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. Glad to be here for another deep dive.
0:10So I want you to imagine something for a second. We usually think of our bodies, especially, you know, our healthy tissues, as these really peaceful, well ordered kingdoms. Right. Like all the cells are just working together in perfect harmony.
0:23Exactly. But right now, inside the seemingly normal tissues of your mouth, there is this invisible microscopic battlefield. Yeah, it's wild to think about. It really is. I mean, 1000000s of mutant clones are constantly fighting for survival, just battling for real estate as you age.
0:42It is definitely not a peaceful kingdom at all. Far from it. It's a vast map of tiny competing territories that are constantly shifting, growing, and colliding. Which brings us to the core questions for today.
0:53What really happens in your cells when you age, drink, or smoke? And how could tracking these invisible battles completely change how we prevent cancer? It completely reframes how we think about biology at a baseline level.
1:06I mean, it really does. Yeah. We tend to walk around assuming that being quote unquote healthy means yourselves are pristine, but beneath the surface. Your tissues are actually these dynamic mosaics, deeply shaped by lifelong wars between different genetic factions.
1:22Today we celebrate the work of the cancer aging and somatic mutation program at the Welcome Sanger Institute, the Department of Twin Research and Genetic Epidemiology at King's College London, and quotient therapeutics, who have advanced our understanding of somatic mutation and selection at a population scale.
1:38And, um, what makes this collaborative effort so special, what really stands out is the sheer ambition of the methodology. Right, they aren't just looking at a small handful of isolated tissue samples here.
1:49Exactly. They are blending massive twin studies, population scale metadata, and just a revolutionary leap in sequencing technology into one massive deep dive. It allows us to observe this hidden cellular world with a clarity that was honestly genuinely impossible just a few years ago.
2:08So to really grasp with the magnitude of that leap. We have to look at the scientific roadblock that has been holding genetics back. Yeah, the technical bottleneck. Right. We've known for a while that as we get older.
2:18Our tissues accumulate somatic mutations, like spontaneous changes in our DNA. And some of these ultimately lead to cancer while others just contribute to general aging. But standard DNA sequencing basically misses them entirely.
2:33And the reason for that is that these mutant clones are incredibly small. They usually make up like less than one% of a given tissue sample. less than one percent. Yeah. That is the core technical challenge.
2:46Traditional sequencing just struggles to confidently identify mutations at those incredibly low frequencies. So scientists develop something called duplex sequencing, right? Right. Which was a huge step forward at the time.
2:57It was. It reads both strands of the DNA double helix to check for errors, cross-referencing one against the other, but even standard duplex sequencing has a stubborn error rate of about one in 10 million.
3:08Which, I mean, that sounds incredibly precise to a layperson. One in 10000000 feels like a rounding error. What's fascinating here is that it does sound precise until you realize you are looking for a needle in a haystack of 1000000000s of base pairs.
3:21Ah, right, because the genome is so massive. Exactly. Standard sequencing is like trying to spot a specific grain of sand on a beach using binoculars. We need a high powered microscope. That one in 10000000 error rate happens because of the chemical preparation of the library.
3:36Wait the library? Yeah, sorry. When you're getting the DNA ready to be read by the machine. During that process, damaged sites on one DNA strand can actually be copied over as errors to the other strand.
3:48So the test itself creates a typo that looks identical to a real mutation. Let me make sure I'm visualizing this correctly. Standard sequencing is essentially like trying to spot a specific typo in a newspaper that has been photocopied a 100 times.
4:02Yes. And the photocopying process itself introduces so many smudges and imperfections that you can't tell what's a real typo from the original author and what's just printer ink from the machine. That is the perfect way to look at it.
4:15The smudges are the chemical damage from the preparation process. Got it It's just not a powerful enough method to filter out the noise. If we want to map the exact genetics of these invisible microscopic territories in our healthy tissues.
4:29We needed a technology that could read the DNA flawlessly. Without introducing its own smudges. But wait, if our healthy tissues are completely full of cancer associated mutations, why aren't we all getting cancer constantly?
4:42Well, to answer that, we really had to get that flawless reading technology first, we have to see the whole board. Enter nanosec or nan sequencing. This is the breakthrough technology they used. And they applied it to 1042 self-collected buckle swabs.
4:58So people simply swabbing the inside of their cheeks to collect oral epithelial cells. Which is a brilliant, non-invasive way to get a massive data set. Yeah, and they also sequence 371 blood samples from the Twins UK registry.
5:11So they have this massive, incredibly clean data set. But how exactly does nanosec bypass that photocopier smudge problem? The data mentions sonication, X and nucleus lunting, and enzymatic fragmentation.
5:24That's a lot of dense terminology. How does this actually drop the error rate? Let's break down the mechanics because the ingenuity is in the chemistry here. When you extract DNA, you have to chop it up into smaller pieces to sequence it, that's the sonication part, using sound waves to break the strands, but breaking DNA normally leaves these frayed damaged edges.
5:45If those edges are damaged, the sequencer misreads them as mutations later on. So the sonication creates the frayed edges, which become the smudges on our photocopy. Precisely. So to fix this, they use a process called exonucleus blunting, you can think of exonucleus as highly precise molecular scissors.
6:04Oh, okay. They systematically snip off those frayed, damaged ends to leave a perfectly clean, blunt cut, or alternatively, they use a specialized enzymatic fragmentation that just cuts it cleanly in the 1st place without transferring errors.
6:18Okay, let's unpack this. We've cleaned up the edges using these molecular scissors. What about the rest of the DNA strand? That is where things get really clever? Even with clean edges, a DNA strand might have tiny, single stranded nicks, like microscopic cuts in the middle of it.
6:33Like a scratch on a record. Yeah, exactly. If left alone, the chemicals used later in sequencing will accidentally feel those nicks in with the wrong genetic letters. So they use specific chemical caps to physically plug up any remaining mix.
6:47So the sequencing chemicals can't accidentally write over them. Right. Right. By combining those clean cuts with those protective caps. They drive the error rate down to fewer than 5 errors per 1000000000 base pairs.
6:58Fewer than 5 errors per billion. Okay, returning to the newspaper analogy. That is like proofreading a 1000000000 word encyclopedia and finding only 5 actual typos. It is an insane level of accuracy. That is mind blowing.
7:13So with this flawless tool, what secrets did they actually uncover in the mouths of over a 1000 twins? When you turn this lens on the human mouth, what does the baseline actually look like? Well, the 1st thing that jumps out of the data is the relentless ticking of the biological clock.
7:28The aging process. Yes. In the epithelial cells of your mouth, mutations accumulate in a remarkably linear, predictable way. It's about 18 single nucleotide variants, or S&Vs, per cell per year. So every year, you're alive.
7:41Every single cell in your cheek adds about 18 new permanent mutations to its genetic code. Exactly. It is an incredibly precise biological metronome. But we know from the analysis that these aren't just random typos accumulating harmlessly, right?
7:54The researchers found a massive landscape of positive selection. They identified 46 different genes that are actively under positive selection in the mouth. And that means when one of those 18 annual typos happens to hit one of these specific genes, it gives that cell a survival advantage.
8:12Just starts dividing faster. Yeah, it starts winning the microscopic turf war. And the scale of this selection is just, well, it redefines our understanding of normal tissue. They cataloged over 62,000 driver mutations across these participants.
8:24Wow, over 62,000. Yeah. The genes that showed the strongest positive selection were NOTCH1, TP 53 and FAT1. Those are heavy hitters. Very much so. In fact, the data reveals that in older individuals, between 10 to 20% of their normal, seemingly healthy cheek cells are carrying a known cancer associated driver mutation.
8:45I have to pause on that because it feels like a paradox. It does. If 20% of my healthy cheek cells carry mutations in heavy hitting tumor suppressor genes, like TP 53, why is it my mouth just completely overrun with tumors?
8:59If these are driver mutations, why aren't they driving? That is the pivotal question of the entire analysis, and it brings us to a concept called in vivo saturation mutagenesis. Okay, what does that mean?
9:11Because they found so many mutations using this flawless nanosect technology, they were able to map exactly where these mutations land on the proteins themselves. Right, because normally to figure out which part of a gene causes disease, Scientists have to manually mutate it in a lab or look at advanced cancer genomes.
9:29But here, the natural aging process of a 1000 people has just done the experiment for us. That's it, exactly. Take TP 53, which is arguably the most famous tumor suppressor gene in human biology. Its whole job is to bind to DNA and stop damaged cells from dividing.
9:44The brakes of the cell. Right. The data shows that the mutations in these healthy mouths don't just land randomly across the TP 53 gene. They heavily cluster right in the DNA binding domain of the resulting protein.
9:59Oh I see. If a mutation lands there, the protein can't grip the DNA, and the cell loses its brakes. And they saw several things with other genes too, right? Yeah, look at another gene they highlighted, RAC1.
10:10The mutations strongly cluster right around its GTP binding pocket, which is essentially the engine switch for cell movement and growth. Here's where it gets really interesting. We are mapping the exact structural weak points of human biology just by swabbing healthy cheeks.
10:27It's incredible, but we still haven't answered why these cells don't become tumors. We have the clock ticking. We have the weak points identified, and we know these clones are expanding. So what stops them?
10:36To understand that, we have to look at the multi-stage model of carcinogenesis. This is the established biological principle that a single cell has to acquire multiple specific driver mutations in a row to finally break free of all controls and become true cancer.
10:53Right, it's a multi-step process. Exactly. What this nanosect data shows is a fascinating phenomenon. In the mouth, these mutant clones grow to a certain point, but then they plateau. They just hit a wall.
11:05What kind of wall are we talking about an immune system response or something physical? It is largely about spatial and biological constraints. The lining of the mouth is an epithelial tissue. It has physical boundaries.
11:17The basement membrane limits how deep a clone can grow, and lateral growth is constrained by neighboring clones pushing back. So they just run out of room. Yeah, plus the cells at the surface are constantly sloughing off.
11:29So a clone takes over a tiny patch of territory, maybe a few square millimeters, and then it's stuck in a stalemate. It just can't grow exponentially. Ah, so because the clone is trapped in this microscopic space, the total number of cells in that clone remains relatively small.
11:45And for that clone to turn into an actual tumor, one single cell inside that trapped population has to get incredibly unlucky and acquire a 2nd or 3rd specific driver mutation. You've nailed the mechanism.
11:57Because the overall population of that specific clone is physically constrained, the mathematical odds of one of its cells taking the next evolutionary step stay extremely low. That makes total sense. The physical architecture of the tissue itself acts as a massive natural barrier to cancer, even though the tissue is absolutely saturated with 1st step mutations.
12:18So we've established this baseline war zone, but we know the environment shapes the battlefield. Definitely. The deep dive into this data explored how our lifestyle choices, specifically alcohol and tobacco, alter this landscape.
12:31Let's start with alcohol. Okay. The researchers found that drinking directly spikes a mutational pattern called signature B. Right. Signature B is the specific chemical footprint left on the DNA by aldehydes.
12:43Which are the toxic byproducts your body produces when it metabolizes alcohol. Exactly. And this is where the twin aspect of the cohort becomes incredibly valuable. By looking at identical twins, They prove that the baseline mutation rate has a highly heritable component.
12:58So your genetics dictate the rules of engagement. Yes, and they found a brilliant example of this with a specific inherited variant in a gene called ALDH2. For context for the listener, the ALDH2 gene produces the enzyme responsible for clearing those toxic aldehydes out of your system, right?
13:16Right, right. And the researchers found that if you inherit a specific variant of this gene, which is known as RS 476-7364, very catchy. If you have that variant, your body's ability to clear those toxins is severely compromised.
13:31Oh wow. For people with this variant, drinking alcohol leaves the aldehydes lingering in the tissue for much longer. As a result, alcohol is biologically far more mutagenic in their mouth. So the exact same glass of wine causes significantly more DNA damage for them than it does for someone without that genetic variant.
13:49Exactly. It is the ultimate combination of genetics and environment playing out at the molecular. That is fascinating. But the findings around smoking are what I found truly paradigm shifting. We know smoking causes oral cancer.
14:01We know tobacco smoke contains dozens of carcinogens. Yet the classic chemical fingerprint of tobacco smoke, which geneticists call SBS 4, was essentially missing from these normal cheek cells. It is a remarkable finding because it challenges our fundamental assumptions about how carcinogens operate.
14:20Wait, so smoking doesn't leave its own unique fingerprint in the mouth. Not directly, no. In the lungs, smoking directly scars the DNA, leaving that heavy SBS 4 signature everywhere, but in the oral epithelium, the mechanism is entirely different.
14:34So it just hits the fast forward button on normal aging. Yes. Smoking doesn't act as a direct mutagen leaving its own fingerprint here. Instead, it acts as a selectogen or a promoter. A select agent? Okay, how does that function mechanically?
14:46If it isn't mutating the DNA directly, how is it driving cancer risk? It alters the tissue environment. First, like you said, it accelerates the normal clocklike aging mutations. What the researchers call signature A.
14:58It makes that 18 mutation per year clock take much faster. But critically, the toxic environment created by the smoke changes the rules of the turf war. How so? It tips the scales in favor of certain dangerous clones, allowing them to grow larger and faster than they normally could in a non-smoker.
15:16Could borrow our earlier concept. It's like smoking isn't just dropping bombs on the territory. It's actively funding the enemy army. Exactly. It creates an environment where aggressive clones have a selective advantage over normal cells.
15:28That is an excellent way to frame it. It is an ecological shift rather than just purely a genetic one. The smoke makes the tissue a harsher place to live, and only the most mutated, resilient clones can actually thrive in that harshness.
15:42So we have this incredible map. We know the mutation clock ticks linearly. We know 46 genes are driving these clones to take over our tissues, but they hit that spatial plateau. Right. We know alcohol leaves a specific aldehyde footprint dictated by your ALDH2 genetics, and smoking acts as an ecological selectogen that alters the battlefield.
16:01It's a lot to take in. It is. The biology is fascinating, but how does this data actually move out of the laboratory and into clinical practice? Right. I mean, how does it change a listener's life? Well, the most immediate application is in classifying what clinical geneticists call variants of uncertain significance.
16:18The US. Yes, V US. When a patient gets a genetic screening today, the results often come back with mutations where the doctor simply has to say, well, we don't know if this is harmless or if it causes disease.
16:30Because it just hasn't been studied enough in a lab setting to make a definitive call. Exactly. But because this nanosect study mapped out 10s of 1000s of mutations in normal tissues, we can cross-reference those findings with clinical databases like Clinvar.
16:44Oh, that's smart. Right. If we see a specific variant, strongly driving clonal expansion and winning the turf war in a healthy person's mouth, it provides massive real-world biological evidence that the variant is actually pathogenic.
16:59It removes the uncertainty. So we are basically crowdsourcing the answers to genetic mysteries using the healthy cells of a 1000 people. We really are. That is brilliant. But let's look further down the road.
17:11What about actually preventing cancer before it starts? If we connect this to the bigger picture? Think about how we handle cardiology today. Okay. We routinely use lipid panels and cholesterol tests to predict heart disease risk.
17:25If your LDL is high, your doctor doesn't just send you home to wait for a heart attack, right? No, they prescribe a statin to alter your biological environment and lower your risk. Exactly. With tools like targeted nanosec, we are looking at the ultimate surrogate risk marker for cancer.
17:41Wait, so what does this all mean for me? Well, I soon go to the dentist, get a swab, and get a quote unquote cholesterol test for my oral cancer risk? We are heading in that direction, yes. Really? But if 10 to 20% of my cells already have these driver mutations.
17:56How does a doctor know which specific clone is actually going to break through that spatial plateau we talked about? Because the test wouldn't just look for the presence of a mutation. It would measure the dynamics of the clones.
18:09I see. In the future, a clinician could swab your mouth, sequence it with near perfect accuracy, and measured the exact size of your clones, the signatures of your exposures, and precisely calculate your trajectory over time.
18:22Wow. If they see a clone that is overcoming those spatial constraints and growing too fast, they could potentially prescribe molecular prevention. Meaning targeted therapies design specifically to shrink those aggressive clones back down.
18:36Exactly. Restoring the natural plateau before a tumor ever has the chance to form. That is the transformative potential of this technology. We are moving from reactive oncology to proactive molecular management.
18:49Lefk is still all this down into a core takeaway. If you remember one concept from this deep dive, let it be this. Our seemingly normal healthy tissues are actually dynamic mosaics of microscopic mutant clones, deeply shaped by our age, our genetics, and our lifestyle.
19:05By reading these changes with near perfect single molecule accuracy, we are unlocking the ability to predict and ultimately prevent cancer at its earliest possible stages. It is a monumental leap forward.
19:17But I think it also leaves us with a really profound philosophical shift to consider. What does this mean for how we define being healthy? When a perfectly normal, healthy mouth is already teeming with cancer associated mutations?
19:32This 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.
19:46If 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.
19:54Thanks for listening and join us next time as we explore more science. base by base.
