Disulfide bonds in synthetic peptides

A disulfide is a covalent bond between two cysteine side chains. It is also the one structural feature of a synthetic peptide that a mass check cannot confirm on its own: the same mass can belong to the correct bridge, to a scrambled bridge between the wrong cysteines, or to a reduced chain with no bridge at all. Purity by HPLC does not settle it either, because isomers often co-elute.

Why the bond matters analytically

Each disulfide forms by oxidising two thiols and losing two hydrogen atoms: a mass decrease of 2.016 Da per bond. That shift is the easy part. The hard part is that several arrangements can produce exactly the same shift.

Species	Mass relative to the reduced chain	What it means
Reduced, free thiols	Baseline	No bridge; cysteines carry −SH
One correct disulfide	−2.016 Da	Two cysteines paired as intended
Scrambled disulfide	−2.016 Da per bond	Same mass, wrong pairing
Two disulfides, one correct and one scrambled	−4.032 Da	Mass agrees; connectivity does not
Mixed disulfide with glutathione or cysteine	Varies with the adduct	Incomplete deprotection or oxidation artefact

This is the specific case of the general rule in mass spectrometry identity checks: a matching mass is consistent with a structure, not proof of one. For a linear peptide the sequence question is answered by fragmentation; for a disulfide the connectivity question needs a different experiment.

How the bond is made

In solid-phase synthesis the cysteines are usually protected with an acid-labile group and the chain is cleaved with the thiols still blocked. Folding — deprotection and oxidation — is a separate step in solution. The common routes:

Air or oxygen oxidation. Slow, cheap, and sensitive to pH, temperature and trace metals. It gives the thermodynamically favoured arrangement, which is not always the desired one.
Glutathione redox buffers. Oxidised and reduced glutathione in a defined ratio set the redox potential; the mixture shuffles disulfides until the stable set dominates.
Iodine or DMSO oxidation. Faster, less discriminating, and more prone to over-oxidation of methionine or tryptophan.
Directed or orthogonal protection. Cysteines are unmasked in a chosen order and oxidised pairwise. This is the reliable route for a specific connectivity, and it costs more.

Two practical consequences for a buyer: a peptide with more than two cysteines cannot be assumed to have a single defined connectivity unless the supplier states that directed oxidation was used, and “HPLC purity 98%” says nothing about which arrangement those molecules have.

Scrambling, and why purification does not fix it

At neutral to alkaline pH, thiolate attacks an existing disulfide and exchanges partners. The result is a mixture of isomers that can equilibrate. Purification removes impurities that differ in hydrophobicity; it does not separate isomers that happen to co-elute, and it does not stop the exchange from continuing in the bottle.

Conditions that promote scrambling:

pH above roughly 8 during handling or storage of a reduced or partially folded intermediate.
Residual reducing agents, or traces of thiol-containing buffers, left in the preparation.
Prolonged exposure to oxygen during lyophilisation and storage, which allows slow re-oxidation in the solid state.
Heat and high concentration during folding, both of which accelerate exchange.

The mitigation is process-side: directed oxidation, a defined redox buffer, a quench step, and a final pH that keeps the product away from the thiolate window. None of that is visible on a certificate unless the supplier chooses to describe it.

What a certificate can and cannot show

A typical COA reports appearance, identity by MS, purity by HPLC and water content. For a disulfide-containing peptide that panel is not sufficient to establish connectivity. The gap is worth stating explicitly, because it is often filled with an assumption:

Question	Answered by a standard COA?	What would answer it
Is the mass right?	Usually yes	MS with observed and expected mass
Is the main peak a single species?	Partly — by HPLC	Chromatogram and method, plus a second selectivity
Are the cysteines paired as intended?	No	Non-reducing digestion with LC-MS/MS, or partial reduction and alkylation
How much free thiol remains?	No	Ellman's assay or alkylation with mass detection
Are there scrambled isomers?	No, unless a method was built for them	Non-reducing HPLC, ion exchange, or digestion mapping

Mapping connectivity: the alkylation experiment

The standard approach is to freeze the arrangement, cut the chain into pieces, and see which pieces travel together.

Block the free thiols. Treat the sample with an alkylating agent such as iodoacetamide or N-ethylmaleimide so that any unpaired cysteine is capped and cannot re-pair. This step also quantifies free thiol if the reagent is used quantitatively.
Digest. Use a protease that cleaves at predictable sites — trypsin at Lys and Arg, or a nonspecific enzyme if the sequence lacks them. Keep the disulfides intact: no reducing agent in the digest buffer.
Separate and analyse. LC-MS/MS of the digest. A disulfide-linked pair of peptides appears as one precursor whose mass equals the sum of the two peptides minus 2.016 Da per bond.
Assign. Fragment the disulfide-linked precursor. The b/y ladder is interrupted at the linked cysteines, and the fragments that remain connected identify the partner.
Compare with a reduced control. Run the same sample with dithiothreitol or TCEP. The linked precursor disappears and the two individual peptides appear; that is the confirmation that the linkage was a disulfide and not an artefact.

This is a more demanding experiment than a mass check, and it is not usually included in a catalogue-grade certificate. It is reasonable to request it for a first lot, for a regulated application, or when a peptide behaves inconsistently and connectivity is a plausible cause. What MS/MS can and cannot resolve in general is covered in sequence confirmation by MS/MS.

Non-reducing versus reducing HPLC

A non-reducing HPLC run keeps disulfides intact; a reducing run adds DTT or TCEP and collapses them. Comparing the two chromatograms is a cheap diagnostic:

A single main peak in both runs suggests a homogeneous, fully oxidised species — but it does not prove the correct connectivity.
A main peak that shifts or splits on reduction indicates that the reduced form differs in hydrophobicity, which is expected; it does not by itself indicate an impurity.
A shoulder or second peak that disappears on reduction is evidence of a disulfide-linked variant or aggregate, and worth investigating.
A peak that appears only after reduction can be a thiol-containing impurity that was co-eluting in the oxidised form.

Ask for both traces when connectivity is critical. A supplier who runs only the non-reducing method has not looked at the question.

Free thiols and Ellman's assay

Ellman's reagent (DTNB) reacts with free thiols to release a yellow anion whose absorbance is measured at 412 nm. It is simple, cheap and quantitative, and it answers a different question from mapping: how much of the material is still reduced.

The assay has caveats worth knowing. It responds to any free thiol in the sample, including residual reducing agent and cysteine from the medium. It needs a buffer at a pH where the thiol is deprotonated, and the extinction coefficient depends on the matrix. And it reports an average over the population — a sample that is 90% correctly oxidised and 10% fully reduced can give the same free-thiol number as one with a different distribution. Use it as a consistency check across lots, not as a connectivity assignment.

Handling material that contains disulfides

Store lyophilized at low temperature and low humidity; slow oxidation continues in the solid state.
Avoid alkaline buffers for prolonged periods, especially once the peptide is in solution.
Do not add reducing agents unless the experiment requires the reduced form; if it does, use them deliberately and record it.
Keep solutions concentrated enough to work with but not so concentrated that aggregation and exchange compete.
Expect the reduced and oxidised forms to behave differently in solubility and in assays; a Method section that does not state which form was used is incomplete.

General storage variables are covered in storage and handling, and the chemical routes by which disulfides and other bonds fail are in degradation pathways.

Questions for an RFQ

How many cysteines does the sequence contain, and how many disulfides are intended?
Is the oxidation directed or random? If directed, what is the protecting-group scheme?
Does the certificate report connectivity, or only mass and HPLC purity?
Can you provide a non-reducing and a reducing chromatogram for the lot?
Is a digestion-based mapping report available for the first lot, and at what cost?
What is the free-thiol specification, and how is it measured?

For a peptide with a single disulfide and no other cysteines, the risk of a wrong connectivity is low and these questions are mostly reassurance. For a peptide with four or six cysteines, they are the difference between a defined molecule and a mixture with a good mass spectrum.

For laboratory research use only. This article describes analytical and manufacturing practice; it makes no claim about any specific product, batch or certificate, and does not support administration to humans or animals.

Editor's take

Disulfide formation is chemistry that fails quietly: the mass can be right and the folding wrong. For cysteine-containing peptides, oxidation state belongs in the specification, not the footnotes.

Limitations

This describes process concepts, not a validated manufacturing process. Scale-up and validation require site-specific data.