 First of all I wanted to thank Oxana and the other organisers so much for organising this meeting. It's so nice to see people if not to see everyone in person. So what I thought I would talk about today would be to talk about this work that we've been doing to try and make a synthetic paired helifocal filament. What I'm going to try and do is to persuade you that we have made one or at least that we are on the road to making one and to explain why we even need to make one in the first place. Okay good so I don't think I need to go through this particularly we've had a really nice introduction already to Alzheimer's disease earlier on today from Professor Taliavini and so just to introduce the two types of plaques we've got so we've got amyloid fibril plaques on the left hand side and then we've got tau neurofibrillary tangles on the right hand side and the only thing I wanted to point out here is that we've got electron microscopy images showing the deposits of fibrils in the brain tissue of humans and what you can see in the lower panel on the right hand side the neurofibrillary tangles you can actually see the paired helical filaments perhaps you can see why they're called paired helical filaments there but the important thing that I want to underline is that A beta and tau both share the structure of amyloid the cross beta structure and this comes in quite it is quite important to point out when at meetings on Alzheimer's disease because there's a temptation for people to refer to amyloid beta as amyloid and as you're all certainly aware amyloid is much more than just A beta and so what I'm showing here is some evidence that we have collected to show the cross beta structure of amyloid fibrils and what you can see here is the cross beta structure cross beta fingerprint that you get from x-ray fiber diffraction and I'm going to show you a quick movie to demonstrate the cross beta nature of amyloid fibrils and many of you may have seen this movie before because I love it so this is the beta strands running perpendicular to the fibraxis and now you can see the hydrogen bonds running parallel to the fibraxis and creating this incredible strong network and now looking down the fibraxis you can see the interdigitation or the steric zipper as it's been called by David Eisenberg whereby the two sheets of this filament interact together and form this extremely stable arrangement. So I'm not going to say too much more about that structure I may come back to a little bit further on but I'm going to focus entirely on tau now and so again this has been introduced briefly already but tau as you will be familiar is quite a complicated protein it has six different isoforms and actually it has a hundred different forms that can be present in vivo and it's been most well known as a protein that binds to microtubules but it also has some other forms and it can also be post-translationally modified it can be acetylated it can be phosphorylated it can be hyperphosphorylated it can be truncated. So what I wanted to point out now is that tau is much more than just a microtubule binding protein and Mahmoud Maynard who works with me did his PhD thesis and investigated the whereabouts of a non-phosphorylated form of tau and he found that it was located within the nucleus of the nucleus and therefore feels a very important function associated with heterochromatin and so I've put some references down here just in case that's something that you're interested in because it's a sideline to what I'm going to be talking about today. So as I've already shown you we are able to do electron microscopy of tissue sections taken from donors who have donated their brain to research and what I'm showing here is a zoom in on the paired helical filaments that are deposited in the tissue of a patient and you can now see the paired helical filament nature of these filaments and interestingly enough there has been recently some controversy about whether paired helical filaments actually are paired helical filaments in vivo and of course you could argue that the processing of the tissue that we've done may create them but I think we can take this as evidence that you can see these filaments in in vivo of course this person is not alive but close close enough and so here what you can see is that what the little black dots are gold particles that we've used to ensure that we're actually looking at tau rather than another filamentous structure. So as you well be familiar cryoelectron microscopy structure was solved in the last few years of the paired helical filaments from patients when their PHF were extracted from tissue and I'm showing an image here and you can see again that the beta strands are running perpendicular to the fibre axis and if we look to the left hand side you can see the conformation of the structure down the fibre axis um my cat keeps meowing so I'm going to have to put her on my lap so she stops sorry um so um when uh the the group of um godet and sharers um went on to solve many structures of tau filaments from different diseases and what's really interesting about this is that although all of these structures share a cross beta structure well we assume that they certainly do um when you actually look at the um conformation of the structures you see that there's a a high degree of polymorphism so here we have ad phf and straight filaments and then we have filaments from um pic disease um so frontal temporal dementia picks um form from chronic chronic traumatic encephalopathy um two types here and then um cortical basal degeneration and so you can see if we look down the fibre axis that we do see some differences between these two different structures and I always wonder if this might be due to um the the earliest changes um that occur within these patients that trigger the self assembly of these structures and create a particular polymorph so what I want to lead on to now is that um each one of these structures actually does share a certain region or fragment of the structure um and it's covered by these repeat regions of the protein the tau protein so um tau is um formed of um imperfect repeats um repeat region two three and four are all found within the structure repeat two is one of these regions that actually can be missing in three repeat tau and it's actually missing from the pics filaments so generally the way that tau has been um examined is been um by using heparin and the reason for this is because uh if you take full length tau it's incredibly soluble so you can put it into solution and no matter what you do to it it appears that it doesn't self assemble so it won't form fibres and I think it was probably 15 or maybe more years ago um it was shown that if you add heparin to your solution then you can template self assembly of the tau and it will form what look like paired helical filaments so that was a great um um advance and meant that people could then make paired helical filaments or so they thought but unfortunately um the advances in cryoam have meant that um groups have shown that the heparin induced tau filaments are very polymorphic for a start and they differ very much from those in Alzheimer's and Pix disease and actually now we know that they differ from all of the cryoam structures that have been taken ex vivo and there is also not very strong evidence that heparin is incorporated into the filaments there's certainly no density observed in the cryoam structures that could be attributed to heparin and so what we wanted to do was to create a new model structure and so we were working with a peptide which we call DJE it's a core fragment of tau and it was first um discovered um in 1988 it was characterized and it's a region that covers mainly R3 and R4 and what you'll realize is that this also correlates nicely with the core structure that's obtained from cryoam structure and they had raised an antibody which recognizes the C terminus of this protein uh which is called AB4223 and a student of mine used this antibody and looked at um paired helical filaments in tissue and found that these stain nicely with this um antibody suggesting that um at least some of the tau that's found in these paired helical filaments is this truncated form uh with a truncation um of the C terminal at 391 and as I've mentioned already this overlaps nicely with the struct with the structures that have been solved by cryoam so that's 304 to 378 um that's shown in gray here with the pink box showing the the um DJE um sequence so then we went on to use um this peptide and we found that it forms filaments similar to uh paired helical filaments in brain without additive so I'm going to try and persuade you um of that um case now so here's some more electron microscopy images of sections of Alzheimer's disease brain and you can see the paired helical filaments here uh zoomed in and again we've used a gold particle this time we've used an antibody T T22 which is an oligomeric antibody uh specific to oligomeric tau and that means that it only sparsely recognizes the tau so it only recognizes oligomers that are associated with the filaments rather than completely coating the filaments meaning that we can now see the structure of these filaments and so a student called um Bronwyn Foster did lots of experiments and she compared our paired helical filaments that we formed in vitro that you can see here and you saw in my very first slide and you can see those individually here and compared them with these um paired helical filaments found in tissue and she looked at the um the width and the um the repeat distance and with uh Wei Feng's group um Wei Feng Zhu's group in University of Kent we also did some atomic force microscopy and you can see here that really nice repeating distance uh which ends up being a really nice correlation between the repeat distance seen in paired helical filaments from tissue and those that we formed in vitro so at least for now we can say that these have similarities although of course they're not identical and so early on today I was talking to um Wei Feng Zhu about um this um this meeting and um he said that links really is very interested in combining lots of different techniques and so I just wanted to show you in the last few moments of my talk and what we've been trying to do so we're trying to develop methods to look at the supramolecular and the molecular structure of synthetic paired helical filaments and of course this may well be applicable to other structural um other um frivolous structures so we're coming back to x-ray fiber diffraction here and here you can see the 4.7 angstrom reflection that comes from the hydrogen bonded uh beta strands and the 9.5 angstrom reflection that comes from the beta sheet spacing and so um we were lucky enough to obtain some Alzheimer's disease uh phs um from michelle growth godos group and although the fiber diffraction pattern is not completely amazing we could collect some data from it and compare the the diffraction pattern that we get from dje with um the diffraction pattern that we get from Alzheimer's disease paired helical filaments and I don't have time to show you lots and lots of fiber diffraction patterns that differ from this uh but for now at least um we're on the right track and so what we can do is we can take um those structures that were solved by cryoelectron microscopy and I'm just showing you a selection here and probably it's too tiny for you to see but here's pics disease here's uh paired helical filaments from ad um and you might be able to see I've tried to uh sort of increase the um contrast so that you can see although you might not be able to see exactly where the reflections are you can see that there's a different character to each one of these diffraction patterns so it's possible then that we can take the structure of paired helical filaments uh or or um filaments from different diseases and we can distinguish between them by uh x-ray fiber diffraction and then we can compare our structures that we create with those um that have been calculated from the original cryoelectron structures and similarly some work done by in Wacang Jews group um by Lisa Lutter who's shown here um has used you have two minutes two minutes I think it's fine uh so atomic force microscopy um to calculate the the atomic force image from um these structures that the their structures that were solved by cryoelectron microscopy so essentially simulating an atomic force microscope image and so um what you'll see here is um a heparin um structure and an ad phf structure and then below again we've got a whole range of these um simulated um atomic force microscope images um similar to the way we've done with fiber diffraction and we can see then how similar this this graph shows how similar um each one of the structures is to dge because we're interested in whether dge is similar to any of the um structures of um of um that have been solved by cryoem and so here you can see that phf is coming up very high so we've got uh good evidence um that might persuade people that this is um this is a good model system so finally I wanted to just summarize and say what I've talked to you about today is that dge overlaps with the phf core it self assembles in the absence of additive so we don't need to use heparin the filaments resemble phf in human brain and of course we need to do more work to to really characterize that structure um and um we think at least from the work we've done so far that the filament may share molecular characteristics of ex vivo phs and below in this little movie I'm showing the other point of all of this which is that this dge without heparin is a useful model system to explore our cellular effects and so shown here is um dge being internalized into cells and you can see um being uh associating with um lysosomes and this work is uh you can explain to more here in polaca tau and also it's a really good system to be able to test potential therapies so we're working with um uh with um tau rx therapeutics to look at inhibition of this um assembly and of course I always get to the end and then I lose time and I haven't had enough time to say uh the important part which is of course the acknowledgments so at least you can see a number of quite happy looking people here I've highlighted people as I went along hopefully and they are shown in bold here so the people that work with me now the people that have worked me with me in the past um on all of this work that I've shown you today and as I've said we work with tau rx therapeutics and with Wei Feng's group at the University of Kent's Lisa and Liam um and um their work was um that was done on um nucleoli was done with um Aiden Doherty at Sussex so I'll finish with a slide of the University of Brighton if you want to come and visit uh the University of Sussex sorry but in Brighton if you would like to come and visit us when there's time and we're allowed again so thank you very much thanks uh thanks a lot for excellent talk um still um do we have questions yeah so there is a question from Matisse Schmidt uh could you find any hints of possible abeta fibers in your tissue analysis uh yes well so actually we yes we've got lots of nice images of um ad filaments too so we can use an antibody against a um out of against abeta um and we also see amyloid plaques it's it's very hard to see the structure of those filaments though because they are very very tangled together so they really are looking like a proper um amyloid plaque it's much harder to see the distribution of them perhaps also if I can ask a slightly more technical question maybe but the the cryoem structures they are often as mentioned previously also incomplete in terms of flexible regions on the surface and things like this how do you handle this in the kind of prediction or the calculated AFM or fiber diffraction studies yes I think that um that generally the field is convinced that there's a what they call a fuzzy coat so certainly there are flexible regions of the um of the towel that are still associated with um the protein within the filaments I think what's important here is that what they've traced is obviously the the highly um stabilized and um non-dynamic regions of the protein and what we're trying to model really is that core so I we're not saying that there's not the rest of you know that there's other parts of the towel um there and that they might have very important um they might have very important biological functions as well but I imagine it also contributes more to the AFM than to the fiber diffraction because if it's fuzzy and so you wouldn't necessarily see it in the diffraction but or does it also become blurring in the AFM yeah that's a very interesting question and actually at the moment we haven't looked at the x-v-v filaments by AFM and so that's something that will be definitely useful to look at so thank you that was a really nice question