Neutrophil extracellular traps mediate m6A modification and regulates sepsis-associated acute lung injury by activating ferroptosis in alveolar epithelial cells

Abstract

Neutrophil extracellular traps (NETs) production is a major strategy employed by polymorphonuclear neutrophils (PMNs) to fight against microbes. NETs have been implicated in the pathogenesis of various lung injuries, although few studies have explored NETs in sepsis-associated acute lung injury (SI-ALI). Here, we demonstrate a major contribution of NETs to the pathology of sepsis-associated ALI by inducing ferroptosis of alveolar epithelial cells. Using both in vitro and in vivo studies, our findings show enhanced NETs accumulation in sepsis-associated ALI patients and mice, as well as the closely related upregulation of ferroptosis, the induction of which depends on METTL3-induced m6A modification of GPX4. Using a CLP-induced sepsis-associated ALI mouse model established with METTL3^-/- versus WT mice, in addition to METTL3 knockout and overexpression in vitro, we elucidated and confirmed the critical role of ferroptosis in NETs-induced ALI. These findings support a role for NETs-induced METTL3 modification and the subsequent induction of ferroptosis in the pathogenesis of sepsis-associated ALI.

Keywords: N6-Methylation; Neutrophil extracellular traps; ferroptosis; sepsis-associated acute lung injury.

Figure 1

Enhanced NETs release in sepsis-associated ALI patients and mouse models. (A) Serum cell-free DNA and (B) MPO-DNA complexes were detected in healthy controls, sepsis patients, and patients with sepsis-induced ARDS. HC (n=20), Sepsis (n=24), Sepsis ARDS (n=22). (C) A positive correlation was found between serum cell-free DNA and (D) MPO-DNA complexes concerning the IL-8 level in healthy controls and patients with sepsis ARDS. HC, Sepsis ARDS; dot in box means Sepsis ARDS patients; (E, F) NETosis representative pictures in the HC group, Sepsis group and Sepsis ARDS group (red: MPO, green: CitH3, blue: DAPI). (G) Paraffin-embedded lung tissue samples were stained for H&E. Representative images of H&E staining are shown at 200× magnification showing lung injury in the HC group, sham group and sepsis ALI group in mice (n=6). (H, I) Cell-free DNA in plasma and BAL in mice of the HC group, sham group and sepsis group (n=6 in each group, BAL: bronchoalveolar lavage). *P<0.05, **P<0.01 (two-way ANOVA with Tukey's correction).

Figure 2

NETs formation inhibition, NETs depletion or NETs degradation protects mice against sepsis-associated ALI. A mouse model of cecal ligation and puncture (CLP)-induced sepsis-associated ARDS was established and then treated with saline, CI-amidine (inhibition of NETs formation), anti-ly6G (neutrophil depletion) or DNase (NETs degradation). (A) Paraffin-embedded mouse lung tissue samples were stained with H&E. Representative histological images are shown at 400× magnification. (B) Pulmonary edema was evaluated by determining the wet/dry weight ratio. (C) The cells in extracted bronchoalveolar lavage fluid (BALF) of mice were analyzed using a cell counter. (D) Representative images of immunofluorescence staining of NETs in lung tissues from HC, sham, and sepsis ARDS mouse models in the presence or absence of saline, CI-amidine, anti-ly6G and DNase (red: NE, green: CitH3, blue: DAPI). Scale bar=50 µm. (E) The cfDNA levels in plasma and (F) BALF were detected in the mouse model. (G) The levels of TNFα, IL-1α, IL-8 and TGF-β in plasma were determined by ELISA in a mouse model (n=6 in each group). *P<0.05. *indicates Sepsis-associated lung injury group versus HC group. # indicates Sepsis-associated lung injury group versus Sham group (Two-way ANOVA with Tukey's correction).

Figure 3

NETs formation inhibition, NETs depletion and NETs degradation attenuate sepsis-induced ferroptosis in an ALI murine model. A mouse model of CLP-induced sepsis-associated ALI was established and then treated with saline, CI-amidine (inhibition of NETs formation), anti-ly6G (neutrophil depletion) or DNase (NETs degradation). (A) The ROS level was analyzed using the DCF-DA assay in mouse lung tissue. (B) The ferritin level was evaluated by ELISA in mouse lung tissue. (C) The ferrous iron (Fe²⁺) level was evaluated using an iron assay kit in mouse lung tissue. (D) The MDA level was evaluated using a lipid peroxidation assay kit in mouse lung tissue. (E) The GSH level was evaluated by ELISA in mouse lung tissue. (F) The mRNA level of GPX4 was analyzed by qRT-PCR in mouse lung tissue. (G) The protein level of GPX4 was confirmed by IHC in mouse lung tissue (n=6 in each group). *P<0.05, **P<0.01. *indicates Sepsis-associated ALI group versus HC group. # indicates Sepsis-associated ALI group versus Sham group (Two-way ANOVA with Tukey's correction).

Figure 4

NETs impair the viability of alveolar epithelial cells by inducing ferroptosis. (A) Representative images of control and NETs-treated HPAEpiC cells with or without the ferroptosis inhibitor ferrostatin-1. (B) Cell viability was evaluated using the CCK-8 assay. (C) The ferrous iron (Fe²⁺) level was evaluated using an iron assay kit. (D) The MDA level was evaluated using a lipid peroxidation assay kit. (E) The GSH level was evaluated by ELISA. (F) The ROS level was analyzed using the DCF-DA assay. (G) The mRNA level of GPX4 was analyzed by qRT-PCR. (H) The protein level of GPX4 was analyzed by western blotting. *P<0.05, **P<0.01, #P<0.05. *indicates NETs+Fer-1 versus control+PBS. #indicates NETs+Fer-1 versus NETs +PBS (Two-way ANOVA with Tukey's correction).

Figure 5

NETs activate METTL3-mediated m6A modification in alveolar epithelial cells. RNA-Seq identified METTL3 upregulation in NETs-treated alveolar epithelial cells. (B) GO and KEGG analyses. (C) m6A modification-related genes were confirmed by qRT-PCR. (D) The upregulated METTL3 protein in NETs-treated alveolar epithelial cells was confirmed by IF (green: METTL3, blue: DAPI). (E) The total level of m6A-methylated RNA was analyzed using a colorimetric kit and (F) the dot blot assay. *P<0.05, **P<0.01. (Two-way ANOVA with Tukey's correction).

Figure 6

NETs promote METTL3 upregulation via activation of the TLR9 pathway. (A) qRT-PCR of TLRs at the mRNA level in alveolar epithelial cells treated with NETs or without NETs. (B) Cell viability was evaluated using the CCK-8 assay. (C) The ferrous iron (Fe²⁺) level was evaluated using an iron assay kit. (D) The MDA level was evaluated using a lipid peroxidation assay kit. (E) The GSH level was evaluated by ELISA. (F) The ROS level was analyzed using the DCF-DA assay. (G) Mev software of protein network predicting METTL3 upregulation via activation of the TLR9 pathway (http://string.embl.de/) (H) NETs significantly increased the expression of TLR9, METTL3, Myd88,p-p65 and METTL3 and the effects were abolished by adding DNAse I and TLR9 antagonist (2.5 µmol/L, ODNTTTGGG).

Figure 7

NETs induced ferroptosis in alveolar epithelial cells. (A) RNA-Seq+M6a-RIP-Seq identified GPX4 modification enrichment in alveolar epithelial cells. (B) METTL3 and GPX4 mRNA were analyzed by qRT-PCR. (C) METTL3 and GPX4 protein were analyzed by Western blotting. (D) m6A-methylated GPX4 mRNA was analyzed by Me-RIP-QPCR. (E) Curve and statistical analysis of GPX4 mRNA half-life after transfection with METTL3 siRNA, METTL3-WT, METTL3-Mut or the negative control after transcription inhibition (TI) are shown. *P<0.05 (two-way ANOVA with Tukey's correction).

Figure 8

METTL3 knockout attenuates NETs-induced ferroptosis in alveolar epithelial cells. Representative images of control and NETs-treated HPAEpiC cells. (B) HPAEpiC cell viability was evaluated using CCK-8 assay (n=3); (C) ferrous iron (Fe²⁺) levels were evaluated using an iron assay kit in HPAEpiC cells (n=3); (D) MDA levels were evaluated using a lipid peroxidation assay kit in HPAEpiC cells (n=3); (E) GSH levels were evaluated by ELISA in HPAEpiC cells (n=3); (F) ROS levels were analyzed using the DCF-DA assay in HPAEpiC cells (n=3); (G) The mRNA levels of GPX4 were analyzed by qRT-PCR in HPAEpiC cells (n=3). *P<0.05. (Two-way ANOVA with Tukey's correction).

Figure 9

METTL3 knockout protects mice against sepsis-associated ALI. (A) Paraffin-embedded mouse lung tissue samples were stained with H&E. Representative histological images were shown at 400× magnification. Scale bar=50 µm. (B) Pulmonary edema was evaluated by determining the wet/dry weight ratio in a mouse model (n=6). (C) The cells in extracted BALF were analyzed by cell counting in a mouse model (n=6). (D) Representative images of immunofluorescence staining of NETs in lung tissues from METTL3^+/+ and METTL3^-/- murine models in the sham and sepsis-associated lung injury group (red: NE, green: CitH3, blue: DAPI). (E) The cfDNA level in plasma (n=6) and (F) BALF was detected in METTL3^+/+ and METTL3^-/- murine models in the sham and sepsis-associated lung injury groups (n=6). (G) The level of TNFα, IL-1α, IL-8 and TGF-β were determined by ELISA in METTL3^+/+ and METTL3^-/- murine models in the sham and sepsis-associated lung injury groups (n=6). *P<0.05 (Two-way ANOVA with Tukey's correction).

Figure 10

PAD4 knockout protects mice against sepsis-associated ALI. (A) Paraffin-embedded lung tissue samples were stained for H&E. Representative histological images are shown at 400× magnification. (B) Pulmonary edema was evaluated by determining the wet/dry weight ratio in PAD4^+/+ and PAD4^-/- mouse models in the sham and sepsis-associated lung injury groups (n=6). (C) The cells in extracted BALF were analyzed by cell counting in PAD4^+/+ and PAD4^-/- mouse models in the sham and sepsis-associated lung injury groups (n=6). (D) Representative images of immunofluorescence staining of NETs in lung tissues from the sham and sepsis-associated lung injury murine models. (E) The cfDNA levels in plasma (n=6) and (F) BALF were detected in PAD4^+/+ and PAD4^-/- mouse models in the sham and sepsis-associated lung injury groups (n=6). (G) The levels of TNFα, IL-1α, IL-8 and TGF-β were determined by ELISA in PAD4^+/+ and PAD4^-/- mouse models in the sham and sepsis-associated lung injury groups (n=6). *P<0.05, **P<0.01 (Two-way ANOVA with Tukey's correction).