Xenon ameliorates chronic post-surgical pain by regulating mitophagy in microglia and rats mediated by PINK1/Parkin pathway

SMIR model construction and Xe treatment

Eighty healthy adult male Sprague-Dawley (SD) rats, 6–8 weeks, 230–300 g, were purchased from Shanghai Laboratory Animal Research Center. Afterwards, they were reared in an alternating day and night environment with free food and water. All animal-related experiments were reviewed and approved by the Fudan University Shanghai Cancer Center Ethics Committee (FUSCC-IACUC-S2022-0346).

The SMIR surgery was performed as previously reported (Pan et al., 2019), shown in Fig. S1. In brief, 60 SD rats were anesthetized with 40 mg/kg 1% sodium phenobarbital (Sinopharm, China) intraperitoneally, and then fixed on the operating table in a supine position. The inner thigh of one side was shaved and then disinfected with alcohol to expose the saphenous vein. A skin incision of 1.5–2 cm is made about four mm medial to the saphenous vein in the middle of the thigh to expose the muscle. A 7–10 mm incision is made in the superficial muscle four mm medial to the saphenous nerve. The superficial muscles were bluntly separated with scissors, and a retractor was placed to retract the skin and superficial muscles to two cm, and exposed the underlying adductor fascia. After 1 h, the skin and muscles were sutured. During the stretch, the rats were given sterile saline gauze to cover the incision for moisturizing and keeping warm.

SMIR Rats were divided into four groups, six rats in each group. In the sham operation group, the rest of the operation procedures were the same except that the skin and muscles were not stretched. Xe group rats received 50% Xe (Wuhan Newread Special Gas Co., Ltd, China) inhalation once for 1 h. CPSP group rats received SMIR surgery. CPSP + Xe group received SMIR operation and 50% Xe inhalation for 1 h, once a day for 5 consecutive days.

Behavioral experiment

Thermal withdrawal latency (TWL) detection

The basal value was detected 1 day before surgery using a paw heat pain tester (Shanghai Yuyan Scientific Instrument Co Ltd., Shanghai, China). TWL was detected on the injured side and the unaffected side within 0–5 h and 2–14 d after Xe treatment based on a design by Hargreaves et al. (1988). The plantar stimulation analgesia was used for detection, the temperature-regulated glass platform was kept at 29 °C, and the animals were acclimated on the platform for 10–15 min. A radiant heat source was placed on the hind sole, and the thermal contraction latency was recorded. Thermal paw withdrawal latency, the time it takes for a rat to remove its hind paw from a heat source, is used to assess thermal hyperalgesia. Left and right paws were tested at random intervals of 1 min. The two posterior grasps were tested 3 times, with a 5-minute interval between the two tests. Data is expressed in seconds.

ELISA kit test

The spinal cord dorsal horn tissue of each group of animals was collected 14 days after operation, and the supernatant was taken after the tissue was homogenized. The oxidative stress indicators malondialdehyde (MDA) superoxide dismutase (SOD), hydrogen peroxide (H₂O₂) and catalase (CAT) in each group were detected according to the instructions of the ELISA kit (Shanghai Enzyme-linked Biotechnology Co., Ltd., Shanghai, China). The specific operations are as follows: Add sample/standard and incubate at 37 °C for 1 h. After washing three times, add enzyme conjugate at 37 °C for 30 min. After washing three times, add chromogenic solution and incubate at 37 °C for 30 min. The reaction was terminated by adding stop solution. Then the absorbance value was detected at 450 nm.

Western blot

Rats were anesthetized with 40 mg/kg 1% sodium phenobarbital intraperitoneally. The spinal cord tissue was quickly removed, lysed with RIPA lysis buffer (Bio-Rad Laboratories, Inc., Hercules, CA, USA) containing 1mM phenylmethanesulfonyl fluoride (PMSF, Thermo Fisher Scientific, Waltham, MA, USA), and homogenized by ultrasound on ice. RIPA lysis buffer was added to cell sample and lysed on ice for 30 min. Centrifuge at 12,000 rpm for 15 min at 4 °C, and take the supernatant. Protein electrophoresis was performed after protein quantification by BCA kit (Thermo Fisher Scientific, Waltham, MA, USA). The total protein was separated by SDS-PAGE gel (MCE, USA.) and transferred to PVDF membrane (Thermo Fisher Scientific) Then the PVDF membrane was blocked with 5% w/v BSA (Sigma-Aldrich, St. Louis, MO, USA) for 1 h at room temperature. After washing three times, add anti-Iba1 antibody (1:1000, Abcam, Cambridge, UK), anti-PINK1 antibody (1:1000, Abcam, Cambridge, UK), anti-Parkin antibody (1:1000, Abcam, Cambridge, UK), anti-ATG5 antibody (1:1000, Abcam, Cambridge, UK), anti-BECN1 antibody (1:1000, Abcam, Cambridge, UK), anti-actin antibody (1:1000, Abcam, Cambridge, UK) and incubate at 4 °C overnight. After washing 3 times, add horseradish peroxidase (HRP)-labeled secondary antibody (Abcam, Cambridge, UK) and incubate at room temperature for 1 h. Finally, add ECL solution (Thermo Fisher Scientific) to exposure for 1∼3 min. Protein bands were analyzed using Image J and optical density was detected. β-actin was used as an internal reference.

Immunofluorescence histochemical staining

Rats were anesthetized with 40 mg/kg 1% sodium phenobarbital intraperitoneally. After laparotomy, normal saline was rapidly perfused through the aorta, followed by 4% paraformaldehyde (Solarbio, Beijing, China) for 30 min. Spinal cords were removed and fixed in 4% paraformaldehyde for 3 h, then transferred to 30% sucrose (Sigma-Aldrich, USA) for 2 days of dehydration. The slices were cut to 25 µm thick under freezing conditions. Spinal cord sections were rinsed 3 times with PBS (GIBCO, Invitrogen Corp, Carlsbad, CA, USA), 5 min each time, and blocked for 1 h at room temperature. Anti-Iba1 antibody (Abcam, Cambridge, UK) was added, overnight at 4 °C. Remove the anti-Iba1 antibody, rinse with PBS for 3 times, add fluorescently labeled secondary antibody (Abcam, Cambridge, UK) for 1 h at room temperature in the dark. The sections were rinsed 3 times with PBS and observed under a fluorescence microscope and photographed.

Microglia extraction

The extraction of microglia refers to the method of Pacheco (Pacheco et al., 2020), and the specific operations are as follows: Take three 1–2 days old neonatal SD rats, quickly cut the skull under sterile conditions, take out the cerebral hemisphere tissue into a petri dish containing HBSS (GIBCO, Invitrogen Corp, Carlsbad, CA, USA) + 10% fetal bovine serum (FBS, GIBCO, Invitrogen Corp, Carlsbad, CA, USA) solution. Then carefully peel off the meninges under the microscope. The brain tissue was cut into small pieces, placed in a centrifuge tube containing HBSS+10% FBS solution, and placed on ice for 1 h. The brain tissue was washed 3 times with three mL of HBSS + 10% FBS solution and HBSS in turn. Add three mL of 0.25% trypsin (GIBCO, Invitrogen Corp, Carlsbad, CA, USA) solution and place in a water bath at 37 °C for 30 min. The brain tissue was washed 3 times with three mL of HBSS + 10% FBS solution and HBSS successively. Add seven mL of HBSS and centrifuge at 450 × g for 5 min at 4 °C. Discard the supernatant and add four mL of HBSS + 10% FBS to resuspend the pellet. Transfer the cells to a T75 culture flask, add 10 mL DMEM (Gibco, Waltham, MA, USA) with 10% FBS and continue to culture in a 5% CO₂, 37 °C cell incubator. On third and fifth day, the medium was replaced with DMEM medium containing 10% FBS. By around day 14, microglia were confluent for purification. Fix the culture flask on a constant temperature shaker at 37 °C and shake (180 rpm) for 1–2 h. The medium containing microglia was collected, centrifuged at 4 °C (800 rpm, 8 min), the supernatant was discarded, and the cells were resuspended in DMEM medium containing 10% FBS to continue the culture.

Construction of CPSP microglia cell model

The CPSP microglia cell model was constructed by collecting the cerebrospinal fluid of CPSP rats and co-cultured with primary microglia for 24 h.

The cells were divided into four groups, the cells in the control group were primary microglia cells, the cells in the Xe group were treated with Xe for 1 h, the cells in the CPSP group were CPSP microglia, and the cells in the CPSP + Xe group were CPSP microglia treated with Xe for 1 h.

Mito-SOX fluorescent staining

Add MitoSOX™ Red (1:1000; Invitrogen, Waltham, MA, USA) to cells at 37 °C for 30 min. After washing with PBS for 3 times, observed with a fluorescence microscope and photographed (OLYMPUS, Tokyo, Japan).

Co-localization of GFP-LC3 and Mito Tracker fluorescent staining

The cells were added with Mito-Tracker Red CMXRos (1:1000, Invitrogen) 37 °C for 30 min. After washing with PBS for three times, add anti-LC3 antibody overnight at 4 °C. Then add fluorescently labeled secondary antibody (Abcam), for 1 h at room temperature in the dark. The cells were rinsed three times with PBS and observed under fluorescence microscope and photographed.

SiRNA design, transfection and screening

Two double-stranded siRNA targeting rat PINK1 mRNA (5′-TGAAGAGACTCAGGCGCTA-3′, 5′-GCGAGGTGGTGAAGAGACT-3′) were designed using siRNA Target Finder and Design software (http://www.genelink.com/sirna/RNAicustomorder.asp) and synthesized by Gemma Gene Company. The control group was Scramble siRNA with no homology.

According to the instructions, Lipofectamine 2000 (Invitrogen) was used to transfect different PINK1 siRNAs into microglia. The transfection experiment was carried out when the cells had grown to more than 80%. First, 20 µL of siRNA was mixed with 250 µL of Opti-DMEM (Invitrogen), and then placed at room temperature for 5 min. Then 10 µL of Lipofectamine 2000 was mixed with 250 µL of Opti-DMEM, and place at room temperature for 5 min. The two solutions were mixed well and left at room temperature for 15 min. Add the mixture to the cell plate and shake the cell plate gently. After transferring the cell plate to the incubator for 6–8 h, replace with new DMEM medium and continue to cultivate for 72 h.

The effectiveness of siRNA was detected by Western blot. Validated PINK1 siRNA were selected for subsequent in vivo experiments.

Statistical analysis

SPSS 16.0 (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 8.3.0 (GraphPad Software, La Jolla, CA, USA) statistical analysis software were used for data analysis and processing of experimental results. Measurement data were expressed as mean ± standard deviation (mean ± SD). The independent sample t-test was used to analyze the data between the two groups, and the one-way ANOVA test was used to analyze the pairwise data of three or more groups. P < 0.05 was considered statistically significant for differences.

Xe improved PWMT and TWL in CPSP rat model

The effect of Xe on pain improvement following SMIR surgery was tested in the rat CPSP model (Fig. S1). The rats were established by SMIR surgery, and then Xe was inhaled for 1 h (50%) immediately, for 5 consecutive days. The PWMT and TWL were measured at BL (baseline), 0, 1, 2, 3, 4, 5 h (short-term) and day 2, 4, 6, 8, 10, 12 and 14 (long-term) after CPSP using the von Frey test and paw heat pain tester. The specific timeline of the individual experiment was shown in Fig. 1A. Behavioral testing showed that PWMT was significantly decreased after CPSP compared with the sham control. After xenon treatment, PWMT increased, but there was no significant difference compared with CPSP group 2 h after xenon treatment (Fig. 1B). Meanwhile, TWL was also significantly increased after CPSP, but no statistical difference after 5 h with xenon treatment compared with CPSP group (Fig. 1C). This result demonstrated that the Xe may be effective in relieving pain caused by SMIR in the short-term at a single treatment. PWMT and TWL were further measured at days 2, 4, 6, 8, 10, 12, and 14. Figures 1D and 1E shown that PWMT and TWL were all significantly increased at these time points compared with CPSP group, revealed that Xe also may be effective in relieving pain caused by SMIR in the long-term at multiple treatments.

Xe repressed oxidative stress and microglial activation in the dorsal horn of rat CPSP model

Considering that oxidative stress plays a critical role in the peripheral nervous system changes in both acute inflammatory pain and chronic neuropathic pain (Shim et al., 2019), we investigated whether oxidative stress was also activated after CPSP and whether Xe treatment was associated with oxidative stress. Spinal cord dorsal horn tissues of each group were collected 14 days after SMIR surgery and Xe treatment, and oxidative stress-related markers (MDA/SOD/H₂O₂/CAT) were detected by ELISA kits. The levels of MDA and H₂O₂ level were significantly increased in the CPSP group, accompanied by a significant decrease in SOD/CAT levels, while Xe inhalation significantly reduced CPSP-induced oxidative stress changes (Figs. 2A–2D). These results demonstrated that oxidative stress was observably activated after CPSP, which was further effectively inhibited by Xe treatment.

Microglia are a type of glial cells found throughout the brain and spinal cord (Yan et al., 2022). Microglial activation changes in morphology and phenotype occur after cerebral ischemic injury, and the severity of cerebral ischemic injury is related to the activation state of microglia (Var et al., 2021). To investigate whether Xe inhibited the activation of microglia in CPSP rats, microglia after Xe treatment were analyzed by Western Blot and immunofluorescence analysis. The Western blot analysis results (Figs. 2G and 2H) indicated that the protein expression level of Iba1 (microglial marker) in the spinal dorsal horn of rats induced by SMIR was significantly up-regulated, while its protein expression was observably down-regulated after Xe treatment. Moreover, Xe inhibited SMIR-induced microglial activation as measured by the intensive ramified Iba1-positive staining (Figs. 2E and 2F). These results suggested that Xe has an important inhibitory effect on oxidative stress and microglial activation in the rat CPSP model.

Xe promoted microglia mitophagy and inhibited oxidative stress in vitro

To explore the role of Xe in activating microglia in vitro, we studied the viability of rat primary microglia incubated with CPSP rat cerebrospinal fluid in the presence or absence of Xe treatment for 0.5 h. Firstly, the content of Mito-SOX, a specific marker of mitochondrial reactive oxygen species, was detected by Mito-SOX fluorescent probe under inverted fluorescence microscope in each group. Compared with the control group, the Mito-SOX positive signal (red) of microglia in the CPSP model cells was significantly increased. However, the positive signal of Mito-SOX was significantly reduced after Xe treatment (Figs. 3A and 3B). Then, co-localization (yellow) of GFP-LC3 (green) and Mito Tracker (red) was used to detect the occurrence of mitophagy. The staining results showed that compared with the control group, the CPSP group was significantly increased the co-localization; It is worth noting that Xe treatment further significantly enhanced CPSP-induced the co-localization (Fig. 4B). Finally, the expression of autophagy-related protein PINK1/Parkin/ATG5/BECN1 was detected by Western blot analysis. The results showed that compared with the control group, the protein expression levels of PINK1/Parkin/ATG5/BECN1 in the CPSP group were significantly increased, and Xe treatment also significantly increased the expression of autophagy-related proteins (Fig. 3C). These results demonstrated that Xe treatment promotes microglia autophagy and may mitigate mitochondrial dysfunction, thereby reducing oxidative stress in microglia.

Xe regulated mitophagy by activating the PINK1/Parkin pathway

Studies have shown that PINK1-mediated mitophagy has been implicated in neuropathic pain (Dionisio et al., 2019). Therefore, we investigated whether Xe induced activation of PINK1/Parkin signaling pathway in a rat CPSP cell model. Firstly, the level of PINK1 in Iba1-positive microglia in the CPSP group was detected by immunofluorescence, and it was found that the level of PINK1 (red) in Iba1-positive cells (green) in the CPSP group was significantly higher than that in the control group, whereas Xe treatment significantly increased PINK1 levels (Figs. 4A and 4B). This result demonstrated that Xe activates the PINK1/Parkin pathway. In addition, we constructed siRNA (siPINK1-1 and siPINK1-2) to knockdown PINK1 in CPSP model cells. The effect of siRNA transfection was verified by Western blot analysis, and siPINK1-1 with the most obvious effect was used as a research tool for subsequent experiments (Figs. S2A and S2B). Then, the effect of Xe on the mitochondrial oxidative stress marker Mito-SOX after knockdown of PINK1 in CPSP model cells was detected. The staining results showed that the positive signal of Mito-SOX was significantly reduced after Xe treatment, while the positive signal was further increased after knockdown of PINK1 (Figs. 4C and 4D). Finally, the co-localization of GFP-LC3 and Mito Tracker was reduced in CPSP model cells after Xe treatment, whereas it was significantly increased after PINK1 knockdown (Figs. 4E and 4F). The above results suggested that Xe can promote microglia mitophagy and inhibited oxidative stress in CPSP by activating the PINK1/Parkin pathway.

Inhibition of PINK1 disrupted the improvement of PWMT and TWL in the rat CPSP model with Xe treatment

The effect of Xe on the pain behavior of CPSP rats after Pink1 knockdown was further studied at the animal level. Intrathecal injection of siPINK1 interfered with lentivirus 48 h before CPSP was used to study the mechanism of Xe therapy (Fig. 5A). Immunofluorescence analysis results (Figs. 5B and 5C) indicated that siPINK1 promoted microglial activation that was inhibited by Xe treatment in the spinal dorsal horn of CPSP rats as measured by the intensive ramified Iba1-positive staining. In addition, the protein expression level of Iba1 was also up-regulated in in the spinal dorsal horn of CPSP rats with both siPINK1 injection and Xe treatment (Figs. 5D and 5E). PWMT was significantly reduced after siPINK1 injection, but no statistical difference after 2 h compared with CPSP group after Xe treatment (Fig. 5F). Meanwhile, TWL was also significantly reduced after siPINK1 injection, but no statistical difference after 5 h compared with CPSP group after Xe treatment (Fig. 5G). PWMT and TWL were further measured at days 2, 4, 6, 8, 10, 12, and 14 (Figs. 5H and 5I). As shown that PWMT and TWL were all significantly reduced at these time points compared with CPSP group after Xe treatment. These results suggested that siPINK1 disrupted the improvement of PWMT and TWL in the rat CPSP model with Xe treatment.