STING aggravates ferroptosis-dependent myocardial ischemia-reperfusion injury by targeting GPX4 for autophagic degradation

Abstract

Despite advancements in interventional coronary reperfusion technologies following myocardial infarction, a notable portion of patients continue to experience elevated mortality rates as a result of myocardial ischemia-reperfusion (MI/R) injury. An in-depth understanding of the mechanisms underlying MI/R injury is crucial for devising strategies to minimize myocardial damage and enhance patient survival. Here, it is discovered that during MI/R, double-stranded DNA (dsDNA)-cyclic GMP-AMP synthase (cGAS)-stimulator of interferon genes (STING) signal accumulates, accompanied by high rates of myocardial ferroptosis. The specific deletion of cgas or Sting in cardiomyocytes, resulting in the inhibition of oxidative stress, has been shown to mitigate ferroptosis and I/R injury. Conversely, activation of STING exacerbates ferroptosis and I/R injury. Mechanistically, STING directly targets glutathione peroxidase 4 (GPX4) to facilitate its degradation through autophagy, by promoting the fusion of autophagosomes and lysosomes. This STING-GPX4 axis contributes to cardiomyocyte ferroptosis and forms a positive feedback circuit. Blocking the STING-GPX4 interaction through mutations in T267 of STING or N146 of GPX4 stabilizes GPX4. Therapeutically, AAV-mediated GPX4 administration alleviates ferroptosis induced by STING, resulting in enhanced cardiac functional recovery from MI/R injury. Additionally, the inhibition of STING by H-151 stabilizes GPX4 to reverse GPX4-induced ferroptosis and alleviate MI/R injury. Collectively, a novel autophagy-dependent ferroptosis mechanism is identified in this study. Specifically, STING autophagy induced by anoxia or ischemia-reperfusion leads to GPX4 degradation, thereby presenting a promising therapeutic target for heart diseases associated with I/R.

Fig. 1

I/R triggers cGAS-STING upregulation in CMs. Male mice (aged 8 weeks) were subjected to the operation and euthanized at 24 h post-I/R or Sham. a Display the mouse I/R model operation flowchart. b Schematic diagram showing the mouse I/R injury model. c Double immunofluorescence analysis for detecting dsDNA and cGAS in the same heart section of the WT border region post I/R or Sham. The positive reactions of tissue sections are displayed in green (dsDNA) and red (cGAS). The positive reaction for co-localisation is displayed in yellow. Scale bar = 20 μm. d, e Western blot and quantification of cGAS and STING in WT cells isolated from border region post I/R or Sham (n = 6). f, g Western blot and quantification of cGAS and STING in various cell types isolated from the heart border region of cgas^−/−, Sting^−/−, or WT mice post I/R or Sham (n = 6). h Double immunofluorescence analysis for detecting markers of CMs and STING in the same heart section of the WT border region post I/R or Sham. The positive reactions of tissue sections are displayed in green (STING) and red (Troponin). The positive reaction for co-localisation is displayed in yellow. Scale bar = 20 μm. i Schematic diagram showing the CM A/R operation flowchart. j Double immunofluorescence analysis for detecting dsDNA and mitochondria (labeled with Mito-Tracker) in MPCs post-A/R. The positive reactions are displayed in green (dsDNA) and red (Mito-Tracker). The positive reaction for co-localisation is displayed in yellow. Arrowheads indicate released dsDNA. Scale bar = 10 μm. Data are expressed as the mean ± SEM. NS non-significant, ***P < 0.001 and ****P < 0.0001 (unpaired two-tailed Student’s t test). I/R Ischemia reperfusion, dsDNA double-stranded DNA, A/R anoxia/ reoxygenation, cGAS cyclic guanosine monophosphate-adenosine monophosphate synthase, STING stimulator of interferon genes, WT wild type, cgas^−/− cgas knockout mice, Sting^−/−, Sting knockout mice, CM cardiomyocyte

Fig. 2

Deletion of cGAS-STING protects against MI/R injury. a Structural pattern diagram of cgas or Sting CM-specific (Myh6-iCre) conditional knockout mice. b–e Effect of cgas-CKO on myocardial infarct size, FS, EF and fibrosis area following I/R or Sham: b Myocardial infarct size (% of AAR) with representative tissue sectioning (n = 6); c Echocardiography and measured EF%, and FS% (n = 7); d Masson staining and measured fibrosis area% (n = 7), scale bar=1 mm; e Western blot and quantification of cGAS in CMs isolated from the heart border region (n = 5). f–i Effect of Sting-CKO on myocardial infarct size, FS, EF and fibrosis area following I/R: f Myocardial infarct size (% of AAR) with representative tissue sectioning (n = 6); g Echocardiography and measured EF%, and FS% (n = 7); h Masson staining and measured fibrosis area% (n = 6), scale bar = 1 mm; i Western blot and quantification of STING in CMs isolated from the heart border region (n = 6). Mean ± SEM, **P < 0.01, ***P < 0.001, and ****P < 0.0001. AAR area at risk, IF infarct area, cgas-CKO cgas^fl/fl Myh^6iCre, Sting-CKO Sting^fl/fl Myh^6iCre, FS fraction shortening, EF ejection fraction, KO knockout

Fig. 3

STING amplifies myocardial ferroptosis through modulation of oxidative stress injury. a Heart region for RNA-seq post I/R. b KEGG Enrichment Bar Plot of RNA-seq. c Statistical chart of immunofluorescence image for detecting Tunel in the heart section of the border region of cgas-CKO or Sting-CKO mice and their control mice post-I/R (n = 6). The corresponding image is supplementary Fig. 2a, b. d GO Enrichment Scatter Plot of RNA-seq. e MDA analysis for detecting lipid peroxidation (n = 6). f Live cell immunofluorescence imaging analysis for detecting ROS in MPC. Scale bar = 10 μm. g, h Western blot and quantification of ACSL4, TFR, SLC7A11 and GPX4 in Sting^fl/fl or Sting-CKO CMs post-I/R (n = 6). i, j Western blot and quantification of ACSL4, TFR, SLC7A11 and GPX4 in Sting^fl/fl or Sting-CKO MPCs post-A/R or normoxia (n = 6). Mean ± SEM, *P < 0.05, ***P < 0.001, and ****P < 0.0001. ROS reactive oxygen species, Fer-1 ferrostatin-1, cGAMP cyclic guanosine monophosphate-adenosine monophosphate, KEGG Kyoto Encyclopedia of Genes and Genomes, GO Gene Ontology, LAD left anterior descending coronary artery, ACSL4 acyl-CoA synthetase long-chain family member 4, TFR transferrin receptor, SLC7A11 solute carrier family 7, member 11, GPX4 glutathione peroxidase 4, MDA malondialdehyde

Fig. 4

Targeting of STING by GPX4. a KEGG displayed ferroptosis pathway enrichment. b Silver staining and LC-MS/MS analysis for tandem affinity purification using STING protein affinity antibodies in MPCs post-A/R. c Images showing ten potential contact modes according to molecular-docking results between STING (PDB ID: 4F5W) and GPX4 (PDB ID: 5L71). d Co-IP assay of MPCs to examine whether endogenous GPX4 interacts with STING and this combination is influenced by short point-stimulation of cGAMP. IB: STING, IP: GPX4. e Co-IP assay of MPCs to examine whether endogenous STING interacts with GPX4 and this combination is influenced by short point-stimulation of cGAMP. IB: GPX4, IP: STING. f Co-IP assay of HeLa cells co transfected with Flag-GPX4 and Myc-STING to examine whether STING interacts with GPX4. IB: Myc, IP: Flag. g Co-IP assay of HeLa cells co transfected with Flag-GPX4 and Myc-STING to examine whether GPX4 interacts with STING. IB: Flag, IP: Myc. h Co-IP assay of HeLa cells co transfected with Flag-GPX4 and Myc-STING under Erastin or cGAMP delivering to examine whether GPX4 and STING interaction is influenced by cGAMP time-grant or Erastin. IB: Flag, IP: Myc. i Double immunofluorescence analysis for detecting STING and GPX4 and their co-localisation in HeLa cells co transfected with Myc-STING and Flag-GPX4 under cGAMP or Erastin addition. The positive reaction for the Myc-label is shown in red, and that for the Flag-label is shown in green. The positive reaction for co-localisation is displayed in yellow. Scale bar = 20 μm. j Double immunofluorescence analysis for detecting GPX4 and STING in the same heart section of the Sting^fl/fl or Sting-CKO border region. The positive reactions of tissue sections are displayed in green (GPX4) and red (STING). The positive reaction for co-localisation is displayed in yellow. Scale bar = 200 μm. LC liquid chromatography, MS mass spectrometry, PDB Protein Data Bank, Co-IP co-immunoprecipitation, IN input, IB immunoblotting, IP immunoprecipitation, Flag-GPX4 Flag-labeled GPX4 plasmid, Myc-STING Myc-labeled STING plasmid

Fig. 5

STING and GPX4 directly interact at the amino acid residues N146 of GPX4 and T267 of STING. a Representative images of molecular docking results to show the potential contact sites between GPX4 and STING. b Co-IP assay of HeLa cells co-transfected with Flag-GPX4 or Flag-GPX4 point mutation (Flag-GPX4[ΔG126], Flag-GPX4[ΔR127] or Flag-GPX4[ΔN146]) and Myc-STING, to examine the potential contact site between GPX4 and STING. IB: Myc, IP: Flag. c Co-IP assay of HeLa cells co-transfected with Flag-GPX4 and Myc-STING or Myc-STING point mutation (Myc-STING[ΔY167], Myc-STING[ΔE260], Myc-STING[ΔY245], Myc-STING[ΔQ266] or Myc-STING[ΔT267]) to examine the potential contact site between GPX4 and STING. IB: Myc, IP: Flag. d Double immunofluorescence analysis for detecting STING and GPX4 and their co-localisation in HeLa cells co-transfected with Myc-STING and Flag-GPX4, Flag-GPX4 and Myc-STING[ΔT267] or Myc-STING and Flag-GPX4[ΔN146]. The positive reaction for the Myc-label is shown in red, and that for the Flag-label is shown in green. The positive reaction for co-localisation is displayed in yellow. Scale bar = 20 μm. e Double immunofluorescence analysis for detecting STING and GPX4 and their co-localisation in HL-1 cells co-transfected with Myc-STING and Flag-GPX4 or Myc-STING and Flag-GPX4 [ΔN146] under cGAMP addition or not. The positive reaction for the Myc-label is shown in green, and that for the Flag-label is shown in red. The positive reaction for co-localisation is displayed in yellow. Scale bar = 5 μm. Scale bar of the enlarged image = 2.5 μm. G Gly, R Arg, N Asn, Y Tyr, E Glu, Q Gln, T Thr

Fig. 6

STING promotes ferroptosis via autophagy-lysosome-mediated degradation of GPX4. a Western blot and quantification of GPX4 expression in MPCs influenced by Fer-1 or cGAMP (n = 5). b Western blot and quantification of GPX4 degradation in MPCs induced by Erastin, cGAMP or H-151 (n = 5). c Western blot and quantification of GPX4 degradation in MPCs induced by cGAMP and its blocking using MG-132, NH₄Cl, chloroquine and calpeptin (n = 5). d Western blot and quantification of GPX4 degradation in MPCs induced by cGAMP and its blocking using Baf A-1, LY294002, 3-MA and Wortmannin (n = 5). e–g Immunofluorescence analysis for autophagic flow presented by mRFP-GFP-LC3 in Sting^fl/fl or Sting-CKO MPCs post-A/R or not. The positive reactions for autophagy are displayed in red dots. Scale bar = 10 μm. h Triple immunofluorescence analysis for detecting GPX4, STING, and LC3B in Sting^fl/fl or Sting-CKO MPCs. The positive reactions for co-localisation are displayed in yellow (LC3B and STING co-localisation) or pink (co-localisation of the three). Scale bar = 10 μm. i Triple immunofluorescence analysis for detecting GPX4, STING, and LAMP2B in MPCs post A/R or not. The positive reactions for co-localisation are displayed in yellow (GPX4 and STING co-localisation) or pink (co-localisation of the three). Scale bar = 10 μm. Mean ± SEM, NS non-significant, **P < 0.01, ***P < 0.001, ****P < 0.0001 and ### P < 0.001. Baf A-1 Bafilomycin A1, LAMP2B lysosomal-associated membrane protein 2B

Fig. 7

AAV-mediated GPX4 therapy shields cardiac function against the severe I/R injury triggered by STING activation. Effect of AAV-cTNT-GPX4 on heart dysfunction therapy in C57BL/6J mice with DMXAA or DMSO addition post-I/R: a Schematic diagram depicting the time course of I/R-induced cardiac dysfunction receiving AAV or DMXAA. b Western blot of verifying successful overexpression of GPX4 in CMs (n = 6). c, d Myocardial infarct size (% of AAR) with representative tissue sectioning (n = 6). e, f Echocardiography and measured EF% and FS% (n = 6). g Double immunofluorescence analysis for GPX4 and 4-HNE in the same heart section of the border region. Scale bar = 20 μm. h Immunofluorescence imaging analysis for detecting ROS in AAV-cTNT-GPX4 or AAV-Control infection mice CMs under DMSO or DMXAA addition following A/R. Scale bar = 10 μm. Mean ± SEM, NS, non-significant, **P < 0.01, ***P < 0.001, ****P < 0.0001, #P < 0.05 and ###P < 0.001. AAV adeno-associated virus, 4-HNE 4-Hydroxynonenal, cTNT-GPX4 AAV-Mus-GPX4-cTNT-C-GFP, DMXAA Vadimezan

Fig. 8

STING is a potential therapeutic target to alleviate cardiac dysfunction post-I/R. Effect of H-151 on heart dysfunction in C57BL/6J mice post-I/R: a i. Schematic diagram depicting the time course of I/R-induced cardiac dysfunction receiving H-151 or DMSO. ii. Representative photographic images of hearts with DMSO or H-151 addition post-I/R. b, c Myocardial infarct size (% of AAR) with representative tissue sectioning (n = 6). d, e Echocardiography and measured EF% and FS% (n = 6). f Double immunofluorescence analysis for GPX4 and 4-HNE in the same heart section of the border region. Scale bar = 50 μm. g Immunofluorescence analysis for GPX4 and STING in DMSO or H-151 addition mice CMs post A/R. Scale bar = 10 μm. h Immunofluorescence analysis for autophagic flux presented by mRFP-GFP-LC3 in DMSO or H-151 addition mice CMs post A/R. The positive reactions for autophagy are displayed in red dots. Scale bar = 20 μm. i Immunofluorescence imaging analysis for detecting ROS in DMSO or H-151 addition mice CMs post A/R. Scale bar = 20 μm. j Schematic diagram showing the mechanism of STING-promoted MI/R injury. Part of the image was drawn by Figdraw. Mean ± SEM, NS non-significant, **P < 0.01, ***P < 0.001, and ****P < 0.0001