 So I think the time is 11.15 so and we have to get started and it's my great pleasure to introduce to you Richard Campbell and Richard came as a postdoc to learn and we did some very nice work together and he was we also started to do some neutron experiment together and Richard then was assigned to be the instrument scientist at Figaro and I don't know how many neutron experiments you had done but not so many and he made a fantastic job in developing the Figaro instrument so the lesson is don't be afraid to start using neutron and neutron scattering don't be afraid of some complex formulas and so on you will master it and no worries at all so and since then Richard and I work together and we are met and we are there together and so on so so I always appreciate Richard's careful analysis of problems and careful in writing a paper writing a paper with Richard here if you don't get above the version 20 I think it's a failure yeah so please Richard okay well thanks for the introduction Tommy and like you mentioned I came to London back in 2006 and had a very enjoyable two years in London and I remember in the first week I was presented with a broken ellipsometer to reconstruct which scared scared me a lot but somehow I got through that and then two years later I went to ILL to take what was bits of a neutron reflectometer and put them together as well which was an even bigger challenge but even though they were both quite scary at the time I think it all came good so yeah when I was with Tommy I did some neutron experiments but also ellipsometry and roof strangle microscopy and these will be themes throughout today and like Tommy said I went on to ILL to commission and to be responsible for the figure of liquids reflectometer so I'm going to give you an introduction to that instrument not only neutron reflect reflectivity as a technique but different ways to use it and how actually you could submit the same beam time proposal and propose to do different types of measurements and learn different types of things so that's what I hope to do today but like the title suggests this is about using the air liquid interface as a platform to study drug interactions and there's I'm going to present three different projects looking at different drug interventions and when we come to those projects they were done in collaboration with Daniela Chiamac and Gianlu at the University of Manchester and I was at the ILL at the time then also Maritay Cardenas who at the time was at the University of Copenhagen and now is in Malmö and I've gone away to Singapore for a sabbatical and then the third project was with Dorota Matizuska and Renata Bilovich at Warsaw University but those are the three kind of research projects I'll present later but to start I'm going to give you an introduction to the neutron technique but this won't be a standard introduction as I say it will be more of a trying to try to get different things out of the same technique so the slides will start with a figure of a cutting-edge neutron reflectometer applied in the biosciences then looking at these three different ways to exploit the same technique and the three projects will be structural disruption by antimicrobial peptides membrane binding of liquid crystalline nanocarriers and interactions of anticancer drugs so let's start by looking at the figure of reflectometer so I was commissioned back in 2008 and it's a time of flight neutron reflectometer it's on a it's on a reactor neutron source it's a continuous white stream of neutrons that come along guides so highly reflective mirrors that come along guides to the to the instrument so in the top right schematic the neutrons come from the right side and probably the most important aspect of a neutron reflectometer is the choppers or the spinning discs so you see some discs like featured after that and these are discs that absorb neutrons but they have a hole in and they spin around and essentially they can create pulses and the special thing about Figaro is that while normally there would be two choppers to form a neutron pulse in this case there are four so at any time two choppers will control the pulse and if they're the two choppers that are closest together you form small but very well-defined packets of neutrons which has lower flux but gives you high resolution in the data or if you use two choppers that differ further apart you have these big big pulses but the starting timing position of these neutrons much less well-defined so the resolution in the data's less you have more flux so after that there's some mirrors which can deflect the neutrons down to the sample in a sample area and then they bounce up to the detector so the principle of the time of flight is that these neutron pulses will travel down the instrument reflect off the sample and up to the detector but the time they take to do that depends on their energy or their wavelength so in Figaro they have neutrons from two to thirty angstroms and the two angstrom neutrons have a low wavelength so their frequency and energy are high so they travel fast and then the thirty angstrom neutrons have a high wavelength their frequency is low and their energy is less they travel slow so essentially once that pulse is made it travels down the instrument because it's because it because they all take different times the the high energy neutrons arrive first and the low energy neutrons arrive last so all the detector needs to do they gets the easy job it just needs to record in time and then you can plot instead of against time you can plot the wavelength so that's the principle of the time of flight technique and the features of the instrument is that it's a powerful and versatile instrument meaning that you have this choice of chopper pair so you can balance the between having high resolution or high flux and actually those mirrors which direct the neutrons to the sample you can imagine with an air liquid interface the neutrons travel better through air than the liquid so you would always reflect up at an air liquid interface the neutrons are good down and bounce up off the sample but there are actually some examples where it's actually better to reflect downwards so on Figaro they've done recently a lot of oil water experiments with bulk oil and bulk water and then when the oil floats above the water it can be actually easier easier for the neutrons to the water than the oil so in that case it's better to reflect down so it's quite versatile and flexible but it's also important to get into perspective the the nature of the instrument so the beam size of several tens of centimeters it's not very not not got good spatial resolution not like a laser or an AFM with like looking at the nanometers scale this is looking at tens of centimeters and the data required are required in a time scale between seconds and minutes so there's a huge range of science that could be done from peptide bindings DNA interactions reaction kinetics drugnena carriers interfacial mechanisms formulations biophysics and various other types of science so it's about designing the experiment it can answer specific questions using the so the the neutron instrument built but even though the infrastructure of the instrument is important you can't do much without really state of the art sample environment so on the top left there you can see adsorption troughs which are really quite quite attractively made and you have these quite large teflon or btfv troughs in which you pour in your solutions so you have quite a large air liquid interface where the neutrons can bounce off and on the upper right upper left side of that assembly you can see this perspex and these there's mirrors where laser will go so precisely align the height of the sample and that's important because it because the the neutron reflectivity is a grazing angle technique the angle of the neutrons can be even less than a degree so it's really important to get the height right so in real time the the laser that the instrument control can speak to the laser and then the whole assembly can move up or down so the neutrons are in the right position. Top right is a Langmuir trough and we're going to look at some examples later of drug interaction solved using a Langmuir trough as a platform to look at the air water interface and there the most obvious thing is to spread a lipid monolayer so you spread lipid and insoluble molecules using an organic carrier solvent and they form a monolayer and then you can use barriers to control the surface pressure. At the bottom left you have a suite of solid liquid interface cells so this is a silicon crystal pressed against some some liquid and then they kind of squeezed together and the neutrons go actually really well through some solids so there you can use sapphire or silicon and the neutrons just travel through the sapphire or silicon reflect off the water and these are used a lot in terms of binding molecules to different model biomembrane and the the last one is a rather specialized environment but it was this overflowing cylinder where water comes from a gravity feed and just goes up a cylinder and spills over it but it's designed to create a flow so you can actually reflect neutrons of a flowing liquid surface and look at formulations relative of relevance to how they're actually used which is not waiting for them to equilibrate but actually under dynamic conditions. Okay so the next three slides are going to consider three aspects of using neutron reflectometry to to address problems that might be of relevance to your research projects and the first one is like the standard way of using the technique but I'll go on to show you two other ways and it's quite interesting to bear in mind. So the first is to use neutron reflectometry to look at the interfacial structure so here you have a relatively collimated fine beam of neutrons reflecting off the sample at grazing angles at a low angle and the nice thing here is there's sensitivity to light elements so if you were to compare with x-ray reflectivity the number of electrons in the in the atom increases with the atomic number so x-ray reflectivity is more sensitive to heavy atoms so actually if you think of the lightest atom hydrogen there's really not much sensitivity at all. The nice thing about neutrons is that it depends on the the coherent neutron scattering length so if you look at the top that like the center image there what what you can see there is some different isotopes so there's there's one h and two h as the first two so that's hydrogen and deuterium and then there's carbon nitrogen oxygen phosphorus and sulfur but you can see that their values they're not that predictable according to the atomic number their their scattering lengths depend very specifically on the isotopic question but the peculiarity is that the neutral scattering length of hydrogen and deuterium have opposite signs so they really stand out relative to each other so actually if you think of organic molecules with a lot of ch2 groups like repeating ch2 along like a hydrocarbon chain if you add one carbon and two hydrants together there so 6.65 minus two times 3.74 you have a very low number so what that means is that hydrocarbons themselves are quite invisible to neutrons so all of a sudden if you deuterate that molecule then you have cd2 instead of ch2 that number goes up massively so it means that you can take individual components in a mixture and if you deuterate them they really stand out. The other special thing about neutron reflectivity is the wavelength so I've already alluded to the fact that the wavelengths around the angstrom layer but if you think of the self-assembly of soft matter systems like liquid monolayers they're the thickness of these layers that are on the angstrom scale too so if you take a surfactant like sds it'll form a monolayer of about one nanometer if you take a lipid like dbbc maybe as closer to two nanometers but these are around the wavelengths that we're looking at but what that means is that if we have a monolayer so look at the the bottom left schematic here and we have this neutron wave reflecting off the top interface and then also it going to refract through the monolayer reflect back off the subphase so reflecting off the second interface those two waves are going to interfere and you'll have regions of constructive and destructive interference so as opposed to if you use light to do that you know the light has a wavelength of hundreds of nanometers so it hasn't got the sensitivity to make interference patterns for such thin films the neutrons do so if you look at the bottom center schematic there this is these are neutron reflectivity profiles where on the left there's the log of the reflectivity and the bottom axis is called q the momentum transfer which is related to the sign of the angle over the wavelength so that's just the way that we normally plot our neutron data but you can see there this is a simulation for films of different thicknesses you can see on the legend there you have films ranging from 15 angstroms up to 50 angstroms and there there's an interference fringe that actually is going from right to left as the film gets different so what that means is that if we do a neutron reflectivity measurement then if the position of the fringe is going to tell us the thickness of the layer but more to the point it'll tell us mostly where the thickness of the deutrated molecules so if you were to measure in different isotopic contrasts like a normal molecule than a deutrated molecule then that fringe position would change and you could model the data together to reveal the interfacial structure so on the right here is is the result of a neutron experiment and it's of a a student that taught me supervised called maria mariana yena sateta and i co-co-supervisor and these are some datas of when a dendromus or hyper branch polyelectrolyte interacts with a surfactant so at the top you see some schematics of different possible interfacial structures formed and at the bottom there's some data so those four datasets are having benedroma with either normal or deutrated surfactant and either a mixture of h2o and d2o that that has zero scattering called contrast matched water or d2o so essentially there's four contrasts so h and d sds and h and d water practically so there's the four contrasts there and you can see that for the for the the red and the green data my apologies if anyone is color-binding the two of the datasets there there's a sharp interference fringe like is shown in the middle picture so what that does is if you can if you apply a common model to that if you to simulate your data and optimize the model for the number of molecules and where they're sitting you can actually work out where the molecules are and all the different components where some other lab techniques they can give you an overall amount of material but they can't tell you where different issues are and what you get out of that is a volume fraction profile which is shown at the top right of that of that image and in the red is the amount of surfactant in the black circle so that's the type of the type of information you get from the experiment and then here's another example from the literature of an antibody binding to a to a peg layer on a on a on a grafting layer on a silicon surface and they just the data aren't shown here but what's shown is the result and really the the level of detail you can get out of these nutrient experiments so again in doing measurements in different isotopic contrasts you get the volume fractions of all these different components there's silicon there's a silicon dioxide layer grafting layer you've got your peg brush your antibody in the water and that's what you get out you get all that information about the amounts of material where they are their volume fraction their thickness okay so that's the normal application of nutrient reflactometry and here we go on to the the second one so this is a technique that emerged over the last five years so it's still quite fresh but it's now really being used a lot so let's look at the top left and this is a simulation of a single layer in air-conscious matched water so the air-conscious matched water has a sketching length of zero has a sketching intensity of zero so on that top left figure if you look at the inset there's a sketching length density graph against the distance down to the interface so I've added just a little bit of smoothing to the data just to just to have some a realistic reference of the interface but essentially one of the curves the blue curves is rather sharp and thin and the green curve the other curve is rather broad and and less intense so essentially what I've done there is I've simulated two different layers one is dense but thin and the other's not dense and thicker but there's the same amount of molecules in each layer so I've there simulated two different systems of the same amount of molecules but for different densities now what you can see from the neutron reflectivity figure on the left is that at the low Q these data coincide so so when I've simulated two two systems of the same number of molecules their reflectivity at the low Q is the same but actually over the whole Q range if you go further to the right then the data are different so that's why in when you're interested in measuring the interfacial structure you need to measure over the full Q range you will get because it's the data at the high Q that has the highest sensitivity to the things but what I'm saying here is if you just measure at the low Q then you don't have sensitive civities of things so whilst that would normally be considered limitation in this new implementation of the technique it acts as a strength so the idea now is to say actually if we measure only at low Q we don't need to worry about the structure we don't need to worry about where the molecules are but we can just measure their amounts so the data at the top right is measuring a polymer and a surfactant in different isotopic contrasts if you if you look at the the three data sets over the full Q range they would take about two hours to result to measure normally about 40 minutes a data set on average and you would get out that the the scattering density profile in the inset so you'd get to know the amounts of the material and where they are in two hours but if you look at the data in the green circle I can measure the upper one in red in about a second or two and if I make the the polymer sorry the surfactant completely invisible by isotopic substitution and just look at the amount of polymer then then just above the background there you have the this tiny signal from the polymer which I can measure in a minute or two so what that means is if I just measure at low Q and I measure a sample in two different isotopic contrasts but both in our air-conscious matched water then in in one or two minutes I can resolve the amounts of the two components and that's what the equations in the middle say there so the scattering density times the thickness is the essentially it's the absorbed amounts the product of the number of molecules in the system so for a single component the scattering density times the thickness is equal to the absorbed amount times the scattering length of that component times Avogadro's number if you have two components you can solve those simultaneous equations you make your two measurements at low Q and the the reflectivity can be solved to get your surface excess of each component so what we're saying here is that before get the surface excess of each component we would measure for two hours and we would get the structure as well but now we can measure for two minutes and we only get the composition but this can really open up lots of studies into dynamics and kinetics that was not possible yeah and the data at the bottom left is some data that was recorded first back in 2016 and in this case we had a polymer surfactant mixture which we we compressed and expanded on a language or five times we wanted to measure the amounts of polymer and surfactant during the compression so before this wouldn't have been possible because the measurements would have taken too long but but now it because I can resolve measurements in two minutes I can I can measure the amount of each component during dynamic experiments and last the last application of neutral activity I'm going to call particle attachment and this is related to the depraction of neutrons through multi-layered samples so if you have a multi-layer sample and you have an instant beam then the reflection at each interface is going to give a reflection and all the waves will interfere and what you get in that case is a bragged diffraction peak so in the bottom left you can see some data of a polymer surfactant mixture at the air-water interface which formed this 2d hexagonal phase so here we had an ordered phase at the air-water interface and this bragged diffraction peak would tell us the the amount of the the ordered material at the interface and then here's some some more data from Tommy's group and one of his phc students back in 2008 Pauline Vanderleg and she was looking at liquid crystalline nanoparticles and in particular she was looking at cubosomes and there she did some some novel neutron experiments looking at the the this bragged peak with time and the bragged peak was the signal that actually the cubosome had attached to the surface so neutron reflectivity most people think of it as one technique but I've made the point here that actually we can think of it as three we can design experiments so we measure multiple contrasts over the full q-range to get the structure or we can do kinetic or dynamic experiments focusing on the low q-range to get the interfacial composition or we can even look at particle attaching so it's almost like three techniques in one so with that introduction to how we're going to use and exploit the technique to understand drug interactions let's move on to three different examples so this first example as I mentioned was done in collaboration with Daniella Chiamac and Gianlu at University of Manchester when I was still at the ILL it started then anyway and the motivation for the research was that we know that antibiotics function by various mechanisms for example penicillins work with interference with cell wall synthesis and there are other mechanisms like interference with nucleic acid synthesis but there's been a discovery avoid so the major antibiotic discoveries were from the 1920s through to the 1980s and therefore because of antimicrobial resistance there's a strong motivation to discover new types of antimicrobial materials this particular project was done to look at some short designed anti-microbial peptides and these have normally at least two positively charged residues and they're usually more than 50% hydrophobic but they can have any kind of structure according to the sequence of amino acids used so this was done in collaboration with Manchester Petroleum University China and the ILL and it was looking at these short designed peptides with the sequence G IIK repeating N I NH2 and the N number really gave different results of the efficacy and of the interfacial behavior so if you look at the top right this is the survival rate of different bacteria there's E. coli or beta subtilis and it was showing that for one of these peptides you could you could get effective suppression of the survival rate with just a few micromolar of the peptide concentration and at the same time that's all very well but if these are so deadly that it destroys all the red blood cells then that's not very good so the hemolytic activity is shown in the bottom right where below about eight micrometers there was negligible hemolysis before four of the different peptides so essentially that was just saying that these have quite some promise to have anti antimicrobial activity but not be too harmful so that was the basis on which to start the project now the reason to do a surface project is to gain insight into the mechanism of interaction of peptides there's still debate over the way in which different peptides interact with bio membranes for example toroidal pores conform or a carpeting mechanism in which which disrupts the membrane so any information we can get from the application of reflectometry or surface sensitive techniques can give some insight into this approach so the the group in I mean quite a few years ago the group looked at the minimum inhibitory concentration to the efficacy of the peptides and it was shown that for the different n numbers the n equals four peptide which was abbreviated to G4 had a very low minimum inhibitory concentration so two micromolar E. coli and 0.5 microbytes of dillist so so then this essentially prompted a led to a PhD project of daniella who then took that G4 peptide and and studied the surface behavior so when designing a model membrane system it's important to essentially according to the type of membrane you want to study you can form a combination of lipids to mimic certain membranes but one feature of the neutrons is it's often useful to start with simple single lipid systems before you can build up to more complex systems and this is something this is you could say it's a limitation of the technique that a lot of the studies of neutrons are done with very simple model systems but I think as you'll see here I hope this is a good example of starting a model systems getting that basic understanding of the molecular parameters and then building up to more realistic model system model systems so this slide was simply comparing dppc with dppg so it says vitrionic saturated lipid with an anionic saturated lipid so very basic model systems but this is first like looking at the different lipid characteristics and in this case a lipid monolayer was spread on a lanyard trough and then peptide was injected underneath the surface starting surface pressure was 15 millimetres per meter and here it can be seen that with the anionic lipid the dppg monolayer it seems that the surface pressure is rising more and therefore it would infer that the extent of interaction of the peptide with the lipid monolayer is higher however surface pressure is not a quantitative direct technique to to work out the number of molecules absorbed for example when when a peptide is to penetrate the chain's layer the surface pressure would probably go up much higher than if it was sit underneath and bind to the head groups because then the pressure of the chains would increase if it's penetrating that layer so just because the interaction of the dppc is lower we can't we can't confirm from these data alone that the interaction is less it might just be different and that's why of course it's good to apply different techniques so at the top left there's some bruce triangle microscopy images so this is an optical technique where essentially p polarized light at 53 degrees for their water interface normally it transmits fully and with some approximations and then when you have a lipid monolayer if there are domains of lipids that are in the condense phase they really should stand out as being very bold and white in the images so for dppg monolayers alone you can see this phase behavior because there's these lateral domains of lipids in the condensed phase but when you inject the peptide underneath you can see that membrane disruption where the phase of the lipid over 25 minutes is completely disrupted then if you look at the top right figure we have some low q data of the kinetics and here we're in the air-conscious matched water and we have an we've made a contrast matched dppg monolayer so the lipids invisible so all we're seeing here is the peptide and the first runs the lighter run at the bottom and the last run is the the darkest run at the top and what that shows is that in real time on the minute time scale we can measure the penetration of the g4 peptide into the monolayer then at the bottom left we have the traditional structural analysis so this is the data that I showed like the type of data I showed in the first example and here we can resolve the the thickness and volume fraction of the asr chains layer then the head group and then principally in this case the low pressure of the peptide so in in summary the interaction with the dppg was shown to be much higher and and this led us then to go on to look at different parameters of the lipid I should just say at the top right here that the using the low q analysis the second implementation of neutral reflectivity then the the interaction of the the peptide was quantified for both the dpc and the dpg and also that the lipid was lost from the monolayer which is something that's often when you do langurizer terms it's often um often researchers would assume that lipid is not lost from the monolayer when when something from the subphase interacts with the lipid because the you keep this this acts you assume full lipid insolubility and that it stays there but this is showing that in interacting the peptide of the lipid monolayer you actually lost lipid and that was also according to the charge of the lipid as well so the second study was to go on to look at the effects of starting surface pressure and of the the saturation of the lipid and um from the surface pressure alone um the the effect of the the saturation was rather minimal so dppg which is what we looked at before was is the saturated anionic lipid and popg is an unsaturated anionic lipid so the head group's the same the comparison what you can do here is have started different surface pressures and look at how much the surface pressure goes up so you can plot the change in surface pressure against the starting surface pressure and if you make a graph of that and extrapolate down to to the zero change in surface pressure that would be the the maximum insertion pressure you can see for both lipids that is approximately the same at 43 millimeters per meter so from surface pressure alone you would you would conclude that the interactions are quite similar but it's not what was concluded from the neutrons which again shows the the benefit of applying more than one technique so the same type of data here in the top left we have the dppg in the in the left panel in the popg in the right panel and and essentially the that fringe in the data on the left panel shows that the form quite a thick layer with the peptide and the extent of penetration is much higher and then we have the low q kinetic analysis on the top right again showing that the higher in this case showing that the higher pressure so those are the black diamond in the left panel you've got a higher penetration of peptide when the surface starting surface pressure is higher so that's interesting given that those are around 30 millimeters closer to the pressure is found in in real cell membranes and here you get a higher penetration of peptide and again the schematic there shows that at the higher pressure and with the with the saturated lipid there was a higher pressure a higher higher penetration of peptide not only into the change but also ahead then lastly just one slide on a study that came out earlier this year and that was starting there to mix the lipids so although previously the two studies had looked at the effect of the lipid charge starting surface pressure and effect of saturation this is now starting to make lipid mixtures to mimic bacterial membranes so in this case it was looked at a mixture of pg with a cardiolipin in the six to four ratio to mimic a gram positive bacterial membrane and pg with p in a three to seven ratio to mimic gram negative bacterial membranes of course these are only approximation the real the real bacterial membranes contains more components like the polysaccharides but this is a basic model of two different types of bacterial membranes as a step forward to the to the to the real systems and here um the comparison of the two systems in the top right shows that um the if you look in the top in the right panel for the the the model gram negative membrane the the two data sets in detour with the critical edge which have the blue squares um they're almost the same whereas in the left panel with the pg they're much different and it showed that with the pg and their uh sorry the cardiolipin with the cardiolipin so the gram positive bacterial membranes there was a greater extent of peptide penetration so this work at least provided a an indication of the possible greater efficacy of this peptide to bind to a gram positive bacterial membrane okay so we go on to the second study now which is uh binding of liquid crystalline nanocarriers um i can't remember whether i stole this from one of tom's papers as well but i may well have done anyway um essentially uh lipids will self assemble assemble into different structures according to the the type of lipid the temperature and various other parameters um and and then when you add components to lipid this can obviously change the phase of the structures in which they self assemble too so there's different structures that can form i mean lipids can self assemble into micelles hexagonal lamella phase cubic phase reverse hexagonal reverse micelles so there's there's various types of structures that that small molecules and fulfilling molecules can self assemble into um this project was rather interesting because it was motivated by some data by uh Maritay Cardenas who was looking at mixes of dendrimers with lipids so dendrimers as i mentioned the hyperperbranche polyelectrolyze and she was looking at their their mixtures with um with p o p c p o p g uh lipid vesicles so what was what was known is that when you add the dendrimers they self assemble into these lamella multi layer stacks and what she found in fact was that when she did different experiments on the systems she ended up with different results so obviously that's not ideal um she would do a technique like ellipsometry or a technique like the quartz crystal micro balance and these results just couldn't be uh understood together and furthermore when she did neutrons when she did it at the isis neutral source in the uk or the ill in france she also got different results so this wasn't this wasn't ideal and then she she actually saw a talk i gave on on a polymers effectant system where i was talking about effects of gravity on the interactions of of liquid crystalline nanoparticles with with surfaces and here the fact that the these particles eventually can um either cream to the top of the sample if they're lighter than the water or they can sink to the bottom of the sample if they're heavier so that that picture of the flask there in the center is uh a picture of one of these samples after some period of time after some hours and what was interesting is the marite was looking at very small samples before i mean two was mixing one or two millilitres or less and you know some creamy liquid would form injected into the measuring cell and it i guess there just wasn't that perspective that gravity could have have an effect on the sample so we designed this experiment on figaro where on the right you can see there are two silicon crystals sandwiched between a bit of teflon which has which actually has the sample the liquid sample in so we could inject the liquid sample between two silicon crystals and as i mentioned that the neutrons travel transmit very well through the silicon so we could reflect them down through the top crystal and reflect up off the liquid but we could also reflect up through the first crystal and down off the liquid so we could actually do solid liquid measurements at an upper and the lower interface on the same cell and the question then is you know immediately from when you inject a sample into and and make it interact with a solid interface would you actually get different styles so the the main data figure there shows with time the data that we got at the two surfaces so the initial data which is lightest and not that dissimilar but you can see already that that lightest green data at the top where the where the black arrow starts going up there is quite a fringe there at low q whereas in the lightest data in the bottom the light blue there isn't so already from the very start from injecting this this mixture of dendrims and lipids into a sample different structures were forming on the top and the bottom interface and then with time I mean look how the look how the data changed the bottom interface made this very sharp interference fringe at low q and the top interface have these four bragg sorry three bragg peaks showing that the lamella particles are actually down to the interface so these were modeled through as a result of the talents of a very skilled uh date uh neutron expert called Eric Watkins and essentially the schematics on the right show how these samples are evolving at the at the bottom interface there's a there's a lipid bilayer but it interlaces dendrims that are sitting on the surface or sitting above the bilayer but with time all the dendrims penetrate through the through the bilayer and the dendrims sit on the surface and the lipid bilayer sits on top whereas on the top surface that's the starting point and with time the lamella particles attach so I think and I must admit I remember one of Maritay's reaction is when she saw that big 100 millilitre sample it's like wow how could how could we ever imagine that was happening you know when when you have less than a millilitre of sample but actually just seeing what happens with the effects of gravity suddenly made it all make sense and now Tommy has a an ellipso-meter where the sample is vertical and the laser beam is horizontal so and then a QCMD probably the I think the the sample would always be horizontal and then for doing a measurement on the the d17 neutron reflectometer ILL the sample's vertical figure out could be horizontal but it could be either above or below the sample but it just shows that these this fact is important it's really important because if you have phase separation in the liquid and you have particles going up or down with time you're going to get different interactions at the surface according to where the surface is and this can really help to explain some unreproducibility in results but it also shows that you need very careful experimental design after this technical you know discovery we then went to look at look at the interaction of of this these mixtures with bilet preform bilay is a different charge and actually what we showed there is that the the particles were attached if the bilayer had a sufficient negative charge but not under so this actually was quite interesting in the sense that we could actually tune the particle interaction with the with with the preform bilayer according to the orientation of the sample but also its charge so I think in there overarching lesson about experimental design okay then so the last study I'm going to introduce is the interaction of cancer anti-cancer drugs and here anthracycline antibiotics include doxorubicin which is a very well used drug in chemotherapy all over the world and edorubicin and they're structurally similar anti-cancer drugs as can be seen at the bottom right so you have the four rings I'm sorry the five four rings joined together and then another ring joined by an ether oxygen but there are different chemical groups in two different places on the molecules so the mode of interaction isn't known so we weren't necessarily looking to get insight into the mode of interaction because that's with interglation to DNA but in any case the outer cell membrane is the first barrier to be crossed so this this this project was to look at the interaction of different anti-cancer drugs with different looking numbers and as I mentioned this was performed in collaboration with Irina Tbilovic and Dorotimus at the University of Warsaw okay so well now we have our usual question of how to design our model system and this is rather crude but simple to start with like I say is the start of a project but here we chose the MPC as as an example of a healthy cell membrane because we know that healthy cell membranes have more vitro ionic PC lipids and we chose the MPS as an analogous lipid to mimic cancer cell membranes and like I say it's a very crude way to model healthy and cancer cell membranes because there are many other things that change like the amount of cholesterol for example this is just one way we did it so I've got the the vitro ionic or neutral PC lipids and the negative discharge PS lipids yeah hey Richard could I just ask you a quick question why are these instruments made so big like it feels like we studied these very tiny small molecules you know and the instruments like feels like so huge why are they not made a slightly smaller yeah you mean like this one this langmuir trough or do you mean yes the langmuir trough yeah yeah okay that's a good example so this actually this is maybe not the best picture of a langmuir trough for an air water interface because you can see in the middle of that langmuir trough there's a hole so this langmuir trough is actually designed for transferring lipid monolayers so you have a solid and that solid like you it goes down into the water as you push the the the lipids onto it or up but essentially there's a transfer of lipids onto the solid so you actually need quite a big surface because you have this large solid going up and down and that's the trough they had in Warsaw and that's the one they used but you do get much smaller troughs these days you can have troughs like this big you need room for a surface pressure monitor if I'm going to use ellipsometry or bruschrangium microscopy I've got to have room for a beam a laser beam to come in and reflect off the middle so in any case I would need it around that size but it's true this one is quite big but I think it's also used for transferring to solids okay thank you so they are available like a much smaller also like okay so yeah thank you great good question by the way great okay so uh so dmpc and dmps lipid monolayers were formed and then doxrubicine or edusrubicine was injected into the subface so the data I'll show you are all on the doxrubicine and this was the initial surface pressure data to show that on the left with the pc the effects of the okay this has the edrusin sorry the effects of the two anthrocycline are relatively small compared with the effects on the dmps so if you look on the right on the dmps those effects of the lipids of the of the drugs are much bigger and they're also very different so the idea was saying well yeah that's really interesting but we can't tell from the from these data alone what's happening so maybe we can get some information as neutrons so this was the analysis applied to the interaction of doxrubicine with the dmpc or dmps monolayers and we could we could guess from the surface pressure data that there would be a lower interaction but we don't really know because we don't know the mode of binding of the drug to the lipid but here we we've proved it so at the top right you can see the the surface excess of the of the lipid and of the of the drug as a as a function of surface pressure and the amount of drug is effectively it's around it's around the sensitivity limits really really low but for the dmps the amount of drugs much much higher and even even higher than the amount of lipid so there's a strong drug binding and it's not just electrostatically driven because there's additional drug over the over the amount of lipid then then we recorded some data this is just a one set of data repeated three times so actually the data is the same in all of those figures but what was different is the models applied so you might say that there's not a huge difference even in the models but you can see some difference on the left at the high q there's there's one data set the green that turns up at the end and the another day set the red that's lower than the the models lower than the points so it's showing that this model of the drug sitting only in the chainslayer is not very good and then on the right one of the drug sitting under the head cruise both of the models in the d2o that's the red and the green they kind of sit together but the data are clearly separated and so actually when you optimize the model which was the central one the dmps was sitting exclusively in the head cruise layer and that that's the best model we can get so in this case compared to some of the data i showed you previously maybe you could be underwhelming because these these differences in the models are not that big because we're really looking at thin films but it's enough to say that there's evidence that the drug sitting in the head cruise layer and then for dmpc this is actually the data with and without the drug and they're almost almost interchangeable we're really at the sensitivity limit in one contrast there's a small difference in the other three there's not a visible difference so um so the the conclusion is that the the interaction with the dmps is much much higher we've quantified that for the first time there's a reduction the lipid surface excess as well so the drug is also solubilizing some of the lipid and uh and there's in cases yeah there's actually the seeds equimolar replacement by the drug and the binding with the dmpc is minimal okay so i should be wrapping up soon because we've only got 10 minutes left so this is just the last summary slide to say that we started this this tutorial really by looking at three different ways we can use the same technique in terms of the interfacial structure the interfacial composition and particle attachment using neutrons so then we looked at three different projects the structural disruption by antimicrobial peptides and really this this is this is an example of taking a system and building up its complexity looking at effects of lipid charge saturation surface pressure and then going on to more realistic model systems then we looked at membrane binding of liquid crystal nanocarriers and electrostatics you would just you would imagine are important they were shown to be important but the effects of gravity were way beyond what we ever thought they really in taking a in taking a a mixture of lipid nanoparticles and in and interacting them with the surface the interactions with the surface above and below the sample were different right from the start so it really shows that in terms of experimental design if you have any sample with phase separation you really need to consider effects of gravity. And last the the interactions of the anti-cancer drugs that some initial studies have been conducted on interactions of doxorubicine and model membranes and we haven't gained insight into the way the drug works but we've gained insight into this mode of interaction with lipid lipid membranes and we've managed to quantify for the first time this extent of interactions so that concludes my talk I hope you've found it interesting enough and I'm very happy to answer any questions