 It's a pleasure to introduce Professor Taikjib Ha, who comes to us from Johns Hopkins today. He's been at University of Urbana-Champaign for 15 years from 2000 to 2015. Before that, he was at Berkeley, where he did his PhD, with the leaders in the field of bringing physics to biology, working with Raymond Genleau's, Steven Chu, Simon Weiss, etc. And before that, he was at Seoul National University, where he did his undergraduate degree in physics. So he's a physicist, a Bonafide physicist, who's come over to biology and has brought with him a variety of concepts, approaches from physics to biology, and now has been able to address some of those mysterious things like helicases and other enzymes that act on genomes, complexes that function within cells, building ever-most sophisticated tools to actually look at functioning molecules in the milieu within which they function. For doing this, pioneering this frontier along the way, he has won every possible award that you can imagine, starting from Seoul scholars, Sloan, early days, to the Ho-Am Prize, which is considered the equivalent of an early Nobel Prize that is given to Korean American scientists, to this year, he was elected to the National Academy of Sciences. We noticed his work early when he started talking about DNA structures and their plasticity, right from the days of early papers, about 200 papers ago, 197th paper, that talked about Stephen Chu, by the way, who's the energy secretary also, has done enormous amount of work in single molecule studies, bringing physics against biology. So how DNA bends, we had studied in school that DNA has a persistence length of about 500 nanometers, and here was someone who's claiming that DNA is far more plastic than that, far more mobile, far more move-up-changeable than that. And since then, we've been following his work in every paper. This year, I believe he's had three or four papers in science, cell, et cetera, already. And I'm looking forward to his stories today, and hopefully he'll continue and collaborate with small molecules that we've been developing in the lab. With that, let me not continue further and hand over the stage to TJ Ha, as he likes to be called. Thank you for coming. Thank you very much for the kind introduction and invitation. Does it work? So I'm between jobs, so it's nice to be invited to get some free food and housing. I'm a physicist by training, I forgot most of my physics and haven't learned enough biology to call myself a biologist. So to paraphrase a bridge in spears who's saying, I'm not a girl, not yet a woman. I'm not a physicist, but not yet a biologist. So it's really fun to be here, lovely, beautiful campus and wonderful colleagues and students. I've learned a lot, thank you very much. Many years ago, Watson Creek published this paper proposing the structure of double helix. The paper ended with a paragraph saying that, well, it has not escaped a notice that the specific pairing suggests a possible copying mechanism for genetic material transfer. And they actually worried that this was not explicit enough, and that maybe someone else may take the credit for the discovery by making the claim more explicitly. So a month later, they published another paper proposing this base pairing, and then the rest became history. That paper ended with a paragraph, a lot to be learned, even if this is correct. For example, what makes the pair of chains unwind and separate when DNA forms a stable double helix? It took more than 20 years until people discovered a class of enzymes called helicases. Helicases for DNA molecules, DNA unminding enzymes, unzipping proteins. They can be also considered as motor proteins because they use ATP energy to power themselves on the DNA tract. In fact, if you give them a single-celled DNA, they will move on the single-celled DNA in a directional manner, 3 to 5 prime or 5 to 3 prime of the DNA. So you can call it a translocase. They function in many different processes inside a cell. They function in essentially every possible imaginable DNA RNA metabolic processes. They unwind DNA or RNA. They can remove proteins bound to nucleic acids. They can cause branch migration, highly resumption. They can do chromatin remodeling, RNA protein complex remodeling, and so on. And when they are not well, if you have mutations in some of these enzymes, you can get serious human genetic diseases. Some helicases form a ring of hexamers, and others are non-hexameric. There is some discussion whether they function as a monomer or dimer that I touched upon today. You know, there are DNA helicases and RNA helicases. And some 5 to 3 prime helicases, meaning that if you want to see unwinding of a DNA, you need to present the enzyme with a DNA with a 5 prime tail so that it can bind to the tail first and then translocate in the 5 to 3 prime direction to unwind the DNA. And vice versa for 3 to 5 prime helicase. Some people are so in love with these enzymes, so they pay additional money to the state to have a license plate like this, it unwinds. So you can actually, if you're not interested in the topic, you can just sit and unwind for the rest of the day here. So this is actually from Missouri, my long-time collaborator, Tim Newman, who has received his PhD in biochemistry here with time record. I am not, I'm too cheap for doing this. So you know, interesting questions are, you know, how do they function and how are their functions regulated? And for example, you know, if you switch, you know, how do you turn on and off the helicase functions when you need them and when you don't need them anymore, because if these enzymes float around inside a cell and unwind every DNA, RNA they see, it can be digested. You need to tightly regulate their functions and how do you achieve this? Okay, this is an important number because we do one molecule experiment, a single molecule experiment. So the main tool that we've been using is called a single molecule fret, where if you want to measure conformational changes of an enzyme, you put green and red dye molecules on the known site so that you can distinguish between closed and open conformations. In the closed conformation, when you excite the green molecule using your laser, energy is transferred to the red and you get red photons instead of green. In the open conformation, you get green photons instead. If you want to measure how quickly the butterfly flaps its wings in an ensemble, you must synchronize a reaction by convincing them to start from one conformation, one sitting on the leaf and then you can frighten them into taking off the leaf and perform relaxation measurements. But as you can imagine, in many cases what you measure is not the wing flapping frequency, but how long it takes for the butterfly to take off the leaf. Ideally, you want to measure this in real time, looking at a single enzyme as a function of time and measure the anti-correlated changes in red and green intensities so that you can deduce the flapping frequency of the wings and also amplitude of the motion. This can be done in many, many different labs in the world these days and can be used for studying amazing variety of different systems. You can also measure intermolecular changes using a single molecule threat. For example, if you have a helicase moving on a single-send DNA from 3 to 5 prime N, putting a green dye on the protein and a red dye on the destination end of the DNA, you can imagine seeing as a function of time, greens it is going down and red is going up due to increase in threat. This is not actual data, it's a powerful animation that I made, but our data is actually almost as pretty, so this is actual data. As a function of time, you see red signal is going up, green is going down, and the surprise here was that once it goes to the end of the track, it somehow goes back to the beginning and repeats the process many, many times. We did a lot of experiments of this kind to get at the physical mechanism of this process and also what it might mean biologically. We wrote up a manuscript and we were able to convince some people that they can publish this paper several years ago. And then I paid a lot of money to a local artist in Urbana-Champaign to draw a cover illustration. This is one example where this old man is sitting on the DNA and then reeling in the DNA loop and then kicking up a protein's bound to the DNA. And unfortunately they didn't take it and instead they picked a picture of a diabetic mouse, so it was very disappointing. It's okay, sometimes you win, sometimes you lose. Here's another picture that they didn't take. My student, Gia Park, is sitting on the DNA and then she's the one who did the work and then reeling in the DNA and then kicking up the asteroid's bound to the DNA. So I spent so much money on these cartoons, I have to show them somewhere, so that's why. Another very popular single-marking method is called optical tweezers and it's basically chopsticks made of light. So if you shine infrared laser in a tightly-focused spot, then a small particle can be sucked into the middle of the spot and then if you move the laser beam around, you can manipulate the position of the bead. If DNA is attached to the bead and then there's an enzyme sitting on the surface pulling on the DNA, you can play a tug-of-war and you can apply very tiny forces, about a peak on the level of forces and then measure the response mechanically using forces and also by measuring the DNA length as a function of time, it can go down to a single base-pill resolution as Stephen Block has shown. The idea of using stretching to extract information is not new and this is an old idea. In medieval times, people are stretching other people to extract information and asking questions such as, what is this? So because my lab was doing just fluorescence measurements for many, many years, I was really envious of people doing optical tweezers and getting this wonderful data and I am a man of many envies so I had this force envy and this I wanted to combine the optical tweezers with single molecule fluorescence and the idea is to measure conformational changes using fluorescence or FRED as a function of applied force and actually there have been many other attempts in this direction and some are listed here so I'm just telling you about our own version of this direction. So if you think about the optical tweezers, you are performing mechanical measurements but with your eyes closed like that but if you do purely fluorescence measurements, you have arms tied in the back but looking at the process with your eyes passively and the idea here is that by combining the two, you can sample the best of both worlds. So Michel Wang was kind enough to host my former post-Saxon Hong for a week, so he spent a week in Cornell University more than 10 years ago to learn about optical tweezers and then he came back and built an instrument combining single molecule FRED and optical trap and he was able to show that you can measure DNA junctions dynamics using fluorescence but as a function of force now he is on the faculty of Seoul National University and another student then continued on the same instrument Rubozo who is now a post-Saxon Hong at Harvard what he did was to work with team Romans group to pull on the single cell DNA wrapped around a protein called SSB and then when you begin with this state of a fully wrapped DNA where two dies are close to each other you get a high threat, may need red signal and then as you increase the force you will gradually peel the DNA of the protein surface giving rise to gradual decrease in threat that you can measure as a functional increasing force. So this is the actual data that he acquired so as a functional force you can increase a force showing gray from low to high value gradually you can return to the low value and then repeat it five times and these are the raw data of red and green intensities of the DNA die and they are changing in an anti-collect matter and the blue is the fred efficiency that is dropping gradually as you increase the force really showing that DNA is getting peeled off gradually little by little as you increase the force showing that we can do these kind of measurements and so on. Actually data was so beautiful that when my students showed me the data for the first time I almost cried but I didn't instead I told them go back and take some more data as every good advisor should do. And then we became more ambitious so Jan Shemla is a colleague a former colleague in Albanian pain and so he came from Carlos Busmante's lab so he's an expert in high-resolution optical tweezers. So we combined together single molecular fluorescence and optical tweezers of dual laser traps you can actually have much higher resolution and sensitivity and here the idea is that if you trap two beads and then you can shine protein in the middle of the DNA you can measure fluorescence from that protein but at the same time measuring the mechanical response with down to single base resolution of tweezers. Here we had to go back and forth between different lasers because if you have the trapping laser on at the same time as a fluorescence laser excitation then you get 10 times faster photobleaching so we have to actually do this like 60,000 times a second to avoid that issue. So it's actually pretty complicated instrument that that Matt Comstock built. Matt is now physics professor in Michigan State University so I'll show you some data that he'll acquire using this instrument later. Okay so coming back to the helicases how does a motor move on the track processively? One possibility is to use the inchworm mechanism. You have two units of a motor protein three different colors and then in the first step you expand the inchworm but in the process if you weaken the binding if you have a weak binding to the front then upon expansion the front head will go forward but next time you compress the inchworm but if you modulate the binding affinity so that now the front head is bound strongly then upon compression the rear head will move forward and you can repeat this process multiple times to move in the track linearly and for non-hexameric helicases the rigorous lab has shown really good data using structural biology that this is how the helicases can move on the single-send DNA track one base at a time. So here's an example of helicase from Hepatase virus NS3 and and this mechanism requires two non-hexameric subunits and these are provided by REC-K4s as you are familiar based due to my my caucus work on REC-K so you have two like REC-K like domains here and the ATB binds in between and the ATB binding then causes the two domains to approach each other compression and weight ATB associated inchworm expands and this can repeat many many times. So here's a structure of the hep C helicase bound to a piece of DNA and if you look at the structure of the protein without any ATP then highly conserved Australian residue is shown here and blue in contact with the phosphate backbone of the DNA three nucleotides apart in this confirmation but if you look at a similar helicase bound to an RNA in the presence of ATP analog then there's compression, inchworm compresses and then there's two Australian residues now two nucleotides apart so that gives the basis for one nucleotide movement along the DNA per ATP hydrolysis. So here's an animation that illustrates how this could happen so two REC-K like domains here and here can move on the DNA backbone one base at a time for ATP hydrolysis. But then how do you couple that translocation into DNA unwinding? One possibility is a following you unwind one base pair every time you translocate on the DNA backbone by one nucleotide. There will be number of base pairs on one versus number of nucleotide translocate one at a time and this was proposed by structure biologists you know this example for example by Wei Young's group where they solve the structures of UBRD helicase in many different forms and then they were able to show that there is a one base pair unwinding for each cycle of ATP hydrolysis. So this model will say that you have continuous process of one at a time indefinitely until the enzyme force of the DNA but we have some interesting single molecule data that we actually use to propose that for at least for NS3 helicase that unwinding is not happening every time you translocate one nucleotide we propose that you wait until three nucleotides have been translocated and then you unwind three base pairs all of a sudden and it continues this way more like a spring loaded mechanism. So in this animation what we are trying to depict is the idea that as a helicase remains move on the DNA backbone the third domain stays behind due to phenylalanine residue stacking against the DNA and residue and then after three base pairs of translocation you unwind three base pairs in a burst. Another data of similar kind was actually was obtained by my former postdoc Wang Lu-kui and in this case we are looking at what's called RIP44 it's an exoribor nuclease that chops the RNA from the end one nucleotide at a time but you can also dice it as RNA that has secondary structure as you flex portion without using an ATP so it uses the hydraulic energy here to cut the RNA one at a time and we also found that in this case our unwinding of the duplex does not happen one base pair at a time although we know that motion itself is prepared one nucleotide at a time and we estimated that it's about four base pairs unwound every after four steps of one nucleotide digestion step so that you can essentially compress the protein with four steps and then it can snap open multiple base pairs at a time and the idea of this type has been actually proposed for other systems too for RNA polymerase initiation there is a mechanism called scrunching where you can build up energy in the system until you make the transition from initiation to elongation phase and escape the promoter and 529 DNA packaging motors have been shown to have multiple layers of actually a stepping not just single size step that repeats indefinitely so so that was my introduction to for the main part of the talk so we are studying super family one helicases named rap UVRD and PCRA at their function in replication we start mismatch repair and basic decision repair and PCRA actually functions in basic replication but they are all bacterial origin they are all stripped to 5 prime DNA helicases again they require 3 prime tail for unwinding to be seen in vitro and they are actually very similar in terms of structures and in fact I have a structure here but I don't even remember which one it is because it doesn't actually matter because they're always similar there are two major confirmations that differ in in the rotation of this domain colors differently here around the vertical axis and this will become important later in the talk so we have interesting questions you know what is a functional form is it a monomer or dimer that is required to unwind the DNA and when you have two forms of the helicase close and open which one is active for DNA winding and also how is unwinding regulated so there's one model that you know Tim normal and I propose based on mostly based on his data but also based on some of our data I share with you shortly that UVRD helicase functions in this manner as a monomer you can move on the DNA three to five prime directionally but it cannot unwind the DNA at least in a detectable manner as as a monomer but when another one comes by then magic happens and you can see unwinding and the other model is you based here that you need just one monomer to unwind the DNA processively and this process can continue you know on and on so my student Kyung Seok Lee built a set up combining optical trap optical tweezers with terp for total internal reflection fluorescence microscopy so the idea is that you tether one end of the long singles and DNA on a surface five prime and and then you stretch the DNA using an optical trap and when the protein lens on it moves from three prime to five prime direction then you will see fluorescence signal moving in the laboratory frame in in your video microscopy and this is how we prepare the DNA so we start with double-send DNA tether to a surface we use an external clays to digest one strand sometimes all the way down sometimes we know not not quite so you can have some meaning double-send the DNA portion and then we can use the optical trap to hold one end three prime and then then stress the DNA vertically in this direction and then then start the experiment so that's the beat at this is a surface attachment point and I will show you a movie and this is a fluorescent junk so you should ignore technical difficulty sorry so what what you would have seen is actually spot appearing in the middle of the DNA and then going down gradually and then in addition the spot becomes brighter and brighter okay and and you can actually plot this in the form of chymogram so you can take the section of a movie apply it as a function of time floating by somewhere here and then it comes down toward the surface for the five prime and and it gets brighter because not because it's collecting cosmic dust but because it's getting closer to the surface you know we have this evidence field excitation or that decays exponential with it as a function distance from the surface in fact you can use this measurement to determine the depths of the laser penetration into the solution and and you can then determine the speed and so on very accurately so we on some DNA molecules we are we are lucky you can see many many evrd helicases binding and translocating to the surface one after the other and I call this my meteor shower because it's so beautiful and you can measure the slope and then the speed that we get is it's almost identical to what Tim Norman measured using his work phase essays in some cases we see that evrd comes down and then actually stops here there's another one coming down but this one is another coming down stopping at the same location here and this is not because the enzyme hit the surface because you are still quite a bit far away from the tethering point right here so idea we have here is that monomer comes down and then it encounters a duplex DNA because of incomplete digestion and just get a stole right there one a support we have for this model is shown here what you can do is to perform mechanical measurements of DNA stretching to estimate how long the doubles and DNA is remaining here and then we can also measure that distance of the store position to that tethering point and you can plot them one at relative to the other and we get a straight line with the slope of one supporting our idea that the enzyme as a monomer stores over it encounters a doubles and DNA because it cannot mind the DNA beyond the junction more direct way of looking at this is the following we can put DNA oligo in a defined location using a particular sequence and you put two dice there just to make it a bit brighter and then here is the DNA location where duplex in a exist and then you feel the binds here and then comes down and it stores in the same exact location and there's another example here this is where we mark the DNA for the duplex junction you really comes down is storage at the same location in fact if you measure that this separation between this and that is essentially zero with plus minus 20 nanometers close to our precision of the measurement so supporting the idea that when you really encounters a double-sand DNA it stops there it doesn't online beyond the junction at least within a resolution of about 20 base pairs if you do experiments in higher protein concentration you do see unwinding and here's an example so you really comes down and stores at duplex junction and then later on then it starts to move again but when it starts to move here fluorescence suddenly becomes brighter so it's saying that there's another protein that just came to the spot at the same moment and so the idea we have is that it's coming down at this rate and then it stops here and then another one comes and then as two proteins then it can unwind the DNA but at a slow rate it's known from teams measurement that unwinding is about three or four times slower than translocation in addition you can measure the force on the bead and then show that at the very moment that force starts to decrease now because now generating additional single-sand DNA so so the kind of data we have here and then other kinds of that is just that UVRD as a monomer is not able to unwind DNA at least not very well you need to have one more than one protein to achieve so the second interesting question was the following you know why do you have two different confirmations open and close for PCIe UVRD rapid cases in in the open confirmation this was called to be domain is so open and the close confirmation is swivels around the vertical axis by 130 degrees to obtain this confirmation and it's a really large movement some pairs of residues on the protein can change the distance by more than 50 angstroms and because this was crystallized with duplex with a 3 prime tail a natural substrate for the enzyme it was natural to propose that this closed form is a functional form for DNA unwinding just want to show you you know all of the structure that have been solved over the years PCIe was the first ever helicase whose structure was determined Apple form in the open form and red was solved in by Tim Norman and Waxman bound to singles and DNA both open and closed and PCIe bound to partial duplex set in the closed very young structure also you know UVRD closed and UVRD as by as Apple is is open so basically all three helicases have two different forms nearly identical and because of the these these structures it was believed that the closed form was the active form for unwinding which is I think very reasonable but but we have some different ideas for example when team moments lab deleted a to be domain then protein still function so that was in contrast to the proposal here where this to be domain in contact with the duplex is essential for unwinding the DNA actively and in addition team group showed that this mutant without the to be domain on my DNA as a monomer not very well but still pretty good you can on my 30 base with DNA as a monomer and third we have some single molecule data where where the rapidly case can move on the three-prime tail but it encounters a duplex and snaps back to the beginning and repeat the process many many times but if you if you label the protein in such a way that high-fret between the two days reports on protein domain closing what we see as a functional time is that as a protein approaches the duplex fret goes up gradually meaning that to be domain closes gradually and eventually then it goes back so we thought that maybe closing of the to be domain is a signal that it has reached the junction that needs to go back as a regulatory mechanism so this is a cartoon that we made is a DNA with three three-prime tail and butterflies approach approaching the junction as you approach the junction it slows down the flapping and then stops at the closed form and the stops there and then it goes back so that was the idea so that's why we thought closed form is inactive for unwinding so so that was our proposal so five years ago I met Mark Dillingham who is a structural biologist in UK at a meeting and we had a small argument that I told him that I think is the open form that is active on winding and Mark told me that well TJ I like you I like your work but I think you're wrong here so he said is it must be the closed form so if I thought well you know we probably need to have actual data right to address this point so I emailed my student Zinan well you know why don't you cross link two systems two systems on the helicase so that you can stabilize a closed confirmation and show that that form is inactive and and that should prove him wrong you cannot prove me right but you can prove him wrong so so that's what he did so he put two systems here on 2B and 1B domains shown in the red residues so that in the closed form there are seven enzymes apart so you can cross link them using bi-functional cross linkers but in the open form there are 42 enzymes apart you cannot cross link so actually if you do it this way then you cannot take the open form so so we called this version rep X for X for cross link then we made a control construct where we put the systems in this manner so that they are next to each other in the open form but 28 enzymes apart in the closed form so that should stabilize a protein in the open form we call this rep oh I miss this spoke I would say rep wide so about that so question was which one is active on winding and I think I'm gonna skip this slide but I was really impressed because my physics student was able to come up with essays to to show that a cross linking is actually intramolecular but not intermolecular and and then he did unsung 80 pH essays and it turned out that 80 pH 80 pH rate didn't really change upon cross-linking so I was very disappointed but I told him there maybe you need to give them some more difficult task of unlining the DNA and then you'll see a difference so the essay here is a flat base essay ensemble fret you have a label DNA initially high fret but if the enzyme on minus the DNA you get a reduction in fret so an ensemble fret measurement as a functional time rep wire type on my DNA very slowly even at high concentrations but rep X on my DNA much faster so I was actually surprised because I had to pay some money with against mark 50 cents and but but I knew that my student was not fabricating the data because he was doing me directly opposite to what I was expecting to see I was right so and then he did some more controls so if you have two systems but do not use cross-linking then you don't get this fast unwinding of rep X and if you do the same with rep Y cross-linked into the open form that is the same as the wild time rep X is much faster so it's not just cross-linking or topological enclosure of the DNA that causes fast unwinding but really the closed form stabilization so the next question is that why is it because you form dimers better with cross-linking or maybe you have now mono-modic helicase that is very active so with this single molecule fret experiments we immobilize the enzyme using a histate histin tag and antibody on a surface and then we added DNA shown here a label in such way that you begin with relatively a low or medium fret and and so you get similar intensities in green and red and when the protein binds DNA binds to the protein on the spot you see a sudden increase in fluorescence and nothing happens until DNA dissociates and then another DNA bind dissociate and so on especially when you have rapidly case bound to the surface and it captures the DNA it doesn't on mine it just you know horizontally forward and then release it releases the DNA but and same for rep Y but if you do rep X you see something different so you see initial binding and then fret actually increases and then fret dropped to zero and then you lose total fluorescence signal and you see three events that happen in succession on the same helicase molecule so what's going on here is that when helicase unwinds the DNA then on one strength becomes coiled up bringing the that die closer to the green on average and then that gives you an increase in fret when you unwind the DNA fully then red strength is gone so we you get zero fret here and then finally when you release the green strength then you go back to the base level fluorescence and you can compete as many times it turns out that 80% of the binding events actually result in full unwinding it's showing that rep X monomer is now active as a monomer and it can unwind you know these duplexes are quite well with high high yield so then we decide to test how how good a helicase is by using optical tweezers experiments so we work with a student in Jan Schemner's group in Urbana so there we have two dual-trap two tweezers and there's a bead with a DNA six kilobase per long DNA with a three-prime tail of various lengths and then second bead is coated with antibodies against the histidine tag so you can grab wrap X on the bead surface and then then once you form a link between the two beads and if you have ATP then helicase will unwind the DNA and then the DNA between the two will become shorter and shorter so distance between the two beads will decrease as a function of time so this is the actual experiment you can in the microfluid chamber you have a capture channel with with ATP gamma s to establish a link and then you move the tethered DNA to the ATP chamber and then when when you move into the ATP chamber shown by this boundary then the length of the DNA starts to decrease and monotonically with some pauses and basically what we find is that every binding event you see result in full unwinding of this long DNA I mean at the bottom we have to stop the reaction because two beads are collided each other where you cannot do measurements anymore but you know we believe that if DNA is much longer than it probably have no problem in unwinding even longer DNA so it's highly processive let me likely buy the lengths of the DNA and this is true for different tail lengths and also different force and with and without SSB so that was really impressive because you know you know because it can be unwind DNA as a monomer for such a long distance and this is comparing wrap X with wrap Y and wrap you know fraction of our but unwinding events if you establish a tether only wrap X show this incredible ability to unwind DNA over long distances you can also perform the measurement where you don't have the force feedback so the force is free to increase and then you can see that enzyme can unwind 30-40 piconewians of forcing force before tether breaks and you can also measure the average velocity normalized velocity as a function of force it doesn't decrease with increasing force up to the moment the DNA undergoes its own transition so we now call this wrap X a super helicase because this is super at least to our mind so the conclusion here is that I think I lost my bet with Mark and so the lesson here is a never bet against a structural biologist now if the closed form is active in unwinding then why did nature create the open form this question can be looked at using what's called the hairpin essay so we we have hairpin in the middle of a DNA and double trap configuration and the protein comes and unwinds the DNA then as you unwind the DNA hairpin the distance between the two base will increase as a function of time and then you can read using the optical trap but because now we have this instrument that can also measure fluorescence in the middle of the DNA you can then use that instrument to measure copying on the above the protein monomer dimer or protein confirmations so this is the experiment where we are measuring our number base based on one as a function of time for the hairpin and you can see that DNA is being getting on run by a single UVR D enzyme labeled with two dies so that high fat corresponds to the closed form and low fat open form and you can see the unwinds at least on the tension you can unwind DNA as a monomer to a limited degree and then it can but instead of continuing fully comes back down and unwinds and receives online receives and it repeats it many many times basically sitting there idling you know without really going very far eventually it dissociates if you measure of a fret from the same molecule simultaneously these are the raw data of green and red intensities and this is a fat efficiency that you calculate and we can begin to assign different fresh stage low fresh stage for the open form and high fat state for the closed form and and then then we can actually color different segments of the trajectories when Fred is high you color this segment in red and for all of the high-fat segments and likewise for the low-fat segments then if you look at the optical trap trajectory when Fred is high DNA is being unbound when you can see them here here and here in here and then when Fred is low generally DNA is getting received so the situation is actually more nuanced that you have the closed form high-fat form is used to unwind the DNA but then after unwinding some number of base pairs then enzyme somehow switches back and then causes re-zipping of the DNA but in that process it takes the open form you can do this from many many molecules and plot a two-dimensional histogram Fred versus velocity so when Fred is high velocity is positive you're unwind the DNA and Fred is low it's negative you are re-zipping the DNA so closed form for unwinding open form for re-zipping so we have this model called a strand reversal model where helicage is moving 3 prime to 5 prime direction to unwind the duplex DNA for after unwinding some number of base pairs it undergoes some kind of transition to strand switch to the 5 prime strand and then it are translocated on the 3 prime to 5 prime direction on this strand and then it causes re-zipping of the DNA upon its awake and they switch back and then repeat the process many many times so that's the idea behind this model and the idea is that when it's unwinding it takes the closed form contacting the duplex DNA with the 2B domain and then when it needs to switch to reverse the direction it maintains a contact with the duplex using the 2B domain but it undergoes a conformational change into the open form so that now you can engage the other DNA tail and translocate on that tail and causing DNA re-zipping so that is the strand reversal strand switching model that we proposed so the idea that we have right now is that in DIBO strand switching and re-zipping after unwinding just a few base pairs can keep the enzyme in the off state because you want to regulate the function again you don't want these enzymes to go berserk and unwind every DNA that they see right so you want to keep them in the off state but in the right place so if it's idling there unwinds a little bit and come back and so on comes back and and then likely in DIBO a partner protein can come into stabilize the closed form and then switching it on and making it a really super active helicase in RepEx we actually make it a super helicase by this allowing the open form required for strand switching so you cannot strand switch anymore now it has to keep going in the same direction we also did the same experiment with the PCRA and likewise Fred and then optical trap measurement showed that PCRA X in the closed form becomes a super helicase and for PCRA we have a known protein that is known to activate this helicase function so what we did was we purified this protein called RepD and then measured the PCRA proteins confirmation using single molecule Fred and without this activating protein we see single molecule Fred histogram shown here in black mainly open form but if you add RepD protein that is known to activate helicase activity then you stabilize selectively the high-fat concoct confirmation so this suggests that perhaps RepD makes the PCRA enzyme a really good helicase by stabilizing the closed confirmation and by preventing strands and switching reaction so we have a monomeric super helicase that's nice because you don't have to assemble multi-meric enzymes and there's no nuclear activities unlike in red big CD so we can envision several useful directions in biotechnological applications one idea is to use this for single market DNA sequencing in nanopores why because nanopores can read the DNA sequence by changing its current through a pool the helicase can provide DNA one base at a time at a measured pace because our helicase is very processive and robust against force you can read thousands of base pairs in one in one DNA second application is called isothermal DNA amplification in PCR you have to go up and down in temperature because you have to melt a double-sand DNA you just made but if you have an enzyme that can unwind the DNA to the melting for you then you don't have to do that you can do everything in one temperature and that can be useful if you want to have essays for detecting pathogenic DNA in Africa or if you want to have a quick test for DNA content in in few situations where you don't need you don't want to wait for an hour before PCR is complete so here's an example you're working with Jens a good luck in University of Washington he's a single model key sequencing platform and so we are using our super helicases to feed the DNA into the nano pool and then read the current as a function of time you see many different levels for different sequences and you can call different levels and then and there is an algorithm to align the current data to the known sequence of this phase DNA then we can actually see pretty good correspondence between our measured current levels in red and then predicted current level so you know it's all the voice shows that super helicase can be used for you know getting information from DNA at the single molecule level over you know many many thousands of base pairs here's another example of doing isothermal DNA amplification so we add this PCR X which is actually from a thermophilic organism so it functions at high temperatures and then you add DNA polymerase and primers and so on with a 1k base 1k or 5 kilobase DNA templates and we can amplify 1 kilobase DNA 5 kilobase DNA at a constant temperature there's no thermocycling if you use wild type then you get a smear here so it's actually beginning to work you can amplify DNA you know even very long DNA and I think there are several interesting applications here let me stop Sinan Arsalan was my main hero for the super helicase story we have a really long standing collaboration with Jan Schoenler's lab metcom stock is the one who built this fancy instrument and measured EBRD commercial changes during unwinding we've been working with Tim Norman's lab for many many years and Chris Thomas was a collaborator on the web D protein thank you very much yeah so on the way back the speed also depends on the 80 concentration so it's powered by ATP not just sleeping so that's a very good question you so published a really beautiful structure this year with the same same screw on the racu helicase in the closed form and there's a very striking change between closed and open forms right so so I think it'll be really interesting to also cross link the protein in the closed form and see if it becomes a super helicase right and my prediction is that it can be done but it'll take a lot of effort because you have to use a different approach racu has a lot of assistance like 30 system so you need five register times to we have to use other other approaches but I think the mechanism like this would be probably I wouldn't call it general but it'll be found to apply in other classes not all but okay so that is is so I think that this is okay so that is a complicated question to answer so wrap why is it in the open form but if you have high high enough concentration that why does unwind DNA not not very well but so that this means that closed form per se is not essential you know essential for unwind thing but it's more that open form is needed for strength switching and we don't see any difference between wrap why and wrap in terms of you know long distance on minding yes so yeah yeah so okay the first question was actually it doesn't slow down we may be reporting to store stores as we see at defined locations so when the enzyme keeps moving there's no slowing down continues to move in the same with the same speed now you you had you have sharp eyes to notice that when you had two proteins appearing to unwind the DNA there's no prior history of following this DNA by the second protein it just seems to appear on the same spot and so we have we have really limited number of molecules that we I think we have maybe less than 10 events like this because if you go to higher concentration to increase this event frequency then then back from that becomes an issue so we don't have a lot of data but we do have some examples where we see another protein following from approaching it by so we see both types