 So, good afternoon everybody. It's my great pleasure to introduce the first distinguished visiting scientists program talk. And it's been a great honor to host Professor Christian Rosamund in my lab. And he is the speaker today. Christian is a really well known figure or person, a scientist studying SNAPTIC transmission throughout his career. And let me introduce his biography. So he is from Germany, and he originally studied pharmacology from Goethe University in Frankfurt, Germany. Then after that, a little bit unusual, I think at that time, he went over to the US to do his PhD at the Wollum Institute with Gary Westbrook. And after that, that's when I got to know him. He went to do a postdoc at the Salk Institute with Chuck Stevens. Yeah, this is way back in 1993, we're just chatting almost 30 years ago. And then subsequently he became a Helmholtz fellow and moved to Germany in Göttingen at the Max Planck Institute for biophysical chemistry with Dr. Alan Marti. And then he subsequently still staying at the MPI by physical chemistry. He became a principal investigator of Heisenberg fellow. And then in 2008, he moved to Baylor back to US now this time in Texas to take up an assistant professorship in the Department of Molecular and Human Genetics, and also in Department of Neuroscience. And then working through the ranks, he became a professor in 2008 at Baylor. And then soon after he moved back to Germany this time to Berlin at Charité, where he is currently to take up a professorship and being involved in multiple responsibilities at Charité. And he's currently a professor in the Center for Basic Sciences at the Institute of Neurophysiology. So apart from these very prestigious fellowships through which he's gained various positions. He's been awarded many grants, prestigious grants, and also since 2014, he's a member of a Board of Trustees of Schramm Foundation in Essen, Germany. And since 2015, he's serving as a scientific advisory board member in the Department of Biomedicine and University of Basel. And for in recognition of his outstanding scientific contributions in 2019, he was elected as a member of National Academy of Sciences Leopoldina in Germany, which is the highest scientific distinction bestowed by a German institution. And he's, as I mentioned, he's an expert in synaptic transmission, and he's really well known for his elegant work throughout his career. I think my very first strong impression was this really highly cited 1993 science paper that he produced as a PhD student, taking advantage of his pharmacology background and really having an elegant method to estimate the diversity of presnaptic strengths, which is a method that's been used throughout, I won't go into the details, but you will hear more from Christian. And I think my favorite is the his identification of cytoskeletal modulation of ion channel. And since then he's been involved in regulation of neurotransmitter release that's very crucial for accuracy of information transmission. And so without further ado, Christian, please. Well, thank you Yuki for introducing me so so graciously. So many honors I nice to hear about. Yeah, I just don't. Yeah, we love the signs. I think this is always the great part. And, you know, I want to maybe start before I go into my research topic would really like to thank OIST to allow me to do my sabbatical here since last September in Yuki's lab. And this was done through the distinguished distinguished visiting scholar program. So again, thank you so much for making that possible for me to come here. I particularly would like to thank Jonas Fischer and his team to to help me go through the moving here part which is, you know, it's it's it was quite a little interesting adventure not not complicated but very interesting and it still remains to be. And yeah, it's a very great honor to be in this very stimulating environment this various prestigious Institute. And yeah, I'm still looking forward for three more months of of interaction with you guys so. Yeah, so what I want to talk to you about is is really kind of at the core of what my laboratories interested in is how synapses actually communicate what's how do they actually do their job. And the, let me just maybe start a bit more broader. The brain, you know, we all know this is probably the most complicated structure in the universe. We know when we look at the anatomy of the human brain. There are approximately 1000 different sub areas within the brain, executing various functions that we performing every day when we think when we sleep when we eat when we have feelings, etc. And that of course is mediated by a whole large host of cells that underlie these structures. There are, of course, mainly to mention the neurons that are astonishingly 86 billion neurons in the human brain. Interestingly, also we see that the number of glia cells and this is something that brought me here to to to for my visit make make almost the same up the same number of cells in the brain and studying synaptic transmission. We really have to start thinking about not start thinking I have to start thinking about it. Yuki is doing that for many years already. What the glia is doing and the interaction and helping synapses to actually do what they're supposed to do. Now, this is just some statistics of the human near cortex. You, when you just look at a microliter of brain volume, you find in there not only about 40,000 neurons, but each of these neurons make approximately 15,000 synaptic connections to other neurons. At the same time, received a similar number of synapses are making this an incredible complex network of interactions. And we find in this microliter of brain volume, four kilometers of axon lengths, which just tells you the complexity of interactions. And if you are just looking at like a cube, that is about 2000 times smaller of this one microliter which is 0.5 nano liter of volume. And you make a reconstruction of all the neurons and connections in the brain in this in this volume of cortex. You can appreciate the complexity of this structures. So you see here in blue axons, later in cyan, you will see the synapses. We're just sort of diving into this tissue through a vessel. See the synapses colored in cyan. And it's just amazing to think about 80 by 80 by 80 micron sized cube how complex that structures here you start seeing an axon. And a dendrite and the synaptic connections in between. This is an individual synapse and of course you can see they are in this cube about 80,000 of those connections. And so as you can, you can think in a way that the brain is a humongous, humongous computational organ that enables us to communicate information across neurons across synapses. So my interest has been, you know, thinking of the brain. Of course, you have to think about the areas and neurons, how they're connected, how the neurons are electrically active. And I have been particularly interested in the synapses, how they actually convey that information from the pre synaptic side to the post synaptic side via action potential and use neuro transmitter release that in turn then activates post synaptic receptors. So just a kind of a scheme of this. One of the home walks of synaptic transmission is really it's incredible speed. But let me show you first what happens really at the synapse we have a pre synaptic depolarization induced by the incoming action potential. You have council channels sitting right next to it that open council fluxes in and then you have fusion competent vesicles that are sort of sitting right next to the membrane, waiting for the, the, the, the go signal. This relates to the stop and go of my title, they're waiting for the go signal to then actually fuses the plasma membrane releasing its neuro transmitter content into the synaptic cleft, and then activating the post synaptic receptors. To open it's, it's gate. So cutting on and cut irons in case of photomate receptors, sort of flow into the post synaptic dendrite, we selecting the pre synaptic electrical activity as a post synaptic depolarization. So here's just to show you the speed questions of when you when you look at functional of neurons like to physiology is a really important technique to do this. You see here the pre synaptic action potential recorded from a from a nerve terminal, and then you see subsequently in the post synaptic membrane, either by looking at current clump recordings of depolarization of the membrane. Or when you look at voltage clump recordings in inward current that occurs on really in a millisecond time scale. So if you then blow this up in terms of the time scale you see here the pre synaptic action potential, you see the pre synaptic counts in current and then you see the post synaptic response. You can start appreciating the speed between the pre synaptic and the post synaptic signal occurring within a millisecond or so depends a little bit on the synapse you're looking at. But this is very important because neurons fire in the brain at very very high frequencies, and the precision of this firing of the neuron is an extremely valuable information content. And so synapses job is to maintain that pre temporal precision, and therefore synapses have to be very very fast. And so that's one of the things that fascinated me in this area. How can this cell biological process of vesicle fusion induced by calcium influx take place in a microsecond time scale. Here perhaps the probably the most fast the fastest cell biological process that I know of but you know given this diverse auditorium yet always maybe someone comes up with something faster. I'm very curious to hear about it but anyway it's one of the fastest things that happen in biology. But the other thing that you have to keep in mind with synaptic transmission is not only speed, but you also have to keep in mind that synapses are have have a very, very precious content that are easily be consumed each other synaptic vesicles. The pre synaptic terminal and the nervous central nervous system has only a few vesicles at hand fusion ready ready vesicles that can be easily consumed if you think of a fifth or sixth action potential. And then once those vesicles are gone they need to be refilled. And so really as an issue of supply demand that the nerves the pre synaptic nervous pre synaptic nerve terminal has to face. When you look at ongoing firing of synaptic transmission, you see that the amplitude of the response for example can go into their knees if you stimulate them too strongly. And you see that of course in certain areas of the brain that you have a not a steady response but kind of a plastic short term plasticity response and synapses. When you look at other synapses, you see different behavior. So there is versatility in the way how nerve synapses communicate the action potential train. So the way I see this is like they speak different dialects. Some synapses tend to facilitate some tend to depress during the train and this kind of code is is imprinted in the structural molecular composition of the individual synapses. And there is more and more evidence for that similar to that neuron have incredibly diverse structure and axons and dendrites fire and very, very different pieces synapses themselves are translating the action potential code into different types of dialects as reflected by this versatility of train responses. And the other thing that the synapses are facing is a signal to noise issue. So when you make a vesicle fusion competent when you make it ready for fusion, it has to be right there because you only have a fraction of a millisecond to release it. And you have to bring it into an energetic state that is kind of just waiting for this last blip of signal which is a calcium influx and that brings it into an energetic state where these vesicles can occasionally also fuse spontaneously and that would create membrane noise. Where you see an absence of stimulation you see this little blips of inward current, which are caused by this so called spent spontaneous release. And if you remove certain proteins for example that we will talk about later the snare complex that bind to the snare complex. If you remove this protein you suddenly see that the synapse faces an incredible problem of producing too much noise by sort of preventing those vesicles to not be not be released an absence of stimulation. This protein cause complex and we've been also studying a lot in the in the laboratory so so there are things to keep in mind synapse synaptic vesicles need to be kept in a certain energetic state. So that we have a maximal response when we give an action potential and minimal activity outside of this outside of this firing of presynaptic nerve terminals. And so that's this isn't kind of a waiting of functions in a nerve system you want to make the vesicle fusion efficient and their fusion properties so that they can fuse efficiently and fast, but don't make it too easy for them to fuse. Otherwise you just get too much membrane noise. Good. So let's look at this on a on a scheme again. What does it take to make a vesicle fusion competent and ready for fusion. So we have this calcine trigger the vesicle sits right next to the membrane we call this docking. They need to be brought down to the membrane and they essentially touch the membrane. And also we have to assemble a fusion apparatus so that, for example, the forces that needs that are required to sort of break two membranes and make that two membranes into one that that these forces are available the energy for that is available. And this, this transition is can be actually triggered by the influx of calcium. Now, now the energy barrier for two membranes to fuse is extremely high. In normal cases, probably would be taking spontaneously without any of this fusion apparatus years. And that's very good because otherwise ourselves would constantly disintegrate through instability of the membrane so membranes intrinsically are very stable. And so you need a catalytic machinery that makes this transition very very easy and fast. And that's what I have been fascinating and what how can you build this machinery. So the system that we use in the lab to study is a very very reduced system. And this sort of is the stink is simplest connectivity you can think of, it's this kind of robins on cruise or situation, you have a single neuron growing on an island, and this neuron can only talk to itself, it has it forms axons it forms dendrites, you can see them here in the dendrites and blue this is so much and you see all that the axon forms, hundreds of synaptic connections onto its own dendritic tree and with this system, we can study the synaptic transmission with a single patch pipette, very easy system You know, Yuki got also very famous when she was in in Chuck Stevens laboratory she used this technology in fact to discover how synaptic attack in the council center we will meet mentioned this briefly don't is actually operating at the synapses so she showed this in knockout mice for snap the tagman using this autaptic system. And, you know, I'm also been very fond of the system because it's just very simple, very highly controllable to study the mechanisms of release. And so what we basically do is we do a simple pre short somatic depolarization of the membrane from minus 70 millivolts to zero millivolts for just two milliseconds. And then along the axon initial segment you get an unplanned action potential sort of distributing the electrical stimulus along the axon, and you're reaching all these hundreds of synapses at once, and then causing a synaptic response that you see here. The beauty of the system is the response size he see here is is quantity quantitatively dependent on factors that we can all collect in our system. We can count the number of synapses, the more synapses the larger the response, we can measure have ways of measuring the number of fusion competent vesicles that these hundreds of synapses make all together it's kind of a sum up. There's also called pool of readily vesicle releaseable vesicles, and we can just simply do that by applying hyper tonic solution to this neuron. And then we look at the transient component of the responses are all sort of these mini blip responses from individual vesicle summed up and count them essentially by looking at the charge current charge and flows and typically for these couple hundreds of synapses we have 5000 fusion competent vesicles. So let's say you have 200 versus synapses and 5000 5000 fusion competent vesicles we can make this quantification for each individual neuron. And we can put this of course in context as an optical response. And we can also look at the spontaneous release from these pool of vesicles so when we have 5000 vesicles available. And the activity of five hertz of five per second that means each fusion competent vesicles that sits here fuses spontaneously every 1000 seconds. So because rates we can directly measure the rate of release of individual vesicles both doing stimuli and spontaneous, we have means of describing the energetic state of these vesicles and quite with quite high accuracy. And that allows us to study, essentially the machinery that underlies this process in great detail and find out what are all these factors these proteins and so on are doing to make this release process possible. So, here is just to briefly mentioned how can we actually measure the efficacy of release itself from Carlson triggered responses. A classical method is to do pet pulse stimulation we give two action potentials right after each other. And then, depending on whether the second response is smaller or bigger. We know that the release probabilities high or low this is just something that has to do with supply demand calcium influx. But we can also directly measure it by quantifying the number of vesicles that are released by an action potential by the number of vesicles that are available. And we call this vesicular release probability again something nailed down to the behavior of individual vesicles. So if you have 5000 vesicles available and 500 are released here, we know that the probability of vesicle to fuse with an action potential is 10%. So it's essentially we know the property of the dice that is rolling when this action potential comes in, we say okay, an action potential makes one out of 10 vesicles to fuse. And this is very important because we can now look at different synapses and we find at the, let's say at the cortical synapse that this likelihood is only 5%, while when we look at an auditory synapses might be the probability might be 15%. And so knowing about this probability of release helps us to understand why some synapses are, you know, depressing during trains of action potentials while others are sort of showing facilitation. So the other thing that we really like to do is to manipulate the molecular content. So if you're really trying to understand how the apparatus of release itself works, we need to go after the proteins. And so for this, we've been, you know, initially, you know, as Yuki said, I started to use a lot of pharmacology to understand the synapses, develop tools to use these autapses. But then, luckily, I was also got in touch with people like Tom Sutoff on the exposure that actually made knockout mice for presynaptic proteins. And our system was perfectly designed to actually study what's wrong with my synapse if I take this protein away. And so that's kind of what we've been doing a lot over the last decades, I would almost say we use transgenic mice that lack essential components of the release machinery, and then characterize the phenotype of this loss of function. But then we can go one step further we can bring back the protein of interest using in particular antivirus we like them a lot by essentially reintroducing this protein using a virus. And then, for example, study mutations in these proteins, or the concentration of these proteins, and how it affects actually release. So the system by putting in reporters or manipulators of neuronal function here, for example, we do have a mixed culture of two different types of neurons. They have green nuclei and red synapses or red nuclear and green synapses. So you can study how neurons form connections with each other under controlled conditions. Here's an example how we use this transducer, the manipulators of electrical activity in our experiments, and brings me also to the second last technique that is important for us is electron microscopy, because synapses are functioning, but they also are structures, and the structure is obviously also involved in controlling function. So in this case, when we had a visiting scientist in this case was Eric Jorgensen came to Berlin, and they had been sort of trying to read and do the version Hoysan Rees 2.0, sort of trying to capture synaptic events on a very on a millisecond time scan using ultra electron microscopy. So the idea is to use a high pressure freezing device, a transfect neuron with a general dobson to control their firing property using optogenetics, and then essentially stimulate an action potential milliseconds. Before we freeze and arrest the structure in its native state. So then when you look at this electron microscope microgram, you see actually a fusing vesicle in the central synapse, which is something that's very, very difficult to do if you do electron microscopy in absence of syncing with the activity of the neuron. It's been extremely powerful novel technique that we've been utilizing now in the lab a lot. And the other thing that you can do is you can sort of look at these structures at the pre synaptic membrane and the position of these vesicles, and sort of ask okay this one is directly attached to the membrane because it's docked. These are the vesicles that are sort of close by. And then we give us in single action potential. And then what we see what what happens with the the number of vesicles we see. And we've what we found is that only the number of vesicles that are directly attached to the membrane is reduced in the in the profile but not the dog desert vesicle so we know for vesicle to be fusing to be able to be released it really has to attach to the membrane. And so that these are kind of tricks how you can use a system to understand it's it's it's architectural principles and and how the positioning of these structures reflect function of these vesicles in the synaptic terminal. Here is just even more advanced things that we currently trying to do is using cryo electron microscopy, which is much more native than even the high pressure reason that I showed you here we basically use no staining no no heavy metals, no fixative really membranes and trying to identify synapses why we are stimulating them electrically. And you might see this here this is a synapse and then you look at carefully to see little bumps here. And we think we don't have proof yet for this is all work in progress. We think that we can actually resolve in this technique in fusion pits and even post synaptic receptors here in quite high resolution and the long term goal of this is that we break things down beyond vesicles and membranes that we actually start seeing the proteins that underlie these activities that we've been interested in. So this brings me to that level, the protein that the core release machinery, which is going to be main focus of my data science here of my talk. These are this is so famous snare complex and the snare complex is a kind of bridging vesicle and plasma membrane. And this complex has to be formed so that vesicles are becoming fusion competent. And I just thought that this apparatus this complex is is providing the energy required for the for the energy for these men that these two membranes use which is with each other. And before fusion this so called trans snare complex because to the two different membranes are involved and after fusion it's so called system snare complex, where you fusion has been finished. And the trick here is how is the snare complex how is the snare complex operating. And how is it actually formed so these are kind of big questions in the field. It's been studied for many, many years, but we still really don't understand this in much detail. Before vesicles come up come in close contact with the membrane, the snare protein syntax and snap 25 and snap the breath and sort of float around and that separate membranes. And then you have for some syntax and being controlled is a kind of a scavenging or like a junk protein monk 18, and then somehow these membranes that inform the snare trans snare complex. Fusion complex ready for for the calcium trigger, we also have to have a calcium sensor in here. And then when calcium comes in, these apparatus allows the the few merging of these two membranes. Okay, so here's just a little bit more detail about syntax and which are mainly going to talk about today is syntax in can be in different confirm confirmations it has kind of the executive part which is a red part, and the regulatory part which is shown here in the end term is and the HPC domain, and we have here among 18 so at the start they form a complex, and then you have to come in with a vesicle snare protein them to us and after Brevin, and that sort of crawls around the snare motif. The two snare motif sort of merge here at the end terminal part. But this is only possible when these, when, when the regulatory domain of syntax and can sort of open up here. And then eventually, you have enough template that also snap 25 which is showing in green is sort of joining the the complex that the end you and ending up with this trimeric snare complex that forms over all these four bundles. And we have these at the end we end up with these complex of vesicle membrane plus membrane and snare complexes. And that's kind of the fusion ready vesicle that awaits the calcium trigger of course what is not shown here is the calcium sense itself. We don't really know where it is exactly it's also an interesting question. But one thing that we can already do in our analysis is to ask how many of those snare snare complexes do we actually need for fusion to take places a single one do we need 10. And so using our genetic tools we can address is simply by taking out to control in the concentration of one of those key proteins in this case and taxon by controlling its amount in the synapse and how do we do this. We take essentially knockouts, heterozygous mice, and then knockdown approaches so we reduce the concentration of the protein using also knockdown constructs and combine those. So essentially we do kind of a titration down of concentration of the protein in the synapse you see go gradually the signal goes down, while the synapse is maintained. And then we can look at these different concentrations how synapse is actually performed. What we find is that we look at for example the number of vesicles that are made a fusion competent is so called RRP charge drops, of course at some point to zero because we need the snares to make vesicle fusion competent. And but looking at the function and the relative concentration or relative expression, we sort of gain an idea about the, the importance of the stoichiometry of this process. And we can also do the same game by looking at vesicle release probably the likelihood of the die throwing what I told you about. And we can look at how when we constant lower the concentration of syntax and that gradually release probably drops. And from this we can be actually found that the slope of this process approximately three. And that could be indicative that essentially there is a transition between vesicles that maybe from one snare complex and multiples that cause an increase of the fusagenicity of those vesicles. And basically illustrating here that we can utilize genetic molecular manipulations to understand better the process of fusion itself. So this is just to show you the modular domains of syntax and I already told you this is a regulatory domains which are sort of opening and closing and enabling the the the business part of the snare domain to to be into interacting with the other snare proteins. And I don't want to we've been working on this quite a lot is very interesting research done here. But I want to mainly talk about to you about executing domains here the snare motif itself is the linker region between the snare motif and the transmembrane domain. This is where they form this bundle. And in between we have a region short region which is important for function and then of course the membrane anchor. So this is what I mainly want to talk to you about what we found in our molecular structural analysis of this region how that actually affects vesicles for in their fusion properties. And to show you this, we're talking about mainly this region here in the trans domain on the system system structure, this gray area here is the linker region transmembrane domain and that's a snare motif. And so the first thing that we did was to ask how important is the length of this structure how is essentially the idea is, how does the snare snare complex couple to the membranes, because something do something with they may do something with the membranes. So there's a pulling in the membranes making them sort of fuse, or making them unstable or forming kind of dimples. This is something that we don't really know yet. But we thought, let's just put a single single helical turn almost it's not even a full turn by just putting GSG into the sequence so we take the syntax and knock out mice. We made them so they are indulgence protein is gone. And then we place it by wire types and taxon or by mutant versions for example that contain additional three amino acids, either at the edge to the transfer the snare motif or to the edge to the membrane domain and then ask, what does this extension of that linker region does to release and does the position of that extension play a role in the fusion process itself. So we must say we were quite surprised to see how dramatic the effect was. In extending the lengths. I will I thought this. I expected that something will happen. But if you look at the logarithmic scale this is a release. This is the amplitude of the synaptic response, typically like five nano amp in our system and that drops by a factor of 100 to 1000. So we are inserting these three amino acids into this linker region. And so to our surprise, the biggest effect was actually putting this linker right next to the edge between the snare motif and the linker and not the linker and this And why do why was it a bit surprising because this region this linker region here has lots of basic residues, and we have the negative phospholipids in the membrane so by making this longer here by pulling this out a little bit. We moving this basic charges further away from the negatively charged phospholipids we thought this would kill the response more likely, but it was actually the other way around. What was also important. What was more important is to put extra distance between the snare motif and the man and the linker and the transmembrane domain, arguing that there must be some kind of spatial or mechanic process sort of force in place when the snare complex sort of forms zippers up up to the end and then somehow create creates force to the to to enable vesicle fusion. So we found that the fusion was essentially gone in this in this mutant. And if you put this linker in here next to the transmembrane domain, we found that the release was larger reduced and also very interesting the release was much much slower. So the release itself slowed down by a factor of about 10. This is something similar that actually Yuki found when she looked at the knockout mice of synaptic tagman so release itself became much much slower. So clearly imprinted in the structure of the snare complex is are the properties like the speed and the efficiency of the release process itself. The other thing that we found was quite surprising despite the fact that this mutant here had a way reduced evoked response and the release was slower. We found a vast increase in spontaneous release activity so the opposite of what was happening with evoked response. So we see here now that the structure of this complex can go and both can have impact in to the two main types of fusion spontaneous and triggered cuts and triggered release in opposite directions. And keep in mind we want in the synapse that evoked release is maximal and spontaneous release is minimal right so to have a big signal to noise increase signal to noise ratio. So the high spontaneous release activity is bad. And so in this tells you that intrinsically fusion is still possible, but we now actually made the release less favored, because we can't really control when fusion occurs with our calcium trigger. And so we then said okay let's look at this linker region important what what is happening in the linker region in terms of evoked and calcium triggered release and by simply making single point mutations and use basic residues to see here. So a lot of basic residues just in this very short linker. And when we made individual mutations in this residues we didn't see much of an impact in evoked release. If you make two or three or four mutations then you start seeing a bigger and bigger effect on evoked release, which is because in positive charges need to interact with negative head groups of the phospholipid membrane, but single residue changes. That's what we do here had a rather little impact on evoked response, but to our surprise, they could they depending on the position they had dramatic effect on spontaneous release. So in particular mutating the position 260 and making a reversal of the positive charge to a negative charge cost a drastic drop in spontaneous release activity, which was telling you again that this region is very very sensitive to controlling the both evoked and spontaneous release activity. So we thought to look at this a bit more careful, but one thing that Gulch who was a postdoc in my laboratory when she did these experiments made you know we typically when we use our viruses we always look for gene expression levels for the viruses, because we want to keep them all very similar. So we don't have these titration effects that I was telling you about right. If you have to low concentration then of course you get different phenotype. So what she found notice is that the band of the syntax and protein that we quantify for expression vary depending on with the mutation you saw that the vital band has the highest molecular weight. And if you have for example having mutation with a 260 we start seeing a lower band and kind of a middle band and some mutations even you see three different band sizes. And that is was a surprise nobody had noticed this before. For example if you look at Hex cells so they are non neuronal cell lines and you express syntax and they all run at a lower band. Okay, but if in neurons actually the band height is higher than in cell lines indicating that neurons make a post translation modification to the syntax and protein. Somehow altering a structure in this case we say depending on the on specific basic residues in the linker region of syntax. So that was the discovery. We thought it's really worth following up what's going on here and how is this change in band patterns related to at all to what we observe this spontaneous release activity. So, good chance sort of made up a code of of band positions going from one to six kind of arbitrary giving them a number. One is the wild type band and six hits, which has a stronger shift to the to a lower molecular weight, and she found that the band position was predominantly important for many frequency activity so a band position one gave you the lowest frequency the highest. So clearly, the post translation modification some is calling is calling very well to spontaneous release activity. So what, what's going on here. So, you know, a one obvious way of changing molecular weight is, for example, to do palmitolation of, of the protein and they are indeed in the transmembrane domain of the snare of the syntax and there are two systems, rather close to our previous lysine at position to 60 which is a very important controlling amino acid for molecular weight. And so what we did we individually mutated the systems and also and compare that the mutation of the lysine. And the result was that the white type band gradually decreased in amplitude from one mutation to double mutation and the double mutation had the similar bandwidth and then mutating this residue and doing the combination of the mutations did not further decrease the molecular weight. So this clearly indicated to us that the we have most likely modification based on the systems in the transmembrane domain that was responsible for the shift and molecular weight. We independently test this within a quite complicated biochemical experiment. Basically, you have the palmitolated residues through a steel ester, you first block all the three system groups in the protein, then you cleave the steel ester with a reducing agency. The SH group now is getting occupied by irreversible covalent bound of a biotin moiety. And then we can use this biotin moiety to be detected with strapped avidine in our western blot. And you can see here that the wild type protein has this band of palmitolated extra band that we see with strapped avidine. So this band of palmitolated extra acidity, but that's the best we could do. We're not biochemists, but we could repeat it and the signal was clear, but it's not super, super clean. And this band disappeared when we made either mutations in the lysine or when we mutated the cysteines in the transmembrane domain. So this band that we expected to see disappeared. You can also see here the shift into them in the molecular weight of syntax. So it's clear then. So from this we were convinced that palmitolation of these transmembrane domain residues are dependent on this basic residue. And so there are transfer races of palmitol groups that are sort of swimming in the membrane and they're sort of looking for motifs. One motif is of course the cysteine in the membrane, but they probably need also basic residues to actually dock on and then enable their enzymatic activity. So the question now is what does palmitolation does it directly? So we can test this by simply looking at the cysteine mutants themselves, not at this mutation and compare them to phenotypes. So again, making the cysteine mutants individually or double didn't really affect the evoked response very much. But again, we saw a drastic reduction in spontaneous release activity. So clearly palmitolation is required for maintaining high level of spontaneous release activity in these mammalian synapses. So it just brings me to the end of my first part and I will have a few minutes to finish up making this a bit more broader this finding. But essentially what we found what Gulchin found is that there are basic residues in the linker region and the linker region itself is very important for both regulating spontaneous release evoked release and release kinetics. And this residue is important for this palmitolation transferase enzyme to actually implement palmitolation fatty acid chains to the transmembrane domain of syntaxin. And having them present here affects the way how vesicles once they are primed can go infusion, they can be calcium triggered or they can be used spontaneously. And the energy barrier for going to these directions can be sort of dependent are dependent on this post translational modifications we see here. Okay, so, so that was one thing I said okay that's new model new mechanism how snare proteins are modulated by nature to have a specific function. But is this really something very specific for here or is it just a kind of a freakish thing maybe it's kind of a household function. So we noticed something looking at a different synapse system. So there are two major types of synapses in the brain. Basically releasing synapses and the chronically releasing synapse the basic release synapses the one that I told you about, they mainly respond to action potentials, the chronically releasing synapse. Essentially the release activity is dependent linearly dependent on the membrane potential of the cell. And this is something you find at the ribbon synapses in the both in the photoreceptors and the bipolar cells of the retina. So then you have essentially the member the photoreceptor measures membrane potential, and then various its release activity according to to the membrane potential so you have a light flash in the photoreceptors this causes a cell to hyper hyper polarized. And because when it's hyper polarized you have less calcium influx of calcium channel you have essentially no release. So in darkness these these ribbon synapses are highly active. Because the membrane itself is depolarized you see the membrane noise here because there's massive release activity happening chronically because the cell is chronically depolarized. So we have here two different synapses here. We have kind of a conveyor belt system where lots of vesicles kind of sort of fed onto the release sites. So these activities in this ribbon synapses, and they only stop when the cell hyper polarized is induced by light flash. And so these ribbon synapses actually use a different syntax and they use syntax and 3b, compared to the syntax and one that we find in basically releasing synapses which are sort of downstream of the photoreceptor activity. And of course, in the rest of most of the rest of the most of the brain. So, we said oh interesting so there are different syntaxes and if you look at the structure of syntax and one versus syntax and three it's isoform that is unique for synapses. At our famous site, we actually have a glutamic acid and not a lysine. And so the prediction would be that syntax and 3b is undergoing different changes in post-translational magnification and compared to syntax and one. So we thought let's look at syntax and 3b behavior in a basically releasing synapse the other way around we haven't done because we're not experts in retinal physiology. It's not trivial to measure from retina, you know light responses and so on you have to do it in the dark it's super complicated. So we stick with our simple cell culture model and studied what is syntax and 3b doing in a regular phasically releasing synapses so we made sure we can express this protein property that it enters the synapses just like the syntax and one. They go where they are supposed to go and we introduced also mutations syntax and 3b to make it like syntax and one a. Okay, so, so making the rule to make it back to lysine and so then we looked at synaptic transmission. And this is a why type evoke response for syntax and one. This is a why type response was in Texas reviews essentially gone, but if you just make the same single more point mutation at position boom, we get back synaptic transmission. And in the case of syntax and 3b, this residue is much, much more important than what we found in syntax and one is because changing this charge at least in a physics synapse is kind of a go no go in terms of release. Okay. So that was clear to us that there has been something going on in evolution, which means that syntax and one and 3b sort of diverted in their structure and then sort of mediating different types of release activity. We could also, we also took them the sister of syntax and 3b the syntax and 3a the splice variant of syntax and 3b. And in fact that splice variant is mainly in the C terminal part in this, for example, lacks the systems it's you see it here 3a has here two valence not two systems. And in fact it has at this position. It's a neutral amino acid and glutamine compared to the lysine and syntax and one. And so we gradually went back and sort of changed link region transmembrane region of syntax and 3a and thought how is it actually affecting protein structure and function. And once we put the linker region and the systems in the position, we got again a shift of the band size of syntax and 3b up indicating we reintroduce parmethylation this protein simply by putting these structures in and release was also very interesting. So syntax and three rescues, you see much smaller response and also the release is much slower. So syntax and three is not designed to be a very rapid releasing snare snare. So the fastest releasing snare is the syntax and one which you find in the basically releasing synopsis. And but we can we we make this synopsis to be reason efficiently and rapid by simply, you know, converting these region here to be syntax and one a like. Okay, so we can really play with this and understand that these difference, the syntax and three a by the way is post synaptically located. So in that post synapse of a neuron, you also have vesicle fusion with this vesicle fusion is actually important for long term potentiation. People have found this. And here we can say that whatever the syntax and three a property is, it is slightly different than what we know from syntax and one a that it is actually not very good in in phasic release. But what it does actually it shows much higher frequency of spontaneous release, compared to there's something in the syntax and three a structure that makes spontaneously is much, much higher than what we see in syntax and one a. And, and if you look at this, the the region this opens up a new air new new domain that makes us interested in spontaneous release which is this last half of the snare motif. Because if you convert syntax and three a to be like syntax and one a by making all these changes to the sea terminus, we have higher release activity compared to syntax and one a. And what is really different here is this region is is the internal part of the snare motif. So somehow in part of this snare motif, we have another region that controls spontaneous release activity. You see, you can almost endlessly study these molecules and find more and more kind of phenotypes, and that helps you ultimately understand how these snare machines actually operate, how they control spontaneous activity and and evoked release. And we even found something similar also for another isoform syntax and two, which is very, very cool to work with because it has even larger effect of spontaneous release. And again here we found that the sea terminal portion of the snare motif is critical in controlling spontaneous release because if you put simply the sea terminal portion of syntax and two into syntax and one, you get a massive increase in spontaneous release activity. And what is different between the syntax and one and syntax and two in this region are only a few amino acids. And you can just look at them they most of them conserved up but a few are different. And they look outside of the snare motif. The snare motif itself is highly charged on the surface, lots of positive and negative charges. Positive and positive negative charges means they are capable of interacting electrostatically with either membranes or proteins. And there's a famous protein that has been actually crystallized to work interact with this net complex which is again coming back to synaptotagmin that Yuki studied. And so synaptotagmin actually interacts exactly with the region that is different between syntax and one and syntax and two here. We know that synaptotagmin clamps blocks spontaneous release. So it's not only a calcium sensor for fast release, it also suppresses spontaneous release as kind of a dual function. So we now testing this hypothesis, we're just doing this now, whether this interaction that we think play a role in spontaneous release is actually mediated by interacting with synaptotagmin. And by, you know, we're in syntax and one you have these crucial to aspartic acid 231, which is mutated in syntax and two. And, you know, enabling the in syntax and one enabling the interaction syntax and one synaptotagmin, and therefore clamping spontaneous release activity. And yeah, so we are this seemed to be the case we have we make mutations in this to make it like syntax and to like and again boom you see the mini frequency goes up. And we are collaborating with people like Axel Brunge or Roseritzu which was really trying to understand this process better on kind of a molecular simulation and structural level. Okay, I'm coming to my last summary slide. Thank you for your patient I know lots of data. So, what I wanted to show you is that synaptic release has really multiple components, we see evoked release we have spontaneous release mostly the spontaneous release is undesirable because it creates noise. And but you can see that these kind of activity differs in different brain areas. Maybe in some areas, spontaneous release might be actually good thing because you don't need a physics nickel you you want to maybe have like a drippling continuous drippling of vesicle fusion happen. And so, you know, this may play a role for example and if you're looking at the eye or at the in the cortex. And these diversity and release contributes to sort of helps to shape how synapse is actually computed. Do we have a basic release activity, do we have high release probability or low release probability depression facilitation, or do we want to actually have chronically is that we need for example for ribbon synapse to convey to do its job. So, this is we know this is taking place and we really think that synapses alike in neurons, and they're firing synapse on speaking different dialects and that has some implication under the of the machinery that underlies these processes. And we know that the core proteins that are forming the release apparatus as well as accessory proteins like complexity knows not to take men are have evolved to adapt to these unique properties to make make let's say basic release very efficient or chronically release very efficient. And you can see that for example comparing different syntax and isoforms. And we think that overall understanding the mechanism, how, how this diversity of function takes place requires that we understand how these proteins also interact with other proteins. Other molecules such as phospholipids, how a fatty acid anchor that we found in syntax and one is sort of interacting with a membrane, or how calcium sensors like synaptic tag when are complex and can interact directly with this new complex. And those are the key modulators for this fusion apparatus to sort of do these jobs up here. Excuse me to the end. This is one, most of the work that I showed you was done by Gülçin Varda. She has Andrea has sort of taken over this project. And we have some older studies and syntax and John Chalra, Marie for until it did the titration experiment. Zina and Estelle are currently working on these interaction between the snack complex and is enough to tag men, and Yana has done this beautiful cry M data that I showed you. Before and we have some really key collaborators for this Erick Jorgensen introduced me to this high pressure fast, fresh and freeze technology. Jose Rizzo is a long time collaborator working particular questions of structure and function. And Tom Zutoff and these roles have been also long term collaborators working with genetic mouse models involving presynaptic proteins. And thank you for your patient. I know it was long, a lot of data, a lot of time and thank you.