The molecular logic of synapse formation in the brain
My, privilege to introduce a, very. Special presenter. Today at the Wednesday afternoon, lecture which, is also one of the NIH directors, lectures. And as, the NIH director I get the treat of being able to introduce the speaker and. Today we're, fortunate to have a Tom suit off with us who's. Going to speak about the molecular logic, of neural circuits. Born. In gutting in, he. Then pursued, an MD degree at the University, of göttingen, Medical, School and, not, long after that came to the United States where. He carried, out a postdoctoral, fellowship, at, a. UT Southwestern, Medical Center, in the laboratory, of none other than brown and goldstein not a bad place to sort of get your momentum going. Somewhere. In there he, moved. In a direction different. Than they did and so, he's not going to talk to you about cholesterol, I don't think. He's. Gonna talk about neuroscience, which is where he, has made a remarkable, series. Of advances, over. Those years at, UT Southwestern. Arising. To become chair of the department, of neuroscience, and then. Since, 2008. As, the Avram. Goldstein professor, Department, of molecular and cellular physiology. At. Stanford, and many. Of you know his name particularly. Because of the remarkable. Achievements, and awards including. The Lasker, DeBakey. Medical basic science award in, 2013. And. Then right, after that that, same year the Nobel Prize in Physiology. Or medicine shared. With James Rothman. And Randy Schekman for. His work on the machinery involved, in vesicle, traffic, but, in this case the, vesicles, he's been particularly, involved in are the ones I'm sure he's going to talk about today. Which involve, synapses. His. Work has, continually. Added, to our understanding of how synapses, form. And function and the role of various molecules and making that happen and increasingly. Also shedding, light on critical. Issues. About autism and schizophrenia and, how this may all be tied together and a better understanding, of the. Synapse and just.
As We are at the moment in very exciting, time of beginning to unravel some, of the mysteries of the brain and certainly, we at NIH deeply. Engaged in this as, part of the brain initiative, the. Work that he is going to talk about is, fundamental. To putting together those, kinds, of insights, understanding. How these, critical. Properties function. Both in health and disease so we're fortunate indeed, to have a Nobel laureate here, this afternoon and, I please. Would ask you all now to welcome professor. Thomas, suit off. Good. Afternoon everybody, it's a tool pleasure. And honor for me to be here, thank, you very much for your kind remarks. Sometimes. When, I hear. The. Story of my life repeated. It, makes me almost feel old seems. Like everything was yesterday. What. I'm going to talk about today, as. Indicated. By the title, regards, the fundamental, question, of how nerve. Cells are connected into, networks. By, synapses. But before, I start let. Me list, my. Disclosures. You can see them here, none. Of the. Stuff. I'm going to talk about has anything to do with. What. I do as, an adviser either to academic, institutions or, to various companies. So. What I'm going to talk about today, is a project or. Series, of projects. That. Is really, at the beginning as, far. As I can see, something, where we are trying to make. The initial. Insights. Into how it works and I wanted to. Explain. This a little bit to you what the actual problem, is. Obviously. This is a human brain a dead human brain and. It always amazes, me. To, consider, how, large, the, human brain is. There's. Trillions. And trillions of, neurons that, are connected, by. Billions. Of synapses. Creating. Vast. Overlapping. Inter, digitated. Networks. That, are communicating. With each other when, you compare, that to, the, human genome. Whose. Sequencing. Francis. Contributed. So much to. It. Is actually, quite amazing. How much, bigger. The, brain is that the genome. Initially. When we. All. Saw, the genome, sequence, we thought my god we're going to understand the human genome, they. Didn't last very long, we. Now. Know. That understanding, the human genome is much more than just having the sequence. Although. It was an absolutely, essential to their cause it, similarly. For the brain. It. Will be very difficult to understand, how it works just, from mapping. Let's say the neurons and their connections, although it will be a. Prerequisite. It. Is clear, when. You look at these numbers however, that. You won't be able to, understand. The brain as a simple. Linear function, of, the genes as, everything.
In Biology this, has to be an indirect, nonlinear, relationship. Understanding. The brain is. Arguably. The most. Challenging. Though. We have now in biomedical, research but. We. In, the neuroscience, community have, very different opinions, about how. Best, to approach such, an understanding. Despite. These opinions I think. Everything, converges. On. To, the concept of neural circuits. What. Is the neural circuit, what, you see here is a very. Simplified. View of a neural circuit. Illustrating. There. A circuit, is formed by a new on, that are connected, by synapses. Synapses. Are junctions. Intercellular. Junctions were, specialized. For. The transfer, of information. And. Also. Maybe more importantly, for. The processing. The computation of that information synapses. Are the fundamental. Computational. Unit of the brain. In. Doing, so synapses. Are not the same. But. Depend, very much on. The. Precise. Nature. Of the pre and postsynaptic, neuron. And on the history, of their, connection. All. Synapses. Operate, by the same, fundamental. Principle. But. The differences between synapses, are, as a center, for understanding. How circuits process. Information, as is. The. Wiring, diagram, that, depicts, where, the synapses, are. We. Need to gain a fundamental. Understanding. Of. How synapses. Are. Formed. And. Specified. Let. Me explain this a little bit better what. You see here is, a schematic view, of how, the brain. Circuits. Develop. In. Necessarily, during development. There's neurogenesis. Axons. And dendrites develop. Most. Of that occurs only during. Development. Synapse. Formation on, the other hand occurs. Throughout life I. Schematically. Divides. In a summation here in the initial. Formation. Of a contact, and, into. The process, that organises, the canonical elements. Such as the release machinery. Of. Synapses. As well as, the. Processes. That, convey. On to synapses, confer, on to synapses, specific, properties, such, as the ability to undergo, LTP. Synapse. Formation, is, in a constant. Continuous. Equilibrium. With synapse elimination. And. Together. They. Determine. How. Information, is, possessed, by. Specific. Circuits, in the, brain. We. Hypothesize. That there must be specific. Molecules. That there must be a molecular logic. That, guides this, overall, process. But. We don't really understand, that logic at this point I. Believe. That, synapse formation, broadly. Defined, here both as the actual formation of the contact, as well, as the process, that, endow synapses. With specific, properties. That. Synapse formation like, that is. A classical. Cell biological. Problem. That. It basically is guided, by a. Signaling. Pathways. That. Are activated. By cell, surface receptors, and then. Result. In the organization. Of specific, structures. Because. Synapses, are intercellular junctions. Such. Cell. Signaling, processes. Most likely. Are. Mediated. By cell adhesion molecules. Which we all know are, signaling. Molecules, that connect. Cells. Edit junction with each other as. It. Turns out we have actually, at this point quite, a few, such candidate. Cells, adhesion. Molecules, at synapses. That. Are illustrated, schematically. Again in this slide here. Among. Others they are the new axons, and, allow. Time for, a thousand fossilized receptors. As shown. Here which. Are presynaptic. In. A kind of hub molecules, that interact with a lot of postsynaptic. Molecules. Their, ton of postsynaptic. Molecules. What. Is obvious, is does that. Already. At this point there's. A wealth of molecules, but. What. We don't know, is which ones of these are actually, going to be central. Which. One's of these are bystanders which. Ones of these do what, and. How. Do. These molecules in. Their. Various, functions. Established. Synapses, how do they decide, what, synapse, properties, should. Be there. In. My talk today I will focus, on. One, particular family, of these molecules, because, I think they, illustrate best. What. I call the, molecular, logic, that determines. Synapse, properties, and also. Because. These molecules, the new axons are arguably. The, best, understood. Molecules. At. The synapse as, cell. Adhesion molecules. So. What. I'm going to talk about is new actions, as anchors. That. Organize. Transana. Pick signaling platforms, which in turn control, synapses, and. To start off with, let, me introduce new, axons, so you have a feeling, for what I'm talking about, new.
Accents, Are Sarkisian. Molecules, that have the two classical, structure, of cell surface proteins, as illustrated, here they. Come in two flavors alpha. New actions and banana axons transcribed. By three genes, in. Addition. One, of the three genes makes a third form gamma. As shown, here at the bottom, these. Three genes have. Independent. Promoters, for these isoforms. So. That there's six, turns. Upon your actions in total new. Actions are localized, to synapses, as I already mentioned this, is illustrated, here in this, slide, which. Shows co-localization. On top. Of an election one with a presynaptic. Marker, and the jacent localization. On the bottom with a postsynaptic, marker, demonstrating. That. These are synaptic. Proteins, at least in part. Recent. Studies. In. Collaboration, with Xiao Wei drugs have revealed. That. Inner synapse, as illustrated. Here by super-resolution. Microscopy. That. Innocent. Herbs new actions, are not. Uniformly. Present. In the synapse but, that they are organized, in a subdomain. Each. Of these green puncta, is. The new action the, red labeling. As, shown. Here is, a postsynaptic marker, called Homer, what. You can see is that the Homer molecule. Is distributed. Across the postsynaptic density, but. The knee reaction is present, as a cluster. This. Cluster, shown, here in more greater detail, contains. Approximately four. To five no X molecules, that. Are then put together in, a micro domain. The. Size of this micro domain as. Analyzed. Here both in, culture. Neurons and in, slices. Occupies. About 10 to 20 percent of the total synapse. And. This. Michel. Domain is positioned, within the synapse in a manner that changes. As a function of the synapse maturation. Which can. You measure as a radial displacement. So. In vitro from div, 18 to 26. It moves outwards, from the center. To. The periphery as, it. Doesn't vivo from P 14 to Pete 28, in hippocampal. Sections. So. These experiments. Revealed that relaxants. Within. Synapse, is part of a subdomain. Suggesting. Or rather confirming. Others people, studies such as Tomblin peds that. Synapses, are not homogeneous, junctions. But instead are. Organized, into. Specific. Segments. Importantly. No. Accidents, have been linked, by, mutations. To. Neuropsychiatric. Disorders. In particular autism. And schizophrenia. What. Is interesting here is that this. Occurs primarily for, the new acts in one gene. Sunil. X in one gene is a huge gene, to.
Mega Bases approximately. And. All, of these mutations are. Technically. Speaking CNVs. Fundamentally. Deletions. The. Distribution of these deletions is illustrated, here in the slide showing, that, they are not uniformly. Present. In terms of the extent but that they vary in extent, over, a great range this. Apparently, is a very dynamic, locus. And. In. These hundreds of mutations, that have been observed the. Extent, of the relation, is quite different, what. Is most important, here however is that, the. Deletions, do not cause a similar. Clinical. Presentation. And different patients, but. I, associated. With a large number of different, presentations. They. Associate with autism, schizophrenia, Tourette. Syndrome, very, importantly, intellectual. Disability, in. A manner that we don't understand. We. Wondered, in fact, whether. These deletions. Would actually, have a functional. Impact, on neurons. And synapses since, they are always hitters, I guess. The. Diversity, of phenotypes suggests there may be genetic, background, effects, all. That, this. May be a rather minor, constituent. Of an, overall, disease process, that's unrelated. So. To test this in more detail. We. Made, conditional, mutations. In embryonic. Stem cells that. Allowed. Us to look, this one in an ISO genic background, as, you can see here in this slide on the top these are images. Stained. Form of neurons, human neurons stained, from that to that either control, contained. And functionally, relaxed in one gene or in your axon one, heterozygous. Loss of function as you. Can see here at the bottom this. Heterozygous, loss of function results, not surprisingly, in a decrease, of the mRNA, levels partial. So. The issue then was would test cause a change, in, synaptic. Transmission. And, it, does despite, the fact that the. Decrease, is. Only about 40 to 50 percent in, the set of expectation. There, is an equivalent. Decrease, in the. Frequency. Of spontaneous. Synaptic, events as shown here without. A change, in the amplitude, and. There's. This equivalent. Decrease, in, the, strength of evoked synaptic, transmission as, shown here on the right, so. These experiments. Validate. The. Functional, importance. Of. These. Heterozygous, mutations, and suggest. Indeed. That. The interaction one mutations. Are. Associated. With clinical presentations. Because they. Cause a, change. In synaptic functions, but, they do not explain, what the actual relation, is and in particular they, leave open the puzzling, question. Of why. The. Same mutations, and different individuals, causes. Different presentations. Another. Feature of new accents that I think is notability, and, in, fact is the central, part of what I'm going to tell you about today is, that new accents, are. Extensively. Alternatively. Spliced, creating. Thousands, of isoforms. These. Isoforms. Have. Been validated, by. Pyrosequencing. Of, individual, mRNAs, so. They truly exist. They. Result, as shown here schematically. From. The alternative, splicing of new actions at six canonical. Sites in. Which, small sequences, are inserted. Or deleted. IPhone. Directions have all six sites Padano accents have only two of these six sites and. As, a result, of combinatorial, diversity. This creates the isoforms, and, we ask the question. Whether. There is a in. Computational. A coding. Sense, of these. Artists. Patching events or whether they are more or less random. In. Other words do. These alternative, splicing events, represent, a code that, we can associate, with a given type of New York Oh. Are. They randomly, distributed. Between your ons and, so to address this question some. Time ago we use single-cell, ammonia, measurements, in. Which, we analyzed at artists pricing, quantitatively. By, rt-pcr, of. The. Asite a zoo of single, cells I won't. Go into detail, here because, it. May be a little bit too much to. Discuss, each individual, splice site, I just, wanted, to illustrate to you that. For the principle, in your accents, or there they always, seem to be Co expressed, as shown, here on the left and individual, cells. Demonstrated. For. Parvalbumin. CCK positive inter neurons of the hippocampus. The. Precise, levels. And the ratio of different, alpha and banana axons, is already characteristic. For, particular. Type of neon. When. We then analyzed, in the same types of neurons these, sites.
Of Alternative, splicing funnel, action 1 & 3, you. Can see that. There is a dramatically. Different pattern. Between. These. Various. Directions. Lines variants. And, neurons. Suggesting. That, each type, of neuron has, one particular pattern. Of alternative. Splicing. What. I would also like you to note, is that, in, each case I, turned, to splicing, in a given cell is not all, or none it, is always, in between this. Is important, because. What. It means is that a, site. Of alternative, splicing is, an, analog, signal it, increases, the information, content of alternative. Splicing dramatically. There is always a certain percentage, of protein that's in a certain percentage that's out for a given site and, thus. The. Coding. Diversity. Is, increased. Significantly. Another. Feature that is important, here is that although new, axons, are highly, homologous and. Have identical. Sites of alternative, splicing they're, not coordinately. Alternatively, spliced in, other words the, regulation, of alternative, splicing is not such that, splice site for in new axon 1 and 3 is always the same. Let. Me illustrate this, point, namely. The lack of coordination, a. Little. Bit further because it is really important. What. You see here in this slide is the plot, of. The. Relation, of a total splicing, at one particular site, splice. At number four in, two. Different directions 1, and 3 and you, can see on the left out and, on the right in 4, no X in one top. In. Bottom. Out for, an erection tree and. You. Can see this is what's done for. Two different types of neurons. Neurons. That project from, the prefrontal cortex to, the nucleus accumbens and, neurons that project from the thalamus to the nucleus accumbens. For. The prefrontal cortex neurons. There's, co-ordinate, alternative, splicing in that new axon 1 and 3 slice at four are predominantly. Out and both, however. For. The ceramic neurons, that, project, to. The nucleus accumbens, no. Action one is predominantly, out but, no accident 3 is predominantly, in and there's, quite a diversity, between yours, this. Illustrates. What. I mean with code. There. Are individual, variations, between specific. Neurons but overall there, is a mean. That. Represents. A specific, pattern, for, each type of neuron and each combination. Of spices inside the. Non coordinate. Of granular, surprising. Oft homologous. Sites, interaction. Suggests, that this is represents. Functional diversification. Among. You except on. Top of this there's. Evidence, from, the laboratories. Of other. Investigators. And fortune of mine that. This our tutors pricing as dynamic, is activity, dependent, that it, changes, as a function of. Activity.
Especially In the upper campus. My. Talk today. Will, in particular deal, with. The function, applications, of the new accelerator spice code. Finally. Not. Surprisingly. The, elections have. Trans, synaptic. Binding. Partners, ligands. As. It. Turns out they, have many, maybe. That is not unexpected. Given the, large size. Of the Nexen 1 2. And 3 alpha, molecules. And they had multi more modular. Domain structure, but. Still. It. Is amazing. What kind of molecular network, emerges, what. You see here again schematically. Is. A. Representation. Of a, subset, of these like. These. Ligands bind, to new. Actions. Via. Interactions. That are often, regulated, by alternative, splicing as. A. Result, the. Precise. State. Of new action complex. Depends. On. Alternative. Splicing and can be modulated by alternative, splicing I won't. Go into the ligands here, today, there's. A subject of another, talk each one of these ligands has, their own history. And their own, biology. I just. Wanted to mention. That. Your, accents, are remarkable. For the fact that none, of these ligands actually. Shares a structural, motif or any, similarity. They. Probably all, signal, in very. Different pathways, as a, result. These, ligand interactions. Are presumed, to. Diversify. The. Out readout, of a transfer, not extinct by new actions. Beyond. What, is often observed in other ligand. Pairs. With. These. Introductory. Remark I hope I have Illustrated. To you that. New elections are likely. Importantly. Synaptic. Cell recognition molecules I don't. Want to tell you about the. Major, story. That I wanted to illustrate to you today. Which. Deals. With this a total, splicing, of no accidents by said number four we're, dealing here with new, action three alpha and three beta and. Specifically. We asked, in this project, what. Is the function, of the tony election gene the, initially, this was not a project geared, toward alternative splicing, but. We, wanted to simply ask why. Does the gene have so. Many isoforms. To. Different promoters, and in order to address this we, generated, conditional, knockouts, that, delete every, transcript. We. Analyzed them in culture hippocampal, neurons and. When, we did this and we looked we. Compared, neurons, that Express, either, the, control, mutant. Clearly companies, or clearly Khamenei's we. Found an unexpected, phenotype. Namely, that.
AMPA Receptor mediated EPS. C's were. Decreased, approximately. 50%. But. NMDA, receptor-mediated, Piercy's were normal, as where, IPS seized this. Was unexpected because. The. Fact that NMDA, receptor, EPS is phenomen, meant that, neurotransmitter, release had to be normal and suggested. That the decrease in alpha receptor, mediated EPS. C's was due to a postsynaptic. Decrease. In amber receptors, further. Studies. Confirmed. This, many. Amplitudes. Were decreased similarly. Moreover. When. We looked at the cell surface at, the synaptic. Surface. Concentration of. An ample acceptor, glue a 1 as shown, here, it. Was also decreased. Specifically. In these experiments, we label cultured, neurons without, penalisation, with, an antibody to blue a1 in AMPA receptor, then, per mobilised. Labor them with additional, inter. Cellar antigens. PST. 95, for the postsynaptic side, we good one for the presynaptic, site of excitatory, synapses. What. You see is the density, of synapses, is unchanged, there. Was no decrease in density but. The puncta size which is a proxy, for the, amount of the protein in the synapse was. Decreased, for. The blue a one subunit. Similar. To the decrease in the AMPA receptor EPS see there, was a trend for PS in 95, there was no change in the good one the, decrease. In AMPA receptor, responses, was not due to a decrease, in. AMPA. Receptor levels because. As shown here by immunoblotting, there was no change, this. Decrease. Correlated. With, increased AMPA, receptor endocytosis. So. We wondered how does. This happen is this a developmental. Phenotype, can we rescue it and what, sequences. Of new axon 3 does it depend on. We. Used rescue, experiments, and cultured neurons and, we found the. New action 3 alpha, n 3 beta could rescue the phenotype fully, as shown, here. However. The. Initial. Experiments. Were performed with, one splice variant, of 3 alpha and 3 beta s s4 - when we did the same experiment, for, the same. CDNA. Is containing, s s 4 plus the, other splice variants, there, was no rescue. So. These experiments showed, that this is not a developmental. Phenotype, and suggests that the possibility. That our turtle splicing, may play a role for. Us exciting, at that time because, all. This enormous. Alternative, splicing highly, regulated of new actions had, never been shown to have any functional, significance and. Now here there was the first time we actually had something. The. Spy side of it was rising in new accents at site, for is extracellular, so. In the next experiment, we, try to test, whether. This was an extracellular. Mediated. Function of, new action 3. We fuse new accent to alpha, in new accent rabida is shown here for 3 alpha 2 a lipid anchor so, there were no intracellular, sequences. No, transmembrane, region, and it's still fully rescued. However. When, we use the splice variant, that is secreted of new accent 3 it.
Couldn't Rescue without. Membrane. Attachment, it didn't. Suggesting. That the cell surface, of. Synaptic, display. Of nerection 3 alpha or 3 beta with. SS 4 - were, sufficient. To, rescue. The phenotype. So. The most important, question at this point was is it alternative, splicing that controls, the AMPA receptor levels in a trance synaptic, manner in. Order to address this, we. Turn to Mouse genetics because, all, rescue, experiments, are always, over expression, exponents, there's an intrinsic problem, with always expression, experiments, because they, can create gain-of-function. Phenotypes. Or loss of functional phenotypes for that matter and so, we, generated. Mice in which we had genetic. Control, over, alternative, splicing, what. You see here is a schematic view of part of the new action 3 gene. Exon. Xx is the, one that's alternatively. Spliced at, splice at number 4 this. Exon, has an unusual. Size acceptor, sequence, basically, as, shown here in the box under wild-type, it has too many AIDS, so. We corrected, that so, to speak and we converted, it into a perfect, canonical. Splice acceptor, sequence, with. The goal of rendering, the exon, constitutively. Included. As. Shown, here with this red I, mean. Nucleotides. That were the canonical sequence. And as, it turns out that worked, no. Excellent three SS. 4 was no longer alternative, splice but constitutively. Included. With this mutation in. Addition, we, flanked, the X envelopes, P sites allowing. Us to a, considerably, excluded. With clearly companies, thus. In these knockout, mice 3. Induces. The SS 4 plus without. Cree. Delta. Cree control, which is a mutant. We have SS for plus as. A result, we have perfect, control of alternative, splicing when. We analyze these neurons, first again, in, cultured, neurons. From. The hippocampus, we obtain exactly the. Same phenotype as, we, had from the three alpha beta knockout, as, though. SS, 4 plus. Completely. Reproduce the total loss of function phenotype, there was a 50%. Decrease, in after receptor EPS C's whereas. The SS 4 - as shown here was the same as an unrelated, wild-type, control, no change in India by septum EPS C's no, change in IPS C's. Moreover. We. Found. That. All other features of these 3 alpha beta knockout, phenotype, in the hippocampus, we were completely, reproduce, and more, finally. That. SS 4 - new actions could, rescue the. Election, 3 SS 4 plus phenotype. So. This seemed, to indicate then. This is indeed a function. Of new action 3 in the hippocampus, to regulate, trance and optically. Postsynaptic. AMPA receptors, but. We wondered is. This really, a trance cinematic, effect is it, possible, that communicated, neurons, are not quite reliable, is it possible, that there is low levels of postsynaptic. Neurons, and. It's, this really something that happens in a vivo in, an animal and so, we went to slice physiology. To address these questions. We. Use the population, that is illustrated here, where, we stereo, tactically. Express the IV's. That.
Synthesize. The either pre recombinase. Or a mutant, clear the community as a control, we. Stereo tactically, inject these IVs, into, the hippocampal, ca1, region at, p21, we. Wait 2 weeks we cut slices and then we perform slice, physiology. What. You see here is that the ca1 neurons, are infected, because they're green and the clearly combination, of the control protein, accused, to egfp. We. Record, from the subiculum because. That, is the major output. Target. Area, for the ca1 neurons. We. Stimulate, the axons, from the ca1 neurons, by stimulating. Electrode. Between. The subiculum and, the ca1 region. The. Subiculum contains. Two classes, of neurons that, differ in intrinsic, electrical. Properties. These. Two classes of neurons are. Either regular. Firing. Neurons that fire action potentials regularly. In response, to current injections. Or burst, firing, neurons that, respond, with births very, surprisingly, with the name when. You fire when, you inject, currents, these, two different types of neurons also. Have. Different. Types of LTP. The. Regular. Fire young have. A postsynaptic. Classical. Form of NMDA, receptor, dependent, LTP, long-term potentiation. Whereas. The, burst firing neurons don't. Cannot, undergo NMDA, receptor dependent entity but, undergo, a presynaptic, form, of LTP that increases, neurotransmitter. Release enter, the cyclic AMP II dependent, so, these two different types of neurons are. Otherwise, quite similar, they, receive inputs, they both received, inputs, from the ca1 region but. They have different intrinsic, properties, in different, forms, of synaptic plasticity, so, this, population allows. Specifically. Analyzing. Presynaptic. Manipulations. Because, you can manipulate the, ca1 region yarn, selectively. And, you, can analyze the effect of such. Manipulations. On post. Synaptic responses. What. We found with, both the. Knockin, ss4, plus as shown. Here or the, knockout, with the 3 alpha beta conditional, is. That. SS. 4 plus in. Input. Output curve had. An approximately. 40 to 50, percent decrease. In amber, receptor, mediated, synaptic. Strength, that. Was completely, rescued. By. Presynaptic. Exertion, of SS. 4 into S s4 - as shown here on the right with, a green bar so, this totally. That produced the, phenotype. Of the. Cultured, neurons. Demonstrating. That this was indeed something, that wasn't a culture artifact. What. Happens to us is that when. We remove presynaptic. Neuron axon 3 SS 4. -. Either by deletion. Of new action tree or by constitutive. Insertion, of SS 4 then. Postsynaptic. AMPA receptors, are. Decreased, they internalized.
The. Postsynaptic, site, is basically, instructed. To, have fewer AMPA receptors. But. It continued the same amount of NMDA, receptors, as from here it changes, the, nature of the post synaptic response, there's, more postsynaptic, AMPA receptors, internally. Based. On this you would expect that LTP. Should be enhanced, because LTP. Consistent. A recruitment, of. Internal. AMPA receptors, to the surface, however. When. We tested, LTP, in, these, same neurons. After. Presynaptic, manipulations. We, found that presynaptic. S s 4 plus new, action three, blocks. Postsynaptic. NMDA receptor, dependent, LTP, and that, this could be fully rescued. By. Presynaptic, excision, of SS, 4 in. Other words. Alternative. Splicing of new accidentally the. Dictates, the, postsynaptic, competence. For. NMDA. Receptor, dependent, LTP, in pretty, much an all-or-none, fashion. The. Presynaptic, form of LTP in the other type of neurons, the burst fire on, was not altered. So. The conclusion, from these studies was that presynaptic. Election, theater surprising has a profound. Function. In, specifying. The postsynaptic. Receptor. Composition. In terms of AMPA receptors, and. Postsynaptic. Long-term. Plasticity, it. Forms a gate for post, to plasticity and, thus. No. Action 3 in this particular synapse, functions, to specify, one particular, property. Of the. Post synaptic response. The. Obvious question, from these experiments, then is. Whether. Other. New actions have the same phenotype in, fact, we could rescue, this reaction, tree knock out phenotype, over expression of other new actions, suggesting. That they should have the same function, but, as you know over expression always has as I, already mentioned, the. Potential, for artifacts. So. We one that if new axon one, has. The same phenotype. And. Maybe. Not very creatively, we. Did exactly the same experiment, from the election one, we. Made exactly, the same.you in from the election one where, we change the terms of splicing as shown here now for new action one its, exon 21 and has one exon more and.
Then. We. Analyze these, mice I'm not going to go into the details, it's exactly, the same as I, described. For you from your excellent 3 I'm, just gonna tell you the results, of this slice physiology. And, this. Result. Was. Completely. Unexpected, and surprising and. It shows you that, you can't rely on predictions. Based on in vitro rescue, experiments. We, found basically. That AMPA receptor mediated appear, sees in this input output curves were. Perfectly normal there, was no change, you could have presynaptic, new. Axon 1 SS 4 + SS for - they, were exactly. The same however. When. We looked at NMDA, receptors. Now. There. Was a totally, different picture now. We. Observed, as shown, here in this input output curves, in the middle in the bog diagrams, in the right both. For burst firing, and for unifying, subiculum neurons, that, the presynaptic, SS, 4 plus. Actually. Enhances. Response. So. You now get an increased. NMDA, receptor, response that, is higher than in the wild-type control, and higher than, the presynaptic, ss4 -. This. Is so because she's, logically. At resting, conditions in, a mouse most, presynaptic. Nerve axon one is ss4 - and. The. Result ESS for plus is a gain-of-function phenotype. As it, is in fact for an erection tree. So. This suggests, that. Very. Very definitely, now the ss4, plus is. Sort of the good one that does it it, enhances. Receptor. Responses, but, not for AMPA receptors, instead, for NMDA receptors. When. We look at LTP, again, we find no change in presynaptic, LTP. For. The for. The best fine neurons but, for the regular firing. Neurons we. Find not surprising. An increase, in Alta P because, if you have an increase and I'm really septal responses, you, get increased, induction, of out of here so you get an increase in LG P so. These experiments. Suggest. That. New. Axon 1 and 3 have completely different functions. Despite, their. Homology. This. Is actually, quite, intriguing. Given, the fact that they are tantalizing. Is not coordinated as. I showed, you and. It's, also intriguing, because it suggests that. You can independently. Modulate. Postsynaptic. Receptor responses, by. Differential. Regulation of, presynaptic. Alternative splicing of nerection 1 and 3 at. This price, at number 4. It. Also suggests. That. Plasticity. Is under, control of a Turkish passing as well which, has. Similar. To the nearest to this receptor, composition. Implications. For circuits. Responses. For how neurons. Respond, to synaptic transmission. As an, aside. What. I find intriguing here, is that as I mentioned earlier. Near election one mutations, have, been linked to schizophrenia in, fact to. The best of my knowledge there, still although, very rare the. Single, most frequent, single, gene mutation, that has been associated with schizophrenia and, it's. Thought that gaddafi, Nia involves, an mba receptors. At. Least that. Is my implication understanding. From some of the literature so. The specific. Regulation, of NMDA receptors, by Nexon one as as far as passing, may. Suggest, a possible hole here, a possible. Link to, schizophrenia. So. What I've told you up to now is. That. The. Next one and three other surprising, at the price at number four, has. Profound. But, very different, functions. In controlling. Postsynaptic. Receptor responses, in. The last bit, of my talk I. Want. To ask the question, whether that is what new accidents do is the. Function, of near action of one and three to, transcend, optically control, postsynaptic. Receptor, composition, and if so why, do new axons have all these other domains why, are there alpha and beta new axons and, moreover. Why. Do often. Been an election knockouts, which I haven't told you before but I'm telling you now have. Such different, phenotypes. So. To. Address this question we, analyzed. The different brain region, for the new accent3 conditional. Deletion, and, we. Find. Surprisingly. That. This there, is a very different, function. In. This preparation we, use culture, neurons from the oil factory bulb, to. Monitor, synaptic, transmission what, you see is an image of such neurons, might, with cell that is huge in the middle surrounded by granule, cells and, we can reliably. Record. Excitatory, inhibitory responses. In these neurons and. When we did this for the new action 3, alpha, deletion, as demonstrated, here, there was no phenotyping, amber receptor responses, no phenotype, in LDL receptor responses, but, it decrease, in IPS C's and inhibitory, synaptic, responses, that was quite profound.
Further. Studies. Demonstrated. That this decrease, was not due, to a change in postsynaptic. Receptors, instead, it, was due to a change and presynaptic. Neurotransmitter. Release. Totally. Different mechanism, that underlies the. Synaptic, phenotype. Of the same mutation. This. Decrease. In cultured. Alpha Trion. Actually. Reflects, a change in. A very specific, synapse. In all factory circles, namely. The reciprocal. Dental dendritic synapses. We. Showed this, by. Using against, their tactic. Injection. Of a. AVS and coding in a white have a mutant EGF. Tagged. 3 recombinase. As shown here in the left, we. Then cut slices 2 weeks later as shown, here and recalled. It from the mitral cells to, monitor, these dendritic. Synapses. What. We found in these input output curve is that there was the same decrease, in IPS, C's as shown, in cultured, neurons. This. Validated. The. Fact that this is a specific, change, in the standard under the synapses, and not surprisingly. When. You analyze the behavior of these mice obviously, before you do the slice physiology, there. Was an impairment, in, smell, as you, would expect. Shown. Here, by. An inability to, find, a buried, cookie poor, Mouse. So. How. Does this arise what is the nature of the impairment, here what is the, specifics. Why do these synapses. Here, have a loss, of. Inhibitory. Neurotransmitter. Release as, opposed, to excitatory, synapse, receptor. Composition. We. Tested, whether, the impairment. Could. Be rescued, by the same. Rescue. Construct, that rescued, the, phenotype in hippocampal neurons, and we, found that it could. Specifically. When. We looked at new action tree alpha, with SS formed, plus, it. Rescued, just fine. Moreover. We, observed no phenotype. Not shown here in the SS for plasmus. So. SS. For plus was, just as activist, as is for - very different from the hippocampal, neurons, moreover. We. Found the GPI anchored, near axon 3 alpha or, 3 beta could, not rescue the phenotype so. In this case you, actually needed. The transmembrane, region, and cytoplasmic, sequence, of no action for the function. To occur so the mechanism, must. Be different. How. Does it work we. Went further in, these rescue experiments, despite, the potential, artifacts. Due to over expression to delineate. The. Phenotype, and we, found, that. New, action 3 beta could. Also not rescue, so, this is an alpha specific. Phenotype, where you need both the cytoplasmic, sequences. And some, alpha, specific, sequences, in fact, what. These rescue experiments, indicated, is that over, expression of new action 3 beta is. Dominant, negative in these neurons as confirmed. Under white when we over expressed it in WyoTech neurons. What. Domains, does. Mediate. The effect now when you look at new action 3 alpha is shown here on the left top, domain, structure. It. Has this 6. LNS domains 3 our beta 3 EGF, like repeats, only, the 6th L&S, domain, is, also. Present and be done directions we've systematically, analyzed. Which Alan s domains might be involved and found. That. The second, Alan s domain single, an S domain, fused. To, the cytoplasmic, sequences. Of new action three and its transmembrane, region. Was. Fully able, to. Rescue the phenotype, so. Here then again it's, a modular function, one lms domain, now. Rescues, the phenotype, it's an LMS to me that's not present in new accent rabida which, explains, the inability of no accident to be there to rescue the phenotype. So. It's just basically. A, completely. Different mechanism completely. Different function. However. Despite being, completely, different, the, lms true domain, shares. With E and an, S six domain the, feature that, it was also alternatively, spliced. In. This case there's two variants, that differ by few amino acids, as shown here to a and, to a B, so. We tested, were, the rescue, of the phenotype, with an election three, alpha and an S to remain dependent, on our Charter surprising, and. We found. Reminiscent. Of, the. AMPA receptor, phenotype, in hippocampal cultures. That. Here for SS to only, there is this true - rescued. But. Not the SS - class. So. What we have here again is. A, complete, control, of a function, by, alternative, surprising, but. It's just the different domain different. Function. How. Does this work. What. Are potential, mechanisms. Of how. It might work well. It turns out that many years ago we. Observed, that new actions, form, a stoichiometric, complex. With district, likened a cell. Adhesion molecule. That is involved, in particular in. Attaching. Cells. To an accessory. Matrix, but, it's also found, in a subset, of inhibitory. Synapses, in brain. This. Complex, interestingly, was, regulated. By alternative, splicing and L&S to what. You see here is that only Ellen s2 -, as.
Shown Here the single I mean. Band. Here, actually. Binds this. Is calcium dependent because, calcium is a structural, component of L&S domains none. Of the other. Splice. Variants, found. So. This was an interaction was, regulated. By total splicing, and, SS - suggesting. That. SS 2 in this case, media. It's trans synaptic. Signaling. By. Binding to, disturb. Lycan. Gratifyingly. Just, like SS for SS, - turns out to be highly regulated. What. You see here is, the, quantification of. SS. - in. Hippocampus. Stradun and. Neurons. That project from the thalamus and medial prefrontal cortex, to the nucleus accumbens in. The stratum, of course and, you can see that in each case there's, a different pattern showing. Their. Regulation. Of this, actor too surprising similar, to what I've told you about aces, form so. This suggests that, tight, regulation. Of new action triode function, in inhibitory, effect about, synaptic. Transmission is. Mediated. By regulate, that alternative, splicing. Often. Election 3 now, at SS number 2. So. To. Summarize, this. What. These data suggest is, that new axons are essential. In, these synapses. Specifically. For, the transom. Optic. Control. Of, either. Post, synaptic, properties. Or, presynaptic. Properties. Depending, on the, specific. Synapse. We, are analyzing. How. Can we. Reconcile. These data, as well. As other data that I haven't had time to discuss, suggesting, additional, functions, for new actions. With. An overall, conceptual. Model and I, would like to propose here, a high resolution model, to. Sort, of illustrate this what you see here in red is. New action and, it. Has all these features sticking. Out. We. Believe that your actions are signaling, platforms. That. Serve to coordinate. Multiple trans. Symmetric, ligands. These. Ligands. Regulate. Among others postsynaptic. AMPA receptors. LTP. Release. Probability. Endocannabinoids. Synthesis. Which I haven't had time to discuss GABA receptors. Potentially. In, a real receptor, trafficking, definitely. And other, things as well. We. Think that that conceptually. Explains, how. New, actions work in, a sinner and, in this manner new actions, contribute. To. The molecular logic. That. Determines, the properties, of synapses. This. Fits into overall, sin affirmation, in. The part of synapse specification. The process. That. Endow, synapses. With. Particular. Properties in. This case. Different. Synapses, different, properties, such as release, probability. And. Receptor. Composition. Thus. New. Actions are central controlled switches for. A facet, of synapse formation, and. That are, complemented. By other control, molecules, that mediate other function, features of synapse formation. There. Are many other features that come to play here beyond, new axons in. Closing. I just want to mention that, one of the new axon ligands, lateral, fill-ins, which. Also bind to other presynaptic. Molecules, such as tenuous and flirts are in. Fact responsible, for another feature of, overall, synapse formation and, I won't go into this let, me just illustrate to, you the natural, fill-ins are distributed. In. In a specific manner, throughout, a particular. Nuan and are, essential. For, synapse. Formation for, the initiation, of synapses. As opposed. To the synapse specification. Function, so there we, go through. Different. Parts. Of the puzzle that, are. Contributed. By different. Molecules, in an overall, Network, molecular, network and, I think that if we go along this path we. Will be able. To. Dissect, deconstruct. The, molecular, signals, that. Are responsible, separately. For. The initiation, in specific, synapses. At specific, locations. For. The organization. Of canonical. Features. Of synapses, such as release probability. So, release machinery, and, the. Specification. Of particular properties. Such. As receptor. Composition. And. Let me close by, acknowledging. Major. Contributors. Most. Recently. The. Work that I discussed, on the erection one alternative, splicing was, performed, by Ginny Addai, the. Work that I discussed, on your accent, three SS.
Two And neurons, were standby Pangu both. With, the help of Justin Tata and, Jason. I Auto. Performed. All the initial, Studies on your action 3 alternative splicing, at splice at number 4 thank. You very much for your attention. And we have time for some questions I know some people may need to go but others I'm sure would like to stay and engage in the conversation their, microphones, in the aisle if you have a question to pose to the speaker please. Go to one of those so people watching my video I. Am. Curious, if you would like to come in besides, supplies, variations. What, do you know about the, the, diversity. Or at the level of post-translational modifications. Are some of these proteins and how that can possibly connect, to the functional, diversity. So, that notifications, are important, especially. Phosphorylation, but we just don't know. In. The first, part of your talk where you spoke, about the single cell sequencing you, suggested. That the effect. Of the splicing, was a sort of analog, signal, but, in the second half of your talk you spoke, about each, splice, site as if, it's a discrete, on and off switch so. Let me ask a specific question when. You turned, on and off, splice. Site number four splice, site number two did you check that the other splice, sites are unaffected, great. Question. So, we, did check we can't find a change in the other splice sites when we turn it on and off for. Experimental. Reasons, we. Have to convert the analog into. A digital signal because otherwise we need to have black and white otherwise, we can't look at the difference. That's. So. I didn't know anything that I want to convey the impression that, physiologically. It's always in or out it's always somewhere in the middle your mouse has two alleles so you could turn one on and the other off. Yes. We could potentially do that very difficult though. Yes. You. Mentioned, about this effect, of splice variants, on the ampa and the nmda, receptors, I was. Wondering, what happens, to the hundreds. Of other ligands. At the. Person, optics I do you think they also might be affected, in some other ways.
So. We. Think that the alternative splicing of presynaptic. New axons. Acts. Differentially. By, changing. The. Complex, is the new actions engagement. So. I think, physiologically. What happens, is that some reactions, are spliced in some are out at. The hippocampal, synapses, in the control, conditions, it's mostly, spliced, out for ss4. Under. Those conditions these. New actions, will, interact with different ligands, then. When they're spliced in we. Think that this switch is, what. Determines. The. Change, in receptor, composition. Postsynaptically. But. Identification. Of the presides, ligands. Is. Not. That straightforward because, there several of them so, we don't actually know which ligands, are involved but. The ligands, the. Nura likens 1 3. Do. They interact with some of the splice variants, good question, they all interact. Differentially. So, new or leggins form. A total. Subject, for another talk because they. Are actually quite complex, they're, postsynaptic. Liens but they're not the only ones as, it. Turns out the or ligand 1 is. Itself. A topically, spliced, it. Has a splice site B the. B plus binds to both plus and minus nu axons to be - only binds to - no accidents, ok. No, 11 - and three never bind to plus they only bind to -, okay. So, it's complicated, it's not that, straightforward. And there's. A diverse. Interaction. Network I think. We have already experienced. From. Beautiful, studies in cancer, biology, how, protein. Networks, can be quite complex, it's. Going to be the same thing at the synapse yeah it's. Not more so since, you're the best expert, could I ask you the last question about the turnover, rate of the synapse, during. Development, I'm. Sorry again turnover. Rate of the synapse. Turnover. Of the synapse so. There's. Beautiful studies, in particularly. From. The Gans, lab and at, NYU. Demonstrating. That, approximately. 10% of, synapses, turn over. 90%. Once, formed, our plastic. In terms, of their properties, but, not in terms of whether or not the synapse gets eliminated they, stay pretty constant. Thank. You. At. The beginning pointed out that. Deletions, in Direction associated. With interesting, phenotypes, that are not necessarily, breeding, true in terms of autism and schizophrenia, is, it possible, to study these same parameters, now with human iPS cells, differentiating. Them into. Organoids, and trying to understand, how, such, changes. In, deletions. Might affect, in, human, cells or what's happening or is that system. Not yet ready for them I think. The organoids. Have a long, way, to go because, the. Neurons and these organoids. Have. To this point never been shown to actually form synaptic. Circuits. They. Have not no so, there is great, excitement I think organoids. Will be fantastic. Models, for early, developmental, impairments. Schizophrenia. And autism are, probably, not early, developmental, impairments, they're. Probably in. Synaptic, networks. That are quite mature as, a, result, I think that. We can use. Cultured. Human neurons. Probably. As better more, reproducible, systems. Especially. Scaling. Up to, examine. Features. To potentially, look for therapies. There's. A chance, that that's going, to be possible to do I think, Organo it's at this, point have a long way to go also because of this incredible. Variability. It's, it's, very very high, variability. Having. Said this however I, do. Think that, we. Need to push Organo it's because, they are so. Powerful. Potentially. If one could get them to the point where they would actually form some kind of circuit, there, would be two or three yeah. Well. We can continue the conversation in. The medical, library for those who interested, in coming to continue. To chat with Professor suit off meanwhile let's thank our speaker again, for a wonderful presentation.