Targeting glial-orchestrated neuroinflammation in migraine pathophysiology

Abstract

Background

Migraine is a chronic neurological disorder characterized by severe headache, nausea, and sensitivity to light and sound, affecting approximately 1 billion people globally. Despite advances in understanding migraine pathophysiology, particularly with the emergence of CGRP-targeted therapies, the mechanisms underlying neuroinflammation and glial contributions remain poorly understood. Current treatments are effective for a subset of patients, yet they don’t tackle the fundamental neurogenic and neuroinflammatory processes that fuel chronic migraine, especially the pathophysiological aspects contributed by glial cells.

Principal findings

This review integrates recent preclinical and clinical evidence to elucidate how diverse glial cells, including central glia (astrocytes, microglia and oligodendrocytes) and peripheral glia (Schwann cells, satellite glial cells), coordinate the neuroinflammation associated with migraine. Evidence shows that astrocytes and microglia are essential to both cortical spreading depolarization (CSD) and mediating the inflammatory cascades that maintain chronic pain. Oligodendrocytes, though less studied, are predicted to affect neuronal excitability and energy metabolism, while Schwann cells and satellite glial cells mediate peripheral nociceptive signaling through their interactions with neural and immune elements. New therapeutic strategies have been put forward. These include targeting glial-specific signaling pathways and employing advanced drug delivery systems such as viral vectors and nanoparticles to improve treatment effectiveness.

Conclusion

Glial cells are pivotal regulators of migraine-associated neuroinflammation. This review underscores their critical role in migraine pathophysiology and highlights glial-targeted therapies as a promising direction for future research and treatment development.

Introduction

Migraine is a prevalent, disabling neurological disorder with prominent sensory and neurovascular features that substantially impairs quality of life and societal productivity [1, 2]. According to the Global Burden of Disease Study, migraine affects approximately 1 billion people worldwide [3], with a cumulative lifetime risk of approximately 33% in women and 18% in men [4]. Migraine is characterized by recurrent attacks of moderate-to-severe headache, often accompanied by nausea, photophobia, and phonophobia. These attacks last 4–72 hours, with premonitory symptoms occurring 2–48 hours before the headache and postdromal symptoms persisting for hours to days [5, 6]. These debilitating attacks leave patients physically exhausted, mentally drained, and socially isolated, highlighting the need to better understand migraine pathophysiology towards developing more precise treatments.

Despite significant advances, the mechanisms driving recurrent migraine attacks remain elusive. There is evidence for both peripheral trigeminal afferent activation and central nervous system involvement regarding brainstem and diencephalon dysfunction [7, 8]. Neuroinflammation is also crucial, contributing to triggering and maintaining migraine attacks [9–11]. Advanced neuroimaging techniques have revealed increased activation in brain regions implicated in pain processing and neuroinflammation during migraine episodes [12]. The trigeminovascular (TGV) system, comprising trigeminal afferents and meningeal vasculature, is central to migraine pain generation [7]. Trigeminal-autonomic circuits involving the superior salivatory nucleus and the sphenopalatine ganglion can modulate meningeal blood flow and neuroimmune signaling, potentially shaping headache expression in susceptible individuals [8, 13].

Sensory neuropeptides, including calcitonin gene-related peptide (CGRP), pituitary adenylate cyclase-activating polypeptide (PACAP), and vasoactive intestinal peptide (VIP), alongside nitric oxide, drive intracranial vasodilation and contribute to peripheral and central sensitization within the trigeminal system [9]. Innate immune cells, such as mast cells and dendritic cells, are activated in response to neuropeptide release, amplifying neurogenic inflammation in migraine models [10]. Preclinical studies and emerging clinical evidence implicate activated glial cells as key mediators in these inflammatory pathways, contributing to both the initiation and persistence of migraine attacks.

Glial cells, including astrocytes, microglia, and oligodendrocytes (OLs) in the central nervous system (CNS), as well as Schwann cells and satellite glial cells (SGCs) in the peripheral nervous system (PNS), play essential roles in maintaining neural functioning [14, 15]. Individual types of glial cell contribute specific functions: i) astrocytes are primarily responsible for maintaining CNS homeostasis, ii) oligodendrocytes and Schwann cells form myelin sheaths around axons providing physical and metabolic support, iii) NG2-glia serve as permanent precursors for oligodendrocytes, iv) microglia act as immune sentinels of the CNS, and v) satellite glial cells both support and regulate neurons within the peripheral ganglia [14]. Current evidence indicates that the various glial subtypes actively participate in neuroinflammatory processes and closely interact with the TGV system to modulate peripheral and central sensitization processes that underlie migraine attacks [16]. Astrocytes and microglia modulate cortical spreading depolarization (CSD), a key mechanism linked to migraine aura, and may influence downstream neuroimmune signaling that activates trigeminovascular pathways. In parallel, glia within trigeminal circuits can shape neuropeptide-mediated nociceptor sensitization and inflammatory amplification (e.g., CGRP-related signaling) [9]. Satellite glial cells within trigeminal ganglia are involved in orofacial pain initiation and maintenance, playing crucial roles during migraine headache phases [17]. This neuroimmune interplay highlights the importance of glial cells in the pathophysiology of migraine, particularly in sustaining chronic inflammation and pain.

While conventional migraine therapies focus on symptomatic relief and acute pain management, treatments generally fail to address underlying neuroinflammatory mechanisms, leading to limited efficacy and refractory symptoms in a subset of substantial patients. Given their central role in neuroinflammation, glial cells have emerged as promising therapeutic targets [18–20]. Candidate strategies include modulating microglial neuron-immune crosstalk (e.g., CX3CL1/CX3CR1 signaling) and purinergic/inflammasome-related pathways (e.g., P2×7 receptor signaling), which have shown promise in preclinical migraine- or trigeminal pain–relevant models [21]. This review provides a comprehensive analysis of glial cell roles in modulating migraine neuroinflammation, illuminates mechanisms through which glial activation influences migraine pathophysiology, and explores emerging therapeutic targets to enhance treatment outcomes.

Immune modulatory roles of glial cells in neural homeostasis

Glial cells are the most abundant cell type in the nervous system and play crucial roles in maintaining neural homeostasis through interactions with neurons, blood vessels, and immune cells. Microglia, further act as resident immune cells, defending the CNS from injury and other insults [22–24]. This section focuses on astrocytes, microglia, OLs, Schwann cells, and SGCs in the context of migraine pathogenesis, emphasizing their essential roles in maintaining neural homeostasis and immune modulatory functions (Fig. 1).

Fig. 1.

Schematic representation of the main glial cell types associated with migraine and their functions. The red lines show the ascending pathways of the trigeminovascular system. Illustrations were created via biorender (http://app.biorender.com)

Astrocytes

Astrocytes, the most abundant glial cells in the CNS, maintain homeostasis by interacting with neurons, synapses, vasculature, and the extracellular matrix [25, 26]. Their perivascular endfeet cover more than 90% of the brain’s microvasculature, thereby supporting the blood-brain barrier (BBB) and facilitating neurovascular communication [16, 25]. Notably, astrocytes are highly enriched in migraine-relevant regions including the cortex, brainstem, and trigeminal nucleus caudalis (TNC), where they respond to disturbances including CSD and neurogenic inflammation [17].

Astrocytes in part regulate synaptic activity by clearing glutamate and GABA through transporters including excitatory amino acid transporter 1(EAAT1)/glutamate/aspartate transporter (GLAST), EAAT2/glutamate transporter (GLT-1), and GABA transporters (GATs), to collectively prevent excitotoxicity [25]. Impaired glutamate clearance has been directly linked to neuronal hyperexcitability in migraine [17]. Astrocytes also support neuronal metabolism through the astrocyte–neuron lactate shuttle (ANLS) and provide emergency energy substrates via glycogen mobilization during periods of metabolic stress [25, 27]. They also regulate osmotic balance, K⁺ buffering, and glymphatic waste clearance through aquaporin-4 (AQP4) and Kir4.1 channels [25]. Disruption of these diverse supportive functions—particularly those mediated by AQP4 and Kir4.1—in migraine models enhances CSD susceptibility [16]. Immunologically, astrocytes express pattern recognition receptors, including toll-like receptors (TLRs), enabling them to detect pathogen-associated and damage-associated molecular patterns (PAMPs/DAMPs) [24, 28]. Activated astrocytes release cytokines like interleukin-1β (IL-1β) and tumor necrosis factor-α (TNF-α), and chemokines like C-C motif ligand 2 (CCL2), which recruit leukocytes and activate microglia [28]. This interaction amplifies neuroinflammation, contributing to migraine pathology. Beyond immune functions, astrocytes modulate synaptic plasticity, circadian rhythms, and neurovascular coupling via calcium-dependent gliotransmitter release [25, 29]. Dysregulation of these processes may promote migraine chronification [30]. Finally, astrocytes participate in BBB remodeling by modulating endothelial tight junctions and promoting angiogenesis, processes subject to disruption during migraine-associated neuroinflammation [16].

Microglia

Microglia, the primary innate immune cells of CNS, arise from yolk sac-derived progenitors and survey the CNS with motile processes [31, 32]. Beyond immunological defense, microglia play critical roles in neurodevelopment and synaptic plasticity. For example, during postnatal maturation, they orchestrate synaptic pruning via complement receptor 3 (CR3)–dependent engulfment of C1q-tagged synapses, while also promoting neuronal survival through insulin-like growth factor 1 (IGF1) release [33, 34]. Furthermore, they facilitate myelination by modulating oligodendrocyte precursor cell differentiation [35].

Microglia act as gatekeepers of neuroinflammatory balance by expressing a diverse sensome that detects pathogen-associated molecular patterns and damage signals via receptors like Toll-like receptor 4 (TLR4), CX3CR1, and P2Y12 [31, 36]. Upon activation, they secrete pro-inflammatory cytokines such as IL-1β, IL-6 and TNF-α that amplify immune responses [36, 37]. In migraine pathophysiology, microglia respond to CSD and oxidative stress by releasing cytokines and activating astrocytes, which together amplify nociceptive transmission [38, 39]. While microglia plasticity supports neurovascular integrity, myelination and neuroprotection functions [31], chronic priming leads to sustained inflammatory states that contribute to migraine and potentially prolonged dysfunction linked to chronic inflammation [40].

Oligodendrocytes

OLs and OPCs comprise the OL lineage in mammalian CNS. Mature OLs form myelin sheaths around axons, facilitating saltatory action potential propagation and increasing signal transmission efficiency [41]. OPCs, characterized by NG2 immunoreactivity, represent a fourth glial cell type accounting for 5–10% of neural cells with proliferation, differentiation, and self-renewal capabilities [42–45]. Recent findings reveal direct synaptic interactions between OL lineage cells and neurons, indicating roles in synaptic regulation [46, 47].

OLs contribute to neural circuit formation by secreting factors including brain-derived neurotrophic factor (BDNF) and IGF-1, essential for neuronal survival and synaptic plasticity [48, 49]. Moreover, OLs may play a role in neuroprotection following CNS injury by enhancing remyelination and reducing neuroinflammation [50].

Schwann cells

SCs, the principal PNS glial cells responsible for axonal myelination, are essential for rapid action potential propagation. They wrap peripheral axons, forming myelin sheaths through proteins including myelin protein zero (MPZ) and peripheral myelin protein 22 (PMP22) [51]. Myelination ensures efficient saltatory conduction by concentrating voltage-gated ion channels at Ranvier nodes, a process critical for the high-fidelity transmission of nociceptive signals that are dysregulated in migraine [52]. Furthermore, nonmyelinating Remak Schwann cells provide essential trophic support to small unmyelinated C-fibers; disruption of this support has been implicated in peripheral sensitization and heightened pain perception, key features of migraine pathophysiology [53].

Beyond myelination, SCs regulate the local microenvironment by modulating glucose and lipid metabolism [54] and respond to nerve injury by dedifferentiating into a repair-promoting phenotype, facilitating axonal regrowth through neurotrophic factor secretion including NGF and BDNF [55]. SCs also contribute to PNS immune modulation by interacting with macrophages to clear myelin debris and reduce inflammation following nerve injury [55, 56].

Satellite glial cells

SGCs encase neuronal cell bodies in peripheral ganglia, including sensory, parasympathetic, and sympathetic ganglia [57, 58]. The 20 nm gap between SGCs and neurons resembles synaptic clefts [58]. Nerve damage, inflammation, and surgical interventions stimulate nitric oxide (NO) production, activating SGCs in sensory ganglia. This activation involves GFAP upregulation, increased SGC coupling, Kir4.1 downregulation, and heightened ATP sensitivity [59]. Activated SGCs release pro-inflammatory cytokines including IL-1β, IL-6, TNF-α, and fractalkine, amplifying CNS inflammatory responses and contributing to neuronal excitation and pain sensitization [11, 59].

Under normal conditions, SGCs store GABA that is released during neuroglial activation. However, in inflammatory conditions, neuronal ATP activates P2×7 receptors on SGCs, upregulating P2Y receptors and maintaining inflammatory environments while enhancing calcium signaling [60, 61]. This process propagates inflammation, often accompanied by IL-1β or TNF-α release, further potentiating P2×3 receptors and contributing to pain sensitization [62]. Additionally, signaling molecules including CGRP, NGFs, cytokines, and prostaglandins activate pain receptors through paracrine pathways [61]. CGRP binds to calcitonin receptor-like receptor (CRL)/receptor activity-modifying protein 1 (RAMP1) complex on neurons and SGCs, facilitating neuroglial communication and pain modulation, while glutamatergic signaling via NMDA receptors contributes to pain events [63].

Critical roles of glial cells in migraine pathophysiology

Astrocytes in migraine: neurotransmission and neurovascular regulation

Migraine-related triggers such as CSD or proinflammatory mediators induce reactive astrogliosis, characterized by morphological changes, increased GFAP expression, and proinflammatory cytokine release [17, 30]. In chronic migraine models using inflammatory soup (IS), the occurrence of astrocyte activation is accompanied by increased GFAP expression in the medullary dorsal horn and distinct morphological changes [64]. CSD triggers neuronal pannexin-1(Panx1) channel opening, leading to inflammasome complex assembly and caspase-1 activation. The resulting release of IL-1β and high-mobility group box 1 (HMGB1) activates the proinflammatory nuclear factor kappa B (NF-κB) pathway in astrocytes [65, 66]. Building on this, Karatzidou et al. [67] reported that CSD induces neuronal HMGB1 release via small extracellular vesicles selectively taken up by astrocytes, leading to NF-κB activation and revealing a novel neuro-astrocytic inflammatory mechanism. Additionally, activation of P2×7 receptors has been shown to facilitate ATP-dependent subcortical propagation of CSD and promote neuroinflammation via astrocyte HMGB1/NF-κB signaling [68]. Once activated, reactive astrocytes secrete proinflammatory mediators such as IL-1β, TNF-α, and prostaglandins, which further exacerbate neuroinflammation and contribute to the sensitization of pain pathways [69]. Astrocytes also interact with microglia to amplify inflammatory responses, reinforcing the chronicity of migraine attacks [17]. Zhang et al. [64] have further demonstrated that IS infusion induces sustained cutaneous allodynia accompanied by upregulation of GFAP and ionized calcium-binding adapter molecule 1 (Iba1) in the TNC and dysregulated cytokine expression, suggesting prolonged astrocyte-microglia coactivation contributes to chronic migraine (CM) pathogenesis. Astrocytes regulate the endothelial cells of blood vessels and maintain the tight junctions of the BBB essential for CNS homeostasis [70]. However, the release proinflammatory mediators from reactive astrocytes compromises BBB integrity, increasing vascular permeability and allowing immune cell, cytokine, and neurotoxic molecule infiltration [71]. BBB permeability changes may contribute to migraine pathogenesis; however, findings vary substantially across study modalities, migraine phenotypes (MA vs. MO), and timing (ictal vs. interictal). Inconsistencies in the literature highlight the need for cautious interpretation regarding a uniform role of BBB breakdown in initiation, chronicity, attack frequency, and pain sensitivity [16, 40, 72].

In migraine, astrocytic dysregulation of glutamate metabolism and homeostasis contributes significantly to pathophysiology. Increased neuronal glutamate release and impaired astrocytic reuptake combine to elevate extracellular glutamate levels [73], activating NMDA receptors and intensifying pain signaling that contribute to trigeminovascular sensitization [73, 74]. In chronic migraine, compensatory glutamate transporter upregulation becomes overwhelmed, leading to persistent glutamate toxicity and neuronal hyperexcitability [74]. Other pathogenic factors include that: i) astrocyte energy deficiency impairs Na⁺/K⁺ ATPase activity, causing extracellular K⁺ accumulation and exacerbating neuronal depolarization and glutamate release [75] and ii) CSD-induced astrocyte end-foot swelling obstructs the glymphatic system, impairing neuroinflammation clearance [17, 74]. Of note, upregulation of astrocytic EAAT2/GLT‑1 enhances glutamate clearance in the TNC while reducing CGRP levels, resulting in decreased nociceptive transmission and attenuated mechanical and thermal allodynia [76]. Petrausch et al. [77] demonstrated that conditional GLT-1 knockout mice exhibit increased CSD susceptibility with accelerated glutamate accumulation, highlighting the role GLT-1 in controlling CSD propagation through efficient glutamate clearance. In familial hemiplegic migraine Type-2 (FHM2) models with mutations in the ATP1A2 gene encoding the α2-subunit of Na⁺/K⁺-ATPase, impaired astrocytic glutamate and K⁺ clearance increases CSD susceptibility [78–80]. Crivellaro et al. [81] reported that selective inhibition of extrasynaptic GluN1–N2B NMDARs, which minimally altered excitatory postsynaptic currents (EPSCs) in wild-type mice, normalized both aberrant NMDAR activation and CSD facilitation in FHM2 mice. They identified extrasynaptic GluN1–N2B NMDAR overactivation as linking astrocytic glutamate dysregulation to CSD initiation and propagation. FHM1 models exhibit astrocytic and microglial activation with inherent neuroinflammatory states [66, 82]. In FHM1-knockin mice, reduced Ca²⁺ responses to somatosensory stimulation in both neurons and astrocytes correlate with impaired neurovascular coupling [83]. Consistent with this hyperexcitable milieu, FHM1 mutants also show enhanced glutamatergic transmission and widespread neuroinflammatory responses after CSD induction [66]. Additional genetic evidence links ion/homeostatic dysregulation to migraine susceptibility: mutations in SLC4A4 (encoding the Na⁺/HCO₃⁻ cotransporter NBCe1) and casein kinase Iδ increase vulnerability to hemiplegic migraine and to CSD [69, 84]. Mechanistic work further implicates astrocytes at the interface between cortical events and meningeal nociception. For example, in a rodent migraine-with-aura model, it was shown that cortical astrocytes mediate mechanical sensitization of meningeal afferents via calcium-independent signaling [85]. Mechanistically, astrocytic suppression selectively reduced CSD-evoked mechanical hypersensitivity without altering initial afferent activation, while pharmacological inhibition lowered baseline astrocyte Ca^{2 +} activity with minimal affects on CSD-associated astrocytic Ca^{2 +} waves. These findings suggest that astrocyte-derived mediators help translate cortical depolarization into meningeal pain pathways.

Complementary evidence from cellular and in vivo studies shows that migraine-related triggers disrupt astrocytic homeostasis to lower the threshold for trigeminovascular activation. In co-cultures of astrocytes, microglia, and meningeal cells, application of glyceryl trinitrate (GTN) to disrupt astrocytic homeostasis resulted in altered iron trafficking and elevated MMP-9 and CRLR/CGRPR1 expression [86]. Corroborating these findings, selective chemogenetic activation of Gq-coupled signaling in visual cortex astrocytes was shown to elicit persistent firing and heightened mechanosensitivity via astrocyte-derived CGRP in meningeal afferents, culminating in cephalic allodynia [87]. Similarly, astragaloside IV attenuated chronic migraine–related central sensitization by restoring astrocyte mitochondrial integrity, suppressing reactive oxygen species (ROS), and downregulating NF-κB and IL-1β signaling. These effects collectively dampen neuroinflammation and normalize synaptic plasticity [88]. Further mechanistic insight comes from studies on the transient receptor potential ankyrin 1 (TRPA1) channel, highly expressed in both cortical astrocytes and neurons. TRPA1 deactivation—via either antibody neutralization or pharmacological antagonism—reduces CSD susceptibility and cortical oxidative stress, whereas its activation facilitates CSD, an effect reversible by antioxidants or CGRP blockade [89]. Together, these findings delineate a ROS/TRPA1/CGRP signaling axis within astrocytes and neurons that drives CSD and sensitization [89].

In conclusion, astrocytes act as a double-edged sword concerning migraine pathology. While they provide neuroprotective functions under normal conditions, their dysfunction during CSD and neuroinflammation exacerbates migraine progression. Disrupted astrocyte-mediated BBB integrity, glutamate homeostasis, and neuroinflammatory modulation contribute to migraine chronicity (Fig. 2). Supported by genetic FHM models that illustrate astrocyte roles in CSD susceptibility and neuronal hyperexcitability it is evident that targeting astrocyte dysfunction holds promise for novel migraine therapies.

Fig. 2.

The role of primary central glial cells in neuroinflammation during migraine. (A) During migraine attack, CSD or synaptic stress induces the opening of neuronal panxl channels, the formation of the inflammasome complex, the activation of caspase-1, and the subsequent release of IL-1β, FKN, and HMGB1, which initiate the parenchymal inflammatory pathway. Simultaneously, activated neurons release substantial amounts of Glu and ATP. (B) There is a significant compromise of the BBB, through which activated immune cells extravasate from the periphery into the CNS. Mast cells release serotonin and ATP for pronociception via 5-HT3 and P2×3 receptors. Both astrocytes and microglia are activated into A1 and M1 states. (C) Glu, ATP, IL-1β, HMGB1, and FKN released by neurons in the trigeminocervical complex activate microglia via CX3CR1; purinergic receptors (such as P2Y12R, P2Y14R, P2X4R, and P2X7R); and S1PR1, leading to the upregulation of TREM-1, PAR2, TLR2, and TLR4. The TREM1 receptor, when it binds with NLRP3, contributes to microglial activation. NLRP3 facilitates the activation of caspase-1 and the generation of mature IL-1β. The activation of TLR2 and PAR2 generates proinflammatory mediators such as TNF-α, IL-6, and ROS. IL-18 is a product of TLR4 activation, and it interacts with IL-18 R expressed in astrocytes, resulting in the activation of astrocytes. P2X4R activation in microglia triggers BDNF release, promoting neuronal hyperexcitability. miR-155-5p promotes neuroinflammation and central sensitization by inhibiting SIRT1, leading to microglial activation. (D) Neuron- or microglia-sourced molecules, including ATP, ROS, C3, C1q, TNFα, IL-1β, and IL-18, act on astrocytes and can induce reactive A1 astrocytes via the regulation of P2X7R, TRPA1, IL-18 R, and α7nAChR. The clearance of K⁺ and glutamate can be reduced owing to the dysfunction of Na⁺/K⁺ ATPase and EAAT2 in astrocytes, which increases reactivity to CSD. Illustrations were created via biorender (http://app.biorender.com)

Microglia in migraine: neuroimmunology and pain sensitization

Microglia generally maintain homeostasis and respond to injury and infection, while in migraine, they become activated by triggers such as CSD and inflammatory mediators. CSD is a primary inducer of neuronal and glial activation, triggering the release of excitotoxic signals, potassium ions, and proinflammatory molecules, all of which contribute to neuroinflammation [17]. In models, CSD activates cortical microglia via the K⁺ inward rectifier (Kir) 2.1 [90], inducing microglial migration and motility to affect the surrounding electrical activity and further increase susceptibility to subsequent CSD [91]. Upon activation, microglia shift from a resting, surveillant state to an amoeboid, reactive phenotype characterized by morphological changes and proinflammatory signaling molecule release [92]. This reactive state plays a critical role in the early stages of migraine attacks, during which microglia release cytokines such as IL-1β, TNF-α, and IL-6, thereby exacerbating neuroinflammation [17]. For example, IL-6, secreted by stimulated microglia through the PAR2-MAPK-NF-κB signaling cascade, enhances neuronal excitability and contributes to dural afferent hypersensitivity. In animal models direct meningeal IL-6 administration induces migraine-like behavioral responses [93].

Although microglial activation is traditionally associated with proinflammatory outcomes, emerging evidence indicates microglia can transition into an anti-inflammatory phenotype under specific conditions. This alternative activation state involves secretion of neuroprotective and anti-inflammatory factors such as IL-10 and transforming growth factor-beta (TGF-β), which suppress proinflammatory signals and facilitate tissue repair [94]. In migraine, this phenotypic shift may be essential for resolving acute inflammation and promoting recovery. Microglial M2a polarization—a hallmark of the anti-inflammatory state—has been linked to reduced susceptibility to CSD [95]. Recent studies have shown that nasal administration of exosomes derived from interferon (IFN)-γ dendritic cells suppress pro-inflammatory M1 microglial polarization, mitigates oxidative stress, and elevates the threshold for CSD in migraine models [96]. In humans, Xiao et al. [97] found galectin-1 levels to be significantly reduced in CM patients and inversely correlated with proinflammatory cytokines while positively correlating with IL-10. Galectin-1 administration attenuated hyperalgesia and photophobia, promoted M2 microglial polarization, and suppressed proinflammatory cytokine expression through PI3K/AKT pathway activation in the spinal trigeminal nucleus. Additionally, in CM models, AMP-activated protein kinase (AMPK) alleviated central sensitization by promoting M2-type polarization in nitroglycerin (NTG)-induced [98]. The AMPK activator AICAR reduced hyperalgesia and enhanced anti-inflammatory cytokine expression through M2 polarization and NF-κB signaling pathway inhibition.

Microglial activation is also regulated by TREM1 and TREM2 receptors expressed on myeloid cells which modulate neuroinflammation in the brain. TREM1 was found to be upregulated in CM, and its activation triggers NF-κB–mediated neuroinflammation, subsequently activating the NLRP3 inflammasome and releasing IL-1β and IL-18, contributing to central sensitization [99]. Moreover, Fan et al. [100] demonstrated that metformin mitigates central sensitization by suppressing microglial-driven neuroinflammation through the downregulation of TREM2–SYK signaling. Bioinformatic analysis of RNA-seq data from migraine patients identified S100A8 as a key mediator of migraine pathogenesis. S100A8 colocalizes with microglia, and its genetic deficiency leads to elevated pain thresholds, reduced inflammatory cytokines and CGRP, and suppressed microglial activation [101]. Purinergic receptor-mediated signaling also plays a role in migraine-related neuroinflammation through the ATP breakdown products ADP and AMP that act as potent microglial activators to initiate inflammatory cascades [102]. The activation of specific purinergic receptors influences migraine pathophysiology differently. For example, activation of microglia P2×4 receptors (P2X4R) promotes the release of BDNF, which in turn enhances neuronal hyperexcitability by upregulating phosphorylated ERK (p-ERK) and CGRP release in the TNC, thus contributing to central sensitization in CM [103, 104]. Similarly, P2X7R activation facilitates neuroinflammation by disrupting autophagy and stimulating NLRP3 inflammasome signaling [105]. P2Y12R activation further aggravates CM-associated hyperalgesia through the RhoA/ROCK pathway. [106] Zhu et al. [107] demonstrated that microglial P2Y14 receptor expression in the trigeminocervical complex (TCC) is significantly upregulated following repeated inflammatory stimulation, and its inhibition reduced periorbital allodynia and suppressed microglial activation. Similarly, Zhou et al. [108] found that electroacupuncture (EA) at GB20/GB34 alleviated cutaneous allodynia in a rat migraine model induced by repeated IS. Mechanistically, EA treatment reduced microglial P2X4R expression, NLRP3/Caspase-1/IL-1β signaling, and c-Fos expression, while inhibiting Iba1-positive microglia proliferation. These findings suggest that EA’s antimigraine effects are mediated by suppressing microglia-driven neuroinflammation, particularly through modulation of the P2X4R–NLRP3/IL-1β pathway.

Additional molecular pathways contribute to migraine-related neuroinflammation. Elevated NLRP3 and IL-1β expression in activated microglia facilitates inflammatory cascade amplification [109] while suppressing the TLR4/MyD88/NF-κB/NLRP3 inflammasome pathway with alpha-asarone alleviates migraine symptoms [110]. This cascade suppression led to decreased levels of IL-1β, IL-18, and caspase-1, further supporting the central role of the NLRP3 inflammasome and related pathways in CM-associated neuroinflammation. Further, microglia TLR4 and IL-18 expression are upregulated in migraine models, and inhibition of TLR4 both reduces IL-18 levels and alleviates tactile allodynia [111]. Chen et al. [112] demonstrated that NTG administration increases BBB permeability, allowing peripheral IL-17A to enter the medulla oblongata, activate microglia, and trigger neuroinflammation. TLR2-mediated microglial activation in the TNC drives central sensitization in migraine, with TLR2 antagonist C29 attenuating NTG-induced mechanical hypersensitivity [113]. Sphingosine-1-phosphate receptor 1 (S1PR1) antagonism significantly reduced microglial activation and NTG-induced hyperalgesia, suppressing CGRP, c-Fos upregulation and STAT3 signaling in the TCC [114]. Yang et al. [115] investigated roxadustat, a hypoxia-inducible factor-1 alpha (HIF-1α) stabilizer, in migraine treatment finding that it alleviated NTG-induced basal hypersensitivity, acute hyperalgesia, and photophobia by downregulating phosphorylated ERK1/2 and c-Fos in the TNC, while suppressing neuroinflammation through the HIF-1α/NF-κB signaling pathway. Ligustrazine (LGZ) was found to mitigate migraine by upregulating and stabilizing sirtuin 1 (SIRT1), suppressing NF-κB signaling in microglia and activating nuclear factor erythroid 2- related factor 2 (Nrf2) via microglia-neuron crosstalk to reduce neuroinflammation and oxidative stress [116]. SIRT1 inhibition abolished the beneficial effects of LGZ, confirming SIRT1 as a key mediator. Exploring overlapping mechanisms between neurological disorders, Zhou et al. [117] examined the shared pathophysiology of epilepsy and migraine comorbidity. Seizures activate thalamocortical and spinal subnucleus caudalis (sp5c) microglia through the fractalkine (FKN)/CX3CR1 chemokine axis. Neuron-derived FKN was shown to amplify microglial BDNF synthesis via CX3CR1 binding, suggesting epileptiform activity primes microglia through FKN/CX3CR1-BDNF signaling, lowering migraine thresholds. In a separate study, Hu et al. [118] identified indoleamine 2,3-dioxygenase 1 (IDO1) as a key contributor to neuroinflammation and pain in migraine. Using NTG-induced CM models, they demonstrated that IDO1 in the anterior cingulate cortex (ACC) mediates pain sensitization and anxiety-like behaviors. IDO1 upregulation disrupts the excitatory/inhibitory (E/I) neuronal balance and promotes microglial synaptic pruning via IFN signaling, revealing the ACC microglial IDO1–IFN axis as a novel therapeutic target for migraine-associated allodynia and affective symptoms. Several studies have identified differentially expressed microRNAs in migraine’s interictal phase associated with immune responses and neuroinflammation [119]. miR-155-5p promotes inflammation by inhibiting anti-inflammatory proteins like SIRT1 [120], and its upregulation has been linked to increased microglial activation and neuroinflammation in migraine models [121]. Inhibition of miR-155-5p alleviates central sensitization and reduces neuroinflammation, suggesting its therapeutic potential [121].

Across repeated NTG/IS paradigms, convergent evidence points to purinergic signaling (eg. P2X4/7) and inflammasome-related pathways (eg. NLRP3/IL‑1β) as recurring drivers of trigeminal sensitization, whereas targets such as HIF‑1α or IDO1 appear model- and region-dependent. Few studies directly test reversal of established hypersensitivity, limiting translational inference.

In summary, microglial activation is central to the neuroinflammatory processes underlying migraine and collectively highlight the central importance of modulating microglial in the regulation of migraine pathophysiology. The selective targeting of pathways involved in microglial activation, including specific microRNAs and cytokine blockers, offers promising strategies for developing more effective migraine treatments.

Oligodendrocytes in migraine: myelin integrity and axonal support

Gene set analysis reveals significant associations between oligodendrocytic gene expression and migraine, suggesting oligodendrocyte dysfunction contributes to migraine pathophysiology [122]. Indeed, oligodendrocytes play dual roles in neuroinflammation, releasing both proinflammatory cytokines (TNF-α, IL-1β) and anti-inflammatory mediators in response to inflammatory stimuli [52]. Proinflammatory cytokines induce oligodendrocyte apoptosis and myelin degradation, processes exacerbated by oxidative stress—a key factor in migraine pathophysiology that impairs neural transmission and intensifies symptoms [40]. A recent study revealed myelin abnormalities in the trigeminal nerves of migraine patients, further supporting the potential link between myelin disruption and migraine pathology [123]. Oligodendrocyte dysfunction also influences central sensitization through glial crosstalk modulation. Dysfunctional oligodendrocytes increase cytokine release (TNF-α, IL-6, IL-33), amplifying nociceptive signaling and contributing to chronic pain development [50, 124]. However, the exact mechanisms linking oligodendrocyte dysfunction to neuroinflammation in migraine remain unclear. Oligodendrocytes are also critical for neuroprotection as they secrete neurotrophic factors such as BDNF and IGF-1, which promote neuronal survival and aid in the repair of damaged tissues [125]. These factors may counteract the damaging effects of neuroinflammation, encouraging remyelination and enhancing neuronal resilience. However, prolonged inflammatory states as observed in chronic migraine can impair the ability of oligodendrocytes to perform these protective functions, potentially leading to further neurological dysfunction and the progression of chronic pain [126]. Compared with the substantial mechanistic evidence implicating astrocytes and microglia in migraine-related neuroinflammation, direct support for oligodendrocyte dysfunction in acute attack biology remains limited. Current associations are derived primarily from genetic enrichment analyses and indirect structural or imaging observations, such as reported trigeminal myelin abnormalities, which require cautious interpretation due to variations in cohort characteristics, imaging versus histological methodologies, phenotypic definitions, and unmeasured confounders [122, 123].

In summary, oligodendrocytes are central to myelin integrity maintenance and contribute to neuroinflammation in migraine through cytokine modulation, glial crosstalk, and neurotrophic factor secretion (Fig. 2). However, it remains to be determined whether selectively targeting these cells might offer therapeutic potential.

Schwann cells in migraine: nociceptor sensitization and neuropeptide modulation

Emerging evidence suggests SCs contribute to the inflammatory environment through interactions with neural and immune components, influencing both peripheral and central sensitization in migraine pathophysiology [127–129]. CGRP released from cutaneous trigeminal fibers binds to CRLR/RAMP1 complexes on adjacent SCs, leading to periorbital mechanical allodynia. This activation induces cyclic AMP (cAMP)-dependent NO production and TRPA1 activation, triggering ROS release and sustaining allodynia via nociceptor TRPA1 activation [128]. Landini et al. [129] further explored ethanol-induced migraine pathways in mice, identifying SCs as central mediators of ethanol-evoked pain. Ethanol administration triggered periorbital mechanical allodynia, a response dependent on acetaldehyde production and TRPA1 activation. Silencing RAMP1 in SCs abolished this response, mimicking the effects of CGRP receptor blockade. In a fibromyalgia (FM) model, reserpine treatment causes periorbital mechanical allodynia, increased oxidative stress, and macrophage infiltration, which are absent upon macrophage depletion or SC TRPA1 inhibition/deletion. Selective silencing of TRPA1 in SCs reduced both allodynia and neuroinflammation, highlighting the role of SC TRPA1 in ROS production and macrophage infiltration [130]. Titiz et al. [131] explored the involvement of SCs in endometriosis-related pain, where silencing the complement component 5a receptor 1 (C5aR1) in SCs blocked inflammasome activation and IL-1β secretion, preventing pain and oxidative stress. These findings underscore the critical role of SCs in amplifying and sustaining neuroinflammation through cytokine signaling and immune cell recruitment (Fig. 3).

Fig. 3.

Summary of the potential role that peripheral glial cells may play in migraine. During a migraine attack, events such as oscillations in hypothalamic activity and/or increased cortical excitability can activate the trigeminovascular system to release CGRP from trigeminal C-fibers. In addition, CSD can release extracellular K⁺ and H⁺ ions as well as proinflammatory substances from meningeal glial cells and nerve terminals. These substances can activate trigeminal nerve endings, resulting in further release of CGRP and substance P. CGRP can cause dilatation of the meningeal vasculature, leading to further vascular leakage and mast cell degranulation. Moreover, CGRP can act directly on adjacent non-CGRP Aδ sensory neuronal cell bodies that express the CGRP receptor, sensitizing them. Extracellular ATP released by neurons in the TG activates SGCs mainly via P2Y and P2Y2 receptors, accompanied by IL-1β and IL-6 secretion and the release of the inflammatory state. ATP also binds to P2Y, P2Y14, and P2×3 receptors on neurons. Stimulation of the TRPV1 receptor on C fibers in the trigeminal ganglion leads to the release of CGRP. Neurally released CGRP is another aspect of the interaction, and it binds to the CRL/RAMP1 receptor on SGCs, causing NO release. In addition, CGRP activates CLR/RAMP1 on schwann cells, triggering the production of NO, which gates TRPA1 to sustain ROS-dependent mechanical allodynia via the nociceptor TRPA1. Neuroglial signaling at peripheral sites induces inflammation and nociception, resulting in an inflammation‒allodynia positive loop in the TG. Illustrations were created via biorender (http://app.biorender.com)

Schwann cells actively engage with immune cells to modulate inflammation and tissue repair. In response to peripheral nerve injury or sustained neuroinflammation, SCs recruit macrophages and T cells, which play crucial roles in clearing myelin debris and amplifying the inflammatory response [54]. SCs also interact with mast cells and macrophages, further enhancing neuroinflammation. This crosstalk is particularly relevant in migraine, where peripheral and central glial cells, including SCs, form a complex immune signaling network that exacerbates neuroinflammation in the trigeminal system [132].

Satellite glial cells in migraine: peripheral sensitization and pain signal amplification

SGCs are critical for maintaining homeostasis, regulating ion balance, and providing metabolic support to neurons [59]. In migraine, SGCs contribute to peripheral sensitization development and maintenance, where heightened neuronal excitability increases pain perception [133]. As noted before, SGCs undergo reactive gliosis under pathological conditions, characterized by morphological changes, increased GFAP expression, and altered gap junction communication. In migraine, this contributes to peripheral sensitization via amplify neuronal hyperexcitability and pain signaling [133, 134].

SGCs maintain bidirectional communication with sensory neurons, influencing pain modulation. Gap junctions, mediated by connexins, facilitate the passage of ions and signaling molecules between SGCs and neurons, enhancing excitatory signal spread and contributing to neuronal hyperexcitability and pain [135, 136]. GPR37L1, a G protein–coupled receptor in SGCs, regulates potassium channels (KCNJ10, KCNJ3) and pain signaling. Reduced GPR37L1 expression impairs pain resolution, whereas overexpression reverses it. The GPR37L1 ligand, Maresin 1 (MaR1), enhances potassium influx while a GPR37L1 genetic variant E296K increases chronic pain risk making this GPCR a therapeutic target [137]. SGCs respond to neuronal activity by upregulating purinergic receptors, particularly P2×7 [138], P2Y2 [139] and P2Y14 [140], which propagate inflammatory signaling and further sensitize pain pathways. Disruption of interactions between SGCs and sensory neurons under migraine conditions reinforces headache persistence [134]. SGCs further contribute to neuroinflammation by releasing cytokines, chemokines, and other mediators that together enhance nociceptive sensitization. For example, activated SGCs upregulate TNF-α, IL-1β, and IL-6, to increase neuronal excitability and prolong pain signaling [19] while ROS produced in SGCs exacerbate inflammation in the trigeminal ganglion [141]. Combined, elevated inflammatory mediators in the PNS sensitize peripheral nociceptors and facilitate central sensitization.

SGCs in the mouse TG are transcriptionally activated after a single CSD, indicating that CSD impacts both cortical glial cells and in distant regions [142]. SGCs also interact with various neuropeptides that play crucial roles in migraine pathophysiology. For example, SGCs in the TG express the canonical CGRP receptor (CLR/RAMP1) and other CGRP receptors such as AM2 (CLR/RAMP3) and AMY3 (CTR/RAMP3). [143] Neuronal CGRP release activates CGRP receptors on SGCs, leading to NO production via MAPK pathway to modulate pain behaviors [144]. This CGRP-mediated positive feedback between TG neurons and SGCs might be involved in the progression of migraine, presenting a therapeutic target [145]. Antibodies against CGRP and its receptors targeting these neuron–SGC interactions have shown promise in reducing migraine symptoms [146]. Additionally, neurokinin substance P and PACAP, released upon the activation of trigeminal neurons, contribute to migraine. Substance P stimulates IL-1 production in glial cells, further amplifying inflammatory pain [147]. PACAP, expressed in both neurons and SGCs, modulates pain through PAC1, VPAC1, and VPAC2 receptors [148].

In conclusion, SGCs are pivotal in migraine development and maintenance through ion homeostasis, neuroinflammation, and neuropeptide signaling. Their interactions with TG neurons and release of inflammatory mediators contribute to peripheral and central sensitization (Fig. 2). Targeting SGC-neuron interactions, particularly via CGRP and other neuropeptide pathways, may offer promising therapeutic strategies for chronic migraine and related conditions. Further research however is needed to fully understand potential SGC targets such as PACAP and its role in orofacial pain.

Strengths and limitations

The studies summarized above collectively support a mechanistic role for glia in migraine-relevant neuroinflammation, with the strongest causal evidence for astrocytes and microglia. Familial hemiplegic migraine models link astrocytic ion/glutamate clearance deficits to heightened CSD susceptibility and neurovascular coupling abnormalities [78–83]. Complementary pathway studies connect CSD to astrocytic inflammatory activation via Panx1/inflammasome signaling and HMGB1–NF‑κB pathways [65–67], while microglial work repeatedly implicates purinergic receptors and inflammasome-related cascades (e.g., P2X4/7/NLRP3/IL‑1β) in sensitization endpoints [103–105, 109]. Targeted manipulations (e.g., conditional GLT‑1 loss; astrocyte chemogenetic activation) further strengthen inference [77, 87].

Key limitations include model-to-disease mapping, marker specificity, and confounding. NTG/GTN and IS paradigms largely model sustained trigeminal hypersensitivity and should not be equated with clinically defined chronic migraine without qualification [64]. Reliance on GFAP/Iba1 and cytokine panels provides limited resolution of functional glial states and is sensitive to timing, stress/anesthesia, and sex/hormonal variables. Directionality may be obscured by systemic effects; NTG-associated BBB permeability permitting peripheral IL‑17A entry suggests secondary microglial activation in some settings [112].

Future studies should use cell- and region-specific approaches with time-resolved designs to distinguish initiation vs maintenance and to test reversal of established hypersensitivity. Convergent purinergic/inflammasome nodes merit rigorous replication across models and anatomical sites [103–105, 109]. Peripheral–central coupling should be directly interrogated, motivated by SGC transcriptional activation after CSD [142].

Mechanisms of glial cell activation in migraine-associated neuroinflammation

Triggers of glial cell activation

Migraine-driven glial cell activation is triggered by factors such as CSD, neuronal hyperactivity, neuropeptides, and inflammatory mediators [13]. CSD induces neuronal and glial activation, releasing excitotoxic signals, potassium ions, and proinflammatory molecules, contributing to microglia and astrocyte activation, neuroinflammation and pain signaling [149]. Glial-neuronal communication, through neurotransmitters and neuropeptides modulates neuronal excitability and enhances synaptic transmission, sustaining the TGV system and contributing to migraine pathophysiology [26, 150]. Dysregulated glial-neuronal interactions drive migraine persistence [151].

Inflammatory mediators, such as cytokines, chemokines, and oxidative stress markers, are key in glial cell activation. Proinflammatory cytokines, including IL-1β, TNF-α, and IL-6, released by glia and neurons, increase neuronal excitability, disrupt the BBB, and facilitate immune cell infiltration [152]. In response to CSD, HMGB1 is released and activates astrocytes [67] and microglia [153], stimulating nociceptors and perpetuating TGV system inflammation [154]. The continuous release of these mediators promotes central sensitization, priming the CNS for future migraine attacks [65]. Recent studies highlight long noncoding RNAs (lncRNAs) and microRNAs in regulating migraine-related neuroinflammation. For example, the LncRNAs PVT1, MEG3, and LINC-ROR are upregulated in migraine, particularly with aura, and regulate inflammation through pathways such as miR-147/NF-κB [155, 156]. MicroRNAs like miR-155 exacerbate inflammation by downregulating anti-inflammatory proteins like SIRT1 [120, 157]. Together, 31 differentially expressed microRNAs have been linked to immune responses, neuroinflammation, and oxidative stress in the interictal phase [119].

Signaling transduction underlying glial activation

Glial ion channels and receptors play crucial roles in migraine-linked neuroinflammation. The P2×7 receptor on microglia, activated by extracellular ATP, triggers inflammatory cascades, including IL-1β and ROS release, exacerbating neuronal excitability and pain [158, 159]. Connexin-43 hemichannels in astrocytes amplify neuroinflammation [160], and the P2Y12 receptor in the TNC contributes to microglial activation through the RhoA/ROCK pathway, impacting migraine pathology [106]. The PAR2-MAPK-NF-κB signaling cascade modulates IL-6 secretion in microglia [161]. Microglial BDNF, released via the FKN/CX3CR1 axis, amplifies neuroinflammatory responses and promotes microglial recruitment [117]. Examining CXCR5 inhibitors in preclinical models shows reduced CX3CR1/FKN/BDNF expression and improved pain thresholds [117]. Another key pathway involves glucagon-like peptide-1 receptor (GLP-1 R), which is colocalized with Iba1 in BV-2 microglia and is upregulated in migraine models where it correlates with increased microglial numbers and neuroinflammation [162]. Xiongzhi Dilong decoction (XZDL) alleviates migraine via α7nAChR-mediated inhibition of astrocytic activation [163]. IL-18 R in astrocytes contributes to neuroinflammation, with microglia producing IL-18 via TLR4/p38 MAPK signaling to amplify inflammatory responses [164].

Interactions between glial cells and neurons

The interactions between glial cells and neurons are central to maintaining and amplifying neuroinflammation in migraine (Fig. 2 and Fig. 3). Astrocytes modulate glutamate clearance and potassium buffering, directly influencing neuronal firing and CSD propagation, which sensitizes trigeminal pathways. Astrocytic dysfunction, particularly impaired glutamate clearance, exacerbates neuroinflammation and pain signaling [151]. Microglial activation through purinergic receptors [104, 106, 159] further amplifies neuroinflammation and perpetuates pain signaling. Microglial-derived BDNF promotes neuroinflammation via the FKN/CX3CR1 axis [117]. Additionally, IL-18 release from microglia enhances nociception and central sensitization through IL-18 receptors on neurons [165]. This neuronal-glial crosstalk perpetuates pain signaling and contributes to migraine chronicity [166].

Microglia and astrocytes communicate through neurotransmitters, cytokines, and ATP, modulating neuronal excitability and enhancing synaptic transmission in the TGV system [26, 150]. Astrocyte–microglia signaling amplifies neuroinflammation through ATP-mediated P2×7 receptor activation [167], IL-18/IL-18 R [164] and NLRP3 [168] inflammasome signaling. A study by Liddelow et al. [169] demonstrated that microglia-derived IL-1α, TNF-α, and C1q collectively drive astrocytes toward the reactive phenotype. Microglia inflammatory mediators also activate STAT3 signaling, promoting astrocyte conversion into the anti-inflammatory phenotype that supports neuroprotection [170]. Conversely, astrocytes induce microglial activation toward the proinflammatory phenotype [171]. Microglia polarized to the anti-inflammatory phenotype release IL-10, which enhances neuroprotection by promoting astrocytic TGF-β secretion [172]. Moreover, astrocytes release IL-33 to modulate microglial synaptic pruning, regulating neuroinflammation [173]. TGF-β from astrocytes suppresses microglial activation and limits inflammation [174]. Supporting the relevance of this glial crosstalk in migraine, pharmacological inhibition of TRPA1 attenuates migraine-related hyperalgesia by suppressing glial activation and reducing CGRP and proinflammatory cytokines [175].

Strengths and limitations

This section compiles convergent preclinical evidence that CSD, neuropeptides, and inflammatory mediators activate glia and amplify trigeminovascular signaling, supported by HMGB1–NF‑κB links [67, 149, 153, 154] and purinergic nodes (P2X7/P2Y12) with downstream cytokines/ROS [106, 158, 159]. Inclusion of noncoding RNA regulation broadens the mechanistic framework [119, 155–157].

However, much of the evidence derives from acute trigger paradigms and surrogate endpoints, with limited validation in human migraine. Several pathways are inferred from pharmacologic interventions with pleiotropic targets, and some data rely on BV‑2 microglia, which may not recapitulate in vivo states [162]. Confounders include anesthesia/surgical stress in CSD models, systemic inflammation and BBB changes, sex/hormonal status, and sampling time relative to attacks [13, 152].

Future studies should prioritize cell- and region-specific perturbations and time-resolved profiling across cortex, TNC/TCC, and trigeminal ganglion to distinguish initiation from maintenance [151, 166]. Human studies are needed to test whether candidate nodes (HMGB1, P2X7, IL‑18) track disease activity and treatment response [67, 158, 164].

Potential therapeutic targets involving glial cells

Clinical evidence linking glial cells and neuroinflammation to migraine

While preclinical models have long implicated glial-mediated neuroinflammation in migraine pathogenesis [176–185] (Table 1), a growing body of clinical evidence has begun to elucidate glial contributions to neuroinflammation in migraine patients [122, 186–193]. These studies have utilized a spectrum of methodologies, including neuroimaging [189, 194, 195], blood-based biomarkers [182, 183, 185, 196–198], CSF analysis [193, 199, 200], and genetic profiling [122, 190, 201–203], offering valuable new insights into the neuroimmune landscape of migraine (Table 2).

Table1.

Summary of potential glia targets and mechanisms in the regulation of neuroinflammation of migraine

Targets	Animal model	Glia Cell	Tissue	Mechanisms of action	Reference
ROS/NF-κB	IS-induced CM rat model	Astrocytes	TNC	Astragaloside IV mitigates CM-related central sensitization by targeting astrocyte mitochondrial dysfunction, suppressing ROS production, and the NF-κB/IL-1β pathway.	Zhang et al. [88]
IDO1	NTG-induced CM mouse model	Microglia	ACC	IDO1 drives microglial activation and aberrant synaptic pruning via the IFN signaling pathway in the ACC.	Hu et al. [118]
Gq-GPCR signaling	Rat model (AAV- based chemogenetic approach)	Astrocytes	Visual cortex	Astrocyte Gq-GPCR activation triggers gliotransmitter release, promoting proinflammatory/nociceptive signaling.	Bree et al.[87]
Galectin-1	NTG-induced CM mouse model	Microglia	TNC	Galectin-1 deficiency exacerbates chronic migraine, while its exogenous supplementation alleviates symptoms by reducing inflammation and promoting microglial M2 polarization via PI3K/AKT signaling.	Xiao et al. [97]
P2X7R	Thy1-ChR2 transgenic mice (for optogenetic CSD induction), and C57BL/6, BALB/c, and Swiss albino mice (for general CSD studies)	Astrocytes	Cortical and subcortical regions	P2X7R activation mediates ATP-driven subcortical spread of CSD and neuroinflammation via HMGB1/NF-κB signaling in astrocytes.	Uzay et al. [68]
α7nAChR	NTG-induced CM rat model	Astrocytes	TNC	XZDL enhances α7nAChR expression, suppressing astrocyte activation via JAK2/STAT3/NF-κB and reducing neuroinflammation, thereby alleviating central sensitization in CM.	Tang et al.[163]
TREM1	NTG-induced CM mouse model	Microglia	TNC	TREM1 triggers NF-κB-dependent NLRP3 inflammasome activation in microglia, promoting neuroinflammation and central sensitization via proinflammatory cytokines and CGRP release.	Sun et al. [99]
AMPK	NTG-induced CM mouse model	Microglia	TNC	AMPK activation attenuated central sensitization by promoting microglial M2 polarization via NF-κB inhibition, reducing proinflammatory cytokines and CGRP expression.	Lu et al. [98]
TLR2	NTG-induced CM mouse model	Microglia	TNC	Microglial TLR2 promotes central sensitization in CM by driving neuroinflammation (IL-6, IL-1β, IL-10, TNF-α and IFN-β1 were increased).	Liu et al. [113]
TLR4	IS-induced migraine rat model	Microglia	Cortex and hippocampus	Alpha-asarone alleviates hyperalgesia and improve behavioral performance in migraine rats by inhibiting hyperexcitability and neurogenic inflammation via TLR4/NF-κB/NLRP3 signaling pathway.	Liu et al. [110]
SIRT1	NTG-induced mouse model of migraine	Microglia	Brain tissue	SIRT1 stabilization in microglia suppresses NF-κB-driven neuroinflammation and reduces oxidative stress via microglia-neuron crosstalk, thereby alleviating migraine.	Chen et al. [116]
S100A8	Migraine rat model	Microglia	TG and medulla oblongata	S100A8 inhibition reduces neuroinflammation and microglial activation, thereby alleviating migraine pain.	An et al. [101]
HIF-1α	NTG-induced CM mouse model	Microglia	TNC	Roxadustat attenuated migraine-like behaviors by inhibiting HIF-1α/NF-κB-driven neuroinflammation in microglia.	Yang et al. [115]
HMGB1	Thy1-ChR2-YFP transgenic mice	Astrocytes	Cerebral cortex	Neurons release HMGB1 via sEVs upon CSD, which are selectively taken up by astrocyte processes, triggering NF-κB p65 nuclear translocation and proinflammatory signaling.	Kaya et al. [67]
CX3CR1	IS-induced CM rat model	Microglia	Temporal lobe cortex, thalamus, and TNC	FKN interacts with CX3CR1 expressed on microglia to promote microglial activation and subsequent BDNF release.	Zhou et al. [117]
S1PR1	NTG-induced CM mouse model	Microglia	TCC	S1PR1–STAT3 axis activation in microglia promotes neuroinflammation and central sensitization.	Pan et al. [114]
CRLR/RAMP1 receptors	Ethanol/CGRP	Schwann cells	TG	CLR/RAMP1 activation in Schwann cells triggers production of NO, which gates TRPA1 to sustain ROS-dependent mechanical allodynia.	De Logu et al. [128]
P2Y14	CCI-ION	SGCs	TG	P2Y14R activation promotes trigeminal neuropathic pain via ERK1/2- and p38-dependent neuroinflammation (elevated IL-1β, IL-6, CCL2, TNF-α).	Lin et al. [140]
TRPA1	NTG-induced migraine rat model	Microglia and SGCs	TNC and TG	Inhibition of TRPA1 modulated glial activation, reduced pro-inflammatory cytokines, and normalized CGRP release, thereby disrupting the neuron-glia inflammatory feedback loop driving migraine pain.	Demartini et al. [175]
IL-17A	NTG-induced migraine rat model	Microglia	TNC	IL-17A crosses the compromised blood-brain barrier to activate microglia-mediated neuroinflammation in the TNC.	Chen et al. [112]
P2Y14	IS-induced migraine rat model	Microglia	TCC	P2Y14 receptor activation in TCC microglia induces mechanical allodynia via ERK1/2 phosphorylation and microglial activation.	Zhu et al. [107]
GLP-1 R	NTG-induced CM mouse model	Microglia	TNC	GLP-1 R activation suppressed central sensitization by inhibiting microglial activation via the PI3K/Akt pathway.	Jin et al. [162]
MicroRNA- 155-5p	NTG-induced CM mouse model	Microglia	TNC	miR-155-5p promotes neuroinflammation and central sensitization by inhibiting SIRT1, leading to microglial activation (M1 polarization), activation of ERK/CREB, and increased release of proinflammatory cytokines and pain mediators.	Tang et al. [121]
P2X7R	NTG-induced CM mouse model	Microglia	TNC	P2X7R contributes to central sensitization in CM by mediating autophagy impairment, which promotes microglial activation and NLRP3 inflammasome-driven neuroinflammation.	Jiang et al. [105]
HMGB1, NF-κB	Pinprick or topical KCl-induced CSD in FHM1 mutant mice	Astrocytes	Cortex and subcortical areas	Enhanced glutamatergic neurotransmission via NMDA receptors triggers Panx1 channel opening, leading to HMGB1 release and NF-κB-driven neuroinflammation in hyperexcitable FHM1 mutant brains.	Dehghani et al. [66]
GluN1–N2B NMDA receptors	FHM2 mutant mice	Astrocytes	Somatosensory barrel cortex	Impaired astrocytic glutamate clearance prolongs extrasynaptic glutamate spillover, preferentially activating GluN1–N2B NMDA receptors to facilitate CSD.	Crivellaro et al. [81]
ATP1A2	FHM2 mutant mice	Astrocytes	Cerebral cortex	Hyperactive astrocytes in Atp1a2± mice exhibited elevated Ca2 + concentrations post-CSD, exacerbating CSD propagation.	Sugimoto et al. [79]
P2X4, BDNF	NTG-induced CM mouse model	Microglia	TNC	P2X4R activation in microglia triggers BDNF release, promoting neuronal hyperexcitability via increased p-ERK and CGRP in the TNC, contributing to central sensitization in CM.	Long et al. [103]
IL-18/IL-18 R	IS-induced migraine rat model	Microglia, astrocytes	TNC	Microglial IL-18, induced by TLR4/p38 MAPK, binds to astrocytic IL-18 R, driving NF-κB-dependent astrocyte activation.	Gong et al. [164]
EAAT2	IS-induced migraine rat model	Astrocytes	TNC	EAAT2 downregulation leads to glutamate accumulation in the synaptic cleft, overactivating NMDA receptors and enhancing synaptic plasticity, thereby promoting central sensitization in CM.	Zhou et al. [76]
GLT-1 (EAAT2)	Astrocyte-specific GLT-1 knockout mice	Astrocytes	Cerebral cortex	GLT-1 deficiency in astrocytes accelerates extracellular glutamate accumulation during CSD, increasing neuronal excitability and CSD propagation.	Aizawa et al. [77]
P2Y12	NTG-induced CM mouse model	Microglia	TNC	Microglial P2Y12R activation promotes CM-associated hyperalgesia via the RhoA/ROCK pathway, leading to microglial activation, pro-inflammatory cytokine release, and subsequent neuronal sensitization.	Jing et al. [106]
NLRP3/IL-1β	NTG-induced CM mouse model	Microglia	TNC	Microglial NLRP3 inflammasome activation mediates IL-1β, promoting central sensitization via neuronal IL-1 R.	He et al. [109]
TRPA1	KCl-induced CSD rat model	Astrocytes	Cerebral cortex and TNC	ROS facilitates CSD induction by activating TRPA1 and promoting CGRP release, forming a positive feedback loop.	Jiang et al. [89]
TLR4/IL-18	IS-induced migraine rat model	Microglia	TNC	TLR4 activation promotes microglial NF-κB-mediated release of proinflammatory cytokines and BDNF, driving peripheral/central sensitization and facial mechanical hyperalgesia.	Su et al. [111]
P2X4	NTG-induced CM mouse model	Microglia	TNC	Microglial P2X4R activation promoted central sensitization by enhancing neuronal excitability via BDNF/CGRP signaling.	Long et al. [104]
Na+/K+ ATPase	FHM2 knocked-in mice	Astrocytes	N/A	FHM2 mutations increases the susceptibility of the brain to CSD by reducing the clearance of K+.	Unekawa et al. [80]
HMGB1/TLR2/4	CSD rat model	Microglia	Cerebral cortex	CSD inductions caused significant HMGB1-dependent microglial activation via TLR2/4.	Takizawa et al. [153]
P2Y2R	CFA-induced rat neuropathic pain models	SGCs	TG	Selective blocking of P2Y2R alleviates allodynia and reduces activation of satellite glial cells in neuropathic pain models.	Magni et al. [139]

Note: Models vary in their ability to recapitulate human migraine features; results should be interpreted with caution regarding glial-specific mechanisms

Table 2.

Clinical evidence of glia cells and neuroinflammation in migraine

Subjects	Research Methods	Results	Reference
58 CM and 69 controls	Plasma analysis	No significant differences were found in plasma GFAP or NfL levels between chronic migraine patients and controls	Colombo et al. [186]
19 CM patients and 10 healthy controls	PET/MR imaging and plasma analysis	[11C] PBR28 binding (a marker of glial activation) was elevated in midbrain/occipital regions, along with higher interictal plasma IL-8/CX3CL1 levels.	Chang et al. [194]
186 migraine hospitalized in-patients (67 MA and 119 MO) and 20 controls.	Blood analysis	MA patients had significantly lower systemic immune-inflammatory indices (SSIIi) compared to those without aura (MO).	Wijeratne. [204]
60 migraine patients and 20 controls	Functional imaging	Migraine patients exhibit altered trigeminal nerve root microstructure (reduced fractional anisotropy), associated with neuroinflammation and diminished brainstem activation.	Tohyama et al. [195]
20 Meniere’s Disease, 20 vestibular migraine, and 10 control patients	PBMC analysis and questionnaire	Menière’s disease patients had higher levels of TNF-α and IFN-γ but lower levels of ENA-78 in PBMCs compared to vestibular migraine patients.	Monaghan. [205]
309 patients with migraine without aura and 199 controls	Blood analysis	Migraine patients without aura exhibit persistent hemoconcentration (elevated hematocrit, platelets, MPV) and chronic inflammation (reduced MLR) even during pain-free periods.	Kömürcü et al. [183]
64 female patients (24 EWM, 20 epilepsy, and 20 MO) and 20 controls	Cross-sectional study	Female patients with comorbid epilepsy and MWoA exhibit synergistic reductions in NAA and myoinositol, indicating more severe injury of neuronal and glial cell.	Wang, et al. [187]
39 patients with migraine (23 during attacks and 16 between attacks), 19 controls	CSF and serum analysis	Elevated CSF CX3CL1 levels (both during and between attacks) and increased serum t-Tau during headache phases.	Süße, et al. [193]
40 CM patients with NSAID overuse headache, 35 EM patients, and 20 healthy controls (all women participants)	Serum analysis	Chronic migraine with medication overuse headache is associated with elevated serum LPS levels, along with increased nociceptive/pro-inflammatory molecules (HMGB1, HIF-1α, IL-6, and CGRP).	Vuralli, et al. [197]
92 migraine patients and 88 healthy controls	Genetic analysis	The IL1A −889C > T genetic variant showed an association with aura presence and osmophobia in migraine patients.	Rocha, et al. [201]
47 patients with migraine and 47 healthy controls	Serum analysis	GSDMD (an inflammatory marker) elevation correlates with migraine symptoms (nausea/vomiting/pain severity) but not with overall migraine diagnosis.	Ocal, et al. [182]
53 migraine patients and 53 healthy controls	Serum analysis	Migraine patients exhibit significantly lower serum levels of the anti-inflammatory mediators PGE2 and L×A4 compared to healthy controls, both during and between attacks	Kocaturk, et al. [198]
285 women with migraine	Cross-sectional study	Higher dietary inflammation scores were significantly associated with increased odds of chronic migraine in Iranian women.	Bakhshimogh, et al. [206]
23 EM, 35 CM, and 29 healthy controls	Plasma analysis	Several plasma proteins associated with inflammation (e.g., haptoglobin, fibrinogen, complement C3) in migraine patients compared to controls or pain-free periods	Togha, et al.[181]
313 consecutive patients with episodic migraine without aura	Genetic analysis	The TNF-α gene polymorphism (−308 A/G) significantly reduces NSAID efficacy in migraine treatment.	Rubino, et al. [202]
4005 patients who visited the neurology outpatient department (OPD) within 30 days after visiting the emergency room (ER) for headaches	Blood analysis	Peripheral inflammatory markers (especially NLR and NMR) were significantly elevated during migraine attacks.	Lee, et al. [180]
251 EM patients and 119 CM patients and 100 healthy controls	Genetic analysis	A proinflammatory/pro-oxidative genetic profile (specifically, LTA rs2071590T and SOD1 rs2234694C alleles and the TGCT haplotype) is associated with white matter hyperintensities in migraine patients.	Ferroni, et al. [203]
120 migraine patients (54 MO and 66 MA patients) and 40 healthy subjects.	Serum analysis	Migraine patients had higher levels of IFN-γ, IL-4, TGF-β, and TNF-α, and lower CXCL8 compared to controls	Taheri et al. [185]
80 EM patients	Serum analysis	Vitamin D3 treatment increased TGF-β and prevented IL-17 rise in EM patients.	Ghorbani, et al. [207]
43 patients with CM, 19 with EM, 29 controls (mostly women), and 22 with CH	Serum analysis	No significant difference was found in interictal serum S100B levels (a marker of activated glia cells) between chronic migraine patients and control groups.	Riesco, et al. [188]
45 non-menopausal women migraine patients	Serum analysis	CoQ10 supplementation significantly reduced CGRP and TNF-α levels and improved migraine in non-menopausal women.	Dahri et al. [208]
13 migraineurs with aura in the interictal state	Functional imaging (PET/MRI brain scans)	Standardized uptake value ratio is elevated in nociceptive processing areas and visual cortex, suggesting glia activation and neuroinflammation in migraine with aura.	Albrecht et al. [189]
43 migraine patients (23 chronic and 20 episodic migraineurs) and 40 healthy controls	Serum analysis	Migraine patients (both episodic and chronic) had significantly elevated serum TNF-α levels compared to controls	Martami et al. [179]
2,039 individual cells from human postmortem brain cortex	Genetic analysis	Migraine susceptibility loci were expressed in glial cells.	Renthal et al. [190]
20 EM patients with/without aura patients and 17 healthy people as controls (all women)	Plasma analysis	Episodic migraine patients have higher levels of TNF-α and IL-12p70, and lower levels of IL-6, IL-8, and IL-10 during interictal period vs. controls	Oliveira et al. [196]
5,954 migraineurs	Genetic analysis	Gene sets containing astrocyte- and oligodendrocyte- related genes are associated with both migraine with aura and migraine without aura.	Eising et al. [122]
33 chronic migraine patients.	Clinical trials	Ibudilast, a potential glial inhibitor, did not improve chronic migraine.	Kwok et al. [191]
72 MA and 31 MO patients and 100 healthy controls	Serum analysis	IL-6 were increased in migraine patients during attacks and pain-free periods vs. controls	Wang et al. [184]
20 pain-free subjects as controls, 39 episodic TTH patients, 34 MA patients, 24 MO patients, and 10 with cervicogenic headache	CSF analysis	TGF-β1, IL-1ra, and MCP-1 were elevated in episodic TTH and MO groups vs. controls; IL-1ra, MCP-1were increased in MA vs. controls.	Bo et al. [200]
21 migraine children (12 MO and 9 MA) and 24 episodic TTH children	Plasma analysis	IL-1α and TNF-α, were elevated in children with migraine compared to those with TTH.	Bockowski et al. [178]
20 NDPH patients, 16 chronic migraine patients, and 2PT patients	CSF and serum analysis	CSF TNF-α levels were elevated in nearly all NDPH, chronic migraine, and PT headache patients; serum TNF-α were normal.	Rozen et al. [199]
25 patients with migraine during attacks and attack-free periods and 25 healthy controls	Plasma analysis	Patients had higher levels of IL-10, IL-6, RANTES, and NO during migraine attacks vs. attack-free periods and controls.	Fidan et al. [177]
15 migraine patients and 10 TTH patients.	Serum analysis	Serum glial S100beta protein levels are significantly elevated in children with migraine.	Papandreou et al. [192]
25 migraine patients (during ictal and interictal periods) and 18 healthy subjects	Plasma analysis	Plasma levels of IL-10, TNF-α, and IL-1β are significantly higher during migraine attacks	Perini et al. [176]

Neuroimaging studies provide compelling in vivo evidence of neuroinflammation. Using positron emission tomography (PET)/magnetic resonance (MR) scanning with the translocator protein (TSPO) radioligand [¹¹C] PBR28 (a marker of glial activation), Chang et al. [194] reported elevated [^{1 1} C]PBR28 binding in the midbrain, occipital lobe, and cerebellar vermis of CM patients compared with controls. These central indications of glial activation were alterations accompanied increased plasma inflammatory cytokines (IL-8, CX3CL1), reduced thalamic volumes, and an inverse correlation between midbrain TSPO levels and headache frequency, linking glial activation to migraine chronification. Similar findings were reported by Albrecht et al. [189], with increased TSPO-PET signals in nociceptive brain regions (thalamus, insula) in migraine with aura (MA) patients, together highlighting the association between glial activation and migraine frequency., with Tohyama et al. [195] 7T diffusion tensor imaging and TSPO-PET further localize neuroinflammation to the trigeminal nerve root in migraine patients as evidenced by decreased fractional anisotropy and increased [^{1 1} C]PBR28 uptake with reduced fMRI response in spinal trigeminal nucleus [195].

CSF and blood analyses substantiate glial involvement with migraine patients showing elevated CSF CX3CL1 levels during ictal and interictal periods indicating persistent microglial activation, and with increased serum t-Tau during attacks suggesting neuronal stress [193]. Papandreou et al. [192] found significantly elevated serum S100β within three hours of pediatric migraine attacks, though Riesco et al. [188] reported no interictal differences in CM patients, suggesting the potential temporal specificity of glial biomarkers. Kocaturk et al. [198] added to this temporal dimension showing reduced levels of both proinflammatory prostaglandin E2 (PGE2) and anti-inflammatory lipoxin A4 (LXA4) in migraine patients during the ictal and interictal phases, implicating glial dysfunction in pro- and anti-inflammatory signaling imbalance. Systemic inflammation also links to peripheral glial function, especially in the trigeminovascular system. Wijeratne et al. [204] demonstrated differential inflammatory profiles using systemic immune-inflammatory index (SSIIi) in MA versus migraine without aura (MO) patients. MO patients and stable outpatients exhibited persistently higher SSIIis, suggestive of ongoing systemic inflammation, whereas MA patients presented a more transient profile. Similarly, Vuralli et al. [197] reported elevated levels of gut-derived inflammatory markers, such as CGRP, IL-6, and HMGB1, in women with CM and medication-overuse headache (MOH). These proinflammatory molecules are known activators of glial toll-like receptors and may contribute to trigeminal sensitization. Monaghan et al. [205] reported altered cytokine secretion profiles from peripheral blood mononuclear cells (PBMCs) in vestibular migraine patients, marked by decreased TNF-α and IFN-γ and increased ENA-78 (CXCL5) release compared with Menière’s disease controls. Genomic studies have begun to elucidate the molecular underpinnings of glial contributions to migraine as evidenced by associations between migraine susceptibility and gene sets predominantly expressed in astrocytes and oligodendrocytes [190]. Single-cell transcriptomics analysis has found that many migraine-associated genes are enriched in glial cells, particularly astrocytes and microglia, in addition to inhibitory neurons [190]. In line with these observations, an association between IL1A genetic variants and the presence of aura has been described, as well as a TNF-α promoter polymorphism linked to NSAID nonresponsiveness, suggesting that proinflammatory glial cytokine pathways may influence the clinical phenotype and treatment outcomes [201, 202].

To date, the therapeutic targeting of glial pathways yielded mixed results. A randomized crossover trial found no improvement with the glia inhibtor ibudilast in CM patients, possibly reflecting inadequate CNS penetration, suboptimal dosing, or patient heterogeneity [191]. However, Ghorbani et al. [207] and Dahri et al. [208] showed immunomodulatory agents (vitamin D3, CoQ10) to modulate glial-associated cytokines and improve outcomes. Furthermore, lifestyle factors can play a role in migraine inflammation. Bakhshimoghaddam et al. [206] reported an association between pro-inflammatory diets and an increased risk of CM in women. Additionally, the concept of a “leaky gut” has been proposed as a factor contributing to chronic inflammation in CM patients. Vuralli et al. [197] observed elevated levels of lipopolysaccharide (LPS) and lipopolysaccharide-binding protein (LBP), indicative of intestinal permeability. These levels correlated with both attack frequency and endothelial damage markers, such as VE-cadherin and HIF-1α, suggesting that gut barrier dysfunction contributes to systemic inflammation.

In summary, clinical evidence confirms neuroinflammation and glial dysregulation as hallmarks of migraine. Central activation (microglia/astrocytes), peripheral immune dysregulation, and genetic variants converge to sustain sensitization and attack recurrence. Emerging evidence suggests that systemic inflammation, gut-derived metabolites, and sex hormones can modulate glial reactivity in migraine, highlighting the need to study glia within an integrated physiological context.

Targeting astrocytes in migraine pathophysiology

Given their central roles in neuroinflammation and migraine pathogenesis, targeting astrocytes may prove promising for migraine therapy. One strategy would be to increase the activities of astrocytic glutamate transporters, such as EAAT1 and EAAT2, in order to clear excess synaptic glutamate and reduce glutamate-induced excitotoxicity and mitigate neuronal hyperexcitability during migraine attacks [76, 209]. In support of this strategy, ceftriaxone is a β-lactam antibiotic that increases EAAT2 expression and shown to restore glutamate homeostasis and confer neuroprotective effects in preclinical Alzheimer’s models [210, 211]. Another therapeutic strategy would be to target the modulation of astrocytic inflammatory pathways via small organic drugs or monoclonal antibodies that suppress IL-1β, TNF-α, or IL-6 [212]. Further, since astrocytic activation in migraine is linked to NF-κB signaling that drive inflammatory mediator transcription, selective inhibitors targeting these pathways may suppress astrocyte-driven inflammation and alleviate migraine symptoms [38, 213].

Metabolic pathways regulating astrocyte reactivity are also potential therapeutic targets. Miglustat, an FDA-approved glucosylceramide synthase (GCS) inhibitor, has shown benefits in models of chronic neuroinflammation by blocking astrocyte immunometabolic pathways [214–216]. Miglustat suppresses cPLA2-MAVS signaling in astrocytes, reducing CNS inflammation and affecting lactate production, which is important for neuronal metabolism [215]. These findings indicate that other GCS inhibitors may modulate astrocytic inflammation effectively. For example, in the early stages of CSD, astrocytes provide neurons with energy and clear K⁺ and glutamate, but as CSD progresses, astrocytic metabolic responses (including altered glycolysis/lactate dynamics) may influence tissue oxygenation and inflammatory mediator clearance. Evidence linking CSD to impaired perivascular clearance (‘glymphatic’) in migraine-relevant settings remains emerging and warrants cautious interpretation. Another approach would be to target mitochondrial dysfunction in astrocytes, previously shown to be relevant to neurodegenerative diseases like MS, AD, PD, and HD [217]. In Huntington’s disease, for example, low glucose triggers metabolic reprogramming in astrocytes, leading to increased ROS production and neuronal toxicity. Mitochondrion-targeted antioxidants like XJB-5–131 limit ROS-induced damage, suggesting the potential of targeting astrocyte metabolism for neuroprotection [214, 216]. Li et al. [218] also found that low glucose states in astrocytes generate increased inflammatory factors and ROS, which may result from mitochondrial dysfunction. As such, improving mitochondrial function in astrocytes could help mitigate migraine attacks by ensuring a sufficient energy supply.

Targeting microglia in migraine pathophysiology

Activation and transition of microglia to neurotoxic, proinflammatory profiles sustains cytokine release and sensitization, suggesting that modulating these processes may offer therapeutic potential. For example, in migraine models, the P2X7R antagonist Brilliant Blue G prevents neuroinflammation and cognitive deficits, supporting target validity [219]. Antagonism of other microglial receptors implicated in migraine such as TAK to block TLR4 and suppress IL-18 expression [164, 220], and MRS2395 to block P2Y12 and reduce IL-1β and IL-6, are effective in pain models [106]. While these preclinical data are encouraging, clinical confirmation remains necessary.

A complementary approach would be to promote microglia anti-inflammatory phenotypes through enhancing IL-10 or TGF-β signaling. In preclinical settings, electroacupuncture inhibits microglial activation, reduces cytokines, and alleviates facial allodynia and hyperalgesia [172]. Minocycline, a well-known microglial inhibitor, similarly diminishes inflammatory cytokine release and improves pain behaviors [221]. Targeting microglial gene programs also shows promise: inhibiting miR-155-5p with SRT1720 attenuates microglial activation and inflammation [121]. Finally, interrupting neuron–microglia communication via the FKN/CX3CR1 axis may prevent central sensitization [117], while dampening NLRP3 inflammasome activity and IL-18–dependent signaling is predicted to further curb nociceptive amplification [222].

In summary, microglia-directed therapies—receptor antagonism, inflammasome and cytokine modulation, promotion of pro-resolving phenotypes, and disruption of pathogenic neuron–microglia crosstalk—represent promising avenues for migraine. Priorities include rigorous clinical testing, dose optimization, and stratification by patient subtype to translate these mechanisms into durable benefit.

Targeting oligodendrocytes and schwann cells in migraine pathophysiology

Therapeutic strategies aimed at protecting oligodendrocytes or enhancing remyelination could help mitigate migraine-associated neuroinflammation. Potential approaches include the antibiotic minocycline to protect oligodendrocytes from inflammatory damage [223], and remyelination-promoting therapies such as the antihistamine clemastine fumarate, which restores oligodendrocyte function and promotes myelin repair [224]. Concerning migraine pain, while emerging studies implicate oligodendrocytes in chronic pain pathogenesis [133], further research is needed to elucidate a specific role in migraine pain and therapeutic potential.

In the PNS, Schwann cells represent a promising target for migraine intervention through reducing activation and/or inhibiting inflammatory mediator release. For example, stigmasterol, a phytosterol with anti-inflammatory properties, alleviates neuropathic pain by inhibiting Schwann cell-induced IL-34 secretion, macrophage activation, and inflammasome activation. In vivo, stigmasterol reduces thermal and cold hyperalgesia and inflammatory cytokines and suppresses the IL-34/CSF1R and NLRP3 pathways, suggesting its potential as a therapeutic for neuropathic pain [225]. Additionally, targeting neurotrophic factors released by Schwann cells, including NGF and BDNF [226], could help reduce nociceptive amplification associated with migraine. Furthermore, inhibiting TRPA1 and ROS production in Schwann cells represents another promising avenue for alleviating migraine-associated allodynia [227]. While oligodendrocyte protection is a promising strategy in demyelinating diseases, its direct relevance to migraine remains hypothetical and requires validation in migraine-specific models. Although research on both oligodendrocytes and Schwann cells in migraine is still in its early stages, accumulating evidence indicates that their modulation holds significant potential for developing targeted treatments aimed at reducing neuroinflammation and preventing disease progression.

Targeting satellite glial cells in migraine pathophysiology

Given their pivotal role in migraine pathophysiology, targeting SGCs may represent a promising approach for novel therapeutic interventions. Pharmacological modulation of SGC activity, including the inhibition of connexin-mediated gap junctions and the blockade of purinergic receptors (e.g., P2×7 antagonists), has demonstrated efficacy in preclinical models of neuropathic pain and migraine [138]. Anti-inflammatory agents that target cytokine release from SGCs, such as TNF-α inhibitors and IL-1 receptor antagonists, may also attenuate neuroinflammatory processes and reduce migraine severity [69]. In addition, emerging strategies such as microRNA-based interventions could modulate SGC reactivity and restore neuronal homeostasis [65]. Further mechanistic studies are needed to clarify the molecular pathways by which SGCs contribute to migraine, thereby facilitating the development of more effective and targeted therapies. Tetrandrine (TET), an alkaloid extracted from a traditional Chinese medicinal herb, has been shown to exert inhibitory effects on glial activation in vitro and has been investigated for its potential therapeutic applications in various neurological diseases. Preclinical studies demonstrate that TET pretreatment produces a dose-dependent reversal of trigeminal nociceptive hypersensitivity induced by NTG [228]. Moreover, TET administration effectively reduces the activation of S100B and p-ERK in trigeminal ganglion SGCs of NTG-treated rats. Given that reduced p-ERK activity is associated with suppression of inflammatory signaling and decreased hyperexcitability of trigeminal ganglion neurons, these findings suggest that TET may represent a safe and effective therapeutic option for alleviating hyperalgesic symptoms in migraine [228].

Clinical applications

TSPO-PET using the [^{1 1} C] PBR28 radioligand shows elevated putative glial signal in CM and MA and, together with ultra-high-field imaging, has begun to localize inflammatory changes to trigeminal pathways—an important recent advance for in vivo stratification and pharmacodynamic monitoring [189, 194, 195]. CSF CX3CL1 and serum S100β provide complementary, phase-sensitive biomarker candidates, although standardization across ictal/interictal windows remains necessary [192, 193].

Direct clinical testing of glia-modulating therapies remains sparse. A randomized crossover trial of the glial inhibitor ibudilast showed no benefit in CM, highlighting challenges of CNS exposure, dosing, and patient heterogeneity [191]. By contrast, adjunct immunomodulatory interventions (vitamin D3, CoQ10) have been associated with symptomatic improvement alongside shifts in inflammatory profiles, supporting feasibility of targeting neuroimmune tone in patients [207, 208]. Future trials should incorporate target-engagement endpoints (e.g., TSPO-PET signal change, CSF CX3CL1 dynamics) and phenotype stratification to align mechanistic hypotheses with clinically meaningful outcomes [193, 194].

Challenges and future directions

Astrocytes, microglia, and SGCs all engage in bidirectional communication with sensory neurons in the TG and CNS, influencing pain transmission and neuroinflammatory processes [133], albeit with cell-type–specific roles often obscured by the inherent complexity of the system. Indeed, one of the primary challenges in studying glial cells in migraine pathophysiology has been isolating their specific contributions amid extensive interactions with neurons and other glia. Standard experimental models often fail to selectively target glial cells without inadvertently affecting neuronal function, leading to confounding results [17]. Another significant challenge lies in translating preclinical findings into effective clinical treatments. Although animal models provide insights into glial-mediated neuroinflammation, they inadequately replicate human migraine complexity. Pharmacological interventions targeting glial cells, including P2×7 receptor antagonists and connexin inhibitors, have demonstrated preclinical efficacy but subsequently fail clinically [105, 229]. Migraine heterogeneity further complicates treatment development, as different subtypes may involve distinct neuroinflammatory pathways, necessitating personalized approaches [17]. Additionally, the absence of specific biomarkers for glial activation limits monitoring treatment responses in clinical settings [13].

Traditional glial inhibitors lack selectivity and often affect other cellular processes, obscuring glial-specific effects [230]. For instance, ibudilast, a phosphodiesterase inhibitor with glial-modulating properties, was evaluated in a double-blind, randomized, placebo-controlled trial in patients with chronic migraine but failed to demonstrate efficacy in symptom relief or reduction of attack frequency [191]. These findings underscore the need for more precise, glial-specific therapeutic strategies. To address these challenges, future research should prioritize investigations into glial cell-specific mechanisms in migraine. Advanced techniques including single-cell and spatial transcriptomics and proteomics together with optogenetics offer promising avenues for elucidating glial-neuronal interactions in migraine [65]. Recent advances in neuroimaging techniques show promise, with the inflammatory tracer [¹¹C]PBR28 detecting meningeal and parenchymal inflammatory uptake in migraine patients [231]. The PET tracer PBR28 binds to the 18 kDa TSPO in the outer mitochondrial membrane, a specific marker of glial cell inflammation, offering insights into inflammatory signaling in secondary headaches such as post-seizure headaches [232].

Novel therapeutic strategies targeting inflammatory mediators released by SGCs, microglia, and astrocytes, including IL-1β, TNF-α, and ROS, could potentially mitigate neuroinflammation and neuronal hyperexcitability [40]. Furthermore, modulating purinergic signaling through P2×7 receptor antagonists [158] or connexin-based interventions [160] may offer new therapeutic strategies for reducing glial-induced sensitization. Delivery innovations aimed to increase specificity such as Adeno-associated virus (AAV) vectors could selectively deliver gene-based therapies to glia [233]. Although clinical approaches have focused mainly on neurons [233], preclinical studies demonstrate successful modulation of glial activity [234]. Nanoparticles engineered to cross the BBB after systemic administration and deliver siRNAs or mRNAs to astrocytes to regulate protein expression represent another possible route [235], although the potential of glial-specific drug delivery systems requires further research concerning clinically feasible therapies for migraine. Cell replacement therapy has been attracting growing interest as a potential strategy for restoring glial function in neurological diseases [236]. Preclinical data demonstrate that the transplantion of healthy astrocytes has the potential to re-establish homeostasis in the diseased nervous system [237]. In a phase 1/2a amyotrophic lateral sclerosis (ALS) study, human neural progenitor cells that had been transduced with glial cell-derived neurotrophic factor (GDNF) differentiated into astrocytes after lumbar spinal cord transplantation, with graft survival and GDNF production and no adverse motor effects [238]. For migraine, however, the invasiveness of intrathecal administration limits applicability, motivating research on less invasive delivery.

In summary, while traditional glial modulators have been instrumental in clarifying functions and mechanisms, progress in migraine therapies will depend on advancements in glia-specific targets, validated biomarkers, and precise delivery platforms (viral vectors, nanoparticles). Future studies using glia-specific transgenic models, in vivo calcium imaging, and spatial transcriptomics will be essential to dissect the temporal and spatial dynamics of glial activation in migraine. Encouraging preclinical and early clinical signals warrant further studies to refine and translate these strategies to effective migraine therapies.

Conclusion

Glia cells play active roles in driving migraine pathophysiology through a bidirectional CNS–PNS inflammatory network. Astrocytes contribute to this process by linking CSD to meningeal nociception through disrupted glutamate/K⁺ homeostasis, BBB instability, and NF-κB–mediated cytokine release. Microglia further reinforce sensitization via purinergic, NLRP3 inflammasome, and chemokine signaling, although they are also capable of switching to pro-resolving states. Oligodendrocytes support myelin-axonal integrity and facilitate cytokine–trophic crosstalk, while Schwann cells and satellite glia drive peripheral sensitization through CGRP–TRPA1–ROS loops, connexin coupling, and purinergic signaling in trigeminal pathways. Positioning glia as key therapeutic gatekeepers is supported by extensive animal models and clinical evidence, including TSPO-PET glial activation, CSF/serum inflammatory biomarkers, and glia-enriched genetic associations (Fig. 4). Future therapeutic strategies may include: i) enhancing astrocytic EAAT1/EAAT2, ii) inhibiting NF-κB/IL-1β/TNF-α, iii) reversing metabolic stress and ROS/TRPA1/CGRP axes, iv) antagonizing microglial P2X7/4/Y12/TLR4, v) dampening NLRP3, vi) reprogramming glia toward pro-resolving phenotypes and vii) disrupting CX3CR1–BDNF crosstalk. In the PNS, beneficial interventions may include: i) inhibiting SGC gap-junctions, ii) attenuating purinergic signaling, iii) curbing Schwann cell TRPA1/ROS output, and iv) supporting remyelination and trophic balance. Additionally, lifestyle and systemic inflammatory factors, such as gut barrier dysfunction, are modifiable and may synergize with pharmacologic modulation of glial activity (Fig. 4).

Fig. 4.

Targeting glial-specific neuroinflammation in migraine pathophysiology. This review explores glial cell roles in migraine pathogenesis, focusing on cell-type-specific functions in central and peripheral pain mechanisms. Astrocytes and microglia drive cortical spreading depolarization and inflammation, while oligodendrocytes, schwann cells, and satellite glial cells modulate neuronal excitability and nociceptive signaling. The review highlights molecular triggers and signaling pathways of glial activation and connects basic glial biology to clinical strategies. Emerging therapeutic approaches, including nanoparticle delivery and glial-specific targeting, offer new avenues for migraine treatment. Illustrations were created via biorender (http://app.biorender.com)

To optimally progress these areas, it will be crucial to perform clinical trials with precision-endotyped patients using biomarker panels and state-of-the-art neuroimaging while employing adaptive trial designs with pharmacodynamic readouts. Both selective glia-targeted AAV and nanoparticles may enhance on-target delivery and efficacy. Further, single-cell and spatial profiling, optogenetics/chemogenetics, and human tissue models will be essential to confirm target engagement and confirm that disease-modifying therapies restore glial–neuronal homeostasis.

Abbreviations

α7nAChR

α7 nicotinic acetylcholine receptor

5-HT3

5-hydroxytryptamine 3 receptor

ACC

Anterior cingulate cortex

AMPK

AMP-activated protein kinase

ATP

Adenosine triphosphate

BBB

Blood-brain barrier

BDNF

Brain-derived neurotrophic factor

CGRP

Calcitonin gene-related peptide

CH

Cluster headache

CLR/RAMP1

Calcitonin receptor-like receptor/receptor activity modifying protein-1

CM

Chronic migraine

CNS

Central nervous system

CRLR

Calcitonin receptor-like receptor

CSD

Cortical spreading depression

CSF

Cerebrospinal fluid

CX3CR1

C-X3-C-motif chemokine receptor 1

EAAT2

Excitatory amino acid transporters 2

ELISA

Enzyme-linked immunosorbent assay

EM

Episodic migraine

EoA

End of the attack

ERK

Extracellular signal-regulated kinase

EWM

Epilepsy with migraine

FHM1/2

Familial hemiplegic migraine type 1/2

FKN

Fractalkine

GFAP

Glial fibrillary acidic protein

GLP-1 R

Glucagon-like peptide-1 receptor

GLT1

Glutamate transporter 1

Glu

Glutamate

GluR1

Glutamate receptor 1

HIF-1α

Hypoxia-inducible factor-1α

HMGB

High mobility group box

HMGB1

High Mobility Group Box 1

Iba-1

Ionized calcium-binding adapter molecule 1

IDO1

Indoleamine 2,3-dioxygenase 1

IFN

Interferon

IL

Interleukin

IL-1ra

IL-1 receptor antagonist

iNOS

Inducible nitric oxide synthase

IS

Inflammatory soup

JNK

c-Jun N-terminal kinase

LTA

Lymphotoxin alpha

LXA4

Lipoxin A4

MA

Migraine with aura

MiR

MicroRNA

miR-155-5p

MicroRNA-155-5p

MO

Migraine without aura

NF-kB

Nuclear factor kappa B

NLR

Neutrophil-to-lymphocyte ratio

NLRP3

Nucleotide-binding oligomerization domain-like receptor (NLR) family pyrin domain containing 3

NMDA

N-methyl-D-aspartate

NMR

Neutrophil-to-monocyte ratio

NO

Nitric oxide

NDPH

Newly developed persistent headache

NTG

Nitroglycerin

P2XR

Purinergic 2X receptor

P2YR

Purinergic 2Y receptor

PAR2

Protease-activated receptor 2

PBMC

Peripheral blood mononuclear cells

p-ERK

Phospho-extracellular regulated protein kinases

PGE2

Prostaglandin E2

PI3K

Phosphatidylinositol-3-kinase

PT

Post-traumatic headache

RAMP

Receptor activity modifying proteins

ROS

Reactive oxygen species

S1PR1

Sphingosine-1-phosphate receptor 1

SGC

Satellite glial cells

SIRT1

Sirtuin 1

SOD1/2

Superoxide dismutase 1 and 2

TCC

Trigeminocervical complex

TGF-β

Transforming growth factor-beta

TG

Trigeminal ganglion

TH

Thalamus

TNC

Trigeminal nucleus caudalis

TNF

Tumor necrosis factor

TNF-α

Tumor necrosis factor-alpha

TNFR

Tumor necrosis factor receptor

TLR4

Toll-like receptor 4

TLR2/4

Toll-like receptor 2/4

TREM-1

Triggering receptor expressed on myeloid cells 1

TRPA1

Transient receptor potential ankyrin 1

TRPV1

Transient receptor potential cation channel subfamily V member 1

TTH

Tension-type headache

Author contributions

W.L.,Y.Z., and J.T., drafted and conceived the initial manuscript. W.L., C.S.,Y.T., and S.L. drew the figures and arranged the tables. J.T., W.L., and T.S. contributed to the design and critically edited the manuscript. Y.S., F.J. and G.C. critically revised and edited the scientific content. All authors have read and approved the final manuscript.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.