 OK, let's get started. So we have today's lecture on development. There may or may not be another partial lecture on development, depending on how ambitious my colleague Professor Lester feels about supplementing development. We used to have at least two, and maybe in the past even more than that, lectures on development. So it's very compressed. This, by the way, is a little video of the change in the thickness of cortex in the brain of a human child as it ages. So lots of people are doing developmental studies at many different levels of analysis. We're not going to talk about this kind of analysis at all, but about mechanisms that are much better understood and much earlier in development. But just to say that this is one thing that people look at. They do longitudinal studies and look at how things change during childhood adolescence in humans. And one point here is the different parts of the brain mature at different points in time. I think I made this point to all of you when you looked at the human brain in your discussions sections yesterday, that this correlates to some extent with the different capabilities that babies, infants, and children have. So different parts of their brain mature at different times. And that's why they're good at some things and bad at other things at certain points in time. They can't control their emotional outbursts very well when they're very young because the frontal cortex matures relatively late. Your problem set, it was already mentioned in the discussion section, is now posted and do some time. It says on the course website. So please take a look. Problem set one, the first real problem set is up there. It will say when you look at the instructions on the first page of the problem set, it will tell you, and I think it's also in the course website, to please type all the answers to problem sets. So we ask you to type them. And then the easiest thing for you to do is to email that problem set by the due time to Jared, your head TA. If you email it in late, you will get decremented in your points according to the formula that's on the website. One quick note that I had here for those of you that are thinking about maybe auditing this course or not taking it and instead taking it next year, this course will not be offered in fall next year. It will be offered in the spring of 2016-17. So this is the reading that you have and you've seen several times. And again, the overall thing happening here is that the reading will get denser and it's denser yet for next week. But again, you're responsible for these specific pages, which cover more than what there is in the lecture. That said, the problem sets and the exams are going to not be limited to what's in the lecture. They're also going to cover other things in the book, but they will emphasize what is in the lecture and in the book, so what you've heard about the most. And for some of this, in particular, the development chapter, I know that this can be overwhelming, especially for those of you that are not native English speakers, because there's a lot of words for different genes and they're all weird words. And so it's very hard to memorize all of them. The most important ones would be the ones in lecture. But please read all other parts in the book that are assigned as well. Any question about the reading, the problem sets? Everybody found their section? I think Geron will be in touch with you about possibly switching sections. It would help if people switched from the 4 to 5 section into the 7 to 8 section to equalize numbers. Yes? So is there a quiz today? There will be a quiz today, but not right now. So pay attention to the lecture right now. OK, so why is development interesting? There are many reasons for this, and I just wanted to motivate why this is interesting and why you should think about it, and read, and think about what's in your book. Wait, what happened here? There we go. OK. One is sort of engineering perspective. Engineers often like to say that you really have to construct something yourself in order to understand how it works. And so understanding the mechanisms by which an adult brain forms, of which there are many, would help you to understand how the adult brain functions. And it turns out that there's no way to build a human brain except to have it develop. And of course, people in robotics and AI are taking some clues from this. There's just too much information, and it's too complicated. If what you want to do is to take, say, an adult human brain by just starting with individual neurons, even if you had an atomic scale manipulation device, and you were just trying to hook up every neuron to every other neuron. First of all, it would be very difficult or impossible to do that in the intact human brain, because you would get in the way of other neurons. And the amount of information required would be absolutely enormous if you had to specify how every neuron is hooked up to every other neuron. Instead, as we'll see, there is a temporal sequence to some extent fairly simple rules and fairly conserved rules that have molecules that are conserved across different species. And it's the sequence of that, the sequence in which genes are expressed in time and where they're expressed in different tissue, and the interaction of all of that machinery, all of that happening with the environment, that results in an adult brain. So if you understand the principles of neural development, you will understand a lot of what it is that gives rise to the adult human brain and a lot of what's not possible, a lot of the constraints that result in features of the adult human brain. Sorry, you're putting the wrong button here. Another reason why development is very important is that it's ubiquitous. And of course, you're developing right now. You continue to develop until you die. And so people call this now lifespan development. Development never stops. There are periods in your life where your brain is changing more than other times. So it's changing a lot right before birth and right after birth. It's also changing quite a bit when you age, if you have a neurodegenerative disease, but it's changing all the time throughout your life. And so to some extent, some of the same processes that we'll be talking about here in development continue throughout your life. You're not going to be making too many more neurons, although even in the adult brain, new neurons are born. We now know, but you are making new connections. There's activity-dependent plasticity. There's plasticity that depends on interaction with the environment, so your brain is always changing. The other aspect of ubiquity is that, as I mentioned, the principles of neural development and the particular molecules that are involved are remarkably conserved across species. So when you look at your book and you get overwhelmed with all the names of the different genes and their protein products that are involved in development, often have very curious histories relating to particular circumstances or labs in which they were discovered, and so they gave them complicated-sounding names, but they're remarkably conserved. So you're not just going to be learning about development in the human brain. A lot of the same molecules, a lot of the same principles apply to monkeys, to rodents, and indeed, to some extent, to flies. Development is important to understand how it can go awry in developmental disorders. So for instance, disorders such as autism spectrum disorder arise, we now know, very early in neural development. The final consequence of them is not diagnosed, typically, until around age three or so, although people are trying to find markers that would be predictive of whether or not you will develop autism at earlier and earlier points, but we know that something goes awry in development and in how neurons make connections very early in life, and so understanding what can go awry is important for understanding disorders as well. And then the final one is just a philosophical one, which is that the environment, and in particular the social environment in which you develop, philosophers think, ultimately has to ground what it is that your brains can represent. So if you were created right now to know which is from a bunch of molecules that happen to coincide in the right way to form a human brain, it sure seems like your brain, your experiences couldn't actually be about anything in the world because there's no causal connection in terms of the history of how they came about and any event in the world. So lots of reasons all the way from very practical engineering ones to clinical reasons to even philosophical reasons to think development is important. Development has a, neural development has a strong history at Caltech. There's Thomas Hunt Morgan, who started the division of biology at Caltech. Seymour Benzer, who passed away a few years ago, was one of the key people looking at the genes that are involved in learning and memory. And then Ed Lewis, who shared the Nobel Prize with Christian Ruslan-Volhardt, and a third person whose name escapes me right now, discovered some genes that are involved in development of the Hawks genes that we'll talk about later in the lecture. All three of them worked on what historically has been the most informative organism for understanding how genes influence development and that's the fruit fly Drosophila. So all three of them worked on that. There's another Caltech professor whose work we'll talk about in just a minute who didn't work on fruit flies for a change, but he worked on frogs and on goldfish and on humans, actually. That's Roger Sperry, who was also here in the division of biology and he developed a theory, a hypothesis called the chemo affinity hypothesis which he then tested that we'll talk about that explains the topographic arrangement of how axons find their targets. So it has something to do with the principle of topography that we already mentioned earlier and links that to neural development and behavior. He won the Nobel Prize, shared with Hewlett and Wiesel like in 1981, not for this work actually, but for another line of work for which he was, he's also very famous. He basically had two separate careers and that second line of work is work in split brain patients. So he studied, he became very interested later in his life in consciousness and the question of whether or not the left and the right cerebral hemispheres if they were disconnected in split brain patients would each have a separate conscious experiences for which he found some evidence. So what are the problems of development? So cells, sorry, keep pushing the wrong button. So cells need to, cells start out very early in life, all being very similar and then they progressively differentiate to form all the kinds of cell types that you find in different organs and the many, many different kinds of cell types that you've heard about that the brain has. So how do they all become different from one another? At what point in time do they make those decisions to differentiate into different cells and how is all of this coordinated in space and time? So these are the basic problems. The mechanisms and again, the very complex patterns and many cell types that you find in the human adult nervous system like in the brain that you saw last night or yesterday afternoon in your study sections arise from a sequence of by themselves fairly simple mechanisms that all interact. So there are basically three. You have to multiply, so cells have to proliferate. They have to divide and you have to make more. They have to be in the right, get to the right place so they have to migrate and then they have to differentiate. They have to decide what it is that they're going to become. What kind of neurotransmitter phenotype are they going to adopt? What kinds of connections are they going to make? And of course this depends on the particular, they all have the same genes but it depends on the particular sets of genes that they decide to express or not. So a lot of this has to do with patterns of gene expression in space and time. So again, remember there's only 20,000 genes in the genome. The information in the genes is not at all sufficient to specify all the information about the wiring in your brain but it is sufficient for a sequence of steps that would during development give rise to the adult brain. Here is how this looks for a human. So if we have conception here at zero and this is to station period in weeks and here is birth and then you would have a baby and an adult here, this is a simplified view of some of the different events that would happen in the development of the human brain. So neuralation happens very early on. We'll take a look at that in just a minute. That's the closing of the neural tube. Remember that your brain is a tube and then different parts of that tube proliferate and become cortex, other parts of the spinal cord and so forth. But that initial folding happens very early and once you have that, different parts of the neural tube proliferate, cells divide, they migrate to different places. There is cell death, apoptosis, some of which is very programmed, some of which depends on competition for targets and for trophic factors. Cells form connections with one another, synaptogenesis and myelin and sheath's axons so that there can be rapid conduction in communication between cells. All of these overlap in time so it's not as if any of these stop, in fact, neural proliferation, as I mentioned, to a much lesser degree, continues on all the way throughout your life. So you're continuously making new neurons in structures like the hippocampus, for instance, even right now, which is helping you to learn the material in the course but at small compared to the massive neural proliferation you have early on here. Same thing with myelination, this continues on, it's dynamic to some extent, but again, roughly, you first have to make a lot of neurons, they have to migrate to the right place and form connections and then they start myelinating, roughly, these all overlap in time but it gives you some idea of the sequences. The way that this unfolds, the precise orchestration of all of these is what constitutes development of the human brain and the way that this happens in detail is what distinguishes the development of the human brain from the development of other brains. So for instance, people in Japan, in whose lab, I will actually be spending a small sabbatical next spring, work with chimpanzees. So here's a pregnant chimpanzee mother being given an ultrasound and she can watch her chimpanzee baby here on the monitor and in this particular paper here, what they did is to look at the change in brain volume of the fetus in utero here with ultrasound. In the human brain, that's the blue curve and in the chimpanzee, that's the red curve. And what you'll see is that these differ. So the human brain volume keeps accelerating right up to birth. So it keeps getting bigger and bigger at a faster and faster rate. The brain, the velocity of growth is plotted here on the right. Whereas the chimpanzee kind of reaches a plateau and then plateaus and starts growing less. So this is linked to what I mentioned earlier that chimpanzees and even more so monkeys at birth are much more mature and have a brain volume that's proportionately larger in relation to their final adult brain volume than do humans who are very immature. So their brains are still really rapidly growing right at the point at which they're born. Okay, so a little preview here and now indeed we will have a quiz. So if TAs, you could pass this around. Do not yet turn this sheet over. It's a double-sided one piece of paper. Put your name on the first page, please. And then when I tell you to, you can turn it over but not yet. So just make sure everybody has a quiz. Put your name on the first page, please. Okay, so again, you will get one of these every week, not necessarily always on Friday but at various points and they're just like what you had. There are simple multiple choice questions. If you attend class, pay attention and just take these. You don't need to get 100% on them but they will cumulatively help and add up to something. The main by far, there's only one factor that really it's kind of a binary thing. One factor that distinguishes cumulative grades and quizzes and that's mainly whether you're here and take them. If you're not here, you get zero, so. And if you're here, then you get three, four, hopefully five points. We will endeavor to take these like we just did now, sort of 10, 15 minutes or whatever into the class so that you're not penalized if you need to rush from another class here. So we don't do them right at the beginning but we'll do them like we did here now. You've heard me say many times that the brain is a tube. Remember that the brain is filled with these fluid filled ventricles that are filled with cerebral spinal fluid and that goes down all the way and into is contiguous, contiguous with the central canal of the spinal cord. And so if you think about this, this is usually elaborated here so that this forebrain really proliferates and differentiates into the cerebellum, the different lobes of cerebral cortex and so forth. But when it started, it was just a tube, the most anterior part of which became hugely encephalized here. It's easiest to see in the spinal cord that still basically looks like a tube but that tube continues on up here. So it started as a tube. How did that happen? So your book goes through this and this is critical for you to know so make sure you study this and read the corresponding pages in the book. So here's a little schematic, I think initially made by Henry Lester, not by me certainly, that beautifully shows you how this happens. So we have the ectoderm, so you have a glastula that has three layers, ectoderm, mesoderm and endoderm. And at a point in development where the mesoderm here gets close to the ectoderm, it induces changes in the neural plate and this folds up and pinches off as a tube from which all of the central nervous system, brain and spinal cord then differentiates. So let's go through how this, oops, let's go through how this looks. So after gestulation, you have these layers approaching, they induce the neural plate here and these folds here, these red folds, so this part up here starts folding and then these folds approach one another and pinch off as a neural tube. And this then rolls into a tube. Here's the neural tube now and these red cells here are neural crest cells that migrate on down and differentiate into a variety of different things. So the number one thing you need to know is the earliest origin of the central and the peripheral nervous system all derives from this early event. All of the central nervous system, all of the neurons and glial in the brain and the spinal cord arise from the neural tube. All of the peripheral nervous system, all of the neurons and glial like Schwann cells in the peripheral nervous system come from these neural crest cells. Any questions about this basic scheme? So memorize this, know this, you'll be very familiar with this. From here on, there are then subsequent stages of differentiation that cause all the different kinds of cell types and different components of the CNS and the PNS. In particular, there's now an orientation here and you can set up gradients of molecules in the neural tube between dorsal parts up here, the neural tube and ventral parts close to the notochord. And so these regions of the neural tube differentiate into different kinds of cells and the neural crest cells that migrate out can adopt different kinds of fates and become differentiate into different cells. The neural crest cells can become many different kinds of cells. So I just told you that neural crest cells migrate and they become all of the peripheral nervous system but not only that, they also become muscle, they can differentiate it into bone, they can become many different kinds of things. So all of the peripheral nervous system comes from the neural crest cells but not all neural crest cells differentiate into the peripheral nervous system. So initially, these are stem cells of a sort that self renew, so they just make more of themselves, these early neural crest cells, they can differentiate then initially into quite a large number of things and progressively their fates gets more and more restricted. So if you have embryonic stem cells, those are totipotent cells which is why there's so much interest in them. Those are cells that can become any tissue in your body. These cells are multipotent, they can adopt many different fates but they can't become anything. They're not gonna become part of your liver for instance or your lung but they can become many different kinds of things. So this is a theme that we see over and over again that the kinds of fates that cells can adopt get progressively more restricted with time. Initially they can become many things but they become determined and then they differentiate and they get much more restricted into what it is that they can differentiate into. There are a bunch of molecules already outlined here and your book has many, many more here. And again, I know from several of you already talking to me that this quickly gets overwhelming for people that are not native English speakers. You know, try and do your best and make a little list from the readings. But again, the particular molecules will try and keep this to a minimum in terms of the particular molecules that we mentioned in lecture and those are the most important to know. These often have abbreviations and the general nomenclature here is if these are all capitals, if you see capital letters that means that they are proteins in humans. If the only the first letter is capitalized they're proteins in mice and if they're not italicized then they're the proteins. If they're italicized they are the genes that encode those proteins. So there are many of them. Now how could a cell be instructed to differentiate? There are many different ways that this could work and you could have a whole collect of different signaling molecules that induce a cell to adopt a certain fate to differentiate into something. So in the simplest case you would have signal molecule A that would bind to receptors in a competent cell that would then trigger some intracellular processes changes in the expression of genes that causes that cell to become the red the red type of cell for instance. Signally molecule B might cause us to become the blue cell and see the green cell. You can quickly see that that's not very efficient it's very precise but it's not very efficient. So typically it's more complicated than that and it's the particular combinations of cells that could induce a certain fate and it's also their concentration as we'll take a look at in just a minute. Nothing would happen of course if this gray cell doesn't have receptors or doesn't have a way of responding to these signaling molecules. So this gray cell needs to be competent that is it has to have a way of responding to the particular inducing factors in order for those to have any effect on it. This just says what I just said. The one thing that we'll take a look at in just a minute is a much more efficient way. So rather than having just a signaling molecule induce a certain fate and that's it and it's the particular identity of that signaling molecule it's concentration could instruct the cell to adopt a certain fate. So I think it's fairly obvious you could have some molecule being secreted and it would have a spatial gradient across tissue and then wherever a cell finds itself in relation to where that molecule is secreted if it's very close by or if it's farther away it will have a different concentration of that signaling molecule and if you have a mechanism that will result in a different fate for instance cell type A or cell type B as a consequence of the concentration of that molecule you can encode much more information here. Indeed, that's what you find that there are gradients of these so-called morphogens that are set up across tissue and so you can quickly see that if I wanted to for instance have a topographic differentiation across XY coordinates of a sheet of tissue rather than having specific molecules that instruct a cell that is specific to a particular XY coordinate all I do is I have a concentration gradient along X a concentration gradient along Y and that's sufficient and by the relative concentration they would encode the same information. Is that clear to everybody? Okay, so how does this actually work? Well, here are two molecules that you do need to know these bone morphogenetic proteins and this sonic hedgehog. These have strange names in many cases. These are two very important molecules that play roles at multiple times in development and indeed, and this is also a common theme that many of these molecules don't only serve one role and that's it. They might play several different roles and so you might encounter them over and over again in different parts of neural development at different points in time. But these ones are particularly important because they are responsible for one, for patterning along one key dimension of the neural tube which is dorsal ventral patterning. We'll talk about anterior, posterior patterning in just a minute. So the dorsal, remember the neural tube pinches off, here it is, it's pinched off. The neural crest cells which are not shown here are feeling migrating around to take on whatever fates they have including to become the peripheral nervous system and now these parts of the neural tube, I told you previously, all of the central nervous system, brain and spinal cord comes from that. How do they know what parts of the central nervous system to become? There's a very complicated series of many, many different decisions but one of the very earliest ones is whether or not they find themselves dorsally here, the roof or in the floor plate down here, ventrally and these different molecules are expressed as gradients, they're morphogens. Sonic hedgehog is down here, ventrally born morphogenetic proteins are up here dorsally and they give rise, they specify, they're one of the earliest molecules that's in a whole sequence that specifies what fate these neurons become. So for instance, in the spinal cord, these down here would become motor neurons, many of them, the ones up here would become more sensory related and so forth. And this just walks you through what I just said. So bone morphogenetic proteins induce dorsal fates, sonic hedgehog proteins induce ventral fates and as a result of this, you get some of the earliest patterning across the dorsal ventral axis of the central nervous system. Any questions about that? Yes. Yeah, how does it work there? Does it go wrong sometimes? Yeah. Well, so it's the relative concentrations of these molecules that play some role. They're secreted not just passively but actually quite actively. So there are a lot of different mechanisms in place that I guess ensure that it is fairly reliable. So for instance, sonic hedgehog protein is, it's very actively regulated by active mechanisms. It's concentration gradient. It's not just that it's put out and sort of passively diffuses and that's the end of it. There are many other molecules that are involved in making sure that its concentration gradient is stable. So the quick answer to your question is there are multiple mechanisms that are to some extent redundant that try to ensure that this is robust and it's not just subject to sort of random diffusion events. Yes. This goes back to the previous slide but where did the neural crest cells originate again? From the neural crest. From this thing that pinches off, so hang on, so here. So these red things up here. So when this closes, the tube closes, these cells that were in the neural crest, they delaminate from the neural crest and they migrate on down. It's a particular, it's a different form of migration than something that we'll see with respect to development of cortex a little later if we get to it. So this is kind of a free migration. So they migrate to many different places in the body. It's not completely free. So there are somites in here. There are molecules that we'll run into in a different role later called efferents that serve to steer these migrating neural crest cells to some extent. But you could think of them as sort of pouring down over into the whole body and then from there they further migrate and they progressively differentiate. But they originate from the neural crest, yeah. Okay, so we had these two molecules, important to remember, or dorsal fates. They're the bone morphogenetic proteins and provincial sonic hedgehog. And again, it's the details of, it's not in one step that these specify neural fates, but they're the earliest step and the earliest sequence in a cascade of steps that gives rise to neurons in the central nervous system brought in dorsal or ventral fates. And again, so for this to work, you can't just have these morphogens, but the cells that sense the morphogens have to have cell surface receptors. That is, they have to be competent to detect them. They then have to trigger intracellular processes and ultimately they have to dictate changes in gene transcription. I mean, there has to be a mechanism of course by which these morphogens can cause the cells to change and to differentiate into whatever it is that they're going to differentiate into. Now, what about the other axis? So the other big axis is front to back. So we've talked about dorsal ventral. The other is anterior, posterior. And again, if you look at the very early development here, the three vesicle stage and later you have five vesicles and you can see how an embryo is starting to develop and the brain would be up here at the most anterior part. But you can see how this is a tube, right? So again, remember the brain's a tube. It's filled with cerebral spinal fluid and different parts of this, these bulges get much, much bigger and elaborate into forming eyes, diencephalon, telencephalon, all the different stages of the forebrain. And then down here, this would be the brainstem and going down into the spinal cord. So the spinal cord looks more closely resembles in the adult, what it looked like in the embryo, whereas the forebrain really looks very, very different because it's undergone such tremendous growth. But so one question is, we talked about dorsal ventral patterning, but what causes the forebrain here to develop into cortex, cerebellum, all of these other things and other parts that are more caudal here to develop into the spinal cord. So that anterior, posterior patterning, again, relies on many different molecules. There are gradients set up on morphogens. There's again a sequence, a sort of hierarchical sequence of steps that initially there's some very coarse changes than this more and more fine grained compartments that are specified. And one part fairly early on, not the earliest on, but one early part of this are a set of genes called Hox genes, which stand for homeotic genes, that are remarkably conserved and whose expression has to do with specifying which of these anterior to posterior segments cells differentiate into. And this is what Ed Lewis and the company won the Nobel Prize for. There's 60 amino acid boxes in there that are very conserved in these Hox genes across many, many different species. The transcription factors, and so again, they bind to particular parts of DNA. They regulate some genes that are turned on, some genes that are turned off, as a consequence of which that cell will adopt a certain fate. And what's shown here is, so these are very conserved from sea urchins on up to humans. And the particular arrangement of these Hox genes along the DNA matches their expression pattern along the anterior-posterior axis of the body. And so you can see that in these color coatings here. Humans, tetrapods, you have four Hox gene clusters. And here's all the different Hox genes. And that's how they're expressed. Again, one big insight that came from these, as well as other genes, is how remarkably conserved they are. So the sequence of these genes in flies and humans and all these other organisms is very, very conserved. In fact, people have done experiments where you can take a Hox gene from a chicken and put it into a fly, and it works perfectly well, which is amazing. So they're very, very conserved across species. And these are responsible in part for anterior-posterior patterning. So now you know about two sets of morphogens, bone morphogenetic proteins, like hedgehog, that have to do with patterning dorsal ventral axis, and a set of Hox genes that have to do with patterning anterior-posteriorly. Again, these are not by any means the only ones, but these are particularly important ones. Here's an overall scheme that gives you a sense for all the different events going on that specify a neuron's identity. So there would be signaling molecules, gradients of morphogens. Again, these are actively regulated in response to your question. So it's not just that they're passive diffusion of morphogens, it's not that simple. So there are concentration gradients, but they're quite actively regulated. So there are a lot of machinery, a lot of regulatory machinery involved in here already. These have to then bind to receptors on cells that are competent. So these cells have to be in the right place and expressing the right kinds of receptors to respond to these molecules. There are, in addition to some of these more long distance mechanisms by morphogens that diffuse, you also have mechanisms that have to do with much more spatially close interactions. For instance, between cells that are touching one another. So you also have cell-cell interactions where one cell can put out a molecule that binds to a receptor on an adjacent cell, and that can also result in pattern formation. One big feature there is that there can be mechanisms that cause patterns through lateral inhibition. So if you have all these interacting, you have cell-cell mechanisms when cells are very close to one another, they interact, so it depends on who your neighbor is, and it depends on just where you're located in general across a big sheet of tissue and what morphogens you're exposed to. Both of those will be operating in parallel to give rise to pattern formation in the nervous system. And then ultimately how all this works mechanistically is that it has to change which genes are expressed and which genes are not expressed. So ultimately all of these molecules have to act on gene transcription in some way, and that then gives rise to the actual changes in the phenotype of the neuron. What kind of morphology it develops, what kind of neurotransmitter phenotype it has. And then of course once you have a neuron that changes, that neuron itself can play a role and could secrete morphogens, could put out ligands that bind to the receptors on other cells and will in turn influence neural development. Okay, does it sort of make sense in terms of the broad mechanism, but you can see the general scheme that there are not, it's not an overwhelming number of different mechanisms. They're fairly simple by themselves, but because of the way they interact and you put them together and they play out in time, you can generate very, very complex patterns like a human brain. Let's take a look at one particular example, which is neocortex. So we didn't mention this, although I think some mentioned it in the discussion section. Neocortex, this wrinkly outer layer of your brain that is gray matter, contains cell bodies, has six layers. Remember this is unique to mammals, birds and reptiles don't have it, flies don't have it. So six layers, they're numbered from the surface of the brain that you would see if you look at it, that's layer one, and as you go deeper and deeper towards the ventricle, down towards the middle of the brain, you would then have layer six. Exactly how these look would vary depending on where you are. Visual cortex looks different from motor cortex. How this develops is rather odd. It develops in an inside out fashion so that the neurons here in the deep layers reach these deep layers and differentiate into those neurons first, and later neurons have to migrate through them in order to populate the higher layers. That's shown how this works here. So you have close to the ventricle down here, you would have on the inner surface of the developing neural tube, glial cells, and these are particular type of glial cells so they can undergo symmetric cell division and make more of themselves. They can also undergo asymmetric cell division and they can make another radial glial cell and a cell that will become a neuron. These cells that will become neurons that are both happens down here, they have to migrate on up to stop in some layer and then they will differentiate into layer six cells, layer five cells, layer four cells, layer three cells, depending on where they end up. And they do this from the deepest layers first, these stop down here, layer six cells stop and form first, and layer five cells, then layer four cells. So the ones that have to get to the upper layers have to migrate through all of the lower ones that have already stopped in the lower layers. Oh, and this I think just says what I said here. And then the radial glial cells eventually later on will differentiate into glial, into astrocytes in your brain. But so that's one particular scheme of development in neocortex. I mentioned briefly, we had, I don't know till when, maybe four years ago, this word definitely was actually read probably because it wasn't clear. It was clear from studies by Fred Gage and others that new cells, new neurons were born in mice. What was not clear is if this happened in primates and in humans. But we now know for sure that it happens in humans, or at least with as much certainty as one can have in these things that happens in humans. In particular in these two regions, olfactory bulb and hippocampus, but probably to a lesser extent elsewhere as well. But in the hippocampus, this is a structure in the brain involved in learning and memory that we'll talk about in later lectures. There are new neurons born even in you right now. And it seems very likely that those neurons that are born in you indeed play a functional role in helping you with learning and memory. One study here, you can look up the details in this paper from Cell that tried to make this argument in humans took advantage of the fact that before the nuclear test ban treaty in 1963, there were above ground, lots of above ground radiation that was released into the atmosphere and into milk and cow meat and stuff that people would consume. And that could be used in the sort of experiment of nature to label cells and label when they divided. They had a very complex model but very roughly just quickly what's shown here is they took people from very young to 90 years old who donated their brains from 20 to 90 years old or 10 to 90 years old. People who died of some cause had agreed to donate their brains. They looked in the hippocampus and they looked for the radioactive labeling in relation to the age of this person. And when they knew back in some time ago, there was this release of radioactive isotopes into the food chain. And so based the bottom line is that they found a population of cells, these gray shaded ones in the hippocampus that continued to be born. These are renewing neurons and they estimated that about 700 neurons per day are born in your hippocampus. So we know that although there is much, much less, there are many, many fewer neurons being born in your brain than say neurons being turned over on an organ like the skin which is continuously renewing all the time. Even in adults, some new neurons are born in particular places. Any questions about this part? So now, so we've gone through this extremely rapidly but you have some vague sense that neurons differentiate. There are morphogens, cell contacts and neurons have differentiated are in the right place. What else do they need to do in order to form a functioning mature nervous system? Well, they need to form connections with one another and this is where the experiment by Roger Sperry comes in. There's an interesting history to this. People used to think that indeed neurons were guided by particular cues to find their targets but then this somehow fell out of fashion and people had this idea that outgrowth was just random and there were some magical kind of resonance that would help to form the specificity of connections that you see. No doubt there are multiple mechanisms but one in particular that has been studied in great detail is from Roger Sperry and here's what he did. He crushed the nerve in a goldfish, the optic nerve in a goldfish. So this is what connects the retina in the eye to the part in the brain that is concerned with vision. In the goldfish, remember fish don't have neocortex so this is not visual cortex but a structure called the optic tectum. Like in visual cortex though, it's topographic. So in this goldfish brain this tectum here has a topographic map of visual space and it has that because the axons that come from the eye make connections to particular locations in the optic tectum and so he found that if you crush, so here's the retina, if you crush the particular part of the optic nerve that corresponds to axons coming from this half of the retina and you let them grow back, you can see which part of the optic tectum they innervate and opposite for this one. So there's a regular topographic arrangement in how the retina innervates the optic tectum. And he postulated, the Keemo affinity hypothesis that this could be set up by some simple gradients like what morphogens, like what we saw in morphogens except that these gradients would just specify when axons grow in, they would attract the axons and axons would form synapses in those regions that had their complement of the particular gradient. So this was the basis of the hypothesis, he didn't know anything about the particular molecules and he also didn't know whether or not these molecules might be attracting or inhibiting neurons growing in. Let me just go straight to the experiment here that he did. So this is an experiment now in the frog. Your book goes through this in detail, be sure that you understand this. In a normal frog, we have the outside world here, here's the grass, here's a fly and this gives an inverted image on the retina just like a pinhole camera except that there's a lens here. Okay, but so you have, here's the fly down here and here's the grass, this is the projection onto the retina in the eye of the frog. And then these colors here, yellow, red, dark blue and turquoise, code regions of the retina that then project to corresponding parts of the optic tectum. So according to the chemo affinity hypothesis, these would have, there would be gradients here that would enable these particular axons to find their match in the optic tectum. So the turquoise goes to the turquoise, the blue goes to blue, the red goes to red, et cetera, et cetera, and then this would be the representation of the tectum that the frog uses for behavior. So this is the connectivity and the frog is able to find the fly. The experiment he did was to do two things. He severed the optic nerve and let it grow back. If that's all he did, the frog would be just fine. The axons would grow back from the optic nerve and there would still be these molecular labels, they would find the corresponding parts in the tectum like what's shown in the normal one in the left and the frog would be fine. But he did a second thing, he rotated the eye. So the molecular markers are still in the same place. These retinal ganglion cells in the retina up here have a marker so they want to find their corresponding target on the optic tectum except they've now been rotated. They grow out just as they would have in the intact eye. They find the same molecular marker but the topography has been flipped. As a consequence of that, the frog now thinks that the fly's actually in the opposite direction that it is and you can never learn to compensate and so it's toned out down here and misses the fly. So that's the fundamental experiment that Roger Sperry did. People have done something similar, not this drastic. This wouldn't work, by the way, in your brain because the axons from the optic nerve, unlike in frogs, wouldn't regenerate. So if you do this in a human, that won't work. People have done much milder experiments and in fact we had a previous TA in this course here, Tegan Wall, who told us that she walked around for three days with these inverting goggles. I don't know if any of you have tried this but you can just do this optically, of course. You can walk around and wear inverting goggles and it flips the world upside down. Unlike with the frog, humans actually learn to compensate for this. So after a while, things look normal. And if you take the glasses off, they flip back again and they're not normal. But there's a remarkable plasticity in the human brain. The frog never learns, so the frog always makes mistakes. I'm gonna run out of time here, so let me end with this. This is now the stage of what we know that Roger Sperry didn't know. There are actual molecules that are on the retinal ganglion cells and that are expressed in the tectum. So molecules in the receptors that implement the chemo affinity hypothesis. These are molecules called efferents. These same molecules that are also involved to help steer neural crest cells when they migrate. They have a different role here and they're receptors, efferent kinases. So these grow out and it's just like what Sperry envisioned with one wrinkle, which is that they actually repel one another, so that rather than being attractant they're repulsive. But other than that, the particular mechanism that Sperry had thought of is in fact implemented and we now know the molecules. Let me just show you one last video here. This is a gross cone, much sped up. Sorry, here we go again. So it just gives you an idea for how dynamic this is. So as an axon tries to find its target, as you can get a sense for in this video, the gross cone, which is this migrating axon that will eventually, once it's reached its target, differentiate it from a very specialized structure, synapse, samples its environment very actively. You can kind of tell it's going back and forth. There are different molecular cues it's picking up. It's integrating all of that in order to steer it towards some target. So a whole bunch of morphogens, other molecules close to cells that may influence its path as it tries to find its target. It just gives you a sense for the fact that this is very dynamic. There are many, many different molecules involved but it's a sequence of decisions that neurons have to make as they differentiate and decide what to become. And the same thing happens as decisions are made in the gross cone of an axon and it has to decide where to go. Once it's found its target, it then has to differentiate and form a particular synapse. And then once that is all set up, you have to have stabilization of that. There are effects that depend on activity between all of this. There is competition between neurons and the connections that they make with their targets. And you have all of these events, again, interacting in the right pattern and over time to result in a fully-formed nervous system. We're out of time with that. Any quick questions about the general scheme of this? I mean, look over the PDF of the lecture, read the readings in your book, which will give you a lot more detail. But I think as long as you have the basic principles here, you should be able to anchor what you read. Two kind of specific questions. You mentioned autism has shown up in development. I will point it out.