In mouse neurons, UFM1 loss or UFM1-R81C expression reduces protein translation, triggers ER stress and PERK activation, impairing dendrite and synapse development.
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. Glad to be here for no one. So, today I want to start you off with a mental image.
0:13I want you to picture the ultimate factory, just the most sophisticated, high stakes manufacturing plant in the known universe. I mean, given our usual deep dives, I am assuming we are talking about the developing human brain.
0:27We are. It is this incredible assembly line that has to run with absolute perfection. It's building neurons, it's connecting circuits. It is practically knitting consciousness out of raw proteins. But now, imagine there is a glitch in the shipping department.
0:42Just a tiny seemingly insignificant error. The shipping department. Yeah. The workers, the proteins, right? They aren't getting the right shipping labels. They exist. They are perfectly ready to be used, but they don't have the tag that tells them where to go, or actually, more importantly, that tells the foreman they passed quality control.
0:59Right, which is a major problem. Exactly. And because of that one missing sticker, The factory foreman completely panics. And by panic, I assume you mean the cellular stress response. I do. The foreman doesn't just slow down the line.
1:13He slams on the emergency brake. He shuts down the entire floor. The machinery overheats, and the product comes out completely stunted. That is a terrifyingly accurate metaphor for the mechanism we are getting into today.
1:26Because when the cells quality control system fails like that, the reaction can be catastrophic. And the stakes here aren't just hypothetical, we aren't talking about a bad day at a real factory. We are talking about microcephaly.
1:40Babies born with significantly smaller brains. We're talking about severe epilepsy, intellectual disability, movement disorders. Right. These are life-altering conditions for the patients and their families.
1:50And for a long time, the link between the genetic defect and the actual physical brain structure was, well, it was a bit of a black box. Right. We knew the what, but not the why. Exactly. Scientists knew that a specific system failure caused these diseases, but the why was really murky.
2:08Like was the brain running out of parts was the machinery literally breaking down. And here's the question that really hooked me when I was looking at the source material for this deep dive. What happens if you try to fix this incredibly complex genetic machinery failure with a drug that is probably sitting in 1000000s of medicine cabinets right now?
2:27You are referring to an antidepressant? I am, because the answer to that question might actually rewrite how we treat rare genetic disorders. It implies we might not even need to fix the broken gene to fix the disease.
2:39It is a phenomenal study, and today we really have to celebrate the work of Catarina Perdigalon, her colleagues. Yes. They are at the Max Planck Institute for multidisciplinary sciences, specifically the Department of Molecular Neurobiology.
2:52And looking at the author list, this was a massive team effort. They had collaborators from the University of Dundee and Charity University States Medicine, Berlin. This is big collaborative cross border science.
3:03Absolutely. Their breakthrough study titled Encephalopathy Linked UFM1 Variants Impede Neuronal Protein Translation, Development and Function, was published in MBO Molecular Medicine in 2026. So let's get into the weeds of this factory floor.
3:19We mentioned a labeling system earlier. What exactly is this system? We are talking about a process called UF Milation? UF milation. Right. And to understand that, you have to remember that when a cell makes a protein, the job isn't done.
3:33That protein is like raw steel coming out of a smelter. It often needs to be modified or folded or tagged to actually function. We call these post-translational modifications. Like putting a stamp on a letter so it actually gets delivered.
3:46Or like a past inspection sticker. Precisely. And one of these stamps is a small protein called UFM1. It stands for ubiquit and fold modifier one. Okay. It is a cousin to ubiquitan, which is the famous tag used for recycling proteins, basically marking cellular trash.
4:01But UFM1 is different. Its specific role has been pretty enigmatic. Enigmatic is a polite way of saying, we knew it was important, but we had absolutely no idea what it actually did day to day. Fair enough.
4:12We knew it was essential for life. But to attach this UFM one tag to a protein. The cell uses a sort of bucket brigade of enzymes. A bucket brigade. Yeah. So you have the E1 enzyme, which is called UBA5, and that activates the tag.
4:25Then it passes it to the E2 enzyme, USC1. And finally, the E3 complex, UFL one, actually attaches it to the target. It's a relay race. So UBA 5 hands the baton to UFC one, who hands it to the target. Exactly.
4:39And if anyone drops the baton, if you have genetic variants in UFM1 itself, or the E1 or the E2, you get these severe human diseases we mentioned earlier. And collectively, those are called UF mylopathies. Right.
4:50Which is a mouthful, but it groups together a lot of human suffering. It does. The symptoms across these encephalopathies are really consistent. Global developmental delay, intellectual disability, and epilepsy.
5:01Now, the study focused on 2 very specific scenarios seen in patients. First, simply not having enough UFM1. The out of stock problem. The factory literally ran out of labels. Exactly. This is UFM one loss.
5:12But the 2nd scenario is much trickier, and, frankly, far more interesting from a biochemical standpoint. It's a specific mutation called R 81C. R 81c. It sounds like a coordinate on a map. In the proteins geography, it basically is.
5:27It means that at position 81 in the amino acid chain, and arginine is replaced by a cystine. The protein exists, the factory makes it, but it's wrong. So you have the tag, but maybe the glue on the back doesn't stick, or the barcode is smudged.
5:42That is exactly what they wanted to find out. Is it just a dead tag? Or does it do something worse? Does it actively gum up the machine? So how do you actually study this? Because I assume you can't just knock on the door of a developing neuron in a human fetus and ask it how it's feeling?
5:55No, obviously not. And you can't just delete the UFM1 gene from a mouse entirely either, because that's lethal. The embryo simply doesn't survive without this system. So the team used a conditional knockout model.
6:07The EKO, right. They used a really clever genetic trick, the Krylock system, to delete UFM1 specifically in neurons. Just in the cortex and hippocampus and right during development. Okay, so the rest of the mouse is fine.
6:21The brain is missing this labeling system. Exactly. And then, to really understand the mechanism, they did something else. They use viral vectors to put UFM1 back in. Like a software patch. Yes. In some neurons, they put back the healthy, wild type, UFM1.
6:35In others, they put in that patient variant we talked about, the R 81 C mutant. Oh, wow. So this allowed them to compare directly. Like, can the mutant fix the damage or is it useless? Precisely. And they threw the kitchen sink at these neurons to analyze them.
6:48I saw they used something called shawl analysis. Yeah, that's a standard, but very powerful way to visually trace how complex a neuron is. Imagine drawing concentric circles around the cell body, like a target.
7:01You count how many times the branches the dendrites cross those circles. It mathematically quantifies how branchy the tree is. Got it. And they also hook them up to electrodes, right? Patch clamp electrophysiology.
7:12They literally listen to the electrical firing and synaptic transmission of single neurons. But my absolute favorite acronym in the steep dive has to be fun and cat. Fun sack cat. Fluorescent, non-canonical amino acid tagging.
7:27Which sounds incredibly complicated, but if I understood the notes correctly, It's basically a way to watch the factory in real time. It is. It allows you to visualize and measure the rate of new protein synthesis.
7:39If the factory line is moving and making product, the cell glows, if the line is stalled, it's dark. It gives you a real-time read on the productivity of the cell. So let's get to the results. They delete UFM1.
7:50What happens to the neurons? Collapse. It is a complete morphological collapse. That sounds dramatic. It is dramatic. The neurons are stunted. The cell body, the soma, is physically smaller. And when they looked at the dendrites, those receiving branches we talked about, they were significantly less complex.
8:07They had far fewer branches to connect with other cells. So if a normal neuron looks like a big, bushy oak tree. These look like what? Like a sapling that's struggling to grow in a drought. And because the structure was stunted, the connections were missing.
8:19They saw a massive reduction in the number of synapses. And synapses are the whole point of a neuron. That's how they talk each other. Exactly. Both excitatory and inhibitory synapses were heavily reduced.
8:30And when they listened to the electrical signals, the EPSC amplitudes, the strength of the signal were incredibly weak. Weak signals, fewer connections. But here's where I found a really surprising detail in the data.
8:43The paper says the probability of release was normal. This is a crucial distinction to make. The synapses that did survive. They function perfectly fine. Structurally under the electron microscope, they looked completely normal.
8:57The pool of vesicles was smaller, but the probability of actually releasing a neurotransmitter was standard. So it's not a quality problem. It's a quantity problem. Precisely. The individual workers. The synapses are competent.
9:10There just aren't enough of them because the factory never built the infrastructure. So why is the factory understaffed? This brings us back to the engine room. What did F1 shit tell them about the assembly line?
9:21It revealed a severe reduction in global protein translation. The neurons essentially stopped building the proteins they needed to grow. The assembly line actually did stop. It halted. And the reason seems to be directly linked to the endoplasmic reticulum or the ER.
9:36The ER is where proteins are folded and processed. It turns out UF milation is critical for ribosomes, the actual protein builders, that are attached to the ER. So when the tag is missing, the ribosomes just jam up.
9:48They stall out. And when ribosomes stall, the cell senses a crisis. It triggers the unfolded protein response or UPR. That sounds like the emergency alarm we talked about. It is, specifically, it activates the PRK pathway.
10:01When the cell senses that proteins aren't being handled correctly, PRK gets activated. It phosphorolates a factor called EIF to alpha, which essentially screams to the nucleus to stop making new proteins until they fix the mass.
10:13Okay, so the lack of the tag causes a traffic jam, the cell panics, pulls the emergency brake via P ear, shuts down production, and as a result, the neuron just never grows up. That is the exact mechanism for UFM1 loss.
10:27The brake is on permanently. But remember the R81 CE mutation, the patient variant. Right, the broken label maker. The researchers found something really nuanced here. R 81c is a hypomorph. Meaning it works, but barely.
10:41It's highly inefficient. They found that the E1 enzyme, UBA5, struggles to charge or activate this mutant UFM1. It takes much, much longer than the wild type. So the relay race is happening in slow motion.
10:52Yes. And when they tried to rescue the knockout neurons with this R81C mutant, it only partially worked. It didn't fix the shape of the neurons effectively at all. But here's the real twist. R 81c isn't just slow.
11:03It might actually be toxic. Toxic, how? Well, when they stress the cells chemically, they used a compound called fapsi gargin to induce ER stress. The neurons with the R2NC mutation had a hyperactive stress response.
11:15They panic more than the neurons that had no UFM one at all. Whoa. So having a broken part is somehow worse than having no part, at least when things get tough. In terms of the stress response, yes. It suggests that the mutant protein is gumming up the works.
11:28Perhaps it's trapping enzymes or confusing the signaling pathways, leading to an aggravated purec response. It's not just that the brake is on. It's like the driver standing on the brake with both feet.
11:39Okay, so we have a clear picture of the disaster. No UFM1 or broken UFM1 leads to a traffic jam in the ER. The cell pulls the emergency brake via the purek pathway. Protein production stops. The neuron stays tiny and unconnected, and brain function suffers.
11:58That is the summary. It's a cascade of failure starting from one single missing tag. So if the problem is that the cell is panicking and pulling the emergency brake, Can we just tell the cell to chill out?
12:08Like can we just cut the brake line? That is exactly what the researchers ask next. They look for a way to inhibit that pure arc pathway because if the ribosomes are stalling, maybe we can force the cell to ignore it and keep building anyway.
12:19And they found a candidate in Trazodone. Trazodone, the antidepressant. The very same. It is a clinically used drug, chemically unrelated to SSRIs, but known to inhibit the perky UPR pathway. It's been shown to boost protein syntases in other models of neurodegeneration, like Alzheimer's and Prion disease.
12:38So they gave the neuronal culture some trazodome. What actually happened? The results were striking. In the neurons expressing that R81C mutant, the patient variant transodone treatment, fully restored protein translation.
12:50Wait, fully Back to control levels. The F1 McCat signal lit right up. The factory started running again. That is incredible. Did the neurons grow back? Did the tree fill out? Well, this is where we have to be scientific and nuanced.
13:04Trazodones significantly increase the number of synapses. The connectivity came back. But I hear a but. But it did not fix the dendritic branding complexity. The trees were still small, even though they had more leaves, so to speak.
13:16The actual skeleton of the neuron remained stunted. It's fascinating. So the drug could fix the connection count, but the physical stunting of the skeleton was irreversible. It suggests that some defects are structural, and perhaps locked in early in development.
13:29Once the tree has grown, you can't easily change the trunk. But others, like synapse number, are plastic. They can be modified much later. Still, increasing synapse numbers in a condition defined by a lack of connections.
13:42That's a huge win, isn't it? It is a major proof of concept. It identifies the UPR and PRK pathway as a druggable target. We can't easily go in and fix the genetic mutation in every neuron of a human brain yet.
13:56Gene therapy is still very difficult for these global brain conditions. But if we can treat the consequence of the mutation, the stress response. We might be able to restore a lot of function. It's like we can't fix the broken label maker, but we can tell the foreman to stop shutting down the factory every time he sees a mistake.
14:13And just keep production moving. Exactly. Just bypass the panic. So what does this mean for the big picture? For you listening right now. Why does this really matter? If we connect this to the bigger picture.
14:23It reinforces a major shifting paradigm in treating genetic diseases. For a long time, we thought broken gene equals broken organism. Fix a gene, or you're out of luck. But the steep dive shows that the pathway from gene to disease has steps, and those steps can be hacked.
14:40Precisely. The UFM1 tag is a quality control manager. When it's gone, the ribosome stalls. But the major damage is largely caused by the cell's reaction to that stall. If we can modulate the reaction, we can salvage the system.
14:54It turns a genetic dead end into a manageable condition. It's really helpful. And it highlights the absolute importance of basic biology. Without understanding UF milation and the perier pathway, without doing that fundamental, how does this work research, we never would have guessed that an old antidepressant could help a rare genetic microcephaly.
15:13It really makes you wonder what other drugs are sitting on our shelves right now, just waiting for us to understand the biology well enough to use them in a totally new way. That is always the dream of drug repurposing.
15:22It speeds up the timeline for therapy from decades to years. So here is the take-home message I'm getting today. This tiny protein tag UFM1 acts as a safeguard for the engine room of the neuron. Without it, or with a faulty version like R81C, the factory shuts down in a massive panic, but that panic is a biological switch we can actually flip.
15:44That's great summary of the mechanism. And here is a thought I want to leave you with. If we can restore synaptic numbers with a drug like Trezodone, even without fixing the core shape of the neuron, how much structure does a brain actually need to function?
15:59That raises a really fascinating question. We always assume form follows function, that you need the big complex tree to have the complex thought. But if function synaptic firing can be restored in a stunted neuron, maybe the brain is far more adaptable than we give it credit for.
16:14Can a small factory produce just as much as a big one if you just keep the lights on and the line moving? That is definitely something to mull over. This episode was based on an open access article under the CCBY 4.0 license.
16:25You 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. If you'd like to support our work, use the donation link in the description.
16:38Now stay with us for an original track created especially for this episode and inspired by the article you've just heard about. Thanks for listening and join us next time as we explore more science based by base.
