ASC-expressing pyroptotic extracellular vesicles alleviate sepsis by protecting B cells

Abstract

Pyroptosis is an inflammatory programmed cell death process characterized by membrane rupture. Interestingly, pyroptotic cells can generate plenty of nanosized vesicles. Non-inflammatory apoptotic cell death-derived apoptotic vesicles (apoVs) were systemically characterized and displayed multiple physiological functions and therapeutic potentials. However, the characteristics of pyroptotic cell-generated extracellular vesicles (EVs) are largely unknown. Here, we identified a group of pyroptotic EVs (pyroEVs) from in vitro cultured pyroptotic mesenchymal stem cells (MSCs), as well as from septic mouse blood. Compared with apoVs, pyroEVs express similar levels of annexin V, calreticulin, and common EV markers, but express a decreased level of apoptotic marker cleave caspase-3. PyroEVs, but not apoVs and exosomes, specifically express pyroptotic maker apoptosis-associated speck-like protein containing CARD (ASC). More importantly, MSC-derived pyroEVs protect B cells in the spleen and bone marrow to relieve inflammatory responses and enhance the survival rate of the septic mice. Mechanistically, pyroEV membrane-expressed ASC binds to B cells to repress cell death by repressing Toll-like receptor 4. This study uncovered the characteristics of pyroEVs and their therapeutic role in sepsis and B cell-mediated immune response.

Keywords: B cells; apoptosis-associated speck-like protein containing CARD; mesenchymal stem cells; pyroptosis; pyroptotic extracellular vesicles; sepsis.

Graphical abstract

Figure 1

Characteristics of pyroptotic MSCs (A) Schematic diagram indicating the procedures of inducing apoptosis and pyroptosis in human BMMSCs. (B) Brightfield images of apoptotic and pyroptotic MSCs. Cell morphology was visualized by ordinary optical microscopy. Scale bar, 10 μm. (C) High-content time-lapse microscopy images of cell membrane eversion and integrity were monitored by annexin V (green) labeling and PI (red) uptake, respectively. Scale bar, 10 μm. (D) Flow cytometric analysis and the corresponding quantification of GSDMD and cleaved caspase-3 expressed in apoptotic MSCs and pyroptotic MSCs. (E) Western blotting analysis showing the presence of GSDMD, N-cleaved GSDMD, ASC, caspase 1, and cleaved caspase 3 in apoptotic MSCs and pyroptotic MSCs. (F) Live-cell super-resolution SIM images of WGA (green) stained MSCs with pyroptosis. Arrowhead indicated bubbling of pyroptotic MSCs. The bottom magnifies the boxed area in the top. Scale bars, 15 μm; 1 μm. (G) Representative SEM images showing the morphological change of apoptotic MSCs and pyroptotic MSCs. The bottom magnifies the boxed area in the top. Scale bars, 10 μm; 1 μm. (H) Representative immunofluorescence images of pyroptotic MSCs. MSCs were stained with WGA (green), and ASC were stained with ASC antibody (red). Colocalization of WGA and ASC was shown in yellow. The right magnifies the boxed area in the left, indicating ASC+ small vesicle-like structures (1), ASC-negative large membrane protrusions (2), and perinuclear ASC speck (3). Scale bars, 10 μm for the left, 500 nm for the right. Error bars are means ± SD. Data were analyzed using one-way ANOVA with Bonferroni correction (D). ns, not significant; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001.

Figure 2

Characteristics of pyroEVs from human BMMSCs (A) Schematic diagram indicating the procedures of isolating MSC-derived pyroEVs. (B) Nanoparticle tracking analysis by NanoSight exhibiting the concentration and size distribution of apoVs and pyroEVs. (C–E) Nanoparticle tracking analysis by Zetaview showing mean particles size, zeta potential, and concentrations for pyroEVs and apoVs. n ≥ 3 per group. (F) EVs protein content determined by BCA. n = 3 per group. (G) Representative TEM images showing the morphology of apoVs and pyroEVs. Scale bars, 200 nm. (H) Western blotting analysis showing the presence of ASC, calreticulin, cleaved caspase-3, syntenin, and CD9, in apoVs, pyroEVs and exosomes. (I) Nanoflow cytometric analysis and the corresponding quantification of GSDMD, ASC, caspase-1, and cleaved caspase-3 expressed in apoVs and pyroEVs. (J) Representative SIM images showing the specific staining of ASC (green) to the apoVs (red) and pyroEVs (red). Scale bars, 5 μm; 500 nm. (K) Representative three dimensional reconstruction images of SIM showing ASC (green) on the surface of pyroEVs (red). Scale bars, 500 nm. Error bars are means ± SD. Data were analyzed using independent unpaired two-tailed Student’s t tests (C–F, and I). ns, not significant; ∗p < 0.1; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001.

Figure 3

Characteristics of pyroEVs derived from septic mice (A) Scheme illustrated that sepsis model was established in mice by CLP. (B) Schematic diagram indicating the procedures of isolating EVs from plasma. (C and D) Nanoparticle tracking analysis by Zetaview showing the size and size distribution of EVs in plasma of control and septic mice. n ≥ 3 per group. (E) Nanoparticle tracking analysis by Zetaview showing the membrane potential of EVs in plasma of control and septic mice. n = 3 per group. (F) Nanoparticle tracking analysis by Zetaview showing the concentration of EVs in plasma of control and septic mice. n ≥ 3 per group. (G) Representative immunofluorescence images of EVs from control and sepsis mice. EVs were stained with PKH26 (red). ASC was stained with ASC antibody (green). Scale bars, 5 μm for the right, 200 nm for the left. (H) Nanoflow cytometric analysis and the corresponding quantification of ASC expressed in EVs in plasma of control and septic mice. n = 3 per group. (I) Nanoflow cytometric analysis and the corresponding quantification of CD19, CD3, and Ly6G expressed in EVs in plasma of septic mice. n ≥ 3 per group. (J) Nanoflow cytometric analysis and the corresponding quantification of CD19 expressed in ASC-positive EVs in plasma of septic mice. n = 3 per group. Error bars are means ± SD. Data were analyzed using independent unpaired two-tailed Student’s t tests (D–F, and H), or one-way ANOVA with Bonferroni correction (I). ns, not significant; ∗∗p < 0.01.

Figure 4

PyroEVs treatment relieved infection phenotypes and tissue injuries of septic mice (A) Scheme illustrating the sepsis procedure and treatment with MSC-pyroEVs. (B) Mouse survival time in sepsis with or without injection of pyroEVs 2 h after CLP, data were shown as percent of animals surviving. n ≥ 3 per group. (C) Quantitative temperature analysis of control group, sepsis group and sepsis treated with pyroEVs group. n = 3 per group. (D) Graph of the bacterial colony-forming area of control group, sepsis group and sepsis treated with pyroEVs group at 24 h after CLP. n = 3 per group. (E) Hematoxylin and eosin-stained images and the corresponding quantitative analysis of tissue injury of kidney, liver, and lung injuries at 24 h after the procedure. n = 5 per group. Scale bars, 60 μm. (F) Quantitative analysis of GSDMD expression in different group of mice detected by confocal microscopy. n = 3 per group. Scale bars, 20 μm. (G) ELISA analysis of the plasma levels of pro-inflammatory cytokines IFN-γ, TNF-α, IL-1β, and anti-inflammatory cytokine IL-10. n ≥ 3 per group. Error bars are means ± SD. Data were analyzed using log rank (Mantel-Cox) test (B), or one-way ANOVA with Bonferroni correction (C–G). ns, not significant; ∗p < 0.1, ∗∗p < 0.01, ∗∗∗p < 0.001, ∗∗∗∗p < 0.0001.

Figure 5

Systemically infused pyroEVs enriched around B cells in the bone marrow and spleen to inhibit B cell death and ASC expression (A) Live animal bioluminescent images of dissected organs from control mice and septic mice after injection of DiR lipid dye-labeled MSC-derived pyroEVs. n = 3 per group. (B) Confocal microscopy images and quantitative analysis of PKH-26-labeled pyroEVs (red) in the liver, spleen, lung, and bone of control and septic mice. n = 3 per group. Scale bars, 10 μm. (C) Confocal microscopy images showing the colocalization of B cell marker CD19 (green) and PKH-26-labeled pyroEVs (red) in spleen and bone marrow of septic mice. Colocalization of PKH-26 and CD19 was shown by arrows. Scale bars, 10 μm. (D) Quantification analysis of the CD19 expression level in the spleen and bone marrow was detected by confocal microscopy. n = 3 per group. Scale bars, 10 μm. (E) Representative immunofluorescence images and quantification analysis of spleen and bone from sepsis mouse. B cells were stained with CD19 (green), ASC was stained with ASC antibody (red). White boxes show the ASC speck. Scale bars, 10 μm for the bottom, 1 μm for the top. n ≥ 3 per group. Scale bars, 20 μm. Error bars are means ± SD. Data were analyzed using independent unpaired two-tailed Student’s t tests (B, D, and E). ns, not significant; ∗p < 0.1, ∗∗p < 0.01, ∗∗∗∗p < 0.0001.

Figure 6

Depletion of B cells blocked the therapeutic effect of pyroEVs on sepsis (A) Scheme illustrated the sepsis procedure and treatment with rituximab and MSCs-derived pyroEVs. (B) Graph of the bacterial colony-forming area of sepsis group, sepsis treated with pyroEVs group and sepsis pre-treated rituximab and treated with pyroEVs group at 24 h after CLP. n = 3 per group. (C–E) Hematoxylin and eosin-stained images and the corresponding quantitative analysis of tissue injury of kidney, liver, and lung injuries at 24 h after the procedure. n = 5 per group. Scale bars, 60 μm. Error bars are means ± SD. Data were analyzed using independent unpaired two-tailed Student’s t tests (B–E). ns, not significant; ∗∗p < 0.01, ∗∗∗∗p < 0.0001.

Figure 7

Interaction between ASC on pyroEVs and TLR4 on B cells inhibited the TLR4-induced cell death in vitro (A) Representative SIM images showing the colocalization of CD19 (green) and PKH-26 (red) in immune cells of spleen. Scale bars, 1 μm. (B) Flow cytometric analysis and the corresponding quantification of the percentages of 7AAD⁺ B cells in vitro. Cells from spleen were incubated with pyroEVs at 37°C for 24 h, then stained with B220 and 7AAD. n = 3 per group. (C) Western blotting analysis of TLR4, GSDMD, N-cleaved GSDMD, ASC, and caspase 1 expressed in B cells. (D) Representative SIM images showing the ASC (purple) expression on the surface of B cells between LPS pre-treated B cells and LPS pre-treated B cells with pyroEVs (red). Scale bars, 1 μm. (E) Nanoflow cytometric analysis and the corresponding quantification of ASC expressed in pyroEVs and pyroEVs treated with triple. n = 3 per group. (F) Representative SIM images showing binding of ASC (green) to the surface of pyroEVs and pyroEVs treated with triple. Scale bars, 5 μm. (G) Nanoflow cytometric analysis and the corresponding quantification of ASC expressed in pyroEVs and si-ASC pyroEVs. n = 3 per group. (H) Representative SIM images showing binding of ASC (green) to the surface of pyroEVs and si-ASC pyroEVs. Scale bars, 5 μm. (I) Representative SIM images showing uncombining of si-ASC pyroEVs to the surface of B cells. After co-culture with PKH26-labeled pyroEVs and si-ASC pyroEVs at 37°C for 24 h, B cells were stained with Alexa Fluor 488-conjugated CD19. Scale bars, 5 μm. (J) Flow cytometric analysis and the corresponding quantification of the percentages of 7AAD⁺ B cells. The B cells were incubated with pyroEVs or si-ASC pyroEVs at 37°C for 24 h, and stained with B220 and 7AAD. n = 3 per group. (K) Western blotting analysis of TLR4, GSDMD, N-cleaved GSDMD, ASC, caspase-1, and HMGB 1 expressed in B cells in vitro. Error bars are means ± SD. Data were analyzed using or one-way ANOVA with Bonferroni correction (B, E, G, and J). ns, not significant; ∗p < 0.1, ∗∗∗∗p < 0.0001.

Figure 8

ASC is responsible for pyroEVs-mediated therapy of sepsis (A) Scheme illustrated the sepsis procedure and treatment with si-ASC pyroEVs. (B) Graph of the bacterial colony-forming area of septic mice with or without rituximab pretreated at 24 h after CLP. n = 3 per group. (C–E) Hematoxylin and eosin-stained images and the corresponding quantitative analysis of tissue injury of liver, lung, and kidney injury score at 24 h after the procedure. n = 5 per group. Scale bars, 60 μm. Error bars are means ± SD. Data were analyzed using one-way ANOVA with Bonferroni correction (B–E). ns, not significant; ∗∗p < 0.01, ∗∗∗∗p < 0.0001.