 And I will introduce you more officially now. So Anna Erzberger is now, at least here, has started a lab of her own in Heidelberg in the Molecular Biology Laboratory. Sorry, I don't remember. E, in the abbreviation. European. European. It's not a German Institute. Very important, non-governmental. So it's in Heidelberg. No German bureaucracy for me. Yes, but it's European Molecular Biology Laboratory. Prior to this shift to Germany, Anna was in New York working with Jim Kutz, but in the Rockefeller University, the Laboratory of Sensory Neuroscience. All right, Anna, please continue from now. Yes, I'm sharing the screen now. Everybody can see and hear. Yes. All right. Yeah, so thank you, Roman. Thank you so much for the invitation. I'm very excited to tell you a little bit about some. So I'm going to talk mostly about work that I did in New York, together with Adrian Jakobo, who's also going to give a talk later. So we split up the contents a little bit. I did the theory and he did the experiments, so it's not going to be so difficult. And I'm also going to talk a little bit later about some future plans that I have in Heidelberg. And I'm going to hijack the last five minutes to shamelessly advertise all the open positions, because we're hiring. And I would be very, very interested in attracting more physicists and theorists to EMBL, because there is a bit of a push for that. And I want to have more of a community of people to talk to. So basically, this work, we were essentially focusing on the question of how the large-scale properties of a system emerge from the properties of its constituents, in particular, the broken symmetries at the level of its constituents. And so this is something that is a very fundamental concept in condensed matter physics. And we wanted to explore how you can apply some of these concepts to biological self-organization, and in particular, the patterning of an organ from the interactions of its cells. And so in the context of cells and the cell biology, there is various ways in which there are broken symmetries and in which cells manifest something that people call polarity. And typically, there is some kind of asymmetry in proteins or other or mechanical properties. And these can give rise to precisely regulated asymmetric cellular structures, but also it can manifest in more dynamic ways, such as directed cell movements or a particular selected axis for oriented cell divisions. And we were looking at polarity patterning in the context of mechanosensory epithelia, which is a very good system for studying this, and it attracted me right away that these epithelia have a very, very precise, for a biological system, they have a very, very precise organization. And you can see here an image of a mammalian cochlea, of a hearing organ, and the sensory hair cells are the ones that carry this V-shaped bundle. And you can see that the organelles are very, very precisely organized, and they are all aligned or all pointing in the same direction, and there is supporting cells that are surrounding each hair cell. And so we were interested in understanding how this architecture arises. And many features of mechanosensory epithelia are also very conserved, and they show up in mechanosensory organs across all sorts of animals and all sorts of organs. And here is a picture from a frog and a picture from a fish. And in all of these cases, you have these intricate hair bundles that are the mechanosensory organelles sitting on the sensory hair cells. And the sensory hair cells are surrounded by supporting cells which fulfill a role that is similar to the glial cells of the nervous system. Excuse me, Anna, can you with the mouse, please show what are the supporting cells and the hair bundles. Yes, so these guys here are the hair bundles. They're sitting on the surface of the hair cells. The supporting cells are covered in these kinds of micro-vely, so they're a bit harder to see in this image. But this is a side view of a neuromast, of a mechanosensory organ in a fish, and you can maybe see here, actually, the lab has done some even better quantifications where you can very precisely measure that. You can see that there is supporting cell around every hair cell. We are probably seeing only the first slide with the title. Oh, wait a second. You're not seeing the slides that I'm showing? No. I think you have to press full screen. Yeah, because I'm sorry. Maybe it's my tiling window manager. Bring your share. Wait a second. Let me just try again. I'm sorry about that. I've had many slides so far. Actually, we saw only the very first slide. Can you zoom on now? Yes, yes. I'm sorry, yeah. Thank you so much. I'm so sorry. I didn't realize. All right. Just quickly maybe review this previous slide. Very briefly. OK, so this was the first slide, which was about broken symmetry at the level of constituting parts. And then we get some macroscopic properties. Here we have a little steering texture example. And then in cells, we have polarized organelles, directed movements, and oriented cell divisions. This is the mechanosensory epithelium. The first one that I showed. And then here is some other examples. Yes, I'm sorry. Yes, just with the mouse, please show where the hair bundle is. These are the hair bundles. These are the mechanosensory organelles of the hair cells that are sitting on the top. So this is the top view. And this here is a side view. And the features that I wanted to highlight, first of all, is that the mechanosensory organelle itself has a very precise substructure. It has an asymmetry with the stereocilia forming this kind of staircase pattern here. And this asymmetry of the organelle is connected to its function because you get the mechanical stimulus that will deflect this bundle. And then you get a directionally, depending on the direction of deflection, you get an electrophysiological response from the cell. And in these epithelial, all of the hair cells, so we now think of these cells as cells that have a polarity, so they have a broken symmetry. And the polarity arrows all point in the same direction. So there is an alignment across the epithelium. That was one of the features that I wanted to highlight. The other feature is that it has this mosaic organization with each hair cell precisely surrounded by the lia-like supporting cells. If this is so far so good. And then there are some very interesting polarity patterns that have evolved as a consequence of this directional sensitivity of the system. Namely, in many mechanosensory epithelia, we have a mirror-symmetric organization of hair cells that point in one direction on one side, and others that point in the other direction on the other side. I'm going to stick with this color code for the rest of the talk now. So there's going to be yellow cells, so there's going to be yellow cells that are defined as negative relatives to the organ axis, or the plantar cell polarity axis. For those of you who are familiar with that, and the blue cells are the positive cells. And in fact, these patterns have evolved to generate organs that can respond very efficiently to stimuli that are mechanical stimuli that are delivered along a particular organ axis. And for example, in our inner ear, our vestibular or balance organs have this kind of pattern. And we were interested in understanding how do you make such a mirror-symmetric, such a beautiful mirror-symmetric polarity pattern in an epithelium. And we were investigating this question in the context of the lateral line organ, which is a mechanosensory organ of fish that they use to sense water currents. But the structure is very similar to the mechanosensory epithelia that we have that mediate our sense of hearing and balance. And they form these little autonomous organs that are arrayed along the side of the fish. And all the hair cells are aligned with the body axis, with the anatomical axis of the fish. And they respond to water currents preferentially occurring in this direction. And we have this beautiful mirror-symmetric organization with the yellow cells on one side and the blue cells on the other side. And what's interesting also is that these mechanosensory epithelia have a very, very great capacity to regenerate. So you can mess up, you can mess with them however you want, and they will grow back. And they will grow back from any kind of damage in a very, very robust way. They also turn over throughout the lifetime of the fish that are quite a dynamic organ. And this is in sharp contrast to our ears. Our ears are not that all able hair cells in mammals in general are not, they never come back if they get damaged. And as a consequence of this, we lose our sense of hearing as we age. And if we experience acoustic trauma, we cannot recover from that in general. And for me, it was an interesting question to understand how do you get this kind of, this kind of robust regenerative self-organizing capacity to create these complicated organs. Just to show you some nice pictures, this was also done in Jim Hotspat's lab. This is a serial block-face reconstruction of one of these neuromasts in 3D. And so you will see how we rotate to the surface, to the apical surface of the organ. And you see now the colors are red and blue. They always form these pairs that are opposites and they're facing the hair bond, the tall ends of the hair bond those are always adjacent to each other. And it is known actually in this system that the way the system makes the same number of positive and negative cells is by having a progenitor event generating two hair cells. So two daughter cells come from one progenitor and they will always form a positive and a negative cell and they will always form a positive and a negative cell in exactly this configuration with the tall ends of the hair bond that's adjacent to each other. So our first question was how do you, how is the initially, how is the symmetry broken between the two daughter cells? Because from what we know from past research is the division itself is symmetric. So initially the two cells are identical. And this is the part of the story that Adrian is also going to talk about some more. There's a biochemical symmetry breaking event that happens here and we got our hints from which pathways involved by looking at mutants. This is a mutant in which the notch 1A receptor is missing the notch pathway is a very important pathway in development of biology does all sorts of things. If you look at these mutants, you see that now suddenly you don't have a mirror symmetric organization anymore. You have only negative cells. And for us it was interesting to think that not signaling might be involved in breaking the symmetry between the two cells because it's a pathway that is typically wired up in such a way that you have a mutually inhibitory interaction between cells that are engaged in the signaling interaction. So here you have receptor ligand binding events. Both receptor and ligand are bound to the membrane. So you require, you need the direct cell-cell contact in order to have signaling. And then what happens is that there is a positive feedback. So that small differences in the expression levels get amplified over time very quickly. And so you have this kind of winner-take-sall scenario in this case. So it's a very plausible mechanism actually for generating positive and negative cells from an initially more or less identical situation in which you might have some stochastic difference that gets amplified. And I thought it was also nice in this situation because we just have two cells, we can actually write down equations for the signaling dynamics simply by considering a single variable. So we can reduce the complexity of the problem if you want. And we can introduce this idea of describing the pair with a signaling dipole where we have a positive cell on one side and a negative cell on the other side. And then the signaling interaction gives rise to the signaling dipole converging to minus one or plus one if we look at it along the axis of the organ, which we call X here. Actually I don't have the X on the slide. And then we can represent the system with this kind of double well potential. And the idea is that we describe the states of the pair with this dipole. We start out at this unstable position at the top of the hill. And then any small difference in the initial levels will make the pair converge either to this point or to this point. And we break the symmetry. We got a positive and a negative cell in the end. Excuse me Anna. Yes. Could you comment so, but in the polarity of the hair cells it's only head to tail. So it's the anterior posterior, right? So there is no polarity up, down dorsal ventral. So it's interesting. So some of the most of the, well, actually some of the neuromasts are polarized along the anterior posterior body axis, but some of the neuromasts are also polarized along the dorsal ventral body axis. And everything that I'm saying here relative to the organ axis is invariant is exactly the same in these two neuromasts. So you have a chosen organ axis. And then along this organ axis, you build this mirror symmetric pattern with positive and negative relative to that axis. So that it becomes a one-dimensional problem, right? Essentially, yes. But you'll see we have to think about the other dimensions in the end. But for now we can think about it as a one-dimensional problem. Thank you. So, oops. Sorry about that. Yes. And then we saw some additional experimental evidence also for the role of Notch and Adrian and Agnick, who is another postdoc that worked with us on this problem. They also did some experiments where they now overexpressed Notch. And then they saw that they could flip the sign of the polarity so we can generate neuromasts with only negative cells and neuromasts with only positive cells. We can even tune these numbers. He's going to talk about this a lot more. So we were quite confident in the end that we had identified the molecular pathway that is involved in breaking the symmetry and creating these two different polarity fates. Excuse me. Yes. Sorry, again. So in this left picture, I see that all the cells really have the orange polarity. But in the right picture, we see that some cells still have the orange polarity. That's true. Yes. So I can tell you what the deal is with this. And it's nice we can actually verify. I left it out here because I didn't want to go into so much detail. The overexpression is done in a slightly more sideway. So not all the hair cells have the construct that overexpresses notch. And we actually have a way of tagging that. And the guys were staining the picture such that in the end, we actually see the cells that have the overexpression. They're pink. And the ones that don't have the overexpression, they are not pink. And so you have to believe me because I didn't put the picture here. But in this image, we take a confocal stack and we can go down in the confocal stack to the nuclei. And then we can see which nuclei are pink. And this guy, for example, is not pink. But the ones that are overexpressed have the overexpression they are pink. I see. So you just cannot really overexpress it in all the cells, right? Yeah, it's a funny thing. Initially, we had some fish that were overexpressing it and all, and then it somehow became mosaic. And it got more and more mosaic, actually. So it looks like it's actually behaviorally problematic for them to have these uniformly polarized neuromasts and maybe larvae that are very bad. They don't make it so that when you maintain this line, you're actually losing the phenotype. But yeah, this is something to ask Adrian also because he knows a bit more about this stuff. Thank you. You're welcome. Yeah, so anyway, basically, we can explain how we get positive and negative cells. But now, if we think about the whole spatial organization of the organ, in particular, this mirror-symmetric pattern that we get, we immediately have a problem, right? Because with this theory, with this idea, we should see positive and negative dipoles. And actually, unless there is something to bias the process, we should see those with equal probability. And nonetheless, what we see in the actual neuromasts, we never see negative dipoles. Once the bundles are formed, they're always facing each other. And there's really this never happens. So it's a process that is extremely robust. I'm impressed because, typically, in biology, things are not this clear cut. But in this case, we monitored hundreds and hundreds and hundreds of pairs. And we never see the system make a mistake. So something is preventing the state from manifesting. And we wanted to understand what's going on. We're clearly missing something. So we wanted to understand what's going on. And Adrian developed a lot of really great tools for imaging the neuromasts and for observing really what happens from the division of the progenitor all the way until they mature. And I'm just going to show you one example. So here, you will see a movie in which we will see one of these neuromasts, one of these mechanosensory epithelia, grow and develop over time. And then there is going to be a number of new pairs that appear, a number of new pairs of hair cells. So the hair cells are the only cells that are labeled in this video. Everything that is black is actually supporting cells, and they are not labeled. And the label is expressed under a hair cell specific promoter. So in fact, the new pairs that show up, they will become brighter and brighter. As time goes on, here is a pair that's relatively young. And there's going to be another one that shows up here. And then there's going to be another one that shows up here. And this is the organ polarity axis, the PCP axis. All right. So here, sorry about that, here is the next pair. And here is the next pair. You see how it becomes brighter and brighter and brighter. This very bright stuff in the middle, those are the hair bundles. They're very bright. These are the hair bundles of this pair. And these are the hair bundles of this pair up here. One thing that we immediately notice when we look at these movies is that the cells actually have a protrusive behavior. They form these active field protrusions. And I will show you again, these guys see that. This is a protrusion. This is a protrusion. Here, these guys also form protrusions. And I worked on cell migration in the past. So for me, it was very interesting to see these protrusions. And they made me think that maybe there is an additional broken symmetry in the system here. Maybe the cells have a particular axis along which they are organizing their cytoskeleton along which they are forming these protrusions. And possibly, this axis is already related to the polarity that is associated with the orientation of the hair bundle. And indeed, when we look at normal wire type situation, this again is a neuromask. There's a new pair here. Actually, this is also a movie I will show you. These guys, again, form protrusions. In a wire type normal situation, the protrusions always form on the opposite poles of a pair. And so if we look at the angles of the hair bundles and we compare the angles, orientation of the hair bundle, the orientation of protrusions, we see that there is a clear correlation. Now, if we go to our mutants, to the notch mutants, which have uniformly polarized pairs, what we see is that indeed the protrusions are formed in the same direction. There's this pair down here and then there's this pair up here. It's not so clear, but they both go in the same way. And in fact, the pair also moves a little bit in the same direction. So when we quantify this, we again see that we have uniform bundle polarity and we have uniform protrusion polarity as well. Yeah, I'm not showing those data here, but we also quantified actually the position and we see very clearly that in the wire type, the two cells move away from each other. In the mutant, the two cells move together and they move to the side in which they form active protrusions. So that suggested that there is an additional, there is a mechanical symmetry breaking that is happening within the cells. So there is a preferred axis along which the cells are organizing their cytoskeleton. And I wanted to understand this a little bit better. And so I found it useful to think, to introduce a generic surface mechanical description of the pair where we introduce this effective cellular surface tension, which is defined simply as an effective energetic cost per unit area that is associated with the different interfaces that these cells form. And so we have three different types of interfaces in for each cell, double it if you want, we have an interface between the cells and then we have an interface of each of the cell with its environment. And I'm saying generic surface tension because we, at this point, we're not distinguishing between the different kinds of molecular sources of this kind of effective surface energy. So this kind of description includes both effects of adhesion, but also effects of the cytoskeleton that is at the surface of the cell that might generate contractility or even act in remodeling that will change the shape of the cell. So all we're saying here is that this is a generic variable that is basically dual to the surface area. And with this description, we can now ask what happens if we introduce a broken symmetry in the surface tension and we can basically simply put in the axis of the organ and explicitly say that now we are going to ask what happens if the surface tension varies along that axis. And I was just interested in very simple, very weak linear gradients, but we can do that and then we can assemble a state diagram that describes the different behaviors of the cell pair. And we basically recapitulate that there's a, so here the state diagram basically has the baseline of the surface tension and the magnitude of the gradient. And we see that there's a line at which is something like an effective wetting transition if you want, if you cross that line, you have regions on the surface of the cell which cross threshold from the region where the surface is trying to basically minimize where the interface is minimized between the hair cells and the supporting cells. Instead now you want the area to be large and so that qualitatively accounts for the appearance of these protrusions. And we also see recapitulate that the pair that the cells detach from one another at a certain point and this interface between them should disappear. But probably the most interesting part is that in this region of the diagram and the yellow region of the diagram, we would expect some interesting behaviors. Namely, what we should see is that we can start out with pairs that are oriented either way with the yellow cell and the blue cell, basically with the blue cell being on the wrong side. And then what we should see is that the cells should self-correct for that and that we should see rotations of the interface essentially driven by these gradients in interfacial tension. And we think that this is basically the answer to the puzzle that I posed earlier and we indeed see this behavior. So now I'm going to show you another movie. This is a progenitor. So you will see this progenitor divide. You will see the interface between the two cells. And then you'll see that this interface rotates here and then after the rotation, the cells form protrusions and they move away from one another. And so we tracked many, many cells and we saw that the behavior was very sharp and very binary. In about 50% of cases, we see a 180 degree rotation of the interface and in the remaining cases, we see no rotation of the interface. And this in fact is consistent with the idea that we have an initial process which is unbiased, which gives us positive and negative that polar configurations with equal probability and then half of them, the ones that are in the wrong configuration, they flip position and they rotate their interfaces. And just to give another sort of mechanical intuition, the idea is that we start out, we can start out either in a situation where the cells are pushing against each other and this is an unstable situation the interface rotates or we start out in a situation where the cells are pulling on each other and then the interface does not rotate, nothing happens. In fact, they move away from one another at a certain point. And we can use this surface mechanical description to write down the surface potential. We see it's this cosine shaped thing for a linear gradient and we have an unstable state up here and the cells transition into the stable state at the bottom of the well. And from this, we can also derive an expression for the angular velocity and we can compare that to the data. And so this is a one parameter fit. So we're pretty good at recovering exactly the kinetics of this interfacial rotation, I would say. And then of course you can make a prediction what should happen in the mutant where we have uniformly polarized pairs and of course we would expect that there should be no rotation in that case. And so in fact, I was quite nervous when Adrian was giving me the data finally because whoever here does theory in collaboration with experiments knows that it's very easy to come up with all sorts of predictions but then you don't know if what actually happens in the system. And so I was quite upset because I watched four or five movies and then the fifth movie was doing this. See, this is the pair of cells and very clearly they are flipping position. And I was complaining to Adrian about this over lunch and then he told me in fact that this mutant that he used for imaging, every once in a while in this mutant there are oppositely polarized pairs. So not all the cells are pointing in the same direction every once in a while you get oppositely polarized pairs. And so when we went back and checked we found that this guy actually those cells later form oppositely polarized hair bundles. And then I thought it was a nice result because it shows that what we see if we pay attention to the polarity of the pairs is that in the uniformly polarized pairs never rotate but we also see that it's not due to a systemic disruption that is a consequence of the mutation. It's really due to the fact that you don't have the opposite polarity because you can have rotations with oppositely polarized pairs in the background as well. And we see that sometimes. Yeah, and so finally I was very happy because in a way these two different symmetry breaking events if you think about it now in 3D if you now think about the whole spatial picture then they unify in a very nice way. And I want to credit Edgar here because he came and we were discussing at Rockefeller and we were talking about promise transitions and biology and it was a very nice discussion. And somehow after that I had this idea that I should think of it in 3D and it was a very satisfying realization that you get sometimes with these things. So thanks for the inspiration. Basically we can now if we now take the double potential that I showed in the beginning. Now we think about it in 3D but we say that the cells have no bias. They don't know anything about which direction they're facing. All they know is they talk to each other and they break the symmetry. Then we can represent our signaling potential in 3D and it should look like a goldstone type potential with a ring of degenerate states at the bottom. And however, in this case what the symmetry breaking actually does what the positive or negative notch fate means is that the cells should pick an orientation for their polarity relative to a global organ axis. And so they start polarizing and they polarize in a mechanical manner in which they start basically exerting oriented forces. So this process is driven by is an active process and we can represent it as a tilting of this potential. It's just a linear tilt if we consider linear gradients. And so what this does is now instead of having a ring of degenerate states we have one unique state and all the cells go to this state. And if you parameterize this ring in terms of the angle then we recover exactly this cosine shape potential that I showed you earlier. And in the real system cells are typically they divide already oriented along the axis of the organ. So most of the cells are either here or here. So of course in reality there is other factors that result in that. We don't have to think about it right now but they are there. And so what happens is that the population that is up here when we tilt they are the ones that rotate. They are the ones that go down here. But every once in a while the divisions are off axis and they are not actually aligned with the axis of the organ. And what we can predict from this is that if that happens, it doesn't matter. They can be oriented however they want in the beginning after we tilt they also go to this position. And this we see. So every once in a while you have an off axis division. I'm sorry these pictures are not that great. It's always a bit tricky because the young cells are the ones that are the faintest. And so you get these super bright parts from the mature cells nearby but maybe you can see a little bit this is a cell division and this is the axis of the organ. And so this division is almost perpendicular to the axis of the organ. And then what happens is that the interface rotates exactly the right amount to allow for the pair in the end to be perfectly aligned with everybody else. And so that essentially corresponds to this. We start with this state here and then we go down only this much. So we see this sometimes. And yeah, I'm basically to wrap up the ideas that we had these two different symmetry breaking events that interact with each other. The first one is biochemical is driven by not signaling the second one is mechanical. And altogether we get this organ scale symmetry that emerges from these cell level interactions. And here are the literature references if you care to check it out. I, we can take a little, I don't know how we're doing time wise. I have a few more slides on some future stuff we want to take this next. I can also answer some questions now about the project. Yes, okay. However you prefer, I don't know, yeah. Yes, we have still a long time to go. Otherwise I just continue. Yeah, but if someone asks questions. Yeah, maybe, let me ask a couple of questions I had on the first part that is more than half of this. So can you show again the plot of the phase as far as of time? And one goes to, okay. Here. Previous one, previous one. It's a previous one. It's the time traces of the angle when they're here. So my question is the following. So you go from zero to pi, but it looks like it's going only in one direction. So you could go zero to pi in two ways, no? Yeah, yeah. So here I'm plotting the rotated angle precisely because of this. So maybe a related question is whether or not there is a chirality in the rotation. Exactly, that was a good. The truth is that we have never rigorously addressed this because in order to address it, you have to keep very careful track of how you mount the fish and so on. And so we had accumulated a huge amount of data where we weren't exactly sure what the mounting was. So there is a confounding factor, but with the data that we have, we see no indication of there being a preferred way for the cells to rotate. So in fact, if I would plot not the rotated angle, but the absolute angle, relative to the actual anatomical axis, I would see one population go up or one population go down and there wouldn't be a difference. Here I'm just plotting the rotated angle, which is defined as relative to the, yeah, just the angle that is growing over time to get around this issue. Okay. And about the model, have you thought about adding fluctuations to this model because it looks so deterministic model? Yeah, I mean, you need, in a sense, you need fluctuations to get out of your unstable states. But I mean, we don't have it explicitly. So to be clear, the picture is that you have a double well and just to put the cell initially here and then you fold somewhere in the minima. But there could be another picture that is that you are in a minimum of a potential and then the potential is raising a barrier. Yes. You mean initially about the notch signaling part? Yeah. Yeah, yeah, yeah. So that would be the spontaneous symmetry breaking classical spontaneous symmetry breaking picture. So here, I think because, yeah, how to say, so the properties or the wiring of the signaling circuit, which is ultimately what gives rise to the bistability, what gives rise to the shape of the potential is, I would say is there already, right? So, yeah, in a way, what happens is that you have a fixed landscape and then you move around the cell pair, it moves around in it. And I'm not, yeah, I'm wondering how you would have a scenario where you have a, where you have a, yeah, maybe if you would start, if you would have a feedback where you, basically we think that the cells are already expressing, not maybe that's the question is, are they already expressed much at the time when they start interacting or is that something that gets amplified as well? Yeah. Actually, I'm gonna talk about that a little bit. Okay. That touches upon that a little bit, yeah. Thank you. Thank you. Yeah, okay, exactly. So, I just want to now give a bit of an overview of what we're planning on doing in the future and also what I have in mind a little bit for my group in Heidelberg. And I want to continue some of the work on the Notch pathway. In general, I'm interested in the aspect that is quite interesting about this pathway is that because we have this requirement that the cells need to be in direct contact with each other, there is this fundamental constraint that gets introduced which couples the morphology and the geometry of the cells, the geometry of the interfaces to the signaling. And so we have this natural way in which we get mechanochemical feedback effect in the context of Notch signaling. Because a lot of the time, what Notch will do is it will change the cell fate and it will change the morphology of the cells, it will change the mechanics of the cells and that will feedback by reorganizing the contact interfaces that will feedback onto the signaling itself by determining who can talk to whom. And this ties into a bit of a general question that I kind of always run into when I read biology literature. I don't know how you guys feel about it, but I often get the sense that there is some crazy molecular multitasking going on in the sense that there is not that many signaling pathways actually. And somehow it's a handful of signaling pathways that are doing everything in a developing organism, which is a lot. And you have studies where people show this and that and it's mediated by this pathway. And then other people show something else and it's also mediated by the same pathway in the same cells at the same time. And so there has to be mechanisms, very robust mechanisms of coordination. And essentially I would like to address very generally what role mechanochemical feedback effects can play in this. And just to stick with the example of the lateral of the neuromasts and the mechanosensory organs, there is a very concrete example of this. Namely, I basically showed you our results, which essentially show that notch signaling is establishing positive and negative polarity fades. But it has been known for a long time that notch signaling also decides sensory versus nonsensory fades. So I showed you in the beginning, this architecture of the sensory cells being surrounded by supporting cells. And this differentiation, the sensory differentiation event is also thought to be mediated by notch signaling in all different mechanosensory epithelia. So in the neuromast at least where you have a very dynamic organ and new cells come all the time, somehow these two signaling processes, some of these two differentiation events are happening at the same time. And we want to see if this can be explained by thinking about the reorganization of the cell-cell interfaces, which constrains who can signal with whom. And in fact, we think that what's going on here is a self-induced, self-regulated reorganization of these interfaces that coordinates these two different differentiation events that are mediated by notch. And here is my little schematic of this mechanochemical feedback loop. And what's very nice is that in our system, we have very nice imaging data from which we can make some inferences on the cell mechanics. And we can, for example, measure the contact angle between the two hair cells. And we can relate this contact angle to the ratio of the effective surface tensions, the same kinds of surface tensions that I introduced earlier. And we can actually do that for our maturing hair-cell pairs. And what we see is that, actually, they follow a very stereotypical dynamics over the course of differentiation. So these cells, as they undergo also this rotation that I showed earlier, and initially they have this contact, and then this contact disappears. And over the course of this time, they are ramping up their hair-cell specific genes and they take on the morphology of a hair cell, which is a process that was presumably triggered before the division of this pair. And we can recover this contact angle dynamics or surface tension, we convert the contact angle into surface tension ratio in this case. And we can recover it with a very, very simple first order linear decrease in the surface tension that is associated with the interface between hair cells and supporting cells with the heterotypical contact. And so what that suggests, basically this is the sort of first order behavior that you would expect if there is something that the hair cells express under their own specific gene expression program that modifies the heterotypic interaction. So this could be some change in the adhesive properties, some kind of heterotypic interaction here. And we don't have to go, this is, again, this is the first parameter fit. So we don't have to postulate anything more complicated in order to recognize, to recapitulate this contact dynamics here. But what that suggests is that indeed we have these two different differentiation events which trigger morphological reorganization, changes in the mechanics, and those in turn self-regulate, they terminate the signaling interaction at the right time, they remove this interface. And once this interface is removed, cells are surrounded by supporting cells and they can engage again in the second modality of not signaling, the one that drives sensory differentiation. And so I'm interested in exploring this idea and within a more generic theoretical framework, exploring what kinds of self-organizing properties you can get from these types of mechanochemical feedbacks and something that inspired this quite a bit also is a very nice recent work where people have made now engineered systems in which there is basically, there is synthetic notch receptors where you can put whatever you want downstream of the notch pathway. And so for example, in this study, they messed with this quite a bit and they put all sorts of adhesion molecules downstream of the notch signal and they see that you can start with a homogeneous population pile of cells and you get the self-organized formation of these layered structures but also these symmetry breaking in different aggregates. And we can, I think this is the way to think about it. And we also want to think about more interesting dynamic scenarios where the time scales of these different elements are now on a comparable scale so that you can get oscillations and other kinds of interesting behaviors. And my group is going to generally study different contexts in which mechanochemical feedbacks play a role and help acquire shapes and self-organize and just to give a few examples, one of the projects that we'll pursue will involve looking at nuclear mechanics and the interactions between cell shape and force on the nucleus. And this is also inspired by quite a bit of recent work that suggests that nuclei can function as universal mechanosensors and that nuclear mechanical state nuclear deformation can be directly linked to cell fate such that you can envision quite interesting situations where you now connect this to larger scale changes in tissue curvature as well. And you can think of curvature sensing mechanisms in which nuclear relocalization mechanical processes will trigger fate changes that then feedback onto these morphogenetic programs and enhance them or suppress them. And so we'll be exploring this in collaboration with the leptin group and look at it in the context of gastrulation of drosophilic restoration. And then another project that we'll pursue is the looking at skin patterning and the formation of these beautiful aggregates here. And this is a collaboration with Amy Shire at Rockefeller. And here we have some interesting ideas about how interactions between the ECM and cell polarity can generically give rise to contractile instabilities that we can describe in these phenomenological hydrodynamic terms. And it helps us basically delineate the regions in which you get different patterns. So yeah, these are just kind of little vignettes and essentially I'm just going to use the last couple of slides to advertise the positions. So I'm gonna be looking for group members at various levels. There's a staff position which is a more senior long-term position. There's various funding opportunities for postdocs as well. And I'm also gonna be looking for PhD students. And EMB has a really fantastic international PhD program which is open to people from all over the world. I should also say in general EMBL is a non-governmental organization. So it's not, you don't have to get a visa or anything like that. It's open to the world. And the program is more known in the biology community. So I'm gonna be doing quite some work to attract physicists and theorists as well. And I already mentioned that EMBL is also going to make a big effort to attract more theorists. And there's going to be various components of that. And they already launched a sabbatical program which is also open to all career stages. So if any of you are looking to establish collaborations with experimentalists, anybody feel free to get in touch and ask me about these things. I also wanna mention the interdisciplinary postdoc program which is also very ideal for candidates who want to combine experiments in theory or want to collaborate with experimentalists. So yeah, and with that, I'm going to just thank Adrian who's been a fantastic, fantastic collaborator and very sad to leave the lab actually because we had a great time there. Agnico also contributed critical molecular biology expertise and Jim and the funding. And of course, you for your attention. Thank you Anna. Thank you. So we solicit some questions from the public. I mean, we already covered the science part, I guess, to some extent. Yeah, well, all the questions you can ask. All sorts of questions. So I don't see any... Hello? Hello? Hello, I have a question. At one time there was an image of the experimentalists that were both configurations of the hair cells. There was a configuration that appears that this is not a bay. I think Roman asks the question, why was that? If you can go back. Probably it was the question where I asked about why not all the mutant cells are there. Exactly that. And you repeat the... Yeah, this one. Yes. Yeah, that, exactly. It's just, I wanted to hear the game because I don't quite understand it at the time. Yeah, okay. Anna, could you please... Yes, yes, yes. Maybe in different words or what? Yes, yes, yes. So I'm not a molecular biologist. So I'm just going to say the way I understand the experiment and then you'll get a chance to talk to the experimentalists themselves. So just a bit of disclaimer. My understanding is that the way this experiment works is that what we're over-expressing here is the intracellular domain of the notch receptor. And basically what happens in notch signaling is you have interactions between receptors and niggans. And when they bind, the intracellular part of the notch receptor gets cleaved, gets cut off. And that intracellular part then goes into the nucleus of the cell and there it binds to DNA and changes gene expression and so on. And so what we're doing here is we're specifically over-expressing this intracellular domain, the one that should change what the cells are doing that should trigger positive or negative. In this case, we've shown that it triggers negative polarity fate. And what we're doing here is we're tagging this notch intracellular domain with a tag which can then look at. And so we can't look at this in these videos that I showed you in live because we have to stain them in order to see where this tag is. But so these images here, for example, in this image here, it's a staining and so we can look at both the orientation of the bundles and also at the nuclei. And so we can look at whether or not this tag is present in the nucleus. And the ones that have the extra, the tag, they have the extra notch and they are pink. So this is what I was saying. And what we find is that not all the cells have the extra notch. Why that is, I don't actually understand completely. Yeah, actually I don't understand that. I don't know why not all of them have it. It's a mutant and I think somehow it gets silenced in some of the cells, but this I have always found that a little bit puzzling. But the bottom line is that we have a mixture of cells that have extra notch and ones that don't have extra notch and we can double check and make sure that we know which one is which. And I haven't added all of that material here to the slide, but in this particular case, we know that this guy, for example, does not have the extra notch. Yes, if I don't mistake, well, in general, right, if we, so basically we do when these mutants are prepared, we do mess up with the genes, right? And sometimes this mechanism which is messed up leads to deterministic to always to a given expression, right? And another mechanism which we can change may not be the only factor which leads to this accumulation of notch, right? Yeah, yeah, yeah. I mean, the other thing that I was saying is that this is a line which has this mutation and so it's maintained, we're maintaining it. So in Jim's app, there's a fish facility and so the fish are swimming around in their tanks. And so the line is maintained by keeping the babies and raising them so that the line is always there. And what we've observed over the time that I was there is that each generation had less of the pink stuff and each generation had more opposite, like that the silencing was maybe selected for. And this is what I was suggesting is that possibly the fish that have only cells pointing in one direction, maybe they're not very good at foraging, maybe they're not very good at eating and maybe the larvae, you have a batch of larvae and some of them are better at silencing, some of them are not as good at silencing, the ones that are better at silencing are going to make it to adults with a higher probability. And that's why we see more, more cytosity in every generation. Yes, yeah, thank you. Hopefully this answers the questions, the question of Alfaro. Gavin, sorry, Gravin. All right, do we have other questions for Anna? Well, Anna, I find that no one has any food questions. Okay, yeah, then I thank you, Anna. Thank you, guys. Very nice talk, we have an interesting talk.