In this study, although mechanical thresholds were similar in the EM and CM models, there were notable neuropathological distinctions in the Sp5C and ACC between groups. Neuronal and microglial markers, including c-Fos, NeuN, and Iba1, showed significantly elevated expression in the Sp5C of the EM group and ACC of the CM group, compared with the VEH group. Proinflammatory cytokines, such as IL-1β, IL-6, and TNF-α, showed significantly increased expression in the Sp5C of the EM group and ACC of the CM group compared with the VEH group; there were no significant differences in the expression levels of anti-inflammatory cytokines, such as IL-4 and IL-10, between the groups. The expression levels of the neuropeptide CGRP were elevated in the Sp5C and ACC in both EM and CM models, compared with controls. Levels of VIP expression were higher in the Sp5C of the EM group and ACC of the CM group, whereas PACAP and substance P showed elevated expression levels in the Sp5C in both EM and CM groups compared with the controls (Fig. 7).
Migraine is classified as either EM or CM. It is very rare for patients to present with primary CM; up to 14% of patients with EM are at risk of developing chronic daily headache, especially CM, over 1 year [12–14]. Patients with EM and CM exhibit differences in clinical presentation and responsiveness to traditional therapeutic options [15–20]. There may also be physiological differences between patients with EM and patients with CM [21, 22]. The overall understanding of migraine and the development of new treatments for its management have advanced through major translational research in humans and experimental animals. Despite the development and use of various preclinical models for migraine-related pain, further progress is required to improve patient quality of life. Additionally, despite clinical evidence regarding differences between patients with EM and patients with CM, there have been few studies to determine whether current animal models of migraine actually represent EM and/or CM [23].
Multiple mouse models of migraine have been developed, including in vivo models of migraine-related pain induced by mechanical, electrical, or chemical stimuli, as well as transgenic mice; all models have unique strengths and weaknesses [24]. As one of the most common preclinical models for studying migraine-related pain, we used mice injected with NTG, which is converted to nitric oxide and vasoactive S-nitrosothiols. NTG is well tolerated, has a very short half-life, and can cross the blood–brain barrier with known and acceptable side effects [25]. We primarily focused on two regions, the Sp5C and ACC, to examine differences in neuropathology between EM and CM models. We found distinct neuropathological differences in the Sp5C and ACC between the two models. Our results revealed increased neural activation and microgliosis in the Sp5C of the EM model and ACC of the CM model. We also found that the degree of neuroinflammation was significantly greater in the Sp5C of the EM model and ACC of the CM model.
The pain pathway in migraine involves input from the spinal trigeminal nucleus and rostral structures; the transmitted information is processed and integrated to generate a migraine headache. Peripheral sensory information is initially collected from the trigeminal nerve and relayed to the trigeminal ganglia, which consists of first-order neurons in the trigeminal system. Subsequently, the trigeminocervical complex, encompassing the Sp5C and the dorsal horn of the first cervical segments, functions as the second-order central nervous system relay within the trigeminal system and receives input from the trigeminal ganglia. The thalamus, functioning as the third-order relay within the trigeminal system, receives direct projections from the Sp5C; it modulates the activities of pain-related cortical regions, including the ACC, insular cortex, and primary and secondary somatosensory cortex [26]. The ACC, which receives sensory inputs from the thalamus and subcortical regions and projects sensory output to numerous regions (motor cortex, amygdala, midbrain regions, periaqueductal gray, rostral ventromedial medulla, and spinal dorsal horn) is a component of the endogenous opioid pain control circuit. It participates in the affective interpretation of pain, cognition, emotion, and motivation [27–32]. The neuropathological changes observed in the ACC of the NTG-induced mouse models were consistent with evidence of ACC involvement from clinical studies involving patients with frequent headaches or CM. Structural analyses using high-resolution magnetic resonance imaging showed that ACC volume was significantly reduced in patients with migraine who exhibited more frequent headache attacks [33]. Functional neuroimaging studies in patients with CM demonstrated increased functional connectivity between the ACC and other regions; they also showed changes in regional cerebral blood flow in the ACC [34, 35]. As EM progressed toward chronification or transforms into CM because of recurrent headaches, we observed more pronounced neuropathological changes in the ACC, which is an upstream structure in the trigeminal pain pathway of migraine. Consequently, we speculate that the primary neuropathological changes occur in the Sp5C (the second-order relay of the trigeminal system) in the EM model, and in the ACC (a higher-level pain modulation region within the trigeminal system) in the CM model.
Neuropeptides, including CGRP, VIP, PACAP, and substance P, have key roles in the pathophysiology of migraine. During trigeminovascular activation, CGRP, PACAP, VIP, and substance P are released; they act on vascular smooth muscle cells (to induce vasodilatation) and endothelial cells (to promote nitric oxide release) [36, 37]. Substance P also plays a role in plasma protein extravasation, whereas CGRP and PACAP contribute to peripheral and/or central sensitization, which are the main mechanisms related to migraine pathophysiology [38]. Furthermore, in addition to the nitric oxide-induced pathway, CGRP release may stimulate the production of inflammatory mediators, such as cytokines [39]. Similar to our findings, previous studies indicated higher expression levels of proinflammatory cytokine genes or proteins (e.g., IL-1β, IL-6, and TNF-α) in the peripheral blood and trigeminovascular region of animals with NTG-induced migraine [40–44]. The neuropeptide CGRP is widely recognized as a critical player in migraine pathophysiology, on the basis of numerous clinical and preclinical studies. Considering the significant impacts of CGRP and its receptor on migraine pathophysiology, monoclonal antibodies targeting these proteins have been developed for use in clinical practice. However, current therapies targeting CGRP, such as monoclonal antibodies that bind to CGRP or its receptor, are reportedly effective in only 50–60% of patients with migraine [45–47] Therefore, the relationships of other neuropeptides (e.g., VIP, PACAP, and substance P) with the pathogenesis of migraine must be elucidated [38, 48]. Among these neuropeptides, PACAP is the predominant isoform of PACAP-38 in nervous tissue; it is found in parasympathetic and sensory neurons of the trigeminal nucleus. PACAP modulates pain processing by increasing trigeminal nociceptor excitability through elevated levels of cyclic adenosine monophosphate [49]. In clinical studies, PACAP levels increased during migraine attacks and decreased after administration of sumatriptan, a drug used to treat migraines. Additionally, injection of the neuropeptide PACAP reportedly triggers headaches in patients with migraine; therefore, research focused on PACAP receptor inhibition for the treatment of migraine is currently underway [50–55].
VIP, released from cranial parasympathetic preganglionic and cerebral perivascular nerves, acts as potent vasodilator [48]. Although VIP infusion did not induce migraine attacks, VIP levels were elevated in patients with CM who exhibited increased cranial parasympathetic system activity during migraine attacks, as well as patients with EM and patients with CM during interictal periods. VIP may play a role in the chronification of migraine [56–59]. Parasympathetic activation can sensitize afferent nociceptors; this hypersensitivity, along with repeated stimulation, may play a role in the conversion of EM to CM. VIP involvement in migraine chronification has been suggested [60]. Among the neuropeptides analyzed in the present study, only VIP exhibited significant differences in expression between EM and CM models, with higher expression in the Sp5C of the EM group and ACC of the CM group. CGRP expression was significantly increased in both the Sp5C and ACC, whereas PACAP and substance P were strongly expressed in the Sp5C in both EM and CM models. However, data in the present study were generated in EM and CM animal models, with a focus on investigating differences in neuropathology between the two models. Therefore, it remains unclear whether differences in VIP expression are the cause or effect of migraine chronification; further research is required to elucidate the molecular mechanisms involved in migraine chronification or transformation. Additionally, further preclinical and clinical studies are needed to gain a better understanding of the roles played by various neuropeptides in the pathogenesis of migraine.
To our knowledge, this is the first study to simultaneously establish animal models representing both EM and CM, then analyze differences in neuropathology associated with their underlying mechanisms. However, this study had some limitations. We primarily examined the results of behavioral tests and brain regions associated with pain processing. Migraine is characterized by recurrent attacks of pain and other associated symptoms. Previous preclinical and clinical studies focused on other components and related brain regions in patients with migraine and in migraine animal models, including regions involved in emotional processing, cognitive components, and memories of pain processing. Further research involving additional analysis, including studies of other regions of the brain and expansion of behavioral tests, is required to evaluate non-pain functions (e.g., cognition). Additionally, this study did not demonstrate pain reduction after the application of migraine drugs in animal models of EM and CM. Therefore, further research is needed to identify changes in EM and CM animal models after the use of various medications indicated for the treatment of migraine.