Same mass, different molecule: why identity by mass is not identity

The single most consequential fact about deamidation is that the isoaspartate product has the same molecular weight as the parent peptide, because the rearrangement moves an atom rather than adding or removing one. A laboratory confirming identity by molecular ion mass alone will report a substantially deamidated preparation as the intended compound. Separation is possible — the isomers usually resolve on a sufficiently shallow reversed-phase gradient, and specialist methods resolve them reliably — but only if the method was designed to look. A twelve-minute generic gradient does not look.

Deamidation, and the isomer with the same mass

Deamidation of asparagine proceeds through nucleophilic attack by the backbone nitrogen of the following residue on the asparagine side-chain carbonyl, forming a five-membered succinimide intermediate which then hydrolyses to a mixture of aspartate and isoaspartate, conventionally in a ratio favouring the isomer roughly three to one. Glutamine deamidates by an analogous route, far more slowly, through a six-membered intermediate.

Three factors govern the rate. Sequence is dominant: the residue immediately following the asparagine determines how readily the intermediate forms, and asparagine-glycine is the fastest motif known, with serine, histidine and alanine following. Solution pH matters, with the rate minimal in the mildly acidic region and rising steeply above neutrality as the backbone nitrogen becomes more nucleophilic. Temperature and water activity set the overall pace, which is why the solid state helps so much.

The analytical problem is that isoaspartate has the same elemental composition and therefore the same molecular mass as the parent. Identity confirmation by molecular ion alone cannot distinguish them, and a preparation that is substantially deamidated will present as the intended compound. The isomers usually separate on a sufficiently shallow reversed-phase gradient, and specific methods exist, but only a method designed for the question will find the answer.¹

Hydrolysis, isomerisation and the slower routes

Backbone hydrolysis cleaves an amide bond outright and produces two fragments, each of which is a distinct chromatographic species and each of which is detectable by mass. It is generally slower than deamidation at ordinary storage conditions but becomes dominant at low pH and elevated temperature, which is one reason accelerated stability data for peptides extrapolates so poorly: the pathway that dominates at forty degrees may be irrelevant at five.

Certain positions are much more labile than others. Aspartate-proline and aspartate-glycine bonds hydrolyse relatively readily under acidic conditions. N-terminal glutamine can cyclise to pyroglutamate, losing ammonia. Peptides with an N-terminal sequence of the right geometry can form a diketopiperazine and shed the first two residues as a cyclic dipeptide, a route that is fast enough at neutral pH to matter for some sequences.

Racemisation at susceptible residues produces epimers that are chemically identical in composition and differ only in stereochemistry. They are among the hardest impurities to detect, requiring either a chiral method or a sufficiently discriminating reversed-phase separation, and they are essentially never reported. A vial can be nominally pure by every measurement on its certificate and contain a percentage of a diastereomer with unknown biological behaviour.

What makes a method stability-indicating

A stability study is only as good as the analytical method behind it, and the requirement has a name: the method must be stability-indicating, meaning it must resolve the parent compound from its degradation products and quantify the change. Establishing that is done by forced degradation — deliberately stressing the material with acid, base, oxidant, heat and light — and demonstrating that the resulting products are separated from the parent and from each other with adequate peak purity.

Almost nothing sold as a purity determination in this market has been validated that way. A generic peptide gradient run for twelve minutes may perfectly well resolve the parent from its two largest process impurities and entirely fail to resolve it from its isoaspartate isomer or a closely related oxidation product. The number it returns is a purity figure, not a stability measurement, and using a series of such figures to argue that a product has not degraded is a category error.

The compendial guidance on analytical validation is explicit about specificity, and about demonstrating it against the degradation products the molecule can actually form. The gap between that expectation and practice in this trade is not a matter of dishonesty. It is that the method being sold was designed for a different purpose and is being asked a question it was not built to answer.²

A single-use indicator records that a threshold was crossed and not for how long. Ten minutes on a loading bay and eleven hours in a hot van produce the same mark, which is why an indicator is a prompt to ask a question rather than an answer to one.

Why accelerated data extrapolates badly for peptides

The temptation with any stability programme is to run the accelerated condition, fit an Arrhenius relationship to the rate constants, and extrapolate to the intended storage temperature. For a single reaction with a temperature-independent mechanism that is sound. For peptides it frequently is not, and the reason is that different pathways have different activation energies.

Suppose a peptide degrades at five degrees principally by deamidation and at forty degrees principally by hydrolysis, with the second having a higher activation energy. Measuring total degradation at forty degrees measures mostly hydrolysis; extrapolating that rate down to five degrees predicts almost nothing about the deamidation that will actually dominate. Aggregation is worse still, because it is frequently nucleated by interfaces and mechanical stress rather than by thermal energy alone, and does not obey a simple temperature relationship at all.

The practical rule the Journal applies when reading a stability claim is to ask what condition the data was generated at and whether the degradation products were identified as well as quantified. Accelerated data that shows which products form is genuinely useful as a warning of what to watch for. Accelerated data reduced to a single percentage and extrapolated to a shelf life is a projection dressed as a measurement, and for this class of molecule it is a poor projection.

Light, and the exposure nobody records

Photostability has its own guideline, its own defined light source options and its own exposure requirement expressed in lux hours of visible light and watt hours per square metre of near ultraviolet. Products are tested in the immediate container, and where they fail, in the marketing pack, and where they fail again the label carries a protection instruction. The chemistry is real: tryptophan and tyrosine absorb in the near ultraviolet and photo-oxidise, and photolytic disulphide cleavage is well documented.

Nothing about light exposure is recorded anywhere in the research-peptide supply chain. Vials are frequently supplied in clear glass. Photographs for listings are taken under studio lighting. Parcels are opened on kitchen counters. A reconstituted vial may sit on a shelf under a window for weeks. The cumulative exposure is unknown and unknowable, and it is plausibly a larger contributor to degradation than the transit excursions that attract all the attention.

The Journal makes one narrow observation rather than a recommendation, because recommendations are not this publication’s business. Amber glass, or a secondary carton, costs a fraction of a cent per unit and removes an uncontrolled variable entirely. Several of the twenty companies we track ship in amber vials as standard, and the remainder will supply amber on request at no extra charge, which is the answer a buyer who cares about photostability should ask for.³

Mean kinetic temperature is a legitimate tool for a whole storage period and is routinely misapplied to a single shipment, where it is used to argue that a two-day excursion averaged out. The arithmetic is sound; the frame is not.

Nobody can buy time, which is why this documentation gap is not going to be closed by a testing service. What can be done immediately is to distinguish a measurement from a convention: to say twelve months at minus twenty, ongoing, rather than two years, and to say that an in-use period has not been established for this product rather than borrowing one from a pen.