1. Periphery: neurogenic inflammation, sensitization of nociceptors
For a long time, several theories described migraine as a vascular disorder with headache attributed solely to the dilation and inflammation of the arteries of the meningeal structures. Over the past decades, data acquired in humans and animals have allowed this view to evolve towards more integrated theories, which involve both vascular and neuronal components (Figure 4) (Malhotra 2016). It has become increasingly clear that meningeal afferent activation, neuropeptide release, and neurogenic inflammation together play a central role in the generation of migraine pain. The inhibition of this neurogenic inflammation has been the most studied therapeutic avenue in recent years, marked mainly by the appearance of CGRP monoclonal antibodies (Xu et al. 2019). Neurogenic inflammation is a process characterized by the release of vasoactive neuropeptides, such as CGRP or substance P, from peripheral nociceptive sensory nerve endings.
These peptides lead to an inflammatory cascade leading to vasodilation of cranial arteries, extravasation of plasma proteins and degranulation of mast cells in their tissue (Malhotra 2016). The release of neuropeptides as well as the degranulation of mast cells, plays a probable role in the activation and sensitization of meningeal nociceptors (Ramachandran 2018). Indeed, the release of inflammatory substances (histamine, prostaglandins, bradykinin, etc.) sensitizes nociceptors. This sensitization increases the reactivity of nociceptive neurons in the periphery and decreases their excitation threshold to stimulation of their receptive field (Woolf 2011). When they are activated in this particular context, the progressive sensitization of the peripheral fibers of the trigemino-vascular system will in turn sensitize the second order neurons within the TSC and third order of the thalamus (Burstein et al. 2010). The intensification of the headache, when the patient tilts the head forward, could be the manifestation of a peripheral sensitization, whereas the cephalic and extracephalic allodynia would be the consequence of a central sensitization (Burstein et al. 2000) .
2. Role of the parasympathetic system in headaches
A. The superior salivary nucleus
Activation of the trigeminal nociceptive system stimulates the superior salivary nucleus and causes autonomic signs that can sometimes be observed in migraine. These are mainly cluster headaches, benign paroxysmal hemicrania or SUNCT (short-lasting unilateral neuralgiform headache with conjunctival injection and tearing) (Goadsby 2002; Vandenheede 2002). The salivary nucleus is partly under the control of the hypothalamus via descending pathways from the lateral and paraventricular nuclei, as well as the parabrachial nucleus. These regions are essentially involved in the regulation of sleep, hunger, stress and can contribute to the vegetative symptoms associated with migraine attacks. Neurons that project from the superior salivary nucleus to the vascular level release VIP and CGRP, mediators of the autonomic response during a migraine attack (Goadsby et al. 1985; Goadsby et al. 1990).
B. Nitric oxide (NO) in headaches
Parasympathetic-induced vasodilation has been shown to be the primary cause of NO release from peripheral fiber endings (Burstein et al. 2005). NO capable of triggering an attack when administered to a migraine patient, and then appears to be involved in the entire migraine process (Olesen 2008). NO is an enzyme-catalyzed gas (NOS) with three possible isoforms, called endothelial (eNOS), neuronal (nNOS) and inducible (iNOS) NOS (Moncada et al. 2009). The eNOS of the endothelial cells of the arteries of the dura mater and of the cerebral arteries, and the nNOS present in the neurons penetrate as close as possible to the arterioles and can act as a vasodilator (Moncada et al. 2006).
In the trigeminal ganglion, they colocalize with CGRP in a small proportion. It is interesting to note that nNOS has also been found in the perivascular fibers of the cerebral arteries which come from the sphenopalatine and otic ganglion of the parasympathetic nervous system (Messlinger et al. 2000; Toda et al. 2000). These fibers are said to be nitroxidergic (Toda et al. 2003) and the specific stimulation of these leads to vasodilation of the cerebral and extra-cerebral arteries, whereas the stimulation of the fibers of the trigeminal ganglion causes vasodilation exclusively via the release of CGRP (Edvinsson et al. 1998). The link between CGRP and NO has long been studied, but no clear answer could be given. Both play a role in triggering headaches and their respective antagonists have shown prophylactic effect in migraine (Lassen et al. 1998; Olesen et al. 2004).
In the trigeminal ganglion, CGRP and NO colocalize in many neurons (Hou et al. 2001). Moreover, the expression of the CGRP gene in the trigeminal ganglion is regulated by the presence of NO (Bellamy et al. 2006). As a result of these results, it has been suggested that NO is the consequence of CGRP release. Unfortunately, other studies have not been able to confirm this (Schwenger et al. 2007). The hypothesis therefore remains to be demonstrated, but it is interesting to note that NOS inhibitors are able to oppose the dilation induced by CGRP (Akerman et al. 2002). Thus, NO could play a role in migraine by three mechanisms: (i) its release by the parasympathetic system via the nitroxidergic fibers at the level of the neurovascular junction, (ii) the release of NO by the endothelium and its activation by a ligand (eNOS), (iii) by central nervous system production affecting the trigeminal or the parasympathetic system (Burstein and Jakubowski 2005).
