The Animal Care and Use Committee of Chongqing Medical University in China gave its approval to all the experiments of this study, which were carried out in line with the NIH Guide for the Care and Use of Laboratory Animals. We used female and male C57BL/6 mice weighed 18–25 g and aged 6–8 weeks old. The mice were kept under a typical laboratory setting with alternating 12/12-h light/dark cycles at 22 ± 2°C and 50 ± 5% relative humidity. Food and water were easily accessible at any time. All experiments were conducted in a blinded manner. Before conducting any experimental procedures, the animals were given at least a week to acclimate to their surroundings and were randomly divided into different testing groups. All mice were kept in groups (6 per cage) to avoid the stress of social isolation.

A MOH model induced by reiterant SUMA injection was developed based on previous studies (Moye et al., 2019; Saengjaroentham et al., 2020; Zhang J. et al., 2020). Prior to being injected, sumatriptan succinate (MedChemExpress/MCE, USA) was diluted in 0.9% saline. Using 0.9% saline as a vehicle control, animals were administered 0.6 mg/kg SUMA or the equivalent vehicle by intraperitoneal injection (i.p.) once daily for 11 days (Figure 1A).

The microglial inhibitor minocycline (MedChemExpress/MCE, United States) was used to study the role of microglia in MOH. Similar to SUMA, minocycline was administered intraperitoneally after being dissolved in sterile saline at a daily quota of 30 mg/kg for 11 days. The function of P2X7R in MOH was investigated using Brilliant Blue G, a P2X7R specific inhibitor (Sigma Aldrich, Hungary). BBG was injected intraperitoneally at a dosage of 50 mg/kg daily for 11 days after being diluted in 0.9% saline. The NLRP3 inhibitor MCC950 (MedChemExpress/MCE, USA) was used to study the function of NLRP3 in MOH. MCC950 was soluble in 0.9% saline and given to mice at a dosage of 10 mg/kg daily for 11 days. As a vehicle control, an identical volume of solvent was used. The literature and early experiments were used to determine the drug dosage and injection method. On the day of use, all chemical solutions were freshly prepared.

From 9:00 to 15:00, all behavioral assessments were conducted in a calm, dimly lit atmosphere. In the present study, behavioral assessments included periorbital and hind paw mechanical withdrawal thresholds, which also served as markers of central sensitization (Bonnet et al., 2019; Tian et al., 2022). All behavioral tests were performed on the same group of animals. All mice were received training for 3 days prior to the experiment and were tested before injection to establish a reliable baseline response. The mechanical withdrawal threshold was measured by the up-and-down technique with a graded von Frey filament (0.02–2 g) applied to the periorbital or hind paw region. Testing was performed on the hind paw every second day after injections were initiated (Figure 1A). For the periorbital area, testing was performed every 1–3 days after injections were initiated, as excessive facial irritation may cause allergic reactions. The evaluators had no knowledge of the experimental groupings, and there were six mice per group.

As mentioned in earlier research, qRT-PCR was carried out (Zhang Y. et al., 2020). The TNC tissues were swiftly removed and temporarily kept in liquid nitrogen after the mice were given 1% sodium pentobarbital to render them completely comatose. RNAiso Plus reagent (TaKaRa, Dalian, China) was employed to purify total RNA, and a NanoDrop spectrophotometer (Thermo, United States) was utilized to quantify the OD260. Using the PrimeScriptTM RT reagent kits (TaKaRa, Dalian, China), cDNA was created by reverse transcription. Quantitative polymerase chain reaction was conducted on a CFX96 Touch thermal cycler (Bio-Rad) using a SYBR Premix Ex Taq II kit (TaKaRa, Dalian, China). A post-PCR data processing program was used to analyze all fluorescence data, and the 2-ΔΔCT method was applied to calculate the relative change of gene expression. There were 5–6 mice in each group. In Table 1, the primer sequences are presented. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was utilized as a control.

The TNC tissues were promptly removed and kept at −80°C after the mice were given 1% sodium pentobarbital to give them a deep anesthesia. The tissue was homogenized for 1 h at 4°C after the addition of radioimmunoprecipitation assay buffer (Beyotime, Shanghai, China) including a protease inhibitor and phenylmethylsulfonyl fluoride (PMSF). The protein concentrations were detected by BCA Protein Assay Kits (Beyotime, Shanghai, China). Equal amounts of protein (20–40 g) were electrotransferred to PVDF membranes (Millipore, United States) after being segregated on 10% or 12% SDS-PAGE gels (Beyotime, Shanghai, China). After being blocked with 5% skimmed milk at room temperature for 2 hours, the membranes were first mixed with primary antibodies (Table 2) for an overnight period at 4°C. The following day, after being rinsed in TBST, the membranes were coated with secondary antibodies (goat-anti-rabbit and goat-anti-mouse, ZSGBIO, China) for 1 hour at room temperature. With the aid of ECL assay kits (Advansta Inc., United States) and an imaging system (Fusion, Germany), immunoreactive bands were visualized and analyzed. There were six mice in each group.

The animals were given 1% sodium pentobarbital to induce a profound anesthesia before receiving 60 mL of precooled phosphate-buffered saline (PBS) and 60 mL of precooled 4% paraformaldehyde transcardially (PFA). The cervical spinal cord and brainstem were taken, covered with 4% PFA, and left to fix at 4°C for 24 h. The tissue was then further dehydrated in sucrose solutions of 20 and 30% until it sank. After being embedded, the tissue was quickly frozen. In accordance with the mouse brain atlas (Pinskiy et al., 2013), TNC tissues were divided into 10–25 μm sections. The slices were repaired with sodium citrate, infiltrated for 10 min with 0.3% Triton X-100 (Beyotime, Shanghai, China), and then sealed for 30 min with 5% goat serum (Boston, Wuhan, China) at 37°C. After that, the slices were reacted with primary antibodies (Table 2) overnight at 4°C. The next day, after being rinsed in PBS, the slices were coated with the appropriate secondary antibodies (Alexa Fluor cy3 or 488) at 37°C for 60 min. The slices were treated with a combination of primary or secondary antibodies from two distinct sources for double immunofluorescence analysis. 4′,6-diamidino-2-phenylindole (DAPI) was utilized to stain the nuclei at 37°C for 10 min. A confocal microscope (LSM800, Zeiss, Germany) was employed to capture images. All images were taken under uniform parameters (100 × objective), and the fluorescence signal intensity was quantified using image analysis software (Image J 1.8.0). The fluorescence intensity of CGRP was calculated using the mean optical density (OD) of CGRP (Zhang Y. et al., 2020). The quantity of Iba1, c-fos immunopositive cells was also counted by ImageJ. Four mice each group, with 4–5 slices per mouse, were assessed. An experimenter who was not aware of the experimental groupings collected and analyzed the images.

For statistical analysis and graph creation, GraphPad Prism version 8.0 (GraphPad Software Inc., San Diego, CA, United States) was applied. The Kolmogorov–Smirnov test was employed to determine the data’s normality before statistical analysis. All data are expressed as the mean ± standard error (SEM). In addition to behavioral data, data from two groups were compared using Student’s t tests. One-way analysis of variance (ANOVA) and Tukey or Dunnett post hoc tests were utilized for multiple comparisons. For the examination of behavioral data, two-way ANOVA and Bonferroni post hoc tests were employed. Statistics were judged significant when p < 0.05.

The female and male mice were intraperitoneally injected with SUMA for 11 consecutive days to establish the MOH mouse model (SUMA group), and the control group (VEH group) was intraperitoneally injected with sterile saline for 11 consecutive days (Figure 1A). Repeated injections of SUMA over the course of 11 days in both male and female mice induced a gradual decline in periorbital and hind paws withdrawal thresholds (p < 0.05, Figures 1B–E). However, the VEH group did not exhibit any appreciable variation. To avoid sex differences, we used only male mice in subsequent studies to avoid possible hormonal effects. c-Fos and CGRP were chosen as the central sensitization indicators based on prior research (Latremoliere and Woolf, 2009; Iyengar et al., 2017). Double immunofluorescence labeling of c-Fos and CGRP in the TNC is shown in Figure 1F. Immunofluorescence staining revealed that the SUMA group had higher levels of c-Fos and CGRP expression than the VEH group did in the TNC (p < 0.0001; Figures 1G–J). Consistent with earlier research, these results demonstrated that repeated SUMA injection causes central sensitization in mice.

To observe changes in the activation of microglia in the TNC after repeated SUMA injection, we assessed the microglial marker Iba1 and the M1 (proinflammatory) microglial marker iNOS (Wang et al., 2021; Armeli et al., 2022). We analyzed the protein expression of Iba1 and iNOS by western blotting. Compared with that in the VEH group, the protein expression of Iba1 and iNOS in the SUMA group steadily raised with repeated injections of SUMA (p < 0.05; Figures 2A,C,D). Consistent with the western blot results, elevated mRNA levels of iNOS were observed in the SUMA group, as determined by qRT–PCR (p < 0.05; Figure 2E). The observed TNC area is shown as the dashed line in Figure 2B. Immunofluorescence staining revealed a spike in the number of Iba1 immunopositive cells after repeated SUMA injection (p < 0.0001, Figures 2F,G) and a shift from the resting (dendritic) to the activated state in microglia in the SUMA group as indicated by fewer protrusions (Figure 2F, rightmost column). The number of branches, average branch length and maximum branch length of microglia were analyzed, but there was no significant difference between the VEH and SUMA groups (p = 0.2939, p = 0.8445, p = 0.5892 respectively, Figures 2H–J) (Young and Morrison, 2018). According to these results, repeated SUMA injection activated microglia in the TNC.

Prior to the daily treatment of SUMA, a microglial inhibitor, minocycline (MINO), was injected intraperitoneally, which resulted in the inhibition of microglial activation. Iba1 and iNOS protein expression reduced dramatically in the SUMA+MINO group compared with the SUMA group (p < 0.01 and p < 0.05; Figures 3A–C), and the SUMA+MINO group also showed a significant decrease in iNOS mRNA levels (p < 0.05; Figure 3D). Similarly, a reduced number of Iba1 immunopositive cells was observed in the SUMA+MINO group, as shown by the immunofluorescence staining images (p < 0.0001, Figures 3E,F). The number of branches, average branch length and maximum branch length of microglia did not differ significantly between groups (Figure 3F, rightmost column). After microglial activation was inhibited, the reduction in mechanical pain threshold caused by repeated SUMA injection was ameliorated (p < 0.0001, Figures 4A,B), and c-Fos and CGRP expression was reduced in the TNC (p < 0.0001, Figures 4C–F). These findings suggest that inhibiting microglial activation can reduce central sensitization and hyperalgesia in MOH.

Previous research has demonstrated that P2X7R is predominantly located in microglia (Jiang et al., 2021). Therefore, in this study, we further examined the function of microglial P2X7R in MOH mice. Double immunofluorescence labeling was used to examine the relationships between P2X7R and different cellular type-specific indicators in the TNC, including Iba-1 for microglia, GFAP for astrocytes and NeuN for neurons. Consistent with previous findings, our results showed that P2X7R colocalized primarily with Iba1 (shown by white arrows in Figure 5A) but not with NeuN or GFAP. The protein and mRNA expression levels of P2X7R were subsequently observed by western blotting and qRT–PCR. After repeated SUMA injection, P2X7R protein expression increased in a time-dependent way (p < 0.01, Figures 5B,C). Likewise, the mRNA expression in the SUMA group was much higher than that in the VEH group (p < 0.001, Figure 5D). Collectively, these findings demonstrated that repeated SUMA injection caused a progressively growth in microglial P2X7R in the TNC.

It has been demonstrated in earlier studies that the NLRP3 inflammasome, which is downstream of P2X7R, is implicated in the pathophysiology of a variety of disorders (Territo and Zarrinmayeh, 2021; Li et al., 2022; Richter et al., 2022). Thus, we further investigated the changes in the NLRP3 inflammasome after repeated SUMA injection in this study. Our data demonstrated that following repeated SUMA injection, the protein expression of NLRP3, Caspase-1, and IL-1β elevated gradually over time in the TNC (p < 0.0001, p < 0.01, p < 0.05, respectively; Figures 6A–D), and the mRNA expression of NLRP3 in the MOH group raised dramatically compared with that in the VEH group (p < 0.0001, Figure 6E). These results suggested that repeated SUMA injections activated the NLRP3 inflammasome in the TNC.

To further determine the role of P2X7R, a P2X7R antagonist BBG was administered intraperitoneally to the mice before each SUMA injection. We discovered that BBG treatment greatly reduced mechanical hyperalgesia induced by SUMA in the periorbital and hind paws regions (Figures 7A,B). In comparison with the SUMA group, the SUMA+BBG group had lower P2X7R protein expression after BBG treatment (Figures 7C,F). Consistent with the decline in P2X7R, BBG treatment markedly reduced the protein expression of NLRP3, Caspase-1, and IL-1β (Figures 7C,G–I). In addition, the mRNA expression of P2X7R and NLRP3 was dramatically cut down (Figures 7D,E). It was also shown that the SUMA+BBG group and the SUMA group significantly differed in the expression of c-Fos and CGRP (Figures 7J–M). Based on these observations, inhibiting P2X7R could prevent hyperalgesia and central sensitization in MOH mice, and this effect was mediated by a reduction in NLRP3 inflammasome activation.

To inhibit the activity of the NLRP3 inflammasome, the NLRP3-specific inhibitor MCC950 was injected intraperitoneally prior to the daily administration of SUMA. The mechanical hyperalgesia caused by SUMA in the periorbital and hind paws was significantly suppressed by MCC950 (p < 0.05; Figures 8A,B). The protein levels of NLRP3, caspase-1, and IL-1β, as well as the mRNA levels of NLRP3, were markedly decreased by MCC950 injection (p < 0.001; Figures 8C–G). Moreover, MCC950 significantly restored the expression of c-Fos and CGRP in the TNC induced by repeated SUMA injection (p < 0.0001; Figures 8H–K). These findings revealed that NLRP3 inhibition could reduce hyperalgesia and central sensitization in MOH mice.