Preclinical studies investigating the neural mechanisms involved in the co-morbidity of migraine and temporomandibular disorders: the role of CGRP

Abstract

Background and purpose: Temporomandibular disorders (TMD) and migraine can be co-morbid. This can be a significant factor in exacerbating and increasing the prevalence of migraine-like symptoms. However, the underlying mechanisms involved are unknown. Our objective was to investigate these neural mechanisms and the role of CGRP as a key modulator in this co-morbidity.

Experimental approach: We combined experimental approaches using CGRP, which triggers a migraine-like response in patients, with that of masseteric muscle injection of complete Freund's adjuvant (CFA), to model myofascial TMD-like inflammation. Using validated electrophysiological methods to assess each of the above approaches independently or in combination, we examined their effects on the response properties of migraine-like dural-trigeminocervical neurons.

Key results: Independently, in ~2/3 of animals (rats) each approach caused delayed migraine-like activation and sensitisation of dural-trigeminocervical neurons. The response to masseteric-CFA was attenuated by a selective CGRP receptor antagonist. The combination approach caused a migraine-like neuronal response in all animals tested, with somatosensory-evoked cranial hypersensitivity significantly exacerbated.

Conclusion and implications: The data demonstrate a neuronal phenotype that translates to the exacerbated clinical co-morbid phenotype, supporting this combination approach as a relevant model to study the mechanisms involved. It provides a pathophysiological rationale for this exacerbated phenotype, strongly implicating the involvement of CGRP. The results provide support for targeting the CGRP pathway as a novel monotherapy approach for treating this co-morbid condition. This has key implications into our understanding of this co-morbid condition, as well as potentially addressing the major unmet need for novel and effective therapeutic approaches.

Keywords: CGRP; central sensitisation; co-morbidity; hypersensitivity; migraine; temporomandibular disorders; trigeminovascular.

Conflict of interest statement

S.A. reports personal fees from Amgen, Allergan, Novartis and GSK, and personal fees from Patent/Legal work in headache and orofacial pain, unrelated to this work. M.R.R. reports personal fees from Amgen, Allergan, Novartis, and GSK, and personal fees from Patent/Legal work in headache and orofacial pain unrelated to this work.

© 2020 The British Pharmacological Society.

Figures

FIGURE 1

Experimental set‐up for electrophysiological studies. (a) Neurons of the trigeminocervical complex (TCC) were recorded in response to electrical stimulation of the trigeminal innervation of the dural meninges, and innocuous and noxious stimulation of the cutaneous facial receptive field (shaded area). (b) Somatotopic representation of the trigeminal territories for receptive field characterisation and an example receptive field region (shaded area); V1, ophthalmic; V2, maxillary; V3, mandibular. (c) Area of masseteric musculature for CFA injection, in the V3 region; MM, medial masseter; LM, lateral masseter. (d) Original tracing of a single sweep (stimulus) of a reproducible, dural‐evoked neuronal cluster classified as receiving Aδ‐fibre input (

FIGURE 2

CGRP mediates delayed activation of central trigeminocervical neurons and neuronal hypersensitivity to cranial somatosensory stimulation. Time course changes in (a) spontaneous trigeminocervical neuronal firing (action potentials per second [Hz]), (b) intracranial dural‐evoked “fast” neuronal responses (3–20 ms or 3–30 ms range) and (c) unitary discharges within only the “slow” C‐fibre latency range (20–80 ms) in response to CGRP (300 ng·kg−1·min−1 for 20 min, i.v., n = 15) infusion. The data are expressed as mean ± SEM. Further, the data are grouped into all data (grouped), and those considered “responders” or “non‐responders” to CGRP. (d) Representative peristimulus time histograms from a single animal demonstrating ongoing trigeminocervical neuronal firing before CGRP, and at 90 min and 3 h post infusion, in a “responder” animal. The numbers indicate the mean firing (spikes per second [Hz]) over the displayed time period, green neuronal firing indicates baseline and no change in responses, red neuronal firing indicates a significant increase in neuronal firing. Overall, both the grouped and “responder” data showed a delayed increase in ongoing trigeminal neuronal firing and hypersensitivity of “fast” neuronal responses to dural‐stimulation after CGRP (*P < 0.05 compared to baseline). In the “non‐responders” there was no change and there was no effect of any group for “slow” neuronal responses. Examples of cutaneous receptive fields in rats that were either (e) non‐responders (no expansion) or (f) responders (expanded) to CGRP. Dark green represents the original receptive field, and light green is the expanded receptive field (including darker green region). Response magnitude to (g) innocuous and (h) noxious cutaneous stimulation of facial receptive field after CGRP. Only “grouped” and “responders” showed a delayed hypersensitive neuronal response to cutaneous stimulation after CGRP. *P < 0.05 represents a statistical significance compared to baseline

FIGURE 3

Masseteric muscle‐CFA mediates reversible activation of central trigeminocervical neurons and neuronal hypersensitivity cranial somatosensory stimulation. Time course changes in (a) spontaneous trigeminocervical neuronal firing (action potentials per second [Hz]), (b) intracranial dural‐evoked “fast” neuronal responses (3–20 ms or 3–30 ms range) and (c) unitary discharges within only the “slow” C‐fibre latency range (20–80 ms) in response to masseteric muscle injection of complete Freund's adjuvant (CFA, 50 μl, i.m. n = 12). The data are expressed as mean ± SEM. Further, the data are grouped into all data (grouped) and those considered “responders” or “non‐responders” to CFA. (d) Representative peristimulus time histograms from a single animal demonstrating ongoing trigeminocervical neuronal firing before CFA, and at 90 min and 3 h post infusion, in a “responder” animal. The numbers indicate the mean firing (spikes per second [Hz]) over the displayed time period, green neuronal firing indicates baseline and no change in responses, red neuronal firing indicates a significant increase in neuronal firing. Overall, both the grouped and “responder” data showed a reversible increase in ongoing trigeminal neuronal firing and hypersensitivity of “fast” and “slow” neuronal responses to dural‐stimulation after CFA (*P < 0.05 compared to baseline). In the “non‐responders,” there was no change in ongoing firing of “slow” neuronal responses, but they exhibited hypersensitivity of “fast” neuronal responses to CFA. Examples of cutaneous receptive fields in rats that were either (e) non‐responders (no expansion) or (f) responders (expanded) to CFA. Dark green represents the original receptive field and light green is the expanded receptive field (including darker green region). Response magnitude to (g) innocuous and (h) noxious cutaneous stimulation of facial receptive field after CGRP. Only “grouped” and “responders” showed a delayed hypersensitive neuronal response to cutaneous stimulation after CFA. *P < 0.05 statistically significance compared to baseline

FIGURE 4

A selective CGRP receptor antagonist prevents the effects of masseteric muscle‐CFA on dural‐responsive trigeminocervical neurons. Time course changes in (a) spontaneous trigeminocervical neuronal firing (action potentials per second [Hz]), (b) intracranial dural‐evoked “fast” neuronal responses (3–20 ms or 3–30 ms range) and (c) unitary discharges within the “slow” C‐fibre latency range (20–80 ms). The data are expressed as mean ± SEM. (d) Examples of cutaneous facial receptive fields (dark green‐original receptive field) in rats. Response magnitude to (e) innocuous and (f) noxious cutaneous stimulation of the facial receptive field. All data are in response to masseteric muscle injection of complete Freund's adjuvant (CFA, 50 μl, i.m.), or intravenous (i.v.) administration of a selective CGRP receptor antagonist, BIBN4096 (900 μg·kg−1, i.v.), followed by CFA. BIBN4096 (n = 12) prevented the increased activation of trigeminocervical neurons mediated by masseteric‐CFA, and the hypersensitive neuronal responses to cranial somatosensory stimulation. Further, there was no expansion of facial cutaneous receptive fields. *P < 0.05 represents a statistically significant difference compared to baseline

FIGURE 5

Masseteric muscle‐CFA co‐morbid with CGRP infusion exacerbates responses of dural‐responsive trigeminocervical neurons to cranial somatosensory stimulation. Time course changes in (a) spontaneous trigeminocervical neuronal firing (action potentials per second [Hz]), (b) intracranial dural‐evoked “fast” neuronal responses (3–20 ms or 3–30 ms range) and (c) unitary discharges within the “slow” C‐fibre latency range (20–80 ms). The data are expressed as mean ± SEM. (d) Examples of cutaneous facial receptive fields (dark green‐original receptive field) in rats, and their expansion after treatment (light green region, including darker region). Response magnitude to (e) innocuous and (f) noxious cutaneous stimulation of facial receptive field. All data are in response to either infusion of CGRP (CGRP for 20 min), masseteric muscle injection of complete Freund's adjuvant, or a combination of both. The combination (n = 10) treatment caused activation and hypersensitive neuronal responses in every animal tested, compared to only a subset of animals with single treatment. Further, dural‐evoked responses were significantly exacerbated in the combination group compared to each group alone. Only one animal did not show expansion of a cutaneous facial receptive field. *P < 0.05 represents a statistically significant difference compared to baseline