Botulinum toxin serotypes: what the science says

This page is for readers who want to go a little further than the concepts presented on our page covering botulinum toxin: the basics. It summarises, in accessible terms, what research knows about botulinum toxins: their diversity, their structure, their target in the nerve, how the nerve recovers, spread and the question of antibodies. It relies mainly on a reference scientific review and on the official labels.

One family of toxins, several serotypes

Botulinum toxins are proteins produced by bacteria of the genus Clostridium, including Clostridium botulinum¹. They are traditionally classified into seven serotypes, labelled A to G: each is recognised by different antibodies¹. Within a single serotype there are also subtypes, and genetic sequencing keeps revealing new toxins whose properties are still poorly understood¹.

Not all serotypes are used as medicines. A 2019 review stated that two serotypes, A and B, were then in clinical use⁵. In aesthetic medicine, toxins have been type A since the first approval in France; since 15 July 2026, a short-acting botulinum toxin (serotype E) has also been approved in the European Union for frown lines in adults^2,6. It has been approved in Canada since 22 June 2026⁸.

A molecule in two parts

Despite their differences, all botulinum toxins share a comparable architecture¹. They are produced as a single inactive chain, which is then cut into two parts that remain linked by a chemical bridge¹:

• the heavy chain, which allows the toxin to bind to the nerve ending and then to carry the active part into the cell;
• the light chain, which is the active part: an enzyme that cuts a specific protein in the nerve ending¹.

In some medicines, the toxin comes with so-called "accessory" proteins, which form a large complex with it; other preparations contain the toxin alone^1,5.

The target: SNARE proteins

To release acetylcholine, the messenger that triggers muscle contraction, the nerve ending uses a set of proteins called SNAREs: VAMP (or synaptobrevin), SNAP-25 and syntaxin¹. They allow the vesicles containing acetylcholine to fuse with the membrane.

Each serotype cuts one of these proteins¹:

• types A and E cut SNAP-25;
• types B, D, F and G cut VAMP;
• type C cuts both SNAP-25 and syntaxin.

Once the protein is cut, acetylcholine release is blocked and the muscle contracts less^1,2. The practical steps of this action, from injection to muscle relaxation, are described on our page explaining how the toxin relaxes a muscle, step by step.

The neuromuscular junction, the toxin’s way in

The toxin acts at the neuromuscular junction, the point of contact between the ending of a motor nerve and the muscle. It binds there in two stages: first to particular lipids in the membrane, then to a protein receptor¹. For types A and E, this receptor is a synaptic vesicle protein called SV2, and the two toxins do not recognise the same forms of it¹.

The toxin then enters the nerve ending by taking advantage of the natural recycling of vesicles, and its light chain is released inside the cell^1,2. The labels describe this sequence: rapid binding to specific receptors, entry by endocytosis, then action in the cytosol^2,3.

Long-acting and short-acting: why such different profiles?

Type A and type E toxins cut the same protein, SNAP-25, but not at the same place. The type A toxin removes only nine amino acids, the building blocks of proteins, from the end of SNAP-25, whereas the type E toxin removes a large part of its second half¹.

The most important difference for duration lies in how long the light chain persists in the nerve ending. That of the type A toxin escapes the cell’s breakdown systems and persists for a long time; that of the type E toxin is tagged and then cleared by these systems¹. A study in cells clarified this mechanism: the active part of the type E toxin is labelled with a small protein, ubiquitin, then quickly destroyed by the proteasome, the structure that recycles the cell’s proteins, whereas that of the type A toxin is much more stable⁹. The authors point out that this work was carried out mainly in the laboratory, on cells, with the limits that implies¹.

In practice, this translates into different clinical profiles. With the short-acting toxin, effects begin to show 8 hours after the injection, peak around day 7 and recovery generally occurs within 2–3 weeks². With a type A toxin, clinical signs appear after 2–3 days and recovery takes place within about 12 weeks³. The practical consequences of these durations are detailed on our page about the duration of the effect depending on the toxin.

How does the nerve recover?

The toxin does not destroy the nerve ending. A study in mice described recovery in two phases⁴. First, the blocked nerve ending produces a network of small branches, called sprouts, which reach the muscle and carry transmission on their own for a while. Then activity returns to the original nerve ending, and the sprouts that are no longer needed are eliminated⁴. The reference review confirms this pattern¹.

The labels describe the same reality: for the short-acting toxin, recovery occurs as primary nerve terminals recover²; for a type A toxin, recovery happens when the nerve endings restore their connections³.

Spread of the toxin

The toxin acts mainly where it is injected, but it can reach neighbouring muscles. In therapeutic use, greater dilution, and therefore a larger injected volume, goes with wider spread towards junctions further from the injection site¹. The labels report that effects away from the injection site have been very rarely reported^2,3.

In the phase 2b trial of the short-acting toxin, funded by the manufacturer, no event suggestive of local or distant spread was reported⁷. The sources consulted do not allow the spread of different toxins to be compared.

Antibodies and immunogenicity

Like any foreign protein, botulinum toxin can, rarely, lead the body to produce antibodies that reduce or cancel its effect⁵. A review identifies several contributing factors: closely spaced injections (including booster injections), high doses per session and a larger amount of antigenic protein; the formulation, manufacture and storage of the product also play a role⁵. Some purified preparations, without accessory proteins, may be less immunogenic⁵. The labels point out that doses given too close together or too high may increase this risk³.

For the short-acting toxin, samples from 2,104 patients were analysed in five studies: none developed neutralising antibodies, but data are still limited². No source allows us to say that one family of toxin is less immunogenic than another. Practical questions about "resistance" are covered in our answers on resistance to the toxin.

What science does not yet know

Several questions remain open. At the doses used in clinical practice, the amounts of toxin are so small that what happens to it in the body could not be studied by standard pharmacokinetic methods^1,2. The trials that led to the approval of the short-acting toxin compare it with a placebo, not with a type A toxin². Finally, the new toxins discovered by sequencing have not yet revealed their properties¹. These limits are worth keeping in mind and discussing with your doctor.