Double-Binding Botulinum Molecule with Reduced Muscle Paralysis: Evaluation in In Vitro and In Vivo Models of Migraine

Abstract

With a prevalence of 15%, migraine is the most common neurological disorder and among the most disabling diseases, taking into account years lived with disability. Current oral medications for migraine show variable effects and are frequently associated with intolerable side effects, leading to the dissatisfaction of both patients and doctors. Injectable therapeutics, which include calcitonin gene-related peptide-targeting monoclonal antibodies and botulinum neurotoxin A (BoNT/A), provide a new paradigm for treatment of chronic migraine but are effective only in approximately 50% of subjects. Here, we investigated a novel engineered botulinum molecule with markedly reduced muscle paralyzing properties which could be beneficial for the treatment of migraine. This stapled botulinum molecule with duplicated binding domain-binary toxin-AA (BiTox/AA)-cleaves synaptosomal-associated protein 25 with a similar efficacy to BoNT/A in neurons; however, the paralyzing effect of BiTox/AA was 100 times less when compared to native BoNT/A following muscle injection. The performance of BiTox/AA was evaluated in cellular and animal models of migraine. BiTox/AA inhibited electrical nerve fiber activity in rat meningeal preparations while, in the trigeminovascular model, BiTox/AA raised electrical and mechanical stimulation thresholds in Aδ- and C-fiber nociceptors. In the rat glyceryl trinitrate (GTN) model, BiTox/AA proved effective in inhibiting GTN-induced hyperalgesia in the orofacial formalin test. We conclude that the engineered botulinum molecule provides a useful prototype for designing advanced future therapeutics for an improved efficacy in the treatment of migraine.

Keywords: Migraine; botulinum; glyceryl trinitrate model; multivalent; neuronal delivery; trigeminal; trigeminovascular.

Fig. 1

Synthesis of BiTox/AA. (a) Schematic showing the native BoNT/A and formation of BiTox/AA from three individual fusion proteins. LHn is the botulinum type A protease with its translocation domain, and HcA is the binding domain of BoNT/A. The three linking peptides form an irreversible complex (light blue). (b) Coomassie-stained SDS-PAGE gel showing the formation of BiTox/AA after the 60-min assembly reaction. The lane indicating 0 min demonstrates the initial protein amounts used in the BiTox/AA assembly reaction. BiTox/AA exhibits higher molecular weight compared to the native BoNT/A molecule. Excess amounts of HcA with linkers 1 and 2 in the lane 60 min migrate at their original positions. Molecular weight (MW) standards are shown on the right

Fig. 2

Functional evaluation of BiTox/AA in neuronal cultures. (a, Upper panel) example immunoblot of human SiMa neuroblastoma cells treated with BiTox/AA or BoNT/A for 72 h using an anti-SNAP-25 antibody (n = 3). Note the downward molecular shift of SNAP-25 upon the action of botulinum protease correlating with its increasing doses. Positions of the intact and cleaved SNAP-25 are indicated. (a, Lower panel) example immunoblot of SiMa neuroblastoma cells treated with either 1 nM BiTox/AA or 1 nM BoNT/A for the indicated duration of time using an anti-SNAP-25 antibody (n = 3). (b) Graph showing the change in CMAP amplitude measured by EMG in rats 72 h after injection with varying doses of BiTox/AA and BoNT/A, compared to the baseline amplitude (left panel). Bar chart showing the dose difference in logarithmic scale required to achieve 50% reduction in CMAP values measured 24 h and 72 h post injection (right panel). (c) Example immunoblot showing the specificity of the cSNAP-25 antibody to the cleaved end of SNAP-25, as compared to antibody raised against the whole SNAP-25 protein. (d) Examples of fluorescent micrographs of cultured rat trigeminal neurons treated with either vehicle or BiTox/AA and co-immunostained using the cSNAP-25 antibody and an anti-tubulin antibody. (e) BiTox/AA-cleaved SNAP-25 co-localizes with a subset of CGRP neurons in rat trigeminal neuron culture

Fig. 3

Inhibitory effect of BiTox/AA on spiking activity in rat trigeminal meningeal preparation. (a) Examples of basal and 4-AP–induced spiking activities in the meningeal afferents in control conditions (left) and after 6 h of exposure to BiTox/AA (right). (b) The time course of 1 mM 4-AP action in control condition and after 6 h of exposure to BiTox/AA (n = 6 for both conditions). (c) Bar chart showing the number of spontaneous spikes in a 20-min recording in control conditions (white) and after application of BiTox/AA (gray, mean ± S.E.M., n = 6, p = 0.046, t test). (d) Bar chart showing the number of spikes during the 20-min action of 4-AP in control condition (white) and after exposure to BiTox/AA (gray, n = 6, p = 0.049, t test)

Fig. 4

Analysis of the inhibitory effect of BiTox/AA on electrical and mechanical activation thresholds of rat primary trigeminal neurons in the trigeminovascular model of migraine. (a) Treatment with BiTox/AA significantly increased the electrical stimulation threshold required to induce an action potential recorded in vivo from trigeminal neurons with Aδ-fiber (p < 0.001, Mann–Whitney U test) and C-fiber (p = 0.002, Mann–Whitney U test) latencies, 7 days post treatment, compared to recordings from trigeminal neurons in animals treated with saline. The whisker plots show the medians with variability outside the upper and lower quartiles. Dots indicate outliers. (b) Examples of traces of post-stimulus recordings with subthreshold (0.1 ms, 11 V) and threshold (0.1 ms, 12 V) electrical stimulations of the periorbital region (assessed as the minimum voltage required to induce evoked action potentials). (c) Treatment with BiTox/AA significantly increased the mechanical stimulation threshold required to induce an action potential recorded in vivo from trigeminal neurons with Aδ- and C-fiber latencies, 7 days post treatment, compared to the mechanical threshold recorded from trigeminal neurons in animals treated with saline. (d) Examples of traces of post-stimulus recordings with subthreshold (0.16, 0.4 g) and threshold (0.6 g) von Frey mechanical stimulation of the periorbital region (assessed as the minimum von Frey force required to induce an action potential when applied on the cell’s receptive field)

Fig. 5

Effects of BiTox/AA in the orofacial formalin test in the glyceryl trinitrate (GTN) animal model of migraine. Bar charts showing the effect of BiTox/AA on total time (in seconds) spent on face rubbing in phases I and II (a) and the time course of the face rubbing (b). Face rubbing was evaluated by counting seconds animals spent grooming the injected area with the ipsilateral forepaw or hindpaw in the periods 0–6 min (phase I) and 12–45 min (phase II) after formalin injection. The observation time was divided into 15 blocks of 3 min each (45 min total). In both phase I and phase II, pretreatment with BiTox/AA significantly reduced the GTN-induced increase in face rubbing time compared to pretreatment with vehicle (GTN + vehicle) (*p < 0.05, **p < 0.01, unpaired t test)