F tier · safety concern
A ClinicalTrials.gov query for colivelin returns totalCount 0 - the compound has never been administered to a human being in a registered trial, and no clinical study of any design exists twenty years after its 2005 description. Every efficacy claim rests on rodent and cell-culture work from a small number of groups.
colivelin
Colivelin is a 26-amino-acid synthetic hybrid peptide created in 2005 at Keio University by Tomohiro Chiba, Ikuo Nishimoto and Masaaki Matsuoka, joining the neurotrophic nonapeptide ADNF-9 to the potent humanin analogue AGA-(C8R)HNG17. In neuronal cultures and transgenic mouse models of Alzheimer's disease and ALS it suppressed neuronal death at femtomolar concentrations, acting through CaMKIV and STAT3. It has never entered human testing, and is today used mainly as a laboratory STAT3-activating reagent rather than as a drug candidate.
// we supply this one
available as a research reagent
≥ 98% HPLC · lyophilised powder · batch certificate published. Grade F above is our own and is not adjusted because we stock it.
the explanation
Colivelin is a lab-made peptide that glues two neuroprotective fragments together, one from a nerve-growth factor and one from a mitochondrial peptide called humanin. In mice and in dishes of nerve cells it stopped brain cells from dying at extraordinarily low concentrations, which generated excitement about Alzheimer's and ALS. But it has never been given to a single person in a clinical trial, so nothing is known about whether it works or is safe in humans.
regulatory status
Unapproved research chemical; no human trials anywhere
Colivelin is not approved as a medicine by the MHRA, EMA, FDA or any other regulator, and no marketing application is documented. A ClinicalTrials.gov search returns zero records, and no registered trial was identified in any other registry; there is no evidence it ever entered formal clinical development. It is supplied only as a research reagent for in vitro and animal use - GenScript now lists its catalogue item as discontinued. No New Dietary Ingredient notification or lawful dietary-supplement pathway was identified for colivelin, and it is not named on the WADA Prohibited List, though as a pharmacological substance with no regulatory approval for human therapeutic use anywhere it falls within WADA category S0 (Non-Approved Substances), prohibited at all times in sport.
how it works · proposed mechanism
Colivelin was engineered on the premise that two independent neuronal survival pathways, tethered into one molecule, would protect more potently than either fragment alone. The N-terminal ADNF-9 segment signals largely through calcium/calmodulin-dependent protein kinase IV, while the C-terminal humanin analogue signals through a cytokine-receptor complex to STAT3. The design proved additive in cell culture, but this architecture has only ever been dissected in rodent neurons and transfected cell lines.
ADNF-9 module and CaMKIV
The nine-residue SALLRSIPA sequence is derived from activity-dependent neurotrophic factor, itself reported to protect motor neurons against mutant SOD1 toxicity. Its antiapoptotic signal is transduced substantially via CaMKIV. Placing it at the N terminus rather than the C terminus was the specific design choice that produced the potency gain.
Humanin module and STAT3
AGA-(C8R)HNG17 is a substituted 17-residue fragment of humanin, the mitochondrially encoded peptide, and carries the STAT3-activating arm of the molecule. Humanin is reported to act through a cytokine receptor complex involving CNTF receptor alpha, WSX-1 and gp130. Colivelin measurably raises phosphorylated STAT3 in cells and in mouse brain.
Dual-pathway redundancy explains the femtomolar figure
In the founding study, colivelin's protection was lost only when dominant-negative CaMKIV and dominant-negative STAT3 were co-expressed together; blocking either arm alone was insufficient. This redundancy is the mechanistic account of full suppression of neuronal death at 100 fM versus 10 pM for the parent humanin analogue. It is a cell-culture inference, not a human pharmacodynamic finding.
Downstream cholinergic and CNS-delivery observations
Intranasally administered colivelin reached the mouse CNS via the olfactory bulb and improved memory performance, with STAT3-dependent transcriptional upregulation of choline acetyltransferase and the vesicular acetylcholine transporter. A separate group reported reduced hippocampal amyloid-beta deposition and restored in vivo long-term potentiation in APP/PS1 mice. No comparable human CNS penetration or target-engagement data exist.
what’s reported
evidence shape
7 sourcesThe composition of the cited literature by study type. This describes the shape of the evidence, not its quality — and it does not by itself set the grade.
⚠ the catch
There is no human evidence for colivelin of any kind - not a phase 1 safety study, not a case series, not a single registered protocol. The headline potency figure, complete suppression of neuronal death at 100 fM, is a concentration in a dish of transfected cells, and cannot be translated into any statement about a dose or an effect in a person. The most rigorous in vivo outcome is a 7.4% extension of mean survival in G93A-SOD1 mice, about 11.6 days, in a model that has produced dozens of compounds that later failed outright in human ALS trials. Critically, the human data people cite in this area belong to humanin, not colivelin: a 2023 systematic review documenting circulating humanin levels in 693 people aged 21 to 113, in centenarians and their offspring, and in coronary and diabetic patients, does not mention colivelin once. Borrowing humanin's observational human literature to support colivelin is a category error, because colivelin is a different, non-endogenous, substituted fusion molecule.
key published findings
- Colivelin is ADNF-9 (SALLRSIPA) fused N-terminally to the humanin analogue AGA-(C8R)HNG17, giving the 26-mer SALLRSIPAPAGASRLLLLTGEIDLP, first reported by Chiba and colleagues in J Neurosci in 2005.
- It fully suppressed neuronal death caused by familial Alzheimer's disease genes (V642I-APP, M146L-PS1) and Abeta1-43 at 100 fM in vitro, roughly 100-fold more potent than the parent humanin analogue.
- Protection required both CaMKIV and STAT3 arms: it was diminished only when dominant-negative CaMKIV and dominant-negative STAT3 were co-expressed simultaneously.
- In G93A-SOD1 transgenic ALS mice, colivelin extended mean survival from 148.7 to 160.3 days, a 7.4% gain.
- Intranasal colivelin reached the mouse CNS via the olfactory bulb and improved memory, with STAT3-dependent upregulation of ChAT and VAChT.
- An independent group reported that colivelin restored hippocampal long-term potentiation, improved object recognition and spatial memory, and reduced hippocampal Abeta deposition in APP/PS1 mice.
- ClinicalTrials.gov returns totalCount 0 for colivelin as both a general term and an intervention; no trial in any registry was identified.
- The compound's dominant present-day role in the literature is as a commercial STAT3-activating tool reagent in unrelated fields, not as a therapeutic candidate.
limitations of the evidence
- Zero human exposure: no phase 1, no pharmacokinetics, no bioavailability, no toxicology, no immunogenicity data in people.
- No human clinical trial has ever been registered, so there is not even a planned pathway to human data.
- Effective concentrations are femtomolar in cell culture, a setting that cannot establish a human effective exposure.
- The ALS survival benefit (7.4%) is small and comes from the G93A-SOD1 model, which has poor historical predictive validity for human ALS.
- Foundational in vitro and mouse work originates largely from one Keio University group (Nishimoto, Matsuoka, Chiba), and Nishimoto died in 2005, ending the primary development programme.
- No head-to-head human comparison with humanin, S14G-humanin or any approved neurology drug exists.
- Plasma or CSF half-life is not established in any species in the sources reviewed, so exposure duration is unknown.
- Independent replications exist for cognition endpoints in mice but not for the femtomolar potency claim in any human tissue.
documented safety signals
- No clinical safety signals are documented - because no human has been studied. Absence of reported harm here reflects absence of investigation, not evidence of safety.
- Colivelin is a pharmacological STAT3 activator, and STAT3 is a canonical oncogenic and pro-fibrotic pathway; colivelin is routinely used in published cancer, fibrosis and osteoarthritis experiments specifically to switch JAK2/STAT3 signalling back on. Sustained systemic STAT3 activation is a theoretical proliferative and pro-inflammatory concern that has never been characterised in humans.
- No repeat-dose toxicology, carcinogenicity or reproductive-toxicity data were identified in any species.
- As a synthetic non-endogenous fusion peptide, immunogenicity and anti-drug antibody risk are plausible but entirely uncharacterised.
- Material sold outside regulated supply chains has no assured identity, purity, endotoxin or sterility profile; the peptide is a research reagent, and one major supplier has discontinued it.
identity
| full name | Colivelin (ADNF-9 fused to the humanin analogue AGA-(C8R)HNG17) |
| category | Cognitive |
| modality | peptide |
| formula | C119H206N32O35 |
| molar mass | 2645.1 g/mol |
| cas | 867021-83-8 |
| sequence | SALLRSIPAPAGASRLLLLTGEIDLP |
laboratory handling
| storage | Lyophilised powder is stored desiccated at -20 °C, protected from light and moisture. Reconstituted solution is aliquoted and frozen for anything beyond short-term use. Handling information only; this compound has never been administered to a human. |
| solubility | Soluble in sterile water and aqueous buffer; DMSO is sometimes used for concentrated stock preparation in cell-culture work. |
Handling information describes laboratory practice for a research reagent. It is not a preparation guide for use in humans.
the receipts
7 citedClinicalTrials.gov, U.S. National Library of Medicine · 2026 · official
CNS Drug Reviews · 2006 · review
Journal of Neuroscience · 2005 · preclinical
Biochemical and Biophysical Research Communications · 2006 · preclinical
Neuropsychopharmacology · 2008 · preclinical
Journal of Alzheimer's Disease · 2017 · preclinical
Biology · 2023 · systematic review
others in Cognitive