Loss of interferon regulatory factor-1 prevents lung fibrosis by upregulation of pon1 expression

Abstract

Background

Interferon regulatory factor-1 (IRF1) is a transcription factor that plays a significant role in various biological processes, including inflammatory injury, viral infection, cell death, and immune responses, and it has been extensively studied in the context of different lung diseases. However, the mechanism underlying its involvement in lung fibrosis remains largely unknown.

Methods

Wild type (WT) mice, IRF1 global-null mice (Irf1^−/−) were subjected to a bleomycin-induced lung fibrosis model to enable examination of the role of IRF1 in lung fibrosis. Proteomic analysis of lung tissue from WT and Irf1^−/− mice treated with saline or bleomycin was performed to explore the mechanism of IRF1 in regulating lung fibrosis.

Results

In the bleomycin-induced fibrosis mouse model, increased expression of IRF1 was observed. Irf1 knockout mice displayed decreased lung fibrosis relative to WT mice following treatment with bleomycin. The protein expression of fibronectin, as assessed by the Western blot analysis of lung tissues, was downregulated in Irf1^−/− mice. We observed a similar reduction in collagen content using hydroxyproline detection. Histologically, there was less collagen deposition in the lungs of Irf1^−/− mice compared with WT mice. Proteomics data revealed that IRF1 may be involved in lung fibrosis via the regulation of ferroptosis. We determined that paraoxonase 1(PON1), a poorly characterized protein in lung fibrosis, was upregulated in Irf1^−/− mice following exposure to bleomycin. In vitro experiments revealed that IRF1 could regulate the level of GSH and MDA through PON1. We also determined that PON1 levels were lower in the plasma of IPF patients compared with healthy controls.

Conclusion

Our data highlight the importance of IRF1 in the fibrotic process, and PON1 may be a potential mediator of IRF1 in the progression of lung fibrosis.

Introduction

Idiopathic pulmonary fibrosis (IPF) is a progressive and devastating lung disease with unknown etiology. Despite the growing interest in IPF, the precise molecular mechanisms underlying the development of fibrosis and leading to the irreversible destruction of the lung are still unknown, and there is a lack of effective therapies for pulmonary fibrosis [1].

Interferon regulatory factors (IRFs) constitute a family of transcriptional factors comprising nine members (IRF1–9) in mammalian cells, which were originally identified as transcriptional regulators of type I interferons. IRF-1 is expressed in various tissues and cells [2, 3] and has been reported to be involved in inflammatory injury, viral infection, cell growth regulation, cell death, and immune responses [3,4,5,6]. Immune abnormalities, inflammation, and cell death are known to be the key pathogenetic mechanisms behind many pulmonary diseases. In our previous study, we found that IRF1 was involved in acute lung injury via oxidative stress, inflammation, and programmed cell death [7,8,9,10]. Although IRF1 has been much studied, its role in fibrosis has not been fully elucidated. Recent studies have found that IRF1 is involved in the fibrosis of important organs such as the heart, kidney, and liver; IRF1 contributes to the process of renal fibrosis through the down-regulation of Klotho [11]. Cardiac-specific IRF1 overexpression has been shown to exacerbate conditions such as aortic-banding-induced cardiac hypertrophy, ventricular dilation, and fibrosis, whereas IRF1-deficient (knockout) mice have been found to exhibit a significantly reduced hypertrophic response [12]. To date, the role of IRF1 in pulmonary fibrosis has not been studied. Furthermore, the mechanism of how IRF1 regulates pulmonary fibrosis remains largely unknown.

In the current study, we demonstrate a key role played by IRF1 in the pathogenesis of lung fibrosis using IRF1 knockout mice. Proteomic profiling in this study reveals a potential novel link between IRF1 and ferroptosis via regulating paraoxonase 1(PON1) in the progression of lung fibrosis.

Methods

Mice

WT C57BL/6 mice were purchased from Silaike Laboratory Animal Co Ltd (Changsha, China). Irf1 knockout (Irf1^−/−) mice were obtained from The Jackson Laboratory (Bar Harbor, ME, USA) and maintained in the laboratory animal center of the Central South University under specific pathogen-free conditions. Seven- to eight-week-old mice matched for age and sex were used in the experiments. All animal experiments and procedures in this study followed protocols approved by the Laboratory Animal Ethics Committee of Central South University.

Bleomycin model of lung fibrosis

Sex- and weight-matched mice were treated with a single intratracheal injection of bleomycin (2.5 mg/mice, Selleck, S1214) to induce lung fibrosis. Mice were anesthetized with an intraperitoneal injection of ketamine/xylazine (100/10/mg/kg) and intubated with a micro-spray (Yuyan company, Shanghai, China) under the assistance of direct laryngoscopy. Bleomycin was diluted in 50 μL phosphate-buffered saline (PBS) and was administrated by intratracheal instillation with a micro-spray to promote widespread distribution in the lungs [13, 14]. Control mice received an intratracheal instillation of 50 μL PBS only. Mice were euthanized 7, 14, or 21 days after bleomycin(BLM) administration.

TMT quantitative proteomic analysis

Lung tissues from WT or Irf1^−/− mice subjected to either PBS or bleomycin instillation were harvested and flash-frozen in liquid nitrogen. Proteomic analysis were processed by Ouyi Biotechnology Co., Ltd. (Shanghai, China). Proteins with quantification changes > 1.5 and a p-value < 0.05 were defined as differentially expressed proteins. After the differentially expressed proteins were identified, Gene Ontology/Kyoto Encyclopedia of Genes and Genomes (GO/KEGG) enrichment analysis was performed to describe the functions of these proteins. Briefly, the fundamental steps of the TMT labeling quantitative proteomics process are outlined as follows: Initially, total proteins are extracted from the samples. A portion is set aside for protein concentration measurement and SDS-PAGE analysis. Another aliquot is subjected to trypsin-mediated enzymatic digestion and subsequent labeling. Equal volumes of each labeled sample are then pooled for chromatographic separation. The final step involves LC–MS/MS analysis and comprehensive data interpretation of the pooled sample. Our bioinformatics approach initiates with the retrieval and quantification of proteomic data, advancing through quality assessment and data preprocessing. This is succeeded by an analysis of protein expression levels and a functional profiling of the identified proteins using multiple databases. Differentially expressed proteins are further scrutinized through Gene Ontology (GO) analysis, pathway analysis, and protein–protein interaction mapping. Comparative data between groups are evaluated through correlation analysis, clustering of expression patterns in heatmaps, and Venn diagram analysis to identify shared and unique features. Moreover, depending on the dataset, we delve into the relevance and significance of our findings, pinpointing key proteins and their associated functions or pathways for deeper investigation and validation.

Hydroxyproline(HYP) assay

Murine lungs were harvested on day 21 after bleomycin or PBS instillation. Left lungs were obtained for acid hydroxyproline quantification, which was performed using a hydroxyproline assay kit (catalog A030-3–1, Nanjing Jiancheng, China) according to the manufacturer’s protocol. This method is predicated on the principle that the oxidation byproducts of hydroxyproline and the oxidizing agent interact with dimethylaminobenzaldehyde, resulting in the formation of a purple-red chromophore. The intensity of this coloration is directly proportional to the concentration of Hyp present in the sample. The subsequent collagen measurement was executed following the protocol established by Crouse et al. in 1986. It was determined that hydroxyproline constitutes 13.4% of the collagen content, which was then utilized to convert the measured values to reflect the total collagen content in the samples. The basic steps are shown as follows, homogenize and hydrolyze samples, then neutralize and centrifuge to collect the supernatant. Centrifuge to collect the supernatant and prepare hydroxyproline standards for the calibration curve. Add oxidizing agent, then developer to form a chromophore. After incubation, read at 550 nm. Use the standard curve to determine hydroxyproline levels, adjusting for dilution.

Glutathione(GSH) and glutathione disulfide(GSSG) assays

Fresh lung tissues were used for GSH and GSSG detection, which was performed with a kit purchased from Beyotime (S0053) following the manufacturer’s protocol. In essence, the GSSG, once formed, can be efficiently converted back to GSH by glutathione reductase in the presence of NADPH. The assay is composed of two parts: the preparation of tissue extracts and the detection of total glutathione (GSH and GSSG). The GSH can be deactivated in the sample by simply adding the Masking Reagent. Therefore, temp-only the GSSG is detected by measuring the absorption (λmax = 412 nm) of DTNB (5,5 Edithiobis (2-nitrobenzoic acid). Also, GSH can be determined the quantity by subtracting GSSG from the total amount of glutathione.

Malondialdehyda(MDA) assay

Fresh lung tissues were used for MDA assays, which were performed using a kit purchased from Beyotime (S0131), and the manufacturer’s protocol was followed. The assay involves the reaction of lipid peroxidation products, primarily MDA, with thiobarbituric acid (TBA), which leads to the formation of MDA-TBA2 adducts called TBARS. TBARS yields a red-pink color that can be measured spectrophotometrically at 532 nm. The TBARS assay is performed under acidic conditions (pH = 4) and at 95 °C. Pure MDA is unstable, but these conditions allow the release of MDA from MDA bis(dimethyl acetal), which is used as the analytical standard in this method. Briefly, tissues can be homogenized using Western lysates, the tissue weight is 10% of the lysate. After homogenization, centrifuge 10,000 g-12,000 g for 10 min to take the supernatant for subsequent assays. Determine the protein content with the BCA Assay Kit to calculate the MDA concentration (expressed in μmol/mg protein units). Prepare TBA stock solution and MDA work solution, the prepared TBA stock solution should be stored at room temperature and protected from light, and will be valid for at least 3 months. The MDA test solution must be fresh. Detect the MDA concentration with MDA work solution as instructed in the protocol.

HE staining and masson staining

Paraffin block samples were used for HE staining and Masson staining, which were performed by the Biossci company.

Immunofluorescence staining

Immunofluorescence staining was performed using frozen tissue slices. Murine lungs were inflated using 2.0 ml O.C.T and PBS confound with two to one ratio, and were then transferred to 4% paraformaldehyde (PFA) and soaked overnight at 4 °C. The lungs were then transferred to a 30% sucrose solution and soaked overnight at 4 °C. The tissue was transferred from the 30% sucrose solution to the pure O.C.T. solution and snap frozen in liquid nitrogen then stored at –80 °C. The protocol for staining is described briefly as follows: Frozen sections were cut and dried at room temperature. The slices were then washed with PBST and fixed with 4% PFA. Tissue slices were permeabilized with 0.3% Triton X-100, then the slices were blocked with serum before being incubated overnight with a primary antibody to IRF1 (1:200, CST, 8478 s) or PON1 (1:100, Proteintech, 18,155–1-AP) in a humidified chamber at 4 °C. Samples were then incubated with the Alexa Fluor-488-conjugated secondary antibody and Alexa Fluor-647-conjugated secondary antibody(Abcam) for 1 h at room temperature. DAPI was used to stain the nucleus, and the slides were mounted using Antifade Mounting Medium (Beyotime, P0126). Images of the slides were obtained by confocal microscopy using a Zeiss LSM 980 laser scanning microscope.

Quantitative real-time PCR analysis

Total RNA was extracted using RNA Isolater Total RNA Extraction Reagent (Vayame, R401), and reverse transcription was performed with the HiScript II 1st Strand cDNA Synthesis Kit (Vazyme, R212-01) according to the manufacturer’s instructions. We performed quantitative analysis of gene expression using SYBR® PCR Master Mix (Vazyme, Q321), with GAPDH expression used as a control. The PCR primers used are described in Supplementary Table 1.

Immunoblots

Immunoblotting was performed using lysates from MLE 12 cells or lung tissues. Either RIPA buffer (Beyotime, P0013B) or tissue protein extraction reagent (Thermo Scientific, 78,510) with protease inhibitor (Thermo Scientific, 78,429) was used to prepare the lysates. The protein concentrations were measured using a BCA protein assay (Thermo Fisher, 23,225). Proteins were resolved by NuPAGE 4–20% gel (ACE, ET12420Gel), followed by transfer to a 0.2 uM or 0.45 uM PVDF membrane. Primary antibodies were used to detect murine IRF1 (1:1000, Cell Signaling Technology 8478 s), murine PON1 (1:1000, proteintech,18,155–1-AP), murine fibronectin (1:1000, Abcam, ab2413), GPX4 (1:1000, Huabio, ET1706-45), GAPDH (1:1000, proteintech, 10,494-1AP), actin (1:1000, Servicebio, GB15003-100), anti-PCNA (1:1000, Proteintech, 60,097–1-Ig) and anti-rabbit or murine IgG (1:10,000, Santa Cruz) were used as secondary antibody. Protein Ladders from Thermo Fisher (26,616,26,634) or Abiowell (AWB0236) were used to show protein sizes, and densitometric quantification of bands was carried out using FIJI running ImageJ software (version b) (https://fiji.sc/), including normalizing to GAPDH as a loading control. All the uncropped gels were attached in Supplementary File 1.

Cell culture and sirna transfection

The MLE-12 murine lung epithelial cell line MLE-12 was purchased from American Type Culture Collection (ATCC, Manassas, VA). Cells were cultured at 37 °C and 5% CO₂ in DMEM/F-12 medium supplemented with 0.005 mg/ml insulin, 30 nM sodium selenite, 10 nM hydrocortisone, 0.01 mg/ml transferrin, 10 nM beta-estradiol, 2 mM l-glutamine, 10 mM HEPES, and 2% fetal bovine serum (FBS; Gibco, California, USA). Small interfering RNAs (siRNAs) targeting IRF1, PON1, or the negative control were transfected with Lipofectamine 2000 (Invitrogen, Shanghai, China). The siRNA sequence were designed by ourselves and produced by Shanghai Shenggong. The primer sequences are shown in Supplementary Table 1.

Chromatin immunoprecipitation(chip) assay

For ChIP assays, we used a Chip Kit (Abcam, USA, #ab500) and followed the manufacturer’s protocol. Each ChIP assay was repeated three times, and we used 1 × 10⁶ cells for each reaction. Firstly, we fixed the cells with Buffer A/Formaldehyde/PBS (the final concentration of formaldehyde was approximately 1.1%). Secondly, we performed cell lysis and sheared DNA using a sonicator to give an optimal DNA fragment size of 200–1000 bp, and we detected the DNA fragment sizes on a 1.5% agarose gel. The chromatin was subjected to immunoprecipitation using an IRF1 antibody (Cell Signal Technology catalog number 8428S, 2 µg), along with a positive control (ab1791) and a negative control consisting of beads only. The incubation was carried out overnight with continuous rotation at a temperature of 4 ℃ to ensure optimal interaction and precipitation of the protein-DNA complexes. We incubated protein A beads with antibody/chromatin samples, then purified DNA and proceeded with qPCR. The primers for ChIP qPCR are shown in Supplementary Table 1.

Enzyme-linked immunosorbent assay(ELISA) assay

The plasma samples were tested in ELISA assays, with each sample tested in triplicate. ELISA assays were performed using a kit (Aifang biologic, AF10504-A) and following the manufacturer’s instructions.

Statistical analysis

All data were analyzed using SPSS 25.0 statistical software or Graphpad Prism 8.0 software, and the results were expressed as mean ± standard deviation. The comparison of mean values between two groups was performed by t-test or Mann–Whitney test, and the comparison of mean values between multiple groups was carried out by one-way analysis of variance; p < 0.05 was considered statistically significant.

Results

Irf1 was upregulated in a bleomycin-induced lung fibrosis model and IPF patients

The elevated levels of fibronectin (FN) and enhanced deposition of collagen type I in the BLM group, as compared to the PBS group, substantiated the successful establishment of our fibrosis model.(Fig. 1A–D). We trialed different time points in the pulmonary fibrosis model before adopting the time point of 21 days for the following experiments (Fig. 1E). We then examined the expression of IRF1 in the bleomycin-induced lung fibrosis model. The lungs of C57BL/6 mice subjected to bleomycin displayed higher expression protein of IRF1 (Fig. 2A) compared with mice subjected to PBS. Meanwhile, we found that IRF1 was mainly expressed in the nucleus. Figure 2B–C shows the relative quantification of IRF1 in WT mice after bleomycin treatment. Consistently, we found that the expression of IRF1 in the lung tissue of IPF patients was also elevated(Supplementary Fig. 1).

Fig. 1

Day 21 post-bleomycin administration was utilized to establish the lung fibrosis model. qPCR detection of fibronectin on 7, 14, and 21 days after intratracheal administration of bleomycin, each plot represents one sample; B qPCR detection of collagen 1 at different time points after bleomycin administration, each plot represents one sample; C Western blot measurement of Fn at day 21; D Bar chart represents densitometry analysis of western blot data(mean ± SEM,3 independent experiments), protein quantification of fibronectin; E Representative images of Masson’s trichrome-stained lung sections of WT mice after intratracheal administration of bleomycin at different time points. (n = 8 for each group). Statistic analysis was performed by T-test, One-way ANOVA, or Mann–Whitney test. “*” indicates p < 0.05; “**” indicates p < 0.01; “***” indicates p-value < 0.001; and “****” indicates p < 0.0001

Fig. 2

IRF1 was upregulated in the bleomycin-induced lung fibrosis model. Representative images of Immunofluorescence staining of IRF1 and Epcam in lung tissue after bleomycin treatment, Scale bars:100um.(n = 3, PBS; n = 6, BLM); B Lung tissue lysates were subjected to Western bTo examine the importancelot analysis for IRF1. GAPDH was the standard. C Bar graph represent densitometric analysis of Western blot data (mean ± SEM, 3 independent experiments). Data were analyzed by Mann- Whitney test. “**” indicates p < 0.01

Loss of irf1 protects against bleomycin-induced lung fibrosis

To examine the importance of IRF1 in lung fibrogenesis, we subjected IRF1 deficient (Irf1^−/−) and WT mice to treatment with bleomycin or PBS; the Irf1^−/− mice had less body weight loss (Fig. 3A) after bleomycin administration. Results generated using a HYP kit showed that the level of collagen in the lung was lower in Irf1^−/− mice than in WT mice 21 days after bleomycin treatment (Fig. 3B). Mice subjected to bleomycin treatment displayed greater levels of injury and increased inflammatory collagen deposition in the WT group relative to the Irf1^−/− group, as detected by HE staining and Masson’s trichrome staining (Fig. 3C, D). The expression of FN and markers of Extracellular Matrix(ECM) deposition were increased in the lung 21 days after bleomycin treatment in WT mice, whereas these effects were suppressed in Irf1^−/− mice (Fig. 3E, F). These results indicate that Irf1^−/− mice are protected against bleomycin-induced lung fibrosis.

Fig. 3

Loss of IRF1 protects against bleomycin-induced lung fibrosis. WT or Irf1^−/− mice were subjected to PBS or bleomycin instillation, and lungs were harvested on day 21 after the procedure. A Body weight changes of WT or Irf1^−/− mice after PBS or bleomycin treatment;B Hydroxyproline levels in the left lung from control and Irf1^−/− mice 21 days after PBS or bleomycin treatment; C Representative images of HE-stained lung sections of WT and Irf1^−/− mice after PBS or bleomycin treatment(n = 3, PBS; n = 6, BLM). D Representative images of Masson’s trichrome staining of WT and Irf1^−/− mice after PBS or bleomycin treatment(n = 3, PBS; n = 6, BLM). E Lung tissue lysates were subjected to Western blot analysis for fibronectin(FN), and GAPDH was the standard. F The bar graph represents the densitometric analysis of Western blot data. Statistic analysis was performed by T-test, One-way ANOVA, or Mann–Whitney test. “*” indicates p < 0.05; “**” indicates p < 0.01; “***” indicates p-value < 0.001; and “****” indicates p < 0.0001

Irf1 regulates ferroptosis in bleomycin-induced lung fibrosis tissue

To explore the mechanism of IRF1-regulated lung fibrosis, we performed proteomic profiling in lung tissues from WT and Irf1^−/− mice. Volcano plots revealed that 123 genes were downregulated and 26 genes were upregulated in IRF1 deficient mice compared with WT mice (Fig. 4A). A Venn diagrammatic analysis revealed an intersection of 51 differentially expressed genes between the comparisons of WT mice treated with BLM versus those treated with PBS, and the Irf1^−/− mice treated with BLM versus the WT mice treated with BLM. The heatmap of these overlapping genes are shown in Fig. 4C. The top five upregulated genes were fatty acid binding protein 1 (FABP1), carbonyl reductase 2 (CBR2), transmembrane protein 38A (TMEM38a), lymphatic vessel endothelial hyaluronan receptor-1 (LYVE1), and PON1. The top five downregulated genes were interferon-induced with tetratricopeptide repeats (IFLIT1), cathepsin S (CTSS), LGK-V19-17, cathepsin D (CTSD), and CD68. The IFLIT1, CTSS, and CTSD proteins are well known pro-fibrotic factors, and qPCR analysis confirmed that they were downregulated in the Irf1^−/− mice (Fig. 4D–F), indicating that the deletion of Irf1 in mice alleviated the process of fibrosis. KEGG pathway analysis of these genes showed that the most significantly changed cellular process pathways included several programmed cell death pathways, including apoptosis and ferroptosis (Fig. 5A). GO analysis showed that most of the top differentially expressed genes were extracellular-space components and components of lipid metabolism pathways involved in the lung fibrosis model (Fig. 5B). Lipid peroxidation is a hallmark of ferroptosis that directly destroys cellular membranes, thereby causing ferroptosis. Lipid peroxidation occurs in polyunsaturated fatty acids on specific phospholipids, and various lipid metabolic pathways, resulting in ferroptosis. To determine whether ferroptosis levels changed in Irf1^−/− mice, we performed GSH and MDA assays and found that the levels of GSH were lower in the lung 21 days after bleomycin treatment in WT mice, whereas these effects were rescued in Irf1^–/– mice; MDA levels showed an opposite trend compared with GSH levels (Fig. 5C, D). Meanwhile, we found that the expression of GPX4 was downregulated in WT mice following bleomycin administration, whereas IRF1 deficiency alleviated this effect (Fig. 5E, F). We found that after bleomycin intervention, the mitochondria of WT mice were significantly shrunk, and the mitochondrial cristae were reduced or disappeared, which were typical characteristics of ferroptosis, while the mitochondrial morphology of knockout IRF1 was improved(Fig. 5G).

Fig. 4

Proteomics analysis of lung tissue reveals that fibrosis was alleviated in Irf1^−/− mice. Volcano plot of proteomics data between WT and Irf1^−/− mice after bleomycin administration; B A Venn diagram analysis  to identify the overlap of differentially expressed genes in two comparisons: wild-type (WT) mice treated with bleomycin (BLM) versus those treated with phosphate-buffered saline (PBS), and IRF1-deficient (Irf1−/−) mice treated with BLM versus WT mice treated with BLM; C The gene names of the above overlapping differentially expressed genes between WT and Irf1^−/− mice after bleomycin treatment; D–F qPCR detection of representative pro-fibrotic genes, each plot represents one sample. Statistic analysis was performed by one-way ANOVA. “*” indicates p < 0.05; “**” indicates p < 0.01; “***” indicates p-value < 0.001; and “****” indicates p < 0.0001

Fig. 5

Ferroptosis was down regulated in Irf1^−/− mice relative to WT mice. KEGG analysis of the differently expressed genes reveals the top 16 pathways of the cellular process; B GO analysis of the top different expressed genes; C GSH levels of WT and Irf1^−/− mice after PBS or bleomycin treatment (n = 5 for each group); D MDA levels of WT and Irf1^−/− mice after PBS or bleomycin treatment(n = 5 for each group); E–F. Lung tissue lysates from WT and Irf1^−/− mice after bleomycin treatment were subjected to Western blot analysis for GPX4, and GAPDH was the standard. G The electron microscopy images after bleomycin intervention in WT mice and IRF1 knockout mice. The left panel represents the WT group, where the mitochondria appear shrunken with a reduction or disappearance of mitochondrial cristae. The right panel depicts the IRF1 knockout mice, where there is an observable improvement in mitochondrial morphology. Statistic analysis was performed by T-test, One-way ANOVA, or Mann–Whitney test. “*” indicates p < 0.05; “**” indicates p < 0.01; “***” indicates p < 0.001; and “****” indicates p < 0.0001

Pon1 might be the molecule linking irf1 and ferroptosis

To delve into the mechanisms underlying IRF1-regulated ferroptosis, we conducted a literature review on the overlapping genes, PON1, a gene that is closely related to lipid metabolism, has caught our attention. qPCR showed that PON1 was upregulated in Irf1^−/− mice (Fig. 6A). Furthermore, the expression of PON1 protein consistently increased in IRF1-deficient mice after bleomycin administration (Fig. 6B–D). We found that in the absence of bleomycin stimulation, the knockout of IRF1 does not alter the expression of PON1(Supplementary Fig. 2A, B). Then we performed immunofluorescence staining of PON1 and found that PON1 was mainly expressed in the cell membrane and was suppressed after bleomycin administration, whereas its expression could be rescued by IRF1 knockout (Fig. 6E). GO analysis has shown that the high-density lipoprotein particle was upregulated in Irf1-/- mice relative to WT mice after bleomycin administration. PON1 is a hydrolase located on High-Density Lipoprotein(HDL) that has been postulated to have a protective effect on low-density lipoprotein oxidation, and PON1 has also been reported to protect HDL from oxidation. In previous studies, PON1 has had anti-inflammation effects, liver fibrosis effects, and lipid oxidation effects [15,16,17]. Therefore, we speculated that PON1 may be a target of IRF1 and might regulate ferroptosis via lipid oxidation. We next explored the effect that knocking down IRF1 in vitro had on PON1. The MLE-12 cell line was treated with bleomycin with or without IRF1 siRNA, and we found that bleomycin could induce the expression of IRF1 both in the nucleus and in the total cell lysis. In addition, PON1 and GPX4 were suppressed by bleomycin treatment but could be rescued by IRF1 knockdown (Fig. 7A–D). We also found that in the absence of bleomycin intervention, the negative control and IRF1 siRNA had almost no effect on the expression levels of GPX4 and PON1(Supplementary Fig. 2C). Meanwhile, we discovered that following BLM intervention in MLE-12 cells, the levels of MDA were upregulated and the levels of GSH were downregulated. The knockout down of IRF1 was able to partially reverse this effect(Fig. 7E, F). IRF1 is a transcription factor; To further analyze the relationship between IRF1 and PON1, we obtained the predicted binding site from the JASPAR website and designed corresponding primers. The detailed sequences are shown in Supplementary File 2. Then we performed ChIP qPCR to validate and found that IRF1 could bind to the DNA sequence of the Pon1 gene (Fig. 7G). The IRF1 antibody exhibited a significant 2.7-fold increase in signal intensity compared to the control group. This elevation in fold enrichment, surpassing the threshold of 1, suggests a notable upregulation of IRF1 binding in the immunoprecipitated (IP) samples relative to the control. To ascertain the role of PON1, we further silenced the expression of PON1 and chosed siRNA-2 for the following experiment(Fig. 7H). Then we conducted an experiment where PON1 expression was silenced in cells with IRF1 knockdown. Our findings indicate that the suppression of PON1 was able to counteract the upregulation of GPX4 that was triggered by the depletion of IRF1 (Fig. 7I). Additionally, the reduction in MDA levels achieved by IRF1 knockdown was counteracted, with MDA levels rising upon PON1 silencing, while GSH levels experienced a decrease again (Fig. 7J, K).

Fig. 6

PON1 expression was rescued by Irf1 knock-out. qPCR detection of mRNA expression of PON1; B Lung tissue lysates were subjected to Western blot analysis for PON1, and GAPDH was the standard; C-D Protein quantification of PON1, the bar graph represents the densitometric analysis of Western blot data; E Immunofluorescence staining of PON1, scale bars:10um.“*” indicates p < 0.05; “**” indicates p < 0.01; “***” indicates p-value < 0.001; and “****” indicates p < 0.0001

Fig. 7

IRF1 regulate GPX4 via PON1. MLE-12 cell line was transfected by IRF1 siRNA with lipo2000, qPCR detection of the efficiency of IRF1 siRNA; B-D.Western blot tests of IRF1, nuclear IRF1, PON1, and GPX4 after bleomycin administration with or without IRF1 siRNA treatment; E Silencing IRF1 partly reverses Bleomycin-Induced MDA Levels in MLE-12 Cells. Bleomycin leads to an increase in MDA levels in MLE-12, but silencing IRF1 can partially reduce MDA levels; F The effect of IRF1 silencing on the GSH Level in the MLE-12 cell line with bleomycin treatment. G ChIP-qPCR to analyze the recruitment of IRF1 onto the DNA sequences of PON1; H qPCR detection of the efficiency of PON1 siRNA; I Western blot tests of IRF1, PON1 and GPX4 after bleomycin and IRF1 siRNA administration with or without PON1 siRNA treatment; J In cells with IRF1 knockdown, further silencing PON1 exacerbates the accumulation of MDA caused by bleomycin; K Knocking down PON1 weakens the effect of reversing GSH levels brought about by silencing IRF1 in MLE-12 cell line with bleomycin treatment. “*” indicates p < 0.05; “**” indicates p < 0.01; “***” indicates p-value < 0.001; and “****” indicates p < 0.0001

A decrease in circulatory pon1 levels may predispose to IPF

We performed ELISA assays to measure PON1 levels in the plasma of IPF patients and corresponding control patients. We found that the IPF group had significantly lower levels of PON1(175.2 pg/ml) in plasma than controls (198.8 pg/ml) (Supplementary Fig. 3). The demographic characteristics of patients were shown in Supplementary Table 2; there were no significant differences in age or gender between the IPF and control groups.

Discussion

In the current study, we have discovered that IRF1 exerts profibrotic effects in bleomycin-induced lung fibrosis. Expression of IRF1 was markedly induced in the lung following bleomycin administration. Moreover, IRF1 deficiency provided protective effects, as evidenced by reduced levels of collagen and fibronectin in the lung and improved body weight loss in response to bleomycin. Proteomic profiling revealed a potential novel link between IRF1 and the PON1/GPX4 axis in the progression of lung fibrosis. Taken together, these results suggest that the IRF1-PON1-GPX4-dependent pathway makes an important contribution to the pathogenesis of lung fibrosis, making it a potential therapeutic target in lung fibrosis (Fig. 8). To our knowledge, this is the first study to reveal that IRF1 is involved in lung fibrosis via regulating ferroptosis, additionally, PON1 was also first reported to be regulated by IRF1 and presented a significant decrease in the plasma of IPF patients compared to controls, indicating that PON1 may involve in pulmonary fibrosis. Its potential linkage to IRF1 and ferroptosis are areas of our interest that warrant further investigation.

Fig. 8

Graphic abstract of the manuscript. IRF1 is upregulated following bleomycin treatment. It directly binds to PON1, resulting in the downregulation of PON1 expression. PON1, in turn, modulates the levels of GPX4 and MDA, ultimately leading to epithelial cell ferroptosis and the development of lung fibrosis

We have observed reduced histological and tissue markers of fibrosis following bleomycin treatment in Irf1^−/− mice compared with WT mice. In support of our current findings, a recent study has shown that treatment with an anti-IRF1 primary antibody inhibited the expression of IFNγ and alleviated senescence-associated pulmonary fibrosis [18]. Similarly, IRF1 has also been reported in the pathogenesis of fibrosis in other organs, including kidney fibrosis [11] and heart fibrosis [12]. To determine the mechanism of action of IRF1, we performed proteomic profiling of lung tissue from Irf1^−/− mice, and we observed that the ferroptosis pathway was downregulated in the lung tissue of Irf1^−/− mice compared with the corresponding control mice. It is known that ferroptosis plays a very important role in the process of fibrosis [19,20,21,22]. Previous studies have shown that IRF1 could regulate several types of programmed cell death, including apoptosis, necroptosis, and pyopotosis [6, 23]. In this study, we found that ferroptosis could also be regulated by IRF1, which is similar to the findings from recent studies [24, 25].

To further investigate the mechanism of IRF1-regulated ferroptosis, we performed GO analysis of 51 differentially expressed genes between the comparisons of WT mice treated with BLM versus those treated with PBS, and the Irf1-/- mice treated with BLM versus the WT mice treated with BLM, and found that high-density lipoprotein particle was upregulated in Irf1^−/− mice relative to WT mice after bleomycin administration. PON1 is one of the top five upregulated genes in this analysis, and it is a hydrolase located on HDL that has been postulated to play a protective role in preventing lipoprotein oxidation [26]. In a previous study, PON1 was found to be involved in liver fibrosis [27, 28]. In the current study, in vivo experiments showed that bleomycin can decrease the expression of PON1; however, knocking out IRF1 can weaken this effect. In vitro experiments have further revealed that IRF1 may regulate the level of MDA and GSH via PON1. As a transcription factor, IRF1 can bind to DNA and interact with the DNA regions of downstream genes. This binding can either promote or inhibit the transcription process, thereby exerting regulatory effects on gene expression. To investigate the mechanism by which IRF1 regulates PON1, we conducted further ChIP-qPCR experiments, which showed that IRF-1 could bind to PON1 directly. Therefore, it is reasonable to speculate that IRF1 may be involved in the development of pulmonary fibrosis by directly regulating the expression of PON1 and influencing lipid oxidation to regulate ferroptosis. the relationship between PON1 and ferroptosis needs further study. PON1 and GPX4 are both enzymes with antioxidant functions, playing significant roles in protecting the body from oxidative stress. PON1 activity levels were inversely correlated with multiple indices of oxidative stress (oxidized fatty acids), while GPX4 helps to reduce the toxicity of lipid peroxides and maintain the integrity of the cell membrane. Furthermore, GPX4 plays a central role in regulating ferroptosis, a form of cell death driven by iron-dependent lipid peroxidation. GPX4 inhibits the occurrence of ferroptosis through its antioxidant activity. In summary, while PON1 and GPX4 may act independently in antioxidant pathways, they together form part of the cell's antioxidant defense system, helping to protect cells from oxidative damage and death. In our study, we found that the knockout of IRF1 affected the levels of MDA, GSH, and GPX4, as well as the expression of PON1. After intervening with PON1 in vitro, the expression of GPX4 changed accordingly. This could be an indirect effect within this complex antioxidant system, and more research will be needed in the future to confirm this.

This study has several limitations. Firstly, the functional validation of PON1 was not performed in vivo. Investigating the effects of PON1 through its knockout or overexpression, followed by the assessment of ferroptosis and pulmonary fibrosis, could provide pivotal insights into the role and underlying mechanisms of PON1 in the process of lung fibrosis. Secondly, this study utilized a global IRF1 knockout mouse model, but it would be more compelling if conditional knockout mice targeting specific epithelial cells were used. Lastly, the relationship between PON1 and lipid peroxidation was not extensively investigated in this study, as previous research had already established the role of PON1 in regulating lipid oxidation.

Conclusion

We have demonstrated a key role for IRF1 in the pathogenesis of lung fibrosis. Our data have revealed a potential novel link between IRF1 and PON1 in the progression of lung fibrosis.