Propentofylline Relieves Incisional Hyperalgesia by Regulating ERK1/2 Phosphorylation in Spinal Astrocytes and Microglia

Summary

This study investigates the effect of intrathecally administered propentofylline on attenuating incisional hyperalgesia and whether this effect is associated with changes in spinal glial markers and in the phosphorylation of extracellular signal-regulated kinase 1 and 2 in rats.

Abstract

Spinal astrocytes, microglia and extracellular signal-regulated kinase 1 and 2 (ERK1/2) have been implicated in pain processing. This study investigated whether preoperative intrathecal propentofylline (PPF) attenuates acute incisional hyperalgesia in rats and whether the effect is associated with spinal glial markers and ERK1/2 phosphorylation. Sixty-six male Sprague–Dawley rats were randomly assigned to blank, incisional pain, normal saline, dimethyl sulphoxide, PPF and U0126 groups. Mechanical and thermal hyperalgesia were assessed at baseline and 2, 4, 8, 24 and 72 h after incision. Spinal neuronal and glial markers, p-ERK1/2 and inflammatory mediators were evaluated by Western blotting and immunofluorescence. Primary spinal astrocyte experiments were additionally used to examine ERK1/2-dependent inflammatory mediator release. Preoperative intrathecal PPF attenuated mechanical and thermal hyperalgesia throughout the 72-h observation period, whereas U0126 attenuated hyperalgesia mainly during the early post-incision period. At 4 h after incision, both treatments were associated with reduced spinal GFAP, Iba-1 and p-ERK1/2 expression and reduced TNF-α and COX-2 immunoreactivity. In cultured astrocytes, PPF reduced LPS-induced inflammatory mediator release, whereas constitutively active ERK1/2 attenuated this effect. These findings support an association between the antihyperalgesic effect of PPF and modulation of spinal glial, ERK1/2-related and inflammatory responses after incision.

Introduction

Incisional pain, a common type of acute pain following surgical procedures, generally does not persist longer than 7 days but has complex pathological mechanisms that differ from those of inflammatory and neuropathic pain^1,2. If incisional pain is poorly controlled during its development, it may further progress to chronic pain syndrome, and its nature may shift from acute tissue injury pain to neuropathic or mixed pain³. Generally, the administration of analgesics, such as opioids and non-steroidal anti-inflammatory drugs, is the most common form of analgesia following incision. However, many side effects may occur with the continuous or high-dose use of these, a circumstance that highlights the need for safer and more effective analgesic strategies.

Reliable evidence indicates that propentofylline (PPF) relieves hyperalgesia across different pain processes^4,5. Furthermore, PPF’s analgesic effect can be partly achieved by inhibiting the spinal expression of mitogen-activated protein kinases (MAPKs), such as p38 and c-Jun N-terminal kinases (JNKs)⁶. Although PPF’s effects on p38 and JNK pathways have been established, its potential modulation of the extracellular signal-regulated kinase 1 and 2 (ERK1/2) pathway, another critical MAPK subfamily implicated in pain signaling, remains unexplored in acute incisional pain. This knowledge gap is particularly important given that ERK1/2 activation shows distinct temporal and cellular patterns compared with p38 and JNK, suggesting potentially unique therapeutic opportunities.

Glial cells, including astrocytes, microglia, and oligodendrocytes, were initially thought to be the connecting and supporting structure of nervous Tissue. However, recent studies have shown that astrocytes and microglia play an important role in regulating synaptic plasticity and transmitting pain information⁷. As the resident macrophages of the brain and spinal cord, microglia can be activated by several extracellular stimulus signals in the very early stage of the pain process and then proliferate and release large amounts of inflammatory cytokines, which increase the conduction of pain signals from the spinal cord to the brain^8,9. Concurrently, activated astrocytes can upregulate the release of the excitatory neurotransmitter glutamate and send information directly to sensory neurons by enhancing intercellular communication¹⁰.

Studies have shown that astrocyte activation occurs in the early stage (within the first 24 h) of incision-induced pain, whereas microglial activation appears much later^11,12. Additionally, one study indicated that activated microglia have a limited effect on mechanical hyperalgesia in incision-induced pain¹³. These findings highlight ongoing questions about the distinct roles of astrocytes and microglia in acute pain. Propentofylline, a glial modulating agent, has been reported to relieve hyperalgesia by regulating the activation of astrocytes and microglia following chronic nerve-constrictive injury, acute nerve injury, and acute Tissue injury^4,5. However, to the best of current knowledge, no study has examined PPF’s effects on astrocyte or microglial activation in acute incision-induced pain.

Extracellular signal-regulated kinase 1 and 2, a major MAPK subfamily activated downstream of mitogen-activated protein kinase 1 and 2 (MEK1/2), have been widely implicated in nociceptive processing and neuronal plasticity^14,15. Previous studies have shown that phospho-ERK1/2 (p-ERK1/2) can be induced in spinal neurons by nociceptive stimulation and can also be detected in astrocytes and microglia in inflammatory or neuropathic pain states^16,17,18,19. In addition, the pharmacological inhibition of MEK1/2–ERK1/2 signaling has been reported to attenuate mechanical and thermal hyperalgesia in several pain models^{20,21,22,23,24,25}.

However, the cellular distribution of spinal ERK1/2, particularly in astrocytes and microglia, activated by an acute incision and its role in incision-induced pain remains insufficiently defined. Previous studies by the research group indicate that PPF relieves acute incisional hyperalgesia in rats by reducing spinal JNK and p38 signaling, but whether PPF also modulates ERK1/2-related signaling in spinal glial cells during incision-induced pain remains unclear. Thus, the present study should be regarded as an extension of the existing glial and MAPK frameworks in pain research rather than a completely independent mechanistic paradigm. Its main distinction from prior work lies in its focus specifically on acute incisional pain, its examination of ERK1/2 rather than p38/JNK as the MAPK component of interest, and its evaluation of the early spinal cellular distribution of p-ERK1/2 in relation to PPF treatment.

This study was designed to investigate whether preoperative intrathecally administered PPF attenuates incisional hyperalgesia in a rat plantar incision model and whether this effect is associated with ERK1/2 phosphorylation or changes in the expression of spinal neuronal and glial markers and inflammatory mediators, tumor necrosis factor-α (TNF-α) and cyclooxygenase-2 (COX-2). More specifically, this study differs from prior reports in its combination of behavioral assessment with temporal Western blot analysis; its pharmacological comparison with the MEK1/2 inhibitor U0126; and its qualitative immunofluorescence localization of p-ERK1/2 in spinal neurons, astrocytes, and microglia within an acute incisional pain model. Although its mechanistic framework builds on previous literature linking glial modulation and MAPK signaling to pain, the present work is intended to provide additional evidence for the possible involvement of spinal ERK1/2-related glial responses in the early phase of incisional hyperalgesia.

Compared with conventional postoperative analgesics, pathway-targeting approaches can test a defined signaling component but may provide only time-limited benefit and can require invasive delivery; MEK1/2–ERK1/2 inhibition has shown antihyperalgesic effects in several experimental pain models^{20,21,22,23,24,25}. PPF differs from a single-pathway inhibitor because it modulates glial responses and has been linked to more than one MAPK-associated process^4,5,6. Practical translation remains limited by the preclinical design, preoperative intrathecal administration, use of male rats only, and the absence of dose-ranging and safety studies in larger surgical models.

Protocol

The experiments conducted were approved by the Ethics Committee of the Second Affiliated Hospital, Lanzhou University, Gansu Province, China (ethics approval no. 2017-070). All procedures followed the International Association for the Study of Pain guidelines²⁶. A total of 66 adult male Sprague–Dawley rats obtained from an accredited laboratory animal centre and weighing 200–250 g were used. Only male rats were included in this study to reduce potential variability related to sex-dependent hormonal influences. All protocols were performed in accordance with the International Association for the Study of Pain guidelines²⁶, and the necessary efforts to minimize suffering were taken. Detailed information on the reagents, consumables, instruments, and software used in this study is provided in the Table of Materials.

Animals

The rats were housed in pairs before operation and individually after incision with food and water ad libitum. The room was maintained in a 12-h light–12-h dark cycle and temperature-controlled at 24 °C ± 2 °C.

Intrathecal injection and incisional pain model

For the intrathecal administration of solutions, rats were anesthetized with 1.5%–3% sevoflurane (delivered through a nose cone) and were placed in the prone position with the vertebral column flexed. The fur over the lumbar region was shaved, and the skin over the L5–L6 intervertebral space was disinfected. A microsyringe was inserted percutaneously at the L5–L6 intervertebral space. Successful entry into the subarachnoid space was indicated by a sudden tail flick or tail movement during needle advancement and was further confirmed by the appearance of cerebrospinal fluid upon gentle aspiration. If these signs were not observed, the needle position was adjusted before injection. Injection was performed only after both confirmation criteria had been satisfied, and animals showing abnormal neurological signs after injection were excluded. The respective solution was then injected slowly, and the skin was cleaned again after injection. The surgical procedure lasted approximately 3 min, and the rats generally recovered from anesthesia within 5–10 min.

The rat model of incisional pain was established as previously described²⁷. After anesthesia, the plantar surface of the hind paw was disinfected with 10% povidone–iodine. A 1-cm longitudinal incision was made in the plantar skin, beginning approximately 0.5 cm from the heel and extending towards the toes. The incision penetrated the skin and fascia. The plantar muscle was gently elevated with ophthalmic forceps and incised longitudinally while preserving overall tissue continuity. The muscle was then repositioned, gentle pressure was applied to achieve hemostasis, and the wound was closed with two mattress sutures using 3-0 nylon. The incision site was disinfected again after wound closure. Following surgery, the rats were housed individually and allowed to recover in a quiet, warm environment, protected from strong light and with free access to food and water. The animals were monitored postoperatively, and those not meeting the predefined criteria were excluded promptly. Wound infection was determined based on redness, swelling, and increased local temperature around the wound, abnormal yellow-green purulent exudate with malodour, delayed wound healing with persistent wound opening, necrosis and enlargement, and systemic signs including elevated body temperature, lethargy, and refusal to feed. Additional exclusion criteria included a failure to recover from anesthesia within 30 min, intraoperative muscle transection or neurovascular injury, postoperative self-mutilation, unsuccessful model establishment, and neurological deficits, such as hind limb paralysis or an inability to bear weight that was unrelated to incisional pain. Rats with wound infection or dehiscence were excluded from the study.

Behavioral measurement

Paw withdrawal mechanical threshold (PWMT) and paw withdrawal thermal latency (PWTL) were measured by an investigator blinded to group allocation. Testing was performed in a quiet room at a consistent time of day after at least 10 min of acclimatisation; the enclosure, mesh or glass floor, ambient light, and heat-source intensity were kept constant across groups. For PWMT testing, calibrated von Frey filaments (0.4 g, 0.6 g, 1.0 g, 1.4 g, 2.0 g, 4.0 g, 6.0 g, 8.0 g, 10.0 g, and 15.0 g) were applied perpendicular to the plantar surface adjacent to the wound until the filament bent by approximately 1–2 mm. Consecutive stimulations were separated by 2 min, and the 50% PWMT was determined using the up-and-down method²⁸. For PWTL testing, rats were placed in individual transparent enclosures on a glass floor, and the radiant heat source was aligned with the plantar surface beside the wound. A 30 s cut-off prevented tissue injury, consecutive exposures were separated by 2 min, and the mean of three measurements was recorded²⁹. Paw lifting, shaking, or licking was considered a positive withdrawal response; trials affected by grooming, locomotion, or loss of heat-source alignment were repeated after reacclimatization.

Western blot analysis

At the preset times, rats were euthanized by cervical dislocation, and the L4–L6 lumbar spinal cord was rapidly harvested into radioimmunoprecipitation assay lysis buffer supplemented with protease inhibitor at a ratio of 100:1. Tissue was homogenised, kept on ice for 20 min, and centrifuged at 15,000 × g for 20 min at 4 °C. The supernatant was mixed with 4× loading buffer at a ratio of 3:1, heated in boiling water for 5 min, and stored at −80 °C. Protein samples (30 µg per lane) were separated on 10% sodium dodecyl sulfate–polyacrylamide gels and transferred to 0.45 or 0.22 µm polyvinylidene fluoride membranes. Successful transfer was checked by confirming the expected migration of the prestained molecular-weight marker and the absence of obvious uneven or interrupted transfer regions. Membranes were blocked with 5% skim milk and 3% bovine serum albumin for 2 h at room temperature and incubated overnight at 4 °C with primary antibodies against NeuN (1:8,000), GFAP (1:4,000), Iba-1 (1:3,000), p-ERK1/2 (1:1,000), t-ERK1/2 (1:3,000), or GAPDH (1:2,000). After washing in Tris-buffered saline containing 0.05% Tween 20 (pH 7.4), membranes were incubated with the corresponding horseradish peroxidase-conjugated secondary antibody (1:5,000) for 1 h at room temperature. Signals were developed with enhanced chemiluminescence and recorded on film using the same exposure range within each target. An investigator blinded to group allocation analyzed band density with image-analysis software. Images were converted to 8-bit grayscale, an identical rectangular region of interest was applied to each band, local background from an adjacent blank region was subtracted, and the integrated density was recorded. The same region size and background procedure were used for all lanes on a blot. NeuN, GFAP, Iba-1, and t-ERK1/2 were normalized to GAPDH; p-ERK1/2 was normalized to t-ERK1/2; and values were then expressed relative to the blank group.

Immunofluorescence

At the preset times, rats were anesthetized with 3% pentobarbital sodium (1.5 mL/kg) and perfused through the heart with 300 mL of precooled 0.01 mol/L phosphate-buffered saline (PBS; pH 7.4), followed by 250 mL of precooled 4% paraformaldehyde. The L4–L6 spinal cord was post-fixed in 4% paraformaldehyde for 6 h, cryoprotected overnight in 20%–30% sucrose–PBS at 4 °C and cut into 10 µm sections at −22 °C. Sections were washed in 0.01 mol/L PBS, blocked in 10% goat serum for 1 h at room temperature and incubated overnight at 4 °C with primary antibodies against NeuN (1:100), GFAP (1:1,000), Iba-1 (1:100), p-ERK1/2 (1:100), TNF-α (1:200) or COX-2 (1:50). After three PBS washes, sections were incubated with the corresponding green- or red-fluorophore-conjugated secondary antibody (1:400) for 1 h at room temperature, washed three further times in PBS and mounted with antifade medium and a coverslip. Negative-control sections processed without the primary antibody were included in each staining batch. Images were acquired by a blinded investigator with a fluorescence microscope. Low-magnification overview images were acquired using a 10× objective, whereas cellular immunoreactivity and co-localization images were acquired using 20× or 40× objectives according to the corresponding figure panel. For comparisons within the same figure and staining batch, objective magnification, illumination intensity, exposure time, detector gain, camera settings, and image-processing parameters were fixed after optimization below saturation. Separate fluorescence channels were acquired sequentially to minimize channel overlap. Three non-overlapping fields in the ipsilateral dorsal horn were captured per animal. Fluorescence intensity was quantified with image-analysis software after 8-bit conversion, background subtraction using a cell-free region, and application of an identical region of interest and threshold within each staining batch; the field values were averaged to obtain one value per animal. Fields with overexposure, tissue folding, tearing, poor focus, or uneven background were excluded and reacquired under the same non-saturating settings. Staining was accepted when tissue architecture was intact, the target signal exceeded the no-primary-antibody control, and no major fold, tear, or saturated region affected the analysis.

In vitro primary spinal astrocyte experiments

Primary rat spinal astrocytes were cultured under standard sterile conditions in complete culture medium at 37 °C in a humidified 5% CO₂ incubator. Cells were used when they reached approximately 70%–80% confluence and were assigned to control, LPS-stimulated, LPS + PPF, LPS + U0126, and ERK1/2 rescue groups. PPF and U0126 were added at 10 µM. LPS stimulation, drug treatment, and harvest conditions were kept identical across parallel wells within each experiment. For pathway-selectivity analysis, cell lysates were collected for Western blotting of p-ERK1/2, t-ERK1/2, p-p38, total p38, p-JNK, total JNK, and the corresponding loading control. Culture supernatants from parallel wells were collected, clarified by centrifugation, and analyzed for TNF-α, PGE2, and IL-6 using enzyme-linked immunosorbent assay according to the manufacturer’s instructions. Western blot densitometry followed the same image-analysis principles described above. For rescue experiments, endogenous ERK1/2 was suppressed by siRNA transfection, and a constitutively active ERK1/2 (CA-ERK1/2) expression construct was introduced before treatment. Knockdown and overexpression efficiency were confirmed by Western blotting before cytokine data were interpreted. Cells exposed to the transfection reagent without active siRNA or construct served as procedural controls when applicable.

Experimental design

Subgrouping of experimental animals

The rats were assigned to six groups using a random number table method, and group allocation was completed before behavioral testing and tissue collection. The six groups comprised a blank group (n = 6), an incisional pain (IP) group (n = 18), a normal saline (NS) group (0.9% NS, 10 µL; n = 9), a PPF group (10 µg in 10 µL; n = 12), a dimethyl sulfoxide (DMSO) group (10% DMSO, 10 µL; n = 9) and a U0126 group (10 µg in 10 µL; n = 12). Rats in the blank group received neither a plantar incision nor an intrathecal injection and served as untreated controls. Rats in the IP group underwent plantar incision without pretreatment by intrathecal injection. Rats in the NS, PPF, DMSO, and U0126 groups received intrathecal administration of their respective agents 30 min before plantar incision. The NS group served as the vehicle control for PPF, whereas the DMSO group served as the vehicle control for U0126.

Doses of propentofylline and U0126

Propentofylline is a glial modulator, and a dose of 10 µg was selected because it had relieved acute hyperalgesia in earlier studies by the research group. U0126 is a non-adenosine triphosphate-competitive MEK1/2 inhibitor used to block ERK1/2 phosphorylation. Effective intrathecal doses of 1–10 µg have been reported across pain models^30,31,32; the 10 µg dose was used in the present study. U0126 was dissolved in 10% dimethyl sulfoxide, whereas PPF was dissolved in 0.9% normal saline. Final concentrations were 1 µg/µL. Drugs and vehicles were prepared before surgery, maintained at 4 °C, and used within 12 h.

Effect of the preoperative intrathecal administration of propentofylline or U0126 on hyperalgesia

Six rats from each group were randomly drawn for behavioral measurement 24 h before the operation, and the resultant PWMTs and PWTLs were designated as baseline. Intrathecal injections were administered 30 min before the operation. To assess the effects of the preoperative intrathecal administration of PPF and U0126 on mechanical and thermal hyperalgesia, PWMT and PWTL were assessed at 2 h, 4 h, 8 h, 24 h, and 72 h post-incision.

Experimental design for evaluating the effects of preoperative intrathecally administered propentofylline or U0126 on spinal neuronal and glial marker expression following incision

To determine the appropriate post-incision timepoint for treatment–effect assessment, a preliminary time course analysis was first performed in the IP group. Specifically, rats in the IP group were euthanized at 2 h, 4 h, 24 h, and 72 h after incision, and the expression levels of the cell markers NeuN, GFAP, and Iba-1 in the spinal cord were measured by Western blot (n = 3 per timepoint). Based on this preliminary analysis, the first 4 h post-incision were selected as an early post-incision observation window because NeuN expression reached its highest level at the 4 h timepoint and because early molecular changes were considered most relevant to the initial development of incisional hyperalgesia. However, the 4 h timepoint was not intended to represent all temporal molecular changes after incision.

At this selected timepoint, rats that had been administered an intrathecal injection were euthanized for Western blot analysis of different cell markers (n = 3 per group) and immunofluorescence evaluation of the morphology of different cell types (n = 3 per group). In the blank group, rats were randomly terminated to determine the basic expression levels of different cell markers (NeuN, GFAP, and Iba-1) in the spinal cord by Western blot (n = 3) and the morphology of different cells (neurons, astrocytes, and microglia) by immunofluorescence (n = 3). Thus, the blank group served as the non-incised, non-treated baseline control.

Treatment effects on spinal p-ERK1/2 and inflammatory mediators

To determine the appropriate post-incision timepoint for evaluating treatment effects on p-ERK1/2 expression and levels of inflammatory cytokines, a preliminary time course analysis was first performed in the IP group. Specifically, rats in the IP group were euthanized at 2 h, 4 h, 24 h and 72 h after incision, and spinal p-ERK1/2 expression was measured by Western blot (n = 3 per timepoint). Based on this time course experiment, together with the timepoint at which NeuN expression was highest after incision, 4 h after incision was selected as the fixed timepoint for the subsequent evaluation of the effects of preoperative intrathecally administered PPF or U0126 on spinal p-ERK1/2 expression and inflammatory cytokines. This timepoint was selected as an early mechanistic observation window rather than as a comprehensive representation of all temporal changes after incision.

At this selected timepoint, rats that had been administered an intrathecal injection were euthanized for Western blot analysis of p-ERK1/2 expression (n = 3 per group) and immunofluorescence evaluation of inflammatory cytokines (n = 3 per group). In the blank group, rats were randomly terminated to determine the basic expression levels of p-ERK1/2 in the spinal cord by Western blot (n = 3) and of inflammatory cytokines (TNF-α and COX-2) by immunofluorescence (n = 3). Thus, the blank group served as the non-incised, non-treated baseline control.

Cellular localization of spinal p-ERK1/2 in neurons, astrocytes and microglia

To further assess the cellular distribution of spinal p-ERK1/2 after incision, immunofluorescence staining was performed in the Blank and IP groups at 4 h after incision (n = 3).

Statistical analysis

Data are expressed as mean ± standard deviation. Sample size was determined with reference to earlier studies by the research group and preliminary experiments. Behavioral data were analyzed by two-way repeated-measures analysis of variance (ANOVA), with group as the between-subject factor and time as the within-subject factor. Time-course Western blot data were evaluated by one-way ANOVA followed by Bonferroni-adjusted pairwise comparisons. Fixed-time Western blot and immunofluorescence data were evaluated by one-way ANOVA with Bonferroni-adjusted planned comparisons; two-group comparisons were evaluated by Student’s t-test. Normality and homogeneity of variance were assessed with the Shapiro–Wilk and Levene tests, respectively. A two-sided p-value < 0.05 was considered significant. Statistical analyses were performed with the statistical analysis software listed in the Table of Materials. For the in vitro astrocyte experiments, independent culture experiments were treated as the unit of analysis; Western blot densitometry and cytokine concentrations were compared using one-way ANOVA followed by Bonferroni-adjusted planned comparisons or Student’s t-test for prespecified two-group comparisons, as appropriate.

DATA AVAILABILITY:

Data supporting the findings of this study are provided in Supplementary File 1.

Results

The effects of incision and of the preoperative intrathecal administration of propentofylline or U0126 on hyperalgesia after incision

Mechanical hyperalgesia

There were no significant differences in PWMTs between the six groups at 24 h before operation (p > 0.05; Figure 1A,B). In the IP group, the PWMT decreased after incision and remained significantly lower than baseline from 2 h ...

Discussion

Most patients suffer from acute pain after surgery, not only physically but even at times mentally³³, yet much is still unknown about the detailed mechanisms of acute postoperative pain. Along with the powerful effects of analgesic medicines come their inevitable side effects, making the exploration of safer, effective analgesics of critical importance. This study was based on a previously validated incisional rat model for simulating acute postoperative pain that was easy and feasible to prepare<...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
0.01 M PBS buffer	Solarbio	P1020
1.0 M Tris-HCl buffer (pH 6.8)	Solarbio	T1020
1.5 M Tris-HCl buffer (pH 8.8)	Solarbio	T1010
10% sodium dodecyl sulfate (SDS)	Solarbio	S1010
10× TBST stock solution	Solarbio	T1081
30% acrylamide	Solarbio	A1010
4× protein loading buffer	Solarbio	P1041
Adobe Photoshop	Adobe	—
Ammonium persulfate	Solarbio	A1030
Automatic flake ice machine	Changshu Xueke	IMS-40
Bovine serum albumin (BSA)	Solarbio	A8020
Cryostat CM150	Leica	CM150
Digital pH meter	—	PHS-3C
Dimethyl sulfoxide (DMSO)	Solarbio	D8371
Electronic analytical balance	Cany	—
Enhanced chemiluminescence (ECL) reagent	Solarbio	PE0010
Film	Kodak	—
Fluorescence microscope DP71	Olympus	DP71
Frozen section embedding medium	Sakura	4583
Gel electrophoresis apparatus	BIO-RAD	—
Glycerol	Solarbio	IG0910
Glycine	Solarbio	G8200
Goat anti-mouse Alexa Fluor 594 secondary antibody	Proteintech	SA00013-3
Goat anti-mouse horseradish peroxidase-conjugated secondary antibody	ZSGB-BIO	ZB2305
Goat anti-rabbit Alexa Fluor 488 secondary antibody	Proteintech	SA00013-2
Goat anti-rabbit horseradish peroxidase-conjugated secondary antibody	ZSGB-BIO	ZB2301
Goat serum	Solarbio	SL038
GraphPad Prism	GraphPad Software	5
High-speed refrigerated centrifuge 3K15	Sigma	3K15
Horizontal orbital shaker TS-1000	Haimen Qilinbeier	TS-1000
ImageJ	National Institutes of Health	1.51j8
Methanol	Tianjin Guangfu	—
Mouse anti-rat GAPDH primary antibody	Proteintech	60004-1-Ig
Mouse anti-rat p-ERK1/2 primary antibody	CST	9101S
Na2HPO3	Tianjin Beichen Fangzheng	—
NaH2PO3	Tianjin Damao	—
Paraformaldehyde	Biosharp	BL539A
Pathology-grade microscope slides	Shitai	—
Phenylmethanesulfonyl fluoride (PMSF)	Beyotime	ST506
Plantar Test 37370	UGO BASILE	37370
Polyvinylidene fluoride membrane	Solarbio	—
Propentofylline (PPF)	Taosu	T19812
Rabbit anti-rat COX-2 primary antibody	Abcam	Ab179800
Rabbit anti-rat GFAP primary antibody	Proteintech	16825-1-AP
Rabbit anti-rat Iba-1 primary antibody	Proteintech	10904-1-AP
Rabbit anti-rat NeuN primary antibody	Abcam	Ab279297
Rabbit anti-rat t-ERK1/2 primary antibody	Proteintech	11257-1-AP
Rabbit anti-rat TNF-α primary antibody	Proteintech	29652-1-AP
RIPA lysis buffer	Solarbio	R0010
SDS	Solarbio	S8010
Skim milk powder	Solarbio	D8340
SPSS	IBM	22
Tetramethylethylenediamine (TEMED)	Solarbio	T8090
Tris	Solarbio	T8060
Triton X-100	Solarbio	T8200
U0126-EtOH	Selleck	S1102
Ultra-low temperature freezer	Haier	—
Ultra-low temperature freezer	Sanyo	—
Ultrapure water purification system	Heal Force	—
von Frey filaments	North Coast Medical	—