 Good afternoon, everybody. Thanks for joining. Actually, sometimes I don't give a webinar, so I need to go back into this format that was pretty familiar, sometimes I go. So I'm going to talk about the structure characterization of messenger RNA and even nanoparticles and just to give a bit of context on why we are interested on mRNA therapeutics in case you're still wondering. Well, basically, if you manage to deliver RNA or DNA directly in the cell, you have a huge advantage compared to when you deliver proteins because you could silence a gene, you could express the protein in the biological environment where it's needed, and you cannot well edit defective genes. So the promise is huge, but there are some challenges to overcome. And RNA is a long chain of negative charges, basically, and it's not that easy to, yeah, on its own, it's not going to cross the membrane. And on top of that, in the body, there is plenty of enzyme that are ready to degrade RNA that is from outside as a protection, of course. So we need to encapsulate RNA for two main reasons. One is to make it cross the cell membrane and get to the side of last, and the other is to protect it from degradation. So we need to find a biocompatible vehicle. Now, as my thought was pretty clear, I'm interested in looking at living nanoparticles to do that. But before going to the work, I wanted to just give you some time to reflect on this timeline, which I find very fascinating. The mRNA field and the living nanoparticle intended as a general live itself assembly kind of started in a similar time back in the 60s, and then they evolved. And very soon after we started to know about RNA and lipids and how to make liposome, there was an attempt to actually encapsulate RNA in liposome already back in the 70s. So there are several points, just a few of the milestones that were reached during the time. And it's not that we suddenly got living nanoparticle to work. There was a lot of effort to get there. And yeah, we all know that these nice parallel lines actually ended up on being a very useful formulation that protected us from the COVID pandemic and kind of made us go back to normal life. So just to now come back to the science and the samples. So living nanoparticle, as the name says, is a multi-component lipid formulation where we have a cationic ionizable lipid, which is let's say the main actor in this formulation, because it's a lipid that has a positive charge when in low pH, which is when you formulate the particle and you will see as well later on in the body. And then while when you increase pH, so in physiological pH, the charge is lost. Well, yeah, it's not united anymore. So it becomes neutral. And that means that this vehicle is less toxic to the cells. So that is the main actor, because it's the one that actually encapsulates mRNA. So the positive charge will bind to the negative charge and encapsulates RNA in low pH environment, and then maintain this protection along the way. Then we have other two very important components, which are a phospholipid, in this case is the SPC and cholesterol. And those are helping stabilizing the particle and help fuse with the endosome and so on. So they help the particle in the function. And then there is some pigmented lipid that helps some colloidal stability to the formulation, helps tune the size, protects or at least reduce a bit protein binding once in body fluids. And that's yeah, basically these are the main functions. And of course RNA, which is the cargo of this formulation. Now, as Vito said, I've been using neutroscattering since basically my master thesis. And this is a very nice system to look at with scattering, in my opinion. And I've decided to use more angle neutroscattering and control variation. I guess this audience does really need a thorough introduction on the technique. But I just want to point out that when you perform SANS, your sample is in solution. So basically you can have conditions that are close to the condition that you store your sample or that the sample can encounter when delivered to the patient. And SANS, as you can see here, basically you have your sample, you have an incident beam that could be extra neutrons. Most of what I will talk about will be neutrons. And you just collect this captere beam on a 2D detector. And then just gradually averaging this captere the intensity you get the, well, pretty probably a very well known intensity versus Q curve that's already just looking at the shape will tell you a lot about what the shape of your particle in solution, what's the size and thanks to contrast variation. So taking advantage of what nature decided to do, so to give a very different perception to hydrogen and deuterium, we can play around substituting hydrogen with deuterium and we can match different parts of what we have in the sample. For example, here there is the scattering density of different biomolecules and the content of D2O in the solvent and changing the content of D2O, you can match the intensity, the scattering density of these different molecules, which means you can either highlight or cover completely the intensity. So a more clear example is like a core shell particle, as you can see here, if every system that has easy solvent, everything has a very different contrast than you can of course look at these as it is without changing, I think, but it will be a bit difficult to get all the information out and understand well the shell, the core, the sizes and all the contrast that describe the system, but if you take advantage of the mixing of deuterium, deuterated water and regular water, you could for example match the scattering density of the core and highlight just the shell and by the way, you could match the scattering of the shell by putting the solvent contrast the same as the shell and then highlight the core and this gives you the possibility to basically have two image of the same system in different conditions and allows you to strengthen your analysis basically. So you can reduce the number of parameters, the three parameters that you have when you have this kind of results. So that's very powerful and that's what we decided to use to look at living on a particle since it's a multi-component system. The work started after was published, the first work, I think is the first work, I'm pretty sure, where Sanz was used to look at living on a particle. In this work, they only had the DSPC deuterated and as you can see here is just a minor, the content is not that high of the DSPC, but still it was the available deuterated version commercially, so they looked at the living on a particle substituting the DSPC with the deuterated version and what they could conclude was that there was this DSPC, which is the deuterated one, is mostly the shell, but now let's maybe go back a little bit. So here you have the scattering course, a different digital content as we have shown before, so taking advantage of contrast variation and here is the scattering density as a function of the radius of the particle. So this is the distance from the center and you can see that there is a core shell and some kind of shell with increasing solvent content. And one thing that we learn from this work is that the SLD of the core actually follows the solvent contrast, which is telling us that the core is not dry, so there is water in the core and that's something that without sense you couldn't really conclude. So a lot of structure work done on this living on a particle was done with cryotium and you can't really distinguish well the content of water if it's not highly hydrated like a liposome, which is basically water so here we learn that we have solvent in the core and we learn that there is a shell that is basically dry and since the only component deuterated there was the SPC, the conclusion of this work was that the SPC is mostly at the shell of these particles and this was the structure proposed by Yan and Zerteta and collaborators. I had the opportunity to work together with Yan and Zerteta and we started together a project that was funded by a funding agency in Sweden where I was working before Malmo University and we decided to continue working on the same formulation but just using different deuterated components because as I said the SPC is just a minor component of the formulation but there are cholesterol and cationic and WP that are present in a large amount so we had the possibility to use cholesterol deuterated which was produced custom made by different deuteration facilities both from ILL and ASTO and that helped us localize cholesterol. Then we had a kind of combined deuteration scheme where we had both the SPC and cholesterol deuterated and in the end we had the deuterated kind of Yan and Zerteta which was produced by the chemist in AstraZeneca which were part of the collaboration and this helped us localize the Yan and Zerteta. So what we did again was to have exactly the same model composition just different deuteration scheme and collect scattering curves in different deuterate solvents and already just looking at the scattering curve we can learn a lot about the sample knowing the deuterated content. So for example if you look at the low Q region in this region here you see that when you have low content of the tool there is a clear indication of the core shell structure and in low content of the tool the mainly the visible component here is the deuterated cholesterol. So that can tell us that cholesterol somehow is not homogeneously distributed across the particle and to confirm if you increase the tool so you go toward matching cholesterol you lose the feature. While if you jump to the formulation where we have only cut the Yan and it's a little deuterated we see that this feature of core shell it's actually most visible in higher deuter content where again cholesterol is visible because it's hydrogenated. So that's a really something kind of pointing out in the right direction where is the component and the other information we can get just looking at the at the sun's curve without any analysis is that we have this kind of broad peak around this Q which corresponds to about the 65 angstrom distance repeating distance and this is visible mostly in this two deuteration scheme where the cationic and azobalipid is hydrogenated sorry and and the detox content is higher so there is where the cationic and azobalipid are the largest contrast while when we have it deuterated we completely lose this feature and that's so by this is because here the cationic and azobalipid is matched in the higher content of d2o while in lower content of d2o is the incoherent background that is to y and just covers up the peak. So again what we kind of our guess just looking at the data as they are is that cholesterol is not homogenous distributed in the particle and cationic the cationic and azobalipid is the one that gives rise to this peak that kind of gives us an idea of what's the internal structure is so there is some kind of repeating distance we cannot say which kind of structure it is if lamellar inverse hexagonal or whatever but for sure there is something that gives rise to the internal structure but we only have one peak so that's not conclusive but of course with the fitting we combine the core shell sphere and the broad peak to catch this feature of the of the peak as I said and from the core shell sphere we get basically four main parameters the radius of the core the thickness of the shell and then the scattering density of the two compartments and by knowing what are the components that we have in there what are what is their volume molecular volume we can understand how they are distributed to result in the scattering density that we get and this is kind of a way to image or well in the proposal the proposal we make to say that our cholesterol so the green the green particle the green component is the one deuterated and then the rest is more or less gray they are very similar in scattering and density when they're not deuterated so what we could say is that the cholesterol is more concentrated in the shell than in the core when we add the deuterated dspc the shell is even more highlighted and then when we have deuterated but again as a believe it we see that it's mostly in the core so this is kind of the picture that we could reconstruct based on the scattering scattering data and and this was done in pH 7.4 in just buffer without anything else in there at room temperature now we wanted to try to yeah learn a bit more about this particle they are they are made through microfluidic mixing and they're kind in a trapped equilibrium state it's not a thermodynamically stable particle so it's kind of expected that they're going to change as soon as they are in different environment so one one thing that is the one mechanism of uptake of this particle when administered intravenously is that apolipoprotein e binds to lipid nanoparticle and apolipoprotein e is a protein that is present in the theorem that binds lipoprotein particle that is responsible for trafficking fat in the body so it doesn't really recognize if it's an endogenous lipoprotein lipoparticle or is external so it can just reversely bind from go from hdl and ldl to lipid nanoparticles and the idea is that as soon as it gets bound to the lipid nanoparticle then the lipid nanoparticle can be taken through the surface receptor to actually recognize apoi bound to lipid particles and once this is bound then it drops in the endosome and there happens the release of mRNA but we can talk about that a bit later now what we were very interested to understand at this point was this stage so once the protein interacts with the apolipoprotein e what happens upon this binding does it change the particle or is just bound is just a corona like not interacting with the delivery system so to investigate this we decided again to go back to ants and contrast matching and we took advantage of the different iterated version of lipids we add to design a formulation that was completely invisible and we needed to be invisible not in a random solvent but in a solvent made up of 46% deuterated water because that's where the protein itself in native states are non deuterated is invisible and you can see here that the lipid nanoparticle with this specific composition the protein alone and the buffer they all overlap and they are all background basically this is just the typical contrast match plot that you would make to find the match point of of your system and this was the lipid nanoparticle so it's clearly yeah minimize the intensity when you are at 46% and now we add everything invisible and we decided okay now let's let's mix the protein with the particle if the two if the protein just binds to the particle without affecting the arrangement of components then we shouldn't see any change you should just give exactly the same it would just be the sum of the two intensity while what we got after the binding was an increase in scattering and this is not the major increase you can clearly see the signal is tiny but that was an indication that something was happening and that the protein was affecting the lipid nanoparticle component distribution so we repeated the incubation with the other formulation so the formulation having different deuteration schemes again where only cholesterol was deuterated where only the cationic and as I believe it was deuterated and where we had both cholesterol and the deuteration and you see in black at the data points of the sample before incubation and in blue is upon incubation with apoE and especially for the sample containing cholesterol deuterated we see a clear shift and again this is not the protein that shows up because the protein is invisible in this contrast is the 46% D2O it's the particle itself that is a redistribution of components so fitting this the data after incubation we could conclude that the binding of apoE affect the distribution and drives the cholesterol even more in the shell while the cationicizable lipid is pushed toward the core so that that's that's a pretty interesting finding because that could mean that the fact that in the somal escape is not so good for this lipid nanoparticle so the fact that you don't really have this positive charge at the outside could bid you as well to what happens to the particle once in the blood interactively protein but of course this is just a single protein is not blood but this kind of points us in the direction that we should try to look at structure and to look at the system in situations that are closer to reality to be able to learn more of what happens to them and just to complement the data with a collected from sans was to measure the sucks of these particles and kind of zoom in in the internal structure and what we could see is that even the internal structure was affected by the binding of apoE and here we tested as well we followed up on the hydrodynamic radius and encapsulation efficiency so that is how much of the RNA present is actually inside the particle and we can see that the hydrodynamic radius doesn't change at all no matter how much protein you add no matter how much time you wait for the incubation to go on so that this is important to show because sometime we like too much on the last so even if the last doesn't show any change it's important to look into the system with more advanced methods because change may be in a other landscape as we as we found and an ear is shown how much of the mRNA is encapsulated upon binding and you can see that the RNA actually the encapsulation goes down as soon as you increase apoE content and as soon as you wait longer for the for the protein and the particle to incubate so what we see is the redistribution of component actually does affect as well the ability to keep the mRNA encapsulated in the particle so that's another important point to keep in mind now if we go back to this team wow the lipid nanoparticle also to be uptaken by cells the the other like point that got our interest is once the lipid nanoparticle is trapped in endosome in the endosome the pH drops and there what is hypothesized is that the lipid nanoparticle becomes positively charged because the cationic anisobal lipid at lower pH is positively charged while the endosomal membrane is negatively charged and then these two should fuse and release RNA now the release of RNA is extremely inefficient is few percent so there must be something going wrong at some point of this process and there is a lot of work ongoing to try to understand and now what we were interested to see was what happens to the structure of this lipid nanoparticle once you lower pH does it remain the same does it redistribute what happens so what we did again we just continue to use the same formulation same content of mRNA and we just add different deterioration deteriorated component yeah just just to have an additional information and again we just collected sunscreen a different content of D2O and again we have a clear signature of the corrosion structure of these particles and then we drop the pH and there is a very clear change so I would say in the low Q we have a clear change that is very very visible a 68 percent D2O where the corrosion structure is lost basically and you can see as well that the size is likely increased it drops a bit a lower Q and then the other clear change is the is the rise of this internal peak so before there was almost nothing is a bit unfortunate that we had the merging of the different Q range but still I wouldn't say there is a clear peak here but as soon as you lower pH that comes out from background and it's very clear and and that means that not only the overall structure of your particle is changing but the internal structure is kind of getting more ordered which actually makes sense since a lower pH they can be charged so it could kind of yeah induce more structure so again I used the core shell sphere model to feed these data and I got very similar radius and thickness and scattering and values for sculptural intensity for current shell to what I found in other deterioration scheme before and then when we lower pH we basically add no shell as it was already visible by just looking at the data and we had a slightly larger radius for the overall particle and the scattering density of the sphere is slightly lower than the core so again kind of suggest that we have a redistribution of this of these components in a homogeneous way across the particle so the pH drop does have a huge effect on the structure of this living on a particle so this brings me to the kind of conclusion of this part and I hope it was clear that using sense and contrast matching you can point out how the living components are distributed and living on a particle and that's going to be translated to many other systems and as well we could see the effect of a protein binding to the living on a particle again thanks to combining sounds and contrast matching and we could as well follow the effect of pH on the particle in C2 and this has been done all like in cuvette without dynamic change of condition but one thing that it's very interesting and opens up a lot of possibility when you use sacks and sounds is that you can actually do time resolved measurements and this is just an example that from a recent review book where they show the aggregation of proteins and they could follow the change of scattering along time and this is something that is extremely powerful and I may have already said what we used to make living on a particle is a microfluidic mixing so the idea that we decided to pursue together with Tommy Nylander and a very bright PhD student was to look at the formulation in in C2 and that was possible together in collaboration with the people at Cossacks in Lund at max four coupling the microfluidic with the with the sacks beam line and this is just a picture of the setup that we had and what we did was to have our microfluidic chip with a with a geometry that is used to make particles and the deformation online and this is work part yeah it's work of Jennifer Gilbert just defended our thesis and what so this is the basically the chip well yeah it's like the microscopic view of the chip and we decided to measure it for different position which actually translate to four different time points along in the formulation and and yeah I reported the scattering curves collected at the different position so each curve is a different formulation because this happens in a very short time so we have to remake the formulation every time and just look at the different point on the chip but I think it's very very fascinating that we can see the building up of the particle so we see this this intentional low q coming up and it's actually the particle being formed and for this the for this cargo which is dna you can see as well that the internal structure forms very quickly this is on the millisecond scale just to give you an idea and and for example we can see differences between targets so this is polyadenilic acid which is a model commonly used for mRNA and and it's clearly different our looks compared to the dna one and then what we did was so um the formulation happens it's a it's a mixture of you have an etanol etanol etanol solution of the lipids and the RNA is an acidic buffer so you mix the aquas and etanol solutions in the chip so what happens are the outlets that you have a mixture of aquas and etanol and you need to get rid of etanol and you need to increase the pH so you do the dialysis to bring to get rid of etanol and increase pH and what we did then was to follow up as well the structure upon dialysis and you can you can see how the the peak evolves so the internal structure evolves along the dialysis process and then after the concentration of the sample so actually the scatter of this data is because the concentration is fairly low and then it comes cleaner once we up concentrate so yeah this is to I think close that I think it's interesting to try to use this technique in in kind of more relevant uh conditions and try to look at what happens over time resolved over time and this brings me to the end and to thank all the people that have been involved in this work and the funding agency and my the group that I am part of of which I already collected quite some data but I don't have time to include unfortunately in this in this seminar and thank you for attention I'm happy to take question