Gasdermin E dictates inflammatory responses by controlling the mode of neutrophil death

Abstract

Both lytic and apoptotic cell death remove senescent and damaged cells in living organisms. However, they elicit contrasting pro- and anti-inflammatory responses, respectively. The precise cellular mechanism that governs the choice between these two modes of death remains incompletely understood. Here we identify Gasdermin E (GSDME) as a master switch for neutrophil lytic pyroptotic death. The tightly regulated GSDME cleavage and activation in aging neutrophils are mediated by proteinase-3 and caspase-3, leading to pyroptosis. GSDME deficiency does not alter neutrophil overall survival rate; instead, it specifically precludes pyroptosis and skews neutrophil death towards apoptosis, thereby attenuating inflammatory responses due to augmented efferocytosis of apoptotic neutrophils by macrophages. In a clinically relevant acid-aspiration-induced lung injury model, neutrophil-specific deletion of GSDME reduces pulmonary inflammation, facilitates inflammation resolution, and alleviates lung injury. Thus, by controlling the mode of neutrophil death, GSDME dictates host inflammatory outcomes, providing a potential therapeutic target for infectious and inflammatory diseases.

Fig. 1. GSDME disruption abolished neutrophil lytic death without affecting overall neutrophil survival.

a Cleavage of GSDME during neutrophil spontaneous death. Bone marrow-derived neutrophils from WT mice were used. Results are representative of at least three biological replicates. All data are represented as mean ± SD, n = 3 independent repeats. b Representative images of neutrophils undergoing apoptosis and lytic cell death. Results are representative of at least three independent experiments. c Representative images of aging neutrophils at the indicated time points. The white arrow heads indicate swollen or “puffed” cells (lytic cell death). PI propidium iodide, BF bright field. Results are representative of at least three independent experiments. d The percentage and number of swollen, AV-positive, and PI-positive cells. All data are represented as mean ± SD. n = 4 independent repeats, *P < 0.05, **P < 0.01, ns non-significant (Percentage of AV⁺ cells: 4 h, P = 0.0273; 8 h, P = 0.0285; 12 h, P = 0.0011; 16 h, P = 0.001; 20 h, P = 0.0004; 24 h, P = 0.0067. Percentage of PI⁺ cells: 8 h, P = 0.0297; 12 h, P = 0.0011; 16 h, P = 0.022; 20 h, P = 0.0168; 24 h, P = 0.0037. AV⁺ cell number: 4 h, P = 0.0184; 8 h, P = 0.0261; 12 h, P = 0.0001; 16 h, P = 0.0001; 20 h, P = 0.0001; 24 h, P = 0.0001. PI⁺ cell number: 8 h, P = 0.0046; 12 h, P = 0.0099; 24 h, P = 0.0012). e The number of intact neutrophils remaining. All data are represented as mean ± SD. n = 3 independent repeats, *P < 0.05, **P < 0.01 (Remaining cell number: 12 h, P = 0.0119; 16 h, P = 0.024; 20 h, P = 0.0001; 24 h, P = 0.0001. Disappeared cell number: 12 h, P = 0.0132; 16 h, P = 0.0207; 20 h, P = 0.0247). f The percentage and number of healthy neutrophils. All data are represented as mean ± SD. n = 4 independent repeats, *P < 0.05, **P < 0.01, ns non-significant (Percentage of healthy cells: 8 h, P = 0.0499; 12 h, P = 0.0115; 16 h, P = 0.0001; 20 h, P = 0.0005; 24 h, P = 0.0008). g The absolute number of healthy neutrophils and neutrophils undergoing apoptosis or lytic cell death. All data are represented as mean ± SD (n = 4 independent repeats). h LDH release from cell culture supernatants. Data are mean ± SD. n = 3 independent repeats, *P < 0.05, **P < 0.01 (4 h, P = 0.042; 8 h, P = 0.0001; 12 h, P = 0.0003; 16 h, P = 0.0023). Statistical significance was examined by unpaired two-sided Student’s t test (d, e, f, h). Source data are provided as a Source Data file.

Fig. 2. GSDME disruption skewed programmed neutrophil death to apoptosis.

a Representative flow cytometry plots of neutrophils undergoing spontaneous cell death. Results are representative of at least three biological replicates. b The percentage of AV⁻PI⁻, AV⁺PI⁻, and AV⁺PI⁺ cells at the indicated time points. All data are represented as mean ± SD. n = 3 independent repeats, *P < 0.05, **P < 0.01, ns non-significant by unpaired two-sided Student’s t test (Percentage of AV⁻PI⁻cells: 8 h, P = 0.0082; 12 h, P = 0.0279; 16 h, P = 0.0083; 20 h, P = 0.001; 24 h, P = 0.0288. Percentage of AV⁺PI⁻ cells: 4 h, P = 0.0064; 8 h, P = 0.0049; 12 h, P = 0.0022; 16 h, P = 0.0007. Percentage of AV⁺PI⁺ cells: 4 h, P = 0.0048; 8 h, P = 0.001). c The number of AV⁻PI⁻, AV⁺PI⁻, and AV⁺PI⁺ cells at the indicated time points. All data are represented as mean ± SD. n = 3 independent repeats, *P < 0.05, **P < 0.01, ns non-significant by unpaired two-sided Student’s t test (Number of AV⁺PI⁻ cells: 4 h, P = 0.0001; 8 h, P = 0.0006; 12 h, P = 0.0004; 16 h, P = 0.0024; 20 h, P = 0.0059; 24 h, P = 0.0455. Number of AV⁺PI⁺ cells: 4 h, P = 0.0017; 8 h, P = 0.0001; 16 h, P = 0.0335; 20 h, P = 0.0008). d Representative scatter plots and AV/PI staining of different populations on scatter plots of WT and Gsdme KO neutrophils cultured for 8 h. Results are representative of at least three biological replicates. e Caspase-3 cleavage during neutrophil spontaneous death. Western blot results are representative of at least three independent experiments. Relative amounts of cleaved caspase-3 were quantified based on densitometry analysis using NIH ImageJ software. Results are the means (±SD) of three independent experiments. *P < 0.05, **P < 0.01 versus WT neutrophils by two-sided Student’s -test (4 h, P = 0.0168; 8 h, P = 0.0032; 12 h, P = 0.0010; 16 h, P = 0.039). f Cleavage of PARP1 during neutrophil spontaneous death. Western blot results are representative of at least three independent experiments. Relative amounts of cleaved PARP1 were quantified based on densitometry. Results are the means (±SD) of three independent experiments. *P < 0.05, **P < 0.01 versus WT neutrophils by two-sided Student’s t-test (8 h, P = 0.386; 12 h, P = 0.004; 16 h, P = 0.0319). g Summary of the mode of programmed spontaneous death of WT and Gsdme KO neutrophils. Source data are provided as a Source Data file.

Fig. 3. GSDME cleavage during programed spontaneous neutrophil death was mediated by PR3 and caspase-3.

a GSDME cleavage in neutrophils treated with caspase inhibitors. Neutrophils from WT mouse bone marrow were treated with pan-caspase (QVD-OPH 100 μM), caspase-3 (Z-DEVD-FMK 100 μM), caspase-1 (VX765 25 μM), caspase-8 (Z-IETD-FMK 10 μM), or caspase-9 (Z-LEHD-FMK 20 μM) inhibitors for 16 h. β-actin was used as a protein loading control. Results are representative of three biological replicates. Relative levels of cleaved GSDME-NT were quantified based on densitometry. Results are means (±SD). n = 4 independent repeats. **P < 0.01 versus DMSO-treated neutrophils by two-sided Student’s t-test (QVD-OPH, P = 0.0001; Z-DEVD-FMK, P = 0.0029). b GSDME cleavage in neutrophils treated with serine protease inhibitor diisopropyl fluorophosphate (DFP). Neutrophils were treated with serine protease DFP (100 μM) or Z-DEVD-FMK (as a positive control) for 6 h. Representative immunoblots of three independent experiments are shown. The densitometry results are the means (±SD) of three independent experiments. *P < 0.05 versus DMSO-treated neutrophils by two-sided Student’s t-test (DFP, P = 0.022; Z-DEVD-FMK, P = 0.0211). c Cleavage of GSDME by serine proteases and caspase-3. Mouse GSDME was overexpressed in HEK293T cells. Cell lysates were incubated with the indicated proteases or caspase-3 for 1 h. GSDME cleavage was assessed by immunoblotting. UT untreated, EL elastase, CG cathepsin G, PR3 proteinase 3. Results are representative of three independent experiments. d Summary of GSDME cleavage during programed spontaneous neutrophil death. In aging neutrophils, PR3 is released from the granules, leading to cleavage and activation of procaspase-3. Active caspase-3 in turn cleaves GSDME to generate GSDME-NT, which targets the plasma membrane to induce membrane pore formation and lytic cell death. e GSDME cleavage in neutrophils during inflammation. Mice were challenged with heat-inactivated E. coli. (HIEC, 10⁷ cfu) for indicated time periods. GSDME cleavage in purified neutrophils was assessed by western blotting as described above. Representative immunoblots of three independent experiments are shown. The densitometry results are the means (±SD) of three independent experiments. Source data are provided as a Source Data file.

Fig. 4. Genetic ablation of neutrophil GSDME enhanced efferocytosis in peritonitis.

a Experimental scheme for assessing efferocytosis of ex vivo-aged neutrophils in peritonitis. Neutrophils from WT and Gsdme KO mice were cultured for 20 h and stained with 1 μM CFSE for 10 min. CFSE-labeled aged neutrophils were intraperitoneally injected into recipient mice challenged with heat-inactivated E. coli. (HIEC, 10⁷ cfu) for 48 h. 1 h post adoptive transfer, the peritoneal cavity was lavaged to collect cells to assess engulfment of CFSE⁺ neutrophils by macrophages. b The percentage of adoptively transferred neutrophils engulfed by macrophages was assessed by flow cytometry. After being engulfed, adoptively transferred neutrophils became ly6g negative but remained CFSE positive. All data are represented as mean ± SD, n = 4 mice, **P < 0.01 (P = 0.0001). c The percentage of macrophages engulfing at least one apoptotic neutrophil was assessed by flow cytometry. In the peritoneal cavity, F4/80⁺CD11b⁺Ly6c⁻ macrophages could engulf apoptotic neutrophils. After engulfing adoptively transferred neutrophils, macrophages became CFSE positive. All data are represented as mean ± SD, n = 3 mice, *P < 0.05 (P = 0.0126). d Experimental scheme for assessing efferocytosis of in vivo-aged neutrophils in peritonitis. Fresh neutrophils from WT and Gsdme KO mice were stained with 1 μM CFSE for 10 min. CFSE-labeled fresh neutrophils were intraperitoneally injected into recipient mice challenged with HIEC for 48 h. 6 h post adoptive transfer, the peritoneal cavity was lavaged to collect cells to assess engulfment of CFSE⁺ neutrophils by macrophages. e The percentage of adoptively transferred neutrophils engulfed by macrophages was assessed by flow cytometry. All data are represented as mean ± SD, n = 7–8 mice, *P < 0.05 (P = 0.0293). f The percentage of macrophages engulfing at least one apoptotic neutrophil was assessed by flow cytometry. All data are represented as mean ± SD, n = 4 mice; **P < 0.01 (P = 0.0002). Source data are provided as a Source Data file.

Fig. 5. Genetic ablation of neutrophil GSDME suppressed inflammatory responses in peritonitis.

a Schematic showing the experimental design for assessing inflammatory responses of WT and Gsdme whole-body KO mice during peritonitis. b IL-10 and IL-1β levels in the peritoneal lavage fluid. All data are represented as mean ± SD, n = 5 mice, *P < 0.05, **P < 0.01 (IL-10, P = 0.030; IL-1β, P = 0.0004). c Neutrophil percentage and number in the peritoneal cavity. All data are represented as mean ± SD, n = 7–8 mice, **P < 0.01 (Percentage of neutrophils, P = 0.0048; Number of neutrophils, P = 0.004). d Schematic showing the experimental design for assessing inflammatory responses of WT (Mrp8-cre(-)/Gsdme^fl/fl) and neutrophil-specific conditional Gsdme KO (Mrp8-cre(+)/Gsdme^Δ/Δ) mice during peritonitis. e In vivo neutrophil death. Lavage fluid from the peritoneal cavity was directly stained with APC-Ly6g to label neutrophils, and with Annexin V (AV) and SYTOX Orange to mark dead cells, all without centrifugation. The cells were subsequently fixed and immediately photographed. The images are representative of at least three independent experiments. Blue arrowheads indicate lytic dead neutrophils and white arrowheads indicate apoptotic neutrophils. The scale bar represents 10 μm. Data are presented as the mean ± SD, n = 3–4 independent repeats. f IL-10 and IL-1β levels in the peritoneal lavage fluid. All data are represented as mean ± SD, n = 6 mice, *P < 0.05, **P < 0.01 (IL-10, P = 0.0108; IL-1β, P = 0.0001). g Percentage and number of neutrophils in the peritoneal cavity. All data are represented as mean ± SD, n = 5–6 mice, *P < 0.05 (Percentage of neutrophils, P = 0.0299; Number of neutrophils, P = 0.0452). h Total protein level in the peritoneal lavage fluid. All data are represented as mean ± SD, n = 4–5 mice, *P < 0.05 (P = 0.0308). i MPO level in the peritoneal lavage fluid. All data are represented as mean ± SD, n = 4–5 mice, *P < 0.05 (P = 0.0368). Statistical significance was examined by unpaired two-sided Student’s t test (b, c, e, f, h, I, l, m, n, o). Source data are provided as a Source Data file.

Fig. 6. GSDME disruption in neutrophils attenuated inflammation and alleviated lung injury during LPS-induced pneumonia.

a Schematic showing the experimental design for LPS-induced acute lung injury. b Quantification of total leukocytes and neutrophils in the BALF. All data are represented as mean ± SD, n = 4–5 mice/group, *P < 0.05 (WBC, P = 0.0247; Neutrophil, P = 0.0194). c Total protein in the BALF. Data are represented as mean ± SD, n = 4 mice/group, *P < 0.05 (P = 0.0350). d LDH release in the BALF. Data are represented as mean ± SD, n = 3–4/group, *P < 0.05 (P = 0.0201). e IL-6 and IL-1β levels in BALF. Data are represented as mean ± SD, n = 4 mice/group, *P < 0.05, **P < 0.01 (IL-6, P = 0.0309; IL-1β, P = 0.0033). f Hematoxylin and eosin (H&E) staining of lung sections. Results are representative of three biological replicates. g Lung inflammation and damage assessed as neutrophil recruitment (vol % of alveoli) and edema area of lung. Data are represented as mean ± SD, n = 4–5 mice/group, **P < 0.01 (Recruited neutrophils, P = 0.0072; Edema area of total area, P = 0.0058). h Experimental scheme for assessing LPS-induced pulmonary inflammation and associated lung injury in WT and neutrophil-specific Gsdme conditional KO mice. i Quantification of total leukocytes and neutrophils in the BALF. Data are represented as mean ± SD, n = 4–5 mice/group, *P < 0.05 (WBC, P = 0.0238; Neutrophil, P = 0.0276). j Total protein in the BALF. Data are represented as mean ± SD, n = 3–4 mice/group, *P < 0.05 (P = 0.0214). k LDH release in the BALF. Data are represented as mean ± SD, n = 3–4 mice/group, *P < 0.05 (P = 0.0487). l IL-6 and IL-1β levels in the BALF. Data are represented as mean ± SD, n = 3–4 mice/group, *P < 0.05, **P < 0.01 (IL-6, P = 0.0123; IL-1β, P = 0.0074). m H&E staining of lung sections. Results are representative of three biological replicates. n Pulmonary inflammation and tissue damage measured as neutrophil recruitment and edema area of lung. Data are represented as mean ± SD, n = 3 mice/group, **P < 0.01 (P = 0.0001). Statistical significance was examined by unpaired two-sided Student’s t test (b, c, d, e, g, i, j, k, l, n). Source data are provided as a Source Data file.

Fig. 7. GSDME disruption in neutrophils enhanced efferocytosis in the inflamed lungs.

a Experimental scheme for assessing efferocytosis of ex vivo-aged neutrophils in LPS-induced acute lung injury. BM neutrophils from WT and Gsdme KO mice were cultured for 20 h and stained with 1 μM CFSE for 10 min. CFSE-labeled aged neutrophils were intratracheally injected into recipient mice challenged with 5 mg/kg LPS. 1 h post adoptive transfer, lungs were lavaged to collect cells to assess the engulfment of CFSE⁺ neutrophils by macrophages. b The percentage of macrophages engulfing at least one apoptotic neutrophil was assessed by flow cytometry as described in Fig. 4. The flow cytometric images are representative of at least three independent experiments. All data are represented as mean ± SD, n = 6 mice, **P < 0.01 by two-sided Student’s t-test (P = 0.0021). c The percentage of adoptively transferred neutrophils engulfed by macrophages was assessed by flow cytometry as described in Fig. 4. The flow cytometric images are representative of at least three independent experiments. All data are represented as mean ± SD, n = 4 mice, **P < 0.01 by two-sided Student’s t-test (P = 0.0077). d Experimental scheme for assessing efferocytosis of in vivo-aged neutrophils in LPS-induced acute lung injury. Fresh neutrophils from WT and Gsdme KO mice were stained with CFSE. CFSE-labeled fresh neutrophils were intratracheally injected into recipient mice challenged with LPS. 6 h post adoptive transfer, lungs were lavaged to collect cells to assess engulfment of CFSE⁺ neutrophils by macrophages. e The percentage of macrophages engulfing at least one apoptotic neutrophil was assessed by flow cytometry. The flow cytometric pictures are representative of at least three independent experiments. All data are represented as mean ± SD, n = 5 mice; **P < 0.01 by two-sided Student’s t-test (P = 0.0035). f The percentage of adoptively transferred neutrophils engulfed by macrophages was assessed by flow cytometry. The flow cytometric pictures are representative of at least three independent experiments. All data are represented as mean ± SD, n = 7 mice, **P < 0.01 by two-sided Student’s t-test (P = 0.0001). Source data are provided as a Source Data file.

Fig. 8. GSDME disruption in neutrophils attenuated inflammation during S. aureus-induced pneumonia.

a Experimental scheme for assessing S. aureus-induced pulmonary inflammation and associated lung injury in WT and neutrophil-specific Gsdme conditional KO mice. Pneumonia was induced in neutrophil-specific Gsdme conditional KO (Mrp8-cre(+)/Gsdme^Δ/Δ) and their WT (Mrp8-cre(-)/Gsdme^fl/fl) littermates using S. aureus. Neutrophil recruitment, pulmonary inflammation, and lung damage were evaluated at different time points post-infection. b Quantification of total leukocytes and neutrophils in the BALF from Mrp8-cre(-)/Gsdme^fl/fl and Mrp8-cre(+)/Gsdme^Δ/Δ mice post-pneumonia induction at indicated time points. Data are represented as mean ± SEM (n = 4–6 mice /group). *P < 0.05, **P < 0.01 by two-sided Student’s t-test (WBC: 24 h, P = 0.0115; 48 h, P = 0.0125. Neutrophils: 24 h, P = 0.0107; 48 h, P = 0.0058). c IL-1β, IL-1, and IL-10 levels in the BALF from mice infected with S. aureus. Data are shown as mean ± SEM (n = 4–6 mice /group). *P < 0.05. ND not detectable (IL-10, P = 0.0221. IL-1β: 24 h, P = 0.0118; 48 h, P = 0.0140. IL-6: 24 h, P = 0.0081; 48 h, P = 0.0435). d LDH release in BALF post-infection. Data are represented as mean ± SEM (n = 4–6 mice /group). *P < 0.05 by two-sided Student’s t-test (24 h, P = 0.0049; 48 h, P = 0.0215). e Total protein concentration in the BALF. Data are shown as mean ± SEM (n = 4–6 mice/group). *P < 0.05 by two-sided Student’s t-test (24 h, P = 0.009; 48 h, P = 0.0453). f Bacterial burden in the lungs of Mrp8-cre(-)/Gsdme^fl/fl and Mrp8-cre(+)/Gsdme^Δ/Δ mice post S. aureus infection at the specified time points. Data are represented as mean ± SEM (n = 4–6 mice/group). **P < 0.01. NS not significant by two-sided Student’s t-test (P = 0.0035). g Bacterial count in the BALF from Mrp8-cre(-)/Gsdme^fl/fl and Mrp8-cre(+)/Gsdme^Δ/Δ mice post S. aureus infection at the specified time points. Data are presented as mean ± SEM (n = 4–6/ mice group). ***P < 0.001. NS not significant by two-sided Student’s t-test (P = 0.0007). Source data are provided as a Source Data file.

Fig. 9. GSDME disruption in neutrophils alleviated lung injury during S. aureus-induced pneumonia.

a Hematoxylin and eosin (H&E) stained lung sections from Mrp8-cre(-)/Gsdme^fl/fl and Mrp8-cre(+)/Gsdme^Δ/Δ mice pre and post S. aureus-induced pneumonia. The scale bar indicates 100 μm. Results represent a minimum of three biological replicates. b Assessment of pulmonary inflammation and damage via neutrophil recruitment (vol % of alveoli) and edema in the lung area. Data are represented as mean ± SEM (n = 4 mice/group). **P < 0.01 by two-sided Student’s t-test. The data represents at least three independent experiments (Recruited neutrophils, P = 0.0074; Edema area of total area, P = 0.0055). c Survival analysis of WT and neutrophil-specific Gsdme conditional KO post S. aureus infection. Mrp8-cre(-)/Gsdme^fl/fl and Mrp8-cre(+)/Gsdme^Δ/Δ mice (n = 8–9 mice/group) were intratracheally infected with S. aureus (1.5 × 10⁸ CFUs/mouse) and monitored for survival over 7 days. Data were analyzed using the Gehan-Breslow-Wilcoxon test. *P < 0.05 (P = 0.0383). Source data are provided as a Source Data file.

Fig. 10. Neutrophil-specific GSDME deletion attenuated inflammation and alleviated lung injury in acid aspiration pneumonitis.

a Experimental scheme for assessing acid-induced pulmonary inflammation and associated lung injury in WT (Mrp8-cre(-)/Gsdme^fl/fl) and neutrophil-specific Gsdme KO (Mrp8-cre(+)/Gsdme^Δ/Δ) mice. Mice were challenged with 0.1 N HCl (2.5 μl/g body weight) and sacrificed after 48 h. b The number and percentage of neutrophils in BALF. WBC and neutrophil counts were determined by FACS using counting beads as described in Fig. S3. Cells were stained with APC-CD11b and PE-Ly6g antibodies. Neutrophils were identified as Ly6g⁺CD11b⁺ cells. All data are represented as mean ± SD, n = 6–7 mice, *P < 0.05 by two-sided Student’s t-test (WBC count, P = 0.0469; Neutrophil count, P = 0.0283; Percentage of neutrophils, P = 0.0407). c IL-10 and IL-6 levels in BALF of WT and neutrophil-specific Gsdme KO mice were determined by enzyme-linked immunosorbent assay (ELISA). All data are represented as mean ± SD, n = 7−8 mice, *P < 0.05, **P < 0.01 by two-sided Student’s t-test (IL-10, P = 0.0099; IL-6, P = 0.0213). d Representative images of the lungs of acid-challenged WT and neutrophil-specific Gsdme KO mice. The lung lobes challenged with HCl are indicated. Results are representative of at least three biological replicates. e BALF total protein level. Protein accumulation in inflamed lungs was measured using a protein assay kit. All data are represented as mean ± SD, n = 6–7 mice, *P < 0.05 by two-sided Student’s t-test (P = 0.0478). f Representative H&E-stained images of acid-challenged lung tissues. Results are representative of at least three biological replicates. g Neutrophil accumulation in alveoli was quantified as volume fraction of the alveolar space occupied by neutrophils. Pulmonary edema formation was quantified as the percentage of edema area in the total parenchymal region. Data are represented as mean ± SD of three experiments. n ≥ 6 mice in each group. **P < 0.01 by two-sided Student’s t-test (Edema area of total area, P = 0.0014; Recruited neutrophils, P = 0.0001). Source data are provided as a Source Data file.