AI Gardner and his brother-in-law built the house in Mount Kisco, New York, back in 1984—two stories, three bedrooms, with a sweet little porch overlooking a sunny backyard. At the time, Gardner had worked in construction management for years. He had, in fact, been mechanically inclined ever since he was a kid, when he helped refurbish a Beaver tractor his dad bought from a neighbor. But the house was the first he ever built from scratch, and he was proud of it. Not many people could say they’d built a house for their family these days. Al Gardner could.
Al has a hard time walking up the stairs to his home’s second floor these days, so he lives on the first. In a lounge chair, surrounded by pictures of his family and the homes he built, he slowly, carefully crosses one knee at the ankle like he’s in a business meeting. His legs are thin and pale and papery. His face, too, has taken on a gauntness since the photo of his daughter’s wedding, mounted on the wall right in front of him, was taken back in 2009. Al lunges forward as if he might stand. But then, when he tries to say hello, all that comes out is a guttural moan. When Al, who is sixty-eight, was diagnosed with progressive supranuclear palsy (PSP) in 2012, he was not guaranteed even this. The disease, caused by degeneration of cells in areas of the brain associated with movement, balance, and thinking, often results in death in about seven years. It has no known cause and no cure. Think of it like Parkinson’s disease but faster, and more horrible. L-DOPA, a drug that can reduce symptoms in Parkinson’s patients and help them move, usually has no effect on PSP. Apart from an aspirin, an antacid, and something for bladder control, Al doesn’t even take any drugs. There aren’t any to give him.
Al’s wife, Fran, ruffles his hair. He stares straight ahead. Al can no longer blink or move his eyes, which Fran says is the worst of it. He has to wear sunglasses just to go upstairs. For now, Al can still communicate in writing: Last week he had a sinus infection, and hadn’t been able to make it out of the house for his usual appointments. On the whiteboard he uses to communicate, he wrote a single word: bored.
As the losses mount, Fran has written out an affirmation she can recite when she needs it. She has joined support groups and is active in the PSP community. She smiles as if making the motion is all that’s keeping her afloat. “We’ve been fortunate to have this time,” she says, upstairs, in the kitchen, where Al will not hear. “It could be a lot worse.”
Washington, D.C. The Neuroscientist Who Wants to Turn Support Cells into Neurons
A disheveled guest emerges from the elevator of the Washington Plaza Hotel in Washington, D.C., carrying a bag of laundry to drop off at the front desk. He looks out the window, where a giant picture of a brain lumbers past on the side of a city bus.
“Are you here for the neuroscience conference?” the man asks a stranger standing next to him. The bus departs, revealing packs of neuroscientists making their way around Thomas Circle, black poster tubes for presentations slung across their shoulders like quivers. “It just seems like everyone is here for this thing,” the man says. “I’m just trying to figure out how far it goes.” Far. This weekend marks the beginning of the Society for Neuroscience’s annual meeting, which attracts more than thirty thousand professors, doctors, graduate students, and postdoctoral researchers from more than eighty countries to discuss the future of the human brain. SfN, as the conference is called, is so enormous that only seven cities in the United States can even accommodate it.
Inside the Walter E. Washington Convention Center, the featured lectures have already begun in the auditorium, which is so capacious (and dark) you could play a game of Marco Polo in it with your eyes open. Up next is Magdalena Götz, a professor who has flown in from Ludwig Maximilian University of Munich to give a talk about treating brain injuries in mice. Her face, complete with blunt, Germanic bangs, is briefly duplicated on a giant screen like an evangelical preacher’s.
There are many reasons that progressive supranuclear palsy, the disease Al Gardner has, is hell on earth, but they can all be traced to one: Generally speaking, neurons don’t grow back. With a few exceptions, when the brain’s primary information-processing cells die, they’re dead. So today, when a doctor encounters a neurodegenerative disease or a brain injury, the strategies are limited: one, do your best to keep the rest of the neurons alive; and two, encourage the brain to work around any sections that are damaged. If someone could persuade neurons in human patients to spontaneously regenerate, it would be one of the most incredible achievements in neuroscience. For now, it remains impossible.
Götz doesn’t study neurons. Or, at least, not at first. She works on another type of cell, called glia. Glia (Greek for “glue”) comprise at least half of the cells in the brain, but scientists thought they were just a supporting framework for neurons for more than a century. Then, in 1990, a Stanford researcher named Stephen J. Smith discovered that a particular type of glia, star-shaped cells called astrocytes, could communicate with each other. It started a race to figure out what these strange cells did. The list keeps growing.
“When an injury strikes, astrocytes become activated,” says Götz. They can kill more neurons, or they can help keep them alive. They can regulate inflammation and control how neurons reconnect after their networks have been decimated. Some stick around and form a scar. Astrocytes are important all the time, but after a brain injury, the scaffolding runs the asylum.
Here’s why any of this matters: By injecting certain proteins (called transcription factors) involved in development directly into the brain, Götz and her team in Munich have figured out how to alter the function of astrocytes after an injury. Like really alter it. Instead of building useless scar tissue, Götz’s astrocytes transform into brand-new neurons to replace the ones that were lost. Götz has done this to human cells in a dish, and she’s done it in living mice. Her team has even convinced the reprogrammed neurons to send little feeler projections out to the places they should go. Now, nearly a year after her talk in Washington, D.C., she has partnered on a review article with a doctor who sees Parkinson’s patients and is working on ways to deliver the transcription factors to mice through oral drugs rather than brain injections.
“Predictions about how long something will take [to be available for humans] are notoriously wrong,” Götz says. “This is how much I can say: We did this for the first time in a living animal in 2005 and that was considered a complete blue-sky approach. Only fifteen years later we’ve reached a stage where clinicians are interested.” That doesn’t mean that fifteen years from now doctors will be able to prescribe a course of transcription factors to cure PSP. But then again . . .
The auditorium is only about half full for Götz’s talk, which seems incredible when you consider the potential impact of her work. Reprogramming support cells to cure brain disease! But SfN is hosting thousands of presentations over the next five days. It’s a halcyon Saturday in Washington, D.C. And in neuroscience, amazing things are happening everywhere you look.
Seattle: The Philanthropist Who Wants to Figure Out How the Brain Works
Nick Dee and Herman Tung, two other members of the Allen Institute’s tissue processing team, use a vibratome to slice the human brain sample.
Past the double doors, Jeffrey Ojemann, a brain surgeon from a family of them (his father, uncle, and brother are neurosurgeons, and his mother is a neurologist), stands over the exposed cerebral cortex of a twenty-three-year-old woman whose epilepsy has become unmanageable. It’s likely that the woman is awake: Ojemann often wakes up patients at this point in the procedure, to apply an electrical current to the brain while the patient names pictures. He wouldn’t want to accidentally remove an irreplaceable chunk of neurons on his way to the knob of tissue that has been causing the woman’s seizures—a spot known as an epileptic focus. Ojemann makes a cut in the side of the woman’s temporal lobe, which is above the ear, avoiding what’s known as the eloquent cortex, parts of the brain that are generally understood to allow people to move, hear, speak, and see. He tunnels in to reach the focus. He will do his best to remove as little as possible.
About twenty minutes later, a young woman wearing dark blue scrubs with a mask drawn down around her neck emerges from the double doors with a lurid pink and white marble in a jar. It’s the manhole cover Ojemann cut from the outer layers of the woman’s brain to get down to the epileptic focus. The marble is healthy, normal brain tissue. Unfortunately, once it was severed from the neurons surrounding it, there was no putting it back. “Sorry it took a little while,” the woman says, handing over the jar. “But it’s literally brain surgery.”
Tamara Casper places the jar in a Styrofoam cooler packed with ice and rolls her cart out into the street. She loads it into the back of an anonymous white courier van, which slips into Seattle’s afternoon traffic. Within fifteen minutes, the marble has been received at the Allen Institute for Brain Science—founded by eccentric Microsoft cofounder and Seattle-based philanthropist Paul Allen—where it will become a permanent part of the first-ever cellular map of the human brain.
Many people think that because neurosurgeons are able to operate on the brain, there must already be a map of how it works, and that is true, to an extent. A structural diagram of the brain has existed since the early 1900s, outlining the regions where cells appear different under a microscope. But most of what doctors know about the function of each of those pieces comes from presurgical electrical recordings, like the ones Ojemann takes; or functional magnetic resonance imaging (fMRI) studies, which show how blood flow changes while people do tasks; as well as from a whole lot of doctors accidentally removing parts they shouldn’t. There is still a lot of empty space on the map. Especially when compared to the rest of the human body, the brain is virtually uncharted.
In 2003, Paul Allen learned that his mother had Alzheimer’s disease. Already obsessed with computers, and determined to donate most of his fortune to charity, he became fascinated with the brain, earmarking an initial $100 million to found an institute that could do the comprehensive, labor-intensive work that would be required to figure out how it works. It would be like Great Britain’s Royal Geographical Society—the world’s first shared observatory for brains.
Over the years, the Allen Institute has used mice and the brains of cadavers to create atlases of where various genes are expressed in the brain. They’ve mapped the spinal cord. They’ve mapped primate brains. In accordance with Allen’s instructions, all of the diagrams and data are available to the entire neuroscience community for free. But even that hasn’t been enough to explain how an organ can process information. Eventually, the Allen Institute’s staff started to wonder if developing a periodic table of brain cells would help researchers figure it out. Just how many different kinds were there?
The short answer: probably more than a thousand. Since 2015, the institute has been working on the first-ever taxonomy of brain cells, sorting them by their electrical activity, the genes they express, and their morphology (how they look). They started with living cells from mouse brains, which are easier to get, but have recently moved over to humans, partnering with six local doctors to pick up leftover cells from surgeries that would otherwise be thrown away or maintained in hospital tissue banks. Christof Koch, the Allen Institute’s chief scientist and president, compares the work to mapping the genome, which has radically transformed medicine since it was completed in 2003. “Today, nothing in biology makes sense anymore without knowing the involvement of genes,” he says. “The same thing is true of cell types. It is going to be an absolute, necessary next step to understand who we are.”
After the sample Tamara Casper collected from Harborview Medical Center arrives at the Allen Institute in the white courier van, she and a team of technicians meticulously slice it and hand it off to researchers who will probe the cells within before they die, which can take anywhere from several hours to three days at the outside. It’s an all-hands-on-deck situation. Some of the staff will remain at the center until one or two in the morning, painstakingly selecting cells that look hardy and poking them with vanishingly tiny glass pipettes to zap them with electricity and record their response. Thankfully for the technicians who perform this work, human cell samples aren’t an everyday occurrence. The Allen Institute’s partnerships net them only about forty a year, each of which is portioned for a half-dozen teams.
But even during the human sample frenzy, the cell census is not the only ambitious project underway at the Allen Institute. On the first floor, a group known as the electron microscopy team is disassembling a cubic millimeter of mouse brain (the size of a grain of sand) into twenty-five thousand slices, taking 250 million microscopic photographs of those slices, and then reassembling the photos into an interactive Google Earth–style street view that will allow researchers to trace a billion connections between roughly a hundred thousand neurons.
Already, the crew has completed an early prototype, and it is incredible, like the first map scribbled by a team of conquistadors returning from South America. Click on one neuron and the software zooms in, showing you everything it is connected to and how. Astrocytes, it turns out, actually look more like sea sponges than pointed stars. Neurons called chandelier cells connect axon to axon, which is weird. Pyramidal cells shoot one thick dendrite up to the brain’s surface, like a periscope. Looking at it, you can see a future in which the cell census, combined with an interactive mapping tool like this, could lead to the kinds of science-fiction tools curing brain diseases will require—cellular surgery, remapping the cortex, even rewiring a damaged brain using Götz’s astrocyte-neurons.
The name of this second project is MICrONS, short for Machine Intelligence from Cortical Networks, and, in addition to Paul Allen’s generous grant, it is supported by $18.7 million from IARPA, the U.S. intelligence agencies’ high-risk, high-reward research program. It is also working with machine-learning researchers from Google. Certainly, IARPA and Google find the medical applications of MICrONS compelling, but their primary interest is in reverse engineering the brain’s information-processing setup to develop ever more powerful machine-learning algorithms. These could, in turn, help the Allen Institute decode more complex strategies the brain uses to process information, forming a feedback loop of computational progress that either ends in an exhaustive characterization of human intelligence and the end of brain disease . . .
Or the rise of sentient death machines, depending.
Baltimore: The Video-Game Designers Who Want to Improve Stroke Care
I Am Dolphin, a video game developed at Johns Hopkins University School of Medicine, retrains stroke patients to move their arms. The machine above helps counteract gravity.
t is impossible to eat the darn fish. I rock my right hand forward and backward and the dolphin on the screen, mimicking me by way of a hacked Xbox Kinect, halfheartedly starts forward and then flops over on its back. I swing my forearm around in a circle, which makes it swim away from me, but upside down, in a way no real dolphin would deign to move. Finally, by undulating my shoulder like I’m playing an octopus in a modern dance piece, I manage to get the dolphin to swim up to a fish, open his mouth, and chomp it. And this is as easy as this game gets.
“There’s about 110 levels in here,” says Promit Roy, the bespectacled software architect who built this game from scratch on a programming engine he also built from scratch. “As you start to get up there, the fish get faster and smarter. And there are sharks.” Sharks?
“They’re going to start to attack you.”
Roy, who used to work for Microsoft and Nvidia and helped ship the game Fracture for Xbox 360 and PlayStation 3, is one of three founding members of the Kata Design Studio at Johns Hopkins Stroke Center in Baltimore. Along with Omar Ahmad, who has a Ph.D. in computer science, and John Krakauer, director of the Brain, Learning, Animation, and Movement Lab at the Johns Hopkins University School of Medicine, he has created one of the most advanced therapeutic video games in the world.
While the neuroscience community painstakingly researches treatments that may be able to physically repair brain injuries and neurodegenerative diseases, most current therapy encourages the brain to rewire itself after it has been damaged, which can be remarkably effective. But it takes time, and an enormous amount of effort. Overall, rehabilitation is a devastatingly boring affair. “There is severe depression across much of the patient population. There’s the question of: Do you even want to do your therapy? No one wants to do their therapy,” says Roy. Many patients register the particulars of their biological catastrophe—the alarming brain scans, the bleak recovery timelines—and then mentally check out.
Roy, Ahmad, and Krakauer identified the main problem as one of motivation. If patients believed they were doomed to a life of approximations—of struggling to lift a cup to their face over and over again, which is what some patients do in occupational therapy—why would they invest the effort required to recover to the best of their ability? What stroke patients needed was a trick—a therapy so entertaining that they would do it until they beat it, no matter how hard it was, or how long it took.
“They’re told eat the fish. That’s the only instruction they get,” says Roy, demonstrating the first round of the game the team eventually created. Called I Am Dolphin, it allows the patient to inhabit a sea creature named Bandit, moving and twisting the damaged side of her body (stroke is a disruption of blood flow that often affects just one side of the brain, causing difficulty moving the opposite side of the body) to make the dolphin flip and glide. “I don’t know how a dolphin moves if I’m coming into this. I don’t know what I’m supposed to do. So I have to figure it out,” says Roy. “I’m not thinking about my disability. I’m not thinking about the fact that I’m in a hospital. It’s just: How do I eat these guys?”
Leave a Reply