Circulating exosomes suppress the induction of regulatory T cells via let-7i in multiple sclerosis

Abstract

Multiple sclerosis (MS) is a T cell-mediated autoimmune disease of the central nervous system. Foxp3⁺ regulatory T (Treg) cells are reduced in frequency and dysfunctional in patients with MS, but the underlying mechanisms of this deficiency are unclear. Here, we show that induction of human IFN-γ^-IL-17A^-Foxp3⁺CD4⁺ T cells is inhibited in the presence of circulating exosomes from patients with MS. The exosomal miRNA profile of patients with MS differs from that of healthy controls, and let-7i, which is markedly increased in patients with MS, suppresses induction of Treg cells by targeting insulin like growth factor 1 receptor (IGF1R) and transforming growth factor beta receptor 1 (TGFBR1). Consistently, the expression of IGF1R and TGFBR1 on circulating naive CD4⁺ T cells is reduced in patients with MS. Thus, our study shows that exosomal let-7i regulates MS pathogenesis by blocking the IGF1R/TGFBR1 pathway.

Fig. 1

Exosomes from patients with MS can selectively decrease the frequency of Treg cells a Representative size distribution of purified exosomes. Using a NanoSight LM10 nanoparticle analysis system, the size was analysed three times for each sample. Red error bars indicate the standard error of the mean. b Western blot analysis for CD9, CD63 and Cytochrome c proteins. The PBMC and exosome samples were collected from HC. Each lane was loaded with 3 μg of protein for blotting. c Dot plot and histogram of flow cytometry data. T cells were prepared from the peripheral blood of a healthy volunteer. They were cultured with PBS as a control or with exosomes derived from HC (HC-exosome) or patients with MS (MS-exosome) under stimulation with anti-CD3 and anti-CD28 mAbs for 48 h. IFN-γ⁻IL-17A⁻CD4⁺ T cells were defined as shown in the left panel, and then the expression of Foxp3 was evaluated as shown in the right panel. d The frequencies of inflammatory and regulatory T cells among CD4⁺ T cells after the culture described above. Among Foxp3⁺CD4⁺ T cells, IFN-γ⁻IL-17A⁻Foxp3⁺CD4⁺ T cells are known to represent most effective Treg population^{, ,} , whereas Foxp3⁺CD4⁺ T cells secreting IFN-γ or IL-17A are supposed to be dysfunctional in the suppressive activity^{, –}. Data are representative of two independent experiments. A one-way ANOVA with Bonferroni’s comparison test was used for statistical analysis. Error bars represent the mean ± s.d. *p < 0.05, **p < 0.01. s.d. standard deviation, n.s. not significant, ANOVA analysis of variance, Cyt c cytochrome c, PBMC peripheral blood mononuclear cell, EV extracellular vesicle, PBS phosphate-buffered saline

Fig. 2

Exosomal miRNA profile differentiates patients with MS from healthy controls a Heat map of miRNA expression profile in the exosomes from HC and patients with MS. RNA was extracted from exosomes, which were isolated from the plasma of four HC and four patients with MS. A 3D-Gene Human miRNA Oligo chip (TORAY) was used for microarray analysis. MiRNAs with assigned identification number lower than 500 and signal intensity higher than 100 were selected and clustered based on the expression patterns of the eight samples. b Volcano plot of miRNAs in the exosomes. The expression difference of each miRNA between MS-exosome and HC-exosome is plotted on the X axis in log2 scale. p value of the difference by t test is plotted on the Y axis. Blue dots represent miRNAs with signal intensity higher than 100. Green dots represent miRNAs with identification number lower than 500 and signal intensity higher than 100. Four miRNAs pointed by arrows (red dots) were identified as candidate miRNAs upregulated in patients with MS. They were selected from green dots. c PC analysis for the MS and HC samples based on the miRNA expression profiles, using the same data as in b. d,e Quantification of the candidate miRNAs by RT-qPCR. The total amount of RNAs in the exosomes from the same amount of plasma was also examined. Based on the clinical information, patients with MS were divided into those with RRMS or SPMS and into those in remission or relapse phase. An unpaired t test was used for statistical analysis. Error bars represent the mean ± s.d. s.d. standard deviation, n.s. not significant, PC principal component, A.U. arbitrary unit, RT-qPCR reverse transcription quantitative polymerase chain reaction, RRMS relapsing-remitting MS, SPMS secondary-progressive MS

Fig. 3

Expression of let-7i in exogenously added exosomes negatively correlates with Treg cell frequency. a Correlation between the frequency of IFN-γ⁻IL-17A⁻Foxp3⁺CD4⁺ Treg cells and exosomal miRNAs. The same experiment as that described in the legend for Fig. 1 was repeated with a different set of primary T cells and exosome samples. Along with the evaluation of the effect of exogenous exosome on Treg cells in the culture, we also quantified the expression levels of miRNA in each added exosome sample. Here, we analysed the relationship between the miRNA levels and the frequency of IFN-γ⁻IL-17A⁻Foxp3⁺CD4⁺ Treg cells. b The amount of total RNA. We also measured total RNA in the exosomes added to the cultures, showing no significant difference between HC and patients with MS. c Expression levels of mature let-7i and primary let-7i were evaluated by RT-qPCR in the T cells after culture with PBS, HC-exosome or MS-exosome. Pearson’s correlation analysis was used in a, and an unpaired t test was used in b and c for statistical analysis. Error bars represent the mean ± s.d. s.d. standard deviation, n.s. not significant, A.U. arbitrary unit, RT-qPCR reverse transcription quantitative polymerase chain reaction, PBS phosphate-buffered saline

Fig. 4

Let-7i transfection decreases the frequency of Treg cells. a T cells were transfected with let-7i, miR-19b, miR-25 or miR-92a and then cultured under stimulation with anti-CD3 and anti-CD28 mAbs for 72 h. The phenotype was determined by flow cytometry based on cytokine expression (IFN-γ, IL-17A) and transcription factor expression (Foxp3). Data are representative of two independent experiments. b T cells were transfected with let-7i inhibitor and cultured in the presence of MS-exosome or HC-exosome under stimulation with anti-CD3 and anti-CD28 mAbs for 72 h. The frequencies of IFN-γ⁻IL-17A⁻Foxp3⁺CD4⁺ Treg cells were analysed and the rates of increase by let-7i inhibitor were plotted. Data are representative of two independent experiments. n = 6 in each group in a, n = 6 in each group with exosomes and n = 3 in the groups with PBS in b. A one-way ANOVA with Dunnett’s comparison test was used in a, and an unpaired t test was used in b for statistical analysis. Error bars represent the mean ± s.e.m. in a, and the mean ± s.d. in b. *p < 0.05. s.e.m. standard error of the mean, s.d. standard deviation, n.s. not significant, ANOVA analysis of variance, PBS phosphate-buffered saline

Fig. 5

Differentiation of Treg cells from naive CD4⁺ T cells is inhibited by MS-exosome and let-7i. a Gating strategy for isolation of naive CD4⁺ T cells (CD45RA⁺CD25⁻), memory CD4⁺ T cells (CD45RA⁻CD25⁻), resting Treg cells (CD45RA⁺CD25⁺) and activated Treg cells (CD45RA⁻CD25^high). b Each population sorted as in a was cultured with HC-exosome or MS-exosome under stimulation with anti-CD3 and anti-CD28 mAbs for 72 h. TGFβ (1 ng/mL) and IL-2 (50 U/mL) were added for the culture of naive CD4⁺ T cells. The frequency of IFN-γ⁻IL-17A⁻Foxp3⁺CD4⁺ T cells was analysed for cultured naive and memory CD4⁺ T cells. The frequencies of dead and proliferated cells were analysed for cultured resting and activated Treg cells. c Each population sorted as in a was transfected with let-7i or a negative control, and cultured in the same way as in b. Data are representative of two independent experiments. d Naive CD4⁺ T cells were sorted from the peripheral blood of HC or patients with MS. Mature and primary let-7i were quantified by RT-qPCR. Correlation between the amount of let-7i in exosomes in the blood and the ratio of mature to primary let-7i in naive CD4⁺ T cells was analysed. n = 3 in each group in c. An unpaired t test was used in b, c and d, and Pearson’s correlation analysis was used in d for statistical analysis. Error bars represent the mean ± s.d. in b and d, and the mean ± s.e.m. in c. s.d. standard deviation, s.e.m. standard error of the mean, n.s. not significant, A.U. arbitrary unit, rTreg resting regulatory T cells, aTreg activated regulatory T cells, RT-qPCR reverse transcription quantitative polymerase chain reaction

Fig. 6

Let-7i transfection decreases expression of IGF1R and TGFBR1 on CD4⁺ T cells. a Induction of Treg cells in the presence of IGF1, TGFβ and IL-2. Naive CD4⁺ T cells were cultured in the presence of various combinations of IGF1 (10 ng/mL), TGFβ (1 ng/mL) and IL-2 (50 U/mL) under stimulation with anti-CD3 and anti-CD28 mAbs for 72 h. The frequency of IFN-γ⁻IL-17A⁻Foxp3⁺CD4⁺ Treg cells among CD4⁺ T cells was determined. b T cells were transfected with let-7i and then cultured under stimulation with anti-CD3 and anti-CD28 mAbs for 72 h. Representative histograms of the expression levels of IGF1R and TGFBR1 are shown. The expression levels of IGF1R and TGFBR1 were determined by measuring the frequency of each receptor-positive cells among CD4⁺ T cells and the MFI of each receptor on CD4⁺ T cells. c Naive CD4⁺ T cells were transfected with siRNAs targeting TGFBR1 and IGF1R as indicated, and then differentiated towards Treg cells with TGFβ (1 ng/mL) and IL-2 (50 U/mL) in addition to stimulation with anti-CD3 and anti-CD28 mAbs for 72 h. The frequency of Treg cells was evaluated. Data are representative of two independent experiments. n = 3 a or 4 b and c in each group. An unpaired t test was used in a and b, and a one-way ANOVA with Dunnett’s comparison test was used in c for statistical analysis. Error bars represent the mean ± s.e.m. **p < 0.01, ***p < 0.001. s.e.m. standard error of the mean, A.U. arbitrary unit, MFI mean fluorescence intensity, ANOVA analysis of variance, IGF1R insulin like growth factor 1 receptor, TGFBR1 transforming growth factor beta receptor 1

Fig. 7

SiRNA-mediated knockdown of IGF1R and TGFBR1 decreases Treg cell frequency. a T cells were transfected with three different siRNAs targeting IGF1R, and then cultured under stimulation with anti-CD3 and anti-CD28 mAbs for 72 h. The expression of IGF1R was assessed by the MFI of the receptor on CD4⁺ T cells. The frequencies of inflammatory and regulatory T cells among CD4⁺ T cells were evaluated. b T cells were transfected with two siRNAs targeting TGFBR1, with or without one targeting IGF1R, and cultured in the same manner as in a. The expression of TGFBR1 and the frequencies of inflammatory and regulatory T cells among CD4⁺ T cells were evaluated similarly. Data are representative of two independent experiments. n = 4 in each group. A one-way ANOVA with Dunnett’s comparison test was used for statistical analysis. Error bars represent the mean ± s.e.m. *p < 0.05, **p < 0.01, ***p < 0.001. s.e.m. standard error of the mean, A.U. arbitrary unit, MFI mean fluorescence intensity, ANOVA analysis of variance, IGF1R insulin like growth factor 1 receptor, TGFBR1 transforming growth factor beta receptor 1

Fig. 8

Treg cell frequency and expression of IGF1R and TGFBR1 by naive CD4⁺ T cells is decreased in MS. a The frequency of IFN-γ⁻IL-17A⁻Foxp3⁺ Treg cells among memory CD4⁺ T cells in circulation. b Representative histogram showing the expression levels of IGF1R and TGFBR1 on CD45RA⁺ naive CD4⁺ T cells. c The expression levels of IGF1R and TGFBR1 were assessed by the MFI of the receptors on circulating CD4⁺ T cells. d The frequency of IFN-γ⁻IL-17A⁻Foxp3⁺ Treg cells and the expression of TGFBR1 and IGF1R on naive CD4⁺ T cells were analysed between the groups with lower and higher amounts of let-7i in exosomes in the blood. The groups were determined so as to be numerically equal. Correlation analysis was also performed. e Correlation analysis between the amount of let-7i in naive CD4⁺ T cells and the frequency of IFN-γ⁻IL-17A⁻Foxp3⁺ Treg cells or the expression of TGFBR1 or IGF1R on naive CD4⁺ T cells in patients with MS and HC. f Correlation analysis between the frequency of IFN-γ⁻IL-17A⁻Foxp3⁺ Treg cells and the expression of TGFBR1 or IGF1R on naive CD4⁺ T cells in patients with MS and HC. An unpaired t test was used in a, c and d, and Pearson’s correlation analysis was used in d, e and f for statistical analysis. Error bars represent the mean ± s.d. s.d. standard deviation, n.s. not significant, A.U. arbitrary unit, MFI mean fluorescence intensity, IGF1R insulin like growth factor 1 receptor, TGFBR1 transforming growth factor beta receptor 1