PE-22-28 peptide: what the mouse studies show and why there is no human evidence

The PE-22-28 peptide is a fragment of a fragment. The story begins with sortilin, a receptor protein also called neurotensin receptor 3, which is made as a longer precursor. When the precursor is processed, a 44-amino-acid propeptide is released. In 2010 Jean Mazella, Marc Borsotto and colleagues at the CNRS in Valbonne reported that a 17-amino-acid piece of that propeptide, which they named spadin, binds the TREK-1 potassium channel with an affinity of about 10 nM, blocks it, and produces antidepressant-like effects in mice within four days rather than the three to four weeks typical of conventional antidepressants ^[1].

Spadin has a weakness as a drug candidate: its activity in mice disappears within about seven hours of dosing. In 2017 the same group studied how spadin is broken down in blood and used that information to design a seven-amino-acid peptide corresponding to residues 22 to 28 of the propeptide, hence the name PE 22-28. On human TREK-1 channels expressed in cells, PE 22-28 blocked the channel with a half-maximal concentration of 0.12 nM, compared with 40 to 60 nM for spadin, and its action in mice lasted up to 23 hours instead of 7 ^[2].

PE-22-28 is therefore best understood as an experimental antidepressant lead compound, not a nootropic in the way Semax or Selank are marketed. Research peptide stores sell it as a freeze-dried vial and sometimes as a nasal spray; PepFinder lists only vials. There is no registered trademark, no drug company sponsor that we could find, and no clinical development programme.

TREK-1 is a two-pore-domain potassium channel that helps set the resting electrical state of neurons. Mice engineered to lack TREK-1 behave as if they had been given an antidepressant, which is why the channel became a drug target ^[1]. Spadin and PE-22-28 are proposed to work by blocking TREK-1 in the same way, which depolarises neurons and, in the dorsal raphe nucleus, increases the firing rate of serotonin neurons ^[1].

The group has also described downstream effects. In cultured neurons, spadin activated the MAPK and PI3K signalling pathways, protected cells against a chemical trigger of apoptosis, and increased two markers of synapse formation, PSD-95 and synapsin; in mice, spadin injections rapidly increased brain-derived neurotrophic factor (BDNF) in the hippocampus ^[3]. PE-22-28 and its derivatives raised PSD-95 in mouse cortical neurons and induced neurogenesis after only a four-day treatment ^[2].

Whether blocking TREK-1 is entirely benign is a fair question, because the channel also protects against pain, seizures and ischaemic injury. A 2012 study from the same laboratory reported that spadin did not increase pain sensitivity, seizure susceptibility or infarct size after experimental stroke in mice, did not block four other potassium channels, and did not alter heart rhythm or blood pressure ^[4]. A separate laboratory reported in 2020 that spadin selectively antagonises the activation of TREK-1 by arachidonic acid rather than blocking the channel outright ^[5], which suggests the picture of the spadin peptides as simple channel blockers is incomplete. A 2019 study found that a low dose of the shortened analogue actually increased TREK-1 activity while a higher dose inhibited it, a biphasic action the authors exploited to protect the brain after experimental stroke ^[6].

Every efficacy result for PE-22-28 comes from mice or cultured cells, and nearly all of it from one group at the CNRS in Valbonne. That does not make the work wrong, but it does mean it has not been independently replicated, and rodent behavioural tests of depression have a poor record of predicting benefit in people.

Antidepressant-like behaviour: in the forced swimming test, mice given PE-22-28 or its derivatives spent significantly less time immobile. In the novelty-suppressed feeding test, a four-day treatment shortened the time mice took to start eating in an unfamiliar arena. The parent peptide spadin had produced the same pattern across five behavioural tests in 2010 ^{[1] [2]}.

Neurogenesis and synapses: a four-day course of PE-22-28 increased the birth of new neurons in the hippocampus, with the strongest effect from a modified version the authors call G/A-PE 22-28, and raised PSD-95 in cultured cortical neurons ^[2]. These are the same markers used to argue that conventional antidepressants work by promoting plasticity over weeks; the claim here is that the peptide does it in days.

Stroke and post-stroke depression: in a 2019 mouse study of focal ischaemia, a single very low dose of the shortened spadin analogue (0.03 µg/kg into the abdomen) given 30 minutes after the onset of stroke and daily for seven days, followed by a higher dose (3 µg/kg) four days a week, prevented weight loss and delayed loss of dopamine neurons, improved motor and cognitive deficits, and prevented depression-like behaviour. Enhanced neurogenesis and synaptogenesis were proposed as the mechanism ^[6].

Other spadin findings, not yet shown for PE-22-28: retro-inverso versions of spadin, built from mirror-image amino acids to resist breakdown, had stronger antidepressant effects in mice ^[7]; spadin increased insulin secretion from pancreatic beta cells ^[8]; and spadin altered the electrical conductance of astrocytes, the brain’s support cells, through TREK-1 channels paired with a related channel ^[9]. A 2019 review by the developers summarises the whole programme, including the retro-inverso, cyclised and shortened strategies that led to PE-22-28 ^[10].

There are none. We found no clinical trial of PE-22-28 or spadin registered on ClinicalTrials.gov and no human study of either peptide in PubMed as of September 2026. The only human data from this research programme concern the natural propeptide as a biomarker: the group has measured serum levels of the sortilin-derived propeptide in people with depression, including a 2018 report that levels rose after electroconvulsive therapy in patients with treatment-resistant depression ^[11]. That is a study of the body’s own peptide as a marker, not a treatment study, and it says nothing about giving PE-22-28 to people.

So does PE-22-28 work? In mice, by the developers’ own tests, yes. In humans, nobody knows. The gap between a fast-acting antidepressant effect in a mouse forced swim test and a treatment for depression in people is one that many compounds have failed to cross, and there is no pharmacokinetic, safety or dose-finding data in humans for this peptide.

In mice, behavioural effects were measured after a single dose in the forced swim test and after four days of dosing in the novelty-suppressed feeding test, and the peptide’s action lasted up to 23 hours per dose compared with 7 hours for spadin ^[2]. The 2019 stroke study dosed for seven days and then four days a week over weeks ^[6]. There is no human information. Reports of effects within hours in people come from users of an unstudied compound, not from research.

These are doses used in animal studies, reported here because people search for them. They are not recommendations and cannot be converted into a human dose; no human dose has ever been tested. The 2017 paper that introduced PE-22-28 gave the peptide to mice by injection and reports that a four-day sub-chronic treatment was enough for behavioural and neurogenesis effects, with the abstract not specifying a per-kilogram dose ^[2]. In the 2019 stroke study, the shortened analogue was given into the abdomen at 0.03 µg/kg once daily for seven days, then 3 µg/kg once daily four days a week; the low dose activated TREK-1 and the high dose inhibited it, so in this model the dose changed the direction of the drug’s action ^[6]. Spadin itself was originally given intravenously for four days ^[1].

Vendor dosing charts for PE-22-28, typically quoting a fixed number of micrograms by nasal spray or subcutaneous injection, are not derived from these studies. Because a dose that is too low or too high can have opposite effects on the channel in mice ^[6], extrapolating from vendor charts is not just unsupported but potentially self-defeating. Our peptide calculator and the PE-22-28 calculator page convert vial contents into concentrations for record-keeping; they do not imply a dose.

The mouse studies gave the spadin peptides by intravenous or intraperitoneal injection ^{[1] [2] [6]}. We found no study, in animals or people, of PE-22-28 as a nasal spray or by subcutaneous injection, which are the two forms sold online. Nasal delivery of small peptides to the brain is plausible in principle, and the developers’ 2019 review discusses improving in vivo stability and bioavailability as the reason PE-22-28 was made ^[10], but no published data show how much of a nasal or subcutaneous dose reaches the brain.

Research stores sell PE-22-28 as a freeze-dried powder, most often in 5 mg or 10 mg vials for reconstitution with bacteriostatic water, and sometimes pre-mixed as a spray. PepFinder lists only vials, since a pre-mixed spray cannot be matched to a certificate of analysis.

No human safety data exist. The animal safety work concerns spadin rather than PE-22-28: in mice, spadin did not increase pain sensitivity, seizure susceptibility or infarct size, did not block the cardiac potassium currents IKr and IKs, and did not change blood pressure, heart rate or blood glucose over three weeks of treatment ^[4]. Whether the more potent seven-residue analogue shares that profile has not been reported in detail, and the finding that low and high doses have opposite effects on TREK-1 ^[6] means the safety margin is not straightforward.

Theoretical concerns follow from the mechanism. TREK-1 is expressed in the heart, the gut, the airways and blood vessels as well as the brain, and it is part of the body’s protection against ischaemia and seizures. A compound that blocks it more potently than spadin, for longer, has unknown effects on those systems in people. The neurogenesis and BDNF effects that make the peptide interesting as an antidepressant also mean it is biologically active in ways that have not been characterised over months of exposure.

Products sold as PE-22-28 are unregulated, and the peptide has no pharmacopoeial standard. Our guide to spotting a fake peptide supplier covers the checks that apply.

PE-22-28 is often listed alongside the Russian peptides because all three are sold as nasal sprays for mood or cognition, but the comparison ends there. Selank is registered in Russia as an anxiolytic and has small human studies; Semax is registered there as a nootropic with small stroke studies; both have decades of use. PE-22-28 was first described in 2017, has no human data of any kind, and targets a potassium channel rather than the enkephalin or BDNF pathways proposed for the Russian peptides. Our Semax vs Selank comparison covers those two in depth. Another peptide sold for sleep and mood, DSIP, likewise has older human studies that PE-22-28 lacks.

PE-22-28 is not an approved medicine anywhere, has never been submitted for approval, and is not in clinical development that we could find. It is not on the US FDA’s list of compounding bulk substances that were nominated and withdrawn, nor in its category 2 list, so it has no compounding status; it is simply an unapproved research chemical. Spadin and its analogues are covered by patents held by the CNRS and associated institutions, which does not affect a buyer but does mean no licensed product exists. See our legal status overview, including the UK, US and European pages, for how unapproved peptides are treated in each market.

PE-22-28 is supplied as a freeze-dried powder. The developers designed it specifically because spadin degrades quickly in blood ^[2], but that says nothing about shelf stability of the powder or of reconstituted solution, which have not been published. General guidance is in our peptide storage guide.

A short peptide like PE-22-28 is easy to synthesise, but easy also to substitute or under-fill, and the sequence sold as PE-22-28 by different vendors is not always stated. A certificate of analysis should show a measured mass close to 774 g/mol and an HPLC purity figure; see how to read a peptide COA and our guide to third-party peptide testing.

You can compare PE-22-28 prices, including PE-22-28 in the UK, in the US and in Europe, and see our list of third-party tested suppliers. A listing is not a statement that a product is legal to buy or appropriate to use.