Fig 2: Inhibition of endogenic EV production and the IBD severity.(a) MC38 cells were treated with the indicated concentration of spiroepoxide for 24 h. The EVs in the supernatant of MC38 cells were absorbed by CD63-coated latex beads and the percentage of CD9-positive latex beads was detected using FACS. (b) The data in a were statistically analysed (n=9). (c) Overall, 1 × 107 MC38 cells were treated with DMSO or 10 μM spiroepoxide for 24 h. The EVs in the culture supernatants were isolated, and the protein measured by a BCA assay and statistically analysed (n=9). (d) Mice were fed with water or 2% DSS solution on day 0 and received intraperitoneal injection of spiroepoxide (2 g kg−1 body weight) or DMSO control. The body weights of mice were measured daily. (e) The mice were killed on day 11 and Li-EVs were isolated and quantified. Quantity of the Li-EVs was normalized according to colon weight and statistically analysed (n=9). (f) Mice were treated as described in d. In addition to spiroepoxide injection, some mice were also intravenously injected with 100 μg A33+ Li-EVs. The body weight of mice was measured daily. Ctrl group, mice that received normal drinking water. (a–c,f) Data are representative of three independent experiments or are shown as mean values±s.e.m. pooled from three independent experiments; (d,f) data are presented as the mean±s.d. from one of the three independent experiments (n=5 per group). P values were generated by one-way ANOVA in b, or two-way ANOVA in d,f, followed by Newman–Keuls multiple comparison test, and by unpaired Student’s t-test in c,e using GraphPad Prism 5 (*P<0.05, **P<0.01, ***P<0.001), DMSO plus DSS versus Spiroepoxide plus DSS in d; Spiroepoxide+A33+ Li-EVs versus Spiroepoxide in f.

Fig 3: ERK and TGF-β1 levels in IBD-A33+ Li-EVs.(a) MC38 cells were treated with 1 mg ml−1 IBD lysates or Ctrl lysates for 24 h. TGF-β1 in MC38 cells or MC38-EVs was measured using western blot analysis. (b) MC38 cells were treated with 1 mg ml−1 IBD lysates or Ctrl lysates for the indicated times. Activation of ERK, JNK, p38 and ATK in MC38 cells was measured using western blot analysis. (c) MC38 cells were pretreated with DMSO, 10 μM ERK-specific inhibitor U0126 or JNK-specific inhibitor SP600125 for 30 min and then treated with 1 mg ml−1 IBD lysates or Ctrl lysates for 24 h. TGF-β1 in MC38-EVs was measured by western blot analysis. (d) Activation of ERK, JNK, p38 and ATK in colon tissues from healthy control or IBD mice was measured using western blot analysis. (e,f) Mice were fed with 2% DSS solution on day 0. Mice received intrarectal injection with 20 μg cholesterol-conjugated ERK or NC siRNA every other day on days 0–11. Twenty-four hours after the last injection, mice were killed. TGF-β1 in colon tissues and A33+ Li-EVs was measured using western blot analysis (e). The body weights of mice were measured daily (f). Ctrl group, mice that received normal drinking water. (a–e) Data are representative of three independent experiments; (f) data are presented as the mean±s.d. from one of the two independent experiments (n=5 per group). P values were generated by two-way ANOVA, followed by Newman–Keuls multiple comparison test using GraphPad Prism 5 (**P<0.01, ***P<0.001), IBD+NC siRNA versus IBD+ERK siRNA.

Fig 4: Li-EVs from IECs inhibit CD4+ T-cell proliferation in vitro.(a) Overall, 1 × 106 ml−1 carboxyfluorescein succinimidyl ester (CFSE)-labelled CD4+ T cells were seeded into 96-well plate and proliferation of T cells was elicited by 1 μl anti-CD3/CD28-coated beads. The indicated concentration of Li-EVs was added. TGF-β1 (10 ng ml−1) was added as a positive control. After culture for 3 days, proliferation of CD4+ T cells was analysed using FACS. (b) The data in a were statistically analysed (n=9). (c) The proliferation assay described in a was performed in the presence of 30 μg ml−1 Si-EVs or Li-EVs. The proliferation of CD4+ T cells was statistically analysed (n=9). (d) CD11c+ and CD11c− Li-EVs were isolated by CD11c+ magnetic beads. CD11c, A33, Alix, TSG101, TGF-β1 and β-Actin in CD11c+, CD11c− and total Li-EVs were detected by western blot analysis. (e) The proliferation assay described in a was performed in the presence of 30 μg ml−1 CD11c+, A33+ and total Li-EVs. The proliferation of CD4+ T cells was statistically analysed (n=9). (f) The proliferation assay described in a was performed in the presence of 30 μg ml−1 A33+ Li-EVs with or without 0.5 μg ml−1 TGF-β1 signalling inhibitor (SB525334). The proliferation of CD4+ T cells was statistically analysed (n=9). Data are representative of three independent experiments or are shown as mean values±s.e.m. pooled from three independent experiments. P values were generated by one-way analysis of variance (ANOVA), followed by Newman–Keuls multiple comparison test using GraphPad Prism 5, versus Ctrl in a (*P<0.05, **P<0.01, ***P<0.001, NS, not significant).

Fig 5: Effects of A33+ Li-EVs on Tregs and the severity of IBD.(a) In all, 1 × 106 ml−1 naive CD4+ T cells were seeded into a 96-well plate and stimulated with 1 μl anti-CD3/CD28-coated beads. Then, 30 μg ml−1 A33+ Li-EVs with or without 0.5 μg ml−1 SB525334 was added. After culture for 3 days, the differentiation of CD4+Foxp3+ Tregs was analysed using FACS. (b) The data in a were statistically analysed (n=9). (c) Mice received one intravenous injection of 100 μg A33+ Li-EVs and were killed, and lymphocytes from the spleen and mLNs were isolated 3 days later. The percentage of CD4+Foxp3+ Tregs was analysed using FACS. (d) The data in c were statistically analysed (n=9). (e) IBD was introduced into the control or CD4+Foxp3+ Treg-depleted mice, and the mice were intravenously treated with 100 μg A33+ Li-EVs on days −2 and 2. The body weights of mice were measured daily. To deplete CD4+Foxp3+ Tregs, each mouse was intraperitoneally injected with 100 μg anti-mouse CD25 monoclonal antibodies or PBS as a control on days −8, −6 and −4. Ctrl group, mice that received normal drinking water; PBS group, mice with drinking water containing 2% DSS and intravenously treated with PBS on days −2 and 2. (a–d) Data are representative of three independent experiments or are shown as mean values±s.e.m. pooled from three independent experiments; (e) data are presented as the mean±s.d. from one of the three independent experiments (n=5 per group). P values were generated by one-way ANOVA in b,d, or two-way ANOVA in e, followed by Newman–Keuls multiple comparison test using GraphPad Prism 5 (*P<0.05, **P<0.01, ***P<0.001), purple or blue colour indicating PBS+anti-CD25 versus A33+ Li-EVs+anti-CD25 or A33+ Li-EVs versus A33+ Li-EVs+anti-CD25, respectively, in e.