M2 polarization of macrophage induced by IL-33 promotes mouse sciatic nerve regeneration

Abstract

Peripheral nerve injury is a common condition that imposes a significant burden on both society and patients, remaining a key focus in basic research. IL-33, a member of the IL-1 family, plays an important role in tissue repair following various injuries. This study aims to investigate the role and mechanism of IL-33 in peripheral nerve regeneration. Transcriptome sequencing, Western blot, and immunofluorescence were used to examine the expression changes and cellular localization of IL-33 after sciatic nerve injury in mice. In a mouse sciatic nerve crush model, exogenous recombinant IL-33 was administered, and its effects on nerve fiber regeneration and functional recovery were assessed through morphological and behavioral analyses. Transcriptome sequencing and in vitro BMDMs culture were employed to explore the mechanism by which IL-33 promotes peripheral nerve regeneration. After sciatic nerve crush injury in mice, IL-33 expression was significantly upregulated from days 1 to 7 post-injury, primarily in Schwann cells. Intraperitoneal injection of IL-33 promoted axonal regeneration, enhanced remyelination of myelinated nerve fibers, prevented atrophy of the affected gastrocnemius muscle, and facilitated motor function recovery. IL-33 treatment significantly increased the gene expression and number of M2 macrophage in the injured nerve without affecting M0 and M1 macrophage populations. Macrophage ablation reduced the pro-regenerative effects of IL-33 on axonal regrowth. In vitro experiments confirmed that IL-33 promotes the polarization of BMDMs toward M2 macrophages. In summary, this study demonstrated up-regulated IL-33 promotes nerve regeneration and functional recovery at least partly through induction of M2 polarization of macrophage after peripheral nerve injury.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13578-026-01536-9.

Introduction

Peripheral nerve injury (PNI), a prevalent neurological condition, primarily results from traffic accidents, iatrogenic surgical complications, mechanical compression, or metabolic disorders such as diabetes mellitus. Despite the intrinsic regenerative capacity of the peripheral nervous system (PNS), PNI frequently leads to persistent neurological deficits with limited functional recovery [1]. Although advancements in microsurgical techniques have improved nerve repair outcomes, surgical intervention remains effective in only a subset of cases, leaving the overall therapeutic efficacy unsatisfactory [2, 3]. Consequently, developing safe and effective treatment strategies for PNI represents a critical unmet need in clinical neurology.

Macrophages of PNS are broadly categorized by origin that tissue-resident macrophages in steady-state nerves and bone marrow-derived macrophages (BMDMs) recruited post-injury [4]. However, upon injury, circulating monocytes are recruited to the lesion site, differentiating into infiltrating macrophages in response to inflammatory mediators (IL-1α, IL-1β, TNF-α) and chemokines (CCL2, CCL3, CXCL10, CXCL1) secreted by repair Schwann cells (SCs). These infiltrating macrophages play a pivotal role in nerve regeneration by collaborating with SCs to facilitate axonal regrowth [5–7]. Macrophages exhibit remarkable plasticity, polarizing into distinct functional phenotypes in response to microenvironmental cues. In the acute phase of injury, classically activated M1 macrophages dominate, driving inflammation and phagocytosis of cellular debris. As repair progresses, a phenotypic shift occurs, with alternatively activated M2 macrophages becoming predominant during the intermediate and late stages. M2 macrophages promote tissue remodeling by secreting anti-inflammatory cytokines and supporting extracellular matrix reorganization. Upon completion of nerve regeneration, macrophages are gradually cleared from the injury nerve [8, 9]. This tightly crosstalk between macrophages and SCs underscores their essential cooperative role in successful peripheral nerve repair [10].

Emerging evidence indicates that IL-33 plays a pivotal role in modulating immune microenvironment by binding to its surface receptor ST2 [11]. The multifaceted immunomodulatory functions of IL-33, particularly its ability to interact with diverse immune cell populations through both ST2-dependent and independent pathways, have been extensively investigated in recent researches [12]. Furthermore, IL-33 has been implicated in the pathogenesis of various central nervous system (CNS) disorders, including traumatic injuries, degenerative diseases, and neuroinflammatory conditions [13–15]. However, the functional significance of IL-33 in peripheral nerve biology, particularly its potential role in PNI regeneration and repair, remains largely unexplored.

In this study, we investigated the role of IL-33 on axonal regeneration after sciatic nerve crush (SNC) injury in mice. Building upon findings demonstrating upregulated IL-33 expression in SCs after PNI, we administered recombinant murine IL-33 (rmIL-33) via intraperitoneal injection in SNC mice. Our results demonstrated that IL-33 treatment markedly accelerated axonal regeneration and improved motor function recovery, which was associated with enhanced polarization of macrophages toward the M2 phenotype. These findings suggest that IL-33 represents a promising therapeutic candidate for promoting nerve repair, offering both efficacy and safety in the treatment of PNI.

Method

Mice

All animal experimental protocols were approved by the Ethics Committee of Guangdong Pharmaceutical University (approval number: gdpulac2024014) and performed in accordance with the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals. Eight-week-old male C57BL/6 J mice weighing 20–25 g were purchased from Guangzhou Ruige Biotechnology Co., Ltd. Mice were housed in a standard specific pathogen-free (SPF) facility with a 12/12-h light–dark cycle at Guangdong Pharmaceutical University. Upon completion of the experimental procedures, mice were euthanized by cervical dislocation.

In vitro Wallerian degeneration (WD)

In vitro WD of nerve was performed as previous reported [16]. In brief, mouse sciatic nerves were dissected under sterile conditions and trimmed to a length of 5 to 8 mm. Then nerve segments were cultured in DMEM/F12 (Corning) medium supplemented with 1% penicillin–streptomycin (Gibco) and 10% fetal bovine serum (FBS, Lonsera) at 37℃ in a 5% CO₂ incubator for 6 h.

RNA sequencing analysis

In brief, total RNA was extracted by phenol–chloroform method using Trizol (ambion) and RNA quality was assayed by Agilent 2100 Bioanalyzer and all samples exhibited an RIN ≥ 8. RNA-seq was performed at a strand-specific 100 cycle paired-end resolution using an Illumina HiSeq 2500 sequencing machine. The read quality was assessed using the FastQC software. Differentially expressed genes (DEGs) were analyzed with DESeq2, and those genes with an q-value ≤ 0.05 and |log2(fold-change)|≥ 1 were considered statistically significant. Figures related to RNA sequencing were plotted by https://www.bioinformatics.com.cn, an online platform for data analysis and visualization.

Mouse SNC surgery and treatment

As previously described [17], we established a SNC injury model in mice. Mice were anaesthetized by an intraperitoneal injection of 1.25% tribromoethanol (Sigma). The hind limbs were shaved and disinfected, and the mice were fixed in the prone position on the operating table. Under aseptic conditions, the bilateral sciatic nerves were exposed by separating the septum between the gluteus maximus and biceps femoris muscle. Then a standardized crush injury was performed at 1–2 mm distal to the sciatic notch using 0.3 mm-wide micro forceps for 10 s. Following injury, the nerves were carefully placed back to the septum, and the surgical site was closed with 5–0 absorbable suture. Animals were then returned to their home cages for watchful recovery. For exogenous IL-33 treatment, 1 μg or 2 μg of the rmIL-33 (Sino Biological) diluted with 200 μl PBS was administered via intraperitoneal injection at indicated times. For macrophage ablation, 200 μl Clodronate Liposomes (FormuMax) was administered via intraperitoneal injection to each mouse one day before SNC surgery. To mitigate post-surgical pain and distress, all operated mice received subcutaneous injections of the analgesic carprofen (5 mg/kg) once daily for three consecutive days.

Western blot

Injured sciatic nerve segments and cultured cells were harvested and lysed using RIPA buffer (Meilunbio) containing 1% PMSF (Meilunbio). The concentration of total protein extract was determined using a BCA protein assay kit (Meilunbio), and then extracts were mixed with 2 × Laemmli sample buffer (Bio-Rad) and denatured by boiling for 10 min. The protein samples were electrophoresed on SDS polyacrylamide gels and then transferred onto PVDF membranes. After blocking with 5% skim milk in TBST, blots were incubated with the indicated primary antibodies, including anti-rabbit IL-33 (1:2000, ab187060, Abcam), anti-rabbit CD206 (1:2000, ab3003621, Abcam), anti-rabbit CD86 (1:1000, ET1606-50, HuaBio), anti-mouse β-actin (1:8000, 660009-1-Ig, Proteintech), individually at 4℃ overnight. Then, membrane was then incubated with the appropriate HRP-conjugated goat anti-rabbit (1:8000, PGAR001, Proteintech) and HRP-conjugated goat anti-mouse (1:8000, SA00001-1, Proteintech) secondary antibodies for 1 h. The immunoreactive protein bands were detected using ECL chemiluminescent substrate (Biosharp) and a chemiluminescence imaging system (Tanon). The band intensity was quantified using ImageJ software and normalized to β-actin expression.

Frozen section preparation and staining

Dissected sciatic nerves and gastrocnemius muscles were fixed with paraformaldehyde at 4℃ for 24 h. Tissues were dehydrated in 30% sucrose overnight and then embedded in optimum cutting temperature compound (O.C.T., Sakura,) before sectioning. Then, sections (7 μm) were cut using a cryostat (Leica, CM1950). For hematoxylin staining, sections of muscles were stained with hematoxylin (Nanjing Jiancheng Bioengineering Institue) for 15 s, and mounted with neutral resin. For immunofluorescent staining, the sections of nerves were incubated with primary antibodies including anti-rabbit IL-33 (1:500, ab187060, Abcam), anti-rabbit GAP43 (1:500, 8945S, CST), anti-rabbit CD206 (1:500, ab3003621, Abcam), anti-rabbit CD86 (1:200, ET1606-50, HuaBio), anti-rabbit IBA1 (1:500, ab178846, Abcam), anti-rabbit S100β (1:500, 66616-1-lg, Proteintech), anti-mouse Sox10 (1:200, 66786–1-Ig, Proteintech), anti-rabbit c-Jun (1:200 ET-1608–4, HuaBio), anti-rabbit Ki67 (1:200, HA71115, HuaBio), followed by incubated with secondary antibodies including donkey anti-mouse Alexa Fluor 555 (1:500, A-31570, Thermo Fisher), donkey anti-rabbit Alexa Fluor 488 (1:500, A-21206, Thermo Fisher). After mounting with anti-fade medium (Vector), immunofluorescent images were acquired using an EVOS M5000 fluorescence microscope (Thermo Fisher) with consistent exposure settings across samples.

GAP43 growth front measurement and immunofluorescence quantification

GAP43 growth front was quantified on longitudinal sciatic nerve sections using ImageJ. The crush site was defined as the 0-mm origin. The distal growth front was defined as the farthest point at which a cluster of ≥ 3 GAP43-positive axons could be identified. For each nerve, 3–5 sections were measured and 5 traces were drawn in each section from the crush site distally, and the mean extension distance was used as a single biological replicate. Immunofluorescence intensity along the nerve was quantified on longitudinal sections using ImageJ. Mean fluorescence was measured in an 8-mm segment starting at the crush site. For each segment, local background was measured using an ROI of identical size placed in an antigen-negative area of the same section and subtracted from the raw intensity values. For each nerve, 3–5 sections were measured and the mean intensity was used as a single biological replicate.

Behavior tests

Balance beam walking

As mentioned earlier [18], the standard experimental equipment was used to assess the recovery of motor function. The mice were trained three days before test, and the test was conducted every other day from 0 to 12 d after SNC in the mice. The 12-day post-injury time point was chosen for tissue collection because it represents the time window for observing the therapeutic intervention’s effects on nerve regeneration and muscle atrophy. Following randomization, the mice received testing and the entire procedure of the behavioral tests was recorded on video. The recordings were then analyzed in a double-blind manner. The times used for passing through a balance beam with 80 cm long and 6 mm wide were recorded. Each animal was measured three times, and the average value was taken as the data for that test session.

Ladder rung walking

As previous reported [19], mice were trained three days before test, and the test was conducted every other day from 0 to 12 d after SNC in the mice. An irregular pattern that was changed randomly from trial to trial prevented the animal from learning the pattern. Following randomization, the mice received testing and the entire procedure of the behavioral tests was recorded on video. The recordings were then analyzed in a double-blind manner. The times used for passing through a 1 m long ladder rungs were recorded. Each animal was measured three times, and the average value was taken as the data for that test session.

Sciatic functional index (SFI)

SFI is used to assess the recovery of motor function of the injured sciatic nerve, and is conducted before and at 0 to 12 d after the SNC surgery. Three parameters, including toe spread (TS, first to the fifth toe), intermediate toe spread (ITS, second to the fourth toe), and total print length (PL, tip of the third toe to heel) were used for determining the SFI. Calculation of SFI was based on a previous report [20].

Transmission electron microscopy (TEM) analysis

The isolated nerve tissue was trimmed into 2–3 mm segments and preserved in a mixed fixative solution containing 2.5% glutaraldehyde and 2% paraformaldehyde at 4 °C. Cut sections were stained with uranyl acetate and lead citrate, and then imaged using a transmission electron microscope (JEM-1400 Flash, Japan). The myelin thickness and axonal diameter were measured using ImageJ, and G-ratio was determined as axon diameter divided by the diameter of the whole myelinated fiber.

Cell culture and treatment

BMDMs were prepared according to the reported method [21]. Briefly, femurs and tibias from both hind limbs of 8-week-old C57 mice were dissected under sterile condition. The epiphyses at both ends of harvested bones were cut off, and bone marrow cells were flushed into a 1.5 ml Eppendorf tube using a 1 ml syringe filled with ice-cold PBS. Then, red blood cells were lysed with RBC lysis buffer (Solarbio). Harvested bone marrow cells were seeded in 6-well plates at a density of 1 × 10⁷ cells per well and maintained in DMEM (Corning) supplemented with 10% FBS (Lonsera) and 1% penicillin–streptomycin (Gibco). Cultures were incubated at 37 °C in a humidified atmosphere containing 5% CO₂. Cells were stimulated with 30 ng/ml recombinant mouse macrophage colony-stimulating factor (M-CSF, Sino Biological) for 5 d to induce differentiation, followed by treatment with rmIL-33 at concentrations of 10 ng/ml or 30 ng/ml for 48 h. BMDMs were verified by immunofluorescent staining of macrophage marker F4/80 (1:500, ab6640, Abcam) and the secondary antibody donkey anti-rat Alexa Fluor 594 (1:500, A-21209, Thermo Fisher).

qPCR

Total RNA was extracted from the sciatic nerve tissue and BMDMs by phenol–chloroform method using Trizol (Ambion). The TB Green® Premix Ex Taq™ II (TaKaRa) was used to measure the RNA levels of Arg1, Mgl2, Chil3, Retnla, Cd206, Cd163, Cd86, Cd80. Primer sequences for each gene are listed in supplementary Table S1. All data were normalized to GAPDH as an endogenous control and analyzed using the delta-delta Ct method.

Statistical analysis

All data are presented as the mean ± Standard Error of the Mean (SEM) from at least three independent experiments. Replicate numbers used for statistical analyses were biological replicates (individual animals or independent culture wells per experiment). Statistical analyses were performed using GraphPad Prism software (version 8.0, GraphPad Software, USA). For comparisons between two groups, the normality of data distribution was assessed using the Shapiro–Wilk test. An unpaired two-tailed Student’s t-test was used for data that passed the normality test; otherwise, the Mann–Whitney U test was applied. For comparisons among multiple groups, the homogeneity of variances was assessed using the Brown–Forsythe test. One-way analysis of variance (ANOVA) followed by Tukey’s post hoc test was used for data with equal variances, while the Kruskal–Wallis test with Dunn’s post hoc analysis was employed for data with unequal variances. A p-value of less than 0.05 was considered statistically significant and exact p-value is labeled in the figures. And the NS indicates not significant.

Result

Transcriptome analysis indicated that IL-33 was upregulated in degenerating nerve

While the role of interleukin (IL) family members in neuroinflammatory responses is well established [22], their specific expression patterns and functions in normal and degenerating nerves remain to be fully characterized. To address this, we performed RNA sequencing analysis on mouse sciatic nerve explants cultured for 6 h in vitro (Fig. 1A). Cultured nerve explants undergo a process similar to WD after injury in vivo with advantage of excluding the influence of infiltrating cells from blood when analyzing transcriptome change of degenerating nerve. Cluster analysis of DEGs revealed distinct expression patterns among IL family members (Fig. 1B, C). Notably, six genes including Il-6, Il-1rn, Il-1b, Il-11, Il-1a, and Il-33 were up-regulated while three genes including Il-34, Il-4, and Il-16 were down-regulated in degenerating nerves (Fig. 1D). FPKM-based quantification of significantly altered IL genes demonstrated particularly striking results for Il33. Il33 exhibited the highest expression abundance post-injury, exceeding other IL family genes by 1–2 orders of magnitude. Interestingly, Il33 also showed detectable basal expression in uninjured control sciatic nerves (Fig. 1E).

Fig. 1.

IL-33 expression was up-regulated in the sciatic nerve of WD in vitro. A Schematic overview of the workflow of RNA-sequencing experiment (n = 2 animals per group). B Heat map showing hierarchical clustering of DEGs between WD and normal control nerves. C Heat map illustrating the expression patterns of interleukin family genes. D Fold-change values of differentially expressed IL family genes between WD and normal control nerves. E Bar chart showing the FPKM expression levels of the differentially expressed interleukin family genes

IL-33 expression is significantly up-regulated in SCs of injured sciatic nerves

To characterize its role in peripheral nerve repair, we analyzed IL-33 expression following SNC injury. Quantitative assessment demonstrated significant up-regulation of IL-33 protein in injured nerves compared to sham controls at 1, 3, and 7 d post injury (Fig. 2A, B). Given that SCs play vital roles in both axonal maintenance and immune regulation in the PNS [23], we performed co-staining analysis of IL-33 and S100, a marker of SC cytoplasm. Results showed that IL-33 signal was located in DAPI positive nuclei surrounded by S100 positive cytoplasm, which indicated SCs was the main IL-33 positive population in injured sciatic nerve (Fig. 2C). These findings suggest that IL-33 may function as a damage-responsive signaling molecule during nerve repair processes.

Fig. 2.

The expression of IL-33 protein is up-regulated after sciatic nerve injury. A Representative Western blot bands of the expression of IL-33 protein in sciatic nerves at 1, 3, and 7 d after SNC. Nerve underwent exposure procedures but not crush served as the sham control. B Statistical quantification of IL-33 in nerves after injury (n = 4 animals per group). C Immunofluorescent co-staining was performed on the longitudinal section of the sciatic nerve at 1, 3, and 7 d after SNC using IL-33 and S100 antibodies. DAPI was used to show cell nuclei

IL-33 treatment accelerates nerve regeneration and functional recovery

To elucidate the role of IL-33 in peripheral nerve regeneration, mice were subjected to SNC and subsequently administered with rmIL-33 intraperitoneally for three consecutive days. Three days after injury, nerve regeneration was assessed by measuring axonal growth distance using GAP43 immunofluorescence [24]. Notably, both 1 μg and 2 μg doses of rmIL-33 significantly enhanced axonal regeneration following nerve injury, with the 2 μg dose exhibiting more pronounced effects (Fig. 3A–C). Additionally, we examined whether rmIL-33 treatment influenced GAP43 protein expression in the neuronal cell bodies of dorsal root ganglia (DRGs) connecting with sciatic nerve. Immunofluorescent co-staining with the cytoskeletal marker β3-tubulin (Tuj1) and GAP43 in L₄ and L₅ DRGs revealed that rmIL-33 administration did not alter GAP43 expression in neuronal somata (Fig. S1).

Fig. 3.

IL-33 treatment promotes axonal regeneration after SNC in mice. A Schematic workflow of the SNC surgery and rmIL-33 treatment experiments. B Immunofluorescence labeling of regenerated axons using GAP43 antibody. The red line indicates the crush site and the yellow line indicates measurement endpoint. C Quantification of regenerated axon length shown as a statistical chart (n = 5 animals per group)

To further investigate the effects of rmIL-3 treatment on functional recovery following sciatic nerve injury, we conducted a comprehensive series of behavioral assessments over 12 days in mice with unilateral SNC in right side (Fig. 4A, B). The behavioral results revealed that rmIL-33 significantly reduced passing time of balance beam walking at 4 and 6 d post-injury compared to the PBS group (Fig. 4C). Similarly, in the ladder rung walking experiment, passing time was significantly shorter in rmIL-33-treated mice than control mice at 4, 6, 8, and 10 d post-injury (Fig. 4D). SFI data showed that although the index of both groups of mice did not return to the normal base level, the SFI of mice treated with rmIL-33 from 4 to 12 d post-injury was improved compared with that of mice in the PBS group (Fig. 4E, F). TEM was used to observe the remyelination and results indicated that compared with the mice in the PBS group, the mice treated with rmIL-33 exhibited a thicker myelin sheath thickness and a smaller G-ratio in myelinated fibers of sciatic nerve at 12 d after injury (Fig. 5A–D). Compared with the sham surgery nerve, the denervated muscle fibers of gastrocnemius muscle exhibited varying degrees of denervation atrophy. However, rmIL-33 treatment significantly protected the mice from atrophy compared to the PBS group (Fig. 5E, F). Above mentioned results indicate treatment with exogenous rmIL-33 significantly accelerates axonal regeneration, promotes remyelination, and enhances recovery of motor function following sciatic nerve injury.

Fig. 4.

IL-33 treatment promotes motor recovery of sciatic nerve after injury in mice. A Schematic workflow of the SNC surgery, behavioral and morphological experiments. B Images of the balance beam and ladder rung walking experiments. C Statistical graph of passing time in balance beam walking tests (n = 8 animals in PBS group, n = 9 animals in rmIL-33 group). D Statistical graph of passing time in ladder rung walking tests (n = 8 animals in PBS group, n = 9 animals in rmIL-33 group). E Representative images of the paw after SNC surgery. F Statistical graph of SFI after SNC surgery (n = 8 animals in PBS group, n = 9 animals in rmIL-33 group)

Fig. 5.

IL-33 treatment promotes remyelination and prevents denervated atrophy in mice after SNC. A Representative TEM images of sciatic nerves at 12 d after crush injury. B–D, Quantification of axonal diameter (B), myelin thickness (C) and G-ratio (D) in TEM images (each dot represents one myelinated fiber, all measured fibers were from 4 animals in each group). E Representative images of the gastrocnemius muscle section stained with hematoxylin at 12 d after sciatic nerve injury. F Quantification of the cross-sectional area of gastrocnemius muscle fibers in hematoxylin staining images (each dot represents one muscle fiber, all measured fibers were from 4 animals in each group)

IL-33 treatment promoted the M2 polarization of macrophages after nerve injury

To elucidate the mechanisms by which rmIL-33 treatment accelerates axonal regeneration in mice with PNI, we performed RNA-seq on sciatic nerves at 3 d post-injury. Comparative analysis of DEGs rmIL-33 treated and control groups identified a total of 257 DEGs, of which 160 were up-regulated and 97 were down-regulated. KEGG enrichment analysis revealed that these DEGs were significantly enriched in terms of cytokine-cytokine receptor interaction and chemokine signaling pathways (Fig. 6A, B). Further analysis using a volcano plot highlighted that the up-regulated genes were predominantly associated with the M2 macrophage signature (Fig. 6C). Evaluation of macrophage phenotype-specific markers by FPKM values showed no significant differences in the expression of markers of M0 or M1 macrophage (Fig. 6D). To validate these findings, M2 markers including Arg1, Mgl2, Retnla, Chil3, Cd206, and Cd163 were tested using qPCR and results confirmed their significant up-regulation following rmIL-33 treatment (Fig. 6E).

Fig. 6.

RNA-seq shows IL-33 induces M2 polarization of macrophages in injured sciatic nerve. A Heat map showing hierarchical clustering of DEGs between rmIL-33-treated and PBS-treated mice (control) (n = 3 animals per group). B KEGG enrichment analysis of DEGs between rmIL-33-treated and PBS-treated mice (control). C Volcano plot of DEGs between rmIL-33-treated and PBS-treated mice, marked points represent genes specifically expressed by M2 macrophages. D Heatmap of expression levels based on FPKM value of markers of M0, M1 and M2 macrophages between rmIL-33-treated and PBS-treated nerves. E, qPCR analysis of the M2 macrophage-related phenotypic genes rmIL-33-treated and PBS-treated nerves (n = 3 animals per group)

In addition, immunofluorescent analysis of nerve sections revealed that mice treated with rmIL-33 exhibited significantly enhanced CD206 (M2 marker) intensity compared with PBS-treated controls (Fig. 7A, B). While no statistically significant differences were observed between the two groups in the expression levels of either IBA1 (pan-macrophage marker, M0 phenotype) or CD86 (M1marker) (Fig. 7C–F). On the other hand, compared with mice in the PBS group, rmIL-33 treatment did not affect the expression changes of genes related to cell proliferation and SC dedifferentiation (Fig. S2A). This result was further confirmed by immunofluorescent staining of cell proliferation marker Ki67, and SC dedifferentiation marker c-Jun. (Fig. S2B-E). Together, these results indicate that IL-33 treatment promotes the polarization of macrophages towards the M2 phenotype.

Fig. 7.

Immunofluorescence shows IL-33 induces M2 polarization of macrophages in injured sciatic nerve. A, B Representative images of CD206 immunofluorescent staining in longitudinal sections of the sciatic nerve, and statistical analysis of CD206 fluorescence intensity (n = 5–7 animals per group). C, D Representative images of CD86 immunofluorescent staining in longitudinal sections of the sciatic nerve, and statistical analysis of CD86 fluorescence intensity (n = 5–7 animals per group). E–F Representative images of IBA1 immunofluorescent staining in longitudinal sections of the sciatic nerve, and statistical analysis of IBA1 fluorescence intensity (n = 5–7 animals per group)

Macrophage ablation reduces the pro-regenerative effects of IL-33 on axonal regrowth

To verify that IL-33 promotes sciatic nerve regeneration by facilitating M2 macrophage polarization, we conducted macrophage depletion experiments using clodronate liposomes, a classical macrophage-depleting agent. Mice were injected with clodronate liposomes one day before the SNC surgery, followed by consecutive three-day rmIL-33 treatment post-SNC. The sciatic nerves were then collected for macrophage marker staining and fluorescence intensity quantification, while axonal regrowth was assessed via GAP43 immunofluorescent staining. In line with previous finding, rmIL-33 treatment significantly increased axon extension distance, while clodronate liposomes significantly suppressed the IL-33-mediated enhancement of axon regeneration (Fig. 8A, B). Clodronate liposomes concurrently reduced the recruitment of IBA1-positive M0 macrophage compared with nerve treated with control liposomes (Fig. 8C, D). In addition, clodronate liposomes suppressed the IL-33-mediated enhancement of CD206-positive M2 macrophage infiltration in the injured nerves (Fig. 8E, F). These findings suggest that IL-33 influences peripheral nerve regeneration at least partly through the macrophage-dependent pathway.

Fig. 8.

Macrophage ablation reduces the pro-regenerative effects of IL-33 on axonal regrowth. A Representative images of GAP43 immunofluorescent staining in longitudinal sections of the sciatic nerve. The red line indicates the crush site and the yellow line indicates measurement endpoint. B Quantification of regenerated axon length shown as a statistical chart (n = 6 animals per group). C, D, Representative images of IBA1 immunofluorescent staining in longitudinal sections of the sciatic nerve, and statistical analysis of IBA1 fluorescence intensity (n = 6 animals per group). E, F Representative images of CD206 immunofluorescent staining in longitudinal sections of the sciatic nerve, and statistical analysis of CD206 fluorescence intensity (n = 6 animals per group)

IL-33 induces bidirectional polarization of BMDMs in vitro

Given previous reports that bone marrow-derived monocytes are recruited to injury sites and contribute to local inflammatory responses following sciatic nerve injury [25], we isolated mouse bone marrow monocytes and differentiated them into macrophages in vitro using M-CSF stimulation for 5 d (Fig. 9A). Cultured BMDMs were verified by staining of F4/80, a classical macrophage marker (Fig. 9B). These BMDMs were then treated with different dose of rmIL-33 for 48 h before RNA and protein extraction. qPCR analysis of M2 markers (Cd206, Cd163) and M1 markers (Cd86, Cd80) showed that, compared to PBS controls, treatment with rmIL-33 significantly increased the expression of both M1 and M2 macrophage markers suggesting the enhancement polarization of BMDMs toward both M1 and M2 phenotypes (Fig. 9C). These results were further validated by Western blot analysis, confirming that rmIL-33 treatment drives dual polarization of cultured BMDMs toward M1 and M2 phenotypes in vitro (Fig. 9D, E). Above results indicated that IL-33 treatment enhanced the macrophages’ responsiveness to both M1 and M2 polarizing stimuli in vitro. However, within the in vivo injured nerve microenvironment, this broad priming effect appears to be selectively channeled, resulting in the net amplification of the M2 program and its associated functional outcomes.

Fig. 9.

IL-33 induces the polarization of BMDMs to the M1 and M2 phenotypes. A Schematic diagram of the procedure for preparing BMDMs. B Immunofluorescent staining of F4/80 of cultured BMDMs. DAPI was used to show cell nuclei. C qPCR analysis of marker genes of M1 and M2 macrophages in BMDMs treated with rmIL-33 for 48 h (n = 6 biological replicates). D Representative Western blot bands of BMDMs treated with rmIL-33 for 48 h. E Statistical quantification of expression of CD86 and CD206 in (E) (n = 4 biological replicates)

Discussion

Extensive evidence suggests that macrophage-SC crosstalk, mediated by intercellular signaling and phenotypic reprogramming, contributes significantly to axonal regeneration after peripheral nerve damage [26]. Here, we demonstrated for the first time that up-regulation of IL-33 expression in SCs after sciatic nerve injury in mice. Based on these findings, we administered rmIL-33 via intraperitoneal injection in a SNC mouse model. This intervention enhanced M2 macrophage infiltration in injured nerve, accelerated axonal regeneration, and improved functional recovery post-injury. Therefore, our findings may provide a novel therapeutic direction for clinical management of PNI.

Macrophages play a crucial role as immune effector cells in the PNS. Following PNI, resident macrophages constitute merely 2–9% of the total cellular population. Notably, infiltrating macrophages begin migrating into the distal nerve segment within 24 to 48 h post-injury, peak in abundance around day 6, and persist at stable levels until day 14. A small but detectable population remains even at day 28 post-injury [9, 27]. Activated macrophages are broadly categorized into pro-inflammatory (M1) and anti-inflammatory (M2) subtypes [28]. Notably, the population of M2 macrophages peaks on day 3 post-injury and subsequently stabilizes at this level [29]. These M2 macrophages play a crucial role in tissue repair through their secretion of anti-inflammatory cytokines and growth factors, which collectively facilitate cellular proliferation, angiogenesis, and extracellular matrix remodeling [30, 31]. In addition, macrophage depletion by administering ganciclovir in CD11b-TK^mt−30 mice or clodronate liposomes in wild-type mice disrupts axon regeneration following sciatic nerve injury [32, 33]. Therefore, activated M2 macrophages have been proven to be one of the key cells for axon regeneration after PNI [34, 35]. Our study demonstrated that rmIL-33 treatment significantly enhanced M2 macrophage infiltration in the sciatic nerve at 3 days post injury.

IL-33 functions as an alarmin that bridges immune activation and tissue repair, can activate various immune cells [36], and it has been reported that IL-33 plays dual-role in inactivating macrophage [37]. This rapidly released cytokine signals through ST2 to coordinate wound healing and inflammatory responses, with its pronounced capacity to induce M2 macrophage polarization being particularly critical for its reparative functions [12, 38]. Emerging evidence that Following anesthesia and surgery, hippocampal levels of IL-33 and its receptor ST2 were significantly downregulated. This decrease was accompanied by elevated expression of pro-inflammatory cytokines (IL-6, IL-1β) and the M1 microglia marker iNOS, along with reduced expression of the M2 marker CD206. Importantly, Intraperitoneal injection of rmIL-33 attenuated neuroinflammation, promoted microglial polarization toward an anti-inflammatory phenotype, and ameliorated anesthesia/surgery-induced dendritic spine loss and cognitive impairment [39]. This finding aligns with the observations from our in vivo experiments. However, in our in vitro experiment of stimulating mouse BMDMs with rmIL-33, we found that rmIL-33 induced BMDMs to polarize towards M1 and M2. A recent report showed that IL-33 preferentially induces M2 macrophage polarization in vivo, however, drives both M1 and M2 polarization simultaneously in BMDMs and RAW264.7 cells in vitro by activating the MYD88/ERK pathway [40]. Interestingly, in another study, treatment of ST2-deficient mouse BMDMs with IL-33 initiated PI3K/AKT-dependent M2 macrophage polarization. This finding suggests that IL-33 may directly induce M2 polarization in macrophages via an ST2-independent mechanism [41]. The mechanisms underlying IL-33-induced macrophage polarization have attracted the attention of researchers in recent years [37, 42–44]. While our in vitro data demonstrate IL-33’s direct capacity to polarize BMDMs, the net pro-regenerative effect in vivo likely results from IL-33 priming macrophages within a complex milieu where interactions with SCs, other cytokines, and temporal dynamics collectively favor an M2-dominant outcome and functional enrichment at the injury site. These studies suggest that the mechanism of action of IL-33 may vary across diverse immune microenvironments.

Our study provide evidence for IL-33 induces M2 polarization of macrophage in injured peripheral nerve and contributes to nerve regeneration. However, some limitations are remained. While we observed significant up-regulation of IL-33 in SCs of injured sciatic nerves, but whether SCs derived IL-33 directly orchestrates macrophage phenotypic reprogramming in this context remains to be classified. Furthermore, our findings demonstrate that rmIL-33 administration in mice with sciatic nerve injury enhances multiple regenerative processes, including accelerated axonal regeneration, improved remyelination, restoration of motor function, and prevention of denervation-induced gastrocnemius muscle atrophy. However, the precise mechanisms underlying IL-33-mediated M2 macrophage polarization and its subsequent promotion of nerve regeneration remain to be elucidated. Therefore, P0-Cre or Plp1-CreERT2 mediated conditional Il1rl1 or Il33 deletion could be used to test SC–derived IL-33 necessity and ST2 dependent mechanism in future study. While our study assessed key histological markers of regeneration, longer-term functional recovery and complementary electrophysiological measures, such as nerve conduction velocity (NCV) and compound muscle action potential (CMAP), were not performed. Incorporation of these assessments in future work will be essential to more completely characterize the functional impact of IL-33 on reinnervation and its potential therapeutic utility. Lasty, while our study provides foundational insights using a male model, future work including female cohorts will be essential to fully generalize IL-33-based interventions to both sexes, particularly in light of established sex differences in immune and inflammatory responses. Although this study has some limitations, it still sheds light on the translational application of IL-33 in the clinical treatment of PNI.

Conclusion

In summary, our study demonstrates that IL-33 expression is up-regulated following sciatic nerve injury in vivo. Treatment with rmIL-33 improved axonal regeneration and functional recovery after peripheral nerve injury, mainly by promoting macrophage M2 polarization. However, the ability of rmIL-33 to induce both M1 and M2 polarization in BMDMs in vitro may be attributed to the simplified culture conditions. In contrast, within the complex in vivo microenvironment, IL-33 selectively promotes M2 polarization, likely due to the interplay of various immunomodulatory factors.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Danyang Xu: Investigation, Visualization, Writing - original draft. Yanyi Li, Qi Zhang, Weiqiang Wu, Yunjing Du: Investigation. Jing Liu, Xiaobao Jin, Xiaochu Li, Juan Shen, Jiasong Guo: Resources. Lixia Li, Jinkun Wen: Conceptualization, Project administration, Funding acquisition, Writing – Review & Editing.

Funding

This work was supported by Medical Scientific Research Foundation of Guangdong Province (A2025031); the Starting Fund for High-level Talent Introduction into Guangdong Pharmaceutical University (51304058053).

Data availability

The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

All animal experiments were approved by the Institutional Animal Care and Use Committee of Guangdong Pharmaceutical University (Approval No: gdpulac2024014).

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.