PE-22-28 is the heptapeptide Gly-Val-Ser-Trp-Gly-Leu-Arg (GVSWGLR), the shortest active fragment of spadin, which is itself a piece of the propeptide cleaved from the receptor protein sortilin. It was described in 2017 by a CNRS group in Valbonne, France, as an inhibitor of the TREK-1 potassium channel, and in their patch-clamp measurements it was far more potent than spadin. It has been studied only in cells and mice, and it is not approved as a medicine anywhere.
The name is a residue range: "PE" is the sortilin propeptide, and 22-28 are the positions of the seven residues within it. This profile follows that lineage from sortilin to spadin to PE-22-28, then covers the channel data, the research models, regulatory status and handling.
PE-22-28 at a glance
| Property | Value |
|---|---|
| Sequence | Gly-Val-Ser-Trp-Gly-Leu-Arg |
| One-letter code | GVSWGLR |
| Length | 7 residues |
| Molecular formula | C35H55N11O9 |
| Molecular weight | 773.9 g/mol |
| CAS number | 1801959-12-5 |
| PubChem CID | 165437303 |
| Parent peptide | Spadin (propeptide residues 12-28) |
| Parent protein | Sortilin (gene SORT1), propeptide residues 34-77 |
| Target studied | TREK-1 (KCNK2), a two-pore-domain potassium channel |
| Oxidation-prone residues | Trp |
| Regulatory status | Not an approved medicine; no human trials registered |
Formula, weight, CAS and CID values are from the PubChem record for PE 22-28. Note that some secondary sources list a different formula and weight (788.9 g/mol); PubChem's value of 773.9 g/mol matches the sum of the seven residues. Catalog material is described on the PE-22-28 page.
From sortilin to spadin to PE-22-28
Sortilin and its propeptide
Sortilin, also called neurotensin receptor 3, is a sorting receptor encoded by the SORT1 gene. UniProt lists human sortilin as an 831-residue precursor with a signal peptide (residues 1 to 33) and a propeptide (residues 34 to 77) that is removed in the mature protein. In the literature on spadin, that 44-residue propeptide is called PE and numbered from 1 (Gln) to 44 (Arg). According to Mazella and colleagues, it is cleaved from prosortilin by the protease furin in late Golgi compartments.
Spadin
In 2010, Mazella and colleagues at the Institut de Pharmacologie Moléculaire et Cellulaire (CNRS) in Valbonne reported that a fragment of PE, residues 12 to 28 (APLPRWSGPIGVSWGLR), binds and inhibits TREK-1. They named it spadin. In their work it bound TREK-1 with a dissociation constant of about 10 nM and inhibited channel currents with an IC50 of about 71 nM. The study used COS-7 cells expressing the channel, cultured hippocampal neurons and brain slices, and mice lacking TREK-1.
Shortening spadin
Spadin is 17 residues long. In 2017, Djillani and colleagues from the same institute made a series of shorter fragments and variants and tested each against human TREK-1. The seven C-terminal residues of spadin, PE 22-28, were the shortest sequence that kept the activity.
| Peptide | Propeptide residues | Sequence | Length |
|---|---|---|---|
| PE (full propeptide) | 1-44 | QDRLDAPPPPAAPLPRWSGPIGVSWGLRAAAAGGAFPRGGRWRR | 44 |
| Spadin | 12-28 | APLPRWSGPIGVSWGLR | 17 |
| PE 22-28 | 22-28 | GVSWGLR | 7 |
| G/A-PE 22-28 | 22-28, Gly22 to Ala | AVSWGLR | 7 |
The PE sequence above is residues 34 to 77 of human sortilin as listed in UniProt Q99523.
What TREK-1 is
TREK-1 is a potassium channel of the two-pore-domain (K2P) family, encoded by KCNK2. K2P channels carry "leak" or background potassium currents that help set the resting membrane potential, and UniProt describes TREK-1 as switching into a leak mode in response to membrane stretch, acidic pH, heat and lipids; arachidonic acid is the usual activator in patch-clamp work. The 2010 spadin paper built on earlier work showing that mice lacking TREK-1 behave differently in standard rodent assays used in antidepressant research, which is why TREK-1 blockers became a research target.
Structurally, each K2P subunit has two pore-forming domains, and two subunits assemble into one channel. UniProt notes that TREK-1 forms both homodimers and heterodimers, including with TRAAK and TRESK, which have distinct gating properties. That matters when interpreting selectivity data: a peptide tested on homomeric TREK-1 in HEK293 cells may behave differently on a heteromeric channel in native tissue, and the published PE 22-28 selectivity panel covered homomeric channels.
The channel data
Djillani and colleagues measured inhibition by whole-cell patch clamp in HEK293 cells stably expressing human TREK-1, with the channel first activated by arachidonic acid.
| Peptide | Reported IC50 on hTREK-1 |
|---|---|
| Spadin | About 40 to 60 nM |
| PE 22-28 | 0.12 nM |
| G/A-PE 22-28 | 0.10 nM |
| Biotinylated G/A-PE 22-28 | 1.2 nM |
On those figures, PE 22-28 is several hundred times more potent than spadin in this assay. These are in-vitro concentration values from one laboratory's patch-clamp system; they are not measures of effect in an animal or a person.
Selectivity. The same study reported that PE 22-28 did not inhibit the related K2P channels TREK-2, TRAAK, TRESK or TASK-1, nor the hERG channel, which is routinely checked because blocking it affects cardiac repolarization.
What PE-22-28 has been studied in
All published work on PE-22-28 is preclinical. In the 2017 study:
- Cell electrophysiology in hTREK-1/HEK293 cells, described above.
- Cultured mouse cortical neurons, where the authors measured the synaptic protein PSD-95 by Western blot.
- Mice (male C57Bl/6J), tested in behavioral assays that are standard in antidepressant research: the forced swim test, novelty-suppressed feeding and learned helplessness, plus a corticosterone model. Hippocampal cell proliferation was measured by BrdU labeling.
- Duration of action in mice, where the analogs acted for longer than spadin had in earlier work.
These are models used to screen compounds; findings in them are not evidence about people. No human trial of PE-22-28 or spadin is listed on ClinicalTrials.gov as of this writing.
Names and variants to watch for
Several closely related molecules circulate under similar labels, and a paper or a supplier document can mean any of them.
| Name | What it is |
|---|---|
| Spadin | The 17-residue parent, PE 12-28 |
| PE 22-28 | The seven-residue fragment, GVSWGLR |
| G/A-PE 22-28 | PE 22-28 with the N-terminal glycine replaced by alanine |
| Biotinylated G/A-PE 22-28 | The same analog with a biotin tag, used for detection and pull-down work |
The single Gly-to-Ala change is the kind of substitution often made to slow breakdown by aminopeptidases, which trim residues from the N terminus. In the 2017 study the shortened analogs were reported to act for longer in mice than spadin had in the earlier work, and the authors attributed that to greater stability rather than to a different target.
When ordering or citing, check the sequence, not the label: GVSWGLR and AVSWGLR differ by one methyl group and by one amino acid name, and they are not the same compound.
Regulatory status
PE-22-28 is not approved as a medicine in the United States or elsewhere, and it has no approved counterpart. It is a research compound. Anhydrolabs supplies it for in-vitro laboratory research only; see research use.
Other short peptides in our catalog have a very different regulatory history. Selank and Semax, for example, are registered medicines in Russia, as described in Selank vs Semax, and the Khavinson short peptides are covered in bioregulator peptides.
Chemistry and handling
Anhydrolabs supplies PE-22-28 as a lyophilized powder in vacuum-sealed vials, and also as 10-vial kits. Each vial is labeled by net content (for example, 5 mg).
Composition. With one arginine and a free N-terminal amine, and no acidic side chains, PE-22-28 carries a net positive charge at neutral pH. Tryptophan and leucine give it some hydrophobic character, but at seven residues it remains soluble in water.
Dissolving. Sterile water is the usual first solvent. If a lot dissolves slowly, a small amount of dilute acetic acid helps, because the peptide is basic. Stock solutions are described by concentration, for example 1 mg/mL, which is about 1.29 mM at 773.9 g/mol before correction for salt and water content. The general method is in how to reconstitute peptides.
Tryptophan. Trp is the most light-sensitive and oxidation-prone residue in the sequence. Oxidized tryptophan products appear in mass spectra as +16 and +32 Da species. Keep solutions protected from light, minimize headspace and use them promptly.
Storage. Keep lyophilized vials at -20 °C, dry and dark. Split dissolved stock into single-use aliquots and freeze them; avoid repeated freeze-thaw.
Purity and identity. Mass spectrometry should show the 773.9 g/mol parent. Purity by HPLC is reported as the area of the principal peak relative to all integrated signals; see peptide purity. A seven-residue sequence is short for solid-phase synthesis, and the usual impurities are deletion sequences, described in how peptides are made.
Assay notes. Picomolar IC50 values mean that very dilute working solutions are involved, and peptides at those concentrations can be lost to plastic surfaces. Low-binding tubes and a carrier protein in the working buffer help keep the nominal concentration honest.


