Three side chains do nearly all of the oxidising in ordinary peptide handling: the thioether of methionine, the thiol of cysteine and the indole ring of tryptophan. Li, Schöneich and Borchardt's review adds histidine and tyrosine to the list of residues with high reactivity towards reactive oxygen species, and both belong there — but the three above are the ones that turn up in real chromatograms and real mass spectra, and the ones worth knowing your sequence for. The signature is simple enough to check in a minute: oxygen weighs about 16, so an oxidised molecule is about 16 heavier than the one you ordered, and on a reversed-phase column it usually comes off earlier.
Glutathione is the extreme case of the cysteine problem and has its own post, why glutathione oxidises in solution. This post is about the general case.
What each residue turns into
| Residue | Product | Nominal mass change | Notes |
|---|---|---|---|
| Methionine | Methionine sulfoxide | +16 | The most common oxidation seen in peptides. A second oxygen gives the sulfone, +32 |
| Cysteine | Disulfide, with another cysteine | −2 | Two thiols lose one hydrogen each. Intramolecular, intermolecular or a dimer — the disulfide post covers which |
| Cysteine | Sulfinic acid, then sulfonic acid | +32, +48 | The route taken when there is no partner thiol available. Not reversible in a vial |
| Tryptophan | Several products rather than one | +16, +32 and others | Torosantucci and colleagues treat characterising the products as a task in its own right, which is a fair description of the tryptophan case |
The mass arithmetic is not a literature claim, it is the periodic table: one oxygen atom is 15.9949 Da monoisotopic. That is why +16 is the number to have in your head, and why it is unmissable on an instrument whose identity tolerance is a dalton or two — the point our LC-MS post makes about what a mass check is good at.
How it shows on a mass spectrum
An oxidised population does not replace the parent peak; it appears alongside it. What you see is the expected mass, plus a satellite about 16 higher, and the relative heights give you a rough sense of how much of the material has gone over. A second satellite around +32 means either two oxidations or one on each of two residues.
One caution worth stating plainly: +16 is consistent with oxidation, not proof of it. Other modifications land on the same nominal mass — a hydroxylation is also +16 — and nominal masses collide more often than people expect. A mass shift tells you to look; the chromatography and, if it matters, fragmentation tell you where.
How it shows on an HPLC trace
Oxidation makes a molecule more polar. A sulfoxide is more polar than a thioether; an oxidised indole is more polar than the ring it came from. On a reversed-phase column, more polar means less retained, so the oxidised species generally emerges before the parent peak. This is the ordinary logic of reversed-phase separation rather than a published figure, and it is worth treating as a strong expectation rather than a law — resolution depends on the gradient, and a small oxidised population can sit as a shoulder on the leading edge of the main peak rather than as a separate peak.
Two practical consequences.
It counts against purity. An oxidised species is a different molecule from the target, so the integration puts it in the impurity fraction — one of the things that can be hiding in the figure our what is in the one percent post takes apart. A certificate reporting 99.8% by RP-HPLC has, in effect, already looked for this in the lot as released.
Detection wavelength matters. Certificates in our library report purity at 214 nm, where the peptide bond itself absorbs. Quantification at 280 nm relies on the aromatic residues instead — as the chromatogram post explains, tryptophan and tyrosine are what make a peptide visible at that wavelength. Modify the indole ring and that absorbance changes, so a tryptophan-containing peptide read at 280 nm can quietly under-report after oxidation for reasons that have nothing to do with how much material is in the tube.
Which compounds in this catalogue carry these residues?
Here is the method, because it is reusable and because the limits of it matter as much as the result.
Step one: count the sulfur atoms in the molecular formula. Methionine and cysteine are the only two standard residues containing sulfur, so a formula with no S contains neither. We read the formulas from PubChem on 29 September 2026; they agree with the figures on our own spec pages.
Step two: decide which residue the sulfur belongs to. A formula cannot tell you — one sulfur is one methionine or one cysteine. That answer comes from the published sequence, and where it does not exist on file we say so rather than guessing.
Step three: tryptophan has to come from the sequence entirely, because it contains nothing a formula makes distinctive.
| Compound | Formula | S atoms | What the sulfur is, and how we know |
|---|---|---|---|
| Sermorelin | C149H246N44O42S | 1 | Methionine. GHRH(1-29) — UniProt P01286 gives the GHRH sequence, whose only methionine is residue 27 and which contains no cysteine |
| Tesamorelin | C221H366N72O67S | 1 | Methionine, the same residue 27, in the (1-44) analogue |
| Semax | C37H51N9O10S | 1 | Methionine, the first residue of Met-Glu-His-Phe-Pro-Gly-Pro, as set out in the Semax post |
| SNAP-8 | C42H72N16O15S | 1 | Methionine, third residue of the acetylated octapeptide in the SNAP-8 post |
| L-glutathione | C10H17N3O6S | 1 | The cysteine thiol, which is the whole subject of its own post |
| MOTS-c | C101H152N28O22S2 | 2 | Two methionines. UniProt A0A0C5B5G6 gives MRWQEMGYIFYPRKLR — and that sequence also carries a tryptophan at position 3 |
| Humanin | C119H204N34O32S2 | 2 | One methionine and one cysteine. UniProt Q8IVG9 gives MAPRGFSCLLLLTSEIDLPVKRRA |
| AOD-9604 | C78H123N23O23S2 | 2 | Two cysteines, closing the loop described in the AOD-9604 post |
| Cagrilintide | C194H312N54O59S2 | 2 | Unresolved. Two sulfurs, but we hold no published sequence for it, so we will not state whether they are cysteines or methionines |
Blends inherit the property of their components, so the Selank + Semax blend, the tesamorelin + ipamorelin blend and the retatrutide + cagrilintide blend each carry a sulfur-bearing partner.
Tryptophan, being invisible to the formula filter, comes from sequences we have already published: DSIP has a tryptophan at position 1, kisspeptin-10 has one at position 3, the GHRP-6 in the CJC-1295 + GHRP-6 blend has two, and the melanocortin agonists — melanotan-1, melanotan-2 and PT-141 — are all built around the His-Phe-Arg-Trp core.
What reduces the risk
Li and colleagues sort the countermeasures into intrinsic, physical and chemical, and the physical one is the whole reason peptides are sold dry: a solid formulation is far less oxidisable than a liquid.
| Habit | What it addresses |
|---|---|
| Keep it lyophilised until needed | Removes the mobility and the dissolved oxygen that the reaction needs |
| Minimise headspace and keep vials closed | Dissolved and headspace oxygen is the supply |
| Keep it dark | Light-induced oxidation is one of the routes Li 1995 names |
| Keep it cold | Ordinary kinetics, applied to every route at once |
| Avoid metal contamination | Transition metals catalyse oxidation at trace concentrations. Clean glassware, and no metal spatulas — the same caution the GHK-Cu handling advice makes for a different reason |
| Do not reach for an antioxidant by reflex | Li 1995 reports that in metal-catalysed, site-specific oxidation, adding antioxidants may accelerate the reaction; careful chelator screening is the alternative they describe |
| Aliquot for single use | Each opening is a fresh air exposure, and a vial opened ten times has a history the label does not record |
| Mind the buffer | pH, temperature and buffer composition all influence the rate, which is part of why buffer choice is not a detail |
None of this is specific to research peptides. It is the same list the protein formulation literature has been publishing for thirty years, and it is short because the chemistry is.
Frequently asked questions
Would a certificate of analysis catch oxidation?
In the lot as released, largely yes — an oxidised species is a separate peak in the purity assay and a separate mass on the identity check, so it is exactly the kind of impurity those two assays are good at. What a certificate cannot do is describe the vial after it has spent six months in your freezer or three weeks reconstituted on a shelf. It is a statement about a sample at a moment, not a forecast.
Is a peak 16 higher always oxidation?
No. It is consistent with oxidation and it is the first thing to suspect, but a nominal +16 can arise from other modifications, and nominal masses coincide more often than the arithmetic suggests. Treat it as a question that the chromatography, and if necessary fragmentation, then answers.
Can oxidation be reversed?
Treat it as permanent in a vial. Cells reduce methionine sulfoxide enzymatically and maintain thiols in the reduced state; a tube on a bench does neither. Disulfides are the partial exception, since they can be reduced deliberately as a laboratory step — which is a decision about your experiment, not a repair.
My peptide has no methionine, cysteine or tryptophan. Is it stable?
More stable against this particular route, which is a real advantage. It says nothing about deamidation of asparagine and glutamine, hydrolysis at acid-sensitive bonds, or aggregation, all of which are covered in how long a reconstituted peptide lasts and the aggregation post. Oxidation is one of four routes, not the whole account.
Every product referenced here is supplied for laboratory research use only and is not for human or animal use.
References
- Li S, Schöneich C, Borchardt RT. Chemical instability of protein pharmaceuticals: mechanisms of oxidation and strategies for stabilization. Biotechnology and Bioengineering 1995;48(5):490-500. doi.org/10.1002/bit.260480511
- Torosantucci R, Schöneich C, Jiskoot W. Oxidation of therapeutic proteins and peptides: structural and biological consequences. Pharmaceutical Research 2014;31(3):541-553. doi.org/10.1007/s11095-013-1199-9
- Manning MC, Chou DK, Murphy BM, Payne RW, Katayama DS. Stability of protein pharmaceuticals: an update. Pharmaceutical Research 2010;27(4):544-575. doi.org/10.1007/s11095-009-0045-6
- PubChem, National Library of Medicine. Molecular formulas used for the sulfur count: MOTS-c CID 146675088; AOD-9604 CID 71300630; Semax CID 9811102; sermorelin CID 16132413; tesamorelin CID 16137828; cagrilintide CID 171397054; humanin CID 16131438; acetyl octapeptide-3 (SNAP-8) CID 71587832; glutathione CID 124886. Read 29 September 2026. pubchem.ncbi.nlm.nih.gov
- UniProt. Somatoliberin (GHRH) P01286, humanin Q8IVG9, MOTS-c A0A0C5B5G6 — sequences used to identify which residue each sulfur atom belongs to. Read 29 September 2026. www.uniprot.org/uniprotkb/P01286/entry
Every product mentioned is sold for laboratory research use only and is not for human or animal use. Nothing on this page describes or recommends use of the material sold here in humans or animals.
