Humanin is a 24-amino-acid peptide encoded by a short open reading frame inside the mitochondrial 16S rRNA gene, not by a nuclear hormone gene. Hashimoto’s group cloned it in 2001 from surviving occipital-lobe neurons of an Alzheimer’s disease brain, where it blocked death from familial Alzheimer’s genes and amyloid-beta. It is the first mitochondria-derived peptide (MDP) described, it is not MOTS-c, and there is no FDA-approved humanin drug.

TL;DR

  • Humanin is a 24-mer encoded in mitochondrial 16S rRNA (MT-RNR2). Sequence: MAPRGFSCLLLLTSEIDLPVKRRA.
  • Hashimoto’s 2001 screen found a cDNA that rescued neurons from familial Alzheimer’s genes and amyloid-beta, not from unrelated death signals.
  • Inside the cell it binds Bax. Outside, it uses a CNTFR / WSX-1 / gp130 complex and formyl peptide receptor 2 (FPR2).
  • Circulating levels fall with age. Children of centenarians run higher. Naked mole rats keep levels steadier than mice.
  • Worm overexpression extended lifespan. Twice-weekly HNG in middle-aged mice improved healthspan measures, not lifespan. A catalog vial is not that experiment.

Educational only. Not medical advice, not a protocol, and not a reason to treat a research vial as a longevity drug.

What is humanin, and where is it encoded?

Most peptides on this site are nuclear genes or lab analogs. Humanin is different. It is translated from a small open reading frame in the mitochondrial 16S rRNA region, the same stretch of mtDNA that later yielded the small humanin-like peptides (SHLPs).

The peptide is 24 residues: MAPRGFSCLLLLTSEIDLPVKRRA, molecular weight about 2,656 Da (Muzumdar et al., PLoS One, 2009). Hashimoto, Niikura, Tajima, and colleagues pulled the cDNA from a functional expression screen of occipital cortex that had survived Alzheimer’s pathology. The polypeptide abolished neuronal death caused by several familial Alzheimer’s disease (FAD) genes and by amyloid-beta, without blocking death from a polyglutamine construct or superoxide dismutase-1 mutants (Hashimoto et al., PNAS, 2001).

That specificity is the origin story. It is not “a general anti-aging hormone.” It was a rescue factor against a defined set of Alzheimer’s-relevant insults.

Endogenous peptide shows up in brain, testis, colon, plasma, and cerebrospinal fluid. The rat homolog is often called rattin. Nuclear mitochondrial insertions (NUMTs) can carry humanin-like sequences, which is one reason circulating immunoassays are messier than an HPLC peak, so start with Peptides 101 for the chemistry.

How does humanin actually signal?

Two layers, and forums usually quote only one.

Intracellular. Guo, Reed, and colleagues showed that humanin binds the pro-apoptotic Bcl-2 family protein Bax. Bound Bax does not undergo the conformational change that lets it dock on mitochondria, so cytochrome c stays put. Knocking down endogenous humanin sensitized cells to Bax-dependent death (Guo et al., Nature, 2003).

Extracellular. Hashimoto’s 2009 loss-of-function work made gp130 (glycoprotein 130, the shared IL-6 family signal transducer IL6ST) essential for neuroprotection, along with CNTFR-alpha (ciliary neurotrophic factor receptor) and WSX-1 (IL-27 receptor alpha). Downstream, labs report JAK2/STAT3, plus AKT and ERK1/2 (Hashimoto et al., Molecular Biology of the Cell, 2009; Miller et al., JCI, 2022). A second, G-protein-coupled route is FPR2, formerly FPRL1.

HNG is the analog most papers actually inject. A glycine-for-serine swap at position 14 (S14G) raises potency about 1,000-fold in the original rescue assays. That is a research tool, not a licensed product.

Did the insulin and longevity studies measure what forums claim?

The metabolic data are rodent clamps, not obesity-drug trials. The longevity data are worms, a mouse healthspan experiment, and human associations.

Muzumdar, Barzilai, Cohen, and colleagues infused humanin into the third ventricle of rats and ran insulin clamps. Whole-body insulin sensitivity rose, hypothalamic STAT3 was required, and intravenous analogs reproduced the effect. Hepatic benefits collapsed when hypothalamic STAT3 was blocked, the peptide did not act on isolated hepatocytes, and circulating humanin was lower in older mice and older humans (Muzumdar et al., 2009).

Yen, Cohen, and collaborators then asked the lifespan question directly (Aging, 2020):

Study Setting Main finding
Hashimoto 2001 Neuronal cells, FAD genes and amyloid-beta HN cDNA and peptide blocked that death, not unrelated insults
Guo 2003 Cultured cells and isolated mitochondria HN binds Bax and stops mitochondrial translocation
Muzumdar 2009 Rat insulin clamps, human plasma Central and analog HN improved insulin action; circulating HN fell with age
Yen 2020 C. elegans, mice, humans Worm overexpression extended lifespan (FOXO-dependent); 18 children of centenarians had higher circulating HN than 19 controls; middle-aged mouse HNG improved healthspan, not lifespan

Mice lose a large fraction of circulating humanin over the first 18 months. Naked mole rats, which age slowly, do not show that cliff. Alzheimer’s cerebrospinal fluid ran lower than control CSF in a small sample, and the mtDNA variant rs2854128 (m.2706A>G) inside the humanin open reading frame has been tied to lower circulating peptide and faster cognitive aging in one cohort (Yen et al., Scientific Reports, 2018, reviewed in Miller et al., 2022).

Read the mouse result the way it was written. Twice-weekly HNG in 18-month-old females improved memory and metabolic measures, but lifespan did not rise. The authors pointed to a plasma half-life of about 20 minutes. Worm overexpression, which keeps the peptide present, did extend life and required daf-16/FOXO.

Association in centenarian offspring is not a dosing protocol. It is a circulating marker that tracks a family-longevity phenotype.

How is a research vial different from the endogenous peptide?

Endogenous humanin is a mitochondrial microprotein you already make, at levels that change with age, mtDNA copy number, and disease context. SS-31 / elamipretide is the rare mitochondrial peptide that later became an approved drug (Forzinity, Barth syndrome). Humanin has not.

Catalog humanin or HNG is a synthetic research chemical. Identity, salt, and endotoxin are COA questions, not label promises. There is no FDA-approved humanin drug, so licensed peptides live in pharmacy and telehealth channels that the provider directory and price index compare, with handling education on the reconstitution guide and the calculator.

If the goal is mitochondrial peptide literacy, read humanin next to MOTS-c (12S rRNA, exercise-responsive metabolic peptide) and the SHLP family from the same 16S locus. They are related by genome geography, not interchangeable vials.

Bottom line: Humanin is a real mitochondrial 24-mer with a 2001 Alzheimer’s-cell discovery, a Bax and gp130 mechanism, and longevity data that are stronger in worms than in mice. A catalog vial is not a lifespan drug.

Educational only. Not medical advice. Not a recommendation to use humanin, HNG, or any research-grade mitochondria-derived peptide.