Attenuated FTO induces necroptosis of alveolar epithelium via the m6A/CYP1B1/ROS/MLKL axis to promote the aggravation of pulmonary emphysema Authors: Mindan Wu, Wenjing Wang, Zemin Chen, Yuanlan Yang, Jiawei Yan, Yachao Cui, Peiyu Huang, Feng Zhang, Wei Tu, Wensheng Chen, Ruiting Sun, Yumin Zhou, Leqing Zhu, Pixin Ran Published: 2026 ABSTRACT AbstractEmphysema is a major phenotype of chronic obstructive pulmonary disease (COPD) and reflects sustained oxidative injury and defective alveolar repair in the distal lung. While cigarette smoke is the dominant environmental trigger, endogenous epithelial programs that restrain reactive oxygen species (ROS) and preserve alveolar integrity remain poorly defined in the adult lung parenchyma. Here we identify that fat mass and obesity-associated protein (FTO) as a redox-responsive regulator in the alveolar epithelium. FTO is predominantly expressed in alveolar type II (AT2) epithelial cells and is markedly reduced in both the epithelium of human COPD patients and the lungs of cigarette smoke-exposed mice. This reduction correlated with worse pulmonary function and was inversely associated with cumulative smoking burden in human. Global attenuation of FTO expression resulted in spontaneous emphysema and impaired lung mechanics in mice. Mechanistically, FTO deficiency promotes cytochrome P450 1B1 (CYP1B1) expression via increased m6A modification and mRNA stabilization, thereby enhancing ROS accumulation and activating the necroptotic effector mixed lineage kinase domain-like protein (MLKL). Genetic silencing of CYP1B1 or antioxidant treatment with N-acetylcysteine (NAC) attenuated ROS and necroptosis. Moreover, conditional deletion of FTO in AT2 cells exacerbated oxidative stress, phosphorylated-MLKL/MLKL levels, and emphysematous injury in mice, indicating spontaneous emphysema driven by necroptosis—effects that were further aggravated in a chronic cigarette smoke-induced COPD model. NAC treatment effectively alleviated the COPD-related deterioration caused by FTO deficiency in AT2 cells. Together, these findings define an alveolar epithelial FTO-m6A-CYP1B1-ROS-MLKL axis that links redox dysregulation to necroptotic cell death and alveolar destruction, and they nominate FTO as a tractable node for redox-targeted intervention in COPD-related emphysema.Keywords: FTO, Alveolar epithelium, Necroptosis, Emphysema, COPD, Oxidative stress Introduction Emphysema is a clinically consequential phenotype of chronic obstructive pulmonary disease (COPD) and contributes substantially to morbidity and mortality worldwide [1]. It is characterized by irreversible destruction of alveolar walls, persistent enlargement of distal airspaces, and progressive loss of gas-exchange surface area. These structural changes promote lung hyperinflation, typically reflected by increases in functional residual capacity (FRC) and total lung capacity (TLC), and they underlie the persistent airflow limitation that defines COPD. [2]. Although COPD encompasses heterogeneous endotypes and can be broadly stratified into emphysematous and non-emphysematous phenotypes, greater emphysema burden is frequently associated with worse lung function, longer disease duration, increased exacerbation risk, and higher mortality [[3], [4], [5]]. Thus, identifying endogenous mechanisms that preserve alveolar integrity remains a central challenge. Alveolar epithelial injury and defective regeneration are core drivers of emphysema. Alveolar type II epithelial cells (AT2s) maintain surfactant homeostasis and serve as progenitors that self-renew and differentiate into type I cells to restore the alveolar barrier after injury. [[6], [7], [8]]. Chronic cigarette smoke (CS) exposure imposes sustained oxidative pressure on this niche by generating excessive reactive oxygen species (ROS) and by perturbing antioxidant defenses. ROS not only damages lipids, proteins, and DNA, but also rewires stress signaling and inflammatory cascades that further amplify injury. In this context, AT2s undergo maladaptive fates—including apoptosis, necroptosis, and senescence—thereby eroding regenerative capacity and accelerating parenchymal destruction. [[9], [10], [11]]. Accordingly, redox imbalance within AT2s is increasingly viewed as a determinant of emphysema progression and severity, yet the intrinsic molecular regulators that maintain ROS homeostasis in these cells remain incompletely defined. Despite advances in anti-inflammatory and bronchodilator therapies, no available intervention halts or reverses emphysematous tissue loss. Multiple efforts have therefore focused on redox-centered strategies, including antioxidant supplementation, enhancement of endogenous detoxification programs, and inhibition of regulated cell-death pathways that are triggered by oxidative stress [12]. However, clinical translation has been limited, in part because the upstream epithelial nodes that couple smoke-induced oxidant stress to irreversible alveolar injury remain unclear. Defining tissue-intrinsic regulators that buffer oxidative stress and preserve epithelial integrity could enable mechanism-based therapeutic targeting and rational patient stratification. Fat mass and obesity-associated protein (FTO) is a conserved Fe(II)/2-oxoglutarate-dependent dioxygenase [13], best known for its role in RNA N6-methyladenosine (m6A) demethylation and for genetic associations with body-weight traits [14,15]. Importantly, the nomenclature and epidemiologic context of FTO can obscure emerging biology that is not directly related to adiposity or to systemic energy balance. Recent studies in diverse cell types suggest that FTO participates in cellular stress adaptation, including modulation of oxidative signaling and survival responses under ROS challenge, in some settings independently of metabolic phenotypes or global m6A effects [[16], [17], [18]]. Consistent with this view, FTO deficiency has been linked to increased ROS accumulation and heightened susceptibility to oxidant-mediated injury in non-adipose tissues. Whether FTO functions as an epithelial redox regulator in the lung, and how it might intersect with smoke-induced injury programs in AT2s, remains unknown. A prominent redox interface between environmental toxicants and epithelial injury is the cytochrome P450 (CYP) system. CYP1 enzymes are inducible by CS constituents and can generate ROS during xenobiotic turnover [[19], [20], [21]]. Excessive CYP activity may translate persistent exposure into sustained oxidative burden, promoting lipid peroxidation, mitochondrial dysfunction, and regulated cell death [22,23]. How AT2s restrain these redox-amplifying programs during chronic smoke stress remains largely unclear. Here we demonstrate that FTO is predominantly expressed in AT2 cells and is markedly downregulated in both the epithelium of COPD patients and the lungs of CS-exposed mice. This reduction correlated with worse pulmonary function and was inversely associated with cumulative smoking burden in human. Global attenuation of FTO expression resulted in spontaneous emphysema and impaired lung mechanics. Mechanistically, FTO deficiency promotes CYP1B1 expression via increased m6A modification and mRNA stabilization, thereby enhancing ROS accumulation and activating the necroptotic effector mixed lineage kinase domain-like protein (MLKL). Moreover, conditional deletion of FTO in AT2 cells exacerbated oxidative stress, phosphorylated-MLKL/MLKL levels, and emphysematous injury in mice, indicating spontaneous emphysema driven by necroptosis—effects that were further aggravated in a chronic CS-induced COPD model. NAC treatment effectively alleviated the COPD-related deterioration caused by FTO deficiency in AT2 cells. Together, these findings define an alveolar epithelial FTO-m6A-CYP1B1-ROS-MLKL axis that links redox dysregulation to necroptotic cell death and alveolar destruction, and they nominate FTO as a tractable node for redox-targeted intervention in COPD-related emphysema. Results FTO expression is downregulated in COPD and correlates with deterioration of spirometry To assess whether FTO is involved in COPD progression, we first interrogated publicly available human transcriptomic datasets from the Gene Expression Omnibus (GEO). In airway/lung epithelial samples, FTO mRNA abundance was reduced in healthy smokers (n = 111) relative to healthy nonsmokers (n = 72) (GSE63127; Fig. 1A). This reduction was even more evident when comparing COPD patients (n = 22) with a pooled control group comprising healthy smokers and nonsmokers (n = 135) (GSE11784; Fig. 1B). In a separate smoker cohort with whole-blood RNA profiles (GSE76705), subjects meeting spirometric criteria for COPD (FEV1/FVC <70%, n = 143) displayed significantly lower normalized FTO expression than those with preserved lung function (FEV1/FVC ≥70%, n = 86) (Fig. 1C). Within this cohort, Spearman analyses further indicated that lower blood FTO levels were associated with worse pulmonary function, including decreased FEV1/FVC and reduced FEV1%Pred. Notably, FTO expression also inversely correlated with cumulative smoking burden (pack-years) (Fig. 1D), implicating smoking exposure as a potential contributor to FTO suppression. FTO expression is downregulated in COPD and correlates with disease severity. (A–B) Relative normalized mRNA expression of FTO in lung epithelium from two GEO database: (A) GSE63127: healthy smokers (n = 111) versus healthy nonsmokers (n = 72); (B) GSE11784: COPD patients (n = 22) versus healthy smokers and nonsmokers (n = 135). (C) The normalized FTO mRNA expression levels in whole blood from the GSE76705 (all smokers with available spirometry data) were stratified into two groups based on the COPD diagnostic criterion (FEV1/FVC <70%; n = 143) versus (FEV1/FVC ≥70%; n = 86). (D) Spearman correlation analysis was performed between whole-blood normalized FTO mRNA expression levels and the following parameters: spirometry indices (FEV1/FVC and FEV1%Pred) and cumulative smoking exposure (pack-years) from the GSE76705. (E) Evaluation of Fto mRNA (by qRT-PCR) and (F) FTO protein (by Western blot) expression in lung tissues of a cigarette smoke-induced COPD mouse model (n = 6/group). Data are presented as mean ± SEM. ∗∗, p < 0.01; ∗∗∗∗, p < 0.0001. Abbreviation: FEV1/FVC, ratio of forced expiratory volume in 1 s to forced vital capacity; FEV1%Pred, forced expiratory volume in 1 s as percentage of predicted value. We next validated these observations in a CS–exposure mouse model. In line with the human findings, CS exposure markedly decreased both Fto mRNA and FTO protein levels in lung tissue compared with air-exposed controls (Fig. 1E and F). Collectively, these results support a conserved downregulation of FTO during COPD pathogenesis across species. Global attenuation of FTO expression induces emphysematous changes and spirometric impairment in mice To define the in vivo impact of FTO reduction, we established a global FTO-deficient mouse (Fig. 2A). Three distinct genotypes were identified by genotyping: wild-type (FTO+/+, WT), heterozygous knockout (FTO+/−, referred to as FTO-KD), and homozygous knockout (FTO−/−, referred to as FTO-KO) (Fig. S1A). FTO-KD mice displayed comparable body size and body weight to WT littermates (Fig. S1B). Consistent with previous reports [24,25], we observed that FTO-KO mice exhibited significantly lower body weight and smaller body size compared with WT littermates (Fig. S1B). We propose that heterozygous FTO deletion (50% dose reduction) may not disrupt body weight regulation due to compensatory mechanisms or a threshold effect, whereby only homozygous loss results in a detectable weight phenotype. Global attenuated FTO expression could cause emphysema and deterioration of spirometry in mice. (A) Schematic map of the generation of FTO global knockout mice, provided by Cyagen Biosciences. (B) Analysis of mouse genotyping (tail DNA; PCR) and FTO protein expression (lung tissue; Western blot) was performed. (C) The efficiency of FTO knockdown in lung tissues was confirmed by quantification of its mRNA and protein levels using qRT-PCR and Western blot, respectively. (D) Body weights of mice. (E–F) Lung morphology and airspace quantification in WT (n = 6) and FTO-KD mice (n = 4). (E) Representative H&E-stained lung sections showing airspace changes. Scale bars: 1 mm, 250 μm, and 100 μm (as indicated). (F) Quantification of MLI. (G) Pulmonary function test was assessed in WT and FTO-KD mice, including the following parameters: FRC, FVC, FEV20/FVC, Cchord and Cdyn. (H) Spearman correlation analysis was performed between quantification of FTO protein levels and FRC. (I) Experimental outline and representative H&E-stained lung sections showing airspace changes. Scale bars: 250 μm. (J) Pulmonary function parameters, (K) body weight and (L) MLI were assessed in WT and FTO-KD mice subjected to NS or PPE treatment respectively (n = 5/group). Data are presented as mean ± SEM. ns, not statistically significant; ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; ∗∗∗∗, p < 0.0001. Abbreviation: WT, wild-type; FTO-KD, FTO-knockdown; H&E, hematoxylin and eosin; MLI, mean linear intercept; FRC, functional residual capacity; FVC, forced vital capacity; FEV20/FVC, ratio of forced expiratory volume at 20 ms to FVC; Cchord, chord compliance; Cdyn, dynamic compliance; FEV50/FVC, ratio of forced expiratory volume at 50 ms to FVC; FEV100/FVC, ratio of forced expiratory volume at 100 ms to FVC; i.t.: intratracheally; NS, normal saline; PPE, porcine pancreatic elastase. However, FTO-KO mice suffered from high perinatal lethality and reduced fertility, which precluded their use in subsequent experiments [25]. Therefore, all subsequent experiments were conducted in FTO-KD mice. Decreased FTO was verified at multiple levels, including PCR-based genotyping and quantification of lung Fto transcripts and FTO protein by qRT-PCR and immunoblotting, respectively (Fig. 2B and C). At baseline, FTO-KD mice displayed body weights comparable to WT littermates (Fig. 2D). Histological assessment of lung architecture revealed overt emphysema-like remodeling in FTO-KD mice, with enlarged distal airspaces evident on hematoxylin and eosin (H&E) staining (Fig. 2E). Consistently, morphometric quantification demonstrated a significant elevation in mean linear intercept (MLI), supporting increased airspace enlargement and alveolar wall loss relative to WT controls (Fig. 2F). Pulmonary function measurements further indicated physiological impairment in FTO-KD mice. Specifically, these animals showed increased functional residual capacity (FRC), elevated forced vital capacity (FVC), and a lower ratio of forced expiratory volume in 20 ms to FVC (FEV20/FVC ratio), consistent with airflow limitation. Mechanical parameters were also altered, including increased chord compliance (Cchord) and reduced dynamic compliance (Cdyn) (Fig. 2G). In addition, Spearman analysis identified a significant inverse relationship between lung FTO protein abundance and FRC, linking lower FTO levels to more prominent hyperinflation (Fig. 2H). We next asked whether reduced FTO predisposes the lung to emphysematous injury by challenging WT and FTO-KD mice with porcine pancreatic elastase (PPE). Following PPE exposure, FTO-KD mice exhibited more severe airspace enlargement together with greater deterioration in lung function than PPE-treated WT counterparts (Fig. 2I and J). Body weight remained similar between genotypes under PPE challenge (Fig. 2K). Morphometric analysis reinforced these findings, showing a larger PPE-induced increase in MLI in FTO-KD lungs (Fig. 2L). Together, these data indicate that systemic attenuation of FTO is sufficient to drive emphysema-like pathology and functional decline, and further heightens susceptibility to a secondary proteolytic insult. FTO is predominantly expressed in alveolar type II epithelium (AT2) and decreased in a concentration- and time-dependent manner in vitro To map FTO distribution within the lung, we performed immunofluorescence on lung sections from the CS–induced COPD mouse model. FTO signal overlapped predominantly with the AT2 marker pro–surfactant protein C (pro-Sftpc), indicating preferential expression in AT2 cells (Fig. 3A). Quantification revealed that CS exposure significantly reduced overall FTO fluorescence intensity and decreased the fraction of proSftpc+ cells positive for FTO (FTO+/proSftpc+) compared with air-exposed controls (Fig. 3B). FTO is predominantly expressed in alveolar type II epithelium and decreased in a concentration- and time-dependent manner in A549 and MLE12. (A) Representative micrographs of immunofluorescence staining for FTO (red), alveolar type II epithelium (pro-Sftpc, yellow) and nuclei (DAPI, blue) in lung sections from WT mice in a cigarette smoke-induced COPD mouse model (n = 4/group). Scale bars: 250 μm, and 50 μm (as indicated). (B) Quantitative analysis of FTO fluorescence intensity and the percentage of proSftpc-positive cells that co-expressed FTO (FTO+/proSftpc+ double-positive cells) in lung sections. (C–F) FTO expression is suppressed by CSE in a concentration- and time-dependent manner in human (A549) and murine (MLE12) alveolar epithelial cells. A549 cells were treated with increasing CSE concentrations for 24 h (C) or with 1.5% CSE for indicated times (D). MLE-12 cells were treated with increasing CSE concentrations for 12 h (E) or with 0.6% CSE for indicated times (F). Analysis of protein and mRNA expression levels was performed by Western blot and qRT-PCR, respectively. Data are presented as the mean ± SEM of three independent experiments. ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001. Abbreviation: WT, wild-type; CSE, cigarette smoke extract. We next tested whether CS directly suppresses FTO in vitro by treating human (A549) and murine (MLE-12) alveolar epithelial cell lines with cigarette smoke extract (CSE). In A549 cells, FTO expression declined in a dose-dependent manner after 24 h CSE exposure (Fig. 3C), and was similarly reduced over time at a fixed CSE concentration (Fig. 3D). Consistent results were obtained in MLE-12 cells, which proliferate approximately twice as fast as A549 cells: both FTO protein and Fto mRNA decreased with escalating CSE doses over 12 h (Fig. 3E) and with extended exposure at a constant CSE concentration (Fig. 3F). Together, these data show that CSE represses FTO expression in alveolar epithelial cells in both concentration- and time-dependent fashions, closely recapitulating the in vivo pattern observed after CS exposure. FTO downregulation promotes oxidative stress by enhancing CYP1B1 mRNA stability in an m6A-dependent manner To elucidate the molecular programs engaged by FTO deficiency in the lung, we conducted KEGG pathway enrichment on RNA-seq profiles from lungs of FTO-KD and WT mice. The ‘Chemical carcinogenesis – reactive oxygen species’ pathway was significantly elevated in FTO-KD lungs, consistent with an intensified oxidative-stress signature (Fig. 4A). We next assessed the epithelial-intrinsic contribution of FTO under CSE challenge using an in vitro model. A549 cells were transfected with negative control (NC) or FTO-targeting siRNA for 24 h and then exposed to 1.5% CSE for an additional 24 h. Efficient silencing of FTO was verified at protein levels (Fig. 4B). KEGG analysis of these cells revealed the same enrichment for the ‘Chemical carcinogenesis – reactive oxygen species’ pathway following FTO knockdown as observed in FTO-KD lungs (Fig. 4C), mirroring the in vivo findings. Apoptosis-related pathways were also significantly enriched in FTO-silenced A549 cells (Fig. 4C). FTO downregulation promotes oxidative stress by enhancing CYP1B1 mRNA stability in an m6A-dependent manner. (A) KEGG enrichment analysis in lung tissues from FTO-KD vs. WT mice. (B) FTO knockdown efficiency was confirmed in human A549 alveolar epithelial cells at protein levels. (C) KEGG enrichment analysis in CSE-treated A549 cells with or without FTO knockdown. (D) CYP1B1 were identified as the overlapping differentially expressed genes in a Venn diagram analysis of four CYP family gene sets, derived from the comparisons: FTOsi vs NCsi, NCsi_CSE vs NCsi, FTOsi_CSE vs NCsi_CSE, and COPD vs Healthy (GSE11784). (E–F) MeRIP-seq analysis revealed that following FTO knockdown in A549 cells, m6A modifications were predominantly enriched in the 3′UTR (49.1%) and CDS (33.2%) (E), with the consensus motif “RRACH” (F). (G) Nine-quadrant plot integrating MeRIP-Seq and RNA-Seq data. Conjoint analysis identifies CYP1B1 in the Hyper-up quadrant (significantly increased m6A modification and RNA expression). (H) IGV visualization of MeRIP-Seq (IP, red) and RNA-Seq (input, blue) data. m6A peaks on Cyp1b1 mRNA were enriched in the 3′UTR and CDS, and peak abundance in the FTO-silencing group was increased along with upregulated transcript levels. (I) Knockdown of FTO significantly prolonged the half-life of Cyp1b1 mRNA as determined by actinomycin D transcription blockade assay. (J) The protein level of CYP1B1 was increased upon FTO knockdown as determined by Western blot. (K) Spearman correlation analysis showed that FTO expression was negatively correlated with CYP1B1 expression in lung epithelium (GSE11784, GSE63127). Data are presented as the mean ± SEM of three independent experiments. ∗, p < 0.05; ∗∗∗, p < 0.001; ∗∗∗∗, p < 0.0001. Abbreviation: KEGG, Kyoto Encyclopedia of Genes and Genomes; WT, wild-type; FTO-KD, FTO-knockdown; CYP, Cytochrome P450; CSE, cigarette smoke extract; NCsi_CSE or FTOsi _CSE, A549 cells were transfected with NC-siRNA or FTO-siRNA and then treated with 1.5% CSE for 24 h. NCsi or FTOsi, A549 cells were transfected with NC-siRNA or FTO-siRNA and without CSE treatment; MeRIP-Seq, Methylated RNA Immunoprecipitation sequencing; 5′UTR, 5′-Untranslated Region; 3′UTR, 3′-Untranslated Region; CDS, Coding sequence; IGV, Integrative Genomics Viewer. To pinpoint candidate drivers of the oxidative response, guided by the gene sets from the ‘Chemical carcinogenesis – reactive oxygen species’ KEGG pathway in A549, we focused on the CYP family, a well-established enzymatic source of ROS. We therefore performed a Venn intersection of differentially expressed genes across four comparisons: (1) CSE-treated versus control A549 cells (NCsi_CSE vs NCsi), (2) FTO-silencing plus CSE versus CSE alone in A549 cells (FTOsi_CSE vs NCsi_CSE), (3) FTO-silencing versus control A549 cells (FTOsi vs NCsi) and (4) COPD versus healthy human airway/lung epithelium (GSE11784). This approach nominated CYP1B1 as shared upregulated genes across all four settings (Fig. 4D). The quantitative results of differential expression analyses demonstrated that CYP1B1 induction was consistently accompanied by reduced FTO expression in each comparison (Table S1). We further validated all CYP450 family members included in the intersection of the four datasets using qRT-PCR. The results showed that CYP1A1 and CYP1B1 were significantly upregulated under CSE exposure, and this upregulation was more pronounced after FTO knockdown (Fig. S2A). To exclude potential contributions from other sources of ROS, we conducted a source-tracing analysis by examining the FPKM (Fragments Per Kilobase of transcript per Million mapped reads) values from our transcriptome sequencing data (A549 cells). It should be noted that FPKM is suitable for comparing relative expression levels of the same gene across different experimental groups, and its absolute value does not directly reflect absolute RNA abundance. We evaluated the contribution of the following families to FTO knockdown-induced ROS production: the CYP450 family, mitochondrial respiratory chain complexes, the NADPH oxidase (NOX) family, xanthine dehydrogenase (XDH), and the lipoxygenase family (ALOX). The results (Fig. S2B) demonstrated that FTO knockdown-induced ROS production was primarily derived from the CYP450 enzyme family (e.g., CYP1A1, CYP1B1), with a minor contribution from the mitochondrial respiratory chain complex (NDUFS1). No significant contribution was observed from the NOX family, XDH, or ALOX family. A volcano plot of the FTO-silenced plus CSE versus CSE-only comparison additionally highlighted CYP1A1 and CYP1B1 among the upregulated genes (Fig. S2C). However, given that FTO functions as an m6A demethylase and may regulate target genes under baseline conditions, we performed MeRIP-Seq (Methylated RNA Immunoprecipitation sequencing) on A549 cells transfected with FTO siRNA to comprehensively detect changes in m6A modification and gene expression profiles following FTO knockdown. The results showed that in the transcripts of A549 cells with FTO knockdown, m6A modification changes were mainly enriched in the 3′ untranslated region (UTR) (49.1%) and coding region (CDS) (33.2%) (Fig. 4E). The m6A methylation motifs shared by the two samples are consistent with the known m6A motif “RRACH” (Fig. 4F). The MeRIP-seq results identified CYP1B1 as a top 50 differentially expressed gene carrying a clear m6A mark (Fig. S2D). We performed integrated analysis of MeRIP-seq and RNA-seq and found that CYP1B1 fell into the “Hyper-up” quadrant (Fig. 4G), indicating that increased m6A modification is accompanied by increased mRNA abundance following FTO knockdown. IGV (Integrative Genomics Viewer) visualization further revealed two distinct m6A peaks within the CYP1B1 transcript, one in the CDS region and one in the 3′UTR (Fig. 4H), aligning with the overall m6A distribution pattern in A549 cells. To assess the functional consequence of this modification, we performed actinomycin D chase assays, which showed that FTO knockdown significantly prolonged the half-life of CYP1B1 mRNA (Fig. 4I), supporting altered mRNA stability as an important post-transcriptional mechanism contributing to CYP1B1 upregulation. Consistent with this, CYP1B1 protein expression was also increased after FTO knockdown (Fig. 4J). We further analyzed two independent human epithelial transcriptomic datasets (GSE11784 and GSE63127), which showed that FTO expression was negatively correlated with CYP1B1 expression (Fig. 4K). Together, these results indicate that CYP1B1 is a direct downstream target of FTO-mediated m6A modification. Although CYP1A1 also showed induction upon CSE exposure and the upregulation was more pronounced after FTO knockdown, neither MeRIP-seq, IGV analysis (Fig. S3A), nor mRNA stability experiments (Fig. S3B) supported CYP1A1 as a direct m6A-dependent target of FTO. FTO knockdown results in elevated expression CYP1B1, along with increased ROS accumulation and necroptosis in alveolar epithelial cells In vitro, silencing FTO in A549 cells under CSE exposure significantly increased the mRNA abundance of Cyp1b1 and CYP1B1 proteins relative to NC-transfected controls (Fig. 5A and B). Concomitant with CYP1B1 induction, intracellular ROS and mitochondrial ROS (mtROS) levels rose sharply in FTO-depleted cells, as quantified by 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA)–based (Fig. 5C) and MitoSOX-based (Fig. 5D) flow cytometry. Functionally, FTO knockdown also increased susceptibility to CSE-induced cell death, evidenced by a higher proportion of Annexin V-positive cells (Fig. 5E). Notably, immunoblotting revealed augmented phosphorylation of MLKL (p-MLKL), a terminal effector of necroptosis, in FTO-silenced cells (Fig. 5F), supporting activation of the necroptotic program. FTO knockdown in A549 cells results in elevated expression of CYP1B1, along with increased ROS/mtROS accumulation and necroptosis. A549 cells were transfected with NC-siRNA or FTO-siRNA for 24 h, followed by stimulation with 1.5% CSE for an additional 24 h. (A) Relative mRNA levels of Fto and Cyp1b1 were quantified by qRT-PCR. (B) The protein level of CYP1B1 was determined by Western blot. (C) Flow cytometry analysis of total intracellular ROS. Representative DCFH-FITC fluorescence histograms were presented; results were quantified as relative mean fluorescence intensity. (D) Flow cytometry analysis of mtROS. Representative MitoSOX-PE fluorescence histograms were presented; results were quantified as relative mean fluorescence intensity. (E) Flow cytometry analysis of apoptosis. Representative Annexin V-FITC/PI dot plots were presented; the total apoptosis rate represented the combined percentage of early (lower right) and late (upper right) apoptotic cells. (F) Analysis of phosphorylation and total expression levels of the key necroptosis mediator MLKL was performed by Western blot. Data are presented as the mean ± SEM of three independent experiments. ns, not statistically significant; ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; ∗∗∗∗, p < 0.0001. Abbreviation: CYP, Cytochrome P450; CSE, cigarette smoke extract; ROS, Reactive oxygen species; mtROS, mitochondrial ROS; MLKL, Mixed lineage kinase domain-like protein; p-MLKL, phospho-MLKL. Furthermore, we extended our analysis to human bronchial epithelial (HBE) cells, which are also highly relevant to COPD pathophysiology. After confirming efficient FTO knockdown (Fig. S4A), we found that FTO silencing in HBE cells increased CYP1B1 mRNA and protein expression, intracellular ROS/mtROS, p-MLKL, and Annexin V-positive apoptotic cells compared with control following CSE exposure. (Fig. S4B–G). Thus, the HBE experiments recapitulate the major oxidative stress and necroptosis-associated features observed in our AT2-based systems, suggesting that the FTO–CYP1B1 axis is not unique to the A549/AT2 context and may represent a broader epithelial stress-response pathway in COPD. Two independent human epithelial transcriptomic datasets (GSE11784 and GSE63127) revealed that FTO expression was negatively correlated with MLKL expression (Fig. S5A and S5B); CYP1B1 expression was positively correlated with MLKL expression (Fig. S5C and S5D). Although FTO knockdown in A549 cells did not markedly alter total MLKL protein abundance, it consistently affected p-MLKL levels. The negative correlation between FTO and MLKL mRNA in transcriptomic analyses further supports an inverse association between FTO status and necroptotic signaling, although this transcriptional relationship may not necessarily translate into detectable differences in steady-state total MLKL protein in A549 cells at the examined time points. Overall, these clinical correlation data, derived specifically from lung epithelial tissues, provide strong supportive evidence for the FTO-CYP1B1-MLKL-necroptosis axis, demonstrating that reduced FTO expression is associated with increased CYP1B1 and MLKL expression in human COPD lung epithelium. Given that CYP1A1 is not a direct m6A-dependent target of FTO, we performed a preliminary investigation into the regulatory mechanisms of CYP1A1. Multiple studies [[26], [27], [28], [29]] have shown that CYP1A1 is primarily induced by polycyclic aromatic hydrocarbons (PAHs) in CSE [30,31] via the aryl hydrocarbon receptor (AhR) pathway. As shown in Fig. S6A–S6D, mRNA and protein levels of both CYP1A1 and AhR were elevated upon CSE stimulation in FTO-silenced A549 cells compared with control, supporting that CYP1A1 induction under CSE exposure is likely mediated through the canonical AhR pathway, which is consistent with previous studies [[26], [27], [28], [29]]. However, MeRIP-seq, IGV analysis (Fig. S6E), and mRNA stability experiments (Fig. S6F) also indicated that AhR was not a direct m6A-dependent target of FTO, suggesting that its regulation by FTO may also be indirect. CYP1B1 knockdown and NAC treatment rescue FTO deficiency-induced ROS/mtROS accumulation and necroptosis in A549 cells To further clarify the contribution of CYP1B1 and ROS to FTO-dependent epithelial injury, we performed rescue experiments under CSE challenge, as oxidative stress and necroptosis were negligible without smoke stimulation. We first asked whether CYP1B1 induction is required for the detrimental effects observed after FTO loss. A549 cells were transfected with a CYP1B1-targeting siRNA, which efficiently reduced CYP1B1 expression at protein levels (Fig. 6A). Cells were then transfected with siRNAs against FTO, CYP1B1, or both for 24 h, followed by exposure to 1.5% CSE for an additional 24 h. As expected, FTO silencing increased CYP1B1 expression, whereas CYP1B1 knockdown did not affect FTO abundance (Fig. 6B), supporting CYP1B1 as a downstream effector of FTO. Importantly, simultaneous depletion of CYP1B1 mitigated the cellular consequences of FTO loss, including significantly reducing ROS/mtROS accumulation, apoptotic burden, and phosphorylation of the necroptosis executor MLKL (Fig. 6C–F). CYP1B1 knockdown and NAC treatment rescue FTO deficiency-induced ROS/mtROS accumulation and necroptosis in A549 cells. A549 cells were transfected with siRNAs targeting FTO, CYP1B1, or both for 24 h, followed by stimulation with 1.5% CSE for an additional 24 h. (A) CYP1B1 knockdown efficiency was confirmed at protein levels. (B)Fto and Cyp1b1 mRNA expression were assessed by qRT-PCR. Knockdown of CYP1B1 attenuated ROS accumulation (C), mtROS accumulation(D), phospho-MLKL levels (E) and apoptosis (F), thereby rescuing the phenotypic effects of FTO knockdown. A549 cells were transfected with FTO-siRNA for 24 h, followed by a 2-h pretreatment with 5 mM NAC prior to stimulation with 1.5% CSE for an additional 24 h. Treatment with NAC eliminated intracellular ROS accumulation (G), mtROS accumulation(H), reduced phospho-MLKL levels (I) and apoptosis (J), and thereby rescued the effects of FTO knockdown. Data are presented as the mean ± SEM of three independent experiments. ns, not statistically significant; ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; ∗∗∗∗, p < 0.0001. Abbreviation: CYP, Cytochrome P450; CSE, cigarette smoke extract; ROS, Reactive oxygen species; mtROS, mitochondrial ROS; MLKL, Mixed lineage kinase domain-like protein; NAC, N-acetylcysteine. We next directly tested whether ROS is a key mediator in this cascade. FTO-silenced cells were pretreated with the antioxidant N-acetylcysteine (NAC), a potent ROS scavenger, prior to CSE exposure. NAC effectively blunted intracellular ROS/mtROS, decreased apoptosis, and suppressed MLKL phosphorylation, thereby rescuing the phenotypes elicited by FTO knockdown (Fig. 6G–J). Collectively, these results demonstrate that CYP1B1 upregulation and the ensuing ROS accumulation are essential downstream drivers of oxidative stress and necroptotic signaling triggered by FTO deficiency in alveolar epithelial cells. Alveolar epithelial-specific FTO deficiency exacerbates lung dysfunction, oxidative stress, and inflammation in a chronic CS-induced COPD mouse model To dissect the cell type–specific contribution of FTO to COPD pathogenesis, we generated an AT2-restricted conditional knockout model by crossing FTOflox/flox mice with Sftpc-CreERT2 (hereafter AT2-Cre) mice (Fig. 7A and Fig. S7). After tamoxifen administration, Sftpc-CreERT2-FTOflox/flox mice (designated AT2-FTO△/△) and AT2-Cre controls were subjected to 24 weeks of chronic CS exposure (Fig. 7A). Immunofluorescence of lung sections from tamoxifen-treated mice maintained in room air for 24 weeks verified efficient and selective loss of FTO within proSftpc+ AT2 cells in AT2-FTO△/△ animals compared with AT2-Cre controls (Fig. 7B). Consistently, immunoblotting of whole-lung lysates showed a marked reduction in FTO protein in AT2-FTO△/△mice (Fig. 7C). Alveolar epithelial FTO deficiency leads to worsened lung function, exacerbated oxidative stress, and increased inflammation in experimental COPD. (A) Generation of the alveolar epithelium-specific FTO knockdown mouse model (Sftpc-CreERT2-FTOflox/flox) and a schematic of the experimental timeline, incorporating the tamoxifen-induced conditional knockdown phase followed by the cigarette smoke exposure phase. (B) Representative immunofluorescence images of lung sections from control (AT2-Cre) and alveolar-epithelial-specific FTO-knockout (AT2-FTO△/△) mice, all treated with tamoxifen and maintained under normoxic (air) conditions for 24 weeks. Sections were stained for FTO (red), the AT2 cell marker pro-Sftpc (green), and nuclei (DAPI, blue). Scale bars: 100 μm. (C) Western blot analysis confirmed reduced FTO protein expression in lung tissues of AT2-FTO△/△ mice compared to controls. (n = 4/group). (D) Δ Body weight was defined as the change in body weight from the initiation of tamoxifen treatment to the end of the experimental model. Δ Body weight, FRC (a pulmonary function parameter) (E), and GSSG (reflecting lung oxidative status) (F) were measured in all experimental groups. (G and H) The expression of CYP1B1 was assessed at both the mRNA and protein levels. (I) The numbers and proportions of total live cells, macrophages, dendritic cells, and neutrophils in BALF were quantified using multicolor flow cytometry. (J) The levels of the inflammatory cytokines IL 6 and CCL2 in BALF were detected by ELISA. (n = 5/group). Data are presented as the mean ± SEM. ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; ∗∗∗∗, p < 0.0001. Abbreviation: FRC, functional residual capacity; GSSG, glutathione disulfide; CYP, Cytochrome P450; BALF, bronchoalveolar lavage fluid; IL, interleukin; CCL2, C–C motif chemokine ligand 2; ELISA, enzyme-linked immunosorbent assay; i.p., intraperitoneally; QOD, every other day. Under chronic CS exposure, AT2-FTO△/△ mice displayed a larger decrease in body weight (Δ body weight) than CS-exposed AT2-Cre controls (Fig. 7D). Lung function measurements further revealed a significant increase in FRC in CS-challenged AT2-FTO△/△ mice (Fig. 7E), indicative of more severe air trapping. These physiological changes were accompanied by an augmented oxidative burden, evidenced by elevated oxidized glutathione (GSSG) in lung tissue (Fig. 7F). It was noteworthy that GSSG levels were elevated even in the absence of smoking stimulation. In parallel, both transcript and protein levels of CYP1B1 were substantially increased in CS-exposed AT2-FTO△/△ lungs (Fig. 7G and H). Because COPD progression involves chronic inflammation in addition to oxidative stress, we next profiled inflammatory cell recruitment and mediators in bronchoalveolar lavage fluid (BALF). Flow cytometry demonstrated higher total live BALF cell counts in CS-exposed AT2-FTO△/△ mice relative to CS-exposed AT2-Cre controls. This increase was characterized by expanded dendritic cell and neutrophil numbers, an elevated neutrophil fraction, and a reduced macrophage proportion (Fig. 7I). In agreement with this inflammatory shift, BALF concentrations of interleukin (IL)-6 and C–C motif chemokine ligand 2 (CCL2) were significantly higher in CS-exposed AT2-FTO△/△ mice (Fig. 7J). Together, these data show that AT2-intrinsic loss of FTO aggravates CS-driven functional deterioration, oxidative stress, CYP1B1 induction, and inflammatory cell influx, underscoring a critical role for alveolar epithelial FTO in restraining COPD progression. Alveolar epithelial-specific FTO deficiency promotes necroptosis and aggravates emphysema in chronic CS-induced COPD To evaluate whether loss of FTO in the alveolar epithelium intensifies cell death and parenchymal injury, we first assessed necroptotic signaling. Immunoblotting showed that CS exposure elicited a significant increase in both phosphorylated MLKL and total MLKL protein in AT2-FTO△/△ mice compared with CS-exposed AT2-Cre controls (Fig. 8A), indicating enhanced activation of the necroptosis pathway. In parallel, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining of lung sections revealed a greater number of TUNEL-positive cells in CS-challenged AT2-FTO△/△ mice (Fig. 8B and C), consistent with increased cell death. Alveolar epithelial FTO deficiency promotes necroptosis and exacerbates emphysema in Experimental COPD. (A) Phosphorylation and total expression levels of the key necroptosis mediator MLKL in lung tissues from each experimental group were analyzed by Western blot. (n = 3/group). Detection and quantification of apoptotic cells in lung tissue by TUNEL assay. (B) Representative staining images (scale bars: 50 μm). (C) Statistical analysis of TUNEL-positive cells, presented as individual mouse data (n = 5/group). Morphological analysis of alveolar airspace: representative H&E images (D) and quantification of the MLI (E) in mouse lungs (n = 5/group). Scale bars: 250 μm, and 100 μm (as indicated). Data are presented as the mean ± SEM. ∗, p < 0.05; ∗∗, p < 0.01. Abbreviation: MLKL, Mixed lineage kinase domain-like protein; TUNEL, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling; MLI, mean linear intercept. We next examined whether these death signals translated into worsened structural damage. Histological evaluation demonstrated that AT2-restricted FTO deficiency aggravated emphysematous remodeling after CS exposure: H&E-stained sections from AT2-FTO△/△ lungs displayed more prominent distal airspace enlargement (Fig. 8D). Quantitative morphometry confirmed this phenotype, with MLI significantly higher in AT2-FTO△/△ mice than in CS-exposed control animals (Fig. 8E). It should be noted that p-MLKL/MLKL and MLI were elevated without CS exposure, indicating spontaneous emphysema driven by necroptosis. NAC treatment effectively ameliorated the exacerbation of COPD-related phenotypes induced by FTO deficiency in AT2 cells We next assessed the therapeutic efficacy of NAC in the AT2-specific FTO knockout mouse model of COPD. As showed in Fig. 9A, AT2-FTO△/△ and AT2-Cre mice were first induced with tamoxifen, followed by 12 weeks of chronic CS exposure. After the initial 12-week exposure, the CS-exposed mice were randomly divided into two groups: one group received regular drinking water, while the other group received drinking water supplemented with 50 mM NAC (pH adjusted to 6.5–7.5, with fresh NAC solution replaced every two days) according to previous research [32]. NAC intervention was continued for an additional 12 weeks of CS exposure, totaling 24 weeks of smoke exposure. NAC treatment effectively ameliorated the exacerbation of COPD-related phenotypes induced by FTO deficiency in AT2 cells. (A) The schematic of the experimental timeline: AT2-specific FTO conditional knockout mice (AT2-FTO△/△) and AT2-Cre controls were induced with tamoxifen, exposed to cigarette smoke for 12 weeks, and then randomly assigned to receive either regular drinking water or drinking water containing 50 mM NAC (pH adjusted to 6.5–7.5, refreshed every two days) for the subsequent 12 weeks of chronic smoke exposure. (B–G) NAC treatment decreased FRC (B), reduced GSSG levels (C), alleviated protein levels of the inflammatory cytokines IL-6 (D) and CCL2 (E) in BALF, and decreased total cell counts (F) as well as neutrophil and dendritic cell counts (G) in BALF. (H and I) H&E staining of lung tissues revealed marked alleviation of emphysema morphology (H), which was quantitatively confirmed by a reduced MLI (I). (n = 5/group). Scale bars: 250 μm. Data are presented as the mean ± SEM. Two-way ANOVA with Tukey's correction was used for comparisons between groups. ∗, p < 0.05; ∗∗, p < 0.01; ∗∗∗, p < 0.001; ∗∗∗∗, p < 0.0001. Abbreviation: NAC, N-acetylcysteine; COPD, Chronic obstructive pulmonary disease; AT2, Alveolar type II epithelial cells; FRC, Functional residual capacity; GSSG, Glutathione disulfide; IL-6, Interleukin-6; CCL2, C–C motif chemokine ligand 2; BALF, Bronchoalveolar lavage fluid; H&E, Hematoxylin and eosin; MLI, Mean linear intercept; AT2-Cre, Sftpc-CreERT2; AT2-FTO△/△, Sftpc-CreERT2-FTOflox/flox. The results demonstrated that NAC treatment effectively ameliorated the exacerbation of COPD-related phenotypes induced by FTO deficiency in AT2 cells. Specifically, NAC improved lung function (as reflected by decreased FRC), reduced GSSG levels, alleviated protein levels of the inflammatory cytokines IL-6 and CCL2 in BALF, and decreased total cell counts as well as neutrophil and dendritic cell counts in BALF (Fig. 9B–G). Furthermore, H&E staining of lung tissues revealed marked alleviation of emphysema morphology, which was quantitatively confirmed by a reduced MLI (Fig. 9H and I). These findings provide supportive in vivo evidences that oxidative stress is a key downstream mediator of the FTO-deficiency phenotype and further strengthen the translational relevance of our study. Discussion Our study establishes that diminished FTO expression correlates with worse clinical outcomes in COPD, including worse lung function, and higher smoking index. This was corroborated in vivo, as global FTO deficiency in mice triggered spontaneous emphysema and lung function decline. Mechanistic insights revealed that FTO knockdown in AT2 cells exposed to CSE upregulates CYP1B1 expression via increased m6A modification and mRNA stabilization, resulting in oxidative stress and necroptosis. Importantly, AT2-specific FTO knockout exacerbated emphysema and other pathological changes in a CS-induced COPD model, and these effects were effectively ameliorated by NAC treatment. These results provide direct evidence for the mechanistic role of FTO in COPD and position it as a promising therapeutic target for emphysema. The decline in FTO expression correlates with disease progression in COPD. Specifically, decreased FTO expression was observed in healthy smokers relative to healthy non-smokers, and was further corroborated in comparisons between healthy individuals and COPD patients, aligning with recent reports [17]. As some healthy subjects in the above cohort were non-smokers while all COPD patients had a smoking history, smoking was considered a potential confounder. To address this, we analyzed a cohort comprising only smokers with comparable lung function to further evaluate FTO expression levels. Using the diagnostic threshold for COPD (FEV1/FVC <0.7), the cohort was stratified into two groups. Consistently, FTO expression was significantly lower in the group with poorer lung function. Furthermore, FTO levels correlated positively with lung function parameters (FEV1/FVC, FEV1%Pred) and negatively with cumulative smoking exposure. These findings directly link lower FTO expression to key clinical features of COPD, including impaired lung function and greater smoking burden, underscoring its clinical relevance. Moreover, global FTO knockdown mice spontaneously developed emphysematous changes and compromised lung function even in the absence of external stimuli, providing robust genetic evidence that FTO deficiency plays a causal role in emphysema pathogenesis, rather than representing a mere association. This finding provides direct genetic evidence that FTO contributes to susceptibility in emphysema development, consistent with previous genetic analyses in population cohorts [14,15,33]. Emphysema, particularly in the context of COPD, is characterized by extensive alveolar destruction driven in part by the necroptosis of AT2s. As alveolar progenitor cells, AT2s are required to proliferate and differentiate into AT1s following injury to restore alveolar structure. However, necroptosis of AT2 cells compromises lung regenerative capacity and exacerbates dysfunction in alveolar repair [6,12,34]. This form of programmed cell death leads to the release of damage-associated molecular patterns (DAMPs) such as p-MLKL, initiating a potent inflammatory cascade that accelerates alveolar breakdown and ultimately promotes emphysema progression [[35], [36], [37]]. Genetic inhibition of MLKL has been shown to attenuate CS-induced airway inflammation and emphysema, suggesting that targeted suppression of necroptosis represents a promising therapeutic strategy for COPD [38]. In our study, FTO deficiency in AT2 cells resulted in a significant elevation of p-MLKL, whether observed in vitro using FTO siRNA-transfected A549 cells stimulated with CSE or in vivo through AT2-specific FTO knockout mice exposed to whole CS. The in vivo findings further confirmed that loss of FTO in AT2 cells promotes necroptosis, thereby contributing to exacerbated emphysema in the COPD model. Collectively, our results support a protective role for FTO in attenuating AT2 cell necroptosis, thereby limiting the progression of emphysema. Increased oxidative stress in AT2s represents another key pathogenic mechanism in emphysema [[39], [40], [41]]. Cigarette smoke serves as the critical exogenous oxidant in emphysema, as chronic exposure results in excessive accumulation of ROS in the lungs, thereby driving dysregulated inflammatory and oxidative stress responses [9,42]. Furthermore, cigarette smoke can activate inflammatory cells such as neutrophils, dendritic cells and macrophages, leading to the release of endogenous ROS and a further amplification of oxidative stress [39,41,42]. This disruption leads to a loss of intracellular homeostasis, resulting in AT2 cell dysfunction, senescence and enhanced necroptosis, which collectively accelerate progression of emphysema [9,43,44]. Our study revealed that in FTO-deficient A549 cells, intracellular ROS levels increased even under basal conditions and were further elevated upon CSE stimulation. Correspondingly, in AT2-specific FTO knockout mice, the oxidative metabolite GSSG accumulated in the air-exposed group, and this accumulation was exacerbated by CS exposure. Excessive ROS production drives intracellular signaling that promotes pro-inflammatory mediator expression, thereby potentiating the inflammatory cascade [45]. As expected, FTO deficiency led to elevated IL-6 and CCL2 levels in the BALF of AT2-specific FTO knockout COPD mice, concomitant with enhanced dendritic cell and neutrophil infiltration. Given that CCL2 potently recruits immune cells like dendritic cells, which themselves can secrete CCL2 during inflammation, this mechanism likely fuels a persistent, self-reinforcing inflammatory cascade. In recent years, antioxidants have emerged as promising therapeutic agents for emphysema, particularly in the context of COPD [39,[46], [47], [48]]. Therefore, we conducted in vitro experiments using NAC, a powerful antioxidant, to treat A549 cells with FTO deficiency, and observed that NAC greatly rescued ROS accumulation and necrotic apoptosis in FTO deficient A549 cells. Two randomized controlled trials (RCTs) have demonstrated that NAC administration reduces the frequency of acute exacerbations in specific COPD populations, particularly in patients not receiving inhaled corticosteroids or in those with moderate-to-severe disease and a history of recurrent exacerbations [49,50]. Our subsequent in vivo NAC treatment effectively ameliorated the exacerbation of COPD-related phenotypes induced by FTO deficiency in AT2 cells. Collectively, both in vitro and in vivo findings consistently demonstrate that FTO mitigates ROS accumulation in AT2 cells, thereby alleviating oxidative stress. CYPs are a major superfamily of heme-containing proteins and serve as the predominant phase I metabolic enzymes in humans, with their core function being the catalysis of redox reactions [19,20,51]. Based on our reanalysis of epithelium in the GEO dataset GSE11784, CYP1A1 and CYP1B1 emerged as the top two differentially expressed genes ranked by log2 fold-change, highlighting their pivotal role in COPD pathogenesis and their promise as therapeutic targets. Further in vitro experiments confirmed that CYP1B1 and CYP1A1 expression were significantly increased in AT2s upon exposure to CSE, which was consistent with prior reports [23]. These two genes are regulated via distinct mechanisms in the absence of FTO. CYP1B1 is a direct downstream target of FTO-mediated m6A modification. FTO deficiency promotes CYP1B1 expression via m6A modification and mRNA stabilization at the post-transcriptional level. Our preliminary investigation revealed that the induction of CYP1A1 under CSE exposure is not dependent on the m6A-demethylation activity of FTO, and appears to be mediated through the canonical AhR pathway, consistent with previous studies [[26], [27], [28], [29]]. However, the precise regulatory mechanism of CYP1A1 remains to be determined in future work. CYP1B1 demonstrates a more defined disease-exacerbating function in COPD [22,23,52]. Using single- and double-knockdown models of FTO and CYP1B1, we confirmed that CYP1B1 acts downstream of FTO. Notably, double knockdown of both genes ameliorated the ROS accumulation and necroptosis induced by FTO deficiency. In vivo analysis confirmed elevated mRNA and protein expression of CYP1B1 in the AT2 cell-specific FTO knockout COPD model compared with controls. The present study delivers several innovations. Firstly, it provides the first integrated evidence from human cohorts and genetic mouse models demonstrating that reduced FTO expression does not merely correlate with, but causally contributes to, adverse COPD-related outcomes such as impaired lung function, higher smoking burden, and spontaneous emphysema. These findings establish FTO as a pivotal disease-modifying molecule in emphysema pathogenesis. Second, mechanistically, we have identified a novel post-transcriptional pathway in which FTO deficiency selectively stabilizes CYP1B1 mRNA, resulting in increased CYP1B1 protein levels. This establishes CYP1B1 as the principal downstream effector of FTO in AT2s. Third, we directly link FTO loss in AT2s (the alveolar progenitor population) to the activation of oxidative stress and necroptosis via CYP1B1 upregulation, thereby connecting epigenetic regulation to core cellular dysfunction in emphysema pathogenesis. Through an AT2-specific in vivo model, we definitively demonstrate that AT2-restricted FTO deletion exacerbates cigarette smoke-induced emphysema, inflammation, and necroptosis, offering direct genetic validation of FTO's protective role in a physiologically relevant context. Finally, we show that the antioxidant NAC rescues ROS accumulation and necroptosis in FTO-deficient A549 cells. Moreover, NAC treatment effectively ameliorated the exacerbation of COPD-related phenotypes induced by FTO deficiency in AT2 cells. Together with clinical RCT evidence supporting NAC use in COPD, these findings translate mechanistic insight into an actionable therapeutic strategy targeting the FTO-m6A-CYP1B1-ROS-MLKL oxidative stress axis, thereby proposing a novel treatment avenue for COPD related-emphysema. This study has several limitations. First, despite observed phenotypes in mice, the clinical translation of these findings requires validation in human cohorts, given the physiological and disease heterogeneity between species. Second, our investigation centered on AT2s, utilizing A549 and MLE12 cell lines as experimental models. Although primary human AT2 organoids represent a more physiologically relevant system, their wide application was limited by current technical barriers in isolation and culture. At present, however, we were not able to perform these experiments because access to fresh, non-tumor human lung tissue suitable for primary AT2 isolation and organoid establishment is extremely limited in our setting. To address human relevance as much as feasible, we further analyzed two independent human epithelial transcriptomic datasets (GSE11784 and GSE63127). In these datasets, FTO expression was negatively correlated with CYP1B1 and MLKL expression, whereas CYP1B1 expression was positively correlated with MLKL. Although these data do not replace direct validation in primary human AT2 cells or organoids, they provide complementary support for the human epithelial relevance of the pathway. Hence, the translational implications of our in vitro data should be interpreted with consideration of this model limitation. Lastly, although we are currently exploring the mechanism likely through the canonical AHR pathway, the precise regulation of CYP1A1 remains to be fully determined in future work. Conclusion Our study identifies the FTO-m6A-CYP1B1-ROS-MLKL axis as a critical regulatory pathway in COPD-related emphysema. FTO deficiency stabilizes CYP1B1 mRNA in AT2 cells, initiating a cascade of oxidative stress, inflammation and necroptosis that drives alveolar destruction. To our knowledge, this is the first report demonstrating the protective role of FTO in mitigating emphysema through its action in AT2 cells. These findings position FTO and its downstream signaling cascade as a promising therapeutic target and provide a novel mechanistic framework for understanding and treating COPD-associated emphysema. Materials and methods Mice WT, global FTO-knockout, FTOflox/flox, and Sftpc-CreERT2 mice, all on a C57BL/6 background, were generated by and purchased from Cyagen Biosciences Inc. To generate mice with conditional FTO knockout in alveolar epithelial cells, Sftpc-CreERT2 (AT2-Cre) mice were crossed with FTOflox/flox mice. The resultant Sftpc-CreERT2/FTOflox/flox transgenic mice were used for the animal experiments. Genotypes were determined by PCR amplification of tail-biopsy–derived genomic DNA. PCR primers are shown in Table S2. To induce specific knockdown of FTO in the alveolar epithelium (AT2-FTO△/△), Tamoxifen (MedChemExpress, #HY-13757A), prepared in corn oil (MedChemExpress, #HY-Y1888), was delivered intraperitoneally at 150 mg/kg on alternating days for four total doses. Animal model In this study, we employed two experimental models. In PPE (sigma-aldrich, #E1250) induced emphysema model, mice (6-8 weeks old) were anesthetized with tribromoethanol prior to PPE administration. Using a mouse intubation needle, PPE (0.5U/50μl/mouse) was instilled intratracheally. During the procedure, the nostrils were temporarily blocked to facilitate inhalation of the solution through the oral cavity into the lower respiratory tract during spontaneous breathing. Control mice underwent the same procedure but received sterile physiological saline instead of PPE. In the CS whole-body exposure model, mice (6-8 weeks old) were placed in the exposure chamber at a density of 5-6 animals per cage (BKALSPS, BIO-KEY). Cigarettes were combusted (12 cigarettes per exposure cycle, 4 cycles per day, with 24 weeks) to maintain a concentration of particulate matter in the chamber between 150 and 200 mg/m3. Air-exposed littermates were processed in parallel as controls. All animal procedures conformed to the Guide for the Care and Use of Laboratory Animals and were reviewed and approved by the Animal Ethics Committee of Guangzhou Laboratory (GZLAB-AUCP-2023-02-A04). Mouse pulmonary function test Pulmonary mechanics were assessed using a computer-controlled small animal pulmonary function test system (DSI Buxco Pulmonary Function Test). Mice were anesthetized with tribromoethanol and underwent tracheotomy, followed by insertion of a metal cannula into the trachea. A series of lung function indices, including but not limited to FRC and FVC, were acquired in compliance with the standardized operating procedures detailed in the Buxco FinePointe™ software user guide. Collection of BALF Following blood collection, bronchoalveolar lavage was carried out through tracheal cannulation by instilling 0.8 mL PBS, repeated for two additional washes, with >85% total fluid recovery. BALF was then centrifuged at 800 × g for 5 min at 4 °C. The clarified supernatant was harvested and stored for ELISA, and the cell pellet was resuspended in PBS containing 1% fetal bovine serum (FBS). Following resuspension, cells were sequentially processed for live cell staining, blocking, and cell surface marker staining with specific antibodies (Table S3). Finally, immunophenotyping and population quantification were conducted using flow cytometry (Sony ID7000). Histology and MLI After euthanasia, mouse lungs were inflation-fixed via tracheal cannulation with 4% paraformaldehyde (perfusion pressure ∼30 cmH2O). Following tracheal ligation, the lungs were excised and immersed in 4% paraformaldehyde (PFA, post-fixation) for 24–48 h. The fixed lung tissues were then paraffin-embedded and sectioned at a thickness of 4 μm. Following the manufacturer's protocol, paraffin-embedded lung sections were subjected to H&E staining and then evaluated for histomorphological changes. Images were analyzed using ImageJ software (open source). From each slide, 10–12 regions, independent of airways and major blood vessels, were randomly selected for analysis. Intersections between alveolar septa and a grid of virtual lines of known length were quantified. An increase of the MLI indicates alveolar airspace enlargement. Immunofluorescence and fluorescence intensity quantification Following deparaffinization, rehydration, and sodium citrate-based antigen retrieval (30 min, boiling), lung sections were blocked (5% serum, 0.1% Triton X-100 in PBS, 1 h) and incubated overnight at 4 °C with primary antibody (1:100–1:500). A Tyramide Signal Amplification (TSA)-based fluorescence multiplex kit (ABclonal, #RK50025P) was used for secondary detection and signal amplification. Nuclei were stained with DAPI, and sections were mounted for imaging. Fluorescent images were acquired on an Olympus FV3000 microscope, and 6–10 random fields per section were analyzed with ImageJ software. GSSG assay GSSG levels were determined by using the GSH and GSSG Assay Kit (Beyotime, #S0053). Lung tissue (10 mg) was homogenized in 100 μL of protein removal reagent M. After centrifugation at 10,000 × g (4 °C, 10 min), the supernatant was collected and incubated with GSH removal reagent at 25 °C for 60 min. The GSH detection working solution was then added sequentially, mixed, and incubated for 5 min. NADPH was subsequently added, mixed thoroughly, and incubated for an additional 25 min. The absorbance was immediately read at 412 nm. Cell lines, cell culture and transfection A549 (human alveolar epithelial) cells were grown in DMEM, and MLE-12 (murine lung epithelial) cells were maintained in DMEM/F-12. Media were supplemented with 10% fetal bovine serum plus penicillin/streptomycin (100 U/mL and 100 μg/mL), and cultures were kept at 37 °C in a humidified 5% CO2 incubator. siRNA delivery was carried out using a Lipofectamine-based protocol. Briefly, siRNA (final 50 nM) was combined with 5 μL Lipofectamine in Opti-MEM and allowed to form complexes for 20 min at room temperature before application to cells. The resulting complexes were then added to cells cultured in OPTI-MEM medium. Following 6 h of transfection, we replaced the medium with complete growth medium. We initiated intervention experiments 24 h later and harvested cells for subsequent analysis within 48∼72 h. The Fto-siRNA (Sense: 5′-ACACUUGGCUCCCUUAUCUTT -3'; Antisense: 5′-AGAUAAGGGAGCCAAGUGUTT-3′), Negative-siRNA, Cyp1b1-siRNA (Sense: 5′- CCCACAGCAUGAUGCGCAATT-3'; Antisense: 5′- UUGCGCAUCAUGCUGUGGGTT-3′) and Lipofectamine transfection reagent (#G04005) are purchased from GenePharma. Acquisition of CSE For in vitro experiments, aqueous CSE was prepared to simulate the effects of cigarette smoke. The CSE was generated using standard research cigarettes. Smoke from one cigarette was slowly bubbled over approximately 5 min through 5 mL of DMEM medium. The resulting solution was then filtered through a 0.22 μm Millex-GS filter, aliquoted, and stored at −80 °C. This stock solution was designated as 100% CSE for subsequent experiments. RNA sequencing To profile transcriptomic changes, we collected RNA from both cellular and animal models: three independent replicates of FTO-knockdown (FTOsiRNA) and control A549 cells (lysed in TRIzol, n = 3 per group), and lung tissues from FTO-KD and WT mice (n = 4 per group). All samples underwent mRNA sequencing at Novogene (Beijing, China). Data were analyzed for differential expression (volcano plot), gene-set overlap (Venn diagram), and pathway enrichment (KEGG) according to Novogene's established workflow. MeRIP sequence RNA was fragmented to 100–200 nt, followed by immunoprecipitation (IP) with an anti-m6A antibody to isolate methylated RNA fragments, alongside input controls. Both IP and input samples underwent library preparation and sequencing. m6A peaks were called using exomePeak to identify significantly enriched methylation sites and annotate their genomic locations and corresponding genes. Genes with |log2FC| ≥ 1 and P < 0.05 in both m6A peak abundance and mRNA expression were considered co-regulated. For example, genes that meet the criteria of (| diffPeak.log2fc |≥1, diffPeak.p-value<0.05) and (| diffExp.gene.log2 (fc) |≥1, diffExp.gene.pval<0.05), were considered as genes with significant differences in RNA transcription and m6A methylation levels between groups. Bioinformatics analysis Public GEO RNA-seq datasets were mined to compare FTO expression across the indicated clinical cohorts. The datasets and their respective groupings were as follows: GSE63127 (healthy smokers vs. healthy nonsmokers), GSE11784 (COPD patients vs. healthy smokers and nonsmokers), and GSE76705 (all smokers, subcategorized based on the COPD diagnostic criterion of FEV1/FVC <70%). Expression data were normalized and differential expression was evaluated in RStudio using the “limma” and “DESeq2” packages. RNA extraction and quantitative real-time PCR (qRT-PCR) Total RNA was isolated with TRIzol (TaKaRa, #9109), converted to cDNA using PrimeScript™ RT Reagent Kit (TaKaRa, #RR037A), and quantified by real-time PCR with TB Green™ Premix Ex Taq™ (TaKaRa, #RR420A) according to the suppliers’ instructions. Gene expression was normalized to Gapdh and reported as fold change using the 2–ΔΔCt approach. Western blot analysis Cellular or lung-tissue proteins were solubilized in RIPA buffer (Thermo Scientific, #89901) containing a protease inhibitor cocktail (MedChemExpress, #HY-K0010). Comparable amounts of protein were resolved on 7.5% SDS–PAGE gels (EpiZyme, #PG211) and electrotransferred to PVDF membranes (Millipore). Membranes were briefly blocked with rapid blocking solution (Servicebio, #G2052) for 15 min at room temperature, then incubated with primary antibodies (Table S3) at 4 °C overnight. After washing, HRP- or IRDye 800CW-labeled secondary antibodies (Table S3) were applied for 1 h at room temperature. Signals were captured using a chemiluminescent imager (BLT, China) or an Odyssey M system (LI-COR), and band intensities were quantified with ImageJ or IPwin software. ELISA The concentrations of IL-6 (Invitrogen, #88-7064-88) and CCL2 (Invitrogen, #88-7391-88) in BALF supernatant were quantified using commercial ELISA kits according to the manufacturer's instructions. Capture antibody was immobilized on the plates by incubation at 4 °C overnight. After washing and blocking, 100 μL of BALF samples and standards were added to each well and incubated either overnight at 4 °C or for 2 h at room temperature. After an additional wash step, plates were sequentially incubated with the detection antibody and then an HRP-linked secondary antibody. Color was developed by adding TMB substrate, and the reaction was stopped with the provided stop solution. Absorbance was read at 450 nm on a microplate spectrophotometer without delay. Measurement of intracellular ROS and mtROS Intracellular ROS and mtROS were assessed with the fluorogenic dye DCFH-DA (Sigma-Aldrich, #D6883) and MitoSOX-PE (Invitrogen, #M36008), respectively. A549 cells were incubated with 10 μM DCFH-DA or 1 μM MitoSOX-PE at 37 °C for 20 min under light-protected conditions. After incubation, cells were washed twice with PBS and resuspended in PBS for immediate analysis. Fluorescence intensity was detected using the FITC channel for DCFH and the PE channel for MitoSOX-PE of a flow cytometer (Beckman CytoFLEX S). Data were analyzed with FlowJo software (version 10) to calculate the mean fluorescence intensity (MFI). Apoptosis We performed Annexin V-FITC/PI apoptosis assay (Immunoway, #AO2001-02P–H) on A549 cells in vitro according to the kit instructions. Cells were detached with trypsin to generate a single-cell suspension, then rinsed twice with pre-chilled PBS. Cells were resuspended in binding buffer, labeled with Annexin V–FITC and PI for 15 min in the dark, and promptly acquired on a Beckman CytoFLEX S flow cytometer. Paraffin lung sections were assessed for apoptotic nuclei using a TUNEL assay kit (Beyotime, #C1088) following the supplier's protocol. TUNEL-positive cells were visualized under an Olympus FV3000 microscope, and 10-15 random high-power fields (HPFs) per section were counted. Apoptosis was expressed as the percentage of TUNEL-positive nuclei among all nuclei counted (TUNEL+/total × 100%). RNA half-life detection Following 48-72 h of gene interference in A549 cells, transcription was halted by the addition of 10 μg/mL actinomycin D. Total RNA was extracted at 0-, 3-, 6-, and 9-h post-inhibition. The expression levels of target genes at these time points were quantified by qRT-PCR, and the RNA degradation rates were subsequently calculated. Statistics Data are presented as mean ± SEM unless specified otherwise. Differences between two groups were assessed using a two-tailed Student's t-test, and comparisons among more than two groups were evaluated by one-way ANOVA with Dunn-Šidák post hoc test, as applicable. Two-way ANOVA followed by Tukey's multiple comparisons test was applied for two-factor designs. Statistical analyses were performed in GraphPad Prism (v9.0), and significance was defined as P < 0.05. Ethics approval and consent to participate The animal study was reviewed and approved by Animal Ethics Committee of Guangzhou Laboratory (GZLAB-AUCP-2023-02-A04). Funding Major Project of Guangzhou National Laboratory (MP-GZNL2025C02006), Foundation of Guangzhou National Laboratory (YW-JCYJ0707,MP-GZNL2023A02001, ZL-SRPG22016, ZL-SRPG22018). CRediT authorship contribution statement Mindan Wu: Data curation, Investigation, Methodology, Validation, Visualization, Writing – original draft. Wenjing Wang: Data curation, Investigation, Methodology. Zemin Chen: Data curation, Investigation, Methodology. Yuanlan Yang: Methodology. Jiawei Yan: Methodology. Yachao Cui: Methodology. Peiyu Huang: Methodology. Feng Zhang: Methodology. Wei Tu: Supervision. Wensheng Chen: Supervision. Ruiting Sun: Supervision. Yumin Zhou: Funding acquisition, Supervision, Writing – review & editing. Leqing Zhu: Funding acquisition, Supervision, Writing – review & editing. Pixin Ran: Conceptualization, Funding acquisition, Project administration, Supervision. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgement We thank Jiajun Yang at Hangzhou LC-Bio Technology Co., Ltd. for his expert support and guidance in the analysis of the MeRIP-seq data. We appreciate the technical support of Xiaoqin Jin from the Public Platform of Medical Research Center, Academy of Chinese Medical Science. Zhejiang Chinese Medical University. Footnotes Supplementary data to this article can be found online at https://doi.org/10.1016/j.redox.2026.104237. Contributor Information Yumin Zhou, Email: zhouyumin410@126.com. Leqing Zhu, Email: zhuleqing1@163.com. Pixin Ran, Email: pxran@gzhmu.edu.cn. Supplementary data The following is the Supplementary data to this article: Multimedia component 1 Data availability Data will be available on request. References Associated Data Supplementary Materials Multimedia component 1 Data Availability Statement Data will be available on request.