Summary of main points:

• Resveratrol is sold as a calorie-restriction mimetic that extends lifespan. The founding evidence behind this is suspect.

• David Sinclair’s 2003 yeast result (Sir2 activation, large lifespan boost) failed to replicate.

• Calorie restriction does not universally increase lifespan. Across 41 inbred mouse strains, 40% restriction shortened lifespan in more strains than it lengthened.

• In genetically diverse mice, calorie restriction can extend lab lifespan, but the lowest-weight animals die sooner; genetics explains one-quarter of the variance, and immune costs could hurt longevity outside laboratory conditions.

• Shorter-lived strains tend to see more longevity benefit; longer-lived strains are more likely to show no benefit or decreases in lifespan.

• Human urinary resveratrol metabolite data: no link to inflammation, disease, or death.

Right now, armies of soulless AI bots are raking in views for health influencers, which pretty much means anyone with an internet connection. Different breeds have evolved, ranging from longevity gurus like Harvard professors to a random Instagram profile repeating the “science-based” talking points shared via email by an affiliate marketing manager at some supplement startup.

Many of the posts you see today are mindless regurgitations of the main published finding of a scientific study, assumed to be fact simply because it was published. A new paper claims that some molecule caused an inbred strain of mice to live 10% longer, or whatever. Ten minutes after publication, Dr. Rhonda Patrick’s social media team has scheduled a month’s worth of posts about adding it to your daily cocktail of fish oil, resveratrol, and 20 other supplements. For longevity. Based on scientific papers published by scientists proving—scientifically—the imminence of your immortality. Buy. This. Supplement.

Speaking of resveratrol…

This supposed longevity molecule has achieved celebrity status due to the work of Harvard professor Dr. David Sinclair, a scientist who has scientifically proven the science behind resveratrol. He supplements with it himself and published many scientific papers about it. At Harvard. In journals like Nature.

In M&M 151 another longevity scientist, Dr. Matt Kaeberlein, told a very different story from the one that Dr. Sinclair tells about resveratrol.

Let’s start with the basics of resveratrol, then look at the papers from David Sinclair’s lab that kicked off the resveratrol craze, considering them in the context of the wider literature. Doing so will provide an important lesson about why you should never take published scientific results to the bank just because they’re published—especially from one lab.

Spoiler alert: You’re almost certainly wasting your money on that resveratrol supplement.

What is Resveratrol?

Resveratrol is a plant polyphenol found in grape skins and sold in capsules. It has a reputation as “the red wine molecule” and is well-known among longevity enthusiasts. That reputation arose from a specific scientific story popularized by Sinclair: activate sirtuins, mimic calorie restriction, live longer.

This story rests on key results from Sinclair’s lab, which are bundled up with the popular belief, even among scientists, that calorie restriction is a surefire way to increase lifespan in animals. As we will see, this belief is premature at best.

In our conversation, University of Washington biologist Matt Kaeberlein walked through what that evidence actually says about resveratrol, sharing some suspicious details about Sinclair’s approach to research. His summary was not subtle: nobody should take resveratrol thinking it’s a longevity drug. In his words:

Learn more in this M&M episode:

• Podcast | Aging, mTOR, the Truth of Resveratrol & Longevity Supplements, Anti-Aging Myths | Matt Kaeberlein | Episode 151

Origins of the Resveratrol Narrative

The founding paper in this saga is a 2003 Nature paper by David Sinclair, perhaps the most well-known longevity scientist working today. In that yeast study, a screen of plant polyphenols found that resveratrol activates a protein called Sir2 in a test-tube assay and extended yeast lifespan considerably, in a Sir2-dependent way. A major and highly influential finding. (The details of sirtuin biology will not be covered here).

Three years later came another Sinclair study. Metabolically unhealthy mice on an unconventional high-calorie diet lived longer and looked healthier when given resveratrol. The popular version of the story wrote itself: the “red wine molecule” will make you live longer.

One of many science-y graphics you can find online, illustrating Sinclair’s basic thesis: resveratrol produces adaptive metabolic responses that mimic calorie restriction, which promotes longevity. The adaptive metabolic response here involves activating signaling proteins known as sirtuins, downstream of a major metabolic signaling pathway, AMP Kinase (AMPK), which acts as a kind of energy gauge for cells (read this article first if you want more detail).

When I asked Dr. Kaeberlein if he thought the original yeast result holds true, he responded bluntly:

The data from the 2003 Sinclair paper shows dramatic results. Resveratrol produced a huge longevity boost in yeast, comparable to the effect of calorie restriction itself.

Share

Experimental Artifacts & Fishy Behavior

Kaeberlein and colleagues tried to reproduce the yeast result, running it more than thirty times, including in the particular strain used by Sinclair’s group. They did not observe Sir2 activation or lifespan extension. How did this failure to replicate Sir2 activation by resveratrol come about? Here’s some of the scientific context, from Dr. Kaeberlein:

So, Kaeberlein first established that the lifespan extension effects of calorie restriction did not require Sir2 activation, which was a big hypothesis at the time. There were apparently two different pathways at work here, each of which could boost longevity: one was a Sir2-independent effect of caloric restriction; the other was Sinclair’s big result of activation of Sir2 by compounds like resveratrol.

If Sinclair’s result was true, it meant Kaeberlein and colleagues now had two independent ways to boost longevity: calorie restriction acting through a sir2-independent pathway; pharmacological activation of Sir2, using resveratrol. Good molecular biologists love having multiple ways of achieving the same end result.

Diligent scientists like to show key results using multiple, independent methods. Everyone with good training in molecular genetics thinks this way. Kaeberlein was simply trying to replicate the Sinclair finding on resveratrol activating sir2—not because they were trying to disprove Sinclair, but because they presumed a high-profile result published in Nature was true, giving them a second means of boosting longevity to test the ideas they were working on.

Another anecdote, illustrating not only that Kaeberlein was acting in good faith and just trying to be a diligent scientist, but also that—perhaps—something fishy was going on with Sinclair’s lab at Harvard.

The devil’s in the methodological details.

Kaeberlein could replicate the test tube experiment that Sinclair had used, but not activate Sir2 in yeast cells. In the test tube experiment, the substrate bound by resveratrol was a labeled peptide. The in vitro “activation” depended on a fluorescent tag on the peptide substrate (Fluor de Lys), not on native histone peptides. That is why the tube assay lit up and the yeast cells did not. Kaeberlein published that result.

Okay, so the test tube experiment is an artifact. That assay does show that resveratrol can activate a structurally similar peptide—Sinclair’s group did that experiment and saw activation, as did Kaeberlein. But Sinclair’s big result was that resveratrol activated Sir2 in yeast, resulting in major lifespan gains. Kaeberlein could not replicate that finding, and Sinclair refused to help him replicate that result by having his scientists work side-by-side with Kaeberlein to reproduce it. That is very suspicious.

Here’s the Kaeberlein paper showing that resveratrol does not improve lifespan in yeast, contradicting Sinclair’s famous paper:

What are we to make of the original Sinclair result, then? Did they actually find that resveratrol activated Sir2 in yeast and extended lifespan? If so, then Kaeberlein and colleagues must have somehow failed to replicate the Sinclair experiment properly despite all efforts to do so, including offering to go to Sinclair’s lab and physically do the experiments alongside his scientists—which, it bears repeating, David Sinclair would not allow for some strange reason.

An obvious alternative explanation: the Sinclair result is fraudulent, or due to some kind of very odd, but honest, mistake. I’m simply stating these as formal possibilities. We have know way to know for sure. But remember: Sinclair did not allow Kaeberlein to work side-by-side with scientists from his lab to replicate Sinclair’s own finding—suspicious.

One way to approach situations like this is to look for patterns of odd results exhibited by the same lab. Recall that the yeast paper by Sinclair was the initial paper that kicked off the resveratrol story. But it was the subsequent mouse paper that really turned resveratrol into the longevity-promoting “red wine molecule.”

Let’s consider that one in more detail…

Morbidly Obese Mice & Weird Science

If the Sinclair yeast result was true, a natural next set of experiments would be to give resveratrol to other creatures, such as rodents. The obvious way to go about this would be to give resveratrol to ordinary lab mice, under ordinary conditions. Just add some resveratrol to their normal diet, measure lifespan, and compare that to the effect of calorie restriction—similar to what was initially done in yeast. Oddly, that’s not what the 2006 Sinclair paper did.

The 2006 Nature result looked at survival on a high-calorie, obesity-promoting diet, not lifespan extension in normally aging mice. Kaeberlein relayed a comment from Rich Miller, who runs the National Institute of Aging Interventions Testing Program: the diet Sinclair used was so extreme that the animals carried enough visceral fat to raise the possibility they were, in effect, suffocating.

The Sinclair paper reported results for resveratrol in mice given that extreme obesogenic diet, showing that those very unhealthy mice lived longer when also given resveratrol.

The 2006 Sinclair paper in mice gave resveratrol (two doses) to mice on both the high-calorie obesogenic diet as well as a standard diet. However, that paper reported only the high-calorie arm. The same colony had standard-diet plus resveratrol groups. Why would they only report the results for the high-calorie arm of the study, but not the standard diet? That’s odd.

The data for mice on the obesogenic diet:

That is not a demonstration that resveratrol slows normal aging. What’s strange is that Sinclair’s group did not also do the same experiment in mice on an ordinary diet, under baseline lab conditions. It’s an easy, obvious experiment to do. Just give normal lab mice normal lab food. Give half of them supplemental resveratrol. You would expect to see some longevity benefit in those mice, not just in the animals put on a diet so fattening that they probably had trouble breathing.

But they didn’t report the results for mice on a standard diet in the first study. Why? Because there was no longevity effect! (See below).

The same guy (Sinclair) who wouldn’t let Kaeberlein come to his lab to reproduce a key, high-profile result chose to “replicate” his yeast finding in a bizarrely unhealthy mouse model, without also showing a longevity effect under baseline conditions.

Sinclair’s group only saw a resveratrol longevity boost in the experiment in morbidly obese mice given a highly obesogenic diet, with 60% of calories coming from fat. But if resveratrol doesn’t increase lifespan in normally-aging mice, that’s not a sexy result. It won’t get you a paper in a brand name journal, or generate headlines. What appears to have happened is that Sinclair’s group sat on the data for the negative result, published the 2006 paper, sitting on the negative result until…

A couple years later, Sinclair’s group published work on the longevity effects of resveratrol in mice eating a standard diet, starting in mid-life. That study indeed found a lack of lifespan extension from resveratrol. However, the paper had more to it than that negative result, which is what made it publishable in a high-profile journal: they found improvements in various aging biomarkers, despite lack of lifespan extension.

Stop here and think for a minute: the observation was no lifespan extension despite “better” aging biomarkers. Well, if the aging biomarkers looked better, why didn’t animals live longer? One possibility is that perhaps the experiment didn’t last long enough, or they didn’t give enough resveratrol, or… something. Well, they looked at mice out to 150 weeks, which is nearly three years old—ancient for lab mice. The doses used were similar to those used in the experiments where they reported lifespan extension in the obese mice on ultra-fattening diets.

Another possibility: we don’t actually understand the relationship between “aging biomarkers” and lifespan as well as we might like to think. (Hold that thought).

Between 2-4 years later, work done at the National Institute of Aging looked at the effects of resveratrol and other drugs on genetically heterogeneous mice—not inbred strains, and not morbidly obese mice. Work was done across multiple labs, in both sexes of animals. Sinclair was co-author but the work was done largely by other scientists, not limited to his lab at Harvard (where Dr. Kaeberlein was not permitted to visit to reproduce Sinclair’s seminal yeast findings).

• 2010 study: Two resveratrol doses given from 12 months of age, in males and females. Rapamycin extended lifespan. Resveratrol had no effect.

• 2012 study: Treatment started at four months of age, but resveratrol still had no effect.

This brings us to a general fact haunting much of the preclinical longevity literature: interventions that are observed to have lifespan-extension effects often only do so in specific inbred laboratory strains—often those with the shortest lifespans. In fact, this is true for calorie restriction itself, which is widely touted as the single best general intervention to promote longevity across species.

Share

Calorie Restriction: Strain-specific Longevity Effects

Calorie restriction is widely believed to be one of the best, most reliable methods for increasing lifespan. Why? Because many labs have shown that laboratory animals put on calorie-restricted diets live longer. The result is highly reproducible. Case closed, right?

Not so fast.

What type of animals are most often used in published studies? Inbred strains that are not genetically diverse as natural populations are. The strains most often used for research are chosen precisely because they’ve been bred for docility and fast reproduction times. Animals with those features allow researchers to publish papers faster.

Across many strains of inbred rodents, average lifespan varies 2-3 fold under normal, ad libitum feeding condition (animals allowed to eat at their own leisure). Imagine three populations of humans: on average, one lives to about 20 years of age while the other two live to be 40 or 60 years old. Now imagine we primarily studied aging in the shortest lived population. You see the problem.

But there are many strains of inbred, genetically homogenous mice. Each has been bred for its own experimental purposes. Some are much more commonly used for research than others, but there are many dozens of them available. What happens when you use the same calorie restriction protocol on a whole bunch of them? If calorie restriction is a general, surefire way to extend lifespan, then calorie restriction should extend lifespan, to some degree, in the majority of them.

This study that looked at 41 inbred mouse strains, measuring the effects of 40% calorie restriction across all of them. 40% is major calorie restriction. This level of restriction is known to drive lifespan extension—that is, in particular strains of lab animals. What was found across the 41 mouse strains tested in this study?

Read more

Early access: Available to paid subscribers here on Substack.

Not medical advice.

Nick talks with Dr. Christopher Walsh about what actually differs between a human brain and a chimpanzee or mouse brain, and why that is still hard to name as a phenotype. Walsh, a neurologist and geneticist at Boston Children’s, walks through scale and association cortex, then the genome-first path: human accelerated regions, genes such as ASPM that can shrink a human brain toward chimpanzee volume, and mosaic mutations that come with building a large cortex.

Forthcoming work he describes finds that about a quarter to a third of late-fetal cortical neurons carry genetic defects, most of them gone by birth, with smaller defects still edited out through adult life. The conversation also covers why the developing brain overproduces neurons, how that editing relates to autism and other highly heritable disorders, and why most human-specific change is in gene regulation rather than new proteins.

ABOUT THE GUEST: Christopher Walsh, MD, PhD is Chief of the Division of Genetics at Boston Children’s Hospital, a neurologist, and an investigator of the Howard Hughes Medical Institute. His lab studies how the human brain develops and how it evolved.

RELATED CONTENT:

• Podcast | Comparative Brain Evolution: Mammals, Primates & Humans | Robert Barton | #171

KNOW YOURSELF:

• Visit the SUPPORT PAGE if you find value in this content. Includes discount codes + links to affiliate partners, including products to help you measure and manage your health.

Listen or watch on your favorite platform:

• Audio version: [Apple] [Spotify] [Elsewhere]

• Video version: [YouTube]

PRACTICAL TAKEAWAYS:

• “The brain is done at 25” is an MRI-scale claim. Neurons are mostly in place much earlier. Myelin still comes in through the teens and twenties. Synapses, and some genetically imperfect neurons, keep changing after that.

• A mosaic chromosome finding on IVF or amniocentesis can be limited to the placenta. Walsh’s point is that early embryos are routinely mosaic, and several correction paths exist, including dumping abnormal cells into the placenta.

• Most of the 80–100 new mutations a child carries relative to the parents are inconsequential. Trouble is concentrated in a smaller set of dosage-sensitive genes, which is why most kids still turn out fine.

SUBSCRIBER CONTENT BELOW: Reference paper + episode transcript.

Read more

Early access: Available to paid subscribers here on Substack.

Not medical advice.

Nick speaks with Dr. Nephi Stella about the endocannabinoid system as a distinct lipid signaling network. Stella explains how anandamide and 2-AG are produced by separate enzymes, trafficked by specialized proteins and microvesicles, and act primarily as retrograde inhibitors through CB1 receptors. The conversation covers developmental risks of adolescent THC exposure, the failure of early FAAH inhibitors, promising new therapeutic approaches, and lessons from rimonabant for metabolism-targeted therapies.

TOPICS DISCUSSED:

• Anandamide vs 2-AG: The two major endocannabinoids with a shared arachidonic acid backbone but distinct head groups biological roles.

• On-demand production: Calcium-dependent enzymes release cannabinoids from membranes in response to neuronal activity patterns.

• Trafficking: Fatty acid binding proteins and extracellular microvesicles traffic endocannabinoids within and between cells.

• Adolescent THC: Global receptor activation during circuit wiring may contribute to cannabis use disorder and psychosis vulnerability in susceptible individuals.

• Enzyme inhibitors: Early FAAH clinical trial halted by off-target toxicity; MAGL inhibitors advancing in phase 2 for anxiety and addiction.

• Peripheral CB1: Liver receptors regulate metabolism; restricted antagonists aim to reduce weight without central psychiatric risks of the weight loss drug rimonabant.

ABOUT THE GUEST: Nephi Stella, PhD is a professor at the University of Washington School of Medicine. He has studied endocannabinoid signaling for more than 25 years, focusing on therapeutic strategies that separate beneficial effects from side effects of cannabinoid-system modulation.

RELATED CONTENT:

• Podcast | Plant Chemistry, Cannabinoids, Cannabis Terpenes & the Entourage Effect | Ryan Vandrey

KNOW YOURSELF:

• Visit the SUPPORT PAGE if you find value in this content. Includes discount codes + links to affiliate partners, including products to help you measure and manage your health.

Listen or watch on your favorite platform:

• Audio version: [Apple] [Spotify] [Elsewhere]

• Video version: [YouTube]

PRACTICAL TAKEAWAYS:

• High-potency THC products, especially concentrates that produce rapid high brain levels, raise cannabis use disorder risk more than lower-potency forms; vulnerability is highest during adolescence.

• Endocannabinoid “tone” may influence longer-term neuronal phenotype (including BDNF expression) while phasic release controls rapid transmission; dietary fatty acid changes could slowly alter tone, but this remains unproven.

SUBSCRIBER CONTENT BELOW: Reference paper + episode transcript.

Read more

Summary of main points (available to subscribers):

Heart disease is one of the leading causes of death. There are different points of view on the major dietary and metabolic drivers. Conventional, mainstream thinking holds that saturated fats of all kinds make atherosclerosis more likely, and polyunsaturated fats of all kinds make it less likely—the former “clogs your arteries,” while all polyunsaturated fatty acids (PUFAs) protect you.

That simplistic viewpoint does not discriminate between the many different saturated and unsaturated fatty acids, none of which have identical metabolic effects. The length of fatty acids, as well as the number and placement of double-bonds they contain, play a major role in the effect they have on the body. Biology doesn’t neatly conform to the convenient conceptual boxes people like to draw, as such “saturated fats = all bad” or “unsaturated fats = all good.”

Different saturated fatty acids, such as those with even or odd numbers of carbons, have distinct associations with cardiovascular health. Likewise, polyunsaturated fats come in multiple flavors; two major classes are the Ω-6 and Ω-3 polyunsaturated fatty acids (PUFAs). They’re not merely different, but often have opposing effects. They also compete for metabolic processing by the same enzymes, which is why the relative balance of Ω-6 and Ω-3 PUFAs is important.

If you’re like me, then you recognize that the non-enzymatic oxidation of PUFAs is a critical step in the pathophysiology of atherosclerosis. Peroxidation of PUFAs produces harmful reactive substances which the immune system naturally and automatically recognized as toxic waste it needs to dispose of. Evidence is mounting that the fundamental, rejuvenating purpose of sleep is to protect mitochondrial metabolism in the brain by removing oxidized PUFAs (i.e. rancid fat) from the nervous system.

And that’s just dietary fatty acids we’re talking about. What about the long, long list of other nutrients and bioactive substances that might promote, prevent, or reduce atherosclerotic plaque buildup? Like it or not, biology is complex. This post is meant as a relatively simple resource—a short primer on non-prescription substances, such as simple dietary nutrients, that show evidence of preventing or even reversing atherosclerosis.

No matter what anyone happens to believe about the causes, atherosclerosis is bad and you want as little plaque in your blood vessels as possible. The more plaque that builds up, the more likely you are to experience heart disease.

A common question I’ve gotten from private clients: how do I prevent or reduce arterial plaque buildup without using prescription drugs?

I will briefly describe the evidence behind each of the listed substances listed below, together with dosing information, whole food sources, and supplements. If you think there are major examples missing, please leave a comment linking to evidence. (I will periodically update this page).

We will start with Ω-3 polyunsaturated fats, where the evidence is fairly clear. This will help highlight the importance of oxidative stress, inflammation, and lipid peroxidation in driving disease generally. As you’ll see, many of the substances below work, at least in part, by protecting against the toxic effects of lipid peroxidation. Understanding the fundamental biology here is crucial for understanding metabolic health and disease generally, not just atherosclerosis.

You will have to decide for yourself what you consume and why. If you want my perspective on your personal situation, you can book time with me here.

EPA: Long-chain omega-3 (Ω-3) fat

The balance and specific fatty acid profile of our cells has health consequences ranging from oxidative stress and toxin production to the regulation of inflammation. Humans display genetic diversity in metabolism, with large differences in PUFA (polyunsaturated fatty acid) metabolism— diversity with major consequences for cardiometabolic health. For example, some people produce very little long-chain Ω-3 PUFAs like EPA from dietary plant Ω-3 fats; for those people especially, dietary intake of EPA is crucial.

Most people alive today consume way more Ω-6 PUFA than Ω-3 PUFA. Intake of dietary Ω-6 PUFAs, such as linoleic acid (“seed oil fat”), increases the ratio of long-chain Ω-6 PUFAs to Ω-3s. Here’s the part that many people miss: this does not happen because linoleic acid substantially increases levels of the long-chain Ω-6 arachidonic acid; instead, it lowers levels of the long-chain Ω-3 polyunsaturated fat EPA (“seafood fat”; more detail here).

The greater one’s intake of Ω-6 PUFAs like linoleic acid, the greater one’s need for direct consumption of the Ω-3 EPA, especially for those with certain genetics. There are many reasons why you don’t want to be deficient in long-chain Ω-3s like EPA and DHA; cardiovascular disease is one of them. EPA in particular (as opposed to EPA/DHA together) has been associated with reduced atherosclerotic plaque burden in human imaging studies.

From this review:

In this trial, a highly purified form of EPA was given to 80 patients with existing coronary atherosclerosis. These people were already on statins and in poor metabolic health. They were given 4 grams/day of purified EPA for 18 months, which resulted in significant regression of arterial plaque.

The overall effect of EPA supplementation from that trial:

An example of one patient’s arterial plaque reduction:

My suspicion is that such patients have poor cardiometabolic health to begin with, at least in part, due to consuming a highly skewed Ω-6:Ω-3 diet (in favor of Ω-6s), as most people do. Based on other data, 2-4 grams of EPA per day is often what is recommended for cardiovascular health.

My view is that amount you need is likely proportional to the Ω-6:Ω-3 profile of your diet: the more Ω-6 linoleic acid you consume, the more Ω-3 EPA you will require to offset this. It’s much better to reduce Ω-6 intake substantially, while ensuring adequate and balanced intake of EPA, than it is to consume high levels of both Ω-6 and Ω-3 PUFAs.

In other words, people without arterial plaque and in good overall metabolic health likely don’t need 4 grams/day of EPA, as in the trial above with unhealthy people. My personal view is that you need sufficient intake of polyunsaturated fat for your body’s needs, but no more than this, with an approximately equal proportion of Ω-6s and Ω-3s. Achieving this is difficult in the modern food environment due to the density of Ω-6 PUFAs in processed foods.

One of the biggest problems with nutrition labels today is that they only tell you the total amount of polyunsaturated fat that processed foods contain, without breaking out the Ω-6 and Ω-3 content. This means that, as the consumer, you are forced to do more work in order to understand the actual contents of processed foods.

Putting all of the above together, there are two ways to increase EPA levels and reduce the Ω-6:Ω-3 ratio in your cells:

• Reduce intake of dietary Ω-6 PUFAs like linoleic acid, which reduce EPA levels;

• Ensure adequate intake of the Ω-3 EPA itself.

EPA intake can occur through two routes: consumption of EPA-rich whole foods; supplementation with EPA directly. As a rule, I prefer whole food sources of nutrients to supplements. Fish oil supplements are big business, and you want to be sure fish oil supplements do not contain rancid oils. For this reason, I prefer liquid cod liver oil stored in the refrigerator to fish oil capsules, as far as supplements go.

2-4 grams per day of EPA would be considered a higher dose of EPA. If you’re eating fresh, fatty seafood two or more days per week, the need for supplementation is less. The total amount of EPA and Ω-3s you need is proportional to how much Ω-6 PUFA you consume. Most people are eating 10s or grams of Ω-6 fat per day. For those people, reducing Ω-6 intake will improve the Ω-6:Ω-3 and EPA profile of their cells on its own, due to the suppressive effect Ω-6s have on EPA—a direct, costless way to improve your omega fat profile.

Share

Whole foods with EPA: Fatty fish & certain seafoods

EPA occurs almost exclusively in marine animal foods, with fatty fish like herring, anchovies, mackerel, salmon, and sardines with the most. Other seafoods with good EPA content often include: Pacific oysters, black cod, some trout, and bluefin tuna. Smaller amounts of EPA can be found in crab, mussels, and shrimp. Overall, seafood is king when it comes to whole food sources of EPA.

How Nick handles his EPA and Ω-3 levels

If you’ve already moved to a lower Ω-6 diet (like me), then 2-4 grams of EPA per day may be more than enough. I don’t supplement at all on days I know I’m eating fatty seafood (twice per week minimum), and take only a small teaspoon of cod liver oil otherwise. I’m also a “strong metabolizer” of dietary PUFAs based on my genetics—my body is more efficient than most at converting dietary Ω-3 PUFAs into longer-chain PUFAs like EPA, which means my risk of deficiency is lower.

If I was in poor metabolic health and had arterial plaque buildup, I’d focus on lowering Ω-6 intake while also getting ~4 grams per day of fresh, unoxidized EPA—especially with a fatty liver, and doubly so if I was a weak PUFA metabolizer. Because I’m in good metabolic health and have different PUFA genetics, I have less need for specific supplementation, especially since I have a balanced intake of Ω-6 and Ω-3 PUFAs from food, consuming no more than several grams of PUFAs total each day.

Learn more about specific dietary fats & human variation in PUFA metabolism:

• Podcast | Ancestry & Genetic Diversity in Human Fat Metabolism: Seed Oils & Omega Fats | Floyd Chilton

Read more

Summary of main points:

• GDF15 is a body-to-brain distress signal: peripheral tissues release it under cellular strain → activates GFRAL in the hindbrain → appetite falls.

• Levels stay low in healthy baseline conditions and do not respond to ordinary meals or short fasts; they rise with sustained mitochondrial stress.

• Exogenous GDF15 reliably cuts food intake and body weight in mice and non-human primates mainly by reducing calories eaten, not by raising energy expenditure; human long-acting agonists produce smaller weight loss.

• GDF15 and leptin potentiate each other: combined treatment yields greater fat loss in obese mice even under leptin resistance.

• Dual GLP-1/GDF15 agonists outperform pure GLP-1 drugs in animals because both pathways contribute roughly equally to the weight-loss effect.

• GDF15 reduces liver inflammation and fibrosis independently of food intake or body-weight change.

• GDF15 is strongly induced when mitochondria struggle to move electrons from food to oxygen; circulating levels serve as a sensitive reporter of energy resistance—the same process that rises with aging and metabolic disease.

Growth Differentiation Factor 15 (GDF15) is a protein hormone released under stressful conditions. It acts as a cytokine (a stress-induced signaling protein) involved in communication between cells; while it can influence immune and inflammatory processes, it is produced by many cell types under stress.

GDF15 in a nutshell:

• Acts like a distress signal from the body to the brain.

• GDF15 levels rise in the blood when tissues experience cellular stress, e.g. inflammation, mitochondrial distress, injury, hypoxia, or certain diseases.

• GDF15 mainly signals through the GFRAL receptor (expressed almost exclusively in the hindbrain), reducing appetite.

Similar to endogenous mitochondrial peptides, such as MOTS-c and others, GDF15 is a stress-induced signal. Unlike endogenous mitochondrial peptides, GDF15 is a full protein encoded in the nuclear genome. But similar to stress-induced mitochondrial peptides, GDF15 levels are low under normal, baseline conditions in healthy animals.

GDF15 levels rise when mitochondria are strained, such as when the body faces sustained nutritional imbalance. Unlike everyday satiety hormones, it does not spike after a normal meal or drop sharply during a short fast. Instead, it acts as a longer-term signal of cellular distress. Because GDF15 is released by peripheral tissues and activates the GFRAL receptor in the hindbrain, it functions as a body-to-brain signal of sustained metabolic distress.

A graphical depiction and description of GDF15, straight from the literature:

GDF15: Basic Biology & Core Effects

• GDF15 is a protein encoded in the nuclear genome. The active protein is a dimer.

• Circulating levels increase with age, obesity, smoking, intense exercise, mitochondrial disease, and many chronic conditions.

• In mice and non-human primates, giving GDF15 consistently cuts food intake and reduces body weight.

• The weight loss comes mainly from eating less, not from major increases in energy expenditure.

• In people, a long-acting GDF15 drug lowered appetite scores and produced some weight loss, but the effect size was smaller than in animals and came with dose-related nausea in some participants.

GDF15 differs from classic appetite hormones in important ways. Short-term overfeeding or moderate caloric restriction barely changes levels, while sustained high-fat feeding or diets low in certain amino acids raise it. This pattern suggests GDF15 reports more serious or prolonged metabolic strain rather than ordinary day-to-day energy balance.

Learn more about gut hormones regulating appetite:

• Podcast | Genetics & Neurobiology of Obesity: Hypothalamus, Leptin, Gut Hormones, Weight Loss Drugs | Giles Yeo

GDF15 Interactions: Leptin & GLP-1

Read more

Summary of main points:

• Most retail avocado oil products fail authenticity tests; many are oxidized, cut with cheaper seed oils, or both. Other cooking oils, such as olive oil, may also not be what they claim.

• Health supplements like melatonin products routinely miss label claims: actual levels range widely and may contain serotonin or other contaminants.

• Limited independent testing of online MOTS-c peptides found one sample with zero active peptide, and others with 20–80% of labeled content.

• Legal marijuana datasets reveal clear statistical fingerprints of lab shopping and collusion: sharp discontinuities at psychologically attractive THC thresholds (especially 20%) that reflect sales incentives, not real biology.

• Similar economic pressures operate across food oils, supplements, peptides, and marijuana; consumers must apply skepticism, weigh personal risk and need, and use direct personal experience as a guide; lab certificates are not a guarantee.

You’re stepping toward enlightenment. Learning. Noticing. Like a friend of mine recently declared, you’re careful to read labels at the grocery store. I’m glad you’re paying attention to what the food label says, but things are far worse than you thought. The words printed on the label may be comforting, but they are probably lies.

Below, we will review four examples of consumer product categories where products may not be what they claim: food, health supplements, peptides, and marijuana. It’s not exactly news that food and supplements may not be exactly what they claim. There are early indications that we should keep a close eye on peptides, too. Behind all of these categories an even darker specter lurks in the shadows: third-party lab testing.

Laboratories themselves can engage in fraud, colluding with unscrupulous consumer brands to separate you from your money. Testing labs can generate Certificates of Analysis (CoAs) that do not accurately represent the underlying consumer product, something I witnessed and analyzed in the legal marijuana industry.

In plain English: even when products are “third-party tested” and come with impressive looking Certificates of Analysis (CoA) that appear to show that a product is legitimate. In reality, it’s possible that the CoA itself is fraudulent, or was generated from a cherry-picked or adulterated sample.

In the marijuana industry, producers of consumer products have conspired directly with third-party laboratories to mislead consumers, even in legal, “regulated” markets. We should assume by default that similar relationships exist in other industries. As usual, incentives drive outcomes. Our supposedly high-trust society is filled with fraud—the private sector, the public sector, and academia are rotten with it.

Share

Your Avocado Oil Probably Isn’t Avocado Oil

Everyone is talking about this new study showing that most food products labeled as made with avocado oil are not made with authentic avocado oil. Those products may be spiked with cheaper oils (e.g. soybean oil) or made with rancid avocados.

It’s not so different from a drug dealer diluting street drugs with cutting agents. The motivations are similar. Even if you’re an honest actor, the next guy may not be. Competition is fierce. Margins are thin. Do you have a family to feed? Shareholders to answer to?

When scientists pulled these popular food products off shelves and tested them, they found:

We’ve known about this problem for quite a while. I previously spoke with the lead author of the new study about similar work. Some highlights:

Back then, Dr. Wang would not reveal the identities of the brands they tested. When I asked why, she gave a lame excuse: “It’s not our job to police the industry.”

By definition, “policing” would involve some authority going in and enforcing the rules on those breaking them. Simply revealing the names of the brands selling bunk products is not “policing,” it’s educating. My take on her comments in 2023: Dr. Wang’s university (UC-Davis) told her not to reveal the names of the brands in question. Why? A reasonable bet is that some of them provide funding to the university, at some level.

(Remember: when academic scientists get research grants, the university take a large chunk of it, “to keep the lights on.” In other words, grants are a major revenue stream for universities).

Whatever the case, things appear to be different now. With the new study, UC-Davis released this list of brands claiming to use avocado oil in their products. It tells you which products by which brands were tested, and whether the composition of the advertised oil was consistent with authentic oil.

Brands + products that appeared to consistently contain authentic oils

• Potato chips:

  • Lay’s, Baked BBQ Flavored Potato Crisps Made with Olive Oil

  • Trader Joe’s, Kettle Cooked Olive Oil Potato Chips

  • Torres, Extra Virgin Olive Oil Potato Chips

  • Graza, Extra Virgin Olive Oil Potato Chips

• Dressings or sauces:

  • Bella Sun Luci, Sonoma Vinaigrette Made with 100% Olive Oil

  • Bragg, Organic Vinaigrette Dressing Made with 100% Extra Virgin Olive Oil

  • California Olive Ranch, Balsamic Vinaigrette Dressing Made w/ EVOO

  • Grove, AvoYeah! Mayo and Garlic Aioli Sauce Made with Avocado Oil

  • Graza, “Original” and “Fancy” Mayo 100% EVOO

A bunch of other brands—Siete, Kettle Brand, Sprouts, etc.—all had products where the tested oil was not compositionally consistent with authentic oil.

In our past conversation, Dr. Wang described similar past research, such as a 2010 survey finding that ~68% of imported olive oil samples failed USDA quality standards, and that ~80% of retail avocado oil samples were oxidized or impure.

Oxidized polyunsaturated fats are basically rancid fats, which animal bodies work hard to dispose of. Cooking oils might contain oxidized fats because the starting material was already rancid fat, or because the fats in the oil were not processed or stored properly. So, there are two broad ways that cooking oils can be “bad”:

1. They contain the correct type of oil (e.g. avocado), but that oil became rancid at some point before, during, or after the oil product was manufactured.

2. The advertised oil is actually some other oil, at least partly. For example, avocado or olive oil might be cut with a cheaper substitute, like soybean oil.

End result: what you’re buying isn’t what you think it is. Your avocado oil may have been made from avocados that were already bad, or the oil was handled in a way that allowed it to oxidize before reaching your kitchen. And of course, in some cases the “avocado” oil was simply spiked with cheap seed oils or whatever.

With academic work such as Dr. Wang’s, we at least know that the scientists have rigorous laboratory training, with fewer direct financial incentives tied to industry. There are certainly financial ties between industry and academia, however. For example, see this article for examples of industry-academia conflicts.

The “lab tested” products that consumers purchase are not tested by scientists in academic research labs—those products are tested by for-profit labs whose revenue depends on having a loyal customer base, just as with any other business. Again, competition is fierce. The “other guy” may very well be undercutting you on price by spiking his “avocado oil” with seed oils.

When products are “lab tested,” how do we know that the lab testing is properly carried out, or that brands aren’t colluding with labs to generate the results they want? As ordinary consumers, we are totally blind to this. To know the answer, you’d have to have direct access to testing data from multiple labs, comparing the data between them (ideally with some of the same products tested in different labs).

After briefly surveying results in the supplement and peptide industry, we will turn to another industry—legal cannabis, i.e. marijuana—where I did extensive research that made an ugly truth crystal clear: third-party lab testing, like so many other elements modern society, is contaminated with fraud.

Health Supplements: How Much Melatonin is in Your Melatonin?

Melatonin is a neurohormone and mitochondrial antioxidant famous for regulating circadian rhythms. Melatonin supplements are a multi-billion dollar industry, which is projected to grow. As with other industries, competition for consumer dollars is fierce.

A number of studies over the years have revealed problems with supplemental melatonin products:

• Across thirty melatonin supplements, this study showed that >70% did not contain melatonin levels within 10% of the label claim. Actual melatonin levels ranged from 83-478% of the labelled amount. 26% contained serotonin, an endogenous neurotransmitter similar to, but distinct from, melatonin.

• 88% of melatonin gummy products analyzed in this study were inaccurately labeled. Only 12% containing melatonin quantities within 10% of advertised levels. One product contained no melatonin at all, and CBD (cannabidiol) instead. Among products containing melatonin, the actual levels ranged from 74-347% of the labelled amount.

I’m not singling out melatonin here for any particular reason, other than already knowing about these findings. Problems have been documented for other types of health supplements as well, including fungal contamination of herbal supplements. My baseline assumption is that many supplement products out there are likely to be more variable in potency (dose) than we might like to admit, and some products may be contaminated with other substances.

Health supplements in general are big business. Similar to the food industry, there is lots of competition. Testing in expensive and often a bottleneck. No producer wants to throw a large batch of product away because some lab geek detected a little bit of contamination, or spend more than they have to meet testing requirements.

There’s probably more literature out there, but from what I’ve seen so far, potency variability and contamination are to be expected, to some extent. In many cases, supplements will deviate at least modestly, and sometimes substantially, from label claims. Sometimes that means you’re getting less than what you paid for; other times, you may be getting much more than what you expect. (When it comes to nutrition, more of something is not automatically better).

Another area I’ve gotten more interested in recently is peptide biology. I’ve heard early indications from scientists on the podcast that there’s reason to be cautious about purity and potency here as well…

Learn more about peptides:

• Article | Metabolic Stress & the Peptide MOTS-c: How Much is Too Much?

Lab-Tested Peptides: Do They Contain the Right Peptide?

Amid the surging popularity of peptides, which represent yet another growing, multi-billion dollar industry. GLP-1 drugs are already massively popular. At the same time, more and more people are purchasing “third-party tested” peptides online, which often exist in a legal gray area. One example is the mitochondrial peptide MOTS-c, which has amazing metabolic effects.

Exogenous peptides often exist in a legal gray area, marketed as being for “research purposes only.” That is not what’s happening, of course. People buy dried peptides in solid form, reconstitute the powder in water, and inject themselves subcutaneously. This route of administration increases the importance of purity and potency. Whether legal or not, anyone injecting peptides will want to know their dosage precisely, with high confidence that their stash is free from contaminants.

Anecdotally, there are no shortage of reports about people’s experience with self-administered peptides. Most reports seem positive or benign, but there are certainly examples of responses requiring medical attention, such as allergic reactions. The extent to which those have been driven by immune responses to peptides themselves or contaminants is not very clear to me.

Most peptide products available online proudly display Certificates of Analysis (CoAs) from third-party testing labs, appearing to prove potency and purity. However, when I spoke to prominent peptide researcher Dr. Pinchas Cohen, he shared that several friends had sent him samples of MOTS-c that they purchased for online. His lab tested the samples directly. Here’s what they found:

In at least one sample a mixture of peptides supposedly containing MOTS-c contained none at all, and in other cases the measured level of MOTS-c was 20-80% of what the label said. As he pointed out, these potency issues can arise from multiple potential sources:

• Stability issues: the peptide degrades over time, at some point between its manufacturing, shipment, and use.

• Cheating: producers may be knowingly diluting samples or putting other peptides in place of the one advertised.

• Incompetence: producers may not be properly synthesizing, testing, or storing products to ensure proper purity and stability.

As the rules and regulations around peptides change and popularity continues to grow, we will likely see people place a strong focus on purity, safety, and lab testing. We will also hopefully see more data on true, off-the-shelf product testing, where third-parties purchase peptides online and test them independently, rather than taking product CoAs at face value.

The last point is important, because most people naturally assume that if a product is “third-party tested” and you have access to a CoA, then everything is good. But what you’re tacitly assuming is that the testing was done properly and in good faith, with no funny business such as collusion between the lab and the producer (i.e. the seller and the buyer) to produce favorable results.

The reason I bring this up comes from my personal experience working in the marijuana industry, where I worked extensively with laboratory testing data. I analyzed large data sets from many different testing facilities across several US states, spanning years of testing and thousands of products.

What do you see when you carefully analyze data from many labs testing consumer products for potency and purity? Testing labs are supposed to be the places watching the producers, keeping them honest. But who is watching the watchers, to make sure they’re being honest?

Learn more about Nick’s research on third-party lab testing of marijuana products:

• Paper | The Cannabinoid Content of Legal Cannabis in Washington State Varies Systematically Across Testing Facilities and Popular Consumer Products

• Paper | The phytochemical diversity of commercial Cannabis in the United States

Marijuana & the Dark Truth About Third-Party Testing

Here’s how it works in the marijuana industry. Profit margins for producers are small. Competition in most markets is fierce. Marijuana can be tricky to grow. Mold. Bugs. Lots of things can destroy a crop. It’s farming.

The product being grown, or any of its derivatives, must legally undergo lab testing for potency and contamination. That five-pound lot of dried weed you just spent countless hours (and dollars) laboriously cultivating—well, guess what? One of the buds you paid to have tested had some mold, a little over the legal limit. Congratulations, you have to throw away all it. Good luck making payroll.

Oh, and those dank, juicy nugs of weed harvested from the prized new strain of marijuana your team labored to nurture to maturity? Well, they just tested at 19% THC. What that means is that consumers are going buy from your competition, which tested at 20% THC and is available for the same price.

Where does this lead? Consumers basically optimize for THC per dollar spent, and people are naturally attracted to psychologically satisfying numbers, like multiples of five or ten. Growing weed that tests at 19% THC is like being a guy who is 5’11’’ tall. You tell everyone you’re 6’0’’, especially women.

And when we analyze large-scale testing data, those human tendencies manifest as clear statistical signals:

Read more

Summary of main points:

• Endogenous peptide variants shape metabolic disease, longevity, and neurodegeneration risk in ancestry-specific ways, raising the practical question of which variants people actually receive when using exogenous peptides.

• MOTS-c variant (~7–8% Japanese/East Asian ancestry): reduces insulin-sensitizing and muscle-protective activity → higher type 2 diabetes and sarcopenia risk in sedentary men; exercise-inducible and tissue-specific.

• Humanin variant (rare, enriched in Ashkenazi Jewish centenarians): exceptional longevity + protection against cognitive decline and Alzheimer’s, including in APOE4 carriers.

• SHLP2 variant (~1% of people with European ancestry): ~50% lower Parkinson’s risk; also linked to higher total and LDL cholesterol.

• SHMOOSE variant (~20–25% of people with European ancestry): ~30% higher Alzheimer’s risk; peptide levels rise with age and boost mitochondrial oxygen consumption.

• MENTSH variant (~20% of people with Indigenous American ancestry): abolishes production → higher type 2 diabetes and metabolic disease risk; likely a “thrifty-genotype” holdover from Ice-Age conditions of food scarcity.

• Ancestral environments selected divergent metabolic adaptations; the modern industrial environment therefore mismatches these predispositions in ancestry-specific ways, making personalized approach to metabolic health critical.

Advances in technology, such as the cost genome sequencing and sophistication of analysis, have enabled us to learn more and more about humanity, including human biological diversity and how it relates to health, disease, and longevity. We are learning more than ever about variation in metabolism among present-day humans, such as differences in fat metabolism, as well as how that variation arose, changed, and spread across the globe among our ancestors. Your particular ancestry, and how it differs from someone else’s, has a lot to do with how your body processes certain food components, how long you might live, and the likelihood you will develop a particular autoimmune condition.

All of this is paving the way for an approach to human health that is more personalized and preventative in nature. Over time, we will be able to tailor everything from dietary interventions to disease treatments to individuals, using their unique genetics and life history to recommend treatments and protocols for them, not someone else. And there is already momentum in the culture to focus more on preserving vitality and preventing illness than the historical focus on treating symptoms after health problems arise.

Peptides are a class of molecules rapidly growing in popularity. Researchers are studying endogenous peptides and how they work. At the same time, regular people are increasingly experimenting with exogenous peptides. The growing consumer demand for peptide therapies is driving the desire for regulatory changes to govern how peptides are manufactured and distributed, as well as the need for more research to understand how specific peptides work.

As with pretty much all things in biology, endogenous peptides that our bodies naturally produce display variation. Biologists have already uncovered some remarkable examples of genetic variation in human peptides which seem to underlie everything from sensitivity to metabolic disease to overall longevity.

Below, a brief survey of known examples of genetic variants in endogenous peptides and aspects of human health or disease they’re tied to. This variation may be relevant to the development of peptide therapies that improve human vitality and longevity, help people prevent or limit the severity of metabolic disease, or influence physical and cognitive performance. The existence of natural variants in endogenous human peptides that have health consequences also begs an important question for anyone who might consider using exogenous peptides: for any given peptide one might try, which specific variants are you getting?

Learn more about the endogenous peptides in this article:

• Podcast | Mitochondria Peptides: MOTS-c, Humanin, Genetic Variation in Health & Disease | Pinchas Cohen

MOTS-c Variants: Greater Risk of Type II Diabetes & Sarcopenia

MOTS-c is an endogenous, 16-amino acid peptide. It’s encoded in mitochondrial DNA (mtDNA), not the nuclear genome. It has striking metabolic effects in animals, including the ability to prevent diet-induced obesity and improve insulin sensitivity in mice.

As reviewed in this article, MOTS-c alters the expression of many genes in response to metabolic stressors like exercise, reprogramming cells so they’re better equipped to deal with future stress. It’s no surprise that variants of the MOTS-c gene are consequential for metabolic health, given how striking the effects of MOTS-c are on muscle physiology, mitochondrial production, and insulin sensitivity.

One of the reasons most people still haven’t heard about the MOTS-c peptide is that it was overshadowed by the emergence of GLP-1 drugs like Ozempic. However, interest in the therapeutic potential of MOTS-c may soon be revitalized, possibly as something used in conjunction with GLP-1 drugs for weight loss (more below on that below).

In humans, there is a relatively uncommon variant that is associated with higher Type II diabetes (adult-onset) in men, especially those that are sedentary. That variant is found mostly in East Asian populations, with an estimated ~7-8% of people of Japanese ancestry carrying it.

MOTS-c production is induced by metabolic stressors such as exercise, which means that a sedentary lifestyle will favor less MOTS-c production overall. The variant in question reduces the insulin-sensitizing and muscle-protective activity of MOTS-c, the effects of which are revealed in people who have the variant and do not engage in regular physical exertion, resulting in lower levels of MOTS-c production overall.

The same variant is also linked to skeletal muscle fiber composition, muscular strength, and a higher risk of sarcopenia (muscle loss) in male carriers, depending on age and physical activity levels. Here’s some data on muscle performance, comparing Japanese men with two different MOTS-c genotypes.

Many of the effects of MOTS-c are context-dependent. Because MOTS-c in induced by physical activity, there is a kind of “use it or lose it” aspect to it: if the MOTS-c peptide is not intermittently triggered by the requisite stimulus, tissues can fail to experience the metabolic benefits of MOTS-c. Animal studies have been used to figure out the molecular details for how MOTS-c works in muscle versus other tissues, like fat, showing that it’s effects differ by tissue type.

MOTS-c aids in muscle preservation, illustrated nicely by studies where rodents wore tiny casts. Like humans, wearing a cast for extended periods leads to muscle atrophy. However, when casted rodents are given MOTS-c injections, muscle atrophy is blunted. Dr. Cohen seems keen on the therapeutic potential for MOTS-c as a way to prevent muscle loss. For example, it’s possible that MOTS-c could be given alongside popular peptide-based treatments for weight loss, like GLP-1 drugs, to help people preserve muscle while losing fat at the same time.

Humanin Variants: Longevity & Protection From Cognitive Decline

Humanin is another endogenous mitochondrial peptide, discussed on M&M 304. Despite being rare overall, there is a humanin variant that highly enriched in centenarians of Ashkenazi Jewish ancestry. These people not only display exceptional longevity, but seem to be protected against cognitive decline and Alzheimer’s disease.

The centenarians who possess this humanin variant, who are also APOE4 carriers, still have reduced Alzheimer’s risk. APOE4 is a variant of the apolipoprotein E (APOE) gene that is otherwise associated with increased Alzheimer’s risk, but even APOE4 carriers 100+ in age appear to be free of this neurodegenerative condition when they carry the this rare variant of the humanin peptide.

A short video clip about this, from M&M 304 guest the Dr. Pinchas Cohen:

Read more

Not medical advice.

Nick speaks with Dr. Pinchas Cohen about mitochondrial-encoded microproteins, including humanin and MOTS-c. Cohen discusses how these small peptides regulate metabolism, muscle mass, innate immunity, and neurodegeneration, and how natural genetic variants influence disease risk and longevity across populations.

TOPICS DISCUSSED:

• Mitochondria Beyond ATP: Mitochondria act as signaling hubs that regulate cell fate, innate immunity, and inter-organ communication.

• Hidden Genes in “Junk” DNA: Small genes in mitochondrial and nuclear DNA encode functional peptides overlooked by classical genomics.

• Humanin Discovery & Function: Independently identified by multiple labs; binds several receptors and shows neuroprotective and metabolic effects.

• Centenarian Humanin Variant: A specific peptide variant is highly enriched in Ashkenazi centenarians and appears to neutralize APOE4-related cognitive decline.

• MOTS-c as Exercise Mimetic: Induced by exercise; preclinical data show muscle preservation and metabolic benefits; limited human trial data exist.

• Ethnicity-Specific Variants: Genetic studies reveal “thrifty-gene” variants (e.g. MENTSH) that were adaptive historically but raise diabetes and obesity risk today.

• Peptides vs Small Molecules: Naturally occurring peptides may offer favorable safety profiles within physiological ranges, though most require injection.

• Emerging Regulatory Landscape: FDA compounding recommendations and voluntary registries aim to improve safety data collection for peptides such as MOTS-c.

ABOUT THE GUEST: Pinchas Cohen, MD is Dean of the USC Leonard Davis School of Gerontology. An endocrinologist by training, he leads research on mitochondrial microproteins and their roles in aging, metabolism, and neurodegeneration.on mitochondrial peptides in aging, metabolism, and neurodegeneration.

RELATED CONTENT:

• Article | Metabolic Stress & the Peptide MOTS-c: How Much is Too Much?

• Podcast | MOTS-c & Mitochondrial Peptides in Health & Aging | David Lee

KNOW YOURSELF:

• Visit the SUPPORT PAGE if you find value in this content. Includes discount codes + links to affiliate partners, including products to help you measure and manage your health.

Listen or watch on your favorite platform:

• Audio version: [Apple] [Spotify] [Elsewhere]

• Video version: [YouTube]

PRACTICAL TAKEAWAYS:

• Endogenous MOTS-c rises with exercise; trained individuals show higher circulating levels than sedentary ones.

• Unregulated “research-only” peptides often contain inaccurate doses or zero active compound; purity and sterility risks are real.

• When the science supports it, combining muscle-preserving agents with GLP-1 therapies is a rational direction, though long-term human data remain limited.

• Observational data link “Mediterranean diet” patterns with higher levels of certain mitochondrial peptides; causality is not yet established.

SUBSCRIBER CONTENT BELOW: Reference paper + episode transcript.

Read more

Not medical advice.

Nick speaks with Harvard neuroimmunologist Dr. Isaac Chiu about bidirectional signaling between nociceptors and immune cells. Dr. Chiu describes how pain-sensing neurons directly detect bacterial toxins, release CGRP to suppress neutrophils, and how this crosstalk regulates barrier defenses, allergy priming, and the resolution—or failure—of inflammation.

TOPICS DISCUSSED:

• Nociception & Pathogens: Sensory neurons detect bacterial toxins within seconds, enabling rapid host-defense coordination ahead of immune-cell recruitment.

• CGRP & neutrophils: Nociceptor-release the peptide CGRP, which blocks neutrophil (immune cell) recruitment; pathogens exploit this in necrotizing fasciitis and meningitis.

• Pain sensitization: Inflammation can lower neuronal thresholds for pain, which sometimes becomes a chronic problem.

• Gut barrier protection: CGRP from gut nociceptors drives goblet-cell mucus secretion, supporting the protective “weep and sweep” response.

• Skin-to-gut allergy priming: Early skin exposure to allergens plus scratching may promote food allergies.

• Vagus nerve circuits: Distinct vagal sensory neurons selectively suppress TNF or induce IL-10, allowing precise inflammatory control.

• Brain inflammatory memory: Insular neurons encode site-specific immune responses that can later be recalled by reactivating those circuits. The brain remembers past inflammatory events.

ABOUT THE GUEST: Isaac Chiu, PhD is a professor in the Department of Immunology at Harvard Medical School. His lab studies nervous-immune system crosstalk in pain, itch, infection, and inflammatory disease.

RELATED CONTENT:

Article | Inflammation Regulation by Omega Fats: How High ω-6 Intake Impacts ω-3 Levels

KNOW YOURSELF:

• Visit the SUPPORT PAGE if you find value in this content. Includes discount codes + links to affiliate partners, including products to help you measure and manage your health.

Listen or watch on your favorite platform:

• Audio version: [Apple] [Spotify] [Elsewhere]

• Video version: [YouTube]

PRACTICAL TAKEAWAYS:

• Avoid habitual daily NSAID or anti-histamine use. Chronic use of anti-inflammatory drugs like NSAIDs may actually promote chronic inflammation in the long run.

• Early oral exposure to potential food allergens may promote tolerance, while repeated skin exposure risks sensitization. Be weary of exposing young children to artificial skin care products early in life.

• CGRP-blocking migraine drugs can slow wound healing and alter gut function; effects extend beyond headache relief. Drugs may produce chronic side-effects that are not apparent with intermittent or short-term use.

SUBSCRIBER CONTENT BELOW: Reference paper + episode transcript.

Read more

Summary of main points:

• MOTS-c is a mitochondrial microprotein (peptide) released under metabolic stress that reprograms cells toward energy conservation and expanded ATP-production capacity.

• Rodent work shows clear benefits with short daily or intermittent protocols: protection from diet-induced obesity and insulin resistance, restoration of youthful muscle sensitivity, increased running capacity, and extended healthspan in aged animals (human-equivalent doses in the low-to-mid tens of milligrams).

• Natural, endogenous MOTS-c release is pulsed, not chronic: prolonged elevation risks sustained stress effects (including elevated homocysteine and impaired methylation); too much, too often is expected to turn benefit into risk.

• No human trials exist for exogenous, injected MOTS-c. Acute exercise raises endogenous MOTS-c and levels naturally decline with age.

• Community use typically involves 5–10 mg subcutaneous injections 2–3 times per week in 4–6 week cycles with washouts, often paired with exercise.

• Mechanistic animal data suggests a cautious human approach: at most one injection per day, 2–3 days per week, for no more than a few weeks; prefer starting in the 1–5 mg range rather than 10+ mg; continuous daily dosing for extended periods will likely become counterproductive.

MOTS-c is a 16-amino-acid peptide (“microprotein) encoded in the mitochondrial genome. Mitochondria produce and release MOTS-c in response to metabolic stressors, such as exercise or nutrient deprivation; it then acts both locally and systemically, entering the nucleus of cells to alter broad patterns of gene expression.

The effects of MOTS-c are context-dependent: under baseline conditions, when cells are “happy” and unstressed, it does little; under conditions of energy deficit, such as vigorous exercise, MOTS-c is produced by mitochondria and reprograms metabolism. In general, this peptide lowers protein synthesis to conserve the currency of cellular energy (ATP), while boosting ATP-production capacity through mitochondrial biogenesis.

Protocol thinking is popular in the health space, but biology is history- and state-dependent. The same molecule can drive adaptive change or risk depending on conditions and peptide dosing schedule. The preclinical literature supplies the clearest data here. Although standardized, evidence-based dosing protocols for humans have yet to be formally developed, there is good reason to think exogenous MOTS-c carries real risk when taken too much, too often.

Let’s first review what’s been clearly shown in animal models, using those results to help us think about exogenous MOTS-c use in humans.

Listen to the podcast that inspired this article, with the scientist who discovered the MOTS-c peptide:

• Podcast | MOTS-c & Mitochondrial Peptides in Health & Aging | David Lee

Rodent Protocols and Outcomes

The foundational paper on MOTS-c is described in this article. In short, mice injected with MOTS-c daily for 8 weeks were protected against diet-induced obesity, fatty liver, and hyperinsulinemia despite the same caloric intake as controls.

Peptides like MOTS-c have a short half-life, especially in species with a very fast metabolism like mice. In these experiments, mice were injected once per day, starting at 0.5 mg/kg of body weight. Based on how metabolic rate scales across species, that roughly equates to something in the range of 2.3 to 4.6 mg/day for an adult human.

In the same study, shorter courses with higher dose (5 mg/kg daily for 7 days) improved glucose tolerance and other aspects of metabolism in skeletal muscle, restoring middle-aged muscle insulin sensitivity toward youthful levels. That dose would roughly correspond to ~23 to 45.5 mg/day in a human.

A 2021 study extended this to physical capacity and aging. Young mice given 5 mg/kg daily for 2 weeks ran farther and longer. Middle-aged (12 months) and old (22 months) mice received 15 mg/kg daily for 2 weeks. Old animals roughly doubled treadmill running time and distance, outperforming their middle-aged counterparts. When treatment began even later (~24 months) and switched to an intermittent schedule of 15 mg/kg three times per week, aged mice maintained grip strength, gait, and the ability to complete a walking test that most untreated, age-matched animals failed. Fat mass was lower and lean mass better preserved, with modestly extended healthspan.

So, between those two studies, we’re talking about several days of once-daily MOTS-c injections, or else three injections per week, using human equivalent doses in the range of 1s to 10s of milligrams per day. Those protocols drove beneficial metabolic and physical performance changes in both young and old mice, without obvious problems. When MOTS-c was given to very old mice, they basically performed like much younger mice. Healthspan was extended, which is what everyone is really after—nobody wants to live longer it means more years of decrepitude.

Despite the lack of clear negative effects in animal studies, the underlying biology MOTS-c taps into clearly suggests, in my view, that you would never want to chronically take exogenous MOTS-c for extended periods. You may even want to avoid very high single doses, multiple doses per day, or multiple days in a row.

Let’s take a look at why.

Learn more about how MOTS-c works at the cellular level:

• Article | MOTS-c & Cellular Energy Homeostasis

MOTS-c: Mechanisms of Metabolic Reprogramming

See this article for a more detailed explanation. MOTS-c is doing a couple of major things at the molecular level:

• Raising AMPK and suppressing mTOR, shifting cells from anabolic “build” toward catabolic “burn” programs.

• Inhibiting key steps of the methionine-folate cycle. This helps drive AMPK activation and changes the cells’ capacity to methylate DNA, important for the gene expression changes driven by MOTS-c.

By stimulating AMPK and driving down mTOR activity, MOTS-c helps preserve ATP by decreasing translation (protein synthesis), which is energetically expensive. At the same time, mitochondrial biogenesis is stimulated. The basic idea is that, in a state of metabolic stress or energy deficit, the cell doesn’t want to “spend” ATP to build up new structures requiring protein synthesis; instead, it wants to preserve ATP and specifically build up production capacity by generating more mitochondria.

That’s all great, but it’s meant to be a temporary metabolic response to stress, not a persistent change. When MOTS-c levels stay elevated for too long, the methionine-folate cycle is impacted for longer than it “should.” If the cell is in a state of metabolic stress like this for too long, homocysteine starts to rise. Homocysteine is usually kept at low levels, as it’s quickly re-methylated back to methionine. As a biomarker, you basically want homocysteine to be as low as possible. If homocysteine accumulates and does not revert to methionine, your cells will be impaired in their ability to methylate other things, such as DNA. Compromised methylation capacity leads to all sorts of problems.

Elevated homocysteine is a risk factor for:

• Cardiovascular disease (e.g. atherosclerosis)

• Cerebrovascular events (e.g. stroke).

• Broader effects such as increased oxidative stress and inflammation.

Think of it this way: some stress is good, but too much stress is bad. Exercise is a form of metabolic stress. Obviously, exercising is generally good for you, but only up to a point. If you train too hard, too often, it can cause problems. At the macroscopic level, this might mean injury, slower recovery, etc. Under the hood, it means excess inflammation, oxidative stress, and so forth. Elevated homocysteine is a major reason why.

You don’t want MOTS-c to be arbitrarily elevated, and definitely not persistently. Too much is expected to cause problems, especially over time, even though the rodent studies above seemed to find mostly metabolic benefits from once-per-day injections. Keep in mind that rodents have a much higher metabolic rate than humans, so they will metabolize peptides like MOTS-c even faster than you or I would. Once-per-day injections might be fine for a rodent, but not for a human. You always have to be very cautious about applying animal results to the human case. Dosing and timing protocols will never translate 1-to-1 across species, even if the overall effects are directionally similar.

Again, the human equivalent doses we’re taking about here would be roughly 1s to 10s of milligrams per injection, given once in a single day. With that in mind, let’s review what people today are reporting when they self-administer MOTS-c.

Endogenous MOTS-c: Human Exercise

I’m not aware of any published randomized trials of MOTS-c in humans. Observational data show that acute exercise raises endogenous MOTS-c in skeletal muscle and circulation in young men. Endogenous MOTS-c levels decline naturally with age.

A small human study indicates that endurance and resistance exercise may stimulate endogenous MOTS-c to different degrees. That work found that endurance exercise stimulated higher increases in plasma MOTS-c compared to resistance exercise. (A similar pattern was seen for the humanin, another mitochondrial peptide).

What about exogenous MOTS-c? How much are people injecting, and how often?

Share

Exogenous MOTS-c: Human Dosing Schedules

Community and clinic reports commonly describe 5–10 mg subcutaneous injections two or three times per week, often in 4–6 week cycles with washouts. Common reports seem to be faster recovery and favorable shifts in body composition (fat loss with muscle preservation) when the peptide is paired with training. These patterns roughly approximate the intermittent rodent schedules rather than daily injections for multiple weeks.

Given the preclinical data and the mechanism by which MOTS-c acts, there is very likely risk from taking too much exogenous MOTS-c, too often. Without well-controlled human data it’s hard to make specific claims here, but it seems prudent to me to avoid injecting MOTS-c more than once per day and perhaps 2-3 times per week. You certainly wouldn’t want to be taking large doses, every single day, for extended periods.

What is a “large dose” of MOTS-c for a human?

Read more

Not medical advice.

Nick speaks with Dr. David Lee about the mitochondrial-encoded peptide MOTS-c. Lee describes its discovery within the mitochondrial genome, its movement into the nucleus under metabolic stress to broadly reshape gene expression, and preclinical evidence that it blocks diet-induced obesity, improves treadmill performance in older mice, and modulates autoimmune responses.

TOPICS DISCUSSED:

• Mitochondrial Roles: Not just ATP production, but “speaker of the house” via communication molecules that “talk” to the rest of the cell.

• Origin of Communication: Mitochondrial peptides were likely co-opted from ancient antimicrobial factors to synchronize nuclear and mitochondrial genomes.

• MOTC Discovery: Identified in mitochondrial ribosomal RNA regions linked to interferon responses.

• Stress Induction: MOTS-c is produced in response to exercise, immune activation, and nutrient withdrawal; circulates as a mitokine.

• Nuclear Regulation: MOTS-c translocates to the nucleus under stress and acts as a nonspecific cofactor that shapes chromatin accessibility and gene expression.

• Metabolic Effects: Prevents diet-induced obesity and fatty liver in mice without altering baseline weight; engages AMPK and suppresses anabolic pathways.

• Immune Modulation: Shifts immune cell programs to reduce autoimmune attack in type 1 diabetes models.

• Aging Phenotypes: Daily or intermittent dosing doubles running capacity and preserves walking ability in old mice.

ABOUT THE GUEST: Changhan David Lee, PhD is Associate Professor at the University of Southern California Leonard Davis School of Gerontology. His laboratory investigates mitochondrial communication and microproteins in the biology of aging.

RELATED CONTENT:

• Article | MOTS-c & Cellular Energy Homeostasis

KNOW YOURSELF:

• Visit the SUPPORT PAGE if you find value in this content. Includes discount codes + links to affiliate partners, including products to help you measure and manage your health.

Listen or watch on your favorite platform:

• Audio version: [Apple] [Spotify] [Elsewhere]

• Video version: [YouTube]

PRACTICAL TAKEAWAYS:

• Intermittent rather than continuous metabolic stress aligns with the MOTS-c peptide’s natural production rhythm.

• Exogenous MOTS-c use should respect short half-life and cycling patterns observed in animal studies rather than continuous elevation.

SUBSCRIBER CONTENT BELOW: Reference paper + episode transcript.

Read more

Summary of main points:

• Copper is a critical component of the mitochondrial electron transport change, affecting production of ATP and Reactive Oxygen Species (ROS);

• Dietary and topical fats reshape skin fatty acid profiles that determine the intensity of lipid peroxidation and inflammation in the skin;

• GHK is an endogenous peptides that, like other natural peptides, declines with age.

• GHK-Cu directly scavenges toxic lipid peroxidation byproducts like 4-HNE, suppresses inflammatory cytokines, and drives gene expression changes that promote collagen synthesis, angiogenesis, and wound repair;

• Dimeric GHK-copper peptide delivered via ROS-responsive hydrogel achieves 97% closure of infected diabetic wounds by synergistically scavenging ROS and boosting antioxidant enzymes.

• Copper’s essential role in mitochondrial metabolism across all tissues indicates GHK-Cu and related peptides limit oxidative stress and inflammation and tehreby drive benefits in tissues beyond skin, including the brain.

The focus of M&M 301 is on peptides like GHK-Cu (“GHK copper”) in skin health. The effects of GHK-Cu on skin biology helps illustrate important themes emphasized in my content:

• Nutrient balance, the relative ratio of nutrients or other substances, has a major impact on cell function and tissue homeostasis. Examples include how the relative balance of Ω-6 and Ω-3 polyunsaturated fats influences patterns of tissue inflammation and susceptibility to lipid peroxidation, or the balance of copper and other ions in mitochondrial energy metabolism. Over time, intake balance drives changes in…

• Tissue composition, a key aspect of biological context. The composition of cells shapes how phenomena like inflammation play out. The same stimulus can trigger a different pattern of tissue inflammation depending on the specific composition of our tissues, such as the fatty acid profile of cell and mitochondrial membranes. Tissue composition shapes…

• Susceptibility to oxidative damage. Depending on biological context, our tissues can be rendered more or less susceptible to oxidative stress. Imbalances in antioxidant capacity can lead to enhanced lipid peroxidation, generating cell toxins that damage or kill cells.

Patterns of energy resistance within our cells determine how fuel is burned by mitochondria to generate cellular energy (ATP), how much oxidative stress is produced, and whether protective mechanisms are able to function properly to limit damage. Patterns of energy flow are constrained by:

• Present context, such as the specific composition of our cells and mitochondria;

• Personal history, such as patterns of nutrient intake and oxidative stress our tissues have experienced over time, which shapes tissue integrity and function;

• Intrinsic factors differ between individuals, such as genetics and mitochondrial haplotype, constrain how energy metabolism functions in any given context, with a given history.

Two people with a similar history, living in a similar context, may nonetheless display large differences in energy metabolism. Likewise, two people with the same “genetic type” of mitochondria can can display very different metabolic health profiles driven by differences in present tissue composition or personal history.

The data below illustrates just how much mitochondria function can vary between two individuals depending on present context and personal history. These two people are from the same maternal lineage and therefore belong to the same mitochondrial haplogroup (“genetic type,” H7 in this case). Nonetheless, they display wildly different functional profiles:

How does copper fit into the picture with respect to mitochondria?

Copper is a mineral that plays a critical role in mitochondrial metabolism and antioxidant systems. For example, copper is indispensable for the assembly and function of cytochrome oxidase C (“Complex IV” in the data above). Copper is also a component of peptides that have become increasingly popular for skin care, such as GHK-Cu.

Below, we will review research on GHK-Cu and copper peptides in skin health. As we will see, copper peptides have effects that depend on their ability to influence inflammation and oxidative stress in skin, likely through copper’s role in energy metabolism. Throughout the article, keep in mind the key themes of balance, context (tissue composition), and oxidative stress.

Learn more about skin, oxidative stress & metabolic health:

• M&M 104 | Benefits & Risks of UV Radiation & Sunlight, Skin Health, Vitamin D, Nitric Oxide, Evolution of Skin Color | Richard Weller

• Article | Dietary Fat, Sunscreen & Skin Photosensitivity: Can Seed Oils Promote Sunburn?

Skin Biology Basics

One of the primary functions of skin is to serve as a protective barrier, helping to preserve internal homeostasis (balance) by preventing harmful substances from entering the body. Skin rapidly regenerates, which is crucial for its ability to maintain a protective barrier. Immune cells patrol our skin, which is subjected to frequent inflammatory triggers and oxidative stress.

As described in more detail here, the fat content of your diet influences the fatty acid profile of skin cells and therefore your susceptibility to oxidative stress. One major exogenous source of oxidative stress in skin is exposure to natural or artificial light (particularly UV). Longer wavelength of light, like red and near-infrared (NIR), penetrate deeply through our skin, which is relevant to how dietary copper and copper peptides influence energy metabolism (more on that, later).

Let’s briefly look at how tissue composition is influenced by diet, shaping how tissues like skin respond to oxidative stress and inflammatory triggers. After that, we will review how copper fits into the picture before looking at GHK-Cu and copper peptides directly.

Subscribe now

Tissue Composition, Oxidative Stress & Inflammation

Read this article for a deeper look at how the fatty acid profile of cells affects how inflammation plays out in tissues. In short, the relative balance of fatty acids, especially Ω-6 and Ω-3 polyunsaturated fats, shapes the magnitude and duration of inflammatory responses.

The balance of Ω-6 and Ω-3 polyunsaturated fats, influenced by dietary intake patterns, sets the tissue context within which inflammation play out. Genetic differences in fatty acid metabolism also play an important role here. In other words, the present tissue context (tissue composition) of your skin is shaped by personal dietary history (fat intake). Along with genetic factors, this shapes patterns of inflammation and oxidative stress in your skin.

You can change the tissue context of your skin by changing what fats you consume. You can also change tissue context by putting fats onto the surface of your skin. The same external stimulus, like UV light, will drive different levels of inflammation and oxidative stress depending on tissue context. For example, topical cholesterol tends to protect barrier-disrupted skin from UV damage, while linoleic acid (Ω-6 polyunsaturated fat) aggravates UV damage.

Where does copper fit in? Copper is crucial for the function of light-sensitive enzyme complexes in mitochondria involved in the production of ATP and Reactive Oxygen Species (ROS). Copper is also critical for antioxidant enzymes that determine how much oxidative stress our cells can tolerate before lipid peroxidation and cell death are triggered.

Share

Copper Deficiency & Homeostasis

Copper is a cofactor for key enzymes involved in antioxidant defense, connective-tissue formation, melanin synthesis, energy metabolism, and iron handling. In skin, copper supports keratinization, normal pigmentation, and extracellular matrix stability. It’s role as a cofactor in major enzymes is not limited to skin. Copper is crucial for mitochondrial metabolism generally, in all tissues.

Read this article first if you want a more basic understanding of the electron transport chain and mitochondrial metabolism. Read this article to more specifically learn about copper function and nutrient deficiency.

Note that copper is part of cytochrome c oxidase, which is “Complex IV” of the electron transport chain, where it’s critical for:

• Efficient mitochondrial ATP production

• Overall redox balance and prevention of oxidative damage

Complex IV within mitochondria absorbs red and near-infra-red (NIR) light, which relates to why photobiomodulation may pair well with use of copper peptides. When mitochondria are exposed to red or NIR light, photons are absorbed by Complex IV. Complex IV activity can be affected by red and NIR stimulation, resulting in:

• Dissociation of inhibitory nitric oxide (NO) that binds and slows Complex IV;

• This relieves NO inhibition, allowing faster electron transfer from cytochrome c, improved O₂ binding/reduction to water, higher oxygen consumption, elevated mitochondrial membrane potential, and increased ATP production;

• Mild, transient ROS production, which can activate pathways for antioxidant defense, mitochondrial biogenesis, and reduced inflammation.

Bottom line: Copper plays a crucial rule in energy metabolism generally. Many of the things we will review below, in the context of skin health, likely apply to other tissues as well. The role of light is also important for tissues beyond skin, as NIR light can penetrate deeply into the body.

Learn more about copper, energy metabolism, and sunlight:

• Article | Nutrition Balance: How Restoring Copper Reversed High Cholesterol Without Pills

• M&M 242 | Sunlight: Cosmology, Energy Metabolism & the Physics of Life | Robert Fosbury

GHK-Cu & Copper Peptides

GHK-Cu is the subject of M&M 301, featuring Dr. Rahi Sarbazhia, who co-authored this introductory paper on copper peptides.

GHK is a natural, endogenous peptide. GHK levels peak in early adulthood and progressively decline with age, not unlike other endogenous peptides like MOTS-c. A 60-year-old will have roughly 40% of the GHK levels of a 20-year-old. This copper peptide has been used in aesthetic medicine for decades and seems to modulate hair growth, wound healing, oxidative stress and inflammation.

Above, we saw how copper is a critical component of Complex IV of the mitochondrial electron transport chain, a light-sensitive enzyme complex that shapes production of ATP and Reactive Oxygen Species (ROS). Copper is also critical for antioxidant enzymes, so it should come as no surprise that copper peptides like GHK-Cu affect oxidative stress and inflammation.

A brief summary of what’s known:

Recall that lipid peroxidation occurs when excess oxidative stress triggers the unregulated oxidation of polyunsaturated fats, generating toxins like 4-HNE. Evidence indicates that peptides like GHK are basically protecting the cell from these toxins.

Read more

Not medical advice.

Nick speaks with Dr. Rahi Sarbaziha about GHK copper, a naturally occurring tripeptide that binds copper and participates in skin wound repair and collagen regeneration. Sarbaziha notes that levels peak in young adulthood before declining, and she outlines how the compound modulates gene expression tied to repair processes in addition to exerting acute effects on oxidative stress. Central concepts include the molecular properties required for topical penetration, distinctions between immediate antioxidant actions and longer-term remodeling, and trade-offs between topical products and injectable administration.

TOPICS DISCUSSED:

• Endogenous Decline: GHK-copper levels peak in young adulthood then fall with age, reducing natural support for skin repair and collagen maintenance.

• Skin Penetration: Effective topical delivery depends on molecular size under and fat solubility, often improved by procedures such as microneedling that briefly disrupt the barrier.

• Mechanisms of Action: The peptide upregulates fibroblasts and vascular endothelial growth factor, stimulates glycosaminoglycan production, and alters expression of genes involved in tissue regeneration and oxidative stress reduction.

• Lasting Remodeling: Gene expression changes produce sustained collagen rebuilding that develops over weeks to months rather than delivering only transient effects.

• Administration Routes: Topical use is accessible and pairs well with hyaluronic acid for hydration; injectables may reach higher local concentrations but lack extensive human trial data and FDA approval for this purpose.

• Safety & Protocols: The compound is generally well tolerated with mild transient injection-site reactions; typical approaches use daily microgram doses for one to two months followed by off periods.

• Hair & Other Uses: Clinical observations indicate potential for hair shaft strengthening comparable to minoxidil with a favorable side-effect profile, alongside exploratory work in wound healing and neurological contexts.

• Research Needs: Expanded human studies are required to standardize dosing, confirm long-term safety, and evaluate value in neurodegeneration, lung repair, and related areas.

ABOUT THE GUEST: Rahi Sarbaziha, MD is a physician in Beverly Hills, California whose practice focuses on regenerative aesthetics, hormone therapies, peptides, and interventions that support vitality, tissue repair, and overall wellness.

RELATED CONTENT:

• M&M 300 | Peptides & Blood-Brain Barrier in Health & Disease | William Banks

KNOW YOURSELF:

• Visit the SUPPORT PAGE if you find value in this content. Includes discount codes + links to affiliate partners, including products to help you measure and manage your health.

Listen or watch on your favorite platform:

• Audio version: [Apple] [Spotify] [Elsewhere]

• Video version: [YouTube]

PRACTICAL TAKEAWAYS:

• Select topical GHK copper serums formulated as liposomal or combined with hyaluronic acid to aid absorption and support skin hydration.

• Use products for at least four weeks before evaluating changes in texture or fine lines, since collagen remodeling and gene-driven effects require time to appear.

• Pair topical application with professional microneedling when feasible to increase delivery into deeper dermal layers where repair activity occurs.

• Approach injectable forms cautiously due to absent dosing standardization and regulatory approval; cycle usage under clinician supervision if this route is considered.

SUBSCRIBER CONTENT BELOW: Reference paper + episode transcript.

Read more

Not medical advice.

Nick speaks with Dr. William Banks about the blood-brain barrier. Banks, who has spent over 45 years studying it, shows that the barrier is a sophisticated, regulated interface that permits bidirectional communication between body and brain. Key concepts include how small lipid-soluble molecules and specific peptide transporters enable entry of signaling molecules, the role of efflux systems, and how barrier changes contribute to aging and neurodegenerative disease. They also discuss practical implications for peptide-based therapeutics and drug delivery to the central nervous system.

TOPICS DISCUSSED:

• Blood-Brain Barrier Structure: Capillary endothelial cells with tight junctions, absent fenestrae, and minimal vesicles create a selective interface; supported by astrocytes and pericytes.

• Crossing Mechanisms: Lipid-soluble small molecules diffuse passively; glucose, amino acids, and regulatory peptides use saturable transporters whose rates vary by need.

• Peptide Transport: Many peptides cross via lipid solubility or specific transporters; rate, not all-or-none presence, determines biological relevance; efflux systems remove some (e.g. certain enkephalins).

• Dynamic Regulation: Transporters for leptin and insulin are modulated by epinephrine, triglycerides, and brain demand; changes occur in aging and Alzheimer’s (reduced glucose uptake, impaired Aβ efflux).

• Circumventricular Organs: Leaky capillary regions allow rapid sampling of blood signals while tanycytes limit spread to surrounding brain tissue.

• Drug Delivery Challenges: Small lipid-soluble drugs, barrier disruption (e.g. focused ultrasound), and “Trojan-horse” strategies each have limitations; intranasal delivery shows rapid brain distribution.

• Peptides & Brain Effects: Gut peptides like GLP-1 analogs, insulin, and ghrelin influence cognition and memory; brain entry predicts central effects.

ABOUT THE GUEST: William Banks, MD is a physician-scientist and leading researcher on the blood-brain barrier. He has focused on humoral brain-body communication, peptide transport, and the barrier’s role in health and disease.

RELATED CONTENT:

• M&M 298 | GLP-1s & Novel Peptides in Obesity, Diabetes & Metabolic Health | Katrin Svensson

KNOW YOURSELF:

• Visit the SUPPORT PAGE if you find value in this content. Includes discount codes + links to affiliate partners, including products to help you measure and manage your health.

Listen or watch on your favorite platform:

• Audio version: [Apple] [Spotify] [Elsewhere]

• Video version: [YouTube]

PRACTICAL TAKEAWAYS:

• The brain receives many gut and fat-derived signals (leptin, GLP-1) that affect appetite, metabolism, and cognition—supporting metabolic health benefits brain function.

• Most small lipid-soluble molecules reach the brain; highly polar or large peptides usually require specific design or delivery methods.

• Aging and inflammation subtly alter barrier function; prioritize sleep, exercise, and metabolic health to support vascular integrity.

• When evaluating peptide supplements or drugs, consider whether they are formulated to cross or act at the barrier; brain effects depend on transport.

SUBSCRIBER CONTENT BELOW: Reference paper + episode transcript.

Read more

Nick speaks with Dr. Pamela Maher, research professor at the Salk Institute, about ferroptosis and its contributions to neurodegenerative disease. Maher’s central argument is that chronic exposure to elevated iron or partial glutathione depletion, unlike acute insults, lowers glutathione peroxidase-4 levels, permits gradual lipid peroxide buildup, and leaves cells more vulnerable to secondary stressors. Key concepts include the Fenton reaction that links labile iron to membrane damage, distinctions among programmed cell death pathways, and evidence that markers of lipid peroxidation appear early in affected brain regions in Alzheimer’s and Parkinson’s disease.

TOPICS DISCUSSED:

• Cell Death Pathways: Apoptosis proceeds through regulated steps with surface “eat me” signals that enable non-inflammatory clearance by phagocytes, whereas ferroptosis and related pathways release cellular contents that can amplify local damage and inflammation.

• Ferroptosis Discovery: Work on glutamate toxicity in neuronal cell lines first identified a glutathione-depletion pathway termed oxytosis; parallel studies in cancer cells renamed and generalized it as ferroptosis to highlight iron’s catalytic role.

• Lipid Peroxidation: Polyunsaturated fatty acids in membranes are oxidized non-enzymatically when ferrous iron reacts with hydrogen peroxide in the Fenton reaction, generating hydroxyl radicals that produce toxic lipid peroxides and reactive carbonyls that modify proteins.

• GPX4 & Glutathione: Glutathione peroxidase-4 is an enzyme that reduces lipid peroxides within membranes and requires glutathione as cofactor; depletion of either component removes this critical line of defense.

• Chronoferoptosis: Nine-day treatment of differentiated neuronal cells with chronic iron or glutathione synthesis inhibitor reduced GPX4 protein, raised basal lipid peroxidation, and increased cell death after addition of otherwise sublethal secondary stressors.

• Protective Compounds: Screening identified flavonoids such as fisetin and sterubin plus synthetic molecules J-147 and CMS-121 that block lipid peroxidation directly or indirectly and, in several cases, also raise glutathione levels.

• Human Disease Evidence: Postmortem brain tissue from Alzheimer’s and Parkinson’s patients shows elevated lipid peroxidation markers and altered iron handling in vulnerable regions; similar changes appear in animal models of multiple diseases.

• Dietary & NRF2 Links: Compounds that activate the NRF2 transcription factor, including sulforaphane from broccoli and certain flavonoids, upregulate antioxidant enzymes including GPX4 and may support long-term cellular resilience.

ABOUT THE GUEST: Pamela Maher, PhD is a research professor at the Salk Institute for Biological Studies in La Jolla, California. Her laboratory studies ferroptosis mechanisms in neurodegenerative diseases and works to identify compounds that inhibit iron-dependent lipid peroxidation.

RELATED CONTENT:

• Article | Burning Down the House: Lipid Peroxidation, Oxidative Stress & Cell Death

KNOW YOURSELF:

• Visit the SUPPORT PAGE if you find value in this content. Includes discount codes + links to affiliate partners, including products to help you measure and manage your health.

Listen or watch on your favorite platform:

• Audio version: [Apple] [Spotify] [Elsewhere]

• Video version: [YouTube]

PRACTICAL TAKEAWAYS:

• Consume a variety of foods daily, such as broccoli and other cruciferous vegetables for sulforaphane and strawberries for fisetin, to activate NRF2-driven production of antioxidant enzymes that help neutralize lipid peroxides.

• Support glutathione status through consistent sleep, regular physical activity, and avoidance of unnecessary oxidative loads, because glutathione is required for GPX4 to detoxify membrane lipid peroxides.

• Recognize that low-grade chronic stressors can erode cellular defenses over days to years without immediate cell death, making sustained dietary patterns that supply protective phytochemicals relevant for long-term neuronal resilience.

SUBSCRIBER CONTENT BELOW: Reference paper + episode transcript.

Read more

Nick speaks with Dr. Katrin Svensson, a Stanford physiologist, about peptide hormone biology and its applications to energy regulation and weight loss. Svensson explains the precursor processing that allows one gene to produce different peptides in different tissues and how this context specificity supports precise physiological responses. The discussion covers why engineered GLP-1 drugs achieve robust brain effects on satiety and food preference that endogenous GLP-1 does not, along with her team’s genomic approach to discovering additional peptides that act on distinct brain pathways.

TOPICS DISCUSSED:

• Peptide Structure & Production: Short amino acid chains (3-50 residues) cleaved from larger precursors by PCSK enzymes, stored in dense-core vesicles, and released upon stimuli rather than acting as folded enzymes like proteins.

• Tissue-Specific Processing: The same precursor yields different peptides by tissue; preproglucagon produces glucagon in pancreas but GLP-1 in gut L-cells due to distinct cleavage patterns.

• Signaling Timescales: Peptides act over minutes via GPCRs and can drive longer transcriptional changes such as FOS expression, yet are cleared rapidly by proteases and kidneys.

• Endogenous GLP-1: Meal-stimulated incretin from gut that boosts insulin secretion with a half-life of only minutes, limiting primary action to periphery rather than deep brain structures.

• GLP-1 Engineering for Brain Access: DPP4 site mutations plus lipid conjugation enable albumin binding, extend half-life, and raise brain exposure.

• Dose-Dependent Effects: Low GLP-1 doses mainly lower blood glucose via pancreas; higher doses engage hindbrain and hypothalamic neurons to suppress appetite and alter food preference.

• Novel Peptide Discovery: Computational scans across the genome flag hundreds of uncharacterized precursors, facilitating new peptide discovery.

• BRP Peptide: This newly discovered brain-derived peptide suppresses feeding in GLP-1 receptor knockout mice, activates non-overlapping neurons, and does not slow gastric emptying.

ABOUT THE GUEST: Katrin Svensson, PhD, is a biochemist and physiologist who leads a laboratory at Stanford University focused on peptide hormone physiology and the regulation of energy balance.

RELATED CONTENT:

• M&M 161: GLP-1, Weight Loss Drugs, Ozempic, Obesity, NMDA Receptors, Metabolism & Brain Health | Christoffer Clemmensen

KNOW YOURSELF:

• Visit the SUPPORT PAGE if you find value in this content. Includes discount codes + links to affiliate partners, including products to help you measure and manage your health.

Listen or watch on your favorite platform:

• Audio version: [Apple] [Spotify] [Elsewhere]

• Video version: [YouTube]

PRACTICAL TAKEAWAYS:

• GLP-1 drugs require higher doses to engage brain satiety circuits for meaningful weight loss; lower doses primarily improve glucose control with little effect on appetite.

• Response varies widely across individuals, with some losing over 25% body weight while others see minimal change.

• Emerging peptides under study may suppress appetite via brain pathways that avoid the gastric slowing responsible for common GI side effects of current GLP-1 drugs.

• Research peptides purchased online frequently show variable purity below pharmaceutical standards and may contain unsafe formulation components or aggregates.

SUBSCRIBER CONTENT BELOW: Reference paper + episode transcript.

Read more

Summary of main points:

• Rigid health protocols often fail because biology is context- and history-dependent. Protocols and strategies—for diet, peptides, or anything else—will rarely be static or universal. They will need to be personalized and dynamic.

• Three core pathways control cellular resource allocation: PI3K routes glucose toward biomass and growth; AMPK senses low energy and promotes ATP production capacity; mTOR commits to protein synthesis and growth when nutrients and oxidative capacity allow.

• AMPK and mTOR are antagonistic—high AMPK generally inhibits mTOR—so their balance (with PI3K) determines whether tissues operate in coordinated “burn mode,” coordinated “growth mode,” or dysregulated states that promote disease.

• MOTS-c inhibits the folate-methionine cycle, raising AICAR levels that activate AMPK; it also translocates to the nucleus to reprogram gene expression for metabolic adaptation.

• Exercise strongly induces MOTS-c production in muscle, amplifying AMPK-driven burn mode; fasting activates AMPK via direct energy deficit but does not increase MOTS-c, highlighting distinct upstream triggers for the same pathway.

• In a key study, exogenous MOTS-c improved whole-body insulin sensitivity, restored youthful sensitivity in older animals, prevented high-fat-diet obesity, reduced fatty liver, and increased energy expenditure despite unchanged food intake and activity.

• Human data show acute exercise raises muscle MOTS-c levels in some groups; lower levels correlate with insulin resistance markers in certain populations; large intervention trials are still lacking.

Biology is complex. It’s easy to get lost in the details. The known details are staggering, and the unknown details outnumber them, constantly haunting any confidence we may build up as we learn. One response is to let your eyes glaze over—trying to memorize everything is futile, so just give up and defer to whomever you deem an “expert.” If that’s you, no judgement from me, but you’ve come to the wrong place. Stop reading and unsubscribe.

Another path: embrace the complexity; admit that we’ll never understand everything. Make it a quest to distill the complexity into general principles and basic frameworks enabling you to understand the general tendencies of biological systems without getting lost in minutiae. We don’t want to lose sight of the forest for the trees. To be sure, we have to study some trees to understand anything about the forest, but always keep in mind that for any patch of woods we’re walking through, we want to understand how that patch fits into the bigger picture.

A popular frame of mind in the “health space” these days is protocols.

Thinking in terms of protocols has its uses, as well as downsides. Protocol thinking assumes our knowledge is comprehensive enough to prescribe reliable solutions. “These are the steps you can take to yield this outcome, period.” That might be true sometimes, but many of the protocols commonly circulated online are constructed according to assumptions that are rarely examined.

In my view, protocol thinking tends to implicitly assume two things with respect to any given area of biology:

• Our knowledge in that domain is greater than our ignorance;

• The protocol is context invariant—if you are experiencing X, Y, and Z, the prescribed protocol will reliably generate the desired outcome for anyone.

It doesn’t matter so much who you are as compared to me—genetic differences, tissue composition, and so on—because we’re all pretty much the same, with differences only skin deep. If we share the same symptom set of X, Y, and Z, the protocol that worked for you will work for me, too. This way of thinking is rooted in the universalist ideology pervasive in the Western world today.

The first assumption is born of hubris: we pretty much know how things work Knowledge > ignorance. With a little self-experimentation we can “optimize” ourselves with supplements and a smart watch. This is why people like the “Don’t Die” guy avoid sunlight. Virtually every practicing dermatologist says UV light is all bad, and 9/10 doctors can’t be wrong—truth-by-expert-consensus. This mindset can blind you to contrary evidence that disagrees with the mainstream “consensus,” driving an exaggerated sense that Knowledge > Ignorance.

Learn more about the problems & pitfalls of relying on mainstream expert consensus:

• Article | The Cholesterol Cult & Heart Mafia: How the process of science evolves into The Science™ of public policy

• Article | Manufacturing Consensus: LDL, Atherosclerosis, Seed Oils & Cholesterol Benefits

The second assumption violates a fundamental principle of complex living systems, which any serious student will know: biology is all about context-dependent adaptation and change. Living organisms dynamically respond to the environment. In different environmental contexts, the system can respond to the same input in a different way. “Context” includes both the external environment in which the organism is embedded and its current internal state.

Internal states can change even when the external world remains constant. These changes can be fast, on the scale of seconds to minutes, as with fluctuations in emotion or basic arousal; they can also drift slowly, over months or years, as with developmental shifts or age-related changes in tissue integrity. Thus, any given health protocol may drive two distinct effects, even for the same person in one environmental context, based on changes in their internal context.

What’s the ideal protocol for losing weight? For who? And when? An obese child suffering from fatty liver disease? Or a young man who wants to drop from 18% to 14% body fat? Does the protocol take into account genetic variation in metabolism? And on and on. Protocols are often prescribed without taking these factors into account.

At this point, some may say that can be addressed by simple first principles thinking: biology is ultimately chemistry in motion. Physical laws to apply to everyone, in all contexts. Energy cannot be created or destroyed, and the laws of thermodynamics always apply. The fundamental formula to lose weight (burn fat) is simply to maintain negative energy balance: Calories In < Calories Out. Different contexts may change the values we input into the energy balance equation, but the equation applies in all contexts, always.

An attractive approach, which seems logical at first blush.

The laws of thermodynamics cannot be violated by living systems, which are physical systems—no disagreement there. What this approach misses is that biological systems are composed of dynamic, nested feedback loops. Inputs affect outputs which feedback to regulate inputs, which may drive different outputs next time. Life is a time series of changing internal states. Each slice of the time series is not independent of the last. The state and trajectory of your tissues is history-dependent.

What this means for calories and body composition is that “Calories In” at the present moment affects both “Calories Out” and “Calories In” in future moments. What you consume now is not merely burned for energy (ATP production), as in a bomb calorimeter. To some extent, what is consumed right now reshapes the dynamics of the organism, influencing things like its energy expenditure (“Calories Out”) and its future tendency to eat (“Calories In”).

Let’s look at a concrete example, where “Calories In” is the same under two conditions, but energy expenditure and body weight nonetheless differ. The difference is driven by MOTS-c, a mitochondrial-derived peptide that has conditional (context-dependent) metabolic effects. The MOTS-c peptide influences how the body utilizes its inputs, which means the same inputs drive different metabolic outputs.

Here’s a preview of the data, showing that MOTS-c protects against diet-induced obesity in animals:

First, we will examine a major cellular pathway MOTS-c taps into. After that, we will fit this particular example of context-dependent metabolic change into a more general framework involving three core cellular pathways, the relative balance of which determines whether tissues are primarily in growth mode or burning mode, and whether that mode of operation is well-synchronized (“healthy”), or a desynchronized (“unhealthy”).

This framework will help us understand the particular MOTS-c example above and can be applied more broadly to anticipate how our tissues manage the fundamental tradeoff between growing and burning in terms of three core metabolic pathways. The goal is to understand enough of the cellular details about these pathways and how they interact (“the trees”) in order to develop a general sense for how our tissues respond across contexts (“the forrest”), in healthy vs unhealthy ways.

MOTS-c Peptide: Basic Cellular Effects

MOTS-c is a short, 16-amino-acid peptide encoded by the mitochondrial genome, not the nuclear genome. Mitochondria produce MOTS-c at some level, depending on conditions, and MOTS-c production tends to decline with age. This peptide is used by mitochondria as a molecular signal to influence the metabolism of the whole cell. Recall the basic fact that proteins are folded polypeptides. So, all proteins are technically peptides, but when we say “peptides” we mean short peptides like MOTS-c, which are like miniature proteins.

The short sequence of MOTS-c is highly conserved across species, with evidence of positive natural selection. This tells us that the structure of the MOTS-c peptide is similar across species, and its function has had adaptive value—what it does made a difference for the survival and reproduction of species. A high degree of evolutionary conservation suggests that the same basic function is served by MOTS-c across species, even if the details differ somewhat.

What are those functions?

MOTS-c affects insulin sensitivity and metabolic homeostasis—how cells respond to and utilize energy inputs, such as glucose (sugar). A major target organ of MOTS-c is skeletal muscle, where it affects cell metabolism through the AMP kinase (AMPK) pathway—one of three major, interacting cellular signaling pathways regulating the extent to which cells are in “build mode” or “burn mode.”

MOTS-c is a stress-responsive signaling molecule that helps coordinate energy metabolism and cellular adaptation to stressors, such as exercise. It operates via two interconnected mechanisms, largely in muscles:

1. Activation of the AMP kinase (AMPK) pathway, a key energy sensor in the cytosol.

2. Reprogramming gene expression in the nucleus, enabling mitochondria to regulate expression of genes involved in metabolic homeostasis.

AMP kinase is a very important enzyme, described as a master regulator of energy homeostasis. You can think of it sort like a cellular “energy gauge.” When the gauge reads low energy, AMPK enzyme activity kicks in, reprogramming cell behavior to boost energy production capacity at the expense of activities that deplete energy.

AMP is adenosine monophosphate. ADP has two phosphates, ATP three. The relative balance of ATP/ADP/AMP is a major signal cells use to track of energy abundance. As ATP is used up to do work, ATP levels decline, and AMP rises. In this context, the AMPK pathway become more active. The cell is basically detecting relatively low ATP and uses this signal to “decide” to reallocate resources to favor ATP-production capacity.

Turning up AMPK has the dual effect of activating pathways that generate more ATP while inhibiting pathways that do not. For example, AMPK activity stimulates glucose uptake and mitochondrial biogenesis. More mitochondria fueled by more glucose, greater ATP production capacity. At the same time, pathways that consume ATP are turned down, such as synthesis of fats and proteins.

If you’re running low on cash, you’re going to shift for time and energy away from activities that cost money (e.g. renovating the kitchen) and focus more on activities that generate more cash (e.g. putting more hours at work). “Build mode” is really the state of using surplus energy (ATP) to build extra stuff, like more muscle, which costs ATP overall. The “burn mode” stimulated by AMPK is about building ATP-generating infrastructure specifically. Spend money to make money, not to acquire new luxury items.

By activating AMPK, MOTS-c shifts cellular resource allocation toward building ATP production capacity (via increased glucose uptake, fatty acid oxidation, and mitochondrial biogenesis) rather than other anabolic processes. Achieving this requires turning down other “build” pathways that cost ATP, such the mTOR-dependent protein synthesis for anabolic tissue growth. Later, we will examine the interplay between AMPK, mTOR, and one more pathway.

Before examining how the MOTS-c peptide affects whole-body metabolism through the AMPK pathway, we need a more general understanding of three key enzyme pathways, of which AMPK is one, that determine whether cells shift into “burn mode” or “build mode.”

Podcast & literature that informs the concepts explored in this article:

• Podcast | Metabolic Effects of Oxidative Stress in Development & Neurodegeneration Disease | Robert Lustig

• Paper | Oxygen as an Organizing Principle of Fetal and Neonatal Development

Growth Mode vs Burn Mode: Three Core Cellular Pathways

To understand how MOTS-c can change metabolic outcomes even when caloric intake stays constant, we need to look at the major control systems cells use to decide how resources are allocated. At any moment, a cell faces a basic choice: use incoming fuel (food + oxygen) primarily to build new cellular material (membranes, proteins, DNA, new cells), or use them to burn fuel more completely for usable energy (ATP) while expanding its capacity to do so in the future.

These two modes cannot be executed at full strength, simultaneously. Life is all about tradeoffs. The cell generally commits to one mode or the other. The long-term consequences of getting stuck in the wrong mode—or having the modes poorly coordinated—show up as metabolic disease.

Three interconnected enzyme pathways act as the main control dials for this decision. The relative activity of each pathway, and whether they are pulling in the same direction, largely determines whether a cell (and by extension a tissue) operates in a healthy build (growth) mode, a healthy burn mode, or a dysregulated state.

• PI3K (PI3-kinase) functions as the cell’s “glucose spigot.” When active, it opens the valve wider, increasing glucose transport into the cell. That glucose can then flow through glycolysis—a fast but low-yield route that produces only a couple of ATP per glucose molecule. However, despite low ATP yield, glycolysis generates valuable intermediates that serve as raw materials for lipids, nucleotides, and other building blocks.

  • Higher PI3K activity therefore biases the cell toward growth and biomass accumulation. This is adaptive during development (fetal cells rely heavily on it) or tissue repair.

  • When stuck inappropriately high, it can drive excessive storage or uncontrolled proliferation, as seen in many cancer cells.

• AMPK (AMP-activated protein kinase) functions as the cell’s primary energy sensor. It is activated when the AMP:ATP ratio rises (signaling low energy charge) and then promotes ATP-generating processes while inhibiting non-essential ATP-consuming ones.. It then does two coordinated things: it stimulates processes that increase future ATP production capacity (enhanced glucose uptake routed toward mitochondria, mitochondrial biogenesis, and fatty acid oxidation) while turning down energy-consuming processes that are not about building that capacity (such as new fat synthesis).

  • Anything that chronically suppresses AMPK makes it harder for cells to shift into efficient burn mode when needed. (Excess fructose can do this via a metabolite that binds irreversibly to AMPK)

  • MOTS-c activates this pathway, as does exercise and fasting.

• mTOR (mammalian target of rapamycin) serves as the growth-commitment pathway. When nutrients and fuel are abundant and the cell has sufficient oxidative capacity, mTOR promotes protein synthesis and other anabolic processes needed for cell growth and division.

  • mTOR sits at the center of much aging and longevity research because dialing it down often shifts cells away from rapid growth, toward maintenance and stress resistance.

An important point is that AMPK and mTOR are usually antagonistic: when AMPK is on, it generally inhibits mTOR. This should make intuitive sense—you do not want to commit heavily to tissue growth while running low on energy.

If we imagine each of these three pathways—PI3K, AMPK, and mTOR—to have two main states, relatively high [+] or low[-], this gives us eight general metabolic states that a cell can theoretically be in. Two of these possible states (where AMPK and mTOR are both high) are unstable and quickly resolve into one of the other states, given that AMPK inhibits mTOR. Two of these states can be considered “well-coordinated” or “healthy”:

• High PI3K + high mTOR + low AMPK supports coordinated growth, as in normal development or recovery.

• High AMPK with mTOR restrained (via AMPK-mediated inhibition) supports coordinated burn mode: enhanced mitochondrial biogenesis and fuel oxidation (glucose and fatty acids), metabolic flexibility, and tissue adaptation to current energy demands.

This table of possibilities is a relatively simple way to model whether cells are in a healthy states of coordinated growth or burning, or an uncoordinated (unhealthy) one. Keep in mind: these pathways do not literally behave binary, on/off switches. Thinking about them each in terms of being relatively high (+) or low (-) is merely to help us think in terms of the general configurations that are possible.

When the three pathways become locked in the wrong configuration for too long, problems accumulate. The unhealthy configurations tend to promote storage, inflammation, metabolic inflexibility, insulin resistance, or other features of chronic disease. When we are metabolically healthy, our cells toggle between the two coordinated growth or burning modes, depending on context.

These three pathways integrate signals from nutrients, energy status, oxygen tension, and other inputs. Their balance is not fixed; it shifts with context. That is why the same caloric load can produce very different body-composition and health outcomes depending on the cell’s current state and history. Rigid protocols that ignore this context-dependence often fail for exactly this reason.

This framework helps clarify what MOTS-c is doing: the peptide primarily activates AMPK while also reprogramming gene expression in muscle. By turning up coordinated burn mode via AMPK activation, MOTS-c reconfigures how tissues process incoming calories. The result, as we will see, is protection against diet-induced obesity and insulin resistance even when caloric intake is not reduced—all because of how the body allocates and oxidizes those calories.

With this map of the core control system in place, we can now examine the specific findings on MOTS-c and see how they fit into the bigger picture of context-dependent metabolic regulation.

Learn more about AMP kinase and cellular energy homeostasis:

• Podcast | AMPK: Biochemistry of Nutrient Sensing, Fasting, Cell Repair & Growth | Greg Steinberg

MOTS-c Peptide: Exercise, Fasting & Behavior

Notice that AMPK leads to coordinated changes across tissues, but does not necessarily have identical effects in each organ. AMPK stimulates tissues like muscle to take up fuel (glucose, fatty acids) and create more ATP production capacity (mitochondria). In the hypothalamus of the brain, AMPK influences circuits regulating hunger in order to stimulate food intake.

AMPK tends to be activated more by energy deficits signals (lower ATP, higher AMP). Fasting and exercise are examples. In contrast, MOTS-c behaves more as an exercise-responsive signaling molecule, rather than a general energy deficit signal. Exercise creates a specific type of metabolic stress (high energy turnover + contraction) that upregulates MOTS-c production.

Fasting creates a different kind of stress (nutrient scarcity with low energy turnover), activating AMPK without driving new MOTS-c production. In fact, prolonged fasting (48 hours in mice) can reduce endogenous MOTS-c levels in muscle and plasma. So, exercise and fasting can both increase AMPK signaling, but only exercise boosts MOTS-c production, which is why this peptide has been described as an “exercise mimetic.” Exercise boosts MOTS-c production by mitochondria, especially in muscles.

Share

In nature, fasting and intense “exercise” (behavioral exertion) tend to be closely linked, because the hunger driven by food deprivation motivates intense efforts to obtain food. In modern humans, exercise is typically done “on purpose” for health reasons, not motivated by life-threatening food deprivation. In fact, many people, such as athletes or body builders, exercise most intensely in the context consuming excess calories and nutrition, for the express purpose of building new tissue, like stronger muscles.

However, if MOTS-c release is stimulated by exercise and thereby increases AMPK (inhibiting mTOR-dependent growth mode), one might expect it to hinder muscle hypertrophy from resistance training. So how then would lifting weights still produce net muscle growth? The resolution lies in nutrient and energy status combined with the timing of exertion versus recovery: high-calorie, high-protein intake around workouts suppresses overall AMPK tone and permits post-exercise mTOR rebound during recovery periods when mechanical tension has already been applied.

Exercise will boost MOTS-c and thereby stimulate AMPK, but high energy intake suppresses AMPK. Someone seriously lifting weights to build muscle will consume substantial calories and protein before and after workouts; this favors lower overall AMPK activity across the training-recovery cycle. Taking muscles “to failure” with high calorie and nutrition intake before and after will stimulate high PI3K and mTOR activity, favoring anabolic growth. This is also why periods of recovery, with much lower exertion, are critical immediately following high-intensity resistance training—it favors the coordinated growth configuration of high PI3K and mTOR, with low AMPK.

For understanding MOTS-c in particular, just know that exercise stimulates MOTS-c release by mitochondria. This turns the AMPK pathway up, favoring energy utilization in muscle tissue for the purpose of boosting ATP production capacity. With that in mind, let’s look at how cell metabolism is reconfigured by exogenous (supplemental) MOTS-c and how that relates to food intake, body weight, and metabolic rate.

Metabolic Effects of MOTS-c

A classic paper studied the effects of MOTS-c on human cell cultures and in animals. Treating human cells with the MOTS-c peptide leads to changes in gene expression within four hours, which continue to unfold for at least 72 hours. Some classes of genes are turned up, such as those involved in metabolic signaling. Others are turned down, such as those involved in inflammation and ribosome production (used for protein synthesis).

An essential take away is that an acute (short-term) application of MOTS-c begins to “reconfigure” cells (change their behavior) quickly, but those changes continue to unfold for hours and days. This is a common feature of peptide signaling: acute increases in peptide levels can trigger a cascade of changes that persist for longer periods, even after the peptide has been degraded.

A brief survey on the major metabolic effects observed for MOTS-c, without diving into the details.

MOTS-c targets the methionine-folate cycle

MOTS-c targets the methionine-folate cycle, which centers on the amino acid methionine and folate (vitamin B9) and supplies carbon units for the epigenetic regulation of gene expression. MOTS-c inhibits key steps in this cycle, leading to accumulation of AICAR, thereby driving AMPK activation (detailed below). This modulates both cytosolic signaling and gene expression programs that influence cellular fuel handling and metabolic adaptation.

AMP Kinase (AMPK) Activation

Higher MOTS-c levels drive greater AMPK activation. As discussed above, this allows MOTS-c to coordinate shifts in cellular glucose uptake, mitochondrial biogenesis, and fatty-acid oxidation. Cells with elevated MOTS-c therefore increase both glucose uptake and fatty-acid utilization—in other words, they burn more fuel. This leads to…

Changes in Whole-Body Metabolism, Insulin Sensitivity & Food Intake

Read more

Summary of main points:

• Meta-analyses are presented as the highest form of evidence, but authors control inclusion/exclusion criteria, weighting, and interpretation—leaving ample room for bias, cherry-picking, and contradictory results from different teams.

• A 2014 meta-analysis concluded higher linoleic acid intake is inversely associated with CHD risk in a dose-response manner and supports replacing saturated fat with PUFAs—yet its overall result hinged on just two heavily weighted studies out of 14 samples.

• A highly-cited 2010 meta-analysis of RCTs concluded replacing SFA with PUFAs reduces CHD events—yet it lumped mixed Ω-3/Ω-6 trials with Ω-6-specific ones, included trials with major (unstated) confounds, and depended on underpowered studies.

• A more meticulous 2010 meta-analysis separated mixed Ω-3/Ω-6 from Ω-6-specific PUFA trials: mixed interventions reduced non-fatal MI and CHD death, while Ω-6-specific diets increased risk of CHD endpoints and all-cause mortality—aligning with distinct biology of Ω-6 vs. Ω-3 PUFAs known for decades.

• Nutrition epidemiology and meta-analyses on this topic routinely feature imprecise surrogate data, statistical voodoo, commercial influences, and selective framing—undermining claims that the science is settled on seed oils being heart-healthy.

Meta-analyses are “studies of studies” where multiple studies are aggregated together. The idea is simple and sensible: compile the relevant published research on a topic, see how all the evidence shakes out. Do most studies show that intervention X drives a positive health outcome, or causes death or disease?

We shouldn’t place faith in any one study. We should look at every relevant study, letting the full weight of evidence point us toward the truth.

Makes perfect sense, right?

But who decides which studies get included and excluded from the analysis, and why? What if two different meta-analyses, by different researchers, come to different conclusions? What if studies that are critically flawed in some way are included in the analysis—how would you know? If you’re like most people, you wouldn’t

Most people—including health influencers, journalists, and even scientists—take the stated results at face value, without scrutinizing the methods or underlying studies. By default, most people place faith in the scholars who publish meta-analyses. The authors of this work are credentialed experts familiar the field. Even if we wanted to, we lack the expertise they have. We should trust them, right?

And why not?

Why not trust professional scientists working in a given field, the people who actually did the work and compiled “all the evidence”?

Did you read all of those studies? I didn’t think so.

The Great Pyramid of Biomedical Research

Meta-analyses and systematic reviews are considered the highest form of evidence in biomedical research, directly informing clinical practice. Scholars within a particular discipline review all of the relevant literature on a topic. Studies are diligently filtered, ranked, and weighted by quality. The findings of every qualified paper are analyzed. Statistics. Stone-cold calculation. The end result is not opinion or conjecture; it’s the science.

That’s the idea, anyway.

The reality: many studies and meta-analyses are garbage.

It’s no secret that the scientific literature is filled with all kinds of irreproducible results and outright fraud. Corruption and deception are not limited to politicians. Smart, credentialed people are not automatically virtuous people; they are just as ambitious and prone to bias as anyone else.

In academia, it’s publish or perish. You don’t have to look far to find examples of scientists who don’t read their own work, or systematic reviews that blatantly misrepresent the clinical trials they cite. The reproducibility crisis is not limited to the social sciences. Clinical research is no exception. Anyone who thinks the existence of a meta-analysis means the science is settled simply isn’t paying attention. As we will see, meta-analytical results can quickly fall apart with some scrutiny, or be contradicted by another analysis published by another team.

Below, we will see how meta-analyses are used and abused by academics with regard to the supposed cardiovascular health benefits of dietary Ω-6 polyunsaturated fatty acids (PUFAs), like linoleic acid (“seed oil fat”).

We will begin by looking at two meta-analyses that claim to support the “official consensus” on Ω-6 PUFAs and cardiovascular health. The first looks at prospective (correlational) data; the second on randomized, placebo-controlled trials. After unpacking the trouble with those meta-analyses, we will survey others that come to different conclusions.

For better or worse, if we really want understand biology, we have to fact-check the fact-checkers, not take them on faith. And if our faith in the mainstream on this consensus begins to crumble, what about other subjects, such as those we’ve never looked at this closely?

Spoiler alert: the great pyramid of biomedical research is a house of cards, at least when it comes to “nutrition science.”

Share

Subscribe now

Trouble With Meta-Analysis: Linoleic Acid & Heart Health

This 2014 systemic review and meta-analysis of prospective cohort studies was published in a American Heart Association journal by respected and highly-cited academics, such as Harvard nutrition epidemiologist Dr. Frank Hu.

Main conclusion of the paper:

The paper reinforces the mainstream consensus on dietary Ω-6 polyunsaturated fatty acids (PUFAs). It has been cited hundreds times, including in the 2015-202 Dietary Guidelines for Americans advisory process and World Health Organization review on fat intake. This work is considered strong evidence in favor of the “heart healthy” designation of polyunsaturated fats, including Ω-6 PUFAs like linoleic acid, the predominant fatty acid in seed oils.

Instead of taking the paper on faith because of it’s high-prestige status markers, let’s take a closer look, beyond the headline conclusions printed on the first page. Arguably more important: let’s look into the underlying studies that most heavily determined the conclusions of the analysis.

Unpacking the meta-analysis

Forest plots are used to depict the effect of each study that went into the meta-analysis, as well as how much each one factored into the equation. Studies are weighted based on overall quality, including sample size and methodological rigor. The authors of any given meta-analysis personally engineer the criteria that determine which studies are included and how much they count. Where there are engineers, there is room for creative engineering.

The overall result is basically a weighted average of studies. Here’s the first forest plot from Frank Hu’s meta-analysis, depicting how dietary linoleic acid “intake” relates to the relative risk of total coronary heart disease events. Numbers less than 1 (left-of-center) indicate that higher linoleic acid “intake” was associated with fewer disease events, such as heart attacks; numbers greater than 1 (right-of-center) indicate that higher intake is associated with more disease events.

Notice the number of studies, where they land, and how they’re weighted.

Notable features:

• Fourteen samples total across ten studies (results for males and females are pulled out as separate samples for some studies).

• 11/14 samples do not show a statistically significant result, i.e. most studies show no significant correlation between linoleic acid and outcome.

• Only 3/14 samples show a significant association, two of which are by far the most heavily weighted samples in the meta-analysis.

• Overall result (diamond): the highest level of linoleic acid “intake” is associated with fewer disease events compared to lowest level.

Share

Don’t let charts and stats overwhelm basic common sense.

The entire result depends on two samples out of fourteen total, heavily weighted. If you didn’t have those two samples, the overall association reported would disappear. These two papers are a great place to start if we want to stress test the overall finding of this influential meta-analysis.

Why were those two trials weighted so heavily?

Ostensibly, the authors of the meta-analysis determined them to be very best studies, which often just means largest sample size. Despite providing more statistical power, larger sample sizes are no guarantee of methodological rigor. Setting that aside, here are the two studies, which indeed had the largest sample sizes.

• NHS-F: 2005 study of n=84,566 females

• HPFS-M: 1996 study of n=41,082 males

Instead of placing faith in Frank Hu because he’s a Professor at Harvard, let’s inspect the cited studies ourselves.

In print, Dr. Hu and his team tell us NHS-F was a study of n=84,566 females. When you look at the study itself, it reports a sample size of n=78,778 women. The meta-analysis must have accidentally over-stated it. No big deal. Let’s give Frank a pass. We’re still talking big numbers here. The study is probably really, really impressive—right?

By the way, why do I keep putting “intake” in scare quotes? Let’s check the Methods section of the study they cite, to see how “intake” was measured:

In 1980, we collected information on usual diet by using a food frequency questionnaire including 61 foods. For each food, a commonly used unit or portion size was specified, and each woman was asked how often, on average, she had consumed that quantity during the previous year. Nine responses were possible, ranging from ‘‘almost never’’ to ‘‘six or more times per day.’’

In 1984, the dietary questionnaire was expanded to include 116 items. Similar questionnaires were used to update dietary information in 1986, 1990, 1994, and 1998. Daily intake of fat and fatty acids was calculated by multiplying the frequency of consumption of each item by its nutrient content and summing the nutrient contributions of all foods on the basis of US Department of Agriculture food composition data.

Once every four years, middle-aged women were asked to provide from memory the average frequency and portion sizes of what they consumed, for dozens of foods.

Imagine your mom walking into an office: “Hi Deb, thanks for coming in. How many pieces of whole wheat bread did you eat each week this past year, on average. How about eggs? And how many apples? Thank you, m’am. We’ll touch base again in four years.”

From there, researchers calculated nutrient intake using a table kept by the USDA. In other words, they assumed middle-aged adults have excellent recall of everything they ate in the past year and that its nutrient content perfectly matched what was in some USDA food database. The database says that eggs contain X grams of saturated fat, on average. Deb said she thinks she averaged ten eggs per week between 1979 and 1980.

X * (Deb’s memory) = Scientific Data

That’s the dataset everything depends on. Nutrient content was calculated from questionnaire results for every person, based on their personal memory, under the assumption it had identical nutrient content to whatever some USDA database says about the cholesterol content of milk, vitamin B content of “meat,” and so forth.

Leave a comment

Who conducted the study, cited by Frank Hu in the meta-analysis? Frank Hu and colleagues! After touching base with Deb and the other women once every four years, they calculate p-values, then write the paper. Later on, he compiles the study together with the next one we’ll look at, publishes the meta-analysis, and for all we know personally chats with someone at the WHO about it.

Eyewitness testimony is far from perfect. Neuroscientists have long known that memories are labile, reconsolidating with every recall. People under pressure to give testimony in murder trials often fail to accurately recall events that unfolded before their eyes. But nutrition epidemiologists build entire careers on asking people to remember how many servings of blueberries they consumed last month, and doing statistics on it.

Could you or I have done such a study any better, back in 2005, with that many people? Perhaps not. But most people don’t even realize this is how “intake” data is collected. You can decide for yourself how much weight to put into such a study, but Frank Hu decided to weigh this Frank Hu study heavily in his meta-analysis.

Remember: the main conclusion of the meta-analysis depends on this study and one other. The next paper is truly remarkable, because it begins to suggest that there’s may be something more going on here.

Learn more about nutrition epidemiology:

• Article | Scientific Junk Food: A brief dissection of nutritional epidemiology

The other study was HPFS, a 1996 study of men. It was also based on asking people to remember what they ate last year, although this one took it a step further:

“We assessed its validity in a random sample of 127 men by comparing their calculated intake of fat with that reported during two diet records over one week taken about six months apart. The correlations between the two assessments of intake, adjusted for energy and corrected for within person variation in the diet records, were 0.75 for saturated fat, 0.76 for cholesterol, and 0.37 for linoleic acid.”

Out of over 40,000 men, they took 127 (0.3% of the sample) and looked at the correlation between their calculated fat intake and what they reported from memory in two weekly diet records. For linoleic acid, the key fat analyzed in the meta-analysis citing this study, the correlation was a measly 0.37—a weak correlation, all assuming memory recall is accurate and reported nutrient content matches what’s in some database.

And it gets much worse.

From the “Statistical Analysis” section of the Methods:

In addition, we conducted analyses with the proportion of calories contributed by different fats as continuous variables. The results were corrected for measurement errors in the assessment of diet, body mass index, and alcohol consumption by using data from the validation study. Intake of polyunsaturated fat, rather than linoleic acid, was used for correction of measurement error because intake of linoleic and linolenic acid could not be obtained from the diet record database.

Read that again.

They couldn’t even obtain estimates for linoleic (Ω-6) and linolenic acid (Ω-3) from the database they used to calculate nutrient content. So what did they do? They just used total estimated polyunsaturated intake, rather than linoleic acid or any specific PUFA, and used that in their “statistical corrections.”

Ω-6 and Ω-3 PUFAs cannot all be lumped together as “polyunsaturated fats” as if they are interchangeable, as they often have opposing biological effects and compete with one another for enzymatic processing and membrane phospholipid incorporation (more detail below).

Share

What we’re looking at here is quantitive voodoo. Statistical gymnastics layered on a ridiculous dataset. And even after knowing that, guess what? This study still found no significant correlation between linoleic acid intake and myocardial infarctions or fatal heart disease.

The relative risk data for each quintile of linoleic acid “intake,” from the study itself:

Wait a minute.

The data table is telling us that, compared to the lowest quintile of linoleic acid intake (column 1 in the table), every other quintile (columns 2-5, higher intake) was associated with a higher relative risk of fatal coronary heart disease. The results are non-statistically significant, but… didn’t Frank Hu’s meta-analysis tell us that this study (HPFS-M) showed lower relative risk of heart disease deaths for the highest compared to the lowest quintile of linoleic acid (Ω-6) intake?

It sure did.

In fact, the study we’ve been looking at was the single most heavily weighted one in the meta-analysis section on heart disease deaths:

The study cited also looked for associations between myocardial infarctions and heart deaths for all the dietary fats analyzed, with each expressed as a proportion of total energy. (Note: this is also an issue, as expressing values as proportions rather than concentrations can alter statistical relationships; see M&M 294; again, this is voodoo).

Here’s the data for all fats, including linoleic acid (Ω-6). Look very carefully:

They are calculating the relative risk of consuming linoleic acid at 5% of energy, as well as other fats, after adjusting for age and other factors like total fat and fiber intake. For linoleic acid, they observe no significant association anywhere, except for one barely significant association between linoleic acid and fatal heart disease, after multivariate adjustment plus total fat adjustment. However, when they adjust for fiber intake on top of that, the barely significant association goes away. (See highlights above).

Remember: the main conclusion of Frank Hu’s meta-analysis rests on the two heavily-weighted studies we just looked at. That paper does not appear to accurately represent the results of one of those studies, which finds no significant association between linoleic acid intake and cardiovascular outcomes after controlling for all other factors in the dataset.

Ladies and gentleman, this is the top of the “hierarchy of evidence” in biomedical science, used to inform clinical guidelines and public policy.

Leave a comment

Consider the time it’s taken to wade through this orgy of methodological absurdity. Epidemiologists asked a bunch of middle-aged adults to recite from memory what they ate last year, cooked up nutrition values based on assumptions, and calculated some p-values. Years later, Frank Hu and his team compiled these studies together and cleverly “weighted” the evidence, which they didn’t even represent accurately.

In fact, the HPFS-M study comes to the exact opposite conclusion of the Frank Hu meta-analysis. Take a look at them, side-by-side, in their own words:

Frank Hu meta-analysis:

In prospective observational studies, dietary LA intake is inversely associated with CHD risk in a dose–response manner. These data provide support for current recommendations to replace saturated fat with polyunsaturated fat for primary prevention of CHD.

HPFS-M study:

These data do not support the strong association between intake of saturated fat and risk of coronary heart disease suggested by international comparisons…They also support a specific preventive effect of linolenic acid intake [that’s Ω-3 linolenic acid, not Ω-6 linoleic acid].

Summary

The HPFS-M study mostly failed to find significant associations between linoleic acid and cardiovascular outcomes. The one barely significant association observed for linoleic acid went away after controlling for dietary fiber intake, as did the associations found for saturated fat and cardiovascular outcomes. The study itself found that it’s own method for calculating linoleic acid intake poorly correlated with reported intake from patients. Statistical voodoo.

The Frank Hu meta-analysis can be tossed in the trash, as it’s conclusions ret on heavily-weighted studies with absurd data and methods. Suspiciously, at least one of the major studies used to justify its conclusions appears to be misrepresented. This work was presumably peer-reviews and read by journal editors, who apparently didn’t look closely.

In the end, it doesn’t really matter: throw out the baby and the bathwater. The baby is imprecise surrogate intake data, not real measurements. The bathwater is statistical mumbo jumbo.

Oh, and the absurd datasets above—you know, where they ask people to remember how many apples they ate last year—is standard practice in nutrition epidemiology. The entire field is a house of cards built on a foundation of low-quality data, churning out results that the heart mafia has used for decades to influence the content of public policy and messaging of public health institutions.

"Heart mafia” is not originally my terminology, by the way. I first heard it here:

Learn more about the checkered history of nutrition research:

The ridiculous Frank Hu meta-analysis only looked at prospective studies, not randomized controlled trails (RCTs) capable of showing cause-and-effect. Many such studies have been done.

RCTs are the “gold standard” in clinical research. At the end of the day, we need to look to those trials, no matter what correlational studies claim to show.

Learn more about the mechanistic basis for how dietary fats and seed oils relate to heart disease:

• Podcast | Seed Oils & Heart Disease: Oxidized LDL, Cholesterol, Fat & Cardiology | Tucker Goodrich

Meta-Analysis of RCTs: Different Studies, Different Conclusions

Read more

Nick speaks with Gábor Somlyai, PhD about the biological role of deuterium, a heavier, non-radioactive isotope of hydrogen present in water and organic molecules. Somlyai’s key argument is that small shifts in the deuterium-hydrogen (D-H) ratio act as a fundamental regulator of cell growth, with deuterium-depleted water (DDW) inhibiting tumor progression. They explore mitochondrial production of naturally deuterium-depleted metabolic water, deuterium’s effects on biochemical reaction rates, evidence from cell culture and animal models, and human clinical data showing improved outcomes in cancer patients using DDW as an adjunct therapy.

TOPICS DISCUSSED:

• Deuterium Basics: Deuterium is twice as massive as regular hydrogen; natural water contains ~150 ppm, with stronger oxygen-deuterium bonds slowing reactions compared to oxygen-hydrogen.

• Mitochondrial Role: Healthy mitochondria produce deuterium-depleted metabolic water (~110 ppm from fat oxidation vs. ~150 ppm from carbs), helping maintain low cellular D:H ratios.

• Cell Growth Regulation: Lower deuterium slows cell division in vitro; cells sense small D:H changes, with higher levels (near/above 150 ppm) promoting growth.

• In Vitro Evidence: DDW (e.g., 25-125 ppm) can reduce proliferation and alters cancer-related gene expression.

• Animal Models: DDW caused tumor regression in mice with transplanted human cancers and showed possible efficacy in veterinary use for dogs and cats.

• Human Prostate Cancer: Phase 2 randomized study showed greater Prostate-Specific Antigen (PSA) reduction, tumor shrinkage, and better 1-year survival with DDW plus standard therapy vs. placebo.

• Dietary Modulation: Ketogenic/high-fat diets lower body deuterium from lipid oxidation.

• Broader Implications: Potential in neurodegeneration models; larger trials needed for registration as adjunct cancer therapy.

ABOUT THE GUEST: Gábor Somlyai, PhD is a molecular biologist who pioneered deuterium depletion studies at the National Institute of Oncology in Hungary. His primary work focuses on the role of deuterium in living organisms and the development of deuterium-depleted water for cancer therapy.

RELATED CONTENT:

• M&M 221: Regenerative Energy & the Light Inside You | Jack Kruse

KNOW YOURSELF:

• Visit the SUPPORT PAGE if you find value in this content. Includes discount codes + links to affiliate partners, including products to help you measure and manage your health.

Listen or watch on your favorite platform:

• Audio version: [Apple] [Spotify] [Elsewhere]

• Video version: [YouTube]

PRACTICAL TAKEAWAYS:

• Drink deuterium-depleted water (DDW) at targeted concentrations (e.g., 85-105 ppm) for months may act as a potential adjunct to standard care in oncology, following established protocols.

• Prioritize high-fat, low-carbohydrate (ketogenic-style) eating to naturally reduce bodily deuterium levels through mitochondrial metabolism.

SUBSCRIBER CONTENT BELOW: Reference paper + episode transcript.

Read more

Nick speaks with Dr. Isaac Harris about the biology of antioxidants and oxidative stress. Harris shares his lab’s recent work showing that breast cancer tumors catabolize extracellular glutathione to acquire cysteine and support growth. The conversation explores how reactive oxygen species (ROS) arise in metabolism, the central role of glutathione as a natural antioxidant, why high-dose antioxidant supplements can backfire, and emerging strategies that target cancer’s reliance on antioxidant systems.

TOPICS DISCUSSED:

• Oxidative Stress: Reactive oxygen species (ROS) generated by mitochondrial metabolism cause damage; context determines whether they promote or suppress cancer.

• Glutathione Biology: Primary cellular antioxidant made from cysteine, glycine, and glutamate; works with peroxidase enzymes to protect against ROS.

• Antioxidant Paradox: Clinical trials (SELECT vitamin E, beta-carotene) showed supplements increased cancer rather than preventing it.

• Cancer Antioxidant Hijacking: Tumors ramp up antioxidant systems to handle high metabolic rates and survive stress including metastasis.

• Extracellular Glutathione Catabolism: Cancer cells break down external glutathione into amino acids, particularly cysteine, to fuel growth (new findings).

• Therapeutic Window: Blocking glutathione breakdown slows tumors in models while sparing normal cells.

• Diet vs Supplements: Balanced diets support glutathione needs; excess isolated antioxidants may enrich the tumor microenvironment.

ABOUT THE GUEST: Isaac Harris, PhD is a biologist working on the role of antioxidants in health, disease, and cancer metabolism.

RELATED CONTENT:

• M&M 229: Linoleic Acid, Seed Oils, mTOR & Breast Cancer | Nikos Koundouros & John Blenis

KNOW YOURSELF:

• Visit the SUPPORT PAGE if you find value in this content. Includes discount codes + links to affiliate partners, including products to help you measure and manage your health.

Listen or watch on your favorite platform:

• Audio version: [Apple] [Spotify] [Elsewhere]

• Video version: [YouTube]

PRACTICAL TAKEAWAYS:

• Obtain amino acids like cysteine through whole-food protein sources rather than isolated antioxidant supplements.

• Exercise caution with high-dose vitamin E, NAC, or glutathione supplements, especially with a cancer history.

• Prioritize evidence-based metabolic approaches over blind antioxidant megadosing.

SUBSCRIBER CONTENT BELOW: Reference paper + episode transcript.

Read more