TRDP with Perturb-seq in human fibroblasts found that manipulating TFs (EZH2, E2F3, STAT3, ZFX) reversed aging hallmarks, and EZH2 overexpression rejuvenated aged mouse livers.
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. Okay, I want you to picture a house.
0:11Not a brand new construction, but one that's been lived in for 50, maybe 60 years. The paint is peeling in the hallway, the plumbing rattles whenever you turn on the tap. The foundation is, you know, settled and cracked.
0:24That house is essentially an aged organ in the human body. Now, if I asked you to fix that house to make it new again, what comes to mind? A massive renovation project. You're thinking about hiring a general contractor, a demolition crew, bringing in new drywall, new pipes.
0:41In biological terms, that is exactly how we usually view the concept of reversing aging. It's complex. It's messy, high risk, and very, very expensive. Right, it feels like a gut renovation. Yeah. But what if I told you that you didn't need the crew?
0:55Okay. What if the house wasn't actually broken, structurally speaking, but just confused? Confused? Yeah, what if you could walk into that living room, flip a single light switch on the wall, and suddenly the paint unpealed itself.
1:07The pipes sealed themselves. The house just remembered how to be new. That sounds less like biology and more like, well, magic. But that is the provocatively simple premise of the paper we're looking at today.
1:21We are diving into a study involving an aged liver, scarred by time, filled with fat, stiff from fibrosis, a truly geriatric organ, and by injecting a single factor, just one, protein researchers reverse the equivalent of decades of biological aging.
1:38In just 3 weeks. In 3 weeks. And crucially, they did it without turning the tissue into a tumor. Which is the plot twist we've all been waiting for in this field. Exactly. This is a breakthrough in what they are calling transcriptional rejuvenation.
1:51The idea isn't to demolish the house and rebuild it. It's to upgrade the software, so the house fixes itself. Exactly. We're looking at a fundamental shift from, you know, hardware replacement to software debugging.
2:04And today we're going to tear apart the computational platform that figured out which line of code to fix. But before we get into the mechanics of this biological time machine, we need to recognize the architects.
2:15This isn't just theory. No, this is hard data from a team at UCSF and the biotech company, Genevity. So who are we celebrating today? Yes, today we are celebrating the work behind the paper titled systematic identification of single transcription factor perturbations that drives cellular and tissue rejuvenation.
2:35It was published in PNAS in January 2026. Okay. And the heavy lifting was done by Janine Sang, Stack, Jesh, and Zang, Taranagayeva, their colleagues. And it's a really dense paper. It's right at the intersection of heavy computational biology and old school mouse models.
2:53So let's set the context here. Our listeners, they know the longevity space. They've heard of Yamanaka factors. We've even covered them before. The Nobel Prize-winning discovery. Four factors that can turn an adult cell back into a stem cell.
3:05So why isn't that the end of the story? Why are we still hunting for these switches? Because Yamanaka Factors art of Ledgehammer? I mean, yes, they rejuvenate the cell, but they do it via a process called Differentiation.
3:17Okay, break that down for us. Think of a skin cell. It has a very specific job, protect the body, whole things together. Right. Yamanaka factors don't just make it a young skin cell. They make it forget it's a skin cell at all.
3:29It reverts to an embryonic state. Ah. Now, if you do that in a Petri dish, it's a scientific miracle. But if you do that in a liver inside an animal, worse, teratomas, these tumors made of just confused tissue.
3:45Oh, wow. You could end up with a liver that starts growing Keith or hair, because the cells have total amnesia. It's the nuclear option. It just wipes the slate too clean. So the goal here was to find a Goldilocks zone.
3:56Can we make the cell young without making it confused? Precisely. Can we keep this identity? I am a liver cell, but just strip away the age. That was the hypothesis. The idea was that the human genome, with its, what, 20,000 genes must contain other transcription factors.
4:12So other proteins that control gene expressions. Yes, ones that can decouple age from identity. They wanted to find the proteins that simply restore the operating system to a previous version without uninstalling all the apps.
4:25But 20,000 genes is a staggeringly big haystack. If you're looking for a needle, how do you even start? You can't just test them one by one in mice. No, you'd run out of mice before you ran out of jeans.
4:36And that's where the methodology really shines. They built something called the transcriptional rejuvenation discovery platform or TRDP. This is the engine of the paper. They didn't start with mice. They started with math.
4:48Okay, walk us through the workflow. How does this TRDP find a rejuvenation switch? It functions like a funnel? Step one, data. They took massive data sets a gene expression from young cells, in this case, early passage, fiber blasts, and old cells.
5:03So they created a digital map of youth and a digital map of old age. A high resolution one, yeah. They can see exactly which genes are turned up and which are turned down in the aged state. Got it. Correct.
5:13Step 2 is the Ensilica screen, the computer part. They used bioinformatics to look at the differences between those maps. But not just which genes were different. They looked for the drivers. Exactly. Yeah. They analyze.
5:28So they're basically going into the fuse box of the cell and flipping switches on and off to see what happens. But here's the kicker. Yeah. They did this in a pooled format, and then use single cell RNA sequencing as the readout.
5:39Ah, so they could see, cell by cell, exactly what happened to the gene expression profile when they tweaked candidate A versus Canada B. Yes. So instead of waiting months to see if a mouse lives longer, they get an immediate readout.
5:52Did this tweak make the gene expression look young again? It's high throughput discovery. That's it. And out of those 200 suspects. Four emerged as the clear winners. These 4 factors successfully reverse the aging signature without erasing the cell's identity.
6:08All right, let's name names. Who are the big four? First, we have E2F3. When they turn this up, overexpressed it, it kicked the cell cycle into gear, it essentially woke the cell up and told it to start dividing again.
6:19Okay. Makes sense. Next. Second is EZH2. This is really the start of the show, and we'll talk about it a lot. It's a histone metal transfer race. Let's pause on that jargon. His stone, methyl transferase.
6:31This deals with the physical packaging of DNA, right? Yes. Think of your DNA as a library. His stones are the schools the DNA is wound around. Methylation is like putting do not read tape over certain chapters.
6:44As we age, that tape falls off or gets put in the wrong places. The library gets messy. So information that should be hidden gets read and information you need gets lost. An easy H2 helps reorganize the library.
6:56It ensures the cell only reads the chapters it's supposed to. So it restores the epigenetic landscape. It tidies up the room. It restores the silence. Youth is organized, aging is noisy, easy H2 quiets that noise.
7:10Fascinating. And the other two? The other 2 involve turning genes down. They repress stat three, which is a major driver of inflammation. Well-known bad guy in aging. A very well known one. And they repressed sad effects, which is linked to stem cell renewal regulation.
7:22So we have E2F3, EZH2, stat 3 and ZFX. Now the paper claims these worked in vitro in the dish, but work can mean a lot of things. Did the cells just look younger on a spreadsheet or were they actually healthier?
7:35They were functionally healthier. This is the crucial validation step. It wasn't just that the RNA looked young. The mitochondria, the power plants ramped up activity. Okay. Proteospaces improved, meaning this all started cleaning up damaged proteins.
7:49out the trash. Exactly. They also saw a reduction in senescence. Those zombie cells that stopped dividing and poisoned their neighbors were reduced. And all this without the identity crisis, the fiber blasts remained fiber blasts.
8:04That's the claim. They didn't turn into stem cells. But we all know the joke. We've cured cancer and mice a 1000 times, or, in this case, in a petri dish. Moving to a living animal is the valley of death for most biotech.
8:18Did they bridge that gap? They did. And this was the main event of the paper. They took this logic into an in vivo mouse model. And they had to pick a champion, right? They didn't use all 4 factors. No, they focused on EZH2.
8:31Why that one? Why not the other? It was strategic. E2F3 pushes the cell cycle very aggressively. That smells a little like cancer risk. A bit scary. A bit. And stat 3 is already well known in liver inflammation, so it's less novel.
8:44Easy H2, sat in this sweet spot. Novel for rejuvenation. But with a safety profile that looked manageable. So set the scene for the mouse experiment. They took 20 month old mice. And that is a senior citizen in mouse hears, rough coat, slow movement, tired organs.
9:01Okay. They injected them with an AAV 8 virus. Which acts as a delivery truck. Right? The virus delivered the instruction to overexpress easy H2, specifically targeting the liver. And then they waited. But here's the shocking part.
9:14The only waited 3 weeks. Three weeks? That sounds incredibly short. Most aging studies I read about span months or even years. That speed is what makes these results so staggering. In just 21 days the gene expression profile of these old livers shifted dramatically back toward youth.
9:29How much? The authors calculated it reversed about 8 months worth of aging. Okay, hold on. 8 months in a mouse. Mice live, what, 2 years? About that yeah. That's a 3rd of their life. That's like taking an 80-year-old human and physically reverting their liver to how it looked when they were 50 in 3 weeks.
9:45The math holds up. And it wasn't just genetic data. The physical structure of the liver transformed. You know how old livers get fatty? Steetosis. Sure. The easy H2 treatment, reduced that fat by at least 50%.
10:00And what about fibrosis, the scarring? Also reduced by halves. Old livers are stiff. These became supple again. Okay, what about function? I mean, a pretty liver is nice, but does it work any better? They texted glucose tolerance.
10:12Old mice usually process sugar poorly. They're borderline diabetic just from age. The treated mice process glucose almost as well as young controls. I have to play devil's advocate here. I'm listening to this and I hear rapid cell growth, reversing aging in weeks, overexpressing genes, my alarm bells are ringing.
10:30This sounds like the perfect recipe for cancer. You are right to be skeptical. Rejuvenation and cancer are arguably 2 sides of the same coin. They both involve unlocking cell growth potential. The researchers were hyper aware of this.
10:43So how do they check? Did they just look for tumors? They went deeper than that. They compared the gene signature of the EZH2 treated livers against databases of known liver cancers. To see if the programs looked similar.
10:56See if the rejuvenation program looked anything like a tumor of genesis program. No resemblance. Cancer is chaotic. Cancer cells rewire their metabolism and stress responses in very specific disorderly ways.
11:07EasyH2 didn't do that. In fact it did the opposite. How so? It reversed something called mesenchymal drift. Mess and chaimal drip. That sounds like a sci-fi novel, decode that for us. It's a hallmark of aging.
11:20Over time, epithelial cells, the cells that line your organs and give them structure, start to lose their crisp, defined edges. They get sloppy. Okay. They drift toward a mezenchiml state, which is fibrous and migratory.
11:33Like a brick wall turning into a pile of clay. That's perfect analogy. Yeah. And cancer loves messenchymel states because it helps cells move and invade. But easy H2 reverse this drift. It forced the cells to crisp up and remember their epithelial structure.
11:48So it restored order, whereas cancer creates chaos. That's a critical distinction. It's not just growth, it's structured restoration. Correct. However, and this is a big, however. We have to look at the limitations.
12:01Of course. Hit us with the reality check. This was a short term study. Three weeks. We know from other literature that if you overexpress EZH2 at high levels for a long time, it can be oncogenic in certain contexts.
12:15It's not a harmless molecule. So you can't just tape the switch to the on position and walk away. Probably not. The future of this therapy likely isn't a permanent gene edit. It's probably transient pulse.
12:25What you mean? Imagine a spa weekend for your liver. You go in, take a drug that boosts easy H2 for a week to clear out the junk and reset the up a genome and then you stop. That aligns with the hidden run approach we're seeing in other longevity research.
12:40But looking beyond the liver, what does this tell us about aging in general? It implies that aging is a conserved molecular program. Meaning it's not just random damage. Right. We often think of aging like a car rusting.
12:52The rust is random. But this data suggests aging is more like a computer operating system getting bogged down with bloatware and corrupted files. The hardware is fine. The instructions are just messy. And if you can run a system restore, which is what easy H2 seems to be doing, The machine works perfectly again.
13:11That's a profound shift in mindset. It means the regenerative capacity is just sitting there dormant inside our 80-year-old cells. They just forgot how to access it. And the TRDP platform gives us a way to find the password.
13:23That's the real value of this paper. It's not just about easy H2 or the liver. You could run this same computational pipeline for the heart, the brain, or the kidneys. A universal remote control for organ regeneration.
13:36Theoretically, yes? You've built the discovery engine. Now they just need to find the specific fuel for each organ. Let's zoom out to the human level. If this translates, and that's always a big if. What does the medical landscape look like in 20 years?
13:48It moves us toward personalized organ maintenance. Right now, if your liver fails, you wait for a transplant. In this future, you might go in for a liver update at age 60. Like getting your oil change.
13:59Exactly. A targeted gene therapy injection that resets the epigenetic noise in your liver, scrubbing away 20 years of wear and tear, and buying you another decade of function. And then maybe you come back 5 years later for your kidneys.
14:12That's the idea. It reframes medicine from fixing what's broken to maintenance of the healthy. But it also raises a philosophical question. If we can reset organs piece by piece. Where is the limit? That is the question.
14:27If you can keep the liver young and the heart young and the vasculature young. Yeah. Do we hit a hard limit on lifespan or do we just keep fixing the car indefinitely? It's the ship of Theseus paradox applied to biology?
14:39If I replace or refurbish every part of the boat? Is it still the same boat? And does the boat ever have to sink? Right. The science here suggests the boat is much more resilient than we ever gave it credit for.
14:49We just need to learn how to speak its language. And thanks to this paper, we've learned 4 new words in that language, EZH2, E2F3, state 3, and ZFX. A small vocabulary, but a powerful start. This episode was based on an open access article under the CCBY 4.0 license.
15:07You 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.
15:20Now 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 base by base.
