Fig 2: M2 macrophage-derived EVs promoted CD8+ T cell exhaustion in mice with primary HCC.The murine model of primary HCC was established by DEN/CCl4 combined induction, and the model mice were injected with EVs via tail vein. A The absorption of EVs by mouse liver tissues was detected using immunofluorescence. Green indicated PKH67-labeled EVs and blue indicated nuclear staining. B Histopathological changes of liver tissues were observed using HE staining at the 40th week after induction. C The number of CD8+ T cells in tumor sections was detected using immunohistochemistry. PBMCs were isolated from the mouse liver and CD8+ T cells were sorted. D The expressions of immune checkpoint inhibitory receptors (PD1 and TIM3) on CD8+ T cells were detected using flow cytometry. E The expressions of effector cytokines (IFN-?, IL-2, and TNF-a) were detected using flow cytometry. F The proliferation ability of CD8+ T cells was measured using flow cytometry. G The ability of CD8+ T cells to kill Hep1-6 cells was measured using flow cytometry. N = 6. The cell experiment was repeated 3 times. Data are presented as mean ± standard deviation and analyzed using one-way ANOVA, followed by Tukey’s multiple comparison test, **p < 0.01.

Fig 3: Overexpression of YOD1 reversed the effect of EVs on CD8+ T cell exhaustion. pcDNA-YOD1-transfected CD8+ T cells were treated with EVs.A YOD1 mRNA expression was detected using RT-qPCR. B The expressions of PD1 and TIM3 were detected using flow cytometry. C The expressions of IFN-?, IL-2, and TNF-a were detected using flow cytometry. D The proliferation ability of PD1-TIM3- and PD1int TIM3+ cells was measured using flow cytometry. E The ability of PD1-TIM3- and PD1int TIM3+ cells to kill Hep1-6 cells was measured using flow cytometry. The cell experiment was repeated 3 times. Data are presented as mean ± standard deviation and analyzed using one-way ANOVA, followed by Tukey’s multiple comparison test, **p < 0.01.

Fig 4: YOD1 inhibited the YAP/ß-catenin pathway.A, B The levels of YAP and ß-catenin in HCC mouse liver and PD1-TIM3- and PD1int TIM3+ cells were detected using Western blotting. N = 6. The cell experiment was repeated 3 times. Data are presented as mean ± standard deviation and analyzed using one-way ANOVA, followed by Tukey’s multiple comparison test, **p < 0.01.

Fig 5: HGF NPs inhibited exosomal PD-L1 and triggered robust immunostimulation.a Western blot analysis of exosome marker CD63 and PD-L1 via collecting the exosomes from the cell-free media of B16F10 cells with various treatments. GAPDH served as a loading control. Images were representative of the three experiments. b Scheme of in vivo experimental design for (c–m). C57BL/6 mice were subcutaneously inoculated with 1 × 106 B16F10 cells. When tumor volumes reached ~ 100 mm3, mice were treated with PBS, HACA-Fe, HACA-GW, or HGF on day 0, day 2, and day 4. Tumor and tumor-draining lymph node (TDLN) were collected on day 7 for further analysis. c Western blot analysis of exosome marker CD63 and PD-L1 in tumor tissues after treatment. Images were representative of the three experiments. d–f Flow cytometric plots (d) and quantification of CD8+ (e) and CD4+ (f) T cells, respectively, gated by CD3+ T cells in the TDLN (n = 5) in groups of PBS (G1), HACA-Fe (G2), HACA-GW (G3), HGF (G4), and HGF + liproxstatin (G5). g–m Flow cytometric analysis of granzyme B+ (GzmB+, g-i), Ki67+ (j and k), and Tim3+ (l and m) in CD8+ and CD4+ T cells. n = 5 biologically independent animals per group. Data were presented as mean ± SD. Statistically significant differences between groups were identified by one-way ANOVA with Tukey’s post hoc test. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.