Humanin: the first mitochondrial-derived peptide — real biology, no human trials

Humanin has one of the best origin stories in mitochondrial biology, and it deserves the attention the science gets — just not the attention the sales pages give it. It is a real mitochondrial-derived peptide with a genuine cytoprotective role, first pulled out of academic work on a degenerating brain. Yet what is marketed as a longevity injectable rests on animal experiments, cell systems and human correlations. The biology is respectable; the therapeutic claims are borrowed from preclinical models, not earned in human trials. Holding those two facts at the same time is the entire job here.

A peptide found in a surviving neuron

Humanin was identified in 2001 from an unusual source: the small population of neurons that survived in the brain of a patient who had died with Alzheimer’s disease. Screening for a factor those cells were making, researchers isolated a 24-amino-acid peptide that rescued neurons from death triggered by amyloid-beta and by a wide spectrum of familial Alzheimer’s mutations — and named it humanin.^[1] What made the discovery doubly remarkable came later: the peptide is encoded not in the cell nucleus but within the 16S ribosomal-RNA region of mitochondrial DNA. Humanin was the first of the mitochondrial-derived peptides (MDPs) to be described, the molecule that opened the field that now includes its sister peptide MOTS-c. That alone is a serious scientific result, and it is worth stating plainly before any of the caveats arrive.

The preclinical biology is real and varied

Across many independent laboratories — almost entirely in animals and cultured cells — humanin behaves as a broad cytoprotective and anti-apoptotic signal. It can block programmed cell death by interacting with pro-apoptotic machinery such as Bax and with IGFBP-3, and it has shown neuroprotective effects in models of amyloid toxicity beyond the original screen. The biology is not confined to the brain. In rodent work, humanin improved insulin sensitivity and altered peripheral glucose handling, framing it as a regulator of metabolism rather than a purely neural factor.^[2]It has also shown cardioprotective and anti-atherosclerotic effects in animal and cell models. Much of this research uses a more potent engineered analog, HNG (S14G-humanin), which is more stable and active than the native peptide — a detail that matters, because a great deal of the “humanin” literature is really about that analog rather than the molecule a vial claims to contain.

The aging signal is intriguing — and only a correlation

The most quoted reason humanin gets pulled into longevity marketing is a set of human observations: circulating humanin levels tend to decline with age, and higher levels have been associated with longevity, including reports of elevated levels in long-lived individuals and in the offspring of centenarians, and across some long-lived species.^[3] In animal work, manipulating humanin has been linked to measures of lifespan and healthspan.^[3] This is a legitimately interesting pattern. But it is associational: a molecule that falls as people age, or runs higher in people who happen to live long, has not thereby been shown to be a cause of healthy aging, let alone something that helps when injected. Plenty of biomarkers track with age and longevity without being levers you can pull. The correlation is a reason to study humanin — not evidence that supplementing it works.

The human-evidence gap

Here is the line the marketing crosses without telling you. There are no adequately powered, randomized, placebo-controlled human trials showing that administering humanin improves any clinical outcome — not cognition, not metabolic disease, not healthspan, not lifespan. The supportive efficacy data are from mice, rats and cell culture, often using the HNG analog; the human data are the observational levels-and-aging associations above. Reviews of the mitochondrial-derived peptide field, written by the researchers who helped build it, are candid that these molecules — humanin among them — remain early-stage, with therapeutic translation to people still unproven.^[4] A peptide that protects a neuron in a dish, rescues a mouse, or correlates with longevity in a cohort has not been shown to be a treatment in a human. Humanin has not crossed that line.

The unregulated-supply problem

Humanin is not an approved drug for any of the purposes it is sold for. It reaches buyers as an unregulated “research” injectable from suppliers operating outside pharmaceutical manufacturing standards — so the actual identity, dose, purity and sterility of what is in a given vial are unverified, and a product labeled “humanin” may not even be the more-studied HNG analog. Those quality unknowns sit on top of the absence of human efficacy and long-term safety data, not in place of it. Modulating apoptosis and cell-survival pathways is also not self-evidently harmless, and no controlled long-term human safety dataset exists for repeated humanin dosing. This is the same structural problem that shadows the rest of the injectable peptide market — the one we describe for NAD+ precursors and for epitalon, where mechanism and lab signals outrun any proven human benefit.

The honest bottom line

Humanin is a molecule worth being curious about. It was the first mitochondrial-derived peptide ever found, it came out of real disease biology, and its cytoprotective and metabolic effects in models are the kind of result that justifies continued research. But “the first MDP, discovered in a surviving neuron” is a story about science, not a license to inject it. The aging-level correlations are suggestive and the rodent data are interesting, and neither is a randomized human trial. Until those trials exist, the longevity pitch is extrapolation, and the product is an unregulated injectable of unverified contents. Respect the biology; distrust the bottle.