Your mitochondria make a peptide that centenarians have more of and Alzheimer's patients have less of. Nobody has tested what happens when you inject it. The gap between that correlation and a cure is the whole story.

Humanin is the peptide that makes you rethink what mitochondria are for. For decades, biology treated mitochondria as the cell's power plant—organelles whose job was to produce ATP and nothing else. Then in 2001, a team led by Ikuo Nishimoto screened genetic material from the brain of an Alzheimer's patient. They found a 24-amino acid peptide encoded in mitochondrial DNA that could prevent brain cells from dying. It was the first mitochondrial-derived peptide ever characterized. It meant that mitochondria were not just making energy. They were making signaling molecules. An entire class of peptides—the MDPs—was discovered in its wake.

The human story is tantalizing. Nir Barzilai's group at Albert Einstein College of Medicine studied over 600 centenarian families. They found that people who live past 100 have unusually high circulating humanin levels. Their children inherit that tendency. Meanwhile, humanin declines with age across every species tested—humans, mice, primates. The pattern is consistent: more humanin correlates with longer life and less disease. Less humanin correlates with aging and neurodegeneration.

The animal data deepens the story. In mouse models of Alzheimer's disease, a potent analog called HNG reduced amyloid-beta levels and improved cognition. In aged mice, 14 months of HNG injections reduced cardiac fibrosis. In C. elegans, overexpressing humanin extended lifespan. The preclinical portfolio is broad and consistent across tissues.

But there is a shadow over the entire program. Humanin's mechanism is anti-apoptotic—it stops cells from dying. In healthy neurons, that is protective. In cancer cells, it is the opposite. A 2020 study in Nature Scientific Reports found that humanin helps triple-negative breast cancer cells survive. It does this by the exact same mechanism that protects brain cells. The peptide does not distinguish between cells worth saving and cells that should die. This is not a theoretical concern. It is published data in a serious cancer model. Any honest assessment of humanin must put this finding front and center, and this article does.

Quick Facts: Humanin at a Glance

What Is Humanin?

Pronunciation: HYOO-muh-nin

Your mitochondria have their own genome—a small circular chromosome inherited exclusively from your mother, containing 37 genes. For decades, biologists assumed those genes only encoded proteins needed for the mitochondrial machinery: energy production, and nothing else. In 2001, that assumption collapsed.

A research team searching for neuroprotective factors in Alzheimer's disease screened a cDNA library built from the preserved brain tissue of an AD patient. They found a short open reading frame in the 16S ribosomal RNA gene of mitochondrial DNA that encoded a 24-amino acid peptide capable of preventing neuronal cell death across multiple Alzheimer's-related insults. They named it humanin. It was the first mitochondrial-derived peptide—the first evidence that mitochondria were not just energy factories but active producers of signaling molecules that circulate in the blood and communicate with the rest of the body.

Humanin exists in two forms: a 21-amino acid version produced in the mitochondrial matrix and a 24-amino acid version produced in the cytosol. Both are biologically active. The peptide is secreted into circulation, where it can be measured in blood and declines steadily with age.

Origins and Discovery

The discovery of humanin was an accident of geography. Ikuo Nishimoto's team at Keio University School of Medicine in Tokyo chose to screen the occipital lobe of an Alzheimer's patient—a brain region relatively spared by the disease—reasoning that surviving tissue might contain protective factors that damaged regions had lost.

They built a cDNA library from the occipital lobe mRNA and screened it against multiple Alzheimer's-related death signals: neurotoxic amyloid-beta peptides, and mutations in familial AD genes APP, presenilin-1, and presenilin-2. One clone protected against all of them. When they sequenced it, they found it did not encode any known protein. Instead, it encoded a short peptide from mitochondrial DNA—a 24-amino acid molecule whose rescue activity depended entirely on its primary structure.

The paper was published in PNAS in May 2001 (PMID 11371646). It was a paradigm shift: mitochondrial DNA was making bioactive peptides, not just mitochondrial machinery. Within a few years, additional mitochondrial-derived peptides were discovered—MOTS-c in 2015, the SHLPs (Small Humanin-Like Peptides) soon after—establishing an entirely new class of biological molecules.

Nir Barzilai's group at Albert Einstein College of Medicine in New York then added the longevity dimension. Studying over 600 families of centenarians through the Longevity Genes Project, they found that people who live past 100 have significantly higher circulating humanin levels, and their children inherit that advantage. This moved humanin from a neuroprotection curiosity to a potential longevity biomarker.

Mechanism of Action

Anti-Apoptotic Pathway (BAX Inhibition)

Humanin's primary mechanism is anti-apoptotic—it prevents cells from undergoing programmed cell death. It does this by interfering with BAX, a pro-apoptotic protein that normally translocates to the mitochondrial outer membrane during cellular stress. When BAX reaches the mitochondria, it triggers cytochrome c release, which activates the caspase cascade and kills the cell. Humanin blocks BAX translocation, preventing this chain reaction. The cell survives.

This is the mechanism that makes humanin both promising and dangerous. In neurons facing amyloid-beta toxicity, blocking BAX is protective. In cancer cells facing chemotherapy, blocking BAX is tumor-promoting. The mechanism does not distinguish between the two.

STAT3 Signaling

Humanin activates STAT3 (Signal Transducer and Activator of Transcription 3) through a trimeric receptor complex consisting of CNTFR, WSX-1, and gp130. STAT3 phosphorylation is required for humanin's neuroprotective effect—blocking STAT3 abolishes the rescue activity. STAT3 activation is also the pathway through which humanin improves insulin sensitivity, acting centrally in the hypothalamus in a mechanism analogous to leptin signaling.

IGFBP-3 Binding

Humanin binds directly to insulin-like growth factor binding protein-3 (IGFBP-3), interfering with importin-beta-1 binding. This interaction inhibits IGFBP-3-dependent cell death—another layer of the cytoprotective mechanism.

Insulin Sensitization

Humanin improves both hepatic and peripheral insulin sensitivity. It increases glucose-stimulated insulin secretion in pancreatic beta cells in a dose-dependent manner. A single treatment with HNG significantly lowered blood glucose in Zucker diabetic fatty rats. Humanin also increases beta cell survival, potentially delaying diabetes onset.

Age-Dependent Signaling Differences

A critical finding for the longevity story: humanin activates different signaling pathways depending on the age of the organism. In old mouse hippocampal tissue, humanin injection increased phosphorylation of AKT and ERK1/2. In young mouse hippocampal tissue, the same injection had no effect. The proposed explanation: young animals have higher endogenous humanin levels that may already saturate these pathways, while aged animals with depleted humanin have room for exogenous supplementation to work. If confirmed, this would mean humanin is most effective precisely when it is most needed—in aging.

Key Research

The Discovery Paper (2001)

Hashimoto, Nishimoto et al., PNAS, 2001 (PMID 11371646). Functional screen of cDNA library from Alzheimer's patient occipital lobe. Found humanin in the 16S rRNA gene of mitochondrial DNA. Demonstrated protection against amyloid-beta, APP mutations, PS1 mutations, and PS2 mutations. The peptide was secreted into culture medium, confirming it is a circulating factor.

The Centenarian Studies (Barzilai Group)

Albert Einstein College of Medicine, Longevity Genes Project. 600+ centenarian families. Centenarians and their offspring have significantly higher circulating humanin levels. The trait is heritable. Higher humanin associated with lower incidence of type 2 diabetes and Alzheimer's. This is observational—correlation, not causation.

The Alzheimer's Mouse Model (HNG Treatment)

Triple-transgenic AD mice (APPswe/tauP310L/PS-1M146V). Three months of intranasal S14G-humanin (HNG) treatment ameliorated cognitive impairment. ELISA and immunohistochemistry showed markedly lower amyloid-beta levels in treated versus vehicle mice. Both male and female mice showed protection. Additional rat studies showed HNG increased dendritic branch density, dendritic spine density, and reduced tau hyperphosphorylation.

The Cardiac Fibrosis Study (2018)

PMID 30004252. American Journal of Physiology—Heart and Circulatory Physiology. Female C57BL/6N mice, aged 18 months. 14 months of intraperitoneal HNG injections (4 mg/kg, twice weekly). Results: increased cardiomyocyte-to-fibroblast ratio, reduced cardiac fibroblast proliferation, attenuated TGF-beta-1, FGF-2, and MMP-2 expression. Inhibited myocardial apoptosis. Upregulated Akt/GSK-3beta pathway.

The C. elegans Lifespan Extension

Overexpression of humanin in C. elegans is sufficient to increase lifespan. The extension is daf-16/FOXO-dependent, linking humanin to the same conserved longevity pathway that mediates insulin/IGF-1 signaling effects on lifespan across species.

The Cognitive Aging Study (2018)

Nature Scientific Reports (PMID 30242290). Cross-species analysis. In mice, humanin prevented age-related cognitive decline. In humans, circulating humanin levels were associated with improved "cognitive age." SNP rs2854128 in the humanin-coding region was associated with 14% decrease in circulating humanin and accelerated cognitive aging.

The Age-Dependent Signaling Study

Published in Oncotarget. Humanin injection increased AKT and ERK1/2 phosphorylation in hippocampal tissue from old mice but not young mice. Proposed mechanisms: blood-brain barrier permeability differences and endogenous humanin saturation in younger animals.

The Tumor Promotion Problem

This section must be read alongside every positive finding in this article. It is not a footnote. It is the central safety concern of humanin biology.

The study: "Humanin Promotes Tumor Progression in Experimental Triple Negative Breast Cancer." Nature Scientific Reports, 2020 (PMID 32444831).

What it found: - Exogenous humanin protected triple-negative breast cancer (TNBC) cells from apoptotic stimuli—the same mechanism that protects neurons - When humanin was silenced using shRNA, TNBC cell viability decreased and chemosensitivity increased - Humanin and its receptors are expressed in human breast cancer tissue (confirmed by TCGA analysis and immunohistochemistry) - Humanin was upregulated in TNBC biopsies compared to normal mammary tissue

Humanin's value proposition is that it stops cells from dying. In neurons facing amyloid-beta, that is therapeutic. In cancer cells facing chemotherapy, that is the definition of drug resistance. The mechanism is the same: BAX inhibition, cytochrome c retention, caspase suppression. Humanin does not ask whether a cell deserves to live. It stops the death machinery regardless.

This is not a theoretical edge case. TNBC is the most aggressive subtype of breast cancer, with the fewest treatment options. Any compound that promotes TNBC cell survival is not just an abstract concern—it is a direct threat to patients undergoing chemotherapy.

Additional cancer data: Humanin also inhibits apoptosis in pituitary tumor cells via NF-kappa-B activation. The pattern is consistent: humanin's cytoprotective mechanism is not tissue-selective. It protects whatever cells it reaches.

Peptidings puts this section prominently—not buried in a safety sidebar—because the tumor promotion finding changes the risk calculus of humanin supplementation. An aging person supplementing humanin to protect their brain is also potentially protecting any undiagnosed cancer cells in their body. That trade-off must be stated plainly.

Claims vs. Evidence

We currently don’t have any vetted partners for this compound. Check back soon.

The Human Evidence Landscape

No human being has received exogenous humanin or HNG in a clinical trial. No Phase I safety study, no pharmacokinetic study, no dose-finding study. After 25 years in the published literature, humanin has zero interventional human evidence. What it has—and what makes the gap so frustrating—is an unusually rich observational human dataset.

The Centenarian Studies (Barzilai Group, Albert Einstein College of Medicine)

Nir Barzilai's Longevity Genes Project studied over 600 centenarian families. The findings are consistent: people who live past 100 have significantly higher circulating humanin levels than age-matched controls. Their offspring inherit this elevated level, suggesting a heritable biological advantage. Higher humanin was associated with lower incidence of type 2 diabetes and Alzheimer's disease. This is the strongest human data for humanin—and it is entirely observational. Correlation with longevity, however robust, does not establish that supplementing humanin would extend life. Centenarians differ from the general population in hundreds of ways.

The Cognitive Aging Study (2018, PMID 30242290)

A cross-species analysis published in Nature Scientific Reports. In the human arm, circulating humanin levels were associated with improved "cognitive age." A SNP in the humanin-coding region (rs2854128) was associated with 14% decrease in circulating humanin and accelerated cognitive aging. This provides genetic evidence linking natural humanin variation to cognitive function—a different type of evidence from the centenarian observations, and supportive of the same direction. But it is still observational.

The Age-Related Decline Data

Humanin declines with age across humans, mice, and primates. This is well-replicated and consistent. Whether this decline is maladaptive (contributing to aging) or adaptive (reducing cancer risk as the anti-apoptotic mechanism becomes more dangerous in an aging body) is unknown—and the answer matters enormously for whether supplementation is wise.

What Does Not Exist

No pharmaceutical company is developing humanin. No IND has been filed. No clinical trial is registered. The absence of commercial development despite 25 years of consistently positive preclinical data is notable. Possible reasons: peptide stability challenges, the cancer safety signal (TNBC study), patent complexity arising from mitochondrial genome origin, and industry preference for small molecules.

What Would Need to Happen

A Phase I safety and PK study in humans—likely in a disease population where the risk-benefit calculation favors intervention (Alzheimer's patients, for whom the cancer risk may be acceptable given the severity of the disease). The cancer safety signal would need to be characterized: does exogenous humanin promote tumor growth in humans, or is the TNBC cell culture finding not clinically relevant? This question must be answered before large-scale supplementation could be responsibly recommended.

Safety, Risks, and Limitations

The Cancer Concern (Central Safety Issue)

The 2020 TNBC study (PMID 32444831) is not an outlier finding. It is the logical consequence of humanin's mechanism. Any compound that inhibits apoptosis broadly—without cell-type selectivity—will protect cancer cells alongside healthy cells. This is different from FOXO4-DRI (which exploits senescence-specific FOXO4 elevation for selectivity) or SS-31 (which targets cardiolipin, a mitochondrial component, not death pathways). Humanin's mechanism has no inherent selectivity. It stops BAX from killing cells, period.

For self-experimenters: if you have an undiagnosed cancer (which is more common with age), supplementing humanin may theoretically protect those cancer cells from your immune system's natural surveillance. This is not proven in humans. It is a legitimate risk extrapolation from published cancer biology.

No Human Safety Data

No human has received exogenous humanin in a clinical trial. Pharmacokinetics (absorption, distribution, metabolism, elimination) are unknown for every route in humans. Maximum tolerated dose is unknown. Drug interactions are unknown. Long-term effects are unknown.

Dose Translation Uncertainty

The HNG analog is 1,000-fold more potent than native humanin. This creates a significant dosing risk: confusing native humanin doses with HNG doses could result in massive over- or under-dosing. Community protocols must specify which form they are using.

Endogenous Does Not Mean Safe to Supplement

Humanin is naturally present in blood. So are cortisol, insulin, and estrogen. Supplementing any endogenous molecule shifts the body's feedback systems. The assumption that "your body makes it, so more must be better" has no evidence base for humanin. The age-related decline may be adaptive (reducing cancer risk as accumulating mutations make anti-apoptotic signaling more dangerous) or maladaptive (contributing to neurodegeneration). We do not know which.

Legal and Regulatory Status

Humanin is not FDA-approved for any indication. No IND application has been filed. No pharmaceutical company is developing humanin therapeutically. No clinical trial is registered.

Synthetic humanin and HNG are available from research peptide vendors as research chemicals. They are not explicitly banned from compounding pharmacies.

Humanin is not listed on the WADA 2025–2026 Prohibited List.

The absence of commercial development despite 25 years of positive preclinical data is notable. Possible reasons include: peptide stability challenges, blood-brain barrier penetration limitations, patent complexity (mitochondrial genome origin), the TNBC safety signal, and pharmaceutical industry preference for small molecules over peptides.

Research Protocols and Formulation Considerations

• Cardiac fibrosis study (PMID 30004252): HNG 4 mg/kg IP, twice weekly, 14 months (aged mice)
• Alzheimer's model: HNG intranasal delivery, 3 months (triple-transgenic AD mice)
• Diabetes model: Single HNG injection, dose-response in Zucker diabetic fatty rats

Storage: Standard peptide storage: 2–8°C (35–46°F) for reconstituted solutions, −20°C (−4°F) for lyophilized powder.

Critical note: Published protocols use HNG (1,000x more potent than native humanin). Dose comparisons between HNG and native humanin are not straightforward.

Dosing in Published Research

Published research doses (all HNG analog in animals): - Cardiac fibrosis: 4 mg/kg IP, twice weekly (mice) - Neuroprotection: intranasal, variable doses (mice) - Metabolic: single injection, dose-response (rats)

There is no established human dose for humanin or HNG. No dose-finding study has been conducted in humans for either form.

Detailed Research Dosing Data (from published studies)

Humanin dosing in published preclinical and observational studies varies widely, reflecting different route, species, and therapeutic goals. Study Type / Model Dose Route Duration / Frequency Notes Cell Culture (neurons, fibroblasts) 10–1000 nM Direct addition to culture medium Acute (minutes to hours) or chronic (24–72 h) Doses are concentration-dependent; physiological relevance unclear. No dose-response plateau typically observed. Transgenic AD Mice (APP/PS1, 5XFAD) 0.5–2 mg/kg Intraperitoneal injection (IP) or intracerebroventricular infusion (ICV) 3–12 months; often chronic (twice-weekly IP injections) IP dose is ~50–100-fold higher than reported endogenous plasma levels; ICV doses much lower due to direct CNS delivery. Long-term safety in these models not well-characterized. Acute Stroke (MCAO) 0.5–1 mg/kg IP injection or intra-arterial infusion Single dose given immediately before, during, or up to 24 h after ischemia Cardioprotective effect seen with single doses; no cumulative data on repeated dosing. Metabolic Studies (obesity, diabetes) 0.5–1.5 mg/kg IP injection 8–16 weeks; 1–3 times per week Doses similar to neurodegenerative models. Insulin sensitivity improves; weight loss modest and inconsistent. Human Observational Studies (centenarians, healthy volunteers) N/A (measurement of endogenous levels) None; blood draw Cross-sectional or longitudinal; plasma humanin measured at single or multiple timepoints Endogenous plasma humanin typically 0.5–3 ng/mL, varies with assay. No exogenous dosing studies in humans. Key Observation: Preclinical doses (0.5–2 mg/kg in rodents) translate to approximately 35–140 mg for a 70 kg human—orders of magnitude higher than the endogenous plasma concentration. Whether such doses are necessary or optimal remains untested in humans.

Dosing in Self-Experimentation Communities

Community use of humanin/HNG is limited but growing. Available reports suggest:

• Typical dose range: 5–10 mg per week, divided into 2–3 subcutaneous injections
• Typical protocol: 2–3 injections per week (e.g., Monday, Wednesday, Friday), cycled 2–4 weeks on, 2–4 weeks off
• Alternative protocol: 1 mg twice daily for 15 days

Critical context: These doses are community consensus, not evidence-based. No dose-response relationship has been established in humans. The 1,000-fold potency difference between native humanin and HNG means that failing to specify which form is being used introduces massive dosing uncertainty. Community protocols rarely specify.

No systematic adverse event reporting exists. The small user base means rare adverse events—including the theoretical cancer risk—would be invisible.

Combination Stacks

Research into Humanin combination protocols is limited. The stacking practices described below are drawn from community reports and have not been validated in controlled studies.

If you are considering combining Humanin with other compounds, consult a qualified healthcare provider. Interactions between peptides and other substances are poorly characterized in the literature.

Frequently Asked Questions

Summary of Key Findings

Humanin is the compound that teaches you why biology is harder than headlines suggest. The centenarian connection is real. The age-related decline is documented. The neuroprotection data is consistent. And the cancer concern is serious enough to rewrite the entire risk calculation. Everything about humanin looks promising until you read the TNBC paper, and then everything looks complicated.

The centenarian data is the strongest human evidence, and it is observational. Barzilai's group found that people who live past 100 have higher circulating humanin, and the trait is heritable. This is the most robust human association in the peptide longevity space. But association is not causation. Centenarians also have different telomere dynamics, different inflammatory profiles, and different genetic backgrounds. Whether humanin drives longevity or rides alongside it is unknown.

The neuroprotection portfolio is broad and consistent. In vitro protection against amyloid-beta. In vivo reduction of amyloid-beta and cognitive improvement in triple-transgenic AD mice. Dendritic spine preservation. Tau hyperphosphorylation reduction. Intranasal delivery that crosses the blood-brain barrier in mice. The data spans multiple models, multiple endpoints, and multiple labs. If you ignore the cancer concern, the neuroprotection story is one of the strongest preclinical packages on this site.

The cardiac data is compelling in a single long-term study. Fourteen months of HNG injections reduced cardiac fibrosis in aged mice. This is the kind of long-duration, age-relevant study that most peptides lack. But it is one study.

The age-dependent signaling finding changes the therapeutic calculus. Humanin activates AKT and ERK1/2 in old hippocampal tissue but not young. If confirmed, this means humanin supplementation is most effective in the population that needs it most—the aged. This is the most encouraging mechanistic finding for the longevity application.

The cancer concern is not a footnote. It is the central safety problem. The 2020 TNBC study showed that humanin promotes tumor progression by the exact same mechanism that protects neurons. Humanin is upregulated in TNBC biopsies. Silencing humanin increased cancer cell death and chemosensitivity. This is not a theoretical edge case. It is the logical consequence of a non-selective anti-apoptotic mechanism. Any honest assessment of humanin must weigh the neuroprotection data against the tumor promotion data, and right now, nobody knows where the balance falls.

Zero clinical development after 25 years is itself a signal. Humanin was discovered in 2001. It has broad preclinical support, a compelling centenarian association, and a first-in-class mechanism. No company has filed an IND. No clinical trial has been registered. The reasons are partly practical (peptide stability, BBB penetration, patent complexity) and partly substantive (the cancer concern). Twenty-five years without a clinical trial is not just a funding problem. It suggests structural barriers that the community should understand.

The HNG potency difference creates dosing chaos. HNG is 1,000-fold more potent than native humanin. Most published studies use HNG. Most community protocols do not specify which form they are using. This is not a minor detail—it is a 1,000-fold dosing variable.

Verdict Recapitulation

Humanin earns Tier 4 because no interventional human data exists. The centenarian studies are observational—they measure blood levels, not therapeutic outcomes. The animal data is strong but entirely preclinical. Twenty-five years of research have not produced a single human dosing study.

The compound earns "Eyes Open" rather than "Thin Ice" because the preclinical portfolio is genuinely impressive. The centenarian association is the strongest human correlation for any Tier 4 compound on the site. The neuroprotection data spans multiple models and labs. The cardiac and metabolic findings add breadth. The mechanism is well-characterized and independently replicated.

But "Eyes Open" also means the cancer concern is real and prominent. The 2020 TNBC study is not a fringe finding—it is a logical extension of humanin's mechanism published in a legitimate journal. A compound that stops cell death broadly will protect cancer cells alongside healthy cells. This is the fundamental trade-off that 25 years of research have identified but not resolved.

For readers considering humanin: you are looking at a compound with one of the most compelling longevity correlations in biology (centenarian levels), one of the strongest neuroprotection portfolios in preclinical research, and a cancer safety concern that has prevented any company from advancing it to human trials for a quarter century. All three facts are true simultaneously. Your eyes should be open to all of them.

For readers considering Humanin, the evidence above represents the current state of knowledge. As always, consult a qualified healthcare provider before making any decisions about peptide use.

Where to Source Humanin

Further Reading and Resources

If you want to go deeper on Humanin, the evidence landscape for longevity & anti-aging peptides, or the methodology behind how we evaluate this research, these are the places worth your time.

ON PEPTIDINGS

• Longevity & Anti-Aging Research Hub — Overview of all compounds in this cluster
• Reconstitution Guide — How to properly prepare injectable peptides
• Storage and Handling Guide — Proper storage to maintain peptide stability
• About Peptidings — Our editorial methodology and evidence framework

EXTERNAL RESOURCES

• PubMed: Humanin — All indexed publications
• ClinicalTrials.gov — Active and completed trials

Selected References and Key Studies

1. Hashimoto, Nishimoto et al., "A rescue factor abolishing neuronal cell death by a wide spectrum of familial Alzheimer's disease genes and Aβ." PNAS, 2001. —Discovery paper PubMed
2. Humanin Promotes Tumor Progression in Experimental Triple Negative Breast Cancer. Nature Scientific Reports, 2020. —Cancer safety concern PubMed
3. Chronic treatment with the mitochondrial peptide humanin prevents age-related myocardial fibrosis in mice. American Journal of Physiology—Heart and Circulatory Physiology, 2018 PubMed
4. Humanin Prevents Age-Related Cognitive Decline in Mice and is Associated with Improved Cognitive Age in Humans. Nature Scientific Reports, 2018 PubMed
5. Neuroprotective Action of Humanin and Humanin Analogues: Research Findings and Perspectives. PMC, 2023. —Comprehensive review PubMed
6. The mitochondrial derived peptide humanin is a regulator of lifespan and healthspan. Aging, 2020. *PMC 7343442*—C. elegans and mouse healthspan data
7. Humanin and Its Pathophysiological Roles in Aging: A Systematic Review. PMC, 2023. *PMC 10135985*
8. The mitochondrial-derived peptide humanin activates the ERK1/2, AKT, and STAT3 signaling pathways and has age-dependent signaling differences in the hippocampus. Oncotarget, 2016
9. Barzilai group—Longevity Genes Project centenarian humanin association studies (multiple publications)
10. Lee et al., "The mitochondrial-derived peptide MOTS-c promotes metabolic homeostasis and reduces obesity and insulin resistance." Cell Metabolism, 2015. —MOTS-c discovery, establishing the MDP family PubMed
11. Humanin and diabetes mellitus: A review of in vitro and in vivo studies. PMC, 2022. *PMC 8984571*
12. S14G-humanin cardioprotective effects against chronic adrenergic and pressure overload-induced heart failure. (Additional cardiac data confirming mechanism.)
13. Apoptosis-related disease treatment review including humanin mechanisms PubMed
14. Potent humanin analog effects on glucose metabolism PubMed
15. Alzheimer's Drug Discovery Foundation—Humanin and Humanin Analogs assessment (https://www.alzdiscovery.org/)

---