This forum reviews evidence that host genetic variants associated with metabolic disease often overlap with loci that shape gut microbiome composition and function. Examples include LCT/MCM6 linking Bifidobacterium to reduced T2D risk, defensin locus variants affecting DEFA26 and Akkermansia abundance, and rs7133214 associating with HbA1c. The authors outline mechanisms, analytic tools, and experimental strategies to resolve causality and call for centralized microbiome–genetic resources.
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. So, um, I want to start with a concept that we all kind of take for granted.
0:13When you think about your genetic risk for something like obesity or, you know, type 2 diabetes. You probably assume your DNA is directly giving faulty instructions to your body cells. Right? Yeah, like, you have a specific gene.
0:27It makes a defective protein in your liver and bam, disease. Exactly. We are always taught to view it as this like straight line. But what if those genes are actually secret instructions for the trillions of bacteria living in your gut?
0:40I mean, that represents a complete shift in how we understand our own biology. It's huge. really is. Think of your DNA as a landlord, okay? And the gut microbiome as the tenant. Okay, landlord and tenant.
0:51like that. Right. So for decades, we've basically operated on the assumption that metabolic diseases were caused by a structurally broken building. Like faulty plumbing or bad wiring in the host's own human tissues.
1:04Exactly. But what if the building itself is actually totally fine. What if the disease is instead caused by the landlord inadvertently leaving the doors unlocked, you know, encouraging the absolute wrong tenants to move in and completely trash the place?
1:21Oh, wow. that reframes the entire problem? Because if the landlord is picking the tenants, then treating metabolic disease by only looking at human tissue is like, I don't know, trying to fix a broken window without actually kicking out the vandal.
1:33That is a perfect way to look at it. Yeah. It makes you wonder how this could change the way we treat these conditions. What if the actual target shouldn't be the human cells at all, but the specific microbes living inside them?
1:45Which is exactly what we're diving into today. Today we celebrate the work of Rebecca C. Simpson. Harry B. Cutler, David E. James, and Stuart W.C. Masson from the University of Sydney, and the Charles Perkins Center, who have advanced our understanding of the overlap between host genetics, the gut microbiome, and metabolic disease.
2:03It is such a remarkable piece of synthesis. They pull together a lot of disparate threads here. They really did. So take us back a bit. Why do researchers even suspect the landlord is manipulating the tenants in the 1st place?
2:15Well, to really appreciate that, we have to look at the historical evidence. For years, genome wide association studies... Or G ways, right? Where they look across the DNA of massive population. Exactly.
2:28G- ways. So G ways and massive twin studies have been pointing to this invisible connection for a while now. We've seen that both the gut microbiomes composition and a person's metabolic health are like inextricably linked to the host's genomic loci.
2:43Wait, so specific regions on our human DNA are somehow pulling the strings of single celled organisms down in our intestines? Precisely. We see very clear evidence of our genetics regulating the gut when we look at the genes governing intestinal barrier integrity.
2:59Like the physical lining of the gut. Yeah, think about them as the border guards of your gut. You have panet cells that secrete antimicrobial peptides, you've got goblet cells producing protective mucus plus your whole systemic immune response.
3:11And these processes fundamentally underpin our immune homeostasis. They control systemic inflammation. And crucially, those host controlled factors dictate not just who gets to live in the gut, but how those microbes eventually impact your overall metabolic health.
3:26Okay, let's unpack this because it's one thing to say that our immune system acts like a bouncer at the club door, you know, checking IDs and deciding which microbes get in. But how does that chain reaction actually lead to something as complex as metabolic disease?
3:40I feel like I need to see the dominoes fall here? Sure, let's look at a 2016 twin study focusing on obesity. to trace those exact dominoes. So researchers identified variants in a gene called TDRG1. TDRG one okay.
3:53Now, this gene is expressed in the mucosal lining of the esophagus. So fairly high up in the digestive tract. But those specific genetic variants were tightly associated with the abundance of gut microbes much lower down in the digestive system.
4:06Specifically, microbes that are linked to leanness. Oh, wow. So a human gene expressing in our own upper tissue is dictating the survival of lean microbes lower down. Exactly. Another really illuminating example is lactase persistence.
4:20You know, the genetic mutation that lets humans digest milk as adults. Oh yeah, that involved the LCT and MCM 6 genetic regions, if I remember correctly. You've hit the nail on the head. So it turns out those genetic variants don't just let you enjoy dairy without a stomach ache.
4:35They actually reduce your overall risk for type 2 diabetes. Wait, how does drinking milk prevent diabetes? That sounds completely counterintuitive? I know, but the mechanism is what is so fascinating. That increased lactose metabolism in your gut specifically provides a food source for beneficial bifidobacterium species.
4:54Oh, so your human genetics provide a very specific substrate like, the lactose byproduct that allows the good tenants to thrive. Yes. And those good tenants then produce metabolites that protect your systemic metabolism.
5:07So the host is basically laying out a highly specific buffet. Only the good bacteria get to eat, so they multiply and pay their rent by keeping us healthy. Exactly. It's a symbiotic loop. That makes total sense.
5:18But, um, the sources also bring up a very famous diabetes gene to illustrate the dark side of this, right? TCF 7L2, I think. Yes, TCF 7L2 is the perfect case study for this whole paradigm shift. Single nucleotide polymorphisms or SMPs.
5:33Which are just the tiny single letter variations in our DNA code. Right. So SMPs in this particular gene are the single strongest genetic links to type 2 diabetes that we know of. Like they've been robustly identified across tons of diverse human populations.
5:47Wow, okay. And for a long time, the clinical dogma was that these genetic variants directly disrupted compensatory beta cell growth in the pancreas. Meaning like when someone starts getting insulin resistant, a healthy pancreas will try to grow more beta cells to pump out extra insulin and compensate, right?
6:04And the assumption was that this gene broke that factory. Exactly. That was the accepted straight line assumption. The gene broke the pancreas. But, and here's the twist. TCF 7L2 also plays a critical, entirely separate role in the development of those paneth cells in the gut we talked about.
6:20Really? Yeah. In fact, it has been genetically linked to Crohn's disease, which is a severe inflammatory bowel condition. Oh, I see where this is going. So if we synthesize these 2 pieces of information.
6:31The revised explanation is that these genetic variants might not be breaking the pancreas first. Exactly. If it alters the panith cells. It's messing with the bouncers at the door. That causes dysbiosis, the wrong microbes move in, and that imbalance triggers chronic systemic inflammation in the gut, which then travels to the bloodstream, and drives insulin resistance elsewhere in the body.
6:53You got it. The pancreas failing is just the end of the domino rally, not the beginning. That maps perfectly onto the new understanding. The host gene causes the dysbiosis and the dysbiosis causes the metabolic dysfunction.
7:05Yeah, but as you can imagine, proving that on a massive scale is incredibly difficult. I mean, it seems like an absolute nightmare. Finding these connections, one gene at a time or one microbe at a time is way too slow.
7:16How do scientists look at 1000000000s of genetic data points across entirely different species like human and microbial without just getting completely buried in statistical noise? It requires some very clever, computational, heavy lifting.
7:28That's for sure. To tackle this, the researchers used a newly developed, high throughput web tool called synony. Syntony. Okay, how does that work? What Sydney does is scrape human genetics summary statistics from a massive repository known as the association to function or A2F portal.
7:46So they essentially build a digital bridge between 2 colossal islands of biological data. But um, how do you actually filter that? Because it feels like trying to find a needle in a haystack by just staring really hard at the hay.
7:59Yeah, they basically needed a very specific magnet. So they took genome wide significant S&Ps that were already known to be associated with microbial abundance. They pulled those from 7 recent microbial GY's papers.
8:12So they started with the known microbial needles. right? Then, they took those specific SMPs and ran a phenomenon Y association study of FEWAS using that syntony tool. Just to clarify the terms for the listener, a regular GS asks, like, what genetic mutations cause this specific disease?
8:31But if fee walls flips it backward, right? It asks, what diseases are caused by this specific genetic mutation? That is the exact distinction. Yes, they were actively hunting for overlaps. They wanted to see if the genetics controlling the microbes were also showing up as the genetics controlling human metabolic traits.
8:48But wait, how do you account for the environment in all this? Because if there's one thing I know about the microbiome? It's that what you eat for dinner on Tuesday completely changes who is living in your gut by Wednesday morning.
8:58Oh absolutely. How do you control for diet in massive human genetic studies? The short answer is you can't. Not perfectly anyway. Diet is the absolute main driver of gut microbiome variation, and humans are notoriously terrible at recording what they eat, let alone strictly adhering to a standardized diet for years.
9:18Yeah, definitely don't remember what I had for breakfast 3 days ago. Exactly. So to find a true underlying genetic signal amidst all that daily dietary noise in humans requires staggeringly large sample sizes, and that is prohibitively expensive.
9:33So what's the workaround? This is why the researchers had to rely on genetically diverse mouse populations, specifically the diversity outbred population in Australia. Oh, because in a lab, the researchers are the chefs, you control the menu.
9:46Exactly the point. The environmental inputs, the food, the temperature, the lighting, the water, all of it can be strictly controlled, that drastically reduces the environmental noise, giving them the statistical power to find the true biological signals.
9:59And then they translate that to humans. Right. Once they map those host microgenetic interactions in mice, they can specifically investigate the equivalent genetic regions in humans. It narrows the search area from the entire genome down to highly probable targets.
10:14Okay, so with this powerful Sydney tool deployed. And the mouse data guiding the way. What did the human data actually reveal? Here's where it gets really interesting. I want to know the scale of what they uncovered.
10:27The scale is really what makes this paper so important. They identify 28 distinct human phenotypic groups that had genome wide, significant associations to these microbe associated SMPs. 28 different categories of human traits linked directly to the microbiome's genetics.
10:44Yes. And intriguingly, about 16% of those significant associations were hematological meaning traits related to the blood, specifically monocite and lymphocyte counts. Which are white blood cells. That perfectly reinforces that connection between the microbiome, the integrity of the gut barrier, and the host immune system that we were talking about earlier.
11:02The bouncers at the door are heavily represented in the genetic data. They absolutely are. But the standout finding, like the most striking overlap they found, was related to blood sugar. The strongest association they discovered was between HBA1C.
11:16Which is the standard clinical marker for long-term blood glucose levels and type 2 diabetes, right? Yes. Between HBA1c and a specific SNP known as RS 7133214. Okay, I'm assuming that random string of letters and numbers means something biologically profound.
11:32It does, I promise. This S&P was originally associated with the microbial mythinine salvage pathway. Okay, let's translate that. Mefining is an amino acid like, a building block of protein. So what are the microbes salvaging it for?
11:45Well, microbes act like tiny chemical factories, right? They primarily use the salvage pathway to take protein scraps, like methianine, and upcycle them to synthesize polymines, such as spermine and spermidine.
11:56Okay. Now, what's fascinating is that the nearest human gene to this genetic variant is called KLHL 42. That gene is part of a protein degradation complex in humans that is linked to systemic fibrosis, inflammation and metabolic dysfunction.
12:10Hold on, you're telling me a human gene, linked to tissue scarring and inflammation, is simultaneously controlling how microbial factories upcycle amino acids. How are those 2 things even talking to each other?
12:23That's the $1000000 question. While the exact mechanistic link isn't fully mapped out just yet. We know that polyamines play central roles in modulating the host's own cellular processes. Like what kind of processes?
12:36They influence autophagy, which is how cells clean out debris and adapogenesis, which is the creation of new fat cells. Wow. Yeah, it strongly suggests the host genetic variant is turning a microbial factory up or down.
12:49And then that factory floods the human host with metabolites that drive diabetic pathology. So the landlord is tweaking the thermostat, which changes with the tenant manufacturers, which eventually burns the building down.
13:00That is incredible. It really is. What about other metabolic markers? I mean, if these microbes are turning out polyamies that affect our blood sugar, What do they do into the fats we eat? Does this dynamic apply to cholesterol, too?
13:11It absolutely applies to cholesterol. They found very strong overlaps between microbial S&Ps and host cholesterol metabolism, specifically high density lipoprotein, or HDL. The biology here really comes down to bile acids.
13:26The digestive juices our liver makes to break down greasy food. Right. So your liver synthesizes primary bile acids from cholesterol and secretes them into the small intestine to help you absorb dietary fats.
13:38But once those primary bile acids hit the gut, specific microbes get to work on them. They chemically alter them into secondary bioacids. The microbes do this partly as a detoxification mechanism for themselves, but also as a source of nutrients.
13:54And then what happens? Do those secondary bylasses just, you know, pass through our system? No, and this is the brilliant part of the biological loop. Both primary and secondary bioacids are recycled. They are absorbed back through the intestinal wall into the bloodstream, and return directly to the liver.
14:09Decobiological recycling program. A highly efficient one. Once those secondary bioacids get back to the liver, they bind to a specific receptor called the Pharnasoid X receptor, or FXR. Okay, FXR. That receptor acts like a sensor, and it directly influences the host's own cholesterol metabolism.
14:29So an overlapping genetic association here suggests a chain reaction. Wait, let me guess. A host gene alters the microbiome. Then the altered microbiome changes how bile acids are chewed up and recycled, and those recycled bile acids return to the liver and alter its cholesterol synthesis.
14:46You absolutely nailed it. It's an entire outsourced metabolic organ. The liver is literally taking orders from the gut bacteria. It really is. And actually, going back to the mouse data for a second, because I know they found something really cool there regarding how the host picks its tenants.
15:00They looked at the defense in Locust, correct? They did. So defenses are those small anti-bacterial peptides we mentioned earlier, secreted by PanF cells to keep the piece. The researchers found that specific genetic variants in the Defense and Locust increased the expression of a protein called DF 26.
15:17Let me guess, this protein acts like a selective filter. Highly selective. It actually boosted the abundance of a very specific metabolically beneficial microbe called achromanzia meesinifila. Okay, acromancia.
15:30Yeah, this microde happens to be naturally resistant to that specific defense in protein. And when the researcher supplemented mice with synthetic EFA 26, it completely recapitulated the effect. So what happened to the mice?
15:42The acromancia level shot up, and the mice showed significantly improved metabolic health. Even when they were fed a terrible, high fat, high sugar diet. That is wild, threading our analogy all the way through.
15:55The host gene is the landlord installing a custom security door that only lets the good tenants, the achromancia inside. And those good tenants pay rent by protecting the host from the consequences of a bad diet.
16:07That perfectly encapsulates what the data is showing us, yes. But looking at all this data, the huge overlaps, the bioacid recycling loops, the polyamine factories. I had to play devil's advocate for a second.
16:18Go for it. In science, correlation is not causation. We see these huge overlaps, but how do we know which direction the river is actually flowing. Like, does the host gene change the microbe, or does the gene change the human tissue first, which then just incidentally changes the environment the microbe lives in?
16:35This raises an important question. And honestly, it is the central challenge facing the entire field right now. Untangling bidirectionality. Because it's a chicken and egg problem. Exactly. It's incredibly difficult to prove causality.
16:50To solve this, researchers need to lean heavily on techniques like mendelian randomization and causal multivariable modeling. Mendelian randomization. That's where you basically use genetics as a natural randomized clinical trial, right?
17:05Yes. Because your genes are randomly assigned at conception, and they don't change based on your diet or your lifestyle. You can use them as a fixed anchor to determine cause and effect. That is the exact logic.
17:15A Mendelian randomization analysis could look at those specific S&Ps associated with both microbes and cholesterol, and mathematically pinpoint the directionality. Ah, I see. It could tell us definitively.
17:27Do these genetic variants primarily affect the microbiome, which then alters the cholesterol, or does the host's cholesterol machinery change 1st, which then alters the microbiome? So what's stopping researchers from running those causal models right now on every single gene we have?
17:42Data silos. The information is just too scattered. There is a desperate need for a massive centralized database that contains both microbe related and metabolism related genetic data sets in one place.
17:54Like the A2F portal they use but bigger. Right. We need something like the A2F portal, but purpose built for the microbiome to run these comprehensive causal models. Furthermore, researchers have to stop just looking at taxonomic abundance, which is just tallying up who is present in the gut.
18:12Oh, right, because just knowing they are, there isn't enough. Exactly. They must combine that data with metabolomics, what those microbes are actually producing. We need to measure the industrial output, not just the head count.
18:24Precisely. Knowing a specific microbe is present isn't enough. We need to know what chemical signals, it is actively sending to the human host. That makes total sense. So bringing this all home, why does this matter for the listener?
18:35How does this jump from a bioinformatic web tool scraping data into a doctor's office actually helping patients? It represents the absolute frontier of precision medicine. Right now, if you go to a clinic for metabolic issues, medicine targets the microbiome, maybe through a diet plan or probiotics, or it targets your genetics and metabolic markers via pharmaceuticals.
18:58But it does so entirely in isolation. Exactly. If they are structurally intertwined? An integrative approach changes the whole game. Meaning we get stopped using the trial and error approach with treatments.
19:10Yes. We could accurately predict how a patient will respond to a specific dietary intervention or a specific drug based on their combined genetic and microbiome profile. We could know ahead of time that a certain medication won't work for you, not because the drug is bad, but because your genetic profile doesn't support the specific microbial tenants needed to process it.
19:30That is incredible. But are there any caveats we should be aware of? You mentioned the mouse models being powerful for controlling the dietary environment, but they aren't perfect proxies for humans, are they?
19:40They certainly have limitations that the field has to navigate. One major biological issue is that mice are coprafagic. They eat their own feces. Oh, well, there goes my appetite. Right. But I assume they do that for biological reason, right?
19:55Like extracting leftover nutrients. Yeah, it's a natural behavior for nutrient absorption in mice, but in a research setting, this leads to a significant statistical problem. Because laboratory mice share cages, this behavior homogenizes the microbiomes of all the mice within that cage.
20:11Oh, I see. It creates a massive cohousing effect. Exactly. Those shared environmental effects can interact with the underlying genetic differences researchers are trying to study, making it very complicated to accurately interpret genotype phenotype relationships.
20:27Right. You can't accurately map a unique host microbiome interaction if all the hosts are, you know, sharing their microbiomes over dinner. The data just blurs together. Exactly. A real hurdle. Well, to distill this entire deep dive down.
20:39The central insight here is profound. The gut microbiome is not just a passive passenger taking a ride in our bodies. It is a highly active, symbiotic, metabolic organ operating under the exact same host genetic control as our liver or our pancreas.
20:54It completely reframes how we define the boundaries of human biology. It really does. The human genes we always thought were breaking ourselves. Might actually just be cultivating the wrong microbial ecosystem.
21:06Which leaves you with its tamol over. What does this mean for the future of genetic testing? Will a routine cheek swab one day tell you not just what metabolic diseases you're physically at risk for, but exactly which specific microbial tenants you need to actively cultivate to rewrite your own metabolic destiny?
21:25It's an exciting future. 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 enjoy this, follow or subscribe in your podcast app, and leave a five-star rating.
21:40If 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.
21:50Thanks for listening and join us next time as we explore more science, base by base.
