Selank: Tuftsin-Derived Chemistry and GABAergic Pharmacology

Selank is a synthetic heptapeptide built from two previously described fragments: tuftsin, the tetrapeptide Thr-Lys-Pro-Arg carried in the Fc region of the immunoglobulin G heavy chain, and Pro-Gly-Pro, a tripeptide of the glyproline family attached at the carboxyl terminus. The full sequence is Thr-Lys-Pro-Arg-Pro-Gly-Pro, and the free peptide has the formula C33H57N11O9.

The parent tetrapeptide was characterized by Najjar and Nishioka in Nature in 1970, who named it tuftsin and described it as a phagocytosis-stimulating peptide. It occupies residues 289 to 292 of the immunoglobulin G heavy chain and is liberated by a two-enzyme route: a spleen endocarboxypeptidase nicks the chain at the Arg-Glu bond spanning residues 292 and 293, and the membrane enzyme leukokininase cleaves at the amino side of the threonine between residues 288 and 289.

The design logic of the heptapeptide is contained in the three added residues. Proline flanks the glycine at both sides of the appended tripeptide, and proline residues resist cleavage by most aminopeptidases and carboxypeptidases. The result is a molecule with three prolines out of seven residues and no cysteine at all, which is why the analytical identity problem for this peptide is sequence and purity rather than disulfide connectivity.

Selank has been characterized in radioligand-receptor work as a positive allosteric modulator of GABA binding rather than as a direct receptor agonist. Vyunova and colleagues, working with isolated plasma membranes from rat brain cells, reported that Selank altered the specific binding of tritiated GABA and that the peptide blocked the modulatory activity of both diazepam and olanzapine on the same membrane preparation.

Their stated conclusion was that the peptide binding site and the benzodiazepine site are apparently not the same, although they may partially overlap, and that one molecular route for the compound's anxiolytic activity is subtype-selective, concentration-dependent allosteric modulation of GABA receptors. That is a mechanistic claim about a binding interaction measured in a cell-free membrane assay, not a claim about any outcome in an intact organism.

A cell-culture experiment published the following year sharpens the picture by returning a negative result. Filatova and colleagues applied 1 nmol per well of Selank, GABA or olanzapine to human IMR-32 neuroblastoma cells for one hour and profiled an 84-gene neurotransmission panel, of which 69 genes had expression high enough to analyse. Selank alone produced no statistically significant change in any gene. GABA alone changed 14 genes, and olanzapine alone changed 25. When Selank was combined with GABA, only JUNB moved, falling 1.7-fold at p equal to 0.034; when it was combined with olanzapine, 35 genes changed. The authors concluded that the peptide has no direct effect on GABAergic gene expression in that cell line while the combination data partially support the hypothesis that it affects the interaction of GABA with GABA-A receptors.

In vivo the picture differs from the cell line. Volkova and colleagues gave rats 300 micrograms per kilogram of either Selank or GABA and assayed the same class of 84-gene neurotransmission panel in frontal cortex, finding significant changes in 45 genes at one hour and 22 genes at three hours, with a positive correlation between the one-hour profiles produced by the peptide and by GABA itself.

Selank applied directly to rat hippocampal slices altered inhibitory synaptic transmission onto CA1 pyramidal neurons. Povarov and colleagues recorded spontaneous inhibitory postsynaptic currents and reported increases in both the amplitude and the discharge rate of those currents. Across the tested range of 1 to 8 micromolar the effect showed no significant concentration dependence, and in some neurons a transient decrease preceded the increase.

Two details of that recording deserve emphasis because they constrain interpretation. A flat concentration-response curve across an eightfold range is not the signature of a classical saturable orthosteric interaction at the recorded synapse, and the transient early decrease indicates the net effect is composite rather than a single monotonic action. Both observations are consistent with the allosteric characterization from the membrane-binding work rather than with direct channel gating.

Selank has been profiled by RNA sequencing in two separate rat models published since 2024. In the acute restraint experiment, 300 micrograms per kilogram given 30 minutes before a one-hour immobilisation session was followed by 549 differentially expressed genes in the hippocampus two hours after the stress ended, at a fold-change threshold above 1.5 and an adjusted p value below 0.05.

The control condition in that experiment is the informative part. When the peptide was given to rats that were not subjected to immobilisation, no significant change in hippocampal gene expression was observed. The functional categories assigned to the 549 genes were antigen processing and presentation and nerve-impulse transmission, and the authors framed the result as regulation of stress-initiated processes rather than a baseline effect on genomic activity.

The second dataset, published in Molecular Biology in 2026, used an MRI-guided transient middle cerebral artery occlusion model in rats and sequenced frontal cortex at 4.5 hours. Comparison against saline controls identified 118 differentially expressed genes at the same fold-change and adjusted p thresholds. Of these, 99 were downregulated, and 60 of those 99 had been upregulated by the ischemic insult alone, which the authors described as a compensatory pattern. Functional annotation pointed to neuroreception, immunity, antigen processing and complement regulation.

Read together, the two RNA-sequencing datasets converge on immune and antigen-processing gene sets in both a psychological-stress model and an ischemic model. That convergence is consistent with the immunoactive origin of the tuftsin core, though neither study tested that connection directly.

Selank resists plasma degradation better than tuftsin does, but it is still cleaved, and the cleavage products have been mapped. Zolotarev and colleagues traced evenly tritium-labelled peptide in blood plasma and identified the pentapeptide TKPRP, the tripeptide TKP, and the dipeptides RP and GP as the major products of Selank biodegradation.

The labelling chemistry behind that result is worth recording because it is what makes the product map possible. The peptides were prepared by high-temperature solid-phase catalytic isotope exchange with spillover tritium at specific radioactivity of 50 to 150 Ci/mmol. Because the label is distributed across every amino acid residue rather than concentrated at one position, all possible hydrolysis fragments carry tritium and can be resolved chromatographically in a single analysis.

A separate enzymatic interaction was reported by Zozulya and colleagues in 2001. Working with human blood plasma, they found that the heptapeptide inhibited enzymatic hydrolysis of plasma enkephalin in a concentration-dependent manner. The same report described a considerable shortening of enkephalin half-life and reduced total enkephalinase activity in blood sampled during generalized anxiety, and proposed inhibition of enkephalin breakdown as a candidate mechanism. Related Russian work has examined the peptide's effect on carboxypeptidases in rat nervous tissue.

Selank has been examined in rodent anxiety and withdrawal models, generally against a benzodiazepine comparator run in parallel. In outbred rats with opiate dependence, a single intraperitoneal application at 0.3 mg/kg lowered the total index of morphine withdrawal by 39.6 percent at p below 0.0001, while diazepam at 2 mg/kg lowered the same index by 49.3 percent in the same experiment.

The component measures in that study separate the two compounds further. The peptide attenuated convulsive reactions, ptosis and posture disorders, and raised the tactile sensitivity threshold ninefold; diazepam raised the same threshold thirteenfold. The authors reported the peptide as weakening the aversive signs of morphine withdrawal in a manner comparable to, though not identical with, the benzodiazepine.

Combination work in mice was published in the Neurochemical Journal in 2025. Kost and colleagues paired the peptide with diazepam and with mezapam in male BALB/c mice and reported that the combinations increased both the number of entries into and the time spent in the open arms of the elevated plus maze.

A 2023 Russian study took a chronic rather than acute approach. Bobyntsev and colleagues exposed male Wistar rats to a 28-day restraint protocol and tested three amount levels of the peptide, reporting anxiolytic-type behavioural changes at the highest level and changes on behavioural despair measures at an intermediate level, both accompanied by lower blood corticosterone.

One older comparison is directly relevant to the chemistry. Semenova and colleagues ran the heptapeptide alongside tuftsin itself in 87 mature Wistar rats pretreated with the serotonin-synthesis inhibitor p-chlorophenylalanine. Thirty minutes later, the heptapeptide had raised serotonin metabolism in the brainstem, while tuftsin produced no change in the brainstem and a decrease in the neocortex. The three appended residues therefore change the pharmacology, not only the metabolic stability.

Handling figures for Selank come from reference-standard and supplier specifications rather than from peer-reviewed measurement, and should be recorded as such. The acetate salt is specified at 98 percent or higher purity, soluble in phosphate-buffered saline at pH 7.2 at 10 mg/mL, and stored as a solid at minus 20 degrees Celsius. The acetate carries the formula C33H57N11O9 plus C2H4O2 with a formula weight of 811.9, against 751.9 for the free peptide.

Salt form is the first identity trap. The free peptide and the acetate are separate registry entries, 129954-34-3 and 2703745-90-6 respectively, and a certificate of analysis that quotes one mass while the material is the other will not reconcile. Documentation should state which form is in hand and which of the two masses the analysis was run against.

The second identity issue is nomenclature, and it has a practical consequence for literature work. The compound is indexed under Selank, Selanc and TP-7, and a bare-name search in PubMed auto-expands the token to TP 7, which returns a large volume of unrelated records. Restricting the query to title and abstract fields, or searching the CAS number and PubChem CID 11765600 directly, returns the actual corpus of roughly 68 records.

Analytically, the molecule is straightforward relative to larger peptides. There is no cysteine and therefore no disulfide connectivity to confirm, so identity work reduces to sequence verification and chromatographic purity against the 751.9 g/mol free-peptide mass. Where degradation is a concern, the published biodegradation map gives the fragments to look for: TKPRP, TKP, RP and GP.