Duplication of clostridial binding domains for enhanced macromolecular delivery into neurons

Abstract

Neurological diseases constitute a quarter of global disease burden and are expected to rise worldwide with the ageing of human populations. There is an increasing need to develop new molecular systems which can deliver drugs specifically into neurons, non-dividing cells meant to last a human lifetime. Neuronal drug delivery must rely on agents which can recognise neurons with high specificity and affinity. Here we used a recently introduced 'stapling' system to prepare macromolecules carrying duplicated binding domains from the clostridial family of neurotoxins. We engineered individual parts of clostridial neurotoxins separately and combined them using a strong alpha-helical bundle. We show that combining two identical binding domains of tetanus and botulinum type D neurotoxins, in a sterically defined way by protein stapling, allows enhanced intracellular delivery of molecules into neurons. We also engineered a botulinum neurotoxin type C variant with a duplicated binding domain which increased enzymatic delivery compared to the native type C toxin. We conclude that duplication of the binding parts of tetanus or botulinum neurotoxins will allow production of high avidity agents which could deliver imaging reagents and large therapeutic enzymes into neurons with superior efficiency.

Keywords: Botulinum; Double; Duplicated; Multivalent; Neuronal delivery; Tetanus.

Fig. 1

Duplication of tetanus binding domain results in increased binding to neurons and augmented cleavage of SNAP-25 in neurons. a) Schematic showing structural features of tetanus and botulinum neurotoxins together with chimeric proteins used. The coloured bridge is formed by three linking polypeptides. Red star indicates Cy3 fluorophore chemically attached to one of the linking peptides. b) SDS-PAGE gel showing 1xHcT-Cy3 and 2xHcT-Cy3 after the 60 min assembly reaction. Proteins were visualised by Coomassie staining (upper panel) and fluorescence (lower panel). c) Examples of fluorescent micrographs (left) and their quantification (right) of cultured rat cortical neurons treated with either 1xHcT-Cy3 or 2xHcT-Cy3 (both 10 nM, red). Nuclear staining was done using Hoechst 3342 stain (blue). Bar chart shows mean Cy3 intensity (red, n = 3). d) Coomassie-stained SDS-PAGE gel showing formation of Bitox/T and Bitox/TT and the difference in their apparent molecular weights. Proteins were analysed in non-reducing conditions and thus exhibit full protein content. e) Duplication of tetanus binding domain increases SNAP-25 cleaving activity of tetanus-botulinum type A stapled chimera. Example immunoblot of rat cortical neurons treated with either Bitox/T or Bitox/TT at the indicated concentrations (n = 3). f) Immunoanalysis reveals significant enhancement of the cleavage of SNAP-25 (S25) for Bitox/TT: BiTox/T = 120 pM, BiTox/TT = 4.4 pM (n = 3).

Fig. 2

Duplication of tetanus binding domain enhances neuronal delivery of Cy3 and botulinum enzyme in rodents. a) Examples of fluorescent micrographs (left) and their quantification (right) of cultured mouse motor neurons treated with either 1xHcT-Cy3 or 2xHcT-Cy3 (10 nM, scale bars, 25 μm). Bar chart shows average intensity normalised to tubulin/TUJ1 signal: 1xHcT-Cy3 = 1 ± 0.1, 2xHcT-Cy3 = 2.5 ± 0.5 (n = 5). b) Schematic showing the injection site in mouse hind paw and the spinal cord area where SNAP-25 cleavage was quantified (left panel). Spinal cord micrographs of mice injected intraplantar with either 1xHcT-Cy3 and 2xHcT-Cy3 indicating an increased accumulation of Cy3 label (red) in the ventral horn of the spinal cord in the latter case (middle panel). Bar chart showing 50% increase in the number of fluorescent motor neurons in the most fluorescent section of the spinal cord in the case of 2xHcT-Cy3 (right panel): 1xHcT = 7.67 ± 0.9; 2xHcT = 11 ± 0.6. c) Axonal retrograde trafficking of HcT-containing carriers is not affected by duplication of tetanus binding domain. Quantification of retrograde axonal transport was performed in cultured mouse motor neurons pulse-labelled with either 1xHcT-Cy3 or 2xHcT-Cy3 (both 10 nM) in microfluidic devices. Cy3-positive carriers were tracked and their frequency and average speed were plotted in the graph (23 signalling endosomes, 217 transport steps). d-f) Example micrographs of whole spinal cords (d) and (e) or dorsal horn region (f) from rats injected intraplantar with either Bitox/T or Bitox/TT. Tissues were immunostained with cleaved SNAP-25 antibody (green) (d-f) and the NeuN neuronal marker (red) (d) and (f) (n = 3). g) Schematic showing the injection site in the mouse visual cortex and the contralateral site where SNAP-25 cleavage was quantified (left panel). An immunoblot of contralateral visual cortex of mice injected with Bitox/T or Bitox/TT. Note increased immunosignal of cleaved SNAP-25 (green) in the case of Bitox/TT compared to Bitox/T (middle panel). Bar chart showing four-fold increase in the quantified immunosignals of cleaved SNAP-25 in the contralateral visual cortex of mouse brain (right panel) (n = 8).

Fig. 3

Duplication of botulinum type D binding domain results in augmented delivery of botulinum enzyme into neurons. a) Coomassie-stained SDS-PAGE gel showing formation of Bitox/D and Bitox/DD and the difference in their apparent molecular weights. Proteins were analysed in non-reducing conditions and thus exhibit full protein content. b) Duplication of HcD within novel botulinum type A stapled chimera increases cleavage of SNAP-25 in both rat cortical neurons (upper panel; BiTox/D EC₅₀ = 637 pM, BiTox/DD EC₅₀ = 6.4 pM) and human neuroblastoma cells (n = 3).

Fig. 4

Recombinant duplication of the receptor-binding domain of BoNT/C (BoNT/CC) results in an enhancement of botulinum functions. a) Schematic (left panel) and Coomassie-stained SDS-PAGE gel (right panel) showing the difference between BoNT/C and the novel BoNT/CC. Proteins were analysed in non-reducing conditions and thus exhibit full protein content. b) Immunoblot of SiMa neuroblastoma cells treated with either BoNT/C or BoNT/CC at indicated concentrations reveals a five-fold enhancement of SNAP-25 cleavage by the duplicated molecule (BoNT/C EC₅₀ = 106 pM, BoNT/CC EC₅₀ = 20.4 pM) (n = 3). c) Quantification of cell death in differentiated SiMa neuroblastoma cell culture treated with 2 nM of BoNT/C or BoNT/CC indicates three-fold increase in cytotoxic properties of the latter (BoNT/C = 13.8%, BoNT/CC = 53.2%) (n = 3). d) Immunoblot of SiMa neuroblastoma cells treated with 3 nM either BoNT/C or BoNT/CC for indicated duration of time, with quantification of cleaved SNAP-25 at 24 h shown in the bar chart (lower panel) (n = 3).