Some sequences are harder than others, and the difficulty is usually predictable from the sequence itself. Aspartimide formation is the clearest example: a largely preventable side reaction — though how well it is prevented depends on the sequence and the conditions — that quietly produces three wrong products alongside the right one, two of which weigh exactly the same as the peptide you ordered.
During standard Fmoc deprotection — 20% piperidine in DMF — the backbone amide nitrogen of the residue following an aspartate attacks that aspartate's side-chain ester. The result is a five-membered imide ring, with the side-chain protecting group expelled.1
The ring does not survive. It is reopened by water or by the piperidine itself, and it can be attacked at either carbonyl. Opening at one gives back the normal α-peptide; opening at the other gives a β-peptide, in which the backbone now runs through the side chain. Piperidine attack gives α- and β-piperidide adducts. And because the intermediate is planar, reopening can scramble the stereocentre, so D-epimers of those products are commonly observed alongside them. How much epimerisation occurs depends on the sequence and the deprotection conditions.1
This is why mass spectrometry alone will not catch it. The β-peptide and the epimers are isomers of your peptide — identical mass, different molecule. Chromatography is the only routine way to separate them, and then only if the analysis is looking. Separation is not guaranteed by a standard analytical gradient either: resolving the beta-peptide and the epimers can require a different column, a shallower gradient or orthogonal characterisation. A shoulder on the main HPLC peak in an Asp-containing sequence deserves attention rather than integration into the main peak.
The attacking nucleophile is the backbone nitrogen of the residue immediately after the aspartate, so the identity of that residue governs the risk. Asp-Gly is the worst case, most likely because glycine is the least sterically hindered residue and its nitrogen has the clearest approach.1 The problem also occurs at Asp-Asn, Asp-Asp, Asp-Gln and Asp-Arg, though less severely; in the Val-Lys-Asp-Xaa-Tyr-Ile model series, appreciable by-product was also seen with cysteine and with unprotected threonine in that position.6 The reported order is Asp-Gly ≫ Asp-Arg ≥ Asp-Asp.22
Sequences containing Asp-Gly should be identified at the ordering stage, since the motif changes how the peptide is assembled.
| Approach | How it works | Limits |
|---|---|---|
| Bulkier side-chain ester OMpe, OEpe, OBno in place of OtBu |
Steric hindrance slows the ring closure. The ranking is consistent: the less bulky the ester, the more aspartimide, with OtBu ≫ OEpe > OBno.2 | Reduces but does not eliminate. Ester-based protection alone has never been shown to stop it completely.1 |
| Backbone protection Dmb or Hmb on the following glycine |
Puts a group on the attacking nitrogen, so the ring cannot close at all. | Coupling efficiency of the protected residue is poor, so it must be introduced as a preformed dipeptide such as Fmoc-Asp(OtBu)-(Dmb)Gly-OH.3,4 These dipeptides are supplied in the form Fmoc-Aaa-(Dmb)Gly-OH, so the approach applies where the residue following the aspartate is glycine.4 |
| Milder deprotection morpholine, or piperidine plus acid |
A weaker base closes the ring more slowly. Morpholine (pKa 8.4) is markedly gentler than piperidine (pKa 11.2), and adding 0.1 M formic acid to piperidine helps.2 | Morpholine is sometimes too weak to remove Fmoc completely. Several acidic additives perform about equally — formic acid, ammonium acetate, HOBt and trifluoroethanol all give similar results.2 |
| Cyanosulfurylide (CSY) a non-ester side-chain mask |
Masks the side chain through a carbon-carbon bond rather than an ester, removing the leaving group entirely. Reported to suppress aspartimide completely, and removed with N-chlorosuccinimide.5 | Oxidation of cysteine, methionine and tryptophan is reported in connection with the method, and chlorination of tyrosine has been observed when cleavage is carried out with a large excess of N-chlorosuccinimide.2 In the original report, N-chlorosuccinimide was described as compatible with all amino acids other than methionine.5 |
No single approach is correct for every sequence. For a short peptide with one Asp-Gly, the Dmb dipeptide is usually the cleanest route. For a long sequence with several aspartates and a tryptophan, a bulky ester with acid-modified deprotection may be the better compromise. The choice is made per sequence.
Boc-based synthesis produced far less aspartimide, largely because cyclohexyl ester protection was used on aspartate alongside in-situ neutralization protocols.1 The move to Fmoc chemistry brought milder cleavage and better functional group tolerance, and brought this problem with it. It is a trade, not a regression — but it explains why older literature may understate the issue.
| Side reaction | Where it appears | Usual control |
|---|---|---|
| Racemization during coupling | Cysteine and histidine are the most vulnerable; prolonged pre-activation and excess base both make it worse | Shorter activation times, milder base, and coupling reagents chosen for the residue |
| Oxidation | Methionine, cysteine and tryptophan, during cleavage and in storage | Appropriate scavengers in the cleavage cocktail; careful handling afterwards |
| Tryptophan alkylation and sulfonation | During cleavage, especially with Mtr, Pmc or Pbf arginine protection | Fmoc-Trp(Boc) at the synthesis stage, rather than a better cocktail |
| Diketopiperazine formation | At the dipeptide stage, particularly with proline or glycine in the first two positions | Resin and linker choice; a different coupling order |
| Aggregation on resin | Beta-sheet-prone and hydrophobic stretches; onset depends on the sequence rather than on a fixed chain length, and is usually first seen as the resin ceasing to swell | Pseudoproline dipeptides, backbone protection, or elevated temperature |
Please email your project details to sales@lifetein.com or use our online quotation form. Difficult sequences are routine work here, and we will tell you before synthesis if your sequence carries a known risk and what we propose to do about it.
Sequences supplied by customers are treated as confidential and are never disclosed or reused.
All LifeTein peptides are supplied for laboratory research use only. They are not for human or veterinary use, and not for diagnostic or therapeutic use.