Carl read the papersWhat it actually is
Humanin was found by functional expression screening for a factor that could stop neuronal cell death caused by familial Alzheimer's genes and amyloid-beta. It turned out to be encoded by a short open reading frame inside the mitochondrial genome, which was a genuinely novel idea at the time and started the whole mitochondrial-derived peptide field. MOTS-c, already on this site, is its best known relative.
It is treated as a mitokine: a signal the mitochondria send out when stressed. Levels fall with age in rodents and in people, and are higher in the children of centenarians than in age-matched controls, which is where most of the longevity interest comes from.
How it’s thought to work
- Cytoprotection — Protects cells against a range of insults in culture, which is the effect it was discovered through.
- Insulin action, centrally — In rodents, infusion into the brain improved whole-body insulin sensitivity through hypothalamic STAT-3 signalling, and blocking that signalling removed the liver effect.
- Mitokine signalling — Circulating levels rise after acute endurance exercise in humans and fall with age, which fits a stress signal rather than a hormone with a set point.
- Analogues do the work — Most of the animal results use HNG, a more potent engineered version, not native humanin. That distinction gets lost in marketing.
What the research says
Plenty of biology, no human dosing. In C. elegans, overexpressing humanin was enough to extend lifespan; in middle-aged mice, twice-weekly treatment with the HNG analogue improved metabolic healthspan measures and lowered inflammatory markers. In rodents, infusion into the brain and intravenous analogues improved insulin sensitivity. In people, the evidence is observational: levels decline with age, are higher in centenarians' children, are lower in Alzheimer's disease and MELAS, and rise measurably after a 45 minute bout of cycling. A rare variant (P3S) found mostly in centenarians carrying APOE4 reduced amyloid pathology in mice. There is no published study in which humanin or HNG has been given to a human being: a ClinicalTrials.gov search returns seven records mentioning humanin, all measuring it as a marker, and a PubMed search for humanin clinical trials returns studies of exercise, hyperbaric oxygen and astaxanthin that measure it rather than administer it.
- Encoded inside the mitochondrial genome, 24 amino acids, discovered in 2001 through a neuroprotection screen.
- Human data is observational: lower with age and in Alzheimer's disease, higher in the offspring of centenarians.
- Plasma levels rose after a single 45 minute endurance session but not after a single resistance session; 12 weeks of resistance training raised humanin protein in muscle by 35% with no change in serum.
- Animal benefits mostly used the engineered analogue HNG, not native humanin.
- Zero human administration studies, and no registered trial that gives it to anyone.
Reported protocols
These are the doses the peptide community actually reports running (Reddit & forums), alongside published research — logged for education, not as a recommendation to use.
| Parameter | Commonly reported range |
|---|---|
| Human dose | Not established. Nobody has published a dose in people, because nobody has given it to people in a published study. There is no pharmacokinetic data, no safety data and no exposure range to copy |
| Animal dosing | The mouse work used twice-weekly treatment with the HNG analogue from middle age. Mouse schedules do not convert into human doses, and this page is not going to pretend otherwise |
| Which molecule | Most published effects are HNG, not native humanin. If a vial says humanin, it is probably not the thing that produced the results being quoted at you |
| Route | Injected in animal studies; no established human route |
| Cycle | Not established |
| Carl's note | The only reliable way known to raise your own humanin is already in the literature: a session of endurance exercise did it in healthy volunteers, and 12 weeks of resistance training raised it inside muscle. |
Common use cases
| Goal | Community protocol notes |
|---|---|
| Research only | the honest category: cell and animal work on mitochondrial stress signalling |
| Longevity marker | measured as a marker in ageing studies, which is not the same as being used as a treatment |
| Exercise response | the one human intervention shown to raise it, and it costs nothing |
Mixing it (reconstitution)
Freeze-dried peptides must be reconstituted with bacteriostatic water before they can be measured. Carl’s calculator turns “mg in the vial + ml of water + target dose” into “units on the syringe” — and tells you how long the vial lasts.
→ Open the Reconstitution Calculator
Injection & handling
There is no human injection protocol to describe, and inventing one would be the dishonest thing to do here. No published study has given humanin or HNG to a person, so there is no dose, no frequency, no route and no safety window to work from. What can be said generally is that it is a peptide, so it would be handled like any other, reconstituted with bacteriostatic water, refrigerated and kept out of the light, and none of that makes the missing dosing data any less missing. Anyone selling a humanin protocol is writing fiction with real-looking numbers in it.
Stacks it appears in
- With MOTS-c — Its best known relative and the other mitochondrial-derived peptide with a page here. They get described as a set, which is a family resemblance rather than a tested combination.
- With SS-31 or NAD+ — Mitochondrial stacks are built this way online. There is no human data for humanin in any combination.
Sources
- Hashimoto Y et al. A rescue factor abolishing neuronal cell death by a wide spectrum of familial Alzheimer's disease genes and Abeta. Proc Natl Acad Sci USA 2001 (PMID 11371646)
- Yen K et al. The mitochondrial derived peptide humanin is a regulator of lifespan and healthspan. Aging (Albany NY) 2020 (PMID 32575074)
- Muzumdar RH et al. Humanin: a novel central regulator of peripheral insulin action. PLoS One 2009 (PMID 19623253)
- von Walden F et al. Acute endurance exercise stimulates circulating levels of mitochondrial-derived peptides in humans. J Appl Physiol 2021 (PMID 34351816)
- Gidlund EK, von Walden F et al. Humanin skeletal muscle protein levels increase after resistance training in men with impaired glucose metabolism. Physiol Rep 2016 (PMID 27923980)
- Miller B et al. Humanin variant P3S is associated with longevity in APOE4 carriers and resists APOE4-induced brain pathology. Aging Cell 2024 (PMID 38520065)
- Burtscher J et al. Mitochondrial stress and mitokines in aging. Aging Cell 2023 (PMID 36642986)
- ClinicalTrials.gov search for 'humanin' (7 records, none administering it)
- PubMed search: humanin with publication type clinical trial (8 records, none administering it)
- SAHPRA registered health products register (no humanin product)
- SAHPRA, Peptide Products public information page (May 2026)
- WADA 2026 Prohibited List, S0 non-approved substances
