Gamma-PNA triplet ligands with Janus bases selectively bind expanded rCUG repeats, displace MBNL1, and show length-dependent cooperativity with partial splicing rescue.
0:00Welcome to Base by Base, the papercast that brings genomics to you, wherever you are. Thanks for listening, and don't forget to follow and rate us in your podcast app. It's great to be here So today, I want to start us off with a concept that, um, it feels a little bit like something out of a sci-fi horror movie.
0:15But it's happening at a microscopic level. Right inside the nucleus of a cell. I want you to imagine something called molecular clutter. Clutter is maybe a polite way to put it Okay, fair enough. Let's call it a disaster zone.
0:30So imagine a genetic error that doesn't just, you know, stop a protein from being made, which is bad enough. But it actually creates a toxic molecule. And this molecule is essentially a trap. It just sits there and actively grabs other essential machinery in the cell.
0:46It's a traffic jam. Exactly. A traffic jam that doesn't just block one street, but it spreads and paralyzes the entire city grid. That is a very, very accurate description of what we call a toxic gain of function.
0:59Gain function. Yeah. Usually in genetics, we worry about a gene breaking and just doing nothing, a loss of function. But here, the gene breaks and creates something actively harmful. It's not just a broken part.
1:12It's a saboteur. And in the case we're talking about today, that harmful thing is a rogue form of RNA that just refuses to behave. Right. And that specific problem, I mean, it's driven scientists crazy for decades because targeting RNA is a complete nightmare.
1:27Why is that? It's shapeshifting, it's slippery, it's fragile. And usually, if you want to drug it, you have 2 pretty bad choices. Let's say choice. You use small molecules, which are basically chemical buckshot.
1:40They can get into the cell easily. theyre tiny, but they hit everything. They often lack the precision to tell the difference between good RNA and bad RNA. So you might hit your target, but you'll break a lot of windows in the process.
1:50You will? Exactly. And choice B. Choice B, you use these massive synthetic strands. They're called anti-sense oligonucleotides. And these are like precision missiles. But they are huge. They often can't fit through the bunker door to get into the cell.
2:04And even if they do, they have to physically wrestle the RNA structure to do their job. It's uh, a brute force approach. So you're stuck. You got low precision or low entry. But here is the provocative question for this deep dive.
2:18What if you didn't have to choose? Go on. What if you could engineer a molecule that acts like a tiny, agile chemical, but reads the genetic code with the absolute precision of a DNA strand? Okay. And what if, instead of trying to, you know, untangle this genetic knot with brute force, you just filled in the gaps.
2:40Now that is the key phrase there. We're talking about a completely new paradigm in how we interact with genetic material. It's not about breaking the structure. about completing it. Yes. Today we are looking at a paper that claims to have done exactly that.
2:51It's titled A pothole filling strategy for selective targeting of RCUG repeats associated with myotonic dystrophy type one. And we really should recognize the team behind this. This is a fantastic work, led by Jay Denithi R. Pereira and Danith H.
3:08Lye at Carnegie Mellon University. Yes, and it's a massive collaboration. They worked with people from Eiser Pune in India, Georgia State, and Nan Yang Technological University in Singapore. It was published in PNAS, the Proceedings of the National Academy of Sciences, in January 2026.
3:24And this isn't just a paper about you know, a new drug candidate. It's a celebration of a new class of nucleic acid legends. They basically solve a decades old engineering problem. They have, by looking at the geometry of the disease.
3:37So let's unpack this. Before we get to the solution, which involves something very cool called Janus bases. We need to understand the villain. Right. We are talking about myotonic dystrophyte type one or DM one.
3:48To give some context, DM1 is the most common form of adult onset muscular dystrophy. It affects about one in 8000 people globally. And the symptoms are pretty tough. They are. Progressive muscle wasting, cognitive impairment and something called myotonia.
4:03Which is where the name comes from. Exactly. Myotonia is a condition where your muscles can't relax after they contract. So, like, gripping a doorknob and then just not being able to let go. Precisely that.
4:13Or imagine shaking someone's hand and being physically unable to release your grip for several seconds. It's debilitating. And the cause is genetic. It is. It's in a gene called DMP pay. Now, in you and me, inside that gene, there is a sequence of letters, C T G.
4:30We might have, say, 5 repeats, maybe up to 35? That's perfectly normal. But in a DM one patient, that number just explodes. It expands massively. You can go to 80, sometimes up to 5000 repeats. 5000 repeats of the same 3 letters.
4:44That seems like an impossible amount of extra code. It is huge. And when that DNA is transcribed into RNA, you get these massive strands of CUG repeats. And this is where the physics comes in. Exactly.
4:54These long strands don't just float freely like normal Messenger RNA, because CNG love to bind to each other, these strands, they fold over on themselves. creating those tight hairpin like structures. So instead of a long string, it's a tangled knot.
5:08And these hairpins are the sponge we talked about earlier. They act like a magnet for a specific protein called MBNL one or muscle blind like protein one. So MBNL one gets stuck to the RNA hairpin. It does.
5:20And ambino one has a day job, right? It's supposed to be off doing something else. It has a very important day job. It's a splicing regulator. Think of it like a film editor. It's job is to cut and paste different parts of RNA to make functional proteins for your muscles, your heart.
5:35But if MB&L1 is stuck to this toxic RNA. The film editor is on strike. Or locked in the basement. Locked in the basement. And the cell starts producing uncut or wrongly cut proteins. And that splicing failure is what causes the muscle wasting, the heart issues.
5:49Okay, so the therapeutic goal seems obvious. Free the MBNL one. Get it off that RNA sponge. Sounds simple, but remember that choice A versus choice B problem. Right, the buckshot or the giant missile. Small molecules they get in, but they're just not specific.
6:03They hit good RNA, bad RNA. Because there's so much RNA in a cell. You target something vague, you're gonna hit innocent bystanders. Precisely. And on the other side, they anti-sense alliga nucleotides or ASOs.
6:17I mean, groups like Thornton have done amazing work with them. ASOs are very specific. But they're huge molecules. They are. So they have trouble getting into the tissues. They have terrible pharmacokinetics, and there's a physics problem.
6:30That toxic RNA is folded into that tight hairpin we mentioned. For an ASO to bind, it literally has to force that hair pin open. That sounds exhausting for the molecule. It's what we call a thermodynamic penalty.
6:43It takes energy to rip that structure apart. So you need a high concentration of the drug to make it work, and that can lead to toxicity. So we needed something in the Goldilocks, though. Small, permeable, but highly specific.
6:55Enter the genus bases. This is the core innovation of the Pereira and light paper. When I read Janus, I immediately thought of Roman mythology. That's the perfect connection, Janus, the Roman god of beginnings, transition, usually shown with 2 faces looking in opposite directions.
7:11And that's what these molecules do. have 2 faces. They are bifacial. Normal DNA or RNA bases, they bind on one side, the Watson Crick face. The standard ATCG pairing we all learned. Right. But these synthetic bases can bind on both sides at the same time.
7:28They have a Wes and Crickface, but also a 2nd phase that can interact with other bases. Okay, so they're like double-sided tape, but what are they attached to? Because usually DNA bases are on a sugar phosphate backbone.
7:41And that backbone is negatively charged, which is a problem because the RNA target is also negatively charged. So they repel each other, like magnets. The drug has to fight just to get close. Exactly. So this team use a different backbone.
7:53It's called gamma PNA, or peptide nucleic acid. We've talked about PNA before. synthetic. It is. The key thing is that it's electrically neutral, no negative charge. So it doesn't fight the RNA. It mimics it, but without that repulsion.
8:06It's also preorganized into a helix shape, so it's ready to go. So we have a neutral backbone and these two-faced Janus spaces. How does this solve the hairpin problem? You mentioned pothole filling. This is the brilliant part.
8:17Remember those toxic RNA hair pins? They aren't perfect ladders. Because it's a CUG repeat folding on itself. The C pairs with G, which is very strong. But then you have a U facing another U. And you and you don't usually pair up.
8:31Correct. They repel each other slightly. It creates a wobble or a gap in the structure. A pothole. Exactly. molecular pothole. The ASOs, they try to rip the road up and pave it new. This new ligin called LG 2B doesn't destroy the road.
8:45Just slides in. It slides into the hairpin structure. It identifies those specific mismatches, the potholes, and it just fills them. That's so much smarter than brute force. It's like using a key that fits the lock exactly as it is instead of trying to break the door down.
8:59And the ligand is tiny. LG2D is only 3 units long. It effectively sandwiches itself into the RNA structure using those Janus spaces to grab the U on one side and the U on the other. So they designed this on a computer first, I assume?
9:13Oh yeah, they used molecular dynamics simulations. And the modeling showed something really fascinating. The ligen binds to the Watson face of one RNA strand and the crick face of the other simultaneously.
9:26It creates the super stable complex. But does it work in reality? I mean, we've all seen perfect drugs fail when they hit actual wet chemistry. The lab data here is incredibly compelling. The most cribble finding was about something called cooperativity.
9:40That sounds like a corporate buzzword. Huh. In chemistry, it's magical. Think about a zipper on a jacket. Getting that 1st tooth to engage is the hardest part, right? you fumble with it. But once that 1st one is in, the 2nd one clicks in easier, and the 3rd is even easier still.
9:56That's cooperativity. So one leggin binds to a pothole, and that changes the shape of the RNA, just enough to welcome the next legend in. Exactly. Binding induces a structural change that makes the neighboring sites more accessible.
10:10And this led to the biggest win of the paper, selectivity. The holy grail. How do you hit the sick gene, but leave the healthy gene alone. Because if you target all CUG repeats, you're gonna shut down genes we actually need.
10:22Remember, DM1 is caused by long repeats. But we all have short repeats in that gene. You don't want to drug the healthy short version. And this study shows. It showed that this ligand LG2B. It essentially ignored the short, healthy repeats.
10:37It just floated on by. Pretty much. Because the run of repeats was too short. It couldn't get that zipper momentum going. The stability wasn't there. But on the long pathogenic repeats, the RCUG 98. That's where it stuck.
10:50It bound tightly. The zipper could run all the way up. That is so clever. It uses the length of the defect against itself. And they compare this to the old tech, the ASO. They did. They compared it to a Morpholino ASO called Sag 25. And the ASO was a blunt instrument.
11:06It bound everything, the healthy RNA, the sick RNA, equally. It couldn't tell the difference. It couldn't, but LG to be could. That is a massive safety damage. If you can ignore the healthy genes, you reduce side effects so much.
11:18Absolutely. And they didn't just infer this from running gels. They actually took pictures. They used atomic force microscopy or AFM. Oh, I love AFM. It's like feeling the surface of the molecule with a tiny needle.
11:29That's a great way to put it. They literally measured the height of the RNA. And when the ligen bound, the RNA height increased by exactly .348 nanometers per base pair. That is an absurdly precise number.
11:42Why does that specific number matter? Because .34 nanometers is roughly the distance between stacked base pairs and a helix, the rise of the ladder. It proves the ligan inserted itself inside the helix stack.
11:54It wasn't just stuck to the outside. It wasn't just stuck on like a piece of tape. It filled the pothole and became part of the ladder. It physically proved the mechanism. Okay, so it binds, it fills the pothole is selective.
12:05Yeah. But does it actually free the hostage? Does it kick MBNL1 off the RNA? It does. They ran displacement tests. And when LG2B binds to that toxic RNA, it occupies the space so thoroughly that MBNL1 simply falls off.
12:18Or if you add the leg in first, MBNO1 can't bind at all. It worked in displacement, prevention, and competition modes. So the trap is disarmed, the film editor is free. But this was all in test tubes so far.
12:32What happened when they put it in real cells? And this is the discussion phase where reality always hits. LG2B is great in a tube, but getting it into a cell nucleus is hard. The cell membrane is a fortress.
12:43So they made a variant called LG2C. LG2C. What's the difference? They added a tail? Specifically, a tale of Argentine amino acids and a disulfide bond. Think of it as a VIP pass to get through the cell membrane and into the nuclears.
12:57And they tested this on patient cells. Yes, Mayartudes derived from DM1 patients. These cells were sick. They had all the splicing errors we talked about. And what were the results? It worked. They looked at 3 specific genes that always go wrong.
13:09Circo one, which controls muscle relaxation. Okay. CTNT, which is crucial for heart function, and the insulin receptor. The big three. Fixing these isn't just cosmetic. It's fixing the root cause of the symptoms.
13:22That's it. Circle one is the calcium pump. If it's not spliced right, you get the myotonia, CTNT affects the heartbeat, the insulin receptor. Well, that's obvious. And the splicing was restored. In all three, yes.
13:35For circo one, the like and improved splicing significantly. It was comparable to existing small molecules, but remember, this leggin has that huge theoretical advantage of being way more specific. Did they see the RNA clums disappear?
13:50The foci? They did. In DM1 cells, you can see these bright spots in the nucleus where the toxic RNA is clumping up. After treatment with LG2C, those focide just vanished, the clutter was cleared. That is incredible.
14:03But I'm guessing there's a butt. There is always a but in science. The limitation is uptake. Getting the drug into the cell is still the hardest part. The LG2C modification helped, but it requires pretty high concentrations to get enough inside.
14:16High concentrations usually mean a higher risk of toxicity. Exactly. And while the PNA backdone is generally safe, you always want to use as little drug as possible. So it needs optimization before it's ready for humans, they need a better Uber to drive the passenger into the cell.
14:29So we aren't going to the pharmacy for this tomorrow. No, but the proof of principle is there. And the implications, I mean, they go way beyond my atonic dystrophy. Right, because DM1 isn't the only disease caused by these repeating stutters in the code.
14:43Not by a long shot. Huntington's disease, some forms of ALS, several of the Ataxias, they all involve similar repeat expansions. And presumably, those repeats also form unique structures. They do. They form their own types of hairpins and their own potholes.
14:59This Janus base strategy, this whole pothole filling technique, could theoretically be tuned to target any of them. You just change the faith of the Janus base to match the specific mismatch of that disease.
15:13Exactly. So you're looking at a platform, not just a single product. We're looking at a shift in philosophy. We really are. For so long, genomic medicine was about reading the sequence, the ACTG. We treated it like a linear code.
15:25This paper is saying, look at the shape. The 3D structure. Yes. The defect creates a shape, a pothole. And if you target the shape, you can be much more selective than if you just target the letters. It's structural biology meeting pharmacology.
15:38Which brings us to our take home message. We've moved beyond that binary choice of dumb small molecule or clumsy anti-sense giant. We can now design synthetic genetic keys. Keys that are small enough to be drugs, but smart enough to read the code by using these bifacial bases that recognize the 3D structure, the potholes, we can target the disease without hurting the healthy genes.
16:01It's a level of precision we honestly didn't think was possible 10 years ago. It's engineering at the nanometer scale. Yeah. And it leaves us with a really provocative thought to chew on. It does. We usually think of genetic defects as just broken code.
16:14But this research turns that on its head. It suggests the defect itself. The shape of the brokenness is actually a unique fingerprint. Right. So the prompt for you, the listener, is this. If we can target the shape of RNA defects rather than just their sequence, what other undruggable genetic structures might act as open doors for this new class of medicine, what other molecular potholes are out there, just waiting to be filled?
16:40That's a whole new map to explore. This episode was based on an open access article under the CCBY 4 license. You can find a direct link to the paper and the license in our episode description. If you enjoyed this, follow or subscribe in your podcast app and leave a 5 star rating.
16:55If you'd like to support our work, use the donation link in the description. Now, stay with us for an original track created, especially for this episode, and inspired by the article you've just heard about.
17:04Thanks for listening and join us next time as we explore more science based by base.
