Using phased long-read nanopore and short-read sequencing across eight trios, the study maps allele-specific DNA methylation and transcription in female human placentas, identifies hundreds of DMRs and novel imprinted genes, and reports somatic placental variants
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. Okay, let's unpack this. Imagine an organ that um, it operates on a deadline.
0:13It's a temporary but absolutely indispensable life support system for 9 months. That's the placenta. Exactly. But it's not some passive filter. It's actually the epicenter of a really fierce molecular tug of war between mother and fetus.
0:28It is. a molecular battlefield. And I think the fundamental question in reproductive biology is how this temporary organ manages such a profound delicate balance. Especially when it comes to deciding which parents' genes get turned on and which ones stay silent.
0:43Right. So here's the surprising fact. The placenta has a molecular blueprint, a unique epigenetic fingerprint that is just completely different from every single other tissue in the human body. It's a genuine evolutionary frontier.
0:56It's packed with genomic mysteries that we are, you know only just beginning to solve. And that's really our mission today. We are diving into a study that used a, well, stunning a technological leap nanopore sequencing to create the most detailed, high definition map yet of this blueprint.
1:14And in doing so, they finally revealed the exact genes that are involved in that molecular tug of war. Yeah, and we're not just talking about finding a few new data points. We are talking about finding the actual battle lines of human development.
1:27This is essential stuff for anyone interested in pregnancy outcomes. So we're launching into a deep dive based on a phenomenal study, an allel resolved nanopore guided tour of the human placental methylome, from a team including Michaela Kenlova, Hannah Byrne, and Adam D. Ewing.
1:43And today, we really want to celebrate the work of the Meter Research Institute at the University of Queensland. They have just dramatically advanced our understanding of this unique epigenetic and transcriptional landscape of the human placenta.
1:55They really have. So let's start with the basics. Why has the placenta been so hard to map? And why does it need its own specific map separate from, say, the liver or the heart? It really comes down to 2 things?
2:06It's structure and it's, um, epigenetics. Structurally, the placenta is this complex mix of specialized cell types, mainly the truffle blasts, which are essential for fetal development. Okay, and epigenetically.
2:21Epigenetically, it's an anomaly. An anomaly in what sense? Well, most of your somatic tissues, your heart, your lungs, they maintain really high levels of something called DNA methylation to keep gene expression tightly regulated.
2:34The placenta, on the other hand. is the opposite. The complete opposite. It has unusually low levels of this chemical modification. We call this widespread hypomethylation. So it's like the rule book for gene expression is just, I don't know, less strict in the placenta than in the rest of the body.
2:49That's a great way to put it. Yeah. And even within that generally relaxed rulebook, the hypomethylation is often clustered into these vast stretches called partially methylated domains or PMDs. And these PMDs give the placenta its unique signature.
3:02They do. They cover huge portions of the genome, giving it a distinctive sort of open access molecular signature. And this unique signature is where that central regulatory conflict, the imprinting all happens.
3:14Precisely. The placenta is the natural epicenter for what we call imprinted genes. These are genes that, you know, they disregard the usual rule of expressing both copies from each parent. Right. Instead, they express predominantly from either the maternal allele or the paternal allele.
3:31This is where that genetic conflict hypothesis comes in, which you mentioned a minute ago. It's a core concept here, isn't it? It's the driving evolutionary force. The maternal genome is, let's say, interested in resource management, optimizing distribution for all potential future offspring.
3:49Okay. But the paternal genome, it's evolutionarily incentivized to push for maximum growth and consumption in the current offspring. So you've got the mother applying the regulatory brakes and the father stepping on the developmental gas pedal.
4:02That's the negotiation. And it happens right in the placental genome. But there was a massive technical barrier, which this paper overcomes. Yes. And until now, separating the maternal copy from the paternal copy, especially for methylation, was.
4:16It was like trying to distinguish two identical puzzle boxes that were already half mixed. You might see a difference in total methylation, but you couldn't say, that part's for mom and that part's from dad.
4:26Not on a genome wide scale. Not at all. That technical challenge is why the methodology here is so important. And this is where it gets really interesting because they used a major leap in sequencing technology.
4:37They did. They employed Oxford nanopore technologies or ONT devices. And this is the key, because ONT offers what we call arbitrarily long reads. So instead of tiny DNA fragments, you get these huge continuous stretches.
4:52Huge. Think of traditional sequencing generating fragments of maybe a few hundred bases. Long reads can span 10s, even 100s of 1000s of bases. Why do those extra long reads matter so much for telling the parents apart?
5:03Because they directly encode 2 critical pieces of information at the same time, the standard DNA bases, A, C, G, T, and the chemical modifications on top of them, specifically 5 methyl cytosine, or 5 mill a C.
5:19That's the methylation mark. So it reads the sequence and its epigenetic state in one go. In one continuous stream. Yes. If I'm following, traditional methods might tell me a reason is methylated, but not which parent's DNA it's on.
5:32Exactly. But the long read, it can capture 100s of known genetic variations we call them SMPs along its length, and these SMPs act like inherited color coding tags. Ah, okay. So if Longread has a set of SNPs that you know only came from the mother, and it also shows a methylation mark.
5:48You can definitively tag that methylation as maternal. So they use these long reads and combine them with a really crucial experimental design. The trio. That's the real innovation. They sequenced 8 female placental trios, so the mother, the father, and the fetus from the Queensland family cohort.
6:05By comparing all three, they could figure out which genetic markers came from which parent. And that let them phase the placental genome. So they didn't just map the placenta's Epigino. They basically created 2 parallel maps.
6:15The maternal map and the paternal map. And then they overlaid them to spot the differences. That level of fidelity is what allows for this comprehensive genome wide investigation. It's a huge step forward.
6:27And with that kind of new technology, you need to be sure it's accurate. Absolutely. They did stringent quality control. They use traditional illuminous sequencing for comparison and something called enzymatic methyl sequencing or EMSEC on 2 samples.
6:41They found a very high concordance about 94%. Which gives us a lot of confidence in the map they built. Immense confidence, yeah. So let's get to the results. What did this high resolution map reveal about the placental methylum starting at the global level?
6:56Okay, so 1st they confirmed the big Fisher characteristic. The placenta is indeed substantially demethylated compared to other tissues like the heart, liver, or hippocampus. It's a truly unique organ. But the real action is where the maternal and paternal maps diverge.
7:10What did they see there? On the big autosomal chromosomes, so the non-sex chromosomes. The methylation profiles were generally highly similar between the 2 alleles, which is what you'd expect in most regions.
7:21But the X chromosome was the big exception. In the female placentas, yes. It was highly variable. They found this variable of mevelation skewing. So some placentas had globally higher methylation on the maternal X, and others had higher methylation on the paternal X.
7:36Why would it be so variable? Isn't X inactivation supposed to be a stable process? Well, it's a process called clonal X inactivation. Early in development, one of the 2 x chromosomes is randomly shut down, transcriptionally repressed.
7:49And because they were analyzing a chunk of tissue, they were basically just getting a snapshot of that randomness. Exactly. The bulk tissue is a mosaic. In some samples, the cell lineages that happen to dominate that particular piece of tissue had silenced the maternal X.
8:04In others, they'd silenced the paternal X. It's like biological Russian roulette captured stochastically. It is, and it's an important reminder that placental function can differ wildly, even between what look like normal pregnancies.
8:17Okay, now let's zoom in from the macro map to the micro discoveries. The differentially methylated regions or DMRs. This is where that conflict hypothesis really plays out. How many did they find? In total, the study identified 723 DMRs.
8:33These are the hotspots where the maternal and paternal copies differed significantly in their methylation. And this is where the power of the new sequencing really shines. How many of them were new? 184 of those DMRs were specific to this study, completely unreported in any prior literature.
8:49Wow. So we're finding new levers that control pregnancy health that we never even knew existed. That's it. But the most striking global characteristic of these DMRs was their bias. The vast majority showed a strong bias toward paternal demethylation relative to the maternal allele.
9:05So where there's a difference, The paternal copy is usually the one with fewer regulatory locks on it, which fits perfectly with the idea of the paternal genome pushing for more expression, more activity, more growth.
9:16It absolutely validates that evolutionary context. And they show that this paternal demethylation bias is a uniquely placental thing. You don't see it in other cell lines. They also gave some fascinating visual examples.
9:27I saw they mapped the known C-19 MC Locus on Chromosome 19. They did. That locust encodes about 50 micro RNAs, which we know are paternally expressed, and they're essential for trophil blast differentiation.
9:41Invasive cells. But even more strikingly, they found an entirely new DMR rich region at the SST1 macro satellite repeats, also on Chromosome 19. This region show this curious periodic pattern of methylation and demethylation, but only on the paternal allele.
9:59And they connected that back to evolutionary innovation, finding that transcription was initiated at HERVH retrotransposen. Yes, it supports this really cool hypothesis that human endogenous retroviruses.
10:10These ancient remnants of viral infections in our geno are acting as transcriptional innovation engines. So evolution is literally using ancient viral code to regulate modern reproduction. Specifically within the human placenta.
10:22It's remarkable. Before we get to the new imprinted genes. We have to touch on the variation within the placental tissue itself. The data confirm the high mutational burden of the placenta. It did. They found an average of 186 de Novo Point mutations per sample.
10:37This is consistent with what we know, that the placentages have a massive number of somatic mutations. Mutations that happen after conception, only in that tissue. And they even caught a concrete example of how one of those random mutations can drastically affect the placental transcripto.
10:53Yeah, that was amazing. They found a somatic 44.7 kilobase pair duplication in one specific sample. This duplication acted like a molecular switch. It fused most of the CNDP1 gene with the very active promoter of another gene, ZNF 407.
11:09Wait, so a random mutation inside that one placenta created a brand new hyperactive gene fusion. In real time. And it drove the expression of CNDP one only in that one sample. It's conclusive proof that even a small fraction of these Somatic mutations can profoundly alter the function of the placenta.
11:26Which could drive variability in pregnancy outcomes. Potentially yes. Okay, this brings us to the biggest payoff of the whole study. All this high-res mapping, all this allegal phasing. It let them intersect their new DMRs with a leal specific gene expression data.
11:41And that's how they found them. Previously unreported imprinted genes. This is the moment they proved the conflict hypothesis exists at this new specific resolution. What is? They found two major examples that just they align with the model beautifully.
11:56Okay, let's start with the paternal side, the gas pedal. That would be a gene called ILDR2. They found it to be paternally demethylated and expressed. Remember, de methylation means the gene is active.
12:07And what does ILDR2 do? It modulates T cell activity. So it's related to the immune system. Exactly. The suggestion is that the paternally expressed ILDR2 plays a role in suppressing the maternal immune response toward the placenta.
12:21By promoting immune tolerance. It allows for deeper invasion. And therefore promotes maximum growth, a clear paternal interest. A textbook case. And then we have the maternal regulatory break. That would be res A1.
12:33And it was the opposite pattern. Maternally demethylated and expressed. Razero one inhibits a major cellular signaling cascade called the Rasmia K pathway. And what does that pathway do in the placenta?
12:44Studies suggest that inhibiting it specifically. tones down excessive trophoblast proliferation and invasion. So it's the perfect snapshot of the molecular tug of war. It is. The mother's gene, Russia one, inhibits excessive growth.
12:58The father's gene, ILDR2, pushes for growth by dampening the immune attack. It's remarkable how neatly their functions align. Beyond those 2 specific genes, the data also confirmed the inherent variability in some DMRs, something known as MP polymorphism.
13:13Right. And jeans like TSC3 and WNT2. And this just reminds us that there isn't one single universal placenta. There's real regulatory variation between individuals, and that variation likely contributes to differences in how pregnancies progress.
13:26So for the research community, now that they have this comprehensive map, what are the next steps? What are the limitations? Well, the study forms a critical basis for targeted functional studies. We can now go after these specific loci, but the researchers noted 2 major limitations.
13:42The 1st being cohort size and diversity, I assume. Yes. While this is an incredibly deep dive. The number of samples is still small. The cohort size needs to be expanded significantly to have enough power to link specific variations in these DMRs to adverse pregnancy outcomes, like pre-eclampsia or growth restriction.
14:01We need the numbers to connect the molecular dots to clinical risk. Precisely. And the 2nd limitation was the scope of the sample set. All female placentas. The initial cohort included only female placentas.
14:12So a crucial next step is to include male placentas. This might offer insights into why females sometimes appear better able to adapt to environmental stressors in utero than males. Which is a common finding in many developmental studies.
14:24It is. Okay, so what does this all mean for you? I think the key takeaway is that cutting edge nanapore sequencing has successfully given us the 1st comprehensive, allele resolved map of the human placental methylome.
14:37It's confirmed its unique, highly demethylated state, and identified 100s of new regulatory sites. And that depth of analysis led directly to the discovery of previously unknown imprinted genes, like ILDR 2 and RASO 1.
14:51And these genes appear to regulate crucial developmental processes, perfectly validating that evolutionary genetic conflict hypothesis. So this new map suggests that we can now track the origins of health and disease right from the moment of conception.
15:05We can start to identify molecular signals that might predict risk factors. How quickly can we leverage this data to actually build those clinical prediction and intervention tools? That's really something for us all to consider.
15:16This 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.
15:29If 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.
15:39Thanks for listening, and join us next time as we explore more science based by base.
