 I know that you who are students in the course have a relatively mixed background so don't be shy to ask a lot of questions and I will try to keep my presentation to a maximum 45 minutes so there will be room for many questions and if you think it's completely ununderstandable just break in while I speak I have no problem with that so yeah I think the task given by Tommy is to try to sort of talk around the neutrons so what do we have to do to complement our neutron experiments what do we need to know before we go and what should we maybe do afterwards to get a more complete picture and I thought this is super open and very difficult to talk about so I decided to take one concrete example to show what did we do before we went to general for the first time and what did we do afterwards and what do the different type of studies what picture do we get from a combination of measurements so the title of my talk could have been alpha synuclein membrane interactions I'm going to talk about the protein called alpha synuclein and how it interacts with membranes and how we have studied this with different biophysical techniques and with neutrons so I will divide my talk in in principle three parts so one part which is the middle part is what did we do I will not explain anything about neutron where if Tommy was captain goes I think you all know that from other parts of the course I will just tell you what we did and what we found show some curves but that is sort of the middle part of the talk so this is the time axis here which is of course a very long time axis I think it's an axis of a decade so I will start to tell you what did we do before so what did we do to learn anything about this interaction before we had our first neutron time so that we would actually spend those I think we had 72 hours Tommy probably remembers exactly how many hours we had I think we had 72 hours and that sounds like a lot of hours but it's not so you have to spend that time really wisely so you get something out of it and then in the end I will also tell you what did we do afterwards and there will actually be two neutral segments so that will be like a neutron and then a little bit in between and a new neutral segment and just to make sure that you don't believe I have done everything even if I do work a lot in the lab I have really not done everything in this project I have done I don't know what I have done I have done some protein expression and purification and a lot of sample making when you are at the neutron time it's a lot of intense sample making because some of our samples are time-dependent and we have to prepare them fresh before we do the measurements so that we know something about the conditions of the samples and many PhD students Eric Heldstrand I think was the first Ricardo Gaspar, Tina has not worked so I remember in my mind but she has done a lot of other things so he will see some of her work Simon Fridol, Kasia McExerich and Maria Dubakig and June Palbo-Arbizon these are the PhD students Marie Kre was a very early postdoc in the project she was also there in the first measurements in Genov and Eric was too and then Eliak we mean a few years later as a postdoc and Katja has been helping us with a lot of alpha silicon purifications and validation of the quality and then there are some other seniors besides me involved in the project and of course collaborators and many more than I have listed so if you listen and your name is not there yeah I hope you are not too much offended because there are of course a lot of more people involved both in Lund and elsewhere in the world and has helped in this project so the project on alpha-cynuclein membrane interactions is part of a wide project we run in Lund and in collaboration with others where we try to understand liquid protein co-aggregation or co-assembly and we call it liquid protein co-aggregation from lipid rich to peptide rich so you may have co-assemblies that are lipid rich like membrane associated proteins here's a lot of lipids relative to the amount of peptide mass in the co-aggregates that you can call this if you want biological membrane there are other features that are relatively liquid rich that are these lipoprotein particles that are present in your blood especially after having had bacon and cheese and stuff like that for breakfast you get an enormous amount of these lipoprotein particles where protein serves to carry the lipids so you would not die from the massive lipid insults it makes more to soluble particles of the lipids and then you have proteins like albumin that you have in your blood which in principle interact with everything it also interact with lipids so you can find crystal structures of the protein with like eight lipid chains inside and then going to even more protein rich we have amyloid lipids that somehow interact and co-assemble with lipids and then you could have these lipid transport proteins for lipo k lipids or like maybe only one lipid per protein in the complex so there's a whole spectrum of ratios of liquid to protein that can exist in co-aggregates and of course many many more examples than those in on this slide and the project we are working on alpha-sanuclein membrane interaction is motivated by the literature and there are lots of reports in the literature so in the beginning maybe I should tell you that in the beginning we wanted to understand the aggregation of this protein which is linked to Parkinson's disease but there's also a lot of literature saying that the aggregates that form in biology contain lipids so it's not only proteins this is some kind of fluorescence microscopy where lipids and protein have different color and you can look at if they are co-co-aggregated or not and there are other other studies also so there are quite a lot of studies where people have tried to look at the composition of those sort of plaques and deposits you have in the brain but what I will talk about here is not so much this amyloid side where we some often have the hourglass or the looking glass we often have the looking glass here also so in this part I'm going to talk about reaction on this side so here we look at protein interaction with membrane so this is more like a lipid rich domain that we're looking at so we study the interaction between alpha and nuclear which in principle is this order protein in solution with phospholipid membrane surfaces and we ask a number of questions some of which can be addressed with neutrons and some cannot so we ask like questions like affinity what is the distribution between free and bound that is affinity, stoichiometry that means how many proteins per lipid or how many lipids per protein I guess you want to say because the lipids are smaller than the protein so how many lipids per protein are there when the surface is saturated so what is sort of the maximum ratio you could have or what is the what is the condition about which you cannot bind more protein to the membrane and then binding mechanism you may want to know do these proteins if there are many of course there would be many on a big membrane are they bound to be independent or one another or is there some kind of cross talk we want to know what is the structure of the bound protein and what is its penetration depth in the membrane does it bind like in this cartoon in the upper acyl layer or does it penetrate in or in the head group area or does it penetrate into the sort of more hydrophobic acyl layer or maybe it even goes through the membrane some of those questions you can address with neutrons we also want to know what are the consequences for the membrane does it disrupt does it thin does it swell that you can also also look at with membranes and you may want to know what are the consequences for the protein the protein may change conformation as it associates with the membrane and it's also relevant to ask does this process have anything to do with the self-assembly of the protein to form amyloid fibres and is there any uptake or napkins that's another question you may want to know and of course there is this difficult question which i think i leave to others i have to confess that i'm a physical chemist maybe i'm also biochemist i don't know but i don't do biological studies so i leave this question to others but it's of course that's what you want to know of course you do all this physical chemistry to address questions that are probably relevant to biology but i can't do those studies someone else had to do them so let's start with one of the players we have two players here we have a liquid membrane and we have a protein so the protein alpha-synuclein has a very i would say i guess you can see the whole protein i see us on top of the protein but yeah i will move us like that just to be sure the protein has an unusual amino acid sequence so here red is negatively short and blue is positively charged and orange is hydrophobic and white is hydrophilic they are the black here hydrophilic parts so sometimes alpha-synuclein is said to have imperfect repeats so there are a number of of short segments that are similar to one another but they are not exactly identical and another very distinct feature is the asymmetric charge distribution so if you look at the negative charges you have an enormous accumulation of negative negatively short residues residues that are negatively shorted neutral pH i would say they are all not all but many of them are in the C-terminal region of the protein so somewhere after residue 97 and up and there is really there is the last lysine and then it's only negative and then you have a core that is super hydrophobic there is a like these two are matching each other plus and minus so this is zero charge for 35 residues and that is called the knack domain and knack stands for non-amyloid core and this is exactly the place that forms the amyloid and that sounds like a silly name but it's called non-amyloid core because it's not amyloid beta so that's how it got its name and then this N-terminal region is what we call anti-phatic so it's like hydrophobic but there is quite a lot of both positive and negative charge residues I think there's even more positive than negative so this one is net positive and this tail is net negative so that's the protein it's a relatively unusual sequence and when it's alone so to say in water solution or in wake buffer it's it is more more or less random it doesn't have any preferred structure so now I would probably move us again over here where we were and what do we do to get this protein so we need a protein for ourselves it's even smaller we express it in ecology so many of you in the audience do that not all of you but when we make our protein we we take the gene we make actually synthetic gene with E. coli optimized codons it's with E. coli we just love to express it's a tRNA composition with much the sequence and we make it with no tags and I think that is very important so we just make this sequence starting with this methionine ending with this alanine so many people who do proteins add tags to them but that is very problematic because tags may help you in the initial isolation step it's well very seldom gives you a pure protein there are other ways to isolate protein and more often the tag you have to clear off the tag because the tag will change the properties of the protein so before you do your studies you have to clean it off but that's often expensive because proteases are expensive especially when you need large amounts of protein and then you need to purify away both the protease and the tag so it actually makes the purification procedure often take more time and cost a lot more so I always recommend express as is and then use the physical chemical properties of the protein to design a purification protocol and in this case we use this fantastic property of the protein being resistant to boiling so after we have laced the cells we just boil the lysate and centrifuge boil cool centrifuge and then we get rid of almost all the economy proteins and then what stays in the supernatant we run through two anionic steps in sequence one that is a coarse pass flow a poor resolution to capture it's more like a capture release step and then we do another one that is high resolution slower where we get the protein really pure and then finally we do even size exclusion to make sure we have the monomer because in this case we want to add a monomer to memories and not aggregated state in our and in other studies we may want to monitor in education and then we need to know that we start for monomer and what we then do after all this purification is we run it on sts page gel and these are just where different molecular weights would run the smaller they are the further they go down down in the gel and this is our purified protein so this is what you want you don't want to see other bands there and this is relatively high loads you have to add high concentration of your protein to be able to judge that there are no other proteins and my recommendation when you want to do any kind of studies is always purify until you see no other bands on silver stained sts page this is comacity you do it even better if you're silver stained because then you can see if there are any weak bands and you don't want small molecule contaminants and that you can easily see by nmr's the trust of other things in the solution that you're protein and you also have to validate that you actually have the protein you think you have by mass spectrometry so intact weight but also fragment just to be sure it's the correct protein it has happened to me that i once got the nicole protein instead of my overexpressed protein and that's why it happened to have exactly spot on the doors on the same intact molecular weight but when we finally did it it was clear that i had purified the chlorine clinical resistance protein and then just start over again you have to do to get your protein so you really have to do this so you know that you study and my advice is that i mean you may study protein you can buy i know lots of people buy proteins but then you do the same you purify it because they are never pure they're often sold very impure there may be 90 so that's a 10 other proteins and then there's a lot of small molecules that they don't tell you so you really have to purify before if you use a protein that is from that reasonable cost in any kind of catalog so that's about the protein you have to be very careful the other player we have are the lipids and in this case when you study protein membrane binding i think you have two choices one is to extract biological membranes from cells or tissues and then purify them which means they will have more or less a natural composition of lipids because the biological membranes are relatively complex but you may also then have other proteins yeah depending on how you purify the lipids so you will then of course have a less pure preparation and you will have the variation both in lipids and acycines on the lipids so that's the disadvantage but the advantage is that it's more representative of a biological membrane although when you purify and then make often liposomal vesicles out of these lipids you often scramble them so biological membrane may be asymmetric but then you make a symmetric one so we often work with lipid model lipid models uses a model lipid membranes where we have defined components and for example zwictorionic which means it has no short zero or negatively short so this is called deolium phosphatidylcholine and this is called deolium phosphatidylcholine diolium means that there are two olio chains which have 18 carbons and one double bond and then we also work with those lipids that are called ganglocytes and I think this is gene one which have they have often a little bit of a mix of acycines because they are purified from actually from I think brain tissue or maybe some other tissue and then they have this very peculiar head group which is huge and sugar-based but this is also interesting and relevant because they are found a lot in neurons where when our vicinity is present so this is our model model lipids and then we mix them in defined ratios so we can have 100% this one get a 100% this and we can have a defined ratio we can have this one and this one so we make them at will with the composition we want and then and you often then drive them as a film and disperse them and and you make basic as proof explosion or sonication depending on what properties you want or electro formation so I'm going to start now before we go to the neutrons to tell you what did we do before so what did we learn before we I think it's even so that you cannot get new from time unless you can show some preliminary data from other methods that you actually know something about your system so after expressing and purifying it we studied the adsorption of the protein to membranes with different techniques circular dichroes and spectroscopy quartz crystal micro balance and confocal fluorescence microscopy so the first method cd i this i would just tell you like in one minute what it is but you use circularly polarized light which is one component of plain polarized light so if you have plain polarized light which is this hour in between it tooks that observates up and down as it propagates you can mathematically divide it into two components and you can also do it by a popular cell if physically divided into two components or one or the other so one that goes in this direction er it goes from here here these are different time points and then you have the other component which is left hand circularly polarized but the some of these two components is always the plain polarized wave and what you do is that you take those two components either you take plain polarized wave and you measure the tilting of the polarization direction or you alternately excite your sample with those two waves and measure how they are differentially absorbed so what happens is that if this correct which has your sample contains chiral molecules one component may be more absorbed than the other one and they may even come out of your face because it's also optically by refringing to this contains chiral molecules and in principle in when you do circular dichroids and spectroscopy there are two ranges of course there are many ranges but there are two ranges relevant for proteins one is called near uvcd which measures spectrofoam aromatic side chains and disulfide bonds and the other range is called far uvcd 185 to 250 nanometers as you say here where you look at the peptide backbone so the peptide backbone form force and you get different types of spectra and they can be quite I wouldn't say feature less feature full lots of little peaks but still relatively feature less given like you have a protein or 140 residue and if it forms an alpha helix you get like two negative peaks and the one positive peak I'll do all of that but the important thing is that different secondary structures give distinctly different spectra and that you can use when you study proteins and you can definitely use this method to study changes in structure so this way employed for alpha celibate and this is what it looks like so the black here is protein alone which looks like a random coil you could go back here it looks like a green spectrum so when the protein is in solution it's more or less a random coil and then we study different lipid to protein ratios so the more towards red we go the more liquid we have for protein molecule so you can see how the spectrum gradually changes to one that looks like an alpha helix and if we take and this has no attunement this is the this is the signal at 222 nanometers which is one of the distinct negative peaks for alpha helix and if we plot that ellipticity here you can see how it goes down more negative as a function of lipid to protein ratio and it clackles somewhere here so that gives us actually a step geometry this curve doesn't give us affinity it tells us that affinity is high because it's a sharp inflection point relative to the protein concentration used here but at least we can see that when membranes are added in the form of liposomes to the protein protein adopts a helical structure and the amount of protein absorbed to the lipid seems to saturate somewhere around the lipid to protein ratio 160 to 200 somewhere here this doesn't tell us at all which part of the protein forms the helix or if it's all of it and then some of it of some fraction of time all of it and some fraction of some none of it because this is just a bulk and time average of what goes on in the solution so we don't get any detailed information but we can tell that helices are formed somewhere in the protein and the other technique we use is called quartz crystal microbalance with dissipation where you in principle have a cantilever with some kind of coating and you measure the vibration frequency of that cantilever and also the damping and that will change I mean this is in principle a little surface if something absorbs on the surface it will be more heavy the frequency will go down and also the damping may change depending on the structure or the absorbed layer here so you can measure in principle something that reports on the mass bounce and when you can inject something and then can inject buffer it goes up again but you can also get some type of structural crude structural information from from the damping and in the case here we deposited a bilayer on the sensor and then we flew over so this is the alpha-cennuc can come on in here over the membrane so we can study how it goes on and off and we actually mainly saw it come off because it could be very slow and here are data for protein so the frequency here is the blue goes down overtones goes down when we inject the protein which means that this cantilever gets more heavy something in the bottom and the damping also changes and here is the same thing at another pH and these are the charge these are the negative charge members we have both the sweet ionic and the negative charge in the mixture here and we repeated the same thing oops i went backwards i should go forward we repeated the experiment with the sweet ionic memories and then we saw no very little absorption so we can really say here that we need some fraction at least 10 to 30 percent of the negative charge lipids to see absorption of the protein which of course is a bit intriguing because the protein as a whole has a net negative charge although i think you saw that it was relatively polarized in its charge distribution and these are just some results from confocal microscopy where we use one flaw for in giant unilamella vesicles so there are many micrometers big so we can see them and we use another flaw in the proteins in principle the membrane is red and the protein is green and so we can actually see here in the microscope that we get co localization of the protein onto the membranes so to summarize what we know before we go to the difference is that the protein absorbs to an ionic memory so that's important so we wouldn't waste time at the neutron beam using sweet ionic bilayers that would have been a bit stupid unless we want the negative control and we can use this one and then set up the exact same experiment so we can do binding studies if we want to know how the protein absorbs and what is the structure of the boundary using these membranes this we cannot study by neutrons but it's good to know that there is helical formation it's not only us not lots of people have reported the same and this we actually see by other measurements i'll show you a little end in our data in the end so we actually know that this is the n-terminal part and others have shown also is the n-terminal central part of the protein that contacts the membrane and this c-terminal part is actually relatively disordered and extends into solution so that may explain actually why it can go down to negatively short membranes we use the part of the protein that is actually net positive and the negative part sticks out from the membrane so what did we then do at the beam we did both neutron reflectometry and scattering and we wanted to know what is the structure of the right bound protein especially what is the penetration depth so i think this is the main question we could address what is the penetration depth for the protein and so how would we address these with neutrons and could we address the same with other methods and what are the consequences of this so we use this technique called contrast matching so we have we do the reflectometry of a neutron beam and then we have surface and on the surface we have deposited our membrane and then we can of course also add protein so we used either protein or lipid deutrated so that we could in H2O we will see both but we can then do what's CMSI where you can actually cancel the lipids or you can use D2O to cancel the protein so you would actually see the structure of one or the other in these measurements and i think these are Thomas little not so little actually as a protein chemist nowadays you typically work in the microliter max millilitre scale and then we get these devices that require 25 millilitre protein and that is when you're super happy that you have a clone you can express it yourself you don't have to buy it you don't have to buy proteases you express it and is so you can actually scale up your protein expression and purification without too high costs and of course I should also say that we got fantastic help from ILL they have a deutration facility so we didn't have to pay for the deutrated protein they express that for us shipped us the cell pellet and we then purified it and learned so that's fantastic service because deutrated media deutrated chemicals are of course expensive so here are some examples of our data so we did all the scattering in three different contrasts and I will not go into details of this how it was recorded or fitted because I know you have that in other parts of the course but in principle what we got here was we used this bilayer slab model we could get the thickness of the membrane and we could also get if there were any kind of features in in the membrane so when the protein is there you can actually see that this becomes asymmetric and what we learned so I will tell you what we learned from this study what we learned was that the protein actually is at the very top of the membrane it doesn't go into the hydrophobic region it doesn't go down it doesn't make a pore it really just sort of absorbs so absorption is actually a good term for this binding interaction it absorbs in the upper layer in the head groove area maybe a little bit into the upper acylane but definitely not very deep into the membrane so that is what we now know if we summarize what we know after the pre-studies and and the neutron studies so we now know that we absorb the anionic membranes we know before the helix we know only n-terminal part contacts the c-throsis we knew before but now we also know this we know how the protein is located in in the membrane in the in the head groove and upper acylayer and not penetrating any deeper and that is very important and we also put some indications on that from confocal because we never saw the protein inside the vesicles so it really didn't go through and was not released on the other side so now I tell you a little bit about what how we have continued what more do we now know about alpha-cellular membrane interactions so from NMR spectroscopy which is still not published we can run NMR spectroscopy data from HSQC NMR where you in principle get one peak for every residue in the protein so you have N15 labeled proteins now it's a different isopropyl it's not due to us anymore and this should say spec-prosperates always misspec and then we do that different liquid to protein ratios and I think the blue is the free protein and you can I think easily see how this region the first hundred residues disappear like at the high liquid to protein ratio where we you remember we have saturation somewhere between 160 and 200 so when we are at the saturation limit where in principle all protein is bound we don't see any of the first hundred residues but when we are below saturation we actually do see depending on the ratio some of those residues at weak intensity but not totally lost so what this tells us is that the protein actually absorbed with different parts so this cd couldn't tell us cd spectroscopy could just tell us that we form helix but combined with this we can say that when we have less liquid to protein the protein does what it can to actually be on the vesicles but only a small part interacts with the vesicle the part that's gone in the NMR spectrum and when you then increase gradually increase the amount of lipids the protein absorbed with a larger and larger fraction formed helix but never more than the first hundred residues because of this this negative tail cannot form helix on the membrane so that's very important information that the binding mode or what you would call it the bound structure changes as a function of the lipid to protein ratio but what happens when you go in excess if you have more lipids or we have more membrane surface then so that you if we actually have absorbed all protein and then add even more membrane what happens then so these are data also from kassar martin sherry very interesting data so she did helpful microscopy in bright field mode where you can see a bunch of vesicles these are just two different areas or the sample these are giant you know lamella vesicles again so you can actually see them scale bar here's 10 micrometers and the super interesting observation she made was that she could see many vesicles in bright field and when then she looked in fluorescence mode where the protein is now the only player that's fluorescently labeled only some of the vesicles were lighting up and they seem to have protein around their whole periphery or whole surface so this is like a cartoon showing there was this vesicle but there was no fluorescent and this is another area where some of the vesicles like this one is fluorescent this one is fluorescent but this big one here is not fluorescent so that was a fantastic and very intriguing observation and what you often see it seems like you often see the old small ones have proteins big ones have not but this is definitely not the carburetor effect because even a five micrometer gv or in these are almost 10 here the protein the protein is here and this is the vesicle so if you look at the protein the protein sees the membrane as a flat surface at this size of vesicles it can't be a carburetor effect so there must be another explanation for the small small or some vesicles being filled first and what Kasia did then to make sure it's not liquid segregation between vesicles she also did just pure pure pure vesicle and saw the same phenomenon and also when she over titrated with protein all vesicles had bound proteins it was not so that there was something strange with some vesicles that they couldn't bind protein it was just that the protein very unevenly distributed when there was excess membrane surface and the conclusion of Kasia's study is that this is due to cooperative binding of the protein to the membrane so cooperative binding means that the protein any incoming protein rather binds to surface where there is already protein compared to the independent binding situation where it would bind to any vesicles and they would be just randomly distributed over the vesicles there's not a random binding it's independent binding this is random this is random because it's random events but if it's cooperative then they accumulate on some vesicles whereas when it's independent they just spread out according to the Pascal triangle and this is some simulations or calculations that Kasia did showing some of course these are fake systems if you have only two binding sites or if you have 10 binding sites per vesicle the vesicles of course are much bigger but what she shows here is that if you have two independent binding means that the half saturation half of the vesicles have one 25 percent have two and 20 percent have zero but when you have exactly half saturation whereas when you have cooperative binding you get less of the half saturated and most of the vesicles have either zero or one so sorry zero or two but but if you go to 10 binding sites this is the independent binding showing all the different states where for example 1 2 3 etc approaches boundary vesicle then the sort of cooperativity although she uses the same free energy coupling between binding events in the 2 and 10 case you can see that when you have 10 coupled binding events you in principle totally suppress any intermediate and you get only vesicles that are completely empty or completely filled and of course a real vesicle may have 1000 binding sites so it's the explanation for this sort of seemingly super strong cooperativity is that it doesn't have to be stronger for binding event but the fact that you have so many cooperating cooperating means that it can actually completely suppress the intimates and yeah i think i just have a little bit more time we also looked at vesicle shape and that's one thing you can also address with cryo but you can also address it with neutron scap then you can't have the depositive membrane so now we instead have vesicles in solution again the constant contrast matching and we want to know is the protein bound with this that we saw on the flat membrane does it go in does it penetrate and mainly we want to know okay are the vesicles deformed when the protein is is bound and we do this trick or contrast matching so we either match out the protein and see only the vesicles or we match out the lipids and see only the protein so we can in principle have a situation that looks like this when you have protein only and vesicle and of course the vesicle is not so big to piece the entire cube back this is a fake image um yeah so what we did here again was this contrast variation uh and we also of course could have been already in the validation what you have to do also when you do this contrast matching status you have to also validate what is your deuteration percentage of the deuterated protein it's not always ideal to have it 100% deuterated in this study we wanted 75% because then it's easier to match out and then we did a mass spec so we have the sort of molecular weight and for the molecular weight we could calculate how many deutras are there on average or how many protons are there on average and then we know the number of residues we know the number of growths that can exchange and then we could actually find out how how much and we aim for 75 and got 76 so that wasn't so bad and then we looked through neutron scattering at the vesicle deformation and the best fit to our data is actually that when you have at some protein liquid or protein ratios you have to get a deformation you need to use an elliptical model to fit the data you can fit it and along with a sphere of the model and that has also been seen using cryo em where you can look at vesicle shape so I will almost stop but I will just tell you also that we don't only look at alpha synergy membrane interaction we also look at the formation of aggregates fibrillar aggregate because that is the process that's going on in Parkinson's disease although nobody knows what's toxic it may be something intermediate we're still very interested in the formation of aggregates and in the formation of aggregates are lipids taken up in this aggregates and we want to know a lot about these things we want to know about their composition we want to know those factors is there any selectivity in lipid uptake is there any kind of mechanism for lipid uptake and what are the consequences so there are lots of questions again and I just want to show this because this is not neutrons but this is one of the coolest confocal images we have again with the red-labeled lipids and if we add aggregates we can see how aggregates accumulate in certain patches of the membrane and the membrane I mean this looks like they have actually taken up lipids and then if we continue and add protein so first we add an unlabeled protein then we add a label protein then we can see how the label protein actually adheres somehow to where there was already a patch or absorb proteins and how you get lipids incorporated in the growing aggregates so that was kind of cool and I think I should stop there the only thing I will tell you actually and then of course you do contrast matching again and try to know something about the structure I will show you this because this is something irrelevant for membranes maybe but probably it is because when you pack all these proteins close on the membrane you probably change the properties of the protein because there will be an electrostatic interaction between all those negatively charged proteins and but in this study that I show you instead looked at what happens when the protein alone without lipids formed with fibroids this is again the sequence and we can forget the mutant for now but this is again the sequence just different color coding showing again this super negative C terminal tail these are models of fibroids that people have acquired so these are taken from the literature but in principle you have a core which is this part and then you have the N terminal tail and you have the C terminal tail and unstructured appendices on the on the fibroid so we asked here what does this mean to the PK values of these acidic groups will they be more prone to actually accept programs when they are packed close together and the experiments that Tina so Tina Palmer doctor did almost all of these experiments she did three types of experiments and one was just using a pH electrode so the wiretrap which is here she formed monomers she desoldered monomers and then she measured the pH of the monomers in solution in water when there is no buffer the only buffer is the protein cell you end up somewhere here 5.6 and then she left fibroids formed and then after fibroids had formed she measured the pH again and then the pH had increased by 0.9 units to 6.5 just because of the self-assembly of the monomers into fibroids she used pH indicator where you get the color change when the fibroids changes pH changes or when the fibroids form and she used NMR spectroscopy where you use the fact that not all monomers are incorporated in the fibroids so there was always some there's a solubility high enough that there are always monomers left in the fibroids and then she looked at the histidines or the monomers in solution we don't see the fibroids you only see the monomers in solution and from the chemical shift change of the histidine she could also calculate the pH before and after so all of this ended up in in a situation where pH increases as fibroids forms that means the proton concentration in solution has gone down because higher pH is less protein lower pH is more protein so something must have taken up protons and that is the fibroids that form they take up protons from the solution because their pk values are effectively pk values are elevated and from the pH change and from knowing how much monomer was left in solution which actually changes its protonation state also because of the pH change so that is like the intrinsic bubble she could calculate the pk value change upon fibroids formation and that was like 1.1 units so that's quite high and after this i will actually stop talking so that you have a chance to ask questions okay i think there is a question already from Radhika who raised her hand so please yes um hi Sara thank you so much for this presentation it was very interesting so i have one question one is that you probably mentioned it but i i guess i missed it like how when you made the vesicles and the lipids and then how did you make sure that it's like a proper bilayer structure yeah so that's a very good question and i didn't say it so when we made the vesicles so we made the vesicles in place at iLL so we made them for extrusion through membranes and then we did actually DLS so they have equipment there so that's also important to know that when you go to neutron source you have to book in advance what other equipment you need so we measured the size distribution and polydispersity yeah so the average size and polydispersity of our vesicles onsite onsite yeah so exactly the ones that go into the cuvette are carried to us all right if i could ask one more so um as as as as you as you've understood that they are just the proteins are just adsorbed they don't penetrate too deep inside the bilayer so in that case would it be possible to come up or target them in such a way that you just kind of can just like take them out from the layer like if they've aggregated too much and we don't want that aggregation because they're like not too deep they're not penetrated too deep inside so in principle would it be possible to kind of take them out and kind of target them like that yeah so would they so i guess the question is would they ever dissolve from the membrane and since it's an equilibrium they will of course be on and off but it seems like the off weight is low but in principle you could if you take away the monomers that are in solution in equilibrium i mean you can principally have them wash you can principally have a bilayer if you have a bilayer deposit and you wash with a stream then you could watch the protein coming off that's one thing you can do but if you want to force them off you can actually add i think there's a thing called proteases i mean let's say you don't want them for purification because they're expensive but in this case you would need very small amounts you can you can start to digest the protein sort of from outside and then it may dissolve right but then that is that is possible chemically like you can add proteases and make them de-absorb chemically but then what if you if you want to make them de-absorb biologically yeah and if you want to do any kind of this treatment that's more difficult but then you also have to keep in mind that and i didn't say that so much we believe that this cooperative binding of the protein relates to the healthy function of the protein which of course is less known than the disease property of the protein or the protein here we have it so if you have this property or property binding so if you think about a membrane and if you want to bud off a vesicle if you bind independently you will be distributed over the whole membrane but if you bind in patches which this would correspond to if the surface is excessive and more more like a flat biode then you would bind in patches and then the patches could then possibly bud off bud off yeah so you probably don't want to interfere too much i think you think it's it's believed at least to be related to synaptic transmission and and you have a lot of others in the synapses but you would like to interfere with the aggregation and aggregation is actually liquid induced so if you have a bilayer that's completely full of protein what happens is that if you have excess protein so not this situation but when all the membrane is coated if you have exactly the protein it seems to be the excess protein that nucleates on the vesicle surface on the protein that's there so then you may want to learn more about the interaction of additional layers so to say or the more weaker interaction of additional layers of protein yeah so that could be the next question to address how are they interacting and how could we possibly interfere with with that interaction hmm don't take the question thank you okay there is another question from Nicole so please yeah hello Sarah thank you very much hello Nicole I haven't seen you in many years it's nice to hear your back thank you and you have been in fiskian for a long time i know yes yes thank you for a great presentation i have question about the neutron reflectometry study so i don't remember if did you use one type of phospholipid in this study and if you use maybe like two different type of phospholipids can you distinguish like where maybe the alpha sonicline is absorbing preferentially like if it is it one phospholipid type or is it like do you think there is a way to distinguish um yeah this is a very good question uh and tommy may correct me if i'm wrong but i have a memory that we used both ps pc and cardiolipid pc yeah and cardiolipid was chosen because that's found that mitochondrial membranes and it's also believed to have a role in Parkinson's disease but we didn't see any difference when we changed so the only difference we saw that we needed a certain amount of negative charge but if there was if it were to penetrate deeper i think you would see it and tommy would be the best one as how much difference would there need to be to get a significant difference in in in the curves so i guess because there are one ocean that's nothing but ten oceans is it there must be a limit how much they would need does it need to go to be recognized yeah and i am thinking like if you mix like two different types or three different type of phospholipids can you then in reflectometry measurement distinguish like what is maybe preferential for the alpha sunicline that's also good questions i think you can see if you have if you have different types of lipids you could do it with one and not the others and so then you would in principle do the measurement three times with different liquids uh visible or different liquids invisible but you probably want them visible one at the time and then you could see because there it could be so that you get you call some kind of asymmetry of the membrane that where the protein is bound maybe one is small common so that is a very that's actually a very good question therefore i could request um because because i will be working with something similar with emma so we will also try to see like the how do you say the how the phospholipids are sorting in a membrane this is why i asked so thank you very much and i think it would be super interesting it is maybe one wants to do that with vesicles i guess because i think it easier exchange of liquids in the in the deposited bilayers but i may be one there but i think if you do that i think it would be super interesting and and especially it will be super interesting with the ganglucide lipids that seem to actually be preferentially taken up in aggregates and is it so that they are if you have a i think if you create vesicles they will be symmetric but then when you add the protein maybe then they become asymmetric in liquid distribution that would be very interesting to know thank you very much for your answer and hope to see you soon i hope so i hope there would be a normal life soon so we can actually see each other okay there is a question also from victoria mecklesh do you want to post it in person or yeah yeah thank you sarah for the presentation i was wondering if you studied the secondary structure of alphas and nucleon with f t i r spectroscopy also mm-hmm oh we did not what others have done but we did not mm-hmm and the other question was i was wondering is a cooperative binding is it because of the electrostatic repulsion between alphas and nucleon and membranes and also if you can study this with afm or i don't know surface force apparatus it's it might be very difficult but yeah but it's a very good question and that's a question we would like to answer the why is it cooperative and we have many speculations but we have not done any experiments yet but so it could be either in cooperativity means that the protein that absorbs the membrane rather absorbs next to another one than on the bare membrane so there is some synergistic effect it's higher affinity next to another protein so there must be some kind of attraction between the protein molecules when they are in the membrane and if it's electrostatic it has to be then that they somehow are organized so that the n-terminal part of one is closer to the z-terminal matter one but you could also do the hydrophobic interactions because the hydrophobic interactions are of course responsible for forming the alpha helix which is very what you call poloic it has a hydrophobic side and the charges are on the upside that there is maybe some hydrophobic also on the sides that may interact we don't know yeah but there are many ways to address this either to try to do more measurements and another way is of course to do mutants or the protein where one thing oh it's probably due to those reasons but when you take them away the problem with mutation studies is that they often open more questions than they answer so you you make mutants to answer one question and you're but I think an important very interesting part to go is to actually look at the familial mutants so there are mutants that are prevalent in families that more easily get part of his disease maybe they have something to do with making this healthy function less and I think we can't actually make quite a lot of them so we still have in our freezer a little collection of of mutants and we have also ordered clones so that we can select the clone for a label some of these mutants so so that's one way to go is to mutate certain types of residues and see if it changes but one can also vary there is also physical parameters like temperature and salt that's cheaper than making mutants over cheaper in time especially because you can vary the hydrophobic effect that is very temperature dependent and electrostatic effect you can swing your salt so there are things that tricks you can play is there anything you can do to get the cooperative to go in a way that would also help to understand what it's due to yeah thank you