Selank (TP-7)

Identity, and a second molecule sold under an adjacent name

PubChem CID 11765600 resolves Selank as the heptapeptide Thr-Lys-Pro-Arg-Pro-Gly-Pro, molecular formula C33H57N11O9, molecular weight 751.9 g/mol, InChIKey JTDTXGMXNXBGBZ-YVHUGQOKSA-N. The record was cross-checked against the FDA Global Substance Registration System, which carries the same substance under UNII TS9JR8EP1G with CAS 129954-34-3, RXCUI 2049928 and EPA DTXSID701029276, and classifies it as a protein. Both databases agree on the CAS number, which is worth stating because they do not always.

The design is additive. Tuftsin is the tetrapeptide Thr-Lys-Pro-Arg, released by enzymatic cleavage from the heavy chain of immunoglobulin G, and Selank is that sequence with Pro-Gly-Pro appended. Czabak-Garbacz and colleagues describe the construction in exactly those terms in their 2006 paper on the compound under its laboratory designation TP-7 (PMID 16963804). The extension does not make the molecule inert to peptidases. Zolotarev and colleagues, tracking uniformly tritium-labelled peptide in blood plasma, identified the pentapeptide TKPRP, the tripeptide TKP and the dipeptides RP and GP as the major degradation products (PMID 16637290). Within minutes of administration the material present is a mixture, and the C-terminal glyproline fragment has its own literature.

A different molecule circulates under a name close enough to be mistaken for this one. PubChem holds no record for N-Acetyl Selank Amidate or for Selank Amidate; the nearest entry is CID 133082488, N-Acetyl Selank, C35H59N11O10, 793.9 g/mol. Its IUPAC name terminates in a pyrrolidine-2-carboxylic acid, meaning a free C-terminal acid rather than the amide the marketed name specifies. That record also carries a defective synonym string, (AC-THR-LYS-PRO-ARG-GLY-PRO), which lists six residues where the structure has seven. A PubMed search for N-acetyl selank returns zero records in any language.

The enkephalinase mechanism, and two IC50 values that do not agree

Zozulya and colleagues published the observation the mechanism rests on in 2001 (PMID 11550013). Examining patients with anxiety and phobic disorders classified by DSM-IV criteria, they reported a shortened enkephalin half-life and reduced total enkephalinase activity in blood during generalised anxiety, but not during panic disorder or agoraphobia. In the same paper, Selank inhibited enzymatic hydrolysis of plasma enkephalin dose-dependently with an IC50 of 15 microM, more potently than bacitracin or puromycin. The inference offered was that a compound slowing enkephalin breakdown would raise the parameter the patients were low on.

Kost and colleagues, an overlapping author group publishing in the same year, ran the assay on human serum and reported an IC50 of 20 microM for Selank against serum enkephalin-degrading enzymes, alongside 10 microM for Semax (PMID 11443939). They also reported that the pentapeptide fragments of both heptapeptides retained inhibitory activity while the tri-, tetra- and hexapeptide fragments did not. The two 2001 figures differ by a third. Neither indexed abstract reconciles them, and the matrices are not identical — plasma in one, serum in the other. The figures are recorded here as published rather than averaged into a single number.

Whether the effect appears at all depends on the animal. Sokolov and colleagues compared BALB/c and C57Bl/6 mice, which differ at baseline in plasma leu-enkephalin half-life, and found that 100 micrograms/kg produced an anxiolytic effect in the open-field test and lengthened leu-enkephalin half-life in BALB/c mice while changing neither behaviour nor enkephalinase activity in C57Bl/6 mice (PMID 12432865). Vasil'eva and colleagues reproduced that restriction fourteen years later, again finding effects only in the higher-anxiety strain (PMID 29787664). A compound that acts in one inbred strain and not another is describing something about baseline phenotype as much as about itself.

The GABA claim rests on binding assays, and a cell study points the other way

Vyunova and colleagues supplied the primary radioligand evidence in 2018 (PMID 30255741). Using isolated brain cell plasma membranes, they reported that Selank affects [3H]GABA binding as a positive allosteric modulator, that its joint action with benzodiazepines is not cumulative and differs from either substance alone, and that it can block the modulatory activity of diazepam and olanzapine — from which they concluded that the binding sites are apparently not the same but may partially overlap. The indexed abstract does not state the species the membranes came from, and that omission matters for a receptor-pharmacology result.

Volkova and colleagues tested the same hypothesis at the level of transcription (PMID 26924987). Thirty male Wistar rats of about 200 g were divided into three groups of ten, given a single intranasal administration of Selank or GABA at 300 micrograms/kg or water, and killed at one or three hours; 84 neurotransmission genes were measured in frontal cortex. Forty-five genes changed at one hour and twenty-two at three hours, with the Selank and GABA profiles positively correlated. The method section carries a constraint the abstract does not: frontal cortex RNA was pooled by group and timepoint into six pools, so the array measured six samples rather than thirty animals, and the design carries no within-group variance.

Filatova and colleagues, largely the same laboratory, then ran the comparison in IMR-32 neuroblastoma cells and found no changes in the mRNA levels of any of the 84 genes under Selank alone (PMID 28293190). Combined with GABA, Selank nearly abolished the expression changes GABA produced by itself. Their stated conclusion is that the peptide has no direct effect on GABAergic gene mRNA in that cell line while partially supporting the receptor-interaction hypothesis. Two results from one group therefore point in different directions depending on whether the readout is binding or transcription, and both are published.

The human record

Zozulia and colleagues studied 62 patients with generalised anxiety disorder and neurasthenia, comparing Selank in 30 patients against medazepam in 32, with Hamilton, Zung and CGI ratings and serum enkephalin measurement (PMID 18454096). PubMed types the paper as a Randomized Controlled Trial and Comparative Study. The reported anxiolytic effects of the two drugs were similar, with additional antiasthenic and psychostimulant effects attributed to the peptide, and the patients were reported to have a decreased leu-enkephalin half-life at baseline that correlated with disease duration and symptom severity.

Medvedev and colleagues account for the two later human studies located. The 2014 paper examined 60 patients with phobic-anxiety and somatoform disorders coded F40.2-9, F41.1-9 and F45.0-1, comparing Selank with phenazepam, and reported that the anxiolytic effect lasted for a week after the last administration (PMID 25176261). The 2015 paper, typed by PubMed as a Randomized Controlled Trial, compared phenazepam alone in 30 patients against phenazepam with Selank in 40, and reported a lower burden of the benzodiazepine's adverse effects in the combination arm on the UKU scale (PMID 26356395). Neither indexed abstract states a dose or a route.

Panikratova and colleagues published the only human imaging study located (PMID 32342318). Fifty-two healthy participants underwent resting-state fMRI three times — before, and five and twenty minutes after, administration of Semax, Selank or placebo — with amygdala and dorsolateral prefrontal cortex as regions of interest. Differences were reported in functional connectivity between the right amygdala and a right-hemisphere region spanning fusiform, inferior and middle temporal and parahippocampal gyri. The abstract describes the administration as an injection and does not specify the route further.

Four constraints apply across the whole human record. Every trial is published in Russian with only an English abstract indexed, so the methods cannot be examined from the retrievable record. The author lists overlap heavily and trace to a single institutional lineage in Moscow. A ClinicalTrials.gov search returns no study with Selank as an intervention; the one text match, NCT01747200, is an unrelated transcranial magnetic stimulation study. And Vanhee and colleagues, writing in Drug Testing and Analysis from an official medicines control laboratory, stated in 2020 that Selank and Semax had not to their knowledge completed any clinical trials (PMID 31667971) — a statement that cannot be squared with the PubMed typing without deciding which record is authoritative. This page does not decide that.

Route is not a detail

Ashmarin and colleagues compared tissue distribution of tritium-labelled Selank and Pro-Gly-Pro in rats after intraperitoneal, intranasal, intragastric and intravenous administration (PMID 18695718). Intranasal delivery was reported as optimal for reaching the central nervous system, and both peptides showed high affinity for gastric tissue by every route studied. The authors linked that gastric accumulation to the separate line of work on glyprolines and the gastric mucosa. No measurement of blood-brain barrier permeability was located in any retrieved paper; a PubMed search pairing the compound with blood-brain barrier returns zero records.

Vasil'eva and colleagues then showed that the route changes which receptor system moves (PMID 29787664). At 300 micrograms/kg per day for five days in BALB/c mice, intraperitoneal administration increased the number of [G-3H]SR 95531 binding sites at GABA receptors in frontal cortex by 38% with no change in hippocampal NMDA binding, while intranasal administration at the same dose and duration increased [G-3H]MK-801 binding site density by 23% with no effect on GABA receptors. Same compound, same dose, same strain, opposite readouts. The authors attributed the divergence to pharmacokinetics and biotransformation.

Findings from one route therefore do not transfer to another, which is a problem for the most-quoted rodent result on this compound. Inozemtseva and colleagues reported in 2008 that intranasal Selank regulates BDNF expression in the rat hippocampus (PMID 18841804). The citation resolves and the title supports the direction of the claim, but PubMed carries no abstract for the paper, and the specific figures that circulate downstream of it could not be confirmed against any retrievable source. They are logged below as untraced rather than repeated.

Regulatory position, and what the anti-doping question turns on

The compound holds no approval from the FDA or the EMA. The FDA GSRS record is a substance registration and carries no marketing authorisation; the word approved in that database describes the status of the data record, not of a medicine. Doyno and White, reviewing GABAergic sedative-hypnotics in 2021, characterised Selank as a poorly studied Russian drug sold to consumers in the United States as a dietary supplement. Whether it is registered as a medicine in Russia is a separate question, and one this record could not settle — see the untraced claims below.

Anti-doping status was checked against the primary document rather than a summary. The text of the World Anti-Doping Agency Prohibited List 2026 was extracted and searched: the strings Selank, tuftsin and TP-7 do not appear anywhere in it, while Bromantan does, which confirms the extraction was working. Absence of the name is not the same as permission. Section S0, Non-Approved Substances, prohibits at all times any pharmacological substance not addressed by another section of the List and with no current approval by any governmental regulatory health authority for human therapeutic use; its illustrative examples include BPC-157 but not this compound. Whether S0 reaches Selank therefore depends on the Russian registration question, which is exactly the fact that could not be verified.

Material sold under the name has been examined once in the forensic literature. Vanhee and colleagues encountered two unidentified seized pharmaceutical preparations at the end of 2017 and 2018, determined that they contained Selank and Semax, and developed a validated LC-MS/MS screen covering ten putative cognitive-enhancing peptides sold online (PMID 31667971). They noted the peptides are freely available as lyophilised powder for injection or in nasal sprays. What a seizure analysis establishes is the content of those samples. No published survey of the identity, purity or content uniformity of material sold under this name more broadly was located.

What is not known

No pharmacokinetic study of Selank in humans was located, so absorption, half-life, clearance, dose proportionality and interactions are unmeasured in the retrievable record for the species the compound is sold for. The rodent distribution work establishes which route reaches the brain best but not what concentration arrives or in what form, and the plasma degradation data show the parent molecule is cleaved into at least four fragments with literatures of their own, so attributing an effect to the heptapeptide rather than to TKPRP, TKP, RP or GP is not possible from the published record. Mechanism is unsettled at the level of the primary sources: one radioligand study describes positive allosteric modulation of GABA binding, and a cell study from an overlapping group finds no direct effect on GABAergic gene expression at all. Nothing establishes an upstream molecular target. The human evidence consists of a small number of Russian-language papers from one institutional lineage, published with English abstracts only, absent from every trial registry searched, never replicated abroad, and with doses and routes missing from the indexed abstracts. Repeated-administration safety is effectively unstudied — one four-week rat experiment, no dependence, withdrawal or receptor-regulation work in any species, and no human discontinuation data. No reproductive, developmental or carcinogenicity toxicology was found. Regulatory status outside Russia is straightforward — no FDA or EMA approval — and inside Russia it could not be verified, which also leaves the anti-doping position undetermined.

Questions