MLKL-mediated endothelial necroptosis drives vascular damage and mortality in systemic inflammatory response syndrome

Abstract

The hypersecretion of cytokines triggers life-threatening systemic inflammatory response syndrome (SIRS), leading to multiple organ dysfunction syndrome (MODS) and mortality. Although both coagulopathy and necroptosis have been identified as important factors in the pathogenesis of SIRS, the specific cell types that undergo necroptosis and the interrelationships between coagulopathy and necroptosis remain unclear. In this study, we utilized visualization analysis via intravital microscopy to demonstrate that both anticoagulant heparin and nonanticoagulant heparin (NAH) pretreatment protect mice against TNF-α-induced mortality in SIRS. Moreover, the deletion of Mlkl or Ripk3 resulted in decreased coagulation and reduced mortality in TNF-α-induced SIRS. These findings suggest that necroptosis plays a key role upstream of coagulation in SIRS-related mortality. Furthermore, using a genetic lineage tracing mouse model (Tie2-Cre;Rosa26-tdT), we tracked endothelial cells (ECs) and verified that EC necroptosis is responsible for the vascular damage observed in TNF-α-treated mice. Importantly, Mlkl deletion in vascular ECs in mice had a similar protective effect against lethal SIRS by blocking EC necroptosis to protect the integrity of the endothelium. Collectively, our findings demonstrated that RIPK3-MLKL-dependent necroptosis disrupted vascular integrity, resulting in coagulopathy and multiorgan failure, eventually leading to mortality in SIRS patients. These results highlight the importance of targeting vascular EC necroptosis for the development of effective treatments for SIRS patients.

Keywords: Endothelial cells; MLKL; Necroptosis; RIPK3; Systemic inflammatory response syndrome (SIRS); TNF-α.

Fig. 1

A–G Heparin alleviates vascular damage and systemic coagulation caused by TNF-α-induced SIRS. WT mice were injected intravenously with PBS or mTNF-α (7 μg/mouse). Low-molecular-weight heparin (200 IU/kg) was administered subcutaneously 30 min before mTNF-α injection. A Representative SD-IVM images of thrombin (green), vascular perfusion (bright red, rhodamine B isothiocyanate-dextran), and platelet adhesion (pink, AF647 anti-CD49b) within the liver (multiphoton microscopy images) microvasculature at 10 h after mTNF-α challenge. Quantitative analysis of thrombin and platelet fluorescence within the liver microcirculation via ImageJ software. White arrows indicate areas of vascular damage. Scale bars: 50 μm. B–E Blood was collected 10 h after PBS or mTNF-α injection. B Cytokines in plasma were analyzed via Luminex liquid chip. C Blood volume, D TF (tissue factor) activity in plasma, and E total platelet count before and after mTNF-α injection were measured. The solid figures represent individual mice; the error bars denote the SEM; n = 3–6 for all experimental groups. F Blood was collected at 0, 2, and 10 h after mTNF-α injection from WT mice pretreated with PBS or heparin. The plasma PAI-1 and FDP concentrations were measured. The error bars denote the SEM; n = 3–4 for all experimental groups. G Representative images of H&E-stained livers and lungs from WT mice at 0, 2, and 10 h after mTNF-α injection. Thrombi in the liver and lung capillaries and sinusoids formed by red blood cells in the intercellular space were apparent in WT mice challenged with mTNF-α. Scale bars: 50 μm. The data are shown as the means ± SEMs. **p < 0.01; ***p < 0.001; ****p < 0.0001. (A) and (G) are from three independent experiments. Unpaired/paired t-test, one-way/two-way ANOVA test

Fig. 2

Nonanticoagulant heparin (NAH) protects against TNF-α-induced hypothermia, mortality, and vascular necroptosis. A–E WT mice were injected intravenously with mTNF-α (7 μg/mouse). Low-molecular-weight heparin (200 IU/kg) and nonanticoagulant heparin (NAH, 100 μg/mouse) were administered subcutaneously 30 min before mTNF-α injection. A, B Body temperature and survival of WT mice injected with mTNF-α after pretreatment with PBS, heparin or NAH. C Representative images of H&E-stained livers, lungs, kidneys, and caeca of WT mice at 10 h after mTNF-α injection and pretreatment with PBS or NAH. Scale bars: 50 μm. D Immunofluorescence staining of α-smooth muscle actin (α-SMA; green) and p-RIPK3 (red) in liver tissues at 0, 2, 6, and 10 h after mTNF-α injection. DAPI (blue) was used for nuclear staining. Scale bars: 50 μm. E Immunofluorescence staining of α-smooth muscle actin (α-SMA; green) and p-RIPK3 (red) in liver tissues 10 h after mTNF-α injection and pretreatment with PBS, heparin, or NAH in WT mice. DAPI (blue) was used for nuclear staining. Areas of vascular damage are indicated by white arrows. Scale bars: 100 μm. D, E Quantitative analysis of vascular damage and p-RIPK3 fluorescence within the liver via ImageJ software. n = 4 for all experimental groups. The data are shown as the means ± SEMs. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Two-way ANOVA with log-rank (Mantel‒Cox) test and one-way ANOVA test were used

Fig. 3

Mlkl gene deletion protects mice against TNF-α-induced vascular damage and lethality. A–F Mlkl^−/− and WT mice were injected intravenously with mTNF-α (7 μg/mouse). A Representative SD-IVM images of thrombin (green), vascular perfusion (bright red), and platelet adhesion (pink) within the liver microvasculature at 10 h after mTNF-α challenge. Quantitative analysis of thrombin and platelet fluorescence within the liver microcirculation via ImageJ software. Scale bars: 50 μm. B, C Survival and body temperature of Mlkl^−/− and WT mice injected with mTNF-α. D Blood was collected at 0, 2, and 10 h after mTNF-α injection. IL-6, TNF-α, IL-1α, and IL-1β levels were analyzed. The solid figures represent individual mice; the error bars denote the SEM; n = 3–6 for all experimental groups. E Blood was collected at 0, 2, and 10 h after mTNF-α injection. The plasma PAI-1 and FDP concentrations were measured. The error bars denote the SEMs; n = 3–6 for all experimental groups. F Representative images of H&E-stained livers and lungs from Mlkl^−/− and WT mice at 10 h after mTNF-α injection. Scale bars: 50 μm. The data are shown as the means ± SEMs. **p < 0.01; ***p < 0.001; ****p < 0.0001. (A) and (F) are from three independent experiments. Unpaired t-test and two-way ANOVA test

Fig. 4

Endothelial cell necroptosis is required for TNF-α-induced vascular damage. A Immunofluorescence staining of PECAM (platelet endothelial cell adhesion molecule; green) and p-RIPK3 (red) or p-MLKL (red) in liver tissues from WT mice treated with PBS or mTNF-α for 10 h. DAPI (blue) was used for nuclear staining. Scale bars: 20 μm. The dotted box (white color) indicates the cross-sectional image of the endothelium. B Schematic figure showing the tracing experimental strategy for endothelial cells. C Immunofluorescence staining of PECAM (green) and tdTomato (tdT; red) in liver sections from Tie2-Cre;Rosa26-tdT mice. DAPI (blue) was used for nuclear staining. Scale bars: 20 μm. D Immunofluorescence staining of tdTomato (tdT; yellow) and p-RIPK3 (red) in liver tissues from Tie2-Cre;Rosa26-tdT mice treated with PBS or mTNF-α for 10 h. DAPI (blue) was used for nuclear staining. Scale bars: 5 μm. (A), (C), and (D) are from three independent experiments

Fig. 5

Mlkl^flox/flox;Tie2-Cre prevents TNF-α-induced SIRS lethality and protects the integrity of the endothelium by inhibiting necroptosis in ECs. A–E Mlkl^flox/flox;Tie2-Cre and WT mice were injected intravenously with mTNF-α (7 μg/mouse). A, B Body temperature and survival of Mlkl^flox/flox;Tie2-Cre and WT mice injected with mTNF-α. C Immunofluorescence staining of PECAM (green) and p-RIPK3 (red) in lung and liver tissues from Mlkl^flox/flox;Tie2-Cre and WT mice treated with mTNF-α for 10 h. DAPI (blue) was used for nuclear staining. Scale bars: 100 μm. Quantitative analysis of vascular endothelium damage and p-RIPK3 fluorescence within the lung and liver via ImageJ software. n = 4 for all experimental groups. D Blood was collected 10 h after mTNF-α injection. Cytokines in the plasma were analyzed via Luminex liquid chip. n = 3 for all experimental groups. E H&E staining of the livers, lungs, and cecum of Mlkl^flox/flox;Tie2-Cre and WT mice at 10 h after mTNF-α injection. Scale bar: 20 μm. n = 3 for all experimental groups. The data are shown as the means ± SEMs. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001. Unpaired t-test and two-way ANOVA test

Fig. 6

Schematic overview showing MLKL-mediated necroptosis of vascular ECs during the development of MODS and death triggered by TNF-α-induced SIRS. In the SIRS model, TNF-α in combination with multiple upregulated cytokines triggers necroptosis of vascular endothelial cells. This process is dependent on RIPK3–MLKL signaling, which causes the destruction of vascular integrity, resulting in coagulopathy and multiorgan failure. This ultimately leads to mortality in SIRS patients