 Okay, so I'm gonna talk about something a lot more focused, but also equally is unknown and so I'm a PhD student in University of Cambridge But I'm gonna talk about the work that I did in Kyoto with Fuyuki Shikawa for the last two years and it was about DNA PK and P's DNA damage induced transcription of stress response genes So even though it's focused, I think that it's dealing with the three pretty interesting concepts So DNA damage repair DNA damage induced transcription and the low-dose stress response and seeing how those three different things are interacting with one another Okay, so I'm gonna talk about a bunch of stuff in the intro I'm not gonna have a lot of time to go over the actual data But I have it all in my poster so we can talk about it if anyone is interested So anyway, so I'm gonna talk about all these things, but I'm gonna start with stress and acquire tolerance So how can we study stress so we can apply stresses to cells or organisms and see what happens So how can we study on a cellular level stress? So we can take ourselves and we can apply a lethal stress and the cells will die And then we have an obvious phenotype so we know what kills the cells and what doesn't But let's say we wanted to study a low-dose stress Which is what we experience more in our daily lives like giving a talk or something and so then you have like cells and You have a low-dose stress and the cells appear to have no obvious phenotype So you can obviously look at all kinds of stuff about what's going in going on inside the cell But let's say you then apply a lethal stress So most of the cells do die, but some of the cells Actually survived and so this is what's called acquired tolerance at least in the lab that I was working in some people refer to it as Hormesis But that has some bad connotations. I think so anyway So how do cells respond to stress? molecularly One way is through cellular response proteins such as the HSP's or heat shock proteins which people talked about earlier in the meeting and during these during cellular stress response these proteins basically transport help facilitate damage-integrated or proteins to the outside of the cell So two examples of HSP I'm gonna be talking I'm gonna be talking about heat reaction Yeah, it is different but a lot of times the stress So I'm gonna be talking particularly about HSP a1a and b which are stress-inducible So not all the HSP's are stress-inducible these two in particular stress-inducible by different kinds of stresses, but I'll talk about heat Okay, so how do cells acquire tolerance to stress molecularly? So one way maybe is through chromatin remodeling for example of the stress response genes So next what's chromatin's chromatin's the way the DNA exists inside of this nucleus And I'm gonna zoom in on a nucleosome and so a nucleosome is the DNA wrapped around the histone octamer And this is a super simplified histone octamer, which is composed of eight histones for two and Four different kinds, but actually just lied to you. They're also histone variants So I'm going to be talking about H33 in particular and H2AZ and these variants can be associated with different chromatin states Such as you chromatin or active or open chromatin or heterochromatin, which is condensed and inactive And then also of course the histone can have histone modification on the tail, which are also associated with different chromatin states All right, so here and its role in acquire tolerance. So what's here? Here as the histone chaperone responsible. It's a histone chaperone complex responsible for Putting H33 into the nucleus Nucleosome and that's not at all how it does it, but anyway so basically The current model of load those stress response in mammalian cells was was put together by Fuyuki Shikawa's lab in this 2012 paper and buddy Fission yeast and then more recently in wi-38 8-H turret and mortalized human lung fire blasts So the current model is that you have a low-dose stress which leads to which induces here complex to localize it a stress response gene such as Hsp1a or B and that induces its expression and brings along RNA polymerase 2 and That induces acquired tolerance. So the question that I originally wanted to ask is what is histone H33 doing? So what's histone H33 so it has a 4 to 5 amino acid difference from the canonical histone H31 and 2 which doesn't seem like a lot But actually it can lead to a lot of different things in terms of its function And the one important thing to think about in terms of its function for me was actually its deposition So the canonical histone H31 or 2 is deposited only during replication by calf 1 So that's again not how it works But anyway, and then histone H33 can be deposited anytime by the here complex So that's important because the stress can happen anytime, right? So that seems like a good model for how you want to maybe remodel the chromatin Okay, so next I'm going to talk about DNA damage and transcription which may not seem like an obvious leeway into that but it I kind of ended up into that going into that direction in terms of my research so Normally we think of DNA damage as deleterious to the genome, but it turns out sometimes it's actually necessary for transcription So one way one model by which it has been shown to be necessary for transcription regulation is in this paper from Nature Communications in 2015 bunch at all and basically they made a model for program DNA damage at the HSPA1A slash B locus So the model goes like this So you have your stress response gene here and RNA pull twos pause downstream of the transcription start site of the gene And then you have a stress or an activator which induces a lot of stuff to happen So you have top-wise summaries to coming here and making a double-stranded break and then you have DNA repair proteins like DNA PK CS and coup which I'll talk about later that form a complex called DNA PK and ATM and then also you get Classic DNA damage markers like gammage 2x showing up and then RNA polymerase 2 is going and transcribing the gene So this has its own controversy and I'm not going to go into that But anyway, so the original purpose of the study was to figure out what h3 3 was doing in this model of Load those stress response, but what ended up happening was that I sort of was Investigating this and so I'll try to connect those two stories So the model that I'm using again is wi-38 h-turn immortalized human lung fervor blasts I also use a couple of the other different cell lines just to make sure that's not going on in just these cells So first step that I did was to identify the interactors of histone H2 3 and I did that by IP mass spectrometry And I'm not going to go over through the data But I'm just going to give you the four top hits that I got and then verified using a immunoprecipitation immunoblotting So first the first proteins are top-wise summaries is one and two So top-wise summaries one makes a single-stranded break at the DNA and top of top of two Makes a double-stranded break and they're also involved in the relegation. Maybe no one really knows and then there's the coup complex Which is involved with sensing DNA damage at an early stage people think and it forms another complex with DNA PKCS called DNA PK and Lastly PARP1 or polyadp ribose polymerase 1 which does a lot of stuff that no one really knows what it does But people think it might have to do with sensing DNA damage Okay, so now I have this kind of image. So h3 3 is binding with all these proteins, but what are they doing? So I looked I wanted to know the role of them and h3 3 in stress response So to do that I asked the question are the interactors deposited at a heat shock locus following a low dose heat stress and to do that I did an experiment called chromatin immunoprecipitation or chip And basically you look for enrichment of a certain protein at a location of your choice So in this case I did QPCR to do that and the locations that I that I was interested in where H is p1a Which again is the stress-inducible gene and gap DH is just a control So these are the different loci looked at so two kilobases before the transcription start site the transcription start site itself and the open Reading frame or gene body of H is p1a and then I just looked at some control regions of gap DH Okay, so now looking at the actual enrichment of protein. So looking at total histone H3 At the H is p1a locus. I found that there is a significant decrease following a low-dose heat stress So this low-dose heat stress is I basically heated the cells to 40 degrees C for two hours So human cells like to be grown at 37 so you don't see any actual Phenotypical change in the cells, but you do see this decrease in total histone H3 at the TSS Next I looked at histone H3 3 and I found that there's a significant increase of H3 3 at the TSS and the orph of H is p1a Following the low-dose heat stress. Okay, so that's interesting. So then I went on to look at the interactors So I found again that PARP is Increased at the orph of H is p1a following a stress and the top two is a Increased at the orph of H is p1a following a low-dose stress So I have this sort of image now after a stress H3 3 top two and PARP are all kind of there and I wanted to ask now about DNA damage So we have top 2a which induces DNA damage probably and PARP which probably repairs it So to look at DNA damage, I looked at gamma H2a x which going back to this original program DNA damage marker I mean model you see that here's gamma H2a x the marker of DNA damage and I found that gamma H2a x is increased significantly at the TSS and the orph of H is p1a following a low-dose heat stress All right So I did some other experiments which are on my poster But I'm not I don't have time to talk about and basically I tried to show that the DNA damage is what's responsible for the transcriptional upregulation of H is P1a and so I found that top 2a Makes a cut and that and maybe results in this DNA damage marker coming coming to this low side Following the low-dose stress. Sorry again Okay, so I think this is probably the last thing I can talk about but the DNA damage So then I looked at the DNA damage response itself So I looked at the DNA damage then I wanted to look at the response and so this is looking at this part of the model So they this group said the DNA pk is responsible for is important for the upregulation of H is p1a and I didn't think that they had really adequate Data saying that so I looked at that so to do that I did a knockdown of DNA pk and I did this RT QPCR experiment where I plated my cells Infected them with a sRNA and then I performed the sub lethal stress and extracted the RNA and then ran on a QPCR So this is just showing that the knockdown was pretty efficient. So I got Knockdown of DNA pkcs No, in particular, which is the catalytic subunit of DNA pk complex And then I found that if you if you knock down DNA pk you get a significant increase in the H is p1a expression following the low-dose stress Okay, so then I did an even a weird weird weird experiment which Actually, no, so then I started using the DNA pk inhibitors because DNA pk knockdown is kind of annoying to do So I just found the inhibitor that would do show the same thing So I did the same experiment but with an inhibitor this time And I found that the inhibitor also increases H is p1a expression following a low-dose heat stress And I found that that's true also in other kind of cells besides wi-38. So I am our 90 and HD 1080 Okay, so then I did a weird experiment an acquired tolerance experiment That's what I called them. So basically I plate my cells I treat them with an inhibitor at a normal temperature Then I treat them in with an inhibitor without an inhibitor at a sub lethal stress And then I give them a lethal stress and then I wait two days and I measure their viability So this is the acquired tolerance experiments on the left This is the control or dms. So on the right if you can see the images of each of these but basically let's just keep to the The graph so the dms. So is the control and you can see that so no stress. There's I normalize it So it's just a hundred and then you look at just a sub lethal stress It's pretty variable But then you look at a lethal stress and you get a depletion in the viability and but then if you look at the Sublethal plus the lethal or the acquired tolerance you get an increase in viability So then if you add the DNA PK inhibitor into that mix you lose that lethal stress You basically lose acquired tolerance, but that's because you're getting a gain in lethal stress tolerance Okay, so then I looked at the effective DNA PK inhibition on different things at the HSP when a locus using chip So I looked at top 2a for example and following a sublethal stress and the DNA PK inhibition I found that there's an increase of top 2a at the orb of HSP when a and I found the same thing for game h2x Even though it looks a little bit less convincing All right, so I found it basically I found the opposite from what the model was showing the DNA PK is negatively regulating the stress response gene HSP when a and I don't have time to talk about I think DNA damage in histone variant h3 3 but essentially I did an h3 3 knockdown and I found I Did similar experiments and I found the h3 3 and DNA PK are counter-regulating the stress response genes HSP 1a and B specifically Okay, so just now looking at this model I just want to show what I did summarize it So I kind of showed that h game h2x is at the HSP when a locus and so is top 2 and that induces this DNA damage Following a low-dose heat stress, but then looking at DNA PK. It looks a little bit different. So based off of My data as well as other people in the field. I made this model so essentially and so you have an inactive stress response gene and You have pull to pause downstream of the transcription start site Then you have a stress and that induces top 2 to localize and make a cut and That now you have two options So you have what I call the slow and safe transcription or the fast interest Fast and risky transcription. So it's slow and safe is DNA PK dependent And this is what is basically the previous model. So I'll go through that first So you have the cut and then you have DNA repair coming here and repairing the DNA and then you have transcription factors and h2 3 and here are localizing and then you have transcription of the gene so you can respond to the stress Okay, then the other side the fast and risky transcription is DNA PK independent. So let's just say in a competition model You have like DNA PK and the transcription factors sort of vying for binding at this site So let's say that the transcription factors are here at or h2 3 get there first and then you have actually Transcription with the damage still there and so this is probably the most controversial part of the model But it there are things that suggest that that might be the case And then maybe we have no idea but maybe then the DNA is repaired afterwards because that would make sense Okay, so and then the question is is actually acquired tolerance not a learned Ability to the lethal stress via the load distress But actually maybe is it just cells that happen to be going through this pathway rather than this one So they can respond to the stress faster since the gene is being upright regulated faster Okay, so that's it So I just want to thank Fuyuki Shikawa and maxed for funding me and Kyoto and then also to my current advisor and Ferguson Smith at Cambridge who let me come here All right, so that's it Top five summaries inhibitors block the stress response transcription So yeah, okay So I've tried that and no I couldn't even get the transcription the top why summaries inhibitors to work at all So none of the commercially available ones seem to actually be blocking all top why summaries activities specific Definitely not top 2a. So can you actually show that each stress inducible gene? Has a cut no made by top two. No, so it's just a model just a model So you don't really know if that if there's damage to the DNA. I do know that there's damage to the DNA at specific stress response genes Not at necessarily all of them So I know that there's definitely DNA damage happening at HSPA 1a and B for sure Based on what okay DNA is one have a sensitivity food printing a taxi Okay, so Okay, besides you mean besides the DNA damage marker. Yeah. Yeah, how do you detect the DNA damage? The DNA damage marker game h2a x, but that's not the only way so yeah, but the gun h2x is actually Not a specific mark of DNA damage. You're right. Yeah, that's true. Okay. It's also when you have increased proliferation Yeah, but maybe that is deep. Maybe that is DNA damage labor's replication bubbles and yeah, but okay So I think that that whole con that whole this whole idea maybe suggests that actually DNA damage is much more Present than people think and that it's actually going on all the time But you're right so game h2x may not be an actual marker of DNA damage But so what I can do is so, you know, do you know what a top side is? So it tops a top a side so it's not actually an inhibitor of top away summaries, too So this is just different so a top side inhibits the relegation mediated by top away summaries to right So I can show that if I add a top side here If I add a top side here, I can get an upregulation of the HSP on a gene So that kind of suggests and then there's also an increase in acquired tolerance and an increase in the game h2x at that low side But yeah, I mean Ideally, maybe we could do something else to show it But if you use any other kind of stress and you look at other stressing just about teams You'll see the same phenomenon. So this model I didn't I wasn't I didn't go into it Basically, this model actually has nothing to do with stress So they initially showed it in HSP 1a, but then they extrapolated and use it for actually serum inducible genes So this is not even stress. So they're they're they're using this model I mean, they did talk about in terms of heat stress initially, but then they use it for serum inducible So if you look at the seafast gene, you'll see the same thing. I'm not yeah panel discussion because we already waiting for our next Speaker but maybe we can take one more but not discussion a short question If we have one, okay, if not, then we should think