 Good morning. We are going to give the second lecture on developmental neuroscience today. The first one was lecture three. Ralph Adolphs did a marvelous job of providing you with an overview. Now that you have a little more experience thinking about molecules and neuroscience, and now that the excitement associated with the Nobel Prize has decreased a bit, I put together this second lecture on developmental neuroscience molecules and mechanisms. Let's go over then the first set of principles that Ralph Adolphs gave you during lecture three. Some of the principles actually derived from the introductory chapters of Kandel. Then he also showed you real human brains during discussion sections, which was as fascinating to me as it was to you, and then began some of the material in chapters 52 and 53. So now let's continue and emphasize these additional points from Kandel. The chapters are very nicely organized by major headings, which are declarative statements in blue. I expect you to be familiar with those declarative statements and to be able to find one example of each declarative statement with the exception that we don't talk about the differences between axons and dendrites. Also in later lectures, we're going to talk about this nifty topic that the development of visual perception requires visual experience, and we will not discuss sexual differentiation of the nervous system at all just because we don't have time. We will certainly talk about repairing the damaged brain in later lectures, and we will certainly talk about the aging brain mostly in the context of neurodegenerative diseases in later lectures. But for today, I'm going to give you a whirlwind tour of these four chapters in Kandel. Going back to this very interesting theme of the Hox genes and homeotic mutations, this word homeotic means a mutation that transforms one body part into another. Here, for instance, is a cross-section of the hind brain of a mouse. There are a total of actually eight segmental units. These are the bulges of the developing neural tube within the hind brain region during development. Now, in a wild-type mouse, in a wild-type mouse, one of the eight units, projecting outside the brain, in this case, the trigeminal nerve, and the facial nerve, the rhombomers are numbered one through eight. A Hox1B mutant mouse has an interruption in the Hox, sorry, B1 gene. Ralph told you, and I will emphasize today, that we have distinct ways of describing genes, which have really developed historically. If the gene is in a fly, it's all lowercase. If it's in a mouse, it is capitalized, the first letter, and if it's in a human, all letters are capitalized. That's the way things have grown up, and you can't change this. Well, I don't have the energy to change this. So, what does a Hox1B mutant mouse means? It means there's been an interruption in the HoxB1 gene, less HoxB1 protein. The result of that mutation, less HoxB1 protein, is that in a rhombomere where HoxB1 is strongly expressed in the wild-type, it is not strongly expressed in the mutant, and the result is that the default pattern for this rhombomere number four is that it has a pattern very similar to that of rhombomere number two. It sends its axons to and from the trigeminal nucleus. So, a homeotic mutation transforms one body part into another. That's the definition of a homeotic mutation. We should point out that homeotic is a general statement for phenotype, that is the phenomenon that results from a genotype. So, that phenotype is homeotic, the genotype is a mutation in the HoxB1 gene. For those of you who have not yet taken the HoxB1 gene by 120, by 122. Of course, the facial, the cranial nerves have been named for hundreds of years. Some, and there are various mnemonics for how to remember those cranial nerves. Most of these mnemonics are scatological, and you can read some of those scatological mnemonics here in Wikipedia, the olfactory, et cetera, down to the 12th nerve, which is, I believe, hyperglossal. No, 12th nerve. I'm not sure what it is. Anyway, you can read and you don't need to memorize this. It's for your own edification, but if you're going to go to medical school, you'll have to learn this sooner or later. So, a rhombomere is one of the segmental units. You could also call it a bulge. The reason that a rhombomere forms, the reason that the bulge forms is that the neurons in the center of the rhombomere proliferate a bit faster than the ones at the edges, so it bulges out. So, these are homeotic mutations and the genes have these names. And let's go on to analyze the homeotic gene a little more. A little bit of terminology, most homeotic mutations, that is, transformation of one body part to another, occur in genes which are transcription factors. A transcription factor is a gene whose gene product, the protein, controls the expression of another gene, typically by binding to DNA. So, all transcription factors are DNA binding proteins. All Hox genes contain a homeodomain, which is in the protein, and it's a type of DNA binding protein domain. And naturally, all homeodomains are found in transcription factors because transcription factors are by far the largest number of proteins that bind to DNA, although some of the chromatin proteins do as well. So, Hox genes have DNA sequences in them called homeoboxes which encode homeodomains. So, we have Hox genes, we have homeotic mutations, homeoboxes and homeodomains, and evidently, the H in Hox gene comes from homeotic mutation. Kandel and many other books show you details of a slide that Ralph Adolfs has always shown you on regionalization of the nervous system, in drosophila, fruit flies, and in mammals. So, there are Hox genes, depending on how you spell them, both among invertebrates and among all known vertebrates. Among vertebrates, there are actually four clusters of Hox genes in mice. They are on different chromosomes. Presumably, chromosomes arose by duplicating each other many times, often in vertebrate genes. So, this is like a figure in Kandel. The Hox genes then control the identity of segments, rhombomeres in the hind brain. Interestingly enough, this in the drosophila, the antedipedia complex is among the Hox genes. There's also the bithorax complex discovered here at Caltech by Ed Lewis. Okay, now, here's a tough one for today. Which body part do you think is duplicated in bithorax? Nobody wants to hazard that one. Yes, it's the thorax. Okay. So, the mutation in the bithorax gene produces two copies of the thorax. And of course, the Hox genes encode transcription factors which bind to DNA and have homeoboxes that participate in the binding to DNA. And Hox gene expression along the anterior posterior axis determines cell fate. And furthermore, Hox gene expression along the body is collinear with the Hox gene order along the chromosome. So, at the three prime end of this chromosome, we have the most anterior genes, and at the five prime end of the DNA, we have the most posterior genes. So, just to put the Hox genes in context, the Hox gene expression patterns produce rhombomere borders. There are other gene transcription factors that control themselves control the expression of Hox genes. And then there are other target genes for the Hox proteins that have rhombomere specific expression patterns. Those include the f-kinases and the efferens. And you may remember that we have discussed the f-kinases and the efferens in the context of axonal pathfinding, more of which we will discuss later today. Okay. So, question for class discussion. I've never asked this before, and so the answer is open. What is the selective advantage of having gene clusters that confer selectivity in neural development? I have not looked up the answer to this question, so let's have some ideas, and I don't even know if the answer exists. Any ideas? Yes, Matthew. So, Matthew has suggested that if the genes are all transferred at the same time, presumably from one organism, which is arising as a species from another, having them all there prevents one from being deleted. Okay. Okay. I like that very much. Any embroidery on that? Any further ideas? Well, wouldn't you say that having just one Hox gene does no good at all, or even two or three? So, we need a number of genes to make a pattern. So, we need a number of genes to make a pattern, and the most robust way for a nature pattern, for instance, to make the eight rhombomeres, we need six or eight genes, and so the most robust way for nature to do that is by transferring them all at once. So, I think you were saying part of the idea, and I simply completed the idea. Any other ideas? Okay. So, let's take that one and use it next year for the course, or write a paper about it. Good. But, we still have this question, number two, which is the selective advantage of co-linearity in the Hox genes. What is the advantage that arranging them from 3.5 that they have the same order from 3 prime to 5 prime as the body segments? I want you to realize that I don't know the answer. So, give me the answer. Come on. You guys could all warm toast with your brains. Give me the answer. All right. Well, if you think of it, email me. Okay. The next 10 slides or so, we're going to discuss growth cones and the molecules that bind them. So, this is somewhere between Ralph Adolf's question. Okay. We got to the vicinity of the target. How do we refine our destination at the target, and how do we make the correct synapses? And so, let's have some terminology here. The leading edge of a growth cone, of a growing axon is called the growth cone. And you've seen some wonderful pictures in Candel. It spreads out and gets very thin. And the growth cone has microtubules and extending into its base. I'll show you a picture of that. But, it's Lamellae in Philadelphia. The end of the growth cone have only actin filaments, which are a lot daintier than microtubules. So, Lamellae is a double parallel membrane. So, it's a thin extension that doesn't have much. And the filipodia is filamentous little feet. That's what filipodia means. Okay. So, when a growth cone is moving rapidly along a pathway, it has a narrow profile. But, when it makes decisions at a junction, it radiates filipodia and becomes wider and more complex. And the adhesion molecules on the filipodia contact the substrate. Now, in an experiment in vitro, the substrate would be a culture dish. In life, it's the surface of other cells. And that helps to nucleate actin polymerization. Then microtubules invade the region with a newly stable actin cytoskeleton. Moves the axonal shaft forward. There are, and I'm going to show you a diagram of this in a moment. So, mini-cell surface molecules function in growth cone guidance. And many are conserved, which is really cool, between invertebrates and vertebrates and among vertebrates and among invertebrates. Now, the context, and Ralph has emphasized this as well, the context of the Q given by a given molecule can either be guidance or guidance receptor. And the result of this contact can be a version or attraction, depending on the context, very much the way a transmitter molecule can be either excitatory or inhibitory, depending on its receptor. Looking greater detail, I've combined several versions of candel to give you what I think is the clearest view. So here is a growth cone. And the theme here is that there are several types of motors. We have a few microtubules invading the growth cone. And if we look in greater detail, here is one of the microtubules. And the microtubule through the membrane of the growth cone has adapters that allow it to contact a substrate, either a culture dish or another cell. The microtubules, as you know, from bite by nest can extend themselves. And this makes the growth cone a little longer. And then desiccalfusion adds membrane to the leading edge of the filipodium. After a while the cytoplasm in the trailing edge of the filipodium contracts. And so we have a new growth cone, new axonal growth, which has moved the, which has moved the growth cone forward. And this is helped by actin polymerization. So we have a couple of engines, motors. We have microtubules being polymerized and deep polymerized, mostly polymerized here. We have actin forming and being dissolved and moving forward. So our growth cone is a very complex machine. And it is constantly sampling the environment and making decisions about where to turn or where to avoid. So among the families involved here are two of the most famous. There are generally called cell adhesion molecules, sometimes just cams or cell adhesion molecules. They have immunoglobulin domains or they have fibronectin repeats. They're all single past transmembrane proteins. So what we are looking here at here is cell number one and cell number two. And here the cell adhesion molecules are extending through the membrane. Some of the cell adhesion molecules called canterans mediate calcium dependent adhesion. They also signal in various ways inside the cell and are attached in various ways to the actin cytoskeleton. So we shouldn't see a cell adhesion molecule is simply anchored in the membrane and allowing the cell to float around it. Actually, these molecules all get anchored to the cytoskeleton and in some places can influence the formation or breakdown of the cytoskeleton. For instance, here's a cyclic AMP signaling system. This is a guanylate kinase signaling system. So in terms of growth cone guidance, growth cones can respond to attractive guidance cues. That's green by increased outgrowth or by turning toward the cue source. We've seen that. They mostly respond to repellent guidance cues by collapsing. The guidance cues can be diffusable and we haven't seen those yet. There are examples in cantile and cell surface and we've seen two examples of those. So they're either proteins. Again, they are single pass transmembrane receptors or small molecules. And I will give you an example of that later in the talk as well. So growth cones can be guided by these various phenomena and they are propelled then by their motors. Let's turn to this very interesting topic that Ralph also brought up, which is chemo affinity. You remember that he discussed Roger Sperry's chemo affinity hypothesis, which was derived from cutting optic nerves to the frog and watching the behavior of the frog. Some of these experiments were also done in Goldfish in the optic tectum, which is the lower vertebrates equivalent of the superior colliculus. And so Sperry said, whenever fibers were disconnected and transplanted or just scrambled, regrowth always led to orderly functional recovery. And so Sperry said, it seemed a necessary conclusion from these results that the cells and fibers of the brain must carry a kind of individual identification tag. And Ralph's heart of the story are the efferents in the eff kinases, which are basically not chemo affinity, but chemo repulsion. And so this is a modification to the Sperry hypothesis in which a gradient of molecules repel each other. And there are various experiments that Ralph told you about, about transplanting one region of the tectum to another and finding that the usual partner avoids that region. And so all of these are phenomena associated with efferents, which are the peptid ligands, and the eff kinases, which are the tyrosine kinase receptors. Well, chemo affinity molecules, not repulsive molecules, but chemo affinity molecules, these would in principle allow a much more selective pairing between cells. And so for this, we turn to the fly visual system run by Kaisen's lab. In fact, these experiments are just going on. They will be published soon. Kaisen is a Caltech professor, and he has collaborated with non-Caltech people. And he has solved the chemo affinity problem, found a group of genes, which were previously called the defective proboscis extension response, DPR. Well, you had to call them something. And you know from looking at the spelling of the gene that this is an invertebrate. And in fact, this is Drosophila. So there are 21 DPR genes, and they interact with nine DPR interaction proteins. And since they're called DPR interaction proteins, they are labeled DIPs or DIPs. You know, the problem is that you have to label, you have to name genes something. And so you sometimes just name them with what they bind to. There are addition, in addition, the DPR and the DIP complex interacts with a leucine rich repeat, CDIP. Okay, so search for the chemo molecules in the fly visual system involves the retina of the fly, the laminar, which is a first stage below it, the medulla in the inner medulla, the molecules that Zin and his collaborators have identified govern interactions among the cells, finding their postsynaptic partners in several of these layers. The interactions between these molecules. So this is a DPR molecule. And here is its binding partner, DIP. This is a particular DPR molecule, cleverly enough called DPR6, and a particular DIP molecule. They make a typical, a pattern that is similar in the nervous system and in other organs throughout the body among vertebrates and invertebrates. For instance, there are C. elegans synapses that have the tips of these two molecules interacting with each other. The function of the gene family in C. elegans has not been described the way it has in flies, but there certainly is a similar molecular interaction. And even in the immune system, molecules that neutrophils that contact other cells have a similar interaction. So again, the DPR6 molecule in blue, the DIP alpha molecule in red, the two of them interacting with each other. So this is the chemo affinity part right here. This is the specific labels on one cell contacting another cell. The interface is interestingly enough almost all hydrophobic. The molecules have to fit together. They have various amino acids in one part and another part that allow them to fit together. And so this then becomes the chemo affinity mechanism at least among fruit flies. Moving on, so that's a rather nice story about Sperry's hypothesis and it really is quite a breakthrough. And those slides were provided by Kaizen. In addition, there are other molecules involved in growth cones. There are the semaphorens and the netrons. Again, they can mediate either attraction or repulsion. So this figure, which is like a figure from Kandel, has a nice growth cone extending its filipods, sensing the environment in here. As a result of a semaphoren interaction, this growth cone has retracted and it's not going to go in this direction anymore. Again, the semaphorens and the net here, the semaphorens are either attached to one of the cells by a single past transmembrane protein or it can be cleaved from that single past membrane protein and diffuse highly locally and interact with the receptors, their heteromultimers, between a molecule called a neuropillan and a molecule called a plexon. And so the semaphorens interact with the receptors in ways that are not yet understood but do cause growth cone attraction or repulsion, either effluorescence of the growth cone or repulsion involves actin depolymerization, the growth cone collapses. Attraction may involve the formation of new actin filaments as the growth cone extends, and there are a number of small molecule GTPases that seem to be involved in making the actin depolymerize or grow. They are called row GTPases and we're going to talk about other G proteins but the small molecule GTPases are different. So simplified view would be that one of these small molecule GTPases row favors collapse and rack favors growth but that may be too simple. So as usual we have complex signal transduction systems, you won't be responsible for knowing these signal transduction systems but as I write the development question for the midterm, I'd like you to be able to find an example of one or more of these. So in addition, let's go to the other topic of the nerve muscle synapse formation. You may remember that in previous lectures we said that many basic principles of chemical transmission were discovered at the nerve muscle synapse and I might add that many basic principles of synaptic development were also discovered at the nerve muscle synapse. So here we have the motor neuron making the end plate or the nerve muscle synapse and this is all a figure that you've seen before from Kandel. We've also seen from Kandel this highly organized figure of the presynaptic terminal, the glial cell in the postsynaptic folds and so now one asks the question how does all this arrive and how does all this arise and the answer is as you several steps. The growth cone reaches the muscle fiber, we've discussed how that occurs, it reforms to make a semi-flat contact with the muscle fiber. A hero of the story very much is the basal lamina that's different from the lamina we discussed in the Drosophila retina which is actually a group of cells. This is a kind of extracellular matrix and it begins to form. Receptors for acetylcholine then cluster under the forming synapse, the nascent synapse. The Schwann cell comes in and wraps the synapse. In the beginning and this is going to be a common theme. Several axons innervate the muscle but later all but one withdraws. So at the mature synapse we have the three cells, the motor neuron, the muscle cell, Schwann cell. Between these cells we have the basal lamina, the postsynaptic membrane is infolded, the acetylcholine receptors are at the top of the folds, sodium channels at the bases and acetylcholine esterase is in the basal lamina. We've said all of this before, acetylcholine esterase. I haven't told you that the acetylcholine channels are at the bases of the folds but they are. So how does all this occur? Acetylcholine receptors which are normally not very well localized before the nerve fiber arrives become localized underneath the nerve fiber. There are several mechanisms. First of all the nerve secretes signals that cluster pre-existing receptors at the synapse and I'll tell you about that in subsequent slides. Also the nucleus as you know a muscle fiber is a fusion of many cells with many nuclei. So this is a cell that has many nuclei and there are signals that tell the nuclei near the synapse to make additional acetylcholine receptors and also receptor expression gets turned off at the other nuclei. All of this is amplified by action potentials in the muscle fiber which presumably occur only if the synapse is working and the presumption in almost every case for activity dependent signaling in the nervous system is calcium influxes either directly toward by the receptor itself or produced by action potentials which activate calcium channels. So here we have the acetylcholine receptor transcription increases but away from the synapse it decreases. Excuse me that phone call was from the person who installed an alarm in our house yesterday so I should tell him. I mentioned the basal lamina of connective tissue collagen like tissue extracellularly. A very interesting set of experiments that go back several decades is that if you cut the nerve destroy the muscle the muscle fiber that's pretty interesting and it turns out that some of the attractive cues occur because while you have destroyed the muscle fiber but you have kept the basal lamina intact and so it is actually the basal lamina a particular laminin variant that is enriched in the synaptic extracellular matrix that attracts the nerve back. How this all occurs is not entirely clear. One of the signals secreted from the nerve is called agrin, eating protein, activates a postsynaptic tyrosine kinase called musk muscle specific kinase which in turn activates another molecule called rapsin and the rapsin actually physically clusters the receptors in response to the nerve. So if we have a cultured muscle fiber with no agrin it has uniform acetylcholine receptor uniform acetylcholine receptor density but if we add agrin those receptors cluster. So rapsin is an abbreviation for receptor associated protein at synapses. Again you have to name these something. There are a few molecules then necessary for clustering acetylcholine receptors and really these clustered acetylcholine receptors and the development of the nerve muscle synapse is actually a lesson for the development of all synapses. So in a wild type mouse you have these wonderful plaques in plates that are nicely compact along the nerve. In an agrin knockout even though the main function of agrin is said to be postsynaptic those exons spread out looking for other synapses and does anybody here want an explanation of what a knockout mouse is? Okay a knockout mouse Ellen is a mouse whose genome has been modified to interrupt a particular gene so that gene has been knocked out. The mutants that we discussed earlier in this lecture such as the homeob the home the hox gene mutants are naturally occurring knockouts typically. Sometimes those knockouts are helped along by irradiating the parents so that there are more mutations caused but in any case those are that's the way that one does forward genetics by knocking out genes. Now in a mouse it's a little more complicated you need to do some fancy molecular biology to knock out a specific gene but you can do it and so if you want to get rid of the action of agrin best way to do it is to knock out the agrin gene this is done with base pairing and other ways to manipulate DNA. A musk knockout gene has a very similar phenotype and the rapson knockout has a bit more compact exons but still nothing like the wild type and so we need the function of all three of these genes to have a nice well organized nerve muscle synapse. Any questions again the phenotype I've just showed you differs from the genotype which is the knockout of the gene so the phenotype is the phenomenon associated with the genotype the knockout of the gene it is possible as many of you know to produce much more subtle changes in genes than simply eliminating them entirely you can make one more active you can change its function etc but for today we're just talking about eliminating the gene yes sir so the question is what is the difference between knockdown and knockout a knockdown experiment is typically done in the animal after birth and it's a partial knockout the way it's done is typically by injecting RNA that base pairs with the RNA made by the animal this causes a set of enzymes to come and destroy the endogenous RNA so the knockdown experiment is easier to do because you don't have to work with embryonic stem cells and with lots of mouse mutations but it takes a lot of experimentation to get the right knockdown simple hairpin RNA so that it does a good job of knocking down a knockdown experiment is rarely complete but usually informative enough to give a phenotype any other questions called acetylcholine receptor inducing activity which actually makes more acetylcholine receptors it was first discovered and called aria acetylcholine receptor inducing activity discovered at the nerve muscle synapse secreted by the nerve it is a member of an extremely interesting family called the nuregulans nuregulans is a transmembrane protein which is proteolyzed to release a growth factor and it is really the founding member of the nuregulans family because this family has been implicated strongly in schizophrenia we'll get to that later on when i talk about schizophrenia egf is epidermal growth factor now we come to the topic of synapse elimination during a baby's first two years of life those 10 to the 11th synapses uh those 10 to the 11th neurons are getting an average of a thousand um synapses a piece 10 to the 14th divided by the number of seconds in two years come you get to the conclusion that you get about a million synapses forming per second actually more than that form because many are eliminated and so as we said a few minutes ago uh growth cones typically get directions not to innovate specific muscle fibers but only a group of fibers so this is called polysynaptic innovation it typically just disappears soon after birth uh and it is a competitive process the neuron that's best able to drive the muscle wins out to become the sole neuron so each neuron then commits its resources to a group of fibers and it abandons the others it had initially innovated and you can bias this with various chemical and activities so obviously this is an extremely interesting phenomenon and one of the hypotheses one of the mechanisms of synapse elimination is called the limiting neurotrophic factor hypothesis here not at the muscle but in a peripheral ganglion when gets the same kind of thing more neurons appear than are required to innovate the target cells and there is a hypothesis that the target cell has a limited supply of a neurotrophic factor that is a factor that allows neurons to form and as ralph told you uh during lecture three one of those neurotrophic factors is nerve growth factor NGF uh the neurotrophin uh then acts uh there are several examples of neurotrophin NGF is one of them another one is called appropriately enough neurotrophin number three another one is called BDNF that's mostly in the brain another one's called glial derived neurotrophic factor uh they distinct sub types of neurons then depend on different neurotrophic factors there are four major ones actually there are many major ones and they are listed here fiberglass fiberglass growth factor glial derived neurotrophic factor number three uh and um NGF ralph showed you this slide which is the tyrosine receptor kinase class of receptors for neurotrophins we've named a couple of these neurotrophins here uh they are uh four uh they are there are four of them recognized by three TRKs tyrosine receptor sub types uh they're actually less specific that specific than one would like there is some crosstalk between them so neurotrophins regulate neuronal survival they regulate proliferation of precursors differentiation growth branching transmitter synthesis synaptic efficacy and rearrangement and uh BDNF is one that is particularly interesting to neuroscientists these days because it may in fact be a pretty good antidepressant but when there is by acting on track B tyrosine receptor kinase B but also too much BDNF causes epilepsy so again the context in which a growth factor a neurotrophin acts is important and the time that it acts is also important so BDNF brain derived neurotrophic so trans signaling pathways are rather complex uh they all involve tyrosine phosphorylations that's why they're called TRK here's the plasma membrane ultimately many uh neurotrophin pathways lead to the activation of transcription factors by phosphorylating them into the activation of target genes that's pretty interesting because in addition other growth factors instruct cells to continue in the cell cycle so we have growth factors that tell a cell to continue in the growth in the cell cycle by activating genes associated with proliferation and with division of neurons and so this can turn in and of course this is the basis for which some uh TRK genes are all oncogenes because they continue to tell a cell to participate in the cell cycle and cells to proliferate even when they should not so some growth factor receptors are indeed oncogenes they can cause cancer we also should mention notch signaling and I'll briefly say that notch and delta signaling occur because the notch and delta uh partners in a cell uh they are polarized but all cells in a region may express both notch and delta however the ones in the middle of the region seem to express slightly more delta and slightly less notch and so a signaling imbalance occurs and the neat thing about notch and delta is that they reinforce each other so that uh notch decreases the activation of delta and as a result one gets a sharpening of a boundary uh so one can now sharpen a boundary in another way we have the f in the efferents the f the efferents in the efferent kinases we have this wonderful set of cell adhesion molecules and then we have boundaries which are sharpened by lateral inhibition uh and michael elowitz at caltech with his mastery of transcription factors mastery of gene activation and of differences between cells is making nice contributions to this pathway even though he does not consider himself a neuroscientist um two more points first there's a famous example discovered by paul paterson here at caltech in which you can actually change the phenotype of a cell that is whether it's cholinergic or adrenergic depending on the target the target secretes neurotrophic factors for instance the most famous example is the sweat gland and the sweat gland is at first during development is adrenergic the sorry the neuron that innervates it is adrenergic but based on the neurotrophic factors that it releases the sweat gland induces the cell to become cholinergic to release acetylcholine and the result is that acetylcholine becomes the major transmitter activating the sweat glands so this switch takes place after birth after the cell has reached the neuron has reached the target and has innovated the target final complication in developmental neuroscience and i promise that this is the final complication at least for today is that cells can actually change whether they are excited whether they are excited or inhibited by GABA we usually think of GABA as an inhibitory transmitter that depends on having a high internal chloride concentration the internal chloride concentration is governed by pumps the cell starts out in many cases with a low chloride concentration sorry just the reverse it starts out with a high internal chloride concentration that means that the chloride in the cell intracellular is equal to the extracellular chloride as a result the Nernst potential for chloride is near zero so that early in life in many cases GABA actually produces depolarizations and activate cells and then when the full complement of pumps for chloride and bicarbonate get established in the cell the intracellular concentration of chloride drops the chloride gets pumped out by transporters as a result we have more chloride outside than inside the same is true for sodium but the sign of chloride differs from that of sodium as a result the Nernst potential for chloride is actually quite negative and so chloride becomes a GABA which activates chloride channels becomes a hyperpolarizing or inhibitory transmitter and the result is that we have a switch from excitatory actions of GABA to inhibitory actions of GABA this is thought to be important in whether early on in life a cell gets activity dependent signals that help it to develop normally so again these are all examples that I'd like you to look at in candel but you won't be responsible for regurgitating many examples for each point and I will have my office hours as usual outside the red door today see you Wednesday