 to understanding chromatin architecture in all life domains as we learned from yesterday's lecture. She is also recognized with many prestigious honors and awards for her research. Now Dr. Luger is a distinguished professor at the University of Colorado Boulder and HHMI investigator. Finally, we would like to thank to thank the support for the steamboat lecture from our department, Wolf and the summer facial. Now let's welcome Dr. Luger for exciting lecture. I feel like deja vu all over again. I feel like I've done this before. I don't quite know how that happened. Thank you for the kind introduction. Thanks again for hosting me. Thank you, everybody who's here for maybe a second time, which is very hard to imagine. So I'm not going to give my talk backwards like I advertised yesterday but I will talk a little bit about more general aspects of my career with the hope that maybe some of you who are earlier in your career stage might find instructive. Let's just ask like how many how many undergraduate students are here? One to a couple, how many grad students? Many, good. Okay, this is for you. And then all the rest of you just have to suffer with me. So, and we'll see how this goes. I don't know lengthwise how we're doing but we'll be done in 50 minutes. I promise I'll just cut it short. When I was like your age, around thereabouts, that was my idea of a typical career path. I thought professors were born like that and then they were of course white and male and they knew everything and that's what happened. And it took me a while to figure out that that's not a typical career path. And in fact, mine really worked out like this. It was a random path, maybe with some attractants and repellents. So I was doing random walk. I had attractants. I knew I was curious about nature. I love planting things and growing seeds and tomatoes and stuff like that. I wanted to know how cells work. I really liked solving puzzles. And I had two older brothers and they were very techie and they always talked over my head and so I wanted to show them up and show them that I could also do science. Repellents were. I did not like sick people. So MD was not an option for me. I did not like teaching teenagers. So high school teacher was not an option that really didn't leave a lot of leeway. I didn't really want to be a pharmacist like my aunt and so I really didn't know what to do. I took myself off to the nearest university, which was three hours away by train to the University of Innsbruck, not a great university, entered the major of microbiology, and I was lucky enough to be adopted by the biochemistry department and biochemistry sounded like you know that might get at how cells work and things like that so kind of a random walk into that department. So I searched there. I found it exhausting, because I was working with Bernd, who was a very senior research fellow there, and is almost like running, running, trying to run an ultramarathon with a trained ultramarathon runner, never having run even a 5k. So, you know, this guy kept going I was exhausted. That was my induction into biochemistry. Nothing worked. I did learn to pour a really mean sequencing gel gradients. I brought them home to my mom the films. She was very impressed. I learned how to do mini preps from scratch so that was good. I did an internship in industry, which I really hated, but I met my future husband there so that was good, I guess, and then I took myself off to PhD work in Basel where my then future husband also resided. So again, kind of directed random walk. My PhD in Cosper's lab was amazing because he was a very hands off mentor and a very creative mentor, and he would let us do whatever we wanted to do. And so I did a really crazy project that amazingly worked. And we could publish it very well. And so that really got me hooked into science and I wanted to do more. We were working closely with the crystallographers. They were all a bunch of really unkempt and unwashed guys who spent many hours in the dark room and in the diffractometer to solve the crystal structures. And I was just really intrigued by how you get from these patterns to a structure. So that's what I wanted to learn. And so I decided to interview with Tim Richmond, who was then a new professor at the eth surgery was a superstar, establishing crystallography there, and also postdocs at the eth paid really well. So that was also an attractant, shall we say. And but before I did that, I decided to in the words of my PhD mentor make a really valid attempt at not having a career, because I took a year off with my then husband. And we rented, we bought an RV and we drove it all across every single National Park in the United States and our piece de resistance was to canoe about 2000 miles down the Yukon River in in in this very trustworthy little rubber to me, which was really amazing. And it was a good thing that I took a long vacation because what I did not know at the time is that I was really in for a real, a really hard project. And we were trying to figure out how the human genome is packaged in the nucleus so like that sounds easy right after canoeing down the Yukon River. It seemed like that was manageable. And just to give you all a little bit of a size comparison, if the nucleus were the size of a golf ball, it would contain about 10 miles of DNA, a really, really thin thread. And you'd have to package that in a way that you could replicate it facefully and you could also find the pertinent information in a spatial and temporal accurate way. I give you a little bit of a history on on on the structure biology and what people did before I entered the field which was nowhere near its infancy. So back in 1974, Roger Kornberg had to do is from a lot of careful biochemistry that chromatin is structured based on a repeating unit of eight histones and 200 base pairs of DNA, which we now know is still true. Francis Crick and Aaron Kluge then deduce that in order to package and compact the DNA in this manner it needed to be kinked rather than curve because the curvature would inflict too much of an energetic penalty which turned out to be wrong but that's okay. In 1976 which is way way way before most of you were born. Finch and Kluge developed the solenoidal model of the of the chromatin structure again based on theoretical considerations as well as whatever experimental data were available which wasn't really very much and this is still one valid model of chromatin higher order structure. This was 1976. In 1977. Finch and Daniela Rhodes and Michael Levitt and Aaron Kluge, but these really pretty amazing electron micrographs of chromatin nucleus almost stacks isolated from I think cuffed timers at that time. This is the electron density map projected and based on that as well as some careful DNAs one digestion of chromatin, they came up with this model of the nucleus on. And if you superimpose what is now known as the nucleus on structure onto this model, you see a pretty amazing correspondence. And this is that yes they were really important and really good papers written before you were born. And, and, and we actually already had a lot of information before I or even Tim Richmond entered the field. So this was 1979 in the early 80s. Tim Richmond then managed to get crystals that diffracted to about seven angstrom resolution. And this is the structure. It's really not something that we would throw on the cover of nature, nowadays, and, but we, we could clearly distinguish a DNA from the protein so it was known that the DNA was on the outside that was actually known before then, and it scattered a little differently. But clearly we had no idea how the histones were arranged at the inside and the, the assignment of the histones on the inside turned out to be completely wrong. But a lot of these model was actually facilitated and supported by very careful cross linking experiments by Mirza Beckhoff and colleagues. And that turned out to be those, those cross linking data turned out to be completely consistent with the structure that we know today so again my point is that yes they did very careful very quantitative biochemistry and structurality in the old days and hybrid approaches, even at that time were a thing and really helped to deduce structures that by current knowledge now seem to be correct. So now enter the early 90s and me, when the dinosaurs work walk the earth. This was a dark ages, all the cool kids were working on transcription this was teach and passion and all these guys working on TF2D and T the whole alphabet soup. Nobody cared about chromatin in fact chromatin was thought to be completely unimportant. The post translational modifications were actually known, but they were largely ignored. So I have a textbook from 1984 where all the modifications are described in full detail, but nobody knew about their biological importance. What was niche for proteins, blabology wasn't really applicable to this third generation synchrotrons were just being built so X ray crystallography was really the only game in town, and we needed very hard x rays to solve structures from for very large complexes. So storage and graphic systems were about the iPhone zero level. And if that we actually had whole data sets we couldn't in process whole crystallography data sets because they were like mega basis in size. And so we had hard drives and tapes we had to shuffle them back and forth so I'm that old yes it's true. Chromatin was isolated from chicken blood and cough time was so natural sources. There was also a structure of the histone octamer that was published in 1991. So this would have been amazing, because this would have provided a lot of really good information for facing, but this group refuse to deposit their structure on the PDB, which at that time was not required so we could not really use it for anything. So there are nuclear some crystal still diffracted to only five angstroms and we still had not solved the face problem, which is a problem that's inherent to crystallography because you can't refocus x rays. So unlike in cryo and where you can just look at things, you have to, you have to solve the face problem that's all I will say about this because it brings back really bad memories. So the path towards higher resolution when I started, it was clear that we needed histones that were homogeneous and not heterogeneous because we knew at that time that there were a lot of post translational modifications. Nobody cared what they did but we knew they existed. We knew there would be a problem. The DNA sequence and length of the nucleosomes when you prepare them from these little arrays isolated from cough thymus were not uniform. And so it was clear that we needed to build nucleosomes with defined DNA sequences and with recombinant histone protein so we needed to make recombinant nucleosomes. Okay, so that was my task. And that was four years of my life in the bucket. And what we did is we we duplicated a DNA sequence that we felt would really like to form nucleosomes in a plasmid we then did 12 liter plasma perhaps to isolate enough of that DNA fragment to, to make nucleosomes for crystallography and the diagram amounts of this 147 base pair DNA fragment, a lot of phenol, a lot of ethanol precipitation, I'm still not sure what it did to my brain cells, and I'm sure I don't want to know. Histone expression, we needed to express each histone in E. coli. This was not a trivial task at that time. Because you couldn't really order genes online so it was a different time. That's all I will say about it. We then isolated them. We had to, we had to refold them to form a histone octomer. And then we had to devise means to combine DNA and protein to form mononucleosomes. And while we're at it we also made nucleosomal arrays for higher order chromatin structure so we could do it on duplicates of, of the positioning sequence. So, having done that, this is all great. I did a lot of quality control to ensure what I've made was really the real deal, because it didn't have any enzymatic activity and so how would I know that I was really making something that was not garbage. So we did that, that was all good. Then we set up crystal trays and that was before robots. So there's a lot of greasing and pipetting. And this is what we got. Super pretty to look at. Really bad crystals. They're actually worse than the crystals we got from chicken blood or cough thymus. They were thin windowpane crystals. If I could mount them they would diffract about five angstroms on a good day. And they were really slow and unpredictable in the crystallization behavior so sometimes they wouldn't crystallize for month. Sometimes they would crystallize in a week it was just a nightmare. So this was about year for ish, I think, great for morale, of course, because you see your career chances slowly circling the drain. But I then had an idea that I think, or I did an experiment that I still think was the most important experiment of my career. What we do with crystallographers when we get crystals, it's always a good idea to take them, wash them, dissolve them and analyze them. So that's what I did. So this here is the input. This is a dissolved crystal. Okay. And what you will see is that the input looked fuzzy, and they just always looked fuzzy. This band is very fuzzy, red kind of bloated. The crystal is really not fuzzy nice crisp and so that was weird. So then I decided to take the mother liquor. What is the mother liquor the mother liquor is the solution that in which the crystal resides. And that's where the leftover protein would be. There was no leftover protein there was no leftover nucleosome and everything is gone and incorporated into the crystal is not fuzzy. So by very sharp deduction. I concluded that the fuzzy material is transformed into non fuzzy. Okay, very scientific explanation. And so what we then discovered subsequently shortly after is that indeed the nucleosomes that we make were not uniformly positioned there was heterogeneity in the positioning. It was DNA sticking out this way or this way, or it was centrally positioned. And it turned out that the crystal only like the centrally positioned species. And so, in equilibrium it just pulled everything out of the solution and that explained all of my fuzzy versus non fuzzy and empty mother liquor data. So that's great. What do you do with this. Now, then again, a stroke of genius. It's a very high tech experiment. I stuck my nucleus on up north tube into a heating block 37 degrees, half an hour. Boom, we're not fuzzy. So this is great. You can then crystallize this material, and you obtain resolution that goes from here, which is really not great to hear, which is good. So this extends all the way to about two point angstrom and beyond. If you push it really hard, it goes to 1.9 angstroms. I just want to show this because it is very, very pretty. Okay. So, check, right sample homogeneity diffracting crystals. Awesome. We still have the face problem. And in order to solve the face problem and I will really spare you the gory details, I had to engineer about 2016 substitutions throughout the histones to then hook up heavy metals to then perform multiple isomorphous replacement. Nowadays, we do these structures with Selena Sistine or Selena methionine wasn't really an option here because the structure was too large. And also there was no system to incorporate Selena methionine. And so this had to be done. A lot of these things did not the fact a lot of these things did make gave me very bad crystals. These things gave very bad phases but finally we managed to pull this off. If you're curious about of a more scientific explanation of what I told you in brief, you can look at this free iBio video that is that that I took with with this with this nonprofit organization. There's actually a lot of really good videos out there by various scientists explaining their key experiments so I really recommend all of the undergrads and grad students to look at this site. Okay, so September 1990 97. Very happy person. This is, this is me at that time already was a big hiker at that time. And so we managed to publish this structure also huge shout out to people who are not on the paper song 10 Thomas rex diner, Yvonne and Andrew my partners in crime so this was a huge team effort. I also encourage you to, to take the call to this this is itoshi Kuro Misaka, and I forced him to write. He's a very talented composer and musician I forced him to write a nuclear zone birthday song and that's what he did. And that was before chat GPT so that's really interesting. So this is a rock song about the nuclear zone. Okay, good. So this is an accomplished we have a structure of the nuclear zone. It's not only beautiful but it also I think explains really well. A lot of the biology and you can you can appreciate how tightly the DNA is wound and bound by the histones and how much of a problem that's going to be for transcription and repair machinery and all that good stuff. I also entered pop culture. It, I'm sure it it inspired the Millennium Falcon this is from Star Wars. I'm sure but that must be that must have been it. And also, I just want to share this because it just gives me so much joy. There's this cancer scientist, Dr Brian Wellman at University of Utah, and he's also an artist and a welder, and he welded giant nuclear zone for me about this big and he gave it to me as a present. I didn't even know the man so this is totally fabulous and and it's in my living room and it gives me joy every day. So, back to back to careers. So eight years one paper. What's next, I should also say that in those eight years. And I don't mean to say this in a mean way, but I received no mentoring whatsoever on the next step. I was there to solve the structure of the nuclear some and I did that. And the rest was my problem. So that was just the way it was at that time. But I did actually around 95 or so Tom check visited our department. And he's very famous, and the last time he showed was the view from his office. So I decided that's where I needed to be. And so I, I moved over, not to Boulder because they didn't want to interview me but to the University of Colorado State University, which was still not well known university but a land grant university with a very strong mission in undergraduate teaching, also a strong vision they wanted to get crystallography going in their department so I thought well I'm, I want to do this the hard way so why go to a department where they already have to set up if I can do it from scratch right it makes perfect sense. So I started in 2000 pretty soon recruited these four rather for Lauren looking graduate students sitting on the side of the road. The lab group pretty rapidly 2005 till 2010 we had a pretty good size already and a lot of fun. A lot of good things happened in 2004. I got a GMI support and also had a baby and at any rate we grew the lab in in many ways. We had a lot of babies, as you can see here. So, so these were really productive times shall we say. At CSU, my first mission was to build an x-ray diffraction facility. And that was good. It was, it was interesting dealing with contractors and such so. And now I was also learning how to teach I was learning how to mentor I was learning how to run a lab I made a lot of rookie mistakes I kind of muddle through it with the help from my friends, and asking a lot of questions. Specifically, we were addressing how epigenetic variations affect chromatin structures so post-insulation modifications histone variants and all that good stuff. And we were really interested in how new plusomes are assembled by histone chaperones. And finally, we started a project that that is also still ongoing and how how the cancer drug target part one senses DNA damage and how it can be inhibited. And how it interacts with chromatin. So these were our project projects in 2015 I felt I needed a little bit of a change of scenery but not too much so I moved down to see a boulder who didn't interview me way back when but now they did hire me which was great. And I pretty soon started hustling and bake sailing and campaigning for a Titan Creos microscope which did not exist at that time and so this is, this is the arrival of Princess Creos on campus in October 2019 about the same time I think your guys is first Creos arrived. This is Shana my administrative assistant who was instrumental in getting her set up and wrestling the contractors and all that good stuff. And we had first slide in 2020 and just to to convey to you how excited we are why we are so excited about this is this is kind of the, the wasteland in Princess Creos sits there's it's the only game in town still and we're very very proud of her she performs really well and we treat her very well. We now have a glass use and and I just don't get over these 2d class images this is not even a particularly good one but now we can just the fact that we can just take a nucleus and put it on the microscope and look at it it's just amazing to me. So, let's talk a little bit about science. So this really allowed us to then study projects that weren't really accessible to x Rikker's biography before and that is that nucleosomes are interaction hubs for nuclear factors so there is about 30 million nucleosomes in the nucleus, a lot. They're very densely packed and they interact with many factors. That's, that's part of their job. And, but the best, the most frequent interact of nucleosomes are themselves because they packed to form higher order structure. So there's primer, you can almost equated with protein structures as primary amino acid sequence. In this case it will be the different flavors of nucleosomes with different ptms and histone variants. These can fold in into secondary structures and that then determines tertiary structure. This is a very simplified way of putting things of course but that's what we were interested in. A really good case point, a really interesting case study would be the centrum or the human centrum. In humans, so the centrum is a specific region on a chromosome to which the mitotic spindle attaches. Okay, it in humans, it's epigenetically and only epigenetically defined by the histone variants and a doesn't really have a special DNA but the only thing that tells a centrum or to be a centrum or is sent a. So we have a nucleosome in which age three is replaced by an age three like protein. Okay, and it's this protein that then recognizes a whole alphabet soup of the inner kinetochore. So a whole slew of proteins converge on this poor sample a nucleosome to then nucleate the outer kinetochore which is again another slew of proteins. And among those proteins there's three different ones that specifically recognize from age three. Now what do they recognize, there is really not a lot to go by. This is the age three nucleosome structure and this is Hitoshi Kouromizaka SEMP-A nucleosome structure, he of the nucleosome song. There's really not that much different they look like nucleosomes. The only difference to write home about is this little dinky loop that sticks out. So we were really interested how does a protein recognize this really minor difference and then nucleate the entire kinetochore on this. And this is a really fundamental question because if you get this wrong you really screw up your cell division and then you're dead. So you have to get this right. So, those were the questions and enter Cody Zhao who said, I will do this by CryoEM. This is when we just started, we didn't have a cryos yet. So Cody said like I will do this with CryoEM. And we were super lucky because CG was at that time in Tom Czech's labs and he was going, he was a little ahead of us with his other project. So Cody and CG teamed up and I don't actually know what you guys did, but you made it work. So huge shout out to CG. This is kind of how our department in Boulder rolls, we really help each other. And what goes around comes around. So thank you for that. And Cody got this to work amazingly. And this is a structure at 2.6 angstroms. And remember, I spent like eight years of my life to get a 2.8 angstrom structure of a nucleosome and now literally undergrads do this on a weekend on a good day. Okay, good. So, so this is our protein kind of stuck and protein stuck to SEMP a nucleosomes and I don't want to go in the detail this is way too much detail here but I just want to say that we now know how SEMP and specifically recognizes this little rinky group that sticks out of SEMP a and that is not present H3. So that's how it discriminates. We've also recognized we've also learned that this protein has a very large protein DNA interaction face interface as you can see here it's almost like it sits with its butt back on the DNA. And the interaction is of course, it can do it with any nucleosome. Okay, so that was interesting and we published this together with Andrea Musakio who helped us with a crystal structure of SEMP a that they had and a lot of other biology. So, what Cody noticed, even in the first days when we weren't really that interested in higher order structure he noticed it form these nucleosomes when he added SEMP and it formed these pesky nucleosome stacks and he tried to avoid them and pick other particles. But then he went back to it and he started to pick dinucleosomes because he figured he could then determine the structure of those dinocleosomes and indeed he did and it turns out that SEMP and bridges SEMP a nucleosomes together so it forms these nucleosome stacks almost like a little bit like a sticky glue or like these. This is one of our collaborators who loves these macarons. We wanted to do a cover for our paper. And so the poor man had to go out and buy macarons to take these pictures. So you know, I was really feeling sorry for him. Okay, so this is a structure of these nucleosomes glued together by SEMP and so very, very cool, and a way to promote chromatin higher order structure, and they do this by employing an additional DNA binding interface that we actually didn't know existed so on the other side of SEMP and that was just like hanging out in the breeze. That also is a very basically charged surface that then is used to interact with a neighboring nucleosome. So this is all good but in real life, nucleosomes do not exist in isolation. They actually exist in nucleosomal arrays. And so remember I told you right at the beginning this approach could also be used to make nucleosomal arrays and so we made 12 more nucleosomes where we just have a very long piece of DNA and then you position 12 nucleosomes on top of that. And you can add SEMP and and you will notice that you find you get very regular, very regular stacks are compacted zigzag arrays of nucleosomes that are promoted by SEMP and you do not get this when you have no SEMP and present and Codi managed to solve the structure of this of this array as well too much lower resolution. To be honest he didn't try that hard to push the resolution because we didn't really need a lot of more additional information and also he was out on job searches and ready to jump to the next adventure. But what we learned from this is that that SEMP and promotes the folding of SEMP A arrays in this manner and these arrays look very different from other arrays that are possibly spawned by the presence of the linker histone H1. So what we think and there's a lot more biology here and we actually teamed up with Aaron straight at Stanford, who is a centrum specialist to test in vivo, whether the second DNA interface was responsible for forming higher order structure and we found that indeed it is and so this second interface really seems to have some critical importance. So we believe that SEMP and is a centrum Eric, kind of a linker histone is like double sticky tape that helps congregate centrum Eric nucleosomes with normal nucleosomes, because that second DNA binding interface does not rely on the presence of SEMP a the only loading function of SEMP and is that little rink it include there's one interaction interface it needs one of those and then it can hook up with a normal nucleosome or any other nucleosome it wishes to do so. Okay. So we think this promotes long range and short range nucleosome interactions and is responsible to form a really specific structure at the center mer. So, so much for that. The point I wanted to make with the stories first of all a huge shout out to city for for helping us. Second, this was our first cryo em structure so we're immensely proud of it. And third, the types of things that you can do with cryo em would literally not be possible with extra crystallography and vice versa. Some of the small things that I showed you yesterday would not have been possible with cryo em so we need them both and we love them both and will continue to use everything, even NMR if we have to. Maybe. I grew up scientifically next to court retreat trick. And so I'm a little shall we say opinionated with respect to anymore. That's all I'll say about this. Okay, so I have a couple more minutes. And I just want to talk about one other big project that really was we've been chewing on for a long time and then really was made possible by cryo em. And that is what happens to paraphrase Roger Kornberg when an irresistible force the RNA polymerase meets an immovable object the nuclear zone something's got to give right. And what gives is that we have to kick out the histones and but we also have to maintain chromatin in the wake of the polymerase. We need histone chaperoning to prevent those histones from aggregating on the DNA on on the RNA so we need his and chaperones. We need these to stabilize these partial nuclear zone so that they don't completely discombobulate and we need to reassemble in the wake. And there's two very diverse classes of protein factors that are responsible for this one is histone chaperones. They are literally like the chaperone at the high school dance. They prevent hanky panky between inappropriate partners. There are also a little bit matchmakers because then they help reassembly which I hope the chaperones at the high school dance do not do. And then we have ATP dependent chromatin remodellers and these guys are big SUVs of machines they use obscene amounts of ATP to shovel nuclear zones out of the way and to remodel chromatin in general. So I'll tell you one story for sure and maybe we'll get to this one because this one is done by John Markert who is a U Wisconsin alum as well I have a great pipeline it seems. So, and the work on fact was done by young Lou, who's now also the who's not the University of Hong Kong. So what do we know about fact fact was originally designed by designed discovered by Danny Reinberg, and also Chris Formosa. It is named very unfortunately because when you search fact in PubMed, you get a lot of hits, obviously. But it's named for facilitates chromatin transcription, presumably by nucleosome disassembly so that's biochemically how it was identified. However, a knockout in yeast leads to nucleosome assembly deficiency so that kind of begs the question, what does it do we now know that is hugely involved in transcription replication DNA repair. And we've renamed it much to Danny Reinberg chagrin to facilitate all chromatin transactions, we've maintained the acronym though so just to keep him happy. It turns out that cancer cells are addicted to fact in most cancer cells fact is hugely upregulated up to one fact molecule per nucleosome. And so it's being explored as a cancer drug target. And we were really interested in how does fact work in these different contexts. And what does it do to new clothes homes without ATP hydrolysis, what's its job. So, our first forays into this was not very exciting, because it turns out that fact actually does not even want to bind to new clothes home so what kind of a factor is that right that's really not very useful. And so when you take fact and nucleosomes or DNA, no binding, but we figured out that if you pre incubate fact with histone H2A H2B. It somehow gets activated, you can then combine it with a partial nucleosome, and it will make a complex that then might be amenable to cry oh yeah, and this is what you see here. So, this is a gel shift. And when you do this experiment, you see some left over his tone at the bottom, but you also see complex one and complex two, there's two bands for crystallography. This would be the kiss of death, like it's a no go for cry and what the heck you just slap it on a grid and see what you can see. In this case, we slept it on a grid. It looked horrible. And we needed a lot of and CT also helped us with this. We needed a lot of help from Bridget Carragher we use spotted on which is now chameleon different freezing techniques so it was really not an easy project. But young and Cody together managed to solve the structure and we fact is a heterodimer, consisting of two multi domain subunits. We, despite the fact that we put the entire fact complex into our complex. We can really see the color bit so there's a large part that is disordered but what we could see was already hugely informative. The resolution is also not amazing but a good starting point. And what we can see is that fact looks, it kind of looks like a, like a unicycle, and it sits astride the nucleosomal diet and each subunit dangles around an either side and there's also this. This is a big terminal domain that seems to hug the histone dimer so just to summarize the findings from this and and the information that we gleaned from this is not so much like where every atom is but really in its mechanism and its regulation. The second point is that. Yeah, we would have liked higher resolution but you should not discard a low ish resolution structure because it can still be very very informative. So fact has an extensive DNA binding interface despite the fact that it does not bind DNA on its own. So that was interesting. And the C terminal acidic domain mimics it acts like it's DNA and tethers the H2H2B dimer to the body of the nucleosome we actually didn't give it enough DNA so usually this dimer would have flown off, but we tether we tether it with a C terminal acidic domain. So this, this allowed us to develop a model for its regulation. So we thought that the C terminal acidic domain is the one that auto inhibits binding when we delete it. We actually can get this thing to bind to DNA. Okay, so this is shown to the gel and the gel on the left it's not great binding, but it is capable of forming a complex with DNA when you delete the tail. What happens is that in the absence of histones when fact is just trundling around and mining its own business. We think this domain is kind of swinging to the inside and binds to this acidic to this basic DNA binding cavity, and it auto inhibits itself. Along comes a free H2H2B dimer because it got kicked off off by a polymerase or something fact fishes it out hugs it to itself, and then can dock it on to a tetrasome and this explains the nucleosome assembly function that we observe in vitro as well as in vivo. Alternatively, and this is also incidentally how we made this complex for cryo EM before knowing anything about this mechanism. Alternatively, fact can also encounter a nucleosome with partially unwrapped DNA and that we know is an intermediate during transcription the DNA has to be peeled off. And this thing is kind of left over fact recognizes that it hugs it grabs a diamond hugs it to itself and this is the complex with structure we actually solve and it prevents the dimer from flying off and this really explains the tethering and stabilization function of fact. So, I just hope you appreciate how much like DNA this the CTD looks like so it's a true DNA mimic. The deletion of the C terminal domain in yeast is lethal and so that really lends credence to the model that we've developed here. I just want to summarize this that fact is a nucleosome maintenance factor it's neither an assembly nor disassembly factor. It's kind of just a repair crew, the cruises around and looks for his stones and bad nuclear zones and just kind of tells them it's okay don't make you hold again and so that's kind of how I see its role. And the regulatory mechanism is such that it only engages partial nuclear zones which is actually a good thing, especially when you have a lot of fact and so fact will be engaging with the DNA that have lost part of their DNA, it will also capture free his stones as they're flying around and reassemble them in the wake of the polymerase. And that model is, and the structure is really highly consistent with all the biochemical data that was in the field at that stage, and was later beautifully illustrated by by some amazing structures by he told she Kormizaka and Patrick Cramer's lab where they actually had fact and nuclear zone and polymerase in a mega complex and they could show it in all in all stages of its life cycle and it's doing exactly what I have actually depicted it's doing here it's doing all of those things. So, all cool explains its central role in all DNA processing pathways because every polymerase whether it's DNA or RNA polymerase needs to peel the DNA off and so we encounter these intermediates. What we do not understand I'm really intrigued in hearing opinions is this dependence of cancer cells and we don't know what the cause or effect is it could be that in cancer cells, you have so much disordered this regulated transcription that you have a lot of disordered nucleosomes and you need a lot of fact to kind of rain them in. It could also be that fact is overexpressed in cancer cells, and too much of a good thing is not good and so factor starts to destroy nucleosomes and that results in destroyed nucleosomes and that then in turn results in in transcription that is this regulated so we do not know what this connection is. So I think I will probably stop here. I just want to give a huge shout out to to john who's been a fabulous graduate student in my lab, who independently worked on a complex that's called smart cat one. I'll show you what what smart cat one does it is involved in transcription replication DNA damage repair. It's a chromatin remodeling factor and john figured out what its structure and its mechanism is and it's really been a tour de force work in the lab we're still working on regulation. So this is an ongoing project and with that I will I will finish and acknowledge the people who did the work so Cody and young who are now in at the University of Hong Kong faculty members, john market who's a postdoc at Harvard but probably will be on the market soon. So if he applies to this place, you should hire him he's amazing. And then, again, a huge shout out to Pamela dire my lab manager of 22 years. So she started with me the year after I started or two years after I started been with me ever since truly indispensable to the success of my lab and with that I'm happy to answer any questions about anything or let you go off and do you think whichever you prefer. Thank you very much. I have a question about the nap one or I'm sorry the fact. Chaperone structure and the extreme sea terminus the acidic. I think you and possibly Rebecca Hill and others have found that other histone chaperones have these poly acidic stretches like nap one that are polyglutamylated and I'm just wondering if you have any evidence from your structures of, of this modification and do you think it might play a role and its chaperoning function. Yeah, that's a blast from the past so so polyglutamylation has been discovered on on jubilant mostly but also on his son chapter on his especially on fact and also on a chaperone called nap one. We did a lot of work on the polyglutamylation in that one. We never published that because there was a key experiment that was missing. And then the student graduated and nobody wanted to pick up the project. What we did find is a polyglutamylation of these of these tales regulates their activity. So, yeah, I actually had forgotten about that thanks for reminding me. Another question about the fact. See terminal domain seem like the knocking that out is obviously lethal in yeast as you described, but is it more the acidic nature or the length of the sea terminal domain that was found that's like really crucial and if so like which residues I guess I'd be curious to know we're found to be really important regarding that. So, it's mostly these and ease with some interspersed other residues. Our density was not good enough that we could have the register we just know it dangles down from there, so we could not build an atomic model of that till. I believe that the team for most of the chopped off the entire sea terminal domain I don't think they did piecemeal deletions, so I can answer that question. If I had to guess, I would say that if you make it much shorter than what we see here. It's probably not going to do. It's probably going to have the lethal lethal effect. Any more questions. Is there something that removes fact from the nucleus home. Or does it just kind of hang out there if it stabilizes the nucleus home. During transcription. He told she Kormizaka like pictured, they were like eight cryo em structures, the whole life cycle so the polymerase eventually shoves it off and we've actually also shown that when you shove it off the DNA it sits on the histones and so it hops over from the DNA to the histone it binds the histones and then it goes off. It's kicked off completely so during transcription the polymerase motor kicks it off when it's just cruising around doing its healing thing if it indeed does that we don't really know we probably. Once it's assembled the nucleus home. I believe what happens is that the DNA competes off the CTD, and that kind of makes it lose its foothold and it's dissociating. And we have in vitro evidence for that but it's not super strong. Do do our key have a version of fact. No they don't know it's a eukaryotic gig. They don't really need it the polymerase doesn't really isn't really that slow down by nuclear something out here. And anymore questions. Like sex doctor look again for the awesome. Thank you.