Kynurenine-AhR-SLC39A10-Zn2+ signaling reprograms macrophages and enhances pirfenidone efficacy in pulmonary fibrosis Authors: Huihui Yue, Ruihan Dong, Jianhan He, Fengqin Zhang, Yan Fan, Xinran Dou, Qing Zhou, Xuan Li, Long He, Shu Zhang, Qilin Yu, Weikuan Gu, Jianping Zhao, Yi Wang, Cong-Yi Wang, Huilan Zhang Published: 2026 ABSTRACT AbstractBackgroundPulmonary fibrosis (PF) is an irreversible and lethal lung disease characterized by progressive scarring lacking safe and effective treatment options. Recent studies have underscored the role of macrophage polarization in fibrotic progression, yet the role of kynurenine (Kyn), a metabolite of tryptophan (Trp), in macrophages during PF progression remains elusive.MethodsLiquid Chromatography–tandem Mass Spectrometry (LC–MS) analysis was used to detect tryptophan metabolism changes in the serum of PF patients and control subjects. Macrophage-specific Ido1 or Ahr deletion mice was utilized to explored the role of Kyn in the bleomycin-induced fibrotic mouse model and ChIP sequence was employed to elucidate the mechanism by which Kyn inhibits pro-fibrotic macrophage activation.ResultsWe identified Kyn, Trp levels and Kyn/Trp ratio (KTR) were notably elevated in the serum of patients with different types of PF and these alterations were inversely correlated with lung function. Although such elevation might appear pathogenic, our functional studies demonstrate that Kyn exerts protective effects in PF, akin to brain natriuretic peptide in heart failure. Macrophage-specific deletion of Ido1 or aryl hydrocarbon receptor (AhR, the receptor of Kyn) exacerbated bleomycin-induced PF, while exogenous Kyn supplementation mitigated disease severity. Mechanistically, Kyn bound to the AhR, facilitating its nuclear translocation, where it promoted Slc39a10 transcription to increase the intracellular levels of zinc ion, thereby inhibiting profibrotic macrophage differentiation. Intriguingly, pirfenidone was noted with high potency to suppress Kyn production and our studies demonstrated that administration of Kyn along with pirfenidone effectively enhanced the therapeutic efficacy against PF.ConclusionsIn summary, these findings reveal a previously unrecognized Kyn-AhR-SLC39A10-Zn2+ signaling axis that governs macrophage polarization in PF, and unveiled the importance of Trp metabolism in PF pathogenesis, which could be novel therapeutic strategies against PF.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12964-026-02900-5.Keywords: Pulmonary fibrosis, Tryptophan metabolism, Kynurenine, Macrophages Introduction Pulmonary fibrosis (PF) stands as a group of heterogeneous lung diseases, characterized by progressive and irreversible destruction of lung tissue structure, ultimately leading to impaired gas exchange and respiratory failure death [1, 2]. Despite extensive research, the pathological mechanisms underlying PF remain inadequately elucidated, resulting in limited effective treatment options [3]. Pirfenidone, a first-line clinical drug approved by the FDA for anti-fibrotic therapy, has been demonstrated to slowed PF disease progression to some extent [4, 5]. It is noteworthy that, its prolonged and high-dose treatment regimen often leads to severe adverse effects in patients, such as diarrhea, gastrointestinal reactions, liver damage, and photosensitivity [6–8], which renders a considerable challenge in the clinical therapeutic management of PF. Macrophages, essential innate immune cells highly abundant in the lungs, are demonstrated to play a key regulator during PF progression [9]. Generally, macrophages can be activated to pro-inflammatory or pro-fibrotic phenotypes response to different microenvironments [10]. Accumulating evidences confirmed increased levels of pro-fibrotic macrophages in lung tissues and BALF lavage fluid of PF patients, characterized by a multi-marker expression profile encompassing CD206, Arg1, Trem2, CD163, TGF-β1, and IL-10 [11–13]. Specifically, pro-fibrotic macrophages have been implicated in the promotion of fibroblast recruitment, proliferation, and differentiation into myofibroblast through releasing pro-fibrotic mediators such as TGF-β, PDGF, interleukin (IL-4), IL-10 and IL-13 [10, 14]. Consequently, targeting pro-fibrotic macrophage activation presents a promising therapeutic approach for PF. Accumulating evidences have uncovered that amino acid metabolism direct the fate decision of immune cells, especially on macrophage function [15]. Tryptophan (Trp), an essential amino acid in mammals, is involved in protein biosynthesis and serves as a precursor for the synthesis of multiple important bioactive compounds [16]. Generally, over 95% of Trp is metabolized through the kynurenine (Kyn) pathway (KP), which is facilitated by the rate-limiting enzymes tryptophan-2,3-dioxygenase (TDO) or indoleamine 2,3-dioxygenase 2 (IDO2) in the liver or kidneys, and IDO1 in other tissues [17]. Importantly, recent studies indicate that metabolites of KP are pivotal in orchestrating immune cells [18], however, the role of tryptophan metabolism in pro-fibrotic macrophages activation during PF progression remains to be elucidated. Interestingly, treatment with pirfenidone, has led to a decrease in the levels of Kyn and KTR in IPF patients [19]. By contrast, our previous data discovered that exogenous administration of Kyn can significantly suppress pulmonary fibrosis [20]. This prompted us to further explore whether the reduction of Kyn limits the therapeutic efficacy of pirfenidone and the supplemental administration of Kyn can enhance the treatment effects of pirfenidone. Therefore, the purpose of this study is to clarify the role of Trp metabolism in macrophages during PF progression, aiming to provide a comprehensive understanding of its pathogenesis and subsequently exploring the synergistic effect of Kyn and pirfenidone against PF. In this study, our investigation revealed increased levels of Kyn and the Kyn/Trp ratio (KTR, reflecting the activation of IDO1) in patients with various types of PF. Notably, the levels of Kyn and KTR exhibited a positive correlation with disease severity. Furthermore, macrophage-specific Ido1 knockout mice demonstrated that the absence of Ido1 significantly enhanced pro-fibrotic macrophage activation and facilitated TGF-β1 production, thereby exacerbating lung fibrosis induced by bleomycin (BLM) in mice. Mechanistically, Kyn bound to the aryl hydrocarbon receptor (AhR) and then promoted transcription of Slc39a10, leading to increased intracellular zinc ion levels and inhibition of pro-fibrotic macrophage polarization. Intriguingly, pirfenidone was noted with high potency to suppress Kyn production and we demonstrated the combination therapy of Kyn and lower dosage of pirfenidone obviously abrogated pro-fibrotic macrophage activation and exhibited a significant therapeutic effect in a BLM induced mouse model as well as in human precision-cut lung slice (hPCLS) model. These findings collectively underscore the crucial role of Kyn in preventing PF progression and highlight the synergistic therapeutic strategy of orally administering Kyn in conjunction with pirfenidone for the clinical management of PF. Materials and methods Human samples Plasma samples were collected from PF patients at the Respiratory Department while plasma samples from the normal control group were collected at the Medical Examination Center of Tongji Hospital. The basic characteristics of the participants including age, sex, lung function and Body mass index (BMI) are presented in Supplementary Table 1. Fibrotic lung samples were acquired from PF patients who underwent lung transplantation (IPF patients = 4 cases, CTD-ILD patients = 3 cases). While the normal control group's lung samples were partly derived from healthy donor’s lung (n = 2), with excised portions unsuitable for transplantation due to size incompatibility or non-transplantable conditions and other control lung samples were gained from para-carcinoma lung tissues (n = 5). Participant clinical characteristics are detailed in Table S2. All procedures involving human subjects adhered to the guidelines established by the National Institutes of Health (NIH) and were approved by the Institutional Review Board (IRB) of Tongji Hospital (TJ-IRB20210966, TJ-IRB20220736). Animals All experimental animals were housed in the Specific Pathogen-Free (SPF) facility at the Animal Center of Tongji Hospital Research Building, where environmental conditions were maintained at 40–70% relative humidity and 20–24℃ ambient temperature, with a regulated 12-h light/dark cycle. Ido1flox/flox mice were generated using CRISPR-Cas9 technology (C57BL/6 genetic background) at the Bioray Laboratory (Shanghai, China) with two loxP sites inserted into the Ido1 gene (Fig. 2C). As for the AhRflox/flox mice, they were obtained from the Jackson’s Laboratory (Stock No: 006203, ME, USA). Then Lyzm-Cre + transgenic mice (Shanghai Model Organisms Center) were crossed separately with Ido1flox/flox mice and AhRflox/flox mice to produce Lyzm-Cre + -Ido1flox/flox or Lyzm-Cre +—AhRflox/flox mice, as indicated in Figure S4A-B. Wild-type C57BL/6 mice (hereafter referred to as WT) were procured from the Jiangsu Jicui Corporation in Jiangsu, China. All animal related experiments were approved by the Animal Experiment Committee (TJH-202205023) and conducted in strict adherence to the guidelines established by the Animal Use and Care Committee of Tongji Hospital, Huazhong University of Science and Technology, Wuhan, China. Specific deficiency of Ido1 in macrophages exacerbates BLM-induced pulmonary fibrosis. A Immunofluorescence co-staining for IDO1 and CD68 (a marker for macrophage) on lung sections from PF patients (n = 5) and control subjects (n = 5). B Results for coimmunostaining of Ido1 and F4/80 in the lung sections from PBS (n = 5) and BLM-treated (n = 5) mice. Cell nuclei were stained blue with DAPI, and images were captured at an original magnification of × 400. C Based on the CRISPR-Cas9 system, the Ido1flox/flox mice were generated by inserting two loxP sequences in the same direction into the intron flanked with the exon 4 of IDO1, which could generate a stop codon in exon 3 to produce a nonfunctional IDO1 protein after Cre-mediated gene deletion. The Ido1flox/flox mice were crossed with the Lyzm-Cre+ transgenic mice to obtain the Macrophage-specific Ido1-knockout mice (Lyzm-Cre+ -Ido1.flox/flox). D Co-immunostaining of Ido1 and F4/80 in lung sections from Ido1-C and Ido1-CKO mice. Nuclei were stained blue by DAPI, and the images were taken at an original magnification of × 400. E LC–MS analysis of Kyn levels in the serum of Ido1-CKO mice and Ido1-C group with or without BLM treatment on Day 21. F–H Representative micro-CT and 3D images of lung tissues from Ido1-CKO and Ido1-C mice with BLM treatment on Day 21 (F). Fibrotic changes were indicated in gray areas (G) and Fibrotic area (H). I Representative images for H&E (top), Sirius red (middle) and Masson staining (bottom). Images were captured at × 200 magnification. J Quantitative mean score of the severity of fibrosis. K The protein expression levels of fibronectin and collagen 1 in the Ido1-CKO and Ido1-C mice with PBS or BLM treatment. Protein expression levels were normalized to β-Actin as an internal loading control. L-N The reverse transcriptase (RT)-PCR analysis of Fn1 (L), Coll1a1 (M) and TGF-β1 (N) mRNA levels in Ido1-CKO and Ido1-C mice with PBS or BLM treatment. Five to six mice were included in each experiment group. Data are presented as mean ± SD and statistical significance was assessed by two-sided Student’s t test. *p < 0.05, **p < 0.01, ***p < 0.001 Bleomycin-induced lung fibrosis model and treatment Ido1flox/flox, Lyzm-Cre+-Ido1flox/flox, AhRflox/flox and Lyzm-Cre+- AhRflox/flox mice were anesthetized using 1% pentobarbital and subsequently administered an intratracheal administration of BLM (1.5 mg/kg, MCE, Shanghai, China) to induce a murine model of PF, following the method described in a previous study [21]. An equal volume of PBS was injected intratracheally into the control group. For Ido1flox/flox and Lyzm-Cre + -Ido1flox/flox mice, on 21 days post-injection, the mice were sacrificed for evaluation. For AhRflox/flox and Lyzm-Cre+- AhRflox/flox mice, Kyn (40 mg/kg, Sigma, Germany) were administrated intragastrically on day 14 following BLM-injection for consecutive 7 days and sacrificed on day 21. For the therapeutic model, Bleomycin (1.8 mg/kg) was administered intratracheally to WT mice. On the 14th day of BLM induction, mice were orally administered with PBS, pirfenidone (50 mg/kg, MCE, Shanghai, China), Kyn (20 mg/kg), Kyn (20 mg/kg) + pirfenidone (50 mg/kg), or high-dose pirfenidone (100 mg/kg) for 7 days and then analyze the PF development and severity on the 21th day. Micro-CT scanning and quantitative image analysis On the 21 st day post-BLM induction, mice were anesthetized with 1% pentobarbital intraperitoneally and placed on a specialized animal holder. Following fixation with medical tape, the mice were positioned within a micro-CT scanner for the purpose of lung imaging. Computed tomography (CT) images of the lungs were subsequently obtained utilizing a CT imaging system (SkyScan 1176, Bruker, Belgium). The imaging parameters were set as follows: pixel size of 18 µm, rotation step of 0.5°, total rotation angle of 360°, and a total scan time of 15 min. The acquired data were reconstructed into three-dimensional images of the lung using scanning software (BRUKER Micro-CT SkvScan 1176, CTvox Version 3.0, CTAn 1.16.4.1, and CTvoi). Lung volume was subsequently isolated and quantified using the CTAn software. Image J is used for analyzing acquired lung micro-CT images. In our study, we employed manual selection in Image J to outline the lung contours of mice in the CT images, allowing us to obtain the grayscale values of the lung region. These values were used for quantitative analysis of the degree of pulmonary fibrosis. Histological analysis and immunofluorescence staining The left lung was infused with 200 μL of 4% paraformaldehyde (PFA) and sliced in half. Following 24-h fixation in 4% PFA, the upper lobe underwent paraffin embedding for preparation of 5-μm-thick histological sections. These sections were then subjected to Masson, Sirius Red staining and hematoxylin and eosin (H&E). For each pathological section, two pathologists blindly selected five random fields at a magnification of 200 × to obtain the average Ashcroft score. For immunofluorescence staining, several 5-μm paraffin sections were prepared, dewaxed, and underwent antigen retrieval. Then relevant antibodies including IDO1, CD206, Slc39a10 (Proteintech, Wuhan, China), CD68, F4/80, Ahr (Santa Cruz Biotechnology, CA, USA) and TGF-β1 (ThermoFisher, MA, USA) were incubated. Subsequently, the sections were stained with 488-conjugated or Alexa Fluor 594 antibodies (Invitrogen, California, USA) and imaged at 400 × magnification using a fluorescence microscope (Olympus, Shinjuku, Japan). Western blot analysis Western blot was conducted following established protocols [22]. Briefly, homogenized samples (cultured cells and murine/human lung tissues) in RIPA buffer (Roche, China) were quantified using BCA assay. Proteins were separated by SDS-PAGE and transferred to PVDF membranes (Bio-Rad). After 1-h blocking with 5% skim milk, membranes were probed with primary antibodies at 4 °C overnight (Table S3). Subsequently, HRP-conjugated secondary antibodies (Servicebio, China) were subjected to 1 h incubation and protein bands were visualized with ECL reagent (Biosharp, Beijing, China), captured on a Gelview 6000Plus smart imaging workstation (Guangzhou, China). Quantitative analysis was conducted utilizing Image J software, with normalization performed using β-actin or GAPDH. Quantitative RT-PCR analysis Total RNA from cells and lung tissues was assessed for concentration/purity using a NanoDrop One spectrophotometer (Thermo Fisher Scientific, USA). Following reverse transcription, quantitative PCR was performed on a CFX96 system (Bio-Rad, USA) with optimized cycling parameters. Target-specific primers (Supplementary Table 3) were used to determine gene expression levels through the 2^(-ΔΔCt) method, normalized against β-actin as an endogenous control. Liquid Chromatography–tandem mass spectrometry (LC–MS) analysis Tryptophan and Kyn levels in plasma and tissues were quantified using an established chromatographic method [23]. Following a series of treatments, the samples were introduced into a high-performance liquid chromatography (HPLC) system (Shimadzu, Kyoto, Japan) for detection and analysis, which was equipped with an Ultimate XB-C18 HPLC column (Shanghai, China). Culture and treatment of BMDMs Mouse bone marrow-derived macrophages (BMDMs) were obtained using established techniques. Briefly, bone marrow cells from mouse femurs were cultured in RPMI 1640 medium supplemented with 10% FBS. M-CSF (30 ng/mL, Biolegend, CA, USA) was added for 3 days, then the medium was refreshed with a lower M-CSF concentration (20 ng/mL) on day 7. Adherent BMDMs were collected on day 7 for further experiments. Quantification of intracellular free Zn2⁺ with FluoZin-3-AM Macrophages were collected and suspended in sample buffer, followed by two washes, and then treated with 1 μM FluoZin-3-AM (American, Thermo Fisher Scientific) at 37 °C for 30 min in a CO2 incubator. Fluorescence intensity was assessed using a FACScan (BD Biosciences, Germany), and intracellular labile zinc ion concentration was calculated using a previously established method and the formula [Zn] = KD*[(F-Fmin)/(Fmax-F)] [24]. Minimal fluorescence was measured with 50 µM N,N,N′,N′-tetrakis-(2-pyridyl-methyl)-ethylenediamine (TPEN), and maximal fluorescence with 100 µM ZnSO4 plus 50 µM sodium pyrithione (both from MCE, Shanghai, China). Chromatin immunoprecipitation (ChIP) sequence The ChIP experiments were conducted by Wuhan Frasergen Biotechnology Co., Ltd., following a previously established protocol with minor modifications [25]. Immunoprecipitation was carried out using anti-Aryl Hydrocarbon Receptor (AHR) antibodies (Endo, USA). ChIP-Seq libraries were generated utilizing the Ultra II DNA Library Kits (NEB, #E7645), in accordance with the protocols specified by the manufacturer. For ChIP-seq analysis, the isolated DNA fragments were ligated to sequencing adapters and analyzed through high-throughput sequencing using the Illumina NovaSeq 6000 platform with a 150 bp paired-end read configuration. Chromatin immunoprecipitation (ChIP) assay and luciferase reporter assay To verify the specific binding of AHR to the SLC39A10 promoter region, we combined ChIP with a dual-luciferase reporter assay. Specifically, macrophages were processed using a ChIP kit (Beyotime): protein-DNA interactions were first crosslinked with formaldehyde, and chromatin was randomly sheared by sonication to approximately 200–1000 bp. An AHR antibody was then added for incubation overnight at 4 °C, with normal rabbit IgG as a negative control. Immune complexes were captured with protein A/G magnetic beads, followed by washing and DNA purification, and enrichment at the SLC39A10 promoter was assessed by ChIP-PCR (primer sequences are listed in Supplementary Table 3). In addition, plasmids carrying mutant SLC39A10 promoters were constructed, and the specific interaction between AHR and SLC39A10 was further validated in a dual-luciferase reporter system. Flow cytometry analysis Single-cell suspension from mouse lung tissues were digested by Collagenase I and Dispase II (Gibco, USA) and stained with flow antibody including BV510-CD45, PE-CY7-CD11b, BV421-F4/80, APC-CD206, PE-TGF-β1(Biolegend, CA, USA). IL-4-stimulated BMDMs (10 ng/mL, 24 h; Biolegend) were analyzed using PE-F4/80 and APC-CD206 antibodies (Biolegend, CA, USA). Flow cytometry data were processed with FACS Express V3 (De Novo Software) following manufacturer protocols. Human precision-cut lung slices (hPCLS) Human Precision-Cut Lung Slices (hPCLS) were prepared following previously established methods [26]. Briefly, non-tumor-adjacent normal lung tissue obtained from lung cancer resections and lung tissue from PF patients undergoing transplantation were placed in an organ preservation solution and thoroughly washed with phosphate-buffered saline (PBS). Then lung tissues were perfused with 1.5% low-melting-point agarose and were rapidly transferred to ice-cold PBS to solidify. The tissue was then trimmed with sterile blades and sliced into 400 μm-thick viable sections using a vibrating microtome (Leica VT1200, Germany). The freshly cut hPCLS were cultured in DMEM containing antibiotics and 0.5% fetal bovine serum (FBS) at 37 °C in a 5% CO2 incubator. After 48 h, the medium was refreshed, and slices were co-incubated for 48 h with a pro-fibrotic cocktail (20 ng/mL TGF-β1: Peprotech, 100–21, 20 ng/mL PDGF-AB: Peprotech, 100-00AB, 10 ng/mL TNF-α: R&D Systems, #P06804 and 5 μM LPA: Cayman Chemical, 62,215) and/or Kyn, pirfenidone (PFD), or a Kyn + PFD combination, after which tissues were collected for assessment of fibrotic indices. scRNA-seq data analysis Publicly available single-cell RNA-seq datasets GSE136831 and GSE135893 were analyzed using Seurat v4.2.2. After quality control and preprocessing, each dataset was normalized, highly variable genes were identified, and data were scaled with ScaleData. Then, principal component analysis (RunPCA) was performed on the normalized data, integration was conducted using Harmony (RunHarmony), and dimensionality reduction and clustering were performed on the resulting Harmony embeddings. Cell subpopulations were annotated based on canonical marker genes, and different cell types were identified by incorporating literature markers. Statistical analysis Data Analysis were conducted using GraphPad Prism 9.0 software (GraphPad Software Inc., San Diego, CA, USA). The data are presented as means ± SD/SEM unless otherwise specified. Comparative analyses employed paired Student's t-test for matched groups and Welch's t-test for independent samples. Multi-group comparisons were assessed by one-way ANOVA with Bonferroni’s correction. Statistical significance was defined as p < 0.05 for all experimental comparisons. Results Serum levels of kynurenine and KTR were inversely associated with lung function To determine the status of Trp metabolism in different types of PF patients, 365 human serum samples were analyzed by LC–MS. Our study cohort included 66 healthy controls, and patients with idiopathic pulmonary fibrosis (IPF, n = 50), interstitial pneumonia with autoimmune features (IPAF, n = 43), non-specific interstitial pneumonia (NSIP, n = 39), and connective tissue disease-associated interstitial lung disease (CTD-ILD, n = 167). Our results demonstrated significantly elevated serum Kyn concentrations in multiple PF subgroups (including IPF, IPAF, NSIP, CTD-ILD) compared to the control group (Fig. 1A). Additionally, in Fig. 1B, the results indicated that tryptophan levels are remarkably increased in IPF, IPAF, and NSIP patients relative to controls, although this elevation is not observed in CTD-ILD. Moreover, the kynurenine-to-tryptophan ratio (KTR), a widely used surrogate for IDO1 enzymatic activity, was elevated in PF patients (IPF, IPAF, NSIP, CTD-ILD) versus controls (Fig. 1C). Consistently, a substantial upregulation of IDO1 expression was observed in lung homogenate derived from PF subjects, coinciding with increased expression of fibrosis related marker, α-smooth muscle actin (α-SMA) (Fig. 1D). Importantly, correlation analysis revealed Kyn, KTR were positively correlated with disease severity, as assessed by lung gas diffusion capacity (DLCO%, Fig. 1E and J), forced vital capacity (FVC%, Fig. 1F and K), forced expiratory volume in one second (FEV1%, Fig. 1G and L), and total lung capacity (TLC%, Fig. 1H and M). Furthermore, the receiver operating characteristic (ROC) curve analysis revealed the area under the curve (AUC) values is 0.809 and 0.714 for Kyn and KTR, respectively (Fig. 1I and N). Serum levels of kynurenine and KTR were inversely associated with lung function. Plasma of Kyn levels (A), Trp levels (B) and KTR (C) in control subjects (n = 66) and IPF (n = 50), IPAF (n = 43), NSIP (n = 39), and CTD-ILD (n = 167) patients. (D) Western blot analysis of IDO1 and α-SMA protein expression in lung homogenate from PF individuals and normal controls (n = 7). (E–H) Correlation analysis between plasma Kyn levels and DLCO% (E), FVC% (F), FEV1% (G), and TLC% (H) predicted in PF patients. (I) ROC curves for plasma Kyn levels. (J-M) Correlation analysis between KTR and diffusion capacity DLCO% (J), FVC% (K) FEV1% (L), and TLC% (M) predicted in PF patients. (N) ROC curves for plasma Kyn levels. (O-Q) Plasma Kyn levels (O), Trp (P) and KTR (Q) in WT mice treated with BLM at day 0, 14 and 21 (n = 10, 10, 12, respectively). (R) Protein expression level of Ido1 and Fibronectin in the lung homogenate between BLM-induced (n = 6) mice and PBS (n = 6) group. Protein expression intensity was measured by Image J, and β-Actin or GAPDH were as an internal loading control. Data are presented as mean ± SD. Significant differences were indicated with p values. *: p < 0.05, **: p < 0.01, ***: p < 0.001 In line with above observations in PF patients, Kyn and Trp levels, along with KTR were increased in the serum of PF mouse model, along with a similar trend in KTR (Fig. 1O-Q). Additionally, RT-PCR and Western blot analysis revealed enhanced expression levels of Ido1 in the lung tissue of BLM-treated mice, accompanied by elevated levels of fibronectin, another fibrosis marker (Fig. 1R, Figure S1). Collectively, our study provides compelling evidence indicating the aberrant activation of the KP during Trp metabolism in PF patients, and the Kyn and KTR in the serum may serve as a putative biomarker for disease detection and monitoring. Specific deficiency of Ido1 in macrophages exacerbates BLM-induced pulmonary fibrosis To further detected the cells exhibiting altered Ido1 expression to confirm the source of Kyn production, we did immunofluorescence co-staining of lung sections from PF patients and the results illustrated the number of CD68+ IDO1+ was dramatically higher than that in control subjects (Fig. 2A). Similar findings were observed in lung sections of BLM treated mice (Fig. 2B), indicating macrophage was also one of the main type cells of Kyn production. Therefore, Macrophage-specific Ido1 deficient mouse model was constructed by crossbreeding Ido1fl/fl mice with LyzM-Cre+ mice (LyzM-Cre+-Ido1fl/fl mice, referred to as Ido1-CKO mice) and their littermate control mice (Ido1fl/fl, referred to as Ido1-C mice) (Fig. 2C). To identify the deletion of Ido1, DNA genotyping confirmed the presence of flox alleles and Cre alleles in mouse tail samples (Figure S2A). Additionally, RT-PCR analysis demonstrated negligible expression of Ido1 in BMDMs from Ido1-CKO mice (Figure S2B). Furthermore, immunofluorescence co-staining of lung tissue sections revealed the absence of Ido1 expression in F4/80+ macrophages of Ido1-CKO mice in comparison to Ido1-C mice (Fig. 2D). Moreover, LC–MS analysis demonstrated a 48.11% decrease in serum Kyn levels in Ido1-CKO mice compared to Ido1-C group (Fig. 2E). Subsequently, both Ido1-C and Ido1-CKO mice were intratracheally administered PBS or BLM for 21 days. Quantitative analysis of lung micro-CT and 3-dimensional reconstruction revealed that the deficiency of Ido1 in macrophages significantly aggravated lung damage, decreased lung volume, and intensified PF induced by BLM in mice (Fig. 2F-H). Consistent with these observations, H&E, Masson's trichrome, and Sirius red staining demonstrated a more severe degree of fibrosis in Ido1-CKO mice compared with control, as indicated by higher fibrotic scores (F ig. 2I and J). Additionally, Western blot and RT-PCR analysis revealed a marked upregulation in the protein and mRNA expression of fibrosis-related markers (Fibronectin and Collagen 1) in Ido1-CKO mice (Fig. 2K, L, M). Notably, the expression of TGF-β1, a key pro-fibrotic cytokine, was robustly enhanced in Ido1-CKO mice following BLM treatment (Fig. 2N). To further demonstrate the impact of IDO1 deficiency in lung macrophages on pulmonary fibrosis, we employed Ido1loxp/loxp mice in combination with intratracheal delivery of rAAV6-F4/80-Cre, which efficiently targets lung-resident macrophages while minimizing off-target recombination in non-myeloid cells. As shown in Figure S2C-H, we observed phenotypic changes comparable to those seen in the Ido1-LyzM-Cre mouse model. Consequently, these data indicate that specific deficiency of Ido1 in macrophages exacerbates BLM-induced PF, highlighting the protective compensatory response of elevated Kyn in the pathogenesis of PF. Kynurenine abrogates the pro-fibrotic program in macrophages Pro-fibrotic macrophages have been demonstrated to be robustly linked to the progression of PF through their sustained production of fibrogenic mediator, such as TGF-β [27–29]. Consistent with previous findings, immunofluorescence co-staining showed the number of CD206+TGF-β1+ was remarkably increased in PF patients (Fig. 3A). Based on this observation, we utilized flow cytometry to assess alterations in profibrotic macrophage phenotypes within lung tissues of Ido1-CKO and Ido1-C groups. The results indicated that Ido1 deficiency in macrophages promotes the activation of pro-fibrotic macrophages, as evidenced by an increased proportion of CD206+TGF-β1+ macrophages(Fig. 3B and Figure S3A). Moreover, co-immunostaining results revealed a statistically significant elevation in the population of F4/80+CD206+ cells in Ido1-CKO mice following BLM induction (Fig. 3C). To further investigate the regulatory role of Ido1 on macrophage activation, the expression of pro-fibrotic macrophage markers was assessed. Western blot analysis demonstrated enhanced expression of Arg1 and CD206 in Ido1-CKO group and rAAV6-F4/80-Cre treated group (Fig. 3D and Figure S3B). Similarly, RT-PCR analysis showed elevated expression of CD206, Arg1, Fizz1, and Ym1 (Fig. 3E-H and Figure S3C-D). Kynurenine abrogates the pro-fibrotic program in macrophages. A Co-immunostaining of TGF-β and CD206 in lung sections from control and PF subjects. B Flow cytometry analysis of macrophages derived from lung tissues of Ido1-CKO and Ido1-C mice after BLM induction. C Co-immunostaining of F4/80 and CD206 in lung sections from Ido1-CKO and Ido1-C mice with BLM induction. D The protein expression levels of CD206 and Arg1 in the lung homogenate from Ido1-CKO and Ido1-C mice with PBS or BLM treatment. E–H RT-PCR results for CD206, Arg1, YM1 and Fizz1. I The CD206 expression by flow cytometric analysis in BMDMs after the indicated treatments. J Representative results for CD206 and Arg1 protein levels in BMDMs with PBS/IL-4 or KYN/IL-4 plus KYN stimulation for 24 h. K-M RT-PCR results for CD206 (K), Arg1 (L) and Fizz1 (M) in BMDMs with PBS/IL-4 or KYN/IL-4 plus KYN stimulation for 24 h. Data are presented as mean ± SD. *: p < 0.05, **: p < 0.01, ***: p < 0.001 Since Ido1 is the key rate-limiting enzyme responsible for Kyn production in lung tissue, we further explored the impact of Kyn on macrophage activation in vitro. Flow cytometry results demonstrated a notable reduction in the number of F4/80+ CD206+ macrophages in IL-4 + Kyn (100 μg/mL)-treated group compared to IL-4 treatment alone in Ido1-knockout BMDMs (Fig. 3I). Similarly, Western blot and RT-PCR analysis confirmed that Kyn (100 μg/mL) markedly attenuated IL-4-induced pro-fibrotic macrophage polarization, with concomitant reductions in CD206 and Arg1 protein expression levels (Fig. 3J), as well as CD206, Arg1, and Fizz1 mRNA levels (Fig. 3K-M). These findings collectively indicate that Kyn, as the direct product of Ido1, abrogates the formation of pro-fibrotic macrophage. The effect of kynurenine on pro-fibrotic macrophage activation is dependent on AhR Recently, studies have identified the AhR as a receptor for Kyn in adipocytes [30]. This discovery was further substantiated by in vitro experiments showing that Kyn significantly facilitates the nuclear translocation of AhR in macrophages (Fig. 4A and B). To examine whether the effect of Kyn on pro-fibrotic macrophage activation is dependent on the AhR, we utilized BMDMs from Lyzm-Cre+ AhRfl/fl mice (defined as AhR-CKO mice, Figure S4A and B). Quantitative analysis of both protein and mRNA expression levels indicated no discernible difference in the pro-fibrotic macrophage markers between groups treated with IL-4 + Kyn and those treated with IL-4 alone, suggesting that the inhibitory effect of Kyn on pro-fibrotic macrophage activation relies on AhR activation (Fig. 4C and Figure S4C-E). Subsequently, AhR-CKO mice and their littermate control mice (AhRfl/fl mice) were subjected to BLM injection followed by treatment with Kyn (40 mg/kg) daily on day 14. Micro-CT images and pathological evaluation demonstrated that the absence of AhR in macrophages remarkably exacerbated BLM-induced lung fibrosis, and exogenous Kyn markedly mitigated BLM-induced pulmonary fibrosis in AhRfl/fl group compared to the moderate reduction observed in AhR-CKO group (Fig. 4D-H). Consistently, the fibrotic markers assessed by Western blot and RT-PCR exhibited similar results (Fig. 4I and Figure S4F). Furthermore, the activation of pro-fibrotic macrophages was evaluated in the lung tissue between these groups. Notably, AhR-CKO group presented enhanced expression levels of pro-fibrotic macrophage markers confirmed by Western blot and RT-PCR analysis, which was not reversed by Kyn treatment (Fig. 4J, Figure S4G). Overall, the findings from both in vitro and in vivo studies indicate that the inhibitory effect of Kyn on pro-fibrotic macrophage activation is dependent on the activation of the AhR. The effect of Kynurenine on pro-fibrotic macrophage activation is dependent on AhR. A Representative confocal results for AhR and phalloidin co-immunostaining after KYN stimulation. B Western blot analysis of AhR expression change in the cytoplasm and nuclear with KYN stimulation on 5, 15 and 30 min. C Protein expression levels of CD206 and Arg1 in BMDMs from AhR-CKO mice with PBS/IL-4 or KYN/IL-4 plus KYN stimulation for 24 h. D-F Representative micro-CT and 3D images of lung tissues from AhR-C and AhR-CKO mice with BLM for 21 days with or without KYN treatment for 7 days, Fibrotic changes were indicated in gray areas (E) and Fibrotic area (F). G Representative images for H&E (top), Masson staining (middle) and Sirius red (bottom). Images were captured at × 200 magnification. H Semiquantitative Ashcroft scores relevant to the severity of fibrosis. I Western blot analysis of fibronectin (Fib) and collagen (Coll) I expression levels in AhR-CKO and AhR-C mice with BLM or BLM plus KYN treatment. J The protein expression levels of CD206 and Arg1 in AhR-CKO and AhR-C mice with BLM or BLM plus KYN treatment. Five to six mice were included in each experiment group. Data are presented as mean ± SD. *: p < 0.05, **: p < 0.01, ***: p < 0.001 AhR mitigates pro-fibrotic macrophage activation via enhancing Slc39a10 expression The aforementioned results demonstrate that Kyn binding to AhR triggers its nuclear translocation, resulting in the inhibition of pro-fibrotic macrophage activation. To systematically elucidate the molecular and functional mechanisms underlying the regulation of profibrotic macrophage activation by Kyn, we conducted ChIP sequencing analysis in IL-4 + Kyn-treated BMDMs and identified the top 20 genes based on enrichment. These genes were then compared with those exhibiting substantial differences between control and IPF group in GSE47460 database (Fig. 5A). Among the overlapping genes, 8 candidate genes emerged as potential AhR target genes (Fig. 5B). Subsequently, RT-PCR was performed to validate these genes, and only solute carrier family 39 member 10 (Slc39a10) demonstrated a significant increase following Kyn treatment (Fig. 5C-D, Figure S5A-G). To validate Slc39a10 as a direct transcriptional target of AhR, we predicted putative AhR response elements and identified a conserved AhR-binding motif located at –646 to –642 relative to the transcription start site using the JASPAR database(Fig. 5E). Chromatin immunoprecipitation (ChIP) assays with an anti-AhR antibody revealed significant enrichment of Slc39a10 promoter fragments containing this motif (Fig. 5F), confirming in vivo binding. To establish direct transcriptional regulation, we generated a wild-type luciferase reporter construct (Slc39a10-WT) encompassing the predicted AhR-binding site and a mutant version (Slc39a10-MUT) with deletion of the motif. Co-expression of AhR markedly increased transcriptional activity from the Slc39a10-WT reporter, but this activation was abrogated in the Slc39a10-MUT construct (Fig. 5G), which demonstrate that AhR directly binds to the Slc39a10 promoter and drives its transcriptional activation. Notably, silencing of Slc39a10 using siRNA in BMDMs significantly enhanced pro-fibrotic macrophage activation and abolished the protective effects of Kyn on pro-fibrotic phenotype after IL-4 stimulation (Figure S5H and Fig. 5H). AhR mitigates pro-fibrotic macrophage activation via enhancing Slc39a10 expression. A The top 20 enriched genes in IL-4 and KYN-treated BMDMs identified by ChIP sequencing were intersected with genes with significant differences between control group and IPF patient group in GSE47460 microarray database as indicated by Venn diagram. B The heatmap of the potential eight candidate genes emerged as AhR target genes between control and IPF group in GSE47460 database. C Quantitative mRNA analysis of Slc39a10 in s from BMDMs after different treatments. D Representative immunoblots of Slc39a10 in BMDMs under IL-4 or IL4 and KYN treatment. E The putative binding site in the Slc39a10 promoter region for AHR was predicted with a relative profile score threshold of 80%. F ChIP-PCR analysis of the AHR binding activity to the Slc39a10 promoter. G Luciferase reporter assays for the Slc39a10 promoter conducted in iBMDM. H Western blot analysis of Slc39a10 and CD206 in BMDMs under the indicated treatments. I Relative intracellular zinc concentrations in BMDMs after different stimulation. J Immunoblot results of Slc39a10 in BMDMs from AhR-C and AhR-CKO mice under IL4 or IL-4 and KYN treatment. K Representative immunofluorescence co-staining images of Slc39a10 and CD206 in the lung section from AhR-C and AhR-CKO mice with BLM or BLM and KYN treatment. (L) UMAP plot showing expression of Slc39a10 in different cell types from control and IPF group at the single-cell level (GSE135893). M Co-immunostaining images showing expressions of SLC39A10 and CD68 in lung tissues from control and PF subjects. Data are presented as mean ± SD of three independent experiments. *: p < 0.05, **: p < 0.01, ***: p < 0.001 Slc39a10, also known as Zinc-influx transporter 10 (Zip10), mediates cellular zinc uptake which could inhibit the pro-fibrotic macrophage activation [31, 32]. Consistent with this, macrophages treated with Kyn demonstrated elevated levels of intracellular Zn2+ while this increase was reversed when Slc39a10 expression was silenced (Fig. 5I and Figure S5I). Furthermore, Western blot analysis manifested supplementation with zinc ions downregulated the expression of profibrotic macrophage activation related markers and the phosphorylation levels of and JAK2/STAT6 was also decreased after zinc ions treatment (Figure S5J-L). Notably, the remarkable elevation in Slc39a10 expression induced by Kyn was abrogated in the absence of AhR, even in the presence of Kyn, suggesting that the expression of Slc39a10 induced by Kyn is dependent on AhR (Fig. 5J). This effect was further corroborated by in vivo data (Fig. 5K). To explore the potential role of SLC39A10 in the development of pulmonary fibrosis, we analyzed GSE47460 RNA sequence database and found a slight upregulation of SLC39A10 in whole-lung homogenates from IPF patients, which was confirmed with BLM-treated PF model (Figure S6A-B). Moreover, the relative SLC39A10 expression was positively correlated with lung function while negatively associated with fibrosis related marker (COL7A1, COL4A6) ((Figure S6C-E), suggesting that SLC39A10 may play a protective role during the progression of pulmonary fibrosis. To address the cellular source of SLC39A10, we performed an in-depth analysis of SLC39A10 expression across major lung cell populations using two independent single-cell RNA-seq datasets from IPF patients (GSE136831 and GSE135893). These analyses revealed that macrophages are the only major cell population in which SLC39A10 expression is significantly reduced in IPF compared with healthy controls (Fig. 5L and Figure S6F). In contrast, other structural and immune cell types, including myofibroblasts, alveolar epithelial cells, club cells, dendritic cells, T cells, and goblet cells, did not show significant changes in SLC39A10 expression (Figure S6H-M). Consistent with these transcriptomic findings, immunofluorescence staining of lung sections from IPF patients showed markedly diminished SLC39A10 protein levels specifically in macrophages (Fig. 5M). In summary, these data substantiate that AhR attenuates pro-fibrotic macrophage activation by upregulating the expression of Slc39a10, and highlight Slc39a10 as a potential therapeutic target for pulmonary fibrosis. Administration of kynurenine combined with pirfenidone reverses the established pulmonary fibrosis The above studies indicate that Kyn holds promise as a potential therapeutic target. Intriguingly, oral delivery of pirfenidone markedly reduced the levels of Kyn in PF mouse model (Figure S7A). Furthermore, pirfenidone showed potential in alleviating pro-fibrotic macrophage polarization only at higher concentrations such as 200 μg/mL, but had no effect at the clinical therapeutic concentration (around 10 μg/mL) (Figure S7B) [33]. Notably, vitro studies revealed that the combination of Kyn (50 μg/mL) and pirfenidone (10 μg/mL) showed enhanced inhibitory effect on pro-fibrotic macrophage polarization compared to Kyn (100 μg/mL) or pirfenidone (10 μg/mL) solely group (Figure S7C). Therefore, this result prompted us to investigate whether a combination therapy involving Kyn and pirfenidone could enhance the therapeutic efficacy of pirfenidone. To test this hypothesis, a BLM-induced PF mouse model was utilized. Firstly, a pilot study was conducted to optimize the Kyn dosage. Oral delivery of Kyn (20 mg/kg) led to a marked increase in circulating and lung homogenate Kyn levels, reaching approximately 1.3 μg/mL and 168.57 µg/g within 1 h respectively, followed by a gradual and steady decline (Fig. 6A-B). Then on the 14th day of BLM induction, mice were administered with PBS, pirfenidone (50 mg/kg), Kyn (20 mg/kg), Kyn (20 mg/kg) + pirfenidone (50 mg/kg), or high-dose pirfenidone (100 mg/kg) for 7 days (Fig. 6C). Administration of Kynurenine combined with pirfenidone reverses the established pulmonary fibrosis. A Plasma Kynurenine concentration at different time points (0, 1, 3, 6 and 24 h) after the mice were given 20 mg/kg Kynurenine by intragastric administration. B The levels of Kyn in the lung after oral administration of Kyn (20 mg/kg) at different time points (0, 1, 3, 12 and 24 h). C Illustration of experimental design for KYN and Pirfenidone treatment in BLM induced PF mouse model. Briefly, the BLM mouse model was constructed by intratracheally administering BLM. Mice were administered with KYN and Pirfenidone every day starting from Day 14 after the BLM challenge. Then, all mice were euthanized with an overdose of anesthetic on Day 21 after the BLM challenge. D The effect of KYN and Pirfenidone on mice with BLM-induced PF. H&E staining (top) and Sirius red (middle) and Masson staining (bottom) analyses were conducted to assess the severity of fibrotic changes in the lungs of mice after BLM administration. Images were captured at × 200 magnification. E Semiquantitative Ashcroft scores relevant to the severity of fibrosis. F, G Western blot (F) and RT-PCR (G) analysis of fibronectin and collagen 1 in different group with the indicator treatment. H The protein expression levels of CD206 and Arg1 in mice with BLM and KYN or Pirfenidone treatment. I RT-PCR analysis of CD206, Arg1 and YM1 expression in WT mice with BLM and KYN or Pirfenidone treatment. Five to six mice were included in each experiment group. Data are presented as mean ± SD. *: p < 0.05, **: p < 0.01, ***: p < 0.001 Initially, histological analysis showed that individual administration of Kyn (20 mg/kg) or pirfenidone (50 mg/kg) did not remarkably improve fibrosis compared to the group treated with BLM. However, the combination therapy of Kyn (20 mg/kg) and pirfenidone (50 mg/kg) exhibited remarkable attenuation of BLM-induced PF. Importantly, the therapeutic effect of this combination therapy surpassed that of high-dose pirfenidone (100 mg/kg) treatment (Fig. 6D and E). These findings were further supported by the protein and mRNA expression analysis of fibrotic markers in lung tissue homogenates (Fig. 6F and 6). The activation of pro-fibrotic macrophages in each group was also assessed. The combination therapy of Kyn and pirfenidone led to a significant reduction in pro-fibrotic macrophage polarization, as indicated by the reduced expression of CD206, Arg1, and YM1 compared to high-dose pirfenidone (100 mg/kg) treatment (Fig. 6H and I). Consistently, immunostaining further demonstrated a decreased number of F4/80+CD206+ and F4/80+Agr1+ in Kyn + PFD treated group (Figure S7E and F). Importantly, the expression of TGF-β1 was robustly blunted in the combination therapy treated mice following BLM injection (Figure S7D). Taken together, these data conclusively demonstrate that the combination therapy of Kyn and pirfenidone exhibits a remarkable ability to reverse BLM-induced PF, potentially by inhibiting the activation of pro-fibrotic macrophages. Pirfenidone in combination with Kyn manifested pronounced therapeutic effects in ex vivo human lung precision-cut tissue slice To evaluate the translational potential of PFD combined with Kyn for fibrotic lung disease, we used an ex vivo human precision-cut lung slice (hPCLS) model, which preserves native lung architecture and cell biology and addresses limitations of mouse models. As outlined in Fig. 7A, freshly resected lung tissue was infused with 2% low–melting point agarose and sectioned into 400-μm hPCLS, then cultured in DMEM with 0.5% FBS for 48 h. For hPCLS derived from PF patients, it was incubated with Kyn and/or PFD drugs for another 48 h. Notably, Western blot indicated that PFD plus Kyn treatment significantly reduced the expression of Fibronectin and Collagen 1 in the lung tissue homogenate (Fig. 7B). Additionally, histopathological analyses showed that PFD plus Kyn could partially rescue the fibrotic phenotype in PF patients (Figure S8A). Moreover, immunofluorescence analysis revealed that the number of Fibronectin + and Collagen 1 + cells in the hPCLS from fibrotic patients was significantly reduced in the PFD plus Kyn treated group (Fig. 7C and D). For hPCLS from normal individuals, compared with the PBS group, Western blot results showed that pro-fibrotic cocktail stimulation robustly increased the expression of fibrotic markers (Fibronectin and Collagen 1) in hPCLS, whereas the PFD plus Kyn treatment group markedly reduced collagen deposition (Fig. 7E). Histological analysis and immunofluorescence staining further corroborated this phenotype (Fig. 7F and G, Figure S8C). Notably, immunofluorescence staining results also observed that PFD plus Kyn treatment remarkedly decreased the number of CD206 + macrophages (Figure S8B and D), indicating that PFD plus Kyn as a promising therapeutic strategy for pulmonary fibrosis. Pirfenidone in combination with Kyn manifested pronounced therapeutic effects in ex vivo human lung precision-cut tissue slice. A A flow chart of the preparation of hPCLS. B In hPCLS derived from PF patients, Western blot analysis of Fibronectin and Collagen 1 protein levels in homogenates after treatment with Kyn and/or PFD. C-D Representative immunofluorescence images of Fibronectin and Collagen 1 of different treatment groups in hPCLS derived from IPF patients. E In hPCLS derived from normal control lung tissue, Western blot analysis of Fibronectin and Collagen 1 protein levels in different treatment groups. F-G Representative immunofluorescence images of Fibronectin and Collagen 1 in hPCLS derived from normal control lung tissue. Data are presented as mean ± SD; immunofluorescence images were acquired at × 400 magnification; GAPDH was used as the loading control for normalization; statistical significance was defined as *p < 0.05, **p < 0.01, ***p < 0.001 Discussion PF is an irreversible and lethal lung disease characterized by progressive scarring. Notably, the imbalance of macrophage polarization renders the clinical treatment of PF still challenging [3, 34]. Herein, this study demonstrates that Kyn and KTR in the serum are remarkably upregulated in different types of PF, including IPF, IPAF, NSIP and CTD-ILD patients. Correlation analysis indicates that Kyn levels and KTR were positively correlated with the severity of PF, suggesting the potential utility of serum Kyn and KTR as valuable biological markers for the diagnosis and assessment of disease severity in PF. Furthermore, macrophage serves as one of the main cell types of Kyn production and specific deficiency of Ido1 in macrophages exacerbates BLM-elicited PF by enhancing pro-fibrotic macrophage polarization. Mechanistically, we discovered that Kyn binds to the AhR, facilitating its nuclear translocation. Within the nucleus, AhR promotes the transcription of Slc39a10, leading to increased intracellular zinc ion levels and decreased profibrotic macrophage markers (Fig. 8). Importantly, our study demonstrates that the combination therapy of Kyn and pirfenidone effectively enhances the therapeutic efficacy of pirfenidone in the treatment of PF. A schematic diagram for mechanisms underlying Kyn regulation of pro-fibrotic macrophage polarization. In macrophages, Kyn binds to AhR and facilitates it into the nucleus, where AhR transcribes Slc39a10 expression, leading to increased intracellular zinc ion levels and inhibition of pro-fibrotic macrophage activation. Importantly, the combination therapy of Kyn and pirfenidone effectively enhances the therapeutic efficacy of pirfenidone in the treatment of PF Recent studies have shed light on the crucial role of aberrant amino acid metabolism in the initiation and progression of PF, offering new therapeutic avenues for this condition. Tryptophan (Trp), an essential amino acid, can be regulated through exogenous supplementation [16]. Previous studies have indicated that IPF, silicosis and PF associated with Sjögren's syndrome patients exhibited increased levels of Trp metabolism, highlighting its potential as a therapeutic target [20, 35]. Consistently, we observed aberrant activation of the kynurenine pathway during Trp metabolism not only in IPF patients but also in IPAF, NSIP, CTD-ILD patients. Importantly, we further identified a positive correlation between Kyn levels, KTR, and the severity of PF, indicating the potential utility of serum Kyn and KTR as valuable biological markers for the diagnosis and assessment of disease severity in PF. Metabolic reprogramming have been showed to be strongly linked with macrophage phenotypes switch, including glucose metabolism, amino acid metabolism, and fatty acid metabolism [36]. Specifically, arginase, serine and glycine were implicated in orchestrating pro-inflammatory and pro-fibrotic macrophage activation [37–40]. This led us to focus on the role of Trp metabolism in macrophages. As previously reported, pro-fibrotic macrophages are characterized by a multi-marker expression profile encompassing CD206, Arg1, Trem2, CD163, TGF-β1, and IL-10 [13, 41, 42]. Consistently, RNA-seq datasets revealed significant positive associations between CD206 expression and established fibrosis markers (TGF-β1, FN1) (Figure S9A, B). Moreover, single-cell RNA-seq analyses (GSE135893, GSE136831) show clear segregation of macrophages into CD206− and CD206+ clusters, with the CD206+ population exhibiting significantly higher expression levels of fibrosis-related genes (TGF-β1, FN1, CD163, Trem2) compared to the CD206− subset (Figure S9C, D), indicating a strong pro-fibrotic signature of CD206+ macrophages. Indeed, in this study, we found that specific knockout of Ido1 in macrophages reduced Kyn levels and exacerbate BLM-induced PF by enhancing CD206+TGF-β1+ macrophage activation. Nevertheless, administration of Kyn abrogated pro-fibrotic macrophage polarization, evident by decreased CD206, Arg1 and TGF-β1 expression in vivo and in vitro, which differs from the previously observed role of Kyn in inhibiting fibroblast activation and collagen 1 formation [20, 43]. The next question to address is the mechanism by which Kyn mitigates the formation of pro-fibrotic macrophages. In previous studies, Kyn exerted its effects through AhR, an endogenous receptor [30]. Specifically, Kyn binding to AhR promotes its nuclear translocation and formation of a heterodimer with AhR nuclear translocator protein (ARNT). Subsequently, the AhR/ARNT complex binds to specific dioxin/xenobiotic response elements (DRE/XRE) and regulates the expression of multiple genes [44, 45]. In our study, ChIP-sequence analysis and luciferase reporter gene assay discovered that AhR can bind to the Slc39a10 gene and increase its expression. The Slc39a10 gene encodes the ZIP10 protein, a zinc ion transporter that enhances intracellular zinc ion concentration. Studies have reported that increased zinc ion concentration in macrophages inhibited the secretion of fibrotic cytokines [46, 47]. Indeed, zinc supplementation significantly inhibited IL-4-induced profibrotic macrophage polarization and IL-4/JAK2/STAT6 signaling activation, while knocking down Slc39a10 expression markedly enhanced IL-4-induced macrophage alternative activation. Interestingly, analysis of publicly available databases (GSE47460) revealed a slightly increased expression level of Slc39a10 in the whole lung tissue of IPF patient, with the expression level negatively correlated with disease severity. Furthermore, single-cell RNA sequencing analysis of IPF patient lung tissues revealed a significant decrease in Slc39a10 expression in macrophages. Therefore, enhancing Slc39a10 expression in macrophages may represent a viable strategy for treating PF. Although pirfenidone, the FDA-approved drug for the treatment of PF, has been shown to act on multiple fibrotic pathways, including reducing fibroblast proliferation and inhibiting the TGF-β pathway [4, 8]. However, the effect of pirfenidone on macrophages remains controversial [41, 48–50], which may impede its effective anti-fibrotic effects [5]. In this study, we found that pirfenidone had no effect at the clinical therapeutic concentration (10 μg/mL), while the combination of Kyn and pirfenidone showed enhanced inhibitory effect on pro-fibrotic macrophage polarization. Similarly, we discovered that this combination therapy not only significantly reduces the dosage of pirfenidone but also effectively enhances the therapeutic efficacy of pirfenidone in treating PF. Intriguingly, previous study and our own result have indicated that oral administration of pirfenidone substantially decreased the levels of Kyn in the serum of PF models (Figure S7A), suggesting that the reduction of Kyn levels may servers as markers for disease remission [19]. However, our study challenges this notion by demonstrating that both lung tissue and circulating Kyn levels rising as fibrosis advances, but the deficiency of Ido1 in macrophages exacerbates BLM-induced PF and Kyn supplementation alleviates pulmonary fibrosis. In other words, in the fibrotic microenvironment, although IDO1-dependent Kyn production is upregulated, this intrinsic increase is insufficient in magnitude to counteract the robust profibrotic signals present in advanced stages of disease. Therefore, Kyn derived from Trp via the Kyn pathway likely acts as a compensatory mechanism in the pathogenesis of PF. This concept parallels the role of brain natriuretic peptide in heart failure, where elevated levels indicate disease stress but also provide beneficial counterregulatory effects [51]. To further explain this phenomenon, we observed that AHR, the receptor of Kyn, expression is markedly downregulated in fibrotic lung tissues compared with control subjects [20]. This reduction in AHR expression likely limits the cellular responsiveness to Kyn, thereby diminishing the anti-fibrotic potential of the endogenous Kyn-AHR axis. This phenomenon is similar to type 2 diabetes, in which serum insulin levels are elevated, yet insulin receptor sensitivity is reduced, leading to impaired insulin signaling and persistent hyperglycemia in patients [52]. To further explore the potential mechanism by which pirfenidone reduces Kyn levels, we assessed IDO1 activity by measured the levels of KTR in the serum of IPF patients before/after PFD treatment. The results showed that KTR and Kyn levels were significantly reduced in the serum of IPF patients after PFD treatment (Figure S10C and D). Next, we explored the binding modes and interactions between IDO1 and PFD via molecular docking. As shown in the Figure S10B, IDO1 exhibited a favorable binding energy with PFD (K = −6.5 kcal/mol) and the compound formed a single hydrogen bond with Thr-383 on the protein. In addition, CETSA analysis showed that pirfenidone significantly increased the thermal stability of IDO1 (Figure S10E). Taken together, these results suggest that PFD may suppress IDO1 activity by binding to IDO1. However, the mechanism by which PFD lowers Kyn remains to be elucidated and requires further study. Moreover, whether Kyn administration can further mitigate PFD-associated adverse effects will require future clinical studies incorporating pharmacokinetic profiling, serum biochemistry, behavioral monitoring, and histopathological examination of gastrointestinal and skin tissues, although side effects on major organs were not observed in mouse model (Figure S11A-E). This study also has some limitations. In this study, we employed LyzM-Cre mice to achieve knockout of Ido1 in macrophages, which is not exclusively restricted to macrophages and can also target other myeloid populations [42, 53]. Although rAAV6-F4/80-Cre, which enables efficient targeting of lung-resident macrophages, was employed to confirm the phenotype, employing more macrophage-restricted Cre drivers (e.g., Csf1r-CreER) would further enhance the specificity of lineage tracing and functional deletion. Another limitation is that we observed that supplementation of Kyn can alleviate pulmonary fibrosis by activating AHR, but the long-term safety and potential off-target effects of systemic Kyn or other AhR agonists warrant further investigation. While CD206 and Arg1 are recognized markers of profibrotic macrophages, relying on them alone may not capture the full complexity and dynamics of macrophage subsets within the fibrotic microenvironment. Higher-resolution approaches—such as single-cell multi-omics and spatial transcriptomics—are needed to more precisely delineate macrophage heterogeneity and to determine how IDO1 regulates specific macrophage subsets in vivo. Conclusion In summary, our study offers significant insights into the role of Trp metabolism in PF. We have demonstrated a negative correlation between Kyn, KTR levels with the severity of PF. Additionally, macrophages have been identified as one of the primary cell types involved in Kyn production, and the specific loss of Ido1 expression in macrophages exacerbates BLM-induced PF and enhances pro-fibrotic macrophage polarization. Mechanistically, our results indicate that Kyn binding to the AhR facilitates its nuclear translocation and upregulates the expression of Slc39a10, leading to the inhibition of pro-fibrotic macrophage activation. Importantly, the combination therapy of Kyn and pirfenidone effectively enhances the therapeutic efficacy of pirfenidone in the treatment of PF. These findings underscore the critical role of Kyn in the progression of PF and highlight the synergistic therapeutic approach of orally administering Kyn and pirfenidone for the management of PF. Supplementary Information Supplementary Material 1. Supplementary Material 2. Acknowledgements Acknowledgements The flow chart in Fig. 7A were created with BioRender.com. Abbreviations Pulmonary fibrosis Aryl hydrocarbon receptor Food and Drug Administration Arginase-1 Transforming growth factor-β Platelet-derived growth factor Tryptophan-2,3-dioxygenase Indoleamine 2,3-dioxygenase Idiopathic pulmonary fibrosis Interstitial pneumonia with autoimmune features Non-specific interstitial pneumonia Connective tissue disease-associated interstitial lung disease α-Smooth muscle actin Diffusing Capacity of the Lungs for Carbon Monoxide Forced vital capacity Forced expiratory volume in one second Total lung capacity The receiver operating characteristic The area under the curve Bone marrow-derived macrophages Human lung precision-cut tissue slice Authors’ contributions Huihui Yue, Ruihan Dong, Jianhan He, Fengqin Zhang, Yan Fan, Xuan Li and Xinran Dou performed experiments. Yi Wang, Jianping Zhao provided human lung tissues and clinical data. Qing Zhou, Qilin Yu, Long He, Shu Zhang, Weikuan Gu, Ping Yang and Huilan Zhang were responsible for the experimental design, data analysis, and contributed to the manuscript preparation. Yi Wang, Cong-Yi Wang and Huilan Zhang led the investigation and wrote the manuscript. All authors reviewed the manuscript. Funding This study was supported by the National Natural Science Foundation of China (82170081, 82570101); Tongji Hospital Clinical Research Flagship Program (No.2024TJCR005); the Tongji Hospital Formulation Research and Development Transformation Fund (No. 24-2ZJZ08002-05) and Tongji Hospital's Open Competition Initiative for Key Projects (No.25-2KYC13066-32); the National Key R&D Program of China 2022YFA0806101; National Natural Science Foundation of China (82130023, 81920108009, 91749207, 82070808 and 82270885); Key Laboratory of Endocrine and Metabolic Diseases of Shanxi Province 202104010910009; Fundamental Research Funds for the Central Universities, HUST (2021yjsCXCY097); Integrated Innovative Team for Major Human Diseases Program of Tongji Medical College, Huazhong University of Science and Technology; and the Innovative Funding for Translational Research from Tongji Hospital. Data availability All data necessary for the evaluation of the conclusions presented in this article are provided within the main text and/or its supplementary materials. Further data contained within the manuscript can be obtained from the corresponding author upon request. Declarations Ethics approval and consent to participate All procedures involving human subjects adhered to the guidelines established by the National Institutes of Health (NIH) and were approved by the Institutional Review Board (IRB) of Tongji Hospital (TJ-IRB20210966, TJ-IRB20220736). All animal related experiments were approved by the Animal Experiment Committee (TJH-202205023) and conducted in strict adherence to the guidelines established by the Animal Use and Care Committee of Tongji Hospital, Huazhong University of Science and Technology, Wuhan, China. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Huihui Yue, Ruihan Dong and Jianhan He contributed equally to this work. Contributor Information Yi Wang, Email: wangyi@tjh.tjmu.edu.cn. Cong-Yi Wang, Email: cwang@hbku.edu.qa. Huilan Zhang, Email: huilanz76@hust.edu.cn. References Associated Data Supplementary Materials Supplementary Material 1. Supplementary Material 2. Data Availability Statement All data necessary for the evaluation of the conclusions presented in this article are provided within the main text and/or its supplementary materials. Further data contained within the manuscript can be obtained from the corresponding author upon request.