Involvement of Interleukin-18-Mediated Microglia/Astrocyte Interaction in Experimental Models of Migraine

Background The exact molecular mechanisms of migraine were not fully understood. Emerging evidence indicated that inflammation had a significant role in pathophysiological mechanisms of migraine. The present study aimed to investigate the role of Interleukin-18-Mediated Microglia/Astrocyte Interaction in the development of hyperpathia or allodynia induced by migraine. Methods Experimental rat model of migraine was established by repetitive inflammatory soup (IS) dural infusions. The expression of Interleukin-18 (IL-18) and IL-18 receptor (IL-18R) was examined by quantitative real-time polymerase chain reaction or Western blot. The expression of activated microglia and astrocytes was examined by Western blot or immunofluorescence. The expression of nuclear factor-kappa B (NF-κB) was examined by Western blot. TAK-242 and anti-IL-18 antibody were used to inhibit the activation of toll-like receptor 4 (TLR4) and IL-18 pathway, respectively. Results IS dural infusions elicited microglial activation and astrocytic activation. Meanwhile, IS dural infusions induced the upregulation of IL-18 and IL-18R in microglia and astrocytes, respectively. Blocking the IL-18 signaling pathway attenuated nociceptive behavior. Meanwhile, blocking IL-18 signaling also suppressed astrocytic activation, as well as activation of NF-κB. And IL-18 dural infusions induced nociceptive behavior and glia activation.


Abstract Background
The exact molecular mechanisms of migraine were not fully understood. Emerging evidence indicated that inflammation had a significant role in pathophysiological mechanisms of migraine. The present study aimed to investigate the role of Interleukin- 18

Introduction
Migraine, a severe and disabling neurologic disorder with quite high morbidity (approximately 15%), has a great detrimental influence on society and individuals. It is characterized by multiple attacks of throbbing and often unilateral severe head pain, usually accompanied with nausea, vomiting, photophobia or phonophobia, and aggravated by physical activity or head movement [1] . Despite decades of clinical and scientific research, the exact molecular mechanisms of migraine were not fully understood. Recently, emerging evidence indicates that inflammation has a significant role in pathophysiological mechanisms of migraine. [2,3] . Glial cells are critical in the inflammation process.
When activated, various substances, such as proinflammatory cytokines, were released. They amplify the nociceptive signals and take part in the sensitization process, leading to enhanced responsiveness of nervous tissue. In this process, glial cells, like microglia and astrocytes, were found in close proximity to each other, facilitating cell-cell communication through ligand-receptor pairings elevated in target tissues [4][5][6] .
However, there has been no study investigating the role of IL-18 signaling pathways in migraine models.
In this work, we show that IL-18 is upregulated in microglia of migraine rat brain, while it's receptor (IL-18R) upregulated in astrocytes. Here, we provide first evidence for the involvement of IL-18 signaling pathways mediating microglia/astrocyte interaction in development and maintenance of migraine.

Materials And Methods
Animals.
Male Sprague-Dawley (SD) rats weighting 200-250 g were used. All rats were housed under a 12 h light/dark cycle at 22 ± 2 °C, with free access to food and water. They were habituated in experimental environment for one week before each experimental procedure. All experiments were approved by the Committee on Animal Use for Research and Education of the Laboratory Animals Center, Renmin Hospital of Wuhan University (Wuhan, PR China), and followed the recommendations of the International Association for the Study of Pain in Conscious Animals [15] . The number and suffering of animals were minimized as much as possible.
Surgical procedure.
The experimental procedures were carried on according to a previous study [16] . Briefly, a midline incision was performed under pentobarbital anesthesia (60 mg/kg; i.p). A guide cannula was implanted in a previously drilled cranial window adjacent to the superior sagittal sinus, to delivery IS or PBS to the dura mater in all rats. We take special care to prevent infection and to minimize the effect of inflammation.

Repetitive dural infusions.
Rats were divided into two groups randomly: the control (CON) group and the migraine model (IS) group. The migraine models were built via repeated IS (30 ul, including 2 mM histamine, 2 mM serotonin, 2 mM bradykinin, and 0.2 mM prostaglandin E2 in PBS) [16] dural infusions for at least 4 days. Rats in the control group received the same dose of PBS.

Drug administration.
A total of 0.8 ml anti-IL-18 antibody (100 ng/µl; R & D Systems) was first administered intraperitoneally 1.5 h prior to the each dural IS infusion [IS + Anti-IL-18(i.p) group]. Then, the intracerebroventricular injection of 5 ul anti-IL-18 antibody was conducted as described previously [17] . Briefly, anti-IL-18 antibody (100 ng/µl; R & D Systems) was injected into the lateral ventricle (-1.0 mm rear from the Bregma and + 1.5 mm lateral, 4.0 mm from the skull plane) 24 h after the 4th IS infusion[IS + Anti-IL-18(i.c.v) group], under adequate anesthesia. TAK (3 mg/kg, Millipore, Bedford, MA), a specific TLR4 antagonist, was administered intraperitoneally 1.5 h prior to the dural IS infusion based on the previous study (TAK group). The 30 µl of IL-18 (5 µg; R & D Systems) was delivered to the dura mater as described above (IL-18 group). TAK was diluted in PBS.
The anti-IL-18 antibody and IL-18 were both dissolved in PBS. Rat in the control group received the equivalent volume of PBS. The dosages of above drugs were determined according to previous studies. Random grouping was used.

Nociceptive behavior
The headache behavior of rats was evaluated by a nociception threshold and face rubbing. A von Frey filament (vFF) was used when measuring the facial mechanical withdraw threshold, as described previously [16] . The nociceptive threshold was taken as the least force applied to make rats quickly retract its head at least three of five stimulus. Face rubbing was a nociceptive behavior when rats received IS. In process testing, the experimenter was blinded to the grouping.

Western blot analysis
Tissue from medullary dorsal horn homogenized in lysis buffer (consisting of RIPA, cocktail PMSF and phosphorylase inhibitor). Then, protein concentrations were measured using a BCA Protein Assay Kit (Beyotime). After heated in loading buffer for 10 min at 100℃, equal amounts of protein were loaded and separated on 10% SDS PAGE gels (Bio-Rad), and subsequently transferred to PVDF membranes (Millipore). After blocked with 5% nonfat milk for 90 min at RT, the membrane was in incubated in primary antibody overnight at 4 °C. The dilution concentration of the primary antibody was applied as follows: IL-18 (1:1000; Abcam), IL-18R (1:50; Abcam), GFAP (1:2500; Gene Tex), p65(1:3000, Abcam), p-p65 (1:50; Santa Cruz) and NAPDH (1:1500, servicebio). The membrane then was washed three times in TBST and incubated with respective secondary antibodies for 60 min at RT. The antibodyreactive bands were visualized using Odyssey CLx Image Studio 3. The gray-scale value of straps was quantified using Image J software (Version1.52, National Institutes of Health). All Western blot analysis were conducted at least three times, and consistent results were obtained.

Statistical analysis
All data are presented as the mean ± SEM. GraphPad Prism 8 (GraphPad Software, San Diego, CA) were used to perform statistical analysis and to generate the graphs. T-test was used to evaluate differences between two independent groups. P < 0.05 was considered significant.

IS dural infusions induced the activation of microglia and astrocyte in the medullary dorsal horn
Glial fibrillary acidic protein (GFAP, astrocytic marker) and ionized calcium binding adapter molecule 1 (Iba1, microglial marker) were examined by immune-fluorescence. Rats in IS group showed higher expression of GFAP (Fig. 2B, D) and Iba1 ( Fig. 2A, C) in the medullary dorsal horn than that in control group.

IS dural infusions induced the upregulation of IL-18/IL-18R in the medullary dorsal horn
The q-PCR results revealed a significantly increase of the IL-18 mRNA level in the medullary dorsal horn of IS group (Fig. 3A). Next, the expression of IL-18 protein was detected by Western blot analysis. Consistent with changes in IL-18 mRNA levels, IL-18 protein was upregulated in the medullary dorsal horn of IS group (Fig. 3B, C). Then, IL-18R upregulation after IS dural infusions was also confirmed by Western blot analysis (Fig. 3B, D).

IL-18 activated in microglia, IL-18R activated in astrocyte
To identify the cell types expressing IL-18 and IL-18R after IS dural infusions, we performed double Immuno-fluorescence staining of the IL-18 or IL-18R with Iba1 and GFAP in the medullary dorsal horn of IS group. Results showed that IL-18 immunoreactivity was predominantly double-labeled with Iba1, but not with GFAP (Fig. 4A). Moreover, IL-18R immunoreactivity was predominantly double-labeled with GFAP, but not with Iba1 (Fig. 4B), suggesting that IS dural infusions induced IL-18 upregulation in medullary microglia, while IL-18R upregulation in medullary astrocyte.
IS dural infusions induced microglia-derived IL-18 upregulation via TLR4/p38 MAPK in the medullary dorsal horn To examine whether TLR4 regulates IL-18 expression in microglia of medulla oblongata after IS dural infusions, we injected TAK (the TLR inhibitor). The result showed that microglia-derived IL-18 induced by IS dural infusions was significantly alleviated by blocking TLR4 (Fig. 5A). Moreover, we found p-p38 entries into nucleus of IL18-immunopositive cells (Fig. 5B), indicating that IS dural infusions induce microglia-derived IL-18 upregulation through TLR4/p38 MAPK pathway in the medullary dorsal horn.

IL-18 dural infusions produced nociceptive behavior
It was revealed that IL-18 occupied a key position in neuropathic pain, cancer pain and morphine tolerance [22,11−14] . To further examine the impact of IL-18 on headache behavior, we conducted repeated IL-18 dural infusions. After the 4th repetitive dural infusions of IL-18, the number of face rubbing in five hours and the facial mechanical withdraw threshold were recorded. We found that IL-IL-18 dural infusions induced the activation of microglia and astrocyte in the medullary dorsal horn Rats in IL-18 group showed higher expression of GFAP (Fig. 7B, D) and Iba1 (Fig. 7A, C) by immunofluorescence in the medullary dorsal horn than that in control group. Furthermore, the results of Western blot analysis verified the GFAP protein upregulation in the IL-18 group (Fig. 7E).

IL-18 blockade attenuated nociceptive behavior induced by repeated IS dural infusions
Next, we tested the effect of anti-IL-18 antibody on the development of IS-induced cutaneous allodynia and hyperalgesia. The injection of anti-IL-18 antibody once daily for 4d with the first injection 1.5 h before IS dural infusion significantly inhibited the development of face rubbing (Fig. 8A) and tactile allodynia (Fig. 8B). And, intrathecal injection of anti-IL-18 antibody after the 4th IS infusion reduced face rubbing induced by repeated IS dural infusions (Fig. 8C), but failed to obviously reverse the tactile allodynia induced by IS dural infusions (Fig. 8D).

Microglia-derived IL-18 mediated astrocytic activation via a NF-κB-dependent manner
A previous study revealed the TLR4-NF-κB signaling pathway an important role in the progress of hyperalgesia following IS dural infusions [21] . In this work, IS dural infusions induced an increase of NF-κB phosphorylation in the medullary dorsal horn (Fig. 9C). Next, to investigate the effect of IL-18 on NF-κB p65 activation, we compared the level of p-NF-κB in the control and anti-IL-18 group. Western blot analysis showed that IL-18 blockade significantly inhibited NF-κB p65 activation induced by IS dural infusions (Fig. 9D), accompanied with the inhibition of astrocytic activation (Fig. 9B).

Discussion
Accumulating evidence supported the glial cell is critical in pain hypersensitivity induced by migraine.
Glial cells were activated in the status of experimental migraine, and subsequently released various proinflammatory cytokines like IL-1β and TNF-α, causing hyperpathia or allodynia [2,21] .
The present study revealed microglia-derived IL-18 signaling a vital role in experimental migraine ( Figure.1). First, IS dural infusions elicited the activation of microglia, followed by increments of IL-18 via TLR4-derived p38 MAPK pathway. Secondly, IL-18 induced nociceptive behavior and activation of microglia and astrocyte. Then, microglia-derived IL-18 acted on IL-18R predominantly expressed in astrocytes, increasing NF-κB phosphorylation, subsequently activating astrocytes. Blockage of IL-18 attenuated nociceptive behavior induced by repetitive IS dural infusions and inhibited the activation of NF-κB phosphorylation and astrocyte. IL-18 might activate microglia and astrocytes in an autocrine or paracrine manner, then synthesized and released pro-inflammatory cytokines, neurotransmitters and neuromodulators, amplifying neuronal excitability [4,5,11,14] and inducing migraine attacks. It indicated a new mechanism underlying experimental migraine that IL-18 signaling mediating microglia-astrocyte interactions, contributing to hyperpathia or allodynia induced by IS dural infusions.
Consistent with previous studies [2,21] , repetitive IS dural infusions induced the activation of glial cells. Activated microglia and astrocytes released various substances, including pro-inflammatory cytokines like IL-1β, IL-6 and TNF-α, subsequently activating the trigemino-vascular system and evoking hyperalgesia and allodynia [21,23,24] . Presently we found that IL-18, a member of the IL-1 family, was upregulated by IS dural infusions. And, double-labelling immunofluorescence indicated microglia is the predominant cell expressing IL-18, similar to previous studies. Multiple studies confirmed that IL-18 has a vital role in the neuropathic pain, cancer pain and morphine tolerance [22,11−14] . However, there was lack of evidence suggested the role of the IL-18 signaling pathway in experimental migraine. Toll-like receptor 4 (TLR4) plays a crucial role in the innate immunity, and is expressed in microglia [19,20] . A previous study reported that TLR4 had a critical role in experimental migraine [21] . Intriguingly, Miyoshi and co-workers found that nerve injury induced the expression of IL-18 in spinal microglia through TLR4/p38MAPK pathway [11] . In a similar vein, the present study hypothesized the molecular events that microglia-derived IL-18 was regulated via a TLR4/p38 MAPK-dependent manner. Blockage of TLR4 attenuated the upregulation of IL-18 induced by IS dural infusions. Furthermore, double-labelling immunofluorescence revealed p-p38 entries into nucleus of IL-18-immunopositive cells, confirming the above hypothesis. IL-18 bound to IL-18R to play a biological role. In line with prior studies [11,14] , we found that elevation of IL-18R induced by IS dural infusions were primarily expressed in astrocytes. IL-18/IL-18R signaling pathway mediated microglia-astrocyte interactions and hence participated in the development of hyperpathia or allodynia induced by IS dural infusions. We pretreated rats with anti-IL-18 antibody to block the IL-18 signaling before each IS dural infusion. Blockage of IL-18 significantly inhibited face rubbing and tactile allodynia produced by IS dural infusions. And, the increments of astroglial markers caused by IS dural infusion was similarly reduced. Inversely, IL-18 dural infusions produced both headache behaviors and astrocytic activation. Such findings indicated that experimental migraine induced the synthesis of IL-18 from microglia through the activation of TLR4/p38 MAPK pathway. Microglia-derived IL-18 stimulated IL-18R expressed on neighboring astrocytes, triggering astrocytic activation.
We subsequently attempted to explore the molecular mechanisms under IL-18/IL-18R mediated astrocytic activation. Previous studies revealed that NF-κB p65 activation was involved in the experimental migraine and targeting NF-κB p65-driven transcription or downstream expression, which provided a potential therapeutic strategy for migraine [21,25,26] . In the present study, IS dural infusions induced an increase of NF-κB phosphorylation in the medullary dorsal horn. Moreover, activation of IL-18 signaling in neuropathic pain promoted the phosphorylation of NF-κB in the astrocytes and activated those cells [11] . So, we detected the effects of neutralizing IL-18 activity on NF-κB phosphorylation. The results shown that suppression of IL-18 bioactivity significantly inhibited the phosphorylation of NF-κB, accompanied by the downregulation of astrocytes, indicating a further establishing linkage between IL-18, NF-κB and astrocytic activation.
Attentionally, we chose peripheral administration intervention to suppress IL-18 bioactivity before each IS dural infusion, by considering the particularity of our animal models. Western blot analysis verified the blockade of IL-18 in the medullary dorsal horn, which could be explained by blood-brain barrier (BBB) breakdown due to repeated IS dural stimulation [27] . Glial activation occupied a key position in the BBB modulation [28,29] . In present works, microglia and astrocyte were activated by IS dural infusion, providing further evidence for the above. It is noteworthy that IL-18 might also play a role through peripheral mechanisms, similar to role of calcitonin gene-related peptide (CGRP) in Migraine [30] . Then posttreatment with anti-IL-18 antibody could not obviously reverse the hyperpathia or allodynia induced by IS dural infusions, which might be explained by that only one dose of anti-IL-18 antibody after successful modeling could not effectively block the IL-18 bioactivity.

Conclusion
The present study revealed a new molecular mechanism for migraine. IL-18-mediated microglia/astrocyte interactions in the medullary dorsal horn may participate in the development of