Lipopolysaccharide-induced histone lactylation mediates m6A RNA modification causing mitochondrial dysfunction and pulmonary fibroblasts activation to exacerbate sepsis-associated pulmonary fibrosis

Abstract

Background

Histone lactylation and N6-methyladenosine (m6A) alteration are epigenetic modifications that have a crucial function in controlling gene expression throughout fibroblast activation and organ fibrosis. However, their roles in sepsis-associated pulmonary fibrosis (SAPF) remain unclear.

Methods

This study established a mouse and cell model induced by lipopolysaccharides (LPS) to investigate the possible mechanisms of lactylation and METTL3-mediated m6A RNA modification in pulmonary fibroblast activation and sepsis-associated PF. The gene expression of m6A modification and lactylation in pulmonary fibroblasts of LPS-induced PF mouse model was examined using scRNA-Seq. Moreover, METTL3 short hairpin RNA (shRNA) and adeno-associated virus (AAV) were employed to knockdown METTL3 expression, and the glycolysis inhibitor Oxamate was utilized to attenuate lactate production and histone lactylation. Furthermore, to confirm the target gene controlled by m6A and H3K18 lactylation (H3K18la), ChIP-qPCR and RNA pulldown investigations were carried out.

Results

Single-cell RNA-sequencing unveiled the promotion of m6A modification and lactylation in pulmonary fibroblasts of LPS-induced PF mouse model. Furthermore, the induction of LPS resulted in an elevation of H3K18la lactylation and METTL3 concentrations, a reduction in PGC-1α levels, and the onset of mitochondrial dysfunction, all of which contribute to the activation of lung fibroblasts and the development of pulmonary fibrosis. Therapeutic effectiveness was observed in both in vitro and in vivo settings through focused rectification of abnormal histone lactylation or by reducing the expression of METTL3.

Conclusion

Our study demonstrates, LPS-induced histone lactylation contributes to sepsis-induced pulmonary fibrosis by upregulating METTL3 expression. Additionally, METTL3 recognizes m6A-modified PGC-1α mRNAs, leading to mitochondrial dysfunction and accelerated fibroblast activation, ultimately driving pulmonary fibrosis. METTL3-mediated m6A modification potently degraded PGC-1α, leading to mitochondrial dysfunction and accelerated fibroblast activation, ultimately driving Sepsis-Associated PF. This suggests that the presence of histone lactylation in the fibrotic microenvironment associated with sepsis plays a crucial role in triggering the expression and activity of the RNA methyltransferase METTL3.

Introduction

Sepsis, an intense multiorgan disorder resulting from an unregulated immune system crisis, may lead to acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) [1, 2]. 40% of patients with ARDS subsequently develop pulmonary fibrosis (PF) [3], a progressive condition marked by characterized by fibrous scarring of the lung parenchyma, irreversible impairment of lung function and ultimately respiratory failure[4]. Pulmonary fibroblasts serve as the primary effector cells in pulmonary fibrosis. Their aberrant proliferation, migration and the overproduction of extracellular matrix (ECM) lead to reduced pulmonary compliance and impaired lung function [5–8]. Our previous research has indicated an association between LPS-induced activation of pulmonary fibroblasts and sepsis-associated pulmonary fibrosis, but the underlying mechanisms remain to be elucidated [9–11].

Recent research has highlighted that cellular metabolic reprogramming played a crucial role in both sepsis and PF [12, 13]. Sepsis disrupts mitochondrial respiration and an accumulation of glycolytic enzymes [14]. Similarly, in numerous diseases involving PF, the activated fibroblast exhibit a glycolytic phenotype, and elevated lactate levels [15].Additionally, our research indicates that the induction of aerobic glycolysis in lung fibroblasts by LPS is linked to the activation of fibroblasts. However, the specific mechanisms behind by which lactate modulates fibroblast activation during SAPF remains unclear.

Growing interest links systemic inflammation and fibrotic diseases to metabolic reprogramming–driven epigenetic regulation, particularly through lactate-induced histone lactylation and RNA N-6-methyladenosine (m6A) modification [16, 17]. Research has suggested that lactylation and m6A alteration might play a role in the mechanisms of arsenite induced Idiopathic pulmonary fibrosis (IPF) [18]. Lactate accumulation during glycolysis promotes lysine lactylation (Kla) on histones, altering transcriptional activity, whereas METTL3-catalyzed m6A methylation at the N-6 position of mRNA adenosine regulates transcript stability and translation via the reader protein YTH N6-methyladenosine RNA binding protein 2 (YTHDF2) [16]. In 2021, Yu et al. revealed that lactylation modification induced by tumor tissue glycolysis can promote the expression of the reader protein YTHDF2, causing the degradation of the anti-cancer gene PER1/TP53 and contributing to the occurrence of ocular melanoma [19]. Similarly, Zhang et al. revealed that the regulatory axis of METTL3-m6A-YTHDF2 mediate the degradation of peroxisome proliferator-activated receptor gamma coactivator-1α (PGC-1α), hindering the process of mitochondrial biogenesis [17]. PGC-1α, a crucial regulatory protein, enhances the generation of new mitochondria and enhances mitochondrial performance [20, 21]. Given that PGC-1α activation mitigates LPS-induced acute lung injury [22], these findings suggest a potential interplay between lactate-induced lactylation and m6A methylation involved in pathogenesis of sepsis-associated pulmonary fibrosis.

In this study, we demonstrate that histone lactylation and m6A modification are increased in both a mouse model of sepsis-associated pulmonary fibrosis and LPS-activated lung fibroblast. Inhibition of histone lactylation or m6A efficiently suppressed sepsis-associated pulmonary fibrosis. Mechanistically, lactate-driven histone lactylation promotes METTL3 transcription, promoting m6A -dependent degradation of PGC-1α RNA, mitochondrial dysfunction, and activation of lung fibroblasts. These findings elucidate the mechanism of sepsis-associated pulmonary fibrosis from the perspective of intracellular epigenetic modification by lactate and showcase the interaction between histone lactylation and m6A methylation in sepsis-associated pulmonary fibrosis.

Materials and methods

Ethics statement and animals

Male mice of the C57BL/6 strain, aged 6–8 weeks and weighing between 18 and 27 g, were obtained from Shanghai SLAC Laboratory Animal in China. The animals were housed in a regulated setting, maintaining a temperature range of 22–24 °C, following a 12-hour cycle of light and darkness, and provided unrestricted availability of food and water. All experiments were approved by the Animal Care and Use Committee of Ren Ji Hospital, Shanghai Jiao Tong University School of Medicine.

Cell lines and culture

The murine lung fibroblast L929 cell line was acquired from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). L929 cells were cultured in Minimum Essential Medium (MEM, Hyclone, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, USA), 100 IU/ml penicillin and 100 IU/ml streptomycin.

LPS model and animal procedures

Male mice of the C57BL/6 strain, aged 6–8 weeks and weighing 18–27 g, were divided into two groups: Sham and LPS, using random allocation. Mice belonging to the LPS group were given a 5 mg/kg dose of LPS (Escherichia coli O127 B8; L3129; Sigma, USA) through intraperitoneal administration for three consecutive days. On the other hand, the sham group received saline injections on a daily basis.

To reduce METTL3 expression in lung tissue, an adeno-associated virus (AAV) containing METTL3 was injected into the trachea, while an inhibitor of glycolysis called Oxamate was injected into the peritoneum to decrease lactate production and histone lactylation (Fig.S1A). After intubating the animals in the animal facility, a week-long period of observation was carried out, during which the animals had unrestricted access to food and water. On the 7_th day, the mice were euthanized using sodium pentobarbital. The right lung was used for Western Blot (WB). The left lung, fixed in paraformaldehyde, underwent histopathology, immunofluorescence, and transmission electron microscopy (ECM) procedures.

METTL3 Inhibition by knockdown AAV transfection

Adeno-associated viruses (AAVs) were acquired from Genomeditech Co., Ltd. (Shanghai, China), and virus infection constructed stable L929 cell lines.

Before intubation, mice were given an AAV that expressed METTL3-shRNA, along with vector controls, which were intratracheal injection into the lungs of mice using 50 µl of PBS containing 1 × 10¹² µg per mouse for a duration of 4 weeks. Random allocation was performed to assign C57BL/6 male mice (6–8 weeks old; weighing 18–27 g) into four groups: Sham, LPS, METTL3-shRNA, and LPS + METTL3-shRNA. Mice in the LPS and LPS + METTL3-shRNA groups received LPS treatment.

Lactylation Inhibition by oxamate

Oxamate (500 mg/kg in corn oil, Selleck, USA) is a substance that inhibits glycolysis with a dosage of 500 mg/kg in corn oil. Before LPS intraperitoneal injection, Oxamate was administered intraperitoneally once. A group of male C57BL/6 mice (6–8 weeks old; weighing 18–27 g) were randomly divided into four groups: Sham, LPS, Oxamate, and LPS + Oxamate. Mice in the LPS and LPS + Oxamate groups received LPS intraperitoneal injection.

LPS model and cell procedures

LPS was used to replicate the sepsis microenvironment in vivo on L929 cells. The cells were incubated with 1ug/mL LPS (L4516; Sigma, USA) for 24 h. Following the completion of the LPS treatment, the cells were collected for Western blotting (WB), immunofluorescence analysis, and extracellular matrix (ECM) examination.

Treatment of cells with METTL3 knockdown

In order to inhibit METTL3, AAVs were introduced into L929 cells. AAVs at a concentration of 5E8TU/mL was utilized for conducting viral infection experiments. Confluent cells were randomly assigned into Sham, LPS, METTL3-shRNA, and LPS + METTL3-shRNA group. LPS and LPS + METTL3-shRNA cells received LPS treatment for 24 h.

Treatment of cells with oxamate

The L929 cells were randomly divided into Sham, LPS, Oxamate, and LPS + Oxamate group. Cells in the Sham group were administered with PBS as a control; cells in the LPS group were subjected to LPS treatment for a duration of 24 h; cells in the Oxamate group were cultivated in a medium containing 5mM Oxamate (Selleck, USA) for a period of 2 h; cells in the LPS + Oxamate group were cultivated in a medium containing 5mM Oxamate (Selleck, USA) for 2 h and subsequently exposed to LPS treatment for 24 h.

Single-cell analysis

We used single-cell RNA sequencing to analyse mouse lung cells. Gene expression data have been deposited into the SRA database at the National Center for Biotechnology Information with accession number SUB11878630. To examine lung cells in mice, we employed single-cell RNA-Seq on both sham (n = 3) and LPS groups (n = 3).The gene expression data has been submitted to the SRA database at the NCBI and can be accessed using the accession number SUB11878630.The cell number and viability in each sample were measured using Rigel S2 (Countstar, China).To generate the single-cell libraries, the single-cell libraries were created using the Gel Bead & Multiplex Kit, Chip Kit, and Chromium Single Cell 3’ V2 Chemistry Library Kit provided by 10x Genomics (Biomarker Technologies, China).To produce GEMs, cellular suspensions were loaded onto the Chromium Controller (10 x Genomics, Pleasanton) for Gel Bead-In-Emulsions generation. The generation of barcoded sequencing libraries was carried out in accordance with the instructions provided by the manufacturer, utilizing the Chromium Single Cell 3’ Reagent Kits v3.1 (10 x Genomics, Pleasanton). After library preparation, every sample underwent paired-end sequencing on a single lane of NovaSeq 6000, with 150nt for each end. The mm10 reference was used to process raw reads through the 10 x Genomics Cell Ranger pipeline, which can be found at https//support.10xgenomics.com/single-cell-gene expression/software/downloads/latest. Cell Ranger allows for clustering of individual cells, identification of marker genes, and exporting of unique molecular identifiers (UMI). To conduct additional analysis, the Seurat R package (version 2.2) was utilized. The majority of Seurat analyses were conducted with default parameters.

Pulmonary histopathology

Lung tissue was fixed overnight using a 4% solution of paraformaldehyde, followed by dehydration and embedding in paraffin. To assess alterations in lung morphology, a lung section measuring 5 micrometers in thickness was treated with hematoxylin and eosin (H&E) stain, while collagen deposition was determined using Masson’s trichrome stain.

Immunofluorescence staining

Pulmonary tissues were fixed with 4% PFA and permeabilized with 0.25% Triton X100. Formalin-fixed, paraffin sections (4 μm) were then stained using primary antibodies (refer to Supplementary Table for details). Donkey Anti-Rabbit IgG H&L (Alexa Fluor 488) (abcam, ab150073) and Goat Anti-Mouse IgG H&L (Alexa Fluor 594) (abcam, ab150116) served as the secondary antibodies. To identify nuclei, DAPI from Santa Cruz Biotechnology in Germany was employed. Images were captured using the Leica TCS SP2 confocal laser scanning microscope.

Transmission electron microscopy

For transmission electron microscopy, 4 μm sections of pulmonary tissues that had been fixed in formalin and embedded in paraffin were collected. These sections were then rinsed with 0.1 M phosphate buffer (PB) and subsequently treated with a solution containing 2% glutaraldehyde and 2% paraformaldehyde in 0.1 M PB for a duration of 3 min. Following this, the sections were further fixed with 1% osmic acid for 1 h. Following the dehydration process in ethanol, the samples were washed with propylene oxide and then embedded in epoxy resin. Subsequently, another round of dehydration in ethanol was performed on the samples. Lead citrate and uranyl acetate were used to stain semithin and ultrathin sections that were cut using a Reichert ultramicrotome on a Hitachi HT7800 TEM.

For the transmission electron microscopy of L929 cells, the cells were treated with 2.5% glutaraldehyde in 0.1 M phosphate buffer, and then exposed to 1% osmium tetraoxide (OsO4) in 0.1 M phosphate buffer. After fixation, cells were embedded in Epon 812 using a standard procedure. The Hitachi HT7800 TEM was used to examine approximately 70-lm ultrasected specimens by Ultracut Reichert–Jung stained with uranyl acetate and lead citrate.

Total protein extraction and BCA

The lung tissues or cells were lysed using RIPA lysis buffer (EpiZyme, China) containing 1% phenyl methyl sulfonyl fluoride protease inhibitor cocktail and phosphatase inhibitor cocktail (EpiZyme, China) for a duration of 15 min while kept on ice. The protein supernatant concentration was determined by using the BCA method and the BCA Protein Assay Kit (EpiZyme, China) after centrifuging at 15,000 × g for 15 min at 4 °C.

Western blot analysis

Proteins were separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and subsequently transferred to polyvinylidene difluoride membranes (PVDF) for further processing. The Supplementary Table listed the main antibodies and the membranes were incubated with the suitable HRP-linked Anti-rabbit IgG secondary antibodies (Cell Signaling Technology,7074 S) or HRP-linked Anti-mouse IgG secondary antibodies (Cell Signaling Technology,7076 S). The detection of blots was performed using Image LabTM software (Bio-Rad, USA) with the aid of the Enhanced ECL Chemiluminescent Substrate Kit (Vazyme, China).

Quantification of m6A in total RNA

Total RNA was extracted from lung tissue or cells following the manufacturer’s guidelines using an RNA purification kit (EZ Bioscience, USA). To measure m6A levels in the ischemic left ventricles of humans, pigs, and mice, whether failing or nonfailing, we employed a previously documented antibody-based technique for m6A capture and colorimetric quantification (P-9005, EpiGentek). The corresponding wells in a 96-well plate received 200ng of RNA per sample. Each experiment was conducted twice. As per the manufacturer’s recommendation, we included both negative and positive controls, along with a standard curve ranging from 0.02 to 1ng of m6A. In summary, following the attachment of RNA to the wells, the anti-m6A antibody was introduced and subsequently rinsed three to four times according to the guidelines provided by the manufacturer. The presence of m6A in RNA, which was bound by antibodies, was identified by introducing a solution for development. Subsequently, a SpectraMax plus microplate reader was utilized for colorimetric measurement. The m6A content in RNA was calculated and compared to the respective control samples, representing the proportion of m6A in the entire RNA.

Biotin-labeled RNAs pull down

The RiboTM RNAmax-T7 transcription kit (Ribo-Bio) was used to transcribe the PGC-1α RNA initially. Next, the enhanced RNA was labeled at the 3’ end with desthiobiotin using the Pierce RNA 3’ End Desthiobiotinylation Kit from ThermoFisher (20163). Ultimately, the Pierce Magnetic RNA-Protein Pull-Down Kit (ThermoFisher, 20164) was utilized to conduct RNA pull-down assays. A mixture was prepared by combining biotinylated RNAs, protein lysates, and streptavidin beads in quantities of up to 50pM, 2 mg, and 50 ml respectively. Following incubation and three rinses, the streptavidin beads were heated and utilized for the western blot analysis.

Chromatin Immunoprecipitation assay (ChIP)-qPCR

The ChIP assay was conducted with the PTM Bio anti-H3K18la antibody, following the instructions provided by the manufacturer of the ChIP Assay Kit [23]. The quantification of fold enrichment was performed using qRT-PCR and expressed as a percentage of the input chromatin (percentage of input).

JC-1 assay

The JC-1 Mitochondrial Membrane Potential assay kit (Beyotime, C2003S) was utilized to quantify the potential of the mitochondrial membrane. The L929 cells were rinsed with PBS once, and then treated with 500µL of staining working buffer per well for 20 min. After washing the cells twice with staining buffer solution, the covering liquid was replaced with MEM. Finally, the cells were examined using a fluorescence microscope (magnification, ×200; Olympus, Japan). Fluorescence microscopy filters were used to capture normal cells (cells that were not treated) containing aggregated JC-1, which emitted red fluorescence (excitation/emission = 540/570 nm). Additionally, mitochondrial dysfunction cells containing monomeric JC-1 were detected using fluorescence microscopy filters that emitted green fluorescence (excitation/emission = 485/535 nm).

Wound healing assay

Culturing of L929 cells took place in 6-well plates. After the L929 cells reached 80% confluence, a scratch ruler was used to create scratches on the plates using a 200µL pipette tip. Images of the identical location were captured 0 and 24 h after treatment and cultivation using a light microscope (Olympus, Japan) with a magnification of ×100.

Transwell assay

Equal numbers of L929 cells were distributed in the upper chamber of Transwell plates (Corning, United States), while 2 ml of ME containing 110% serums were added to the lower chamber. Following 24 h of incubation, the cells on the lower side of the membrane were immobilized using a 4% solution of formaldehyde and then permeated with 0.3% PBST for a duration of 30 min. Subsequently, the cells were washed three times with 0.1% TBST for 15 min each. L929 cells were then subjected to staining using crystal violet (Beyotime, China). The images of each membrane were captured using an inverted light microscope (magnification ×100, Olympus, Japan), and the number of transferred cells was assessed to determine the migration ability of L929 cells.

Statistical analysis

The data is presented as mean ± SEM. One-way analysis of variance (ANOVA) was conducted to analyze the means of three or more treatment groups. Student’s t-test (two-tailed) was used to compare the differences between the two groups. Statistical significance was determined by GraphPad Prism 9 (GraphPad Software Inc, CA) if the two variables showed significant differences with a p-value < 0.05.

Results

Elevated histone lactylation levels are associated with LPS-induced pulmonary fibrosis

Given the findings from our prior investigation, it is indicated that LPS stimulates aerobic glycolysis in pulmonary fibroblasts, resulting in the generation of significant quantities of lactate as materials for histone lactylation. Thus, we proceeded to assess the extent of protein lactylation in LPS-induced pulmonary fibrosis.

In order to establish the in vivo model of mice with LPS-induced pulmonary fibrosis, the mice were monitored for a week following the administration of 5mL/kg LPS for a duration of 3 days. Histological evaluations showed that LPS treatment worsened pulmonary injury, interstitial leukocyte infiltration, alveolar edema, and hemorrhage in comparison to the sham group. Masson staining revealed an augmentation of collagen deposits in the pulmonary interstitium following LPS administration. In addition, alveolar walls and patches of tissue were thickened (Fig. 1A).

Fig. 1.

Elevated histone lactylation levels are associated with LPS-induced pulmonary fibrosis. A, Lung injury was accessed by Hematoxylin and Eosin staining. Collagen deposition was assessed with Masson’s trichrome staining and evaluated through the Ashcroft fibrosis score. Original magnification ×200. Scale bars correspond to 100 μm (n= 6). B, Fibrosis was also quantified by the determination of collagen-I α1 (COL1A1), and α smooth muscle actin (α-SMA) in lung tissues by Western-Blot(WB). Relative densitometry of the protein bands of COL1A1 and α-SMA over Tubulin is displayed in bar graphs (n = 6). C, Lung tissues were stained with fluorophore-labeled antibodies against COL1A1 (Alexa Fluor 488, green) and α-SMA (Alexa Fluor 594, red). 4',6-diamidino-2- phenylindole (DAPI) stain was used to detect nuclei (blue). Original magnification ×200. Scale bars correspond to 100 μm (n = 6). D, lactylation was also quantified by the determination of global lactylation levels (Pan-Kla) and H3K18la levels in lung tissue by WB. Relative densitometry of the protein bands of Pan-Kla and H3K18la over Histone3 is displayed in bar graphs (n = 6).E, Lung tissues were stained with fluorophore-labeled antibodies against Pan-Kla (Alexa Fluor 488, green) and lung fibroblast marker α-SMA (Alexa Fluor 594, red). 4',6-diamidino-2- phenylindole (DAPI) stain was used to detect nuclei (blue). Original magnification ×200. Scale bars correspond to 100μm(n = 6). Data are expressed as means ± SEM. * p < 0.05, **p < 0.01, *** p < 0.001,**** p < 0.001

Western blot analysis revealed an elevated expression level of α-Smooth Muscle Actin (α-SMA, a marker for activated fibroblasts) and collagen α−1 type I (COL1A1) as well (Fig. 1B, Fig.S1B). Additionally, the immunofluorescence staining demonstrated that LPS increased the proportion of α-SMA⁺/COL1A1⁺ cells in comparison to the sham group (Fig. 1C).

The elevated levels of both global lactylation and H3K18la in the LPS group were verified through western blot and immunofluorescence staining (Fig. 1D-E, Fig.S1C). The data suggest that a significant proportion of pulmonary fibrosis induced by LPS is marked by increased histone lactylation, which is probably implicated in fibroblast activation.

LPS-induced pulmonary fibrosis is concomitant with an increase in m6A levels, a decrease in PGC-1α levels, and mitochondrial dysfunction

In order to clarify the underlying mechanisms of LPS-induced lung fibrosis, we carried out an extensive study utilizing single-cell RNA-Seq analysis. Figure 2A displays the tSNE plot representing all cell types. Figure S1D displays the cluster marker genes along with Fig.S1E showed DEGs in fibroblast cluster, whose GO enrichment analysis revealed histone modification process, lactylation and mitochondrial function were significantly changed (Fig. 2B).

Fig. 2.

LPS-induced pulmonary fibrosis is concomitant with an increase in m6A levels, a decrease in PGC-1α levels, and mitochondrial dysfunction. A, The UMAP plot of all cell types in lung tissue. B, GO enrichment analysis with DEGs between fibroblast from Sham and LPS group. C, EpiQuik M6A RNA Methylation Quantification Kit was used to detect m6A levels in lung homogenates(n = 3). D, Lung tissues were stained with fluorophore-labeled antibodies against METTL3 (Alexa Fluor 488, green) and lung fibroblast marker α-SMA (Alexa Fluor 594, red). 4',6-diamidino-2- phenylindole (DAPI) stain was used to detect nuclei (blue). Original magnification ×200. Scale bars correspond to 100 μm (n = 6). E, Protein expression of METTL3 in lung homogenates was determined by WB. Relative densitometry of the protein bands of METTL3 over Tubulin is displayed in bar graphs (n = 6).F, Lung tissues were stained with fluorophore-labeled antibodies against PGC-1α (Alexa Fluor 488, green) and lung fibroblast marker α-SMA (Alexa Fluor 594, red). 4',6-diamidino-2- phenylindole (DAPI) stain was used to detect nuclei (blue). Original magnification ×200. Scale bars correspond to 100 μm (n = 6). G, Protein expression of PGC-1α in lung homogenates was determined by WB. Relative densitometry of the protein bands of PGC-1α over Tubulin is displayed in bar graphs (n = 6).H, Representative transmission electron microscopy(TEM) images of mitochondrion in lung fibroblast cells of lung tissues (n = 6). Data are expressed as means ± SEM. * p < 0.05, **p< 0.01, *** p < 0.001, **** p < 0.001

In vivo experiments provided additional confirmation of the single-cell RNA-Seq data. M6A assay demonstrated m6A level increased in LPS group (Fig. 2C) while immunofluorescence staining of lung tissue showed significantly higher levels of METTL3⁺/α-SMA⁺ fibroblast in LPS group than in sham group (Fig. 2D). WB showed level of METTL3(m6A writer protein) was increased in LPS group (Fig. 2E, Fig.S1F).

Our investigation has unveiled an association between reduced PGC-1α expression and mitochondrial dysfunction. Immunofluorescence staining shows the decreased amounts of PGC-1α⁺/α-SMA⁺ were observed in the LPS group (Fig. 2F) while WB showed PGC-1α were decreased (Fig. 2G, Fig.S1G). Furthermore, we conducted transmission electron microscopy (TEM) to examine the ultrastructure of mitochondria. Fibroblasts in the sham group exhibited a typical mitochondrial morphology, whereas fibroblasts in the LPS group displayed pronounced mitochondrial dilation (Fig. 2H).

The results of our study indicate that a significant portion of LPS-induced lung fibrosis is characterized by an increase in m6A levels along with a decrease in PGC-1α, resulting in impaired mitochondrial function. The involvement of this mechanism is expected to have a crucial impact on the development of pulmonary fibrosis induced by LPS.

LPS-induced METTL3-m6A mediates PGC-1α mRNA translation to cause mitochondrial dysfunction to promote lung fibroblast activation

METTL3 is a well-known m6A writer protein that regulates the stability of m6A-deposited mRNA, thereby regulating their abundance and availability for translation. In order to further examine the function of METTL3, an AAV containing METTL3-shRNA was created to suppress the expression of METTL3 in L929 cells (Fig. S1H). Streptavidin RNA pull-down assay showed that METTL3 enhanced PGC-1α expression in L929 cell line (the murine lung fibroblast cell line) (Fig. 3A). By utilizing METTL3-shRNA, we effectively suppressed the expression of METTL3, resulting in a noticeable decrease in the protein level of PGC-1α as depicted in Fig. 3B and Supplementary Fig. 2A.

Fig. 3.

LPS-induced METTL3-m6A mediates PGC-1α mRNA translation to cause mitochondrial dysfunction to promote lung fibroblast activation. A, WB analysis of METTL3 after RNA pull-down assay with biotinylated-PGC-1α, and beads only (NC) in cell lysates of L929 cells. B, Protein expression of PGC-1α and METTL3 in L929 cells was determined by WB. Relative densitometry of the protein bands of PGC-1α and METTL3 over Actin is displayed in bar graphs (n = 3). C, Representative TEM images of mitochondrion in L929 cells (n = 3). D, JC-1 assay was performed to evaluate the mitochondrial function of L929 cells (n=3). E, Transwell assay was performed to evaluate the migration ability of L929 cells (n=3). F, Protein expression of COL1A1 and α-SMA in L929 cells was determined by WB. Relative densitometry of the protein bands of COL1A1 and α-SMA over Tubulin is displayed in bar graphs (n = 3). G, L929 cells were stained with fluorophore-labeled antibodies against COL1A1 (Alexa Fluor 488, green) and α-SMA (Alexa Fluor 594, red). 4',6-diamidino-2- phenylindole (DAPI) stain was used to detect nuclei (blue). Original magnification ×200(n = 3). Data are expressed as means ± SEM. * p < 0.05, **p < 0.01, *** p < 0.001, **** p < 0.001

The protein levels of METTL3 increased significantly in LPS group while the protein level of PGC-1α decreased significantly in the LPS group (Fig. 3B, Fig.S2A). Meanwhile, we measured mitochondrial function in L929 cells after LPS treatment. As illustrated in Fig. 3C, alterations in mitochondrial ultrastructure were visualized through TEM. JC-1 experiments also showed mitochondrial dysfunction after LPS treatment (Fig. 3D). Enhanced fibroblast migration was evident in the LPS group, as substantiated by both transwell and wound healing experiments (Fig. 3E, Fig. S2B).

Simultaneously, we observed fibroblast activation in the LPS group. After treating L929 cells with LPS, Western blot analysis showed a significant rise in the levels of COL1A1 and α-SMA (Fig. 3F). Furthermore, consistent with the Western blot results, the immunofluorescence staining revealed a notable increase in the proportion of cells positive for both COL1A1 and α-SMA following the administration of LPS (Fig. 3G, Fig.S2C).

Notably, all observed effects were mitigated when pretreated with METTL3-shRNA before exposure to LPS (Fig. 3B-G, Fig.S2A-C). These findings collectively suggest that the LPS-METTL3-PGC-1α-mitochondrial dysfunction may play a significant role in L929 cell activation in vitro.

LPS upregulates lactate-induced METTL3 via histone lactylation and causes mitochondrial dysfunction to promote lung fibroblast activation

Subsequently, we sought to ascertain whether histone lactylation represents a potential mechanism contributing to the upregulation of METTL3 expression. To reveal the regulatory function of histone lactylation on METTL3 expression, Chromatin immunoprecipitation (ChIP) was conducted using anti-H3K18la antibodies, followed by qPCR (ChIP-qPCR). The results demonstrated the enrichment of H3K18la in the promoter regions of METTL3 (Fig. 4A).

Fig. 4.

LPS upregulates lactate-induced METTL3 via histone lactylation and causes mitochondrial dysfunction to promote lung fibroblast activation. A, H3K18la relative occupancy with METTL3 promoter in L929 cells was analyzed by ChIP-qPCR. B, Protein expression of H3K18la and Pan-Kla in L929 cells was determined by WB. Relative densitometry of the protein bands of H3K18la and Pan-Kla over Histine3 is displayed in bar graphs (n = 3). C, Protein expression of PGC-1α and METTL3 in L929 cells was determined by WB. Relative densitometry of the protein bands of PGC-1α and METTL3 over Actin is displayed in bar graphs (n = 3). D, Representative TEM images of mitochondrion in L929 cells (n = 3). E, JC-1 assay was performed to evaluate the mitochondrial function of L929 cells (n=3). F, Protein expression of COL1A1 and α-SMA in L929 cells was determined by WB. Relative densitometry of the protein bands of COL1A1 and α-SMA over Tubulin is displayed in bar graphs (n = 3). G, L929 cells were stained with fluorophore-labeled antibodies against COL1A1 (Alexa Fluor 488, green) and α-SMA (Alexa Fluor 594, red). 4',6-diamidino-2- phenylindole (DAPI) stain was used to detect nuclei (blue). Original magnification ×200(n = 3). H, Transwell assay was performed to evaluate the migration ability of L929 cells (n=3). Data are expressed as means ± SEM. *p < 0.05, **p < 0.01, *** p < 0.001, **** p < 0.001

In order to assess the impact of inhibiting histone lactylation on m6A levels in LPS-induced pulmonary fibrosis, we employed glycolysis inhibitors Oxamate to decrease the overall intracellular level of histone lactylation in L929 cells. The administration of Oxamate led to a reversal of the increased protein levels of H3K18la and global lactylation in the LPS + Oxamate group (Fig. 4B, Fig.S2D). Moreover, the administration of Oxamate successfully suppressed the enhancement of METTL3 and the mitochondrial dysfunction induced by LPS in L929 cells. This was supported by the observed changes in the protein levels of PGC-1α and METTL3, as shown in Fig. 4C, Fig.S2E), a reduction in observed mitochondrial ultrastructural changes (Fig. 4D), and decreased alterations in JC-1 fluorescence (Fig. 4E). Subsequently, we observed that the reduction in histone lactylation effectively suppressed fibroblast activation. In the LPS + Oxamate group, WB showed a reduction in the heightened levels of COL1A1 and α-SMA, as illustrated in Fig. 4F. Simultaneously, immunofluorescence staining demonstrated a notable decrease in the proportion of COL1A1+/α-SMA + cells in the Oxmate-treated group in comparison to the LPS group (Fig. 4G, Fig.S2F).Additionally, transwell and wound-healing assays indicated a weakened migratory ability of cells in the LPS + Oxamate group (Fig. 4H, Fig.S2G).

Consequently, our hypothesis is that the lactic acid induced by LPS might result in the increase of METTL3 through histone lactylation, which in turn leads to mitochondrial dysfunction and coordinates the activation of lung fibroblasts.

Inhibiting METTL3 activation alleviates mitochondrial dysfunction and LPS-induced pulmonary fibrosis

To evaluate the significance of elevated METTL3 levels in LPS-induced pulmonary fibrosis, we employed WB, histopathological examination, and TEM. Intratracheal injection of METTL3-shRNA AAV was employed to infect pulmonary tissue in mice, resulting in the reduction of METTL3 level in the LPS group (Fig. S3A). In line with the alterations in pulmonary histopathology (Fig. 5A), the immunofluorescence staining reveals a notable rise in the proportion of lung fibroblast expressing COL1A1⁺/α-SMA⁺ after LPS exposure(Fig. 5B, S3B). Additionally, western blot analysis demonstrates an increased protein expression of COL1A1 and α-SMA in the pulmonary tissue (Fig. 5C).

Fig. 5.

Inhibiting METTL3 activation alleviates mitochondrial dysfunction and LPS-induced pulmonary fibrosis. A, Lung injury was accessed by Hematoxylin and Eosin staining. Collagen deposition was assessed with Masson’s trichrome staining (n = 6). B, Lung tissues were stained with fluorophore-labeled antibodies against COL1A1 (Alexa Fluor 488, green) and α-SMA (Alexa Fluor 594, red). 4',6-diamidino-2- phenylindole (DAPI) stain was used to detect nuclei (blue). Original magnification ×200. Scale bars correspond to 100μm (n = 6). C, Protein expression of COL1A1 and α-SMA in lung homogenates was determined by WB. Relative densitometry of the protein bands of COL1A1 and α-SMA over Tubulin is displayed in bar graphs (n = 6). D, Lung tissues were stained with fluorophore-labeled antibodies against METTL3 or PGC-1α (Alexa Fluor 488, green), lung fibroblast marker α-SMA (Alexa Fluor 594, red). 4',6-diamidino-2- phenylindole (DAPI) stain was used to detect nuclei (blue). Original magnification ×200. Scale bars correspond to 100μm (n = 6). E, Protein expression of METTL3 or PGC-1α in lung homogenates was determined by WB. Relative densitometry of the protein bands of METTL3 or PGC-1α over Actin is displayed in bar graphs (n = 6). F, Representative TEM images of mitochondrion in lung fibroblast cells of lung tissues. Data are expressed as means ± SEM. * p < 0.05, **p < 0.01, *** p < 0.001, **** p < 0.001

In the LPS group, the immunofluorescence staining and WB analysis demonstrated an upregulation of METTL3 protein levels and a downregulation of PGC-1α protein levels (Fig. 5D-E, Fig.S3C). Moreover, alterations in mitochondrial ultrastructure in lung fibroblast cells following LPS treatment were observed (Fig. 5F). Significantly, METTL3 downregulation attenuated all of these responses in the LPS + METTL3-shRNA group (Fig. 5A-F, Fig.S3B-C).

The findings suggest that blocking LPS-induced activation of METTL3 could potentially improve mitochondrial dysfunction and alleviate the progression of pulmonary fibrosis.

Suppressing histone lactylation relieves LPS-induced upregulation of m6A levels, mitochondrial dysfunction and pulmonary fibrosis

In order to examine the impact of lactylation inhibition in vivo on LPS-induced pulmonary fibrosis, mice were administered Oxamate through intraperitoneal injection for pretreatment. As depicted in Fig. 6A, pulmonary histopathological changes were attenuated by Oxamate pretreatment, while immunofluorescence staining indicated an increase in COL1A1⁺/α-SMA⁺ lung fibroblasts, which were subsequently reversed after Oxamate pretreatment (Fig. 6B, Fig.S3D). The levels of COL1A1 and α-SMA proteins in pulmonary tissue were increased as shown by WB, but this effect was reversed after pretreatment with Oxamate (Fig. 6C, Fig.S3E).

Fig. 6.

Suppressing histone lactylation relieves LPS-induced upregulation of m6A levels, mitochondrial dysfunction and pulmonary fibrosis. A, Lung injury was assessd by Hematoxylin and Eosin staining. Collagen deposition was assessed with Masson’s trichrome staining (n = 6). B, Lung tissues were stained with fluorophore-labeled antibodies against COL1A1 (Alexa Fluor 488, green) and α-SMA (Alexa Fluor 594, red). 4',6-diamidino-2- phenylindole (DAPI) stain was used to detect nuclei (blue). Original magnification ×200. Scale bars correspond to 100 μm (n = 6). C, Protein expression of COL1A1 and α-SMA in lung homogenates was determined by WB. Relative densitometry of the protein bands of COL1A1 and α-SMA over Tubulin is displayed in bar graphs (n = 6). D, Lung tissues were stained with fluorophore-labeled antibodies Pan-Kla (Alexa Fluor 488, green), lung fibroblast marker α-SMA (Alexa Fluor 594, red). 4',6-diamidino-2- phenylindole (DAPI) stain was used to detect nuclei (blue). Original magnification ×200. Scale bars correspond to 100 μm (n = 6). E, Protein expression of Pan-Kla, H3K18la and Histone3 in lung homogenates was determined by WB. Relative densitometry of the protein bands of Pan-Kla or H3K18la over Histone3 is displayed in bar graphs (n= 6). F, Lung tissues were stained with fluorophore-labeled antibodies against METTL3 or PGC-1α (Alexa Fluor 488, green), lung fibroblast marker α-SMA (Alexa Fluor 594, red). 4',6-diamidino-2- phenylindole (DAPI) stain was used to detect nuclei (blue). Original magnification ×200. Scale bars correspond to 100 μm (n = 6). G, Protein expression of METTL3 or PGC-1α in lung homogenates was determined by WB. Relative densitometry of the protein bands of METTL3 or PGC-1α over Actin is displayed in bar graphs (n = 6). H, Representative TEM images of mitochondrion in fibroblasts of lung tissues. Data are expressed as means ± SEM. * p < 0.05, **p < 0.01, *** p < 0.001, **** p < 0.001

Immunofluorescence staining revealed an increase in Pan-Lac⁺/α-SMA⁺ lung fibroblasts, which was subsequently reversed after Oxamate pretreatment (Fig. 6D, Fig.S3F). Additionally, WB demonstrated an increase in H3K18la and Pan-Lac expressions, which were then decreased following Oxamate pretreatment (Fig. 6E, Fig. S3E).

Furthermore, immunofluorescence staining demonstrated an increase in METTL3⁺/α-SMA⁺ lung fibroblasts and a decrease in PGC-1α⁺/α-SMA⁺ lung fibroblasts, both of which were reversed after Oxamate pretreatment (Fig. 6F, Fig.S3G). The Western blot results were consistent with the immunofluorescence findings (Fig. 6G, Fig.S3E). In parallel, we observed mitochondrial ultrastructure changes in lung fibroblast cells following LPS treatment, which were subsequently reversed after Oxamate pretreatment (Fig. 6H). Taken together, these results indicate that the suppression of histone lactylation may be a potential target for the treatment of LPS-induced pulmonary fibrosis.

Discussion

In our prior research, we showed that LPS could trigger aerobic glycolysis in lung fibroblasts, leading to lactate production and facilitating the progression of pulmonary fibrosis [9]. Although previous studies have elucidated the significant role of lactate as a cell metabolism byproduct in creating an extracellular pro-fibrotic microenvironment which is crucial for the formation of sepsis-associated pulmonary fibrosis. However, the mechanism by which lactate mediates epigenetic modification intracellularly during sepsis-associated pulmonary fibrosis is not well understood. This study reveals that during sepsis, LPS induces lactylation of histone H3K18 in pulmonary fibroblasts, enhancing METTL3 expression. METTL3, in turn, inhibits the expression of PGC-1α by m6A methylation modification, leading to mitochondrial dysfunction. Consequently, this cascade of events promotes the activation of pulmonary fibroblasts, facilitating the progression of sepsis-related pulmonary fibrosis (Fig. 7).

Fig. 7.

Lipopolysaccharide-induced Lactylation and METTL3-mediated m6A RNA modification cause mitochondrial dysfunction and pulmonary fibroblasts activation to exacerbate sepsis-associated pulmonary fibrosis. Figure 7 is created with Biorender.com

Pulmonary fibroblasts serve as the primary effector cells in pulmonary fibrosis. Activated lung fibroblasts aggregate to form fibrotic foci which constitutes crucial mechanisms underlying the occurrence of pulmonary fibrosis. Mitochondrial dysfunction acts as a pro-proliferative marker which could induce fibroblast activation and promote organ fibrosis [24, 25], cancer [26] or other autoimmune disease [27]. Mitochondrial biogenesis is a crucial mechanism for self-renewal and damage repair within mitochondria, serving as an essential pathway for mitochondrial quality control.PGC-1α plays a crucial role in enhancing mitochondrial biogenesis and enhancing mitochondrial function [20, 21]. Furthermore, research has indicated a strong correlation between the inhibition of PGC-1α gene expression and cellular movement [28]. During the course of sepsis-associated pulmonary fibrosis, our examination uncovered a decrease in the expression of PGC-1α. Conversely, upregulation of PGC-1α demonstrated the ability to ameliorate mitochondrial function, thereby reversing fibroblast activation and the sepsis-associated pulmonary fibrosis process. These findings suggest that strategies focused on modulating mitochondrial function hold significant therapeutic potential for the treatment of pulmonary fibrosis.

Epigenetic elements create a gene-specific epiregulome for controlling gene expression after transcription [29]. The epiregulome defines RNA N6-methyladenosine (m6A) modification as the predominant modification in mRNA. The m6A alteration is introduced by ‘writers,’ specifically, the methyltransferase (MTase) complex consisting of methyltransferase-like 3 (METTL3)/METTL14/Wilms’-tumor-1-associating protein (WTAP) [30, 31]. The main role of METTL3 is as the catalytic center in the complex, whereas METTL14 and WTAP serve as subunits responsible for regulation.The zinc-finger domain (ZFD) of METTL3, known as CysCys-Cys-His (CCCH), selectively attaches to an RNA that includes a consensus sequence of 50-GGACU-30. Conversely, it does not attach to RNAs lacking this sequence, indicating that ZFD functions as the target recognition domain (TRD) [32]. In 2021, a study discovered that METTL3 functions in monocyte inflammation induced by oxidized low-density lipoprotein (oxLDL). In this process, METTL3 and YTH N6-methyladenosine RNA binding protein 2 (YTHDF2) work together to alter PGC-1α mRNA, leading to its degradation. Consequently, the levels of PGC-1α protein decrease, ultimately intensifying the inflammatory response [17]. This study shows that the administration of LPS enhances the expression of METTL3, resulting in the deterioration of PGC-1α and subsequent impairment of mitochondrial function. The series of occurrences boosts the stimulation of lung fibroblasts, ultimately advancing the advancement of sepsis-associated pulmonary fibrosis. The present study unveiled that METTL3 in lung fibroblasts decreases the level of PGC-1α, leading to subsequent mitochondrial dysfunction. The progression of sepsis-associated pulmonary fibrosis is ultimately promoted by the activation of lung fibroblasts, which is enhanced by this series of events. Moreover, our results indicate that the suppression of METTL3 using METTL3-shRNA reduces the activation of lung fibroblasts, thus mitigating the progression of sepsis-associated pulmonary fibrosis. Hence, our findings indicate that the degradation of PGC-1α caused by METTL3 might contribute to the development of sepsis-associated pulmonary fibrosis.

Gene transcription from chromatin is directly stimulated by the epigenetic modification known as lactate-derived lactylation of histone lysine (K) residue, as recently discovered [33]. H3K18 lactylation was reported to induced METTL3 upregulation in Tumor-infiltrating myeloid cells [16]. In our prior investigation, we showed that the administration of LPS could enhance the concentration of lactate in lung fibroblast cells and facilitate the progression of pulmonary fibrosis [9]. Consistent with these findings, we observed the involvement of H3K18 lactylation in sepsis-associated pulmonary fibrosis, enhancing the degradation of PGC-1α through increased METTL3 expression. Specifically, the inhibition of histone lactylation decreased METTL3 expression and increased the expression level of PGC-1α, thereby mitigating sepsis-associated pulmonary fibrosis.

The gathering of proof has indicated that the interplay of RNA m6A methylation and histone/DNA epigenetic mechanisms is responsible for determining transcriptional outputs [29]. Recent reports indicate that METTL3-mediated m6A modifications on chromosome-associated regulatory RNAs (carRNAs), such as promoter-associated RNAs, enhancer RNAs, and repeat RNAs, have the potential to enhance the open chromatin state and subsequent transcription [34]. A different article examined the connection between Arsenite-idiopathic pulmonary fibrosis and the promotion of Ythdf1 transcription through hyper-H3K18la, which also facilitated the m6A methylation of Nrep mRNA [18]. Our research revealed the importance of lactylation of H3K18la and its validation for METTL3.The lactylation-m6A modification-PGC-1α regulatory axis potently induces mitochondrial dysfunction in the LPS-induced fibrotic environment. Based on the results of this study, we can expect that targeting the suppression of histone lactylation or m6A modification could be a promising approach to intervene in sepsis-associated pulmonary fibrosis. As a limitation of this study, we solely employed RNA pull-down to identify the decreased expression of PGC-1α mediated by METTL3 after LPS treatment. Hence, our strategy involves overcoming this constraint through the implementation of Methylated RNA Immunoprecipitation Sequencing sequencing (MeRIP-seq) and integrative genomics viewer (IGV) examination on lung tissue.

Although our study was conducted in murine models and lung fibroblast cell lines, the findings have meaningful translational implications. Sepsis-associated pulmonary fibrosis (SAPF) remains underdiagnosed and understudied in clinical settings, partly due to the lack of specific biomarkers and targeted therapies. Our data suggest that histone lactylation and METTL3-mediated m6A RNA modification could serve as mechanistic biomarkers for early detection or disease stratification. Furthermore, both lactylation and m6A pathways are potentially druggable, with emerging small-molecule inhibitors of METTL3 and metabolic modulators such as Oxamate showing efficacy in preclinical models. These findings raise the possibility that targeting the lactylation–m6A–PGC-1α regulatory axis may represent a novel therapeutic strategy for halting fibrosis progression in sepsis survivors. Future studies in human cohorts and patient-derived samples will be essential to validate these epigenetic signatures and assess their diagnostic and therapeutic potential.

Conclusion

We unveiled the fibrotic role of histone lactylation and METTL3-mediated m6A methylation in sepsis-induced pulmonary fibrosis. Lactate accumulation in the sepsis-induced fibrotic microenvironment potently induces METTL3 upregulation in fibroblasts through H3K18 lactylation, which, in turn, mediates m6A modification leading to PGC-1α mRNA degradation in lung fibroblasts. Our findings underscore the significance of lactylation-driven METTL3-mediated RNA m6A modification, offering fresh insights into the epigenetic regulation of sepsis-induced pulmonary fibrosis.

Abbreviations

ALI

Acute Lung Injury

ARDS

Acute Respiratory Distress Syndrome

COL1A1

collagen α-1 type I

ECM

Extracellular Matrix

IPF

Idiopathic Pulmonary Fibrosis

m6A

N6-methyladenosine

PF

Pulmonary Fibrosis

PGC-1α

Peroxisome proliferator-activated receptor gamma coactivator-1α

α-SMA

α-Smooth Muscle Actin

Authors’ contributions

Conceptualization, Zhengyu He and Shuya Mei; Methodology, Zhengyu He, Shuya Mei, Ri Tang, and Shaojie Qin; Software, Ri Tang, Wenyu Lin, Qiaoyi Xu, Shuyi Zhang, and Shuya Mei; Validation, Ri Tang, Shaojie Qin, Yawen Peng, Jinhua Feng and Qiaoyi Xu; Formal Analysis, Ri Tang, Shaojie Qin, Shuyi Zhang, Qiaoyi Xu and Shuya Mei; Investigation, Ri Tang, Shaojie Qin, Wenyu Lin, Shuyi Zhang, Jinhua Feng, Qiaoyi Xu, Yawen Peng, and Shuya Mei; Resources, Zhengyu He, Yuan Gao, and Shunpeng Xing; Data Curation, Ri Tang, Shaojie Qin, Jinhua Feng, Qiaoyi Xu and Shuya Mei; Writing original Draft, Ri Tang, Shaojie Qin, and Shuya Mei; Writing, review and editing, Ri Tang, Zhengyu He, Qiaoyi Xu, Yuan Gao, Jinhua Feng, Shunpeng Xing, and Shuya Mei; Visualization, Zhengyu He and Shaojie Qin; Supervision, Zhengyu He, Shuya Mei, Yuan Gao, and Shunpeng Xing; Project Administration, Ri Tang, Shaojie Qin, Jinhua Feng, Qiaoyi Xu, and Shuya Mei.

Funding

This study was supported by the National Natural Science Foundation of China (NSFC, Grant Nos. 81970059 and 82300086), the Shanghai Top Talent Program of the Eastern Talent Plan (BJWS2024012), the Fundamental Research Funds for the Central Universities (No. 24X010202059), and the Shanghai Engineering Research Center of Peri-operative Organ Support and Function Preservation (20DZ2254200).

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

Ethical approval and consent to participate All experimental protocols were approved by the Animal Care and Use Committee of Ren Ji hospital, Shanghai Jiao Tong University School of Medicine.

Consent for publication

Neither the entire paper nor any part of its content has been published or has been accepted elsewhere. All authors agreed to the publication of this manuscript.

Competing interests

The authors declare no competing interests.