 The next speaker is Pavel Janosz. Pavel, are you here? Yes, so you have to share. Yes, let's see. Yeah, it looks like it's working. Okay, perfect. So greetings everybody, my name is Pavel Janosz. Sorry, Pavel Janosz, so you have about 25 minutes. Yeah, I will try to keep the time limit. Okay, so greetings, my name is Pavel Janosz. I'm a postdoc in the group of Alessandro Magistrato here in Sierra Eome within CSI. And my talk would be, let's say, application or talk, you will not hear anything really interesting or useful. Pavel, sorry, your voice sounds a little bit like, maybe you can put the mic closer, I don't know. No, maybe far away, I think. It looks like you are far away, yes. No, I think the opposite, no, it's just, it's, try maybe. Is this good or not the idea? Okay, maybe. I'm limited by the lack of microphones, so I don't know what to. Okay, go ahead, maybe it's okay. Okay, go ahead, solve it. Okay, so I will talk about one of our application projects that we did in the group regarding NKCC-1 ion co-transporter. And I put a bit of a spoiler in the subtitle in the four ions on the channel. So let's go. To introduce, the NKCC stands for sodium potassium chloride co-transporter. It is a membrane transporter that transports chloride, sodium and potassium ions across cellular membranes, and it does so in the 211 stoichiometry. So it transforms two chlorides, one sodium and one potassium. It is schematically shown here. It is a very, very important transporter, it plays a role with various biological processes that regulates the homeostasis of these ions, which then is linked to the cell room regulation, and so on and so on. And we, in mammalian organism, at least, recognize two isoforms, the NKCC-1, which we are more interested in, which is ubiquitously expressed in pretty much all tissue, and find in bosalt-round membranes for gowns, in inner ear, or in neurons. And then there is the NKCC-2 isoform, which is specific for kidney issue, which on the right I'm showing the schematic representation of nephron, which is the building block of a kidney. It is where the NKCC-2 is positioned, and there it is responsible for salt reabsorption from urine. And as you can imagine, this transporter has been implicated in variety of human diseases from hypertension and renal disorders, which are linked to the like salt concentrations in your blood and urine, to stuff like death less, no robotic pain, those neurological disorders, and also various cancers. And it is the last part that was interested us in this project, because we are part of a larger project called ARIES that's targeting brain cancer, and experimental collaborators in this project alerted us to the role of this transporter in glioblastoma, which is the most common malignant brain tumor, and the treatment for these brain tumors is rather limited in that the surgical removal is not enough, and then the chemotherapy is associated with various side effects and efficiency issues. And it was shown that the NKCC-1 is linked to the tumor progression in that they see an upregulated NKCC-1 activity is then linked to the cell migration and resistance to a cell death. And some preliminary studies have shown that the inhibiting the NKCC-1 in conjunction with applying more traditional chemotherapy is shown to reduce the growth of the cancer. The problem here is that the existing inhibitors of NKCC-1 are very, very poorly traveled across the blood-brain barrier. And then you are introducing very severe side effects because we are targeting to transport in all the tissues, and you are causing issues. So that is a ruling for development on better inhibitors. It would be better targeted for the brain cancer, and that's what we wanted to do. And as with all projects that we start in the group, we first look at the mechanistic study of what we are looking before we start actually doing potentially drug discovery kind of studies. So you can answer also some biological questions and provide some complementary information to existing experimental data. So one thing that I want to first clarify is that I was using the term transporter. And there are these two differentiated from channels. So channels are also transformed by proteins that can facilitate transport of substrate across the membrane. They do it in a passive 3-year manner, just adding an endochemic gradient suitable for the substrate. So these are essentially static structures or as static as these large biomarker systems can be. And then transporter on the other hand transports the substrate across the membrane. Bit-associated conformation change. So typically we would recognize three states of these transporters. An outward open state where the substrate is where the substrate can attach to the transporter from the external space. When there is a qubit state when the transporter cannot see any more substrate. And then inward open state where the substrate is released into the intercellular region. And we are limited by the available experimental structures with transporters. And in our case, as is in case for most transporters, we really only have the inward open state. So we are limited to studying at this point just the release of the ions into the intercellular space. But that still leaves us with plenty of questions that we can answer. First of which is in what order do these ions exit the transporter. We are talking about four ions. They cannot all move at the same time. So is there an order in the diffusion? And if so, what is the order? And the question is also do they use the same channel? They have ions with the opposite charge. So it may be possible that different channels. In fact, when the first structure was solved, they identified three possible exit channels. And this is something we can also answer from simulations. And then of course, we want to know the mechanism of the individual ion transporters and what are the key residues involved, which against our information that can drive us in the further design part of the study. So here I'm showing the full structure of the NKCC one transporter, which exists as a homodimer with each one of them having the transparent brain domain, and then a rather large C-terminal domain, which are in the domain's web configuration. We saw that they are in the course from each other. But for our purpose, for the study, we built a more simple, smaller model where we only took the transparent brain domain of one of the monomer units. We of course ran a benchmark simulation of course of the full model to confirm that our smaller model is stable. It is. And so we move on with it. Here I was also showing the schematic of the transporter with highlighted inner cavity of the transporter with the ions in their approximate orientation. And because we are dealing with two high ions, to distinguish them, I'm labeling one as top chloride, TCL, and one as bottom chloride, BCL. And I will use this schematic representation of the inner cavity throughout the slides. So this is the close-up of the iron binding sites. I will not do too much in the video, but I will just highlight that the chlorides are bound by the amino group coming from the backbone of the protein. And the positive ions, the sodium and potassium, are bound by the carboxyl oxygen coming from the backbone, as well as oxygens coming from various side chains. Now for the computational details, again, nothing super exciting. We are using molecular dynamics and metagenomics, which are here, the Warhawks workhorse methods in our field. Here are the first few details, which if somebody is interested, they are there. And our system, even though we use the smaller model, is still about 140,000 atoms. That includes the transmine domain, the membrane itself, and the full expected solvent. For the algorithmic study of these processes, we use two set of collective variables. First is just simple distance collective variables that we are monitoring the distance of the ion to its respective binding site. And then we are using what's called volume-based TVs or volume-based metagenomics, where you take the center of mass of your protein, and then you calculate the spherical coordinates of your ligand with respect to the center, and you use these spherical coordinates to sample the unbinding. This was method that was proposed specifically to study ligand binding and unbinding processes. The only potentially issue that we are facing is that we are working with three dimensional CV space. So for the analysis and visualization, we are elevating the free energy surface onto just a 2D system, and we take the row, which is the distance from the center, as one CV, and then we use a coordination number of formalism to represent the number of contacts the ion makes in the protein. So this is our CV2. And first, the initial simulation that we did was very simple, and actually I called it here a quick and dirty, and the idea was here to figure out the order of the dissociation. So we did it in a very simple way, and then we took the transporter with all four ions bound, and then we were using metagenomics to buy us all four of the ions using the distance CV, and seeing which ion dissociates first. We are not really looking here for a nice converged free energy surface information, but we're looking at just quite actively which ion dissociates first, and to make sure that it's not just random chance, we repeated all these three times. And from this, we have our answer to our first question, it is the order. So we see that, and started with all four ions, the first ion to dissociate is the bottom chloride in the pink, then if you move to the transporter with the three remaining ions, the sodium potassium and the top chloride, we see that the potassium is the next one dissociate. But then we come to the last one, the sodium and chloride, the top chloride, and we see that now the situation is less simple. But we see that now it's not clear which one is first. In fact, they look like their movement is interconnected, which makes it a bit more difficult, but we'll come into it later. So now that we have the information of the order, we can immediately look into the dissociation of the individual ions. And for this, we use the volume based CV. So here we are looking at the dissociation of the bottom chloride from the transporter in their cavity. And here is the free energy surface with the CV one is the distance from the center and then CV two essentially number of contacts of the ion with the problem. And you can follow up all this process. And we see that at the beginning, the chloride starts to lose its coordination with the original binding site. So it loses the protein contacts and becomes more solvent exposed. Here, the water molecules are represented only by the oxygen atoms as these are at squares. So the chloride becomes more solvent exposed, which is stabilized by this potassium in the bottom. And then it's from the inner cavity outside and dissociation. I want to go too much into the designs, but I want to highlight the role of asparagine specifically the asparagine 220, 298, and 427 specifically because we will see them later on. And we also see that the dissociation free energy of this ion is very relatively low, which makes sense that it will be the first one to dissociate. So now we move to the next step in the transfer process, which is the potassium dissociation. And again, the process starts with the potassium losing its coordination with the original binding site, but still remaining in the inner cavity and becoming slightly more solvent exposed. This process is associated with relatively low energy, but then it has to leave the inner cavity and move to the exit part at the channel, which is then associated with relatively higher energy biome, so it has to escape. And here I will again highlight the same asparagines that we saw before, before they were involved in the chloride dissociation. Here they are helping the potassium dissociation. Then the next step, the potassium sketched by this glutamic acid, then this is a relatively strong interaction with the negative protein residue and the positive ions. So this is associated with relatively high dissociation free energy. Then it dissociates to the exit channel. And overall, we see that the barrier here is higher than the bottom chloride association. And now we finally move to the last two ions. As I said, the situation is a bit more complicated. So we are talking about dissociation of the top chloride and the sodium. And because we saw previously that these are somewhat interlinked, we cannot really just sample one of them at the time. So we have to sample both of them at one simulation, which again, we can't really use the volume base because there will be too much CVs for us to handle. So we are back to using that just distance, simple distance CVs. So for on X axis is the distance of the top chloride to its binding site. And on Y axis is the distance of the sodium to its binding site. And the reagent surface is kind of complex as we can identify three possible paths which are schematically shown here. So in the first path, the sodium doesn't move. It remains in this binding site. It is just the top chloride that moves and dissociates on its own. This is associated with rather large free energy volume. And the second path is more interesting because we see first the movement of sodium towards a new stable, metastable minimum. And then from this, the chloride can move. Which is this, let's say, branching point from which, again, the path split into the path two, which just the chloride associates or path three where the chloride and the sodium dissociates together. I will focus on more details on this first path because for us it's the most probable one, one more interesting one. So as I said at the beginning, the sodium moves from its binding sites to a new stable minima. And this is interesting because this represents the potassium binding site. The potassium dissociates previously so the binding sites are empty. So the sodium can move to this originally potassium binding site. And the chloride can now associate, now we can associate with lower free energy than if the sodium didn't move. It moves past the sodium and it reaches this point, which is the, let's say, the branching point, the P23 state. And this is interesting because now the top chloride occupies the original bottom chloride binding site. So we are now in the state where the sodium is in the potassium binding site and the top chloride is in the bottom chloride binding site. And then here the top chloride can now continue dissociation and finally dissociate using the same pathway as the bottom chloride dissociated, which we have seen previously. Or we can see the movement of the sodium, which now moves away from now the potassium binding site and then sort of step wise the sodium potassium, sodium and top chloride moves away until they dissociate. And again, in the meantime, again highlighting the asparagines, which I told you before, which have so far been improved in all the dissociation processes. Now, this here I'm showing just a summary of the process, which is a schematic diagram of the whole against the C1 ion transfer process. So it starts with the dissociation of the bottom chloride with relatively low free energy barrier. Then potassium dissociates with, let's say, slightly larger, but there's a barrier. And then we come into the mess that is the dissociation of the top chloride and sodium, with the most overall path being that the sodium and top chloride move together in some interconnected way. Of course, we also investigated the possibility that it's the provider dissociates first, and then the sodium dissociates on its own, but I'm not showing this time constraints, but all these options were investigated and most probably is this path highlighting the red that represents the ion transport process. Now, as I have slightly spoiled in the subtitle, all four ions travel through the same channel, which is interesting because we are here dealing with ions with opposite charge. Here I'm showing the figures, hopefully it's visible, but the surface is presenting the volume occupied by the ion, so this is for the bottom chloride, potassium, top chloride and sodium. And the question might be asked, how is it that in one channel you move both positively and negatively charged ions? And here we are going to highlight these asparagens. You have seen be part of the process of all the ions, and it makes sense that as the asparagens sort of ambivalent nature, they have partial positive charge, which through which they can interact with the chloride ions with the negative charge, and they have partial negative charge through which they can interact with the positive charge ion, the sodium, and potassium. So we were optimizing that this may be like the feature of these ion transporters, maybe even shared in other families. So we took a look at the simple sequence alignment of the two isoforms in KCC as well as the KCC transporters. The KCC transporters are a related family of transporters that transport potassium and top chloride, but no sodium. What we saw is that two of the asparagens identified as key are conserved among both of these families, while one of them was still conserved only within the KCC one. So we are optimizing that this is potentially a common and important feature of it is ion transport proteins. So to conclude what we have done, we have really used a very simple and straightforward workhorse method of biomorphic simulations to answer, I would say, really interesting biological questions of this transporter. We identified the order of the existing ions as bottom chloride, potassium, top chloride, and sodium, or top chloride, and sodium together. And this is also reflected in the fact that the later ions, the sodium, visits the now empty potassium binding site, while the top chloride visits the bottom chloride binding site, which is now now in empty. We also see that all the ions have a full sense channel, which will be facilitated at this partially by the unprovided nature of the asparaginous use. Of course, no scientific project is ever finished. They are just abandoned, and we are not abandoning this project yet. So they are ongoing efforts. First is to look at the full dynamic model. Because while we think that the smaller model we use is suitable for what we have done so far, there is still the question of what is the role of the demonization, what is the effect of the demonization of the dynamic transport, and what is the role of the large seed term domain. And of course, because ultimately, the goal is to discover a study finding new novel inhibitors of the system. We are also looking at the inhibitor interaction with the transporter. And for this, I have just a small sneak peek of what we are doing here. So we are looking at the role of the Bumetanid, which is a ribbed neurotycline, which is a drug that I think used to be prescribed against swelling and high blood pressure. I think it's no longer as commonly prescribed. And here I'm showing, we have a preliminary binding pose of this drug inside the transporter, inside the inner cavity. What is interesting, and what we see from the simulations, this partially collaborated on the literature, is that this inhibitor outcompetes or competes with the bottom-hole right ion, so the bottom-hole right is no longer bound, but it requires the presence of the other free ions for the interaction, which is, which makes the, let's say, vitro-aschistic and other studies a bit more, more challenging. Again, the workflow for this was, again, very simple. You start with the docked pose of your ligand, which you then can refine for the micro-dynamic simulations or metadynamics to obtain, like, the true optimal minimal inter-pose of your inhibitor. And then you can do magnetic energy calculations and look at the important interactions that you can then target in more a rational drug design approach. So I will not go into any details or time constraints, but I will be available for the gather session if anybody has any questions about this type of studies, what you do here. So with that, I will end the talk. I will acknowledge the supervision of Alessandro Vemistrato and the nice friendly collaborative atmosphere in our group with all the members. Thank you. Thank you. Thank you very much, Pavel, also for staying perfectly in time. So we have five minutes, let's say, for a few questions. So please go ahead. I have a couple of questions, if I may. Yeah, go ahead. Yeah, sure. Ciao, Pavel. Thanks a lot for the very nice talk. It was super, super interesting. I was curious about the first part where you did the quick and dirty simulations to get an understanding of what was the order of the exit of the ions. So I was wondering is going to the second steps and then the third, for the exit of the second or third ion, did you start your simulation from where the first ones ended up or from the first initial one equilibrated and you removed the ions yourself? To just understand it practically. What you've already done is around the metadynamics, then take the final structure from this with the ion that left, let it equilibrate again for a couple of nanoseconds, maybe even up to 100 nanoseconds, and then there was the starting point for the next metadynamics. Amazing, okay. And also another question I have is like during the metadynamics run, where you basically like at every round you choose like the proper variables to observe the exit of like, I don't know, like the bottom chloride and then the potassium and then the other two. In the case of the first one, for instance, you set your metadynamics and enhancing the sampling of this ion here. I don't know, does this affect in any way the binding of the other ions or you had to put some restraints on those or it went like smoothly? No, no restraints were really needed for the other ions. Okay, so you were like it was working rather well just putting the yes. Okay, interesting. And also this is just like a stupid curiosity probably. At the end, you showed like for the for the next projects you have like this inhibition, yes. So I was curious how you plan to like after after docking, which I more or less know how it works. I've used it a little in the past. I was curious about how you plan or if you have some ideas on how to do the metadynamics pose refinement in this case. This is what I've already done actually. I'm just not showing you. And for this I have again used the volume based metadynamics. Okay, so similar practice to the one used for the ions. Yeah. Okay, cool. Thanks. Thanks. Okay, one more question maybe. So there he is. I think Christian had his hand up. So yeah, thank you for the presentation. My question is, no. So John, and what I shown you are transported by a gradient of concentration or by an energy needing on a transport port like at the base or what is the dynamic for some drop when they went on that transport like by releasing what is dynamic of that transportation. I'm sorry, I didn't understand properly. I'm asking for sodium amputation a transport that what is the dynamic of the transportation. Is it a gradient of concentration or is it by an energy you don't want to at the base part of some of this. There is no ATP used here. It is driven by a the ion concentration and be by the conformational change in the transporter. And there is a there's a regulation mechanism where the transport is phosphorylated for it to function. But there is no use of ATP or anything like that anywhere in this process. If that's the question you are asking, I don't know if I understand correctly. Yeah, thank you. Okay. Okay. Thank you very much. So thank you very much Pablo again for the witness talk. So