Large-scale exome sequencing shows CFTR risk variants, including deltaF508, reduce susceptibility to inflammatory bowel disease, suggesting targeted CFTR modulation as a potential IBD therapy.
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, it's really great to be back in the studio.
0:10Yeah. I'm excited for this one Today we are looking at a biological mystery. That is, honestly, It's massive. It feels like one of those foundational questions that once you ask it, you can't really un-ask it.
0:23No, you really can't. It completely changes how you look at human history and human health too. Right? Because it touches on that tension between, uh, between survival of the fittest and the reality that our bodies are actually full of what look like mistakes.
0:39Yeah, exactly. We have this idea of evolution as this master sculptor, right? It just ships away everything that doesn't work. If a gene kills you or makes you sick, natural selection is supposed to just grab the red pen and edit it out of the population.
0:51That's evolution 101. In a perfect theoretical vacuum, yes. Harmful mutations that reduce your reproductive success should eventually just disappear. They get weeded out because the people carrying them don't pass them on as often.
1:02But, and here is the hook for today. We know that isn't always true. No, it's not. We have these devastating, lethal genetic disorders that persist in the human population. And they persist at rates that are way too high to be just random bad luck.
1:18Exactly. When you see a bad gene sticking around for 1000s of years, typically at frequencies of like one or 2 or 3% of the population, you have to stop assuming it's an error. You have to start assuming it's a strategy.
1:31A biological strategy, yeah. A trade-off. It's like the classic example everyone learns in biology class is sickle cell anemia. Right, the double-edged sword. Two copies of the gene give you a terrible blood disorder, but one copy.
1:43One copy makes you resistant to malaria. Precisely. So the gene is a killer in one context, but an absolute lifesaver in another. And today, we are looking at a deep dive that suggests this same double edged sword dynamic is happening with one of the most famous genetic diseases in the Western world.
2:00Cystic fibrosis. Cystic fibrosis. We're talking about a disease that destroys lungs, blocks digestion, and historically killed children before they even reached adulthood. It's a brutal disease. And yet the mutation is surprisingly common, especially in people of European descent.
2:15It is, and the question has always been why. What is the hidden benefit here? Right, what's the malaria equivalent for CF? Exactly. And today's deep dive suggests the answer lies in this brutal physiological conflict between the lungs and the gut.
2:30The lungs versus the gut. The hypothesis is fascinating. What if a mutation that destroys your lungs when you have 2 copies, acts as a shield for your intestines when you have just one? So can a broken gene actually protect you from one of the most complex, miserable inflammatory diseases we deal with today?
2:49Specifically, we are talking about inflammatory bowel disease. Can carrying the cystic fibrosis mutation stop you from getting Crohn's disease or ulcerative colitis? That is the big question. Before we get into the answer, which involves some serious data crunching, by the way.
3:04Right. We need to acknowledge the team behind this because this isn't some small observational study. No, this is what we call big science. We are looking at a paper titled cystic fibrosis risk variants, confer protection against inflammatory bell disease.
3:18This was published in Cell Genomics on December 1, 2025. It's a massive collaborative effort. You have the Broad Institute of MIT and Harvard, the Welcome Sanger Institute in the UK, and the International Inflammatory Bowel Disease Genetics Consortium.
3:32I want to shout out the lead authors today. Mingli, Yu-Kian Zhang, and Kai Yuan. And the correspondence team of Carl A. Anderson, Mark J. Daly, and Helian Kwang. These are heavy hitters in the world of genomics.
3:46And honestly, to answer a question this complex, you need heavy hitters. You need access to 100s of 1000s of genomes, which is exactly what they had. Let's set the stage here. We have 2 antagonists in this story.
3:56On one side, we have cystic fibrosis. On the other, inflammatory bowel disease. Let's start with CF. Most people know it has something to do with the lungs, but what is actually happening at the cellular level?
4:07To understand CF, you really have to understand salt. Salt. Like table salt. Simple table salt. Chloride ions. Cystic fibrosis is an autosomal recessive mendelian disorder caused by mutations in a gene called CFTR that stands for cystic fibrosis transmembrane conductance regulator.
4:26That is quite the mouthful. Let's just stick to CFTR. Agreed. The CFTR protein is essentially a gatekeeper. It sits right on the membrane of the epithelial cells that line your lungs, your intestines, your pancreas, and its job is to open up and let chloride ions flow out of the cell.
4:41And in biology where salt goes. Water follows. Osmosis 101 by pumping salt out onto the surface of the airway or the gut. The CFTR channel draws water out with it. And this hydrates the mucus layer that coats our organs.
4:55So it keeps things moist and slippery? Exactly. It makes the mucus runny enough that little hairs called cilia can sweep it out of the lungs, that clears away dust and bacteria. It also keeps the gut lubricated so food could just move through.
5:07But in cystic fibrosis, the gate is broken. The gate is shut tight, or sometimes it's missing entirely. Salt stays trapped inside the cell. Water stays inside the cell, so the mucus on the outside dries out.
5:19It becomes this thick, sticky dehydrated sludge. And that sludge is deadly. In the lungs, it traps bacteria like pseudomonas leading to chronic infections and eventually respiratory failure. In the gut specifically for newborns, it can cause something called maconium ileus, which is a severe bowel obstruction.
5:36Later in life, it causes massive constipation and malabsorption because digestive enzymes basically can't get through the sludge. So CF is defined by too much thickness. Things aren't moving. Now let's pivot to the other character here.
5:49Inflammatory bowel disease or IBD, which includes Crohn's and all sort of colitis. Right. IBD is almost the opposite problem. It's an autoimmune condition where the body attacks its own gut lining. And what's the mechanism there?
6:03Well, normally you have that mucous layer acting as a sort of demilitarized zone. It keeps the trillions of bacteria in your your microbiome away from the actual cells of your intestinal wall. Good fences make good neighbors.
6:16Exactly. But in IBD, that fence is often broken. The mucus layer might be too thin or too permeable, bacteria just swim right through, they touch the intestinal wall, and the immune system absolutely freaks out.
6:29Launches an attack. A massive inflammatory response to kill the invaders, but it ends up destroying the gut tissue in the process. So on one hand, we have CF, where mucus is too thick and too sticky. On the other hand, IBD, where mucus is too thin, leaky, and inflammation runs wild.
6:45And historically, there has been a 3rd player in this theory, cholera. The bacterial infection. Right. Vibrio cholera. Cholera kills you by causing massive, uncontrollable diarrhea. It literally sucks the water out of your body until you die of dehydration.
6:59And the way it does that is by hijacking the CFTR channel. It forces the gate open. It basically kicks the door off the hinges. The cholera toxin overstimulates CFTR, causing it to just dump all the chloride and water into the gut.
7:12So for decades, the running theory was, if you are a CFK, if you have one broken gene, maybe you are resistant to cholera. Your gait is a bit sticky, so you don't lose as much water. You survive the plague, and you pass the gene on.
7:25That's been the textbook explanation. But this paper we're looking at today says, wait a minute. What if it's not just about surviving cholera? What if it's about surviving inflammation? The idea being that having a sluggish CFTR channel puts you in a sort of biological Goldilocks zone.
7:41Exactly. You aren't fully broken, like in CF, but you aren't fully open either. You are somewhere right in the middle. Maybe your mucus is just sticky enough to trap bacteria and stop them from triggering IBD, but not so sticky that it clogs up your lungs.
7:55It makes intuitive sense, but intuition is really dangerous in science. People have actually tried to prove this before. And what happened? The results were a complete mess. Some earlier studies said yes, carriers are protected.
8:06Others said no carriers actually have worse outcomes, and some found no link at all. Why was it so hard to figure out? I mean, we have the DNA. We have the patients. It's a numbers game. The effect we are looking for is likely subtle.
8:19And remember, IBD is complex. It's not a Mendelian disease caused by just one gene. It's polygenic. Lots of genes involved. Right. Plus being a carrier isn't just one thing. There are 1000s of different mutations in the CFTR gene.
8:36If you only look at 500 people, you might not have enough carriers to actually see the pattern. So the team behind this paper decided to stop messing around with small cohorts. They went for the nuclear option of data sets.
8:47They really did. Let's look at the methodology. They used a massive 2 stage design. First, a discovery stage using the broad institute's data. How many people are we talking about? 38,558 IBD cases and 66,945 controls.
9:03That's over 100,000 people just to generate the hypothesis. And then they replicated it. They went to the Sanger Institute and other biobanks and pulled in another 42,000 cases and 192,000 controls. So we were talking about basically a quarter of a 1000000 people analyzed in this study.
9:20Yes. They used both XM sequencing and whole genome sequencing. And that scale allows you to do something very specific called a burden test. I love this term, but it sounds like something from a stress exam.
9:33What is a burden test in genomics? Okay, imagine you want to know if a specific gene is broken. Usually in genome wide association studies or GWAKS, we look at single variants. We look at, say, typo A at position 100.
9:47And we ask, does type O A cause the disease? Right. But what if typo A is super rare? What if only one person in 10,000 has it? You can't get a statistical signal from that, but what if there is also typo B at position 200 and typo C at position 300?
10:03They are all different typos, but they all break the gene in the exact same way. So the burden test groups them. It collapses all these rare variants into a single score. It basically asks if we treat everyone who has any gene breaking mutation as one single group, do they have a different risk profile?
10:20It's like the broken car analogy. It doesn't matter if your transmission fail, your engine block cracked, or your tires were slashed. If the car doesn't run, it's in the broken car category. Perfect analogy.
10:30They aggregated all the mutations that are known to cause cystic fibrosis and treated them as a single burden. But wait, how do they know which mutations actually break the car? There are 1000000s of changes in our DNA that don't do anything at all.
10:44That is the tricky part. You really have to filter the noise from the signal. So they used a clinical database called CFTR2. This is the gold standard. It's a list of variants that doctors know cause cystic fibrosis in actual human patients.
10:58And they also had a control group of variants, right? Yes, and this was so clever. They took variants in the CFTR gene that are known to be benign, changes that don't break the protein, and they ran the burden test on those too.
11:10As a negative control. Right. If the benign variants also showed protection against IBD, you'd know your math was wrong or there was some other bias in the data, but they found no association there. The protection was only linked to the broken variants.
11:23Now, I want to applause on something they did regarding ancestry because usually genetics has a terrible bias problem. It does. The vast majority of genomic studies have been done on people of European ancestry.
11:35It's a huge issue because it means our medical insights are just completely skewed toward one population. And CF is often thought of as a European disease. Historically, yes, because the most common mutations are found in northern Europeans, but this team made a massive effort to look beyond that.
11:52They included African American, East Asian, and Southeast Asian populations. How do they make sure they are comparing apples to apples, though? If I grab a group of people from Tokyo and compare them to a group from London, their DNA is going to be different for a 1000000 reasons that have absolutely nothing to do with IBD.
12:10That's a problem called population stratification. If you don't control for it, you get false results. So they used a technique called principle component analysis or PCA. Let's break that down for the listener.
12:20PCA is one of those terms that gets thrown around a lot, but what does it actually do? Imagine you have a room full of people. You measure their height, their weight, hair color, shoe size, 1000s of variables.
12:32PCA is a mathematical way to squash all those variables down into a simple 2D map. When you do this with genetics, people with similar ancestry naturally cluster together on the map. So all the people with British ancestry land in one corner of the map and people with Han Chinese ancestry land in another.
12:50Exactly. Or they use random forest models to assign ancestry groups. So the researchers could use this to say, okay, we are only going to compare the cases in the East Asian cluster to the controls in that exact same East Asian cluster.
13:03Ensuring that any difference they find is actually due to the IBD risk, not just because they have completely different genetic background. Precisely. It's incredibly rigorous and it's necessary for a study this big.
13:13Okay, let's get to the juicy part, the findings. They crunch the numbers on 250,000 people. Does being a CF carrier protect you from IBD. The answer is a resounding yes. Let's start with the most famous mutation, Delta 508.
13:28This is the deletion of a single amino acid phenoline at position 508. The one responsible for the vast majority of CF cases. If you carry just one copy of Delta 508, your risk of developing IABD drops significantly.
13:44The overall odds ratio was 0.82. Talk to me about odds ratios. If one.0 is no difference. What does 0.82 actually mean for you? It means roughly an 18% reduction in risk. Now, 18% might not sound like a miracle cure to some people, but in the world of complex polyogenic diseases.
14:04That is a massive signal. It is highly statistically significant. The P value was something like 10 to the -11, right? 8.96 times 10 to the negative 11. Yes. That means the probability of this result happening by random chance is effectively zero.
14:17It's incredibly real. Did it matter if it was Crohn's or ulcerative colitis? It worked for both, but the protection was definitely stronger for Crohn's disease. The odds ratio there was 0.79. So a 21% risk reduction. For colitis, it was 0.87, which is still nominal significance.
14:36Why the difference between the two? We are 100% sure, but Crohn's affects the entire digestive tract, right? It goes deeper into the tissue, whereas colitis is restricted just to the colon and the surface lining.
14:48The CFTR mechanism might be more relevant to the specific deeper pathology of Crohn's. Now, skeptic listening to this might ask, wait, people with full-blown cystic fibrosis have terrible gut issues. Are we sure the researchers aren't just confusing undiagnosed CF patients in the data?
15:07That's a crucial check. The researchers rigorously excluded anyone who was homozygous, meaning anyone who had 2 bad copies of the gene. They removed anyone who actually had cystic fibrosis. So this result is strictly for the carriers.
15:19You don't have the disease. You don't have the lung rot. You just have the one copy. And you get the shield. You get the shield. I remember we talked about ancestry. This is where it gets really cool. They found this exact same protective effect in the African-American and East Asian populations too.
15:32Yes. Why is that cross ancestry replication so important? Because if it only happened in Europeans, you could argue it was just a coincidence. Like a fluke Right. Maybe the CFG is sitting next to another gene on the chromosome that actually confers the protection and they just get inherited together in Europeans.
15:50genetic linkage. Right. But if you see the same protection in African and Asian populations who have totally different genetic backgrounds and different linkage patterns, it proves that the protection is coming from the CFTR gene itself.
16:04It's a biological universal. Powerful. It means the mechanism is fundamental to human biology, not just a European quirk. It really is. Now, I want to take a slight detour into a part of the paper that I found absolutely fascinating.
16:18It's a bit inside baseball, but it's a showdown between human clinical curation and artificial intelligence. I always love a good man versus machine story. So in the last few years, Google Deep Mind released alpha missens.
16:31It's an AI model derived from alpha fold. It predicts whether a genetic mutation is pathogenic, meaning bad or benign, meaning harmless, based purely on how it changes the 3D shape of the protein. And alpha missens has been hailed as a total revolution in genetics.
16:47It has. It's an incredible tool. So, the researchers thought, hey, instead of just using the list of mutations doctors already know are bad from CFTR too. Let's use the AI to find all the bad mutations, even the ones doctors haven't seen in the clinic yet.
17:00That's logical. The AI should be able to find more broken cars than the doctors can. That was the hope. They ran the burden test using the AI's predictions and it barely worked. Wait, the AI failed. It performed significantly worse than the human clinical list.
17:14The statistical signal was just incredibly weak. Why? Is the AI just not good enough yet? No, the AI was actually too smart or rather too sensitive. It was flagging tons of variants as pathogenic because they slightly altered the protein structure.
17:28It gave them a score of greater than 0.8. But structure isn't everything. Exactly. In reality, those structural changes weren't enough to actually cause the disease in a living, breathing human being. So it was crying wolf.
17:40It was crying wolf. It was diluting the pool. It was tossing a bunch of working cars into the junkyard. So when you ran the stats, the broken car group was actually full of working cars, so the connection to IBD just completely disappeared.
17:53That is such a critical lesson. We tend to think AI is magic, but structural change doesn't always equal functional failure. Context is everything. A protein might look a little weird to an AI, but still do its job perfectly fine in the cell.
18:07The clinical data, the actual history of human patients getting sick, was orders of magnitude more powerful than the AI prediction in this context. So don't fire the doctors just yet. Not quite yet, no.
18:19Let's move to the discussion. We know that this protection happens. The data proves it. Now we need to understand why. We touched on the Goldilocks mucus earlier. Let's go a bit deeper into the mechanism.
18:29This brings us to the biophysics of the gut. We talked about how CFTR pumps chloride and water. In a carrier, you basically have 50% of the normal CFTR function. So you were pumping half the water. Roughly, yeah.
18:41This means your mucus is slightly more viscous. It's stickier, it's less hydrated. And in the context of IBD, Y is sticky good? Think about the enemy in IBD. Bacteria trying to breach the wall. If the mucus is runny, they can just swim right through it, if the mucus is a thick, sticky gel, it acts like quicksand.
19:01It physically traps them. It traps them and keeps them away from the epithelial cells. This is crucial because the moment a bacterium touches an epithelial cell, the cell screams for help and the immune system drops a bomb.
19:14And that bomb is inflammation. That's the mucous hygiene hypothesis. By slowing the bacteria down, you prevent the alarm from ever being tripped. But there is another layer to this involving a very specific type of cell called the best T4 plus Enterocite.
19:29Best T4 plus and terasite. Sounds like a specialized droid from Star Wars. They are very cool cells. They are specialized pH sensors and electrolyte transporters in the gut. It turns out CFTR is highly expressed in these specific cells.
19:42And what happens to them in IBD? In active alternative colitis. We see that these cells get totally messed up. They lose their function. The study suggests that in CF carriers, because these cells have lower CFTR activity, naturally, they might be, let's call it preconditioned.
20:00Preconditioned how? They might be less reactive to stress. A wild type or normal cell might absolutely panic when the pH changes, or when it senses a microbe triggering that runaway inflammation, a carrier cell might be more chill.
20:12It's like, oh, the pH is off, whatever. I'm used to it. So it raises the threshold for freaking out. Exactly. It dampens the inflammatory reflex. You also mentioned salmonella earlier. Is there a direct link there with CFTR?
20:23Yes. There are some really fascinating mouse studies showing that if you have CFTR mutations, salmonella Thai Phi, the bacteria that causes typhoid fever has a much harder time crossing the gut wall. It effectively closes the door the bacteria uses to break in.
20:38Right. And since IBD is partly driven by bacteria crossing the gut wall, the mechanism really overlaps, the same lock that keeps typhoid out might keep your own microbiome at a safe distance preventing Crohns.
20:48This brings us to the most practical question for anyone listening. Therapeutics. If having a lazy CFTR gene protects you from IBD, can we fake it? That is the $1000000000 question. Currently, all our CFTR drugs are designed to do the exact opposite.
21:06Drugs like Tricofta are potentiators. They force the channel open to help CF patients breathe. But for an IBD patient, we would want to see FTR inhibitor. We would want to close the gate. But isn't that dangerous? If you close the gate too much, don't you just give them cystic fibrosis?
21:22That is the massive risk here. You are walking a serious tightrope. If you develop a pill that inhibits CFTR all over the body, You might cure their colitis but destroy their lungs. You'd fill their airways with mucus.
21:34So you'd just be trading one awful disease for another. Exactly. So the drug would need to be incredibly specific. Maybe it's a pill that doesn't get absorbed into the blood at all. It just stays in the gut and only acts on the intestinal lining.
21:46Or maybe a drug that is a dimmer switch, not a total off switch. Just lower the activity by 50% to mimic a carrier, but don't kill it entirely. It's a huge pharmacological challenge, but knowing the target is half the battle.
21:59Before this paper, we didn't really know that CFTR Inhibition was a viable broad strategy for IBD. Now we do. It kind of flips the script on drug discovery. Instead of trying to guess what pathways are involved, we are looking at what evolution has already successfully done.
22:14Evolution is the biggest R&D lab in the world. It has been running clinical trials on humans for 100s of 1000s of years. And the results of those trials are the survivors. Exactly. The high prevalence of the CF mutation in Europe isn't an accident.
22:28It's a direct result. It means that for 1000s of years, carrying this mutation helped people survive something. Call her a typhoid dysentery. And now, even though we have clean water and antibiotics, that ancient survival mechanism is still hanging around, protecting us from modern inflammatory diseases.
22:47It's a beautiful example of the heterozygot advantage. It reminds us that health is always context dependent. A gene that is bad in the Ice Age might be good in the Bronze Age and bad again in the 21st century.
22:58Or in this case, surprisingly good for IDD. It really makes you rethink the whole term genetic defect. It does. It not a defect. It's a variation with consequences depending on your environment. So let's wrap this up with a take-home message.
23:12If a listener is stuck in an elevator and someone asks, what was that deep dive about, what should they say? They should say, we used to think the cystic fibrosis gene was just a killer, but by analyzing a quarter 1000000 genomes, we now know that being a carrier provides a powerful shield against inflammatory bowel disease.
23:30It proves that the body is a system of trade-offs where what hurts the lungs can actually save the gut by changing the physics of our mucus. And it suggests that the next blockbuster drug for Crohn's might actually be a controlled targeted dose of cystic fibrosis light.
23:45Exactly. I'm going to leave everyone with a provocative thought as we sign off. We spend so much energy trying to fix our genome, Chris Bergene editing. We wanna wipe out disease. Which is obviously a noble goal.
23:57It is. But if we had magically wiped out the CF mutation 5000 years ago, would the human population of Europe have been completely wiped out by cholera? And if we wipe it out now, what other unseen protections are we losing that we don't even know about yet?
24:12It raises a really profound question. Could the key to curing cancer or Alzheimer's be hidden inside the biology of other genetic diseases we currently just label as defects? A library of solutions written right there in our DNA that we were only just learning how to read.
24:27It's very humbling thought. On that note, we are going to close the book on this deep dive. Thanks for having me. 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 a license in our episode description.
24:42If you enjoy 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.
24:56Thanks for listening, and join us next time as we explore more science, based by base.
