Anti-Inflammatory Macrophage-Derived Exosomes Modified With Self-Antigen Peptides for Treatment of Experimental Autoimmune Encephalomyelitis

Abstract

Current treatments for autoimmune diseases often involve broad-acting immunosuppressants, which carry risks such as infections and malignancies. This study investigates whether exosomes derived from anti-inflammatory macrophages (AE) and decorated with myelin oligodendrocyte glycoprotein (MOG) peptide (AE/M) can induce immune tolerance in autoimmune diseases. Experimental autoimmune encephalomyelitis (EAE), a mouse model for multiple sclerosis, serves as the autoimmune disease model. Exosomes derived from myoblasts or fibroblasts are also modified with MOG peptides for comparison. Unlike their myoblast or fibroblast counterparts, exosomes from anti-inflammatory macrophages demonstrate a targeted capacity toward antigen-presenting cells. Moreover, AE/M uniquely promotes the differentiation of dendritic cells (DC) into a tolerogenic phenotype. When splenocytes are treated with AE/M, an increased population of tolerogenic DC (tolDC) is observed, even under proinflammatory stimuli. Subcutaneous administration of AE/M in the EAE mouse model results in MOG peptide-specific immune tolerance and preserves motor coordination. In contrast to treatments with fibroblast- or myoblast-derived exosomes modified with MOG peptides, AE/M treatment provides complete protection from EAE in mice. These findings highlight the potential of self-antigen modified AE as a versatile and adaptable nanoplatform for the treatment of various autoimmune diseases.

Keywords: anti‐inflammatory macrophage‐derived exosome; autoimmune disease; encephalomyelitis; immune tolerance; self‐antigen modification.

Figure 1

Proposed mechanism of AE/M to induce antigen‐specific immune tolerance. A) Anti‐inflammatory macrophages were generated from naïve macrophages using interleukin 4 (IL‐4) stimulation. Subsequently, anti‐inflammatory macrophage‐derived exosomes were isolated from these anti‐inflammatory macrophages. B) MOG peptide, serving as a model self‐antigen, was conjugated to the surface of AE using maleimide‐(PEG)₂‐NHS ester. Cysteine‐modified MOG peptide (Cys‐MOG) was conjugated to the exosomes via the thiol group and maleimide moiety, resulting in the formation of AE/M. C) Exosomes were also isolated from myoblasts (MyE), fibroblasts (FE), and naïve macrophages (NE). Following MOG peptide conjugation, these exosomes were termed MyE/M, FE/M, and NE/M, respectively. D) A schematic illustration outlines the proposed mechanism of action for AE/M in protecting mice from EAE. The treatment with AE/M is proposed to induce MOG‐presenting tolDC. The tolDC migrated to lymph nodes and facilitated the generation of T_reg cells, which play a key role in suppressing autoimmune activation and protecting neurons.

Figure 2

Characterization of MOG‐conjugated exosomes. A) Size and zeta potential of different exosomes (n = 5). B) Size distribution of AE/M obtained by nanoparticle tracking analysis. C) Concentration of exosomes counted through nanoparticle tracking analysis (n = 5). D) 2D AFM image of AE/M. E) 3D AFM image of AE/M. F) The height and length data for the AE/M highlighted in (E). G) Morphology of AE and AE/M was visualized using transmission electron microscopy (TEM). Scale bar: 200 nm. H) A schematic illustration of AE/M‐bound aldehyde/sulfate latex beads which was used for exosome visualization and analysis of exosome surface markers. I) Observation of AE/M‐bound aldehyde/sulfate latex beads by scanning electron microscopy (SEM). AE/M was allowed to be adsorbed on aldehyde/sulfate latex beads followed by SEM observation. Scale bar: 500 nm. J) SEM image of AE/M that was adsorbed on aldehyde/sulfate latex beads. Scale bar: 100 nm. K) Visualization of exosome surface markers by confocal microscopy. Scale bar: 2 µm. Anti‐CD9 antibody, anti‐CD63 antibody, anti‐CD81 antibody, and anti‐LFA‐1 antibody were used to stain exosomes after the adsorption of exosomes onto aldehyde/sulfate latex beads. Primary anti‐MOG antibody was used to stain MOG peptides followed by the second antibody staining for the detection. L) Relative mean fluorescence intensity of exosome surface markers obtained by flow cytometry (n = 5). (Data are presented as the mean ± standard deviation (SD). n.s., not significant; ***p < 0.001).

Figure 3

Induction of tolDC by AE/M. A–D) The mean fluorescence intensity of proinflammatory markers, including TNF‐α A), MHCII B), CD80 C), and CD86 D) was measured (n = 5). E–G) The mean fluorescence intensity of anti‐inflammatory markers, including PD‐L1 E), IL‐10 F), and TGF‐β G) was measured (n = 5). H) Summary of the expression of inflammation‐relevant markers after treatment. I) Visualization of the expression of MHCII, CD86, and PD‐L1 in DC after AE/M treatment via confocal microscopy. Scale bar: 20 µm. J) A schematic illustration showing that AE/M treatment induced tolDC as manifested by the increase in anti‐inflammatory markers and the decrease in proinflammatory markers. (Data are presented as the mean ± SD. **p < 0.01; ***p < 0.001).

Figure 4

Uptake of MOG‐modified exosomes by DC and macrophages. A) Cellular uptake of MOG‐modified exosomes by DC was visualized by confocal microscopy. Scale bar: 40 µm. B) 3D surface plot of DiD‐exosome obtained from confocal images shown in (A). C) Histogram of cellular uptake of MOG‐modified exosomes by DC was assessed by flow cytometry. D) Mean fluorescence intensity of DC uptake of MOG‐modified exosomes (n = 5). E) Cellular uptake of MOG‐modified exosomes by macrophages was visualized by confocal microscopy. Scale bar: 20 µm. F) 3D surface plot of DiD‐exosome obtained from confocal images shown in (E). G) Histogram of macrophage uptake of MOG‐modified exosomes was determined by flow cytometry. H) Mean fluorescence intensity of macrophage uptake of MOG‐modified exosomes (n = 5). (Data are presented as the mean ± SD. ***p < 0.001).

Figure 5

Biodistribution of MOG‐conjugated exosomes. A) The schedule of the biodistribution study is outlined. DiD‐labeled MOG‐conjugated exosomes were administered subcutaneously. At different time points, in vivo fluorescence imaging was performed using the IVIS instrument. Lymph nodes were harvested for further cellular analysis. B) Images show the biodistribution of different MOG‐modified exosomes as determined by the IVIS instrument. C) The fluorescence intensity of exosomes in inguinal lymph nodes was measured at different time points (n = 5). D) Immunofluorescence images show the accumulation of MOG‐modified exosomes in lymph nodes 24 h after administration. Tissue sections were prepared, and immunostaining was conducted using 4′,6‐diamidino‐2‐phenylindole (DAPI) (blue), anti‐CD11c antibody (green), and anti‐F4/80 antibody (gray). The scale bars indicate 0.25 mm for the upper and lower images. E,F) Cellular uptake of MOG‐modified exosomes by DC in lymph nodes: histogram plot E), and mean fluorescence intensity F). DiD‐labeled MOG‐modified exosomes were subcutaneously injected followed by collection of lymph nodes at 24 h. DiD⁺ immune cells were analyzed by flow cytometry (n = 5). G,H) Cellular uptake of MOG‐modified exosomes by macrophages: histogram plot G), and mean fluorescence intensity H) (n = 5). I) t‐SNE visualization of different immune cell populations from lymph nodes isolated from AE/M‐treated mice. J) t‐SNE heatmap statistic of DiD⁺ cells in lymph nodes from AE/M‐treated mice. (Data are presented as the mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001).

Figure 6

In vivo prophylactic effect of AE/M on EAE mice. A) A schematic illustration of the schedule of in vivo study. Two days after EAE model induction (prior to EAE onset), various exosome formulations were administered through subcutaneous injection on days 2, 6, 9, and 13. Clinical score and body weight were recorded. B) Clinical score of the mice treated with different MOG‐conjugated exosomes (n = 5). C) Clinical score at day 15 post‐EAE induction (n = 5). D) The percentage of EAE‐free mice after immunization in each group (n = 5). E) Body weight of the mice treated with different exosome formulations (n = 5). F) Photographs of the mice from the untreated group and the AE/M‐treated group showing walking behavior. Live movies are provided in Video S2 (Supporting Information) (Data are presented as the mean ± SD. n.s., not significant; **p < 0.01; ***p < 0.001).

Figure 7

Induction of antigen‐specific immune tolerance by subcutaneous administration of exosomes. A) A schematic illustration of the in vivo study. Subcutaneous administration of MOG‐conjugated exosomes started 2 days after EAE model induction. The administration interval was 3 or 4 days. 4 weeks after immunization, spleens were harvested for further analysis. B) Photographs of red spots generated by MOG‐responsive splenocytes from mice treated with different formulations. 28 days after EAE model induction, spleens were harvested, and splenocytes were stimulated in vitro with 5 µg mL⁻¹ of MOG_35–55 for 24 h followed by IFN‐γ enzyme‐linked immunosorbent spot (ELISpot) assay. C) The number of IFN‐γ spots generated by MOG‐responsive splenocytes from mice treated with various exosomes (n = 5). D) The concentration of IL‐17 in the medium collected from splenocytes restimulated with 10 µg mL⁻¹ of MOG_35–55 (n = 5). E–J) Immunophenotyping of spleens was performed using flow cytometry. E) Plots show CD86⁺ CD11c⁺ DC in the spleens. F) Plots show CD80⁺ CD11c⁺ DC in the spleens. G) Representative plots show the population of anergic T cells in the spleens following exosome treatment. H) The frequency of CD86⁺ CD11c⁺ DC in the spleens was measured (n = 4). I) The frequency of CD80⁺ CD11c⁺ DC in the spleens was measured (n = 4). J) The percentage of anergic T cells in the spleens was measured (n = 4). (Data are presented as the mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001).

Figure 8

Immune microenvironment of CNS following exosome treatments. A) A schematic illustration of the in vivo study. Subcutaneous administration of MOG‐conjugated exosomes started 2 days after EAE model induction. The administration interval was 3 or 4 days. 4 weeks after immunization, spinal cords were harvested for further analysis. B) Representative flow cytometry plots show the level of CD86 on DC. C) The mean fluorescence intensity of CD86 on DC was measured (n = 4). D) Representative flow cytometry plots show the level of CD80 on DC. E) The mean fluorescence intensity of CD80 on DC was measured (n = 4). F–H) The number of immune cells in 5 × 10⁵ spinal cord cells analyzed by flow cytometry: macrophage number F), CD8⁺ T cell number G), and CD4⁺ T cell number H) (n = 5). At the end of the in vivo experiment, spinal cords were harvested and stained with various antibodies followed by analysis of immune cell profiles by flow cytometry. I) Plots showing IL‐17A⁺ CD4⁺ T cells in spinal cords. Cells from spinal cords were restimulated with 5 µg mL⁻¹ of MOG_35–55 and analyzed by flow cytometry. J) The percentage of IL‐17A⁺ CD4⁺ T cells in spinal cords (n = 3). K) The percentage of IFN‐γ ⁺ CD4⁺ T cells in spinal cords (n = 3). L) The frequency of CD25⁺ FOXP3⁺ T_reg cells among CD4⁺ T cells in spinal cords analyzed by flow cytometry (n = 5). M) Plots showing CD25⁺ FOXP3⁺ T_reg cells among CD4⁺ T cells in spinal cords. (Data are presented as the mean ± SD. *p < 0.05; **p < 0.01; ***p < 0.001).

Figure 9

Infiltration of leukocytes to CNS. A) A schematic illustration depicts the schedule of the in vivo study. Subcutaneous administration of various exosome formulations began 2 days after EAE model induction. At the end of the study, brains and spinal cords were harvested to prepare tissue sections. B) Immunofluorescence images show leukocytes in the midbrain. Immunostaining was performed with DAPI (blue), anti‐MBP antibody (green), and anti‐CD45 antibody (red). Scale bar: 0.5 mm. C) Immunofluorescence images of leukocytes and demyelination in spinal cords, were analyzed using the THUNDER imaging system. Immunostaining was performed with DAPI (blue), anti‐MBP antibody (green), and anti‐CD45 antibody (red). Scale bar: 0.7 mm. D–E) The density of CD45⁺ cells in different CNS regions was analyzed using the Vectra tissue analyzer: midbrain D) and spinal cord E) (n = 10). F) The percentage of MBP⁺ area was calculated by dividing the MBP⁺ area by the total area of the selected tissue regions (n = 10). (Data are presented as the mean ± SD. **p < 0.01; ***p < 0.001).