This study shows that CpG methylation at proximal HBG promoters causally enforces perinatal silencing and that targeted epigenome editing can reverse that silencing in cell models and primary erythroblasts. UHRF1 and the methyl-CpG reader MBD2 are key mediators of repression.
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 your body has a perfectly functioning backup system for a vital protein, right?
0:13But it permanently locks that system away just days after you're born. Even if your primary system is failing. Exactly, even if it's causing a severe life-threatening disease, it's um, it's just this incredibly frustrating biological reality for 1000000s of people.
0:30It kind of like having a pristine backup generator in your basement during a permanent freezing blackout. You know, the power is right there. The generator works flawlessly, but the door to the basement is welded shut.
0:41You absolutely cannot access it, no matter how desperately your body needs it. Yeah, and that structural lockdown is really the defining tragedy of beta chemoglobinopathies. For listeners familiar with conditions like sickle cell disease and beta thalacenia.
0:55You know the adult beta globin gene called HBB is mutated. Right. So it produces defective red blood cells. Exactly. But these patients actually possess completely healthy fetal hemoglobin genes, HPG1 and HPG2.
1:08The backup generator. The backup generator, yeah. The issue is just that human biology naturally silences these fetal genes perinatally, so right around the time of birth. The backup is flawless, but the cellular machinery seals it away just as the patient transitions to using the defective adult version.
1:24Man, and picking the lock on that basement door has been like the holy grail of hematology for decades. So before we get into how a new breakthrough actually picks that lock, we need to acknowledge the minds behind it.
1:36Today we celebrate the work of Henry W. Bell, Rupeng Feng, Mixel J. Weiss, Merlin Crossley, and their collaborative teams at UNSW, Sydney, and St. Jude Children's Research Hospital, who have advanced our understanding of gene silencing in beta hemoglobinopathies.
1:51Yes, and they're open access research titled Removal of Promoter CPG methylation by Epigenome editing, reverses HPG silencing, was published in Nature Communications in July 2025. Which is super recent.
2:03Yeah. And what makes this paper so compelling is that it finally solves a mystery that has divided researchers since, well, since the late 1970s. Wow, the 70s. Okay, let's contextualize that mystery for a moment.
2:14As a quick refresher for our listeners, methylation acts as a tiny molecular, do not read tag on the DNA. You attach a methyl group to a cytocene base, and it generally tells the cell to shut that gene down.
2:29Right, and we've known about this correlation since 1979, actually, when scientists noticed heavy methylation on silenced globin genes in chickens. Chickens, go figure. But the problem historically wasn't that we didn't know about the tags.
2:41It's that our only tool to remove them was essentially a chemical sledgehammer. Drugs like 5 visa acidine. Well, they do write. They remove those methyl tags and wake up the fetal hemoglobin, but they do it globally.
2:53They just strip methylation across the entire human genome. Which, I mean, sounds pretty messy. Highly toxic, yeah. Because if you consider what methylation actually does, it keeps latent retroviruses suppressed.
3:05It keeps onca genes turned off, it maintains cellular identity. So stripping it away and discriminately is bad news. You're turning on the backup generator, but you're also basically setting fire to the rest of the house.
3:17Exactly. Which creates a massive blind spot for researchers. Because of that blunt instrument approach, the field debated whether the local methylation right at the fetal hemoglobin promoter was the direct physical cause of the silencing, or if it was just a correlative byproduct, right?
3:33Like, we saw the methyl tags when the gene was off, but we couldn't prove the tags themselves were flipping the switch. Right, because other repressor proteins were always swarming the area, too. You couldn't tell who was actually in charge.
3:43So how did they finally separate correlation from causation? Well, the research team took a step back and ran a massive forward genetic screen using CRISPR Cast 9. They used HUDF 2 cells, which are an adult type red blood cell precursor line.
3:58Okay, so these cells normally express the adult HBB gene and keep the fetal genes shut down. Correct. And the team systematically knocked out 1000s of individual genes to see which absence would suddenly wake the fetal hemoglobin backup.
4:11Among a few known players, the screen flagged a protein called UHR F1. U-H-R-F-1. So they slag this protein, but if I remember my cellular biology correctly, UHRF1 is essentially a maintenance worker, right?
4:25Yeah, that's a good way to put it Like, it partners up with DNA methyl transfer races to copy existing methylation patterns across the genome every time a cell divides. It acts as the global architect for maintaining those repressive tags.
4:38So when the researchers disrupted UHRF1 in these adult type cells, they observed global DNA demethylation. The maintenance worker was off the job. The methyl tags faded away as the cells divided, and as a result, the fetal hemoglobin genes roared back to life.
4:53But wait, if UHRF one maintains methylation everywhere, aren't we just back to square one? What do you mean? Well, knocking it out is just a genetic version of the 5A's a cedine problem, isn't it? It causes the same messy global chaos as the old chemo drugs.
5:08We still haven't proven that the tags directly on the field gene are the sole key to the lock. Spot on. And that is the exact realization that forced the team to pivot to a groundbreaking precision tool epigenome editing.
5:21Oh, this is a good part. Yeah, to prove local causation, they needed to surgically alter the epigenetic tags at one single genetic address without touching a single base pair anywhere else in the genome. So they built a specialized fusion protein called Tet TV 4.
5:38Let's break down how Ted TV 4 actually functions because this is where the engineering gets so brilliant. They took a cast 9 enzyme, which is the standard CRISPR protein we normally use to cut DNA. And they essentially killed it.
5:49Right, De Cast Nine. Yeah, Deadcast nine. They mutated its cutting domains. It can still use a guide RNA to find a highly specific 20 letter DNA address, but if chemical scissors are broken, It just parks there.
6:01Then they attached a payload to that dead cast nine. They fused it to the catalytic domain of an enzyme called TT1. And what does TT1 do? In nature, TT enzymes are the body's native to methylators. They oxidize those methyl tags and initiate the process of erasing them.
6:16I love this. So a standard CRISPR is a pair of molecular scissors. This decast 9 Tet1 fusion is like sending a highly trained locksmith with exact GPS coordinates to spray rust remover on one specific gear rather than dunking the entire cellular engine in acid.
6:33That's a great analogy. You leave the surrounding machinery completely untouched, and that directed precision allowed them to run the ultimate test. They program this Tet TV 4 editor to target the promoters of the fetal hemoglobin genes in those U-Dep 2 cells.
6:49The adult type cells. Right. And the baseline expression of fetal hemoglobin in these cells was a microscopic 2.2%. But when they deployed the targeted eraser to strip the local methyl tags away, that expression skyrocketed to 86%.
7:03Wait, from 2% to 86%. That's, I mean, that's not just a subtle shift in the background noise of the cell. That is kicking the biological backup generator into full roaring overdrive. It's a massive targeted induction.
7:16But, you know, in science, proving you can turn something on is really only half the battle, to establish absolute ironclad causality, they needed to prove they could actively throw the switch in reverse.
7:26Okay. So they built the mirror image tool, instead of an eraser, they used a targeted pen called D3 AL, which is designed to write methyl tags back onto the DNA. This is the real aha moment for me when I was prepping for this deep dive.
7:41They didn't just break the lock. After successfully turning the fetal gene on and hitting that 86% mark, they sent the D3AL remephilator back to the exact same promoter. And they watched the fetal hemoglobin expression plummet all the way back down to 23%.
7:56It's totally reversible. Exactly. They essentially built a volume knob for a human gene. The expression directly and undeniably responds to the presence or absence of those specific local chemical tags.
8:08That's wild And as they map this out, they discovered the lock isn't operated by a single switch. It's highly nuanced. Right. It functions more like a combination lock. Yeah. Like they couldn't just demethylate one specific cytosine and expect the door to swing wide open.
8:21Yeah, when they used individual guide RNAs to strip the tag from just one specific CPG site, the gene only partially activated. By mapping the entire region, they found a strong negative correlation between expression and the methylation of 4 specific CPG sites located upstream of the transcription start sites.
8:41So full robust activation actually required clearing multiple tags in that upstream promoter region. Exactly. And for anyone wondering if this is just, you know, an artifact of immortalized lab cells, because we all know cell lines like U-Tep 2 can sometimes act a bit weird after years and dish, didn't stop there.
8:58No, they didn't. They validated this entire dynamic in primary erythral blasts. They basically grew red blood cell precursors directly from the CD 74 positive blood stem cells of healthy human donors. And applying that exact same TED TV 4 targeted demethylation to those primary donor cells yielded the same robust activation of the fetal genes.
9:18It confirms that this combination lock is a fundamental feature of human biology, not just a laboratory quirk. So we've established that the methyl tags are operating the lock. But here is where we need to look at the physical mechanics of the door itself.
9:32Earlier, we mentioned that there are repressor proteins swarming this area. Anyone who follows genomics knows about a protein called BCL Bran A. It's famous for being the primary repressor that shuts down fetal hemoglobin.
9:45So naturally, you would assume that if you erase the methyl tags and the gene suddenly activates, it's because you've evicted BC 11 A from the premises. I mean, that assumption makes intuitive sense. If the gene is transcribing massive amounts of protein, the repressor shouldn't be there.
9:58But when the researchers ran an assay to physically check what proteins were bound to the chromatin after demethylation. BCL Avenet was still sitting right there on the gene. I really want our listeners to absorb how bizarre that is.
10:11The gene is pumping out fetal hemoglobin at an 86% clip, but the main repressor protein known to silence it is still parked directly on the DNA strand. How is the DNA being read if the roadblock is still in the way?
10:25It's crazy, right? It forces us to rethink what BC 11A actually does, and it shines a spotlight on the real muscle of the operation, which is the MBDD tunerd complex. The nerd complex. Yeah. BCL 11 A might bind to the DNA, but it doesn't act alone.
10:41It serves as an anchor for this massive multi-protein co-repressor complex called nerd. The nerd complex physically remodels the chromatin. It decidillates the area and physically slides the nucleosomes together, essentially crushing the DNA so tightly that the cellular machinery can't read it.
10:57So BCL of NA is just the scout, but the nerd complex is the army that actually barricades the road. That's a great way to put it. But how does nerd know when to attack the chromatin? It relies on a specific subunit called MBD2.
11:09MBD2 is the translator. It has a highly socialized structural pocket designed to physically bind to methylated DNA. If the tags are there, MBD 2 binds to them, recruits the massive nerd complex, interacts with BCL 11 A, and the chromatin gets crushed closed.
11:24Which leads to just the most elegant molecular experiment in this entire deep dive. The researchers wanted to prove that MBD 2's ability to read those mental tags is the linchpin of the entire silencing operation.
11:37So they introduced a highly specific mutation into the MBD 2 protein, the Y 178 F mutation. Right. They altered a single amino acid in MBD 2's binding pocket. They basically blinded the translator. I love that Yeah, the mutated MBD 2 protein was fully functional in every other way, but it could no longer read or bind to the methyl tags.
11:57And the results of blinding that single reader protein are stunning. In the cells with this mutated MBD2, the fetal hemoglobin turned on at full blast. It didn't matter that the DNA was still heavily methylated.
12:09It didn't matter that BCL 11 A was still anchored to the strand. Nope. Because MBD2 couldn't read the tags, it couldn't stabilize the nerd complex, the entire massive co-repressor complex just falls apart.
12:19It proves definitively that local methylation is absolutely required for the repressive machinery to function. Without the methyl tag stabilizing the MBD 2 nerd complex. BCL 11 A is left stranded and impotent on the DNA, and the door swings wide open.
12:35So we have this incredibly elegant molecular mechanism that tags the lock, MBD2 is the key, and nerd is the heavy door. But if we connect this to the reality of clinical practice for patients suffering today, how does this epigenome editing approach actually compare to what's happening right now in hematology clinics?
12:52Well, it represents a massive potential paradigm shift in terms of safety. Currently, there are approved CRISPR therapies for sickle cell disease, but those therapies rely on the active cast 9 enzyme? They literally cut the DNA double strand to destroy the binding sites for these repressors.
13:07And if you're wondering why cutting DNA is a problem and it cures the disease, you have to remember that cutting DNA is a highly violent event for a cell. Standard CRISPR doesn't technically fix the gene.
13:17It breaks the DNA and relies on the cell's natural, highly error prone repair mechanisms to hastily glue the strand back together with missing pieces. Exactly. Anytime you deliberately shatter a DNA double strand, you introduce the risk of off target mutations, unwanted large scale genomic deletions, or even chromosomal rearrangements.
13:38Which is terrifying. Yeah, I mean, the current therapies are miraculously effective for patients who need them, but they do carry inherent long-term genetic risks. But because the decast 9 tool used in this study has broken chemical scissors.
13:51Ebigenome editing leaves the actual genetic code pristine. You alter the expression of the gene purely by erasing chemical tags without ever breaking the structural integrity of the patient's DNA. It's safe, reversible lock picking, instead of blowing the door off its hinges with explosives.
14:07That preservation of the genome is the ultimate advantage. However, the researchers are careful to highlight a significant unknown, which is the long term durability of the treatment. Right. In their Huda piece cell cultures, the fetal hemoglobin remain highly active for over 3 months after just a single transient exposure to the Tet TV 4 editor.
14:27The cells didn't seem to natively erm anothylate the promoter as they divided in the dish. Which is great, but 3 months of stable activation in a Petri dish is just a start. A patient with sickle cell disease needs this activation to last for decades.
14:40Precisely. The open question is how primary hematopoetic stem cells in a living human bone marrow will behave over a lifetime. Stem cells have robust maintenance mechanisms. So they might try to undo the fix.
14:53Exactly. As they divide over years, they might slowly attempt to fix the missing tags and remethylate that promoter, silencing the backup generator once again. If that natural fading occurs, the therapy might require periodic targeted delivery of the epigenomic editor to keep the fetal genes active.
15:10It's a clinical hurdle that will require years of follow-up to understand. But even with that hurdle, the foundation this team is built is extraordinary. So to distill all of this complex biology down for you.
15:21This paper proves that localized CPG methylation at the fetal hemoglobin promoters is the direct causal mechanism of gene silencing. It's not a byproduct, it is the lock itself. Yes, and by deploying precision epigenome editors to selectively erase those molecular tags, scientists can dismantle the nerd repressor complex and safely turn healthy fetal genes back on, all without ever cutting a single strand of DNA.
15:46By identifying the exact combination lock of upstream CPG sites and mapping how MBD 2 translates those sites into physical silencing, we now have a precise blueprint for waking up healthy genes that biology has prematurely shut down.
16:00It's an incredible leap forward for hematology. But it also leaves you with a much bigger, more provocative question to think about. If we can theoretically cure a severe life-threatening blood disorder, simply by erasing a few chemical tags, leaving the DNA completely intact, how many other perfectly healthy locked genes are hiding dormant in your genome right now, just waiting for the right molecular locksmith.
16:22That is a fascinating thought. 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. If you enjoyed this, follow or subscribe in your podcast app and leave a 5 star rating.
16:38If 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.
16:48Thanks for listening and join us next time as we explore more science base by base.
