Key points
- A disulfide bond is a covalent S–S link between two cysteine residues. Forming it is an oxidation and removes two hydrogen atoms, lowering the mass by 2.016 Da per bond.
- Disulfides form readily outside the cell and are actively prevented inside it, where the cytoplasm is kept strongly reducing by glutathione and thioredoxin.
- With 2n cysteines there are (2n − 1)!! possible pairings: 3 for four cysteines, 105 for eight. Getting the right one is the central problem in making disulfide-rich peptides.
- Disulfides rarely change a peptide's function directly. They lock a conformation, which is why reduced peptides usually lose activity.
The chemistry of the bond
Two cysteine thiols react with an oxidant to form a disulfide and release two protons and two electrons:
The reacting species is not the neutral thiol but the thiolate anion (R–S⁻). Cysteine's side-chain thiol has a pKa near 8.2 in a free peptide, so only a small fraction is deprotonated at pH 7, and the fraction rises sharply above pH 8. This single fact explains most practical rules about handling cysteine-containing peptides: oxidation is slow in acid, fast in base, and the pH used for storage or for deliberate folding is chosen accordingly. The pI calculation treats the thiol as one of the ionisable groups for the same reason.
The S–S bond is about 2.05 Å long, considerably longer than a C–C bond, and it can rotate. Its preferred dihedral angle is near ±90°, which gives disulfides their characteristic twisted geometry in protein structures.
Disulfides also exchange. A free thiolate can attack an existing disulfide and displace one of its partners, transferring the bond. Thiol–disulfide exchange is fast, reversible, and is what allows a misfolded peptide to reshuffle its bonds towards the most stable arrangement. It is also why a single free cysteine in a sample can scramble the disulfides of everything around it.
What a disulfide does to the mass
Each disulfide bond removes two hydrogen atoms: −2.01565 Da monoisotopic, −2.0159 Da average. This is how disulfides are counted in practice. Measure the mass, reduce the sample with DTT or TCEP, measure again, and divide the difference by 2.016.
The MW calculator computes the mass of the linear, fully reduced peptide, so the correction has to be applied by hand. Insulin is the standard illustration: its A and B chains are held together by two interchain disulfides, and the A chain carries a third one within itself. The intact hormone is therefore 3 × 2.016 = 6.05 Da lighter than the sum of its two reduced chains.
Free cysteines are often blocked before analysis so that they cannot oxidise or scramble. The standard reagents add a fixed mass:
| Treatment | Added group | Monoisotopic Δ per Cys |
|---|---|---|
| Iodoacetamide | Carbamidomethyl | +57.0215 |
| Iodoacetic acid | Carboxymethyl | +58.0055 |
| N-ethylmaleimide | NEM adduct | +125.0477 |
| Disulfide formation | — | −2.0157 per bond |
Carbamidomethylation with iodoacetamide is the default in proteomics, which is why +57.021 on cysteine is treated as a fixed modification in almost every database search, as described in the guide to peptide mass spectrometry.
Where disulfides occur, and where they do not
Disulfide bonds are largely absent from cytoplasmic proteins and common in secreted and cell-surface ones. The reason is the redox environment. The cytoplasm is kept strongly reducing by glutathione, present at millimolar concentrations and mostly in its reduced form, together with the thioredoxin and glutaredoxin systems. Any disulfide that forms there is rapidly reduced again.
Outside the cell, and inside the endoplasmic reticulum and the bacterial periplasm, conditions are oxidising and disulfides are stable. Cells do not leave their formation to chance: in the ER, protein disulfide isomerase (PDI) both introduces disulfides and, crucially, reshuffles incorrect ones until the native pattern is reached. Bacteria use the Dsb system in the periplasm for the same purpose.
This is why the peptides richest in disulfides are venoms, toxins, antimicrobial peptides and hormones, all of which act outside the cell that made them. Conotoxins from cone snails typically carry two or three disulfides in fewer than 30 residues. Defensins have three. The cyclotides from plants have three arranged in a knotted motif described below.
Connectivity: which cysteine pairs with which
Knowing that a peptide has two disulfides does not say how they are arranged. For a peptide with four cysteines there are three possible pairings, conventionally named:
The number of possibilities grows steeply. For 2n cysteines it is (2n − 1)!!, the product of all odd numbers up to 2n − 1:
| Cysteines | Disulfides | Possible connectivities |
|---|---|---|
| 2 | 1 | 1 |
| 4 | 2 | 3 |
| 6 | 3 | 15 |
| 8 | 4 | 105 |
| 10 | 5 | 945 |
All of these isomers have identical masses, so mass spectrometry of the intact peptide cannot distinguish them. Connectivity is determined by digesting the peptide with a protease under conditions that prevent scrambling, usually acidic pH, and identifying which fragments remain linked. Partial reduction followed by stepwise alkylation with different reagents is another common approach.
Cystine knots
Some arrangements are so useful that evolution has found them repeatedly. In the inhibitor cystine knot motif, two disulfides and the backbone between them form a ring, and a third disulfide threads through it. The result is exceptionally rigid and resistant to heat, proteases and extremes of pH. It appears in spider and cone snail toxins, in plant protease inhibitors, and in the squash family of inhibitors.
Cyclotides go one step further: they combine a cystine knot with a head-to-tail cyclic backbone, giving molecules that survive boiling and digestion. Kalata B1, isolated from a plant used in Central Africa in traditional medicine, was the first of these to be characterised and remains the best studied.
Making disulfide-rich peptides
After solid-phase synthesis, the peptide is obtained fully reduced, and the disulfides have to be formed afterwards. Two strategies are used.
Oxidative folding
The reduced peptide is diluted into a buffer, usually around pH 8, and allowed to oxidise and find its own arrangement. Typical conditions are a peptide concentration of 10–100 µM, which keeps intermolecular reactions rare, and a redox buffer such as reduced and oxidised glutathione in a ratio of about 10:1. The oxidised component introduces disulfides, and the reduced component allows exchange so that incorrect pairings can be undone. The process often takes hours to days and is monitored by HPLC, where each isomer typically elutes at a different time.
This works well when the native fold is clearly the most stable, which is usually the case for natural sequences. It works poorly for designed peptides or fragments with no dominant fold.
Directed, stepwise formation
When the arrangement must be controlled, each cysteine pair carries a different protecting group, removed in sequence so that only one pair is free at a time. Common orthogonal sets in Fmoc chemistry use Trt for the pair to be oxidised first, then Acm, and sometimes a third group such as tBu or Mmt. Each deprotection is followed immediately by oxidation of the newly freed pair. The approach is laborious but gives a single defined isomer, and it is how peptides with three or more disulfides are usually made.
Handling cysteine-containing peptides
Free thiols are among the most reactive groups a peptide can carry. A few precautions prevent most problems:
- Keep solutions slightly acidic unless oxidation is wanted. Around pH 5–6 the thiolate concentration is low and both oxidation and exchange are slow.
- Avoid DMSO as a solvent. It oxidises thiols, which is sometimes exploited deliberately to close disulfides.
- Exclude oxygen and trace metals. Copper and iron ions catalyse thiol oxidation at very low concentrations. Degassed buffers and a chelator such as EDTA help.
- Use TCEP rather than DTT where a reducing agent is needed alongside maleimide chemistry: TCEP is effective at acidic pH, does not carry thiols of its own, and is more stable in solution.
- Watch for dimers. A peak at roughly twice the expected mass, minus 2.016 Da, is an intermolecular disulfide, not a contaminant of a different peptide.
The wider picture of handling, including oxidation of methionine and other degradation routes, is covered in the guide to peptide stability and storage.
Disulfides in peptide drugs
Many long-established peptide drugs depend on a disulfide-closed ring, among them insulin, oxytocin, vasopressin and desmopressin, calcitonin, somatostatin and octreotide, and ziconotide, a synthetic conotoxin. In several of these the ring is what holds the pharmacophore in the right shape.
Because a disulfide can be reduced in the bloodstream or inside cells, medicinal chemists often replace it with a more stable mimic once the structure is known. Common replacements include a lactam bridge between a lysine and an aspartate or glutamate side chain, a thioether, a diselenide, and a carbon–carbon link introduced by ring-closing metathesis. These are discussed further in the guide to cyclic peptides.
Frequently asked questions
What is the difference between cysteine and cystine?
Cysteine is the free amino acid with a thiol side chain. Cystine is the dimer of two cysteines joined by a disulfide bond. The distinction matters when reading older literature and extinction coefficient tables, where "cystine" contributes about 125 M⁻¹cm⁻¹ at 280 nm while free cysteine contributes essentially nothing.
Can a disulfide form between two cysteines that are far apart in the sequence?
Yes. Sequence distance is irrelevant; only three-dimensional proximity matters. Insulin's A and B chains are separate molecules, and cystine knots are built from disulfides that connect distant parts of the chain.
Why do reduced peptides often lose activity?
Because the disulfides were holding a specific conformation. Once they are broken the peptide becomes flexible, and the surface that its receptor recognises is no longer presented correctly. The chemistry of the side chains has not changed; the shape has.
How can I tell an intramolecular disulfide from an intermolecular one?
By the mass. An intramolecular bond gives the monomer mass minus 2.016 Da. An intermolecular bond gives roughly twice the monomer mass minus 2.016 Da. Size-exclusion chromatography or non-reducing gel electrophoresis will also separate them.
References
- Anfinsen CB (1973) Principles that govern the folding of protein chains. Science 181:223–230.
- Hogg PJ (2003) Disulfide bonds as switches for protein function. Trends in Biochemical Sciences 28:210–214.
- Craik DJ, Daly NL, Bond T, Waine C (1999) Plant cyclotides: a unique family of cyclic and knotted proteins. Journal of Molecular Biology 294:1327–1336.
- Moroder L, Besse D, Musiol H-J, Rudolph-Böhner S, Siedler F (1996) Oxidative folding of cystine-rich peptides vs regioselective cysteine pairing strategies. Biopolymers 40:207–234.