Humanin

Humanin is encoded in the mitochondrial 16S rRNA gene and was found in 2001 by screening for factors that protected neurons from amyloid-beta toxicity. It has two arms of action. Extracellularly it signals through a receptor complex of gp130, IL-27R and CNTFR, engaging STAT3 and ERK survival pathways. Intracellularly it binds BAX and prevents its translocation to the mitochondrial membrane, blocking the intrinsic apoptosis cascade; it also binds IGFBP-3. Circulating humanin declines markedly with age in humans and is higher in the offspring of centenarians, which is the observation that put it on the longevity map. The analogue HNG, a single serine-to-glycine swap at position 14, is roughly a thousand times more potent and is what most research and most grey-market product actually is.

Targets: BAX, gp130/IL-27R/CNTFR receptor complex, STAT3, IGFBP-3

• · This range reflects what vendors and forums suggest, extrapolated loosely from rodent mg/kg work. There is no human dose-finding study for humanin or HNG, and the honest answer is that nobody knows the right dose.

Work out your exact syringe units →

• commonInjection-site irritation
• commonUnknown long-term profile— This is the honest entry: humanin is broadly anti-apoptotic, and no human has taken it long enough for anyone to know what that means over years.

• Active malignancy — a systemically administered anti-apoptotic signal is theoretically the wrong thing to give a tumour.
• Pregnancy and breastfeeding — no data.

• synergymots-c — Frequently stacked as the two best-known mitochondrial-derived peptides, though they act on quite different pathways.
• synergyshlp-2 — Same peptide family with overlapping metabolic effects; combining them is speculative.

• · No established markers. Fasting glucose and lipids are reasonable if you are using it for metabolic reasons.
• · Nothing about humanin gives you a usable feedback signal, which is worth knowing before you spend money on it.

Not approved anywhere; sold as a research chemical.

• Hashimoto et al. 2001, PNAS — identification of humanin as a factor suppressing neuronal death by Alzheimer's-related insults (preclinical)
• Yen et al., humanin levels across ageing and in offspring of long-lived individuals (other)
• Lee, Yen & Cohen, review of humanin and the mitochondrial-derived peptide family (review)

Humanin has two mechanistically separate arms, and conflating them is the commonest error people make when reading about it. The extracellular arm signals through a trimeric receptor complex assembled from gp130, IL-27 receptor alpha (WSX-1) and ciliary neurotrophic factor receptor alpha. That is a cytokine receptor architecture, and it engages exactly what you would expect: JAK/STAT3 and ERK survival signalling. It is the arm responsible for the systemic, endocrine-like effects — the insulin sensitisation, the cardioprotection, the effects at tissues distant from where the peptide was made. Humanin also engages formyl peptide receptor-like 1 (FPRL1/FPR2), which is how it competes with amyloid-beta at the neuronal surface and where the original Alzheimer's protection story comes from. The intracellular arm is completely different: humanin binds BAX directly and prevents its translocation from cytosol to mitochondrial outer membrane, which blocks the intrinsic apoptosis pathway before it can commit. It also binds tBid and IGFBP-3. So humanin is simultaneously a secreted cytokine-like signal and an intracellular apoptosis brake, and which arm dominates depends entirely on how you deliver it. The ageing relevance rests on two solid observations: circulating humanin declines with age across species, and it is markedly higher in the children of centenarians than in age-matched controls. A rare P3S variant is enriched in centenarians carrying APOE4 and appears protective against amyloid pathology. Both of those are association findings, and neither tells you what happens if you inject the peptide. The clinical honesty here is that humanin is a broad anti-apoptotic signal, and 'stop cells dying' is not unambiguously good — that is precisely the phenotype a tumour wants.

The first problem is that you cannot tell whether it worked. Humanin produces no reliable subjective effect and moves no marker you can easily track, so people run it for a month, feel nothing definite, and either quit or double the dose on no information. The second is the identity problem: 'humanin' on a vendor site may be native humanin, HNG, a C8A variant, or the nuclear-encoded MT-RNR2-like sequence, and these are not interchangeable. HNG is reported to be far more potent than native, so the same milligram dose is not the same pharmacological dose. Ask for the exact sequence, and if the vendor cannot give it to you, you are guessing. The third is handling: the free cysteine at position 8 means the peptide can dimerise, and dimerisation state affects activity, so a badly stored solution is not simply weaker, it is a different mixture. The fourth is the one that actually carries risk: humanin is a systemic anti-apoptotic and pro-survival signal with no human safety data over any duration, and the theoretical concern with an undetected cancer is not boilerplate — it follows directly from what the molecule does. Nobody has run this long enough in humans for anyone to know.

• gp130 / IL-27Rα (WSX-1) / CNTFRα trimeric receptor complex — Complex formation demonstrated; no consensus Kd for humanin at the assembled trimer that I could resolve.

  STAT3 and ERK activation — the cytoprotective and metabolic arm. This is a cytokine-family receptor, which is why humanin behaves like an endocrine signal rather than a local factor.

• BAX — Direct binding; classical in vitro result. Affinity constants vary by assay and are not consistently reported.

  Prevents BAX translocation to the mitochondrial outer membrane, blocking intrinsic apoptosis before mitochondrial outer membrane permeabilisation. This is the intracellular arm and it requires the peptide to be inside the cell.

• Formyl peptide receptor-like 1 (FPRL1 / FPR2) — Not resolved.

  Competition with amyloid-beta at the neuronal surface — the mechanism behind the original 2001 neuroprotection finding.

• IGFBP-3 — Direct binding reported.

  Modulates IGF-1 bioavailability, which links humanin into the insulin/IGF-1 longevity axis rather than acting purely as an apoptosis regulator.

• Autophagy induction (downstream, mechanism partly defined) — Not a receptor interaction.

  Humanin-induced autophagy was shown to be necessary for its effects on skeletal muscle function and lifespan extension in C. elegans, which is a distinct pathway from the anti-apoptotic one.

There is no established human time course, and stating one would be fabrication. What can be said honestly: in rodent injury models HNG effects on tissue markers appear within days of dosing, and in the metabolic models insulin sensitivity changes appeared over one to several weeks. Whether any of that transfers to a subcutaneous dose in a person is unknown. Users report nothing consistent and nothing acute. If you are expecting to feel something, you are probably going to be disappointed, and that expectation is itself the reason so many people conclude their vial was underdosed.

Humanin is most often stacked with MOTS-c on the reasoning that they are both mitochondrial-derived peptides. That reasoning is weak — they share a genomic neighbourhood and nothing else mechanistically. MOTS-c is an AMPK-route metabolic signal; humanin is a cytokine-receptor and BAX-binding cytoprotectant. Stacking them is not synergy in any demonstrated sense, it is just buying two things. The pairing with SHLP-2 is more coherent, since both are anti-apoptotic and both improve insulin sensitivity, but that also means you are doubling the same signal rather than covering different ground. The combination that should give you pause is humanin with anything else anti-apoptotic or pro-survival — an anti-apoptotic peptide plus IGF-1 analogues plus growth hormone secretagogues is a stack whose entire theme is telling cells not to die, and that theme has an obvious downside. Note also the S14G confusion: most product sold as 'humanin' is actually HNG, and HNG is described in the literature as a substantially more potent analogue. If you dose HNG as though it were native humanin you are not running the dose you think you are.

Against MOTS-c: MOTS-c has a clearer mechanistic story, an identified direct protein partner in CK2, and human polymorphism data tied to a functional phenotype. Humanin has better human epidemiology — the centenarian-offspring association and the P3S longevity variant are genuinely interesting human genetics — but weaker interventional grounding. Against SS-31: not comparable. SS-31 has a label and phase 3 data; humanin has never been given to a human in a published study. Against SHLP-2: same family, overlapping anti-apoptotic and metabolic actions, but humanin has an order of magnitude more literature and a defined receptor complex, so if you are picking one it should be humanin. Against SHLP-6: opposite direction entirely — SHLP-6 promotes apoptosis where humanin blocks it, and running both is self-cancelling. Against doing nothing: for a healthy person, unknown, and honestly this is one where the absence of a usable feedback signal is a strong practical argument against spending money on it.

$80–$250/month. Not verified against live vendor pricing this session. Based on typical 5-10 mg vial pricing and community dosing of 2-10 mg daily, a month consumes roughly 60-300 mg. The wide range reflects both dose spread and the fact that product sold as humanin varies in what it actually is — genuine HNG synthesis is more expensive than native humanin, and cheap product is a signal worth reading.

• ?No human pharmacokinetic data exist by any route. Half-life, tmax, bioavailability and blood-brain barrier penetration are all unmeasured in people.
• ?No human interventional trial of humanin or HNG has been completed or reported.
• ?The relative potency of HNG versus native humanin is described in the literature as large — the commonly repeated figure is around a thousandfold — but I could not resolve a primary source quantifying it in this session. Treat the number as folklore until you see the paper.
• ?The dimerisation state of humanin in a reconstituted vial, and how much it matters for activity, is not characterised for grey-market product.
• ?Which of the two arms — the extracellular receptor arm or the intracellular BAX arm — dominates after a subcutaneous dose is unknown, and they have quite different implications.
• ?The community dose range of 2-10 mg daily has no derivation from any published animal dose-scaling I could identify.
• ?The theoretical cancer risk from sustained systemic anti-apoptotic signalling has never been characterised in any model at therapeutic exposure.
• ?Cost figures are estimates and were not verified against live vendor pricing.

• A rescue factor abolishing neuronal cell death by a wide spectrum of familial Alzheimer's disease genes and Abeta Hashimoto Y, Niikura T, Tajima H, Yasukawa T, Sudo H, Ito Y, Kita Y, Kawasumi M, Kouyama K, Doyu M, Sobue G, Koide T, Tsuji S, Lang J, Kurokawa K, Nishimoto I, Proceedings of the National Academy of Sciences USA, 2001 · PMID 11371646

  The discovery paper. Humanin was found by screening for factors that protected neurons against familial Alzheimer's mutations and amyloid-beta — the first mitochondrial-derived peptide ever identified.

• The mitochondrial derived peptide humanin is a regulator of lifespan and healthspan Yen K, Mehta HH, Kim SJ, Lue Y, Hoang J, Guerrero N, Port J, Bi Q, Navarrete G, Brandhorst S, Lewis KN, Wan J, Swerdloff R, Mattison JA, Buffenstein R, Breton CV, Wang C, Longo V, Atzmon G, Wallace D, Barzilai N, Cohen P, Aging (Albany NY), 2020 · PMID 32575074

  The paper behind the two claims that made humanin famous in longevity circles: circulating humanin declines with age across species (but not in naked mole-rats), and levels are much higher in children of centenarians than in age-matched controls. Both are associations.

• Humanin variant P3S is associated with longevity in APOE4 carriers Miller B, Kim SJ, Mehta HH, et al., Aging Cell, 2024 · PMID 38520065

  A rare humanin coding variant enriched in centenarians carrying APOE4, with neuroprotection against amyloid pathology. Human genetic evidence that humanin sequence matters in people — which is a different and stronger kind of evidence than a rodent injection study.

• Humanin-induced autophagy plays important roles in skeletal muscle function and lifespan extension Kim SJ, Miller B, Mehta HH, et al., Biochimica et Biophysica Acta - General Subjects, 2022 · PMID 34624450

  Shows the lifespan effect in C. elegans runs through autophagy, not only through apoptosis blockade — a mechanistically distinct arm that is often left out of summaries.

• Mitochondrial-derived microproteins: from discovery to function Yen K, Miller B, Kim SJ, Cohen P, et al., Trends in Genetics, 2025 · PMID 39690001

  Current authoritative review of the whole family from the discovery group. The right place to calibrate how much of humanin biology is established.

• S14G-humanin alleviates acute lung injury by inhibiting NF-κB activation Wu Y, et al., Aging (Albany NY), 2023 · PMID 38054825

  Representative of the large and growing HNG animal literature across organ-injury models — renal, pulmonary, ovarian, retinal. Illustrates that the analogue is broadly cytoprotective in rodents, and that none of this work is in humans.

• Humanin (UniProtKB entry Q8IVG9) UniProt Consortium, UniProtKB

  The source used to verify the 24-residue sequence MAPRGFSCLLLLTSEIDLPVKRRA in this record.