In mESCs and defined in vitro assays, EPOP and MTF2 stimulate PRC2 methyltransferase activity and promote de novo H3K27me3 deposition with GA- or GCN-rich DNA preference.
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. Imagine having a master off switch for your DNA.
0:10A molecular machine inside you right now that dictates whether a newly forming cell becomes a piece of your brain, a section of your liver, or a muscle fiber in your heart. Right. It essentially the ultimate architect of your body.
0:24Exactly. Now, imagine that for years, scientists studying this incredible machine thought a vital part of it was broken. They thought it was a break, holding the machine back. Yeah, that it was actively inhibiting the whole process.
0:36But in reality, it was the turbocharger all along. So welcome to the deep dive. Today, we are unpacking a massive biochemical mystery about how your cells permanently turn off specific genes during development.
0:48And before we get into the weeds, today we celebrate the work of Jeffrey Granite, sanction Lou, Danny Reinberg, and their incredible research team, who have really advanced our understanding of epigenetics.
0:59Yeah, their 2026 biochemistry research article in PNAS, which is titled EPOP and MTF2, activate PRC2 activity through DNA sequence pestericity. It basically rewrites the textbook on this. It really does.
1:13The mission of this deep dye is to understand how this vital silencing machine, which biologists call polycomrepressive complex too. Or PRC 2. Right, PRC 2. We're looking at how it actually navigates the incredibly complex landscape of your genome, because it doesn't just wander around randomly.
1:30It uses a highly specific kind of rotating cast of copilots to find exactly which genes need to be shut down. And to really grasp what Granite and his team discovered. We need to completely change how we visualize DNA.
1:43I mean, most of us grew up picturing the human genome as a static, perfectly organized blueprint. But I want you to imagine your DNA, not as a blueprint, but as a sprawling ancient library. A library that contains, you know, every single instruction manual needed to build and operate you.
1:57Exactly. Every single cell in your body has the exact same library, the exact same collection of books. But obviously, a heart cell doesn't need the instruction manual for making tooth nail. No, if it starts reading that book, you get biological chaos.
2:09Right. So the cell employs a librarian, a very strict, incredibly efficient librarian whose entire job is to go through specific wings of this vast library, lock the doors, and stamp 1000000s of books with a heavy chemical padlock.
2:23And in the microscopic world of your cells, that strict librarian is the PRC2 machine. And that padlock it uses is chemical tag called H3K 27E3. Which is quite the mouthful. I know, it sounds like absolute alphabet soup, but biochemically, it's just a specific methyl group attached to a protein spool that your DNA wraps around.
2:43Once PRC2 stance that tag onto the spool, the gene is locked down, it is completely silenced. Okay, let's unpack this. Because before we get to the massive discovery in this paper, we have to look at the massive paradox that set the researchers on this path in the 1st place.
2:57The librarian, PRC2, has to find specific starting lines in the genome to begin silencing genes. Yeah, it doesn't just start stamping anywhere, right? Right. Those starting lines are called nucleation sites.
3:10You can think of them as specific aisles in the library where the librarian knows to begin, and these aisles are often marked by regions called CPG islands, which are loaded with specific repeating genetic sequences, specifically GCN and GA tandem repeats.
3:25Okay, so a lot of repeating letters. Exactly. And to find these specific repeating sequences, the core PRC2 machine uses accessory proteins. You can think of them as specialized tool the librarian carries, 2 very well known tools are called MTF 2 and JRD.
3:41So MTF 2 is like the librarian's aislefinder. But then researchers kept noticing this other accessory tool, a protein called EP op. And EPIB is highly clustered right at these starting lines, right in these nucleation sites where the heavy stamping needs to happen.
3:56Yeah, it's everywhere the machine is working. But here is the paradox. Previous studies looked at EPOP and concluded it was an inhibitor. They thought EPO's job was to actively restrain the librarian from stamping the books.
4:07Which created a severe logical bottleneck in the field of epigenetics. I mean, EPOP is heavily enriched precisely in the locations where PRC2 needs to be working at maximum capacity to start the silencing process.
4:20I mean, logically, that makes 0 sense. If you need a massive amount of work done in a specific aisle of the library, why would you send an assistant whose only job is to aggressively pull the ink pad away from the librarian?
4:32Right, exactly. Why would evolutionary biology waste energy putting a break exactly where the gas is needed most? It's a brilliant question, and it is the exact logical knot that Granette Lou and the team wanted to untangle.
4:45They suspected that the previous studies were, well, misinterpreting EPiopes function because observing these delicate biochemical processes inside a living cell is incredibly difficult. A living cell is just a chaotic, dynamic soup of 1000000s of interacting molecules.
5:00Wait, if a living cell is that messy and chaotic, how do you even begin to isolate this one specific librarian machine to see what its tools actually do without breaking the whole thing? You have to strip away the noise.
5:12You have to take the experiment in vitro meaning outside the living organism and into the test tube. Ah, okay. Yeah, Grenett and his team realized they needed to build the PRC2 machine entirely from scratch in a perfectly clean environment to see what EPI was truly doing.
5:28But how do you build a microscopic biological machine from scratch? I mean, you can't just order the parts on the Internet? Actually, in modern biochemistry, you sort of can. Wait, really? Yeah. They use recombinant proteins.
5:39This means they took the specific genetic instructions for the individual pieces of the PRC 2 machine, inserted them into easily controlled cells like insect cells or bacteria, and force those cells to manufacture massive quantities of just those pure proteins.
5:54Oh, wow. Then they purified them and painstakingly assembled the core PRC2 complex in a test tube. Okay, so they have a perfectly clean custom built librarian. But a librarian needs books to stamp. You can't just throw naked strands of DNA at it, right?
6:10Because in our bodies, DNA is packaged. Precisely. In our cells, DNA is tightly wound around protein spools called histones. This wrapped structure is called chromatin. So the researchers artificially constructed chains of DNA wrapped around these histone spools, creating what they call oligonucleusum arrays.
6:31So it's essentially a pristine, highly controlled mini library. Exactly. They gave the strict librarian a stack of completely blank, perfectly formatted books, and basically said, let's see how fast you can stamp these.
6:44That's amazing. It is. They performed in vitro histone, methyl transphrase essays. First, they let the core PRC 2 machine work by itself. Then they added the MTF 2 aislefinder tool. And finally, they added the EPOP tool, the supposed break.
6:58And what happened? Did EPOP hit the brakes? The results completely overturned the old assumption. When they added NTF2, the stamping activity increased, which was expected. But when they added EPOP, the activity became incredibly robust.
7:11Both MTF 2 and EPOP directly and powerfully stimulated PRC 2's activity. So EPOP is not a break at all. It's a massive gas pedal. Or, sticking with our metaphor, EPOP is an automated turbo stamper for the library.
7:24Yeah, exactly. Away from the chaotic traffic of a living cell. When you just look at the raw mechanics, EPOP is clearly accelerating the silencing process, not stopping it. But this brings up a crucial scientific hurdle.
7:37Proving EPOP is a gas pedal and a clean test tube is fantastic, but it does not explain why the older studies got it so wrong when looking at living cells. Right, because if you want to rewrite the textbook, you have to prove why the old textbook is wrong.
7:50Exactly. To truly rewrite the textbook on EPO, Granite and his colleagues couldn't just stay in the test tube. They had to go back into the messy living cell and prove exactly why the old experiments were creating an illusion.
8:04And honestly, the experimental design they used here is one of the coolest things I've read. To figure this out, they use mouse embryonic stem cells, but they utilize something called an ED rescue system.
8:14Can you walk us through what that actually is? Yeah, it is a phenomenal genetic trick. ED is a core structural pillar of the PRC 2 machine. Without the ED protein, the entire librarian complex physically falls apart.
8:27Yeah, and when that happens, the cell loses its ability to place those chemical padlocks. Over time, as the cells divide, the genome is wiped completely clean of that specific silencing mark. So by knocking out this one pillar, they force the cell to unlock every single door in the library.
8:45They create a total, completely bare blank slate. Yes. And then here is the magic trick. They flip a genetic switch and turn ED back on. The librarian machine rapidly reassembles itself, and the researchers get to watch it go to work de novo, meaning entirely from scratch.
9:02And if you're listening to this, this blank slate rebuilding process is essentially what your own cells had to master when you're just a developing embryo, trying to figure out which parts the genome to lock down to create your different organs.
9:14Exactly. So the researchers initiate this blank slate rebuilding process. But in one group of cells, they specifically delete the EPOP protein. They want to see what fails to happen during the rebuilding phase when the librarian is missing that specific tool.
9:27Right, because the old studies use what are called steady state sales. In those old studies, the library was already built, the padlocks were already mostly in place. Correct. In those old steady state cells, if you remove EPO, the cell compensates.
9:40Another tool called Jared 2 steps in to try and fill the gap, which creates a messy, overlapping signal that made EPP look like an inhibitor. Ah. But in this de Novo blank slate system, you watch the actual real-time construction process.
9:56And they found 2 distinct things when EP pop was missing. First, the initial recruitment to the DNA was perfectly fine. PRC2 still found the correct aisles in the library. So EPF doesn't help the librarian find the aisle.
10:08MTF2 does that. Right. But here's the critical 2nd finding. Without EPOC, the actual deposition of those silencing marks, the actual locking of the padlocks dropped by about 50%. Wow, 50%. Yeah. PRC2 found the right spot, but it was incredibly sluggish at actually doing its job.
10:24So EPLP really is the turbo stammer. MTF 2 drives the truck to the site or guides the librarian to the exact right shelf, but without EPOP, the librarian's arm gets tired incredibly fast. EPOP is a tool that makes sure the lock actually snaps shut.
10:41That's a great way to put it. MTF 2 is heavily involved in binding the complex to the DNA, while EPLP is absolutely crucial for stimulating the catalytic activity once the machine arrives. But the genome is huge.
10:53I mean, there are 1000000000s of base pairs. Do MTF 2 and EPLP work in the exact same aisles or do they have preferred territories within the library? Well, this is where the researchers had to look closely at the secret code written in the DNA itself.
11:05Earlier, we mentioned that the starting aisles for silencing are rich in specific repeating sequences, GCN repeats, and GA repeats, to test if the librarian's tools care about these specific letters, the researchers built custom dinucleosomes.
11:19Okay, let's visualize that for a second. A nucleosome is essentially a spool of thread. The spool is the histone protein, and the thread is the DNA. So a dinucleosome is just two spools next to each other.
11:29Precisely. Two spools, connected by a tiny 40 base pair string of DNA called linker DNA. Which is incredibly short. I mean, 40 base pairs is just a tiny fraction of the genetic code. It is, but it's enough to test a hypothesis.
11:43Into that tiny linker string connecting the 2 spools. They inserted specific genetic codes. They made one version with just random DNA letters. They made a 2nd version heavily enriched with GCN repeats, and a 3rd version heavily enriched with GA repeats.
11:58And then they unleash the PRC 2 librarian on these custom spools. Yes. And when PRC 2 was paired with the MTF 2 tool, its silencing activity doubled on the GCN Rich DNA. It recognized that specific repeating sequence and worked much harder.
12:12But when PRC2 was paired with the EPOP tool, its activity doubled on the GA rich DNA, so they have distinct territory, EPOP binds tighter and works much harder specifically on those GA genetic sequences.
12:25What's fascinating here, and what is going to keep biochemists up at night, is that this completely challenges our structural understanding of EPP. From everything we know, EPP is largely an unstructured protein.
12:36Under a microscope, it basically looks like a loose floppy piece of string. Okay. It has no known DNA binding motifs. It doesn't have the biological fingers we usually see in proteins that are designed to read DNA letters.
12:48Here's where it gets really interesting. And frankly, kind of mind-bending. Wait, if EPAP doesn't have the biological hardware, like the fingers to physically read the DNA letters, A, C, T, and G, how on Earth?
13:02Is it feeling the GA bumps on the DNA track? How is a blind protein reading the code? That is the $1000000 question. And Granite and his team found a vital clue. They describe that EPOP's preference for GA sequences only works when the DNA is wrapped up in those histone spools.
13:17So if you just give EPIP a naked, unwrapped string of GA rich DNA, it doesn't care. It doesn't turbocharge anything. Exactly. On naked, unspooled DNA. EPIP is effectively blind to the sequence. It shows absolutely no preference.
13:32It only activates its turbo boost when that specific GA sequence is physically wrapped around a nucleus spool. That is wild. It really is. EPOP requires the three-dimensional architecture of the package DNA, not just the two-dimensional text.
13:47Let make sure I'm grasping this. Wrapping DNA around a spool creates physical tension. Different letters bend differently. So a string of GA repeats wrapped tightly around a pool is going to create a very specific physical bump or a specific structural shape compared to a random sequence.
14:03It's like a librarian who can't actually read the titles on the spines of the books, but they know they are in the correct aisle just by feeling the physical texture and shape of how the books are stacked on the shelves.
14:12That is a brilliant analogy. Yeah, EPOP appears to be reading the structural geometry of the chromatin, not the sequence of the letters itself. And if we connect this to the bigger picture. This solves a massive puzzle in epigenetics and developmental biology.
14:27We've always wondered how a general universal silencing machine like PRC2 knows exactly how to customize its behavior for all the different developmental pathways. Right. Like, how does it know to sculpt a liver cell today and a white blood cell tomorrow?
14:43Because the core machine, the strict librarian doesn't actually change, right? It's the exact same machine in every cell. Right. But the tools change. The accessory proteins like EPOP and MTF 2 are expressed at different levels depending on the cell type.
14:57So a cell can essentially swack out the tools the librarian carries to dynamically change which genes get logged down. Precisely. The researchers describe the system as acting like a rheostat, a biological dimmer switch for selfie decisions early in embryonic development.
15:12For example, if a very early developmental cell is expressing a massive amount of EPR, but very little MTF2, the PRC2 librarian is going to robustly shut down all the genes located near those GA repeats, but it might leave the GCN repeat regions relatively online.
15:30Oh, wow. By subtly swapping these copilots, the cell shifts the balance of silencing power across the entire library, which ultimately determines what kind of specialized tissue that cell will become. So by simply changing which accessory proteins are floating around in the nucleus, the cell can instantly reprogram the entire physical landscape of the genome.
15:49Exactly. It's an incredibly elegant system. So what does this all mean for us? If we pull ourselves out of the test tubes and the custom DNA spools, Granite Lou and the team have completely rehabilitated EP Ops reputation?
16:01It is not a cellular brake holding back development? Not at all. It is a highly specialized, sequence specific gas pedal. It drives the heavy silencing of GA rich regions of our DNA. And in doing so, it acts as a master sculptor, helping to physically shape who we become at a fundamental cellular level.
16:20It really is an elegant, beautifully complex system. And I think it leaves us with one final, incredibly provocative thought to ponder. We just discussed how EPP has no physical way to read the genetic text, and that it seems to be reading the 3D geometry, the physical shape of the wrapped GA sequences.
16:38But let's take that one step further. We know that physical forces, environmental stress, and even cellular crowding can alter the mechanical tension inside a nucleus. Wait, are you saying the physical environment of the cell could warp the shape of those DNA spools?
16:51Exactly. If EPP is reading the physical tension or the flexibility of those wrapped GA repeats rather than the letters, then what happens when environmental factors change that tension? Could the physical physics of the cell, the literal mechanical pressure dictate where EPOP binds and therefore dictate which genes get silenced?
17:09That's fascinating. It suggests that our cellfates might not just be driven by a chemical code, but by the physical mechanical biology of the genome itself. The idea that the strict librarian isn't reading the books, but is reacting to the physical pressure and tension of the library shells themselves, that is wild.
17:29The shape of our genome dictating the function of our genes. That is something that is going to stick with me for a long time. It's definitely a whole new frontier for biology. Well, this episode was based on an open access article under the CCBY 4.0 license.
17:42You 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.
17:54Now 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.
