Untargeted plasma metabolomics (1,930 features) in 1,244 participants of the Swiss HIV Cohort Study were paired with genome-wide genotypes to map genetic influences on metabolite levels, test colocalization with eQTLs, and apply Mendelian randomization to probe causal links with aging-related biomarkers and diseases.
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. It is great to be here So I want you to imagine that you take your car, um, to the mechanic, right?
0:13Because of a catastrophic engine failure. Oh, a nightmare scenario. Total nightmare. But they work an absolute miracle. They replace parts, fine tune the system, and suddenly the engine is purring perfectly.
0:26Okay, sounds good so far. Right. But then you drive off. And within a week, your brake pads dissolve, the transmission slips, and like your tires go completely bald. Oh wow. So the main problem is solved, but the arrest of the car is just falling apart.
0:40Exactly. The wear and tear on the rest of the system is inexplicably in overdrive. And, um, in human terms, that exact scenarios playing out right now for a 1000000s of people. Yeah, it really is. Today, if you're living with HIV and receiving optimal anti-retroviral therapy, the virus can actually be suppressed to totally undetectable levels.
1:01The engine is fixed, basically. Right, which means life expectancy is practically approaching that of the general population. But despite that effective viral suppression, the body still seems to age faster.
1:11People with HIV face this disproportionately high risk for aging related conditions. We're talking cardiovascular kidney and liver diseases. The health span gap is just still wide open. Yeah. So how could mapping the tiny molecules in the blood change our understanding of this accelerated aging?
1:31How do we find out why those metaphorical tires and brakes are degrading so quickly? That is the $10000 question. And that is exactly what we are exploring with you today in this deep dive, really looking into the hidden architecture of human metabolism.
1:44I'm really excited to get into this. Me too. Today, we celebrate the work of the Swiss HIV cohort study, the Equalty Polytechnique Federal de Los Anne, and the Swiss Institute of Bioinformatics, who advanced our understanding of the metabolic and genetic factors, driving accelerated aging in people with HIV.
2:01Yeah, and to really grasp the global relevance of this puzzle, you have to look at the scale. HIV currently affects over 42000000 individuals worldwide. a staggering number. It is. And for a long time, the sole focus of medicine was just keeping the virus from replicating.
2:19And anti-retroviral therapy or art E is genuinely a lifesaver there. But keeping the virus at bay doesn't erase the history of the infection. Because, what, the virus already did some foundational damage before the drugs even kicked in.
2:34Yeah, that's a huge part of it. RT only partially restores the profound metabolic disturbances that the initial infection caused, you know, all throughout the body. And on top of that, the ART medications themselves are powerful drugs, they can introduce entirely new metabolic changes.
2:50Like side effects from the drugs themselves. Exactly. So the body is constantly navigating this really complex triangle. The lingering effects of the virus, the ongoing presence of the medication, and well, just the natural aging process.
3:02Okay, let's unpack this. How do we actually look beyond the virus itself to find out what's causing these specific diseases. The virus is asleep. The drugs are doing their job. So where do researchers even begin to look for the source.
3:15of this accelerated aging. They look at the metabolone. The metabolism. Okay, yeah. If you think of your genome as the biological blueprint of what, like what could happen? Your metabolum is the real time on the ground report of what is actually happening right now in your cells.
3:31Right. And the metabolum is made up of metabolites. What are those, practically speaking? They are very small, low molecular weight molecules floating around in your biological fluids, mostly blood plasma.
3:43Like what? We were talking about things like lipids, amino acids, carbohydrates, and even xenobiotics, which are basically chemical compounds foreign to the body, like medications or environmental substances.
3:54So they're essentially the exhaust fumes, the spare parts, and the fuel left over from all our cellular processes. Which is exactly why they are so incredibly valuable to researchers. They act as proximal reporters of disease.
4:07Proximal reporters, meaning they tell us what's happening immediately. Yes. Because they are the immediate downstream result of your body's physiological functions, their levels in the blood will shift the moment a biological pathway is disturbed.
4:21Oh, wow. So the metabolite levels will show that something is wrong. long before a disease like, say, kidney failure or heart disease is fully clinically manifest. Okay, that makes perfect sense. We want to read the real time report.
4:34But just knowing what these molecules are isn't really enough to solve the aging puzzle, right? No, not at all. We need to know who or what is pulling the levers. We need to connect the genetic blueprint to that real-time report.
4:47And achieving that requires an incredibly robust study population and some highly advanced analytical technology. You really need enough data to separate the signal from the noise. Let's get into the people behind the data then.
4:59This deep dive relies on 244 participants from the Swiss HIV cohort study. A fantastic cohort to work with by the way. Yeah, they were all age 45 or older. Most were male, about 79.9%. And overwhelmingly of European descent, so around 90%.
5:15But the most crucial detail here is that the vast majority had their viral loads successfully suppressed. Right. This isn't a study about active HIV replication. Exactly. Which is great because it isolates the exact variable we want to study.
5:28The aging process under viral suppression. Another critical detail is that these participants provided overnight fasting plasma samples. Which is standard for this kind of thing. Yes, that fasting aspect is essential for metabolic research.
5:42It gives you a clean baseline reading without the immediate interference of, you know, whatever the person just ate for breakfast. You'll want to spike in blood sugar from a muffin moneying up the data.
5:51Exactly. So then the researchers took these pristine plasma samples and used what is called untargeted metabolomics. Untargeted. So they weren't looking for anything specific. Right. Utilizing mass spectrometry.
6:04The researchers measured 1930 putative metabolites across these 12,244 individuals. Let's pause on the word putative for a second. I'm assuming that means we have like a rough sketch of the molecule, but not the final high definition portrait.
6:20That is a highly accurate way to frame it. The researchers know the molecular formulas and the exact mass of these 1930 molecules with really high confidence. They know they exist and they can track their levels.
6:32But they don't know exactly what they look like. Right. They're precise structural identities, how all the atoms are arranged in three-dimensional space. Those are still pending confirmation. But even so, they were able to map these molecules across major biological superpathways, like lookids and amino acids.
6:49So we have 1930 of these molecules, and we need to figure out which human genes control their levels in the blood. It sounds like a massive, computationally heavy matchmaking exercise. It is an enormous undertaking.
7:01To do it, they performed a genome wide association study, frequently referred to as a G double. Ah, GWS. We hear that a lot. Yeah. They tested the fluctuating levels of these 1930 metabolites against 1000000s of individual genetic variants across the DNA of those 12,244 participants.
7:19So you are basically looking to see if a specific tiny tweak in a person's DNA code consistently matches up with a higher or lower level of a specific molecule floating in their blood. Yes, exactly. And when they find a strong statistical match between a genetic variant and a metabolite, they don't just stop there.
7:38Because correlation isn't causation. Right, exactly. So they run 2 innovative follow-up analyses to actually prove the underlying biology. The 1st is EQTL colloquialization. Okay, you're going to have to walk me through EQTL.
7:53Yeah, it stands for expression quantitative trait locus. That's a mouthful. It really is. But in simple terms, the researchers want to prove that the genetic variant they flagged isn't just a passive landmark on the DNA.
8:05They look into actual human tissue data to see if that exact same genetic variant actively regulates the expression of a gene nearby. It proves the variant is physically pulling the levers of biology, not just, you know, sitting close to the action.
8:19Okay, so that locks down the biological mechanism. We know the gene changes the molecule. But how do we link that molecule to the accelerated aging and the diseases we talked about earlier? Well, that brings us to the 2nd follow up step.
8:32Mendelian randomization, or MER. Ah, Mendelian randomization. Yeah. This is how you prove cause and effect in human observational data. I always found MR fascinating, but the logic can get a little dense.
8:45How do they use it here to actually prove causality? Think about the fundamental nature of your DNA. Your genes are randomly assigned at conception, right? And they are fixed. They don't change because you started eating poorly or because you developed heart disease at age 60.
9:01The genetic variant was there first. Right. The blueprint is set before the building is built. Exactly. So the researchers took the specific genetic variants they just proved were linked to these metabolites, and they cross-reference them with massive external health data sets.
9:15Like which data sets? Specifically, they use the UK biobank, which contains genetic and long-term health data for 100s of 1000s of people, tracking 18 different aging related diseases. So the logic is, if you have a genetic variant that naturally gives you a higher level of a specific metabolite from birth, and that exact same genetic variant is definitively linked to a higher risk of heart disease in the UK biobank.
9:40You can confidently say the metabolite drives the heart disease. Yes, exactly It's not just a side effect of getting older. That is the core of it. Mendelian randomization basically cuts through the noise of lifestyle factors and establishes a one way street of causality.
9:56Okay. So with that robust matchmaking and causality engine in place, what actually bubbles to the surface? Like when the dust settles on all that data, what do they find? Out of the initial 1930 molecules, they identified 27 putative metabolites that were significantly associated with 12 specific genetic low sci.
10:16That feels like a pretty solid yield. It is a very solid, high confidence yield for a cohort of this size. And looking at the breakdown of those results, the data on lipids really caught my eye. Yeah, the lipids are super interesting.
10:27The researchers looked at heritability, which is, um, how much of the variance in a metabolites level across the population is purely explained by genetics. They found that lipids had the highest genetic heritability of all the categories with a median of 21.01%.
10:43That is wild. Over a 5th of the reason your blood fat concentrations are what they are is just hardwired into your DNA. Completely independent of your diet or the anti-retroviral drugs. Yeah, it highlights a profound baseline genetic influence on cardiovascular risk factors, which is incredibly crucial when managing aging populations.
11:05So looking at the specific findings. Here's where it gets really interesting. The notes on the NTA gene really jumped out as a major focus of the study. Oh definitely. It looks like a breakthrough for kidney health, but I'm trying to connect the dots.
11:15How does a single genetic tweak actually protect the kidneys? Let's trace the pathway. The Net8 gene is primarily involved in how the body processes amino acids. The researchers identified a specific genetic variant within this gene that actually reduces the function of the NT8 enzyme.
11:32How does it do that? It alters the enzyme's binding site. So the enzyme is basically less efficient. What happens to the metabolites then? Well, because the enzyme is sluggish, people who carry this genetic variant end up with lower levels of a very specific metabolite in their blood called anacetyl cituline.
11:51Okay, so the NT8 variant equals lower anacetyl citroleum, where does the kidney protection come in? This is where they deployed mandelian randomization. They took that genetic variant and looked at the UK biobank data.
12:03They proved that genetically lower levels of anacetyl citrillin causally lead to higher serum creatinine concentrations. And for anyone who has ever looked at a routine blood panel, creatinine is a classic biomarker for impaired kidney function.
12:18The higher you're creatinine, the worse your kidneys are at filtering waste out of your blood. Exactly right. By mapping the pathway from the NAT 8 gene to the reduction in the anacetal citrilline molecule to the resulting spike in the creatine biomarker, they demonstrated that anacetal citrilline is fundamentally protective of kidney function.
12:37So when your genetics cause a lack of it? Your kidneys are just more vulnerable to wear and tear. Exactly. That is an incredible insight. We literally have a protective molecule mapped right back to the DNA code.
12:48Just amazing. But as fascinating as that is. Not all the discoveries were about the body's internal defenses, right? The finding about cores mate is honestly a little unsettling. Oh, absolutely. What's fascinating here is that choresmate is a key intermediate molecule in something called the shike metabolic pathway.
13:07And according to the source material, that pathway doesn't exist in humans. No, it doesn't. Wait, if it's a bacterial molecule. Why on earth is a purely bacterial molecule floating around in the blood plasma of these patients?
13:19It is a critical question, and it points to a phenomenon highly prevalent in HIV infection known as microbial translocation. You will often hear it referred to colloquially as leaky gut. Leaky gut. So the intestinal barrier, which is supposed to kept the bacteria in the digestive tract gets compromised, and these microbial byproducts just leak directly into the bloodstream.
13:42Yes. The gut associated lymphoid tissue is severely damaged during the early stages of an HIV infection. Oh, wow. And even after anti-retroviral therapy suppresses the virus in the blood. That gut barrier often remains highly permeable.
13:58The immune system in the gut just doesn't fully recover. So you have this chronic leakage. Exactly. Chronic leakage of bacterial molecules, like core is made, into the systemic circulation. But it doesn't just float there harmlessly, does it?
14:12Far from it. The Mandelian randomization analysis, showed something really striking. Genetically elevated levels of chorus mate, causly drive up human cholesterol and triglycerides. Seriously. So a bacterial molecule leaking from a damaged gut is directly spiking the patient's bad cholesterol, which accelerates their cardiovascular risk.
14:32It completely changes how you view heart disease in this population. It really does. And the interaction goes even deeper than that. The genetic association they found for this lipid trait actually overlaps with a human gene called A2.
14:44Then two, okay. Inette T2 encodes an enzyme expressed in the liver and intestines that is responsible for xenobiotic detoxification. Its entire job is to clear out foreign chemicals. Let me make sure I'm synthesizing this correctly.
14:58And the viral history causes the gut to leak a bacterial molecule. Core is made. Right. This molecule actively raises the patient's cholesterol, and how effectively the patient's body can fight back and clear that molecule out, depends entirely on the genetic lottery of their net to gene.
15:14You've captured the exact dynamic. Your host genetic variation in detoxication capacity acts as a modulator of the cardio metabolic damage caused by the microbial intruder. That is wild. It is a stunning multi-layered example of host microbiome interaction driving accelerated aging.
15:31There's one more specific finding on what to dig into, and that is the F11 gene variant. This one really stood out because it illustrates why we can't just rely on general population data. Yes. Why is it so important to study a specific cohort like people living with HIV?
15:45The F 11 variant is the perfect illustration of that need. In the Swiss HIV cohort, a specific genetic variant near the F11 gene, was strongly linked to plasma levels of 5 hydroxytryptophan. And 5 hydroxytryptophen is a direct precursor to serotonin, right?
16:02The neurotransmitter that regulates mood, sleep, digestion. Correct. But here is the catch. When the researchers looked for this exact genetic association in massive general population studies, so people without HIV, they didn't find it.
16:16Wait, really? It wasn't there. It wasn't there at all. Why would a genetic link controlling a serotonin precursor only show up if you have HIV? I mean, your DNA didn't change when you acquired the virus.
16:25No, it didn't. It comes down to the unique systemic stress the virus place is on the body. Chronic HIV infection specifically perturbs tryptophan metabolism. How so? The constant, low grade immune activation associated with the virus shifts the body's resources.
16:41It induces biological pathways that rapidly break down tryptophan. Oh I see. So because the whole metabolic system is under this immense, unique stress burning through tryptophan, the underlying effect of that F11 genetic variant suddenly becomes visible.
16:58Exactly. It's almost like a stress test for the genome. That's a great way to put it. The genetic variance effect is magnified by the viral environment in a way it just isn't in someone without the virus.
17:07And if you're a patient whose system is constantly breaking down tryptophan because of this chronic immune stress, that's not just an interesting data point on a spreadsheet. Absolutely not. That could be actively impacting your mood, your sleep quality, your overall quality of life as you age.
17:22Yeah, and if we connect this to the bigger picture, this study fundamentally proves that despite the massive complexities of an HIV infection, you know, the history of the virus, the damage gut, the lifelong anti-retroviral drugs fundamental human genetics, still robustly control our metabolic pathways.
17:40Right. The genetic architecture doesn't just disappear under the weight of the disease. It continues to actively shape our baseline health and our aging process. So bringing this back to the human element.
17:52What does this all mean for the patients? For someone walking into a clinic tomorrow. Well, how does mapping these metabolites actually change the way we approach care? It opens the door wide for true precision medicine.
18:04Let's look back at the NXT 8 variant and the kidneys. If we sequence a patient living with HIV and see they carry that specific genetic variant, We know immediately that their kidneys are at a higher genetic risk, because they naturally lack that protective anacetal cetraline molecule.
18:22So you don't have to wait for the creatinine levels of spike on a blood test 10 years down the line. You just know from day one. Precisely. And knowing that allows for proactive, highly targeted care. Like what?
18:33What could a doctor do? A clinician might choose an anti retroviral regimen that is known to be less taxing on the kidneys, they might monitor kidney function more aggressively, or intervene with dietary and lifestyle recommendations long before any clinical symptoms of kidney failure appear?
18:49We can manage the wear and tear based on a patient's unique genetic code. I am fully on board with the incredible potential here. But I have to push back a little on a potential flaw in the logic. Specifically regarding the Mendelian randomization.
19:04Okay, let's hear it. You mentioned they use health outcome data from the general population in the UK biobank to prove cause and effect for these diseases. Yes I did. But we just talked about how the F11 gene behaves completely differently in people with HIV doesn't using the UK biobank assume the genetics work exactly the same in people with HIV as they do in the general public.
19:25You're hitting on the crucial limitation here, and it is a very sharp observation. Okay. The researchers are actually highly transparent about this. The Mendelian randomization relies on the assumption of transferability, basically, the assumption that a genetic variant causing kidney disease in the general UK biobank population is causing the disease via the exact same pathway in the Swiss HIV cohort.
19:50And while human biology is generally consistent, the HIV environment is a massive wild card, as we've seen. It is. So for the metabolites that were not able to be cross-referenced with external HIV specific data sets, we have to interpret these as putative or presumed causal associations.
20:08Okay, that makes sense. They are incredibly strong leads, but they aren't definitive proof of HIV specific disease mechanisms until they are replicated in other HIV cohorts. Right. What other limitations in the study design do we need to keep in mind?
20:22Because no single study has all the answers. Another major limitation goes back to our definition of putative metabolites. The exact structural identities of many of these 1930 molecules still need to be rigorously confirmed in the lab.
20:37So the high definition portraits are still pending. Exactly. Furthermore, this was a single cohort heavily skewed toward European males. Yeah, 90% European, nearly 80% male. Right, which severely limits how broadly we can generalize these genetic findings to diverse global populations, especially considering the global footprint of HIV.
20:55And finally, while they used fasting samples, they lacked detailed long-term dietary and lifestyle data, which can, you know, exert confounding effects on metabolism over a lifetime. Which just means there is plenty more work to do.
21:09Where does the research go from here? What are the next steps to bring this closer to the clinic? The immediate priority is funding and building larger, highly diverse longitudinal cohorts. We need varied ethnicities, more women, and matched control groups of people without HIV.
21:24That makes total sense. Moving forward, researchers also want to integrate what are known as epigenetic clocks. Epigenetic clocks. Yeah, these measure exact biological aging at the cellular level, totally independent of your chronological age.
21:38Oh, wow. Combining epigenetic clocks with this metabolomic data could help us precisely measure exactly how much of the accelerated aging is driven by the viral history, how much by the medication, and how much by the patient's genetic baseline.
21:51We have covered a massive amount of complex ground today. How would you distill this entire deep dive into a central takeaway for the listener? I'd say this. By integrating genetics and untargeted metabolomics, this research uncovers how specific host DNA variants and even microbial molecules leaking from a damaged gut directly shape the risk of kidney and cardiovascular disease in people with HIV.
22:15It proves that aging with HIV is a complex interaction between the virus, the medication, the microbiome, and the patient's own genetic code. You know, it brings us right back to that mechanic analogy we started with.
22:27We figured out how to keep the engine running with anti-retrovirals, which is just one of the greatest scientific triumphs of our lifetime. Absolutely. But now we finally have the tool, the metabolomics, the genetics, to look at every single bolt brake pad entire on that chassis.
22:43We are learning how to maintain the whole vehicle. It's a whole new frontier. So as you go about your day, I leave you with this. What does this mean for the future of personalized medicine? Not just for viral suppression, but for managing the long-term metabolic health of aging populations everywhere?
22:58It's a lot to think about. It really is. This episode is 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.
23:15If you'd like to support our work, use a 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.
23:24Thanks for listening, and join us next time as we explore more science, based by base.
