Bi-allelic FSD1L variants cause retinitis pigmentosa; FSD1L localizes to the photoreceptor axoneme and a deep intronic deletion abolishes retina-enriched exon 10b inclusion.
0:00Welcome to Base by Base, the paper cast that brings genomics to you wherever you are. Thanks for listening, and don't forget to follow and rate us in your podcast app. Always great to be back. So, um, I want to start today by putting you in a very specific headspace.
0:14It's a it's a scenario that plays out in clinics constantly. And frankly, it's heartbreaking. Just imagine you're a parent. You notice your, say your nine-year-old is stumbling a lot when the sun goes down.
0:27They uh, they can't find their seat in a dark movie theater. The classic presentation. Nyctalopia. Right, night blindness. So you take them to the specialist, and they run the tests, they map out the visual field, and they give you a name.
0:41Retinite is pigmentosa. It's a progressive diagnosis. The vision is going to constrict, basically like a tunnel closing in, but then comes the next step. You want to know why? Naturally. You want the cause?
0:52You want to know the prognosis. Exactly. So you sign up for the genetic testing. You give the blood sample, you wait months. And you think, you know, we sequenced 3000000000 letters of DNA. Surely the answer's in there.
1:03And then, yes, silence. The report comes back saying negative. Or the even more frustrating uh, variant of uncertain significance. The dreaded VUS. It's a massive issue in the field. I mean, we are technically capable of reading the entire genome.
1:19Yet for inherited retinal diseases or IRDs, somewhere between 40 and 50% of patients still walk away without a molecular diagnosis. That number is wild to me, nearly half. Nearly half, yeah. It's what we call the missing heritability.
1:32We know it's genetic. We can clearly see it tracking through families, but we just, we can't find the typo. So that is the mission for this deep dive. We are going hunting for those missing answers. And we're asking, what if the answer isn't actually missing at all?
1:48What if it's just hiding in the dark matter of the genome that most standard tests just ignore? And perhaps even more intriguingly, what happens when a gene we thought we knew? A gene we thought belonged firmly to the brain turns out to have a secret life in the eye.
2:01Yes. Today we celebrate the work of Lynn at all. And a really massive collaborative team. We're talking researchers from the UCL Institute of Ophthalmology, Morfield's Eye Hospital, and the Institute of Molecular and Clinical Ophthalmology, Basil.
2:16Heavy hitters in the field, for sure. Absolutely. Their paper, which was published on March 5, 2026, in the American Journal of Human Genetics, it effectively solves one of these cold cases. It's just a masterclass in how to look where others haven't.
2:31It really is. It highlights that the low hanging fruit in genetics has basically been picked. If you want to solve that remaining 50%, you have to climb higher and you have to dig deeper. Right. But before we get into the actual detective work, let's just level set on the condition.
2:44We mentioned retinitis pigmentosa or RP, biologically, what is breaking down here? So RP is the most common form of inherited retinal disease. It affects about one in 3500 people. Globally, IRDs affect around 5.5 million people.
3:01Wow. Yeah, it's significant. You have to picture the retina at the back of the eye, like a digital camera sensor. It's lined with photoreceptors. The rods and cones. Exactly. In RP, usually the rods die off first.
3:13Those are your low light sensors, hence the night blindness we mentioned earlier. And as they degenerate, the overall structural integrity of the retina starts to fail. Eventually, the cones which give you your color and sharp central vision, they die too.
3:27And that's when you go blind. And the unmet need here isn't just about academic curiosity. I mean if you don't know the gene, you're just stuck. You are totally stuck. You can't enter clinical trials. You can't access the new wave of gene therapies, well, those are gene specific.
3:43If I don't know what to fix, I can't give you the tool to fix it. Plus you can't even give the patient an accurate timeline. Some forms of RP are slow, others are rapid, without the gene you're really just guessing.
3:54Okay, so enter the suspect, a gene called FSD1L. Now I look at genomic papers all day, and I'll be honest, this one was not on my radar for eye disease. No, and it shouldn't have been. That's the twist.
4:04FSD1L stands for get ready for this, fiber nectin type 3 and SPRY domain containing one like. Oh very catchy. Rolls right off the tongue, doesn't it? But until this paper, literally nobody thought it was an hygiene.
4:17Its cousin, FSV1, is a known microtubule gene associated with the brain. So finding FSD1L as a definitive cause for blindness is a huge paradigm shift. So how did they catch it? If you have 3000000000 letters of DNA, and you're looking for a typo in a gene nobody even suspects, where do you start?
4:36They started with a very exclusive club, a cohort of just 6 individuals from 4 unrelated families. We had families from the UK, the US, and Switzerland. Just 6 people. That sounds incredibly small for a study like this.
4:49It is small, but that just speaks to the sheer rarity of this mutation. To find them, they utilized massive data sets like the UK 100,000 genomes project just to filter through 1000000s of variants. They were hunting specifically for ultraware biololic variants.
5:04Let's pause on biolect for a second. We're talking about recessive traits here, right? Correct. You need 2 broken copies, one from mom, one from dad to get the disease. If you have one working copy, you're a carrier, but your vision is fine.
5:15These 6 patients had mutations in both copies. But simply finding a mutation isn't enough to prove anything. I mean, everyone walking around has mutations. How did they prove this specific gene was actually doing something in the eye?
5:28That is where the deep dive methodology comes in. They couldn't just rely on reading the DNA sequence. They had to look at the context where it operates. So they use single cell RNA sequencing. Which basically lets them eavesdrop on individual cells to see what genes they are actively using.
5:43Exactly. They compared human retinal data with mouse data. And they found something really fascinating. This gene, FSD1L, was being express, it was being turned on specifically in the photo receptor. And not just any photo receptors.
5:58Right. It was highly expressed in the cones, and to a slightly lesser extent in the rods, which is super interesting because as we said, RP is usually a rod 1st disease. It suggests the biology is way more complex than we originally thought.
6:11Perhaps involving the interplay between how these cells structurally support each other. Okay, so the gene is in the crime scene. It's in the eye. But what is it actually doing there? They use a technique called UXM or ultrastructure expansion microscopy.
6:26This completely blew my mind when I read it. Oh, UXM is brilliant. It's so cool. One of the biggest problems in cell biology is simply that cells are tiny. The machinery inside them is even small, actually smaller than the wavelength of light.
6:40You just can't see it clearly with a normal light microscope. It's just blurry It's highly pixelated, essentially. UXSM solves this by physically expanding the sample itself. Imagine drawing a highly detailed picture on a deflated balloon.
6:54It's tiny, it's crammed, you can't make out the lines. Now below the balloon up, the picture stretches, and suddenly you can see every single detail. So they literally inflate the eye tissue. In a way, yes.
7:04They use a polymer gel to physically expand the tissue about 4 times its normal size. This let them take super high resolution photos to see exactly where the FSD1L protein was sitting inside the cell.
7:15And where was it? It was sitting right on the connecting cilium and the microtubule axinome. Okay, let's demystify those terms for a second. Connecting cilium. Right. Think of a photoreceptor like a factory.
7:26The main body of the cell makes all the energy in the proteins, but the antenna, the outer part that actually catches the light, is physically separated from the main body. The connecting cilium is the bridge between them.
7:37like the logistics highway. Precisely. It's a very narrow bridge where massive amounts of cargo have to travel back and forth every single 2nd to keep your vision working. And the microtubial axonym is basically the steel skeleton inside that bridge holding the whole thing up.
7:52So if FSD1L is broken. The bridge becomes unstable, the crucial cargo can't get to the antenna, the antenna starves, and the cell dies. That is the fundamental mechanism of the blindness here. It's a structural failure of that bridge.
8:06That makes so much sense. But here is where the story gets really complicated. And frankly, really elegant, because they didn't just find one uniform type of patient. No. And this was the puzzle that drove them crazy.
8:20In family A, the patients had retinitis pigmentosa, but they also had mild learning disabilities, and something called spastic diplegia, which is stiffness in the legs. It was a neural eye syndrome. Right, neurological involvement.
8:33But then you look at family D, they had the severe blindness, but absolutely 0 neurological issues, no learning disability, no motor issues, just the eye disease. How is that even possible? If it's the exact same gene causing it?
8:48Shouldn't it cause the exact same disease? That is the big question. And the answer lies in a really crucial concept called isoforms. I love this concept. It's the idea that a gene isn't just one single static instruction, right?
9:00Exactly. We often teach genetics in school, like one gene equals one protein. But it's actually more like one gene equals a recipe book. Depending on which cellular kitchen you're cooking in, you might skip a page or add an extra ingredient.
9:12Those different modified versions are called isoforms. So the brain uses one recipe from the FSD1L book, and the eye uses a totally different one. You hit the nail on the head. The research has discovered a specific retina enriched isoform.
9:24The I's recipe includes a very specific unique chunk of genetic code called Exxon 10B. Exon 10B, the secret ingredient. The secret ingredient that only the eye seems to crave. Now, here is the real Sherlock Holmes moment.
9:40In Family D, the family with only blindness, where was their mutation located? It wasn't in the main coding part, was it? No, it wasn't. It was a deep intronic deletion. It was buried in the intron, which is the part of the DNA we historically used to call junk DNA or just spacer material.
9:56The dark matter of the genome. Exactly. But this specific patch of dark matter contained the vital instructions for the cellular splicing machinery to say, hey, make sure you include Exxon 10 B. Because that specific instruction was deleted, the cell couldn't make the eye recipe.
10:11The brain recipe. The brain recipe doesn't even use X on 10B. So the brain recipe was spliced together perfectly fine. The brain was completely spared. That is just incredibly elegant. So by that logic, in family A, where they had both the brain and the eye issues, I assume their mutation broke something common to both recipes.
10:29Spot on. Family A had mutations like frame shifts or miscent mutations that wrecked the core protein structure regardless of which isoform is being built, so both the brain and the eye systems failed. This completely redefines how we look at the disease.
10:43It's not just a binary is the gene broken or not. It's which version of the gene is broken, and in which specific tissue? It turns the disease into a spectrum. It perfectly explains why clinical presentations can vary so wildly from patient to patient.
10:57And to actually verify this mechanism, they use something called minogene assays. Which sounds kind of adorable, but I assume it's highly technical. It's a very powerful tool. Since the whole gene is huge, they can't easily test the entire thing in a petri dish.
11:12So they literally build a mini version containing just the relevant exxons and introns. They put that synthetic gene into cells, specifically ARPE 19 cells, which are retinal pigment epithelial cells, and they just watched how the splicing machinery handled the code.
11:27Kind of like running a spell check simulation on the DNA. Yes. And it confirmed that the deep intronic mutation in family D specifically caused the machinery to skip Xon 10B. It proved the mechanism definitively.
11:41It wasn't just a hypothesis. They literally watched the cell fail to read the eye recipe. Wow. So zooming out for a second, we have a new gene, FSD1L. We know lives on the connection bridge of the photoreceptor.
11:53We know it has a special eye only mode involving Exxon 10 B. Why does this matter to the listener today or to the doctor seeing a patient tomorrow? The immediate impact is totally diagnostic. As we said, we have 1000s of people in that 40 to 50% unsolved category.
12:09We need to start screening them for FSD1L mutations immediately. That could literally end the diagnostic odyssey for a lot of families right there. Immediately, yes. But the much bigger lesson here is about how we fundamentally do genetic testing.
12:21Standard whole XO sequencing, which is the absolute workhorse of clinical genetics right now, only looks at the Exxons, the coding parts. So it would miss family D entirely. It would miss them, 100% of the time.
12:33This paper is a massive wake-up call that we need to be looking at the introns. We need whole genome sequencing. And more importantly, we need to know what we are looking for when we stare into those non-coding regions.
12:46It's like searching for your keys, right? If you only look under the street lamp because the light is better. Your game miss them if you drop them out in the shadows. Xome sequencing is just the street lamp.
12:57That's a perfect analogy. And biologically, this connects 2 totally different medical fields. You have neurologists seeing kids with learning delays and ophthalmologists seeing kids with blindness. This gene proves they need to be talking to each other.
13:12The symptoms are connected at the root. It's what we call a syndromeic versus non-syndromeic presentation. But the root cause is the exact same gene, just slice differently by the cell. What about treatment, though?
13:24Does knowing it's a transport bridge problem, help us actually fix it? It's the essential 1st step. You can't fix a car if you don't know which part is smoking. Now we know it's a failure of the sillier transport.
13:36Researchers can start looking at drugs that stabilize microtubules, or looking at gene therapy if we want to treat family D, we need to deliver a healthy copy of the gene to the retina. But we better make absolutely sure we deliver the version with Exxon 10B.
13:52Exactly. If you just deliver the standard brain version to the eye, it might not work at all. Or worse, it might interfere with the remaining function. The details matter. The ISO forms matter. This level of specificity is crucial for the next generation of genetic medicine.
14:08It really feels like we are just scratching the surface of the dark genome. Every time we think we've mapped it out, we find a trap door leading to a whole new room. The genome is definitely not a static document.
14:20incredibly dynamic. It's a fluid library of recipes that changes based on whether you're in the heart, the liver, or the eye. And we are just now learning how to read the marginalia, the tiny notes in the margins that tell the cell how to actually read the book.
14:32So, to bring this all together. What's the core takeaway for today's deep dive? I'd say it's twofold? We've identified FSD1L as a significant new cause of blindness. It was hiding in plain sight because we weren't looking at the right isoforms, specifically the retina enriched version containing Xon 10B.
14:50And we've proven that looking at the junk DNA, the introns can finally explain why a disease hits one organ but completely spares another. And biologically, we've reinforced that the photoreceptor connecting celium is a critical choke point for vision.
15:03If that bridge fails, site fails. It's a tragedy for the cell, certainly, but a real triumph for science that we finally understand how it happens. It solves the mystery for these specific families, which is amazing, but I feel like it opens up a much larger, almost philosophical question for the field.
15:19It really does. And this is what I want you, the listener, to really chew on after this. We know half of these retinal cases are unsolved. We just found one answer hiding in a tissue specific isoform. So what does this mean for all the other conditions out there?
15:35The heart conditions, the kidney failures, the liver diseases that are currently labeled idiopathic or unknown cause? Are they really unknown? Or are we just reading the wrong recipe? Are the answers for 1000000s of patients hiding in alternative isoforms that we haven't mapped yet simply because we're only looking at the standard canonical version of the gene?
15:55That is a very provocative thought. The idea that the cure, or at least the cause, is sitting right there in the data we already possess, but we just haven't learned the specific dialect of that tissue yet.
16:06It's the next frontier. Yeah, absolutely have to stop looking at the genome as a one size fits all blueprint. It is bespoke, it is tailored to every tissue, and our diagnostics need to be just as tailored if we want to solve that remaining 50%.
16:18Well, on that note, we have reached the end of our deep dive into the dark quarters of the genome. This 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.
16:32If you enjoyed this, follow or subscribe in your podcast app, and leave a 5 star rating. If 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.
16:45Thanks for listening and join us next time as we explore more science base by base.
