#147 - Does Sleep Really Clean Your Brain?

0:02Hi there, it's Matt here and welcome back to the podcast. Today, we are going to talk about the glymphatic system, the network that removes the metabolic waste the brain produces each day, and that appears to do much of its work while you sleep. It is one of the best detective stories in modern neuroscience, and it is still being written. But we begin 240 years ago with a man who was right and with what it cost him to be right too early. In 1787, an Italian anatomist named Paolo Mascagni published the first great atlas of the human lymphatic system, the body's drainage, the fine mesh of vessels that carries waste out of living tissue.

0:46Mascagni had mapped it with a scalpel, a lamp, and extraordinary eyes, and in his lifetime, he was the world's foremost authority on the plumbing of the human body. And in among those pages, on the tough outer membrane that wraps the brain, Muscagny described lymphatic vessels. Now, his atlas was in Latin, and it was never translated into English, so for a long while, it simply went unread. But in 1875, a Swedish anatomist named Gustav Retsius put the matter to rest. Mascagny, he suggested, had grown so enamored of the lymphatic system that he had started seeing lymph vessels even where they did not exist, in the brain, for instance.

1:31That was more or less the end of it. A handful of anatomists did report similar structures over the following century and were similarly ignored, but the consensus had set and the vessels on the surface of the brain stayed off the map for the next 140 years. Hold on to Mascagny. We will come back to him twice because you cannot understand why he was dismissed until you understand the thing everybody knew for certain. The thing that made dismissing him the careful, rigorous, sensible choice. Everybody knew that other tissues and organs in your body produce waste. Your muscles, your liver, the skin on the back of your hand, all of them generate cellular detritus, and all of them have vessels to carry it away, clearing out the residue of the day's work.

2:21Every tissue except the brain. We didn't think the brain had such a service crew for cleanup, which is strange given what the brain actually does. Neurons burn fuel at a rate that would embarrass a Formula One engine. Your brain is about 2 % of your body weight and it takes around 20 % of your energy. So picture the most expensive restaurant in the city. Michelin stars, tasting menus, the works, a kitchen running full tilt every night for 80 years and no backdoor, no dumpster, no dishwasher. That was the brain according to the textbooks I trained on. The brain is privileged, we said. The brain is special.

3:04And once a thing has a name, it stops being a question. It stayed quiet for decades, sitting in plain sight inside every anatomy textbook in every medical library in the world. Then a Danish neuroscientist went looking. Her name is Mikan Nedegaard. And at the University of Rochester, she and her colleagues did something so simple, it is faintly embarrassing nobody had done it sooner. They injected fluorescent tracer into the cerebrospinal fluid of living mice. Molecular dye, glow-in-the-dark confetti, essentially. Then they looked down into the living brain and watched where the confetti went.

3:44Based on everything the textbook said, it should have drifted the way a drop of ink spreads through a still glass of water. Slowly, aimlessly, wherever chance took it. But it did not. Instead, it traveled. fluid was pouring into the brain along the outside of the arteries running through sleeves wrapped around the blood vessels as they dive down into the tissue it flushed through it mixed with the fluid bathing the neurons then it swept out along the veins and carried the waste with it a river with tributaries and a current and a mouth at the sea which raises the obvious question and it is the one they asked next.

4:24Rivers need something to move them. So what was doing the pushing? The answer was a water channel called aquaporin-4 studded across the end feet of astrocytes, the star-shaped cells that clasp onto blood vessels throughout the brain. Take the channel away, and in theory, you break the pump. So they bred mice without it. In those mice, clearance of waste from the brain fell by 70%. Compared, that is, with littermates who had the channel and were otherwise identical. Then they ran it again with the one waste product my field worries about most, amyloid beta, the protein fragment that aggregates into the plaques found in the brains of people with Alzheimer's disease.

5:07In mice missing that single channel, amyloid clearance fell by 55 % against the same comparison. So the brain had not been excused from the economy of waste after all. It had simply concealed the plumbing where nobody thought to look. Not a separate system of dedicated vessels, which is what everybody had been hunting for and failing to find, but a system folded around the blood vessels the brain already had, using its own glial cells as the pipework. And Nadergaard gave it the name I used at the top of this episode, which is a small act of genius. Glymphatic glial. for the cells that run it, lymphatic for the job it does.

5:48It had been there the entire time. Somebody just had to inject the dye and watch. But the part of this that reaches out of the laboratory and takes hold of you personally was still a year away.

6:03I had a realization about a year ago. I, perhaps unsurprisingly, go on and on about what happens during sleep for recovery. things like growth hormone release and muscle repair. But I don't think I talk enough about what's happening in your body in the hours before sleep that sets up recovery or sabotages it during the night. And poor circulation, and here we're talking sort of swollen legs and inflammation in the lower extremities, all of that makes it harder for your body to do its job overnight. And especially if you're like me, I can be standing a lot of the time. I can be lecturing, I'm in and around the lab.

6:40I can be standing at my office desk. And furthermore, I do travel a fair amount. And those long haul flights can result in blood pooling in the legs. And as a result, you kind of almost feel as though you get leg claustrophobia, if that's almost a word. I don't think it is, but I think you probably know what I mean. That's why I ended up switching to something called hollow socks. And here's their secret sauce. Their compression socks are made from baby alpaca fiber, which is really soft, breathable. But the critical part is it has a graduated compression that actually promotes blood flow without strangling your legs to that point of suffocation.

7:20And I wore them recently on a 10-hour flight and genuinely forgot I'd had them on. Whereas with the compression socks normally, I always have that sense that they're there. And over 2 million pairs sold and known medically that they are trusted by doctors. They are made in the United States. And for a limited time, Hollow Socks is having a buy to get to free sale. So head over to hollowsocks.com. That's hollowsocks.com for 50 % off. And if you can just tell them that I sent you here and that will help the show a little bit. But overall, just go and buy the socks. They're fantastic. hollowsocks.com

8:04In 2013, a member of Niedergaard's lab named Lulu Shi ran an experiment that sounds on paper like it should not have worked. To study sleep in a mouse brain, you have to let the mouse sleep. And mice do not sleep on command in a busy laboratory with the lights on and a microscope pressed to their heads. Z spent a considerable amount of time teaching mice to relax enough to doze off under the lens. Sometimes the most consequential data in the world sits behind the least glamorous task imaginable, which in this case was waiting for something very small to fall asleep. When it finally did, the mouse brain changed shape.

8:46That is not a figure of speech. As the animals dropped into sleep, the interstitial space, the narrow gaps between brain cells, opened up. Measured against those same animals awake, the channels widened by more than 60 % and through them the fluid surged, carrying amyloid out roughly twice as fast as it had during waking. Picture an office tower at 3 in the afternoon. Every floor occupied, desks pressed together, corridors so thick with foot traffic that the cleaning carts cannot get through. Now it is midnight, the workers have gone home, the floors have emptied, and only now can the crew move down every corridor and carry out what the day left behind.

9:28That is what sleep looked like inside those mice, not an absence of activity, a shift change. This is the finding that usually gets repeated without its species attached, so let me attach it. Everything so far happened in rodents. Mouse brains are marvels. They are not human brains, and the history of neuroscience is littered with elegant rodent findings that failed to survive the journey into us. Still, it spoke to the oldest and most awkward question in my field. Why do we sleep at all? Every animal with a nervous system does it, and sleep is expensive. Hours spent unconscious and defenseless, not eating, not mating, not watching for the thing that wants to eat you.

10:13That cost is a puzzle, not a proof, and I won't run it backwards into an argument. But any function worth that price is a function worth taking seriously, and cleaning the brain is one. Hold on to that too. It is going to get complicated. First, the map needed one more piece. A river needs a mouth. If this system was flushing waste out of brain tissue, where did the waste go? Out of the skull, presumably. But through what? The old dogma still insisted there were no lymphatic vessels up there to receive it. In 2015, two laboratories answered independently, publishing within weeks of each other.

10:56One of them, led by Jonathan Kipnis at the University of Virginia, was not even looking. They were studying immune cells in the meninges, the membranes wrapping the brain. A researcher in the lab, Antoine Louvaux, worked out how to mount those membranes intact on a slide, and what they saw were vessels. Vessels carrying the molecular signatures of lymphatics, running alongside the great blood-draining sinuses, ferrying fluid and immune cells down toward the lymph nodes of the neck. Kipnis has said they checked it again and again before they were willing to believe the textbooks were wrong. At almost the same moment in Helsinki, a team led by Kari Alitalo with Alex and Terry Asperlund as lead author came at the same network from another direction and traced its plumbing out of the skull and into the deep cervical lymph nodes.

11:50They had found lymphatic vessels in the tough outer membrane that wraps the brain, which is to say they had found Mascagny's vessels, the ones the old man described in 1787, and that Gustav Retsius waved away in 1875 on the grounds that Mascagny was seeing things. Retsius was not a crank. He was one of the most accomplished anatomists of his century. He simply looked, did not see them, and drew the reasonable conclusion. So the picture was complete and it was beautiful. Fluid flushes in along the arteries, sweeps through the tissue clearing waste, drains out along the veins and a true lymphatic network carries it out of the skull into the lymph nodes of your neck.

12:36A closed loop, a brain that cleanses itself. That is the point at which the story went viral. And I understand why. Your brain washes itself while you sleep is an irresistible headline. But a good scientist holds two things at once. The beauty of a story and a stubborn willingness to ask whether it is true. In 2019, a young researcher named Nina Fultz in Laura Lewis's laboratory at Boston University found a way to ask it in humans. She put 13 people inside an MRI scanner and had them fall asleep. No small feat, given that an MRI scanner produces precisely the kind of noise that seems engineered to prevent it.

13:18Then she measured three things at once. The electrical brainwaves of sleep, the ebb and flow of blood, and the movement of cerebrospinal fluid. What came back was choreography. As a person descended into the deep slow wave stage of non-REM sleep, great slow tides of cerebrospinal fluid began washing into the brain roughly once every 20 seconds. 13 people is a small study and I would want it replicated at scale before building anything permanent on it. But the tides were exquisitely timed. Each pulse of fluid locked to the electrical slow waves of deep sleep and to a matching pulse of blood. Blood volume ebbed.

14:03Fluid rushed in to fill the space it left. Then the whole thing reversed. A tide chart written in the language of neurons and blood and water, all keeping time. Though be careful what you take from that. Fultz measured fluid moving. She did not measure waste leaving. That is still true today, but it was the bridge my field needed. Something moves fluid through the sleeping human brain in lockstep with the deepest stage of our sleep, the very stage that thins as we age, the very stage that collapses in dementia. Which brings us to why any of this matters beyond the elegance of the biology. If sleep drives the cleaning and the cleaning removes amyloid, an amyloid is one of the proteins that accumulates in Alzheimer's disease.

14:53An alarming possibility comes into focus. Poor sleep may not simply be a symptom of dementia. It may be one of the accelerants. Bad sleep impairs clearance. Impaired clearance lets waste accumulate. Accumulating waste damages the very circuits that generate deep sleep, which degrades clearance further and around it goes. In 2020, Nadergaard and Stephen Goldman argued exactly this in the journal Science, under a title that does not flinch. Glymphatic failure as a final common pathway to dementia. In humans, poor sleep is associated with amyloid burden. Associated, the randomized experiment that would settle causation has not been done and may never be done because you cannot assign a thousand people to two decades of bad sleep and weight.

15:45So we have mechanism in animals and correlation in people and a gap between them that nobody has closed. And amyloid is itself disputed territory. A good many researchers now think the amyloid cascade has been oversold as the engine of Alzheimer's, which means this entire chain of reasoning rests on a link that may not bear the weight. Anyone who tells you a good night's sleep is a vaccine against dementia is selling you something, and that would be a tidy place to stop. Except the science here is still in motion, and the next chapters are the interesting ones.

16:25A long-standing partner of today's podcast is Element, spelled L-M-N-T. Element is a sugar-free electrolyte drink mix. And for me, it was the sugar-free part that made me start using it a while back. And again, I buy my own supply just to stay objective. Beyond having the electrolytes in the right balance, the other reason I'm a fan of Element is that it's scientific. If you look up the people who created the company, you'll find some pretty heavy-hitting physiology, biochemistry, and pharmacology knowledge baked into the product. And again, And all of that comes without any sugar. Now, here's the update.

17:07Element's lemonade salt flavor, which used to be seasonal, was so popular that it's now permanent. You can get it year round. Happily, I badgered them and they have agreed to give you, find folks, a free eight count sample pack with any purchase of Element. If you use the link drinklmnt.com slash Matt Walker. So for free product with your purchase, just visit drinklmnt.com slash Matt Walker. And as I said, you'll get some free product.

17:45The first serious challenge went straight for the mechanism. In 2017, a team led by Alex Smith and Alan Verkman at the University of California, San Francisco, published a direct test. Verkman is not a bystander here. He has spent much of his career studying aquaporins, the very channels at the heart of this story. If anybody had standing to say the pump was not a pump, it was him. They ran their own tracer experiments and concluded that solute movement through brain tissue was explained well enough by plane diffusion, molecules drifting down a gradient, no current required, and that deleting aquaporin-4 did not impair it.

18:29Nadergaard's camp answered in 2018 in the same journal led by Humberto Mestre. They argued the null result came down to methodology, the choice of anesthetic in particular, and a flawed way of reasoning about how a particle's size ought to affect its transport. And then in 2024 came the most striking result yet. A team at Imperial College London, led by Nicholas Franks and William Wisden, published a paper in Nature Neuroscience with a title that leaves nothing to interpretation. Brain clearance is reduced during sleep and anesthesia. Reduced, not enhanced. They injected dye deep into one brain region and tracked how much was still sitting there hours later.

19:17and reported the precise opposite of the 2013 finding. Clearance slowed when their mice were asleep compared with when they were awake. I will note the study used only male mice, which the authors state and defend. And if they are right, part of the story I told you 20 minutes ago has the polarity backwards. Here is where things actually stand. The two camps are measuring different things in different ways. Watching a die enter and spread through a brain is not the same as watching a die finally leave it. The anesthetic can tip the result. So can the injection site. So can the way the optical measurements are corrected.

20:00Some of this disagreement is not about biology at all. It is about the difficulty of measuring invisible fluid moving through a living pulsing brain without disturbing the very thing you are measuring. rather like trying to establish the natural current of a river by wading into the middle of it. What almost everybody still agrees on is the architecture. Cerebrospinal fluid exchanges with the fluid around brain cells. Waste is cleared along vascular routes. There is true lymphatic drainage out of the skull. Muscany was right. The arguments are about the details, and the details are the interesting part.

20:41How fast, by what force, and whether sleep speeds the process up or slows it down. Then, just as the sleep story looked most vulnerable, the plot turned again. In early 2025, a study led by Natalie Hauglund, who has worked in Nadergaard's orbit, appeared in the journal Cell. It pointed at an unlikely hero. Norepinephrine, the arousal signal, the molecular cousin of adrenaline, the chemistry of the jolt you feel when a door slams. You would assume it goes quiet in deep sleep. It does not. In sleeping mice, this molecule rises and falls in slow waves, released in pulses from a small brainstem structure called the locus coeruleus, one wave roughly every 50 seconds, all night.

21:27And each wave makes the small arteries of the brain constrict, then relax. A slow, rhythmic squeezing the researchers call vasomotion. That squeezing, they argue, is the pump. Interfere with the waves and the cleaning falters. Deep sleep, on this account, is not an off state. It is a rhythm, and buried inside it is a slow tide, a heartbeat beneath the heartbeat. It is worth knowing, As you weigh all of this, that a good number of the key papers in this field come from Niedergaard's laboratory or from scientists she trained. That is how young fields often begin. The pioneers do the early running.

22:08And it is also exactly why the independent tests now arriving matter so much. Replication from outside the founding lab is what turns a discovery into a fact. Now back to the 2025 study. The researchers then gave those mice zolpidem, which millions of people know as Ambien. The drug worked in the sense that the mice fell asleep faster. It also flattened the norepinephrine waves and glymphatic flow fell with them. Now, I need to be careful here about scope. This was mice, one paper, one drug. It does not mean you should stop a prescribed medication tonight. If you take a sleep aid, please do not change anything on the say-so of a podcast.

22:51talk to your doctor. With that said, this speaks to something I have long suspected. Sedation is not sleep. A drug can switch off consciousness without delivering everything natural sleep delivers. The lights going out is not the same as the night shift arriving. So where does this leave us tonight? Somewhere more honest than the headline and more interesting. Hold firmly to this. Your brain has a waste clearance system that the textbooks I trained on flatly denied existed. It took a handful of people willing to inject the dye and question the blank space on the map. Hold this nearly as firmly.

23:30Waste is cleared from the brain. That clearance depends on the health of the brain's fluid systems, and its failure is entangled with the diseases we fear most. The entanglement is real, even while the mechanism is contested. And hold this loosely. Exactly how sleep and the brain's cleaning rhythm fit together is still being worked out by very good scientists who do not yet agree. The story may survive intact. It may need rewriting. You are watching science happen, not science finished. And I find I love this story more, not less, for being unfinished. Because consider what it took to lose Muscanyi.

24:11Not one careless anatomist, a book nobody translated, one witty dismissal from a distinguished man who had looked and honestly seen nothing, then generations of careful people who trusted him and had no particular reason not to, and a scattering of others who saw the vessels, said so, and were not believed either. Every one of them doing the job properly, all of them wronged together for 140 years. The vessels were there the whole time. So was the plumbing wrapped around your arteries. So was the tide moving through your head every night of your life, roughly once every 50 seconds, whether or not anybody had a name for it.

24:51Thanks so much for listening. I really do appreciate you giving me your time and your ears. Please take care of yourself. Enjoy your waking hours and sleep a sound sleep tonight. Bye for now.

25:18Pray after by the Relevance of the duet