A peptide in a vial changes over time. Some changes break the chain, some rearrange it without changing its mass, and some make the peptide stick to the wall of the tube or clump together. How fast this happens depends less on the peptide as a whole than on a few short stretches of its sequence, and on the conditions around it: water, pH, temperature, oxygen, light and the buffer[1].
The short answer
- Dry is usually more stable than dissolved. Many peptides keep better in a suitably formulated freeze-dried (lyophilised) state than in aqueous solution, which is why many are supplied as powder. Reconstitution, however, adds a risk of contamination[1].
- For a medicine, the label decides. Storage conditions belong to a specific formulation, not to the peptide in general. Of three peptide medicines discussed below, one must be refrigerated before first use, one is kept at room temperature, and one must not be refrigerated after mixing[2][3][4].
- There is no universal best pH. Asparagine deamidates fastest at neutral and basic pH, while aspartate bonds break fastest in acid[5].
- The sequence shows the weak points. Asn followed by Gly, Asp followed by Gly, Pro at position 2, and Met, Cys or Trp anywhere are the usual suspects[6][5][7][1].
- Not all losses are chemical. At low concentrations, much of a peptide can simply stick to the container[8].
Three different questions
"Is it still good?" hides three separate questions. Chemical stability asks how much of the original molecule is left: covalent bonds are made or broken, and new compounds appear. Physical stability asks whether the molecule is still in the right state: dissolved rather than aggregated, precipitated or stuck to a surface[1]. Microbiological safety asks whether the solution has been contaminated, a risk that appears as soon as a vial is reconstituted or opened[1].
The three do not predict each other. A clear solution can hold a peptide that has partly deamidated, a fully intact peptide can be lost to the tube wall, and neither tells you whether the solution is sterile. This is why multi-dose diluents for injection contain a preservative, such as the benzyl alcohol in the bacteriostatic water supplied with tesamorelin[4].
Asparagine: the fastest clock
Deamidation is the loss of the side-chain amide group of asparagine (Asn) as ammonia. It runs through a ring: the nitrogen of the next peptide bond attacks the Asn side chain and forms a five-membered succinimide. The ring then opens with water, giving either a normal aspartate or an isoaspartate, in which the chain continues through the side chain. The succinimide can also flip its configuration, so D-aspartate and D-isoaspartate appear as well[9].
The model peptide Val-Tyr-Pro-Asn-Gly-Ala shows how fast this can be. At pH 7.4 and 37 °C its half-life was 1.4 days. Its succinimide was opened with a half-time of 2.3 hours and changed configuration with a half-time of 19.5 hours, so the product was a mixture of L- and D-, normal and iso forms[9].
The neighbour on the carboxyl side matters most. A survey of 306 Asn sequences found half-times between 1 day and 455 days. Glycine after Asn gave about a day; bulky branched residues such as valine and isoleucine gave 250 to 320 days; proline, which cannot form the ring, gave more than 1000 days[6]. Replacing the glycine in the hexapeptide above with leucine or proline slowed degradation 33- to 50-fold[9]. Glutamine deamidates much more slowly, with half-times extending beyond 10 years[6].
To limit deamidation, peptide formulations are preferably kept between pH 3 and 5[1]. In the Asn-Gly hexapeptide, degradation at pH 7.4 was about 25 times faster than for the same peptide with Asp in place of Asn[5].
Aspartate: isomerisation and cleavage
Aspartate (Asp) can form the same succinimide by losing water, but about 34 times more slowly than Asn in the same sequence[9]. The result is the same mixture of normal and iso forms. Because Asp and isoAsp have the same mass, this change is invisible to a simple mass measurement; it has to be detected by separation, for example by HPLC, where the isoAsp form elutes at a different time[5].
In acid, aspartate does something else: its side-chain carboxyl group most likely helps to break the peptide bond that follows it. In the hexapeptide Val-Tyr-Pro-Asp-Gly-Ala at 37 °C, the main reaction between pH 0.3 and 3 was cleavage of the Asp-Gly bond into two fragments. At pH 4–5 the peptide both cleaved and isomerised, and above pH 6 it only interconverted slowly and reversibly with its isoAsp form[5].
The two residues therefore pull in opposite directions. The Asn peptide was most stable at pH 3–5, exactly where the Asp peptide degraded fastest; at neutral and basic pH the Asp peptide was the more stable of the two[5]. The best pH for a solution is a property of the particular sequence.
Other reactions
Cleavage next to serine. The hydroxyl group of serine can attack the neighbouring peptide bond. Gonadorelin and triptorelin, two hormone analogues, break on the N-terminal side of serine at pH 5–6 and lose the configuration of serine above pH 7[1].
Diketopiperazines. When proline is the second residue, the free N-terminal amino group can attack the bond between residues 2 and 3, cutting off the first two amino acids as a cyclic dipeptide, a diketopiperazine. In model compounds the uncharged amino group was the reactive form, so the reaction speeds up as pH rises towards neutral; the residue before proline also changed the rate, Ala-Pro reacting twice as fast as Val-Pro[7]. The same reaction has been reported in recombinant human growth hormone and in substance P, whose second residue is proline[7].
Oxidation. Methionine, cysteine, tryptophan, tyrosine and histidine are the residues most prone to oxidation. Methionine forms methionine sulfoxide and, further, methionine sulfone. Oxygen, traces of iron and copper ions, light and peroxides from excipients all drive these reactions; formulators counter them by excluding air, adding antioxidants or adding chelating agents that bind metal ions[1].
The buffer itself. Buffers are not inert. The somatostatin analogue octastatin degraded faster in phosphate than in glutamate buffer, and faster still as the phosphate concentration rose; oxytocin was most stable at pH 4.5[1].
Physical losses
Aggregation. Peptides can associate into dimers, larger soluble aggregates and finally insoluble precipitates. Higher concentration, freezing, heating and agitation all promote this. Some peptides, including calcitonin, leuprolide and β-amyloid, can form gel-like aggregates as their structure shifts towards β-sheet[1].
Adsorption. Peptides stick to surfaces, and the loss can be large. In a study of eight radiolabelled hormones kept for 48 hours in different tubes, recovery depended on both the peptide and the material: GLP-1, for example, was recovered at 3.7% from polystyrene and 20.5% from polypropylene. Coating glass with silicone made recovery worse rather than better, while adding 1% albumin improved it; freeze-drying the peptides with albumin in the best-suited tubes gave more than 89% recovery for all eight[8]. These experiments used trace amounts of peptide, where adsorption matters most, but they show that the container is part of the storage conditions.
Storage of approved medicines
For approved peptide medicines, the storage conditions are established by stability testing of the exact product and printed on the label. They differ considerably, and they apply to the specific presentation named: even different delivery formats of the same medicine can have different storage instructions. Ozempic in single-dose prefilled syringes, for example, may be kept at 8–30 °C for up to 28 days, whereas the pen may be used for 56 days after first use[2].
| Medicine | Before use | After opening or mixing |
|---|---|---|
| Ozempic multi-dose pen (semaglutide) | Refrigerator, 2–8 °C | 56 days at 15–30 °C or in the refrigerator |
| Insulin vial (Humulin R) | Refrigerator until expiry, or 31 days at up to 30 °C | 31 days, refrigerated or at room temperature |
| Tesamorelin powder (Egrifta WR) | Room temperature, 20–25 °C, protected from light | 20–25 °C for up to 7 days; do not refrigerate |
None of the three may be frozen. Sources: Ozempic[2], Humulin R[3] and Egrifta WR[4] prescribing information.
The tesamorelin example shows why a rule such as "keep peptides in the fridge" does not transfer from one product to another. Its powder contains a cyclodextrin and mannitol as excipients, and it is mixed with a preserved diluent[4]; the conditions on its label were set for that combination. A different formulation of the same peptide can have different conditions: the label notes that the two tesamorelin formulations differ in their storage requirements and are not interchangeable[4].
In the vial, in a blood sample, in the body
Stability during storage is a different problem from stability in the body, where peptides meet enzymes that cut them. This starts even in a blood collection tube: in blood samples left without protease inhibitors, the enzyme DPP-4 removed the first two amino acids of GLP-1 and related hormones, while tubes containing inhibitors kept GLP-1 intact for more than 96 hours at room temperature. Antibody-based assays did not always detect the breakdown that mass spectrometry revealed[10].
In the body, fast clearance is one of the main obstacles for peptide medicines, and replacing natural amino acids with unnatural ones does not always give the protection hoped for[11]. Examples on this site range from sermorelin, cut by DPP-4 within minutes, to retatrutide, designed to last about a week. How these differences arise is explained in peptide half-life.
In short
- Peptides change in storage chemically (new compounds form), physically (aggregation, adsorption) and microbiologically (contamination); the three are separate questions.
- Asn-Gly is a particularly deamidation-prone motif: in model peptides it deamidated with a half-time of about a day at pH 7.4 and 37 °C. Actual rates depend on the surrounding sequence, the structure of the molecule and the formulation.
- Asp bonds break mainly in acid, Asn deamidates mainly at neutral and basic pH, so there is no single best pH for every peptide.
- Pro at position 2, serine, and oxidisable residues such as Met, Cys and Trp add further, predictable risks.
- For medicines, follow the label: storage conditions belong to the formulation, and the three examples here range from "refrigerate" to "do not refrigerate after mixing".
References
- Nugrahadi PP, Hinrichs WLJ, Frijlink HW, Schöneich C, Avanti C (2023). Designing formulation strategies for enhanced stability of therapeutic peptides in aqueous solutions: a review. Pharmaceutics 15(3):935. DOI: 10.3390/pharmaceutics15030935
- Novo Nordisk (2026). Ozempic (semaglutide) injection, for subcutaneous use: prescribing information. Revised 05/2026. novo-pi.com (PDF)
- Eli Lilly and Company (2023). Humulin R U-100 (insulin human) injection: prescribing information. Revised 6/2023. uspl.lilly.com
- U.S. Food and Drug Administration (2025). EGRIFTA WR (tesamorelin) for injection, for subcutaneous use: full prescribing information. Revised 3/2025. accessdata.fda.gov (PDF)
- Oliyai C, Borchardt RT (1993). Chemical pathways of peptide degradation. IV. Pathways, kinetics, and mechanism of degradation of an aspartyl residue in a model hexapeptide. Pharmaceutical Research 10(1):95–102. DOI: 10.1023/A:1018981231468
- Robinson NE, Robinson AB (2001). Molecular clocks. Proceedings of the National Academy of Sciences USA 98(3):944–949. DOI: 10.1073/pnas.98.3.944
- Goolcharran C, Borchardt RT (1998). Kinetics of diketopiperazine formation using model peptides. Journal of Pharmaceutical Sciences 87(3):283–288. DOI: 10.1021/js970325m
- Goebel-Stengel M, Stengel A, Taché Y, Reeve JR Jr (2011). The importance of using the optimal plasticware and glassware in studies involving peptides. Analytical Biochemistry 414(1):38–46. DOI: 10.1016/j.ab.2011.02.009
- Geiger T, Clarke S (1987). Deamidation, isomerization, and racemization at asparaginyl and aspartyl residues in peptides. Succinimide-linked reactions that contribute to protein degradation. Journal of Biological Chemistry 262(2):785–794. DOI: 10.1016/S0021-9258(19)75855-4
- Yi J, Warunek D, Craft D (2015). Degradation and stabilization of peptide hormones in human blood specimens. PLoS ONE 10(7):e0134427. DOI: 10.1371/journal.pone.0134427
- Al Musaimi O, Lombardi L, Williams DR, Albericio F (2022). Strategies for improving peptide stability and delivery. Pharmaceuticals 15(10):1283. DOI: 10.3390/ph15101283