Michael Levin | Taming the Collective Intelligence of Cells for Regenerative Medicine

Welcome everyone to the Fawcett Health and Biotechnology Outreach Group sponsored by 100 plus capital. I'm very excited to have Michael Levine here today. He was recommended by someone whose recommendations are greatly appreciated, so I'm very excited about his talk. I think he's bringing together a variety of different super interesting fields, I definitely think you're pushing the boundaries with the research that you do. I won't say more and will let you guide the conversation. We will have a short break halfway through where we will have some announcements from people in the community and then we will have a longer discussion.

I think Michael already said that he can stay a little longer, but this is just a reminder to me at the beginning that I'll do it on second notice. in the chat to ask questions uh and um usually what happens at the end of these talks is that there are a lot of questions, we have a lot of questions and we don't have time to cover them all, so don't be shy, ask them early and then we can start moving up. downloading questions because usually in the end we always try to keep them as short as possible, but why not start early by asking questions?

Thank you so much for joining today and Michael, it's really a pleasure to finally have you on our programming. This has been going on for a while now, and you're going to tell us a lot of things about

regenerative

medicine

that I think a lot of people here don't even have on their radar, but you're leading the Taft Center for Regenerative and Developmental Biology. I will publish it. more in your bio in chat for now please remove it. I'm very excited to have you here and thank you for joining. Thank you very much and yes, thank you for having me and for the opportunity to talk to you about all of this. um, so let's see, uh, can everyone see my slide?

Yeah, perfect, okay, so you'll see two websites here. Please don't hesitate to contact me later if you have any questions that we don't cover. everything else is there and I have to make a revelation. I have a spin-off company called morphaceuticals inc with david kaplan and I'll tell you a little bit more about that, so I want to give you the main point of the talk right at the beginning. Front um and here are the four things that I'm going to try to convey today. I believe that solving the problem of anatomical homeostasis that I will discuss is the key to truly transformative

regenerative

medicine

that will address aging and many other things. and that current approaches that focus on stem cell biology and genome editing of things like this are limited by themselves and that what's really key is understanding the software of life and I'll show you some examples of that, a key means by which

cells

compute and make decisions in vivo is what we call non-neuronal bioelectricity and now it turns out that we have learned to read and write various target states in the

collective

intelligence

of tissues and I will show you some examples of that and I think Fundamentally , cracking this bioelectric code, which is a kind of evolutionary precursor to the electrical code our brain uses, will enable a really profound new class of electroceutical products with many new applications in birth defects, regenerative medicine, cancer, aging and synthetic bioengineering, as well If we boil the whole talk down to one sentence, what I'm here to tell you is that, just like the brain, all the

cells

and tissues in your body make decisions and these are decisions about structure and function and that now we can, we have the techniques. really read the information processing that goes into those decisions and modify it rationally and what you see here is one of our five-legged frogs and this is my reminder to tell you that you will see all kinds of strange looking creatures today and that is not Photoshop, these are real animals that we produce as a means of testing our theories about where shape comes from and how to control it, so let's first talk about some fundamental knowledge gaps that I want to highlight.

What do we not know despite all our incredible advances in molecular biology? So what we would like to have is something we call the anatomical compiler. The idea is that one day you can sit in front of a computer and draw the animal or plant you would like to have, so don't try to create it from the bottom up from a description of the pathways, you shouldn't need to think about the pathways that you should be able to, like we do for computer-aided design, if the when the field is mature, you should be able to sit down and say: I want a three-headed flatworm with a shape that looks roughly like this and what the compiler would do is converting that anatomical description into the set of stimuli that they have that will be given to those cells so that they produce whatever you have asked for, okay, that is the ultimate goal of this field, is the complete and total control of morphology and it could be a liver and it could be a heart and it could be a whole organism or it could be something that never existed on earth before, it shouldn't matter, we should be able to take a description of the anatomy we want and figure out what stimuli would have to be given. to the cells so that they are capable of producing it and the fundamental thing is that if we could do this we would immediately have all the answers to birth defects, to the regeneration of lost or damaged organs, to the desertion of the body plan known as cancer, to aging , to all of these challenges to the normal, healthy morphology of a body and therefore this is fundamentally an information processing problem, as I will argue, and the question is why don't we have this yet.

You know, genomics has advanced very rapidly. Why don't we have this? Of course we don't have anything remotely close to this, our control over anatomy is still extremely poor and the reason is that there are fundamental gaps in our understanding and I'll just give you a very simple example, so here on the left you see a axolotl and axolotl larvae have legs and on the right you see a frog larva and frog larvae do not have legs so tadpoles do not have legs and one can do a thought experiment and say that while going to the early embryonic stage I'm going to take 50 percent of the axolotl embryo 50 percent of the frog embryo I'm going to mix them together and I'm going to make a very nice frog and it turns out that the cells work perfectly well they actually work together, but now I ask you a simple question: Will the frogalado have legs?

You have access to both genomes You have access to the axolotl genome You have access to the frog genome Will the frogalot have legs? and if so the legs will be made entirely of axolotl cells or will recruit frog cells. The point is we have no idea and until we do the experiment, of course, well, even having the genomic sequence, these basic, fundamental questions about what shape it will turn out. are something that we can't answer yet that really holds back regenerative medicine, so let's just ask the fundamental question of where does the form come from and we all start life as this kind of ball of embryonic blastomeres and then here you see. a cross section through the human torso also look at this amazing arrangement all the organs are in exactly the right place orientation size position relative to each other where does all this come from and if you know, you might be tempted to say the genome and therefore The genome certainly produces the cellular hardware that you need to create this, so the genome gives you all the protein sequences, but the genome doesn't say anything directly about this kind of order, it's just, it's just not there. in that way and then we really need to understand how cell groups know what to do and when to stop.

It is a

collective

intelligence

problem in the sense that all these cells are active agents that do various things and the result of their activity is to self-assemble this amazing organism and as part of understanding this process we must ask ourselves, first of all, In regenerative medicine, if part of this is missing, how do we convince the cells to rebuild it again? and as engineers we might wonder what else we could build. Could we build something completely different from these same cells? So this central question of where anatomical order comes from and how we control it is the key, I think really really profound changes in the way we do medicine, so the important thing is that individual cells are incredibly competent, so here you go This creature is called lacrimal and you can see what it is doing, it is hunting, it manages all its physiological functions. its anatomical and behavioral needs are all in a single cell there is no brain there is no nervous system there are no stem cells there is no communication between cells it is a cell and it is doing and it is doing everything it needs on the scale of a cell now when the cells I understand when when cells come together, they actually increase this capacity and work on much larger goals and these goals are the production of complex invariant anatomy like this here and we know that simply differentiating stem cells into particular tissue types is not is enough because then you can get something like this down here, which is a teratoma, so this is a tumor that could have skin, hair, teeth, bones and muscles, and the reason why this is not a good embryo is because those three are missing. -dimensional uh organization that structure all the stem cell differentiation proceeded well you have your cell types you have your mature derivatives but they are not organized in the right place so it is not enough to have the building blocks, you actually need to specify how they come together, so there's kind of a generalized paradigm that thinks about an emergent feed-forward process where you have gene regulatory networks and these genes turn each other on and off. proteins, some of these proteins are sticky or they exert force or they diffuse and so on and as a result there is this process that they call emergence and in the end something really complex comes out and the difficulty with this paradigm for regenerative medicine is that if you want to make changes here, if you want to make repairs in the anatomy, you have to somehow reverse this chain to figure out what you are going to do here, what genes you are going to edit and how to create complexes. anatomical changes and that inverse problem is fundamentally untreatable, it's extremely difficult and this is basically what's holding back regenerative medicine, we just don't know how to determine what changes to make at the lower levels to give you the desired uh.

The effects on this at the system level and what we're very good at in the field is figuring out things like this, what genes and proteins control each other, etc., these pathways, what we really want to understand is where these originate. different ways. they come and what do we do when we want to rationally alter them um okay so um uh I should point out that it's not just about creating a complex body during embryogenesis, but also about the amazing ability of cellular collectives to repair Over time, what you see here is an animal called an axolotl, so this is a Mexican salamander.

These animals throughout their lives regenerate their eyes, their limbs, their ovaries, parts of their heart, brain and spinal cord, so, in a way, they bite. each other's legs all the time and you know that if you keep them in an aquarium and the legs grow back and this is what happens, it doesn't matter where you amputate or where they lose it along this axis, they will make exactly the right amount . of growth and pattern and then you have an indistinguishable limb at the end and then, what's remarkable about this is not only that it quickly creates the right tissues that it actually stops at the right time, the fact that the regeneration stops is really surprising. , think about it, when does it stop?

It stops when a correct salamander limb has formed and until then you have really rapid cell proliferation and movement and then it changes in gene expression, but all of that stops when a correct salamander limb is formed. has been formed, how do you know what the correct salamander limit is? How do you know when to stop? Because this is not simply a process that always starts in the egg, continues through a series of programmed steps, and then stops. This is a flexible process that you can interrupt, you can make changes, you can cut at any point and it will continue in exactly the right place, do what it needs to do and then stop when it's done, so there is a degree of plasticity and problem solving adaptive here, so of course.

Um, axolotls aren't the only creatures that regenerate, so the human liver is regenerative and even the ancient Greeks knew, I don't know how they knew, but they did, as you see here. Deer are a large adult mammal that regenerates every year. amounts of bone and skin and innervation and vasculature, these antlers grow at a rate of one and a half centimeters per day. Consider that one and a half centimeters of new bone per day is possible in mammals and even in human children, generally below the age of between 7 and 11 years, if the fingertips are amputated, if they are not sewn with skin but that keep themClean, they will regenerate, so even human children initially have some regenerative capacity to regenerate their fingers. and then and then it disappears okay so these here are, I think, the champions of regeneration and this should be a really motivating and aspirational animal for all of us who are interested in health.lifespan this is a planarian these animals have a couple of interesting properties first of all they are a complex animal with a true brain and bilateral symmetry they are not like earthworms they are similar to our direct ancestor well first of all they have the ability to regenerate so you can cut them up In any piece, along any plane, the disk is something like 276 pieces, each piece will regenerate exactly what is missing, no more, no less, and give you a perfect little worm in each piece, they regenerate their brain every time. . another organ not only that they are intelligent so they can learn, but they have the ability to learn in a variety of behavioral trials and, best of all, they are immortal, there is no such thing as an old planarian, so the cells Individuals age and fall and regenerate what's missing and they just carry on and what these guys are telling us is that it is possible to be an immortal, highly regenerative creature that also has intelligence.

This isn't just for sea anemones and things like this. actually being a bilateral with a true brain, the same neurotransmitters that we have and the immortal, okay, so this is a proof of principle right here, a fruit of existence, so one of the really important concepts to try to understand. How this all works is really the idea of ​​problem solving through the collective intelligence of cells and I want to show you just one example. There are many examples. I want to show you an example. here is a tadpole face and here are some eyes and the brain. and the nostrils and the griffins need to become frogs and to become frogs they rearrange their face so that the nostrils have to move the eyes have to move forward the jaws have to move all these things have to move and it used to be thought that in some ways what genetics does is it gives you a programmed set of moves, after all all tackles are the same and all frogs look the same so as long as everything moves the right way you should pass from a normal tackle to a normal front.

So what we did was we did the so-called picasso tackles where everything is in the wrong place, so the eyes are out of place, um, the nose is not where it's supposed to be, the jaws are on the side of the head , everything is, yes, everything is not where it is. That's supposed to be the case when this happens, what we find is that they still make largely normal frogs, because all the different components are moving in novel paths, okay, and in fact, sometimes they go too far and they have to backtrack and come back, but everyone moves until they practically reach a normal frog phase.

This is remarkable. This means that what Generation and when you start it in the wrong configuration, something that never appears during normal embryogenesis, uh, no problem, they can still navigate more space, they can navigate a path to where they need to be from this completely unexpected starting position, so this if If we had a swarm of robotics, an instance that was capable of doing this, you know, instead of cells, we had little robots that could do this, this would be a remarkable example of collective intelligence, which is why there are groups of cells here that we are able to solve this problem from different initial configurations and this is the process in which that collective intelligence and that problem solving is what we need to solve, in addition to all the mechanistic advances on the component pathways to be able to really control our anatomy. and to really solve regenerative medicine, then what we would like to do is try to get to a point where we really understand this on an algorithmic level, not only the mechanisms that are required to make this happen, but also the algorithms that are sufficient for this.

This is going to happen and we really need to understand how these cellular collectives process information and how reprogrammable it really is, so computing gives us an interesting analogy in the 1940s and 1950s, this is what programming was like to be able to program this computer. you had to physically move the cables, you had to, you had to physically change the structure of the machine and since then, the journey that computing has taken is to realize that if your hardware is good enough, if it's reprogrammable, then you don't need physically change the hardware, what you can do is interact with it through inputs or experiences and take advantage of the software that is in the system and the plasticity that is in the system, okay, so of course, what do you want?

You'll see what you'll notice that all of modern biology is basically down here at this point we're um all the exciting advances are single molecule approaches uh you know we're reconfiguring pathways genome editing um it's all about getting more and more specific at the low level. of hardware and these are very important things, but they have to be combined with an understanding of what the software looks like and how much we can do with stimuli, not with rewiring, whether genetic or otherwise, the framework we work in looks a bit similar more to this, where, yes, there is a feedback component, but there is also a very important set of feedback loops where any challenge to the body and this could be a traumatic injury, it could be aging.

These could be keratogenic, pathogenic, anything that triggers a set of feedback loops both at the physics and genetics level that serve to return to the correct target morphology. Anyone who has studied engineering will immediately recognize this as a homeostatic cycle, it's basically what you have in your home thermostat, where this is a system that is able to regulate itself back to the correct state now, of course, if the feedback It is not, it is not a new biology. All biologists understand the importance of feedback, but two things are unusual. The things here in this model that I'm going to talk about first is that the set point for this homeostatic process is not a simple number or a scalar like temperature or ph or hunger level or something like that, it's actually a lot more complex. structure that is a coarse grain of anatomy, so large scale pattern information is what is being what guides this homeostatic process and the second part that is a little unusual is that this is a process that has a goal in the when I say objective this.

It's not in some mystical sense, it's basically cybernetic, in the sense of cybernetics and control theory, there is a set point and this is a system that is fundamentally capable of expending energy and effort to return to the correct region. of state space by uh, activating all kinds of cellular behaviors to get as close as possible to minimize the error and minimize the difference between the current shape and the shape it's supposed to be, so if you make that hypothesis, it has a couple of strong predictions A strong prediction is that you should be able to find the medium in which the target morph is encoded, so the set point should be recorded somewhere and then it means that you could alter that set point, you could rewrite it and ask machine. build something different without having to rewire it the same way you can set your thermostat to a different temperature and the exact same system will now maintain a different temperature because there is a distinction here between data and execution so this is what we've been doing in my group for a few years: saying: can we find the representation of the target morphology and exploit it for regenerative medicine?

I think we found an important part of it, so to tell you. um, how that target morphology is encoded, I have to introduce you to the bioelectricity of development and all the cells in the body are located in a morphogenetic field of information that comes in all kinds of different media, so there are chemical gradients and stresses and strains in the extracellular matrix, etc. And then the one we're interested in is bioelectricity, so bioelectricity is not the only important layer, but I think it's exceptionally powerful and interesting and that's the one we'll talk about now. And?

It's bioelectricity, well let's think about what happens in the brain, in the brain, you have networks of these cells called neurons, they all have ion channels, they create voltage potentials by passing ions in and out of these ion channels and they can go through those states electrical to their neighbors through these small electrical synapses, the synapses are known as gap junctions and the networks of these cells have this really interesting property that they can perform physiological information processing that we associate with memory, behavior, goal-directed activity, preferences, etc. So what you can see here is an example of tracking electrical activity in the brain.

It turns out that this is a zebrafish thinking about whatever the zebrafish is thinking about and they can literally see the electrical activity and neuroscience has the following commitment: we think that if we understood the coding, if we understood the neural electrical code, we could observe this electrical activity and extract semantic information. We would know what the meaning is for the animal, what the content of that memory is of whatever it is thinking. Well, people are working and have had some success with all kinds of decoding methods to try to extract that information from the electrical activity in the brain.

Well, you may wonder where all this came from, certainly. It just evolved out of nothing, it must have had a simpler evolutionary precursor, and it did. The evolutionary precursor is the fact that all cells do this, not just neurons, so every cell has ion channels, most cells have gap junctions with their neighbors, and this, um, this. This kind of scheme of putting electrically active subunits into networks is as old as bacterial biofilms, it's an extremely old evolutionary thing and all cells do this, so what you're looking at here is an early frog embryo in a span of time where the cells are having discussions with each other electrically about who is going to be anterior posterior left right dorsal and so on because while the electrical activity in the brain is designed to move and activate the muscles to move the body to solve problems in space three-dimensional, what developmental bioelectricity does is process information to move the body through morphospace to change and control morphogenesis and we can do this exactly the same thing here we can we can we can try to decode this to try to understand how specific patterns arise and how we can change them Okay, so we have developed, my lab has developed some tools to do this, so first, here are voltage sensitive fluorescent dyes and this is, my colleague danny adams perfected this particular technique of using voltage reporters, so this It is not a model. a real embryo using this voltage dies to read in real time all the electrical conversations these cells have.

Of course we do a lot of computational modeling and then more importantly we have functional techniques where you can go in and change the electrical information processing in the tissue we don't use electrodes, applied electric fields, there's no electromagnetism, this is pure physiology molecular to target the native processes by which these cells communicate electrically, so there are ways to control the topology. of the network, so the cells communicate with each other through these separation junctions and we can open them, we can close them, we can mutate them, etc., and the same for ion channels, we can go in and literally set the electrical states. using optogenetics or drugs or mutations that change the properties of the channel, okay, exactly, basically, taking all the tools of neuroscience and using them to control morphogenetic decision making outside of the nervous system, and this is, um, this is just a simple example of what some of these endogenous patterns look like here you see a time lapse again this embryo is putting its tadpole faces together you can see all the dynamic electrical patterns here is a frame from that movie and the amazing thing is why I love it so much this example is that the coding is really simple here the electrical we call it the electric face the bioelectric pattern of the face actually looks like a face and this appears before the genes are activated uh to um to regionalize the face and then here you can see this is where the eye will be here the postersthey use where the mouth will be all these bioelectric gradients are a pre-pattern they are a subtle scaffold that directs gene expression and anatomy downstream and if we go in and change these uh the voltage distribution here you will change the gene expression and you will change the anatomy which is one way to make those Picasso tadpoles that I showed you so this is a native pattern that is required it is a normal and natural part of specifying what the tackle's face will look like there is a pathological pattern what we have done here is oops what we have done here is to inject a human oncogene into the sample, they will make tumors like this but before they do that, you can already see using this bioelectric map, you can already see that these cells have a really aberrant electrical potential, they are dissociating from the electrical grid , they're basically going back to their single-celled state and they're going to treat the rest of the animal as a simple external environment that has been disconnected from the collective and this is metastasis, this is just um, this is a lack of participation in embryogenesis because the first thing that oncogenes do is to uncouple you from the electrical grid that normally tells each cell what it should be doing, so the reason we know this is important is that if you then inject tumors like nasty chaos mutations and so on, if you inject these things and causes tumors, you can prevent actual tumorigenesis even though the oncogene is expressed very strongly, so here this is the same animal, so here is the oncogene that we have labeled it with very strong red. expressed and yet there is no tumor because what we have done is coinjected an ion channel that forces these cells into a voltage state that keeps them coupled to the rest of the electrical network and they simply do not progress to the genesis of the tumor or the metastasis, so we know this is instructive, now we go beyond the single cell property of the conversion to cancer.

We could look at morphogenesis and we can say, "Okay, here's a tadpole, let's take a region of the intestine." So here are some cells in the embryo that will become intestine and we are going to inject some ion channels that will establish a voltage state that is normally associated with making an eye and guess what happens to these cells once you give them instructions. that way they will actually build an eye from cells that died out in the intestine and you can make these eyes anywhere in the tail of the spinal cord in the intestine, whatever they have all the right layers, the nerve retina optical, all that and also beautiful.

The cool thing is that they instruct their neighbors, so these blue cells here are the ones whose voltage we changed and they recruited a bunch of other cells that we never touched directly. These are just neighboring cells that they recruit to make this nice lens that's sitting around. in the tail of this uh of this tadpole and then we can use this this technique we can induce eyes brains um hearts and uh and and other things and then there are a lot of things that we can't, we don't know how to do So the point is that these bioelectric gradients are instructions for anatomy that control on a large scale and one of the most important features of this is that we don't have to micromanage this process, so what we basically find is an instructional hook in subroutines that say make an eye here or make a heart here or make a brain here we don't have to go in and give all the information about how an eye is made what are all the layers or all the cellular differentiation cascade, all the genes that have to be expressed, that would be incredibly difficult, we are not not even close to being able to do it as a field, but what you can do is find the native, modular, modular control structure, in the, in the in the bioelectrical layer that determines which organs are placed in which regions and then the cells will take it from there once you've instructed them what to do they'll turn on all the things they need so let's go look at the planarian because this is really interesting in a planarian you cut off the head and the tail you have this middle fragment the middle fragment um it has You have to know how to build a new head here and a new tail here.

How do you know? To do it right, you have this interesting voltage gradient that says a head here in this depolarized region a tail here and if we go in and artificially manipulate the voltage here to depolarize it, then surely these cells will build what you said and create a two-headed animal and these are perfectly viable, so you can make a two-headed animal, you can make a headless animal, okay, so again it's instructive and all we've done is activate the build-ahead subroutine and we've done Here, in this particular region, without knowing too much about how a head is built in the first place, the cells will turn on and off all the genes they need once they have received this high level of instruction about which organ. are supposed to build, not only can ectopic heads be built from the same animal, but heads appropriate for other species can also be built.

Well, if we do this, we take this planarian with a nice triangular head, cut off the head and disturb the power grid. and you can get flat heads like a pifilina, you can get round heads like a mediterranean or you can get your normal triangular heads and this is because you not only get the shape of the head but also the shape of the brain, uh, changes in the distribution of the stem. Cells become like these other species because the space of possible outcomes of this electrical circuit contains attractors that correspond to other species that evolution has encountered in the past and you can mark them without any change in genomics, so both here and then the two-headed animals that I showed you have nothing wrong with the genome the genetics are completely wild this modifies the information stored in the electrical network and you can access other attractors that belong to other species with exactly the same genetically normal cells and in fact , you can access regions of that state space that evolution doesn't use because they aren't viable.

You can take flat planarians and turn them into these strange pointy shapes. You can make these three-dimensional things that look like hats. You can do these kinds of things that are sort of combinations so that flatworms no longer need to be flat, so the idea here is that you can really have drastic control over large-scale anatomy given the same genome, because that? The genome is built are cellular collectives that use electricity to store their set points, their target morphologies and this sounds like a very familiar story to neuroscientists because this is exactly how brains work, so we come to the A kind of applied regenerative medicine To this, unlike salamanders, frogs do not regenerate their legs and what we have done is that we took little frogs and designed them.

We used this information to design a cocktail of medications that changed. the electrical state of the wound and you can see here normally on the frog, you amputate the leg and basically there's nothing in 45 days, um, we added the cocktail and the applications really only for 48 hours, sorry, 24 hours, so one day and after. that day the exposure starts the whole cascade so here the pro-regenerative genes are activated so this is msx1 and then immediately you start growing this leg, 45 days later you have some toes, You have a nail, eventually a very nice leg and you can see that the leg is sensitive to touch and mobile, so what you can do is override the normal growth program, which is normally very small, and give everything with a simple signal for um 24 hours. then it gets picked up by all the transcriptional cascades and everything else that will work for months, in fact, our last article 13 months of leg growth after one day, um, one day of treatment, okay, so we know we've looked at this in the following way.

These are frogs and worms. We have observed this in human mesenchymal stem cells and cardiomyocytes, etc., and it is highly conserved. It works throughout the tree of life. We know it is relevant to humans because of many human channelopathies. of human ion channel mutations lead to different types of birth defects, which tells us that even in humans these electrical properties are very important in establishing the correct morphologies and what we do now and this is our company with uh david kaplan the tufts we have a company that focuses on uh david's lab builds these bioreactors that basically deliver um uh uh ion channel drug cocktails that are selected by us to induce pro-regenerative states and now we're having more or less salty in frog now we're in um in mammals we're doing this in mice and then hopefully one day hopefully humans of course so just so you know it starts to end here the point of all of this is we're now starting to We understand circuits to the point where we can make models at multiple scales, so we start with knowing what genes are there to produce various types of ion channels, but these ion channels open and close, so the simple fact of knowing the genes does not tell you what the electrical current is.

The states are going to be, you have to simulate the physiological tissues and those give rise to larger organ structures and eventually you can get to algorithmic types of control that show you how the electrical activity in these tissues makes decisions about what's going to happen. and where. you go from genomically specified hardware to the low level and then to the higher level software that allows these cells to make the kinds of decisions to robustly create and repair complex morphologies, okay, and in that process, of course, there's a huge um opportunity to apply machine learning and so we've done some of this using machine learning to discover electrical circuits and discover the modulations that are needed to use them therapeutically and very simply, I'll just close.

Up here with a couple of slides to show you a very simple example. This is what a normal tadpole brain looks like and if you disturb it with a mutation, say in a really important gene like notch, that basically the forebrain disappears. The midbrain is a bubble, behind the brain, you know, it's also a bubble, but they don't have any behavior, this is there and what you can do is create a model of the bioelectric circuit that determines the shape of the brain and ask it in circumstances like this . Either there is a mutation or they have been hit by alcohol or nicotine or some other type of poison.

What would you have to change? What channels would you open and close to return to this correct bioelectric pattern that is responsible for normal life? brain, so the model is now detailed enough to tell us what channel it is and using it, we can choose drugs known as ion channels and when you do this, you get the right brain. mold the right brain size and you get their good behavior, you get their IQ back so their learning rates go back to normal even though they've had really debilitating mutations or exposure to teratogens like nicotine or alcohol etc.

Well, this is an example of starting with a basic understanding of bioelectric circuits and how they work to a computational model that actually helps you design therapies to solve a fairly complex outcome, which is brain development. Okay, so what are we doing? Now we're creating a complete process where you start with existing information about what ion channels are in the tissue of interest, what bioelectrical state you want, and then use the model and simulations to inform you. What channel openers and blockers do you need? Something like 20 of all drugs are ion channel drugs, so there's a huge set of tools that we call electroceuticals that with the right, correct computational simulations, you can start to implement.

Outside of things like, you know, cardiac arrhythmias and epilepsy, all of those drugs can now be repurposed for regenerative medicine, so this is an example of our software. Everyone can help play with it. It's online and The idea is that you choose the tissues you want and that would really help you choose the right medications. Okay, so I'll close here and summarize what I have. What I told them is that there is an incredibly powerful layer of physiological software that sits between genotype and anatomy and it is a really tractable target for the evolution of regenerative biomedicine.

I discovered very early on that electrical signaling is a really convenient medium for global computing and decision making. why brains exploit it, this is why our computer technology exploits it and this is why morphogenesis runs in the system cracking this code. You can really give us the opportunity to rewrite the pattern memories that allow us to control shape on a large scale when I say. pattern memories, I mean, these are literally those two-headed worms that I showed you, yeah nowYou cut them into pieces without further manipulation of any kind, they continue to create two-headed animals even though their genomes are wild type, okay?

It's um, this is that memory, actually that electrical circuit actually holds the memory, so you can reconfigure it to be an aggregate one, but until you do that, the pieces will continue to form two-headed worms, so now there's new artificial intelligence tools. We are online to help us design these types of strategies, so it is now a rational design to address birth defects, cancer and regeneration, and also our latest work has been to create synthetic living organisms, so I would like to thank the people who contributed to all this. work, our sponsors, um and uh, and I'll thank all of you for listening to us and if we could just take a five minute break, then I'll come back and I'll be happy to answer any questions.

Wow, that was a lot. I think we need a break, we'll let you go for now and have you back in five minutes. Thank you very much in advance and there are many questions in the chat, but I will moderate them and we will see you at a perfect time. five minutes, okay, sounds great, okay, lovely, well that was a lot to digest, continue with the questions and also upload the ones you want to push. He already accepted that he can stay for half an hour up to half an hour more. So for those of you who want to learn more about this, feel free to join for a longer time, for now we have many interesting community updates, many projects, events, scholarships, hackathons and so on in the community that we will start. with elena elena if you are here for the future, I am silent and tell us about your next conference yes, hello everyone, it is very nice to see many useful developments in the field and listen to all those amazing talks, I am actually a very big fan of this group , although I can't always attend anyway, I just wanted to let you know that we will soon have our fourth annual conference of uh and in Adriatic diseases 2021, it will be held virtually this year again because we are not so sure that it will be safe for all of us together.

We expect that to change in the coming years, of course, so the conference will be held in August 1922 and in the Eastern time zone, so we will focus on the four traditional topics which will be the mechanisms of aging, the human trials of biotechnological rejuvenation therapies, biomarkers of aging and of course traditionally investment issues and regulatory issues, as usual we have several leading researchers giving their talks, some of the information is available. to be pretty fresh right out of the lab, so I hope you take a look at the conference program and probably get inspired and join us this year for more information, visit our website lifespan.other conference and as a way to Support this community.

You can apply a promo code for the 2021 site to get a discount. I'll leave a link that will automatically apply a discount in the chat for this call in case you want to continue right away. Thank you so much. Ok, and next we have Anastasia Anastasia, are you here? Oh can you hear me? Yes, beautiful background, oh thank you, it's Idaho. Yes, thank you, Alison, and it's a pleasure to meet you all at these regular meetings that are really great and important and informative. Some of you may know me and I work with Ranjan um Rajon is the founder of the group r42.

We are a fund and we also invest in AI and longevity and also an institute. Part of our institute that we have is a scholarship, so it is a scholarship and now we are in our cohort four and our demo day is actually tomorrow between 8 a.m. and 7 p.m. and 12 p.m. noon, Pacific, some really cool projects. Students from all over the world were working on these projects over the 12-week period, so they include venture capital math, no-code machine learning platform for biomedical researchers, cancer biology, a wide range of really interesting projects, as well that you are all welcome to come and I will send you the schedule and also the registration link in the chat as well as yeah feel free to come and go and come see all of our new performances so yeah great and yeah thank you here, you could add some words, let us know in case you want to do it, ask and otherwise we will wait for you.

Yeah, okay, it's free, I'll just add that it's free, so tomorrow we'll have the first holographic phone call between England and Silicon Valley, so it'll be around 11am. m., exciting Pacific stuff you're taking. ai very meta topic okay lovely next we have dima yeah hi everyone so we are organizing a longevity hackathon so it's a 48 hour event to make ideas in longevity a reality and the idea is to bring together to longevity researchers, developers, um. Entrepreneurial AI scientists bring in people from outside the field and also help them build teams and address the kind of long-term fruits where innovation can happen in just 48 hours, so there are issues of biotechnology and pharmaceuticals, we already have people. from all continents, except maybe Antarctica and Australia, but there are people joining from all over the world.

There are longevity technology funds that will invest a hundred thousand in the winning teams. So I'm also sure this will continue afterwards. and then the Foresight Institute is also supporting with the scholarship for 2022. Oh yes, please join, we are still looking for sponsors, mentors, feel free to contact us and I will leave it and chat also about the information about it. Thanks, okay, I have one more left, which is the conference on aging research and drug development that will be held in Copenhagen from August 31 to September 3. I'll leave a link to it and as you can see there's a lot of interesting things happening, I think it's happening Nathan. the travel scholarship I contacted you before if you want to say something feel free to do it now or in the chat otherwise I think we have Michael back okay Michael are you back?

I'm back, okay, lovely, okay, great, I think let's see, let's maybe ask some questions on the record and if anyone would prefer to have theirs on the record, then I'll turn off the recording. I think some people prefer that for a more open discussion, but the first one that was uploaded a bunch is called call us I think three questions so you can make your choice now don't pressure you this is totally good these are good questions to get on the record because um Well, I mean, first of all, let me congratulate you on this. It's amazing, thank you, it sounds and it blows my mind a little bit, so before we get into the topic of regenerative medicine and aging, even this is more basic, you know, I have a lot of more basic background questions about these networks bioelectric. the meta question in you know, I'll just give two example questions the meta question, you know, the huge number of questions it asks is what is the best modern review or textbook that presents to all those people who don't know anything about these bioelectric networks. to that, but the two example questions I had right away were: what is the down?

Do you know what is inside a cell? What is the path for these bioelectrics to truly control gene expression? So what is it? happening in the cell that way, how does it work epigenetically and um and then the other question I wrote was um in non-regenerative species after development, what do these bioelectrical networks do with anything? Yes, yes, great. Okay, three three questions, let's see. As for reviews, I have a lot, I've written a lot, a couple of other people have written some good ones. Anyone interested please email me or if you have some kind of forum I can post on.

I'll do this, post it and tell Allison later offline. She can, Florida, sure, yes, yes, yes, yes, we have written tons of very basic reviews on this stuff, taking you from zero to what is known today. the second question is transduction to gene expression, so at the single cell level we know that there are about half a dozen ways that voltage change regulates downstream gene expression, so this includes things you're familiar with with neuroscience, for example. Voltage-dependent calcium signals followed by calcium transduction. This includes the control of the movement of neurotransmitters like serotonin between cells through gap junctions or through the serotonin transporter and this includes some more exotic things like voltage-sensitive phosphatases.

It includes things like voltage-dependent butyrate transporters that then press hdac and those kinds of people know the chromatin modification pathways, that's all known, I should say, although it's very unsatisfactory because what I want to say is good information, it's good knowing that the problem is all that is data at the single cell level and if you want to know why your hand doesn't look like your foot, it's good to know the transduction at the single cell level, but you still have to ask about the global dynamics that determine size and shape and scale it's not enough to know what transduction looks like in a single cell, I mean what's the third oh what does it do in adults?

I think what it does in adults is morphostasis, so it holds them together against senescence and turnover of individual cells. What it is, and by the way, the interface between bioelectricity and aging is generally not well understood at all, no one has studied it, right, we haven't even touched it, so it's not like we know what the answer is. , But but. Basically, I think probably what it's doing is just suppressing cancer and morphostasis and holding tissues together and it's really activating those very dynamic morphogenetic rearrangement modules that are needed to really rejuvenate and restore after injury, are you satisfied?

I sneaked in three questions and called, okay, next one we have Ronald, we do well, fascinating stuff, so what controls the appearance of these electrical zones, what determines their dispersion as well? Yes, so, there are two, there are two, there are two, two important pieces to this. An important piece is what underlies the hardware itself and so you have to ask in order to have any kind of electrical state, you have to have the right ion channels that are capable of producing them, so one thing you have to understand is what channels there are. and what their properties are, for example they themselves could be voltage sensitive or they could be ph regulated or whatever, and this controls the properties of these channels, they are the excitable medium in which these electrical phenomena now propagate the patterns specifics, the specifics, let's say that the electric face pattern is an emergent pattern in the same way that tour patterns emerge from a chemical substance from an excitable chemical medium or the way that, you know, when we talk about This with my students.

I start and say that you visualize that you have a collection of electrical parts that make a calculator and when you turn it on it has a predetermined behavior, since the parts were chosen so that when a juice is extracted. on and the thing is it's running, by default it will start in a constant reference emergency state which is zero now it has cool properties of reprogrammability and all sorts of other things it can do, but there's this. This basic state is that the parts that are formed by evolution, um, there is an emerging pattern and we can study that symmetry breaking, so it is a symmetry breaking process of long-range inhibition amplification, short-range activation that establishes, you know. set exactly two eye points at a particular distance from each other, all these things can be modeled, are you satisfied?

I don't know, yeah, cool, lovely, cool, next up is Josh, hey, thanks again for the chat. super interesting my question was that the worms put planarian is how you pronounce it um if you cut you mentioned that they could have learned behaviors. I was curious if worms cut off multiple pieces if once those pieces grow if all the pieces retain those learned behaviors like the original worm yeah that's a great question um it's uh it was first addressed in the 60's by a guy called mcconnell and he published his observations that, in fact, they do, you can take a worm and you can train on a particular task cut the tail raise a new worm from that tail so in 2013 we reproduced that whole aspect of his work , he did other things, but that's what we reproduce and it's true that they do, so what that tells you is that uh even The kind of learned behavioral information, not just morphological information, is stored outside of the brain that you knows in some way and can be imprinted on the new brain as it forms.

It's quite remarkable because that tail doesn't do anything until the new brain is formed.Anyone in this community, generally speaking, what we've been doing for years to try to lower the barrier for other people to enter is we post a lot of reviews. and a lot of procedural protocols, so we put everything online to the extent that you want to go into this from any system, whatever you're interested in here it's literally step abc, like what do you do next correctly? So we send um gift baskets with reagents that we provide um all the plasmids, the dye protocols, uh, you guys know all the data interpretation stuff, we, all of that is available, so anyone who's interested, I would be more than happy. to help them do it.

This would be great, fantastic, thank you and, well, you also mentioned that there is a company, right, and that you could clarify quickly at the beginning of the talk, and what happens with that next, what are your next steps in this field. Yeah, it's basically the company, the initial focus is pretty narrow, it's limb regeneration, so what we're doing is testing our various cocktails on mice using David Kaplan's portable bioreactor, the short-term application of our ion . channel cocktail can we trigger, um, can we trigger limb regeneration? That's the beginning, obviously, my dream is to grow this a little bit, um, you know, a fundamental approach to choosing electroceuticals for all kinds of indications, from birth defects to regenerative medicine. from aging to cancer, but right now it's just about limb regeneration, okay, and anything that people in this group can do to help, for example, are you looking for collaborators for that company or investors or are there something?

What are the next steps? I'd sure like you to know that we're happy to hear from investors. um sure uh anyone who wants to collaborate send me an email and tell me what you do and we'll see what you know we'll see what. which makes sense yeah, yeah, we're our lab, it's very collaborative, we work with all kinds of people, so, sure, I'd love, I'd love to hear from this community, cool, it sounds like you're already collaborating on at least the team unknowingly around the corner. One last thing, you know, we ask people to submit a longevity channel challenge that you've done and this is just another package where people can respond to this challenge and actually again up to 250. dollars to the do it, do you want to maybe say a few words about the challenge that you presented?

Yeah, well, I'm serious, I wasn't very familiar with how this all works, but basically, the big challenge in this field that I see is just uh, what I think, let's put it this way, what I think would make a real advance and would advance the field. Kind of a watershed effect would be the ability to actually visualize these bioelectric states in all kinds of important conditions and not because the images alone you know, I really focus on functional experiments. I mean, seeing the beautiful patterns is nice, but the most important thing is that they are instructive.

Can I show that when you alter the pattern, the gene expression changes and the anatomy changes and we do that and that's what my lab focuses on, but it would certainly be much easier to develop those kinds of functional approaches if we had the reference data from other systems model other conditions various diseases conditions various human organs this would be a real game changer, so that's what I put.