Mesenchymal stem cell-derived extracellular vesicles subvert Th17 cells by destabilizing RORγt through posttranslational modification

Abstract

Mesenchymal stem cell (MSC)-derived small extracellular vesicles (MSC-sEVs) are known to exert immunosuppressive functions. This study showed that MSC-sEVs specifically convert T helper 17 (Th17) cells into IL-17 low-producer (ex-Th17) cells by degrading RAR-related orphan receptor γt (RORγt) at the protein level. In experimental autoimmune encephalomyelitis (EAE)-induced mice, treatment with MSC-sEVs was found to not only ameliorate clinical symptoms but also to reduce the number of Th17 cells in draining lymph nodes and the central nervous system. MSC-sEVs were found to destabilize RORγt by K63 deubiquitination and deacetylation, which was attributed to the EP300-interacting inhibitor of differentiation 3 (Eid3) contained in the MSC-sEVs. Small extracellular vesicles isolated from the Eid3 knockdown MSCs by Eid3-shRNA failed to downregulate RORγt. Moreover, forced expression of Eid3 by gene transfection was found to significantly decrease the protein level of RORγt in Th17 cells. Altogether, this study reveals the novel immunosuppressive mechanisms of MSC-sEVs, which suggests the feasibility of MSC-sEVs as an attractive therapeutic tool for curing Th17-mediated inflammatory diseases.

Fig. 1. MSC-sEVs were characterized and were taken up by T cells.

A Transmission electron microscopy (TEM) images of MSC-sEVs. The sEVs were observed with low power (20,000×, left) and high power (60,000×, right). sEVs exhibited a lipid bilayer structure and cup-shaped morphology. Scale bars, 1 µm (left) and 500 nm (right). B sEVs were visualized under a confocal microscope. sEVs were detected with confocal microscopy by Alexa Fluor 555 bound to the murine CD9 and DIC (differential interference contrast) channel (10,000×). C Immunoblots of MSC-sEVs lysates against Alix, CD9, and calnexin. The EV-specific markers Alix and CD9 were detected in proportion to the number of proteins. In contrast, calnexin, an ER membrane-bound protein, was not detected in sEVs. D Determination of the specific gravity. MSC-sEVs were resuspended in 2.5 M sucrose-containing HEPES buffer before ultracentrifugation. A 0.25–2 M sucrose gradient was applied to the suspension (fraction 1–6 = 1.05–1.19 g/ml; 1 = 1.05 g/ml, 2 = 1.10 g/ml, 3 = 1.13 g/ml, 4 = 1.15 g/ml, 5 = 1.1.7 g/ml, and 6 = 1.19 g/ml). The range of the specific density of sEVs measured by sucrose density gradient centrifugation corresponded to 1.15–1.19 g/ml. E Evaluation of EV internalization into T cells. The CD4+ T cells were cocultured with MSC-sEVs labeled with 5 µM CFSE and were trypsinized to remove surface-bound sEVs. The uptake of MSC-sEVs by T cells at different time points was analyzed by FACS. Representative results are shown here. F Confocal microscopy image of MSC-sEVs. CD4+ T cells were incubated with GFP-tagged MSC-sEVs for 17 h, and trypsin was added to remove surface-bound sEVs before observation. Z-axis serial sections of confocal microscopy revealed the ingestion of MSC-sEVs by T cells.

Fig. 2. MSC-sEVs specifically inhibited Th17 cells by targeting RORγt.

A Effect of MSC-sEVs on Th1, Th17, and Treg cell differentiation. Naïve CD4+ T cells differentiated into Th1, Th17, and Treg cells in the presence of MSC-sEVs. The expression of transcription factors (T-bet for Th1, RORγt for Th17, and Foxp3 for Treg) was analyzed by FACS. MSC-sEVs significantly reduced the number of RORγt+ T cells but not Foxp3+ and T-bet+ T cells. B Effect of MSC-sEVs on the cytokine secretion by Th17 cells. The levels of IL-17A, IL-4, IFN-γ, and IL-6 secreted from Th17 cells cocultured with MSC-sEVs were measured by the cytometric bead array (CBA) assay. Treatment with sEVs decreased the production of IL-17 but not other cytokines, including IFN-γ, IL-4, and IL-6, by Th17 cells. C Analysis of RORγt expression levels after Th17 differentiation in the presence of MSC-sEVs. sEVs-depleted culture serum (EDCS) and sEVs were used to treat differentiated Th17 cells. EDCS failed to reduce the number of RORγt+ T cells, indicating that MSC-sEVs were key players in the suppression of Th17 differentiation. D, E Assessment of the proliferation level of Th1 and Th17 cells treated with MSC-sEVs. Naive CD4+ T cells were labeled with CFSE (5 µM, 5 min) and differentiated into Th1 and Th17 cells under MSC-sEVs treatment conditions. sEVs specifically suppressed the differentiation of Th17 cells, and their mechanisms were independent of cell cycle arrest. A representative image from the three independent assays is shown in the dot plot. The percentages of RORγt- and T-bet-positive cells among the proliferating cells are shown in the plot. The mean of triplicates is displayed as a bar graph, and the error bars indicate SEM. ****p ≤ 0.0001.

Fig. 3. MSC-sEVs destabilized RORγt at the protein level in Th17 cells.

A The phosphorylation level of STAT3 examined by immunoblotting. The normalized level of phospho-STAT3 (P-STAT3) was measured using individual band intensity (p-STAT3/STAT3 ratio, right panel). MSC-sEVs did not affect the phosphorylation of STAT3, indicating that MSC-sEVs regulate the expression of RORγt downstream of STAT3 phosphorylation. Representative images from the three independent immunoblots are shown. The mean of triplicates is shown as a bar graph, and the error bars indicate SEM. ***p ≤ 0.001. B Mean fluorescence intensity (MFI) of RORγt in MSC-sEVs-treated Th17 cells. MSC-sEVs were added at various time points after the initiation of Th17 differentiation (5, 24, 48, 72, and 96 h). The MFI of RORγt assessed by flow cytometry at 120 h of differentiation significantly decreased. C RORc mRNA expression levels were confirmed using real-time PCR. D Cycloheximide chase analysis of RORγt protein degradation. MSC-sEVs and 0.5 μg/ml cycloheximide were added to the T cells at 12, 24, and 48 h before the endpoint of Th17 differentiation (5 days). Zero hours indicates the positive control, which was not treated with MSC-sEVs and cycloheximide. Western blotting was conducted to visualize the expression level of RORγt. MSC-sEVs significantly reduced RORγt in the absence of de novo translated RORγt. A representative image from the three independent western blots. The RORγt/GAPDH ratio is shown as a bar graph. *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001. E Assessment of IL-17 expression levels using intracellular cytokine staining of MSC-sEVs-treated Th17 cells. MFI of IL-17A expressed in differentiated Th17 cells is shown with varying MSC-sEVs treatment time points. Reduction of RORγt by MSC-sEVs treatment resulted in a decrease in IL-17 production by Th17 cells. ****p ≤ 0.0001. F The ratio of RORγt (+) CD4+ T cells under Th17 polarizing conditions following treatment with MSC-sEVs. MG132 (2 μM). The proteasome inhibitor inhibited the reduction in the ratio of RORγt (+) cells mediated by MSC-sEVs. The mean of triplicates is shown as a bar graph, and the error bars indicate SEM. G Immunoblot to detect the K63-linked polyubiquitination of RORγt in Th17 cells. MG132 (2 μM) was added to prevent the proteolytic degradation of RORγt. RORγt was pulled down from MSC-sEVs-treated Th17 cell lysates using an anti-RORγt antibody. Immunoprecipitates were separated by SDS‒PAGE, transferred to PVDF membranes, and probed with anti-RORγt and anti-K63 ubiquitin antibodies. MSC-sEVs treatment destabilized RORγt through K63-deubiquitination.

Fig. 4. Eid3 from MSC-sEVs destabilized RORγt in Th17 cells.

A Volcano plot of the differentially expressed mRNAs in MSC-sEVs-treated Th17 cells compared with untreated control Th17 cells. The dashed horizontal line indicates a p value of 0.05. The Eid3 gene, proteasome-related genes, and the other genes are colored red, blue, and gray, respectively. B The expression level of Eid3 in MSC-sEVs-treated T cells. The protein level of Eid3 was increased in MSC-sEVs-treated T cells. A representative image from three independent immunoblots is shown. The eid3/actin ratio is shown as a bar graph. No sti: without CD3/CD28 stimulation; Sti: with CD3/CD28 stimulation. ***p ≤ 0.001. C The protein level of Eid3 is expressed in each cell lysate. The relative expression of Eid3 in MSC-sEVs was the highest, and a negligible amount of Eid3 was observed in T cells and splenocytes, meaning that Eid3 was not intrinsic to activated T cells but came from MSC-sEVs. A representative image from three independent immunoblots is shown. The Eid3/actin ratio is shown as a bar graph. hADSC: human adipose-derived stem cell. ***p ≤ 0.001; ****p ≤ 0.0001. D mRNA level of Eid3 by real-time PCR (normalized by GAPDH). MSCs and MSC-sEVs both contained Eid3 mRNA. NSC: neural stem cell; MSC: mesenchymal stem cell. ***p ≤ 0.001. E Expression of RORγt after transduction of lentiviral Eid3-ORF (GFP-tagged) into Th17 cells (5 MOI). A lentiviral control-ORF particle (GFP-tagged) was used as a control (Lenti-control, 5 MOI). The downregulation of RORγt expression was also found in the lentiviral Eid3-ORF-transduced group. F The expression level of Eid3 in MSCs, Eid3-knockdown MSCs, MSC-derived sEVs, and Eid3-knockdown MSC-derived sEVs. Knockdown of Eid3 was performed by transducing the lentiviral Eid3-shRNA construct and was confirmed by western blotting. G Expression of RORγt in naive CD4+T cells, Th17 cells, MSC-sEVs-treated Th17 cells, Eid3-knockdown MSC-sEVs-treated Th17 cells, and C646-treated Th17 cells. The expression level of RORγt was assessed by immunoprecipitation with an anti-RORγt antibody. Eid3-knockdown MSC-derived sEVs failed to reduce the expression of RORγt. H Expression of p300 and RORγt in Th17 cells and MSC-sEVs-treated Th17 cells detected by western blotting. The expression levels of both p300 and RORγt were significantly decreased in MSC-sEVs-treated Th17 cells. A representative image from three independent immunoblots is shown. The p300/GAPDH and RORγt/GAPDH ratios are shown as bar graphs. *p ≤ 0.05; **p ≤ 0.01. I Regulation of p300 and RORγt by Eid3. Immunoprecipitation was conducted using an anti-p300 antibody, and immunoblotting was conducted using anti-p300, Eid3, and RORγt antibodies. The upregulated expression of Eid3 by lentiviral transduction or delivery of Eid3 through MSC-sEVs downregulated both p300 and RORγt. J Assessment of the polyubiquitination of K63 and the acetylation of RORγt by coimmunoprecipitation. Thymocytes treated with MSC-sEVs or C646 were immunoprecipitated using an anti-RORγt antibody. Immunoblotting was conducted using anti-p300, K63 polyubiquitin, acetyl-K, and RORγt antibodies. Lysate refers to thymocyte lysate without immunoprecipitation used as a positive control. Eid3 derived from MSC-sEVs destabilized RORγt by suppressing K63 polyubiquitination and acetylation through the inhibition of p300.

Fig. 5. Injection of MSC-sEVs in mice ameliorates EAE.

A Effect of MSC-sEVs treatment in a mouse EAE model. EAE was induced by MOG/CFA coinjected with pertussis toxin (PTx). Here, 25 μg of MSC-sEVs or sEVs-depleted supernatant (EDCS) was intravenously injected into EAE mice once a day for 7 days. Clinical scores of the diseased mice showed no change in the EDCS-treated group (n = 8) and a decrease in the sEVs-treated group (n = 6). The average body weights of the control, EAE mice, and EAE mice treated with MSC-sEVs were measured. The body weight increased in the control group (n = 3) and decreased in the EAE group (n = 3). Treatment with sEVs partially compensated for the weight loss from EAE (n = 3). ****p ≤ 0.0001. B Changes in the number of IL-17-, IFN-γ-, and IL-6-producing splenocytes (1 × 10⁵) from EAE mice analyzed by the enzyme-linked immunospot (ELISpot) assay. IL-17-producing splenocytes were significantly reduced in the MSC-sEVs-treated group but not in the EDCS-treated group. The means of triplicate experiments are shown, and the error bars indicate SEM. ***p ≤ 0.001. C Expression of secreted IL-17A, IFN-γ, and IL-6 from EAE-induced splenocytes (5 × 10⁵) restimulated with MOG measured by the Cytometry Bead Assay (CBA). IL-17 secretion by splenocytes was significantly reduced in the MSC-sEVs-treated group but not in the EDCS-treated group. *p ≤ 0.05. D H&E staining of mouse brain tissue filtration. Infiltration of CD3+ T cells was decreased in MSC-sEVs-treated EAE mice compared to EAE mice without MSC-sEVs treatment. E Flow cytometry analysis data showing a significant decrease in CNS tissue infiltrating CD45+ leukocytes in the MSC-sEV-treated group. F Flow cytometry analysis of the expression of RORγt and IL-17A in CNS-infiltrated CD4+ T cells. The expression of RORγt and IL-17A in CNS-infiltrated CD4+ T cells was downregulated in the MSC-sEVs-treated EAE group. G Size of the secondary lymphoid organs (spleen and lymph node) from WT, EAE, and EAE treated with the MSC-sEVs mice. H Analysis of the expression of RORγt and IL-17A among the CD4+ T cells in the draining lymph nodes from the WT, EAE, and EAE treated with the MSC-sEVs groups by flow cytometry. Compared to the control group, RORγt and IL-17A-producing CD4+ T cells were significantly decreased in the MSC-sEVs-treated EAE group.

Fig. 6. Schematic explanation of the underlying mechanism of MSC-derived sEVs in Th17 cell suppression.

RORγt is stabilized by K63 polyubiquitination and acetylation by p300. Eid3 derived from MSC-sEVs suppresses the K63-linked polyubiquitination and acetylation of RORγt by p300, resulting in proteolytic degradation of RORγt. As a result, MSC-sEVs inhibit RORγt expression and IL-17 production in Th17 cells. MSC-sEVs exert their therapeutic effects by degrading RORγt at the protein level, resulting in the depolarization of Th17 cells. From this unique finding, MSC-sEVs are suggested to be useful in treating Th17-associated autoimmune diseases.