What published research reports on Semax in neuroprotection models

Published August 26, 2026 · Artemis Labs

Semax — answer capsule: Semax is a seven-amino-acid research peptide (PubChem CID 9811102) studied in laboratory models of nerve-cell injury, almost entirely by Russian groups. Nearly all of the Semax neuroprotection record comes from cultured cells and rodents, not from people. The results are mixed: Semax improved survival in cultured neurons under glutamate stress, but in a mouse model of dopamine-neuron damage its authors called the Semax effect slight while a comparator drug did far better. The only randomized controlled trial indexed for Semax was run in Russia in motor neuron disease, and it reported no effect on its primary endpoint. For Semax there are no meta-analyses, no systematic reviews, no double-blind studies, and no independent replications outside Russia.

Key findings

• In cultured rat cerebellar granule cells under glutamate toxicity, Semax “improved neuronal survival by on average 30%” (PMID 18239779). Cell culture, not an animal and not a person.
• In a mouse model of dopamine-neuron damage, Semax “slightly but reliably” increased striatal dopamine while the comparator drug Nomifensine “almost completely” protected dopaminergic fibers (PMID 34707903).
• In an Italian cell study, “Semax acetylation did not protect from Cu(II) induced toxicity on a SH-SY5Y neuroblastoma cell line, thus demonstrating the crucial role played by the free NH2 terminus in the cell protection” (PMID 27586814).
• The one randomized controlled trial in the record, run in Russia in motor neuron disease, found Semax “does not influence either the course of CPD or the dynamics of clinical estimates” (PMID 18379501).

What does “neuroprotection” mean in these studies?

Neuroprotection is a laboratory idea before it is a medical one. It means keeping nerve cells alive, or working, while something damages them. In the Semax literature that stress is usually too much of the signalling chemical glutamate, a shortage of oxygen, a metal ion such as copper, or a toxin that kills one type of neuron. Each is a model, and a model stands in for a disease rather than being it. A dish of neurons flooded with glutamate is not a stroke, so every sentence below names the system its result came from.

What did the cell-culture work find?

The clearest positive cell result is a 2007 study from the Russian Academy of Medical Sciences in Moscow. In cultured rat cerebellar granule cells, Semax and its Pro-Gly-Pro fragment delayed calcium dysregulation under glutamate neurotoxicity and improved survival by about 30 percent on average (PMID 18239779). The authors hedge their own explanation: the effect “can be due to” better mitochondrial resistance to calcium stress.

Two of the very few non-Russian Semax papers are Italian, and both are in vitro. Researchers in Catania reported in artificial membrane models that Semax prevented formation of amyloid-beta complexes with copper, concluding only that the results “suggest” an effect on fiber formation (PMID 35080861). No animal and no patient was involved.

The second Italian study is more consequential, and it is negative for the acetylated analog. In an SH-SY5Y neuroblastoma cell line, acetylating the free amino terminus abolished protection against copper toxicity, and the authors credited protection to the free NH2 terminus that acetylation removes (PMID 27586814). It is the only head-to-head comparison of the two forms PubMed indexes, and it is narrow: one endpoint, one cell line. Both facts belong together.

What did the animal models find?

In a mouse MPTP model, used to damage dopamine neurons, the authors described Semax as “slightly but reliably” increasing striatal dopamine, against a comparator, Nomifensine, that “almost completely” protected dopaminergic fibers (PMID 34707903). A Moscow rodent study reported the peptide “alone failed to alter the tissue and extracellular concentrations of dopamine and its metabolites” (PMID 16362768).

Oxygen-deprivation models produced a null. One Russian study reported that “neither mexidol nor semax upon single administration were effective on the models of acute hemic and histotoxic hypoxia”, and that Semax showed “bell-shaped” reversible dose-effect relationships (PMID 20486550). A bell-shaped curve means the response rose and then fell again across the range tested. More was not better.

The only non-Russian, non-Italian primary animal study located comes from a neurosurgery department in Surabaya, Indonesia. In male Sprague Dawley rats with spinal cord injury, the peptide “can increase anti-inflammatory cytokine expression in Sprague Dawley rat models with mild and severe SCI” (PMID 41179234). Results elsewhere were strain-dependent and once contradictory: Semax reduced predisposition to audiogenic epilepsy “in only one-month-old DBA/2J mice” of five inbred strains tested (PMID 19145359), while an earlier report found it “increased the sensitivity to sound” (PMID 14714455).

Has any of this been tested in people with a neurodegenerative disease?

Once, and the primary result was negative. A 2007 trial in a Russian neurology journal, in 27 patients with motor neuron disease, is the only PubMed record for this peptide carrying the Randomized Controlled Trial publication type. Its own abstract calls it an “open-label clinical trial” run “in sequential groups of patients”, using the Russian intranasal 1% formulation. The primary finding was flat. Semax “does not influence either the course of CPD or the dynamics of clinical estimates, in particular the terms of ensuing marked functional deficits on bulbar, cervical and lumbosacral levels of segmental innervation” (PMID 18379501). One secondary measure moved, a quality-of-life estimate the authors tied to “the improvement of emotional state and motivation”.

The other human study here is older and weaker. A 2000 report in a Russian ophthalmology journal, in patients with optic nerve disease, said Semax “effectively protected nervous tissue from aftereffects of injury” (PMID 10741256). It gave the peptide as an add-on, “in parallel with basic neurotrophic and antiinflammatory therapy”, reports no patient count, and describes no randomization or blinding. The result cannot be separated from that underlying therapy.

Ischemia work is covered on our Semax stroke and ischemia research page, and the neurotrophin mechanism on our Semax BDNF and neurotrophin research page.

What do recent reviews say about the evidence base?

They say it is thin. A 2026 US review says of these neuroactive peptides that “although preclinical studies are promising, there is a current lack of clinical trials” (PMID 41490200). A 2026 gerontology review lists Semax under neuroprotection among “non-approved peptides”, which “lack long-term safety data and systematic validation” (PMID 42021992). A 2025 review of peptides in neurodegenerative disease says “future clinical validation and advanced strategies are essential” (PMID 41004910).

What the research does not show

No Semax study in humans has been conducted outside Russia. Every human record in the PubMed index is Russian-affiliated or published in a Russian-language Russian journal, and there is no meta-analysis, no systematic review, no double-blind study, and no independent replication of any Russian result in either direction.

Four limits are specific to this page. The positive results are in cells and rodents, and a cultured neuron surviving glutamate stress is not evidence about a person. The one trial in a human neurodegenerative disease found no effect on what it set out to measure. The acetylated analog lost protection in the only head-to-head test that exists, so claims that an acetylated version is stronger are contradicted rather than supported. And effects in a healthy brain are uncharacterized: the originating Russian group wrote that “it is necessary to study influence of the peptide on the brain cells under normal physiological conditions, including understanding the risks of their use”, reporting immune-gene suppression in normal rats (PMID 39418522).

Frequently asked questions

Has Semax been shown to protect nerve cells in people?

No. The protective findings are from cultured cells and rodents. The only randomized controlled trial, run in Russia in motor neuron disease, found no effect on its primary endpoint (PMID 18379501), and the older Russian report claiming nerve-tissue protection gave the peptide alongside other therapy (PMID 10741256).

Does the amyloid work mean Semax is relevant to Alzheimer’s disease?

No. That study used artificial membranes in a test tube, and its own conclusion is that the results “suggest” the peptide interferes with amyloid-beta fiber formation (PMID 35080861). No animal and no patient was studied, so it says nothing about a disease outcome.

Is N-acetylated Semax more protective than Semax?

The only primary comparison indexed in PubMed found the opposite. Acetylation removed protection against copper toxicity in a neuroblastoma cell line, and the authors credited the free NH2 terminus that acetylation eliminates (PMID 27586814). One endpoint in one cell line is not a general verdict, but it is the only head-to-head data there is.

Why are almost all Semax studies Russian?

Semax was developed in Russia and studied there as a domestic pharmaceutical, and the research programme largely stayed there. The non-Russian primary research is not in humans at all: two Italian in-vitro studies (PMID 35080861, PMID 27586814) and one Indonesian rat study (PMID 41179234). This describes what is indexed in PubMed, not what exists.

References

1. Effects of semax and its Pro-Gly-Pro fragment on calcium homeostasis of neurons and their survival under conditions of glutamate toxicity. Bull Exp Biol Med, 2007. PMID 18239779 · doi:10.1007/s10517-007-0192-x
2. Influence of the N-terminus acetylation of Semax, a synthetic analog of ACTH(4-10), on copper(II) and zinc(II) coordination and biological properties. J Inorg Biochem, 2016. PMID 27586814 · doi:10.1016/j.jinorgbio.2016.08.013
3. Semax, a Synthetic Regulatory Peptide, Affects Copper-Induced Abeta Aggregation and Amyloid Formation in Artificial Membrane Models. ACS Chem Neurosci, 2022. PMID 35080861 · doi:10.1021/acschemneuro.1c00707
4. Mouse MPTP model of dopaminergic damage, 2021. Cited by identifier. PMID 34707903 · doi:10.32607/actanaturae.11433
5. Rodent study of dopamine and its metabolites. Neurochem Res, 2005. Cited by identifier. PMID 16362768 · doi:10.1007/s11064-005-8826-8
6. Rat spinal cord injury study, Surabaya, Indonesia, 2023. Cited by identifier. PMID 41179234 · doi:10.12688/f1000research.127413.2
7. Study of acute hemic and histotoxic hypoxia models. Cited by identifier; no DOI in the PubMed record. PMID 20486550
8. Changes of Transcriptomic Activity in Rat Brain Cells under the Influence of Synthetic Adrenocorticotropic Hormone-Like Peptides. Biochemistry (Mosc), 2024. PMID 39418522 · doi:10.1134/S0006297924090104
9. [The study of chronic partial denervation and quality of life in patients with motor neuron disease treated with semax]. Zh Nevrol Psikhiatr Im S S Korsakova, 2007. No DOI in the PubMed record. PMID 18379501
10. [Evaluation of therapeutic effect of new Russian drug semax in optic nerve disease]. Vestn Oftalmol, 2000. No DOI in the PubMed record. PMID 10741256
11. Audiogenic-epilepsy predisposition across five inbred mouse strains. Cited by identifier; no DOI in our source record. PMID 19145359
12. Report of increased sensitivity to sound. Cited by identifier; no DOI in the PubMed record. PMID 14714455
13. Review, 2026, United States. Cited by identifier. PMID 41490200 · doi:10.5435/JAAOSGlobal-D-25-00236
14. Gerontology review, 2026. Cited by identifier. PMID 42021992 · doi:10.3389/fragi.2026.1790247
15. Review of peptides in neurodegenerative disease, 2025. Cited by identifier. PMID 41004910 · doi:10.1016/j.npep.2025.102563

Methodology: this page draws only on primary records verified against PubChem and the PubMed index for the Artemis Labs Semax fact base, last verified August 26, 2026. Records without a verbatim title in that fact base are cited by PMID and DOI.

All compounds sold by Artemis Labs are for laboratory research use only. Nothing on this page is medical advice, and no statement has been evaluated by the FDA.