Schwann cell-derived exosomes containing MFG-E8 modify macrophage/microglial polarization for attenuating inflammation via the SOCS3/STAT3 pathway after spinal cord injury

Abstract

Macrophage/microglia polarization acts as an important part in regulating inflammatory responses in spinal cord injury (SCI). However, the regulation of inflammation of Schwann cell-derived exosomes (SCDEs) for SCI repair is still unclear. Therefore, we intend to find out the effect of SCDEs on regulating the inflammation related to macrophage polarization during the recovery of SCI. Firstly, the thesis demonstrated that SCDEs could attenuate the LPS- inflammation in BMDMs by suppressing M1 polarization and stimulating M2 polarization. Similarly, SCDEs improved functional recovery of female Wistar rats of the SCI contusion model according to BBB (Basso, Beattie, and Bresnahan) score, electrophysiological assay, and the gait analysis system of CatWalk XT. Moreover, MFG-E8 was verified as the main component of SCDEs to improve the inflammatory response by proteomic sequencing and lentiviral transfection. Improvement of the inflammatory microenvironment also inhibited neuronal apoptosis. The knockout of MFG-E8 in SCs can reverse the anti-inflammatory effects of SCDEs treatment. The SOCS3/STAT3 signaling pathway was identified to participate in upregulating M2 polarization induced by MFG-E8. In conclusion, our findings will enrich the mechanism of SCDEs in repairing SCI and provide potential applications and new insights for the clinical translation of SCDEs treatment for SCI.

Fig. 1. Identification and uptake of SCDEs.

A Schematic diagram of BMDMs and SCDEs. B Morphology of SCDEs under TEM. Scale bars: 100 nm. C The diameter distribution of SCDEs. D Western blot analysis of the positive expression of Alix, CD9, and CD63 in SCDEs. E Representative images of S100 (red). Nuclei were labeled with DAPI (blue). Scale bars: 20 µm. F Representative images of F4/80 (red) and CD11b (green). Nuclei were labeled with DAPI (blue). Scale bars: 20 µm. G Representative images of PKH26 (red)-labeled SCDEs absorbed by BMDMs. Nuclei were labeled with DAPI (blue). Scale bars: 20 µm.

Fig. 2. The regulation of SCDEs on M1/M2 polarization in LPS-stimulated BMDMs.

A Representative images of iNOS (red) and F4/80 (green). Nuclei were labeled with DAPI (blue). Scale bar: 20 μm. B Representative western blots showing the reduction of iNOS. C The image analysis results were presented as the relative mean intensity of the fluorescence of iNOS (n = 5). D Quantitative analysis of the iNOS/GAPDH ratio (n = 3). E Representative images of CD206 (red) and F4/80 (green). Nuclei were labeled with DAPI (blue) in each group (n = 5). Scale bar: 20 μm. F Representative western blots showing the increase of CD206. G The image analysis results were presented as the relative mean intensity of the fluorescence of CD206 (n = 5). H Quantitative analysis of the CD206/GAPDH ratio (n = 3). I–J Flow cytometry assay detected CD86 + M1 phenotype, indicating that SCDEs treatment suppressed the M1 polarization in LPS-stimulated BMDMs (n = 3). K, L Flow cytometry assay detected CD206 + M2 phenotype, showing that SCDEs treatment promoted the M2 polarization in LPS-stimulated BMDMs (n = 3). Data were presented as mean ± SEM. Results were analyzed by One-way ANOVA. Significance: *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 3. Function recovery of rats by SCDEs after SCI in vivo.

A The Flow chart of animal experiments. B, F The BBB scores of Sham, SCI group, and SCI rat group treated with SCDEs (n = 5). C Representative paw step images and limbs’ supporting timing view of CatWalk gait analysis. D The H&E staining of the bladder showed a lesser thickness of the bladder in the SCI rat treated with SCDEs. Scale bar: 500 μm. E, H Analysis of motor evoked potential (MEP) was performed as an electrophysiological assessment in both groups at day 28 post-injury in both groups (n = 5). G The wall thickness of the bladder in each group, showing that the bladder function recovered faster in the SCDEs treatment group (n = 5). I Quantitative analysis of catwalk at day 28 post-injury, including regularity index, print position, and stances of the hindlimb (n = 5, RH: right hindlimb, LH: left hindlimb). Data were presented as mean ± SEM. Results were analyzed by One-way ANOVA. Significance: *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 4. SCDEs improved spinal cord repair after SCI in vivo.

A Representative images of immunofluorescence staining of ED1 (red) and GFAP (green) in each group at day 3 after SCI (n = 3). Scale bars: 1000 µm. The bottom row of images were high-resolution versions of the boxed regions in the top row of images. Scale bars: 50 µm. B Representative images of immunofluorescence staining of GFAP (red) and NF-200(green) in each group at day 14 after SCI (n = 3). Scale bars: 1000 µm. The right images were high-resolution versions of the boxed regions in the left images. Scale bars: 20 µm. C Representative images of immunofluorescence staining of ChAT(red) in the SCI group and SCI + SCDEs group (n = 3). Scale bars: 20 µm. D The H&E staining of the spinal cord showed the reduction of the damaged area in the SCI rat treated with SCDEs. E Quantitative analysis of the number of ED1 + cells, showing that being treated with SCDEs can reduce the number of ED1 + cells in the rostral, lesion, and caudal. F Quantitative analysis of the relative mean intensity of the fluorescence of NF200. G Quantitative analysis of the number of ChAT+ cells, showing that the number of ChAT+ cells in the SCI rat treated with SCDEs was higher. Data were presented as mean ± SEM. Results were analyzed by Student’s t-test. Significance: *P < 0.05, **P < 0.01.

Fig. 5. The effect of SCDEs on the regulation of M1/M2 polarization in vivo.

A Representative images of PKH26 (red)-labeled SCDEs absorbed by macrophages/microglia in vivo. Scale bars: 20 µm. The right images were high-resolution versions of the boxed regions in the left images. Scale bars: 2 µm. B Representative immunofluorescence staining images of iNOS (green) and ED1 (red). Nuclei were labeled with DAPI (blue) in each group (n = 3). Scale bar: 20 μm. D Quantitative analysis of the ED1+ cells and iNOS+ cells (n = 3). F Representative western blots showing the reduction of iNOS in vivo (n = 3). C Representative immunofluorescence staining images of CD206 (green) and ED1 (red). Nuclei were labeled with DAPI (blue) in each group (n = 3). Scale bar: 20 μm. E Quantitative analysis of the ED1-positive cells and CD206-positive cells (n = 3). G Representative western blots showing the increase of CD206 in vivo (n = 3). Data were presented as mean ± SEM. Results were analyzed by One-way ANOVA. Significance: *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 6. MFG-E8 was the key component of SCDEs in the regulation of inflammatory responses.

A The differentially expressed proteins between SCDEs and SCs were analyzed by proteomics, and the up-regulated or down-regulated proteins of SCDEs were screened and compared with SCs. B Representative western blots showing the knockout of MFG-E8 in the SCs (shNC) and SCs (shMFG-E8-KO) group. C Quantitative analysis of the MFG-E8/VINCULIN ratio in the SCs (shNC) and SCs (shMFG-E8-KO) group (n = 3). D Representative fluorescence images of SCs (green) after the MFG-E8 gene was knocked out. Scale bar: 100 μm. E Representative immunofluorescence staining images of Iba-1 (red) and MFG-E8 (green). Nuclei were labeled with DAPI (blue) in each group (n = 5). Scale bar: 20 μm. F The image analysis results were presented as the relative mean intensity of the fluorescence of MFG-E8. Data were presented as mean ± SEM. Results were analyzed by Student’s t-test and One-way ANOVA. Significance: **P < 0.01, ***P < 0.001.

Fig. 7. MFG-E8 knockout suppressed the M2 polarization in vitro and vivo.

A Representative immunofluorescence staining images of iNOS (red) and F4/80 (green) in vitro. Nuclei were labeled with DAPI (blue) in each group (n = 5). Scale bar: 20 μm. B Representative immunofluorescence staining images of CD206 (red) and F4/80 (green) in vitro. Nuclei were labeled with DAPI (blue) in each group (n = 5). Scale bar: 20 μm. C, D The results of the iNOS and CD206 relative mean intensity of the fluorescence (n = 5). E Representative immunofluorescence staining images of iNOS (green) and ED1 (red) in vivo. Nuclei were labeled with DAPI (blue) in each group (n = 3). Scale bar: 20 μm. F Representative immunofluorescence staining images of CD206 (green) and ED1 (red) in vivo. Nuclei were labeled with DAPI (blue) in each group (n = 3). Scale bar: 20 μm. G, H Quantitative analysis of the positive cells of ED1, iNOS, and CD206 (n = 3). I, J Flow cytometry assay detected CD86+/CD11b + M1 phenotype, indicating that MFG-E8 knockout promoted the M1 polarization (n = 3). K, L Flow cytometry assay detected CD206+/CD11b + M2 phenotype, showing that MFG-E8 knockout inhibited the M2 polarization (n = 3). Data were presented as mean ± SEM. Results were analyzed by One-way ANOVA. Significance: *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 8. Improved inflammatory microenvironment inhibited neuronal apoptosis.

A Schematic representation of co-culture of PC12 cells and BMDMs. B, E Representative western blots of Cleaved Caspase-3 and Bcl-2 in PC12 cells. Quantitative analysis of the Cleaved Caspase-3/GAPDH ratio and Bcl-2/GAPDH ratio in PC12 cells (n = 3). C, F Flow cytometry assay detected apoptosis of PC12 cells in each group, showing that SCDEs inhibited LPS-induced apoptosis and that the knockout of MFG-E8 reversed the positive effect (n = 3). D, G The TUNEL staining of PC12 cells showed that SCDEs reduced TUNEL + cells while MFG-E8 knockout increased TUNEL + cells (n = 3). Scale bar: 50 μm. Data were presented as mean ± SEM. Results were analyzed by One-way ANOVA. Significance: ns-not significant, *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 9. M2 polarization induced by MFG-E8 in SCDEs was upregulated via SOCS3/STAT3 signaling pathway.

A Representative western blots showed iNOS was reduced, and CD206 was increased, while MFG-E8 knockout reversed the effects of SCDEs in vitro. B Quantitative analysis of the iNOS/GAPDH and CD206/GAPDH ratio (n = 3). C Representative western blots showing the reduction of iNOS and the increase of CD206, while MFG-E8 knockout reversed the effects of SCDEs in vivo. Representative western blots showed that SOCS3/STAT3 signaling was activated via MFG-E8 in SCDEs, while MFG-E8 knockout reduced the SOCS3 and increased the STAT3 in vivo. D Quantitative analysis of the iNOS/GAPDH and CD206/GAPDH ratio in vivo (n = 3). E Quantitative analysis of the MFG-E8/GAPDH, SOCS3/GAPDH, and STAT3/GAPDH ratio and CD206/GAPDH ratio in vivo (n = 3). Data were presented as mean ± SEM. Results were analyzed by One-way ANOVA. Significance: ns-not significant, *P < 0.05, **P < 0.01, ***P < 0.001.

Fig. 10. The mechanism of SCDEs containing MFG-E8 modify macrophage/microglial polarization through SOCS3/STAT3 pathway after SCI.

Spinal cord injury induced activation of STAT3 phosphorylation in macrophages/microglia, led to M1 polarization and exacerbated inflammatory damage at the site of spinal cord injury. After treatment with SCDEs, they were phagocytosed by macrophages/microglia at the injury site and began to release MFG-E8. This increased the protein levels of SOCS3, which inhibited the phosphorylation of STAT3 to promote the conversion of the macrophage/microglia phenotype to M2 for alleviating inflammatory damage. Moreover, the improved inflammatory microenvironment at the SCI site reduced neuronal apoptosis and promoted nerve regeneration.