PE-22-28 is one of the thinnest records this site covers: one paper, in mice, from the laboratory that designed it. That makes the dosage question short to answer and easy to misread, because the paper tested several related peptides at several amounts, and the figures that circulate for the research vial have been drawn from different parts of it. This page sets out what the paper gave, to which molecule, by which route, where it contradicts itself, and what the vendor figures look like on a syringe (sources opened 2026-10-22). It reports figures; it suggests none.
Any vial and water volume is worked by the PE-22-28 calculator, and the all-compounds dosage chart sets this row beside the rest. Medibact's PE-22-28 guide covers the compound itself.
Where PE-22-28 comes from
Sortilin, also called neurotensin receptor 3, is made with a 44-amino-acid propeptide that is cut off as it matures and released into the blood. In 2010 a French group at the CNRS Institute of Molecular and Cellular Pharmacology named one piece of it spadin (residues 12 to 28) and reported that it blocks TREK-1, a potassium channel linked to depression in mouse models. Spadin's effect faded within about seven hours.
The 2017 paper looked for something longer-lasting. Incubating spadin in mouse serum, the group identified its breakdown fragments, tested them on human cells carrying TREK-1, and found that the seven residues 22 to 28 blocked the channel at an IC50 of 0.12 nM, against 40 to 60 nM for spadin. That seven-residue fragment is PE-22-28. The paper then made modified versions — a glycine-to-alanine swap (G/A-PE 22-28) and biotinylated forms — and tested them alongside.
What the paper gave, and to which molecule
| Experiment | Molecule | Amount and route | What was measured |
|---|---|---|---|
| Forced swim test, single dose | PE-22-28 and analogs | 3.2 to 4.0 mcg/kg intraperitoneal, test 30 min later | Immobility time: 91.8 s for PE-22-28 against 161.7 s for saline |
| Same test, comparison | Spadin | 100 mcg/kg intraperitoneal | Reference |
| Four-day course | PE-22-28 and analogs | 3.0 mcg/kg intraperitoneal (text) or 1 mg/kg by gavage | Still active in the swim test; novelty-suppressed feeding |
| Duration of effect | G/A-PE 22-28, biotinylated G/A-PE 22-28 | 3.2 and 32 mcg/kg; 4.0 and 40 mcg/kg | Half-effect times 14, 17, 21 and 23 h |
| Neurons in a dish | PE-22-28 and analogs | Concentrations in culture | PSD-95 expression (synapse marker) |
Three things in that table matter more than any single number. The injected amounts are in micrograms per kilogram of mouse, so a 25 g mouse received 75 to 100 nanograms; a whole 5 mg vial is 50,000 to 67,000 such injections. The oral arm needed about 250 to 330 times the injected amount (1 mg/kg against 3 to 4 mcg/kg) to stay active, which is the paper's only statement about swallowing it. And the duration figures — the basis of the "lasts all day" line on vendor pages — were measured on the modified molecules, at two amounts each; the paper gives no half-effect time for plain PE-22-28.
The paper's own inconsistency
The text of the results says the four-day injected course used 3.0 mcg/kg. The legend of the figure that reports that course says the treatments "were administered either by ip injection (100 μg/kg) or by gavage (1.0 mg/kg)". The two cannot both describe the same mice, and 100 mcg/kg is the spadin amount used elsewhere in the paper. This page reports both and cannot settle which is the typo. It is worth knowing because this site's own dosage chart had recorded "rodent work at 100 to 500 mcg/kg" for PE-22-28 on 2026-09-18; on re-reading the full text, no 500 mcg/kg figure appears for any molecule, and the chart row is corrected with this page.
The vendor figures, and how they compare
Vendor pages recorded on this site's chart on 2026-09-18 cite 50 to 200 mcg under the skin once a day. For a 70 kg adult that is 0.7 to 2.9 mcg per kilogram — a per-kilogram figure of the same order as the mouse injections. That resemblance is not evidence: per-kilogram amounts do not carry over between species, the route differs (intraperitoneal in the mice, subcutaneous on the vendor pages), and the outcome measured in the mice was swimming behaviour 30 minutes later. No human pharmacokinetic study exists to say what any amount does in a person.
The arithmetic per vial
| Vial | Water | Concentration | mcg per U-100 unit | 50 mcg | 100 mcg | 200 mcg |
|---|---|---|---|---|---|---|
| 5 mg | 1 mL | 5,000 mcg/mL | 50 | 1 unit | 2 units | 4 units |
| 5 mg | 2 mL | 2,500 mcg/mL | 25 | 2 | 4 | 8 |
| 5 mg | 3 mL | 1,667 mcg/mL | 16.7 | 3 | 6 | 12 |
| 10 mg | 1 mL | 10,000 mcg/mL | 100 | 0.5 | 1 | 2 |
| 10 mg | 2 mL | 5,000 mcg/mL | 50 | 1 | 2 | 4 |
| 10 mg | 3 mL | 3,333 mcg/mL | 33.3 | 1.5 | 3 | 6 |
One U-100 unit is 0.01 mL. At the circulating figures the draws are small: in 1 mL of water a 10 mg vial puts 50 mcg on half a unit mark, where a half-unit reading error doubles or removes the amount. More water spreads the same amount over more of the barrel; the insulin syringe units page shows how 30-, 50- and 100-unit barrels read. A vial holds 25 to 100 of the circulating amounts at 5 mg and 50 to 200 at 10 mg, whatever the water volume.
What is not established
No human dose, route, half-life or safety record exists for PE-22-28. The single published study is a mouse study by the group that designed the molecule, with its strongest duration data on modified versions and one internal inconsistency in its methods. The antidepressant-like effects are behavioural tests in mice. Nothing in the record says what PE-22-28 does in a person. Decisions about administering anything belong with a clinician; comparisons with other compounds sold for mood or cognition are on the compound dosage directory.
Sources and dates
Opened 2026-10-22: Djillani A, Pietri M, Moreno S, Heurteaux C, Mazella J, Borsotto M, "Shortened Spadin Analogs Display Better TREK-1 Inhibition, In Vivo Stability and Antidepressant Activity", Front Pharmacol 2017;8:643 (doi 10.3389/fphar.2017.00643, full text via PubMed Central); PubMed title/abstract searches for "PE 22-28"/"PE-22-28" (1 record) and spadin (41); ClinicalTrials.gov search for PE-22-28 (0 studies); DailyMed (no label). Vendor figures as recorded on this site's dosage chart on 2026-09-18. Concentration, unit and per-vial arithmetic by this site. Corrections go to the contact page.