Selective inhibition of meningeal nociceptors by botulinum neurotoxin type A: therapeutic implications for migraine and other pains

Abstract

Background: Meningeal and other trigeminal nociceptors are thought to play important roles in the initiation of migraine headache. Currently, the only approved peripherally administered chronic migraine prophylactic drug is onabotulinumtoxinA. The purpose of this study was to determine how botulinum neurotoxin type A (BoNT-A) affects naïve and sensitized meningeal nociceptors.

Material and methods: Using electrophysiological techniques, we identified 43 C- and 36 Aδ-meningeal nociceptors, and measured their spontaneous and evoked firing before and after BoNT-A administration to intracranial dura and extracranial suture-receptive fields.

Results: As a rule, BoNT-A inhibited C- but not Aδ-meningeal nociceptors. When applied to nonsensitized C-units, BoNT-A inhibited responses to mechanical stimulation of the dura with suprathreshold forces. When applied to sensitized units, BoNT-A reversed mechanical hypersensitivity. When applied before sensitization, BoNT-A prevented development of mechanical hypersensitivity. When applied extracranially to suture branches of intracranial meningeal nociceptors, BoNT-A inhibited the mechanical responsiveness of the suture branch but not dural axon. In contrast, BoNT-A did not inhibit C-unit responses to mechanical stimulation of the dura with threshold forces, or their spontaneous activity.

Discussion: The study provides evidence for the ability of BoNT-A to inhibit mechanical nociception in peripheral trigeminovascular neurons. These findings suggest that BoNT-A interferes with neuronal surface expression of high-threshold mechanosensitive ion channels linked preferentially to mechanical pain by preventing their fusion into the nerve terminal membrane.

Keywords: Pain; TRPA1; TRPV1; headache; high-threshold mechanosensitive ion channels; nociception.

Figure 1.

Peripheral innervation of intracranial and extracranial structures relevant to migraine. (a) Schematic illustration of peripheral nerves that carry sensory/nociceptive information from pericranial muscles, fascia and periosteum through the greater occipital nerve (blue) and from the intracranial dura and pia and extracranial periosteum through meningeal nociceptors that travel along intracranial nerves (red, such as the tentorial nerve). (b) Possible scenario of extracranial origin of intracranial pain. In this scenario, action potentials generated at extracranial collaterals of meningeal pain fibers (1) spread antidromically to collaterals that terminate inside the cranium, resulting in local release of proinflammatory neuropeptides and activation of neighboring meningeal nociceptors (2). (c) Possible scenario of intracranial origin of extracranial pain. In this scenario, action potentials generated at intracranial meningeal pain fibers (1) spread antidromically to collaterals that terminate outside the cranium (2), resulting in local release of proinflammatory neuropeptides in the scalp and activation of neighboring somatic nociceptors (3). Asterisk marks original site of activation. Red dots represent local release of inflammatory neuropeptides. Red cylinder depicts a blood vessel. SpV: spinal trigeminal nucleus; SSS: superior sagittal sinus; TG: trigeminal ganglion. Source: Panels (b) and (c) adapted with permission from Kosaras B, Jakubowski M, Kainz V, et al. Sensory innervation of the calvarial bones of the mouse. J Comp Neurol 2009; 515: 331–348 (4).

Figure 2.

BoNT-A does not reduce the spontaneous activity of naïve meningeal nociceptors. (a) Peristimulus-time histograms showing firing rate of an Aδ unit before (green) and one, two and three hours after (blue) BoNT-A administration to the dural receptive field. (b) Peristimulus-time histograms showing firing rate of a C-unit before and one, two and three hours after BoNT-A administration to the dural receptive field. (c) Mean firing rate of 17 Aδ-units. (d) Mean firing rate of 21 C-units. Numbers in parentheses depict mean spikes per second for the illustrated 900 seconds. Bars represent SEM. BoNT-A: botulinum neurotoxin type A.

Figure 3.

BoNT-A suppresses mechanical nociception in naïve C- but not Aδ-units. (a) Peristimulus-time histograms showing individual Aδ unit responses to mechanical dural indentation with threshold and suprathreshold forces before (green) and one, two and three hours after (blue) BoNT-A administration to the dural receptive field. (b) Peristimulus-time histograms showing individual C-unit responses to mechanical dural indentation with threshold and suprathreshold forces before and one, two and three hours after BoNT-A administration to the dural receptive field. Black traces on top show timing and force of each mechanical stimulus. (c) Mean response to threshold (open circles) and suprathreshold (filled circles) mechanical stimuli of 17 Aδ-units. (d) Mean response to threshold (open circles) and suprathreshold (filled circles) mechanical stimuli of 21 C-units. Note that BoNT-A suppresses responses only in C-units and only to dural stimulation with suprathreshold mechanical forces. Numbers in parentheses indicate mean spikes per sec for the duration of the stimulus. BoNT-A: botulinum neurotoxin type A.

Figure 4.

Selective suppression of naïve C-unit meningeal nociceptors by BoNT-A. (a) C-units exhibiting reduced response magnitude to suprathreshold mechanical stimulation of the dura following BoNT-A administration. (b) C-units exhibiting no reduction in response magnitude to suprathreshold stimulation of the dura following administration of BoNT-A. BoNT-A: botulinum neurotoxin type A.

Figure 5.

IS-mediated increased spontaneous activity is partially decreased (though not significantly) by delayed administration of BoNT-A. (a) Peristimulus-time histograms showing firing rate of an Aδ unit before (green) and after (red) sensitization, and two hours after BoNT-A administration (blue) to the dural receptive field. (b) Peristimulus-time histograms showing firing rate of a C-unit before (green) and after (red) sensitization, and two hours after BoNT-A administration (blue) to the dural receptive field. (c) Mean spontaneous firing rate of six Aδ-units. (d) Mean spontaneous firing rate of six C-units. Numbers in parentheses depict mean spikes per second for the illustrated 800 seconds. Bars represent SEM. IS: inflammatory soup; BoNT-A: botulinum neurotoxin type A.

Figure 6.

BoNT-A reverses mechanical nociception in sensitized C- but not Aδ-units. (a) Peristimulus-time histograms showing individual Aδ unit responses to mechanical dural indentation with threshold and suprathreshold forces, before (green) and after (red) sensitization, and three hours after BoNT-A administration (blue) to the dural receptive field. (b) Peristimulus-time histograms showing individual C unit responses to mechanical dural indentation with threshold and suprathreshold forces, before (green) and after (red) sensitization, and three hours after BoNT-A administration (blue) to the dural receptive field. (c) Mean response to threshold (open circles) and suprathreshold (filled circles) mechanical stimuli of six Aδ-units. (d) Mean response to threshold (open circles) and suprathreshold (filled circles) mechanical stimuli of six C-units. Black traces on top ((a), (b)) show timing and force of each mechanical stimulus. Numbers in parentheses indicate mean spikes per second for the duration of the stimulus. Note that BoNT-A significantly suppressed mechanical sensitization in C- but not Aδ-nociceptors. BoNT-A: botulinum neurotoxin type A.

Figure 7.

When given early, BoNT-A prevents IS-induced increase in spontaneous activity of C- but not Aδ-meningeal nociceptors. (a) Peristimulus-time histograms showing firing rate of an Aδ unit before (green) and three hours after (blue) BoNT-A administration, and 30 minutes after IS. (b) Peristimulus-time histograms showing firing rate of C unit before (green) and three hours after (blue) BoNT-A administration, and 30 minutes after IS. (c) Mean firing rate of 10 Aδ units. (d) Mean firing rate of 13 C units. BoNT-A: botulinum neurotoxin type A; IS: inflammatory soup.

Figure 8.

When given early, BoNT-A prevents IS-induced increase in C-unit responses to suprathreshold mechanical stimulation. (a) Peristimulus-time histograms showing individual Aδ-meningeal nociceptor responses to threshold and suprathreshold stimuli of the dura before (green) and three hours after (blue) BoNT-A administration, and 30 minutes after IS. Note that BoNT-A did not prevent the IS-induced increase in response magnitude. (b) Peristimulus-time histograms showing individual C meningeal nociceptor responses to threshold and suprathreshold stimuli of the dura before (green) and three hours after (blue) BoNT-A administration, and 30 minutes after IS. Note that BoNT-A prevented the IS-induced increase in response magnitude to suprathreshold, but not threshold mechanical stimulation. (c) Mean responses of 10 Aδ units to threshold (open circles) and suprathreshold (filled circles) mechanical stimulation of the dura. (d) Mean responses of 13 C units threshold (open circles) and suprathreshold (filled circles) mechanical stimulation of the dura. Black traces on top ((a), (b)) show timing and force of each mechanical stimulus. Numbers in parentheses indicate mean spikes per sec for the duration of the stimulus. IS: inflammatory soup; BoNT-A: botulinum neurotoxin type A.

Figure 9.

Selective inhibition of mechanosensitivity of suture branches of C- but not Aδ-units. (a) Experimental setup. Initial identification of a meningeal nociceptor was conducted by stimulating the dura electrically through two small holes in the skull. Identification of a suture branch was performed by applying mechanical pressure (using VFH) along the superior sagittal and transverse sutures. Neuronal spontaneous activity and mechanosensitivity were determined before and after application of BoNT-A to the extracranial suture. To determine mechanosensitivity of the suture branch, suprathreshold mechanical stimuli were applied to the most sensitive site along the suture (defined as the site from which the smallest force induced the largest number of spikes). To determine mechanosensitivity of the dural parent axon, suprathreshold forces were applied to the bone overlying the dural receptive field. (a) Mean spontaneous firing rate of six Aδ units before (green circles) and up to four hours after (blue circles) administration of BoNT-A to the suture. (c) Mean spontaneous firing rate of 10 C-units before (green circles) and up to four hours after (blue circles) administration of BoNT-A to the suture. Note that suture application of BoNT-A did not reduce the spontaneous firing of the nociceptors in the four-hour period in which it was measured. (d), (e) Peristimulus-time histograms showing C-unit responses to mechanical stimulation of the suture (d) and bone overlying the dural receptive field of the unit. (f), (g) Peristimulus-time histograms showing loss of mechanosensitivity in the suture branch of the meningeal nociceptor (f) but not in the parent axon in the dura (g). Trace in (e) (upper right) identifies the studied neuron as a C-type meningeal nociceptor (conduction velocity = 0.95 m/sec). VFH: von Frey hair; BoNT-A: botulinum neurotoxin type A.

Figure 10.

Synaptic vesicle membrane delivery of ion channels and receptors. Cell surface proteins are delivered to the plasma membrane by either a constitutive or regulated synaptic vesicle (SV) pathway whereby proteins channels, receptors and transporters associated with the SV lipid bilayer are inserted into the nerve terminal. SVs form a reserve pool at the nerve terminal and may be filled with neurotransmitter(s). Most SVs are decorated with multiple proteins (70); delivery of membrane-associated protein receptors TRPV1 and TRPA1 are depicted. SVs dock adjacent to the nerve terminal inner membrane “active zone,” and undergo an adenosine triphosphate (ATP)-dependent “priming” step that enables responding to the Ca⁺⁺ signal that triggers fusion, exocytosis and consequent delivery of not only SV contents into the extracellular space but also lipid membrane and associated proteins into the cell surface. Successful fusion requires an interaction between the vesicle-associated membrane protein (VAMP)/synaptobrevin with those on the internal membrane surface, namely synaptosomal-associated protein of molecular weight 25 kDa (SNAP-25) and syntaxin, which together form the soluble NSF (N-ethylmaleimide-sensitive factor) attachment protein receptor (SNARE) complex; other associated proteins (e.g. Munc 18, Rab) are also involved (not depicted) (85). The SV membrane may fully fuse into the terminal membrane (“full collapse fusion”), thus delivering the protein receptors, e.g. TRPV1 or TRPA1, into the cell surface. Excess terminal membrane is recycled through one of the endocytosis pathways (86). BoNT-A cleaves SNAP-25, impairing SV fusion and the regulated delivery of receptors TRPV1 or TRPA1 to the terminal membrane, thus down-regulating receptor activity. BoNT-A: botulinum neurotoxin type A; TRPV1: transient receptor potential cation channel vanilloid subfamily, member 1; TRPA1: transient receptor potential cation channel ankyrin subfamily, member 1.