GSDME-mediated pyroptosis promotes the progression and associated inflammation of atherosclerosis

Abstract

Pyroptosis, a type of Gasdermin-mediated cell death, contributes to an exacerbation of inflammation. To test the hypothesis that GSDME-mediated pyroptosis aggravates the progression of atherosclerosis, we generate ApoE and GSDME dual deficiency mice. As compared with the control mice, GSDME^-/-/ApoE^-/- mice show a reduction of atherosclerotic lesion area and inflammatory response when induced with a high-fat diet. Human atherosclerosis single-cell transcriptome analysis demonstrates that GSDME is mainly expressed in macrophages. In vitro, oxidized low-density lipoprotein (ox-LDL) induces GSDME expression and pyroptosis in macrophages. Mechanistically, ablation of GSDME in macrophages represses ox-LDL-induced inflammation and macrophage pyroptosis. Moreover, the signal transducer and activator of transcription 3 (STAT3) directly correlates with and positively regulates GSDME expression. This study explores the transcriptional mechanisms of GSDME during atherosclerosis development and indicates that GSDME-mediated pyroptosis in the progression of atherosclerosis could be a potential therapeutic approach for atherosclerosis.

Fig. 1. Indications of occurrence of pyroptosis in human and mouse atherosclerotic plaques.

a Electron microscopic appearance of macrophage death in human atherosclerotic plaques. Fragments of a dying macrophage with still recognizable lysosomes (L). Macrophages with normal-looking nucleus (N) and numerous lysosomes (L) containing inclusions of phagocytosed cell remnants. Representative electron microscopic analysis of six different human plaques from one experiment. Scale bar: 5 μm; Scale bar: 2 μm (magnification). b Transmission electron microscopy (TEM) images of a suspected pyroptotic macrophage in human carotid artery atherosclerotic plaques. The arrows indicate pore formation and discontinuity of the cell plasma membrane. Scale bar: 2 μm (right), Scale bar: 1 μm (middle), Scale bar: 500 nm (right); Asterisks indicate cell membrane thickening. Representative electron microscopic analysis of six different human plaques from one experiment. c Immunohistochemical staining of GSDME, caspase 3, IL-1β in human carotid artery compared with control vessels (human normal abdominal artery derived from autopsy) and the protein integrated optical density (IOD)/area between the two groups are shown (mean ± SEM, n = 6). GSDME (**P = 0.004), caspase 3 (**P = 0.004), IL-1β (**P = 0.004). Scale bar: 200 μm; Scale bar: 100 μm (magnification) . d In situ detection of the interaction of GSDME and caspase 3 in human carotid atherosclerotic plaques and control vessels (mean ± SEM, n = 5) using a generalized proximity ligation assay compared with immunohistochemical staining of IL-1β. **P = 0.009. For all panels, P value was determined by two-tailed Mann–Whitney U test. Scale bar: 100 μm; Scale bar: 50 μm (magnification) . Source data are provided as a Source Data file.

Fig. 2. GSDME is mainly expressed in atherosclerotic macrophages.

Cells with over 10% mitochondrial gene content were removed. Single-cell transcriptomic profiling and dissection of the cellular heterogeneity of 5370 cells from human carotid artery advanced atherosclerotic lesions of patients undergoing carotid endarterectomy. a Uniform manifold approximation and projection (UMAP) dimensional reduction of the 5370 atheroma cells. Main cell types were identified (upper panel). UMAP distribution of clustering revealed four distinct myeloid populations (lower panel). Population identities were determined based on marker gene expression. b Biaxial scatter plots and Pie graphs show the expression pattern of GSDME among the different subgroups in the total atheroma cells. Color scale represents expression levels; gray: low, red: high. Atherosclerotic cells and myeloid cells subgroups are labeled by colors (Atherosclerotic cells subgroups: blue, myeloid cells; orange, fibroblasts; green, endothelial cells; red, proliferating cells; purple, T cells; brown, mast cells; pink, B cells. Myeloid cells subgroups: blue, M1 macrophages; green, monocyte; orange, M2 macrophage; red, dendritic cells. c The myeloid cell development trajectory visualization in a biaxial scatter plot. Color scale represents the development stage, dark colors indicate early development (left upper panel). Pseudo-time developmental analysis demonstrated a branched single-cell trajectory of myeloid cells beginning with monocytes and dendritic cells ((left lower panel). Myeloid cells are labeled by colors. Two-dimensional plots showing the dynamic expression of myeloid cells marker genes (right panel). Scatter plots for example DEGs of myeloid cells depicting expression level as a function of pseudo time score. Each point represents a single-cell. The color scheme depicts a cluster. d Representative immunofluorescence image of CD68 and GSDME co-staining in human carotid artery atheroma and normal aorta, n = 4/4 atheroma/normal aorta. The arrows indicate nonspecific staining of elastic fibers. Scale bar: 100 μm; Scale bar: 50 μm (magnification). Source data are provided as a Source Data file.

Fig. 3. GSDME expression is increased in atherosclerosis.

a GSDME gene expression in human carotid atheroma (stage IV) and distant macroscopically intact tissue (stage I and II) (accession no. GSE43292, n = 32/group). Boxplots span from the first until the third quartile of the data distribution, and the horizontal line indicates the median value of the data. The whiskers indicate the minimum and maximum values found within 1.5 times the interquartile range (***P < 0.001). b Western blotting analysis of GSDME and cleaved caspase 3 expression in human carotid artery atherosclerotic lesions derived from carotid endarterectomy. GSDME (*P = 0.036), N-GSDME (**P = 0.003), cleaved caspase 3 (P = 0.303). n = 4/group. c GSDME protein (*P = 0.045), N-GSDME protein (P = 0.125), and d GSDME mRNA (***P < 0.001) were determined by western blotting and quantitative polymerase chain reaction (qPCR), respectively, in the aortas of male atherosclerosis-prone ApoE^−/− (apolipoprotein E deficient) mice fed with an ND (normal diet) or WD (western diet) for 12 wk. n = 4/group from one experiment. ns, not significant. e Quantitative polymerase chain reaction (qPCR) analysis of the gene expression relative to inflammation as well as GSDME in aortas of ApoE^−/− mice fed with ND or HFD and in control aortas from WT mice. WT, wild type. ^#P < 0.001 WD vs ND or control aortas from WT mice. n = 4/group. P value was determined by unpaired two-tailed Student’s t test (a–d) or one-way ANOVA with Bonferroni post hoc test for multiple comparisons (e). Data are expressed as mean ± SEM in b–e. Quantification on the blots derive from samples of the same experiment and gels/blots were processed in parallel. Source data are provided as a Source Data file.

Fig. 4. GSDME deficiency attenuates atherosclerotic lesion area and size.

Male 8-wk-old ApoE^−/− and GSDME ^−/−/ApoE^−/− mice were fed with a high-fat diet for 12 wk. a Serum total cholesterol level. P = 0.549. b Serum total HDL level. P = 0.874. cSerum total LDL level. P = 0.796. (n = 7/group). ns, not significant. d Representative image of Oil Red O staining of the whole aorta. Plaque area was quantified as the percentage of the total surface area of the aorta, n = 8/7 (ApoE^−/− mice / GSDME ^−/−/ApoE^−/− mice), **P = 0.009. e, f H&E (hematoxylin-eosin) and O.R.O (Oil Red O) staining image of lesions in aortic sinus sections (*P = 0.031) and brachiocephalic artery (**P = 0.005, ***P < 0.001), respectively, and quantitative data for plaque size and necrotic area, n = 10/8 (ApoE^−/− mice /GSDME ^−/−/ApoE^−/− mice in e), n = 6/6 (ApoE^−/− mice / GSDME ^−/−/ApoE^−/− mice in f). NC, necrotic core. g PCR analysis of gene expression related to inflammation in the aortas harvested from male 8-wk-old ApoE^−/−mice and GSDME ^−/−/ApoE^−/−mice fed with high-fat diet for 12 wks. n = 3/3 (ApoE^−/−mice /GSDME ^−/−/ApoE^−/− mice). IL-1β (**P = 0.008), TNF (*P = 0.042), MCP-1 (*P = 0.010), IL-6 (*P = 0.024). Scale bars are 1 mm (e) or 500um (f). All panels, data were expressed as mean ± SEM. P value was determined by unpaired two-tailed Student’s t test. Source data are provided as a Source Data file.

Fig. 5. GSDME is induced by ox-LDL in macrophages.

a Western blotting analysis of GSDME and N-GSDME expression in mouse PM treated with ox-LDL (100 μg/ml) for 24 h. GSDME (*P = 0.025), N-GSDME (*P = 0.050), n = 4 independent experiments. b Real-time polymerase chain reaction (PCR) analysis of GSDME mRNA in level in PMs. **P = 0.005, n = 4 independent experiments. c Western blotting analysis of GSDME and N-GSDME expression in BMDMs treated with ox-LDL (100 μg/ml) for 24 h. GSDME (**P = 0.001), N-GSDME (P = 0.688), n = 4 independent experiments. d Real-time polymerase chain reaction (PCR) analysis of GSDME mRNA in level in BMDM cells. n = 4 independent experiments. ***P < 0.001. e GSDME and caspase 3 in whole-cell lysates were pulled down by the appropriate primary antibody and subjected to western blotting analysis to detect GSDME and caspase 3. Representative for three independent experiments. f The LDH content in the culture supernatants of WT and GSDME^−/− peritoneal macrophages treated with ox-LDL of indicated concentrations for 24 h. The data shown represent one of four separate experiments (n = 4). *P = 0.018, ***P < 0.001. The fraction of cells was normalized by the experimental LDH release as a percentage of the positive controls. For all panels, unpaired two-tailed Student’s t test was used for between-group comparisons, and one-way ANOVA with Bonferroni post hoc test for multiple comparisons. Data are expressed as mean ± SEM. Con indicates control. ox-LDL indicates oxidized Low-density lipoprotein. Source data are provided as a Source Data file.

Fig. 6. Macrophage GSDME deficiency represses macrophage inflammation and pyroptosis.

a Transcriptomic analysis of oxLDL-stimulated WT and GSDME^−/− macrophages. Peritoneal macrophages were isolated from WT and GSDME^−/− mice respectively and treated with oxLDL for 24 hours. Differentially expressed genes (DEGs) were shown by a cluster heat map. Color scale represents relative expression levels; blue: low, red: high. b Volcano plots showed the DEGs between WT and GSDME^−/− macrophages. Each dot represents a specific gene, with red dots indicating significantly up-regulated genes and green dots indicating significantly downregulated genes. DEGs were identified by a fold change>2 (Log₂FC > 1 or Log₂FC < −1) and P adjust <0.05. c GO Chord plot of top 10 ranked overrepresented GO terms. Chords represent the detailed relationship between the expression levels of DEGs (left semicircle parameter) and their enriched GO terms (right semicircle parameter). Color scale represents Log₂FC of the DEGs. blue: low, red: high. d Representative images (left) of transwell migration assay showing migration of WT and GSDME^−/− macrophages with the addition of MCP-1 (600 ng/ml) to the medium in the lower chambers. Bar graph (right) shows the quantitative estimation of the number of migrated cells. n = 6/group. *P = 0.027. Scale bars are 50 μm. e Representative images(left) of the wound healing test are shown for WT and GSDME^−/− macrophages. Bar graph (right) shows the migration index. n = 8/group. *P = 0.049. Scale bars are 100 μm. f Real-time polymerase reaction (PCR) analysis of gene expression related to inflammation in WT and GSDME^−/− mice peritoneal macrophage treated with oxLDL(100 μg/ml) for 24 hours. n = 8/group. TNF (***P < 0.001), IL-1β (***P < 0.001), MCP-1 (**P = 0.001 WT Con vs WT ox-LDL; **P = 0.004 WT ox-LDL vs GSDME^−/−, ox-LDL). g Western blotting analysis of WT and GSDME^−/− peritoneal macrophages treated with ox-LDL. The experiments were repeated three times and the results were similar. h WT and GSDME^−/− peritoneal macrophages were treated with TNF (100 ng/ml) +cycloheximide (CHX; 20 μg/ml) for the indicated time. LDH activity in the medium was measured. The fraction of cells was normalized by the experimental LDH release as a percentage of the positive controls. n = 8/group. ***P < 0.001. Unpaired two-tailed Student’s t test was used for between-group comparisons (d, e). One-way ANOVA with Bonferroni post hoc test for multiple comparisons (f). Repetitive Measure Analysis of Variance ANOVA was used in analyzing LDH at different time h. For all panels, data are expressed as mean ± SEM. Each experiment was repeated independently three times for d–h. Source data are provided as a Source Data file.

Fig. 7. STAT3 targets the GSDME promoter and activated GSDME transcription.

a Dot plots showing the expression pattern of STAT3 and GSDME genes among the myeloid cell subgroups in human atherosclerotic plaques. Color scale represents relative expression levels; blue: low, red: high. b Representative western blotting of p-STAT3, STAT3, GSDME, and N-GSDME in mouse PMs treated with ox-LDL (100 μg/ml) for 24 hours. c Western blotting analysis of p-STAT3, STAT3, GSDME, N-GSDME, cleaved caspase 3 and caspase 3 in PMs treated with TNF (100 ng/ml) for 24 hours. d PMs were transfected with siRNA control (Si-NC;100pmol) or siRNA-STAT3(Si-STAT3;100pmol) for 72 hours, and GSDME protein were determined by western blotting (n = 4/group, **P = 0.001). e PMs were transfected with FLAG-pcDNA3.1-STAT3(5 μg/ul) or pcDNA3.1-NC(5 μg/ul) for 48 hours, and GSDME protein were determined by western blotting (n = 4/group, **P = 0.001). f PMs were transfected with siRNA control (Si-NC;100pmol) or siRNA-STAT3(Si-STAT3;100pmol) for 72 hours, and GSDME mRNA was determined by quantitative polymerase reaction (n = 3/group, *P = 0.012). g PMs were transfected with FLAG-pcDNA3.1-STAT3(5 μg/ul) or pcDNA3.1-NC(5 μg/ul) for 48 hours, and GSDME mRNA was determined by quantitative polymerase reaction (n = 3/group, P = 0.220). h Depiction of 4 putative stat3 binding sites, at -1373/-1364(site1), -936/-927(site2), -366/-357(site3), and -60/‐51(site4) bp upstream of transcription initiation site in the mouse GSDME promoter. i Chromatin immunoprecipitation analysis with antibodies against pSTAT3 or IgG, soluble chromatin from PMs, and primers targeting the region spanning the 4 binding sites in the GSDME promoter (n = 4/group, **P = 0.007 site1, **P = 0.008 site3, ***P < 0.001). j 293 T cells were transfected with a mouse GSDME promoter-driven luciferase vector (0.4 μg) and pcDNA3.1-STAT3C or pcDNA3.1-NC(0.2 μg). After 48 hours, the luciferase activity was measured and normalized to Renilla activity (n = 4/group, ***P < 0.001). For all panels, data are expressed as mean ± SEM. Unpaired two-tailed Student’s t test was used for between-group comparisons. Each experiment was repeated independently three times. Source data are provided as a Source Data file.

Fig. 8. Graphic summary of the mechanisms for GSDME accelerating inflammatory response in atherosclerosis.

Transcriptional activation and the upregulated GSDME augment the activity of caspase 3 and promote apoptosis converts to pyroptosis in macrophages during the process of atherosclerosis.