Fatty acid synthesis promotes mtDNA release via ETS1-mediated oligomerization of VDAC1 facilitating endothelial dysfunction in sepsis-induced lung injury

Abstract

Sepsis involves endothelial cell dysfunction leading to the development of lung injury. Fatty acid synthesis contributes to the development of inflammatory injury in sepsis. However, the regulatory mechanisms of fatty acid synthesis-related endothelial activation remain unclear. In this study, we found that fatty acid synthesis in patients with sepsis was greatly disordered. Inhibition of fatty acid synthesis significantly alleviated sepsis-induced endothelial damage and lung injury both in vitro and in vivo. We further found that the release of mtDNA participated in fatty acid synthesis-related regulation of endothelial inflammatory and coagulation activation. Mechanistically, fatty acid synthesis promoted the oligomerization of voltage-dependent anion channel 1 (VDAC1) via ETS proto-oncogene 1 (ETS1)-mediated inhibition of VDAC1 ubiquitination, thereby leading to the increased release of mtDNA and subsequent activation of cGAS-STING signaling and pyroptosis in endothelial cells. Our findings revealed that fatty acid synthesis promoted endothelial dysfunction through mtDNA release, providing new insight into the therapeutic strategies for treating sepsis-associated lung injury.

Fig. 1. The dysregulation of fatty acid metabolism in septic patients.

A Partial least squares discriminant analysis (PLS-DA) showing the separation of septic patients from healthy controls. B Pie charts of Superclass differential metabolites. C, D Heatmap and KEGG enrichment of metabolic analysis data in septic patients. E–G The levels of FASN, sVCAM-1, and sE-selectin were measured by ELISA in healthy controls (n = 10) and septic patients (n = 18). H, I The correlation analysis of FASN with sVCAM-1 and sE-selectin. All data were expressed as the mean ± SD. Student’s t test was used for (E–G). Simple linear regression and Spearman correlation coefficients were used for (H, I).

Fig. 2. Inhibition of fatty acid synthesis ameliorated endotoxemia-induced lung injury and EC activation in vivo.

The C57BL/6N mice were i.p. injected with LPS (10 mg/kg) with or without C75 (10 mg/kg, i.p.) pretreatment for 1 h (n = 5). A–C The levels of TNF-α, IL-6, and TF in mice plasma were measured by ELISA (n = 5). D H&E staining showing the tissue injury in the lung (n = 5). E, F IHC staining showing the levels and distribution of VCAM-1 and E-selectin in the lung tissues (n = 5). G, H Immunofluorescent staining showing the levels of fibrinogen (anti-fibrinogen, Red) in lungs from septic mice pretreated with or without C75 (n = 5). CD31 (anti-CD31, Green) was used to label endothelial cells and nuclei were stained in blue with DAPI. I, J Immunoblot showing the expression levels of VCAM-1, E-selectin, PAI-1, and TF in the lung tissues in mice (n = 3). All data were expressed as the mean ± SD. Comparison among three or more groups was analyzed by one-way ANOVA. ns no significance, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig. 3. Inhibition of fatty acid synthesis ameliorated LPS-induced EC dysfunction in vitro.

A, B HUVEC were pretreated with FASN inhibitor C75 for 30 min before LPS stimuli for 4 h. Protein levels of VCAM-1, E-selectin, PAI-1 and TF were analyzed by western blot (n = 3). C–I The mRNA levels of E-selectin, VCAM-1, ICAM-1, TF, PAI-1, MCP-1 and IL-8 with C75 treatment were detected by RT-qPCR (n = 3). J, K HUVEC were transfected with siRNA specific to FASN before LPS stimuli for 4 h. Protein levels of FASN, VCAM-1, E-selectin, PAI-1 and TF were analyzed by western blot (n = 3). L, M Immunoblot showing the expression levels of VCAM-1, E-selectin, PAI-1, and TF with PA (10 µM) or PAS (20 µM) (n = 3). All data were expressed as the mean ± SD. Comparison among three or more groups was analyzed by one-way ANOVA. ns no significance, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig. 4. Cytosolic mtDNA release mediated fatty acid synthesis-related EC activation in sepsis.

A KEGG analysis of genes regulated by both LPS stimulation and C75 treatment. B The levels of mtDNA in the plasma of septic patients (n = 18) were measured by RT-qPCR. C–F The correlation analysis of mtDNA with APACHE II score, SOFA score, sVCAM-1, and sE-selectin. G HUVEC were pretreated with C75 for 30 min before LPS stimuli for 4 h. Cytosolic mtDNA/nDNA was detected by RT-qPCR (n = 5). H, I HUVEC were pretreated with C75 before LPS stimuli. Confocal laser microscope assay was used to assess the levels of mtDNA (anti-dsDNA, green). Mitochondria were stained in red with Mito-Tracker and nuclei were stained in blue with DAPI. Quantitative results from 8 cells for each condition were reported. J The levels of mtDNA in the plasma of septic mice (n = 5) were measured by RT-qPCR. All data were expressed as the mean ± SD. Simple linear regression and Spearman correlation coefficients were used for (C–F). Unpaired t-test was used for the comparison between two groups. Comparison among three or more groups was analyzed by one-way ANOVA. ns no significance, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig. 5. Fatty acid synthesis promoted mtDNA-related activation of cGAS-STING signaling and pyroptosis in ECs in response to LPS.

A Heatmap of DEGs enriched in cytosolic DNA-sensing pathway. B–E HUVEC were pretreated with C75 before LPS stimuli. Protein levels of cGAS, p-STING, and p-p65 were determined by western blot (n = 3). F, G Representative immunoblots of cGAS, p-STING, and p-p65 in the lung tissues of septic mice pretreated with or without C75 (n = 3). H, I Confocal images of p65 nuclear translocation in LPS-challenged HUVEC with or without C75 pretreatment. Quantitative results from 8 cells per group are reported. All data were expressed as the mean ± SD. Comparison among three or more groups was analyzed by one-way ANOVA. ns no significance, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig. 6. VDAC1 was associated with fatty acid synthesis-related mtDNA release and EC dysfunction during sepsis.

A, B HUVEC were pretreated with C75 before LPS stimuli. The levels of VDAC1 oligomerization were detected by western blot (n = 3). C, D VDAC1 were knocked down with siRNA before LPS stimuli. The levels of VDAC1 oligomerization were detected by western blot (n = 3). E, F Immunoblot showing the expression levels of VCAM-1, E-selectin, PAI-1 and TF in HUVEC pretreated with VBIT-4 (40 µM) before LPS stimuli (n = 3). G, H HUVEC were stimulated with LPS in the presence or absence of VBIT-4 (10 µM, 40 µM). The protein levels of VDAC1 oligomerization were analyzed by western blot (n = 3). I VDAC1 in HUVEC was knocked down with siRNA before LPS stimuli for 4 h. Cytosolic mtDNA/nDNA was detected by RT-qPCR (n = 5). J, K VDAC1 was knocked down before LPS stimuli. Confocal laser microscope assay was used to assess the levels of mtDNA (anti-dsDNA, green). Mitochondria were stained in red with Mito-Tracker and nuclei were stained in blue with DAPI. Quantitative results from 8 cells for each condition were reported. L HUVEC were stimulated with LPS in the presence or absence of VBIT-4 (40 µM). Cytosolic mtDNA/nDNA was detected by RT-qPCR (n = 5). M, N HUVEC were pretreated with VBIT-4 (40 µM) before LPS stimuli. Confocal laser microscope assay was used to assess the levels of mtDNA (anti-dsDNA, green). Mitochondria were stained in red with Mito-Tracker and nuclei were stained in blue with DAPI. Quantitative results from 8 cells for each condition were reported. All data were expressed as the mean ± SD. Comparison among three or more groups was analyzed by one-way ANOVA. ns no significance, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig. 7. ETS1 participated in fatty acid synthesis-related mtDNA release and EC damage induced by LPS.

A Volcano plot of differently expressed genes. B, C HUVEC were transfected with siETS1 before LPS stimuli for 4 h. Expression levels of ETS1, VCAM-1, E-selectin, PAI-1, and TF were analyzed by western blot (n = 3). D ETS1 in HUVEC was knocked down with siRNA before LPS stimuli for 4 h. Cytosolic mtDNA/nDNA was detected by RT-qPCR (n = 5). E, F ETS1 in HUVEC was knocked down with siRNA before LPS stimuli. Confocal laser microscope assay was used to assess the levels of mtDNA (anti-dsDNA, green). Mitochondria were stained in red with Mito-Tracker and nuclei were stained in blue with DAPI. Quantitative results from 8 cells for each condition were reported. G–J HUVEC were pretreated with siRNA specific to ETS1 before LPS stimuli. Protein levels of cGAS, p-STING, and p-p65 were determined by western blot (n = 3). K, L Confocal images of p65 nuclear translocation in LPS-challenged HUVEC with or without ETS1 knockdown. Quantitative results from 8 cells per group are reported. All data were expressed as the mean ± SD. Comparison among three or more groups was analyzed by one-way ANOVA. ns, no significance, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig. 8. ETS1 promoted VDAC1 oligomerization to facilitate mtDNA release and EC activation via limiting VDAC1 ubiquitination.

A CoIP-MS image of VDAC1. B Molecular docking showing the interaction between ETS1 and VDAC1. C Immunoblot showing the binding of ETS1 with VDAC1 by GST pull-down (n = 3). D, E HUVEC were transfected with siRNA specific to ETS1 before LPS stimuli. The levels of VDAC1 oligomerization were detected by western blot (n = 3). F KEGG analysis of proteins binding to ETS1. G, H HUVEC were transfected with siRNA specific to ETS1 or pretreated with C75 before LPS stimuli. The ubiquitination levels of VDAC1 were detected by western blot (n = 3). All data were expressed as the mean ± SD. Comparison among three or more groups was analyzed by one-way ANOVA. ns no significance, *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig. 9. Mechanistic diagram.

In sepsis, overwhelmed synthesis of fatty acids promotes ETS1 expression which limited the ubiquitination of VDAC1 and promoted its oligomerization. VDAC1 oligomerization facilitates the release mtDNA into the cytosol to activate cGAS-STING signaling, leading to endothelial dysfunction in sepsis-induced lung injury.