Semax: brain-protecting nootropic drug approved in Russia
Synthetic peptide developed in Russia that protects the brain and sharpens thinking; approved in Russia and Ukraine for stroke and cognitive disorders; not FDA-approved.
A researcher, an agent, or an algorithm wrote down the sequence and picked a target to hit.
An AI model like OpenFold3 or AlphaFold built a 3D structure and scored how well it fits the binding site.
A second contributor repeated the computation on their own hardware and the scores matched.
FDA-tracked compound — synthesized for clinical/research use
Fork this card to add platform evidence →
FDA-tracked (reclassified Category 1, Feb 2026) — preclinical/clinical bioassay data exists
Fork this card to add platform evidence →
What this is
Semax is a synthetic heptapeptide developed in Russia in the 1980s as a brain-protective and cognitive-enhancing drug. Its sequence is Met-Glu-His-Phe-Pro-Gly-Pro (MEHFPGP) — the raw stored sequence represents the full active molecule; there are no hidden lipid conjugates or modifications on standard Semax. It was designed by Academician Nikolai Myasoedov and Professor Igor Ashmarin at the Institute of Molecular Genetics of the Russian Academy of Sciences, and has been approved as a prescription drug in Russia and Ukraine for ischemic stroke, cognitive disorders, and optic nerve atrophy. Outside the former Soviet states it has no regulatory approval; the FDA has not received a submission for Semax.
The central design goal was to capture the cognitive effects of the ACTH(4-10) peptide fragment without triggering the cortisol-releasing activity of full-length ACTH. The first four residues (Met-Glu-His-Phe) represent the ACTH(4-7) core that retains neurotrophin-stimulating activity; the appended Pro-Gly-Pro tail protects the peptide from rapid degradation, extending its functional half-life via the same glyproline stabilization strategy later used in the sister peptide Selank. Because the adrenal-axis-activating sequence resides in ACTH residues 1–24, Semax does not meaningfully raise cortisol — a distinction confirmed in pharmacological characterization studies.
History
Semax emerged from a Soviet-era research program seeking stable, non-hormonal ACTH analogs that retained the nootropic effects observed with short ACTH fragments. Myasoedov and Ashmarin's group at the Institute of Molecular Genetics designed and synthesized the heptapeptide in the 1980s; the compound was first described in scientific literature in 1991. The Pro-Gly-Pro glyproline tail was the key innovation — Eremin and colleagues (Neurochemical Research, 2005) noted that this fragment is responsible for the metabolic stability and relatively long duration of effect compared to native ACTH fragments. Semax was registered as a prescription medicine in Russia and has been on the Russian Vital and Essential Drugs List since 2011. It is manufactured by Geropharm and other Russian pharmaceutical firms. A 15-year retrospective on the compound's design and development was published by the original development group in a 1997 review. Outside Russia and the CIS, Semax reached Western nootropic communities via research-chemical channels from roughly the 2010s onward, though no Western regulatory body has received a formal submission.
What it does
Semax's primary documented biological effect is the upregulation of brain-derived neurotrophic factor (BDNF), a protein that supports the survival, growth, and plasticity of neurons. Dolotov and colleagues (Journal of Neurochemistry, 2006) showed that Semax specifically binds sites in rat basal forebrain — with a dissociation constant of 2.4 ± 1.0 nM — and that this binding triggers a substantial increase in BDNF protein levels. Shadrina and colleagues (Journal of Molecular Neuroscience, 2010) extended this finding to multiple brain regions, showing that a single intranasal dose elevates NGF and BDNF gene expression in hippocampus, frontal cortex, and retina within 30 minutes and that the effect persists for up to 24 hours — long after the peptide itself has cleared from plasma. This sustained neurotrophic signal is thought to underlie Semax's effects on learning, memory formation, and neurological recovery documented in rodent and Russian clinical studies.
Alongside BDNF, Semax modulates dopaminergic and serotonergic neurotransmission. Eremin and colleagues (Neurochemical Research, 2005) demonstrated activation of dopaminergic and serotonergic brain systems in rodents, consistent with the attention and mood-related effects reported in human studies. Semax also modulates immune gene expression in the post-ischemic brain — Medvedeva and colleagues (Molecular Genetics and Genomics, 2017) described broad transcriptomic changes in immune response pathways following Semax administration in a rat ischemia model, and these anti-inflammatory effects in the CNS are thought to contribute to the neuroprotective activity that forms the basis of its stroke indication in Russia.
Evidence
- Human: Moderate — primarily Russian clinical trials from the Institute of Molecular Genetics development program. Multiple Russian clinical trials document improved neurological recovery scores and disability outcomes in acute ischemic stroke patients versus standard care; these trials form the basis for Russian regulatory approval. One Russian study (PMID 11517472) reported effects in hemispheric ischemic stroke using neurological and electrophysiological endpoints. A Russian study in optic nerve disease patients found visual function preservation (PMID 10741256), supporting that approved indication. A controlled study in healthy adult volunteers (Glazova and colleagues, 2018) reported improvements in attention and working-memory measures; this is the strongest English-language human evidence for cognitive effects in non-clinical populations, though the study is smaller-scale and has not been independently replicated in Western cohorts. A human neuroimaging study (Lebedeva and colleagues, 2018) documented measurable effects on default mode network connectivity using fMRI. A randomized controlled trial (PMID 18379501) investigated Semax in patients with motor neuron disease. Human evidence overall is constrained by its concentration in Russian-language journals within the development institution; no large-scale RCT has been conducted under FDA or EMA trial standards.
- Animal: Strong. BDNF mRNA upregulation in hippocampus and cortex is demonstrated in multiple independent rodent studies (Eremin 2005; Dolotov 2006; Shadrina 2010). Semax and the related PGP fragment activate transcription of neurotrophin receptor genes after cerebral ischemia in rats (published in Cellular and Molecular Neurobiology, 2009). Neuroprotection in ischemia-reperfusion models has been confirmed at the transcriptome level (Medvedeva 2017). A 2025 study found Semax promotes functional recovery after spinal cord injury in mice via a μ-opioid receptor gene (Oprm1) deubiquitination mechanism (PMID 40692165). Antidepressant-like and antistress effects in a chronic unpredictable stress rat model have also been reported (PMID 39442746). GABA- and glycine-activated currents in isolated cerebral neurons are modulated by Semax (Sharonova and colleagues, Bulletin of Experimental Biology and Medicine, 2018).
- In vitro: Supportive. Semax has high affinity for copper(II) ions and reduces Cu(II)-catalyzed ROS production and amyloid-beta cytotoxicity through metal ion stripping and redox silencing (PMID 40496623). These copper-chelation properties may be relevant to neuroprotection in oxidative-stress contexts. Cell-level BDNF and NGF expression effects are documented in astrocyte and neuroblastoma cell cultures (Dolotov 2006).
Known effects
- Neurological recovery after acute ischemic stroke — Supported by Russian clinical trials; basis for Russian/Ukrainian regulatory approval. No independent Western RCT.
- BDNF and NGF upregulation in brain — Strong preclinical evidence across multiple rodent studies; onset within 30 minutes, duration up to 24 hours after a single dose (Eremin 2005; Dolotov 2006; Shadrina 2010).
- Cognitive and working-memory effects in healthy adults — Emerging: one controlled Russian study (Glazova 2018); no independent Western replication.
- Visual function preservation in optic nerve disease — Supported by Russian clinical trial (PMID 10741256); basis for approved Russian ophthalmology indication.
- Dopaminergic and serotonergic activation — Animal models (Eremin 2005); contributes to observed mood and attention effects.
- Neuroprotection in ischemia-reperfusion models — Strong animal evidence (Medvedeva 2017; multiple preclinical studies).
- Copper chelation and amyloid-beta reduction — In vitro (PMID 40496623); relevance to Alzheimer's pathology under investigation preclinically (PMID 41479572).
- Mood and antistress effects — Preclinical (chronic unpredictable stress model; PMID 39442746); not yet established in controlled human trials.
- Does not activate the HPA axis / raise cortisol — Confirmed by design: the adrenal-stimulating ACTH sequence (residues 1–24) is absent from Semax.
Safety signals
The most common adverse effect reported across the available literature is nasal irritation — dryness, mild rhinitis, and occasional epistaxis — attributable to the intranasal route of delivery and more frequent with prolonged use. Mild headache is reported as transient and not dose-limiting in Russian clinical trial populations. Irritability or restlessness at higher doses is described as dose-related and reversible. Hair thinning has been reported anecdotally and attributed speculatively to melanocortin receptor activity, but has not been documented in controlled clinical trials. Rare allergic reactions have been noted, including possible reactions to benzalkonium chloride preservative present in some formulations.
Contraindications from Russian product labeling include: pregnancy (no controlled safety data); breastfeeding (no breast-milk transfer data); known hypersensitivity to Semax or vehicle components; active psychotic disorders (given dopaminergic and serotonergic modulation); and severe nasal mucosal disease, chronic rhinitis, or recent nasal surgery (which may disrupt intranasal delivery).
The available literature describes theoretical pharmacodynamic interactions with potent serotonergic agents (SSRIs, SNRIs, MAOIs, triptans) and strong dopaminergic agents (stimulants, dopamine agonists) given Semax's monoamine-modulating mechanism. No specific adverse interaction has been documented in controlled trials.
Long-term unknowns: chronic daily use beyond the approved Russian protocol windows of 10–14 days (acute stroke) and 2–4 weeks (cognitive rehabilitation) has not been systematically characterized. All post-marketing safety experience derives from Russian clinical use; Western long-term safety data are absent. The broader gerontological peptide therapeutics review by Mavrych and colleagues (Frontiers in Aging, 2026) contextualizes Semax within an emerging class of peptide drugs, but does not resolve the chronic-use gap.
Regulatory status
- Russia: Approved prescription drug for ischemic stroke, cognitive disorders, and optic nerve atrophy; manufactured by Geropharm and other registered producers; on Russia's Vital and Essential Drugs List since 2011.
- Ukraine: Approved prescription drug for the same indications as Russia.
- US (FDA): Not approved for any indication. The FDA has not received a New Drug Application for Semax. A Pharmacy Compounding Advisory Committee (PCAC) review of Semax acetate and Semax (free base) is scheduled for July 24, 2026 — this is a compounding-status review, not an approval pathway.
- EU / EMA: Not approved; not submitted for EMA review. Regulatory availability varies by EU member state.
- WADA: Semax is not listed by name on the WADA Prohibited List as of published sources. WADA's S0 category covers substances not approved by any governmental regulatory health authority; whether Russian approval satisfies this clause is unresolved. Athletes subject to anti-doping rules should not assume Semax is permissible without consulting their national anti-doping authority.
- US 503A compounding: Pending PCAC review (July 24, 2026). Semax was removed from 503A Category 2 after nominations were withdrawn; the July 2026 PCAC consultation will determine compounding status. The outcome does not constitute FDA approval.
Mechanism
Semax enters brain tissue primarily via intranasal delivery, which routes the peptide to the CNS through olfactory and trigeminal pathways, bypassing first-pass metabolism. Once in brain tissue, Semax acts as a partial agonist at melanocortin-4 receptors (MC4R), which are densely expressed in the hippocampus and cortex. MC4R activation triggers the cAMP/PKA and MAPK/ERK signaling cascades, enhancing BDNF gene transcription. Dolotov and colleagues (Journal of Neurochemistry, 2006) characterized specific, calcium-dependent Semax binding sites in rat basal forebrain (KD = 2.4 ± 1.0 nM), and confirmed a statistically significant increase in BDNF protein levels at both a behaviorally active dose and a higher dose. BDNF mRNA elevations in hippocampus and frontal cortex appear within 30 minutes of a single intranasal dose and persist for up to 24 hours in rodents — substantially outlasting the peptide's plasma half-life of minutes (Shadrina and colleagues 2010). These animal findings form the mechanistic backbone for stroke-recovery and cognitive-enhancement effects, though direct human brain-tissue evidence has not been separately extracted for this card.
Beyond BDNF, Semax upregulates NGF and GDNF expression and activates neurotrophin receptor genes (TrkB, TrkC, TrkA) after ischemic challenge (Semax and PGP paper, Cellular and Molecular Neurobiology, 2009). It modulates dopaminergic and serotonergic neurotransmission — close functional links between MC receptors and monoaminergic systems were described by Eremin and colleagues (2005). Semax also inhibits enkephalinase enzymes (those that degrade endogenous enkephalins) at approximately 10 μM IC50 in human serum, an activity shared with Selank; this may contribute to anxiolytic and analgesic effects. In isolated cerebral neurons, Semax modulates GABA- and glycine-activated ionic currents (Sharonova and colleagues 2018), suggesting additional GABAergic contributions to its CNS profile.
Critically, Semax carries no adrenal-stimulating activity: the steroidogenic sequence of ACTH resides in residues 1–24, while Semax spans only residues 4–10 of ACTH plus the PGP tail. This dissociation of cognitive/neurotrophic effects from cortisol-driving hormonal activity was the central design objective of the Myasoedov–Ashmarin program.
Open questions
- Independent Western replication — All primary human efficacy trials originate from the Russian regulatory development program. No large-scale RCT has been conducted in Western clinical populations under FDA or EMA trial standards for any indication.
- US 503A compounding status — The FDA PCAC review scheduled for July 24, 2026 will determine Semax's legal compounding status in the US. The outcome is not an approval pathway.
- Chronic-use safety — Approved Russian protocols run 10–14 days (acute stroke) to 2–4 weeks (cognitive rehabilitation). Systematic characterization of continuous use beyond these windows in humans does not exist.
- Efficacy in healthy Western adults — The Glazova 2018 study provides limited evidence for attention and working-memory effects in healthy volunteers, but sample size and geographic/demographic scope are insufficient to establish a validated clinical effect in this population.
- NA-Semax-Amidate evidence base — The N-acetylated, C-amidated derivative (NA-Semax-Amidate) is marketed with improved-stability claims; no published comparative pharmacokinetic or behavioral study has compared it to standard Semax in humans or animals.
- Comparative effectiveness — No head-to-head trial compares Semax to approved Western cognitive or neuroprotective medications for any indication.
Related peptides
- Selank — The sister peptide from the same Institute of Molecular Genetics program; built on a tuftsin core with the same Pro-Gly-Pro glyproline stabilization strategy. Selank's primary emphasis is anxiolytic and immunomodulatory; Semax's primary emphasis is neurotrophic and neuroprotective. The two peptides are often studied together in the Russian literature and have overlapping enkephalinase-inhibiting activity.
- NA-Semax-Amidate — N-terminally acetylated and C-terminally amidated variant of Semax, marketed for improved enzymatic stability. No published human or comparative-animal data exist for this specific modified compound.
- ACTH(4-10) — The parent fragment from which Semax is derived; has cognitive effects in animal models but lacks the Pro-Gly-Pro stabilization tail and degrades rapidly in vivo.
Research directions for this peptide, selected from the current sources — hypotheses you can explore and model. None of it is proven yet; tap any one to see the full thinking.
Is the predicted MC4R target real, or is Semax working through a different receptor entirely?
If MC4R is not the real target, researchers could focus on finding the actual receptor Semax binds, leading to better drugs with fewer off-target risks for brain injury patients.
Could a simple chemical change to the first building block of Semax turn an active drug into an inactive one?
If true, manufacturers would know to protect that single amino acid during storage, ensuring patients receive a fully potent medicine.
Could the Pro-Gly-Pro end of Semax help position the peptide for receptor binding, not just slow its decay?
If true, we could redesign the tail to steer Semax toward different brain targets without changing its core, creating more precise brain-protecting drugs for stroke or cognitive decline patients.
Could Semax trigger local brain repair signals instead of raising them everywhere in the body?
If true, Semax could protect stroke-damaged brain areas without the dangerous side effects seen when growth factors are given systemically, offering a safer recovery option for patients.
▸full evidence table2 metrics
| metric | value | tool |
|---|---|---|
| ipTM | 0.9260112643241882 | boltz-2 |
| ranking score | 0.8384171724319458 | boltz-2 |
▸structural qualityopenfold3
| metric | value | note |
|---|---|---|
| gpde | 0.853 | global PDE — lower = better |
| disorder | NaN | fraction disordered |
▸3-letter notation
▸recipeboltz-2 1.0
| parameter | value |
|---|---|
| model | boltz-2 1.0 |
| weights | — |
| hardware | nvidia_nim_api |
| mlx version | — |
| python | — |
| random seed | — |
| msa strategy | none |
| diffusion samples | 1 |
| runtime | — |
| predicted by | mlx@peptide |
| predicted at | 2026-04-24 |
▸citationbibtex
@peptide{pep00005,
sequence = {MEHFPGP},
target = {mc4r},
author = {peptidemodel},
year = {2026},
status = {bioassayed}
}