Semax and Selank get name-checked together so often that people assume they are twins. They share exactly one feature: the same Pro-Gly-Pro tail, tacked onto two completely unrelated parent peptides. Past that, they diverge sharply. Different parent molecules, different proposed mechanisms, different experimental literatures. Treating them as an interchangeable pair of "cognitive peptides" hides what each one was actually built to test.
One programme, two parent sequences
Both compounds came out of the Institute of Molecular Genetics of the Russian Academy of Sciences, from a research line built on a simple premise. Endogenous regulatory peptides often show interesting central activity, but they get destroyed too fast in plasma to study systematically. Take a short active fragment, stabilise it, and you have something you can dose and measure. Semax and Selank apply that recipe to two unrelated parent molecules.
Semax is built on ACTH(4-7), the sequence Met-Glu-His-Phe, extended to Met-Glu-His-Phe-Pro-Gly-Pro. That parent fragment sits inside adrenocorticotropic hormone but carries none of its corticotropic function, since the region needed for melanocortin-2 receptor driven steroidogenesis is absent. That is why the animal literature reports central effects without adrenal activation. Interest in ACTH(4-10) as a behavioural peptide predates Semax by decades. The Russian work simply took the neurotropic end of that fragment and made it survivable.
Selank starts from tuftsin, Thr-Lys-Pro-Arg, a tetrapeptide derived from the CH2 domain of the IgG heavy chain and long characterised as a macrophage and neutrophil activating factor. With the same tail appended, it becomes Thr-Lys-Pro-Arg-Pro-Gly-Pro. The immunological pedigree matters here: much of the Selank literature reads as neuroimmunology rather than classical neuropharmacology, with cytokine measurements sitting alongside behavioural endpoints.
What the Pro-Gly-Pro tail is for
Short linear peptides clear from plasma in seconds to minutes, mostly through exopeptidases working inward from the termini. Proline is the standard defence against that: aminopeptidases and most carboxypeptidases handle proline-containing bonds poorly, so a C-terminal Pro-Gly-Pro cap slows degradation of the sequence sitting in front of it. The same trick shows up throughout peptide medicinal chemistry.
Two honest qualifications belong here. First, the tail extends stability, it does not confer it: measured plasma half-lives for both peptides remain short, and pharmacokinetic work depends on rapid sampling. Second, Pro-Gly-Pro is not inert. Free PGP and its cyclic derivatives have documented activity of their own in gastric and haemostatic models, so metabolite effects cannot be cleanly separated from parent-peptide effects. That is a genuine interpretive problem whenever an outcome gets measured hours after administration. Published work in this tradition generally used intranasal administration in rodents, on the argument that a fraction reaches the CNS along olfactory routes and avoids first-pass loss. The quantitative brain-exposure data behind that argument are thinner than the behavioural literature built on top of them.
Semax: neurotrophin signalling and ischaemia models
The most consistent molecular finding is neurotrophic. In rodent studies, Semax increases BDNF and TrkB expression in hippocampus and basal forebrain, with parallel changes in NGF signalling, and cortical transcriptome work after middle cerebral artery occlusion shows broad modulation of neurotrophin, vascular and inflammatory gene sets rather than one clean target. That pattern, wide transcriptional shifts with no identified high-affinity receptor, is characteristic of regulatory peptides, and it is the main reason the mechanism remains unsettled. For experimental design it also means single-marker readouts are a poor test: a compound that moves whole gene sets needs array or panel-level endpoints to be assessed fairly.
Ischaemia is the best-developed model context. Transient occlusion studies report reduced infarct volume and improved sensorimotor recovery, and the compound is registered as a medicine in Russia for ischaemic stroke and optic neuropathy on the strength of that domestic evidence package. Cognitive endpoints form a secondary strand: passive-avoidance and maze retention in rats, plus small human studies using selective-attention paradigms and P300 event-related potentials. That human work is old, small, and published largely in Russian-language journals.
Selank: enkephalin turnover and GABAergic tone
Selank's proposed mechanism runs through the enkephalin system. It inhibits enkephalin-degrading enzyme activity in plasma, extending the measurable lifetime of Leu-enkephalin, which places it among indirect opioidergic modulators rather than receptor agonists. Alongside this, rodent studies report altered GABA-A receptor subunit expression in hippocampal tissue and shifts in serotonin metabolism, plus a reasonably consistent immune signal: modulation of IL-6 and of Th1/Th2 cytokine balance in treated animals.
The behavioural literature leans heavily on anxiety models: elevated plus maze, open field and conflict paradigms in rodents, with anxiolytic-type profiles at the milligram-per-kilogram exposures specified in those protocols. Russian clinical trials compared it with benzodiazepines in generalised anxiety disorder and reported comparable rating-scale outcomes without sedation or memory impairment. Those trials were small, largely open-label, and have not been replicated by independent groups outside that tradition. That last point is the single most important thing to know about them.
Semax vs Selank: where they actually differ
- Parent sequence: Semax from ACTH(4-7), a melanocortin fragment. Selank from tuftsin, an immunoglobulin-derived immunomodulator.
- Primary mechanism claim: Semax works through neurotrophin (BDNF/TrkB) and monoamine modulation. Selank works through enkephalinase inhibition with downstream GABAergic and cytokine effects.
- Dominant model type: Semax sits in cerebral ischaemia, neuroprotection, learning and attention tasks. Selank sits in anxiety and stress paradigms with immune endpoints.
- Composition: both heptapeptides, both hydrophilic, roughly 750-815 Da. Semax carries oxidation-prone methionine and histidine, while Selank contains no aromatic residue at all.
- Analytical consequence: neither sequence contains tryptophan or tyrosine, so neither has a usable 280 nm chromophore. Semax offers only weak phenylalanine absorbance nearer 257 nm. Peptide-bond detection near 214 nm or mass spectrometry is therefore required for identity and purity work on both.
- Evidence base: Semax has the larger animal dataset, Selank the more distinctive proposed mechanism. Both sit almost entirely within one national research literature.
Weighing the evidence honestly
The provenance issue is not a rhetorical hedge, it is a design constraint. When nearly all primary data come from one institutional tradition, the cross-checks that make an effect trustworthy (independent replication, differing model systems, differing analytical methods, adversarial reanalysis) are largely absent. Neither peptide has been through a large multi-centre programme under EMA or FDA standards. Reporting detail on randomisation and blinding in the older clinical literature is often sparse by current expectations, and translation quality complicates secondary reading.
The reasonable response is not dismissal but stricter internal controls. Published effect sizes are best treated as hypotheses to test rather than benchmarks to reproduce. Vehicle controls and, where feasible, a free Pro-Gly-Pro comparison arm help separate parent-peptide activity from metabolite activity. Identity and purity of the test material should be established independently of any supplied paperwork, since a mis-synthesised or partially oxidised heptapeptide will happily generate a clean-looking null result. The principles for reading those reports are set out in our guide to verifying peptide purity and interpreting laboratory analyses.
Handling short peptides in vitro
Both peptides are small, charged and freely water-soluble, so preparation is straightforward. The failure modes lie in stability, not solubility. Keep lyophilised material cold, dry and dark, and treat reconstituted solutions as short-lived working stocks. General procedure is covered in our reconstitution and storage guide. Semax deserves particular care because methionine oxidises to the sulfoxide readily: avoid peroxide-contaminated solvents, minimise headspace and repeated freeze-thaw cycles, and consider an oxidised-species check if results drift between batches.
Selank is strongly basic thanks to its lysine and arginine residues, with a high isoelectric point, so adsorptive loss to glass and untreated plastic becomes significant at low working concentrations. Low-binding tubes and, where the assay tolerates it, a carrier protein reduce that error. In culture, the larger confound for both peptides is serum: aminopeptidase activity in serum-containing medium degrades short peptides measurably across an incubation, so nominal concentration and actual exposure diverge. Serum-free or reduced-serum conditions, shorter exposure windows, mid-experiment medium replenishment, or direct chromatographic measurement of medium content all narrow that gap, and any of them beats assuming the amount added is the amount present. Published cell-culture work with both compounds commonly operates in the micromolar range, close enough to degradation-driven error that this genuinely matters.
Both are stocked as lyophilised research material, Selank included.
All products are supplied strictly for in-vitro laboratory research use only. Not for human or veterinary use.