Fig 2: FOXN3 is required for PARP1-mediated pulmonary fibrosis.(A) Western blot analysis of FOXN3 protein levels in the lungs of Parp1fl/fl and Parp1fl/flFoxn3LSL/+ mice following BLM treatment (1.5 mg/kg, 21 days) (n = 4). The mice were treated with or without Cre recombinase. (B) Histological trichrome staining assessed lung fibrosis in Parp1fl/fl and Parp1fl/flFoxn3LSL/+ mice with or without Cre recombinase treatment following BLM treatment (1.5 mg/kg, 21 days). Scale bar, 100 μm. (C) The collagen content was quantified using ImageJ software for (B). Four distinct fields for each sample were analyzed and expressed as a percentage of the total area. (D) H&E staining analysis was performed to evaluate the inflammatory response in the lungs of Parp1fl/fl and Parp1fl/flFoxn3LSL/+ mice with or without Cre recombinase treatment following BLM treatment (1.5 mg/kg, 21 days). Scale bar, 100 μm. (E) ELISAs were performed to measure the levels of TNFα and IL-6 in BAL fluid from the lungs of Parp1fl/fl and Parp1fl/flFoxn3LSL/+ mice with or without Cre recombinase treatment following BLM treatment (1.5 mg/kg, 21 days) (n = 4). (F) Quantitative PCR analysis was performed to assess the RNA levels of profibrotic factors in the lungs of Parp1fl/fl and Parp1fl/flFoxn3LSL/+ mice with or without Cre recombinase treatment following BLM treatment (1.5 mg/kg, 21 days) (n = 4). (G) Hydroxyproline measurements comparing the collagen content in the lungs of Parp1fl/fl and Parp1fl/flFoxn3LSL/+ mice with or without Cre recombinase treatment following BLM treatment (1.5 mg/kg, 21 days) (n = 4). (H) Anti-ProSPC and anti–α-SMA immunofluorescence staining analysis of lung sections from BLM-challenged (1.5 mg/kg, 21 days) Parp1fl/fl and Parp1fl/flFoxn3LSL/+ mice with or without Cre recombinase treatment (n = 4). Fluorescence intensity from four distinct fields for each sample was quantified using ImageJ software and expressed as the relative fluorescence intensity. The data in (D) to (F) were assessed via one-way ANOVA and are shown as the means ± SD. The blotting data for (A) were quantified as the means from three independent experiments using ImageJ software and analyzed using one-way ANOVA. NS means not significant; *P < 0.05 and ***P < 0.001.

Fig 3: PARP1 inhibits p38 expression by mediating the FOXN3-dependent repression of Smad transcriptional activity.(A and B) Western blotting analyzed p38 levels and FOXN3 phosphorylation at S83 and S85 in (A) A549 cells treated with TGF-β (10 ng/ml) over a time course (0, 16, 24, 48 hours) and (B) mouse lungs following BLM administration (1.5 mg/kg) at days 0, 12, and 21. (C) Western blotting evaluated p38 levels and FOXN3 phosphorylation in A549 cells treated with or without BYK204165 (20 μM, 4 hours) or PJ34 (5 μM, 4 hours). (D and E) Western blot analysis of p38 levels and FOXN3 phosphorylation in A549 cells with or without PARP1 knockdown (D) or KO (E) after TGF-β treatment (10 ng/ml, 16 hours). (F) Quantitative PCR detected the RNA levels of p38 in A549 cells treated with or without BYK204165 (20 μM, 4 hours) or PJ34 (5 μM, 4 hours). (G) Quantitative PCR detected p38 RNA levels in WT or Parp1−/− A549 cells treated with TGF-β (10 ng/ml, 16 hours). (H) CUT&Tag analyses showing the binding profiles of FOXN3, PARP1, and Smad4 to the p38 promoter in A549 cells treated with or without BYK204165 (20 μM, 4 hours). (I) ChIP assays in A549 cells treated with or without BYK204165 (20 μM, 4 hours) assessed FOXN3 and Smad protein binding to the p38 promoter. (J) ChIP assays measured FOXN3 and Smad protein binding to the p38 promoter in WT and Parp1−/− A549 cells after TGF-β treatment (10 ng/ml, 16 hours). (K) ChIP assays evaluated Smad protein binding to the p38 promoter in A549 cells with or without FOXN3 knockdown after TGF-β treatment (10 ng/ml, 16 hours). Data from (F), (G), and (I) to (K) were analyzed using two-tailed Student’s t tests and presented as the means ± SD. The blotting data for (A), (C), (D), and (E) were quantified as the means from three independent experiments using ImageJ software. Data for (A), (C), and (D) were analyzed using one-way ANOVA, while data for (E) were assessed using two-tailed Student’s t tests. The blotting data for (B) were also quantified as the means and analyzed by one-way ANOVA. *P < 0.05, **P < 0.01, and ***P < 0.001.

Fig 4: The expression levels of PARP1 and FOXN3 are significantly decreased in clinical cases of lung fibrosis.(A and B) Representative images from the histological immunohistochemical staining analysis showing the expression levels of PARP1 (A) and FOXN3 (B) in the fibrotic lesion areas and the corresponding adjacent normal tissue of patients diagnosed with lung fibrosis. Representative images from four distinct fields for each sample are shown and quantified. Scale bars, 100 μm. (C) Results from the quantitative analysis of (A) and (B) are shown as the immunoreactive score (IRS) (n = 10). (D) A proposed working model clarifies the molecular mechanism by which PARP1 regulates the p38-FOXN3 feedback loop to modulate Smad signaling. The schematic representation was created in BioRender. J. Lian (2025); https://biorender.com/p73d358. The data for (C) was assessed via two-tailed Student’s t tests and are presented as the means ± SD.

Fig 5: FOXN3 forms a complex with PARP1 and colocalizes at the promoters of Smad response genes.(A) Affinity-purified immunoprecipitates from HEK293T cells via anti-Flag immunoprecipitation (IP) were analyzed by mass spectrometry to identify unique peptides associated with Flag-tagged FOXN3. The sequences of the PARP1 peptides linked to FOXN3 are presented. (B) A co-IP assay was carried out in A549 cells to examine the endogenous association between FOXN3 and PARP1. (C) An anti-Flag co-IP assay was performed in A549 cells stably expressing Flag-tagged FOXN3 to assess the association between FOXN3 and PARP1 upon TGF-β treatment (10 ng/ml) at the indicated time points [0, 16, 24, and 48 hours (hr)]. IB, immunoblot. The cells were treated with MG-132 (20 μM, 4 hours) before collection. (D) An immunofluorescence assay was performed in A549 cells to assess the colocalization of FOXN3 and PARP1. Scale bar, 10 μm. (E) A deletion-mapping assay was performed in HEK293T cells to define the region within FOXN3 associated with PARP1. Flag-tagged FOXN3 WT or its various mutants were cotransfected with HA-tagged PARP1 into HEK293T cells (ΔNTD: Δ1-113aa; ΔForkhead: Δ114-199aa; ΔCTD: Δ200-490aa). (F) An in vitro GST pull-down assay was performed to examine the direct interaction between bacterially expressed GST-tagged FOXN3 and His-tagged PARP1 purified from Sf9 insect cells. (G) Venn diagrams illustrating the overlapping genes cotargeted by FOXN3 and PARP1 in A549 cells, as determined via CUT&Tag data analysis. (H) The normalized distribution of reads for the FOXN3 and PARP1 peaks at the transcription start site (TSS) is presented. (I) The binding profiles of FOXN3 and PARP1 at representative Smad response genes in A549 cells were analyzed via CUT&Tag data. The blotting data for (B), (C), (E), and (F) were quantified as the means from three independent experiments using ImageJ software. The data for (C) and (E) were analyzed using one-way ANOVA. **P < 0.01 and ***P < 0.001.