Conjugation results are usually determined by reagent selection rather than by the chemistry itself: a reagent chosen for the bond it forms most readily may not form the bond the experiment requires, and reacting both ends of a heterobifunctional crosslinker at a single pH suits neither end. This page sets out the three decisions that govern the outcome — which functional groups to target, how far apart to hold them, and how permanent the linkage must be — with the published conditions for each.
In a peptide or protein the realistic handles are lysine ε-amines and the N-terminal α-amine, cysteine thiols, and carboxyl groups on aspartate, glutamate and the C-terminus. Amines are abundant, which makes them easy to hit and hard to hit selectively. Thiols are usually rare, which makes them the preferred handle whenever site-specificity matters — and the reason a cysteine is so often engineered into a sequence that did not have one.
An NHS ester acylates a deprotonated primary amine to give a stable amide. The competing reaction is hydrolysis, and it is fast enough to dominate if the conditions are wrong: the hydrolysis half-life of NHS ester compounds is 4–5 hours at pH 7.0 and 0 °C, but only about 10 minutes at pH 8.6 and 4 °C1. Raising the pH speeds the wanted reaction and the unwanted one together, which is why a practical window of pH 7.2–8.5 is used, with around 7.5 a common optimum1.
Two consequences follow. First, dilute protein solutions conjugate poorly, because hydrolysis consumes the reagent before it reaches an amine. There is no general concentration rule to apply here: the optimal crosslinker-to-protein molar ratio has to be determined empirically for each pairing1. Second, the buffer must not contain a primary amine, because Tris and glycine compete for the reagent. That competition is precisely why adding Tris or glycine at the end of the procedure is the standard way to stop an NHS reaction1.
Maleimide is valued for the narrowness of its selectivity window. It reacts specifically with sulfhydryl groups when the pH of the reaction mixture is held between 6.5 and 7.5, and conjugations are run in that range for this reason1. Above about pH 8.5 two effects appear together: the reaction begins to favour primary amines, and the maleimide ring hydrolyses to a non-reactive maleamic acid1. Running a maleimide conjugation in a carbonate buffer therefore sacrifices much of the selectivity while hydrolysis consumes the reagent.
Haloacetyl reagents — iodoacetyl and bromoacetyl — alkylate the thiol instead, giving a simple thioether. The optimal range for iodoacetyl is pH 7.5–8.5, and these reactions are run in the dark because free iodine can be generated1. They react more slowly than maleimides, and the bond they form is not reversible: see our page on bromoacetyl peptides for how this chemistry is used to build defined conjugates.
The maleimide–thiol bond is not permanent. The thiosuccinimide formed by Michael addition can undergo a retro-Michael reaction, releasing the maleimide and the thiol again; the freed reagent then exchanges onto other available thiols, typically glutathione or a solvent-exposed cysteine on serum albumin2,3. In plasma the thiosuccinimide undergoes two competing reactions: elimination of the maleimide by the retro-Michael route, which loses the cargo, and hydrolysis of the ring, which gives a derivative resistant to that elimination3.
This matters far beyond antibody–drug work. Any conjugate that will sit in serum, in a cell lysate containing millimolar glutathione, or simply in storage alongside excess thiol can lose its cargo by the same route. The established remedies are to hydrolyse the ring deliberately after conjugation — a controlled step in mild alkaline buffer that adds 18 Da, opens the ring and removes the elimination pathway4 — or to use a maleimide carrying a basic amino group adjacent to the ring, which provides intramolecular catalysis so that the adduct hydrolyses rapidly at neutral pH once the thiol has added3. Ring opening should not be assumed to happen on its own: with the maleimides in common use it is too slow to prevent thiol exchange, but electron-withdrawing N-substituents accelerate it substantially, and the ring-opened products have half-lives of over two years5. A further approach closes a six-membered ring by transcyclization, which locks the thioether in place and resists glutathione exchange6.
Where the conjugate must simply remain intact, a haloacetyl thioether avoids the question entirely: the reaction is an alkylation rather than an addition, so there is no reverse reaction.
Spacer arm length is set by the geometry of the question, and the practical approach is comparative: the optimal length is established by running a panel of similar crosslinkers of differing lengths1. A spacer that is too short will fail to span the distance; one that is too long gives the two partners enough freedom that the linkage no longer reports proximity, which matters when crosslinking is used as a structural measurement rather than as a means of attachment. Success with a short crosslinker is itself informative, since it is a strong indication that the two molecules are interacting1. PEG-containing reagents extend the reach considerably — BS(PEG)5 spans 21.7 Å against 11.4 Å for BS3 — and the sulfonyl group on a sulfo-NHS ester makes the reagent water-soluble1.
Order of operations matters for heterobifunctional reagents, which are designed for sequential, two-step conjugation1. Because the NHS ester is the hydrolytically labile end, it is reacted first with one partner; excess reagent and hydrolysed by-product are then removed by desalting; finally the maleimide-activated intermediate is combined with the thiol-bearing partner at its own optimal pH. Reacting both ends in a single pot at one pH gives a heterogeneous product containing homodimers.
| Target | Chemistry | Working pH | Bond formed | Main caveat |
|---|---|---|---|---|
| Amine | NHS or sulfo-NHS ester | 7.2–8.5 | Amide, stable | Hydrolysis competes; no amine buffers |
| Thiol | Maleimide | 6.5–7.5 | Thioether, reversible | Retro-Michael exchange unless stabilised |
| Thiol | Iodoacetyl / bromoacetyl | ~7.5–8.5 | Thioether, not reversible | Slower; run iodoacetyl in the dark |
| Thiol | Pyridyldithiol | ~7–8 | Disulfide, cleavable | Reduced by any thiol, by design |
| Carboxyl | Carbodiimide, usually with sulfo-NHS | 4.5–7.2; most efficient near 4.5 | Amide, zero length | Intermediate is short-lived; polymerisation if both partners carry both groups |
For conjugating a peptide to a carrier for antibody production, see carrier protein conjugation; for the design of the peptide itself, see epitope mapping and peptide scanning.
| NHS ester | Amines, pH 7.2–8.5; hydrolysis t½ 4–5 h at pH 7, ~10 min at pH 8.6 |
| Maleimide | Thiols, pH 6.5–7.5; selectivity lost above pH 8.5 |
| Reversibility | Maleimide thioether can exchange; haloacetyl thioether does not |
| Spacer | Determine the optimal length with a panel of differing lengths |
| Sequence | NHS end first, desalt, then the thiol end |
LifeTein synthesises peptides with the handle your conjugation needs — an added terminal cysteine, a bromoacetyl or maleimide group, a defined spacer — and carries out carrier protein and other conjugations in house. Tell us what you are joining and how long the bond has to last, and we will recommend the chemistry before anything is made.