What "This Supplement Has a Solid Mechanism" Actually Means

What would you say if the only controlled human trials of one of the most popular peptides online were done by squirting it up people's butts, while longevity influencers tell you to inject it or swallow it?

Yes, you read that right. The peptide is BPC-157, sold online as an injection or an oral "supplement" for healing and longevity.

Not injecting it, and not swallowing it as a supplement. BPC-157 was being studied for a claim that goes around often: gut health, and specifically ulcerative colitis. Putting it straight into the large intestine is the most direct way to get it there, and that area is full of blood vessels.

Even so, the blood tests couldn't detect it.

And the ulcerative colitis trial? It didn't clearly beat the placebo.¹

Welcome to the wild world of molecular mechanisms dressed up as marketing claims.

This isn't new, either.

In 2008, one of the biggest drug companies in the world paid about $720 million for a small company whose molecules seemed to switch on an enzyme linked to aging. Two years later, scientists at a rival drug company tested those molecules a different way.

In the standard lab test, they only switched on the enzyme when a glowing dye was attached to the test material. Take the dye away, and the effect disappeared.²

Was it a $720 million lab mistake?

Not exactly. The scientists behind the idea came back with real evidence that the enzyme can be switched on directly, under the right conditions.

Proving it took about a decade, and even with $720 million and all those years spent working out whether the mechanism was real, the compound hasn't turned out to be a proven longevity intervention. Very little of the form we swallow reaches the blood intact, and the jury is still out on whether the doses on the shelves can engage the mechanism scientists are still arguing about anyway.

You've likely seen this molecule on the supplement shelves and listened to countless podcasts and advertising reels about it: resveratrol, the red wine compound.

I spent fifteen years running companies in the supplement and wellness industry, right through the years resveratrol went from a lab finding to a supplement aisle.

I took resveratrol myself for a while, until I learned how little of it reaches the blood unchanged when you swallow it, how unsettled its proposed mechanism still is, and how far the doses used in studies are from what you can afford to take in a capsule.

Both of those stories run on one word: mechanism.

It's the word a longevity influencer uses on a podcast when they say something "has a solid mechanism," and it's the word behind the majority of claims and marketing copy that probably landed a supplement bottle in your cabinet.

By the end of this post you'll know what a solid mechanism actually is, why it's so hard to prove, how my own scoring uses it, and four questions that separate a real mechanism from a sales pitch.

The Bottom Line

A mechanism is the chain of steps between taking something and a change in your cells, and for most supplements it's the main evidence there is.

Proving one is hard enough that drug companies with hundreds of millions on the line spent years arguing over a single link.

Even a real mechanism still has three more gates to clear before it does anything for you:

• The dose.
• The route you take it by.
• How much survives the trip into your blood, since much of it can be broken down in the intestines and liver before it ever reaches your blood and the tissues where the mechanism is supposed to work.

So what is a mechanism of action?

It's the chain of steps from swallowing something to a change inside a cell, and every link has to hold.

Almost every real drug starts there. For a lot of supplements, it's also where the science stops, because the human trials that would come next rarely get funded.

How a red wine molecule got to $720 million

It started in a single cell

In 2003, a team reported that resveratrol switched on a sirtuin, one of a family of enzymes that help cells respond when food is scarce.

Sirtuins sit inside one of the hallmarks of aging I covered in my hallmarks post: deregulated nutrient sensing, the cell's ability to read how much fuel is around and adjust.

In yeast, resveratrol extended lifespan by 70%.³

Yeast is a single cell. It shows something can happen, not that it happens in an animal with a liver, a gut and a bloodstream.

Two years later, the first crack

The standard test gives the enzyme a small piece of protein with a glowing dye attached. When the enzyme does its job, the dye lights up.

In 2005, a lab ran the test without the dye:

• With the dye, resveratrol boosted the enzyme.
• With a natural protein fragment and no dye, it had "no effect."⁴

The same lab also went back to yeast and, in three different strains, found resveratrol had no detectable effect on lifespan.

Then a mouse, and the money

In 2006, middle-aged mice eating a high-calorie diet were given resveratrol. It shifted their physiology "towards that of mice on a standard diet" and they lived longer.⁵

That's the next link: a living mammal, not a dish. Notice the detail, though: these were mice on a junk-food diet.

In 2008, the company built on this idea sold for about $720 million.

However, scientists making a living in the world of biochemistry and mechanism science were still arguing about the dye. That's normal. Big bets get made while the science is still open, because waiting for it to close can take a decade.

The rival lab, and the answer

In 2010, a team at a rival drug company tested resveratrol and the acquired company's drug versions with natural proteins. There was no activation. The compounds were binding the dye-tagged test fragment itself.²

The scientists behind the original idea didn't walk away. In 2013 they came back with real data.

The dye, it turned out, resembled something real: a bulky, water-repelling patch that some of the enzyme's natural targets also carry.

Then they ran one of the strongest tests available in the world of biochemistry and molecular mechanism science. They changed a single building block in the enzyme, one amino acid, and the activation largely disappeared. In cells carrying that changed enzyme, "the metabolic effects" of these activator compounds "were blocked."⁶

And the links that didn't carry over

Remember the junk-food mice? When the National Institute on Aging's testing program gave resveratrol to mice on a normal diet, at three separate labs, it "did not have significant effects on survival in male or female mice."⁷

Then there's the blood:

• When six volunteers swallowed 25 milligrams of resveratrol, at least 70% was absorbed, and the body broke it down so fast that only "trace amounts" of resveratrol itself ever showed up in their blood.⁸
• At a single 5-gram dose, which is 5,000 milligrams, blood levels peaked at 2.4 micromolar.⁹

Now check the label on your bottle. When an independent lab tested resveratrol products, servings ranged from about 100 to 600 milligrams: a tenth of that dose, or far less.

The same 5-gram study notes that resveratrol's anti-cancer effects in cells need "levels of at least 5 micromol/L." So even with the huge dose given to human volunteers, it still fell short of what those cell experiments used.

So what came of the $720 million bet on resveratrol science?

In 2013, five years after the purchase, the drug company closed the acquired company's office and folded what was left into its own research.

What a solid mechanism looks like

When a scientist says something "has a solid mechanism," they mean the chain has been checked link by link, ideally with one of three kinds of proof that the compound really acts on its target:

• It sticks to the target, measured directly on the purified protein.
• The knockout test. Remove the target, or change it, and the effect disappears. That shows the effect needs that target, which is why the 2013 result carried so much weight.
• A picture of it bound. An image, at near-atomic detail, of the molecule sitting in its target.

When I worked in epigenetics and cancer research, we had access to a cryo-electron microscope, which flash-freezes molecules so you can see their shape, and seeing proteins at that level of detail is pretty cool.

The microscope alone costs roughly $5 million, before the specially built room it needs. That's the price of this kind of proof.

Rapamycin is the textbook example of mechanism science that held up. It grips its target, a growth switch called mTOR, and there are pictures of it doing exactly that.

In the same three-lab testing program where resveratrol did nothing, rapamycin started in mice at 9 months of age extended median survival by an average of 10% in males and 18% in females.⁷

Rapamycin is a prescription drug that suppresses the immune system, and these were mice. It's here as the benchmark for mechanism science.

I'll save my analysis of rapamycin for a future post. It's a great example of how, even when the mechanism is no longer in dispute, the human evidence and outcomes don't always cleanly follow.

You've already seen a simplified example of a knockout test in people on this site, too.

In my mouthwash post, the bacteria in your mouth turn nitrate from vegetables into nitrite, which your body uses to make nitric oxide and relax your blood vessels.

"Knock out" the beneficial bacteria with a week of antiseptic mouthwash and that link breaks: the mouth makes far less nitrite, and in a week-long study, blood pressure went up a few points. That's a mechanism someone broke on purpose, in people.

Why mechanism carries so much weight in my scoring

I have an MS in bioinformatics and computational biology that was based around epigenetics, cancer therapies, and utilizing advanced computational methods in the analysis of hundreds of studies and massive amounts of biological data derived from mRNA sequencing and mass spectrometry.

My current PhD project is based around these same techniques, which I am applying to the biology of aging, via the framework known as the Hallmarks of Aging.

I've been building out a consumer version of my thesis project I've come to call the Longevity Intelligence Report, which scores the supplements, drugs and habits in your routine and opens to everyone in mid-December (early access). Mechanism is half of how it grades each one.

What I want to know about anything in your stack is how it connects to the biology of aging, and which parts of that connection anyone has actually tested.

Trials long enough to show a supplement slows aging in people don't exist, and studies that measure the hallmarks of aging directly in people are rare, because diagnostics specific to the individual hallmarks either don't exist or are only in their infancy.

So the mechanism, and the lab markers associated with each hallmark, are the best stand-ins we have to examine an intervention's potential effect on aging biology.

Every supplement gets two scores for each hallmark of aging:

• The mechanism score asks: if the biology plays out, could this touch that part of aging, and how much of that has been shown for this compound?
• The evidence score asks: has anyone seen it happen in a real study? Cell and animal studies can earn a 1 or 2, and human studies 3 to 5, based on criteria I'll dig into in future posts and that are already spelled out on our methodology page.

The two results (mechanism and evidence) are multiplied, so weak evidence drags down even a great mechanism, and a perfect mechanism with no evidence at all scores zero.

Back to resveratrol, and how it scores:

• Nutrient sensing, the piece of aging biology it's supposed to act on most directly and the hallmark this whole story is about: a 4 out of 5 for mechanism. It acts on the cell's fuel gauge (AMPK) and on the sirtuins, without the nailed-down proof a 5 requires. It also has more human evidence there than anywhere else, but the results are mixed: pooled analyses of human trials point to modest improvements in blood-sugar control in people with type 2 diabetes,¹⁰ and no effect in people without it. So that hallmark is rated Emerging, not Supported.
• Absorption: the picture changes when you get to its bioavailability: in plain terms, how much of it reaches your blood unchanged after you swallow it, and from there the tissues that matter. Its grade is Low, and the report flags it.
• Stem cell health, another hallmark of aging: human and animal studies have reported changes tied to that hallmark, but nothing explains how resveratrol would act there, so it scores zero.

The whole method is public: how we grade every intervention, with the mechanism step here.

It's still happening

The resveratrol tale is an old story now, but the same pattern repeats itself every day as new supplements come out, and it's especially rampant in the world of peptides.

Peptides often have fairly well-understood mechanisms and almost no human evidence. The doses are often understudied, and the best forms and routes are still up in the air.

To make matters worse, peptides are powerful compounds, and plenty of people are taking research-grade versions with almost no safety studies behind them.

Insulin was the first peptide ever used as a drug, and it's still one of the most dangerous drugs in the hospital when the dose is wrong. I see that play out with diabetics in the emergency room every day, where low blood sugar can kill as fast as a heart attack or stroke if it isn't corrected quickly.

So how do you spot a mechanism being oversold? Watch for these five moves.

1. The hub trick

"Activates AMPK." "Supports your sirtuins." "Turns on Nrf2."

These are master switches, and a long list of compounds touch them. Acting on a master switch can be the real thing: it's how rapamycin works.

The trick is when that's the whole case, when nobody has measured what the compound does to the part of aging on the label. In my scoring, I give a connection inferred that way a 1 out of 5, the lowest score above zero.

On a supplement label, it often leads the marketing story and is communicated as established science.

2. Can BPC-157 be taken orally? The route doesn't match

In the animal studies, it was mostly given by injection into the belly cavity, by tube into the stomach, or in drinking water.¹

The only controlled human trials with results gave it as an enema, one in healthy volunteers and one for a bowel disease, and in the blood tests afterward the peptide wasn't detected.

The FDA's own review found no human data on how much reaches the blood when it's injected, swallowed or sprayed in the nose, which are the ways it's sold.¹ A trial of BPC-157 as a pill was registered in 2015 and never posted results.

I'll cover BPC-157 in detail in an upcoming post. Not just one or two studies: every study we could find on this compound has been run through the computational biology engine I use to pull together the evidence and the conflicts of interest behind it.

3. Apigenin: the result came from an injection

Apigenin is sold as a way to raise NAD+, a molecule cells need to make energy, because it blocks an enzyme called CD38 that uses NAD+ up.

The study behind that claim injected mice with apigenin into the belly every day for a week, and their liver NAD+ went up.¹¹

I couldn't find a single controlled trial of an apigenin pill on its own, in people, that measured NAD+, sleep or aging. Not one.

4. The acai berry and the minefield of proprietary formulas

Around 2010, acai was being talked about like a miracle from the Amazon, with claims ranging from better metabolism to wild ones about cancer.

One company announced its sales had passed a billion dollars, built on the story of acai as the antioxidant that could cure almost anything.

But the problem runs deeper. These formulas are often proprietary, so you have no idea how much of the ingredient is in the bottle, or even what form it's in. Comparing the label to actual study data is impossible.

My approach to proprietary formulas? Steer clear. If a company won't show you what's in the capsule you're swallowing, and how much of each ingredient, don't bother. I don't. I can't analyze something I can't match to the evidence.

5. How it gets on the label

The federal rule for supplement labels allows statements that "characterize the documented mechanism" of an ingredient.

The company is supposed to have evidence for it, but nobody at the FDA checks before it's printed. That's what the familiar disclaimer means: "This statement has not been evaluated by the Food and Drug Administration."

A label that lists an ingredient studied at a big dose, then includes a sliver of it, has a name in the trade press: fairy dusting.

“

They list the ingredient by name, leave off the dose, and then make marketing claims off the study evidence, even when they know damn well the dose in the bottle is a fraction of what the studies used.

Four questions to ask about any mechanism claim

1. Where was it shown: in cells in a petri dish, in an animal, or in a person?
2. Was this compound tested, or is the link borrowed from a pathway it touches?
3. Was it shown at the dose, in the form, and by the route you're actually taking?
4. Who paid for and conducted the research?

If the label can't answer the first one, look up the study it's leaning on. If the answer is "a dish," the next two usually settle it.

The fourth one matters just as much, and I'll cover it in detail in future posts. In the Longevity Intelligence Report, we check every human study in a dozen places for conflicts of interest and follow the money, so you can weigh that when you look at an intervention and its evidence.

A conflict of interest doesn't automatically mean bias or foul play. But even if the mechanism and evidence check out, raise an eyebrow if the only studies were paid for and run by the company that holds the patent or sells the compound. Raise it higher if a couple of independent studies found no effect.

What we don't know yet

Mechanism scoring is a stand-in, which is why I multiply it by evidence.

A high score doesn't tell you that your bottle, at your dose, will slow your aging, and my absorption grade can't tell you whether your particular product reaches the target.

Resveratrol does show modest benefits in some people with metabolic problems. Nobody has shown it helps a healthy person live longer.

What I changed

I stopped taking resveratrol once I saw how little of it reaches the blood unchanged and how far the study doses are from what's in a capsule.

And I don't buy proprietary blends anymore: if a label won't tell me how much of each ingredient is inside, I can't match it to the evidence, so it doesn't go in my cabinet.

The Calibrated Claim Audit

Claim	Evidence	My read
"It has a solid mechanism"	Means proven link by link; the strongest proofs are binding, a knockout or a picture of it bound	Ask which links were checked, and in what
"Activates longevity pathways"	Often a master switch inferred from other compounds, with nothing measured for this one	The weakest kind of mechanism claim
Resveratrol extends life	Yeast in one lab and junk-food mice, yes; normal-diet mice at three labs, no7; only trace amounts reach human blood unchanged8	Not shown in mammals on a normal diet, never tested for lifespan in people

Commercial distortion risk: high. The main incentive ecosystems here are the supplement industry and drug companies betting on aging biology. That doesn't make a claim wrong, but it changes how carefully I read the evidence.

The Final Signal

• What the claim gets right: mechanism is real science, and for most supplements it's the main evidence there is.
• What it gets wrong: a mechanism shown in a dish says nothing about the dose, the route or how much reaches your blood.
• What I changed: I dropped resveratrol, and I skip proprietary blends.
• What would change my mind: a controlled human trial, at a dose people actually take, showing a supplement sold on mechanism moved a hallmark of aging.
• What's next: evidence. What counts when the big trial never happens, and the grid that puts mechanism and evidence on one chart. Until then, the rest of my evidence critiques are here.

---

References

1. US Food and Drug Administration. Pharmacy Compounding Advisory Committee briefing document: BPC-157. 2026. fda.gov/media/193343 [Finding: The only controlled human studies with results gave BPC-157 as an enema, and the peptide was not detected in blood afterward; in the ulcerative colitis trial the difference from placebo was not significant; FDA found no human data on blood levels after injection, oral, nasal or skin use.]
2. Pacholec M, Bleasdale JE, Chrunyk B, et al. SRT1720, SRT2183, SRT1460, and resveratrol are not direct activators of SIRT1. J Biol Chem. 2010;285(11):8340-8351. PMID: 20061378 [Finding: Resveratrol and three drug-company activators switched on the enzyme only when the test protein carried a fluorescent dye, not with natural proteins, and the compounds bound the dye-tagged protein itself.]
3. Howitz KT, Bitterman KJ, Cohen HY, et al. Small molecule activators of sirtuins extend Saccharomyces cerevisiae lifespan. Nature. 2003;425(6954):191-196. PMID: 12939617 [Finding: Resveratrol activated the human enzyme SIRT1 in the lab test and extended yeast lifespan by 70%.]
4. Kaeberlein M, McDonagh T, Heltweg B, et al. Substrate-specific activation of sirtuins by resveratrol. J Biol Chem. 2005;280(17):17038-17045. PMID: 15684413 [Finding: Resveratrol boosted the enzyme only on test proteins carrying a non-natural fluorescent tag, and had no detectable effect on lifespan in three yeast strains.]
5. Baur JA, Pearson KJ, Price NL, et al. Resveratrol improves health and survival of mice on a high-calorie diet. Nature. 2006;444(7117):337-342. PMID: 17086191 [Finding: Middle-aged mice on a high-calorie diet given resveratrol shifted toward the physiology of mice on a standard diet and lived longer.]
6. Hubbard BP, Gomes AP, Dai H, et al. Evidence for a common mechanism of SIRT1 regulation by allosteric activators. Science. 2013;339(6124):1216-1219. PMID: 23471411 [Finding: Natural targets with a bulky water-repelling patch allowed direct activation; changing one amino acid in SIRT1 abolished it, and blocked the compounds' metabolic effects in cells.]
7. Miller RA, Harrison DE, Astle CM, et al. Rapamycin, but not resveratrol or simvastatin, extends life span of genetically heterogeneous mice. J Gerontol A Biol Sci Med Sci. 2011;66(2):191-201. PMID: 20974732 [Finding: At three test sites, resveratrol in food did not significantly change survival in male or female mice, while rapamycin extended median survival by an average of 10% in males and 18% in females.]
8. Walle T, Hsieh F, DeLegge MH, Oatis JE Jr, Walle UK. High absorption but very low bioavailability of oral resveratrol in humans. Drug Metab Dispos. 2004;32(12):1377-1382. PMID: 15333514 [Finding: In six volunteers, at least 70% of a 25 mg dose was absorbed, but only trace amounts of unchanged resveratrol (under 5 ng/ml) were found in blood.]
9. Boocock DJ, Faust GE, Patel KR, et al. Phase I dose escalation pharmacokinetic study in healthy volunteers of resveratrol, a potential cancer chemopreventive agent. Cancer Epidemiol Biomarkers Prev. 2007;16(6):1246-1252. PMID: 17548692 [Finding: A single 5 g dose produced peak blood levels of 2.4 micromolar; the authors note anti-cancer effects in cells need at least 5 micromolar.]
10. Liu K, Zhou R, Wang B, Mi MT. Effect of resveratrol on glucose control and insulin sensitivity: a meta-analysis of 11 randomized controlled trials. Am J Clin Nutr. 2014;99(6):1510-1519. PMID: 24695890 [Finding: Across 11 trials (388 people), resveratrol improved fasting glucose, insulin, HbA1c and insulin resistance in people with diabetes, and had no significant effect in people without diabetes.]
11. Escande C, Nin V, Price NL, et al. Flavonoid apigenin is an inhibitor of the NAD+ ase CD38: implications for cellular NAD+ metabolism, protein acetylation, and treatment of metabolic syndrome. Diabetes. 2013;62(4):1084-1093. PMID: 23172919 [Finding: Apigenin blocked CD38 in cells, and mice injected with it daily for a week had higher liver NAD+.]