Lighting up the Brain | Adam Cohen | TEDxCambridge

Take your hand and put it on the side of your head. Your fingers are now within one centimeter of the most complex and amazing piece of matter in the universe. You can take your hands down now. Inside our brains we have about 86 billion neurons, and these neurons are each connected to each other by about 10,000 synaptic connections. These neurons signal to each other with little electrical blips about a tenth of a volt high and one thousandth of a second long. Every thought, feeling, dream, sensation that we have is represented by these electrical blips. From the sound of my voice to the memory of your first kiss, it's all electrical blips. That's all there is. Now these blips are the internal language of our neurons, and they represent our thoughts. And these blips are misregulated in every disease of the nervous system. Everybody in this room knows somebody who has suffered a disease of the brain. To understand how the brain works and what happens when it goes wrong, we'd really like to see these blips whizzing around in the neurons in our brain. The problem is that we can't see the electricity in a neuron anymore than you can see the electricity in this wire. Traditionally, scientists have measured these electrical blips by poking cells with tiny physical probes. This is incredibly slow and labor-intensive work. It usually kills the cell eventually, and it only probes the voltage at one point in space. It's like watching opera through a straw. For decades, scientists have been trying to find a way to convert this electrical activity into light. I'm going to tell you a story about my lab's efforts to solve that problem. The story starts here. This is a picture of the Dead Sea between Israel and Jordan. There's a single-celled microorganism that lives here that produces a protein which it uses to capture solar energy. This protein is called archaerhodopsin 3. It sits in the membrane of the protein and it absorbs sunlight and it uses that solar energy to pump a charge from inside the cell to outside the cell. This builds up a voltage which this creature uses to power its metabolism. You can think of this as the world's smallest solar cell. Light comes in and electricity comes out. So a few years ago, I said to myself, I wonder if we can run this thing in reverse. Instead of having light come in and electricity come out, maybe we can use the voltage in a cell to produce an optical signal that we can see. And you can see this works in PowerPoint, and so that's very encouraging. In this cartoon, we're shining red light on the protein and we're getting fluorescence out only when there's a voltage applied. Now I'm going to skip several years of really hard work and tell you that we found a way to turn this cartoon into reality. Here's a picture of a rat neuron in a dish. We took the gene for this protein and we put it into a virus and we infected the neuron with this virus, so the neurons saw the genetic material and it started to produce the protein. In this picture, there's a little electrode going to the back of the cell and we'll trigger the cell to fire with little pulses of current. I'll show you a movie now at a thousand frames a second of what we saw. So here you can see for the first time what a neuron looks like when it fires. Now to understand how neurons work, it's very useful to stimulate them in different ways and to see how they respond. Here we're using this electrode to stimulate the cell, and that partially defeats the purpose of having this nice optical readout because we have to go and poke the cell. It would be nice if we could stimulate the neuron to fire with light as well. Turns out we can do that, and I'll tell you the story about that. That story starts here. This is a pond in England. There's a freshwater alga which lives in this pond, and this alga likes to swim towards the sunlight. It senses the sunlight because it has a little protein in its membrane called channelrhodopsin. This protein is a light-activated ion channel. When it absorbs blue light, it opens a pore in the membrane and then ions can flow through that membrane to generate a little electrical impulse which tells this creature it's been exposed to light. If you take the gene for this protein and you put it into a neuron, then when the neuron is exposed to blue light, an electrical impulse will trigger that neuron to fire. Here's an example. This is from my friend Feng Zhang when he was a graduate student in Karl Deisseroth's lab. He took the gene for this channelrhodopsin and he put it into a virus and he injected it into the motor cortex of a mouse. For the non-biologists in the audience, that's a mouse. Now this mouse is in a bucket and you can't quite see it, but there's a little optical fiber going into its head, and the mouse is exploring the bucket. At a certain point, the blue light will go on and the mouse will change its behavior. It starts to run in circles, and it'll continue to run in circles as long as that blue light is on, activating the neurons in its brain and its motor cortex, making it run. At a certain point, the blue light will go off and the mouse will return to its previous behavior. Okay, so now we have one protein where we can trigger a neuron to fire with blue light, and I told you about another protein where when the neuron fires, we can see its response with red light. So I said, why don't we just put both proteins into the same cell? That way we'll have a bidirectional optical interface to this neuron. We can stimulate it with blue light and record from it with red light. Now one of the neat things about using light to stimulate neurons is that you can control when and where you deliver the light to stimulate the cell in different patterns of space and time. We developed an optical system that has about a million individually tunable points of light that cover the sample, and so this way we can stimulate a cell in any pattern that we can dream of. Here's an example of a neuron where we're optically stimulating it just in the center of the cell. We also developed some algorithms for processing the data to map the propagation of these electrical impulses with extremely high time resolution. I'll show you a movie of the electrical impulse propagating at 100,000 frames a second, and this is real data, it's not made up. Okay, so for the first time you can really see how these impulses propagate through a cell. We can also do this on large fields of view. Here there's about 50 neurons from a rat. I'll show you a movie at 500 frames a second where at a certain point we'll turn on a blue light, a little icon will light up, and all the neurons will start to fire. So here we can see every impulse in every cell in the culture, and we can segment these images and pull out the firing patterns of each of these cells. And encoded within these flashes of light coming from the cells is a huge amount of information about the underlying physiology of the cells. This movie was seven-tenths of a second long, and we recorded from about 50 neurons. To do this with manual recordings would have taken about one hour per cell. So instead of one hour per cell, it's seven-tenths of a second for 50 cells. Now these are rat neurons, and we're also interested in looking at human neurons. So I said to my students, would anybody please donate a little chunk of brain for our experiments? And they were strangely reluctant. This brings us to another problem in studying human neurons: we can't get to the tissue. Cancer researchers have loads of the cells they're trying to study because doctors love to chop out tumors, but we don't biopsy patient brains. How can we study a tissue that we can't even get to? In recent years, scientists have discovered a remarkable workaround. All the cells in our body are pretty much genetically identical. All that distinguishes skin from blood from brain is which genes are turned on. But every cell has the information for the whole body. Scientists can now take some skin from anybody and take these skin cells back through embryonic development into a stem cell-like state, and then take them forward down a different developmental pathway and turn them into neurons. And these neurons will grow in a dish, and they'll wire up, and they'll start to fire, and they're genetically identical to the neurons in the brain of the person who gave the skin. You can do this on neurons derived from a healthy person or on neurons derived from somebody with a genetically based disease of the nervous system, and you can compare and contrast. It turns out that everybody's neurons are a little bit different because we all have a unique genetic background. When a patient is, for example, diagnosed with epilepsy, the diagnosis is a description of the symptoms, not the underlying cause of the disease. There are hundreds of mutations that can cause epilepsy, and the same is true for most other diseases of the nervous system. The ability to study neurons from individual patients gives us the hope that we can understand in that particular patient what the basis of the disease is, and maybe we can develop therapies that are targeted for that individual patient. To pursue this vision, I teamed up with my good friend and colleague Kevin Eggan. We've started to apply this technology to study devastating diseases of the nervous system. This is a picture of a patient who has ALS. This is a neurodegenerative disease of motor neurons. It causes paralysis and eventually death. There is no cure for this disease. In Kevin's lab, they've been taking skin from patients like this, little pieces of skin, and converting these skin cells into neurons, into motor neurons, which are the cells afflicted in the disease. And then in my lab, we study the firing patterns of the neurons. These are human motor neurons lighting up in the dish, and we see differences in the firing patterns between neurons derived from ALS patients and from controls. And these differences tell us about the molecular mechanisms which are causing the disease. In earlier work that Kevin had done with his colleague Clifford Woolf at Boston Children's Hospital, they inferred from the firing patterns of neurons that there was a particular ion channel which was defective in at least some ALS patients. And it turns out there's a drug which is already approved to treat epilepsy, which is a modulator of this channel. When they put this drug on the neurons in the dish, it made those cells survive longer in the dish. And on the basis of that, there's now a clinical trial going on for that drug in ALS. So the ability to study these human neurons in the dish using these advanced optical tools which make the cells light up gives us the hope that we can get new insights into these devastating diseases of the brain. Kevin and I are now applying these tools to many different diseases: ALS, epilepsy, pain, Alzheimer's disease. Before I end, I want to say a little bit about where these tools come from that we're working with. The voltage indicator comes from this microorganism up here which lives in the Dead Sea. The light activator protein comes from this alga from a pond in the south of England. We fuse that protein to another protein derived from a coral. We fuse our voltage indicator to a protein derived from a jellyfish. And then we actually link these proteins together with a peptide derived from a pig virus. It's just incredible to me that we can take these genes which evolved under totally different circumstances and staple them together to come up with new functions which have never been selected for in the history of evolution. And the only reason we can do this is that ecologists have been wading through the muck for decades, classifying these seemingly useless organisms. The Israeli ecologist who discovered this organism in the early 1980s never would have guessed that 30 years later, a gene from that organism would lead to new insights into a neurodegenerative disease. It's also important to conserve the places where these organisms live, because we can only borrow from nature. We could never invent these functions on our own. So as you leave here tonight, I invite you to think about the beautiful colors and forms of the natural world, and to contemplate the inner workings of these incredible biological machines, and to imagine how fundamental explorations of the natural world can lead in totally unexpected ways to life-changing discoveries. Thank you.