Size-exclusion chromatography for aggregates and conjugates
The only routine method that sees an aggregate, and the only sensible way to characterise a polydisperse conjugate.
Alternative techniques
Disulfide mapping and the scrambling problem
A correct mass proves nothing about how the cysteines are paired. What does, and how rarely it is done.
A peptide with two cysteines has one possible disulfide. With four cysteines there are three possible pairings, all with identical mass. With six — the case for IGF-1 — there are fifteen. Only one is correct, all fifteen are isobaric, and a mass measurement is silent on the question.
The three states
| State | Mass vs correctly folded | Retention on RP | Activity |
|---|---|---|---|
| Correctly paired | reference | reference | Full |
| Reduced (open chain) | +2.016 per disulfide | Earlier — more exposed hydrophobic surface | None to low |
| Scrambled (mis-paired) | identical | Slightly later or earlier, usually 0.3–0.6 min | Reduced, sometimes to zero |
| Intermolecular dimer | +M (per partner) − 2.016 | Much later | None |
The reduced form is easy: +2 Da, earlier retention, plainly visible. The scrambled form is the hard one, and it is the reason a cyclic peptide can pass every routine test and be inactive.
Methods that actually resolve pairing
- Non-reducing versus reducing comparison. Run the sample intact and again after reduction with dithiothreitol or TCEP. A species that changes retention and gains 2 Da per disulfide on reduction was disulfide-bonded. This distinguishes reduced from oxidised, and dimer from monomer. It does not distinguish correct pairing from scrambled.
- Peptide mapping after proteolysis. Digest under non-reducing conditions with a protease that cuts between the cysteines, then identify the disulfide-linked fragment pairs by LC-MS/MS. This is the definitive method. It requires a suitable cleavage site between every cysteine pair, which is not always available.
- Partial reduction and alkylation. Reduce stepwise, cap each newly freed thiol with a distinguishable alkylating agent, then digest. Assigns pairings in order of reduction potential. Laborious.
- Comparison against a correctly-folded reference. Co-injection and retention matching. Cheap and effective where a reference exists, which for research peptides it usually does not.
The compounds this affects here
Cagrilintide, pramlintide, setmelanotide, oxytocin, desmopressin, AOD-9604 and adipotide each carry one disulfide; IGF-1 LR3 and IGF-1 DES carry three; follistatin-344 is a disulfide-rich glycoprotein. For the single-disulfide compounds a non-reducing/reducing comparison is close to sufficient, because with one bond there is only one way to be right. For the IGF variants it is not, and no report in this index establishes their pairing.
Normative references
Cited because they are the documents the acceptance criteria above are taken from. The index applies them as written and states every deviation.
- ICH Q6B — Specifications for biotechnological products
- Ph. Eur. 2.2.55 — Peptide mapping
- USP <1055> — Biotechnology-derived articles: peptide mapping
Related methods
Identity by LC-MS
What a mass measurement establishes, what resolution buys you, and the specific compound pairs where unit resolution is not enough.
Identity
Forced degradation and stability-indicating capability
Deliberately damaging a sample to prove the method can see damage — and what the resulting degradant profile says about storage.
Physical tests