Selank: a complete guide to the anxiolytic heptapeptide

Selank is a seven-amino-acid peptide with the sequence Thr-Lys-Pro-Arg-Pro-Gly-Pro ^[1]. It was synthesized at the Institute of Molecular Genetics of the Russian Academy of Sciences in the 1990s, designed as a stabilized analog of tuftsin ^[2], an endogenous tetrapeptide your body makes as part of an immunoglobulin fragment.

The classification is awkward. In the Russian Federation, Selank is a registered prescription drug with an approved indication: generalized anxiety disorder and anxiety-asthenic conditions ^{[3] [4]}. In the United States and the EU, it has no regulatory status as a medicine at all. It exists in commerce as a "research compound, not for human use," which is the polite fiction surrounding most peptides that lack FDA approval but get bought anyway.

So the honest answer to "what is it" depends on which jurisdiction is asking.

Tuftsin is the parent molecule. It's a four-residue fragment (Thr-Lys-Pro-Arg) cleaved from the heavy chain of immunoglobulin G ^[5], and it has documented immunomodulatory activity, mostly involving macrophage and natural killer cell function. The trouble with tuftsin as a drug candidate is that it falls apart fast. Endogenous peptidases chew through it before it can do much therapeutically.

Selank addresses that limitation. The Russian chemists appended a Pro-Gly-Pro tail to the tuftsin sequence, a configuration that resists proteolytic cleavage and extends the molecule's working half-life. The tetrapeptide became a heptapeptide. The immune activity stayed. New central nervous system effects emerged.

That dual heritage, immune fragment plus CNS-active peptide, is why Selank's pharmacology reads strangely to anyone expecting a clean anxiolytic. It does several things at once because it descends from a molecule that already did.

There are two mechanistic stories worth understanding, and they aren't redundant.

The GABAergic story. Selank appears to modulate GABA-A receptor signaling, but not by docking into the benzodiazepine site the classical way. Gene expression work in rat frontal cortex (PMID: 26924987) found Selank shifted a notable subset of genes in GABAergic neurotransmission pathways an hour after dosing, with fewer showing changes by the three-hour mark. In IMR-32 neuroblastoma cells (PMID: 28293190), a cluster of GABA-related genes shifted as well. The pattern tracked positively with what GABA itself does to the same cells, which hints that Selank engages the system upstream rather than parking itself at a specific receptor site.

That matters. It's a different shape of pharmacology than diazepam. Benzodiazepines are positive allosteric modulators of a specific GABA-A subunit, they hit fast, and in practice they also produce tolerance, sedation, and the kind of dependence the receptor biology more or less predicts. Selank's effect looks transcriptional and indirect, which would, on paper, explain why its clinical profile doesn't carry those liabilities.

On paper.

The enkephalin story. Selank dose-dependently inhibits enkephalin-degrading enzymes in plasma, though the precise potency figures reported in the literature are worth scrutinizing before citing them. Leu-enkephalin, an endogenous opioid peptide involved in stress and mood regulation, gets degraded more slowly when those enzymes are blocked. So endogenous enkephalin hangs around longer.

The biochemistry makes a prediction: Selank should work better in GAD than in panic. The clinical data, what little of it exists, tracks with that prediction.

Add to this: documented effects on serotonin turnover, BDNF upregulation in hippocampus and frontal cortex (with TrkB blockade reversing some cognitive effects in rodents), and modulation of HPA axis activity under chronic stress. The mechanism isn't one thing. It's a network.

The key human trial is Zozulia et al. (2008), published in Zhurnal Nevrologii i Psikhiatrii ^[6]. It enrolled 62 patients with GAD or neurasthenia, randomized them to Selank (n=30) or medazepam, a benzodiazepine (n=32) ^[7], and tracked them on Hamilton anxiety, Zung depression, and Clinical Global Impression scales ^[8].

The reported conclusion: Selank's anxiolytic effect was comparable to medazepam, with additional anti-asthenic and mild psychostimulant properties, no sedation, no dependence, and no withdrawal in the study period.

That sentence is doing a lot of work. The English-language abstract doesn't report numeric scale deltas, effect sizes, p-values, or responder rates. The trial was short. The sample was modest. Independent Western replication, as far as anyone can tell, isn't there. If a U.S. registration submission landed at FDA looking like this, it wouldn't advance.

But as mechanistic plus clinical signal in a peptide that's been kicking around Russian psychiatry for years, it's more than nothing.

A 2020 fMRI study (PMID: 32342318) ran a group of healthy participants through a crossover design comparing Selank, Semax, and placebo on resting-state functional connectivity. Selank altered connectivity between the right amygdala and right temporal cortex within minutes of intranasal administration, distinct from both placebo and Semax. The amygdala result is biologically suggestive. Anxiety circuits run through it. Whether that fMRI signal predicts clinical anxiolysis in patients is a separate question, and the study can't answer it.

Short version: the anxiolytic signal exists, but the file is thinner than the marketing suggests.

This is where the literature gets noisier.

Selank gets sold and described as a cognitive enhancer. The basis is partly mechanistic (BDNF upregulation, serotonergic modulation) and partly behavioral data from rodents showing improved performance on learning and memory tasks. Cognitive improvements in animals are sometimes reversed by TrkB antagonism, which is consistent with a BDNF-mediated mechanism.

The gap is the human evidence. Controlled cognitive trials in healthy human volunteers are essentially absent from the peer-reviewed Western literature. The fMRI work suggests acute brain effects; it doesn't measure memory or executive function in any standard cognitive battery.

A 2016 route-of-administration study (Vasileva et al.) found something worth noting: intranasal Selank in mice biased toward anxiolytic effects, while intraperitoneal administration biased toward cognitive effects ^[9]. If that holds up, it would mean the nootropic claims and the anxiolytic claims aren't quite the same drug in practice. They might depend on how you give it.

For a Western researcher, the honest read is this: the cognitive claims are plausible mechanistically, suggestive in animals, and unproven in controlled human studies.

The tuftsin lineage is not decorative. Selank inherits real immune activity.

A rat study using a social stress model (PMID: 32621722) found that daily Selank administration attenuated stress-induced elevations in pro-inflammatory cytokines associated with the central inflammatory response to chronic stress. A separate liver morphology study (PMID: 31243679) in stressed Wistar rats found Selank reduced hepatocyte degeneration at an optimal dose of 300 µg/kg ^[10].

These are preclinical findings in stressed animals. They suggest Selank does something to the inflammatory tone of the stress response, which fits the tuftsin pedigree and is consistent with reports of anti-asthenic effects in patients. Whether it functions as a meaningful immunomodulator in non-stressed humans, or in conditions where immune dysregulation is the primary issue, is unestablished.

Here is the part Western researchers need to read with both eyes open.

What exists: - A small number of controlled clinical trials, mostly published in Russian-language journals, mostly with sample sizes under 100, mostly conducted at Russian academic institutions. - One head-to-head against a benzodiazepine (medazepam) in GAD. - Mechanistic human work, including the fMRI study and the enkephalin plasma study. - Decades of accumulated clinical use under Russian Ministry of Health approval.

What doesn't exist: - Large, multi-center, placebo-controlled, double-blind RCTs published in major Western psychiatric journals. - Independent replication of the key clinical findings by research groups outside Russia. - Long-duration safety data beyond a few weeks of controlled administration. - Standardized reporting of effect sizes, confidence intervals, and responder definitions in the English-language abstracts.

This isn't a claim that Russian clinical research is unreliable. It's a claim that the file as it reaches Western readers is incomplete, and that incompleteness has real consequences for how confident anyone can be about efficacy, dose-response, or patient selection.

The honest take is that Selank has clinical evidence, just not the kind regulators in the US or EU are built to evaluate.

The preclinical file is broader and more mechanistically detailed than the clinical one.

A 2022 study (PMID: 36322304) in outbred rats tested Selank against morphine withdrawal. At 0.3 mg/kg intraperitoneal, Selank reduced the total withdrawal index by 39.6% (p<0.0001). Diazepam at 2 mg/kg did slightly better, at 49.3% ^[11]. That's a large effect for a non-sedating compound in an addiction model, and it tracks with the enkephalin mechanism, since opioid withdrawal is partly an enkephalin-system event.

Other rodent work has shown anxiolytic activity in elevated plus maze and open field paradigms, cognitive enhancement in learning tasks, and the cytokine and BDNF effects already noted.

Two things to flag. Rodent anxiety models predict human anxiolytic activity inconsistently. Drugs that pass them sometimes fail in patients. Drugs that pass them in some labs sometimes fail in others. And the dose ranges used in rodents, often 100 to 1000 µg/kg intraperitoneal, don't map cleanly onto human intranasal protocols. Translation requires assumption stacking.

Selank is administered intranasally in essentially all human clinical work. This isn't incidental.

Peptides this size, seven residues, don't survive oral administration. The gut tears them apart. Injection works but is inconvenient for chronic anxiolytic use. Intranasal delivery exploits two routes simultaneously: systemic absorption through nasal mucosa, and direct nose-to-brain transport via olfactory and trigeminal pathways that partially bypass the blood-brain barrier.

For a CNS-active peptide, that second route is the interesting one. It allows central concentrations to rise without proportional systemic exposure, which in principle reduces peripheral side effects and improves the therapeutic index.

Precise human pharmacokinetics for Selank (plasma half-life, central exposure, dose linearity) are not well-characterized in the available English literature. The Pro-Gly-Pro tail extends metabolic stability relative to tuftsin, but specifics on humans remain a gap. Anyone telling you they know Selank's exact brain concentration after a 500 µg intranasal dose is filling in numbers the published file doesn't support.

A mechanistic comparison only. Not a recommendation, not a substitution claim.

Benzodiazepines (diazepam, alprazolam, lorazepam) act as positive allosteric modulators at a dedicated benzodiazepine site on GABA-A receptors. Onset is fast. The anxiolytic pull is strong, sedation tracks the dose, and chronic use brings the familiar problems: tolerance, dependence, and withdrawal syndromes that are clinically real, not theoretical. The mechanism is direct, and the liabilities fall out of it.

SSRIs (sertraline, escitalopram, paroxetine) block presynaptic serotonin reuptake, leading to gradual receptor adaptation over weeks. Slow onset (typically four to eight weeks for anxiety response), no acute anxiolysis, no dependence in the classical sense though discontinuation syndromes are real, with sexual and GI side effects common. First-line for chronic anxiety in most Western guidelines.

Buspirone: 5-HT1A partial agonist. Slower onset than benzodiazepines, no sedation, no dependence, modest effect size in GAD, well-tolerated but often underwhelming clinically.

Selank: indirect GABAergic modulation via transcriptional changes, plus enkephalinase inhibition, plus serotonergic effects and BDNF upregulation. Reported as non-sedating and non-dependence-forming in the available trials. Effect size relative to SSRIs or benzodiazepines in head-to-head Western trials is unknown because those trials haven't been run.

The mechanism is genuinely different from any of the approved classes. Whether the difference translates into a meaningfully better clinical profile in real patients is the question the existing literature can't answer.

The published safety signal is reassuring on its face. Russian clinical trials, mostly running only a few weeks, describe Selank as well-tolerated, without sedation, dependence, or withdrawal, and without notable adverse events. Preclinical toxicity work hasn't identified an LD50 in tested ranges and hasn't found organ histopathology in standard rodent studies.

What's missing is the long view. No published controlled human studies have run Selank past a few weeks. Drug interactions in people? Poorly mapped. Safety during pregnancy, in kids, and in patients with kidney or liver problems hasn't been worked out either.

The FDA's Adverse Event Reporting System shows no Selank-specific signal, which tracks, since the drug isn't approved and doesn't move through the channels that would generate one.

Worth flagging: the absence of adverse signals in short trials is not evidence of long-term safety. It's evidence that short trials didn't find anything. Those are different statements.

Russia. Registered prescription pharmaceutical, approved by the Russian Ministry of Health for generalized anxiety disorder and anxiety-asthenic conditions. Prescribed in clinical psychiatry. Manufactured under Russian GMP.

United States. Not FDA-approved for any indication. As of current FDA bulk substances guidance, Selank is not on the list of substances permitted for compounding under 503A or 503B pathways ^[12]. It's effectively off-limits for U.S. licensed compounding pharmacies. It circulates in U.S. commerce as a "research compound, not for human use," a labeling convention that has limited legal weight but defines the market.

European Union. No EMA approval. No EU member-state approval that we're aware of. Status similar to the US in practical terms.

Other. Selank does not appear to be on the WADA prohibited substances list based on available data, though athletes subject to anti-doping rules should verify independently before assuming.

The asymmetry is real. A Russian psychiatrist prescribes it on Monday. A U.S. physician writing that same script on Tuesday has no legal pathway to fill it through a compounding pharmacy.

Same molecule. The regulatory environment isn't.

The research base is concentrated. The Institute of Molecular Genetics in Moscow, where Selank was developed, has continued to publish on it. Russian psychiatric research groups have run the clinical work. A handful of Western groups have engaged with the molecule, mostly on the mechanistic side (BDNF, GABA gene expression, neuroplasticity), often in collaboration with Russian co-authors or using cell-culture systems.

The reasons people are still looking at this peptide are reasonably specific.

The enkephalin mechanism is novel. No approved anxiolytic works primarily through enkephalinase inhibition. If the GAD-specific normalization of plasma enkephalin half-life holds up, it suggests a biomarker-defined subpopulation of anxiety patients who might respond to a mechanism the current pharmacopoeia doesn't address.

The non-sedating, non-dependence-forming profile, if real and durable, addresses the most persistent clinical complaints about benzodiazepines.

The intranasal delivery route and rapid onset (the fMRI signal at 5 minutes is genuinely fast) suggests a use case the slow-acting SSRI class can't fill.

And the immunomodulatory inheritance from tuftsin opens questions about neuro-immune crosstalk in anxiety disorders that the field has only recently started asking seriously.

What's striking here is what would shift the picture: a properly powered, placebo-controlled, multi-center trial run to Western regulatory standards, with pre-registered endpoints and transparent reporting. That trial hasn't been run. Why it hasn't is partly economic, since the patent landscape and commercial incentives don't favor it, and partly geopolitical, since the research base sits in a jurisdiction whose drug regulators don't currently have reciprocity with the FDA or EMA.

So Selank lands where it lands: a registered drug somewhere, a research chemical elsewhere, a real but partial clinical file, and an open question about whether the West will ever bother evaluating it on its own terms.

Both things can be true.

The honest read of the data goes like this. Selank looks more intriguing than the size of its English-language literature would predict, and less proven than the wellness marketing claims.