View transcript
- Dr. Muotri has brought something very special tonight with him. He's brought some cortical organoids for you to look at.

- Wow!

- And so we're gonna look at those as part of the program. What we're gonna do is we're gonna hear from him for about 30 minutes, and then he'll be joined on stage by my colleague, Rob Semper, who's our senior scientist in charge of research and development here at the Exploratorium. So a little bit about Dr. Muotri. He's a geneticist originally and then went into neuroscience. And he's gonna tell you a little bit about that himself. We thought he was really interesting to bring to you because he asks questions, and then he's making a tool to answer it. So these organoids are crazy compelling. They are alive, and they're both an instrument and a new set of questions.

- Thank you so much for the invitation. It's my pleasure to share a little bit of our research that we're doing in the lab. I'm actually from San Diego in California where I have a stem cell biology lab, and we work on conditions like autism or schizophrenia, and we try to understand how those conditions actually happen in the brain and if there is ways to fix that. But before I jump into what we're doing in the lab, I can tell you a little bit on how I get involved with neurosciences. So I'm a geneticist by training, and so I have a PhD in genetics. And I was kind of looking for opportunities, and one day I was invited to talk to the European Parliament about my work. And after the talk, there was a bunch of people in the room talking about what are the biggest challenges of our world, so this is where we live. And I have a challenge for you. Think about what you think would be the biggest problem that we are facing nowadays? So if you're like me, you're probably thinking about one of those. And that's what those people were thinking, as well. But defining the problem is just the beginning. I think another huge challenge is how to fix it. And I have a challenge for you. I mean, how are we gonna fix them all, all those problems? How, there's only one way to fix all those problems. So just pause and imagine, how are we gonna fix all those problems? So that was the exercise that I did, and I realized that the only way to solve those problems is to use the human brain. So there is no other species who are gonna save this planet. It is the human brain who's gonna save this planet. So in here, we have a conundrum with the human brain neuroscience. Most of the neuroscience today, I mean, we work on trying to understand the human brain as it is. So we put subjects on MRI scans, and we analyze their brain waves, and we try to understand how the brain works. But the brain's such a complex machine that we are just observing that complexity, and we don't know how the brain get to that stage. So we're turning to animal models. There's a bunch of my colleagues who study mice and trying to understand how the human brain works is studying the mouse brain. But here is the thing. The mouse brain is so tiny it fits in my eyeball. So it's millions of years of evolution that changes those two brains. So good luck trying to understand the human brain by looking at a mouse. So we really need a model, a human model to really understand how the brains form. And now just to give you like an idea of how little we know, I mean, you all realize that this is a car because all the pieces are very familiar to you. If I break the brain apart and put all the pieces like that, there is no neuroscientist today that will tell that this is a brain because we just don't know all the parts. We don't know how it's formed, so we have no idea. And you might ask, so, I mean, why don't you learn how to do it? And here's the biggest challenge. The human brain is formed inside a uterus. And we don't have access to that human embryo brain. It's gonna be unethical to do an experiment with a healthy human brain, so we just don't study. Instead, we rely on no invasive technologies such as ultrasound to kind of realize how the brain is going. But we don't fully understand or appreciate how the first neuron fire, how the first networks are formed and turn into the complex brain that we all have. So that challenge has been a major roadblock for all neuroscientists because we really can't understand the human brain. So that's why we're turning to models. So when I give talks to lay public, I realize that people have a problem with the word model. And I'm from Brazil. When I go back and talk to my fellow Brazilians and they ask me, "What are you doing?" And I say, "Oh, I use the stem cells as a model "to understand the human brain." And said, "Why do you think it's a model?" And I get all kind of crazy stories. I mean, people think that this is a model, right? And I said, "Well, you're right, this is a model." So, I mean, a child might think that this is a model of a train or of a flying machine is also a model. And everybody's right because a model is an imperfect representation of the reality. So this very simplified representation of the reality or models that we all use, and the best models are the ones that you can increase in complexity to get closer to the reality, which is the human brain. So I mentioned to you that we use the human brain tissue, mostly postmortem tissue as a model, but the cells are dead, so we cannot hear from a single neuron or how the networks are formed. We study cells from individuals such as blood, but blood don't form synapses so they are not the brain cells that we want. And again, I mean, we, most of us, are actually turning to mouse models, and again, I mean, there's millions of years of separation between those species. So what my lab has decided to do is to use stem cells to recreate the brain from scratch, and you'll probably hear from the media that what we're doing is we have a mini-brain in a dish. And this is the image that most of you might have, and I'll tell you that that's not what we have, and we have it right here so you can see it, exactly how it looks like. So the word mini-brain is misleading because we don't have a fully formed brain there. Instead, what we have is something way more simpler, and there is a bunch of limitations to this model. I'll list them to you before I continue. For example, most of those neurons that we can form are immature neurons. These what we call brain organoid, that would be the correct scientific term, it's a brain organoid, are not vascularized so they cannot grow that big. We don't all cell types represented. We don't even know if we're growing them in the right condition. We're gonna see here that they're floating in a solution that we think or we optimize for the growth of those human neurons, but we actually don't know. It's empirically defined, so there is room for improvement. And so my lab over the past five years have create a recipe to generate these brain organoids that is fully optimized and very simple way to do it, so we start with human pluripotent stem cells. These human pluripotent stem cells are similar to the human embryonic stem cells. And where do we get them? Well, the embryonic stem cells you get from the human embryo. So couples that go to clinics for in-vitro fertilization, they donate the leftover embryos, and we can extract these embryonic stem cells, and we can use for this purpose. But we have a more powerful way to get those cells because we can get those cells from anybody who's alive. So it means that if I take cells from my skin or my hair or my dental pulp, I can go back to my lab, reprogram the cells to make this pluripotent stem cells. And you know what happens, you capture the genetics or the genome of the person where the cells were derived from. So the donor cell type will carry your genome here, and when I make these protocol to create these brain organoids, guess what? I'll have a brain organoid from you because it contains all your genetic information. So this would be interesting in the future when I talk about disease modeling. I will get there. So the protocol is very simple. It's a series of repetition. You just grow the cells and let them self aggregate into , you self aggregate, start to migrate out and form different brain structures. You can actually guide them to the different types of brain cells that you want to have in there. So this is a protocol for a cortical organoid. We mostly focus on the cortex because that's the region where we see problems and conditions such as autism, schizophrenia, dementia, Alzheimer's. So it's a very important brain region. So what you have at the end is a brain organoid that in a couple of months, will start to have these cortical plate. So this is more or less what brain organoids look like. This is a post-doc in the lab, a plate with six wells, and each white dot in those wells is one of these brain organoids. So they can grow up to .5 centimeter, so you can actually see them at naked eye. And once I make a batch of brain organoids from someone, I don't only end up with one or two, I end up with hundreds if not thousands of them so I can use for different experiments. So the way these happen, I think that's the most interesting way. It's all genetically programmed. The cells do what they're supposed to do. So I hope you can see here that there are cells migrating out from the core on top. So this is one of the brain organoids that we just played in a two-dimensional structure and I hope you can see that there are some cables. This is what we call radio glial cells where the progenitor cells migrate out to form the cortex. So in a 3-dimensional structure, this will be from inside out. That's how your cortex is formed, that's how your brain is formed from inside out. So if we light one of these organoids, what you have is a structure like that as a ventricular zone in the center where the progenitor cells are located. And as they migrate out, you're gonna have deformation of the cortical plate. So the cells that we have in there are different cell types. This is just one way to identify all the cells type you have, glutamatergic neurons, these are the first type of neurons that appear during human development. These are excitatory neurons. We have glia cells, These are cells that support the formation of synapses, gluten GABAergic neurons, these are your inhibitory neurons. They appear a little bit late during development and they will help to fine-tune the networks. And finally your pool of progenitor cells that continues to produce these neurons and glia cells so you can have more cells in these organoids. We can keep them for several years in the lab and I'll tell you why this is not a good idea but these progenitor cells, they will kind of stop growing, producing more neurons when they reach about one year and mostly because they are not vascularized so there is no nutrients going inside of these organoids, so they stop periphellating. So the good news is that we have all the cell types that we need for the brain to work. All the major cell types are there and we have the perfect anatomy. We have a ventricular zone, we have the cortical plate. As soon as scientists start to realize that we have that tool that mimics human embryonic development, we start asking can we use these tool for something useful? And I'll give you one example from my lab. I was in 2015, a colleague of mine called me and said "Have you ever heard about the Zika virus? And I had never heard about the Zika virus before and said you know we think that it's causing some kind of microcephaly outbreak here in Brazil but we don't know if that's the causal agent or if it's something else. So what we did was we collect a little bit of Zika virus and we exposed to these brain organoids and we see that the brain organoids shrink. So that was the first causal evidence to suggest that the Zika virus was leading to microcephalic cases in the northeast of Brazil. We couldn't do that with mouse models because the mouse brain developed so fast, the Zika virus didn't have the time to kill the cells so we can see a microcephalic brain. So you need these low developing human brain to model a condition like that. But most importantly, once we figured out that the Zika virus was killing the cells we could use that model to try drugs that could block the action of the virus. And two years later, we found the drug called sofosbuvir. It's a drug that's already approved by the FDA for another retro, another virus. And this drug was able to block viral replication. So right now, if we have another Zika virus outbreak and there is a woman who is pregnant and it's positive for the virus, she can take this drug and the baby will be safe. So this is one of the few cases where we have the right to in the right biological question that we can use to solve and understand the virus biology and provide a solution for the problem. We were so happy with that idea that we decided to expand that idea to everything else that is exposed in uterus. So we think that the uterus is just like an incubator, right, I mean it's just growing the babies in there. But it's not. It actually shapes who you are and women that are pregnant, they are exposed to thousands of chemicals. Some of these chemicals might not be good for the developing brain and unfortunately there is no way to know because all the toxicology is done in mouse brains. So if the mouse brain does not recapitulate that, we're gonna miss it. So you probably heard about Tylenol increasing the chances for you to have a kid with autism or other neurological disorders. So why we took so long to realize that Tylenol was bad for pregnant women because it needs to catch up with the epidemiology so those kids have to grow up and receive the diagnose so we can do the correlation with the use of Tylenol. So we think that we can use this model now to create what we call a brain safe label. So everything that goes in uterus, we can test and put a red flag so we can investigate better or block if those medications or drugs goes into the body of a pregnant woman. So that's one of the applications. The other application that I think was really cool and this is a recent work, we can take some of these brain organoids then we can seed them with glioblastoma cells. Glioblastoma is one of the nasty cancer types that really kills children. So here what you see, these big balls are our brain organoids and the green cells are glioblastoma that we took from a patient. This patient could not respond to any drug. So we used this model to find a potential treatment that would wipe out all the cancer cells and leave the brain intact. So the good news here is that if we did it for one person, we can do it for everybody and this can be a personalized treatment. So you can have your own mini brains with your own cancer and we can figure out what kills your cancer cells and leave your brain intact. So this is lowly moving into clinics and we're gonna get there. So all those applications were so nice because we have these brain organoids and again I mean the shape of these brain organoids and the cell types really mimics the human developing brain. But there is one thing that we never had. None of those brain organoids actually function like the human brain. So we could never had neurons firing in a synchronized network as the human brain does. And we think this was the case because the protocols we had was not that good but as we improved the protocols, we thought that perhaps we have a chance. And the way to measure those networks is to place some of those brain organoids on multi electrode arrays. So these are dishes where you have the letters printed in the bottom and you place the organoid in there and we start recording from their activity. So you have activity maps like that or raster plot where each channel would have different traces representing when neurons fire. So you can record over time. So when we start doing that, I was, to be honest, not very excited and I'll show you why. So if I plot a brain activity here over time, you can see that most of the work is below five Hertz. And these are human neurons that we used to have or even though if you can keep them a long time up to 32 weeks, you never pass this five Hertz. But we want to get to these 20 Hertz that it's close to the primate brain. So it means that there is more neurons active and the brain is just firing in a frequency that is more close to the human brain. So now I'm gonna show you how this new protocol work and how it changes everything. So what you see is weekly measurements of these organoids over time and you can see that when they reach about 25 weeks, they start to peak in activity and becomes almost exponential reach close to that 20 Hertz. So that was unprecedent. We never imagined that these will happen and make us really happy. But also open the possibility to start seeing if those brain organoids will actually show neural oscillations. So neuro oscillations are these waves that you can record once you have electrodes through your skull. So this is what we call a lateral encephalo gram and it's easy to do, easy to record and it's important because those brain waves correlates with different states of being. So if you're paying attention, your brain wave is gonna have certain shape. If you're sleeping, your brainwave will have a different state. If you have a seizure, your brainwave will dramatically change. So they correlate with different diseases' state, different developmental time and there's a lot of studies correlating these neural oscillations with human behavior. The one that I like the most is when they put these electrodes in a group of people watching a movie and they start recording all at the same time. And all the time, when there was a social interaction in the movie, a dialogue between two people, the neuro oscillations from everybody in the audience synchronized. Isn't that amazing? So it tells how much social our brains are programmed to behave. So we are social beings so our brainwaves organoids synchronize when we are talking to our friends. So it's much higher chances to synchronize with a familiar face then with an unfamiliar face. So that's why it's important to have these brainwaves when I'm studying autism for example. So and we do see this brainwaves in this organoids. That was the biggest surprise. They actually appear when they are about four month of age. By six months, you see that they are very synchronized and by eight months, they become highly complex. So that complexity, it's interesting because it says that the networks are getting in a stage of where they become very plastic. And plasticity in the brain or in the network is something desirable because it makes you change ideas, makes you make different interpretations of the word. So once we have these waves, we start asking what is the age of our organoids? How can we compare these brain waves that we are recording from these structures to the human brain? And since this start from very early embryonic stage and I told you that we cannot place electrodes in a developing human brain, so we have to rely on something else. We have to rely on babies that are born prematurely And what happens when a baby is born premature, you place some of these electrodes in their heads just to monitor their overall brain health, brain activity. And what we learn by doing that is that as we develop, we pass to a state that's called trace discontinue. It means that in the brains of these babies, you have a peak of activity followed by a chi ascent a silent time in another peak of activity so you can see those peaks of activity over time. So that's why we call trace discontinue. As we grow older and this happens to everybody, the period between the activity, the silent chi ascent and time becomes shorter and shorter and shorter so our EEGs, or our brainwaves become highly complex. So the organoid shows brain waves that are very similar to these premature human baby brains and then I mean to ask if they are very similar to in age we decided to create on a machine learning where we use EEG data from several premature babies from 25 weeks to 38 weeks. We train the machine to determine the age of the subject, base it on EEG features. And once the machine was really good, we start feeding the machine with the data coming from the organoid and ask if the machine was able to interpret or to predict the age of the organoid. And the data is just amazing, I'll show you. So the perfect data is gonna be on the dot bar here. You have age in weeks and you have the prediction age by the machine learning. So I hope you can see that the black dots are the real data. These are the data from the human brain and the blue dots are the ones from the organoids. So they overlap. The machine gets confused, it cannot distinguish between the organoid and the real human brain. So that prediction happens after 25 weeks. Before 25 weeks, there is poor correlation and there is one reason for that. We don't have any human data before 25 weeks. The human brain would just not survive. Right now, there are more and more neonatologists that are able to keep the human brain alive even at 23, 24 weeks but this is really pushing the limits. Most of the premature babies before 25 weeks would just die. So we don't have any data to compare, so we cannot train the machine. So we still don't know what happens in the first 25 weeks of humans gestation in the brain. So that data suggested the organoids that we create in the lab follow the same trajectory in neuro oscillation as the human brain. So you know what happens if you create an artificial network that mimics the human brain like that, if you're a scientist and you have a lab? You end up in the New York Times. So that was last year. We'd just finished that, we'd just complete that. So now we can recreate a brain organoid from any of you and understand how your brain oscillations actually happens in these very early stages of life. So one thing that I didn't mention is that after nine months, the level of these networks stay kind of constant. It plateaus. And we were expecting them to continue to evolve. So we thought that the reason why these networks plateaus it's because they don't have any stimulus. Right now, we grow this brain organoids inside an incubator inside the lab and they just stay there. Nobody is stimulating them, they don't see anything, they are not receiving any input. So that's very different when a baby is born. By nine months when you deliver it the first thing that happens is the baby experiences the world. It cries and then opens the eye and see mom. And then mom sees the baby and a strong connection happens. And the network's at that stage are just crazy for stimulus. So we are trying to mimic that and one way to mimic that is to create a visual system. So what if we add the eye to the brain? So what you have here is three types of organoids. The first organoids is a retina organoid where the photo receptors are in green. In the middle, we have a Ptolemaic organoid that receives the ganglion neurons from the retina and project to the cortex. And you can follow the trace of those neurons in red. So we have markers so we know that we have the right structures. So these structures are now just maturing in the lab as we speak and what we hope for is that they would mature to a level where the photo receptors here would be stimulated by light so we can post images, we can post light and we can record from the cortex to see if the cortex now acquire a more visual identity. Then if that's the case, we can ask questions like can these brain organoids restore our memory? Can we give them any kind of experience? So hopefully, we can cross that plateau activity and have the brain organoids to continue to evolve so we can see different types of oscillatory waves appearing. So that's mimicking biology but I was very anxious on stimulating these organoids and I thought what if we give them not only an eye but an entire body, right? Why not, right? So we decided to create an interface with a robotic machine. So what you are seeing here is a robot that is learning how to walk by using the electrical information coming from one of these organoids. So the way this works is the following. The organoids are in the multi electrode. We capture the electrical activity, we send electrical activity to a computer who decodes that, and tells to the robot how to coordinate the four legs. That's what you are having here. Now we add those two eyes. These are not real eyes, these are infrareds. The infrared helps the robot to detect when it's approaching an obstacle such as a wall. And as it approaches the wall, when it gets really close to the wall, it stimulates back the organoid. The organoid receive a new stimulus. It changes the network, the network group will become something else. The computer will now realize that while the network changes so now instead of moving forward, you give a second command to the robot such as this talk and walk back. So the idea is to have these robots exploring the environment, while they stimulate the organoid and they are doing this all the time. So again, it's an artificial system that we're creating to stimulate these organoids and helping them to mature even further. We're also using these machine learning to create ways to understand how the neurons in these organoids self arrange and understand how the human brain learns so you can create better algorithms for artificial intelligence for example. So this is a new partnership that we have with Microsoft. So that's one thing that I found amazing but there is another thing that I think these organoids are very useful, which is to understand our own oranges. So when I started my lab, the biggest question that I have was to understand what makes the human brain so unique? We are an outlier over all other species so we are highly social and we can create arts and technology and no other species have done that. So why is that? Most of evolutionary neuroscientists, we are used to compare our brain with our closest primate relatives, the chimpanzees. But again, when I look to a chimpanzee brain, I looked at the human brain, they are so different that I cannot even compare. So instead, I was thinking about why not comparing the brain of our ancestral, for example, the Neanderthals. And I don't know how much you guys know about human evolution but there was lots of human types living on earth at a certain time. The ones that we know the most are really the Neanderthals because then we have lots of fossil records so we learn a little bit about their lifestyle, how they live in community. But one thing that is to me most appealing is that even though they have lived on earth for a period that is longer than what we have lived, their tools and technologies are limited. So someone might even argue that whoa, there is a huge advancement here over several hundreds of years but it is still a stone tool. We modern humans, we're changing the world, right? I mean we started with stone tools and now we have all these arts and technologies and we are here having this conversation because we learn how to live in highly complex society, we learned how to take advantage of human imagination, we learn how to communicate to each other in a very efficient way so we can advance all those things. And again, as I mentioned, no other species have done that. So, all right. So Neanderthals are like us humans, so would be better to compare their brains with the human brain. But there's a problem here. Although we can have bones from the Neanderthals, we can even extract their genome and sequence their DNA, we don't have live cells. And I need live cells to reprogram so I can make these brain organoids. So, since we don't have live cells, we decided to take a genomic perspective. So we aligned the genome of Neanderthals and modern humans and we ask what is so unique about us? So if you look for the protein coding genes which is about 20,000 genes and you ask how many of those are different between us and the Neanderthals, we end up with 61 genes. And then if you ask what are these 61 genes doing, you realize that there's a couple of them that are involved in the immune system, some of them are involved in bones, they have like strong bones, muscles. But what about the brain? How many of those are involved in brain development? And that's fascinating. It's only four genes. So if it's only four genes that have point mutations that make us different from the Neanderthals, I don't need live cells from the Neanderthals. I can just edit a human genome to carry the ancestral genes of the Neanderthals and then I can create Neanderthalized cells and create Neanderthalized organoids or Neandertoids, how we call them in the lab. So nobody in the lab wanted to do that because they said ah, this is such a high-risk project. I mean what if I do all these work and at the end there is nothing? So this is a project I actually started myself but finally, I convinced someone to help me and the data pays off. So these are the organoids carrying the ancestral variants compared to the modern humans that I just show you. So I hope you can appreciate how the morphology changes a lot. So this is all about how the progenitor cells migrate to form the microcircuitries of the cortex. And these markers microcircuitries are really important because when we measure their activity, they are very different from modern humans. So what does it mean? When we compare those networks, they are very similar with the networks coming from being organized from autistic individuals that we have in the lab. So wait a minute. Can you say that the Neanderthals are autistic? No, we cannot say that. That would be like a huge extrapolation. But we can say that there is something about autistic neural development that might be part of human evolution. It could be hyper focus, it could be lack of socialization, it could be difficulty with language. So probably those four genes have helped us on our evolutionary path. So now think about it. If I'm using these organoids in genome editing to look back into our past, why not looking to our future? Why not manipulating the human genome to create other kind of brains that never exist that evolution has never had the time to do it so we can accelerated human evolution in a dish. So that's what we are doing. We can speed up human evolution. We're looking for regions in the genome that are highly polymorphic among us and create very different brain organoids. So what kind of brain organoids are we creating? We're creating brain organoids that are probably, that's gonna be probably resistant to Alzheimer's. We're creating brain organoids that will have more synapses. We are creating brain organoids that are more plastic. So I mean these are all useful things and as we reach to that level, we as a society, we have to decided if those genetic variants should be passed on to everybody. So that is starting to create some ethical dilemmas and we can talk about that later. But this is the kind of project that makes me think about other futuristic projects. For example, last summer we sent these brain organoids to the space station. So they were growing and developing inside a space station and we were inspired by this program from NASA who compared genetically identical twins that one stayed on the space station for about a year, the other one is stayed here on earth. When he came back, they did the comparison and they realized that there are several circutries or system in the body that was affected. Some of those changes are reversible, others are permanent. Things like cognitive decline are very important for NASA because I mean if you are planning of sending someone to Mars, you need the astronaut to have a really good mental health. Problems with the retina, so the astronaut must see what's going on. So we thought that perhaps we have the right system to help on space exploration. So we had to build a machine that create these organoids so this is the kind of machine we have. It's a shoe box and the organoids draws inside those tubes here and we keep all the fluid that's coming from these bags and keep feeding the organoids as long as we wish. So we went to Cape Canaveral. We put everything in a box, we got the seal of approval from NASA, we put in a SpaceX. We helped it to launched and if it launches you celebrate with your friends. About 32 hours later, I mean the mechanic Canadian arm from the space station will get the capsule, the Dragon capsule and we put in the state station and the only thing that the astronaut has to do is to take that shoebox put in a porta foam like that and plug in one of those compartments. As soon as the astronaut plug, I receive a ping on my computer saying that it was plugged so I have immediately access to everything inside so I could monitor how the organoids were growing, how was the temperature, things like that. I could even take a picture. So this is one of the pictures. We need to improve the camera by the way but nonetheless I mean I hope you can see some spheroids in there. So these are the brain organoids and there's a couple of things that we're learning for example. Look how perfect as spheres they are. So when we grow organoids here on earth, they are always lightly different, they are not perfect. But microgravity makes them really perfect. They are perfect as spheres. Also it seems like they grow a little bit bigger. So what is the consequence for space colonization? So if you have a baby in space, you'd probably need the C-section because your baby is gonna be like a big brain. That is true. It's so funny because I never seen a science fiction book that talks about delivering a baby in another planet. But that's a real problem I mean I should talk to Elon Musk about that. But anyways, so yeah, so these are important considerations for astronauts and in long-term spaceflights. But we also want to use this technology to learn something about human health here on earth. And one thing that we and others have observed is when you bring those cells back, there is something really funny that happens which is accelerated aging. Yeah. Why you're in microgravity the cells are beautiful nice but as soon as you bring them back to earth, in about two weeks, we have these speed-up accelerated aging. So the cells just aged faster. Nobody fully understand the mechanism. By the way, this is an open field but we know it happens. So perhaps now, can I use this brain organoids to model diseases of late onset so I don't have to grow these organoids for several years to see for example plagues in Alzheimer's brain organoid. Maybe I can just keep them over there for a while then bring them back so I can study dementia, Alzheimer's, Parkinson's, some things that requires a long time to develop. And another consequence of this project was that as soon as we have those shoebox and grow in the organoids in a fully autonomous ways. I mean people in my labs started saying why do you need me to change the media? Just buy one of those boxes for me. And they're right because we keep doing these tissue cultures. It's like a hundred-year-old technology. But what if we have all these boxes growing all these organoids for all the different projects? So the landscape of my lab of how we do science is changing, so we think that in the future we're gonna have these brain farms, lots of those boxes growing different brain organoids. And I tell you what, when I show you that experiment from cancer. So it's cancer centers it starts to pay attention on that and they realize hmm, maybe we also need our own brain farms so we can grow these organoid from our patients and test the different drugs before we testing the patients. So that might speed up all the clinical trials so it might save lives. So these technology might be very transformative in that sense. All right. So I think I talked too much already, so I'll stop here just to acknowledge people in my lab. So these are folks who actually do the work except for the Neanderthal thing that I started. But most of the time, they're the ones that do the work, the families that contribute to my research and funding agencies. Thank you very much. Wanna pick it up?

- So actually what I think we should do, this is the Exploratorium after all. So why don't we show an organoid?

- Yep.

- So let's go to the microscope as they say. So Dr. Muotri brought these up with him on the plane in his pocket. Right through TSA, right? And what we're gonna do is get them focused and start to look at a real organoid.

- So I have a couple of them here. This is a small one. So these are about six months of age so they grew for six months. It's a little bit dark inside because six months is already like kind of old. I hope what you can see is there are some structures inside and these are the ventricular zone that I showed you in the neurons that migrate out and form the cortical plate here. So there are some of them that are better to see than others. Eventually, and I remember seeing one of them in here, we end up with structures that are or cells that are very dark inside these organoids. These are unintentional retina cells. So sometimes the brain and the eye, they develop more or less at the same time and we end up with some photo receptors in there, but unintentional. Most of the time we can grow them in a very simplistic cortical fashion. So yeah, progenitor cells will be inside. So as we see that it becomes darker in the center, It means that in the center we're not receiving the nutrients anymore from this media. So this is a media that contains all the nutrients and factors that these cells needs to be alive. And that's why they stopped growing. So they reach about 0.5 centimeter and then they will stop grow. There are some people in the lab that are creating a vascular system so hoping they can at least double in size. Here you are seeing about 2.5 million neurons in here plus all the glial cells and progenitor cells.

- [Woman] So you were saying that's like a bee?

- Yeah, so two million neurons is about like the brain of a bee. Yeah, yeah.

- And how many neurons in the human brain?

- The human brain has 86 billion neurons. Yeah, so what's fascinating you might ask, "Come on, I mean you just showed "that those million neurons are similar to billion neurons?" And that's one of the fascinating things that we are learning that the brains to form these neural oscillations, we don't actually need that many neurons. We can do it with two million neurons. And I think when we think about consequences for example, neural networks for artificial intelligence, that's good news, so we don't need that many inputs or nodes to creating this kind of networks. Let's see if we can find something else.

- So one of the things that's fascinating to me is what really are these things? I mean, they're not brains but it's kind of an interesting question particularly because of ethics. What is this and what isn't it in terms of an object, a biological object?

- So we got was so funny because we were having this discussion with different philosophers and that question appear. What are those things? And those philosophers they said they're not in their philosophy books. So it's something else. We have to create a new way to understand them or to give them some kind of status. Are they humans? Are they cells in a dish? What are those? So that's open question and I think I was joking with them that I carried those organoids in my pocket in a tube like that. We're having like lots of ethical and philosophical discussions related to organoids becoming conscious or self-aware and really there is no way for me to know. But let's suppose that in the future we will figure out that they are. So it might be telling the future. I might not be able to do that. That might be like a crime to keep them here without the proper conditions because they will receive a moral status similar as an animal or a mouse in the lab so that might be the future, yeah.

- So you're using this as a model really to learn about the brain as well as to do some biological work. How do you think about the fidelity question? How do you deal with the fact that this is I mean, you saw those, the electronic, the electrical signals coming together which must have been very exciting by the way. I mean it must have an extraordinary discovery. How do you do tiers and figure out well, how much of this is really something unique to this kind of object and something as opposed to telling you something about the brain itself or on a trajectory of development? I mean are there some artificial things that you've seen that you have to worry about that may not be true in terms of the real brain development?

- Yeah. So I think there are two possibilities, right? One is we're mimicking human biology in a very precise way here and we have evidence that that might be the case, right. The trajectory is the same, the cell types, the identity. So they're doing what they're supposed to do. It seems all genetically programmed to do that. But on the other hand, we know that these are growing outside the body. They're not connected with other brain regions, they're not in the uterus so they probably are not doing what they're supposed to do because they don't have all the right conditions. So, I think scientists now like me I mean we see things that are similar and we see things that are very different from the human brain. It might change in the following years. As we improve the technology, those things will become more and more close to the reality which is the human brain. But I think the important take-home message here is that even though if they're not similar, even if we've already have, we can learn so much about the human brain, how it forms in that very early stages, how the first neuron polarize, how the first networks start to get together, that I think it's a tool that in the future, we're gonna be looking back and said, "Oh my gosh, how did we survive "in neuro sciences without this tool?"

- So you have such passion about this work and it's interesting to think about what's really driving you to do this research. I mean what's really, you're doing so many different things here. How would you describe what's the inner drive for this? What's making you wanna do this?

- Yeah, speaking as a scientist, I think it's really the possibility of discover something that nobody has seen before. I mean we all have this ego of being the first of doing something. This is part of the science, part of the motivation. I had that at one point, now I think in a different way. I have a kid with autism, a very severe autism and epilepsy and seizures and there is no medication, there is nothing that can helps him. So I think this is a tool that we need to attack or to improve the life of people with neurological disorders. And when we think broadly about neurological disorders, we're talking about millions of people. So I think that's my main motivation is the translational aspect of that. How can I move from this very simplistic model to helping people with neurological conditions?

- So I have one more question but before I do I wanna put an ad in for people to go and look at the exhibit that's in the back part of the museum cells to self. This lecture series is really to talk about some of the current breaking edge science that's happening about that same topic you can't really show easily in an exhibit although pretty soon I think we can have a organoid in our exhibit. So go and see that exhibit and then come to these other shows. I think my final question is a little bit something we talked about earlier which is why is it so critical for people to learn about this, I mean this is kind of a pretty advanced scientific topic? But you said this is really something that one needs to know about deeply.

- [Dr. Muotri] Yeah, yeah.

- So what are your comments on that?

- And that's why I applaud your initiative to have these conversations. I've been in the stem cell field for a while and I remember when scientists cloned, I mean Dolly. Right? I mean we cloned an animal, a large animal and everybody was taken by surprise, I mean you could see people freaking out. So now it's Dolly, tomorrow it's gonna be a human so how come I mean, we're cloning people already? But nobody was informing the community or the society of the scientific advancements that was happening along the way to get to that stage. And I think if we know everything that happens along the way, we as a society we're much better informed of the uses and consequences of that. So we can support this kind of science or we can control it better. So I don't wanna to have these technology to become the next cloning. So even the name is wrong, you see? I mean we use clone and it's not the proper scientific name. Here people will start calling them mini brains. It's a wrong name. Again, we are making the same mistakes again. So having this opportunity to really talk, I think it's important for people to realize what is these stages, what is the possibility, what this technology can bring to the table that we never had before? Well with that, join me in thanking Dr. Muotri

- for a wonderful talk.

- Thank you.

- And we encourage you to go out and have fun we're still open till 10 tonight and thank you for coming.