Review shows how alternative splicing, via TE exonization and cis-regulatory changes and revealed by long-read RNA-seq, reshapes gene regulation and drives phenotypic evolution in mammals.
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 is great to be here I want to start today by asking you to close your eyes for a second.
0:12Well, maybe not if you're driving or operating heavy machinery, but, you know, use your imagination. I want you to picture the silhouette of a monkey. Any monkey. A macaque, a baboon, maybe a little squirrel monkey running along a branch.
0:27Got it I've got it. I'm picturing a macaque. Long tail for balance, very distinct profile. Right. Now, right next to it, picture the silhouette of a human, or even a chimp, or a gorilla. What is the single most obvious glaring difference between those 2 outlines?
0:42Hmm. Well, aside from the posture, you know, standing upright versus being on all fours. has to be the tail. Exactly the tail. Monkeys have them, we don't. Hominoids, that's humans and apes. Lost them about 25000000 years ago.
0:55A long, long time ago. And for a long time, this was just viewed as a cosmetic change. You know, we came down from the trees or we changed how we moved. We just lost the tail. Right, like it just fell off one day.
1:07An evolutionary afterthought. But when you think about the biology of that, it's actually insane. It really is. It's not just a cosmetic difference. It's a fundamental alteration of the skeletal structure.
1:17I mean, you're talking about changing the number of caudal vertebrae, completely reworking the musculature and rewiring the nervous system pathways. It's not just a rope attached to your back. Not at all.
1:29It's an extension of your spine. To lose a tail requires a massive rewrite of the developmental program in the embryo. You don't just drop a limb, you have to stop it from ever starting. Right. And here is where the mystery has been driving scientists crazy for decades.
1:45Because logic and classical genetics would tell you, if you lose a whole body part, you must have lost the genes for it. That's the intuitive leap. Or you must have had some massive mutation that just, you know, broke the tail gene beyond repair.
1:59That was the prevailing assumption for a very long time. The architect model of evolution. If you want a different building, you change the blueprints. You delete a section. So for a long time, geneticists were hunting for these lost genes.
2:13Exactly. They were comparing the human genome. to the monkey genome, looking for big deletions or, you know, broken instructions. They thought, surely there is a gene that says build tail, and we just deleted it.
2:25But they didn't find them. No, and that's the kicker. The gene that builds the tail, specifically a gene called TBXT, is still there. It's still in our DNA right now. It is largely intact. We have the blueprint, we have the bricks.
2:40just don't build that particular extension on the house anymore. That is the hook for today. How do you have the blueprint for a tail? Sit on it for 25000000 years and never build a tale. It turns out the difference isn't in the blueprint itself, the DNA sequence, but in how the instructions are read.
2:56Precisely. And the culprit or the hero, depending on how much you miss having a tail, is what we used to call junk DNA. Chunk DNA. specifically a jumping gene that inserted itself into a developmental gene.
3:07So a piece of genomic junk jumped into a vital gene and convinced the cellular machinery to just skip a page. That is exactly what happened. It essentially edited the instruction manual without changing the original text, just by changing which paragraphs get read.
3:22And that brings us to the central theme of our deep dive today. It really does. Our complexity as humans, our brains, our hearts, our immune systems, doesn't come from having more genes than a worm or a fly.
3:33Not at all. It comes from our ability to edit the genes we have. Yeah. in wildly complex ways. I love that concept. It's not about how much stuff you have. It's about how you use it. It shifts the view of evolution from an architect who builds new rooms.
3:48As an editor who rearranges the scenes to make a better movie. That's the perfect analogy. And we are looking at the masterclass and editing today. Okay, before we unravel how we lost our tales and how we built our brains.
3:58We need to give credit where credit is due. Absolutely. Today we are doing a deep dive into a review article published in the EBO Journal. It's titled The Splice of Life, how alternative splicing shapes regulatory and phenotypic evolution.
4:13And the authors. Carissa Emerson Hunter and Yishing from the Children's Hospital of Philadelphia and the University of Pennsylvania. And they've done a lot of heavy lifting here. This isn't just one. Oh, no.
4:23They've synthesized decades of research to show that this mechanism, alternative splicing, isn't just a side effect of biology. It is a primary engine of evolutionary innovation. And in doing so, they've really helped to address one of the biggest embarrassments in the history of genomics.
4:39Oh, yes. The C value enigma. Which sounds like a Robert Ludlum novel, but I gather it's a scientific problem. It is a historical puzzle that really humbled the scientific community. Back in the day before we sequenced the human genome, so pre 2000s, there was this, this very confident assumption.
4:59Which was? that organismal complexity must correlate with genome size. Basically, if you are a complex creature like a human, capable of language, math, and, well, building podcasts. Then you must have a massive number of genes.
5:12A huge number. And if you are a simple creature like a roundworm, you must have very few. It makes intuitive sense. I mean, a big complex brain needs more instructions than a microscopic worm. You'd expect the manual for a space shell to be thicker than the manual for a toaster.
5:26Right. And the estimates for human genes were all over the place. I've seen papers suggesting 100,000, maybe more. Scientists replacing bets. Literally. There were betting pools. But then we actually sequenced the genomes, and it was a shock, a real profound shock to the system.
5:42What did we find? It turns out humans have roughly 19,000 protein coating genes. 19,000? That sounds low. It felt very low. And then to really rub it in, we sequence the worm. And the worm. The nematode worm canerhabitis elegance, which is about a millimeter long and has exactly 959 cells, also has about 19,000 to 20,000 genes.
6:06You're kidding. I am not. That is underwhelming. That actually kind of insulting. It is very humbling. We are working with the same size parts list as a microscopic worm that lives in the dirt. So the question became, how do you get a human brain, a sophisticated immune system, and a beating 4 chambered heart from a worm sized parts list?
6:24That is the C value enigma. That's it. And the answer isn't more genes. No. The answer lies in regulation. Exactly. Evolution didn't just add new jeans. It changed when and where those genes are used. We've known about this for a while.
6:38I mean, think about Darwin's finches. The birds from the Galapagos with a different beak shapes. Yes. Their beaks change shape not because they got new beak genes, but because of changes in the expression levels of a gene called BMP4.
6:51So evolution just turned the volume knob up or down? It turned the volume up or down on that gene, and the beak got thicker or thinner. It's a beautiful example of regulatory evolution. Okay, so that's volume control.
7:05Gene expression. That explains some things. But volume control doesn't explain how you get a totally new function. Right. Turning up the volume on a radio doesn't turn it into a TV. And it doesn't explain how you lose a tail while keeping the gene intact.
7:18Exactly. Volume control gene expression is only one layer. It tells you how much protein to make. But to solve the enigma of complexity, we need to look at alternative splicing, or AS, this allows one gene to produce multiple different MRNA isoforms, it allows you to change what protein you are making, not just how much.
7:37Isoforms. That's a key word today. Let's break down the mechanism for everyone, because if we don't get the mechanics right, the rest of this won't land. Good idea. How does this actually work at the molecular level?
7:48Okay, so imagine a gene as a long string of DNA. But it's not a continuous sentence. It's interrupted. It has Exxons, which are the coding parts, the actual instructions for the protein. And introns. And introns, which are noncoding spacers in between.
8:04Like commercials and a TV show. That's a decent analogy. The introns are the commercials. When the cell copies the DNA into RNA. It makes a pre-MRNA that includes everything, the show and the commercials.
8:15The raw footage, exactly. Splicing is the process of physically cutting out the commercials, the introns, and pasting the show segments, the exxons together to make the final message. Okay, that's standard splicing.
8:28The cell cleans up the tape. It removes the junk and keeps the story. What is alternative splicing? Alternative splicing is when the cell decides to edit the show differently. It gets creative. Maybe in one version, you keep scenes one, two, and three.
8:42The standard version. Right. But in another version, you keep scene one, skip scene 2 entirely and jump straight to scene three. So you create a shorter version of the movie. A shorter and maybe functionally different version.
8:53Or maybe you retain a commercial break and treat it like part of the show that's called intron retention, or maybe you only use the 1st half of scene two. There are all these different ways to mix and match.
9:03So from one original recording, one gene, you get 2 or 3 or 10 different movies. Precisely. The best analogy I've heard is a Lego set. A gene isn't a pre-built model. It's a box of bricks. If you use all the bricks, maybe you build a car.
9:17The default model on the box. Exactly. But if you skip the wheels and use the wing pieces instead, that same box builds a plane. If you use the whole pieces, it builds a boat, one gene, multiple final product.
9:30And these final products, the car, the plane, the boat, those are the ISO forms. Yes. And in humans, this is we've been. I mean, it's not a rare thing. Almost all multi-exon human genes undergo alternative splicing.
9:44All of them. Pretty much. It creates this incredibly versatile toolkit. It allows a single gene to do one thing in the brain, something else in the liver, and something totally different in the heart, just by swapping out Exxons.
9:56That is just, it's incredibly efficient. It solves the space problem. Completely. You don't need a separate gene for every single function. You just need a modular one. You just need a gene with a lot of optional parts.
10:06But who is making these decisions? Who's the editor in the cutting room? It's a mix of cis and trans elements. And this distinction is crucial for understanding how evolution works. Okay, definition's time.
10:19Let's start with sis. What does that mean? Sis refers to the regulatory elements on the RNA itself. Think of these as road signs or traffic lights embedded in the sequence. There are little codes that say, cut here, start here, or maybe more subtle sign like slow down, splicing side ahead.
10:36So they're a part of the molecule being spliced. They're part of the script. The script has notes in the margins. Delete this paragraph Keep this sentence. Okay, that makes sense. So what are trans elements?
10:46Trans refers to the factors that come from elsewhere to read those notes. These are proteins called splicing factors. They are part of a big machine called the Spleesusum. They're the drivers of the car.
10:59They are the drivers or the editors who see the sign that says stop and hit the brakes. They float around the nucleus looking for their specific signs to bind to. So if you want to change the movie. You have 2 options.
11:12You can either change the notes in the script. That's a cist change, mutation in the DNA. Or you can change the editor who is reading them. A transchange. Evolve the protein machinery itself. Okay, and evolution uses both.
11:24It uses both strategies, but as we'll see, it really loves tinkering with the notes in the script. Why is that? It's just, it's easier. It's safer. If you change the editor, you might mess up the editing for every movie in the studio, but if you change the note on one script, you only change that one movie, it's more targeted.
11:41Before we get to the specific examples, which are mind blowing, by the way. I want to ask about how we even know this because for a long time, we were just guessing, right? We were to some extent. We knew splicing happened, but we couldn't see it in detail.
11:54We were limited by our tools. This is the technological evolution part of our story. Yes. For about 15 years, the gold standard was short red RNA sequencing, mostly on alumina platforms. This is what most people think of when they hear genome sequencing.
12:10High throughput. very accurate, widely used. Right. It's incredibly accurate, but it reads short chunks. 50 to 200 base pairs. Why is that a problem for splicing? If you have the sequence, you have the sequence.
12:22Imagine you take a book or gene and you shred it into confetti. Okay. You can read every individual piece of confetti perfectly. You can see the word the, and the word cat. But if you're trying to figure out if the sentence was, the cat's out on the mat, or the cat ate the rat, and those words are on different pieces of confetti.
12:40You have a hard time reconstructing the full sentence. A very hard time. You lose the connectivity. You can't tell which Exxons were linked together. You know X on A is in the mix, and X on B's in the mix, but are they in the same molecule, in the same ISO form?
12:53You can't see the long range coupling, as the paper calls it. Exactly. You might infer it statistically, but you can't prove it. You can see the ingredients, but not the final recipe. So what changed? What was the breakthrough?
13:05Long read sequencing. Technologies like Oxford, Nanopore, and Pacific Biosciences. These machines don't shred the book. What did they do? They thread the DNA or RNA through a tiny pore, literally a nanopore, and read the whole chapter at once.
13:19They can generate reeds spanning more than 10,000 base pairs. So instead of confetti, you get the whole scroll. Yes. You capture the full length isoform. You can see exactly which Exxons are included and which are skipped in a single molecule.
13:32You can see if Exxon one is connected to Exxon 4, or if it skips all the way to Exxon 7. It's been a revolution for understanding complexity. A total game changer. And I understand we're also moving from smoothies to fruit salads.
13:45Yes, that's the single cell revolution. Traditional sequencing was bulk. You take a piece of brain tissue. You grind it up into a smoothie and you sequence it. So you get the average of everything. The average.
13:57But a brain has neurons, glial cells, blood vessels, all sorts of different cells. If the neuron is splicing one way and the blood vessel another, the smoothie just shows you a muddy, averaged out mix, you lose the specific identity.
14:12Single cell sequencing, lets you pick out the strawberry and the grape separately. Exactly. We can now see if a neuron splices a gene differently than a skin cell. And now we're combining long reeds with single cell tech.
14:26The best of both worlds. It's giving us an unprecedented view of this Lego building process in action. We can see exactly which cell is building the car and which is building the plane. And just to be thorough, the paper mentions they're using proteomex to check their work.
14:40Why is that important? Yes, because just making the RNA doesn't mean it does anything. The cell makes a lot of trash, a lot of transcriptional noise. So you want to see if the blueprint actually gets built.
14:51Precisely. Proteomics using mass spectrometry confirms that these weird alternative RNA versions actually get translated into real functional proteins. It proves they aren't just mistakes. They're the finished product.
15:04Okay, so we have the tools when we have the concept. Let's talk about what we've found. What are the main ways evolution uses this splicing toolkit to change us? The review outlines 3 main modes. The 1st is Exxon Creation or Exxonization.
15:18Exonization. This is basically turning junk into gold. Evolution takes a piece of non-coding intron, often a transposable element or jumping gene, and mutates it just enough so the cellular machinery recognizes it as an Exxon.
15:32So a piece of gibberish in the middle of the text suddenly becomes a new, meaningful paragraph in the story. Exactly. And because it was junk before, it's a great source of raw material. It wasn't doing anything important, so you can mess with it without breaking the organism.
15:44You can test out new ideas. You can try out new protein sequences without destroying the old ones. It's a very low risk way to innovate. Okay, so that's mode one. What's two? The 2nd mode is Xon loss. This is the opposite.
15:57This is what happened with our tales. A constituent of Exxon one that was always there, always part of the final product gets deleted or more often skipped. The cell just starts ignoring it. Yes, and the third.
16:07The 3rd mode is splicing level changes. This is more subtle. The Exxon is still there, but the frequency changes. Maybe in a mouse, an Exxon is included 100% of the time. It's constitutive. But in a human, maybe it's only included 50% of the time.
16:23It becomes alternative. That subtle shift can have massive effects. It's about tuning the ratio, not just an on-off switch. I want to go back to that Assis versus Trans thing for a second. You mentioned a study involving a mouse and a human chromosome that settled a big debate.
16:37This seems important for understanding who drives evolution. Ah, the TC one mouse study. This is a classic experiment. It really gets at the heart of the question. Is the difference between a human and a mouse due to the machinery, the trans factors, or the genetic instructions, the cis elements?
16:53So they put a human chromosome inside a mouse? They created a mouse strain that carried human chromosome 21. It's a humanized mouse. That sounds like something out of science fiction. But useful. Incredibly useful.
17:06It allows for a perfect experiment. You have human DNA floating inside a mouse cell nucleus, surrounded by mouse splicing factors. The question is, will the human genes on that chromosome splice, like they do in a human, or will the mouse machinery force them to splice, like a mouse?
17:25It's a showdown. The mouse machinery is the trans driver. The human DNA has the cyst signs. Who wins? The signs win, the cis elements. The human exxons inside that mouse spliced almost exactly like human exxons.
17:38Wow. So the instructions are baked into the sequence itself. The contractor follows the blueprint no matter who the contractor is. Predominantly, yes. This implies that evolutionary changes in splicing are mostly driven by mutations in the sequence itself incis.
17:51Which is safer, like we said. Exactly. It's great news for evolution because again, changing the machinery is dangerous. Changing one blueprint is specific and targeted. But the machinery does evolve too, right?
18:01It can't be totally static for 1000000s of years. It does, but more slowly and often in interesting ways. Splicing factors like Novia or ESRPs are ancient. They are conserved from sea urgins to humans.
18:15But what changes is where they are used. Like repurposing a tool for a new project. Exactly. In a sea urchin, the Nodier factor is used in the gut, in a fruit fly, it's in the salivary gland. In humans, it's absolutely critical for the brain.
18:29Same tool, different job site. And sometimes the tool itself gets an upgrade? Yes. There's a fascinating case with factors called SRRM 3 and SRM 4. They acquired a specific domain called e-make early invertebrate evolution.
18:43This upgrade allowed them to recognize tiny, tiny exxons called microexxons. How small is a microxon? Sometimes just 3 to 15 nucleotides. That's one to 5 amino acids, a tiny glip. Tiny blip. But these microexons are critical for neurodevelopment.
18:58They fine tuned protein interactions in the synapse. Without that tool upgrade, our brains wouldn't wire up correctly. It's amazing that something so small can be so important. This leads into the playground hypothesis mentioned in the paper.
19:10I really like this idea because it explains how evolution can take risks. It's a beautiful concept. Evolution needs a way to experiment. If you mutate a vital gene, the organism usually dies, that's bad for evolution.
19:23Right, a dead end. But alternative splicing offers a safe space or a playground. Because you keep the original version. Exactly. The gene can keep making its main essential isoform, the car that runs perfectly, and at the same time produce a new alternative isoform, the plane, at very low levels.
19:41A little side project. Right. If the plane crashes, it's fine. You still have the car, but if the plane flies, then evolution can select for it and ramp up production over generations. It's like beta testing software, you don't delete the old operating system, you just run the new one on a few computers to see if it works.
19:57Precisely. And this is why that so-called junk DNA is so important. It provides the raw code for these beta tests. Okay, let's get to the main event. the showstoppers. These are the specific examples from the paper that prove splicing isn't just noise.
20:11It changes bodies, organs, entire species. ready. We started with a tail. Let's dive deep on that because the mechanism is wilder than I expected. The tail loss. This was a paper by Shia and colleagues, it really shook things up.
20:25We talked about the Gene TBXT. This is a crucial developmental gene. In monkeys, this gene builds a tail. Correct. Now, in the ancestor of humans and apes, something happened. A retro transposen, specifically an Alu Y element inserted itself into an intron of the TBXT gene.
20:43Okay, retransposen is a jumping gene. It copies and pastes itself throughout the genome. Primey genomes are full of them. About 10% of our genome is just alu elements. Usually they land in an intron and do nothing.
20:56They are just ignored commercials. But this one was different. This time there was a twist, literally. There was another Alu element, an older one in AluS, already sitting in the intron on the other side of Exxon 6, and it was inverted, meaning its sequence was backwards compared to the new one.
21:10So you have 2 matching sequences facing each other, bracketing this one particular Exon. And because their sequences are complementary, they acted like Velcro. The RNA strand folded over and the 2 alu elements stuck together.
21:22forming a very stable stem loop structure, a hairpin. And Exxon 6 was trapped inside the loop. Exactly. When this place is from the editor came along, it couldn't see Exon 6 because it was physically hidden in this hairpin loop.
21:34So it just skipped it. It's spliced Xon 5 directly to Exxon 7. So the instruction for Build Tail was on Exxon 6 and it got skipped because the pages of the instruction manual got stuck together by these bits of junk DNA.
21:47essentially, yes. This caused the production of a specific isoform called TBXT Delta X on 6. And how do we know this caused the tail loss? It sounds like a just so story. A good explanation, but how do you prove it?
21:59They proved it experimentally. They used CRISPR to engineer mice to have this specific skipping event. They forced the mouse genome to skip Exxon 6 of its TBXT gene. And the mice. They failed to grow tails, where they have very short, stumpy ones.
22:13That is incredible. A random piece of jump DNA jumps in, Velcro to another piece, hides a page of the manual, and boom, 25000000 years later, humans lose their tails. It highlights how fragile and dynamic our genome is.
22:25It wasn't a grand design. It was a molecular accident that stuck because, presumably, there was an advantage to being tailless for our ancestors. Maybe for sitting upright or walking differently. Let's move from the tail to the heart.
22:39This one blew my mind because it involves a completely different strategy and solves a huge physiological problem. The superheart. This is about the gene, TNNI3, which encodes cardiac proponent array. This protein is part of the machinery that makes heart muscle contract and relax.
22:56Okay, so it's a fundamental part of the heartbeat. Absolutely. Now, think about the heart rate of a human. 60, maybe 80 beats per minute at rest. Unless you've had too much coffee, like me right now. Even then, you're not hitting the numbers of a shrew or a bat.
23:08Some of these small mammals have heart rates up to 1500 beats per minute. One,500. That's not a beat, that's a hum. That is 25 beats per second. It's incredibly fast. And there's a physics problem here.
23:21To beat that fast, the heart muscle has to relax instantly after contracting to fill up with blood for the next beat. If it stays tight too long, a problem called diastolic dysfunction, the heart stops filling and pumping effectively.
23:33So they need a high performance relaxation system. In humans and larger mammals, Exxon 3 of this gene is usually included. It acts as a break. It makes the muscle sensitive to calcium, keeping it ready to contract.
23:46When we need to speed up, like during exercise, we phosph relate the protein, we add a chemical tag that effectively presses the gas pedal to override the brake. So we have manual transmission. We have to actively push the gas using chemistry to go faster.
24:01But moles, shrews, and bats. They can't wait for that chemical signal. It's too slow for 1500 BPM. So evolution did something drastic. did it do? It removed Exon 3 entirely. They cut the brake line. They removed the brake pedal from the car altogether.
24:16In these animals, Exxon 3 is either lost from the genome or consistently skipped via alternative splicing. The resulting protein mimics the gas pedal down, state permanently. It allows for rapid fire relaxation, enabling that incredible heart rate.
24:30So by skipping one axe phone, they turned a family sedan heart into a Formula One engine. Exactly. And what's really cool is that this happened convergently. Bats did it one way, moles did it another, but the result was the same.
24:42Lose X on 3, go fast. It's a beautiful example of evolution, finding the same solution to the same problem. That is wild. It shows how splicing can solve physiological limits. Okay, number three, the immune system.
24:57We've got decoys. This is a great example of exonization making a new Exxon out of junk. The gene is IFNAR 2. It's a receptor for interferon, which is a key alarm signal for viruses. So when a virus hits, Interferon is the signal that tells the cell, we are under attack.
25:14Activate defenses. Right. But you don't want that alarm ringing forever. An overactive immune response is dangerous. You get inflammation, cytokine storms. It can be more damaging than the virus itself.
25:23You need to dampen the signal eventually. You need a way to turn the alarm off. Or at least turn it down. Enter the junk DNA. Our old friend. A primate specific alu element again, an alu mutated into a coding X zone.
25:35When this new exon is included, it changes the shape of the receptor protein. The normal receptor sits on the cell surface. It has a part outside to catch the signal and a part inside to ring the alarm.
25:46This new exon introduces a stop signal, creating a trundated version. It loses the anchor that holds it to the cell. It falls off. It becomes soluble, it gets secreted and floats around in the blood, but and here's the key.
25:59It still has the catcher's mint part. So it catches the interferon signal, but it's not connected to the fire alarm anymore. Exactly. it's a soluble decoy It floats around, soaking up excess interference like a sponge before it can reach the real receptors on the cells.
26:13It dampens the immune response to prevent overreaction. So primates evolved away to chill out their immune system using a jumping gene. Yes. And there's a similar story with IL 13 RA1, another immune receptor.
26:25It seems the immune system loves using these decoy isoforms to fine tune sensitivity. It's not an on off switch. It's a dimmer switch created by splicing. That is so clever. Evolution is a genius. Or a very persistent tinkerer.
26:38finds what works. A tinkerer with unlimited time and a giant box of spare parts. Okay, last case study, the brain. And this one connects to something a lot of people worry about. Alzheimer's. Yes, the MAPT gene, which produces the Tao protein.
26:55We know Tao clumps up into those famous tangles in Alzheimer's disease, but naturally Tao is crucial for stabilizing microtubules, the skeleton of the neuron. Think of microtubules as the train tracks that transport supplies up and down the long axon of a nerve cell.
27:09Tao is the railroad tie holding the tracks together. Got it. A critical job. Very critical. Now, the MAPT gene has a very famous Exxon 10, famous to us nerds anyway. Whether you include or skip Exxon 10 determines whether Tao protein has 3 binding repeats, 3R or 4 binding repeats for.
27:263R versus 4R. What's the functional difference between those 2 isoforms? Four are Ben's tighter to the microtubules. It makes the tracks very rigid and stable. Three R is looser, allowing for more plasticity, more flexibility in the track.
27:37And you need both? A healthy human brain needs a very specific balance of these two. Roughly 50-50. Perfectly balanced, is all things should be. If you throw off that balance, if you have too much 4R or too much 3R, the tracks destabilize, you get neurodegeneration, you get a whole class of diseases called tauopathies.
27:55So how did evolution handle this delicate balance? Humans seem to have fine-tuned this balance to a higher degree than other primates. The paper discusses how we acquired mutations that weakened a binding site for a splicing factor called MBNL near Exxon 10.
28:10By weakening the magnet for this factor, We changed how often it bound. Right. And that shifted the splicing ratios to get that precise 50-50 balance required for our complex long-lived neuronal architecture.
28:22So it wasn't about inventing a new Tau protein. It was about tweaking the ratio of the 2 existing versions to make the brain more stable and perhaps more complex. Exactly. It suggests that our brain complexity relies on these subtle quantitative shifts in splicing, not just new genes for smartness.
28:39We didn't get a smart gene. We just tuned the instrument we already had better. This brings us to the discussion section. We've seen tales, hearts, immunity brains. What does this all mean for the big picture of evolution?
28:50It forces us to completely rethink junk DNA. For years, we dismissed key ease, transposable elements and introns as garbage, evolutionary debris. The stuff we just had to carry around. But this paper, and the research it summarizes, argues they are a reservoir.
29:07The parts bin. A parts bin, a scrapyard of potential parts. They are sequences that are almost Exxons. They just need one or 2 mutations, a splice site here, an enhancer there, and suddenly they're exenized.
29:18They become part of the protein. It explains why we have so much non-coding DNA. It's not a bug, it's a feature. It's future potential. It's not trash. It's the R&D department. It's the raw material for the playground we talked about earlier.
29:31Without that junk, evolution wouldn't have anywhere to experiment. If every piece of DNA was vital. You couldn't change anything without dying. But there are limitations here, right? We have to be careful.
29:42Not every splice variant we find is a genius invention. No, absolutely not. And that's the hard part for scientists now. Distinguishing noisy splicing, just random molecular errors or sloppy machinery from functional splicing, real adaptation is really difficult.
29:58Because if you look deep enough with these new long read sequencers, you find weird isoforms everywhere. 1000000s of them. Many are present at very very low levels. Are they the beta tests we talked about, the early experiments?
30:10Or are they just the cellular equivalent of a typo that gets immediately thrown in the trash? We don't always know We don't. Just because it's there doesn't mean it's doing something useful. Proving function is the next big challenge.
30:22And even with the new tech, it's hard to do this at scale. Current long read sequencing is still a bit of a bottleneck for massive studies. It's expensive and the data is huge and complex to analyze. We can't yet sequence every single cell in a human body to find every isoform, not yet.
30:40So where is this going next? What's the future? Spatial Transcriptomics is the next frontier. Spatial. As in, where things are in space. Yes. Right now, even with single cell, we dissociate the tissue, we turn the brain into a fruit salad, but we lose the shape of the fruit bowl.
30:57We don't know which cells were neighbors. We lose the architecture. We want to know. Is this specific splicing event happening in the den right of a neuron in layer 5 of the cortex or in the cell body?
31:08Location matters? Location, location, location. Exactly. Understanding where the splicing happens within the tissue structure will help us understand what it's doing, and also, as we discussed with cow, looking beyond binary changes.
31:21It's not just X on in or X on out. It's Exxon in 40% of the time versus 60% of the time. Those subtle analog shifts might be where the real magic happens. It's the analog nuance in a digital world. Beautifully said.
31:35So we've covered a lot. From the C value enigma to the loss of our tales. If you had to boil this down for someone at a dinner party, what is the take-home message? I would say this. Evolution is rarely an architect that builds from scratch.
31:48It is a remodeler. It takes the existing structure, the genome, and uses alternative splicing to rearrange the rooms, knockdown walls and build extensions. It finds new uses for old parts. It turns a static code into a dynamic adaptive system.
32:01It creates a car, a plane, and a boat from the same box of Legos. And that flexibility, that editorial power, is why we are here. It's a huge part of the story. I want to leave our listeners with a thought that came to me while reading this.
32:14We talked about how a single splicing change, an accident with 2 alu elements, lost us our tales. We talked about how another one rewired our brains. Right Small changes huge consequences. If that potential is sitting in our junk DNA right now.
32:31What else is in there? What dormant potential is lying hidden in your genome, just waiting for a single letter mutation to create a new splice site? That is the question. Are we the final product? Or is there a super brain Exxon just waiting to be switched on in one of our introns?
32:47A super heart Exxon? Maybe the X-Men aren't so far fetched after all. Just a few splice variants away. Just a few splicing errors away from telepathy. Who knows? Who knows? It's an amazing thing to think.
32:59This episode was based on an open access article under the CCBY 4.0 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.
33:13If 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.
33:23Thanks for listening and join us next time as we explore more science, base by base.
