Methionine metabolite spermidine inhibits tumor pyroptosis by enhancing MYO6-mediated endocytosis

Abstract

The connection between amino acid metabolism and pyroptosis remains elusive. Herein, we screen the effect of individual amino acid on pyroptosis and identify that methionine inhibits GSDME-mediated pyroptosis. Mechanistic analyses unveil that MYO6, a unique actin-based motor protein, bridges the GSDME N-terminus (GSDME-NT) and the endocytic adaptor AP2, mediating endolysosomal degradation of GSDME-NT. This degradation is increased by the methionine-derived metabolite spermidine noncanonically by direct binding to MYO6, which enhances MYO6 selectivity for GSDME-NT. Moreover, combination targeted therapies using dietary or pharmacological inhibition in methionine-to-spermidine metabolism in the tumor promotes pyroptosis and anti-tumor immunity, leading to a stronger tumor-suppressive effect in in vivo models. Clinically, higher levels of tumor spermidine and expression of methionine-to-spermidine metabolism-related gene signature predict poorer survival. Conclusively, our research identifies an unrecognized mechanism of pyroptotic resistance mediated by methionine-spermidine metabolic axis, providing a fresh angle for cancer treatment.

Fig. 1. Methionine restriction promotes GSDME-mediated pyroptosis.

(A) Schematic depicting the workflow for model-building and the screening method. (B and C) Effects of individual amino acid deprivation on Dox-induced PI uptake (B) and LDH release (C) in GSDME-NT^Tet-On HONE1 and HCT116. EAA, essential amino acid; NEAA, non-essential amino acid. (D) Immunoblot of cleaved caspase-3 in Dox-treated GSDME-NT^Tet-On HONE1 and HCT116 cultured in normal or Met-omitted medium with or without ZDEVD-FMK (30 μM) treatment. (E and F) Effects of zDEVD-FMK on PI uptake (E) and LDH release (F) in Dox-treated GSDME-NT^Tet-On HONE1 and HCT116 cultured with normal or Met-omitted medium. All p-values comparing the uptake of PI and the release of LDH in cells subjected to Met deficiency, in response to Dox, with those in cells subjected to deficiency of other amino acids, are less than 0.0001. (G) Effects of MRD on serum Met level of BALB/c-nu/nu mice (n = 6). (H) Immunofluorescence of HMGB1 in GSDME-sh^Tet-On HONE1 tumors of indicated groups. Scale bar: 100μm. (I) Quantification of nuclear HMGB1^- cells shown in (H) (n = 6). Data are represented as mean ± SD. Two-way ANOVA with Bonferroni’s multiple comparisons test (B, C, E, F); Two-tailed Student’s t test (G); One-way ANOVA with Tukey’s multiple comparisons test (I). NS, not significant. The results are representative of three independent experiments (B–F). Source data are provided as a Source Data file.

Fig. 2. Met- derived metabolite SPD mediates pyroptotic resistance.

(A) Methionine metabolism schematic. (B) Heatmap of changed metabolites in HONE1 with or without Met deprivation (n  =  4). (C and D) Effects of addition of SAH (10 μM), SPD (50 μM), SPM (50 μM) on Dox-induced PI uptake (C) and LDH release (D) in GSDME-NT^Tet-On HONE1 and HCT116 starved of Met. (E and F) Effects of DFMO (500 μM) on Dox-induced PI uptake (E) and LDH release (F) in GSDME-NT^Tet-On HONE1 and HCT116. (G and H) Effects of SRM or AMD1 knockdown on Dox-induced PI uptake (G) and LDH (H) release in GSDME-NT^Tet-On HONE1 and HCT116. Data are represented as mean ± SD. Two-way ANOVA with Bonferroni’s multiple comparisons test (C–H). NS, not significant. The results are representative of three independent experiments (C–H). Source data are provided as a Source Data file.

Fig. 3. SPD promotes endolysosomal degradation of GSDME-NT.

(A) Immunoblot of GFP-GSDME-NT in Dox-treated GSDME-NT^Tet-On HONE1 and HCT116 in response to Met deprivation with or without SPD supplement. (B) Transcriptional level of GFP-GSDME-NT detected by qPCR in Dox-treated GSDME-NT^Tet-On HONE1 and HCT116 in response to Met deprivation with or without SPD supplement. (C) Immunoblot of GFP-GSDME-NT in Dox-treated GSDME-NT^Tet-On HONE1 and HCT116 in response to CHX treatment (100 μg/ml) for the indicated time points. (D) Quantification of GFP-GSDME-NT level shown in (C). (E) Immunoblot of GSDME-NT oligomer and monomer in Dox-treated GSDME-NT^Tet-On HONE1 in response to Met deprivation with or without SPD supplement. (F) Immunoblot of GFP-GSDME-NT in Dox-treated GSDME-NT^Tet-On HONE1 and HCT116 in response to Dyngo4a (30 μM), CQ (50 μM), Pitstop-2 (30 μM) or MG-132 (10 μM) with or without Met deprivation. (G and H) Effects of Dyngo-4a, CQ and Pitstop-2 on Dox-induced PI uptake (G) and LDH release (H) in GSDME-NT^Tet-On HONE1 and HCT116. (I) Immunofluorescence of GFP-GSDME-NT and EEA1 in Dox-treated GSDME-NT^Tet-On HONE1 in response to Met deprivation with or without SPD supplement. Scale bar: 10μm. (J) Quantification of colocalization of GFP-GSDME-NT with EEA1 (n = 30 cells) shown in (I). (K) Immunofluorescence of GFP-GSDME-NT and lysotracker in Dox-treated GSDME-NT^Tet-On HONE1 in response to Met deprivation with or without SPD supplement. Scale bar: 10μm. (L) Quantification of colocalization of GFP-GSDME-NT with lysotracker (n = 30 cells) shown in (K). (M) Immunoblot of GFP-GSDME-NT of isolated endosomes and plasma membrane in Dox-treated GSDME-NT^Tet-On HONE1 in response to Met deprivation with or without SPD supplement. WCL, whole cell lysates; PM, plasma membrane. Data are represented as mean ± SD. One-way ANOVA with Tukey’s multiple comparisons test (B, D, J, L); Two-way ANOVA with Bonferroni’s multiple comparisons test (G and H). NS, not significant. The results are representative of three independent experiments (A–M). Source data are provided as a Source Data file.

Fig. 4. SPD directly interacts with MYO6 to regulate endocytosis of GSDME-NT.

(A) Schematic depicting the production of SPD FG bead. (B) Venn diagram displaying the intersection among SPD-binding proteins and GSDME-NT-binding proteins (cells fed with CM or Met-omitted medium) in Dox-treated GSDME-NT^Tet-On HONE1. (C) Immunoblot of GFP-GSDME-NT in Dox-treated GSDME-NT^Tet-On HONE1 and HCT116 with MYO6 knockdown. (D and E) Effects of MYO6 knockdown on Dox-induced PI uptake (D) and LDH release (E) in GSDME-NT^Tet-On HONE1 and HCT116. (F) Immunofluorescence of GFP-GSDME-NT and EEA1 in Dox-treated GSDME-NT^Tet-On HONE1 with MYO6 knockdown. Scale bar: 10μm. (G) Quantification of colocalization of GFP-GSDME-NT with EEA1 (n = 30 cells) shown in (F). (H) Immunoblot of GFP-GSDME-NT of isolated endosomes and plasma membrane in Dox-treated GSDME-NT^Tet-On HONE1 with MYO6 knockdown. WCL, whole cell lysates; PM, plasma membrane. (I) High concentration of SPD (2 mM) addition to lysates from Dox-treated GSDME-NT^Tet-On HONE1 inhibited capture of MYO6 proteins by SPD-conjugated beads. (J) CETSA displaying an increase thermal stability of MYO6 by SPD addition. (K) In situ Duolink-PLA assay displaying the molecular interaction between SPD and MYO6 in Dox-treated GSDME-NT^Tet-On HONE1. Scale bar: 10μm. (L) Quantification of PLA signal per cell (n = 30 cells) shown in (K). (M) BLI analysis of the molecular binding between SPD and MYO6. Data are represented as mean ± SD. Two-way ANOVA with Bonferroni’s multiple comparisons test (D and E); Two-tailed Student’s t test (G and L). NS, not significant. The results are representative of three independent experiments (C–L). Source data are provided as a Source Data file.

Fig. 5. MYO6 bridges GSDME-NT and the endocytic adaptor AP2.

(A) Co-IP with an anti-HA antibody using lysates from Dox-treated GSDME-NT^Tet-On HONE1 undergoing Met deprivation with or without SPD treatment. (B) Schematic of MYO6 and its truncation mutant structure. (C) Co-IP with an anti-mCherry antibody using lysates from Dox-treated GSDME-NT^Tet-On HONE1 with transfection of MYO6 or its truncation mutants. (D) The tendency of the root mean square deviation (RMSD) plot for two indicated complexes. (E) The RMSF plot for MYO6 in two indicated complexes. (F and G) The simulation of the interaction and surface binding model between MYO6 and GSDME-NT in the absence (F) or presence (G) of SPD. (H) Co-IP with an anti-FLAG antibody using lysates from Dox-treated GSDME-NT^Tet-On HONE1 with transfection of FLAG-MYO6 and/or MYC-AP2M1 undergoing Met deprivation. (I) Co-IP with an anti-HA antibody using lysates from Dox-treated GSDME-NT^Tet-On HONE1 with transfection of MYC-AP2M1 and MYO6 knockdown. The results are representative of three independent experiments (A, C, H and I). Source data are provided as a Source Data file.

Fig. 6. High tumor SPD level and expression of MEPO signature predicts a poor prognosis.

(A) SPD level is evaluated by IHC staining in 220 NPC tumor tissues. Scale bar: 100μm. (B) Association between SPD level and death status in a cohort of 220 NPC samples (two-sided χ² test). (C) Kaplan–Meier analysis of overall survival based on the SPD level (log-rank test). (D) The correlation between MEPO signature expression and activated DC signature expression in head and neck squamous carcinoma (HNSC), lung adenocarcinoma (LUAD) and esophageal carcinoma (ESCA) (The correlation coefficient and two-tailed p value were calculated using Pearson’s correlation analysis). (E) The correlation between MEPO signature expression and effector T cell signature expression in HNSC, LUAD and ESCA (Pearson’s correlation analysis). (F) Kaplan–Meier analysis of overall survival based on the SPD level in HNSC, LUAD and ESCA (log-rank test). Source data are provided as a Source Data file.

Fig. 7. Met restriction therapy boosts cetuximab efficacy via enhancing anti-tumor immunity.

(A) Effects of Met deprivation on ADCC in HONE1 and HCT116. (B) Effects of Met deprivation on HMGB1 release by ELISA in HONE1 and HCT116. (C) Immunoblot of GSDME cleavage in HONE1 and HCT116 under different conditions indicated. (D) HONE1 tumor growth in BALB/c-nu/nu mice. Mice were fed with normal diet or Met-restricted diet (n = 6). Drug regimen is described in methods. (E) HONE1 tumor growth in BALB/c-Nu/Nu mice (n = 6). Drug regimen is described in methods. (F) CT26-hEGFR tumor growth in BALB/c mice (n = 6). Mice were fed with normal diet or Met-restricted diet. Drug regimen is described in methods. (G) Immunofluorescence of HMGB1 in CT26-hEGFR tumors of mice receiving indicated treatments. Scale bar: 100μm. (H) Quantification of nuclear HMGB1-negative cells shown in (G) (n = 6). (I) Activated DC cell infiltration in CT26-hEGFR tumor of mice receiving indicated treatments (n = 5). (J) CD8⁺ T cell tumor infiltration in CT26-hEGFR tumor of mice receiving indicated treatments (n = 5). (K and L) Percentage of IFN-γ⁺ (K) and TNF-α⁺ (L) CD8⁺ T cell in CT26-hEGFR tumors of mice receiving indicated treatments (n = 5). (M) The signaling transduction pathway related to Met-mediated pyroptotic resistance, clinical relevance and potential applications are illustrated. This figure was created in BioRender. Wu, J. (2025) https://BioRender.com/x24v758.” Data are represented as mean ± SD. One-way ANOVA with Tukey’s multiple comparisons test (A, B, D, E, F). One-way ANOVA with Bonferroni’s multiple comparisons test (H–L). NS, not significant. The results are representative of three independent experiments (A–C). Source data are provided as a Source Data file.