416 ‒ The science of mitochondria: their role in health, aging, disease, and much more
Audio Brief
Show transcript
Does mitochondrial dysfunction actually drive normal human aging, or is that popular health buzzword misleading? On Peter Attia's show, released on November 16, 2026, Attia and researcher Navdeep Chandel examine what mitochondria really do and the clinical evidence behind popular therapies. By the end, you will know how mitochondria work, why most supplements fail, and how to test your own cellular fitness.
Mitochondria are far more than simple cellular powerhouses. Beyond making adenosine triphosphate, or ATP, for cellular energy, their metabolic cycle acts as an anabolic engine that makes building blocks for DNA, RNA, lipids, and amino acids, while also generating signals to guide cell decisions and releasing cytochrome c to trigger cell death. Liver cells, for instance, are packed with thousands of mitochondria not for energy, but to build precursors like glucose.
Those cellular signals depend on a surprising culprit: free radicals. As mitochondria pass electrons to oxygen, they leak a few, creating superoxide that turns into hydrogen peroxide; in genetic mouse experiments, decreasing mitochondrial hydrogen peroxide stopped T-cells from becoming memory T-cells. Skip high-dose antioxidant supplements, which worsened cancer and sepsis outcomes in clinical trials, and rely on a proper diet so you do not block this vital immune signaling.
Mitochondria also control your epigenetic clocks and age-related inflammation. Their cycle produces alpha-ketoglutarate, a metabolite that fuels 75 enzymes regulating DNA and histone demethylation, with Vitamin C acting primarily to recycle iron for those reactions, while Chandel hypothesizes that chronic, low-level leakage of mitochondrial DNA or RNA over 20 to 30 years may drive age-related inflammation. Eat two oranges a day to get enough Vitamin C to keep those 75 epigenetic enzymes running.
Even when mitochondrial energy fails, your cells have a massive backup capacity. In human mitochondrial diseases, you must lose 70 to 80 percent of mitochondrial protein function before a tissue shows a clinical deficit, which is why mouse hearts with knocked-out electron transport chains still pumped for months purely on glucose fermentation, or glycolysis.
Because of this massive reserve, mitochondrial decline likely does not cause normal aging. Mutator mice require ten-fold higher mitochondrial DNA mutations than occur in normal aging before degenerating, though in exercise, a recent longitudinal study showed that while trained people are limited by heart output, unfit, non-exercising individuals are actually rate-limited by their mitochondria. Measure the power output you can sustain on a bike or treadmill while keeping steady-state blood lactate below 2 millimoles per liter to gauge your true mitochondrial fitness without a tissue biopsy.
When it comes to drugs like metformin, its effects happen mostly in the gut, not the muscle. Because metformin requires specific Organic Cation Transporters to enter cells, oral doses accumulate at up to 1 millimolar in the gut, where grad student Zach Zubo found in human data that the number one metabolite that drops on metformin is gut-derived citrulline, which Chandel hypothesizes blunts exercise adaptations alongside reduced Vitamin B12 absorption. If you take metformin, take it about 30 minutes before eating to blunt glucose spikes, and supplement citrulline and Vitamin B12 to help offset its exercise blunting.
Those same cellular transporters explain why cancer trials and popular supplements often disappoint. A CRISPR screen of 3,000 genes in Chandel's lab showed cancer cells survive metformin simply by losing the OCT3 transporter, while in supplements, free plasma urolithin A is undetectable in picograms because the liver modifies it, and NAD, CoQ10, and cardiolipin do not drop to rate-limiting levels in normal aging. Before buying any compound, ask whether the pathway is rate-limiting in a healthy human, where it goes in the body, and what its target is, and choose cheaper, regulated metformin over berberine since mouse experiments prove they work identically.
Before spending money on non-invasive mitochondrial panels or longevity hype, remember that healthy human mitochondria already have massive reserve capacity, and most oral compounds never reach your muscles or lack a rate-limiting target. Two oranges a day is enough to make those enzymes run.
Episode Overview
- Peter Attia and Navdeep Chandel examine what mitochondria do beyond generating ATP, whether mitochondrial dysfunction actually drives normal human aging, and the mechanisms and clinical evidence behind popular mitochondrial drugs, supplements, and therapies.
- The discussion reframes mitochondria from simple cellular powerhouses into anabolic engines and signaling hubs that regulate epigenetics, immunity, and cell death.
- Chandel challenges the mainstream longevity narrative by explaining why human mitochondria possess massive reserve capacity and why sufficiency experiments in mice do not prove that mitochondrial decline causes normal aging.
- A deep dive into metformin reveals how its reliance on specific cellular transporters concentrates the drug in the gut rather than muscle, explaining its effects on glucose, exercise adaptations, cancer, and inflammation.
- The episode concludes with a rigorous evaluation of commercial mitochondrial tests and interventions, including berberine, urolithin A, NAD precursors, CoQ10, methylene blue, hyperbaric oxygen, and red light therapy.
Key Concepts
- The Four Functions of Mitochondria: Beyond generating ATP via the electron transport chain (ETC), which generates a proton-motive force or voltage harnessed to make ATP (while glycolysis also makes ATP, mitochondria are more efficient), mitochondria serve three other vital roles. Second, the coupled Krebs/TCA cycle is an anabolic engine that produces metabolites used as building blocks for new lipids, nucleotides (DNA and RNA), and amino acids. Third, over the last 30 years, research has shown mitochondria act as decision-making organelles that generate signals to the rest of the cell. Fourth, as discovered by Xiaodong Wang in the mid-1990s, when cytochrome c is released from mitochondria, it activates caspases and triggers mammalian apoptosis (cell death). Red blood cells are the only cells in the body without mitochondria, while other cells have thousands or more (though Chandel admits he does not know the exact number of mitochondria in a neuron); for instance, hepatocytes (liver cells) are packed with them not for ATP, but for anabolic precursors like making glucose.
- Electron Transport, Reactive Oxygen Species (ROS), and Ferroptosis: Glucose converts to pyruvate, enters mitochondria, and becomes acetyl-CoA (with oxaloacetate) to jumpstart the TCA cycle, converting NAD to NADH. NADH carries two electrons to Complex I to regenerate NAD and pass electrons down the chain to oxygen to form water ($H_2O$) and $CO_2$. Because the ETC is slightly imperfect, it leaks electrons at Complex I and III, converting oxygen to superoxide ($O_2^{\bullet-}$), which superoxide dismutases (SOD) rapidly convert into hydrogen peroxide ($H_2O_2$). $H_2O_2$ has three fates: catalase and peroxiredoxins can detoxify it into water; it can oxidize specific cysteine residues on proteins to drive essential cellular signaling (in genetic mouse experiments, decreasing mitochondrial $H_2O_2$ caused T-cells to proliferate normally when exposed to a pathogen, but they failed to differentiate into memory T-cells); or it can react with iron via the Fenton reaction to form toxic hydroxyl radicals that react with polyunsaturated fatty acids to create lipid peroxides, triggering a chain reaction called ferroptosis (a form of cell death discovered about 12 years ago). Vitamin E and selenium-dependent proteins like GPX4 detoxify lipid peroxides (the U.S. has high soil/dietary selenium, whereas northern New Zealand, Sweden, and Finland are somewhat deficient). Clinical trials of high-dose antioxidants failed and worsened outcomes in cancer and sepsis because a proper diet provides sufficient antioxidant capacity and excess antioxidants block normal $H_2O_2$ signaling.
- TCA Metabolite Signaling, Epigenetics, and Inflammaging: Mitochondria regulate epigenetics through TCA metabolites. They control acetyl-CoA levels, which acetylate lysines on proteins and histones to keep chromatin open or closed and drive autophagy. They also produce alpha-ketoglutarate (AKG), which fuels a class of 75 enzymes called alpha-ketoglutarate dioxygenases (which use oxygen, AKG, and iron to make succinate, while Vitamin C acts primarily as a cofactor to recycle iron $Fe^{2+}$ rather than as an antioxidant). These 75 enzymes include collagen prolyl hydroxylases, histone demethylases, and DNA demethylases that directly control the reactions behind epigenetic clocks. Additionally, because eukaryotic cells evolved from a symbiosis between an archaea and an alpha-proteobacteria before oxygen existed, innate immune sensors (cGAS-STING, MAVS) view leaked mitochondrial DNA or RNA in the cytosol as a foreign pathogen and trigger Type I interferons. In a study with a Northwestern colleague, poxviruses caused cells to release mitochondrial DNA to amplify the immune response. Chandel hypothesizes that chronic, low-level leakage of mitochondrial DNA or RNA over 20 to 30 years may drive age-related autoimmunity and inflammaging, though he emphasizes that how it leaks out of the double membrane without killing the cell remains a mechanistic "black box."
- Massive Reserve Capacity and the Survival of Cells on Glycolysis: While cyanide is rapidly lethal to an organism because it binds cytochrome oxidase, genetic knockout experiments in mice show that if you knock out the ETC in heavy bioenergetic tissues, AMPK is activated, mTOR is turned down, and cells survive for months purely on glycolysis. A Northwestern colleague knocked out ETC components in mouse hearts, which pumped on glycolysis for months before signaling caused hypertrophy and death. In Chandel's lab, knocking out Complex I in mouse dopaminergic neurons (studied via grant funding from the Michael J. Fox Foundation for Parkinson's disease and Complex I gene therapy, alongside NIH funding) allowed the neurons to live on glycolysis for 3 months before dying, though they stopped making dopamine during those 3 months. Furthermore, human genetics of mitochondrial diseases show you must lose 70% to 80% of mitochondrial protein function (leaving 20% to 30% remaining) before a tissue shows a clinical deficit. Chandel compares this to a new Aston Martin (217 mph) versus a 1964 Goldfinger Aston Martin (140 mph) driving in Chicago where the speed limit is 50 mph: top capacity is not rate-limiting.
- Why Mitochondrial Dysfunction Likely Does Not Drive Normal Aging: Chandel notes it is almost a reverse conflict of interest for him to state he is skeptical that mitochondrial dysfunction drives normal aging. He explains that the "Hallmarks of Aging" rely on sufficiency experiments (breaking an organelle in a mouse makes it look old) rather than proving necessity in normal aging. Inducing nuclear DNA mutations in mice at levels seen in normal aging causes aging hallmarks, whereas mitochondrial DNA mutator mice require 10-fold higher mitochondrial DNA mutations than occur in normal aging before degenerating, and knocking out mitochondrial function in mouse microglia at 1 year did not produce an aged phenotype. Chandel also doubts human heart failure is caused by intrinsic mitochondrial dysfunction because SGLT2 inhibitors act on the kidney to excrete glucose and globally reprogram metabolism to rapidly improve heart function, which would not work if heart mitochondria were intrinsically broken. In exercise, Attia notes $VO_2$ max declines ~10% per decade (falling below 28 to 30 mL/kg/min would ruin his quality of life, and 15 to 16 mL/kg/min is the threshold for independent living); while maximum aerobic output in trained people is rate-limited by cardiac output and oxygen delivery, a recent longitudinal study showed unfit, untrained, non-exercising individuals are rate-limited by peripheral oxygen extraction (mitochondria).
- Metformin's Gut-Specific Mechanism, Citrulline Depletion, and Inflammation: Attia reviews the Bannister paper (claiming diabetics on metformin had better survival than non-diabetics) versus the more rigorous Keys paper ~8 years later (which challenged those findings), alongside failed Phase 3 breast cancer trials, mixed prostate cancer trials, and the NIA Interventions Testing Program (ITP), where canagliflozin and acarbose extended mouse lifespan while metformin did not (Chandel states he does not have a good explanation for why metformin failed in the ITP). Metformin requires Organic Cation Transporters (OCT1, OCT2, OCT3) to cross the plasma membrane (though Chandel notes we still do not know how it crosses the mitochondrial membrane, where its positive charge causes it to enrich 100x to 500x inside the negatively charged matrix). $C^{11}$-labeled PET imaging shows oral metformin accumulates at up to 1 millimolar (1 mM) in the gut, drops 20x in the portal vein/liver (50 to 100 $\mu$M), circulates in blood at 10 to 20 $\mu$M, and does not enter the brain, heart, or muscle at appreciable levels (though macrophages and T-cells take it up). Oncologists stop metformin before FDG-PET scans because it inhibits Complex I in the gut, forcing enterocytes to take up massive amounts of glucose for glycolysis and lighting up the gut. Mammalian Complex I is a 45-subunit boot-shaped complex (whose metformin binding site has been mapped by cryo-EM), whereas yeast have a single-protein homodimer called NDI1 that performs the same function but is refractory to metformin. In Chandel's lab, Colleen Reczek put NDI1 into a whole mouse, abrogating metformin's glucose-lowering effect, and grad student Zach Zubo expressed NDI1 only in the mouse gut (preprint), which completely abrogated the gut FDG-PET signal and blocked metformin's ability to raise GDF15, GLP-1, and Lac-Phe. Zubo also found in human metabolomics data that the #1 metabolite that drops on metformin is citrulline (made by enterocytes using mitochondrial ATP to drive kidney arginine and nitric oxide vasodilation). Chandel hypothesizes metformin blunts exercise $VO_2$ max and hypertrophy not by entering muscle, but by lowering circulating citrulline and Vitamin B12 absorption (raising methylmalonic acid, a Complex II inhibitor). When Chandel mentions metformin improves the gut microbiome (like Akkermansia muciniphila), Attia pushes back strongly, assigning "very low confidence" to microbiome claims due to scientific illiteracy in the field, and Chandel agrees it is not a primary reason to take the drug. However, metformin shows strong anti-inflammatory effects: Northwestern colleagues Scott Budinger and Gokhan Mutlu (Cell Metabolism) showed that Beijing-level $PM_{2.5}$ air pollution entered mouse alveolar macrophage mitochondria, increased $H_2O_2$ signaling, and drove up prothrombotic IL-6, while metformin in drinking water inhibited Complex I, blocked $H_2O_2$ release, and lowered IL-6. Additionally, in Clonal Hematopoiesis of Indeterminate Potential (CHIP), where 3% to 20% of aging hematopoietic stem cells pick up mutations and form rogue clones driving inflammation and cardiovascular disease, three recent papers (two in Nature, one in Nature Communications) showed that in the 450,000-person UK Biobank, people on metformin had lower rates of CHIP, and in mouse models metformin (at 50 $\mu$M in vitro) blocked rogue clone outgrowth via Complex I inhibition.
- Cancer Metabolism and the Pharmacokinetics of Mitochondrial Supplements: Although Otto Warburg won the Nobel Prize for discovering cytochrome c oxidase and proposed the "Warburg effect" in the 1920s (that cancer cells make lactate in the presence of oxygen due to "injury to respiration"), 95% to 98% of cancers actually have intact ETC and TCA cycles because they need TCA metabolites for anabolic growth. In Chandel's lab, Karthik Vasan used CRISPR to knock out 3,000 metabolic genes in a metformin-sensitive cancer cell line; the only clone that survived high-dose metformin had lost OCT3. Clinical trials of metformin in cancer (like breast cancer) failed because most tumors lack OCT transporters and trials never stratified patients by OCT positivity or tumor FDG-PET response; when tumors do have OCTs, metformin inhibits Complex I, depletes TCA building blocks, and stops oxygen consumption (reducing tumor hypoxia to aid radiation and immunotherapy). Evaluating supplements (regarding which Chandel states he has no conflict of interest), Chandel notes:
- Urolithin A: Rapidly conjugated with glucuronic acid by the liver; supplementary data from a human inflammation paper showed free plasma urolithin A is in picograms (undetectable) while the conjugate is at 5 to 10 $\mu$M and cannot enter cells unless cleaved by enzymes near immune cells. After 10 years, no direct molecular target has been identified for how it induces mitophagy, though a mouse study showed improved anti-tumor T-cell immunity and a small human trial showed reduced CD8+ T-cell exhaustion in the elderly.
- AKG: Has 75 enzyme targets, though intracellular levels (~100 $\mu$M) may not be rate-limiting in normal aging; Brian Kennedy's group showed lifespan/healthspan effects in female mice, and another study showed AKG protected mice against colitis and 5-FU chemotherapy gut damage by promoting secretory over absorptive stem cells.
- NAD Precursors (NR/NMN) and SS-31: Referencing a recent New York Times article, Chandel notes NAD does not decline in normal aging to rate-limiting levels, and NR failed in the ITP, though a study by Joe Baur at UPenn showed NAD supplementation boosted liver regeneration after surgery. SS-31 (elamipretide) is FDA-approved for Barth syndrome (a rare genetic mutation in tafazzin causing cardiolipin deficiency), but cardiolipin is not rate-limiting in normal people.
- Oxygen and Light Therapies: Hypoxia therapy activates the HIF-prolyl hydroxylase-VHL pathway (2019 Nobel Prize to Semenza, Ratcliffe, and Kaelin) to boost EPO and hematocrit, while red light therapy likely hits cytochrome c oxidase (potentially popping off nitric oxide or mildly inhibiting it to trigger a small $H_2O_2$ signal for skin collagen synthesis, and Attia cites a human study where red light therapy on the back reduced the glucose area-under-the-curve during an oral glucose tolerance test).
Quotes
- At 0:01:10 - "I think it's not only mitochondria, but the other big buzzword is mitochondrial dysfunction. Right? That's the real buzzword which you hear over and over." - Chandel introducing the central skepticism of the episode regarding how mitochondrial decline is framed in popular health.
- At 0:02:51 - "And it turns out that the TCA cycle is the anabolic engine also in your cells. So mitochondria do two things: they make ATP, and they can make the building blocks, the metabolites that are the building blocks for new lipids, new nucleotides like DNA, RNA, right, amino acids for the building blocks of the protein." - Chandel explaining the essential biosynthetic role of mitochondria beyond energy production.
- At 0:06:20 - "And that third function is that mitochondria are constantly generating signals to the rest of the cell, and these signals are optimized for cellular function." - Chandel describing the paradigm shift that redefined mitochondria as cellular decision-makers.
- At 0:13:20 - "That can, with a polyunsaturated fatty acid, can make something called lipid peroxide and start a chain reaction which is now called ferroptosis. It's a new form of cell death discovered about 12 years ago." - Chandel detailing what happens when hydrogen peroxide reacts with iron instead of being detoxified or used for signaling.
- At 0:16:20 - "Two oranges a day is enough to make those enzymes run." - Chandel explaining that small dietary amounts of Vitamin C are sufficient to recycle iron for the 75 alpha-ketoglutarate dioxygenase enzymes.
- At 0:17:50 - "And we've done genetic experiments where we decrease mitochondria's ability to make H2O2, for example. A T-cell, when it sees a pathogen, will proliferate okay, but it won't make the memory T-cell." - Chandel illustrating why reactive oxygen species are required for adaptive immunity and why high-dose antioxidants can be harmful.
- At 0:21:06 - "The people who are so into their epigenetic clocks, alpha-ketoglutarate is controlling those clocks because they're needed to keep the demethylase reactions going properly." - Chandel connecting TCA cycle metabolites directly to DNA and histone methylation.
- At 0:22:02 - "So there's a bacteria within us, and if that bacteria releases its DNA or its RNA, it's viewed as a foreign, and the same inflammation pathways..." - Chandel explaining how the evolutionary origin of mitochondria can drive innate immune activation and inflammaging.
- At 0:30:43 - "A colleague of mine at Northwestern knocked out components of the electron transport chain; the heart continued to pump on glycolysis." - Chandel demonstrating that even highly bioenergetic tissues can survive for months without mitochondrial ATP.
- At 0:34:25 - "I have none without taking a biopsy." - Chandel answering what non-invasive blood or breath test a patient can take to directly measure mitochondrial function.
- At 0:35:12 - "You have to lose 70% of mitochondrial protein function. It is that much excess capacity." - Chandel quantifying the massive redundancy built into human mitochondria before clinical deficits appear.
- At 0:38:29 - "Because as long as lactate is below 2 millimole in steady state, this person is maximizing mitochondrial capacity. Once lactate starts to rise, you know that they have hit that threshold and now they are shunting through glycolysis." - Attia describing his practical clinical exercise test for assessing peripheral mitochondrial fitness.
- At 0:49:10 - "The hallmarks, by the way, they're instructive, but they're misleading, all the hallmarks. They are what's called a sufficiency experiment." - Chandel critiquing the Hallmarks of Aging framework for confusing artificial genetic damage with natural aging processes.
- At 0:57:33 - "It doesn't get in the brain, it doesn't get in the heart, it doesn't get in the muscle at appreciable levels. Why is that? Because metformin needs a transporter, the organic cation transporters." - Chandel clarifying the pharmacokinetics of oral metformin and why it acts primarily in the gut and liver.
- At 1:04:23 - "He was looking at metabolomics of people on metformin. Do you know what the number one metabolite goes down when people are on metformin? Citrulline." - Chandel revealing the gut-mediated drop in citrulline that may explain how metformin blunts exercise adaptations.
- At 1:15:11 - "One guy lived. One clone lived. It lost the organic cation transporter. The plasma membrane transporter, if you lose it, it laughs at metformin. It doesn't care." - Chandel explaining his lab's CRISPR screen showing why tumors without OCT3 are completely resistant to metformin.
- At 1:42:11 - "In fact, 95, 98% of cancers have a very intact electron transport chain, TCA cycle, and the reason is because they use it to make ATP, but also to make those anabolic precursors, the metabolites for growth." - Chandel debunking the popular misconception of the Warburg effect that cancer cells have broken mitochondria.
- At 1:44:11 - "If you look at the supplementary data, how much is in the plasma of urolithin of those people who took it? Picograms. Undetectable. Because the liver puts glucuronic acid on it." - Chandel pointing out the bioavailability limitations of oral urolithin A supplements.
- At 1:48:58 - "So berberine, as far as I can tell, works exactly like metformin, and gee whiz, metformin is probably cheaper than berberine and is regulated. So use metformin if you're going to use berberine. Sorry." - Chandel summarizing his lab's yeast NDI1 mouse findings on berberine's mechanism of action.
- At 2:12:39 - "We have never had a moonshot for metabolism. Whether it's the government, whether it's a foundation, whether it's philanthropy." - Chandel advocating for a massive increase in government, philanthropic, and biotech funding for metabolism research.
Takeaways
- Evaluate mitochondrial supplements using three strict criteria: Before buying any compound, ask whether the targeted pathway is actually rate-limiting in a healthy human, where the oral compound goes in the body, and what its exact molecular target is.
- Skip commercial blood and breath tests for mitochondrial function: Do not waste money on non-invasive organic acid or acylcarnitine panels, as measuring steady-state metabolites cannot tell you whether metabolic flux is fast or backed up without a tissue biopsy.
- Use Zone 2 exercise output to gauge functional mitochondrial capacity: Measure the power output (in watts or METs) you can sustain on a bike or treadmill while keeping steady-state blood lactate below 2 mmol/L (which occurs at 60% to 70% of $VO_2$ max; for context, Attia's lactate rises above 2 mmol/L at ~230 watts, whereas Tour de France cyclist Tadej Pogačar maintains 7 W/kg and stays below 2 mmol/L up to 350 watts).
- Avoid high-dose antioxidant supplements: Rely on a proper diet (for example, eating two oranges a day provides enough Vitamin C to keep your 75 alpha-ketoglutarate dioxygenase enzymes running) rather than taking massive doses of antioxidants that interfere with beneficial $H_2O_2$ cysteine-oxidation signaling and memory T-cell formation.
- Supplement citrulline and Vitamin B12 if you take metformin: Because metformin inhibits Complex I in the gut and lowers enterocyte ATP, it reduces circulating citrulline (needed for arginine and nitric oxide vasodilation) and lowers Vitamin B12 absorption (which raises methylmalonic acid, a Complex II inhibitor); supplementing both may help offset metformin's blunting of exercise performance.
- Time metformin ~30 minutes before meals and weigh its use carefully if you are insulin-sensitive: Taking metformin ~30 minutes before eating helps blunt post-meal glucose spikes in the gut, though both Attia and Chandel remain "on the fence" about whether healthy, insulin-sensitive people should take it (Chandel, at age 50, takes a low-dose statin, exercises, and sleeps, and wrestles daily with whether to take metformin for anti-inflammation, while noting neither he nor Attia has an informed enough opinion on whether it interferes with GLP-1 agonists).
- Choose regulated metformin over berberine: Because genetic experiments in yeast NDI1 mice prove berberine lowers blood glucose via the exact same gut Complex I inhibition mechanism as metformin, opt for metformin, which is cheaper and FDA-regulated.
- Do not take carnitine, CoQ10, NAD precursors, or SS-31 expecting an energy boost in normal health: Carnitine (especially in meat-eaters), CoQ10, NAD, and cardiolipin are not rate-limiting in healthy individuals; however, CoQ10 may help statin-induced muscle symptoms anecdotally (since statins inhibit the upstream CoQ10 pathway, though trials have largely failed and bioavailability is poor).
- Exercise caution with methylene blue, especially if taking SSRIs: While methylene blue shuttles electrons from Complex I directly to Complex IV to reduce $H_2O_2$ generation, it has a narrow therapeutic window and inhibits monoamine oxidase (MAO), which can trigger severe serotonin toxicity in people on SSRIs.
- Avoid hyperbaric oxygen therapy (HBOT) for general longevity: Outside of clinical wound healing, avoid HBOT; the longevity hype stems from a flawed Israeli study on telomere length, whereas high oxygen lacks a cellular receptor, generates damaging free radicals, and causes lung epithelial cell senescence and cell death in mice and ICU patients.
- Be cautious with hypoxia therapy in older adults: While low-oxygen therapy activates the HIF pathway to boost EPO and hematocrit for athletes, it is not a general longevity panacea and carries the risk of hypoxia-induced pulmonary edema in the elderly.
- Stratify cancer trials by OCT transporter status if testing metformin: For researchers and clinicians investigating metformin in oncology, patients must be stratified by whether their tumors express Organic Cation Transporters (like OCT3) or light up on an FDG-PET scan after metformin administration, as tumors lacking these transporters cannot take up the drug.