it about each individual species, how they forage, how territorial they are, how they’re preyed upon, that links to sleep. And I think this is true for lots of different model organisms. You know, the more we study sleep in different animal models, the more we’ll learn about the links between their ecology to how much they need to sleep.
Strogatz (42:26): Wow, that’s a fantastic note to end on. Thank you so much for joining us today, Alex.
Keene (42:32): Thanks so much for having me. This was really fun.
Announcer (42:38): If you like The Joy of Why, check out the Quanta Magazine Science Podcast, hosted by me, Susan Valot, one of the producers of this show. Also, tell your friends about this podcast and give us a like or follow where you listen. It helps people find The Joy of Why Podcast.
Steve Strogatz (43:01): The Joy of Why is a podcast from Quanta Magazine, an editorially independent publication supported by the Simons Foundation. Funding decisions by the Simons Foundation have no influence on the selection of topics, guests, or other editorial decisions in this podcast, or in Quanta Magazine. The Joy of Why is produced by Susan Valot and Polly Stryker. Our editors are John Rennie and Thomas Lin. Our theme music was composed by Richie Johnson, and I’m your host, Steve Strogatz. If you have any questions or comments for us, please email us at firstname.lastname@example.org. Thanks for listening.