 Well, can you hear me? Bonjour, no Bonjour, no amigos so Each year we go to the Houston Natural Science Museum because they have this beautiful exhibit The National Geographic's photographs of the year and if I hadn't been to the ICTP I would never know Immediately what this was but I saw this I couldn't believe it. I think I emailed it to you Christian because it's so beautiful The Miramarae Castle So this is one of the best photographs of the year So if anyone wants to go scuba diving in front of Miramarae Castle, I brought my gear I would love to try to shoot a picture like this So as Marielle said, I'm gonna give a talk today about how we need to move beyond the static structures of Watson and Crick and think about dynamic supercoiling and I've shown this slide before but it always serves as a really nice introduction So this is linear DNA. This is what we know a whole lot about. This is that elastic rod worm-like chain Persistence length DNA that people talk about Here it is. And if you circularize it, it's a relaxed DNA. Here you go This is where most of our experiments are done But if you look at consider what DNA activities are going to go on Not much is going on at that level of supercoiling Across the bottom of this axis here. I have increasing supercoiling So here is the relaxed DNA that we know a lot about. This is the Watson Crick Helix And if you just underwind this thing relative to itself and allow it to ride up That's what's shown increasingly across that bottom axis Okay, till this point of highly writhed DNA right here. Okay, very good model right here So what you see and I think it's amazing. I still find it amazing It's what I've been working on for years is this amazing phase transition where there's no activity at all Until a critical level of supercoiling is reached at which point it turns on and this was first measured for site-specific recombination Then it was second measured for transcription initiation Then it was measured for initiation of replication. These are all in E. Coli the bacterium They were then later measured this exact curve was seen in yeast and Even recently this kind of activity curve has been seen in human cells So across all the kingdoms of life, it seems that DNA holds the key Something about the structure that we know a lot about Transitions to a structure. We know almost nothing about and it's almost magical and it's a turn a switch on As you might imagine I when I'm lecturing in undergrads I ask I ask for people to volunteer. Where do you think cancer cells have their DNA? Way up here. So in a cancerous cell it can't turn the switch back off And cancer is in fact uncontrolled growth So I started working on this problem years ago because I was thinking if we could only understand what transition point happens Then maybe we can be able to turn it back down in cancer or in infectious diseases because an infection Those DNAs are also very very active more active than the normal So as I mentioned most of our experiments are done here and what we know a lot about DNA this lovely DNA That's linear and dead is here But what I want to know is what is this phase transition? What happens here? So what happens to activate DNA? We've proposed it could be kinking Base flipping and I'll show you pictures of these things Denaturation this one's sort of what everybody thought but if it were gen denaturation in general what we all thought When you would get this nice kind of monatonic increase with increasing supercoiling because that's what we thought Denaturation should look like if it were acting like an elastic medium such as this foam cord So not only is DNA kept in all cells at this critical transition point Where it can be turned on or turned off very rapidly and thus controlled Transiently through the cells through all cells There are there are waves of positive supercoiling and waves of extreme negative supercoiling So homeostatically it's kept at one state But constantly as the polymerases go zipping through you get Extreme overwinding in front of this is RNA polymerase and this is DNA polymerase When you're making duplicates, this is replication making copies You're not necessarily getting increased negative supercoiling behind the fork there and expect there But here as this thing zips through the double strands you have transiently very negative supercoiled here very positive supercoiled here and Again back to this cancer angle The most important antibiotic drug targets so also for antibiotic resistance, which we heard a little bit from the twas man yesterday The most important antibiotic drug targets in the world are the DNA gyro asymptote by summaries for they Prefer to act at the positive supercoiled level and we know nothing about positive supercoiled DNA We don't have a crystal structure. We don't know anything and people don't study that of course And also the anti-cancer drug target the human topoisomerase to alpha it also preferentially acts here So it's thought that somehow the topoisomerase drug DNA ternary complex is in the positive supercoil state So I'm gonna give a little bit of an introduction to nomenclature that I hope is Routine for most of you But if we consider the basic state of of linking so that would be this most relaxed state or the one we know a lot about right? The okay not This equals in the number of base pairs divided by h the helical repeat and sometimes people get confused about helical repeat So let's just count these so one two you see how I'm counting this one complete turn to another All right, how many of them are there in this particular circle? Well, there are 30 here and We relate this to length by sigma, which is basically just a normalization of all right if this is completely relaxed this should have a Linking number a sigma of zero. Okay, it's not really fair, but zero But in fact this particular example should be 32, but it's 30 because we counted them and So therefore the delta okay is minus two The sigma in this case would be about six and a half percent underwound So think of sigma is just percent underwound all right or overwound and that's important because sigma normalizes for size So we can compare something in a chromosome that's very long or a plasma That's very long to a tiny tiny bit of DNA such as our crystal structures that we have of the linear linear DNA So a while ago and this is when physicists started talking to me We did some molecular dynamics simulations of just twist trying to get at that question of what is that Transition that happens when you go from inactive DNA with the persistence length of 150 base pairs To active DNA. What is that transition? So to try to get at that we simulated multiple different States of supercoiling and I'm not going to talk about this. This is published work now But I just want to point out that the elastic rod holds true for relaxed DNA and for a little bit of Overwound DNA it does not at all hold true for underwound negatively supercoiled DNA so biology is kept as I've already pointed out all Life seems to keep our DNA at this level around nine percent eight eight percent underwound But transiently we can get to even these extreme levels. So what did we see? We saw base flipping So what is base flipping? I'll show a picture of it in a second But the base just flips out spontaneously boom no proteins are doing this no ATP is binding nothing It's just spontaneously flipping out We also saw denaturation But not in this monotonic denaturation that we kind of thought with the elastic rod model, but with this very sudden burst of Of activity so there'd be a base flip and a denaturation We never saw those things in the positive supercoiling and then in the most positive supercoiled DNA The entire thing flipped inside out so that the phosphates were on the inside and the bases were splayed outside Now we thought that the that the thing must have blown up, but we got it repeatedly and For every bit of supercoiling over a critical threshold threshold level Well, that was pretty cool, but it's just simulation and you know how I think some of you know How people are well about just simulations you must prove it, right? But I just want to point out again that persistence length that we know Happens in a regime. That's not necessarily active and the persistence length is not defined here And in fact may be infinitely flexible All right, our data Perfectly aligned with these beautiful force extension measurements done by the bin samon and croquet groups where they Overwound or underwound DNA and stretched it and let it come back the elastic test, right? How elastic is this medium and what I want to point out is that it's only elastic you can stretch it And it'll come back stretch it it comes back, but it's only elastic and only comes back to a very little bit of underwinding and then you stretch it and it stays it does not bounce back and Then you do a little more Torque before you can stretch and it doesn't bounce back in the positive direction So this is something pretty dramatic. You don't recover the elastic Properties at a certain level of supercoiling just like you don't hear I want to point out Also that one polymerase can put 15 piconewtons of force just one and in bacteria You can have a whole slew of polymerase is lined up and acting together We don't know that they're necessarily additive, but there's at least 15 piconewtons put on DNA constantly and 15 piconewtons means DNA doesn't ever see this elastic regime and This is put nicely into a diagram of phase because not actually a phase transition But it acts a lot like one here We're looking at negative supercoiling here positive supercoiling. Here's beloved B. DNA B form DNA And you see just a little bit left of negative supercoiled. You're in something pretty much undefined So physicists called it linear Because what they envisioned was that the DNA became train tracks So it became so unwound that it just the two things laid out I think but that doesn't ever happen But they were just trying to explain this very sharp transition It's reminiscent of the first slide. I showed you DNA activity is dead dead dead active Right, so the two things to me coming together said there was something really interesting about the transition point from B. DNA to something else more active and In the positive supercoil direction to something and they called it pauling like because there was such a sharp phase Transition and then this one is supercoil because you get it now. I just want to point out that Oh, so B is rare to non-existent and the phase diagram is distinctly Asymmetric, so it's not the perfect elastic medium. All right and This oh, here's some pictures. This is what Pauling like DNA would look like and this is would be form of course looks like But this has been updated I'm not going to go into it, but they've gotten more and more sensitive with their measurements And so now they can break down these these these phase transitions into subdomains But basically the truth that I just showed you remains So the summary of our molecular dynamic simulations was that the localized structural failure So the failure could be a base flipping or a denaturation Allowed the rest of the molecule to adapt B form the form we know and love So there was a complete mess and then B form The denaturation always started with a base flip So it went base flip base flip and then denaturation So the denaturation didn't just pull apart like we envision with our with our heating measurements So here's a picture of a flipped out base crystal structure of it This is one of the first ones ever seen and I think these are enzymes that are knotted Sophie, I think this is one of your knotted enzymes. So these really cool important enzymes that have knots in them themselves Somehow stabilize this base flip transition and it was a big mystery for everyone Where does the energy come from because thermodynamically you're not going to pull a base pair in its lovely Environment out into a solvent. It's just not going to happen, but yet it does. It's been captured now 300 times with crystallography But we find that torsional stress the underwinding of the DNA can pop that DNA base out and that the protein then can bind to it so we also found that the the Overwound and underwound DNA profoundly differ and the torsional stress is not Uniformly distributed over the length of the helix and that means that it's not an elastic medium Here a perfect elastic medium if I unwind here, I feel it over here, right? But no you have to picture what's happening in the real underwound DNA as you underwind you base flip here You base flip there you denature there from here on you're just gonna stay be DNA This end doesn't feel it once that it has flipped out. So that's actually pretty interesting and it's a kind of a patchy idea instead of this elastic medium idea and I didn't show this at all, but that we looked we actually had the counter ions in these simulations And we watched where the sodium was so Manning's theory said that it would be this nice gradient coming in this nice kind of Normal-looking gradient, but what we found is if you unwound and wound the DNA you rung out those sodium ions Almost like a sponge. So it was very sharply dependent on the degree of underwinding so even the counter ion field changes with superquiling and What I found kind of interesting is that these counter ions were never found Around the Pauling like DNA and here's actually from our simulations this inside-out structure with the bases splayed For whatever reason and maybe it's because these are very hydrophobic the counter ion field Was not present here and now it was kind of bumpy here So this landscape that we envision of this perfect, you know nice landscape in fact is sort of bumpy and thumpy and has a lot of interesting features to it and We postulated that that could be One of the things that proteins or other nucleic acids or even drugs could interact with DNA So these simulations they provided Atomic explanations for the very sharp transitions in force extension experiments measured by others and They reveal the myriad of potential features that other nucleic acids proteins or drugs might recognize But as the reviewers of our paper said so this paper that was published in NAR, which is a journal I love Sad at science for a year and the reviewers said you must prove it. You must Test it with real DNA as if that's easy, right? So we didn't do that we published it in NAR and it took me now seven more years to do the proof And that's what I'm going to talk to you about today So how do we do the proof so in our simulations? We had only simulated twist We didn't allow the DNA to rise. We were only unwinding it and winding it as it was held, okay? So we need to get writhe in there because that's real DNA, right? So the way we did it is we needed to make a whole lot of tiny circles big plasmids are just Very difficult for many many reasons. I could give you an hour lecture on that But I'm not going to remember that in these diagrams of plasmids underneath it is double-stranded DNA Okay So we used just took advantage of all the knowledge that people had generated over the years for recombination and we put these two site-specific recombination sites here at P and at B and we just move these arrows closer and closer and closer together and We said how small can we go and there was a different question and we published that of course But what we were really trying to do was make tiny tiny circles of DNA So we could do these real experiments to test the the hypotheses put forth from our molecular dynamic simulations So lambda into grace will take these sites and recombine and make a catnated Intermediate so here's the the rest of the plasmid and here's this little catnane and topa for normally Decatonates that and so you have a little circle and this big thing Well, the trick was how in the world do we separate this little from all the rest of this mess And that took us several years to figure out that we now have done it And what's really fun is when people try to do for example of these J factor measurements You can put certain sequences into linear DNA and have it ligate, right on a pretty short-length scale But we didn't want to be caught by certain sequences We wanted to be able to put any sequence in there because we wanted to study different biological processes Maybe initiation of replication or initiation of transcription or site-specific anything that we wanted to do So it was really gratifying to us that we can put any sequence in red We can clone it in in this parent plasmid which is very easily to Manipulated since the 80s, so we did not invent this. This is easy to do. What's hard is this part for us? We can put any sequence in the world here And that's right here if you get rid of all the supercoiling and just lay this thing out as a circle Here's this sequence and so you can put any DNA sequence So unlike when you're going to ligate linear DNA and try to see oh, you know Will it what's the persistence length in fact? That's where it comes from is how much how long can it be before the two ends will come together Unlike that method this method tells you well This is what DNA persistence link really isn't real biology because this is done in a live bacterium We let the bacterium do the work and then we pull the DNA out and say wow it's pretty tiny and it's pretty supercoiled and What's really been very very exciting is it because we can put any DNA sequence in this and it doesn't have all the Negative things that come with a plasmids such as antibiotic resistance genes such as bacterial sequences This in fact has turned into a very important gene therapy vector So this is being used for important gene therapy I'm not going to talk about that today, but if you want to talk to me about that. I'm happy to to talk about it If I have any voice left, I hope you can hear me. I Usually can talk and I don't know what happened to my health these last few days Okay, we've already done a little introduction of this, but I'm going to add this component Remember that linking number is twist plus writhe and in the previous time I told you this the twist was the only thing we had changed I didn't allow any writhe and this in fact was truly planar if you looked at it there was no writhe at all from the side and We're going to let writhe happen now. So now we have to consider the writhe component So delta okay is the change in writhe plus the change in twist So if you're going to change twist you have to overcome the torsional rigidity of DNA If you're going to change writhe, you have to overcome bending rigidity of DNA these are two very different forces and So this twist and writhe question is is pretty fundamental and pretty cool and goes back to the my first love which is math But we're going to get there So what is it? What is it about making these tiny circles that's advantageous? Well, here is an atomic force micrograph of a typical size plasmid and you can see that while there they are and they're super coiled and they're kind of big and There's a lot of different conformations even in a pretty You know tight distribution on a gel Here is our new little mini circle. So they're very tiny. They're in the nanometer range and If you're going to ask for example about enzymes binding, let's say my favorite enzyme topoisomerase 2 Well, here the topo 2 has been drawn to scale If you were going to do an experiment, how many topo 2 would you add per one of these molecules? Well, you might say one two three four maybe ten right maybe ten combined You want to do one-to-one stoichiometry? I mean it's impossible and you can't do real kinetics So a dirty little secret in the topoisomerase world is that there are zero real kinetic data None of those are real kinetic data They're all relative data because you can't do it because you can't even start But look at the mini circles. They're exactly the same size So we've shown by analytical ultracentrifugation that one topo binds to one Mini circle boom we have defined our stoichiometry. We defined our binding sites There's never two to one and now we can finally for the first time do real kinetics And I'm going to digress from my question of what is the structural transition that happens with supercoiling? Just for a second to show you a little bit of binding data and just to show you what people Maybe you've never done these kinds of experiments, but this is what people are typically stuck with So here is a plasmid run in a gel and the gel electrophoresis we heard yesterday about So here's no enzyme in this enzyme. It's just a boring enzyme a restriction enzyme a core five So if you add one to four It core five to DNAs here you get some smeary stuff And if you add one to one you have some smeary stuff And if you add ten to one you finally see the supercoiled bands start to shift a little bit That's what people are stuck with quantifying They say oh here it's looking as ten to one binding or maybe something like that. It's terrible. It's ugly. It's hideous Finally at a hundred to one It's all bound so maybe you'd say oh a hundred to one is the answer. What's the answer? What's the question? It's very difficult. So here with the little tiny circle. Here's the supercoiled circle We don't have any of this other mess in our contamination no contaminants in our preps and if you add one Molecule of ecore for the core five to DNA you see 25% binding if you add one to one You have one hundred percent binding if you had ten to one It's a big smeary ugly mess and if you have a hundred to one is just digs in the well So we can very precisely Determine stoichiometry and binding and you can see that the resolution here is really nice compared to this So that's just an example of why we wanted to go smaller. It's also faster to run So here's real data with human topos topoisomerase 2 alpha the Target for about 40% of all Anti-cancer drug is used worldwide. So this is a very important drug target and we don't know so much about this enzyme So here is free Mini-circle this one is nicked and all we do is we start adding more and more topoisomerase until it shifts and we can look at this binding curve and very precisely get a KD a Rate constant for how well that thing will bind and We do that over and over again for different degrees of supercoiling because I think I've mentioned to you We can start here. We can add one One one one one or go to the positive for the first time one one one And so we did that over the whole regime using these many vectors and Here's what we see and know that there is a four law difference in the KD of Human topo 2 alpha binding to DNA Depending only on the degree of underwinding or over winding So it seems that it binds the positive and the negative supercoiling relatively equally and Much more tightly. It is now officially the tightest DNA binding protein ever measured. I Think they're going to be a lot more when they can be able to measure it precisely now with supercoiled DNA and Here is the KD for the relaxed DNA so the one we know and love up here Of course the enzyme doesn't bind to that. It's not a substrate. What's it going to do with relaxed DNA, right? Unless it's knotted and then it would unknot it, but if there's no knot in there There's nothing to do and the reason that this is an important thing to start with Before I show you the picture of the 3d structures of these things is that we wanted to know Were they biologically active? Because one of the reviewers of a grant I wrote I've never been funded for this work until very recently and The reviewer said you've made these tiny circles. They can't be biologically relevant So here is an enzyme. That's very important binding to it and here's the same enzyme acting on it and what's amazing even to us is And this looks complicated, but it's so simple Here is relaxed DNA if you add topoisomerase so here's minus topo plus topo nothing happens because it has nothing to do with relaxed DNA Here is minus one if you have the minus one and you add topo two It's gonna change in okay of two. It can't do anything with this. So it makes a mixture of minus and plus one Minus two goes to zero because it's an okay change of two minus three Goes to a mix of minus and plus one minus four goes to zero minus five minus and plus one minus six goes to zero Even the crazy Hyper-negative supercoiling twenty percent underwound topoisomerase relaxes Beautifully has no problem. So this is not some kind of freaky thing In fact, it's active and let's look for the first time in the positive supercoil arena So it plus one you get the mix of plus and minus one because plus one You know can't go to zero from one if you're doing okay change it to so the plus two goes to zero and Interestingly, it has a hard time with the plus three It doesn't act very well on it But it does do some and it only takes it to the plus one It doesn't get the mix of the plus one and minus one Which tells us something really interesting about these enzymes that was never ever known before So everyone thought that it would be able. I mean, why doesn't it go to what it did this? Why doesn't it make a mix? There's something about the loading the initial loading That then sets how far it can go. So there's this was an unexpected but all we wanted for this talk was that every one of these things is active and Every linking number when we purify them is absolutely pure. You don't see any cross Contamination you don't see one in the two or two in the three And that's important because we start to see some overlap in structure But in fact, that's true overlap and not contamination So to get at this question of what does it look like in three dimensions we used cryo electron tomography There have been major advances in the field that now allow us to do this maybe two years ago We could not have done this so maybe you know how this works So it's just like cryo electron microscopy Except the tomography comes when you take this the field and you tilt it in Increments of two degrees let's say and you can go all the way 70 degrees one way and 70 degrees the other way and now you can reconstruct like in this picture a three-dimensional image from multiple two-dimensional projections and So this is really a good measure for potentially heterogeneous specimens Which ours ended up being? So this is just a movie showing us the raw data For this is the LK equal to so this has two overlinks topo loved it relaxed it to zero and That's six percent Overwound so let's have a look at what this look like So this is a slice of the ice and we're looking at every single thing that we could see And what we did is we put little dark spots on any density We didn't care if it was a circle or not any density got spots And this is what we ended up with at the end of the day So this is what a typical field looks like Of the raw data, and that's the only raw data. I'm going to show you And this is the Summary and this is now published so you are welcome to look at it and the rest of the stuff that I'm showing you is not published So here's a gel that you you're kind of used to looking at by now So here's relaxed DNA plus one plus two plus three here's minus one minus two minus three minus four minus five minus six So this is how they run on gels. These are acrylamide gels by the way, not augurus gels So representative Pictures of these three-dimensional objects are shown here and very commonly we saw these ellipses Okay ellipse never was it a perfect round circle. That was surprise number one We thought these are three hundred thirty six base pairs. That's two That's a coon statistical length or two persistence lengths We thought that thing wanted to be open so much that it was just going to be this ready to burst into a perfect circle But it's elliptical. So that was a surprise never was it a circle. So very commonly we saw these ellipses and We also saw some kind of kinked ellipses and we can look in Rasmol We can look in in three dimensions. So we look very carefully in all dimensions for these calls We see the figure eights and we better have seen those This is where everybody says we should see right where the things seem to actually be touching in the middle And then we saw things that looked like rackets where there's a handle here in a large loop or a loop on two sides Or a loop on one side and then a long string and then just a complete rod And if you turn these 90 degrees because as I said, we're looking in three dimensions here This is what they look like. You can see this is very planar. This one has a little bit of of writhe to it and This one's planar, etc, etc If we quantify this and as far as we know, this was the first time anybody had quantified Cryo em images using individual particles So what you see people do is when they get these low resolution images as they average a whole lot of them And if you average 8,000 of these, you're gonna get a perfect circle, right? Because it's all gonna cancel out. So we did not average anything. What we did is we counted three people blind They didn't know what okay, they were looking at said, what do I see? And they you know, we looked at it in three dimension on this big big screen and Everybody went blind And anyway, did I ever see it look like this? Well, it relaxed DNA you mostly see looked like that But sometimes you see relaxed DNA with a little bit of a bend in it But you never see relaxed DNA in a figure eight or a racket or handcuffs or a needle or a ride ever Well, what about Nick's DNA? That's sort of our control Nick DNA and relaxed DNA might be different Right, so Nick DNA can have no torsional stress because there's a Nick there The Nick's DNA looked exactly like the relaxed DNA Exactly in here is how many individual particles were counted So if we go in the negative direction and that's this way What you see is when you go from from relaxed to minus one you see this huge transition point we're suddenly the DNA is majority a racket and a Handcuff and then some other stuff. We'll get back to this other You add another super coil and you get every single possible confirmation ever seen What? So that's why I thought these things have contamination and that's why that previous experiment is so important That DNA is exactly the DNA that we froze. So there isn't a different, you know, some other hidden Contaminants So in the minus three we finally start pushing this thing into the rise confirmation You might notice something different about the riots that we saw versus the riots that most people draw Because of the the charge of DNA Most people when they draw a rise DNA Kind of draw it like that Right with this kind of openness. It doesn't want to touch. It's highly positively charged what we noticed with these small circles right off is that The inside was absolutely smashed closed. So it would rather rise Than have that circle be bent So that circle wants to be open and that dominates it seems for these linked scales and Finally with increasing negative supercoiling we finally get rise DNA and then more and more and more Of this other that we'll get back to What about the positive super quiet? I told you nobody's ever looked at that before so this is a whole new world Well the positive superquiling had a pretty dramatic electrophoretic shift We predicted that this would look just like the perfect little figure eight because we that's the only structural transition We could imagine that would cause a gel electrophoretic shift. In fact it looks exactly like the relaxed DNA and We did this over and over because we couldn't believe this so What you see is it's mostly of the ellipse with a little bit of the bend and then the complete writhed What we hypothesize is that these things are writhing very there's a transition very rapidly between complete writhed, but it can't stay there and complete open and we can't Get the stuff in the middle, but we're trapping those two transitions and that's telling us again about this incredible dynamic structure of DNA and Somehow that transition is what's captured by gel electrophoresis because this is pretty major But people would say because this is not a bleary band. It's not a doublet look like a single band What in the world right so that cross must be very rapid and enough to be causing a gel electrophoresis change Finally when you go to the plus two Just like when you went to the minus two you open up every single possible confirmation except no of these other none Then at the next transition point to the plus three We see some of the open and then a lot of this writhed and the majority other What is other? Other are things that have never been predicted by elastic rod theory or by anybody or anything They don't make sense. They're branched. They're three-legged. They're s-shaped They have four lobes L shaped and we repeatedly saw these really freaky shapes What are they? Are they base flipped because if you have a base flip you're now a pivot point at the other base So you're infinitely flexible. We think at the end of each of these dramatic bends is a base flip or a kink Or perhaps there's even Pauling like DNA and here it's just a little movie of the negative super coil compared to the positive super coil in Three dimensions you can see the handedness very easily and you see that this the bend here must be incredibly tight Right incredibly tight so that how do you facilitate such a dramatic bend? I think it has to be a base flip All right, so we always did molecular dynamic simulations. We either did explicit or implicit Solvated simulations To accompany all of these. I'm not going to show you those So if it's base flipping then that means bases are exposed and bowel 31 is an enzyme that probes for exposed bases And what you can see is as you increase this is the same kind of gel We've looked at before Relaxed DNA doesn't react with bowel 31 because there's no exposed bases. So nothing if It's underwound by one it like it gets clipped immediately if it's minus two so on Very very rapidly is it clipped by bowel 31 and then the positive direction not until you start getting these other Do you now have? The bases exposed So if you look at the threshold of well bowel 31 will cleave There's a very strong threshold of around guess what? Biologically relevant supercoiling Bases are exposed at the point where DNA starts to become active It's pretty cool in the positive direction. It takes a lot more positive supercoiling to finally expose bases But again, you can do it but if you put these on the same scale this shows you the kinetics of base Exposure of these circles and I'm not going to go into the details of this But we did this with chemical probing as well. So this chemical will bind to exposed bases exactly the same effects So we mapped where the cleavage sites were and that's really cool, but if you know Craig Benham's predictions He he has this beautiful algorithm. We plugged in our sequences into it here It is online He said that we should have a bowel 31 cleavage site right here Here it is with very very high confidence But we mapped our bowel 31 site right here and our molecular dynamic simulation said right here is where the base flip was so this makes argument for this cooperative kinking model of Stasiak And again to in the talks about how DNA supercoiling can communicate across Circles, which is pretty neat I'm not going to go into that because Marielle is telling me that I'm out of time But you physicists are always asking me for effective salts on these things So we haven't done this for all the structures that we have looked at the effect of salts on gel electrophoresis and what you can see is if you don't have Salt in the negative supercoiling it just flattens out and doesn't run. So there's not a lot of rye But if you add the salt boom you can flatten out So this is a very sensitive way to compare across salt So we did this over and over and over and over and here's some the bottom line for all of these data Is it in the positive supercoiling domain it doesn't matter if you have salt or magnesium or calcium or what it is impervious The structure seems to be impervious at least if as measured by gel electrophoresis Mobility, but negative supercoiling is highly dependent on the salt concentration and here it is for magnesium physiologically Magnesium is around five millimolar But at the point of mitosis It suddenly jumps to 50 millimolar. So we're thinking that magnesium influx and calcium influx could have dramatic effects on Shapes of supercoil So this is how we explain all of these data That the relaxed DNA can become positively supercoiled and more positively supercoiled But in the negative domain it can become negative or it can start to denature in low salt But with higher salt you can force that to be more rye and that can somewhat act elastic rod-like And in summary each mini circle that we looked at each topoisomer has a very reproducible unique fingerprint of its distribution of confirmations and Each topoisomer is exquisitely sensitive to Monovalent and divalent cations and drugs. I didn't show you that today The positive and negative are dramatically different DNA behaves like an isotropic elastic rod for the very little bit of positive supercoiling and other than that it does not ever look like that so I showed you that about 31 cleaves that's indicate indicative of Basis exposed and glyoxal binds all the negative supercoil topoisomers and the most extreme positive and The negative supercoil DNA demonstrates this cooperative kinking so in the future we're going to start adding enzymes to this drugs and See which things are bound by what and These are the wonderful people I get the privilege of working with today I talked mostly about work from Jonathan Fogg a new assistant professor in my lab Jamie Cadney is a former Postdoc has now taken a job as a professor at Rice University across the street Steve Ludkey is The creator of E-man and he's really wonderful with his student Muyan Sarah Harris could continue to do the molecular dynamic simulations of the circles and watch you and his long-term associate Mike Schmidt and her hit their graduate student Rossi Did all this work and this is our funding? We're very grateful for and here is my awesome lab, and I hope I can take some questions