Studies of Botulinum Neurotoxins Mechanism of Action

Botulinum neurotoxins (7 serotypes of BoNTs, named from A to G) and tetanus neurotoxin (TeNT) are the most powerful clostridial toxins (CNTs). They are responsible for botulism and tetanus respectively. TeNT and BoNTs bind to peripheral nerve terminals and inhibit neurotransmitter release from presynaptic neuronal cells by proteolytic cleavage of proteins involved in the fusion of synaptic vesicles with the cell membrane. BoNTs act at the level of Peripheral Nervous System (PNS) causing flaccid paralysis whereas TeNT acts at the level of Central Nervous System (CNS) causing spastic paralysis. In particular BoNT A cleaves and disables SNAP25 (synaptosome-associated protein 25), impairing the release of acetylcholine at neuromuscular junction. Structurally, CNTs are composed of two polypeptide chains linked by a single disulphide bond: the 50 kDa Light Chain (LC), which acts in the cytosol as a metalloprotease; and the 100 kDa Heavy chain, which includes a translocation domain (HN) and a receptor binding domain (HC). The three functional domains are structurally distinct and arranged in a linear fashion, such that there is no contact between the LC and HC domain. HC is further composed of two distinct subdomains HCN and HCC. These neurotoxins act at femtomolar concentration and the high affinity binding is due to multiple binding sites, either for membrane ganglioside and neuronal specific membrane proteins. BoNT/A binds to SV2 (synaptic vesicle 2) and to the ganglioside GT1b. It was thought that these two binding sites were located one in HCN subdomain and the other in the HCC subdomain. HCN share some sequence homology with lectins so it was a good candidate to bind ganglioside. Recently by crystallographic analysis it has been shown that both the protein receptor and ganglioside sites of BoNT/B are in the HCC domain. Due to the high homology between all CNTs it is likely that also the SV2 and GT1b sites are in the BoNT A HCC domain. If this is the case the role of N-terminal subdomain of BoNT A is still unknown. The aim of this project is to investigate the role of HCN in the binding of BoNT A to the plasma membrane. It is important to notice that the sequence of this toxin portion is conserved among all CNTs. The sequence of BoNT/A coding for the HCN domain (aa from 855 to 1093) has been cloned as His-Tag fusion protein, and fused to the Enhanced Green Fluorescent Protein (EGFP) and to the monomeric cherry red fluorescent protein (mCherry). By fluorescent microscopy observations we have shown that both the fluorescent chimera were able to bind to the plasma membrane of epithelial and neuronal cells. The fluorescent HCN domain remains at the plasma membrane during incubation times that allow the internalization of whole binding domain, HC. The fluorescent staining is not homogenous on the plasma membrane but is enriched in bright spots. For TeNT binding a role of lipid raft have been establish but for BoNTs the question seems to be still open. Ours data show that the sphingomyelin binding toxin lysenin, colocalized with HCN staining and treatment with sphingomyelinase diminished the HCN binding on epithelial cells. Moreover, in dot blot analysis HCN was able to directly interact with anionic lipid in particular phosphatidylinositol 5 phosphate (PI(5)P). A role for negative charged lipid in the binding of BoNTs and TeNT to lipid bilayer, it was already suggested; our hypothesis is that the N-terminal portion of the binding domain is able to bind anionic lipid in the environment of lipid raft. We suggest that these additional interactions with the membrane surface may play the role of positioning the toxin on the membrane surface ready for membrane insertion.