Deamidation At Asparagine And Glutamine

The mechanism above pH 5 begins with the backbone nitrogen of the following residue attacking the side-chain amide carbonyl. Ammonia leaves, and a five-membered succinimide ring closes across the backbone and the side chain. That ring is strained and hydrolyses, and because water can open it at either carbonyl the products are two isomers: normal aspartate, and isoaspartate, in which the backbone now runs through what used to be the side-chain carboxyl. The isoaspartate form typically predominates, in roughly a three to one ratio.

Glutamine goes through the same chemistry but must form a six-membered glutarimide ring instead of a five-membered one, which is entropically and geometrically less favourable. Glutamine deamidation is consequently slower by roughly an order of magnitude or more under the same conditions, and in most stability work on synthetic peptides it is asparagine that dominates the observed change.

The net mass change is small and exact. Replacing NH₂ with OH removes 17.027 and adds 18.011, for a monoisotopic increase of 0.984 Da regardless of the size of the peptide.

The rate is set by the next residue

Because the attacking nitrogen belongs to the residue on the C-terminal side of the asparagine, that residue's side chain controls access to the reaction centre. Glycine, with no side chain at all, offers the least hindrance and gives the fastest deamidation known in a peptide. Proline, whose nitrogen is tertiary and has no amide proton to lose, cannot form the succinimide at all and effectively blocks the pathway.

Two further variables move the rate independent of sequence. pH matters strongly: the succinimide route is base-catalysed, so rates climb above pH 6 and fall to a minimum somewhere around pH 3 to 4, below which a slower direct hydrolysis route takes over. Temperature follows ordinary Arrhenius behaviour, which is the basis on which accelerated conditions are used to project ambient behaviour.

The failure mode: a 0.984 Da shift that hides in plain sight

Deamidation is the degradation route most likely to pass a routine release panel unnoticed, for two reasons that compound each other.

The first is chromatographic. A deamidated peptide differs from its parent by one amide converted to an acid, which changes charge at neutral pH but barely changes hydrophobicity. On a reverse-phase separation the deamidated species often elutes within a minute of the parent, sometimes inside its peak width, and is integrated as main peak. It shows up as a shoulder on the trailing edge, or as a tailing factor drifting from the method's historical value, rather than as a new peak.

The second is spectrometric. A 0.984 Da increase is not resolvable at nominal mass, and even at high resolution it sits close enough to the parent's isotope envelope to be missed without deliberate looking. Worse, the aspartate and isoaspartate products are structural isomers of one another with identical elemental composition, so no mass measurement of any resolution distinguishes them.

Establishing which of the two formed requires a method aimed at the isomer specifically: an ion-exchange or mixed-mode separation developed to resolve them, electron-transfer dissociation fragmentation, which produces a diagnostic fragment from isoaspartate, or an enzymatic assay using protein isoaspartyl methyltransferase. None of these falls out of a standard purity and identity panel.

What a sequence reveals before any testing

Deamidation liability is readable from the primary sequence, which makes it one of the few degradation routes that can be assessed on paper. An asparagine followed by glycine is a known hotspot; an asparagine followed by proline is effectively inert to this route. Counting Asn-Gly and Asn-Ser motifs in a sequence gives a first-order expectation of where change will appear on stability, and where it will not.

• Asn-Gly present: expect this to be the dominant degradation route in aqueous solution near neutral pH.
• Asn followed by a beta-branched residue such as Ile, Val or Thr: slow, and often outranked by oxidation at methionine or cysteine.
• No Asn and no Gln: the route is unavailable, and hydrolysis of the backbone or oxidation becomes the leading concern instead.
• C-terminal amide present: a distinct hydrolysis liability, chemically similar in outcome and also a 0.984 Da shift.

The last line matters for interpretation. A C-terminal amide hydrolysing to the free acid produces the same mass shift as side-chain deamidation, so a 0.984 Da species observed on an amidated peptide has two candidate origins and locating it requires fragmentation.

Where the published record stops

Material listed by Aurum is independently assayed for purity by reverse-phase HPLC, and identity is assayed by mass spectrometry. Those two records are the basis of the published figures, and there is a specific limit on what they can say about this degradation route: a chromatographic purity figure will only resolve a deamidated species if the method separates it, and a mass identity result cannot distinguish aspartate from isoaspartate under any circumstances.

Isoaspartate content, and deamidation rate under defined conditions, are not among the specifications Aurum publishes. Published purity figures are HPLC purity and are not net peptide content, which is also not published. Sterility, endotoxin and pyrogen testing are not performed and are not claimed.

Common questions

Is deamidation reversible?

No. The succinimide intermediate is transient and its hydrolysis is not reversible under ordinary conditions. Enzymatic repair of isoaspartate exists in biology, via a methyltransferase that re-forms the succinimide, but it is not a laboratory procedure for correcting material.

Does the lyophilized solid deamidate?

More slowly than solution, and the rate depends on residual moisture and on storage temperature relative to the glass transition. It is not eliminated by drying, only suppressed.

Which is worse, aspartate or isoaspartate?

Isoaspartate is the larger structural change, since the backbone gains an extra methylene and the original side chain becomes part of the main chain. Both are distinct chemical entities from the parent peptide and both are quantified as related substances where a method resolves them.

Can deamidation be told apart from oxidation by mass alone?

Yes, easily. Deamidation adds 0.984 Da and oxidation of methionine adds 15.995 Da. The two are not confusable at any useful resolution, which is a point in favour of looking at both.

Does freezing a solution stop it?

Lowering temperature slows it substantially, but freeze-concentration in a partially frozen solution can raise local solute concentration and pH at the same time, so repeated freezing and thawing is not a neutral operation.