Endothelial cell pyroptosis plays an important role in Kawasaki disease via HMGB1/RAGE/cathespin B signaling pathway and NLRP3 inflammasome activation

Abstract

Kawasaki disease (KD) is the most common cause of pediatric cardiac disease in developed countries, and can lead to permanent coronary artery damage and long term sequelae such as coronary artery aneurysms. Given the prevalence and severity of KD, further research is warranted on its pathophysiology. It is known that endothelial cell damage and inflammation are two essential processes resulting in the coronary endothelial dysfunction in KD. However, detailed mechanisms are largely unknown. In this study, we investigated the role of pyroptosis in the setting of KD, and hypothesized that pyroptosis may play a central role in its pathophysiology. In vivo experiments of patients with KD demonstrated that serum levels of pyroptosis-related proteins, including ASC, caspase-1, IL-1β, IL-18, GSDMD and lactic dehydrogenase (LDH), were significantly increased in KD compared with healthy controls (HCs). Moreover, western blot analysis showed that the expression of GSDMD and mature IL-1β was notably elevated in KD sera. In vitro, exposure of human umbilical vein endothelial cells (HUVECs) to KD sera-treated THP1 cells resulted in the activation of NLRP3 inflammasome and subsequent pyroptosis induction, as evidenced by elevated expression of caspase-1, GSDMD, cleaved p30 form of GSDMD, IL-1β and IL-18, and increased LDH release and TUNEL and propidium iodide (PI)-positive cells. Furthermore, our results showed that NLRP3-dependent endothelial cell pyroptosis was activated by HMGB1/RAGE/cathepsin B signaling. These findings were also recapitulated in a mouse model of KD induced by Candida albicans cell wall extracts (CAWS). Together, our findings suggest that endothelial cell pyroptosis may play a significant role in coronary endothelial damage in KD, providing novel evidence that further elucidates its pathophysiology.

Fig. 1. Levels of pyroptosis-related proteins were increased in sera from KD patients.

a–f Enzyme-linked immunosorbent assay (ELISA) was performed to assess the levels of ASC (a), caspase-1 (b), IL-1β (c), IL-18 (d), GSDMD (e), and LDH (f) in sera from age-similar acute KD patients and healthy controls (HCs). Significance: *P < 0.05, **P < 0.01, and ***P < 0.001. g, h Expression of GSDMD and mature IL-1β was analyzed in HC (n = 4) and KD (n = 4) sera. Transferrin was applied as a loading control (g). The right panel is the statistical analysis of immunoblots (h). Data were expressed as mean ± SD. *P < 0.05

Fig. 2. Pyroptosis occurred in human umbilical vein endothelial cells (HUVECs) after exposure to KD sera-treated THP1 cells.

a, b HUVECs were co-cultured with THP1 cells that were treated with HC or KD sera (pooled from eight patients) for 24 h. Protein levels of NLRP3, caspase-1, mature IL-1β, and IL-18, and cleavage of GSDMD were examined by western blot analysis. GAPDH was used as an internal control. Data in (b) were shown as mean ± SD from three independent experiments involving different batches of cells but the same pooled HC or KD sera. *P < 0.05 and **P < 0.01. c, d Caspase-1 fluorescence intensity and the percentage of TUNEL-positive cells were respectively evaluated using immunofluorescent assay and TUNEL staining. The nuclei were stained blue using DAPI. Magnification: × 200. Scale bar = 100 μm. **P < 0.01. e LDH release was determined in HUVECs after HC or KD sera treatment (n = 5). *P < 0.05. f Percentage of PI-positive cells was detected in HUVECs using Hoechst33342/PI staining. Magnification: ×100. Scale bar = 500 μm. Significance: **P < 0.01

Fig. 3. NLRP3 inhibitor repressed KD-treated EC pyroptosis.

HUVECs were pretreated with NLRP3 inhibitor (MCC950, 10 μM) for 30 min, and then ECs were co-incubated with pooled sera-treated THP1 for 24 h. a Effect of MCC950 pretreatment on the protein expression of pyroptosis were analyzed by western blot analysis. GAPDH was used as an internal control. b–h Quantitative analysis of pyroptosis-related protein expression was conducted. Data were presented as mean ± SD (n = 3). i–k Caspase-1 expression, DNA fragmentation and LDH release were determined using immunofluorescent assay, TUNEL staining and LDH activity assay kits, respectively. Magnification: ×200, Scale bar = 100 μm. l Double staining of Hoechst 33342 (blue) and PI (red) was conducted in the treated endothelial cells (Left: the representative photographs, Right: the quantification evaluation of PI-positive cells). Magnification: ×100, Scale bar = 500 μm. Significance: *P < 0.05 and **P < 0.01 indicates significant difference between KD group and HC group, and ^#P < 0.05 indicates significant difference between KD groups with and without MCC950 addition

Fig. 4. Inhibition of cathepsin B alleviated KD-treated EC pyroptosis.

HUVECs were pretreated with a cathepsin B inhibitor (CA074-Me, 25 μM) for 30 min, and then ECs were co-cultured with pooled sera-treated THP1 for 24 h. a Endothelial cells were stained with Magic Red Cathepsin B detection reagent to examine the activation of cathepsin B. Magnification: ×200, Scale bar = 100 μm. b Effects of CA074-Me treatment on the expression of pyroptosis-associated proteins were assessed in the treated endothelial cells. c–i Quantitative analysis of pyroptosis-related protein expression, and GAPDH was used as an internal control. Data were exhibited as mean ± SD (n = 3). j–m Caspase-1 expression, TUNEL staining, LDH release, and Hoechst 33342/PI double staining were conducted as described in Fig. 2c–f. Magnification: ×200, Scale bar = 100 μm for (j) and (k). Magnification: ×100, Scale bar = 500 μm for (m). Significance: *P < 0.05, and **P < 0.01 indicate significant difference between KD group and HC group. ^#P < 0.05 indicates significant difference between KD groups with and without CA074-Me addition

Fig. 5. KD-treated EC pyroptosis was dependent on HMGB1/RAGE signaling.

a HMGB1 contents were evaluated in the supernatant medium of pooled sera-treated THP1 cells by ELISA analysis. b, c The mRNA and protein expressions of RAGE were determined in ECs. d Effects of a RAGE inhibitor (FPS-ZM1) or the anti-HMGB1 antibody on the activation of cathepsin B were assessed. e Protein expression levels of NLRP3, caspase-1, GSDMD, IL-1β, and IL-18 in the treated endothelial cells were determined. f–l Expression of pyroptosis-related proteins was quantitatively analyzed, and GAPDH was used as an internal loading control. Data were exhibited as mean ± SD (n = 3). m–p Caspase-1 expression, TUNEL staining, LDH release, and Hoechst 33342/PI double staining were performed according to the above descriptions. Magnification: ×200, Scale bar = 100 μm for (m) and (n). Magnification: ×100, Scale bar = 200 μm for (p). Significance: *P < 0.05, **P < 0.01 vs. the HC group. ^#P < 0.05 vs. the KD group

Fig. 6. NLRP3-mediated endothelial cell pyroptosis in a KD mouse model.

Mice were continuously injected i.p. with 4 mg CAWS for 4 days to induce coronary arteritis. Twenty-eight days later, mice were sacrificed and the heart tissues were harvested for follow-up examinations. a Representative images from H&E staining at 28 day post-CAWS injection. Scale bar = 100 μm. b Expression of macrophage marker F4/80 was determined using IHC staining in the endothelium of coronary arteries. The histogram showed the area percentage of the endothelium positive for F4/80 in coronary arteries. Enlarged images of area of interesting (AOI) were indicated with a red arrow. Scale bar = 100 μm. c The expression of neutrophil marker was examined in the endothelium of coronary arteries. The histogram exhibited the area percentage of the endothelium positive for neutrophil marker in coronary arteries. Scale bar = 100 μm. d Expression of pyroptosis-related proteins was examined by western blot analysis. e–k Quantitative analysis of pyroptosis-related protein expression was conducted. Data were expressed as mean ± SD (n = 3). l Comparison of expression and subcellular distribution of caspase-1 by double staining of caspase-1 (green) and CD31 (red) in coronary artery of control mice, CAWS mice, and CAWS mice pretreatment with MCC950. The presence of caspase-1 in endothelial cells was indicated by the co-localization of caspase-1 and CD31 (an endothelial marker). Magnification: ×200. Scale bar = 50 μm. m Identification of DNA fragmentation by co-localized staining of TUNEL (green) and CD31 (red). The nuclei were stained blue with DAPI. Magnification: ×200. Scale bar = 50 μm. Significance: *P < 0.05, **P < 0.01 vs. the control group. ^#P < 0.05 vs. the CAWS group

Fig. 7. HMGB1/RAGE/cathepsin B signaling was implicated in the endothelial cell pyroptosis in the KD mouse model.

Mice were pretreated with a cathepsin B inhibitor (CA074-Me), a RAGE-specific inhibitor (FPS-ZM1), and the anti-HMGB1 antibody 60 min before CAWS injection. a Heart tissues were analyzed for inflammatory infiltration by H&E staining. Magnification: ×200. Scale bar = 200 μm. b Expression of macrophage marker F4/80 was determined using IHC staining in the endothelium of coronary arteries. The histogram showed the area percentage of the endothelium positive for F4/80 in coronary arteries. Enlarged images of area of interesting (AOI) were indicated with a red arrow. Scale bar = 100 μm. c The expression of neutrophil marker was examined in the endothelium of coronary arteries. The histogram exhibited the area percentage of the endothelium positive for neutrophil marker in coronary arteries. Scale bar = 100 μm. d Western blot analysis was used to determine the expression of pyroptosis-related proteins. e–k Quantitative analysis was performed to detect pyroptosis-related protein expression. Data were shown as mean ± SD (n = 3). l Caspase-1 expression was analyzed in coronary endothelial cells using caspase-1/CD31 double staining. Magnification: × 200. Scale bar = 50 μm. m DNA fragmentation in endothelial cells was identified by co-localization observation of TUNEL and CD31. The nuclei were stained blue with DAPI. Magnification: ×200. Scale bar = 50 μm. Significance: *P < 0.05, **P < 0.01, ***P < 0.001 vs. the control group. ^#P < 0.05 vs. the CAWS group

Fig. 8. Schematic model for endothelial cell pyroptosis in Kawasaki disease.

High level of serum HMGB1 interacts with its receptor RAGE, enters lysosomes, and induces Cat B (cathepsin B) activation and release from the ruptured lysosomes. This is followed by NLRP3 inflammasome activation and subsequent induction of caspase-1-dependent pyroptosis