Patient data, population genetics and iPSC-derived βcell models show INS R6C impairs preproinsulin ER translocation and causes recessive insulin-deficient diabetes in homozygotes.
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. So let's dive in. We need to talk about this, this whole dogma of toxic gain function.
0:12In rare disease genetics. It's sort of the default, right? The go-to explanations. And what we're all taught. Yeah. You find a mutation, the protein misfolds, it stresses the cell, and boom, the cell dies.
0:23A clean, logical story. But what if that story is just... Well, too simple. There's a new paper out that suggests this whole narrative might be leading us to, I mean, misdiagnose 1000s of patients. It's a classic case where the textbook theory was so strong, we almost missed what the biology was actually telling us.
0:42We're looking at a scenario where it's not about toxicity at all. It's more about a threshold. Specifically we're talking about the INS gene. insulin production. For years, we thought a certain mutation here was like a dominant bully.
0:54You know, one bad copy and it just crashes the whole system. But this new work shows that the body is surprisingly resilient. It can handle a broken gene as long as you've got a backup. It basically rewrites the whole story of this type of diabetes from a killer disease to more of a logistics failure.
1:13A traffic jam, a logistics failure that went unnoticed in parents for decades just because we weren't asking the right question. We weren't looking at Zygosity. And that's the hook for today's deep dive.
1:23We're looking at how a single amino acid change can trick us into thinking a cell is dying when really it might just be on strike. Exactly. Today we're diving into a paper titled A new form of diabetes caused by INS mutations defined by zygosity, stem cell, and population data, which was published in MBO molecular medicine on January 3, 2026.
1:44And this was a massive effort from Utah, Miriam Snop, and their team at the ULB Center for Diabetes Research in Brussels. They also had collaborators from Luxembourg, Exeter, and Helsinki, a big team. What I really appreciate here is the way they tackled it.
1:58It's this methodological triad. They didn't just stop at what they saw in the clinic. No, not at all. They integrated huge population scale data, CRISPR corrected, stem cell models, and even alpha fold predictions to really build their case.
2:12It's the modern genomic toolkit in action. I mean, they essentially had to build a human pancreas in a dish to solve a puzzle that had confused clinicians for years. Okay, so let's set the stage. We're talking about monogenic diabetes.
2:24Right. So this isn't type one, which is autoimmune or type two, which is more about metabolic resistance. This is caused by a defect in a single gene. And in this case, the gene is INS. The gene that codes for pre-pro insulin, it's the blueprint.
2:39If you break the blueprint, you break the body's ability to regulate its blood sugar. The specific mutation we're focused on is called R6C, originine to cystine, and historically, this has been stamped as autosomal dominant.
2:51That's right. And the logic, you know, it made sense based on the biochemistry. The mutation swaps an arginine for a cystine at the 6th position of the protein. And Sistina sticky. Sistine is very sticky.
3:02It forms to sulfide bonds. So the assumption was that this extra cystine just wreaks havoc with how pro insulin folds inside the endoplasmic reticulum, the ER. So the theory was, this misfolded protein builds up.
3:15It triggers the unfolded protein response, the UPR, and that just stresses the beta cell until it commits suicide, a poptosis. That's the classic toxic gain a function model. One battle eel poisons the whole well.
3:27But the authors of this study saw something that just didn't fit. They had these pro bands, these kids, with really severe early onset diabetes, diagnosed at, like, age 9 or 11. But their parents, who carry the exact same variant, were fine.
3:42Clinically, you couldn't tell them apart from the general population. And there's the paradox. If R6C was really a dominant toxic agent, those parents should have had significant beta cell loss, but they didn't.
3:53They were walking around with normal fasting glucose. Which really implies that zygosity is a key here. The kids were homazygus, 2 bad copies. The parents were heterozygis, one bad one good. But to prove that, you have to do more than just point that out.
4:06You have to prove that the single meat and allel isn't stressing the cell. You have to actively disprove the old toxic theory. You need to see the machinery in action. You really do. So let's get into how they did that, their methodology.
4:18They started with big data, right? to check how common this thing even is. Yep. They scour the UK biobank, Geisinger, No Me D, these massive databases. And they found that R6C is present in about one in 25,000 people.
4:32So it's rare, but not, you know, once in a lifetime rare. Okay. And here's the crucial part. The heterozygous carriers in the UK buyer bank did not have a higher rate of diabetes than non-carriers. Wow.
4:45Okay, so that's a massive statistical blow to the dominance theory right there. It's a huge one. Statistically, having one copy doesn't seem to make you sick, but, you know, statistics aren't mechanism.
4:54To understand why, they went for the gold standard. Isogenic IPSEs. Induced plural potent stem cells. Exactly. They took blood cells from the patients, reprogram them back into stem cells, and then pushed them to become pancreatic beta cells.
5:09But here is the really elegant step. They used CRISPR Cast 9 to correct the mutation in those same cells. And that's so critical because it gets rid of all the genetic background noise. You end up with 2 cell lines that are perfect genetic twins, except for that one single letter change.
5:25It's the perfect controlled experiment. You have the patient's homeazyga cells, and you have the CRISPR corrected version as your isogenic control. They also made heterozygous lines to mimic the parents.
5:36And they didn't just watch them in the dish. They actually put these human beta cells into mice to see if they worked in a living system. Plus they used alpha fold 3 to model the protein structure. I mean, they threw the kitchen sink at this problem.
5:49So let's get to the findings because this is where that traffic jam idea really comes into play. When they looked at the homozygus cells, the ones with 2 bad copies. What did they see? They found that insulin secretion was just slashed.
6:02It was down by about 50%. The cells were barely putting out any functional protein. Okay, but why? Well, the reason was completely unexpected. It wasn't about misfolding inside the ER. It was about getting into the ER in the 1st place.
6:14The signal pep type. The signal picked it. Yes. So pre-pro insulin had a little tag on the front, a sequence of amino acids that's like a VIP pass, a molecule called the signal recognition particle. The SRP sees this pass and basically drags the whole protein making machine to the ER membrane.
6:33Like the bouncer at a club. It's the bouncer. And it docks with a channel, a little door called the Sex 61 TransLicon and shoves the protein through. Into the factory. into the factory. But what Alpha Fold 3 predicted was that the R6C mutation, that single swap it, drastically lowers the hydrophobicity of the signal peptide.
6:53It messes up the VIP pass. It changes the physical properties of the key, so it doesn't fit in the lock. Pretty much. The model suggested the pass either fails to be recognized by the bouncer, the SRP, or it tries to go through the door, the Sex 61 channel, all wrong, maybe even upside down.
7:08So the protein just never enters the factory. Exactly. It piles up in the cytosol, which is the fluid outside the factory. It's a translocation failure, a traffic jam at the front door. And wait, this explains why the old toxic gain of function theory was wrong.
7:21If the protein never even gets into the ER. It can't cause ER stress. Spot on. They looked for all the classic markers of ER stress and apoptosis, you know, cell death, and they found nothing. The homes Iga cells weren't dying.
7:36They were just sort of dormant. That's fascinating. So they're not dead. They're just on strike. They can't do their job because the raw materials can't get through the door. Or maybe idling is a better word.
7:46The study showed that protein translation gets repressed. It's a mechanism called the integrated stress response to stop that pile up in the cytosol from getting out of hand. The cell basically says, whoa, stop the assembly line, but the cell itself is perfectly viable.
8:00Okay, so that brings us back to the parents. Yeah, heterozygotes. Why are they walking around perfectly healthy? Because they have one good copy, one functional INS allel. And that one good gene produces enough correct signal pectides to keep the bouncer happy and the doors open, the traffic flows.
8:19So one lane on the highway is enough. It is. The study showed that the heterziga cells had insulin secretion levels that were pretty much comparable to the healthy controls. The bad protein that gets made just gets cleaned up by the produceum in the cytosol, and the good protein from the other allele keeps the lights on.
8:37This completely flips the script for genetic counseling then. We're moving R6C from dominant to recessive loss of function. It's a huge relief for these families. I mean, parents no longer have to feel like they've passed down some kind of poison that's also hurting them.
8:51And clinically, it tells us that heterozygous carriers are mostly fine, but their, let's call it their reserve capacity, is a bit thinner. The study mentioned environmental stress. Yeah, like obesity or pregnancy.
9:02Exactly. This fits perfectly with a 2 hit hypothesis. The genetics kind of load the gun. You have a slightly smaller engine, but it's the environment that pulls the trigger. The parents in the study were generally healthy, but some did develop mild diabetes later in life, especially if they became obese or for the mothers during pregnancy.
9:21Gestational diabetes, that makes total sense. Pregnancy puts a massive demand on insulin production. If you're already running at, say, 50% capacity because one of your genes is down, you're much more likely to hit that wall.
9:33You are, but it's not the inevitable, severe, early onset disease we once thought. It's a susceptibility, not a sentence. So what about the kids, the homozygous patients? The study confirmed they had very low C peptide, so almost no insulin production.
9:47Did the researchers try any treatments on their models? They did. They treated the mice, the ones carrying the human homozygus beta cells, with GLP1 receptor agonists, drugs like Xenden 4. So, similar to the Ozempic or Truicity class of drugs we hear so much about.
10:03Did it work? Did they kickstart the factory? Well, it improved their blood glucose a little, but mostly by suppressing glucagon. It did not restore insulin secretion. Because the door is still jammed. The door is still jammed.
10:16You can press the accelerator all you want, but if the fuel line is cut, the engine won't start. GLP1 agonists stimulate the secretion machinery, but if the protein can't get into the ER to become insulin in the 1st place, all the stimulation in the world isn't going to help.
10:32That's a really critical insight for treatment. It tells doctors that pushing these beta cells harder with certain drugs just isn't going to work. The problem is structural. It confirms that for these homzygous patients, insulin replacement therapy is, for now, the only way to go.
10:48But on the bright side, because the cells are alive and not dead, it theoretically opens the door for something like gene correction down the road. If you could crisper that typo back to normal, those cells should wake right up.
11:00Looking at the bigger picture, this makes me a little worried about our genetic databases. I mean, we all rely so heavily on things like Clinvar to tell us if a mutation is pathogenic or not. It should make us worry.
11:11This study is a warning shot. How many other variants are sitting in those databases classified as pathogenic or dominant simply because we haven't looked closely enough at the parents? We could be overdiagnosing risk in 1000s of carriers, while completely missing the real mechanism in the patients who are actually sick.
11:31It's a call to action. We need to revisit all these variants of uncertain significance, the VUS pile, with real functional assays. We can't just rely on the old dogma of dominance anymore. We need to look at the protein mechanics.
11:45We need to verify zygosity. It really highlights that tension between what you see in the clinic, the phenotype, and the molecular reality. The disease looked dominant because you saw it in every generation with sick kids.
11:56But the genotype required a recessive explanation. And honestly, without the big population data from the UK biobank, we might never have spotted all those healthy carriers who were the 1st clue that the story was wrong, it just shows how essential big data is for understanding rare disease.
12:11So if we boil this entire deep dive down, what's the one message you want our listeners to take away? The central insight is that the INS R6C mutation is not a dominant toxic killer. It is a recessive traffic jam caused by a faulty address label.
12:27The signal peptide on the insulin protein. One good copy of the gene is enough to keep the factory running, but 2 bad copies shut down production, though they leave the cell itself intact. It's a fundamental shift from cell death to cell dysfunction, and that shift really, really matters.
12:44It matters immensely. And it leaves us with this one last provocative thought. If we start reanalyzing other so-called dominant monogenic diseases with this level of scrutiny, you know, combining stem cells and population data.
12:57How many other incurable degenerative diseases will turn out to be treatable trafficking defects? I mean, are we misinterpreting the silence of the cell for its death? That is a fascinating and maybe slightly unsettling possibility.
13:09A humbling reminder that biology is rarely as simple as our textbooks make it safe. Absolutely. 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.
13:23If 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.
13:38Thanks for listening, and join us next time as we explore more science base by base.
