Renal tubular GSDME protects cisplatin nephrotoxicity by impeding OGT-STAT3-S100A7A axis in male mice

Abstract

Gasdermin E (GSDME) is known as a key executive protein of pro-inflammatory pyroptosis. However, the function diversity of GSDME needs further investigation. Here, we show that GSDME expression is downregulated in kidney tissues after cisplatin treatment without detectable N-terminal fragment. Global and tubule-specific Gsdme deficiency aggravates cisplatin-induced renal injury. Mechanistically, loss of GSDME in proximal tubular cells facilitates the recruitment of OGT to the CUL4B-DDB1-WDR26 E3 ubiquitin ligase complex, promoting OGT degradation and subsequently reducing STAT3 O-GlcNAcylation. This post-translational shift enhances STAT3 phosphorylation and induces upregulation of its downstream target gene, S100a7a. Elevated S100A7A promotes macrophage infiltration via RAGE activation, amplifying renal inflammation. Tubule-specific depleting S100a7a improves renal function and reduces renal injury and inflammation. These findings uncover a protective, non-pyroptotic function of GSDME in modulating O-GlcNAcylation and STAT3-S100A7A-RAGE signaling to maintain renal homeostasis under cisplatin stress in male mice.

Fig. 1. GSDME deficiency aggravates cisplatin nephrotoxicity.

a Representative immunofluorescence images showing co-staining of GSDME (red) with LTL (green) in renal sections from mice on days 0-3 after cisplatin treatment. b Quantification of GSDME immunofluorescence intensity in panel (a) (n = 12 biological replicates). c RT-qPCR analysis of Gsdme in renal tissues from mice on days 0-3 after cisplatin treatment (n = 12 biological replicates). d, e Western blot analysis (d) and densitometric quantification (e) of GSDME protein levels in renal tissues from mice on days 0-3 after cisplatin treatment (n = 12 biological replicates). f Serum creatinine levels in WT and Gsdme-KO mice on days 0-3 after cisplatin treatment (n = 12 biological replicates). g,h Representative images of H&E staining (g) and corresponding renal tubular injury scoring (h) of renal sections from WT and Gsdme-KO mice on days 0-3 after cisplatin treatment (n = 12 biological replicates). i,j Western blot analysis (i) and densitometric quantification (j) of NGAL and GSDME protein levels in renal tissues from WT and Gsdme-KO mice on day 3 after cisplatin treatment (n = 12 biological replicates). k RT-qPCR analysis of Tnf, Ccl2, Il6, and Tlr4 mRNA expression in renal tissues from WT and Gsdme-KO mice on day 3 after cisplatin treatment (n = 12 biological replicates). l–n Representative immunohistochemistry staining (l) of F4/80 and MPO in renal sections from WT and Gsdme-KO mice on day 3 after cisplatin treatment, with quantitative analysis of F4/80 (m) and MPO (n) positive cells (n = 12 biological replicates). Data were expressed as means ± SEM. ^##P < 0.01, ^###P < 0.001 versus WT mice treated with saline. ^&P < 0.05 versus cisplatin-treated WT mice (day 2). ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 versus cisplatin-treated WT mice (day 3). P-values were determined by one-way ANOVA (with Least-Significant Difference [LSD] or Games-Howell post hoc tests) in panels (b, c, e, f, j, k, m, n). P values were determined by a two-tailed nonparametric test (with Wilcoxon test) in panel (h). All scale bars, 100 μm. Source data are provided as a Source Data file.

Fig. 2. Cell populations and cellular attributes of WT and Gsdme-KO mouse kidneys revealed by scRNA-seq analysis.

a UMAP visualization of single-cell transcriptomes from kidneys of Gsdme-KO mice and their WT littermates. Colors represent distinct cell clusters; each dot corresponds to a single cell. b UMAP plot displaying the subclusters of PTCs. c Bubble plot showing the gene expression profiles of Gsdme across different renal cell populations. PTCs Proximal Tubular Cells, MPC Mono-nuclear Phagocyte, T cells T lymphocytes, Neutro Neutrophil, B cells B lymphocytes, CD Collecting Duct, LOH Loop of Henle, ECs Endothelial Cells, DCT Distal Convoluted tubule, NK cells Natural Killer cells, Myofib Myofibroblasts, PECs/Podo Parietal Epithelial Cells/Podocyte. d Bubble plot showing the gene expression profiles of Gsdme across different WT PTCs subpopulations. e Proportions of different PTCs subpopulations in different groups. f Pseudotime analysis illustrating the differentiation trajectory of PTCs. g Visualization of PTCs differentiation trajectories stratified by the indicated experimental condition. h Heatmap showing differentially expressed genes (DEGs, rows) along pseudotime (columns), hierarchically clustered into four distinct expression patterns, each corresponding to a specific trajectory phase. Right panel: GO-BP terms associated with each gene expression pattern.

Fig. 3. GSDME deficiency in bone marrow-derived cells has no impact on cisplatin-induced renal tubular injury and inflammatory responses.

a Schematic of the experimental timeline: Gsdme-KO mice and their WT littermates were irradiated, followed by bone marrow transplantation from donors of different genotypes, and subsequently subjected to cisplatin. The illustration elements were created in BioRender. Chen, Q. (2025) https://BioRender.com/k19218v. b Serum creatinine levels in chimeric mice on day 3 after cisplatin treatment (n = 8 biological replicates). c, d Representative images of H&E staining (c) and corresponding renal tubular injury scoring (d) of renal sections from chimeric mice on day 3 after cisplatin treatment (n = 8 biological replicates). e RT-qPCR analysis of Tnf, Ccl2, Il6, and Tlr4 mRNA expression in renal tissues from chimeric mice on day 3 after cisplatin treatment (n = 8 biological replicates). f Representative immunohistochemistry staining of F4/80 in renal sections from chimeric mice on day 3 after cisplatin treatment. g Quantitative analysis of F4/80-positive cells from experiments in panel (f) (n = 8 biological replicates). h Representative immunohistochemistry staining of MPO in renal sections from chimeric mice on day 3 after cisplatin treatment. i Quantitative analysis of MPO-positive cells from experiments in panel (h) (n = 8 biological replicates). Data were expressed as means ± SEM. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 versus WT → WT chimeric mice. ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 versus KO → WT chimeric mice. P-values were determined by one-way ANOVA (with LSD or Games-Howell post hoc tests) in (b, e, g, i). P-values were determined by a two-tailed nonparametric test (with Wilcoxon test) in (d). All scale bars, 100 μm. Source data are provided as a Source Data file.

Fig. 4. GSDME depletion exacerbates cisplatin nephrotoxicity by upregulating S100A7A expression in PTCs.

a Venn diagram illustrating the overlap of upregulated DEGs among three comparisons: KO ctrl vs. WT ctrl, WT cisplatin vs. WT ctrl, and KO cisplatin vs. WT cisplatin. b, c Western blot (b) and densitometric quantification (c) of S100A7A in primary PTCs (n = 4 biological replicates). d, e Western blot (e) and densitometric quantification (d) of S100A7A in renal tissues from cisplatin-treated WT and Gsdme-KO mice (n = 12 biological replicates). f S100A7A concentrations in the culture medium from primary PTCs (n = 3 biological replicates). g Gene expression z-scores of S100a7a in PTCs from WT cisplatin kidneys in scRNA-seq data. h, i Representative S100A7A immunohistochemistry staining (h) and H-score analysis (i) in cisplatin-treated WT and Gsdme-KO mice (n = 12 biological replicates). j Survival analysis of cisplatin-treated WT and Gsdme-KO mice injected with lenti-NC or lenti-S100a7a-shRNA. (k) Serum creatinine levels of cisplatin-treated WT and Gsdme-KO mice injected with lenti-NC or lenti-S100a7a-shRNA (n = 12 biological replicates). l, m Representative H&E-stained (l) and histopathology scoring (m) of cisplatin-treated WT and Gsdme-KO mice injected with lenti-NC or lenti-S100a7a-shRNA (n = 12 biological replicates). n RT-qPCR analysis of Tnf and Il6 from cisplatin-treated WT and Gsdme-KO mice injected with lenti-NC or lenti-S100a7a-shRNA (n = 12 biological replicates). o–r Representative F4/80 (o) and MPO (p) immunohistochemistry staining of cisplatin-treated WT and Gsdme-KO mice injected with lenti-NC or lenti-S100a7a-shRNA. Quantitative analysis of F4/80 (q) and MPO positive cells (r) (n = 12 biological replicates). Data were expressed as means ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001 versus WT cisplatin group. &P < 0.05, & &P < 0.01, & & &P < 0.001 versus WT ctrl group. #P < 0.05, ##P < 0.01, ###P < 0.001 versus Gsdme-KO mice treated with cisplatin and lenti-S100a7a-RNA. P-values were determined by one-way ANOVA (with LSD or Games-Howell post hoc tests) in (c, d, f, i, k, m, n, q, r). P-values were determined by a two-tailed nonparametric test (with Wilcoxon test) in (m). All scale bars, 100 μm. Source data are provided as a Source Data file.

Fig. 5. Tubule-specific knockout of S100a7a mitigates cisplatin nephrotoxicity.

a Schematic illustration of renal tubule-specific S100a7a knockout (S100a7a^Ksp-KO) mice generation by cross-breeding Ksp-Cre mice with S100a7a^flox/flox mice. The illustration elements were created in BioRender. Chen, Q. (2025) https://BioRender.com/k19218v. b Representative immunofluorescence images of S100A7A (red) and LTL (green) from WT and S100a7a^Ksp-KO mice on day 3 after cisplatin treatment. Scale bars, 100 μm. c Serum creatinine levels in S100a7a^Ksp-KO mice and their WT littermates on day 3 after cisplatin treatment (n = 10 biological replicates). d,e Representative images of H&E staining (d) and corresponding renal tubular injury scoring (e) of renal sections from S100a7a^Ksp-KO mice and their WT littermates on day 3 after cisplatin treatment (n = 10 biological replicates). Scale bars, 100 μm. f RT-qPCR analysis of Tnf and Il6 mRNA expression from S100a7a^Ksp-KO mice and their WT littermates on day 3 after cisplatin treatment (n = 10 biological replicates). g,h Western blot analysis (g) and densitometric quantification (h) of NGAL, cleaved-caspase-3, caspase-3, and S100A7A protein levels in renal tissues from S100a7a^Ksp-KO mice and their WT littermates on day 3 after cisplatin treatment (n = 10 biological replicates). i Representative images of TUNEL staining of renal sections from S100a7a^Ksp-KO mice and their WT littermates on day 3 after cisplatin treatment. Boxed areas were shown at higher magnification in the lower panel. Scale bars, 200 μm. j Quantitative analysis of TUNEL-positive tubules from experiments in panel (i) (n = 10 biological replicates). k–m Representative immunohistochemistry staining (k) of F4/80 and MPO in renal sections from S100a7a^Ksp-KO mice and their WT littermates on day 3 after cisplatin treatment. Quantitative analysis of F4/80-positive cells (l) and MPO-positive cells (m) (n = 10 biological replicates). Data were expressed as means ± SEM. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 versus WT cisplatin mice. P-values were determined by one-way ANOVA (with LSD or Games-Howell post hoc tests) in (c, f, h, j, l, m). P values were determined by a two-tailed nonparametric test (with Wilcoxon test) in (e). Source data are provided as a Source Data file.

Fig. 6. S100A7A exacerbates cisplatin nephrotoxicity by recruiting inflammatory cells via its receptor RAGE.

a, b Western blot analysis (a) and densitometric quantification (b) of RAGE protein levels from cisplatin-treated WT and Gsdme-KO mice (n = 12 biological replicates). c RT-qPCR analysis of Ager, Tnf, Il6, Cd86, and Mrc1 mRNA expression in BMDMs treated with or without recombinant S100A7A protein (n = 4 biological replicates). d Representative transwell images showing migration of BMDMs treated with the indicated concentrations of recombinant S100A7A protein. e Quantitative analysis of BMDMs migration from experiments in panel (d) (n = 4 biological replicates). f Representative transwell images showing BMDM migration in response to culture supernatants collected from PTCs of different genotypes with various treatments (cisplatin, S100a7a siRNA, and FPS-ZM1). g Quantitative analysis of BMDMs migration from experiments in panel (f) (n = 4 biological replicates). h Serum creatinine levels in cisplatin-treated WT and Gsdme-KO mice, with or without FPS-ZM1 pretreatment (n = 8 biological replicates). i, j Representative images of H&E staining (i) and histopathology scoring (j) of renal sections in cisplatin-treated WT and Gsdme-KO mice, with or without FPS-ZM1 pretreatment (n = 8 biological replicates). k RT-qPCR analysis of Tnf and Il6 mRNA expression in renal tissues from cisplatin-treated WT and Gsdme-KO mice, with or without FPS-ZM1 pretreatment (n = 8 biological replicates). l–o Representative immunohistochemistry staining of F4/80 (l) and MPO (m) in renal sections from cisplatin-treated WT and Gsdme-KO mice, with or without FPS-ZM1 pretreatment. Quantitative analysis of F4/80 (n) and MPO positive cells (o) (n = 8 biological replicates). Data were expressed as means ± SEM. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 versus cisplatin-treated WT mice. ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 versus cisplatin-treated Gsdme-KO mice pretreated with FPS-ZM1. ^{& &}P < 0.01, ^{& & &}P < 0.001 versus BMDMs ctrl group. P-values were determined by Student’s two-tailed t-test in (c) and one-way ANOVA (with LSD or Games-Howell post hoc tests) in (b, e, g, h, k, n, o). P values were determined by a two-tailed nonparametric test (Wilcoxon test) in (j). All scale bars, 100 μm. Source data are provided as a Source Data file.

Fig. 7. Tubule-specific knockdown of Gsdme aggravates cisplatin nephrotoxicity.

a Schematic illustration of Ksp-Cre-driven Gsdme knockdown (KD) virus administration in Ksp-Cre mice. The illustration elements were created in BioRender. Chen, Q. (2025) https://BioRender.com/k19218v. b Representative immunofluorescence images of GSDME (red) and LTL (green) in renal sections from Gsdme^Ksp-KD and negative control (NC) mice on day 3 after cisplatin treatment. LTL was used to stain renal proximal tubules. DAPI was used to stain nuclei (blue). c Quantification of GSDME immunofluorescence intensity from experiments in panel (b) (n = 8 biological replicates). d Serum creatinine levels in Gsdme^Ksp-KD and NC mice on day 3 after cisplatin treatment (n = 8 biological replicates). e, f Representative images of H&E staining (e) and corresponding renal tubular injury scoring (f) of renal sections from Gsdme^Ksp-KD and NC mice on day 3 after cisplatin treatment (n = 8 biological replicates). g, h Western blot analysis (g) and densitometric quantification (h) of NGAL, S100A7A, RAGE, and GSDME protein levels in renal tissues from Gsdme^Ksp-KD and NC mice on day 3 after cisplatin treatment (n = 8 biological replicates). i RT-qPCR analysis of Tnf, Ccl2, and Il6 mRNA expression in renal tissues from Gsdme^Ksp-KD and NC mice on day 3 after cisplatin treatment (n = 8 biological replicates). j–m Representative immunochemical staining of F4/80 (j) and MPO (l) in renal sections from Gsdme^Ksp-KD and NC mice on day 3 after cisplatin treatment. Quantitative analysis of F4/80-positive cells (k) and MPO-positive cells (m) (n = 8 biological replicates). Data were expressed as means ± SEM. ^**P < 0.01, ^***P < 0.001 versus Cre⁺ mice treated with cisplatin and NC siRNA. ^###P < 0.001 versus Gsdme^Ksp-KD mice. P-value was determined by Student’s two-tailed t-test in (c). P-values were determined by one-way ANOVA (with LSD or Games-Howell post hoc tests) in (d, h, i, k, m). P values were determined by a two-tailed nonparametric test (with Wilcoxon test) in (f). All scale bars, 100 μm. Source data are provided as a Source Data file.

Fig. 8. GSDME inhibits STAT3 phosphorylation and S100a7a transcription by stabilizing OGT protein.

a,b Western blot analysis (a) and densitometric quantification (b) of GSDME-FL, S100A7A, p-STAT3, and STAT3 in primary WT PTCs treated with increasing concentrations of cisplatin (0, 5, 10, 20 μM) for 24 h (n = 4 biological replicates). c,d Western blot analysis (c) and densitometric quantification (d) of p-STAT3 and STAT3 in renal tissues from cisplatin-treated WT and Gsdme-KO mice (n = 12 biological replicates). e,f Western blot analysis (e) and densitometric quantification (f) of p-STAT3, STAT3, and S100A7A in primary WT and Gsdme-KO PTCs treated with or without cisplatin and/or Stattic (n = 4 biological replicates). g Co-IP assay examining the O-GlcNAcylation of STAT3 in cisplatin-treated primary WT and Gsdme-KO PTCs (n = 4 biological replicates). h Densitometric quantification of O-GlcNAcylation of STAT3, OGT, and OGA in panel (g) (n = 4 biological replicates). i RT-qPCR analysis of S100a7a mRNA expression in primary WT and Gsdme-KO PTCs treated with or without cisplatin and/or Thiamet G (n = 4 biological replicates). j Co-IP assay examining the O-GlcNAcylation of STAT3 in primary WT and Gsdme-KO PTCs treated with or without cisplatin and/or Thiamet G (n = 4 biological replicates). k Densitometric quantification of O-GlcNAcylation of STAT3, p-STAT3, and S100A7A protein levels in panel (i) (n = 4 biological replicates). l,m Western blot analysis (l) and densitometric quantification (m) of OGT in primary WT and Gsdme-KO PTCs treated with CHX for the indicated time (n = 4 biological replicates). n, o Co-IP assay examining the ubiquitination of OGT in primary WT and Gsdme-KO PTCs (n = 4 biological replicates). Data were expressed as means ± SEM. ^#P < 0.05, ^##P < 0.01, ^###P < 0.001 versus KO cisplatin group. ^*P < 0.05, ^**P < 0.01, ^***P < 0.001 versus KO ctrl group. ^&P < 0.05, ^{& &}P < 0.01, ^{& & &}P < 0.001 versus WT ctrl group. P-values were determined by Student’s two-tailed t-test in (o) and one-way ANOVA (with LSD or Games-Howell post hoc tests) in (b, d, f, h, i, k). Source data are provided as a Source Data file.

Fig. 9. GSDME interacts with WDR26 and inhibits CUL4B-DDB1-WDR26 complex mediated OGT ubiquitination and degradation.

a Co-IP assay examining the interaction between endogenous WDR26 and FLAG-GSDME in HK-2 cells. b Co-IP assay examining the interactions among endogenous GSDME, OGT, CUL4B, DBB1, and HA-WDR26 in HK-2 cells. c Co-IP assay examining the interaction between endogenous WDR26 and GSDME in primary WT PTCs treated with or without cisplatin. d Densitometric quantification of GSDME protein levels from experiments in panel (c) (n = 4 biological replicates). e Co-IP assay examining the interaction between endogenous OGT and WDR26 in primary WT and Gsdme-KO PTCs with or without cisplatin treatment. f Densitometric quantification of WDR26 protein levels from experiments in panel (e) (n = 4 biological replicates). g GST pull-down assay examining the direct interaction between GST-WDR26 (1-274) and His-GSDME. h Whole-cell lysates from HEK-293T cells transfected with HA-WDR26 were incubated with purified His-tagged proteins. His pull-down assay was utilized to examine the direct interaction between HA-WDR26 and His-GSDME-N. i In vitro ubiquitination assay: Recombinant FLAG-OGT protein was incubated with recombinant CUL4B, DDB1 protein, ubiquitin, E1 (UBA1), E2 (UBCH5C/UBE2D3), Mg²⁺ and ATP, with or without GST-WDR26, His-GSDME, His-GSDME-N. Samples were analyzed by Western blotting. j Schematic illustration of the proposed role of GSDME in PTCs and its mechanism in promoting the pro-inflammatory response during cisplatin-induced nephrotoxicity. The illustration elements were created in BioRender. Chen, Q. (2025) https://BioRender.com/k19218v. The results in (a, b, g, h, i) are representative of three independent experiments. Data were expressed as means ± SEM. ^***P < 0.001 versus WT Ctrl group. ^###P < 0.001 versus WT Cisplatin group. P-values were determined by Student’s two-tailed t-test in (d) and one-way ANOVA (with LSD post hoc test) in (f). Source data are provided as a Source Data file.