Semax vs Selank: How the Two Compounds Differ

Research Use Only. This article is written for laboratory and research professionals. The compounds discussed are supplied strictly for in-vitro laboratory research. They are not drugs, dietary supplements, or foods. They are not for human or veterinary use, and not for diagnostic or therapeutic use. Nothing here is medical, legal, or regulatory advice.

The short answer

The query runs both ways in search, selank vs semax and semax vs selank, and both point at the same structural question. Semax and Selank are seven-residue synthetic peptides that share an identical three-residue C-terminal extension, Pro-Gly-Pro, and share nothing else in their sequences. Semax carries residues 4 to 7 of adrenocorticotropic hormone at its N-terminus; Selank carries the whole of tuftsin, a tetrapeptide released by proteolysis of the immunoglobulin G heavy chain. One derives from a melanocortin hormone, the other from an antibody. The shared extension is a stability device, not a shared pharmacophore. The two are studied against different receptor systems, in different tissues, with different readouts, and direct head-to-head experiments are rare.

This article compares structure and mechanism. It contains no use-case guidance and no recommendation between the two.

Structural and mechanistic comparison

Property	Semax	Selank
Sequence	Met-Glu-His-Phe-Pro-Gly-Pro	Thr-Lys-Pro-Arg-Pro-Gly-Pro
Single-letter code	MEHFPGP	TKPRPGP
Parent molecule	ACTH(4-10), a proopiomelanocortin fragment	Tuftsin, an IgG heavy-chain fragment
Parent family	Melanocortin	Immunoglobulin-derived
Residues retained from parent	4 (ACTH 4-7)	4 (complete tuftsin)
C-terminal extension	Pro-Gly-Pro	Pro-Gly-Pro
Molecular formula	C₃₇H₅₁N₉O₁₀S	C₃₃H₅₇N₁₁O₉
Average mass	813.9 g/mol	751.9 g/mol
Monoisotopic mass	813.348 Da	751.434 Da
CAS number	80714-61-0	129954-34-3
PubChem CID	9811102	11765600
Sulfur-containing residue	Yes (methionine)	None
Aromatic residues	Phenylalanine, histidine imidazole	None
Ionisable side chains	Glutamate (acidic), histidine (weakly basic)	Lysine and arginine (both basic)
Approximate net charge at pH 7	Near neutral	Cationic, roughly +2
Computed XLogP	-2.8	-6.1
Topological polar surface area	312 Ų	322 Ų
Proline residues	2	4
Principal neural pathway examined	Neurotrophin transcription (BDNF, NGF, Trk receptors), monoamine metabolites	GABA radioligand binding, neurotransmission gene panels
Principal non-neural readouts	Immune and vascular gene sets in rodent ischemia models	Complement, chemokine and cytokine gene expression in spleen; serum cytokines
Characterised receptor	None identified	None identified
Shared in-vitro mechanism	Enkephalin-degrading serum peptidase inhibition, IC₅₀ near 10 µM	Same assay, IC₅₀ near 20 µM

Values are drawn from PubChem CID 9811102 and PubChem CID 11765600.

Different parent molecules

Semax comes from a hormone fragment

Adrenocorticotropic hormone is a 39-residue peptide cleaved from proopiomelanocortin. In the human precursor catalogued as UniProt P01189, the ACTH segment reads SYSMEHFRWGKPVGKKRRPVKVYPNGAEDESAEAFPLEF. Positions 4 through 10 of that segment, Met-Glu-His-Phe-Arg-Trp-Gly, form the fragment written ACTH(4-10). Semax keeps the first four of those residues and replaces Arg-Trp-Gly with the extension.

That substitution has a specific structural consequence. The melanocortin peptides share a conserved His-Phe-Arg-Trp message sequence at ACTH positions 6 through 9, and it is the element the melanocortin receptors MC1R through MC5R recognise. Semax retains His and Phe and loses Arg and Trp, so it carries half a pharmacophore. It also lacks the basic Lys-Lys-Arg-Arg address region at positions 15 to 18 that adrenal steroidogenesis requires, which is why it is described as melanocortin-derived without corticotropic activity.

Selank comes from an antibody fragment

Tuftsin is Thr-Lys-Pro-Arg, catalogued as PubChem CID 156080 with an average mass of 500.6 g/mol. Najjar and Nishioka described it at Tufts University and reported it in Nature in 1970 as a stimulator of phagocytosis.

Tuftsin has no gene of its own. The sequence sits at residues 289 to 292 of the CH2 domain of the immunoglobulin G heavy chain and is released when that protein is cut. In the human IgG1 constant region catalogued as UniProt P01857, the motif appears within the CH2 stretch WYVDGVEVHNAKTKPREEQYNSTYRVVS. Selank retains the entire tetrapeptide, which is a different relationship to its parent than Semax has: Semax is a truncated hormone fragment with a piece of its recognition element missing, while Selank is a complete natural signalling peptide with a tail added.

The shared Pro-Gly-Pro motif, and why two unrelated peptides converged on it

Both compounds emerged from the same Russian research programme, centred on the Institute of Molecular Genetics of the Russian Academy of Sciences, and both faced the same engineering problem. Short linear peptides are cleared fast: aminopeptidases work inward from the free N-terminus, carboxypeptidases from the free C-terminus, endopeptidases internally.

Proline is the general-purpose answer. Its side chain loops back and bonds to its own backbone nitrogen, forming a pyrrolidine ring. The amide nitrogen carries no hydrogen and backbone rotation is constrained, so most peptidases cannot productively bind a bond adjacent to it. Appending Pro-Gly-Pro puts one proline immediately after the active portion, shielding that junction, and one at the free C-terminus where carboxypeptidases would otherwise start, with glycine as a minimal spacer between them.

So the shared motif is a convergent design decision, not evidence of a shared mechanism. Two different starting sequences received the same protective tail from the same group of chemists. Content that treats "both are PGP peptides" as though it implied related activity has the logic backwards.

There is a real complication in that reasoning, and it applies to both compounds. Pro-Gly-Pro and its fragments are not inert. Dmitrieva and colleagues ran Semax against free PGP in rat cortex after middle cerebral artery occlusion and found that both altered transcription of neurotrophins and their receptors, with partially overlapping profiles, though they read the Semax effect as selective to ischemic tissue and the PGP effect as largely non-selective. On the Selank side, the dipeptide Gly-Pro reproduced much of the parent peptide's gene-expression profile in mouse spleen. For either compound, attributing a measured effect to the parent-derived portion requires a fragment control arm, and much of the literature does not include one.

Different physicochemical profiles

The two molecules behave differently in the flask, and the sequences say why.

Semax contains methionine, so it has an oxidation-labile thioether that converts to the sulfoxide and shows up 16 Da heavy on a mass spectrum. Selank has no cysteine and no methionine and therefore no comparable oxidation liability.

Semax carries one acidic side chain and one weakly basic one, which roughly cancel, leaving no strong net charge at neutral pH. Selank carries lysine and arginine and no acidic residues, so it is persistently cationic. The charge difference drives the polarity difference: computed XLogP is -2.8 for Semax and -6.1 for Selank. A doubly cationic peptide is more prone to adsorption onto glass and some plastics at low working concentration.

The analytical consequence is the one most often missed. Semax contains phenylalanine and histidine and has some ultraviolet absorbance in the aromatic region. Selank contains no tryptophan, tyrosine or phenylalanine at all, so it has no aromatic chromophore and is effectively invisible at 280 nm. Purity assays for Selank must be run at low wavelength, typically near 214 nm, where the peptide bond absorbs. A certificate of analysis for Selank that does not state its detection wavelength has left out something material.

Different receptor systems and pathways studied

Neither compound has a characterised high-affinity receptor. That is worth saying first, because it means every mechanistic account for either is inferential.

Semax. The dominant thread is neurotrophin expression. Reported measurements include large increases in BDNF and NGF messenger RNA in rat glial cultures and smaller increases in BDNF protein and trkB phosphorylation in rat hippocampus. A second thread is monoaminergic: rat striatal work reported a rise in the serotonin metabolite 5-HIAA, no direct change in dopamine or its metabolites, and amplification of the dopamine response to D-amphetamine. A third is descriptive omics in rodent vessel-occlusion models, where the largest coordinated changes fell in immune and vascular gene sets.

Selank. The dominant thread is GABAergic. Radioligand work on isolated brain cell plasma membranes has been interpreted as positive allosteric modulation of [³H]GABA binding, with non-additive effects when Selank is combined with benzodiazepines. Gene-expression work is mixed rather than uniform: an 84-gene neurotransmission panel changed substantially in rat frontal cortex, while the same panel in cultured IMR-32 human neuroblastoma cells showed no change under Selank alone. A second thread follows tuftsin's immunological pedigree into complement, chemokine, cytokine and interleukin-receptor gene expression in mouse spleen.

The one mechanism measured for both in the same experiment is peptidase inhibition. Kost and colleagues reported that both inhibited enkephalin-degrading enzymes in human serum in vitro, with half-maximal inhibitory concentrations near 10 µM for Semax and 20 µM for Selank, and found that pentapeptide fragments of each retained the activity while shorter fragments did not. Micromolar potency against a serum peptidase is a modest, indirect mechanism, and the authors present it as a contributing factor rather than an explanation.

Different research domains

The literatures barely overlap, and the reason is historical rather than principled.

Semax work grew out of the ACTH-fragment tradition and concentrated on neurotrophin biology and on rodent models of experimental vessel occlusion, which is where most of the transcriptomic and proteomic datasets come from. Selank work grew out of tuftsin immunology and split between GABA-receptor pharmacology and immune gene panels, with spleen tissue appearing as often as brain tissue.

That divergence means an apparent contrast between the two compounds is frequently an artefact of what each was tested for. Nobody has run the Selank GABA-binding assay on Semax at scale, and nobody has run the Semax neurotrophin experiments on Selank at scale. Absence of a finding for one compound in a domain usually reflects absence of the experiment.

What the comparative literature does and does not establish

Direct head-to-head work is scarce. The peptidase study above is the clearest example: both compounds, one assay, one set of conditions, quantitative results for each. A mouse embryonic stem cell study also tested Semax, Selank and several Pro-Gly-Pro fragments in parallel, measuring proliferation, survival under serum deprivation and differentiation into neuronal precursors, and reported different profiles for the two compounds on those endpoints. Beyond a handful of such papers, and one small resting-state imaging study in human volunteers that included both compounds and a placebo arm, side-by-side data are thin.

Shared authorship limits independent contrast. Both literatures come substantially from the same institutional network. Comparisons drawn within a single research programme are useful but do not carry the weight of independent replication.

Language and indexing limit appraisal. A large share of the primary work on both compounds exists in Russian with English abstracts only, so methods, controls and statistical detail are inaccessible to most readers, and reporting quality is uneven enough that at least one published cytokine study names the wrong compound in its own results text.

No comparative mechanism is established. Because neither peptide has an identified receptor, and because the shared Pro-Gly-Pro fragment has measurable activity of its own, the two cannot presently be placed on a common mechanistic axis. Any claim ranking them, or assigning them complementary roles, is running well ahead of the data.

What can be stated with confidence is structural: different parent molecules, different residue chemistry, different charge and polarity, different analytical requirements, an identical stability motif, and two research literatures pointed at different targets.

Regulatory status of both compounds

Both were developed in Russia and both are registered as medicines in the Russian Federation. Neither has any United States marketing authorisation.

• Neither Semax nor Selank has ever been approved by FDA.
• Neither appears on the binding 503A bulks list at 21 CFR 216.23, which contains six substances and no peptides, unchanged since 2019.
• Neither is among the Category 1 entries, which include VIP and GHK-Cu except injectable.
• The Pharmacy Compounding Advisory Committee meeting of 23 to 24 July 2026 produced a non-binding recommendation. No proposed rule and no final rule has followed, and the legal status of these compounds did not change.

Russian registration categories translated as "nootropic" and "anxiolytic" describe classifications in another jurisdiction, operating under different evidentiary standards. They are not FDA determinations and they do not transfer.

Handling notes for either compound

Both are lyophilized powders in research supply, both are hygroscopic, and both are materially more stable dry, cold and dark than in solution. Aliquoting avoids repeated freeze-thaw.

The differences that matter in practice: Semax needs protection from oxidation because of its methionine, and its identity trace should be checked for a +16 Da satellite. Selank needs attention to detection wavelength because it has no chromophore, and to adsorptive losses because it is doubly cationic.

For both, the documentation is the product: identity by mass spectrometry against the monoisotopic mass, purity by reversed-phase HPLC with column, gradient and wavelength stated, counterion identified, water content, net peptide content, and a named lot with the issuing laboratory, method and date.

Research-grade Semax and Selank sit in the neuro research collection. Fuller treatments of each compound are in the Semax structure and mechanism article and the Selank structure and mechanism article.

Working with this material

The certificates of analysis for Semax and Selank, from Horizon Analytical, are on the Semax and Selank product pages and on Lab results. Our own lot records appear at the lot lookup as lots are released. Supplied for laboratory research use only. Not for human or veterinary use.

Compliance notice. Nothing in this article should be read as guidance for administering any compound to a human or animal subject. "Research Use Only" describes the conditions under which a material is supplied; it is not a determination of legal status and confers no exemption from the Federal Food, Drug, and Cosmetic Act. Researchers are responsible for compliance with all applicable federal, state, and institutional requirements, and for obtaining any approvals their work requires.

Frequently asked questions

What is the difference between Selank and Semax?

They are both seven-residue synthetic peptides that end in the same Pro-Gly-Pro extension, and their first four residues are entirely different. Semax is Met-Glu-His-Phe-Pro-Gly-Pro, built on residues 4 to 7 of adrenocorticotropic hormone. Selank is Thr-Lys-Pro-Arg-Pro-Gly-Pro, built on the complete tuftsin tetrapeptide. They differ in molecular formula, mass, charge, polarity, aromatic content and the pathways their literatures examine.

Do Semax and Selank come from the same molecule?

No. Semax derives from adrenocorticotropic hormone, a melanocortin peptide cleaved from proopiomelanocortin. Selank derives from tuftsin, a tetrapeptide released by proteolysis of the immunoglobulin G heavy chain at residues 289 to 292 of the CH2 domain. One parent is a pituitary hormone, the other an antibody.

Why do Semax and Selank both end in Pro-Gly-Pro?

Because both were developed by the same Russian research programme to solve the same stability problem. Proline's side chain bonds back to its own backbone nitrogen, so most peptidases cannot cleave adjacent bonds. Pro-Gly-Pro places one proline after the active portion and one at the free C-terminus, blocking carboxypeptidase attack. The shared motif is a convergent design choice, not evidence of shared mechanism.

Do Semax and Selank act on the same pathways?

Not as studied. The Semax literature concentrates on neurotrophin transcription, monoamine metabolites and transcriptomic responses in rodent vessel-occlusion models. The Selank literature concentrates on GABA radioligand binding, neurotransmission gene panels and immune gene expression in spleen. The one mechanism measured for both in a single experiment is inhibition of enkephalin-degrading serum peptidases in vitro, at micromolar concentrations. Neither compound has a characterised receptor.

Has any study compared Semax and Selank directly?

Very few. Both were tested in one serum peptidase assay, with reported half-maximal inhibitory concentrations near 10 µM for Semax and 20 µM for Selank, and both appeared alongside Pro-Gly-Pro fragments in a mouse embryonic stem cell study measuring proliferation, survival and neuronal differentiation. Beyond a small number of such papers, side-by-side data are thin, and most of it comes from the same institutional network that produced the individual literatures.

Are Semax or Selank approved in the United States?

No. Neither has ever been approved by FDA and neither has any United States marketing authorisation. Neither appears on the binding 503A bulks list at 21 CFR 216.23, which contains six substances and no peptides, and neither is among the Category 1 entries. The non-binding advisory committee recommendation of July 2026 changed no legal status.

Primary sources

The remaining literature cited above, for direct verification:

• reported it in Nature in 1970: https://www.nature.com/articles/228672a0
• UniProt P01857: https://www.uniprot.org/uniprotkb/P01857/entry
• found that both altered transcription of neurotrophins and their receptors: https://pmc.ncbi.nlm.nih.gov/articles/PMC11498467/
• changed substantially in rat frontal cortex: https://pmc.ncbi.nlm.nih.gov/articles/PMC4757669/
• inhibited enkephalin-degrading enzymes in human serum in vitro: https://pubmed.ncbi.nlm.nih.gov/11443939/
• tested Semax, Selank and several Pro-Gly-Pro fragments in parallel: https://pubmed.ncbi.nlm.nih.gov/29063333/