Humanin is unusual among peptides because of where its instructions live. Most of the proteins your cells make are encoded in the nucleus. Humanin is encoded inside mitochondrial DNA, in a short open reading frame that overlaps the 16S ribosomal RNA gene (MT-RNR2). That makes it one of the first described mitochondrial-derived peptides, a small class of signaling molecules read out of the mitochondrial genome itself.
The peptide is 24 amino acids long. It was first reported in 2001 by Nishimoto's group, who were screening a cDNA library from the brain of an Alzheimer's patient, looking for genes that kept neurons alive under stress. The clone that protected cells turned out to be humanin (Hashimoto et al., PNAS; review in Yen et al., 2013).
What it is and how it is thought to work
Humanin is best described as a cytoprotective and anti-apoptotic signaling peptide. The proposed mechanisms come mostly from cell and animal work:
- It binds IGFBP-3 and interferes with the pro-apoptotic role of that protein.
- It interacts with Bax and related BCL-2-family proteins to block the mitochondrial apoptosis pathway.
- Outside the cell, it acts through a receptor complex (CNTFR / WSX-1 / gp130) and formyl peptide receptors, switching on STAT3, ERK1/2 and AKT signaling (Cobb et al., Oncotarget 2016).
A synthetic variant called HNG (a serine-to-glycine swap at position 14) is reported to be roughly 1,000-fold more potent than native humanin in protective assays, which is why a lot of the animal literature uses HNG rather than the natural sequence.
State of the evidence
There are two very different bodies of evidence here, and it helps to keep them apart.
Observational human data are real. Circulating humanin can be measured in human blood, and several studies report that levels fall with age. Children of centenarians, who are themselves more likely to reach very old age, tend to have higher humanin levels than age-matched controls. Work linking humanin to cognitive aging in both mice and humans was published in Scientific Reports in 2018 (Yen et al.). So as a biomarker and a piece of mitochondrial biology, humanin is genuinely studied in people.
Therapeutic data are not. The protective effects in Alzheimer's models, diabetes, atherosclerosis, heart injury and macular degeneration come from cell cultures and animals. A 2023 review of humanin and its analogues for neuroprotection states plainly that the evidence is preclinical and that substantial work remains before any clinical use (Mamais et al., *Biology* 2023). We did not find completed human therapeutic trials of humanin or HNG.
Regulatory and quality notes
Humanin is not an approved drug. It has no FDA approval and no established legal pathway for human use, which means material sold to consumers is research-use-only and not intended for human consumption. peptideone publishes this as neutral information and gives no medical, dosing or efficacy advice.
Because humanin is sold as a research chemical rather than a regulated pharmaceutical, there is no official manufacturing standard behind any given vial. For research purposes the things worth checking are the same as for any peptide: a recent certificate of analysis with identity confirmation (mass spec) and a purity figure (typically by HPLC), ideally from a third-party lab rather than only the seller. Independent vendor raters that this site aggregates can help cross-check who actually publishes testing versus who only claims it.
Nothing here is medical advice.