What humanin is
Humanin is a 24-amino-acid peptide with the sequence MAPRGFSCLLLLTSEIDLPVKRRA. Unlike most bioactive peptides, which are translated from nuclear genes, humanin originates from a small open reading frame within the 16S ribosomal RNA gene of mitochondrial DNA (mtDNA). The peptide is secreted into circulation and is detectable in human plasma at picomolar concentrations.
Humanin is the founding member of a family now called mitochondrial-derived peptides (MDPs). This family also includes MOTS-c and the small humanin-like peptides (SHLPs 1 through 6), all encoded in the 16S or 12S rRNA regions of the mitochondrial genome. The identification of humanin in 2001 was the first evidence that the mitochondrial genome encodes functional signaling peptides beyond its role in oxidative phosphorylation.
The full-length peptide is 24 residues and has a molecular weight of approximately 2.8 kDa. A potent synthetic analogue, HNG, carries a single glycine substitution at position 14 in place of serine. This change confers roughly 1000-fold greater potency in cell-based assays; most research published after 2005 uses HNG or other analogues rather than the native sequence.
Discovery
Hashimoto and colleagues at Osaka Bioscience Institute identified humanin in 2001 by screening a cDNA library built from postmortem brain tissue of an Alzheimer's disease patient. The library was constructed from the occipital lobe, a region without significant amyloid pathology in that patient. Functional expression screening against neuronal cell death assays identified a short open reading frame whose translated product blocked cell death caused by mutant presenilin-1, mutant presenilin-2, mutant amyloid precursor protein, and amyloid-beta peptides.
The results appeared in Proceedings of the National Academy of Sciences (Hashimoto Y et al., PNAS 2001;98:6336-6341). The same reading frame has since been confirmed in all mammals examined, and homologous sequences appear in zebrafish, suggesting conservation over several hundred million years of vertebrate evolution. The discovery prompted the renaming of the mitochondrial genome region as a functional peptide-coding locus rather than purely structural RNA.
Molecular mechanisms of cytoprotection
Humanin prevents programmed cell death through two intracellular pathways and one membrane receptor pathway. These are mechanistically distinct and have been studied separately in cell-free preparations and intact cells.
The first intracellular pathway involves direct binding to Bax, a pro-apoptotic member of the Bcl-2 family. Guo et al. published mechanistic data in Nature (Guo B et al., Nature 2003;423:456-461) showing that humanin physically interacts with Bax and blocks its translocation from cytosol to mitochondrial outer membranes. Preventing this translocation stops cytochrome c release and the downstream caspase cascade. Humanin also blocks Bax association with isolated mitochondria in cell-free preparations, confirming that the interaction is direct rather than cell-signaling-mediated.
The second intracellular pathway involves IGFBP-3 (insulin-like growth factor binding protein-3), itself a pro-apoptotic signaling molecule. Humanin binds IGFBP-3 directly and reduces its apoptotic activity. A single amino acid substitution at position 6 (phenylalanine to alanine) abolishes this binding without affecting the Bax interaction, which established that the two intracellular mechanisms are structurally separable.
The membrane receptor pathway operates through a trimeric complex of WSX-1, CNTF receptor alpha, and gp130. Humanin binding to this complex activates three parallel intracellular cascades: PI3K-AKT, MEK-ERK1/2, and JAK-STAT3. Studies in hippocampal tissue have measured increased phosphorylation of all three effectors following humanin application, with the relative contribution of each branch varying by cell type and by the age of the tissue.
Humanin research findings by organ system
Neuroprotection
The original 2001 paper established protection against Alzheimer's-relevant insults as humanin's defining activity. The compound blocked cell death from at least six distinct familial Alzheimer's disease gene mutations and from exogenous amyloid-beta peptides in neuronal cultures, suggesting broad coverage of AD-relevant cell death pathways rather than a single pathway-specific interaction. Subsequent work extended neuroprotective activity to oxidative stress and excitotoxic damage in multiple neuronal cell types.
Beta cell survival and metabolic effects
Humanin protects pancreatic beta cells from apoptosis through STAT3-dependent signaling and separately improves insulin sensitivity in metabolic models. In an in vivo experiment using 9-week-old female NOD mice (a standard type 1 autoimmune diabetes model), daily intraperitoneal humanin injections at 0.7 mg/kg for 6 weeks (n=12 per group) improved glucose tolerance, reduced insulitis severity in pancreatic tissue, and delayed diabetes onset relative to saline controls (PMC2932671). A STAT3 inhibitor abolished the protective effect when co-administered, confirming that STAT3 activation is required for the beta cell effect.
Gonadal protection during chemotherapy
Research from the USC Leonard Davis School of Gerontology showed that humanin and HNG protect male germ cells from apoptosis induced by cyclophosphamide and doxorubicin, both ex vivo in rat seminiferous tubule cultures and in vivo in the rat testis (PMC3982780). The same research group subsequently documented protection of female ovarian function from cyclophosphamide-induced damage. These findings opened a line of inquiry on humanin analogues as adjuncts during cancer chemotherapy to preserve reproductive function in patients undergoing gonadotoxic treatment regimens.
Lifespan and healthspan
Yen et al. published a cross-species study in Aging in 2020 (Yen K et al., Aging 2020;PMC7343442) showing that humanin overexpression in C. elegans extends lifespan in a daf-16/FOXO-dependent manner. Humanin transgenic mice showed increased protection against toxic insults. Twice-weekly treatment of middle-aged mice with HNG reduced circulating inflammatory markers and improved metabolic healthspan parameters. In humans from the same study, children of centenarians had measurably higher circulating humanin than age-matched controls drawn from the general population. Naked mole-rats, which live well beyond the expected lifespan for rodents of their body size and show negligible age-related disease, maintained stable humanin levels across their adult lifespan in contrast to the progressive age-related decline seen in conventional laboratory mice and rats.
Age-related decline and disease associations
Circulating humanin falls with age in blood samples from rodents and humans. In diseases associated with mitochondrial dysfunction, including Alzheimer's disease and MELAS (mitochondrial encephalomyopathy, lactic acidosis, and stroke-like episodes), humanin levels are reduced relative to healthy age-matched controls. Whether the decline is a driver of age-related pathology or a downstream marker of mitochondrial health remains an open question.
The lifespan data from transgenic mouse and C. elegans overexpression studies argue for a causal role in health maintenance rather than a purely passive biomarker. But causality has not been tested directly in humans. Longitudinal studies tracking humanin levels and health outcomes in aging cohorts are needed before the relationship can be described as anything stronger than association.
Relationship to MOTS-c and the broader MDP family
Humanin and MOTS-c are structurally and functionally distinct despite sharing a mitochondrial genomic origin. MOTS-c is 16 amino acids, encoded from the 12S rRNA gene, and acts primarily on skeletal muscle and adipose tissue to improve glucose metabolism, with notable exercise-mimetic effects in rodent studies. Humanin is 24 amino acids, encoded from the 16S rRNA gene, and its primary characterized activities involve cell death prevention across neural, pancreatic, gonadal, and retinal cell types.
The two peptides use different receptors and activate partly overlapping downstream signaling networks. Researchers working on the MDP family increasingly study humanin and MOTS-c together as complementary secreted signals from the same organelle, with distinct tissue targets and physiological roles. The SHLP family (SHLPs 1 through 6), encoded from the same 16S rRNA region as humanin, shows overlapping cytoprotective activities with different relative potency across subtypes; SHLP2 has the highest activity in published metabolic studies.
Research status and limitations
The body of humanin research is preclinical. Neuroprotection, beta cell survival, gonadal protection, and lifespan extension have each been demonstrated in cell culture and rodent models, but no human therapeutic trials for these indications have been completed and published as of mid-2026. The absence of human trial data is a substantial gap for a compound with over two decades of preclinical literature behind it.
Comparing results across the literature is complicated by the use of native humanin, HNG, and multiple other synthetic analogues in different studies. HNG is not pharmacokinetically equivalent to native humanin, and the 1000-fold potency difference in cell assays does not translate linearly to in vivo dose predictions. Standard research practice for reconstituted humanin peptides involves verifying concentration against the molecular weight of the specific form being used; the dosing calculator on this site covers concentration, volume, and unit calculations for peptide research setups.