Semax Peptide: Structure, ACTH Origin and the Research Record

What Semax is

Semax is a synthetic heptapeptide, Met-Glu-His-Phe-Pro-Gly-Pro, derived from adrenocorticotropic hormone. PubChem lists it under CID 9811102 with the molecular formula C37H51N9O10S, a molecular weight of 813.9 and CAS number 80714-61-0 ^[1]. Like several compounds developed at the Institute of Molecular Genetics of the Russian Academy of Sciences, it pairs a fragment of a natural signalling molecule with a proline-rich tail intended to slow enzymatic breakdown.

Precise Pep Labs sells Semax 10mg for in vitro research use only. Nothing here describes or recommends use in people or animals, and nothing here is a protocol.

The naming problem: ACTH(4–10) or ACTH(4–7)?

Semax appears in the literature under two descriptions that look contradictory and are not. Papers variously call it "a synthetic ACTH(4–10) analogue" and "ACTH(4–7)PGP", and its full chemical name is written ACTH4-10Pro8-Gly9-Pro10 ^[16].

Both describe the same molecule. Take the seven-residue ACTH fragment numbered 4 to 10, keep the first four residues, and replace residues 8, 9 and 10 with proline, glycine and proline. What remains is four ACTH-derived residues followed by Pro-Gly-Pro — which is why the same compound can accurately be called an ACTH(4–10) analogue and described as ACTH(4–7) carrying a C-terminal Pro-Gly-Pro tripeptide ^[2]. The chemical name spells the substitution out explicitly.

This matters when searching the literature. A database query for one form will miss papers indexed under the other, and a vendor listing that uses one name is not describing a different product from one that uses the other.

Why the free N-terminus is not incidental

Semax is sold both in its ordinary form and as "N-Acetyl Semax", in which the free amino group at the N-terminus is acetylated. Those are different molecules, and there is direct chemical evidence that the difference is functional rather than cosmetic.

Magrì and colleagues characterised copper(II) and zinc(II) coordination by both forms. At physiological pH, Semax forms a Cu(II) complex with a CuN4 chromophore, while acetylated Semax forms a distorted CuN3O chromophore with a weak apical interaction from the methionine sulphur. The reduced ligand field shifts the formal redox potential of the acetylated complex to a more positive value, and the two behave differently towards ascorbic acid: the amino-free complex is redox-stable and unreactive, the acetylated one is not. In SH-SY5Y neuroblastoma cells, acetylation did not protect against copper-induced toxicity, which the authors read as showing that the free N-terminus plays the crucial role in that protection. Both forms bound Zn(II) with comparable strength ^[2].

Two things follow. A result obtained with one form does not automatically describe the other, and a product name alone does not establish which form is in a vial. That is an identity question, and identity is a measurement, as our guide to reading a certificate of analysis sets out.

What the research investigates

The published work is overwhelmingly preclinical, with a smaller clinical literature discussed separately below. Each result below carries its model.

Neurotrophin gene expression

The most developed line concerns nerve growth factor and brain-derived neurotrophic factor. Shadrina and colleagues compared the time course of NGF and BDNF gene expression in rat hippocampus, frontal cortex and retina under Semax ^[4].

A study in a rat permanent middle cerebral artery occlusion model then compared Semax with its own C-terminal fragment, Pro-Gly-Pro. Both activated transcription of neurotrophins and their receptors in the cortex, and the two transcription profiles partially overlapped. Semax raised Bdnf, TrkC and TrkA three hours after occlusion, Nt-3 and Ngf at 24 hours, and Ngf at 72 hours; PGP raised Bdnf and TrkC at three hours and Ngf, TrkB, TrkC and TrkA at 24 hours. The authors' own distinction is the important part: Semax affected transcription selectively in ischaemic cortex, whereas the influence of PGP was "mainly unspecific" ^[3].

That is a useful caution against attributing everything to the intact heptapeptide, and equally against dismissing it as merely a PGP delivery vehicle. The published comparison supports neither simplification.

Ischaemia transcriptomics

A more recent series used RNA sequencing in a rat transient middle cerebral artery occlusion model. At 4.5 hours after occlusion, Semax was associated with 131 differentially expressed genes in dorsolateral frontal cortex, against 322 for a related peptide, ACTH(6–9)PGP. Both partially prevented the ischaemia-induced disruption of immune- and neurosignalling-related expression profiles, while differing in their action on predominantly immune genes ^[5]. Companion work extended the comparison to a day after occlusion ^[12], to glyproline peptides more broadly ^[13], and to brain regions beyond the frontal cortex ^[14].

These are gene-expression profiles in a rodent injury model. They describe transcriptional correlates, not demonstrated clinical benefit.

The Pro-Gly-Pro tail, and what it is doing there

The Pro-Gly-Pro tripeptide on the end of Semax is usually described as a stability measure: proline-rich sequences resist peptidase cleavage, so the tail is intended to keep the molecule intact for longer. That is a chemical rationale, not a biological one, and the published work complicates the tidy picture in two directions.

First, PGP is not inert. Released as a free tripeptide it belongs to the glyprolines, a family studied in its own right, and glyproline peptides have been examined for effects on inflammatory and neurosignalling gene expression after cerebral ischaemia and reperfusion ^[13]. In the direct comparison described above, PGP on its own altered neurotrophin and receptor transcription in rat cortex ^[3].

Second, the intact peptide and the fragment did not behave identically. Semax acted selectively on ischaemic cortex while PGP's influence was reported as mainly unspecific ^[3], and in the RNA-sequencing work two closely related ACTH-derived peptides produced markedly different numbers of differentially expressed genes in the same model ^[5].

The practical reading is that neither "Semax is just a PGP carrier" nor "the tail is only there for stability" survives contact with the cited evidence. Both the parent fragment and the tail appear to contribute, and the published comparisons are about differences in selectivity rather than presence or absence of an effect.

Cellular and behavioural models

• Intracellular calcium dynamics in rat brain neurons have been examined directly ^[6].
• In rodent behavioural work, Semax and Melanotan II were reported to have antidepressant-like and antistress effects in one 2024 study ^[7], and Semax was reported to attenuate behavioural and neurochemical alterations following early-life adverse exposure in another ^[8].
• A 2017 study examined both Semax and Selank in rats with 6-OHDA-induced parkinsonism ^[9].

An independent line of work

Almost all of the above comes from the originating institute or collaborators, which is a real limitation on how much apparent replication is worth. One recent exception is useful: an Indonesian group studied the compound, under its chemical name, in a rat spinal cord injury model, reporting increased anti-inflammatory cytokine expression (IL-4, IL-10, IL-13) in treated animals relative to placebo at three and six hours, with the effect larger in mild than severe compression. The authors concluded that further research is needed to determine optimal conditions and in vivo outcome ^[16].

One independent rodent study does not settle anything, but it is worth more than another paper from the same group, and it is the kind of evidence the field needs more of.

Human research, described precisely

Semax is registered as a medicine in the Russian Federation and has been used there in stroke care. It holds no such approval elsewhere, and nothing below describes an approved use in any other jurisdiction.

Four limitations are visible from the records themselves:

1. Design. The 2018 report describes subgroup comparisons within a rehabilitation cohort, not a randomised, blinded, placebo-controlled trial.
2. Scale and setting. 110 participants, single-country, and the comparator is the absence of the compound rather than an active control.
3. Language and access. Both reports are in Russian in one national journal; the internationally available records are abstracts.
4. Independence. The clinical, mechanistic and developmental work share authors and institutions.

Correlation between a blood marker and a recovery score is not evidence that one caused the other, and the cited authors do not claim a controlled demonstration. We make no efficacy claim.

Analytical identification and purity testing

What the methods show

Reversed-phase HPLC with ultraviolet detection, the technique behind most peptide certificates, separates components and reports the main peak as a share of total detected peak area. It measures purity, not identity: it says how much of what was detected is the principal component, not what that component is. Compendial chromatography practice is set out in the United States Pharmacopeia's general chapter on chromatography ^[15]. Identity requires a separate measurement, normally mass spectrometry.

For Semax specifically, identity carries the extra weight described above: a mass measurement would also distinguish the ordinary form from the acetylated one, since acetylation adds 42 Da.

Our Semax certificate

The certificate on file for Semax 10mg is Kovera Labs report KVR-2026-433CF4, certified 18 July 2026 and added to our COA Library on 5 August 2026. Those are two different dates, and neither is the date the laboratory performed the analysis, which the certificate does not state.

Three observations follow. Each records what the document does and does not state.

The printed formula and CAS match the public record. Both agree with PubChem CID 9811102 exactly ^[1]. That confirms the document describes the compound it claims to describe. It does not confirm what is in the vial; only a measurement on the material can speak to that.

The identity line prints no observed mass. It records confirmation by LC-MS without a measured value or a theoretical value to compare against. For this compound that absence is more consequential than usual, because an intact mass would also separate Semax from N-acetyl Semax. A reader cannot check the conclusion from the document.

The report number is not a lot number, and the net content sits above nominal. The Batch field reads "N/A", so no lot identifier is recorded and KVR-2026-433CF4 should not be read as a verified lot. The reported net content of 10.81 mg is 8.1 % above the nominal 10 mg, and the laboratory marked that row as meeting its reference. That reference is printed as "~10mg" — a nominal value, with no tolerance, upper limit or lower limit anywhere on the certificate — so the difference cannot be assessed against a documented acceptance criterion, and this guide does not describe it as passing, failing or otherwise. Purity and net content answer different questions in any case; our guide to purity versus net content explains why.

Endotoxin is a separate question

Bacterial endotoxin testing is a different measurement from chromatographic purity and answers a different question. This guide presents no endotoxin result and makes no endotoxin claim for Semax, because our editorial review of the available endotoxin documentation is unresolved. A purity percentage says nothing about endotoxin content, and the reverse is equally true.

Limitations and open questions

• The evidence is preclinical. Rodent gene-expression, cellular and behavioural models dominate; none of it demonstrates a clinical effect.
• The fragment complicates attribution. Pro-Gly-Pro alone altered neurotrophin transcription, though less selectively than the intact peptide ^[3].
• Acetylated and non-acetylated forms differ chemically, including in copper coordination and redox behaviour, and one did not reproduce the other's protection in a cell model ^[2].
• Human evidence is limited in design, scale, language and independence, as set out above ^[10][11].
• Independent replication is thin, with ^[16] a recent exception.
• Our certificate records no lot identifier and no observed mass, and its net content is 8.1 % above nominal against a reference that states no range.
• Sources read. Findings were verified against the published abstracts and bibliographic records through PubMed, and identity values against PubChem. Full texts were not obtained for this article, so results reported only in figures or supplementary material are not represented here.

A note on Selank

Selank, another peptide developed at the same institute, carries the same C-terminal Pro-Gly-Pro tail but is built from an unrelated parent sequence, the immunoglobulin fragment tuftsin. The shared tail is a stabilisation strategy, not a shared mechanism, and results for one compound do not describe the other. Selank has its own guide: Selank peptide.

Where to find the certificate

The Kovera Labs report for Semax 10mg is in our COA Library, and it is linked from the product page. If the lot printed on a vial you hold does not appear there, contact us with that lot number and we will locate the corresponding report.