Hello everyone. Welcome to day three of our virtual event Illuminating New Frontiers Cracking the undruggable code. During day one, we discussed a notorious formerly undruggable target, Ras, along with targeting protein protein interactions. And on Day 2, we tackled induced proximity and targeted protein degradation modalities that are showing promise and modulating previously intractable targets. All these presentations are available on demand if you are interested in learning more. And today we're joined by a panel of experts to discuss another modality showing promise and drugging the undruggable, and that's targeting RNA. Before we get into the agenda for today, like to go over a couple housekeeping things. So you have a screen with multiple windows. So all of these windows are movable and resizable. So feel free to customize it for your experience. I'd like to draw your attention to the Ask a Question window. So at any time during the event, please go ahead and submit questions for our presenters. We'll either answer them in the chat or after their presentation, or we'll save them for the live panel discussion at the end of the event. There's also a resource library available with a bunch of helpful materials. You can also share the webinar with anyone else that you think might be interested, and we will send a survey at the end of the presentation. So we'd love your feedback and what you thought of the event in the format and that will help us in future events. And then throughout the event, we're going to have some polling questions sprinkled in. So these are kind of fun and we appreciate your engagement in those. So here is our agenda for today. Our list of stellar speakers that we have for targeting RNA. We'll have each speaker give about 30 minute presentation and then again at the end we'll have a live Q&A session with the speakers. And to kick it off today we have our first polling question. So go ahead and select all that apply. And we know in targeting RNA there's a number of different strategies that are being used. So we're we're curious what you're currently using or what you're interested in learning more about. So we have small interfering RNA antisense oligos, small molecule RNA binders, RNA editing technologies like ADAR and CRISPR cast systems, RNA aptomers, micro RNA modulation and long non coding RNA targeting or other. We know there's lots of things happening, so please go ahead and select any that apply and that you're interested in learning about. I'll give it just a moment here. All right, number answering. So let's take a look at what we've got. All right. We have pretty decent spread, so a lot of algo work with SIRNAS or ASOS as well as a good portion at small molecule RNA binders also looking at RNA editing aftermers really across the whole board. All right. Thank you all for participating in that. And now we'll go to our first speaker. So Professor Matt Disney is an Institute professor and chair of the Department of Chemistry at the Herbert Wertheim University of Florida Scripps Institute for Biomedical Innovation and Technology. His laboratory has pioneered the development of small molecules targeting RNA. The resulting data have informed the design of chemical probes to study problems of biomedical importance and to advance a gene to RNA to precision medicine. Paradigm Labs Research has garnered various awards including the ACS Nobel Laureate Signature Award for Graduate Education and Chemistry, along with Alicia Angebello, the NIH Director Pioneer Award, the Tetrahedron Young Investor Award, and the Eli Lilly Award in Biological Chemistry, among others. Laboratories work has also spurred small and large pharmaceutical companies to invest in small molecule targeting of RNA. And with that, welcome, Matt. Thanks, Sarah. Thanks for that very nice introduction. It's nice to talk to you all today. I'm I'm, I'm going to do my best to muddle to get through this. I'm not feeling great. That's an excuse to not doing as good of a. Job as I need to do all right. So I'm not. I need to backtrack from there anyway, so all right, so I'm going to talk about ways that the lab has taken an RNA sequence and designed various small molecules that modulate RNA function. What I chose in this talk was more to focus on some work that we've recently published and some soon to be published work on drugging undruggable proteins by targeting their messenger RNAs. OK, So I want to start by acknowledging my group, some that are currently in the lab, but some, all of them have made a pretty large impact on what we're doing. And I'm going to discuss various work of these group members during this talk. So Sybil Gaputi was a graduate student in the lab. He's now a scientist at Expansion Therapeutics and him and I wrote code that we call in Forna, which we're currently updated that updating that can help you drug RNA with small molecules. I'm going to describe some work by Peiwan Zhang and Tang Wei Wong on targeting Tau pre messenger RNA splicing and alpha synuclein mRNA. And then last is I'm going to describe a huge amount of work that was pioneered by Yukon Tong, Raphael Ben Hamou, Hafiz Hanif Haru Akawa on targeting CMIC messenger RNA. So I want to thank them. One of the things I want to sort of point out here is that size from India, Taiwan and Tang, we are from China. Raphael is from Israel. He's currently a faculty member at the University of Jerusalem in Israel. Yukwan or Hamish is is from China. And I think one of the things in science that's for me has been really fulfilling when I take a step back, is being able to interact with all these people that have various perspectives. And that's something I think that we'd like to think in America is not unique to scientists. But I think we're in a privileged position to be able to collect data sets on talking to these people, not based not just about science, but also pay one about how he grew up in China and what his experiences. And last but not least, I want to thank seriously, the US taxpayer, someone that currently resides in Florida probably isn't talking about paying taxes. Traditionally, they generally move here to not pay them. But I think it's important to note that the US taxpayer, especially now and we're in the heat of a presidential election, has really funded innovation that saved us countless times. It's put a man on the moon. It's we're in an era now where the disease was an RNA, SARS COV 2 and the cure was an RNA. And we would never have been able to have position so quickly to be able to make a vaccine that could save people, politics aside, without the work of the US taxpayer and our government officials making that a priority. And frankly, students like these working in labs 50 years ago, sequencing T7 RNA, plum rays and figuring how to make large RNA by in vitro transcription. OK, so this is an RNA talk. After all, you had your protein talks yesterday. So if you look at the the human genome, one of the most striking things is that when people have done analysis, about 2% of our genome gets made into protein and yet 90% gets made into RNA. What our lab and what I've been focused on for my whole career is thinking about how both coding and non coding RNAs fold and being able to determine folds by computation. And so when we think about this is that, and you guys have had a series of talks on this in the past few days. When you consider a protein druggable, it means that you have a modulator of a protein family. And it turns out only about 15% of protein families have a modulator leaving the rest on druggable. Yet when we consider not our RNA, generally about 0% of them have been drugged with a small molecule. The only one that's been drugged is wrist a plan which stable wrist a plan which targets SMNSM N2 messenger RNA and it's stabilizing RNA protein complex. And So what our lab central focus has been on trying to figure out how to increase the drug ability of both coding and non coding RNAs. In fact, there's been a lot of criticism about non coding RNAs not driving phenotypes. There's a genetic disease that was just published in New England Journal of Medicine that's due to a mutation. It's a nerve developmental disease. It's a mutation of a long non coding RNA called chaser. And as what always happens is the discovery from the human genome, oftentimes to identify sequence, the ability to understand the biology of these functions takes time, right? That's not just a data set, that's in depth, detailed mechanistic work. And as you'll hear today, you know, coding and non coding RNA plays roles in all, all diseases, cancer, heart disease, diabetes, stroke, dementia, ALS, you name it. I can tell you an RNA that's causing that disease. Oligonucleotide based modalities are powerful. Most of you wanted to hear a talk on sirnas or we're most interested about sirnas and antisense oligonucleotides. They are game changing medicines. They are very powerful tools and probes for functional genomics. However as medicines they have limited tissue penetrance and so one of the desires for having a small molecule approach that could modulate RNA in a similar manner that oligonucleotides can would be 1. You can get into tissues that oligonucleotides wouldn't traditionally be penetrant to, which could even be solid tumors. And 2nd is you can deliver these compounds to the brain by injecting them peripherally or ultimately these compounds could be oral, oral medications. And so the reason that we think we can pull off as, as the speakers today will show, RNA targets, target RNA with small molecules is because RNA plate has pervasive structures. Here's just some structures that it took 5 minutes to figure out to and I'll go from left to right and then I'll go to the bottom. So at the five Prem UTR of oftentimes messenger RNAs have highly robust structures that modulate translation. What I'm showing here is the iron responsive element, which it controls translation of the alpha synuclein protein from its messenger RNA by binding the iron regulatory protein, and you can have a triplication of that gene and that can cause genetically defined Parkinson's. You can also have RNA structures that regulate splicing at both Exxon and Tron junctions and within the exons themselves. And diseases that are manifest by either mutation or these splicing events can include cancers, Alzheimer's dimension, and Parkinson's. To the far right on this slide is an example of a triplet repeat expansion disorder. I've spent most of my career on targeting rare genetic diseases, part because my niece has a rare genetic disease. And this RNA structure is ACUG repeat expansion that causes the most common form of adult onset muscular dystrophy called myotonic dystrophy type 1. And that disease is caused because you form a toxic RNA protein interaction. And what you do is the RNA gains of function by sequestering the protein to then not allow that protein to do its normal function which is involved in RNA metabolism. And then in the bottom here is the most famous sub molecules. No, it's not alpha fold. No, it's not RNA alpha fold. It's micro RNAs. So the bottom of this slide is a micro RNA. So as as we all should be aware, micrornas I would say earned a Nobel Prize and were awarded it last week. But micrornas can play major roles in a wide variety of diseases, including inflammation, cancer. And if you and if you can silence a micro RNA, you have potential to treat a disease that's caused by hapulence efficiency. And so we've worked broadly on targeting these RNAs with small molecules. And I'm going to show you some examples. But as I mentioned earlier and I reiterate here, my focus is going to be on CMAC, OK. And so basically what we do is we can scan these messenger RNAs and then we want to stick a small molecule into them. So this is an example of if we look at that whole messenger RNA and we look at that pre messenger RNA struck the structure at the Exxon intron junction and we stick a small molecule into it that's shown on the right. That small molecule can actually change what proteins that pre messenger RNA encodes by affecting exclusion of an Exxon. And so that's one mode of action that we've been able to work in broadly for RNA targeted small molecules. OK. So I'm going to go through the platform that we've identified without boring you to tears and then I'm going to give you biological examples with my focus being on the MIC. So how does how does our lab look at drugging RNA? So my training is on RNA structure prediction, secondary structure prediction and chemical modification reagents. And so the thesis thesis with that was if we had a database or an encyclopedia or whatever you want to call it, information on RNA fold small molecule interactions, then what we could do is we could take that information and mine it against the RNAs that cause disease or that we're interested in modulating across the human genome to try to identify what are RNA folds in biological RNAs that we can target. Once we have the ability to do 1, collect the database of RNA fold ligand interactions and two, use computation against those interactions to find biological targets, then we can understand the ways in which we can perturb RNA function with small molecules. And I'll talk about some of that. There's ways you can do it. There's ways that are harder than others. And then last but not least, you want to study it in patient cells to animal models, which would be the last slide of my talk. All right, so here's the way that we do it. So we have about 10 to the 6th RNA small molecule pairs. And so they can be the colors here in this scheme are meant to represent RNA structure, RNA secondary structure manifolds, but they're certainly 3 dimensional structures and the colors within those letters are meant to represent small molecule ligands that bind to that RNA. So we have a selection experiment where we can identify in a massively parallel format these interactions. We can even use DNA encoded libraries to do it in a library versus library format and sequence directly on DNA expressing deeds. What are the RNAs that are bound and what the ligands that are bound? But I can spend the whole talk on this. You're not interested in that, you're interested in the function. So we can mine those interactions against say a disease coding up causing RNA. So for example, if I look at the CUG repeat that causes the most common form of adult onset muscodystrophy, myotonic dystrophy, we quickly identified this blue compound sits in the CUG repeat. And then what we can do is because that's a repeat, you have that monomeric compound. We could enable design to have the thing have the compound bind with higher affinity and specificity. Or early on in our career, my career, we linked these compounds together to bind to that RNA target in a programmable way that reads out not just the identity of the loops, but the distance between them. And a bunch of other groups have picked this up. And what that'll do is it'll free the protein that's bound to this target and that that improves, improves disease. And then we can take these compounds and throw it into cells and animals. All right, So what are some examples of this? So we want to apply this to to diseases with poor prognosis or no cure. Here are three examples that we've done and then I'm going to go to Mick. So by using I just showed you my atonic dystrophy type 1 and we've been able to target a wide variety of repeat expansion disorders including C9, ALS, Huntington disease, autism which is called fragile X syndrome. The only single known single gene cause of autism is caused by Narnia repeat. But we've been also able to target micro RNA. So here's an example of targeting a non coding RNA. So micro RNA's are made as precursors and the precursor has the precursor is shown here and it gets cleaved by these nucleases Dicer androsia that are meant to represent meant to be represented by the orange and the purple Pac-Man. And so we really identify this chiral ligand that's shown in the blue highlight that binds to the Drosia site or where the purple Pac-Man is to inhibit Biogenesis of the micro RNA. And it turns out and the papers are here if you want to read them. This is a nice paper by Rafael Benjamin in PNAS. And what we can do is, is we can find this microni target with about 100 nanomolar KD. We can inhibit the processing of this MIC RNA at about 500 nanomolar. And the molecule targets an invasive phenotype and triple negative breast cancer or breast cancers that don't have a precision medicine that can target them. Here's alpha synuclein structure, this iron responsive element. We've been able to target the iron responsive element and alpha synuclein here. That compound binds to that RNA structure that's near the start codon and it thermodynamically stabilizes the RNA. And what it does is it has the pre initiation complex of the ribosome sitting on the five prime end of this alpha synuclein messenger RNA and the compound binding thermodynamically stabilized the RNA. So the ribosome cannot as efficiently unwind that structure to initiate translation. And then also we can target RNA structures at Exxon and Tran junctions which is shown here for this RNA in the microtubule associated protein Tau. This was a large collaboration that we had with Pfizer. And what happens is this RNA structure can be mutated and thermodynamically destabilized with that mutation and it results in including too much of Exxon 10 in the spliced isoform of Tau. And what that does is it creates A4R or aggregation prone version of Tau that gets deposited in brains for people that have this genetically defined dimension parkinsonism or regular Alzheimer's. And so the small molecule binds of this RNA structure and thermodynamically stabilize it and limits the inclusion of Exxon 10 in the mature messenger RNA. So we can base, we can change the protein content of a messenger RNA by using a small molecule targeting an RNA. And we have a paper that's submitted where we've converted these lead molecules into oral blood brain penetrant molecules that affect how splicing. And you can do that by looking at the physical chemical properties of the compounds and we think integrating some unique aspects of design. So now what I'm going to talk about for the last part of my talk is using small molecules to subject RNA's to quality control. And so, you know, all that I showed you before was figuring out ways to modulate RNA by simple binders. But are there ways where you could bind an RNA and then summon natural enzymes that facilitate quality control to eliminate the RNA? And so there's three examples that we've done and these are some later examples of this work. One is there, it turns out there are many RNA mediated diseases where you have a toxic RNA is in an intron and these are mainly due to RNA repeat expansions. And one of them is C9 or ALS. So the most common form of ALS and what happens is the rest of the RNA gets properly spliced, but the intron in the intron that has the repeat gets bound to proteins and it gets harbored as a retained intron. And So what we've been able to do is show that pathway by gain and loss of function by for example, taking SIRNAS to eliminate or reduce that protein that binds to the RNA and that will facilitate splicing of that toxic intron, which then gets degraded. Or we can transfect plasmids into cells that have that protein that bind the repeat, which will increase the amount of that protein that the amount of the retained intron. And So what we did, this is work by Ali Angevello that was followed up by Jessica Bush for a peripheral injected molecule that crosses the blood brain barrier is that you can have a small molecule that binds us, intron kicks off the protein, that intron will be subjected to normal splicing, it'll be degraded by the nuclear RNA exosome. So you can degrade RNAs by using small molecules that have very drug like physical chemical properties. We've also done a lot of work on hetero bi functional compounds. I'm giving you 2 examples here, but we've also done cross linking molecules, molecules that can change RNA sequence, etcetera. But here are two examples where we've been able to summon RNA quality control on RNAs. And all these have been have been demonstrated from cells to animals. So for example, we can do direct cleavage of RNA. So in this we take natural products that damage DNA and we disable them from binding DNA. And so when you throw them in cells, they can't bind and they don't damage DNA. But when you pin them onto an RNA binder, they'll damage RNA and subject the RNA to quality control, facilitating targeted degradation of the RNA. This work by Ali was done in Type 1 myoponic dystrophy and she did an unbiased RNA seek assessment and you can improve 98% of disease associated pathways and eliminate in an allele specific manner the toxic RNA. The other work that we've done is induced proximity of a new ribonuclease. This is the first example of this is 2018, 2017, although we have earlier work on degraders before this 2010, 2012. I think in cells where what happens is you have these hetero by functional compounds or one end binds an RNA target and the second end binds a ribonuclease to facilitate an unnatural association of the quality control protein to the quality control ribonuclease to eliminate the RNA. OK, so I'm going to talk about targeted small molecule degradation of semic RNA. This is this is a riff off. I mean, some of this data is is unpublished but soon to be published and some is from a paper that we had in 2000 23 But this is basically work that we've done in collaboration with Nikhil Moonchy's lab at Dana Farber Institute working with Domenico Eugenio Mamet Hamish and Tangli Hamish goes by U Kwan. And so as many of you are aware, CMIC is considered to be undruggable because it looks like a a tennis ball basically. So the classic way that undruggability. So you just got you just talked to black belts about on drug ability in the last two days. I'm I'm a, I'm a yellow belt or white belt on that. So I shouldn't even be talking. So basically C MIC protein, it's a transcription factor. Transcriptions are factors are not druggable because they they don't have pockets where you can sit small molecules in. And so the thesis in this paper was and, and we've done other work on this is what if we look at the messenger RNA that encodes for a non druggable protein? Do they have pockets that a small molecule combined? And so for MIC, it turns out it has an internal ribosomal entry site at its five prime end. And we're able to identify ligands that bind to that RNA and it and make them heteroby functional, recruit an effector protein to cleave the messenger RNA and thereby reduce CMIC protein levels. Yeah. So here's the actual structure that we go about targeting in CMIC. So it's this RNA structure and the Ribotac that we use. I'm, I'm, I'm being very not good here in the depth because I don't have a lot of time. But basically we can recruit ribonuclease L with some of the molecules that we produced and RNA cell is present as an inactive monomeric RNAs that we recruit and then dimerize and make it active to degrade a target. And so we can identify in CMIC a small molecule that binds to this site and then it degrades the whole message. And many high priority undruggable targets have highly structured five premier TRS in their mrnas to affect translation. So here's data with this compound. So the binder and this is a theme, the binder has no effect by itself. It's only efficacious effective when we make it into a header by functional degrader. So the binder has no effect up to 10 micromolar where's the degrader can reduce about 60% of the RNA target at the same dose. So a lot of RNA binders are inactive as binders. You need to have effector functions be brought in for them to be bio active. We can talk about that this molecule reduces proliferation and apoptosis in cells. And we've done a full RNA seek assessment. The RNA seek is to the left and the full proteomics is to the right, which the RNA seek is. CMIC and EGR one are the two most affected RNAs, which CMIC obviously, but EGR 1 is a Cmic's, a transcription factor. So EGR, one message is made by MIC, and then the proteins that are affected, it turns out the proteins that SEMIC facilitates transcribing its mRNA are diminished. Whereas if we compare it to HIF 1A, which is also transcription factor that has one base pair change, there's no effect. All right, so here's here's where I'm going to go for the next 3 minutes. And then I'm going to give you guys some peace from having to hear me talk. So we generated resistance mutants for this with Nikhil Munchi and we can get a tenfold decrease in the activity of these compounds. When we did that, not surprisingly, the first thing that we found is that we increase the expression of multi drug resistant protein like in a crazy way, right? So we could increase pumping, even increase breast cancer resistant proteins for cells that actually expressed them. The most interesting thing that we found is we lost 30% of the level of RNA cell. And we found for the RNA cell that remained, there were mutations within it, about 17 mutations into that RNA cell site, including four mutations at the catalytic site. The other 13 mutations we think might be functional, but they also have synonymous codons. So we don't know if RNA cell regulates itself with its RNA synonymous codons, but we're we're looking into that. This might be important and it might be not. We never ever found mutations in semic messenger RNA, so why could it not? Matt, you didn't look hard enough. I got you. Why could it be important, Matt? Holy RNA is made of 4/4 bases. Proteins are made of 20 amino acids. Maybe it's much harder for an RNA to change a structure to disable the binding of the MIT compound, but still function as an IRS. Could be, we don't know. We'll tell you later. Then what we also did is we studied these compounds against the panel of patient multiple myeloma cells and it turns out all of the multiple myeloma cells that expressed both MEC and RSL, we were able to to diminish the viability of those cells. So it was very clear that it had to have RNA cell and C MEC. And what we also did is we took patients, patient samples from three patients. So here's patient one, patient two, patient three. This is like I'm getting like nervous even talking to you about this. I think it's hard to be a clinician and be dispassionate about this. So here's, here's 3 patients. So patient one, you can see the Western blot from this is the left flannel. They have C MEC, they all have C MEC, but only patient one has RNA cell. And when we deliver the drug, we can get 80% knockdown of C MEC protein and RNA in those cells. And then to the right, when we deliver that to mice that have the human tissue implanted, we can reduce C MEC, which is the Western blot to the far right and we can decrease the tumor burden. All right, Sarah, how much time do I have left? I don't want to over talk. I have like 2 more minutes. Yeah, you do go for. It all right, so I want, I, I want to end my talks now by saying we have a lot of work to do, but I want some there are things to think about and study going forward. So what are the best RNA targets for small molecules? Are they the structures that we talked about? I don't know. But we have unbiased covalent approaches to be able to study that that we published 2 papers on. And there's more on it where we basically take fragments like the Encyclopedia Cravatica is what I refer to it in the protein space. And we react ligands with their RNA targets and pull them down. The other question we have is how is how can one best lead optimize compounds or potency and selectivity? It's established maybe for protein, but it's not established for RNA. And RNA has a lot of electrostatic potential in binding pockets. So there's a lot of dynamics and a lot of electrostatic potential that we don't know how to deal with. What are the best ways to affect the target? Some might be binders, some might be degraders. What's the best effector protein to degrade a target? We have work to figure that out. And what are the best diseases to put to clinic? I think the ones where you need a cure and then already I think doing science is it's pretty fun. It's pretty amazing to come into work every day and learn something new that no one else has done before. So I'm very happy with the choice of science, despite sometimes I can grouse like I did in the closed door panel before we open the live session here. Sorry, Amanda. Sorry, Sarah. Sorry, sorry Sherry. So I think undruggable proteins can be inhibited by targeting the mrnas that encode them. And we should, we should start to be considering RNA as a way to expand the view of drug ability. And I want to thank the people that did the work. I mentioned them during the course of this and I'm happy to answer any questions knowing that I probably went over in time. Thank you. Thank you, Matt. I appreciate your presentation. Great example of how targeting RNA can go after undruggable targets like like CMIC. So I think we will hold any questions for the panel discussion and move on to our next speaker. Thank you, Matt. Thank you. Before we do that, another poll question, so this one, what do you believe is the most significant challenge in RNA targeted drug discovery today? So go ahead and select the one you believe we have structural complexity and dynamics of RNA molecules, delivery mechanisms for RNA therapeutics, off target effects and specificity, limited availability of robust screening assays or regulatory and approval hurdles. So I'll give it just a moment here. You guys can please select your answer. All right, see those responses coming in. Thank you. All right, go ahead and take a look at the results. Structural complexity and dynamics of RNA molecule seems to be a clear winner along with limited availability of robust screening assays and delivery. OK, our next presenter, Professor Amanda Gardner, is a professor in the Department of Medicinal Chemistry and director of the Interdepartmental Program in Medicinal Chemistry at the University of Michigan. She received her PhD in chemistry from the University of Pittsburgh and completed her postdoc studies at the Scripps Research Institute. Her research integrates chemical biology, medicinal chemistry, and molecular and cellular biology approaches for early stage drug discovery efforts with a primary focus on validating new therapeutic targets in RNA biology. She was awarded the David West Robertson Award for from the ACS Division of Medicinal Chemistry and the Ono Pharma Breakthrough Science Initiative award, among others. Welcome, Amanda. Awesome. Thank you, Sarah so much for the kind introduction and thank you also for hosting this event. I think this is a great way to, you know, increase knowledge and also, you know, I guess support the fields, right, support our on drug fields since a lot of times, you know, people can shy away from these things due to the challenge. And so I also appreciate the poll question. So limited availability of robust assays because I'm going to talk a lot about enabling technologies specifically for another area of RNA biology that Matt, you know, didn't talk about, right. And so we talk about RNAs as individual molecules as if they're just on their own floating around in the cell carrying out function. And certainly in some cases this might be true, but one of the disclaimers I like to make about cellular RNA's is that they're never naked in the cell. And so the cell has both specific and nonspecific binding proteins called RNA binding proteins that bind nearly every cellular RNA. So shown on this side is the example of messenger RNA's, but certainly this applies to non coding RNA's as well. And these NA binding proteins regulate nearly all cellular RNA's throughout their entire life cycle from birth, a transcription all the way to death or decay. And so these RNA binding proteins can both positively and negatively affect these processes, transcription, alternative splicing, translation and decay. And what's really cool about RNA binding proteins is the fact that the cell makes a lot of them. And so there's been over 2500 RNA binding proteins identified to date and consider about we have 20 or so 1000 proteins in the proteome. This is a sizable fraction that goes into regulating cellular RNA is again showing the significance of RNA biology and controlling RNA biology in the cell. And these RNA protein interaction networks do play many critical roles. So RNA binding proteins can affect many aspects of an RNA and so in this case would be quite general to both coding and non coding RNAs. So the processing which we'll talk about modification was not mentioned, but that's another, you know, kind of unknown aspect about RNAs, their stability translation in the case of messenger RNAs as well as their localization. RNAs in turn can also affect the protein biology affecting the RVP's function, interaction, stability and localization, especially for localization in the context of the nucleus. And there's so many sub compartments in the nucleus and this networking interaction is very, very important in in affecting the function of the nucleus. And so in addition to the nucleus, just general disruption of RNA protein interaction networks has been linked to many diseases, cancers, neurodegeneration, a lot of the diseases that Matt has talked about. And so this became, you know, to us a very exciting area of RNA biology and really thinking about how we could use chemical biology approaches and medicinal chemistry approaches to really shed light on RNA protein interaction networks for drug discovery. But also, you know, I think kind of Matt alluded to, you know, there's still a lot of unknowns about the biology of RNAs in general. And so, you know, approaches that would enable to probe further RNA biology and learn more about human goings on, right, would be very, very important as we try to, you know, translate and learn more about what was revealed from the human genome sequencing. And so, you know, our lab over the years has always taken a pretty broad approach in terms of how we strategize as to how we might best study and manipulate the activity of these RNA protein interactions. And so we've spent a lot of time in rational drug design, specifically looking at a translation initiation factor EI 4 E, which is important in cap dependent translation. We've complemented that with structural biology as well as doing some prior OEM work as well. But what I'm going to talk to you about today will be, you know, more of the focus on these enabling technologies, both in the realms of high throughput screening as well as chemical biology approaches that it will allow us to both validate and screen RNA protein interactions. And so how I've set this up is really kind of helping us evolve our thinking, right? So the more experiments we do, of course, we recalibrate our, our, our learnings and think about, you know, how we can always push forward in the future. And so I've tried to to organize this like that. And so as our model system, our lab when we started was very interested in mechanisms of translation of control. And this is what drew us to micro RNAs, which of course is Madelow 2. Now, you know, very well deserved Nobel Prize. But these are small non coding RNAs, about 20 or so nucleotides in length that function in gene silencing. And so micro RNAs can do a lot of things. So they can induce target cleavage, deodentilation and decapping as well as inhibit the translation of of target transcripts. And so as Matt also already alluded to, the processing of micro RNAs is complex. So micro RNAs are transcribed by RNA polymerase 2 from our genome into a longer PRY micro RNA that then goes undergoes a first processing step by the RNA 3 enzyme Drotia in the nucleus to form a 60 to 80 nucleotide hairpin precursor micro RNA or pre micro RNA, which then gets processed by a second RNA 3 enzyme in the cytoplasm called Dicer, which then forms the mature micro RNA duplex, which goes loaded into this RNA induced silencing complex to carry out the functional relevance of that micro RNA. And so in addition to these processing enzymes, what became really exciting to us in this idea of thinking about how to develop approaches for looking at RNA protein interactions was the finding through both in depth biological studies as well as proteomic analysis that many specific and nonspecific micro NA binding proteins. Can regulate Micron A Biogenesis and function. And fortunately for us, one of the best characterized example of a Micron A binding protein is Lynn 28, which of course also has relevance in human cancers. And so Lin 28 is an RNA binding protein that regulates the maturation of AI. Guess one of the first micro RNAs to be discovered and described, the let 7 family of micro RNAs and cancer. These micro RNAs are known to exist as tumor suppressors. So let 7 targets include Mick Ras, HMG, as and cyclin D transcripts. And so loss of let 7 through the binding by Lin 28 can drive a cancer phenotype. And so Lynn 28 is a recruiter. And so I should say this because I think Matt kind of talked about this. So RNA binding proteins can recruit degraders. And so Lynn 28 recruits terminal urinal transferases which are not shown on this slide that ultimately results in the degradation and loss of this left 7. And so for us, we thought, you know, this is very well characterized example of an RNA protein interaction. There's structures, there's a lot of biology. Could we use this as a model to develop assays and figure out, you know how do we discover small molecules that would be able to inhibit this Let 7 Lin 28 RNA protein interaction And with this more largely, you know, for micro RNAs that are lost in cancer, which is is more common than those that are up regulated. Would it provide a strategy that would allow us to reinstate the production of those left those microtase like let 7 to create a new mechanism of of tumor suppression. And so how we first tackled this problem was a biochemical assay. And so this is an essay technology that I've talked a lot about South our catalytic enzyme link click chemistry assay where we can have an RNA binding protein in this case Lynn 28 that we can immobilize on a strapped Ave. encoded well plate using Halo tag. So I'll drop some Probega technologies we can which we can biotinulate for the immobilization. We can incubate that with a pre let 7 substrate that we chemically synthesize and modify to bare a click chemistry handle, in this case transcyclo octane. We can then detect the RNA protein interaction which will be on the surface using horseradish peroxidase which we also chemically modified to bear the corresponding click chemistry handle and TET which then undergoes inverse electron demand deals all the reaction to covalently link that RNA protein complex with horseradish peroxidase which we can detect like we detect any other HRP reaction which is commonly used in like Western blot. And so with this assay technology, we ran a pretty large screen and found a HIT compound, in this case this bis sulfonamine compound that I'm showing here on the slide. And so in Catalka, this had an IC50 of about 8 micromolar and then subsequent SVR experiments, we were able to show that this bound to the protein and had no RNA binding affinity whatsoever. But I also like to include this example. So one is the technology, but 2 is, is the the challenge that we had. And so this molecule ended up being a zinc binder because what I didn't tell you about Lin 28 is that it contains zinc knuckle domains. And so we did lots of medicinal chemistry and one could say waste our time, but we spent a lot of time trying to optimize this compound really to no avail. And so this really caused us to pause and also, you know, think about our asset technology. Was this the right approach, right? What are limitations of doing biochemical assays for RNA protein interactions? And so, you know, we do have, there are still a lot of unknowns about RNA biology and so we should always premise with that. But even in the the realm of RNA protein interactions, you know, inside of a cell, occupancy of that protein is not going to be 100%. But of course in our biochemical test tube, we add, you know, 1 to one or excess of one of these components and that's really challenging. And that, you know, we might be trying to compete for binding, which, you know, is not really happening to the extent that it is in our test tube, which will, you know, and of course there's lots of other challenges and in terms of structure. And so RNA structure has been known to have unique aspects. I'd say the structural ensemble can change both between in cells and in our test tube. The RNA may be post transcriptionally modified. We don't know that RNA binding proteins also contain high degrees of disorder which is known that can affect the overall structural complexity of that RNA binding protein. They can be post translationally modified. And these of course, are all things that we cannot really recapitulate in our biochemical assay. But also even for lint 28 and other RNA binding proteins as well, these often, you know, exist in larger protein complexes. And so you know, how we could model that in a biochemical system when sometimes we don't know the other protein components really became a challenge to us. And so we kind of thought, you know, why try to recapitulate, you know, biology that may not be translationally relevant when we could just go ahead and said develop a cellular assay. And so that's what we did. And so this is the work of Dan Lorenz and Sydney Rosenblue of what they were graduate students in my lab. And so they really took this this challenge to heart and created a new assay system that we call RNA interaction with protein mediated complementation assay or Ripka that also I will always give a plug makes a lot of use of the highly relevant of Promega tools that they have developed over the years. And so how this assay works is that we have flippin hack 293 cells that are stably expressing small bit Halo tag. And so this is assay is going to be based on complementation of large bit and small bit for the split nano luciferase technology. We can translate Co transpect these cells with an RNA binding protein that will be fused to large bit, which is the other part of the nano luciferase and an RNA probe that bears a chloroalkane tag. And so here's it's just kind of the chemistry that we do to do to modify these RNA. So simple NHS chemistry works pretty well. And so how we think this assay works is that since we're a small bit is already expressed in the cell, the first thing that we expect to happen and we do have some experimental evidence for this as well, is that the RNA will become labeled with Halo tag, a small bit conjugate. Then when the RNA binding protein large bit is is actually expressed and the protein is made, the RNA binding protein can bind to the RNA large bit and small bit can reassemble to form the functional nano luciferase enzyme that we can detect. And so I won't talk too much about our our early studies since this has been published for quite some time. But using let 7 and 28, we were able to show that we could actually this technology was quite successful. We could differentiate between binding and non binding RNA substrates as well as look at individual domains and make this a nucleus exclusive assay. And so of course RNA binding proteins don't just play in the cytoplasm, which is how the first iteration of the assay was set up. They play in all the different compartments of the cell. And so a lot of them are nuclear and it was really important to us that we could do that. We also further went on and and tried to validate micro RNA protein interactions that were identified using proteomics technologies. And so we were able to do that and to develop a whole suite of different micro RNA protein interaction Ripkas. But of course, our goal in all of this was not to just detect RNA protein interactions, but really to enable probe and drug discovery in the space of targeting RNA protein interactions. And so Sydney and and later Dalia did a lot of really nice work and further optimizing Ripka, miniaturizing it, automating it with liquid handling. And this was the collaboration that we did with Vlad Simoth and George Yambasu at Merck. We were able to take our Ripka assay technology and screen library of molecules provided by Merck and this worked quite successfully. And so I'll share some of the details of, of our, our screen and, and what we found with RIPCA. And so our high throughput screening campaign kind of overview is shown here. And so from Merck, we got an RNA knowledge base set. And so there's a lot of like in forna, right? There's a lot of effort in creating these libraries of molecules that you know are predicted to bind RNA. So we screen that as well as we also added in virtual screening hits of molecules that we predicted to bind to Lynn 28. So that's about 18,000 compounds in total. And so we took this through Ripka and integrated with Cell Titer Glow because we expected some compounds to be toxic and not necessarily the signal loss was due to loss of that RNA protein interaction. We took this all the way through and in replicates and dose response and got finally 23 compounds after we integrated with our CAD Alka biochemical assay and did some SIM expansion on HIT molecules to get down to 23 that had PAC50 values about what we would expect for a starting hit. And today what I'll share with you is the data for one of those compounds, Sid 41526 O, which had an IC50 of about 7.6 micromolar and RIPKA. And so the structure of that molecule is shown here. We were able to test it against multiple Lynn 28 LET 7 Ripcas and it had IC 50s of about 10 micromolar against each of these systems. We were also able to show using a qPCR assay that we could see some restoration of mature Let 7 levels in line with what we expected at the beginning of this project that we will to restore Let 7 levels and tumor suppression. We were also able to show in a luciferase based assay that we could silence or that Let 7 levels were increased by silencing luciferase with Let 7 binding targets in the three prime untranslated region. Of course, a big, you know, aspect of this assay is, you know, and something that we were always, you know, quite curious about is, you know, what's the best way to target an RNA protein interaction? Is it binding to the RNA? Is it binding to the protein? And we got asked that a lot. And so we had to mechanistically characterize this molecule in terms of, you know, how it might be functioning in RIP guns. So we use SPR experiments as well as STD NMR experiments and collaboration, most seroquene that I'm not showing to show that in fact this molecule does bind to the RNA as expected in line with the use of this RNA bias library. And we saw no binding whatsoever to the RNA binding protein. And so this molecule structurally was quite interesting to us both based on its similarity to RISA plan, what you heard about, right, This is an RNA binding splicing modulator, but also the fact that it has a lot of similarity to other now as I would say putative known established depending on who you're talking to, RNA binders including kinase inhibitors. And so Matt's lab has already shown nice work and in showing that you could repurpose kinase inhibitors. This example of a poblocyclo that was from Gabe Veroni's lab and you can see the high degree of structural similarity between our Sid molecule and palocyclo. And if you're a good medicinal chemist, you will say your compound is a kinase inhibitor and you would say, I would say you are correct. And so Sid 41526 O was predicted as an RNA binding ligand, shown to be an RNA binding ligand, but was originally developed as AV1 kinase inhibitor. And so we did subject this molecule to kinome profiling and it exhibits an IC50 of 35 nanomolar for WE one. And so because we're doing a cell based assay, my Spidey sense went, you know what if this is all just a phosphorylation effect and it has nothing to do with RNA binding or direct disruption of this RNA protein interaction. And so it was really important to us to go in and make a control compound. And so we were able to make an inactivated version of this this compound. So CCG 386224 which we confirmed lacked kinase activity for WE 1 using Western blot. And fortunately, fortunately this molecule retained activity in our RIPKA assay. Actually it had some modestly improved IC50. We think this is also due to the fact that it improved in cell viability. So of course we expect that we want is a cell cycle kinase that there would be some toxicity with these molecules at high concentrations of Sid 41526 O and cell titer glow. Of course we see that versus our CCG 386224 kinase dead compound that that is non-toxic in these cells that we wouldn't expect to have their viability affected by. And so that's really exciting to us. And so, you know, to kind of quickly summarize this, you know, in our biochemical assay we found what I'll call a crap, even 28 compound with a lot of, you know, liabilities in terms of its activity and specificity and it's highly toxic molecule. And now using our cellular assay, we actually found an RNA binder, so opposite of what we found in our biochemical assay with decent activity. And we're looking to further optimize this. But one of the things I want to point out about this molecule is the big discrepancy we saw between activity in cells and activity in our biochemical cattle gas. We don't have a good grasp on this, why this might be the case if there are other aspects of this RNA protein interaction that, you know, we are more potently able to disrupt in cells. This has been observed with other ribosome binders and translation modulators or modulators more generally of RNA biology that in vitro or in a biochemical assay, they're much less active than they are in cells. But I think this raises the point and something that you know, certainly Matt's lab is involved in and helped to pioneer is we really need to have a better sense of, of how to measure target engagement inside live cells and really understand what's happening at the RNA level for these RNA protecting interactions and and the manipulation of them with small molecules. And certainly we're doing our own kind of work and effort in this area inspired by this finding of this molecule. And so this is the work of Jose and Brandon in the lab in collaboration with Jace Weidman, who's a faculty here at Michigan. And so in using pre Q1 as a model, we're able to show that we can, you know, enrich these RNA's, we can integrate our probes with structure mapping and actually use them for transcriptome wide sequencing profiling. And we'll hopefully we'll have this out later this year. But I think, you know, the more information we can get is exactly as Matt said about these molecules inside cells and what they're doing, it will really help to catalyze this field more on the Ripka side. You know, we're still working to develop lots of assays and having a lot of fun doing that, you know, and so we started in non coding RNA, specifically micro RNA's, but you know, they're one of a small 11 family of a very large number of different types of RNA's. And so we've already published our work and some of these systems. So Dalia has done a lot of really great work in adapting this for mRNA protein interactions, showing that we can specifically detect real, you know, RNA ligands versus, you know, mutant ligands that shouldn't have binding. We've demonstrated the hook effect, you know, kind of proving mechanism of how this assay works. Gabriella is working really hard to expand this to reader proteins and looking at RNA modifications. And so we're, we're having a lot of fun, but in having a lot of fun, you know, we're always looking to, you know, understand and, and know that we've developed a rigorous and robust assay. And so, you know, the more we develop, you know, we come to models that of course don't work. And so, you know, in our future work, we really think, you know, about some of these complexities of cellular RNA protein interactions, right? And I am selling why this cell based assay is so much better than a biochemical assay, right? We can, you know, hopefully control stoichiometry and structure and all of these things. But one of The Dirty secrets about RNA binding protein specifically in a test tube is that they can be quite nonspecific. I don't think this happens in the cell because you know, there's compartmentalization, there's a heavy level of regulation. But in RIPKA we are over expressing these RNA binding proteins sometimes on top of already high levels of that RNA binding protein. And so for some, you know examples specifically Alf one which we published in our mRNA protein interaction paper, this is really challenging to actually get that assay to work. And so Gabriella has been working really hard to try to now develop the next generation of RIPKA assays using endogenous tagging of the RNA binding protein at the at the endogenous locus. And so we call this CRISPR RIPKA. And essentially our goal is for it to work exactly the same as regular Ripka. In this case, it would already be making the large bit RVP and using the small bit as our detection region. And so Gabriella has successfully done this with eif 4E as a model. So ER 4E is one of the RNA binding proteins we've studied for a long time and especially in developing inhibitors for this assay. But we can, you know, successfully make these cell lines, we can engineer substrates for EI 4E, so it binds to the M7G cap structure present at the five prime end of coding messenger RNAs. And we can also show that we can inhibit signal with some of the inhibitors in our lab. And so this and some other, you know, optimization, some very, very exciting optimizations that I unfortunately won't have time to talk to you about today. We hope we'll really be able to expand this technology to any RNA protein interaction that you or I may care about to really be able to dig in and, and probe these these interactions with chemically or with other reagents as, as you know, as we see fit. OK. And So what I've hopefully showed you today is through our our work and develop these technologies that were, you know, beginning to have glimpses that we might have a, you know, useful pipeline for enabling probe and drug discovery in this area of targeting RNA protein interactions. So we can either take discovered RNA protein interactions using proteomics, which we've also done, or known interactions we can screen and RIP, validate and cadalk or other methods that we have as well identify small molecules and really get to this point of medicinal chemistry that I think Matt talked about. And we really want to push forward so we can learn new rules for targeting these RNA protein interactions. But I think what Matt talked about and you know, I would definitely agree, is it's going to take a village and maybe a world to accomplish all of these goals. And so, you know, we really are in need of new methods for characterizing these RNA small molecule interactions rigorously inside cells, prioritizing functional interactions or creating new modalities like Ribotax to go after the function of these RNAs. And maybe diverse classes, right, You know, so I'm showing you inhibition, why not glues or stabilizers like wrist a plan, you know, and so I think we need to think broadly about how we might want to do this because every RNA is different. And so, you know, I always, I'll end with this, like my pet peeve is, you know, sometimes we refer to RNA in the singular, which we would never do for a protein because we talked about specifically kinases or GPCRS and we very specifically about their function. You know, for some of these RNAs, we don't necessarily have that information yet, but you know, certainly we don't want to treat them all the same because they're not. And so I think it's going to take lots of different approaches to finally, you know, maybe crack the code of all the RNAs. And with that, I will end with my most important slide, which is my acknowledgement slide. And so I always like to say this for students and trainees, you know, I'm an RNA convert as an assistant professor, having never done an RNA experiment. You know, my background is in in synthetic chemistry and chemical biology. And so I am always indebted to the hard work, creativity of all of my students, past and present that have worked with me and really can put their sweat and blood and tears into getting this work to go. I want to thank all the funding agencies, especially Pro Mega for giving us all these wonderful technologies and being a great supporter. I thank you for your attention and I'd be happy to take any questions now or in the panel discussion. Thanks. Thank you so much, Amanda. Great talk. We always are excited to see the clever ways that people come up with using our tools and Repco's a great example, super excited to see what comes with all the advancements you're making to that in your small molecule RNA target engagement work too. That's super interesting, something we're exploring as well. So all right, Q and AI think we will hold till panel discussion. A reminder to all the attendees, you can pop a question and that ask a question box at any time and we're collecting those and then we can discuss them later in our live panel discussion. Now we'll move into our final poll question of the day. So what do you consider the most promising technology for advancing RNA targeted drug discovery? So again, select your favorite answer here. We have high throughput screening assays, computational modeling and AI driven design, CRISPR and gene editing tools, advanced RNA structural analysis techniques and novel delivery systems for RNA therapeutic. So go ahead and put your favorite answer in. I'll give it a moment as we collect those. All right, looks like we're getting a good amount. Let's see how they look. High throughput screening assay is about 43%. Think this is a very promising technology for any advancing RNA targeted research, also computational modeling and AI driven design, RNA structural analysis and delivery systems. All right, thank you all for participating in our poll. On to our next speaker, Doctor Jay Schnikla. So he is a a senior investigator and head of the Chemical Genetics section in the Chemical Biology Laboratory at the National Cancer Institute. He earned his PhD at Yale University under the direction of Professor Craig Cruz, where he designed the first cell permeable Protac molecules. He then pursued an NIH postdoc fellowship with Eric Sorenson at Princeton University. At the NCI, his research focuses on understanding nucleic acids as targets for small molecules. His work encompasses development of high throughput screening techniques. Chemical Biology. Probe Design, biophysical characterization of target ligand interactions, target validation, and studying the structure and conformational dynamics of nucleic acids. He is currently the Chair of NCIS Medicinal Chemistry Accelerator, a program aimed toward translational development of novel biologically active small molecules discovered within the NCI. And now for Professor Sneakloss presentation. Hello everyone, My name is Jay Schneckloth. I'm a senior investigator at the National Cancer Institute in Maryland. I want to start off today by thanking our hosts at Promega for putting together this exciting event. I also wanted to apologize for not being able to be here Live Today. I had a conflict with something at the NCI, but I hope you're all enjoying the day and, and the rest of the talks. Since I'm not the first talk, I'm, I'm sure we've had plenty of introduction on why RNA is important as a target for small molecules. And I wanted to spend some time talking about how my lab has been approaching what we view to be an extremely important problem in the field. And so, you know, I think there's a lot of unanswered questions in this space, and it's an exciting time to be working on RNA as a target for small molecules. My lab has been working on some what we consider to be fundamental problems in the space. And the things we think about are what type of RNA makes a suitable target for small molecules. Often if you look in the RNA literature, you'll see structure drawings illustrated like this up here. These are secondary structure drawings that illustrate base pairing events and accurately reflect the the secondary structure of the RNA. But as medicinal chemist, we really want to think about RNA in three dimensions. As you can see on the bottom and this RNA here, it has a complex 3 dimensional fold as well as a hydrophobic pocket that might be a good target or a good binding site for a small molecule ligand. We'll talk more about that in a moment. The next fundamental challenge is what kind of chemical matter binds to RNA. Again, historically, you might see look in the literature and see molecules like neomycin or other aminoglycoside antibiotics that bind tightly but not specifically to RNA. We're more interested in developing molecules that are closer to targeted therapeutics or targeted probes at least. An example of that might be the molecule in the bottom ribosil. And this is a a molecule discovered by a group of Merck to buy that binds to a bacterial riboswitch and regulates gene expression by altering the confirmation of the RNA. Finally, and this is actually where we spend most of our time, we've been thinking about what types of tools are needed to better understand RNA Liga interactions. And here we've been working a variety of areas looking at different tools, things that like X-ray crystallography or advanced molecular dynamics, advanced molecular dynamic simulations or single molecule techniques. This type of this type of technique might allow us to understand an individual RNA in high detail, high precision. But we also work on techniques like developing tools like these photo affinity labeling probes shown on the bottom that allow us not not to look at a single RNA in high detail, but to look at RNAs in the context of their native environments and cells. And we can use these to understand or map binding events throughout whole transcriptomes in lysates or whole cells. So I'm going to start off really rather broadly talking about some of these questions. And then I'll go through a few vignettes of individual systems we've looked at on the in the later part of the talk. So the first thing we did, or one of the first things we did when I started the lab many years ago was to ask questions about the three-dimensional structure of RNA. And what we did was we downloaded the entire PDB and looked at the structures of every RNA in the PDB, even when you delete all the proteins if it's an RNA protein complex. And what we found by analyzing these structures using commercially available software from ICM Molsoft that allows you to look for hydrophobic pockets. We look for pockets and RNA that were likely to be binding sites for small molecules based on physical properties that are predictive for for binding sites for small molecules and proteins. And when we analyze these pockets in property space, here you're looking we're looking at properties like buried nest, hydrophobicity and volume 3 dimensional volume. You can see that the pockets on RNA are very similar to the pockets on proteins. RNA pockets are illustrated in red and protein pockets are in blue. This was really a surprise to us because the prevailing sentiment in the field at the time was that hydrophobic, or was that pockets on RNA were too shallow solvent exposed in polar to really be suitable targets for small molecules. And we observed that that wasn't the case. Even more than that, something like 82% of the highest quality pockets already contained ligands using a very loose definition of ligand. And what I'm talking about here could be a small molecule bound in the RNA, but it could also be a buffer molecule, protein side chain, or even an ion crystallizing for example, in a biting site. But regardless, 82% of the high quality pockets already contains some sort of ligand and was used for recognition of the RNA by another molecule. We can also say now we know very well that pockets in RNA are both diverse and distinct from grooves. If you look at a classic duplex structure, you can see that this pocket is indeed shallow and solvent exposed and very extended makes it not necessarily an ideal pocket for recognition by small molecules. In contrast, you can look at this example of the HIV core packaging signal in the structure that was solved by Mike Summers lab at UMBC and you can see that this falls into an intricate 3 dimensional structure that has a globular hydrophobic pocket that might be a great binding site for small molecules. So there are several really important caveats with this type of approach though. First of which is that there's no connection to function. Just because you have this beautiful 3 dimensional pocket that might love to bind to small molecules doesn't mean that a molecule binding to that site will do anything. And in fact we find that many of these binding events are functionally silent. There's also no consideration of dynamics. We know that RNA is a very flexible biopolymer, really exists as a an ensemble of confirmations rather than the individual static structure as is shown here. And so this is an incomplete representation of structure. And so these are some of the limitations. But what we can take away from this type of analysis is that all sorts of diverse RNAs contain these hydrophobic pockets. And this is really contrary to the dogma at the time. So the next question we had is what type of chemical matter binds to these structured RNA's? Here we've been using a an approach that was originally developed by Stuart Schreiber and Angela Kohler, also been used extensively by one of our other speakers today, Matt Disney. Although our approach is a little bit different than his, but we use this microarray based approach to rapidly screen for and identify small molecule ligands. What we do here in this sort of incarnation of this approach is we use a robotic microarray or to spatially array covalent link thousands of molecules, 10s of thousands of molecules to glass slides. We can then take any RNA we're interested in, label it with fluorophore. We like to use red fluorophores like SCI 5 to avoid auto fluorescence by small organic screening compounds. You incubate that RNA onto the glass slide and then you generate array data. And what you're looking for here is individual binding events. In this case we see that this RNA illustrated here or binds to the compound that's printed at these locations on the on the array. And we can use this to validate and identify and validate binding interactions between the molecules printed these locations and the and the RNA of interest. We can screen about 44,000 drug like compounds in one day and requires a relatively small quantity of labeled RNA. This is an amount that you can make yourself pretty easily, or you can buy if it's rather short from some commercial vendor. And you know, the real power for these approaches is that you can do screens very quickly, which means that you can compare the results of many screens against each other. So we can not only look at individual binding interactions, but we can compare the binding of individual compounds across many RNAs and get insights into both selectivity of binding events, but also targetability of classes of structure. So we've now looked at all sorts of diverse individual RNA and DNA targets using this SMM approach. I'm showing you here examples of compounds that we published over the last few years. I'm not going to go through all these, but these are targets, These are small molecules that bind to different nucleic acid targets, mostly for cancer and infectious disease targets, Cancer and infectious disease relevant targets. And what I hope you'll take away from this is that all sorts of drug like chemical matter are at least relatively accessible. Chemical matter binds to binds to RNAs of interest. One of the important things that we noticed was that the hit rates for these RNA's are relatively comparable to unbiased protein screens. And we typically identify small molecule ligands with affinities ranging in the one to 25 micromolar range for decent hits. This is about what you might see for a protein screen. And so this is a really encouraging observation. And with with those, most of the compounds we've seen are in the micromolar range, but in several cases we've been able to get down to the nanomolar range. I'm showing you here examples of compounds that break the sort of sub micromolar barrier and also one that's a low nanomolar binder. And in fact in some cases now with design ligands, we can be able to get to the picomolar affinity regime. So we can identify compounds that have relatively specific interactions and relatively tight interactions, what you might need for a probe or a therapeutic lead. Not only looking at individual compounds, but we look, we can look at classes of RNA structure to ask which types of RNA structure like to be targeted by small molecules. In this example, I'm going to show you a profiling experiment that we do with G quadriplexes, a class of structure I'll be talking about more today. These are enriched in DNA and oncogen promoters and in in RNAs. They're enriched in splice sites and UTR's here. This is a heat map of compounds that bind. Each row in the heat map is a different RNA or DNA in each columns, different small molecule and the more red the tighter the binding event. So this is a subset of 13 quadruplexes versus 26,500 small molecules. And from these experiments we can get insights into both selectivity of the interactions but also targetability of classes structure. So we can say that all sorts of diverse drug like chemical matter binds to G quadruplexes and we can say that the pan quadruplex binders tend not to bind to other classes of RNA like triple helices, three-way junctions, bulge or bulge structures. We can also say something of course that's well widely accepted in the field. It's very difficult to achieve selectivity. Most of the compounds that we look at look like this on the left where you can see they bind to many, most or most if not all of the quadruplexes. Whereas in rare cases we can find selective binding events that are shown on the right and we think these are more interesting from a targeting perspective. We can also say that small molecules recognize classes of structure. We can do the same type of analysis for things like pseudo knots or three-way junctions. And, you know, we can really get insights about the types of chemical matter that like to bind to different types of, of, of, of RNA structures or DNA structures as well. And this is really complementary to the, to the sort of chemi informatic type work that people like Matt, who's also speaking today, have done really beautiful work. And also Amanda Hargrove has made important contributions to this, to this space here. So I wanted to acknowledge their work. And I think from our large data set and our analysis here, we can get some really interesting insights into RNA binding chemical space. We've now gotten actually this is a little update we need. We now screen over 90 different targets, mostly cancer. I work at a Cancer Institute, but we've also looked at infectious disease targets. So viral and bacterial RNA's and this really represents highly diverse structures ranging from 19 to 600 nucleotides, simple, simple RNA hairpins all the way up to ribo complex, ribozymes and even entire UTR's. We generated a data set in House of over three and a half million interactions probed. And we can say from this analysis, something that I think is a consensus beginning to emerge in the field is that RNA binders do exist within drug like chemical space. And so if we look at properties of the compounds that bind RNA here you're looking at histograms or KDE estimates of histograms in blue is RNA binding compounds and oranges protein binding compounds. You can see looking at parameters like medicinal chemists think are very important things like molecular weight or rotatable bonds or predicted solubility. You can see that the RNA binders we identify are similar in many ways to protein binders. And whenever you have a large data set these days, I think you begin to think about how to can you, can you use machine learning to apply that to the data set to learn anything? We've been working here with a a small startup company known as Ladder Therapeutics for several years. What we did was we took a well characterized or well annotated data set of 26,000 small molecules verse 38 different RNA or DNA targets for a total of 1.6 million interactions probed and a collection of 2000 new RNA binders that had not been reported in the literature before. We annotated 1700 different chemical properties for each one of these compounds, as well as a another set of compounds that were known to bind to proteins. So these are medicinal chemistry compounds from the literature like enzyme inhibitors. We were able to ask which of these chemical properties are more likely to to make the compound bind to RNA versus proteins. And we used a variety of different binary classification algorithms, both both interpretable and black box algorithms, things like LASO, logistic regression or multi layer perceptron neural network algorithms. We found that a variety of algorithms have high predictive power for guessing whether or not or predicting whether or not small molecules are likely to bind RNA or protein. Just showing you an example of one of these algorithms here. And from this work, we were able to say that you can develop a set of a complex set of physical chemical properties that are that are highly predictive for RNA binding within a library of compounds. So we can do things like taking test sets of compounds, things that are known to bind RNA in the literature like this compound 80 Q compound four that came out of our lab screening Ribosel I mentioned earlier from a Merck group or tetracyclins. These compounds are predicted to have high by probabilities of binding to RNA. Whereas other compounds like imatinib, ibrutinib, lovastatin and European are much less likely to be binding to class, much less likely to be binding to RNA. We can also identify functional groups and scaffolds that were enriched in RNA binders and say that that, you know, we can really get begin to get insights into the types of compounds that like to buy an RNA. We can say that for example, in at least in the, in the sort of solution we came to is that things like high nitrogen content, low oxygen content and properties like Vanderwall's surface area, something demand or Hargrove has also noted. These properties can be important and powerfully predictive for binding to RNA. You can also say that RNA binders are highly diverse in chemical space. So here I'm showing you AT map representation. This is a tree like representation of chemical space and compounds that are on branches that are close to each other or compounds that are close to each other and branches are more structurally similar to each other. And what's interesting about this if you look at it, you don't see much in the way of clustering means that all sorts and I should mention in blue are are protein binding or known protein binders black or FDA approved drugs and and. Orange are the RNA binders that we've identified and you can look here. I'm sorry this isn't showing up very well, but if you look at this one particular branch, you can identify all three classes in structurally similar examples. Here in the bottom we have osanomide, Osanomide which is Fingerstein phosphate receptor agonist. And you can see here that compounds 6 and four have similar chemical chemical structures. Those are known enzyme inhibitors from literature, whereas compounds 1-2 and three are new RNA binders that we've identified through through screens. So from this analysis we can identify next drug like RNA binders or at least compounds that have physical properties that medicinal chemists like to think make them drug like. It's a very loose definition of drug like. We might also be able to use this to identify FDA approved drugs with RNA off target binding and possibly off target pharmacology. If we're able to study them enough, it can inform next generation high throughput screening library design. And this is now I think the largest fully public repository of for RNA binding ligands and I think the only one that also contains information about things that didn't bind, at least in the screens we tried. So if you're interested in looking at this, you can download the whole library and look at it and develop your own conclusions if you think we're not right about something. But it's in this paper here that we published last year. So I wanted to spend the rest of the time talking about individual systems we've been looking at. Early on when we started the lab, we were inspired by, you know, the work of people like Ron Breaker and a huge community of other folks who have been studying structured RNAs and bacteria particularly, particularly these ribo switch systems. And for many years now we've been publishing on the pre Q1 ribo switch. This is a small RNA that exists in the five prime UTR of bacterial ribo switches and it binds to this hyper modified nuclear base known as pre Q1. When the RNA binds to pre Q1 it undergoes a dramatic conformational rearrangement going from a hairpin like structure to an H type pseudo knot. And actually the small molecule binds where that red dot is in a hydrophobic channel right in the center of the pseudo knot and makes a network of hydrogen bonds that leads to highly tight and specific binding interaction on some cases as low as A2 nanomolar KD. Now a variety of different organisms have evolved pre Q1 ribo switches independently. So an example of convergent evolution. We thought this is intriguing as an example of different times that nature has evolved RNA aptomers for a metabolite. It's an example of a ligand induced conformational change of an RNA and it's a small tractable aptomers. So it's relatively straightforward to study structurally, which we thought was really attractive conceptually. We also thought it was intriguing that this is sort of how nature has solved the RNA small molecule recognition problem. So we think we could learn a lot from that. So what we did was we designed a screening construct and we performed a small molecule microwave screen. We identified compounds that could bind to the pre Q1 ribo switch and not other Ribo switches like Sam, Sam or TPP ribo switches. And it didn't also also didn't bind to like broad broadly speaking, other RNA structures like hairpins, triple Ulysses or Quadra plexes for example. So this is the chemical structure of one of the lead compounds I'm going to be talking about today or hit compound that I'm going to be talking about today. In collaboration with Adrian Ferry de Mari, my friend and longtime collaborated collaborator and HLB I, we're able to put this compound after extensive biophysical analysis, we're able to put this compound in a functional assay. What I'm showing you here is a single round transcriptional termination assay showing that the compound, actually the synthetic ligand has actually the exact same functional consequence as the native ligand. When you put it in these termination assets. You see here that pre Q1 as well as our synthetic ligand have actually the exact same functional consequence, which is give rise to this terminated transcript. You can actually quantify this and see that it's a saturable but relatively weak potency effect, but a reliable and reproducible effect. Now real advance came, we were able to get an X-ray crystal structure of this compound bound to the RNA, again in collaboration with Adrian's lab. You can see here in cyan is the ligand. It sits in that hydrophobic cleft to the center of the pseudo knot and in fact it does bind to the same binding pocket as the cognate ligand, which gave us a plausible mechanism for how this compound was working. It's a 1.8 angstrom resolution structure, relatively high for an RNA. Although the contacts differ from the cognate ligand, it does bind to the same binding site. And so you can see here that it's mostly shape complementarity and stacking interactions instead of that network hydrogen bonds. In fact, this compound only makes one hydrogen bond with a 29 still on. The access to a high resolution structure meant that we could possibly use it for structure guided improvement. We published this now several years ago in Nature Communications and since then we've been working on trying to improve activity in functional assays. Although this compound bound at about 1 micromolar KD to the RNA as a very weak activity, functional activity. And we were able to use computational modeling, pretty straightforward computational modeling as well as rational design to identify multiple new chemical scaffolds that also bind to this, to this hydrophobic pocket. And in fact, a variety of these scaffolds have dramatically improved activity and functional assays. In fact, in one case down to less than 10 micromolar, which is which is really good, a really big improvement. I'm going to show you an example of one of these scaffolds here. We had hypothesized that if you replaced, if you inserted a carbonyl group in between the aromatic rings of the dibensifteran scaffold, you might be able to leverage a new hydrogen bond or at least a bonding interaction with the RNA on the far side of the binding pocket. And in collaboration with Tom Numata, a former student, a former postdoctoral fellow of Adrian's who's now begun his independent career at Kyushu University in Japan, we're able to, Chris, get an X-ray Co crystal structure of this compound bound to the RNA. And we could find that indeed as we had anticipated or I should say hoped this compound actually makes now a second bonding interaction with the two prime hydroxyl of G11 on the far side of the binding pocket. So we hypothesized that this sort of new bonding interaction might be why the compound, even though it's not more tight binding, it's about a one micromolar KD as well, but it's dramatically improved in functional aspects. So it's not just the binding event, but the nature of the interaction that impacts the activity. We wanted to dig a lot deeper in mechanism and we've we've been collaborating now with a company called the Pyxis from France who does this, has developed an instrument to do single molecule magnetic force spectroscopy. And the way this assay works is that you put your oligo in between a bead and attach to the surface using complementary oligos and using a magnet, it can detect both the force it requires to unfold the RNA and the amount of time it spends in folded and unfolded states. And so we did this with the pre Q1 ribo switch and we were able to show that you can see here that in the presence of DMSO, we have, roughly speaking, 2 state populations. Here you can see on the top this is the unfolded state. And here on the bottom, this is the folded state. And if you look at this histogram here on the right, you can see that the compound spending more time in the unfolded state on the top than it is on the on the bottom in the folded state. Now when we add pre Q1 to this system, you can see that it stabilizes the pseudo knot and causes the formation of persistent folded pseudo knots. OK. And see this is reflected not only by these long periods of time when it spends in the folded state, but also in the histogram where the population shifts to the folded state. Now we were surprised to see that our synthetic ligand had a very different effect. Here you can see when we add our synthetic ligand, we do not see the presence of these persistent folded pseudo knots. In contrast, what we see, but we do what we see, is that the the synthetic ligand actually alters the rate of refolding of the RNA. So it makes it actually fast fold and unfold much more rapidly. But in the end the aggregate is that the compound still causes the RNA to spend more time in the folded state. And so this really gives us a plausible mechanism again for why the compound is is is working in the aggregate result is that both ligands induce transcriptional termination, even though the cognate ligand has a thermodynamic mechanism that caused these persistent folded pseudonauts. And our synthetic ligand actually has a kinetic mechanism that alters the rate of refolding. So dramatically different mechanisms for two compounds that bind to the exact same binding site of the RNA. We wanted to go forward and ask whether or not these compounds have activity in cells in live bacteria. Here we developed a, a reporter system. We got this reporter system from Joe Wetican's lab, who in turn got it from the Beatty lab who originally signed it, where you put the Ribo switch in front of GFP. And the idea here is that inhibitors would block fluorescence of the bacteria if it if it's successful. And we were able to show here that if we grow bacteria in Agri, you can see here that both that in the presence of DMSO, the, the bacteria are flushing green in the presence of are are are the cognate ligand, you can see that it's silent and you can see that our synthetic ligand also silence is expression of the GFP. So finally, when we put the xanthone in with what we considered an empty vector or really a riboswitch lacking vector that expresses GFP, our compound has no effect. And then similarly, the compound that's a negative control that has no binding to the RNA in biochemical assays also has no effect in in the bacteria. So you know, an example of where we can use a bacterial system to try and better understand how mechanistically 2 very different liggets can achieve the same result. A really intriguing system. If you're interested in reading more about this, it just came out a few weeks ago in Nature Communications. Shefali Parmar is the first author of this, although it was a really collaborative effort. So I'm going to end the talk with a story about a human system that we've been thinking about for a long time and that's NRAS. I work at a Cancer Institute, so we're very interested in how we can develop compounds with new mechanisms of action and anti cancer activity. NRAS, in case you're not familiar with it, is sort of related, of course, to the K Ras protein small. It's a small GTP ACE. It's involved in regulating cell growth. It's highly dysregulated a variety of cancers, including neuroblastoma, which is where the N comes from, and it's a classically undruggable oncogene for, for lack of a better term, unlike K Ras, which has now been drugged, there simply aren't any drugs yet for people who suffer from cancers that are caused by deregulated N Ras. So we were really intrigued by a report from Shankar Bell Supermaniums Lab illustrating that if you looked at the sequence of the NRS mRNA, the very five prime end of the five prime UTR, there was this G quadruplex structure. This is really a landmark paper the first time showing that a structured G quadruplex can influence translation of the gene. And we thought it might be really attractive to try and target this G quadruplex using our platform. I'm going to go through this relatively quickly, but we did a small molecule microarray screen, found a compound that had some selectivity for the Nrash quadruplex over other quadriplexes as well as other RNA structures. And we use extensive biophysical analysis, things like thermal melting out analysis in the circular dichroism to show that the compound actually thermodynamically stabilizes this quadruplex. We can also use things like fluorescence titration and SPR biophysical techniques to show that the compound binds reversibly to the RNA. Though there is a big discrepancy between the affinity we see by fluorescence titration or SPR, We think that the actual binding Infinity of this compound is about a micromolar, is about what the KD is. Sumi Valoratnam, who's a postdoctoral fellow in the lab, was able to do RNA, a footprinting to indicate a site of interaction with the COMP, with the compound, with the RNA. And as you can see here as a function of dose, we can see protection occurring at C-17 as well as C-12, but not at U16. So we're a little surprised by this. Initially we found it to be a little bit unusual that a compound might be blocking are protecting or interacting with two nucleotides that are are not next to each other in in sequence space. It might mean that they are close to each other in three-dimensional space. And so in collaboration with Adrian's lab again, they were able to solve an X-ray crystal structure of the RNA and you can see that it adopts this canonical G quadruplex fold. We were not able to get a crystal structure in complex with a small molecule, but we were able to see here through the crystal structure that as we had hypothesized C12 and C-17 are close to each other in 3 dimensional space, but that residue that is not affected is actually pointing away. So we think this yellow highlighted area is likely the binding site of the small molecule. In collaboration with Danny Incarnado, we're actually, we're also able to use shape map structure probing on the reporter gene that contained the the NRAS 5 prime ETR. And we're able to illustrate that in comparison to all the other G's spread throughout this or distributed throughout this mRNA, we could see that the G's present in this G4 are far more ordered and and structured and less reactive to to shape map than the the rest of the G's. And so this indicates to us that not only is our quadruplex found in the in, in the in isolation and tested, but even the complex of a long mRNA, we could see folding of this G quadruplex. We're also able to use a reporter system that we got from the Bell Subramanian lab where we had paired luciferase reporters, one of which contains the G quadruplex and the other is mutated. So it does not. We use these reporters for in vitro translation assays. And we're able to show here on the bottom that our compound inhibits translation only of the reporter that contains the G quadruplex. So evidence that this compound is blocking translation of the gene only by by binding to the G quadruplex and not other structured elements within the mRNA. Now a surprise came when we put these compounds in cancer cells. We were able to show we were we were very surprised to see by both RNA seq and global proteomics that NRAS mediated signaling and NRAS levels did not change. However, we did see that 70 different RNA, DNA associated quadruplex associated genes were not perturbed as well. So the compound is selective and in functional assays it at least in, in lysates and reporter genes, it it impacts NRAS, but not in the endogenous cancer cells. And we were trying to figure out what might be happening here and we were quite confused by this. And it was only when Sumi developed a, a PCR assay where she designed an amplicon that would span the quadruplex region as well as span the coding sequence either of NRAS or of the reporter system that we're able to figure out what's going on. And it turns out here that if you amplify these two these these two regions, you can see here that both the, the quadruplex, which is red, and the coding sequence, which is on the right here, we can see that we amplify both regions very well in the reporter gene. But when we go to endogenous cancer cell lines, you can see that that's not the case. And So what we're observing here is the fact that most of the NRAS transcripts that are found in in in these cancer cell lines, the majority transcript actually lacks the quadruplex and is forming is, is expressing a shorter transcript. And so most of the NRAS transcripts in these cancer cell lines are free to express NRAS independent of whether or not the compound is binding the quadruplex. We analyzed today 14 different cell lines. In all cases, the major transcript lacks the G4. We confirm this mapping by race analysis, 5 prime and mapping. So we still think this would be a really attractive way to go after NRAS if we can identify cancer cell line that does express this longer transcript. But so far we haven't been able to find one. So to wrap up this section, I'm just about done with my talk. We were able to use our microarray strategy to identify compound that chemically inhibits the translation of of an RNA by targeting a structured region of A5 prime UTR. In the specific case of NRAS, transcript heterogeneity really impacts the target ability, an important lesson for the field to learn and certainly a tough one for my lab to learn. But it does provide a road map for how to target a gene selectively with a small molecule binding to a structured portion of that mRNA. Now we've been working on finding other structured regions within the NRAS RNA itself, but we think the path forward for NRAS might be other targetable structures or more broadly to look at other genes that have structured elements in their five prime UTR's such as other quadraflexes. This is my last data slide and and we published this earlier this year, so I won't go through it. But we've been able to show that, for example, the DHX 15 mRNA contains a quadruplex and we're able to show that you can use small molecules in the similar approach to identify compounds that will block at least in vitro translation assays will only inhibit the translation of quadruplex containing reporters and not mutated reporters. So I'm just going to wrap up here and say that I told you today about how my lab is approaching nucleic acids as targets for small molecules. We use this microarray screening approach to identify and profile targets by studying the molecular functional and functional bases of nucleic acid interactions. We can do ligand optimization and cover new functions for ligands. And finally, we can use machine learning as a powerful tool leverage towards drugging RNA. I'm going to end here by thanking the people in the lab who did the work. I'm really fortunate to have an absolutely spectacular group of people. I've listed their names along the slides and tried to mention them during the the real leaders of the projects who I talked about during the talk. But you know, really, we do interdisciplinary science, so almost everyone in the lab contributes to most projects. So I really love this collaborative environment. I'm also really fortunate to have a spectacular group of collaborators and colleagues who I've listed along the way, but they're also listed here. Thanks to the NCI for funding, and thanks again to Promega for putting together a beautiful symposium today. I'm thrilled and honored to be a part of it. Again, apologies that I can't be here, but I'd be happy to respond to questions by e-mail after the fact. So thank you again and I hope you enjoy the rest of the the seminars. Thank you to Professor Shikla for joining us and sharing his work. As he mentioned, if you do have any questions for him, we will be sure to get them to him. We will move on now to our final speaker of the morning. After the talk, we'll have our panel discussion live. So if you have any questions for the presenters or any topics that you are interested in them diving deeper on, please do put that in the Ask a question box and we will talk about it during the panel discussion. So next we have Brad Swanson. He's a strategic Collaborations Manager at Per Mega. He works with our R&D and business teams to connect scientists with Per Mega technologies and services with an emphasis on per Mega cellular drug discovery products for both large and small molecule drug discovery. Prior to joining PER Mega, Brad was Vice President of Life Science R&D and Strategic R&D programs at Cellular Dynamics International or CDI, and before CDI, he was a senior scientist at both Roche Nimblegen and at Promega. Brad received his PhD from the University of Wisconsin, Madison, and he performed his postdoc research in the laboratory of Philippa Merrick at the National Jewish Medical and Research Center as a Howard Hughes Medical Institute Postdoc Research Fellow. So welcome, Bran. Thank you for attending day three of our own Druggables conference. My name is Brad Swanson and I'm a Strategic Collaborations Manager here at Permega. For my talk, I'm going to provide an overview of two of Permega's drug discovery technologies and how they can be used in the RNA drug discovery space. The first example that I will discuss is our CRISPR Hyde gene tagging technology to enable precise real time quantitation of the small molecule and siRNA modulation of protein production. For the second-half of the talk, I will discuss our Nano Brett target engagement technology to measure modulation of RNA protein binding complexes. Both technologies that I'm going to discuss today are based on Promega's Nano Luck luciferase, the small, stable, and extremely bright luciferase derived from a luciferase present in a deep sea shrimp. Nano Luck is roughly 19K ohms in size and produces almost 100 times more light compared to Firefly luciferase or renewal luciferase. It's also highly thermally stable and has many other excellent properties for a reporter molecule. We've used nano log to develop many very valuable and very widely used cell based assays at Promega in the last decade plus. However, despite its small size, there are some applications or an even smaller the Safaris tag is advantageous. With the desire for an even smaller lumigenic tag in mind, our advanced technologies group of Promega developed the nano bit split luciferase system by splitting nano log luciferase between amino acids 156 and 1:57. This site minimized self association of the two split portions producing 2 components, large bit, which is a stable bright large subunit bright when it's complemented with the short peptide and the and their to complement large bid. We identified several small peptides with varying affinities for large bid that are required for the complementation in the enzymatic activity of Nano Lock. We used two different peptide versions depending on the assay, small bid when small lower affinity interactions are required for the assay and high bid for higher affinity detection and assays that require high affinity interactions. So we used Danabit and specifically high bit to develop the CRISPR high bit knock in cell model to allow analysis of endogenous protein expression in high throughput simple workflow using CRISPR CAS 9 gene editing. The small bioluminescent high bit tag is inserted at the endogenous locus in the gene of interest, either at the C terminal or N-terminal region of the protein. This enables quantitation of the endogenous protein expressed within the cells using either kinetic or endpoint assays, depending on the application that you're looking at. An example, some of the examples shown in the cartoon demonstrate how you can use hybrid tag genes to measure relative just general protein expression. You can use it to measure receptor internalization, you can use it to study protein secretion, and it's often widely used to study targeted protein degradation. The hybrid tag when it's used in this way is a very versatile tag with a wide application base. By using the CRISPR cast 9 gene editing to knock in this bioluminescent tag at the endogenous locus you use, you can use these cells to study proteins expressed from the native promoter with a very simple assay using light output as a surrogate for the protein level expressions. And the small tag allows for a really high knock in efficiency and and reduces the requirements for cloning and other molecular biology requirements to do gene tagging, tagging and endogenous loci. So it's a really nice simple system to get up and working in your lab. Now the first example I'm going to talk about is work from Matt Disney's lab studying using a CRISPR hybrid model to study regulation of the alpha synuclein protein in the SCNSNCA gene at the endogenous locus to study to try to identify small molecules that modulate protein expression from the gene. Now, the goal of these studies was, as I mentioned, to identify small molecules that might bind and disrupt production of alpha synuclein from the mRNA. Since it's been difficult to drug alpha synuclein due to its unstructured nature. Previous studies have identified some molecules that bind to the five prime structured UTR of the alpha synuclein gene. One of these proteins or one of these small molecules is called synucleosid and it selectively binds the mRNA the at that five prime UTR structured region. And the Disney lab used CRISPR hybrid knock in for alpha synuclein to study how the small molecule modulates its expression in real time. And that's the data and the slides that I'm going to go over in the next couple slides. So the first study that I've shown here is looking at how this synucleosid molecule selectively inhibits protein production from the the mRNA, the SC, the the alpha synuclein mRNA produced within the hybrid engineered cells. On the left, you see measured CRISPR hybrid alpha synuclein protein levels with increasing amounts of the synucleosid molecule and you can see it higher concentrations of this molecule of both 24 and 48 hours. You see a reduced expression level of the protein. And what's important to also incorporate in these assays, we are treating the cells with a small molecule to influence the cells behavior and in this case production of Alka synuclein protein. You should also do an assay that makes sure that the treatment of the cells of the small molecules not influencing general cell health. You can very easily do this, do this with an orthogonal assay in this case. And that data is shown on the right where you see the treatment with the same treatment of those cells measuring cell viability with cell tiger floor, you see no effect on general cell viability. The previous slides data shows readouts using an endpoint assay looking at either 24 or 48 hours. One of the benefits of the CRISPR hybrid system is the ability to actually do kinetic measurements over the course of multiple days with the same cells using cells that express large bit inside the cells that are that harbor your CRISPR hybrid knock in. And the same study that was done on the previous slide was done but in a kinetic manner to look at the kinetics of of reduced alpha synuclein protein levels in those same cells. And as you can see here with increasing treatments of the of the of the cells with synucleosid, this nucleosid molecule, you can see within the 1st 20 hours of the kinetic, you really don't see much reduction in the protein levels. However, at 50 hours, you see a significant decrease at most if not all the different concentrations of the treatment. And this, this slide really demonstrates the power and the additional information that you get when you use the CRISPR hybrid models in a kinetic mode. At the same, in the same way, you should always be checking your cells to make sure that your your treatment is not influencing the cell viability. And they did those studies which are easy to do in the system. And you can see that at all concentrations in this particular study, you see no influence of cell viability with increasing treatment with cynically, is it? Now the other important aspect to keep in mind here is looking at, you're looking with the CRISPR hybrid model of endogenous levels of the mRNA. Depending on the cell model that you use, you might have some different differential expression levels of your target protein or your target mRNA. And that's something that was seen here. When you compare the ability of the small molecule to knock down out this nucleon expression and SYSSYSHY 5 cells on the left compared to the Heva hybrid CRISPR Nodkin cells, you see a more robust ability to knock down the expression in the neuronal cell model. And the reason that that is, is because there's a higher level of of, of the the gene expressed in the HEBA cells relative to the SHS Y5 cell line. And that's shown here from the quantitation of the alpha synuclein transcripts and data from the protein Atlas as well. So the HEBA cells have more mRNA to inhibit. So it's harder for them to knock it down with this molecule, which is why you see a slightly smaller reduction and and no reduction and lower concentrations of the small molecule. Now the next couple slides, I'm going to talk about a slightly different way to use the same methodology to kind of study RNA as a molecule for treatment or or modulating protein levels. And that's using SIRNAS to knock down expression of protein and using CRISPR hybrid model to read that out. This next this slide shows an HEKH EK293 cell line that stable expresses large bit and has the MIT gene tag with the hybrid peptide. And that was made with the purposes of studying how you can knock down, knock down target genes with SIRNAS and read that out using the Christopher hybrid system. And you can see in this slide, both cells, Christopher hybrid cells tag with the hybrid tag at the MIC locus either treated with srnas, a clonal srna or a pool of srnas. Either one of them produces a significant reduction in the levels of MIC protein in those cells over the course of 28 hours. And this, again, this is a readout that's not an endpoint asset. This is an actual kinetic we read where we're reading the same plate every 10 minutes to get a real time measurement of the protein levels as is, as is the luminescence signal as a surrogate for protein levels. Now using the same setup, this slide shows how you can use endpoint readings to demonstrate the knockdown of the protein of interest with, again, either a pool of SIRNAS to MIC or a single version of an S single sequence of the MIC siRNA. And it also demonstrates, again, as I mentioned before previously, that you should always be looking at the cell, the health of your cells when you're treating them to make sure that the reduction in the protein levels you're seeing is not due to cell health and viability. And that's what's shown here on this slide. You can see again the pool siRNA for C MIC is slightly more robust and it's knocked down capabilities and the single srna for MIC and for all three treatments of the cells including the negative control, you see a consistent level of cell viability suggesting that the knockdown is not due to an off target effect and cell viability. Now it's also important and I want to spend just a second to plug our Maxwell instrument here for all these experiments. It's obviously really nice to have a very robust and easy way to purify RNAs to verify the levels of knockdown that you're seeing, especially in these siRNA type experiments is is easily quantifiable by downstream analysis of the RNA level. First step in that is having a great way to purify your RNA and our Maxwell instrument is really good for that. You know, this is an instrument that lets you quantify 16 samples at once. It has an integrated fluorometer for sample quantitation and really easy to use software and a lot of different chemistries that allow you to purify RNA from most different biological samples. In this case, you're going to be looking typically at the chemistry for purifying RNA from cell cell culture samples. And we use that actual purified RNA to actually do quantitative gene analysis to show that the knockdown we see from the SIRNAS leads to actual gene expression for MIRN AM RNA level reduction using this analysis. And that's what's shown here. The cell, the same cell line was treated with the different SIRNAS and then the RNA was collected using Maxwell. We used our Dotac RTPPCR system to quantify the gene and see the gene expression. And you can see that the gene expression levels by ARTS and PCR closely mimic what we see for the protein expression levels by the high bit measurements. In addition to that, here's the nut there are the the next step that you may want to take to really verify your data would be to look at actual protein using a blotting technique. And you can actually use the the hybrid tag to do hybrid blotting similar to a Western blot, but you're detecting actual light production from proteins tag with hybrid. And that's shown here on the left where we've treated cells with different Mick srnas treated as Christopher hybrid cells that have had Mick gene tag with hybrid and then treated with different Mick srnas and then action aided on SDS page gel and then perform I did Western blotting and you can see with the different levels of the different srnas of Mick applied to the cells, you see a nice reduction in the protein levels. If you see a less intent and with ischemic treatments and the control treatments have basically very little and no change compared to to non srna treated cells. So this is another way that you can use the hybrid tag to explore your protein levels. If you want to verify something about the size of something like that, hybrid blotting is very convenient. So in summary, using CRISPR hybrid cell lines, you know, it's a great way to study protein levels and using live cell assays if that's of your, if that's something that's of interest to you, really simple high throughput bioluminescent tag Multiplex. These assays as I mentioned for studying things like cell viability after your compound treatment. Kromega has actually generated over 200 different CRISPR hybrid cell lines that we have in our catalog that are readily available for purchase. And we're always making new targets or we can make a specific target for you and our tailored RND solutions group. But that's something that's of interest to you as well. So for the next part of the talk, I'm going to discuss how you can use nano bread target engagement to measure RNA protein binding, kind of a unique application, some of the nano bread target engagement application that's been around for the past almost 10 years now. Now kind of as an aside, just as a little background, typically when we're talking about nano Brett, we're talking about bioluminescence resonance energy transfer using nano luck luciferase as the light donor. So I've talked a lot about nano luck, but the other component in that assay is often our Halo tag, which has been labeled with the fluorescent ligand to be the fluorescence light acceptor Halo tag. Just as an aside, I want to give a little plug for it. It's a great tag for your studies as far as being able to do a lot of different things as far as microscopy, protein, protein, protein interactions, different types of microscopy. You know, some super resolution microscopy studies have been enabled by recent new ligand developments that we've gotten recently placed in our catalog or or for sale. So it's something to keep in mind when you're thinking about doing on Nanobret target engagement type studies, but more specifically just nano bread protein protein interaction studies. Nano bread works typically with when you're trying to study two proteins coming together and one protein is labeled nano lock and the other is labeled with Halo tag that's been labeled with a floor to make it fluorescent. Then the presence of these two proteins coming together, protein A and protein B, you'll bring together nano lock close enough to that fluorescent Halo tag to produce a signal from the floor on the Halo tag that's been derived from the light donated by the Halo tag model. So you have energy transfer from the nano lock to the fluorescent Halo tag to produce signal. That's kind of a typical way that you would use nanobrick technology typically in protein interactions. But today I'm going to talk about a slightly unique the way that you use nanobret detection technologies to measure a small molecule binding the protein part. And this is a mock method that's been developed almost 10 years ago now by Matt Roper at Omega, described here by by Matt Work and others, Matt, from 2015. And this is a methodology that led to actually study small molecule binding and to target proteins in live cells. In the past, this has been done typically by biochemical studies studying binding to purify proteins. There's many advantages doing this in cells, which is what this method allows you to do. I'm not going to get into that. That's another topic for another day. But the system itself is described on this slide where you have your target proteins used in analog and you express that in the cell. And then you also have a a tracer molecule that's labeled with a fluorescent ligand and that's shown here, shown here in the cartoon. When you apply that ligand or that tracer to the cell, you get a breath signal because comes in close to the contact with the nano log signal with the fluoresce. And then what you're really trying to measure in this system is the ability to displace that fluorescent probe with your molecule whose binding you want to understand that binding to that target protein. And that's shown in the next step of the cartoon where you apply your drugs and then you abrogate that Brett signal with increasing amounts of your molecule's concentration. When you do that, you can actually develop some nice dose curves that show binding affinity of your small molecule for that target protein. We first applied this technology to kinases. It's been applied to many different cellular classes over over recent years. And I'm going to talk in future slides about how you can use this concept to actually look at all of those binding to target proteins within a cell. So the studies I'm going to talk about for the rest of my talk have been done in a collaboration with a professor Francois Hughes in Belgium. And his lab is interested in studying RNA binding proteins and their effect on cancer. And the targets in this study are the splicing factor SRSF 2 and how it is binding differentially, preferentially to RNAs that contain an epigenetic mark called Plaid methyl sided. And really this SRS of two molecules and and genes that have been and in its target genes that's involved in splicing have been implicated in a number of different cancers. It's got a binding motif that's shown here on the slide and in previous studies that that the lab has performed one of the the only one of 12 binding proteins that have been identified to preferentially bind this consensus sequence that contains A5 methyl cytidine in the two positions. So the goal of these studies was to try to develop a cell based system to understand preferential binding of all the nucleotides contained as 5 methyl cytidine to the SRS 2 protein. And a cartoon in the upper right kind of shows this in a very simple fashion. And the components of the assay are a fluorescently labeled all of the nucleotide in this particular setup with or about A5 methyl cytidine mark at the consensus binding sequence and the ability to load cells with that and measure nanobret signal and see if you can see any difference in all of those that have the five methyl cytidine and versus those that don't. And the chart or the graph in the bottom of this slide shows that five methylphytidine containing all the nucleotides in this system have a preferential affinity for the SSRS RF2 protein and the spell based nano bread target engagement asset. And this is a pretty simple application of this concept. And the next slide, they took it to a little bit greater detail and you can actually look at look into more detail under these studies through their publication and molecular Scalk in 2003. Now the little the slightly more in depth studies actually use kind of the more traditional workflow for target engagement studies where you have a tracer Sir, tracer RNA molecule that's labeled at the five prime end with the floor and you're trying to compete away it's binding to the target protein with non labeled oligonucleotides with or without that bimethyl cytidine mark in the consensus bindings. You're measuring displacement of A4 labeled oligo with these two different versions of the cold we call the cold oligo. And you see the graphs that you obtained from this type of study on the top of this slide, the version the data shown in blue are where you are competing away the tracer binding with non bimethyl cytidine all of the nucleotides and on the right the same sequence but with five methyl cytidine incorporated in that consensus finding sequence. When you calculate the IC 50S from these curves and do a Chang Crusof analysis with this graph on the bottom, you see in the red the five methyl cytidine incorporated all of the nucleotides have a lower apparent Ki for the protein relative to the non marked all of the nucleotides with the same sequence, suggesting again a higher affinity with this particular sequence in this particular consensus sequence for the target. Actually did the same studies in more depth with a little more variation of a sequence of the nucleotides in the in the consensus sequence and that's shown in the panel of four figures on the left. Bottom line here is that all the different variants that incorporate 5 methyl cytidine with the different sequence variants of the consensus sequence outside the two position in this case in the three position all have a higher affinity than the non 5 methyl cyanine all of the nucleotides in the system. So this is a really nice way, a unique way to use an existing technology to measure how all of the nucleotide RNA binding can be measured and quantified in live cell. Pretty exciting. There's you know, in summary, and it's a, it's a really interesting work, a novel approach for studying this type of complex interaction in cells. However, you know, you get very quantitative data. The workflow is very simple. You get one of the benefits always for bioluminescence resonance energy transfer readout is that the numbers we get are ratio metric, which make makes the data somewhat more robust. So I think this is a great kind of first step for using nano bread target engagement and in a unique way. And there's plenty of follow up studies to be done, including what are the best ways to continue to optimize this assay work in cells. This first, this first attempt generated some really nice data and we're expecting there to be even more interest in the system and we want to improve and help researchers improve the workflow and the data that you get out of of these types of studies and even additional studies. We're hoping these studies were hope, were hopeful for people to wanting to use this system for looking at small molecule binding to RNA and not just protein RNA binding. So a lot of interesting future potential for nano bread target engagement in RNA biology. Now as the last slide, I'd like to highlight Promega's tailored R&D solutions group and their capabilities to really help you get in the ecosystem or generate data using any of the assays that I mentioned or any of our other assays for that manner. It's a, it's a group that really is enables you to kind of use us as an extension of your research lab to do any number of services including making custom vectors, developing custom cell based assays as well as lumic immuno assays. We also have significant cell manufacturing capabilities for developing fawn yourself for assays and we'll actually bring in your compounds as well if you're interested in using any of our existing assays to screen with your compound of interest. So if you have any interest in using our tailored RnB solutions group, we'd be happy to have a follow up discussion with you on that. So with that, I'd like to thank you for your attention. I'll be looking forward to answering your questions in the Q&A session. Thanks. Thank you, Brad for the great talk and highlighting some of our newer technologies that can be applied to this space. And now we're going to move into our live panel discussion. A reminder that you can continue to answer or to insert questions into your Ask a Question box or topics you're curious to learn more about and we can discuss them during the panel discussion. OK. We have a few questions here, some philosophical ones and future looking questions. It seems. So. The first one, RNA has traditionally been considered a challenging target for drug discovery due to its dynamic nature, structural complexity. We saw this a bit reflected in the poll questions as well. What do you see as the most significant challenges currently facing RNA targeted drug discovery and how are recent technical advancements helping to overcome these? Amanda, you want to start us off? Sure. I don't know what happened to Matt. Yeah, are. You here, Matt, do you want to go first? I'll. Figure out how to fix the video? You're fine. Just go. I'll figure it out. Oh, I mean, yeah. So you know about the dynamic. I mean, I guess my answer is like, let's do everything in cells and real systems and figure it out later. I mean, but I don't know if that's the the right approach either, right? Because I think they'll definitely be more stable structured elements. And then I think the question is, you know, will those be functional or will they need to be outfitted, you know, as a degrader or with some other functionality to be functional? I mean, I think, I don't know, I would say like compared to, you know, I would say when I started the field in 2013, you know, I think before people thought you couldn't get, you would never get a molecule other than like an aminoglycoside or a host. I to bind RNA, right, Matt? I mean, and now I think you see things can bind. It's just how do you get them to be functional? Matt, do you want to go? Yeah, I agree with that. I I. Think, you know, the way that I, I view it as we have to sort of compartmentalize maybe a riboswitch ligand that binds in a very evolved complicated structure that bacteria have where they don't have, you know, years of evolution that humans do to bind a pocket versus some of the other lower complexity or maybe higher complexity structures that are harder to detect. Yeah, I think I and I think the the, the problem is you also have to start to drop the, the potency has to get in an animal arranged so that there are things are more predictable so. Yeah. And a good handle on the biology to know what to assay, right? You know, I think the proclivity for mrnas or even micrnas is we know what to assay, right? You know, once we go further, we know what to assay. But like say it's an RNARNA interaction or you know some higher order structure that becomes challenging. Yeah, RNA binding proteins isn't there. I mean, there's so many, right? But we have very little idea of what many of them are doing, right? So there's a lot of space, a lot of biology there to explore as well. Oh, yeah, this is, I mean, I, I feel like we should encourage basic scientists who want to do biology. I mean, I think like this example of the chaser, right? You know, pick a RNA on RNA protein interaction and actually characterize it. Well, I mean, we're, we're kind of missing some of those those systems. Yeah. And Brad, I'm curious since you're you're a technology guy, is there any different technologies you spoke about a couple of them that you see really being helpful for this RNA field? Yeah, I would say from like the technology or assay provider perspective, we're always trying to enable researchers to do these studies in a cell based background instead of using biochemical methods. That's pretty can be very complicated at times. Obviously it's a whole nother huge level of complexity being able to do target engagement that you know, and out that assay being invented really demonstrated the quality of the data, the relevance of the data being better when you can do those binding studies in a cell versus with biochemical studies. Both methods definitely have their purpose kind of understanding when the throughput of is needed and might prefer or require a biochemical method versus maybe a little more biological rich readout and then trying to provide asset technologies that you know, make both of those easy and give you the most relevant data. I think, you know, the couple of things that I touched on are really nice starts for looking at, you know, these interactions in a more, you know, cellular relevance. And, you know, we're going to continue to work with researchers to enable that and hopefully continue to develop technologies that make that easier. Yeah, I thought it was really compelling. Amanda, your data you showed with a big difference between your biochemical and cell based results for that one. It might not be totally fair though, because we screened different libraries. I mean like, no, I mean, you know, if we did the app, I would love to do, you know, again, it'll come down to funding, right, to do the actually head to But you know, I, I do think there is a value to doing cell based acids. I mean, even if it's eliminating toxic molecules, right, that like downstream you don't want to work on it. And certainly you will find a lot of. But I but I. Think another count. Yeah. I, I think there has to be in, in terms of there's chemical probe discovery and then there's drug discovery, right? So for drug discovery, I think you have to quickly know if your compound occupies the target in cells and use a biochemical assay and sell your assays to know the mode of action. Because one of the big challenges in RNA targeted small molecules is that the structure activity relationships tend to be flat and which means you can't have subtle changes in the activity of the compounds, which could then be massive differences in the distribution, metabolism and pharmacokinetics properties of it. And so I think there has to be, you know, more of an effort on a cell based assay and then quickly figuring out early on, quickly figuring out what's a biochemical assay that can really drive drive the compounds. Now if if we're going to go through Ristaplan, I mean Ristaplan was a phenotypic screen by stabilizing RNA protein complex. I don't ever think that there was a huge investment made by PTC and Roche in that approved molecule in a biochemical assay. What they did do, which I found was interesting is they prioritize cell based assays, but in a way where they were reading out selectivity and on target activity by next generation sequencing. And so you know, the way that I view RNA is we have a lot of tools actually some of the tools to implement for RNA targeted drug discovery in terms of omics are easier than doing chemo proteomics, right. We can amplify our targets, we can do seek more cost effectively. And so a question is how, how do you integrate that and do not just RNA targeted drug discovery, but I think all drug discovery? I guess the challenge is we just can't get the level of precision like amino acid labeling that they can get. Yeah, nanopore is still developing, right? So I think, yeah. Yeah. So that that actually sags into another question I had, which is, you know, what do you wish you could do now that you can't? Like what are the technology advancements that you guys are looking for in this field? It looks like sounds like some of more cell based assays on being able to more directly and specifically label RNAs. Anything else? Where, Where, Where's your wish list for tools and technologies? Easy, easy binding assays. Doing binding assays for some of these compounds as a peanut, peanut is, is not trivial. I mean, there, there's a lot, yeah. I mean, having having as Amanda said, long reading and of course sequence thing for these targets where you could identify with single nucleotide resolution where compounds are binding. We think we have something like that also before I get murdered in this talk is we also need to have a sobering view of these structures that are predicted by chemical probing because they're still predicted structures. They're just predicted structures with chemical with chemical probing data on top of that structure. And so, you know, one of the things that would be very nice to have, I think in, in the RNA area and we, we've talked about this in the lab is if there was a reference to a line or a reference line where all the RNA isoforms were annotated, all the RNA structures were annotated. And with a sobering view of, you know, it's not every RNA has a defined structure. It's what regions are robustly folded and what are not. And then elaborating onto those structures, what are the RNA binding, protein capacity within those cells. And so I do think having a reference, reference system like that, I would prefer to be human for a variety of reasons, although bacterial ones would be easier. That could be really enabling because you would know what the on and off targets are for a molecule, and you would have some data to reliably get what the structure is. And then I'll get out. You'll be able to do a whole bunch of other things. I agree with that. I mean I think just in general, right like so for RNA protein interaction, I mean our verifications until we get better nanopore or other methods for detecting all of them there are low abundance, right. I mean, often the occupancy of those things are is low. And so, you know, in the way we do sequencing now, it's indirect, right? We're actually converting into cDNA and losing all that information. So I feel like, you know, that certainly would be very enabling and how that might affect structure. But even for the RNA binding proteins where we have Eclipse, which is just a piece, right, like a nucleotide piece, but you could find that anywhere and you know, where are the transcripts is actually folded where it would be recognized by an RNA binding protein versus, you know, where it's not accessible. I mean, that still hasn't been, you know, solved. All right, very interesting. And final, final question of the day, kind of a looking ahead, a future looking, how do you envision the field of RNA targeted drug discovery evolving and say like the next 5 to 10 years? Yeah. What are some of the key paradigm shifts you think we might see or that people are working towards? Look, I think it's pretty obvious that these molecules that affect splicing are going to continue to be put into clinic. The MIB splicing modulators by Remix Regenta, the Huntington splicing modulators by PTC mean those molecules also act in a somewhat related way to the C9 or F compound where they're basically binder and then they they include a premature termination codon. I don't know mechanistically how much of those individual targets have been elucidated, but it's very likely that they stabilize RNA protein complex to effects effects splicing. Those are very different outputs than you know, targeting a repeat expansion or targeting an RNA only. And so I think the question is going to be is, you know, with with those compounds going into the clinic and you know, God. We all should. Say a prayer. I'm not making sure you're religious, but you know, you know, hopefully they get approved for cancer patients and patients with Lunmington disease like they transform medicine for humans that have spinal muscular atrophy. The question is, can the RNA only targets, which you can just look at repeat expansions, Can we get a molecule that targets those? And I think to something that Amanda had said earlier is a lot of the biology for these things is pretty much nailed. So I think my atonic dystrophy, there's oligonucleotides that have very favorable phase three clinical. Or. Phase two emerging phase three clinical data on oligonucleotide antibody conjugates. But the problem is that a lot of these diseases are multi systemic, so they affect the brain, they affect the muscle. And so the requirement to treat the patient with one drug is going to necessitate using a small molecule. So, you know, the question is what are what are the targets that could be pushed to clinical success sooner rather than later for chemical biology? Who knows? That's a different ball of wax. I agree. I mean, even you know, I think following on the oligos, right, I mean, I think that's the risk of plan story, right. Nusunersen validated in some ways the mechanism, right? And you could, I mean, it was serendipitous that the small molecule did that. But are there other, you know, waste? And I think Matt, you've done that over the years, right? Try to integrate SIRNAS and small molecules, right, in terms of like target validation and where are RNAs that might be, I guess, higher priority in the absence of detailed, you know, structure and other types of information that we typically like to have for proteins. I need our army. Right, we need armies of people to do all this. But but no, it's going to get better. We'll have the RNA alpha fold. So OK, I get this asked all the time. So here's the difference between RNA and a protein. 90% of the folding free energy of an RNA is the helical base secondary structure and very little of it is the 3D structure. Their efforts like Atomic AI and others and even some of Baker work with Riju Das when Riju was his posts postdoc or graduate, I think he was postdoc. He's a Nacho. No, he was a graduate student. He was a Nacho. Tinoco postdoc is they're trying to better understand how to do an RNA alpha fold RNA. But the problem is the protein, the PDB is the protein databank. It's not the RNA databank. It will never be. And so that means that in order to do AI learning and the evolutionary covariation at scale that was able to pull off alpha fold is going to require a lot more structure than what we currently have. So we need more, as Amanda said, we need more basic science and a diversity of structures. But it could also be that RNA is so dynamic that there isn't going to be a diversity of structures. It's going to be a. BLOB. So there's a lot, a lot to. A lot to learn on this. So I get that question asked all the time. So and it's the stacking interactions for RNA that's really important and and we under represent them and and stocking procedures. So anyway. There was a chat. Question that came in, if you have time. OK, I lied. That was not the last question. We have one more question from the audience. If I have an RNA binding protein, what is the workflow to characterize the function of the protein in RNA? Like broadly characterized like what it does. Perhaps a new RNA binding protein? The workflow? How to go about figuring out the importance of it? The function of it? Knock it out, Knock it. I don't think there's any difference with protein biology. If you want to know that, if you want to know finds, I mean if you have an RNA that you have you know of interest in, I mean I think any of these assays you could do to know. Right. So, so one of the ways you can, you can do it is you can do this clip, which is Bob Darnell's cross linking and then it was evolved by Gene Yao and others to, to cross link or any protein complexes. Nova is the example of this which Bob Darnell found that Nova, which is a a brain associated protein that causes neurological defects. Don't the details of that is going to be wrong and it might even not be Nova, but nonetheless the function of that protein was determined because it could cross link that it found to near Exxon intron junctions and when it bound it actually affected splicing. So genes were found in one position, the Exxon was occluded, genes were found in another position, it was included. And so having the binding capacity might be able to give you clues if it's a premium RNA splicing regulator. If it binds a three prime UTR it could be a stabilization but God just use an siRNA and get rid of it and see what happens to your cells and then add it back and see if you get the opposite effect. All right. Well, I think that concludes day three of Undruggables. Thank you very much to our presenters. Really appreciate you guys joining today, sharing your expertise. Thank you to all our attendees as well for hanging with us today and we'll follow up with any questions. All of these presentations will be available on demand as well as days one and two, which you can see the agendas for now. And with that, thank you all again. Have a good rest of your day. Thank you. Thanks guys. _1734848777922

Time: 10am CDT / 8am PST / 4pm London / 5pm Amsterdam-Paris

"Illuminating New Frontiers: Cracking the Undruggable Code" invites you on a groundbreaking journey across the evolving landscape of drug discovery. This three-day virtual event unites trailblazers in science and industry to explore how innovative approaches are reshaping our ability to tackle "undruggable" targets. From the complexities of the RAS pathway and the intricacies of protein-protein interactions to the pioneering fields of targeted protein degradation/induced proximity and RNA targeting, each day illuminates a different frontier of therapeutic potential.

We stand at the cusp of a new era, where challenges become opportunities for breakthroughs that extend beyond conventional boundaries. This event is a testament to the power of collaboration and innovation in unveiling new therapeutic strategies, offering hope for untreatable conditions. Join us as we navigate the cutting-edge of science, inspiring a future where every target is within our reach, and no one is left without options.

Day 3:Unlocking the Potential of RNA: New Paths for Targeting RNA

This session delves into the innovative realm of RNA as a key therapeutic target in modern drug discovery. Long considered a challenging prospect due to its dynamic nature and critical role in gene regulation, RNA offers unique opportunities for intervention. We will explore cutting-edge advancements in directly targeting RNA molecules and RNA-protein interactions, which are opening avenues in treatment strategies. Discover how novel RNA-targeting techniques are transforming our ability to influence genetic and cellular functions, presenting potential cures for previously untreatable conditions. Join the conversation and engage directly with the experts in a live round-table discussion following the presentations.

_1734848778106