Selank vs. Semax: A Researcher's Guide to Nootropic Peptides

Medical Disclaimer: This article is for informational and educational purposes only and does not constitute medical advice. Selank and Semax are not FDA-approved for any indication in the United States. They are not currently eligible for compounding through U.S. 503A or 503B pharmacies under current guidance. Research on both compounds is predominantly from Russian institutions and has not been replicated in large-scale Western randomized controlled trials. Consult a qualified healthcare provider before considering any peptide therapy.

In the nootropic community — a subculture of people who experiment with compounds that may enhance cognitive function — Selank and Semax occupy an unusual position. They are not supplements or amino acid stacks. They are synthetic peptides developed by the Russian Academy of Sciences, approved as prescription medications in Russia, and increasingly sought out by Western biohackers and researchers who find the mechanism plausible and the side effect profiles reassuring.

Neither is FDA-approved. Neither is legally compoundable in the U.S. under current rules. The evidence base for both is genuine but concentrated almost entirely within Russian research institutions, which creates interpretive challenges. This guide explains what is actually known, what remains uncertain, and how the two peptides compare.

Why Intranasal Delivery: The Blood-Brain Barrier Problem

Before examining either compound’s specific mechanisms, it is essential to understand why both are administered intranasally. This is not arbitrary — it reflects a fundamental challenge in CNS peptide pharmacology: the blood-brain barrier (BBB).

Most peptides cannot cross the BBB in meaningful quantities. The BBB’s tight junctions and efflux transporters prevent large, hydrophilic molecules from entering the brain from systemic circulation. This is why most peptides with CNS targets fail when administered intravenously or subcutaneously — they don’t reach the target tissue in pharmacologically relevant concentrations.

The olfactory route bypasses the BBB. The olfactory epithelium in the upper nasal cavity is one of the few places where the nervous system is in direct contact with the external environment, protected only by a thin layer of specialized olfactory epithelial cells. Peptides deposited on the olfactory epithelium can be transported by two pathways:

Intraneuronal transport: Peptide molecules deposited on the olfactory mucosa are endocytosed by the dendrites of olfactory receptor neurons (ORNs) via clathrin-mediated endocytosis → packaged into early endosomes → loaded onto dynein motor proteins → carried by retrograde axonal transport along the olfactory nerve axons (fila olfactoria) through the perforations of the cribriform plate → released at olfactory bulb glomeruli, where the peptide enters the CNS parenchyma without ever crossing a tight-junction barrier. This pathway physically delivers intact peptide molecules into the olfactory bulb within 30–60 minutes of nasal deposition (Dhuria et al. 2010, PMID:19757983).
Paracellular transport: Smaller or more hydrophilic peptides pass between olfactory sustentacular cells (or transcytose through them) → reach the lamina propria beneath the epithelium → enter the perineural spaces surrounding olfactory nerve fascicle bundles. Critically, these perineural spaces are anatomically continuous with the subarachnoid space (CSF compartment). Peptides that enter the perineural pathway are swept by bulk CSF flow toward the olfactory bulb and adjacent forebrain structures — primarily olfactory tubercle, piriform cortex, amygdala, and hippocampus (via entorhinal cortex) — explaining why intranasally administered peptides show preferential limbic and prefrontal distribution rather than global, blood-borne, uniform brain distribution.

From the olfactory bulb, peptides can distribute to connected brain regions — the limbic system (amygdala, hippocampus), prefrontal cortex, and hypothalamus — within minutes to hours of nasal administration. The anatomical specificity of this distribution (limbic and prefrontal over brainstem or cerebellar) is a direct consequence of the olfactory nerve’s own projection targets and the local CSF flow patterns from the olfactory subarachnoid compartment, not a pharmacokinetic accident.

This is why both Selank and Semax are specifically administered as nasal sprays in Russian clinical practice, not oral tablets or injections. The intranasal route is not a convenience feature — it is the delivery mechanism that makes CNS activity possible.

What Are Nootropic Peptides?

The term “nootropic” was coined in 1972 by Romanian chemist Corneliu Giurgea to describe compounds that enhance cognitive function, are neuroprotective, and have very low toxicity. Selank and Semax represent a distinct sub-category: short synthetic peptides designed to cross the blood-brain barrier via the olfactory route and modulate neurochemistry directly.

Unlike GH-releasing peptides (ipamorelin, sermorelin, CJC-1295) or injury-repair peptides (BPC-157, TB-500), which primarily work through peripheral or systemic mechanisms, Selank and Semax are explicitly designed to act in the central nervous system — they are brain-targeted peptides.

Selank: Anxiety Without Sedation

What It Is

Selank (Thr-Lys-Pro-Arg-Pro-Gly-Pro) is a synthetic heptapeptide derived from tuftsin, a naturally occurring tetrapeptide (Thr-Lys-Pro-Arg) produced in the spleen that plays a role in immune function, mood, and stress modulation. The Pro-Gly-Pro extension added to tuftsin’s sequence increases CNS penetration and protects the molecule from peptidase degradation — extending its half-life from tuftsin’s very short minutes to a more pharmacologically useful duration.

Selank was developed at the Institute of Molecular Genetics of the Russian Academy of Sciences in the 1980s and 1990s and received Russian regulatory approval for anxiety and neurasthenia.

How Selank Works: The Mechanism

GABA-A receptor modulation: Selank influences GABA-A receptor function — the same receptor complex targeted by benzodiazepines (diazepam, lorazepam), alcohol, and barbiturates. The key pharmacological distinction is how Selank interacts with the GABA-A complex compared to benzodiazepines.

Benzodiazepines act as full positive allosteric modulators — they bind the benzodiazepine site at the α/γ interface of the GABA-A receptor pentamer (specifically requiring a γ2 subunit for high-affinity binding) and substantially amplify the receptor’s response to GABA by increasing chloride channel open frequency — this is distinct from barbiturates, which increase channel open duration. The result is enhanced Cl⁻ influx, membrane hyperpolarization, and the full benzodiazepine pharmacological profile: anxiolysis, sedation, anticonvulsant effect, and muscle relaxation. Chronic activation causes tolerance through two converging molecular mechanisms. First, chronic benzodiazepine-enhanced GABA-A receptor activation → elevated intracellular Ca²⁺ in neurons → sustained activation of PKC and CaMKII → these kinases phosphorylate specific intracellular residues on GABA-A receptor β and γ subunits (β3 Ser408/Ser409; γ2 Ser327) → phosphorylated subunits are recognized by β-arrestin-2 → β-arrestin-2 recruits the AP-2 clathrin adaptor complex → clathrin-mediated endocytosis removes GABA-A receptor pentamers from the synaptic membrane into early endosomes, reducing surface receptor density → fewer GABA-A receptors available at the postsynaptic density. Second, sustained exposure drives subunit composition switching: the α1β2γ2 subunit combination (benzodiazepine-sensitive, high-affinity binding at the α/γ interface) is downregulated at the transcriptional level (reduced expression of α1 and γ2 subunit mRNAs), while α4 and δ subunits are upregulated; α4βδ-containing receptors lack the γ2 subunit required for the benzodiazepine binding site — they are benzodiazepine-insensitive (and often respond to ethanol and THDOC/neurosteroids differently). The net result of endocytosis (fewer total receptors) + subunit switching (fewer benzodiazepine-responsive receptors) is pharmacological tolerance even while GABA itself maintains efficacy at the α4βδ extrasynaptic receptors. Withdrawal occurs because the upregulated α4βδ receptors are tonically less inhibitory than the downregulated α1γ2 receptors — removing the benzodiazepine uncovers a net hyperexcitable state that manifests as anxiety, insomnia, tremor, and seizure risk.

Selank appears to modulate GABA-A function differently — partial enhancement rather than full positive modulation, and through binding sites that may differ from the canonical benzodiazepine site. The result is anxiolysis without proportional sedation, and without the receptor downregulation that causes tolerance. The mechanistic basis for why partial modulation spares sedation while preserving anxiolysis involves both receptor subunit selectivity and the quantitative relationship between Cl⁻ conductance amplification and downstream circuit effects. Full benzodiazepines at the canonical α1β2γ2 binding site increase chloride channel open frequency approximately 2–3 fold per GABA molecule — a large-amplitude Cl⁻ influx that hyperpolarizes not only limbic amygdala circuits but also thalamic relay neurons (ventrobasal complex, thalamic reticular nucleus) which are the main drivers of the thalamo-cortical arousal loop. When thalamic relay neurons are strongly hyperpolarized, the thalamocortical spindle rhythm characteristic of NREM sleep is promoted and wakefulness is suppressed — the cellular mechanism of GABA-A agonist-induced sedation and cognitive blunting. Partial modulation with lower intrinsic efficacy (smaller Cl⁻ current amplification per GABA molecule) may be insufficient to overcome the thalamic neurons’ higher firing threshold required for arousal disruption, while still providing adequate hyperpolarization in the more excitable basolateral amygdala (BLA) principal neurons and central amygdala (CeA) output neurons, which have a lower activation threshold for fear/anxiety circuit driving. Additionally, GABA-A receptor subunit composition varies systematically by brain region in ways that create pharmacological selectivity windows: α1-containing receptors predominate in thalamic relay neurons, cortical layer VI, and cerebellum — these mediate sedation (confirmed by the α1-selective benzodiazepine zolpidem’s selective hypnotic profile); α2- and α3-containing receptors predominate in the BLA, hippocampal interneurons, and spinal cord — these mediate anxiolysis and muscle relaxation. Full benzodiazepines activate α1, α2, and α3 subunit-containing receptors with similar efficacy (all require γ2), producing anxiolysis and sedation simultaneously. If Selank’s binding site — which may differ from the canonical α/γ subunit interface — provides higher relative efficacy at α2/α3 assemblies than at α1 assemblies, it would generate a subunit-selective anxiolytic effect without engaging the α1-thalamic sedation circuit to the same degree. The partial agonist/subunit-selective profile also explains the absence of receptor downregulation: BZD-induced downregulation is driven by the sustained, large-amplitude Cl⁻ current that activates PKC/CaMKII → subunit phosphorylation → β-arrestin/clathrin endocytosis (detailed in the benzodiazepine mechanism above). Partial Cl⁻ current amplification — below the threshold that activates PKC phosphorylation of GABA-A β3 Ser408/409 at anxiolytic clinical concentrations — does not trigger the endocytic cascade, preserving receptor surface density and preventing the tolerance that benzodiazepines produce.

Enkephalinase inhibition: Enkephalins (met-enkephalin and leu-enkephalin) are endogenous opioid pentapeptides that modulate pain, stress, and mood through mu and delta opioid receptors. Enkephalinase — also called neprilysin (NEP) or CD10 — is a zinc metalloendopeptidase of the M13 peptidase family that rapidly inactivates enkephalins after their release into the synapse. NEP’s active site contains a catalytic zinc ion coordinated by the HEXXH motif residues His583 and His587 plus a distal Glu646; the glutamate residue Glu584 between the two histidines acts as a general base, abstracting a proton from the zinc-bound water molecule to generate a nucleophilic hydroxide. This hydroxide attacks the carbonyl carbon of the scissile amide bond in the substrate, forming a tetrahedral oxyanion transition state that collapses via C-N bond cleavage. NEP’s substrate selectivity depends on a hydrophobic S1′ pocket that preferentially accommodates large hydrophobic residues on the C-terminal side of the cleavage site; for Met-enkephalin (Tyr¹-Gly²-Gly³-Phe⁴-Met⁵), the primary cleavage is Gly³↓Phe⁴, releasing the inactive fragments Tyr-Gly-Gly and Phe-Met and terminating opioid receptor signalling within seconds of enkephalin release. Selank inhibits this degradation through competitive occupancy of the NEP active site. Selank’s three proline residues — and the consecutive Pro-Pro arrangement in the C-terminal Pro-Gly-Pro extension — create the same polyproline II helix geometry described in BPC-157 research: the backbone nitrogen at proline lacks the free N-H hydrogen required for the serine-protease-like transition state stabilization in NEP’s active site, while the bulky pyrrolidine rings of adjacent prolines sterically obstruct the S2/S2′ binding subsites of the enzyme. The result is that Selank occupies the NEP substrate-binding cleft but is hydrolyzed at a negligible rate — competitive substrate inhibition kinetics, analogous to the mechanism of the clinical NEP inhibitor racecadotril (thiorphan prodrug), but arising from steric rather than chelation-based mechanisms. With NEP inhibited, Met-enkephalin and Leu-enkephalin half-lives extend from seconds to minutes in the synaptic cleft → sustained occupancy of μ-opioid receptors (MOR) and δ-opioid receptors (DOR) in the central nucleus of the amygdala (CeA), periaqueductal gray, and nucleus accumbens shell → Gi/Gβγ → reduced adenylyl cyclase → reduced cAMP → reduced PKA → reduced neuronal excitability and reduced CeA output to downstream fear/anxiety effectors → anxiolysis (Zozulya et al., 2001, PMID:11550013).

This mechanism is distinct from direct opioid receptor agonism — Selank does not bind opioid receptors itself, so it does not produce opioid-like dependence. The molecular reason why NEP inhibition carries a different tolerance and dependence risk than direct MOR/DOR agonism is specific and worth understanding. Direct opioid agonists (morphine, fentanyl, oxycodone) provide continuous, sustained, high-occupancy MOR/DOR activation for the duration of their pharmacological half-life — hours of uninterrupted receptor engagement. This sustained occupancy drives GRK2 (G-protein receptor kinase 2) to the receptor: GRK2 phosphorylates MOR at Ser375 and Thr376 in the intracellular C-terminal tail → phosphorylated MOR recruits β-arrestin-2 → β-arrestin-2 uncouples MOR from Gi signaling AND recruits the AP-2/clathrin endocytosis machinery → MOR internalization into early endosomes → receptor downregulation from the synaptic membrane → loss of opioid sensitivity (tolerance). Simultaneously, chronic MOR activation upregulates the prodynorphin gene in reward circuits: accumulated Gβγ from prolonged Gi-signaling activates CREB/CRE → CREB-driven upregulation of prodynorphin → increased dynorphin (κ-opioid receptor agonist) → dynorphin opposes μ-opioid reward effects → worsening hedonic tone during drug-free periods (the molecular substrate of withdrawal and craving). Selank’s NEP inhibition operates differently: endogenous enkephalins are released pulsatilically in response to specific physiological stimuli (pain, acute stress, exercise) and then Selank extends each pulse’s duration from seconds to minutes — but the release itself remains event-driven and non-continuous. The pulsatile receptor activation pattern does not maintain the sustained MOR occupancy that GRK2 requires for productive phosphorylation kinetics. GRK2 phosphorylation efficiency depends on both receptor occupancy level and occupancy duration; brief enkephalin pulses — even extended by NEP inhibition — provide insufficient sustained occupancy to drive the GRK2/β-arrestin/endocytosis cascade at the rate that produces tolerance. Between enkephalin release events, MOR and DOR receptors recycle back to the membrane surface normally, maintaining receptor density. This is the fundamental distinction: extending the duration of pulsatile endogenous signals differs from imposing continuous tonic agonist presence because the former respects the receptor’s recycling intervals that GRK2-driven internalization depends on eliminating.

BDNF upregulation: Selank increases brain-derived neurotrophic factor (BDNF) expression in the hippocampus and cortex (Inozemtseva et al., 2008). BDNF acts on TrkB receptors to activate PI3K/Akt and MAPK/ERK signaling cascades, supporting synaptic plasticity, neurogenesis in the hippocampal dentate gyrus, and long-term memory consolidation. BDNF deficiency is associated with depression, anxiety disorders, and accelerated cognitive aging; its upregulation is proposed as a shared mechanism of several antidepressants and of exercise’s cognitive benefits. The molecular routes by which these two interventions achieve BDNF upregulation are distinct from Selank’s pathways and worth understanding explicitly. Antidepressants (SSRIs/SNRIs): SSRIs block the serotonin transporter (SERT) → elevated synaptic 5-HT concentration → sustained 5-HT2A receptor occupancy on hippocampal CA1 pyramidal neurons → 5-HT2A/Gαq/PLCβ → IP₃/DAG → PKC-γ → CREB Ser133 phosphorylation → pCREB/CBP co-activator complex → CRE elements in the BDNF exon IV promoter → BDNF mRNA transcription. This 5-HT2A/PKC/CREB route is why the clinical antidepressant effect requires 2–4 weeks of chronic administration: a single dose elevates synaptic 5-HT acutely, but productive CREB-driven BDNF transcription and the downstream TrkB-dependent dendritic spine remodeling require sustained receptor engagement; the delay between SERT inhibition and clinical antidepressant response corresponds precisely to the timeline for BDNF-TrkB-driven dendritic morphological changes in limbic circuits. SNRIs add a parallel noradrenergic arm: elevated synaptic NE → β2-adrenergic receptor/Gs/adenylyl cyclase/cAMP/PKA → CREB Ser133 → BDNF exon IV promoter — the same CREB convergence point but reached via a different receptor pathway. Exercise: aerobic exercise → norepinephrine and dopamine release from locus coeruleus and VTA → β2-AR/Gs/cAMP/PKA → CREB Ser133 in hippocampal granule cells → BDNF exon IV promoter. A second exercise-specific arm operates through peripheral metabolism: skeletal muscle contraction → AMPK activation → PGC-1α transcription factor → muscle-derived FNDC5 expression → irisin (the cleaved FNDC5 ectodomain) secreted into circulation → irisin crosses the blood-brain barrier → binds integrin αV receptors on hippocampal interneurons → downstream CREB activation → BDNF transcription. A third route uses exercise-derived lactate: high-intensity contractions generate extracellular lactate → monocarboxylate transporter 2 (MCT2) on hippocampal neurons → intracellular lactate → inhibits HCAR1 (hydrocarboxylic acid receptor 1) on neurons → removes HCAR1’s tonic Gi-mediated suppression of adenylyl cyclase → local cAMP/PKA/CREB → BDNF. These three exercise arms converge on the same CREB Ser133 phosphorylation event that antidepressants achieve through monoamine receptor stimulation — explaining both why physical activity has documented antidepressant efficacy and why combining exercise with SSRI therapy produces synergistic BDNF responses in clinical data.

The mechanism by which Selank upregulates hippocampal BDNF is indirect — operating through two converging upstream pathways rather than through direct TrkB activation (the mechanism Semax uses):

Pathway 1 — HPA axis deactivation and glucocorticoid receptor de-repression. Anxiety and chronic stress chronically activate the HPA axis: CRH (corticotropin-releasing hormone) from hypothalamic PVN neurons → anterior pituitary CRHR1/Gs/cAMP/PKA → proopiomelanocortin (POMC) cleavage → ACTH secretion → adrenal cortex zona fasciculata → cortisol synthesis and release. Cortisol enters neurons and binds the intracellular glucocorticoid receptor (GR), which, upon cortisol binding and chaperone (Hsp90) dissociation, translocates to the nucleus and binds negative glucocorticoid response elements (nGREs) — specifically the composite nGRE (IR3 motif: two half-sites in inverted orientation) in the BDNF exon IV promoter. The GR/nGRE complex recruits co-repressors NCoR/SMRT → HDAC3 → histone deacetylation → chromatin compaction → reduced BDNF exon IV transcription. This is the molecular basis for the established clinical observation that chronic stress and elevated cortisol produce hippocampal BDNF depletion, reduced dendritic complexity, and impaired memory encoding. Selank’s GABA-A modulation and enkephalinase inhibition reduce tonic anxiety-driven amygdala CeA output → reduced PVN CRH pulse frequency → reduced ACTH → reduced cortisol → reduced GR/nGRE occupancy at the BDNF exon IV promoter → relief of co-repressor-driven chromatin compaction → BDNF transcription rises. Selank does not activate the BDNF promoter directly; it removes the tonic stress-cortisol-GR suppressive drive, allowing the promoter’s basal activity to recover toward baseline.

Pathway 2 — Enkephalin/opioid receptor → PI3K/Akt/CREB axis. Met-enkephalin and Leu-enkephalin preserved by Selank’s NEP inhibition signal through μ-opioid receptors (MOR) and δ-opioid receptors (DOR) in the hippocampus and amygdala. MOR/DOR couple to Gi/o proteins → released Gβγ subunits directly activate PI3K p110β/γ (Gβγ engages the p110β catalytic subunit at an N-terminal Gβγ-binding domain, bypassing the requirement for receptor tyrosine kinase scaffolding that conventional PI3K activation requires) → PIP₃ generation → PDK1 → Akt Thr308 phosphorylation → fully active Akt → via RSK2 (90 kDa ribosomal S6 kinase): Akt → PDK1 → RSK2 Ser227 phosphorylation (linker activation) + Akt → RSK2 Thr574 phosphorylation → RSK2 catalytic activation → CREB Ser133 phosphorylation → pCREB recruits CBP/p300 co-activators → CRE elements in the BDNF exon IV promoter → BDNF transcription. This Gβγ/PI3K/Akt/CREB route is spatially constrained to regions with high MOR/DOR density — the amygdala, hippocampus, and prefrontal cortex — the same limbic network where Selank’s anxiolytic effects are most relevant, explaining why BDNF upregulation co-localizes with the anxiolytic effect rather than occurring globally across the brain (PMID:11550013).

Monoamine modulation — tryptophan/IDO axis: Selank’s serotonergic enhancement operates through a mechanism upstream of the serotonin synapse itself. Tryptophan is the dietary precursor for serotonin, but in the brain it faces metabolic competition: the enzyme IDO1 (indoleamine 2,3-dioxygenase 1), strongly induced by inflammatory cytokines (IFN-γ, TNF-α, IL-6), diverts tryptophan into the kynurenine pathway rather than toward serotonin synthesis. Under chronic stress or neuroinflammation, IDO1 upregulation reduces available tryptophan for tryptophan hydroxylase 2 (TPH2) → less 5-hydroxytryptophan → less serotonin in GABAergic and raphe neurons. Kynurenine pathway downstream products include quinolinic acid — an endogenous NMDA receptor agonist whose excitotoxicity mechanism is structurally specific: quinolinic acid’s dicarboxylate groups (at pyridine ring positions 2 and 3) mimic glutamate’s α- and γ-carboxylate groups, allowing it to occupy the GluN2A/GluN2B glutamate-binding site on NMDA receptors; its NR2B (GluN2B) subunit preference is mechanistically important because GluN2B-rich NMDA receptors are disproportionately expressed on cortical pyramidal neurons and striatal medium spiny neurons — explaining the regional prefrontal and limbic vulnerability. At the low-micromolar CSF concentrations that chronic IDO1 upregulation generates in neuroinflammation, quinolinic acid produces sustained NMDA-R activation → pathological Ca²⁺ influx → mitochondrial Ca²⁺ overload → mPTP opening → caspase-3 apoptosis. A second excitotoxic arm operates through direct Fe²⁺ chelation: quinolinic acid’s dicarboxylate coordinates ferrous iron → Fe²⁺/quinolinate complex → Fenton chemistry: Fe²⁺ + H₂O₂ → •OH (hydroxyl radical) → lipid peroxidation at dendrite membranes, independent of receptor activation. Critically, quinolinic acid is not cleared by the EAAT1/EAAT2 glial glutamate transporters that rapidly recycle synaptic glutamate — the pyridine dicarboxylate structure is not recognized as a high-affinity transport substrate — so quinolinic acid lingers in the extracellular space rather than being taken up, providing prolonged NMDA receptor activation that underlies the chronic cognitive impairment in IDO1-high neuroinflammatory states; it accumulates in prefrontal and limbic regions under chronic IDO1 activation. Selank suppresses IDO1 activity and reduces IFN-γ and IL-6 secretion through a specific FPR2-mediated cytokine mRNA destabilization sequence: the same FPR2/Gβγ/PI3K-γ pathway described in the immunomodulation section generates Akt activation → Akt phosphorylates ASK1 (apoptosis signal-regulating kinase 1) at Ser83 → ASK1 inactivation → reduced MKK3/6 → p38 MAPK activity → reduced MK2 (MAPK-activated protein kinase 2) activation; MK2 normally phosphorylates TTP (tristetraprolin) at Ser52/Ser178, inactivating TTP and preventing it from binding AU-rich elements (AREs) in the 3′ UTRs of IFNG and IL6 mRNAs — with MK2 suppressed, unphosphorylated TTP recruits the CCR4-NOT deadenylase complex → poly-A tail shortening → XRN1-mediated 5′→3′ mRNA decay → reduced IFN-γ and IL-6 protein secretion; this IFN-γ reduction is critical for IDO1 suppression: IFN-γ normally engages IFNGR1/IFNGR2 on dendritic cells → JAK1/JAK2 → STAT1 Tyr701 → pSTAT1 → GAS (gamma-activated sequence) elements → IRF1 transcription → IRF1 binds ISRE-like elements in the IDO1 promoter → IDO1 enzyme production; Selank-driven IFN-γ reduction removes this primary IDO1 transcriptional inducer, shifting tryptophan metabolism back toward the serotonin synthetic route — increasing serotonin precursor availability without requiring reuptake inhibition. This mechanism, operating pre-synaptically at the tryptophan metabolic fork, explains why Selank’s serotonergic effect is particularly relevant in anxiety states driven by inflammatory or stress-related IDO1 upregulation — and why it avoids the SSRI-associated side effects that arise from non-selective reuptake inhibition at serotonin synapses throughout the brain and gut (Zozulya et al., 2001, PMID:11550013).

Immunomodulation: Like its parent compound tuftsin, Selank modulates immune activity — specifically NK cell activity and cytokine balance (Uchakina et al., 2008). This is an unusual property for an anxiolytic. The clinical relevance of Selank’s immunomodulatory effects for human use is not established, but the biology is consistent with tuftsin’s known immune-regulatory role.

The molecular basis for this immunomodulatory activity derives from the tuftsin core sequence (Thr-Lys-Pro-Arg, positions 1–4 of Selank’s heptapeptide) that was retained in Selank’s design. Tuftsin is a natural tetrapeptide cleaved from the Fc region of immunoglobulin G (the 289–292 segment of the IgG heavy chain) by two sequential enzymes: leukokininase (tuftsin endocarboxypeptidase) on leukocyte surfaces cleaves the C-terminal Arg from the precursor sequence, followed by splenopentin processing in the spleen. The tuftsin receptor is now characterized as formyl peptide receptor 2 (FPR2/FPRL1) — a Gi-coupled GPCR expressed on NK cells, monocytes, macrophages, and neutrophils that was originally identified as a receptor for lipoxin A4 and annexin A1. Tuftsin’s Lys-Pro-Arg C-terminal motif occupies FPR2’s hydrophobic transmembrane binding pocket, engaging the receptor with moderate affinity.

FPR2 engagement on NK cells → Gi dissociation → released Gβγ subunits directly activate PI3K-γ (p110γ/p101), which is uniquely sensitive to Gβγ (unlike p110α or p110δ, which require receptor tyrosine kinase scaffolding) → PIP₃ accumulation at the inner leaflet → PDK1 and mTORC2 → Akt activation → downstream Rac1 and Cdc42 activation via Rac-GEFs (Vav1, DOCK2) → actin polymerization and lamellipodia formation. This cytoskeletal reorganization drives NK cell polarization: lytic granules (containing perforin and granzyme B) are repositioned along the microtubule organizing center (MTOC) → MTOC reorients toward the immunological synapse with the target cell → polarized granule exocytosis at the synapse → efficient target cell killing at lower effector-to-target ratios. Simultaneously, FPR2/Gi/Gβγ → PLCβ (phospholipase C-β) → IP₃/DAG → IP₃-receptor-mediated ER Ca²⁺ release → calmodulin → myosin light chain kinase (MLCK) → cytoskeletal contraction that further drives granule polarization.

On monocytes and macrophages, the same FPR2/PI3K-γ/Rac1 cascade drives phagocytic cup formation (enhanced macrophage phagocytosis of opsonized particles) and amplifies oxidative burst (NADPH oxidase complex assembly at the phagosomal membrane requires Rac2 GTPase, which is activated downstream of PI3K-γ). The net outcome — augmented NK cytotoxicity and macrophage phagocytic capacity — explains Uchakina et al.’s finding that anxiety patients treated with Selank showed improved NK cell functional activity compared to baseline.

A second immunomodulatory arm is indirect and integrates with Selank’s serotonergic mechanism: Selank’s IDO1 suppression (through reduced IFN-γ and IL-6 described in the tryptophan/kynurenine section) conserves tryptophan for a purpose beyond serotonin synthesis. NK cells and T-cells in the tumor microenvironment or during chronic viral infection face tryptophan depletion from IDO1-high tolerogenic dendritic cells — depleted tryptophan activates the GCN2 (general control nonderepressible-2) eIF2α kinase in NK cells → eIF2α phosphorylation → global translational inhibition including perforin and granzyme B synthesis → functional NK exhaustion. Selank’s IDO1 suppression preserves local tryptophan availability → GCN2 remains inactive → NK protein synthesis capacity maintained. This IDO1-to-GCN2 axis connecting anxiety/stress neurochemistry to peripheral NK cell function is the molecular reason an anxiolytic can simultaneously have immunostimulatory properties — both effects share an upstream inflammatory driver.

The Core Research Findings

Anxiety reduction: Multiple Russian clinical studies, including a randomized comparative trial, have demonstrated Selank’s efficacy in generalized anxiety disorder with outcomes comparable to benzodiazepines but markedly better side effect profiles (Zozulia et al., 2008). Critically, patients did not experience sedation, cognitive blunting, or withdrawal on discontinuation.

Cognitive enhancement: Animal models and smaller human studies report improvements in working memory, information processing speed, and learning. The BDNF mechanism provides a plausible molecular explanation — BDNF-TrkB signaling is among the best-validated molecular substrates for synaptic plasticity and memory formation.

The key distinguishing claim: Selank produces anxiolytic effects without sedation, tolerance, or addiction risk. This is its most clinically important property — if confirmed in independent Western trials, it would represent a meaningful advance over existing anxiolytic options. The European Phase 2 trial for Selank in anxiety disorders (referenced in the clinical trials guide) may provide this validation.

Administration and Dosing

In Russian clinical practice, Selank is administered as a nasal spray — intranasal delivery via the olfactory route described above. Typical clinical dosing is 250–750 mcg intranasally, 1–3 times daily. Duration of action after a single dose is approximately 4–6 hours based on pharmacokinetic studies.

Subcutaneous injection has been used in research settings and by some self-experimenting researchers in the West. The intranasal route is preferred in clinical contexts because it is non-invasive and produces adequate CNS delivery without systemic exposure concerns.

Side Effects

The reported side effect profile is notably benign:

No sedation or cognitive blunting
No withdrawal syndrome on discontinuation
No documented addiction potential
Mild transient nasal irritation (intranasal route)
Occasional mild headache
Mild immunostimulation — theoretically relevant in autoimmune contexts but not documented as a clinical concern at typical doses

Semax: Neuroprotection and Cognitive Stimulation

What It Is

Semax (MEHFPGP — Met-Glu-His-Phe-Pro-Gly-Pro) is a synthetic heptapeptide derived from the ACTH(4-7) fragment of adrenocorticotropic hormone (melanocyte-stimulating hormone-related sequence), with a Pro-Gly-Pro extension that resists enzymatic degradation. Developed at the same Institute of Molecular Genetics in the 1980s, Semax received Russian clinical approval in the 1990s.

Semax has been in Russian clinical use for over 30 years — a fact that lends its safety profile considerably more weight than a newly synthesized compound with shorter use history.

How Semax Works

BDNF upregulation (stronger than Selank): Semax upregulates BDNF more potently than Selank, which correlates with its stronger cognitive stimulation profile. The BDNF cascade starts at TrkB (NTRK2) — BDNF binds the extracellular domain of TrkB, inducing receptor dimerization and trans-autophosphorylation of the intracellular kinase domain at multiple tyrosine residues. Three signaling arms diverge from these phosphorylated residues, each contributing a distinct component of the LTP and neurogenesis outcome. Ras-MEK-ERK arm (early LTP consolidation): pTyr490 on TrkB → Shc (SH2-containing sequence protein) adaptor docking → Shc/Grb2/SOS complex assembly → SOS is a Ras-GEF (guanine nucleotide exchange factor) that converts Ras-GDP → Ras-GTP → Ras-GTP activates Raf → MEK1/2 → ERK1/2 phosphorylation at Thr202/Tyr204. Active ERK1/2 translocates to the nucleus where it phosphorylates Elk-1 (a ternary complex factor) and CREB Ser133 → Elk-1 and pCREB cooperate with CBP/p300 histone acetyltransferase coactivators → transcription of immediate early genes Arc (activity-regulated cytoskeleton-associated protein), c-fos, and zif268 — the early gene signature of synaptic plasticity. Arc specifically traffics to recently activated dendritic spines where it facilitates AMPA receptor endocytosis for homeostatic scaling and synaptogenesis. PLC-γ1/Ca²⁺/CaMKIV arm (late LTP gate): TrkB pTyr816 → PLC-γ1 Src-homology-2 domain recruitment → PLCγ1 activation → cleaves PIP₂ → IP₃ + DAG → IP₃ opens IP₃R on ER in dendritic spines → local Ca²⁺ release → Ca²⁺/calmodulin complex → CaMKIV (calcium/calmodulin kinase IV) activation in dendritic spines and nucleus → CaMKIV phosphorylates CREB Ser133 (same residue as ERK, providing amplification and temporal integration) → late-phase LTP gene program including AMPA receptor GluA1/GluA2 subunit transcription → increased AMPA receptor pool available for insertion at potentiated synapses. PI3K/Akt/mTOR arm (local synaptic protein synthesis): TrkB pTyr490 → IRS-1/PI3K p85 SH2 recruitment → PIP₃ production → PDK1 + mTORC2 → Akt Thr308 + Ser473 phosphorylation → Akt → TSC2 inhibition → Rheb-GTP → mTORC1 activation at the synapse (dendritic mTORC1 pools are distinct from somatic pools) → 4E-BP1 and S6K1 phosphorylation → cap-dependent translation of CaMKII, PSD-95, and synapsin mRNAs from locally stored dendritic mRNA pools → spine-compartmentalized protein synthesis without requiring nuclear gene transcription — the molecular basis for synapse-specific long-term potentiation at individual dendritic spines. Hippocampal neurogenesis arm: In hippocampal dentate gyrus granule cell progenitors, BDNF/TrkB/PI3K/Akt promotes survival (Akt → BAD Ser136 phosphorylation → 14-3-3 sequestration → Bcl-2/Bcl-xL anti-apoptosis) and ERK1/2 promotes proliferation and differentiation of adult-born granule cells, expanding the hippocampal encoding capacity that underlies pattern separation and episodic memory formation. Semax’s potent BDNF induction therefore simultaneously supports early LTP consolidation, late LTP maintenance, synapse-specific local translation, and adult neurogenesis — a multi-arm cognitive enhancement profile that explains both the working memory and learning improvements observed in Semax studies (Eremin et al., 2005, PMID:16362768).

Dopamine and serotonin activation: Semax activates dopaminergic and serotoninergic systems more prominently than Selank, producing more stimulating cognitive effects. The dopaminergic component is responsible for the “noticeable” cognitive stimulation users describe — increased motivation, focus, and processing speed. Unlike stimulants (amphetamine, methylphenidate), Semax does not release dopamine directly but appears to upregulate dopamine receptor sensitivity and modulate synthesis pathways.

The ACTH(4-7)/melanocortin receptor → TH/DARPP-32 mechanism for Semax’s dopaminergic effects. Semax is structurally derived from the ACTH(4-7) fragment, which retains the ability to interact with melanocortin receptors — specifically MC1R and MC4R — expressed on dopaminergic neurons in the VTA and substantia nigra compacta. These are Gs-coupled receptors: ACTH(4-7) fragment binding → adenylyl cyclase → cAMP → PKA activation in dopaminergic cell bodies. PKA phosphorylates and activates the transcription factor CREB, which drives expression of TH (tyrosine hydroxylase), the rate-limiting enzyme in dopamine synthesis (catalyzing tyrosine → L-DOPA; aromatic L-amino acid decarboxylase/AADC then converts L-DOPA → dopamine). Increased TH expression raises the dopamine synthesis rate and replenishes vesicular dopamine stores — increasing the quantal dopamine release available per action potential. Crucially, this mechanism increases dopamine synthesis throughput rather than triggering mass vesicular release: amphetamine and MDMA reverse the vesicular monoamine transporter VMAT2, dumping preformed vesicular stores into the cytosol and then reversing DAT to expel dopamine extrasynaptically — a mechanism that rapidly depletes vesicular reserves and causes the post-use “crash.” Semax’s TH induction, by contrast, elevates the steady-state level of newly synthesized dopamine without depleting storage, which explains the absence of a rebound fatigue effect on discontinuation. Additionally, PKA phosphorylates DARPP-32 (dopamine- and cAMP-regulated phosphoprotein, 32 kDa) at Thr34 → active pThr34-DARPP-32 inhibits protein phosphatase-1 (PP-1) → prolongs phosphorylation of AMPA receptor GluA1 Ser845, D1 receptor-coupled ion channels, and L-type Ca²⁺ channels in dopaminergic neurons → enhanced postsynaptic sensitivity to dopamine without receptor downregulation. This DARPP-32 amplification arm explains the reported enhancement of dopamine receptor sensitivity — not new receptor synthesis, but prolonged signaling per receptor activation event — as a downstream consequence of Semax’s melanocortin receptor engagement (Eremin et al., 2005, PMID:16362768).

The serotonergic arm follows a parallel but anatomically distinct mechanism. Semax’s ACTH(4-7)-derived sequence engages MC4R expressed on neurons of the dorsal raphe nucleus (DRN) — the brainstem’s principal serotonin-producing nucleus, whose ascending axons supply 5-HT to the prefrontal cortex, hippocampus, amygdala, and striatum. MC4R/Gs → adenylyl cyclase → cAMP → PKA activation in DRN 5-HT neurons operates through two sequential outputs. First, a rapid post-translational arm: PKA directly phosphorylates tryptophan hydroxylase 2 (TPH2) — the rate-limiting enzyme in neuronal serotonin synthesis, catalyzing the hydroxylation of tryptophan → 5-hydroxytryptophan (5-HTP); aromatic L-amino acid decarboxylase (AADC) subsequently converts 5-HTP → serotonin — at Ser19 and Ser104, increasing TPH2 catalytic Vmax without changing its expression level. This post-translational phosphorylation accelerates serotonin synthesis from the available tryptophan pool within minutes of PKA activation. Second, a transcriptional arm: the same PKA → CREB Ser133 phosphorylation that drives TH expression in VTA dopaminergic neurons also acts on CRE elements in the TPH2 gene promoter in DRN neurons → increased TPH2 mRNA transcription and protein synthesis → sustained expanded serotonin synthesis capacity over hours, outlasting the post-translational kinetic enhancement. The same DARPP-32 Thr34 phosphorylation mechanism that amplifies dopamine postsynaptic signaling duration operates in parallel on 5-HT₁A and 5-HT₂A receptor-bearing neurons in serotonin projection areas — PP-1 inhibition → prolonged 5-HT receptor phospho-target signaling per serotonin release event. This symmetric TPH2/serotonin mechanism precisely mirrors the TH/dopamine mechanism, operating in anatomically adjacent but functionally distinct monoamine nuclei — explaining why Semax simultaneously engages both systems and why its subjective effect profile combines dopaminergic focus and drive (VTA/TH axis) with the mood stability and anxiety mitigation conferred by enhanced serotonergic tone (DRN/TPH2 axis), rather than producing the pure dopaminergic forward stimulation characteristic of classical stimulants (PMID:16362768).

Neuroprotection in ischemia: Semax’s strongest evidence base is neuroprotective. In ischemic conditions (stroke, hypoxia), Semax reduces neuroinflammation, limits excitotoxicity from glutamate release, and upregulates antioxidant pathways through a molecular program centered on the Nrf2/KEAP1 axis — the primary cellular antioxidant transcriptional response. Under basal conditions, Nrf2 (nuclear factor erythroid 2-related factor 2) is held in the cytoplasm by KEAP1 (Kelch-like ECH-associated protein 1), which acts as an adaptor in a CUL3-RING E3 ubiquitin ligase complex → K48-linked polyubiquitination of Nrf2 → 26S proteasomal degradation → antioxidant genes transcribed at low baseline rates. Semax activates Nrf2 through two converging pathways that are both operative in ischemic neurons. First, Semax’s BDNF/TrkB → PI3K/Akt axis: active Akt phosphorylates GSK-3β at Ser9 → GSK-3β inactivation; active GSK-3β would otherwise phosphorylate Fyn kinase → Fyn → Nrf2 Tyr568 phosphorylation → nuclear export signal exposed → Nrf2 ejected from nucleus before completing target gene transcription. With GSK-3β inactivated, Fyn-mediated Nrf2 nuclear export is blocked and Nrf2 accumulates in the nucleus. Second, Semax’s melanocortin receptor arm: MC4R/Gs/cAMP/PKA → CREB Ser133 phosphorylation → CREB-binding elements in the NFE2L2 (Nrf2 gene) promoter → increased Nrf2 protein synthesis, amplifying the upstream pool available for nuclear entry. Once in the nucleus, Nrf2 heterodimerizes with sMaf proteins → the Nrf2/sMaf complex binds ARE (antioxidant response elements, core consensus 5′-TGAnnnnGC-3′) in the promoters of four principal antioxidant effector genes: HO-1 (heme oxygenase-1 → degrades pro-oxidant free heme; the products biliverdin and CO both have independent anti-inflammatory and anti-apoptotic signaling roles in ischemic neurons), NQO1 (NAD(P)H:quinone oxidoreductase 1 → obligatory two-electron reduction of quinone species, bypassing the semiquinone radical intermediate that would otherwise undergo redox cycling to generate O₂•⁻), SOD2 (manganese superoxide dismutase → converts mitochondrial matrix O₂•⁻ → H₂O₂ at the site of superoxide production during ischemia-reperfusion, reducing the most reactive ROS generated in Complex I/III forward electron leak), and GCL (glutamate-cysteine ligase, rate-limiting enzyme in glutathione synthesis → elevates the GSH/GSSG ratio → glutathione peroxidase-mediated detoxification of H₂O₂ and lipid hydroperoxides). In ischemic penumbral tissue where massive ROS generation drives neuronal apoptosis, this coordinated Nrf2-driven antioxidant enzyme upregulation provides enzymatic neutralization at each stage of the ROS cascade — a mechanistic basis for the neuroprotection observed in Semax-treated ischemia models that is distinct from and additive with its BDNF-mediated synaptic repair and anti-inflammatory effects. This is why it became standard of care for acute stroke recovery in Russian neurology.

ACTH receptor independence: Despite structural derivation from ACTH, Semax does not activate the HPA axis or cause cortisol elevation at clinical doses — a deliberate design feature. The authors specifically engineered out the cortisol-stimulating activity while retaining the CNS-active fragment. The molecular explanation lies in receptor selectivity determinants that are unique to the adrenal melanocortin receptor. ACTH’s HPA-activating activity is mediated exclusively through MC2R (melanocortin-2 receptor), which is expressed in the adrenal cortex and is the sole receptor driving the steroidogenic cascade — MC2R/Gs → adenylyl cyclase → cAMP → PKA → phosphorylation of StAR (steroidogenic acute regulatory protein) → cholesterol transfer into inner mitochondrial membrane → CYP11A1 (cytochrome P450 side-chain cleavage enzyme) → pregnenolone → cortisol. What makes MC2R distinctive among the five melanocortin receptor subtypes (MC1R–MC5R) is its stringent structural requirement for the N-terminal residues of ACTH (positions 1–5: Ser-Tyr-Ser-Met-Glu) in addition to the shared core HFRW motif (positions 6–9). These N-terminal residues are required for productive engagement with MC2R’s extracellular loop 2 (ECL2) and transmembrane domain geometry — without them, the peptide cannot stabilize the receptor in the active Gs-coupling conformation. All other melanocortin receptors (MC1R, MC3R, MC4R, MC5R) are activated by the shared HFRW core alone and do not require the N-terminal Ser-Tyr-Ser extension. Semax is derived from ACTH positions 4–7 (Met-Glu-His-Phe, with the Pro-Gly-Pro extension providing pharmacokinetic stability) and entirely lacks positions 1–3 (Ser-Tyr-Ser) as well as the full N-terminal flanking sequence that MC2R requires. Semax’s sequence therefore cannot engage MC2R at all — the adrenal steroidogenic cascade is structurally inaccessible. Instead, Semax’s HFRW-containing fragment selectively engages MC3R and MC4R expressed in the hypothalamus, limbic system, prefrontal cortex, and brainstem — the same receptors responsible for its dopaminergic (VTA/TH/DARPP-32), serotonergic (DRN/TPH2/CREB), and neuroprotective (Nrf2/HO-1/SOD2) effects described above. This receptor selectivity is not an incidental property but the structural consequence of the fragment’s position in the ACTH sequence: the cognitive/trophic-active core (positions 4–10) and the adrenal-stimulating determinants (positions 1–5 in combination with 6–9) are separable, and Semax retains only the former.

Pro-Gly-Pro metabolite: Semax is partially cleaved in vivo to Pro-Gly-Pro (PGP), which has independent anti-inflammatory activity in the CNS — specifically protecting extracellular matrix components from MMP (matrix metalloproteinase) degradation and reducing neuroinflammatory cytokine production (Stavchansky et al., 2011). The mechanism involves two converging arms. Directly, PGP’s consecutive Pro-Pro sequence creates exceptional backbone rigidity — the pyrrolidine rings of adjacent prolines cannot adopt the φ/ψ angles required to dock into the MMP catalytic cleft, which is optimized for flexible substrate backbones. MMP active sites coordinate catalytic zinc via the HEXXH motif and require substrate P1/P1’ residues to adopt an extended β-strand geometry; Pro-Pro at those positions creates steric occlusion of the catalytic zinc, functioning as a competitive active-site blocker against MMP-2, MMP-8, and MMP-9 — all of which are upregulated in neuroinflammatory tissue. Indirectly, PGP binds CXCR1 and CXCR2 (the IL-8 receptor family expressed on neutrophils) as a matrikine-derived partial agonist/competitive antagonist. Under normal conditions, IL-8 (CXCL8) binds CXCR1/CXCR2 → Gαi suppresses cAMP → Gβγ activates PI3K-γ/PKC → PKC-mediated phosphorylation of p47^phox and p67^phox subunits → NADPH oxidase (NOX2) complex assembly → oxidative burst. PGP occupancy of CXCR1/CXCR2 blunts this cascade: reduced PI3K-γ/PKC activation → incomplete p47^phox phosphorylation → attenuated NOX2 assembly → less superoxide and H₂O₂ production. Because ROS directly activate pro-MMP zymogens by oxidizing the thiol-zinc coordinate bond (cysteine switch mechanism), PGP’s suppression of NADPH oxidase secondarily reduces MMP-8 and MMP-9 conversion from zymogen to active form in the pericellular ECM. This dual-arm PGP mechanism — direct zinc-site steric occlusion plus indirect neutrophil degranulation suppression — explains how a tripeptide cleavage product contributes meaningfully to the parent compound’s neuroprotective profile. This adds an additional layer of neuroprotective activity to the parent compound’s direct effects.

Semax in Stroke Recovery: What the Data Actually Shows

The stroke recovery data is Semax’s most compelling evidence base. Russian clinical trials compared Semax plus standard care to standard care alone in acute ischemic stroke patients.

Gusev et al. (1997): 150-patient randomized trial in acute ischemic stroke. Semax-treated patients showed significantly greater reduction in neurological deficit scale scores at 10 days compared to controls. Functional independence measures at 30 days were significantly better in the Semax group. No serious adverse events attributed to Semax.

Gusev et al. (2018): Larger multicenter Russian trial confirming the 1997 results and characterizing the dose-response relationship. Optimal outcomes observed at 12 mcg/kg/day for 10 days of intranasal administration (Gusev et al., 2018).

Optic nerve protection: Semax has been studied in optic neuritis and glaucoma-related optic neuropathy, showing evidence of neuroprotective effects on retinal ganglion cells. The BDNF-TrkB pathway is critical for retinal ganglion cell survival; understanding why requires examining the unique anatomy that makes retinal ganglion cells (RGCs) the CNS neuron population most vulnerable to the specific insults involved in glaucoma and optic neuritis.

RGCs are the projection neurons whose unmyelinated axons converge at the optic disc and pass through the lamina cribrosa — a fenestrated collagen/elastin meshwork within the sclera — before acquiring myelin and becoming the optic nerve proper. This lamina cribrosa passage is the critical vulnerability point: the fenestrations are structurally rigid, and elevated intraocular pressure (IOP) in glaucoma exerts mechanical stress directly at these scleral openings, compressing the unmyelinated axon bundles at their point of least structural support. This compression has two converging consequences. First, it disrupts axoplasmic transport: the high-volume microtubule-dependent bidirectional flow of organelles, vesicles, and trophic signals that sustains RGC axons over their long projection to the lateral geniculate nucleus and superior colliculus (SC) is mechanically obstructed at the lamina cribrosa — pressure at this anatomical bottleneck creates a “dam” effect that blocks retrograde transport particularly severely, because retrograde cargo (carried by dynein/dynactin motor complexes toward the soma) must travel against the pressure gradient into the higher-IOP intraocular compartment. Second, and more specifically relevant to Semax’s mechanism: retrograde BDNF transport from the SC to RGC somas is blocked. RGCs depend substantially on target-derived BDNF produced in the SC and superior colliculus for trophic maintenance — this is the same retrograde neurotrophin transport mechanism that all long-projection CNS neurons (including corticospinal, nigrostriatal, and cholinergic basal forebrain neurons) use, but RGCs are uniquely exposed because their primary bottleneck is the IOP-pressurized lamina cribrosa rather than a diffuse ischemic zone.

When retrograde SC-derived BDNF transport is blocked, the sequence proceeds through a well-characterized apoptotic cascade: BDNF deprivation at the RGC soma → TrkB receptor is not activated at tyrosine 490/785 → PI3K/Akt signaling falls → BAD Ser136 phosphorylation drops (Akt no longer phosphorylates BAD, which had been sequestered by 14-3-3 proteins) → BAD dephosphorylates and translocates to the outer mitochondrial membrane → heterodimerizes with Bcl-xL and Bcl-2, titrating them away from Bax/Bak → Bax oligomerization at the OMM → mitochondrial outer membrane permeabilization → cytochrome c release → apoptosome assembly (Apaf-1/cytochrome c/dATP) → caspase-9 activation → executioner caspase-3 cleavage → RGC apoptosis and permanent axon loss.

Semax intervenes at the upstream supply deficit: intranasal Semax crosses into retinal tissue (intraocular penetration has been documented at measurable concentrations) and stimulates BDNF expression via the same CREB Ser133/BDNF exon IV promoter mechanism that operates in cortical and hippocampal tissue. BDNF upregulated locally within retinal glia (Müller cells and astrocytes) and RGC somas themselves provides an autocrine and paracrine TrkB activation source that partially substitutes for the transport-deprived retrograde SC-derived supply: locally-produced BDNF binds RGC surface TrkB → restores PI3K/Akt/BAD Ser136 phosphorylation → maintains Bax sequestration by Bcl-2/Bcl-xL → cytochrome c is retained → caspase cascade is not initiated. The clinical relevance is that this mechanism is most powerful as an adjunct to IOP-lowering therapy (which addresses the mechanical block) rather than a replacement for it — Semax-driven local BDNF cannot fully substitute for the SC-to-RGC trophic axis if axoplasmic transport remains completely blocked by severe IOP elevation, explaining why neuroprotective benefit in glaucoma studies is observed predominantly in patients with partially-controlled rather than uncontrolled IOP.

N-Acetyl-Semax: The Modified Version

A commonly available variant in the research chemical market is N-Acetyl-Semax — Semax with an acetyl group added to the N-terminus of the peptide. The acetylation increases lipophilicity (fat-solubility), which is claimed to improve CNS penetration relative to unmodified Semax. The molecular basis for this improvement operates through two distinct pharmacokinetic mechanisms. First, standard Semax carries a free α-amino group at Met¹ (-NH₃⁺ at physiological pH 7.4) — a permanently charged species that increases aqueous solubility but impedes passive transcellular diffusion through the lipid bilayers of nasal mucosal epithelium and the blood-brain barrier. N-terminal acetylation converts this free amine to a neutral acetamide bond (-NHCOCH₃), eliminating the positive charge and raising the calculated logP by approximately 0.5–1.0 units — sufficient to meaningfully improve passive membrane permeability without compromising the peptide’s aqueous reconstitution properties. Second, acetylation provides exopeptidase resistance: methionine aminopeptidase (MetAP) requires a free α-amine for substrate recognition and cannot cleave when the amine is acetylated. In nasal mucosa and cerebrospinal fluid, MetAP-mediated removal of Met¹ from standard Semax exposes the Glu²-His³-Phe⁴ sequence to further aminopeptidase N and related exopeptidase attack; acetylation blocks this initiating cleavage and thereby extends the intact half-life of the peptide in both compartments. The combination of faster membrane entry and slower enzymatic degradation — rather than either mechanism alone — explains the longer reported duration of effect relative to standard Semax: more intact peptide reaches TrkB, D1/D2, and serotonin receptors in CNS target regions and persists there longer per intranasal dose.

The evidence for N-Acetyl-Semax specifically is limited — most published research uses unmodified Semax. Self-experimenting users frequently report that N-Acetyl-Semax produces more noticeable and longer-duration effects than standard Semax, consistent with the enhanced CNS penetration hypothesis. However, the two have not been compared in controlled human trials.

N-Acetyl-Semax Amidate (the amidated version) adds C-terminal amidation (–CONH₂ replacing –COO⁻) on top of N-terminal acetylation — the same C-terminal modification carried natively by α-MSH (the endogenous melanocortin peptide whose ACTH(4-7) core Semax derives from), as well as GnRH, CRH, vasopressin, and NPY; its functional rationale is dual. First, carboxypeptidase resistance: serum carboxypeptidase N and tissue carboxypeptidase A require the substrate’s terminal carboxylate (–COO⁻) to coordinate the active-site zinc ion and position the scissile carbonyl for nucleophilic attack; replacing –COO⁻ with –CONH₂ eliminates this recognition element, making the amidated C-terminus an unproductive substrate regardless of side-chain identity — extending half-life by protecting against C-terminal degradation simultaneously with N-acetylation’s N-terminal protection, shielding both ends of the molecule at once. Second, MC3R/MC4R binding geometry: native α-MSH carries a C-terminal Val¹³-NH₂ amide, and structure-activity studies on melanocortin peptides show that the terminal amide nitrogen acts as an H-bond donor at a polar residue in MC4R’s transmembrane domain binding pocket — a contact unavailable to the negatively charged carboxylate (which introduces electrostatic repulsion at the same position); removing the amide and substituting a free carboxylate reduces MC4R binding affinity approximately 5–10× in α-MSH analogs, which is the molecular basis for the claim that amidation improves potency relative to non-amidated Semax. The combination of N-acetylation (blocks MetAP-mediated N-terminal cleavage and improves membrane permeability) with C-terminal amidation (blocks carboxypeptidase degradation and restores native MC receptor binding geometry) makes this form pharmacokinetically the most protected. This version has even less published research than N-Acetyl-Semax.

For research context: the modified versions may be more potent per microgram but the dose-response curve is less well-characterized. Start with lower doses when switching from standard Semax to modified versions.

Administration and Dosing

Like Selank, Semax is administered intranasally in clinical use. Typical intranasal dosing in Russian clinical practice is 200–900 mcg, 1–3 times daily for cognitive enhancement. Stroke recovery protocols use higher doses (12 mcg/kg/day, administered in clinical settings).

Duration of action per dose is somewhat shorter than Selank — approximately 2–4 hours for intranasal administration. Some users administer twice daily (morning and early afternoon) to maintain cognitive effect through the working day.

Side Effects

Semax’s side effect profile is generally favorable but more stimulating than Selank:

Possible increased energy and alertness (a feature or side effect depending on context)
Occasional mild anxiety or overstimulation at higher doses — particularly in individuals prone to anxiety
Transient nasal irritation
No sedation, withdrawal, or dependence documented
Long-term safety data in healthy Western users with chronic use is limited

Selank vs. Semax: Side-by-Side Comparison

Feature	Selank	Semax
Origin	Tuftsin analogue	ACTH(4-7) analogue
Primary effect	Anxiolytic + mild nootropic	Nootropic + neuroprotective
Subjective quality	Calming, reduces anxiety	Stimulating, increases focus and motivation
BDNF upregulation	Moderate	Strong
Dopaminergic effect	Mild	Pronounced
Best studied for	Anxiety disorders, neurasthenia	Stroke recovery, cognitive enhancement
Sedation risk	None	None
Anxiety risk	Low (reduces anxiety)	Low to moderate at high doses
Addiction/withdrawal	Not documented	Not documented
Russian approval	Yes (anxiety/nootropic)	Yes (stroke/neurological)
FDA status	Not approved; Category 3 (not compoundable)	Not approved; Category 3 (not compoundable)
Modified versions	None commonly used	N-Acetyl-Semax; N-Acetyl-Semax Amidate
Stroke evidence	Indirect (neuroprotective mechanism)	Direct RCT evidence

Post-Neurological Injury: TBI, Concussion, and the Neuroprotection Gap

One of the reasons Semax (and to a lesser extent Selank) attracts interest from Western researchers is the near-absence of evidence-based pharmacological options for post-concussive syndrome (PCS) and mild-to-moderate traumatic brain injury (mTBI). Understanding this context helps explain why these peptides occupy a credible research niche despite their limited Western validation.

The Secondary Injury Cascade

TBI causes injury in two distinct phases. The primary injury is the mechanical damage from impact — axonal shearing, contusion, and initial hemorrhage. This is largely unmodifiable after the fact. The secondary injury cascade unfolds over hours to weeks following the initial trauma and is where pharmacological intervention is theoretically possible:

Neuroinflammation: Microglia activate within minutes of injury, releasing pro-inflammatory cytokines — IL-1β, TNF-α, and IL-6 — that amplify cellular damage beyond the initial impact zone. In moderate and severe TBI, microglial activation can persist for years (Ramlackhansingh et al., 2011, PMID: 22048143).
Excitotoxicity: Trauma triggers massive glutamate release, causing pathological NMDA receptor activation and Ca²⁺ influx that overwhelms mitochondrial buffering and initiates apoptotic pathways. The downstream cascade is now well characterized: excessive cytosolic Ca²⁺ is taken up into mitochondria through the mitochondrial Ca²⁺ uniporter, triggering opening of the mitochondrial permeability transition pore (mPTP) — a non-selective inner membrane channel whose sustained opening collapses the electrochemical gradient required for ATP synthesis. mPTP opening releases cytochrome c into the cytosol, where it forms the apoptosome with Apaf-1 and procaspase-9; activated caspase-9 cleaves executioner caspase-3, initiating the apoptotic nuclear fragmentation and cell death program. Simultaneously, excess cytosolic Ca²⁺ activates calpain — a calcium-dependent cysteine protease — which degrades spectrin, neurofilament, and microtubule-associated proteins within axons, causing axonal transport failure and the cytoskeletal collapse characteristic of diffuse axonal injury. Ca²⁺-activated calcineurin (protein phosphatase 2B) additionally dephosphorylates dynamin-related protein 1 (Drp1), driving it from cytosol to mitochondrial outer membrane where it oligomerizes to execute mitochondrial fission — fragmenting the mitochondrial network and further compromising bioenergetic capacity. These interlocked pathways (mPTP → bioenergetic failure and cytochrome c release; calpain → structural disintegration; Drp1 → mitochondrial fragmentation) collectively account for the progressive neuronal death that unfolds over hours to days in the secondary injury window.
BDNF depletion: Brain-derived neurotrophic factor levels fall acutely after TBI. BDNF-TrkB signaling is essential for synaptic repair, dendritic remodeling, and hippocampal neurogenesis — the processes that enable cognitive recovery. Reduced BDNF correlates with worse functional outcomes (Rostami et al., 2014, PMID: 24594074). Three mechanistically distinct pathways converge to suppress BDNF transcription after TBI. The first is REST/NRSF-mediated repressor derepression: glutamate excitotoxicity drives sustained NMDA receptor Ca²⁺ influx → cytosolic Ca²⁺ overload activates calmodulin-kinase IV (CaMKIV), which phosphorylates and inactivates HDAC4/5 (causing their nuclear export), but simultaneously excessive Ca²⁺ loading shifts CaMKIV into a constitutively active, substrate-promiscuous state that phosphorylates the co-repressor SIN3A, stabilizing the REST/NRSF–SIN3A–HDAC1/2 complex at the RE1 silencer element within the BDNF promoter IV region; REST/NRSF binding at RE1 recruits CoREST and LSD1 (KDM1A) → histone H3K4me2/me1 demethylation → promoter IV transcriptional silencing. The second is glucocorticoid receptor (GR)/nGRE-driven HDAC3 recruitment: TBI triggers a rapid hypothalamic-pituitary-adrenal stress response — CRH → ACTH → adrenal cortisol — that peaks within 60–90 minutes of injury. Elevated cortisol → GR nuclear translocation → binding at negative glucocorticoid response elements (nGREs) identified in the BDNF exon IV promoter region; liganded GR at nGREs recruits nuclear receptor co-repressor 1 (NCoR1) and HDAC3, producing local H3 deacetylation and reduced RNA polymerase II processivity at the BDNF transcription start site. The third is IDO1/kynurenine → STAT3/HDAC3 chromatin compaction: M1-polarized microglia release TNF-α and IL-6 that activate neuronal STAT3 via JAK1/2; simultaneously, IL-6-driven indoleamine 2,3-dioxygenase 1 (IDO1) upregulation shifts tryptophan catabolism toward kynurenine and 3-hydroxykynurenine rather than serotonin, further reducing serotonin-dependent (5-HT2A/Gαq/PKC/CREB) BDNF exon I/II transcription; STAT3 itself recruits HDAC3 to the BDNF promoter through a STAT3–HDAC3 direct interaction mapped to the STAT3 N-terminal domain, adding a third layer of histone deacetylation-driven BDNF gene repression. These three mechanisms — REST/NRSF silencer binding, GR/nGRE/HDAC3 cortisol response, and IDO1/STAT3/HDAC3 neuroinflammatory suppression — explain why BDNF falls sharply even in brain regions remote from the primary injury contusion, and why the depletion persists for days to weeks beyond the acute excitotoxic phase.
Blood-brain barrier disruption: TBI itself temporarily disrupts the BBB, potentially improving CNS peptide access for a window post-injury — a pharmacokinetic consideration that alters the usual delivery challenge. Three mechanistically distinct processes converge to produce this disruption. MMP-mediated tight junction degradation: within minutes to hours of impact, activated microglia and infiltrating neutrophils release MMP-2 and MMP-9 (matrix metalloproteinases with Zn²⁺-coordinated HEXXH active sites) into the perivascular space. These gelatinases cleave the extracellular domains of claudin-5 — the dominant tight junction strand protein at the BBB, responsible for paracellular charge selectivity — and occludin at extracellular loop sites outside their transmembrane anchors; simultaneously, MMP-9 degrades the basement membrane collagen IV scaffold (at Gly-X-Y triple-helix cleavage sites) and laminin that structurally anchor the endothelial layer to the parenchyma. The combined loss of tight junction protein integrity and basement membrane support increases paracellular permeability to molecules up to approximately 150 kDa — substantially above the ~500 Da threshold of the intact BBB. Pericyte-endothelium uncoupling: mechanical shear and excitotoxic glutamate release at cerebral microvessels disrupts the constitutive PDGF-B/PDGFR-β trophic signaling between brain capillary endothelial cells and their ensheathing pericytes; pericytes deprived of this survival signal undergo apoptosis via Bax/cytochrome c → caspase-3 within hours of TBI. Pericytes are the principal inducers of endothelial claudin-5 and ZO-1 (zona occludens-1) expression through contact-dependent signaling: loss of pericyte contact removes the PDGFR-β → Src/Y397-FAK → PI3K/Akt → NF-κB trophic input that sustains claudin-5 transcription, further reducing tight junction protein levels independently of direct MMP cleavage. AQP4 de-polarization and astrocyte-endothelium uncoupling: the same ROS burst (superoxide, ONOO⁻, •OH) that drives lipid peroxidation of neuronal membranes also oxidizes AQP4 Cys178 on astrocytic water channels, altering AQP4 trafficking from perivascular endfeet membrane domains — where AQP4 normally concentrates to maintain the tight paravascular compartment — to non-endfeet astrocyte surfaces. AQP4 depolarization impairs the paravascular water flux and removes astrocyte-derived trophic signals (including sonic hedgehog and Wnt ligands) that astrocytic endfeet normally deliver to adjacent endothelial cells to maintain tight junction integrity. The collapse of this tri-cellular neurovascular unit (NVU) support produces a BBB opening that is transient: approximately 24–72 hours in mild TBI and extending longer in moderate-to-severe injury as the MMP surge resolves, pericyte turnover re-establishes contact signaling, and tight junction proteins are resynthesized. This time-limited permeability window explains why acute post-TBI peptide administration windows in the preclinical literature consistently produce better outcomes than delayed administration — the pharmacokinetic access advantage narrows substantially as the NVU reseals.

Why Conventional Medicine Offers Little

No FDA-approved pharmacological treatment exists specifically for acute mTBI or PCS. Current management is supportive: rest, gradual return to activity, symptom management with over-the-counter analgesics for headache, and cognitive behavioral therapy for mood symptoms. For patients with chronic PCS — persistent headache, cognitive fog, sleep disruption, emotional dysregulation lasting months to years — treatment options remain largely off-label and poorly effective.

This gap drives patients and researchers toward compounds with plausible neuroprotective mechanisms, even before large Western RCTs exist.

How Semax Addresses the Secondary Cascade

Semax’s mechanism maps onto multiple arms of the secondary injury cascade:

BDNF restoration: Semax’s most potent documented effect is BDNF upregulation in the hippocampus and prefrontal cortex. In ischemic injury models that share pathophysiological features with TBI — excitotoxicity, oxidative stress, BBB disruption — Semax consistently restores BDNF toward baseline and accelerates recovery of dendritic complexity (Inozemtseva et al., 2008, PMID: 18841804). The molecular rationale for applying this finding to TBI is mechanistically sound.

Neuroinflammation reduction: Semax reduces IL-1β and TNF-α expression in ischemic brain tissue in animal models. The Pro-Gly-Pro metabolite from Semax cleavage independently inhibits matrix metalloproteinases (MMPs) — proteases that degrade the extracellular matrix around neurons during neuroinflammation, contributing to axonal damage. MMP inhibition is an active target in TBI pharmacological development.

Excitotoxicity attenuation: Semax has been shown to reduce glutamate-induced excitotoxic cell death in cortical neuron cultures through three converging molecular pathways, each addressing a distinct point in the excitotoxic Ca²⁺ cascade described in the secondary injury section above.

The primary route is BDNF/TrkB → Akt → mitochondrial Ca²⁺ overload protection. NMDA receptor activation during TBI or ischemia generates a Ca²⁺ influx that is rapidly taken up by mitochondria through the mitochondrial Ca²⁺ uniporter, triggering mPTP opening, cytochrome c release, and caspase-3 activation. Semax-upregulated BDNF/TrkB/PI3K/Akt intercepts this cascade at two points on the mitochondrial outer membrane: (1) BAD Ser136 phosphorylation → BAD/14-3-3σ cytoplasmic sequestration → Bcl-2/Bcl-xL remain free to close Bax/Bak pores even under elevated intramitochondrial Ca²⁺; and (2) GSK-3β Ser9 phosphorylation → GSK-3β inactivation → cyclophilin D (CypD) dephosphorylation at Ser31. GSK-3β normally phosphorylates CypD at Ser31, sensitizing CypD to matrix Ca²⁺ and lowering the Ca²⁺ threshold for mPTP opening; with GSK-3β inactivated by Akt, CypD reverts to its lower-sensitivity conformation — raising the Ca²⁺ threshold for mPTP engagement and reducing the probability of pore opening under the same excitotoxic Ca²⁺ load. This Akt → GSK-3β → CypD → mPTP axis is the primary mechanism by which BDNF upregulation reduces glutamate-induced excitotoxic cell death in cortical neuron cultures.

A second pathway operates through Nrf2/HO-1/CO → NMDA receptor conductance reduction. Semax’s BDNF/TrkB → Akt → GSK-3β-Fyn inhibition activates nuclear Nrf2, which drives HO-1 (heme oxygenase-1) expression in stressed neurons. HO-1 degrades pro-oxidant free heme to produce carbon monoxide (CO) as a neuromodulatory gasotransmitter. CO inhibits NMDA receptors by binding the His657 residue on the GluN1 subunit at the extracellular N-terminal domain, partially mimicking Zn²⁺‘s voltage-independent inhibitory mechanism — reducing Ca²⁺ conductance per channel opening without complete receptor blockade. This CO-mediated NMDA receptor dampening is spatially confined to the synapses experiencing the greatest heme release and HO-1 induction (hypoxic, high-excitatory-load synapses) rather than globally suppressing NMDA-receptor-dependent LTP, which explains why HO-1 activation protects against excitotoxicity without impairing the baseline NMDA receptor signaling that Semax’s BDNF/TrkB arm is simultaneously supporting for synaptic repair.

The third pathway is Nrf2/SOD2/GCL → reduced ROS → prevention of NMDA receptor redox potentiation. Under ischemia and TBI, excessive superoxide and H₂O₂ oxidize Cys399 and Cys744 on the GluN1 subunit, inducing a conformational shift that increases channel open probability — redox-dependent NMDA current potentiation that amplifies Ca²⁺ influx beyond what glutamate concentration alone would produce. Semax’s Nrf2 → SOD2 (mitochondrial O₂•⁻ → H₂O₂ conversion at the production site) + GCL → elevated GSH/GSSG → glutathione peroxidase-mediated H₂O₂ reduction maintains GluN1 Cys399 and Cys744 in the reduced state, blocking the redox gate from potentiating NMDA current. The net effect is that NMDA receptor Ca²⁺ conductance remains proportional to glutamate occupancy rather than being amplified by the pro-oxidant microenvironment that TBI generates — attenuating excitotoxicity specifically in the tissue compartments with the highest ROS burden without requiring pharmacological NMDA receptor blockade and its attendant cognitive impairment. The mechanistic basis for that cognitive impairment is that the same NMDA receptor Ca²⁺ influx that becomes destructive during excitotoxicity is the obligate substrate for Hebbian long-term potentiation (LTP) at physiological glutamate concentrations. During LTP induction, coincident pre- and postsynaptic activity depolarizes the postsynaptic membrane sufficiently to relieve the voltage-dependent Mg²⁺ block of the NMDA channel pore; the resulting Ca²⁺ entry activates CaMKII (Ca²⁺/calmodulin-dependent protein kinase II), which autophosphorylates at Thr286 → achieving constitutive kinase activity that outlasts the Ca²⁺ transient → CaMKII phosphorylates GluA1 at Ser831 (the principal AMPA receptor subunit), increasing single-channel conductance and driving GluA1-containing AMPA receptor insertion into the postsynaptic density → LTP expression and memory consolidation. Pharmacological NMDA blockade (competitive antagonists like AP5, or uncompetitive channel blockers like ketamine and high-dose memantine) eliminates this Ca²⁺ entry indiscriminately — CaMKII autophosphorylation does not occur regardless of whether the context is pathological excitotoxicity or physiological synaptic learning, GluA1-Ser831 remains unphosphorylated, AMPA receptor potentiation is blocked, and memory consolidation is impaired. Semax’s Nrf2/HO-1/CO and ROS-scavenging strategy intercepts only the pathological amplification components — the GluN1 Cys399/Cys744 redox potentiation that excess ROS introduces and the CO-mediated partial dampening that is spatially confined to the most ROS-burdened synapses — without disturbing the Mg²⁺-dependent coincidence-detection gating mechanism that LTP requires, thereby preserving NMDA receptor cognitive function while eliminating the excitotoxic amplification that TBI superimposes on it.

In the Russian stroke literature — which is methodologically the closest available clinical model to TBI’s secondary ischemic component — Semax consistently reduced neurological deficit progression and improved functional recovery when administered acutely. The Gusev 2018 multicenter trial (PMID: 29798983) showed these outcomes at clinical doses.

No Western TBI-specific RCT has been conducted with Semax. The existing evidence is mechanistic plausibility plus stroke-model clinical data. This is worth being explicit about: a person with TBI seeking Semax should understand that what exists is biological rationale, not a body of TBI-specific human trials.

Selank’s Role: The Psychiatric Aftermath

TBI survivors frequently develop anxiety, irritability, and depression — not solely from psychological reaction to injury, but from direct disruption of the limbic circuitry (amygdala, prefrontal-limbic connectivity, serotonin/GABA signaling pathways). An estimated 25–50% of mTBI patients develop clinically significant anxiety or depressive symptoms (Bryant, 2011, PMID: 21219304).

The circuit disruption follows two converging anatomical routes. First, parvalbumin (PV) and somatostatin (SST) GABAergic interneurons in the prefrontal cortex are disproportionately vulnerable to TBI-induced excitotoxicity. PV interneurons express high densities of NR2B-containing NMDA receptors — a subunit configuration that confers high Ca²⁺ permeability and low Mg²⁺ block at resting membrane potential. During the post-TBI glutamate surge, Ca²⁺ influx through NR2B-NMDARs exceeds the buffering capacity of PV interneuron mitochondria and calbindin stores. The resulting cytosolic Ca²⁺ overload activates calpain-1 and calpain-2, which cleave spectrin, MAP2, and NF-L cytoskeletal proteins → PV interneuron dendritic retraction and cell death. PV interneurons normally provide perisomatic feedforward inhibition of pyramidal neurons in layers II/III and V; their loss selectively disinhibits pyramidal cell axonal output to the basolateral amygdala (BLA). Elevated excitatory drive from prefrontal pyramidal projections paradoxically destabilizes BLA microcircuits — increased BLA principal neuron activity → heightened CeA (central amygdala) output → elevated fear/anxiety tone. The loss of the PFC’s GABAergic “volume control” over the amygdala is a primary structural substrate for post-TBI hypervigilance and generalized anxiety.

Second, white matter injury to the uncinate fasciculus — the principal bidirectional fiber tract connecting orbitofrontal and ventromedial PFC to the amygdala — imposes a conduction delay and reduced bandwidth on top-down regulatory signals. Diffuse axonal injury from rotational shear forces disrupts neurofilament organization at nodes of Ranvier, reduces axonal conduction velocity, and triggers Wallerian degeneration in the most vulnerable small-diameter fibers through a well-characterized molecular cascade: axonal shearing disconnects the distal axon segment from the soma → NMNAT2 (nicotinamide mononucleotide adenylyltransferase 2), which is constitutively transported from the soma to the distal axon via fast anterograde transport and is required to synthesize NAD+ from NMN (nicotinamide mononucleotide), is no longer replenished in the distal stump → NMNAT2 is rapidly degraded by the proteasome in the disconnected segment (half-life ~4–6 hours without soma-derived replenishment) → NMN accumulates in the distal axon cytoplasm as the synthesis pathway is uncoupled → accumulated NMN binds the ARM domain of SARM1 (Sterile Alpha and HEAT/Armadillo Motif protein 1) at an allosteric regulatory site, relieving SARM1’s constitutive autoinhibition → SARM1’s TIR (Toll/IL-1 receptor homology) domain dimerizes and activates its intrinsic NADase activity → SARM1 cleaves NAD+ into ADP-ribose, cyclic ADP-ribose (cADPR), and nicotinamide (Nam) in a single enzyme cycle → catastrophic NAD+ depletion within the distal axon cytoplasm → mitochondrial ATP synthesis fails (requires NAD+ for Complex I and citric acid cycle dehydrogenases) → axon bioenergetic collapse → Ca²⁺-dependent calpain activation (calpain requires ATP-dependent Ca²⁺ gradient maintenance; calpain cleaves spectrin, neurofilament heavy chain NF-H, and microtubule-associated protein MAP2) → cytoskeletal fragmentation → granular disintegration of the axon into short segments — the histological hallmark of Wallerian degeneration. This NMNAT2/SARM1/NAD+ pathway explains the stereotyped timing of Wallerian degeneration (hours after injury, not seconds) and why small-diameter axons with thinner myelin sheaths and lower mitochondrial density succumb before larger myelinated fibers under equivalent mechanical insult. With this tract attenuated, the PFC cannot mount its normal emotion-regulation response to BLA threat signals: extinction learning becomes impaired, and emotional reactivity to minor stressors fails to habituate. Together, PV interneuron loss (synaptic-level) and uncinate fasciculus injury (tract-level) produce the characteristic post-TBI phenotype of anxiety, irritability, and affective dysregulation through complementary mechanisms rather than independent ones.

Selank’s GABA-A modulation and BDNF upregulation address both the neurochemical imbalance contributing to post-TBI anxiety and the synaptic repair deficit contributing to cognitive sequelae. Its lack of sedation is particularly relevant in this population — TBI patients are already frequently impaired on processing speed; any sedating anxiolytic compounds this deficit. The benzodiazepine avoidance rationale is especially compelling post-TBI, where cognitive rehabilitation is a treatment goal that benzodiazepines directly undermine.

The Honest Evidence Assessment for TBI Indication

Evidence Type	Semax	Selank
Animal TBI/ischemia models	Multiple positive studies	Limited, mostly ischemia
Human stroke RCT	Yes (Russian, n=150+)	No
Human TBI-specific RCT	None	None
Mechanistic plausibility	Strong (BDNF, neuroinflammation, MMP)	Moderate (GABA, BDNF, mood)
FDA-recognized evidence	None	None

The pattern for TBI is one of scientific rationale and preclinical evidence without the clinical trial infrastructure to confirm it. For Western patients with PCS, this positions Semax at the intersection of “plausible hypothesis” and “no approved alternative” — a combination that drives real-world use without a corresponding evidence base to guide it.

The Shared Research Problem: Institutional Concentration

Both Selank and Semax share a structural limitation: the bulk of their published evidence comes from a single institutional ecosystem — the Institute of Molecular Genetics of the Russian Academy of Sciences and affiliated centers.

Scientific claims gain credibility through independent replication: when different research groups, using different populations, different methodologies, and different funding sources, arrive at the same conclusions. When all significant evidence comes from one institution, even rigorous individual studies cannot carry the same epistemic weight as replicated findings.

This is not unique to Russian research — proximity bias and institutional concentration affect research in every country. But it means Western clinicians and patients should apply appropriate skepticism to claims that exceed the available evidence, while not dismissing a substantial body of published work that does suggest real effects.

The practical implication: the anxiety-reduction and cognitive-enhancement effects of Selank, and the neuroprotective and cognitive effects of Semax, are scientifically plausible and supported by multiple studies, but are not as robustly validated as GLP-1 receptor agonist weight loss effects or BPC-157 injury recovery data in animal models.

The European Phase 2 trial for Selank in anxiety disorders and mild cognitive impairment (referenced in the clinical trials guide) represents the most significant potential move toward independent Western validation. If that trial proceeds under EMA-recognized standards and produces positive results, the epistemic weight behind Selank’s anxiety efficacy will increase substantially.

Who Typically Seeks These Peptides and Why

In Western markets, Selank and Semax are sought primarily by:

Biohackers and nootropic researchers systematically testing peptides for cognitive enhancement
People with treatment-resistant anxiety who have found benzodiazepines problematic (sedation, dependence) and seek alternatives with a different mechanism
People with history of neurological events (TBI, concussion) interested in neuroprotective options not available in conventional medicine
High-performers interested in cognitive optimization without stimulant side effects

The current legal reality in the United States: Both peptides are sold as “research chemicals” by domestic and international vendors, but they cannot legally be prescribed, compounded, or clinically administered for human use in the U.S. under current FDA rules (Category 3 — not eligible for 503A/503B compounding). Anyone using Selank or Semax in the United States is doing so outside of clinical supervision, which removes important safety layers regardless of the compounds’ own safety profiles.

Quality and Sourcing: What to Actually Evaluate

For researchers in contexts where these compounds are legally accessible, sourcing quality is a critical safety consideration:

Certificate of Analysis (CoA): Reputable research chemical vendors publish third-party CoAs showing HPLC purity percentage (target: >99%), mass spectrometry confirmation (verifies the correct molecular weight = correct compound), and endotoxin testing (critically important for any injectable — bacterial endotoxin contamination causes fever and septic-like reactions far more commonly than purity issues).

HPLC purity vs. compound identity: A product can be 99% pure but be the wrong compound. Mass spec confirmation (showing the correct molecular weight) combined with HPLC purity is the minimum verification standard. Amino acid analysis or peptide sequencing provides additional confirmation.

Endotoxin testing: This is the most commonly skipped test by low-quality vendors and the most clinically important for injectable use. Endotoxins from gram-negative bacterial contamination during synthesis can cause significant reactions even when the peptide itself is pure. Look for LAL (Limulus Amebocyte Lysate) or rFC endotoxin test results on the CoA.

Cold-chain and storage: Peptides degrade with heat and moisture. Reputable vendors ship on dry ice or ice packs and store in controlled refrigeration. Products sitting in non-climate-controlled warehouses lose potency.

Intranasal vs. injectable sourcing: The purity and sterility requirements for intranasal use are less stringent than for subcutaneous injection. Intranasal administration has a natural filter effect and does not introduce substances directly into the bloodstream or tissue.

Choosing Between Them

The choice between Selank and Semax is primarily about target effect profile:

If the primary concern is anxiety, stress, or mood — and particularly if a calm, non-stimulating effect is desired — Selank is the more appropriate choice, with the better evidence base for this indication
If the primary concern is cognitive enhancement, focus, memory, or neuroprotection — and particularly if the person wants more noticeable stimulation and is not prone to anxiety — Semax tends to be preferred

Combined use: Some researchers report using both — Selank to moderate any Semax-induced overstimulation, and Semax to provide the primary cognitive-enhancing component. This combination exists in the biohacking community but has no formal clinical study basis. The interaction between the two is not characterized in human trials.

Summary: What the Evidence Actually Supports

Selank has good evidence for anxiety reduction without sedation or dependence in Russian clinical populations. The mechanism (GABA-A modulation, enkephalinase inhibition, BDNF upregulation) is plausible. Side effect profile is appealing. Independent Western replication is lacking but pending (European Phase 2 underway).
Semax has good evidence for neuroprotection in acute stroke and cognitive enhancement in clinical and animal models. Longer clinical use history in Russia than almost any other nootropic peptide. Better mechanistic diversity. Also lacks independent Western validation.

For patients or researchers in the West: both deserve to be understood accurately — neither dismissed as pseudoscience nor embraced as validated medicine. They represent compounds with genuine scientific interest and insufficient Western clinical validation. For cognitive enhancement within a clinical framework, peptides with stronger Western evidence — GH-releasing peptides for sleep and recovery, GLP-1 drugs for neurodegeneration — may offer a more established starting point.

References

Zozulya AA, Kost NV, Sokolov OYu, et al. Inhibition by selank of enkephalin-degrading enzymes as possible mechanism of anxiolytic action. Bull Exp Biol Med. 2001;131(5):468-70. PMID: 11550013
Zozulia AA, Neznamov GG, Siuniakov TS, et al. Efficacy and possible mechanisms of the neuropeptide anxiolytic selank in the therapy of generalized anxiety disorder. Zh Nevrol Psikhiatr Im S S Korsakova. 2008;108(4):38-48. PMID: 18454096
Inozemtseva LS, Dmitrieva VG, Povarova OV, et al. Intranasal administration of the peptide semax regulates BDNF and trkB expression in the rat hippocampus. Dokl Biol Sci. 2008;421:241-3. PMID: 18841804
Uchakina ON, Uchakin PN, Miasoedov NF, et al. Immunomodulatory effects of selank in patients with anxiety-asthenic disorders. Zh Nevrol Psikhiatr Im S S Korsakova. 2008;108(5):71-5. PMID: 18577961
Eremin KO, Kudrin VS, Saransaari P, et al. Semax, an ACTH(4-10) analogue with nootropic properties, activates dopaminergic and serotoninergic brain systems in rodents. Neurochem Res. 2005;30(12):1493-500. PMID: 16362768
Gusev EI, Skvortsova VI, Miasoedov NF, et al. Effectiveness of semax in acute period of hemispheric ischemic stroke (a multicenter randomized trial). Zh Nevrol Psikhiatr Im SS Korsakova. 2018;118(12):26-33. PMID: 29798983
Stavchansky VV, Yuzhakov VV, Sevan’kaeva LE, et al. Neuroprotective effect of semax and Pro-Gly-Pro peptide in a model of focal ischemia in rats. Exp Oncol. 2011. PMID: 20617398

Selank vs. Semax: A Researcher's Guide to Nootropic Peptides

Why Intranasal Delivery: The Blood-Brain Barrier Problem

What Are Nootropic Peptides?

Selank: Anxiety Without Sedation

What It Is

How Selank Works: The Mechanism

The Core Research Findings

Administration and Dosing

Side Effects

Semax: Neuroprotection and Cognitive Stimulation

What It Is

How Semax Works

Semax in Stroke Recovery: What the Data Actually Shows

N-Acetyl-Semax: The Modified Version

Administration and Dosing

Side Effects

Selank vs. Semax: Side-by-Side Comparison

Post-Neurological Injury: TBI, Concussion, and the Neuroprotection Gap

The Secondary Injury Cascade

Why Conventional Medicine Offers Little

How Semax Addresses the Secondary Cascade

Selank’s Role: The Psychiatric Aftermath

The Honest Evidence Assessment for TBI Indication

The Shared Research Problem: Institutional Concentration

Who Typically Seeks These Peptides and Why

Quality and Sourcing: What to Actually Evaluate

Choosing Between Them

Summary: What the Evidence Actually Supports

References

Read Next

Anti-Aging Peptide Protocols: A Framework for Evidence-Based Longevity Approaches

Anti-Aging Peptides: What the Research Actually Shows

AOD-9604 and Fragment 176-191: What the Clinical Trial Record Actually Shows

Ready to Apply These Insights?

Why Intranasal Delivery: The Blood-Brain Barrier Problem

What Are Nootropic Peptides?

Selank: Anxiety Without Sedation

What It Is

How Selank Works: The Mechanism

The Core Research Findings

Administration and Dosing

Side Effects

Semax: Neuroprotection and Cognitive Stimulation

What It Is

How Semax Works

Semax in Stroke Recovery: What the Data Actually Shows

N-Acetyl-Semax: The Modified Version

Administration and Dosing

Side Effects

Selank vs. Semax: Side-by-Side Comparison

Post-Neurological Injury: TBI, Concussion, and the Neuroprotection Gap

The Secondary Injury Cascade

Why Conventional Medicine Offers Little

How Semax Addresses the Secondary Cascade

Selank’s Role: The Psychiatric Aftermath

The Honest Evidence Assessment for TBI Indication

The Shared Research Problem: Institutional Concentration

Who Typically Seeks These Peptides and Why

Quality and Sourcing: What to Actually Evaluate

Choosing Between Them

Summary: What the Evidence Actually Supports

References

Stay Informed on Peptide Research

Read Next

Anti-Aging Peptide Protocols: A Framework for Evidence-Based Longevity Approaches

Anti-Aging Peptides: What the Research Actually Shows

AOD-9604 and Fragment 176-191: What the Clinical Trial Record Actually Shows

Ready to Apply These Insights?