 Yeah, so good morning everyone. So I'm Lawrence Reynolds working in CNRS in Montpellier and I'm a soft matter scientist, experimentalist and I started working on let's say food-related materials a few years ago. So I wanted to thank the organizer for this setting of this nice meeting and giving me the opportunity to put on this work and this work is about gluten and I try to show you how we can get very useful information by combining sites and sands. And I will show you pieces of work of a bunch of students and postdocs, Mohsen Daesh, Yusin Pesnay, Salvatore Consenso, Amar Luishi and all this work has been done with my two close collaborators, Thors, Marie-Laine Morel and Amélie Bon. So what is gluten? So gluten is the protein that we get, the storage protein that we get from in the grid grains. So we you probably all know gluten either because you want to avoid it for a health issue or because you want to take advantage of very unique properties like being a very viscoelastic network that is really an added value if you want to make bread or croissant in a very efficient manner. And it's also a very fascinating material for a scientist because there's still a lot, it's unique in the plant world, this kind of viscoelastic network and there's still a lot of questions regarding the relationship between the structure of these materials and their mechanical properties both in the linear regime and especially in the non-linear regime. So a fascinating subject but very complicated and it's very complicated because it's a huge mixture of several proteins, more than 60 types of polypeptides and which can be classified, let's see, can be classified into two main families, gliadin which are essentially monomeric protein and glutinous which are units which are linked together by disulfide bond and you can reach very, very long polymer to seven million square per mole. And you see here this domain in blue, actually there are intrinsically disordered domain made of repetitive units and they provide a lot of flexibility to this protein which in fact be as more like polymer than like colloid. And on top of that this gluten are insoluble in water and you can see it here in these two images so here you just have gluten when you add a little bit of water so you get this very viscoelastic mass but if you add more water you have a phase separation you cannot swell this network so which this kind of material are not easy let's say to make a rationalized studies because you basically you're stuck with one concentration. So to overcome this limitation we choose to work on a more model like system using a good solvent-possessed protein which is a mixture of water and ethanol. So the extraction procedure we have developed is based on gluten, this industrial gluten that we first dissolve in these water ethanol mixtures and we just keep the soluble part because there is one part where the polymer are so long that they're not soluble in any solvent. So we keep this soluble part and then we take advantage of this liquid-liquid phase separation that you could have when you crunch the sample and what we found is that depending on the crunch temperature you can actually control and tune the composition of your protein extract and you can vary the gluten in content of your protein extract from virtually zero up to something like 70 percent and what you see in this kind of images here is that with this model protein extract you can get nice and homogeneous samples over a very broad range of concentration for very dilute here it's then really going to melt till a very concentrated system and you got from liquid-like samples up to gels. So you have homogeneous samples which are let's say more convenient to investigate experimentally. So the first thing we did was to look at what we have in a very dilute regime. So we use asymmetrical flow-field flow fractionation in combination with multi-angle static and dynamic light scattering and we found that we have different three families of species in our materials. We have what we expect because we put monomers which are in fact the gliadins, so small molecules of a few nanometer size. We have also the glutamine, the polymer that we put inside where we saw behave as polymer chain in good solvent condition and with intermediate molecular weight and a few tens of nanometers size and what was interesting to us also that we have a third families of species which are assemblies. So they are big objects of about 100 nanometer size and they have a very large molecular weight something like 30 million so much larger than all the species we put in our solution and we can show that they are made of they have a polymeric branch and loose structure. So in fact they are assemblies of different kind of proteins. And what was interesting is that if you look at these assemblies the amount of assemblies is correlated with the amount of glutamine we have in the sample. So we are really interested in this assemblies and we were first asking whether we could have some hint of the presence of these assemblies if we go to more concentrated samples now. So this is just a schematic and very rapid conclusion. So this is a phase diagram so we have a one phase at high temperature and then a two phase version when we point to sample and if we quench to an intermediate temperature what we see with live microscopy is this development and coarsening of structure which are in fact spinodal decomposition and using USACs, Ultrasmolangal X-ray scattering performed at ESRF in Grenoble in France. And as a result we can really follow the the characteristic size of this of this structure as it develops with time and show that it really behaves as spinodal standard spinodal decomposition. We also found that if we use a sample which a larger quantity of glutamine is a more viscoelastic sample we can have also small anomalous spinodal decomposition but I don't have time and I'm fortunate to talk more about that. What I just wanted to say here is that if we quench further the system and not that's up here but quenching deeper what we see that the droplet that forms here we have another phase separation inside and which is due to the hidden phase diagram of gliadin that we have at lower temperature. So all this to say that the fact that we have this multiple liquid-liquid phase separation is a hint that we also have the presence of this big object assemblies at first separate at relatively low temperature. So the question is do we see more directly these assemblies in these kind of samples? So the first thing we do we did was to look at the scattering. So here you have a curve of the scattering profile in scattered intensity as a function of the wave vector and we combine different techniques wax, sacks and very small angle neutron scattering this time to cover very broad range of concentration and what we see in fact and I'm going to detail a little bit this curve is that we have different features with all leaders to the conclusion that it's really the scattering of a polymeric network in good solvent condition with some large scale heterogeneities probed here. So what we see is that at large quills of small distance we have the rod-like structure of individual protein and then we have up to the cure which corresponds to the persistent lines and then we have the random work of this rod-like molecules which is exactly what we observe with these minus five cells cell power law which is what you expect and predict for a polymer in good solvent condition until you have the crossover when you reach a semi-dilute regime and you get the correlation length of block size and then at smaller quills you have this power law decrease with the exponent of minus two which show you that you have fractal organization at very large length scale and heterogeneities at different scale but we don't have any clear signature of these assemblies up to the moment and the surprise went when we did small angle neutron scattering so for some for scattering whatever it is neutron or x-ray you expect the scatter intensity to be proportional to the form factor the scattering factor and times this factor which is delta rho square which is the contrast between the scattering object in a solvent so usually and it's a standard way to avoid incoherent scattering you replace in your sample when you do neutron you replace your hydrogenated solvent by your deuterapid solvent and therefore what you expect you just change the contrast so instead of having this contrast of your let's say your polymer gel in hydrogenate solvent you will get this contrast but basically you expect the same scattering profile with simply different intensity because you have different contrasts and what we observe is totally different so here in light blue you see what we have observed what I showed you before where you have this fractal like evolution at very low q so at large length scale and here we recovered the the polymeric like structure we have at large q so small length scales and but when we do this neutron scattering what we find is that also at small length scale we see like the signature of our polymer in good solvent condition we have a totally different system here with this q minus four dependent and this whole of a small q and of course you could say okay you change h by d so you might change the interaction if your material but not at all in fact if you look if you look we did also using the dissolvant in sax so this is this we got the same system if you do sands with the etch solvent you got this so it's only when you have when you do neutron scattering with a deteriorated solvent that you have this totally different scattering profile so what do you see if you do if you do um neutron scattering basically with uh a deteriorated solvent you you basically you see the contrast between h and d and of course and we are we have talked yesterday on this you have some h and d exchange when you use the deteriorated solvent because you so we use this mixture of water ethanol with the d instead of h and you have this d deterioration from d to o and the one od here which are labile and we can exchange with some of the edge that you have on the proteins and which are the one which are linked to this s n or o atoms so you have this exchange but if you have the exchange you go from this situation to this situation when you reduce the contrast but see you get the same contrast if you assume you have an homogeneous exchange so the only way with we well the only hypothesis we have to explain our data is that in fact we don't have an homogeneous exchange but we have a very inhomogeneous exchange uh implying that in some part of the in some domains in our material we don't have this hd exchange and basically we have here the domain which is enriched in h uh in a in a in a c or enriched in d a solvent and we see this contrast when you when we do a small angle neutron scattering so what in a more physical point of view well first you can of course extract the characteristic size at this uh domain so we use a standard model which have been built up to uh to to to model uh microphase separation solitism this is the library model which is nothing else that implying that you have sharp interfaces between two two domains so you have this q minus four and you have a cutoff at a given wave vector give you the size of this domain so you have here the fit of this uh with this model and from that you can extract the characteristic size of this domain which is of the other 16 nanometers so on a more physical point of view what happens is that in this domain you don't have hd exchange and you don't have this because you believe you have a very strong edge bond interaction between the proteins which prevent this exchange um so to prove that um and we believe that this domain in fact has really uh as the signature of the protein assemblies that we have in the dilute regime and why do we believe so first we believe so because we think we we measure that this character size is uh independent of of the protein concentration of our sample independent on the fact that uh of the composition we use to make this data so as we have to observe for this assemblies in the dilute regime they have more or less the same size of the protein assembly we measure in the dilute regime and if we look uh quantitatively and quantify because you have the same size basically the the intensity you have at low q is directly proportional to the amount of these domains and if you plot this the this uh intensity at small q as a function of what you expect for the assemblies that you have measured the dilute regime we have a super nice correlation which in fact implies that the two are correlated and what we see here uh in this domain which are held by edge point is really the signature of the protein assembly that we uh we are able to highlight thanks to this contrast variation technique so just um one or two slides to finish so just to make the correlation with the mechanical property here are the um frequency suite that the elastic modulogy prime and the loss modulogy double prime are the function of a frequency for different samples uh with uh rather concentrated but with different um quantity of glutinine for the protein extract and we go from liquid like samples when we have essentially a system very poor in glutinine up to gel when we have a sample enriched in this glutinine and uh if we have this sample so I recall you that this is uh this kind of sample where we have this assemblies in dilute regime and the heterogeneous deterioration in semi dilute regime that we have interpreted as being due to this uh strong edge bond between the the the proteins and in fact if you take oops if you take this gel and dilute it um in a mixture of water ethanol and urea and urea is a compound that just this that just uh disrupts all edge bond in the material and what you you see that if you dilute your sample in this urea based material you that you dissolve your gel while you you don't do it if you just use a simple water ethanol so it's another uh evident that edge bond is also important to control the the the gelation property of this network so to conclude uh I have uh I hope I can with you that uh both edge bonds are very important to control the gelation of this material and also uh that they they are uh they are the uh the main interaction uh interaction that that keep the proteins together and they are both in the dilute regime and concentrated regime and something I wanted to stress here in this audience is that thanks to a combination of uh sex and contrast variation we are able to get unique information and visual well visualize indirectly visualize and have a clear signature of this protein assemblies in um in a concentrated regime and with that I thank you for your attention thank you laurence um I don't know if there's any questions for you so I'm just gonna ask you one uh just to start um how do you know that you have uh these uh anomogenates when you're when you're doing the exchange um I'm stuck I cannot transfer my slides um actually it took us a long time I mean this was that was this experiments that you I don't know if you see my pointer probably not no okay here um when we did this experiment we were expecting to get the same results in sense and actually we just did sense because we wanted we wanted to extend the the web vector that we could access and it was easier that time to do sense and sex and then we got these totally different results so you had to explain them and then you're certain we have to explain so maybe the law for contrast variation to change the contrast from pure D solvent to pure F solvent and yeah we have to to expand and that was the only explanation I think it's reasonable but maybe someone we came up with another explanation well but what is this difference maybe this follows uh Tommy's question that he just put in the in the chat and he says how does the size of the aggregate vary with the preparation and if you if you have these domains can you change the size of the domains somehow no why for the moment no the way we prepare the sample is that we have this uh well we have this sample preparation with um when we have this liquid pressure liquid liquid pressure and then we we first try the samples and then we hydrate them again and once we do that we always get here I don't show the size but basically whatever the the composition we have we always get the same size and this is a size you observe also systematically when you when we do sands experiments the question we are actually the biochemist we're working with us is whether the this phase drying procedure also uh change something in in the in the assemblies and we have some things that it might change a little bit the size of this zone but it's something we haven't explored that much for the moment yeah actually I have a similar uh idea about how we rehydrate powders and if is it really what we see part of the fact that we cannot rehydrate fully and in some places you cannot exchange with deuterated water so it's probably a similar uh interesting okay so if nobody else has any questions is coffee time