All the animal procedures were performed in accordance with the guidelines of the Institutional Animal Care and Use Committee of Anhui Medical University (No. LLSC20160052). C57BL/6J female mice at 6–8 weeks of age were purchased from the Experimental Animal Center of Anhui Medical University. The mice were housed in a temperature- and humidity-controlled room with a 12:12-light/dark cycle and allowed free access to food and water.

All the surgical procedures were performed under pentobarbital anesthesia. The skin around the lesion core was shaved and disinfected using iodophor. Subsequently, the T10 spinal cord was exposed via a dorsal laminectomy, which was located spinous process of mice, cut open the back skin about 2 cm, separate soft tissue to posterior lamina and open with rongeur to expose the spinal cord, and moderately severe crush SCIs were made using No. 5 Dumont forceps (Fine Science Tools, 11,252–20, Heidelberg, Germany) ground down to a tip with a width of 0.5 mm by compressing the cord laterally from both sides for 5 s (Wanner et al., 2013). Then, twitching of the hind limbs and movement of the tail were observed, which indicated that the SCI model was successfully established. Finally, the wound was sutured with 3–0 silk threads. The mice with SCI were examined daily to monitor their recovery, and their bladders were expressed manually three times a day until the return of reflexive bladder control. The sham group were subjected to laminectomy alone. The mice were sacrificed at 3, 7 and 14 days after SCI.

To eliminate microglia, mice were administered PLX5622 (MedChemExpress, HY-114153) at 130 mg/kg by oral gavage once a day for 17 consecutive days. PLX5622 was diluted in 5% DMSO, 40% polyethylene glycol 300, 5% polysorbate 80, and 50% saline, according to the manufacturer’s instructions. The total volume of the liquid is 200 μl An equal volume of vehicle was used as the control. SCI was established on the third day after gavage.

Myelin debris was isolated as previously described (Rolfe et al., 2017). Briefly, 6- to 8-week-old mice were euthanized, and their brain tissue was harvested and homogenized in ice-cold 0.32 M sucrose. Myelin debris was isolated from the brain tissue by sucrose density gradient centrifugation. The endotoxin concentration of the myelin debris was below the limit of detection of the Limulus Amebocyte Lysate assay (Lonza, Switzerland). Myelin debris was added to cells at a final concentration of 1 mg/ml in all the experiments.

BV-2 is a commonly used microglial cell line that has been widely used in vitro experiments to explore inflammatory responses in many studies (Riedl et al., 2009; Basso et al., 2017; Yoshizaki et al., 2021). The BV-2 microglial cell line was obtained from the American Type Culture Collection (CRL-3265, ATCC, Manassas, VA, United States) and cultured in Dulbecco’s modified Eagle’s medium (DMEM, HyClone, SH30021) supplemented with 10% fetal bovine serum (FBS, Gibco, 10,270,106), 100 U/ml penicillin and 100 g/ml streptomycin (Gibco, Grand Island, NY, United States). The cells were incubated in a humidified chamber at 37°C in a 95O₂ and 5% CO₂ atmosphere. Small interfering RNA (siRNA) was transfected into these cells using jetPRIME (Polyplus Transfection, 114–15), according to the manufacturer’s instructions. siRNA targeting mouse Fascin-1 (siRNA: 5′- GAU​GCC​AAC​CGU​UCC​AGU​UTT -3′) and nonspecific control siRNA (NC) were purchased from GenePharma (Shanghai, China).

BV-2 cells were plated in poly-d-lysine (PDL, Sigma, P7280)-coated 6-well plates at a density of 1 × 10⁶ cells/ml and cultured overnight. After serum deprivation for 24 h, the BV-2 cells were polarized toward the M1-like phenotype by treatment with lipopolysaccharide (LPS; 100 ng/ml, Beyotime Biotechnology, ST1470, Shanghai, China) and IFNγ (20 ng/ml, Beyotime Biotechnology, P6137) or toward the M2-like phenotype by treatment with IL-4 (20 ng/ml, Beyotime Biotechnology, P5916), and the cells were cultured for 24 h (Kigerl et al., 2009; David and Kroner, 2011). The supernatant was taken as the conditioned medium (1200 rpm, 5 min) after 24 h culture in serum-free medium. The M1 and M2 polarization of BV2 cells was identified by subsequent Western blot analysis and immunofluorescence analysis.

For Western blot analysis, the mice were anesthetized and transcardially perfused with 0.1 M phosphate-buffered saline (PBS) to remove the blood. Spinal cord tissue of 5 mm centered at the lesion core were harvested. For histological analysis, once the blood was removed, the mice were transcardially perfused with 4% paraformaldehyde (PFA), and a 5 mm portion of the spinal cord encompassing the injury epicenter was extracted and embedded in paraffin. Then, the tissue was sagittally sectioned at thicknesses of 6 μm on a microtome (Leica RM2235). Every 10th section was collected and mounted onto a series of slides.

Immunohistochemistry. Six-micrometer, paraffin-embedded spinal cord sections were dried, dewaxed, hydrated and subjected to antigen repair. Next, 10% donkey serum albumin (DSA, Solarbio, SL050) containing 0.3% Triton X-100 (Solarbio, T8200) was added and incubated at room temperature for 1 h. Then, primary antibodies were added and incubated at 4°C overnight. The primary antibodies used were as follows: mouse anti-Fascin-1 (1:50, Santa Cruz, sc-21743), rabbit anti-Fascin-1 (1:100, Abcam, ab126772), goat anti-Iba1 (1:200, Novus Biologicals, NB100-1028), rabbit anti-CX3CR1 (1:500, Abcam, ab8021), rabbit anti-Tmem119 (1:100, Synaptic system, 400002), rabbit anti-GFAP (1:100, Proteintech, 16825-1-AP), mouse anti-GFAP (1:100, Proteintech, 60190-1-Ig), goat anti-PDGFRβ (1:40, R&D Systems, AF1042-SP), rabbit anti-PDGFRβ (1:200, Abcam, ab32570), mouse anti-Mac2 (also known as galectin-3, 1:100, GB12246, Servicebio), rabbit anti-NG2 (1:100, Proteintech, 55027-1-AP), rabbit anti-iNOS (1:100, Affinity, AF0199), rabbit anti-CD206 (1:100, ab64693, Abcam) and rabbit anti-NeuN (1:500, Abcam, ab177487). The secondary antibodies were diluted in 1% donkey serum in PBS and incubated for 1 h at room temperature. The following secondary antibodies were used: Alexa Fluor 488 and Alexa Fluor 594 (1:500, Thermo Fisher Scientific, A-21206, A-21202, A-21203, A-21207, A-11058). The nuclei were stained using 4’,6-diamidino-2-phenylindole (DAPI) (1 μg/ml, Thermo Fisher Scientific). The fluorescence signals were obtained using an Axio Scope A1 microscope (Zeiss, Germany). ImageJ software (National Institutes of Health, Bethesda, MD, United States) was used for quantitative analysis.

Immunocytochemistry. Cells were fixed with 4% PFA for 10–15 min, permeabilized in 0.5% Triton X-100 in PBS for 10 min and blocked with 5% donkey serum in PBS for 30 min at 20–25°C. The primary antibodies (as listed above) were diluted in 1% donkey serum in PBS and incubated overnight at 4°C. The secondary antibodies (as listed above) were diluted in 1% donkey serum in PBS and incubated for 1 h at room temperature. Images were acquired as described above.

Every 10th 6 μm-thick, paraffin-embedded section was quantified, resulting in five analyzed slides per animal that included the entire injured spinal cord. The total numbers of Fascin-1⁺, CX3CR1⁺, and Fascin-1⁺CX3CR1⁺ double-positive cells in each of the sagittal sections of the spinal cord was counted under a ×20 objective lens with Zeiss ZEN imaging software (Zeiss, Germany).

The injured spial cord tissue was homogenized in radioimmunoprecipitation assay (RIPA) buffer (Sigma, R0278) supplemented with protease inhibitors (Roche, 04693124001) and phosphatase inhibitors (Roche, 04906845001). The cells were washed with cold PBS, homogenized in RIPA buffer on ice for 30 min, and then centrifuged at 12,000 rpm at 4°C for 30 min. The protein extracts were quantified by using a Pierce BCA protein assay kit (Beyotime Biotechnology, P0010S). Aliquots of the protein samples containing equal protein concentrations were separated on 10% sodium dodecyl sulfate (SDS)-polyacrylamide gels and subsequently transferred to polyvinylidene (PVDF) membranes. The membranes were blocked with 5% nonfat milk in Tris-buffered saline with 0.5% Tween-20 (TBST) at room temperature for 1 h and then incubated with primary antibodies, including mouse anti-GAPDH (1:2000, Proteintech, 60004-1-Ig), rabbit anti-Fascin-1 (1:5000, Abcam, ab126772), rabbit anti-iNOS (1:3000, AF0199, Affinity) and rabbit anti-CD206 (1:1000, ab64693, Abcam), at 4°C overnight. After three washes with TBST, the membranes were incubated with horseradish peroxidase (HRP)-conjugated goat anti-mouse secondary antibodies (1:10,000, Sigma, A4416) and HRP-conjugated goat anti-rabbit secondary antibodies (1:10,000, Sigma, A0545) for 1 h at room temperature. The protein band signals were obtained using an ECL detection kit (ECL, Thermol Biotech Inc., United States) and Tanon 5,200 system (Tanon, Shanghai, China). ImageJ was used for quantification analysis. The intensity of the GAPDH bands was used for normalization.

Cell migration was assessed by performing a scratch assay. Briefly, BV-2 cells were seeded into PDL-coated 6-well plates at a density of 2 × 10⁵ cells/well and incubated for 24 h. Then, BV2 cells were transfected with NC or siFascin-1 according to different design groups. After 24 h with transfection, BV2 cells were cultured with serum-free medium and were treated with or without myelin debris at a final concentration of 1 mg/ml for another 24 h. Then, BV2 cells were washed away with PBS for three times to remove non-ingested myelin debris. The cell layers were scratched using a 200-µl pipette tip to form a wound-like gap. The cells were then maintained in DMEM with 2% FBS, and images were captured at 0, 24 and 48 h after cell scratching. ImageJ was used to analyze the wound width.

A 24-well plate containing a 3 μm chamber (Costar, 3415) was used to assess the migration abilities of the cells. BV-2 cells were transfected, treated with myelin debris and activated in a PDL-coated 6-well plate before inoculation into the chamber. Then, the cells were suspended in serum-free media and seeded into the upper chamber (5 × 10⁴ cells per chamber). The lower chamber contained complete medium. After incubating for 12 h, the nonmigrating cells on the inside of the membrane were carefully removed with a cotton swab, and the migrating cells on the outside of the membrane were fixed with 4% PFA at room temperature for 20 min and stained with 0.1% crystal violet for 15 min. The cells were observed under a microscope and counted. At least five random fields were photographed, and the cells in each field were counted.

All the experiments were independently performed with at least three replicates and quantified in a blinded manner. The data are presented as the mean ± standard error of the mean (SEM). The statistical analysis was carried out with SPSS version 19.0 (SPSS Inc., Chicago, IL, United States) software. Student’s t test was used to compare the difference between two groups. One-way analysis of variance followed by Tukey’s post hoc test was used to compare differences among multiple groups. Differences were considered statistically significant if p < 0.05.

To determine the changes in Fascin-1 expression during SCI, Western blot was used to detect the relative expression levels of Fascin-1 before and 3–14 days after SCI. As shown in Figures 1A,B, Fascin-1 expression was significantly increased at 7 and 14 days after SCI compared with that before injury (^* p < 0.05 and ^*** p < 0.001, respectively). To further analyze the distribution of Fascin-1, we carried out immunofluorescence detection of Fascin-1 and GFAP, which is used to label the astrocytic scar formed around the lesion core. The fluorescence signal of Fascin-1 was also obviously increased at 7 and 14 days after SCI and gradually accumulated outside the lesion core, intermingling but not colocalizing with GFAP⁺ astrocytic scars (Figure 1C). These results indicate that the expression of Fascin-1 is prominently upregulated and mainly distributed outside the lesion core, Fascin-1 expression has a spatiotemporal distribution pattern similar to that of microglia after SCI (Bellver-Landete et al., 2019).

The cellular localization of Fascin-1 was then detected by immunofluorescence staining. We found that Fascin-1 was highly colocalized with CX3CR1-labeled microglia. The percentage of Fascin-1⁺CX3CR1⁺ costained cells relative to the total number of Fascin-1⁺ or CX3CR1⁺ cells in the injured spinal cord reached 94.06 ± 0.82% or 87.65 ± 1.08%, respectively (Figures 2A,E). These costained cells were mainly located around the SCI site (Figure 2A), which was consistent with previous studies (Bellver-Landete et al., 2019). Fascin-1 was not expressed in the GFAP⁺ astrocytes surrounding the lesion core or in the PDGFRβ⁺ pericytes accumulated in the epicenter (Figures 2B,C). We further examined the co-expression of Fascin-1 and Tmem119, a specific marker of homeostatic microglia (Bennett et al., 2016), the results showed that they were partially costained and confined outside of the lesion core (Figure 2D). The percentage of Fascin-1⁺ Tmem119⁺ costained cells relative to the total number of Fascin-1⁺ or Tmem119⁺ cells in the injured spinal cord reached 29.38 ± 1.19% or 59.0 ± 12.13%, respectively (Figure 2F). Considering the high co-staining of Fascin-1 and CX3CR1 (Figure 2A), these suggest that Fascin-1 may be significantly up-regulated in the activated microglia.

Iba1 is a common marker of microglia, but during SCI, both resident microglia and infiltrating macrophages can be identified by Iba1 (Wang et al., 2015). Our results showed that costaining of Fascin-1 and Iba1 was mainly confined around the lesion core, which was consistent with the localization of microglia (Figure 3A). Since Iba1 can also mark infiltrating macrophages, we used the area stained by Fascin-1 to identify the lesion border and circled it with the blue dotted line. We counted the costained of Fascin-1 and Iba1 outside the blue dotted line. The percentage of Fascin-1⁺Iba1⁺ costained cells relative to the total number of Fascin-1⁺ or Iba1⁺ cells in the injured spinal cord reached 88.79 ± 1.09% or 63.81 ± 1.93%, respectively (Figure 3E). Moreover, Fascin-1 was not colocalized with Mac2⁺ macrophages located at the lesion core (Figure 3B). In addition, Fascin-1 was not expressed in NeuN⁺ neurons or NG2⁺ cells in the injured area (Figures 3C,D). Taken together, these results show that Fascin-1 is highly expressed specifically in activated microglia after SCI.

To confirm that Fascin-1 is derived from microglia, PLX5622, a selective inhibitor of colony-stimulating factor 1 receptor (CSF1R) that crosses the blood–brain barrier (BBB) and eradicates nearly all the microglia in the central nervous system (CNS), was administered to mice (Huang et al., 2018). A solution containing PLX5622 was administered to mice by gavage starting 3 days prior to SCI and continuing until sacrifice. As shown in Figures 4A,C, the number of Fascin-1⁺CX3CR1⁺ microglia in the PLX5622 group was significantly decreased compared to that in the control group (****p < 0.0001). The expression of Fascin-1 decreased accordingly with the depletion of CX3CR1⁺ microglia. The percentages of Fascin-1⁺ cells and CX3CR1⁺ cells in the PLX5622 groups relative to those in the untreated control groups were 26.14 ± 0.61% and 27.38 ± 1.07%, respectively (Figure 4D). After the depletion of microglia, we observed that both GFAP⁺ astrocytic scars and PDGFRβ⁺ fibrotic scars became less compact and disorganized, and infiltrating Mac2⁺ macrophages were diffusely scattered outside of the lesion core (Figure 4B). Fascin-1 may be involved in microglial scars formation, which needs to be confirmed by specifically knocking out microglia-derived Fascin-1 instead of depleting microglia.

To assess the effect of Fascin-1 on the function of microglia, we cultured the BV-2 microglial cell line in vitro. Western blot (Figures 5A,B) and immunocytochemistry (Figure 5C) analyses revealed that compared to the negative controls, Fascin-1 siRNA (siFascin-1) could knockdown the expression of Fascin-1, while myelin treatment could upregulate the expression of Fascin-1 (^* p < 0.05). The migration ability of the microglia was reduced after Fascin-1 knockdown but enhanced by myelin addition, as observed by scratch and Transwell assays (^* p < 0.05, Figure 6). Furthermore, the inhibition of Fascin-1 expression and microglial migration using siFascin-1 could be reversed by treatment with myelin (Figures 5, 6). These results suggest that Fascin-1 is required for typical microglial migration.

Previous studies have shown that microglia can be polarized into either a pro-inflammatory (M1-like) or anti-inflammatory (M2-like) phenotype and involved in the regulation of the SCI microenvironment (Francos‐Quijorna et al., 2016). Next, we detected the effect of microglial polarization on Fascin-1 expression. Western blot analysis showed that M1-like microglia significantly expressed iNOS and M2-like microglia significantly expressed CD206 (Figure 7A). In addition, immunocytochemistry was used to further confirm the reliability of microglial polarization (Figures 7C,D). However, the expression of Fascin-1 in the polarized microglial populations was not significantly different (p > 0.05, Figure 7). The results show that microglial polarization has no effect on the expression of Fascin-1.