Myeloid Fbxw7 Prevents Pulmonary Fibrosis by Suppressing TGF-β Production

Idiopathic pulmonary fibrosis (IPF) is a group of chronic interstitial pulmonary diseases characterized by an inexorable decline in lung function with limited treatment options. The abnormal expression of transforming growth factor-β (TGF-β) in profibrotic macrophages is linked to severe pulmonary fibrosis, but the regulation mechanisms of TGF-β expression are incompletely understood. We found that decreased expression of E3 ubiquitin ligase Fbxw7 in peripheral blood mononuclear cells (PBMCs) was significantly related to the severity of pulmonary fibrosis in IPF patients. Fbxw7 is identified to be a crucial suppressing factor for pulmonary fibrosis development and progression in a mouse model induced by intratracheal bleomycin treatment. Myeloid cell-specific Fbxw7 deletion increases pulmonary monocyte-macrophages accumulation in lung tissue, and eventually promotes bleomycin-induced collagen deposition and progressive pulmonary fibrosis. Notably, the expression of TGF-β in profibrotic macrophages was significantly upregulated in myeloid cell-specific Fbxw7 deletion mice after bleomycin treatment. C-Jun has long been regarded as a critical transcription factor of Tgfb1, we clarified that Fbxw7 inhibits the expression of TGF-β in profibrotic macrophages by interacting with c-Jun and mediating its K48-linked ubiquitination and degradation. These findings provide insight into the role of Fbxw7 in the regulation of macrophages during the pathogenesis of pulmonary fibrosis.


INTRODUCTION
Idiopathic pulmonary fibrosis (IPF) is a lung disease characterized by fibroblast hyperproliferation and extracellular matrix (ECM) deposition (1), which can cause pulmonary fibrosis and eventually lead to respiratory failure. However, the etiology of IPF remains unclear. Transforming growth factor-b (TGF-b) is one of the best-characterized pro-fibrotic factors (2). In mammals, there are three isoforms of TGF-b (TGF-b1, TGF-b2, and TGF-b3), of which TGF-b1 most closely related to the development of IPF (2). Disturbances of the pulmonary microenvironment induced by TGF-b are critical to promote the recruitment of circulating fibrocytes and bone marrow-derived progenitor cells to the lung (3), and enhance the pro-fibrotic capacities of these cells or induce their differentiation into activated myofibroblasts (4), eventually leading to the typical IPF symptoms like ECM deposition and abnormal collagen accumulation (5). Moreover, TGF-b1 induces macrophage recruitment and stimulates the expression of several proinflammatory and profibrogenic cytokines in macrophages, such as TNF-a and IL-1b (5). IPF patients have much higher TGF-b1 mRNA expression in profibrotic macrophages compared to normal volunteers (6,7). Previous studies have reported that TGF-b1 expression can be induced by cytokines, such as IL-1b, TNF-a (8), IL-13 (9,10) and IL-24 (11), the increased generation of ROS also induces TGF-b1 production (12). Moreover, the activation of TGF-b1 in fibrotic disorders can be induced by multiple factors, such as matrix metalloproteinases (MMP) 3 and MMP9 (13), thrombospodin-1 (14) and integrins a V b 6 (15). TGF-b1 signaling pathways and their downstream signal transduction mechanisms have been extensively studied, but there are few reports involved in the regulation mechanism of TGF-b1 expression.
Pulmonary macrophages derived from blood monocytes and classified into CD64 + SiglecF + alveolar macrophages (AMs) and CD11b + CD206 + interstitial macrophages (IMs) based on the different anatomical locations (16), both of those macrophage subsets derived from blood monocytes have been known to play a crucial role in IPF pathogenesis. Pulmonary macrophages polarized into alternatively activated anti-inflammatory phenotypes to promote tissue repair and fibrosis after lung injury (17). Aberrant immune function of anti-inflammatory macrophages present profibrogenic phenotypes, which induce the dysfunction of wound repair and accumulate in fibrotic lung, further support fibrogenesis of lung by generating reactive oxygen species (ROS) (6), contributing to collagen synthesis and ECM remodeling. Increased accumulation of profibrotic AMs have been detected in lung of IPF patients (6). During lung fibrogenesis, profibrotic macrophages release soluble factors to create and sustain the profibrotic lung microenvironment (18)(19)(20), such as TGF-b1, CCL8 (21), IGF-I (22), PDGF (23), and arginase 1 (24). In contrast, loss of profibrotic macrophages promotes the resolution of established pulmonary fibrosis (25) and the reduction of lung collagen deposition (26). Importantly, in rodents with bleomycin-induced lung fibrosis, pulmonary macrophages have been demonstrated to be the predominant source of profibrogenic mediator TGF-b (27). However, the specific mechanism of TGF-b1 secreted by macrophages in the pathogenesis of IPF is remain poorly understood.
F-box and WD repeat domain-containing 7 (Fbxw7) is a member of the F-box protein family, which is a component of the SKP1-cullin-F-box-protein (SCF) ubiquitin ligase complex. Many studies have shown that Fbxw7 is a tumor suppressor by targeting c-Myc, Notch, MCL1 and c-Jun (28). Fbxw7 mutation was detected in squamous cell carcinoma developed on IPF (29). Fbxw7 also participates in the regulation of Parkinson's disease (30) and circadian rhythm (31). Moreover, Fbxw7 controls neural stem cell differentiation (32) and myelination (33). Fbxw7 has immunoregulatory capacities, it controls effector T cell polyfunctionality and survival in tumor environment by inhibiting Notch activation (34). Our previous studies have shown that Fbxw7 enhances antiviral innate immune response (35) and aggravate intestinal inflammation (36). These observations implicate that Fbxw7 may act as a transducer of tissue microenvironment, regulating the plastic function of macrophages. In this work, we further investigated the function of Fbxw7 in pulmonary macrophages during the pathogenesis of IPF. Our findings demonstrate that Fbxw7 expression decreased significantly in peripheral blood mononuclear cells (PBMCs) of IPF patients and lung tissues of the mice model of bleomycin-induced fibrosis. Fbxw7 deletion in myeloid cells increased the release of fibrogenic factors and the recruitment of monocytes into lung tissues. In terms of the mechanism, the Fbxw7 deletion promoted the TGF-b expression in macrophages through reducing c-Jun ubiquitination, thereby aggravating the severity of pulmonary fibrosis.

Mice Information
Fbxw7 fl/fl mice (C57BL/6J background) were obtained from Jackson Laboratories. LysM-Cre mice C57BL/6J were kindly provided by Dr. Ximei Wu (Zhejiang University School of Medicine, Hangzhou, China). LysM + Fbxw7 fl/fl mice were generated by crossing Fbxw7 f/f mice with LysM-Cre transgenic mice as previously described (36). All mice were bred at the laboratory animal center of Zhejiang University under specific pathogen-free conditions. In all experiments, age-matched (8)(9)(10)(11)(12) week) mice were randomized into the experimental or control groups. All animal experiments were performed according to the protocol approved by the Animal Ethics Committee of Zhejiang University and were in compliance with institutional guidelines.
Recombinant vectors encoding mouse c-Jun (NM_010591) were cloned into pcDNA3.1-c-Myc eukaryotic expression vector. The RAW264.7 cells (3 × 10 5 cells/well, in 24-well plate) were incubated with NATE ™ (In vivogen, lyec-nate) for 30 min to enhance the transfection efficiency and transient expression of target plasmid. Then RAW264.7 cells were transfected with c-Jun plasmid (Roche, FuGENE HD Reagent) and treated with IL-4 for the indicated time.

Co-Culture Experiments
MEF cells (2×10 5 ) were cultured in the lower chamber and macrophages (5×10 5 ) were added to the upper chamber of a 12-well transwell apparatus (0.4 mm pore size, Costar, Cambridge, MA). The MEF were subjected to the further analysis at 48 h after co-culturing.

Induction of Pulmonary Fibrosis
An experimental pulmonary fibrosis model was established using the direct endotracheal injection of bleomycin (37). Briefly, 6-to 8-week-old C57BL/6 mice were anesthetized with pentobarbital by intraperitoneal injections. Neck area of mice was sterilized using povidone-iodine swabs after depilation of hairs, and 1 cm midline incision was made with sterile scissors. Intratracheal administration of 3.5 mg/kg bleomycin hydrochloride (Hisun-Pfizer Pharmaceuticals, Taizhou, China) in sterile PBS, an equal volume of sterile PBS for controls by using 1 ml syringe fitted with a 26 G needle (carefully insert the needle into the visualized trachea and rapidly inject the bleomycin solution during a single inspiration). The incision was closed with surgical sutures after withdrawing the needle. The mice were placed on warmer pads and monitored until fully awake. Mouse lung tissues were sacrificed at 14 days and 21 days post-bleomycin injection for subsequent experiments.
LPS induced pulmonary injury model (38,39) were established using the direct endotracheal injection of 5 mg/kg LPS (Sigma-Aldrich, St. Louis, Missouri, USA). Mouse lung tissues were sacrificed at 10 days post-LPS injection for subsequent experiments.

Flow Cytometry
Cells were stained with fluorochrome-labeled mAbs and analyzed with an LSRII flow cytometer (BD). Flow cytometric analysis was performed using FlowJo software. The following fluorochrome-labeled mAbs were purchased from Biolegend, Inc. and used according to the manufacturers' protocols: anti-

Lung Histology
Lungs from bleomycin-treated or control mice were dissected, fixed in 10% phosphate-buffered formalin, and embedded into paraffin. Tissue sections were stained with H&E for morphological analysis and with Masson's trichrome for the detection of collagen fibers according to the manufacturers' instructions (Nanjing Jiancheng, Nanjing, China). The Szapiel's score was used to evaluate the severity of fibrosis using the following criteria (40): none (0), no evidence of fibrosis; mild (1), <20% of the lung affected; moderate (2), 20%-50% affected; severe (3), >50% affected.

Hydroxyproline Assay
To assess collagen synthesis in the lung, pulmonic hydroxyproline levels were determined using a hydroxyproline assay kit obtained from Jiancheng Bioengineering Institute (Nanjing, China), according to manufacturer's instructions. Hydroxyproline content was expressed as mg/mg lung wet weight.

Quantitative RT-PCR
Total RNA was extracted from tissues or cells using RNAiso Plus (TakaRa), and then reversely transcribed to cDNA using the PrimeScript RT reagent kit (TaKaRa) with random primers according to the manufacturer's instructions. Quantitative PCR was performed using KAPA SYBR Green (SYBR Green Fast qPCR Master Mix) on CFX96 Touch ™ Real-Time PCR Detection System (BioRad). Expression data were normalized to the mRNA levels of the b-actin gene and calculated using the 2 −DDCt method (41). The sequences for primers were listed in Supplementary Table 1.

ELISA
The levels of mouse TGF-b in bronchoalveolar lavage fluid (BALF) or culture supernatants were measured with ELISA kits (Invitrogen, 88-8350-88) according to the manufacturer's instructions.

Statistical Analysis
Statistical analysis was performed with GraphPad Prism 7.0. All data are shown as mean ± SD. The comparisons of two experimental groups were determined using an unpaired, 2-tailed student's t-test. Multiple comparisons were assessed by one-way ANOVA with Tukey's multiple comparisons test. P-values less than 0.05 were considered significant.

Decreased Expression of Fbxw7 Is Related to IPF
We first investigated whether Fbxw7 was related to the development of IPF, the expression level of Fbxw7 from IPF patients was analyzed in the data from Gene Expression Omnibus (GEO) database (GSE70867) (42). The result showed that Fbxw7 mRNA expression in bronchoalveolar lavage (BAL) cells of IPF patients was significantly lower than that in healthy controls ( Figure 1A). In another database of microarray expression profiling of PBMCs from IPF patients in the training cohort (GSE132607) (43), Fbxw7 mRNA expression level in 12-months IPF patients was significantly lower than that in baseline, 4-months, and 8months IPF patients ( Figure 1B). We then used the bleomycin-induced mouse model of pulmonary fibrosis to observe the expression of Fbxw7 in lung tissues ( Figure 1C) and peripheral blood monocytes ( Figure 1D). qRT-PCR results showed that the expression level of Fbxw7 was significantly reduced after 21 days of intratracheal bleomycin administration compared to the PBS controls. Moreover, the expression of Fbxw7 in the macrophages in mice pulmonary fibrosis tissue was markedly decreased compared with those in healthy control subjects detected by immunofluorescence staining ( Figure 1E). These results suggest that the decreased expression of Fbxw7 is associated with the formation of pulmonary fibrosis.

Myeloid Deficiency of Fbxw7 Aggravates Bleomycin-Induced Collagen Deposition in Lung Tissue
Augmented collagen deposition within the interstitium (44,45) and the remodeling of ECM (46)  LysM + Fbxw7 fl/fl mice and their Fbxw7 fl/fl littermates were subjected to establish experimental pulmonary fibrosis model. Masson staining of lung tissues showed that collagen deposition was significantly increased in LysM + Fbxw7 fl/fl mice 14 days after bleomycin administration, compared with Fbxw7 fl/fl mice (Supplemental Figure 2A), and further increased after 21 days (Figure 2A). Szapiel scores also showed that Fbxw7 knockout exacerbated the degree of fibrosis (Supplemental Figure 2B and Figure 2B). The expression of collagen III and a-SMA observed by immune-histochemical staining was also significantly higher in the fibrotic lung tissue of LysM + Fbxw7 fl/fl mice compared with Fbxw7 fl/fl littermates ( Figure 2C). Furthermore, LysM + Fbxw7 fl/fl mice had significantly higher levels of hydroxyproline in lung tissues compared with Fbxw7 fl/fl mice ( Figure 2D).
Moreover, we examined the mRNA expression levels of collagen-related genes in the lung tissues after bleomycin administration. Fibronectin 1 (Fn1), a pro-fibrotic marker, is an important member of the fibronectin family (47). Col1a1 and Col3a1 is component of the ECM in IPF. qRT-PCR results showed that the expression of Fn1 and Col3a1 increased significantly in LysM + Fbxw7 fl/fl mice (Supplemental Figure 2C and Figure 2E). Furthermore, as a surface molecular marker of myofibroblasts, the mRNA expression of a-smooth muscle actin (a-SMA or Acta2) was also significantly increased in LysM + Fbxw7 fl / fl mice (Supplemental Figure 2D and Figure 2F). Matrix metalloproteinases (MMPs) and their specific inhibitors (tissue inhibitors of metalloproteinases, TIMPs) participate in the formation and degradation of ECM, and the imbalance between MMPs and TIMPs plays a pivotal role in pulmonary fibrosis pathogenesis. qRT-PCR results demonstrated that Fbxw7 deletion significantly increased the expression of Mmp9 and Timp1 ( Figure 2G). These results suggest that the myeloid cell specific Fbxw7 knockout mice have increased collagen deposition in their lung tissue that may aggravate the degree of fibrosis induced by bleomycin.

Myeloid Deficiency of Fbxw7 Aggravates Bleomycin-Induced Lung Injury and Monocytes Recruitment
Tissue damage and inflammation are important triggers for pulmonary fibrosis (48). To determine whether Fbxw7 deletion in myeloid cells aggravates lung injury and inflammation, we evaluated the morphological characteristics in mouse lung tissues. The H & E staining results showed that the lung tissues of Lysm + Fbxw7 fl/fl mice were more severely damaged 14 days (Supplemental Figure 3A) and 21 days ( Figure 3A) after bleomycin administration. Lung tissues of LysM + Fbxw7 fl/fl mice appeared to be denser, congestion and edema, as well as loss of alveolar structures compared with Fbxw7 fl/fl mice. Moreover, we examined the mRNA expression of surfactant protein-D (Sftpd) and qRT-PCR results showed that myeloid deficiency of Fbxw7 resulted in higher levels of Sftpd in fibrotic lung tissues (Supplemental Figure 3B and Figure 3B). Furthermore, LysM + Fbxw7 fl/fl mice had higher degree of inflammatory cell infiltration in the alveolar space observed by H & E staining ( Figure 3A). Flow cytometric analysis was used to investigate the changes in mononuclear phagocyte subpopulations in bronchoalveolar lavage fluid (BALF) and lung tissue during pulmonary fibrosis development. After 21 days of bleomycin-induced pulmonary fibrosis in mice, Fbxw7 knockout did not increase CD64 + SiglecF + AMs accumulation (Supplemental Figure 3C), but significantly increased CD11b + Ly6C + monocytes recruitment ( Figure 3C) in BALF. Moreover, a single-cell suspension was prepared by digesting lung tissues and analyzed by flow cytometry. Compared with the PBS group, the percentages of CD11b + CD206 + IMs ( Figure 3D) and CD11b + Ly6C + monocytes ( Figure 3E) in lung tissues of mice treated with bleomycin were increased. Importantly, LysM + Fbxw7 fl / fl mice showed an obviously increased accumulation of IM and monocyte cell populations compared with Fbxw7 fl/fl littermates ( Figures 3D, E). These results suggest that the deletion of Fbxw7 exacerbates injury-induced lung fibrosis by promoting the recruitment and accumulation of mononuclear phagocytes.

Fbxw7 Deficiency Upregulates TGF-b Expression in Macrophages
Macrophages are the major source of multiple profibrotic mediators in pulmonary fibrosis (49). We next investigated whether myeloid Fbxw7 deletion altered the cytokine profile in fibrotic lung tissues. The qRT-PCR results showed that the mRNA expressions of Tnf-a, and Il1b in lung tissues of LysM + Fbxw7 fl/fl mice was significantly higher than that of Fbxw7 fl/fl mice 14 days after bleomycin administration, but the expression of Il6 and Il10 was not significantly different (Supplemental Figure 4A). In contrast, LysM + Fbxw7 fl/fl mice had significantly higher levels of Il6 and Il1b, but not Tnf-a and Il10, compared with Fbxw7 fl/fl mice 21 days after administration of bleomycin (Supplemental Figure 4B). Furthermore, we evaluated cytokines expression in alveolar macrophages and found that the mRNA expressions of Il1b, Tnfa, Il6 and Il10 were significantly increased after bleomycin administration, however, there were no significant differences in the expression level between LysM + Fbxw7 fl/fl mice and Fbxw7 fl/fl littermates ( Figure 4A). Interestingly, mRNA expression of Tgfb1, the macrophage-derived fibrogenic cytokine, was significantly increased in lung tissues and alveolar macrophages in LysM + Fbxw7 fl/fl mice compared to Fbxw7 fl/fl littermates ( Figure 4B). ELISA further confirmed that the concentration of TGF-b protein in BALF from LysM + Fbxw7 fl/fl mice was significantly higher than that from Fbxw7 fl/fl littermates ( Figure 4C). In lung injury model established by LPS, more severe lung injury, fibrosis associated collagen deposition as well as higher expression of Tgfb1 were found in lung tissue and BAL cells from LysM + Fbxw7 fl/fl mice compared to Fbxw7 fl/fl littermates (Supplemental Figures 4C, D and Figure 4D). The significantly elevated concentration of TGF-b protein in BALF and serum from LysM + Fbxw7 fl/fl mice was observed by ELISA compared with that from Fbxw7 fl/fl littermates ( Figure 4E). These data suggest that deficiency of Fbxw7 in macrophages exacerbates pulmonary fibrosis, which may be related to the up-regulation of TGFb expression.
IL-4, as a Th2 cytokine, was reported to induce TGF-b expression in profibrotic macrophage to activate fibroblasts and promotes extensive tissue fibrosis (50,51). qRT-PCR confirmed that the expression of Il4 was abundant in fibrotic lung tissues, but the level of Il4 was similar between LysM + Fbxw7 fl/fl mice and their Fbxw7 fl/fl littermates (Supplemental Figure 4E). Next, we performed a Transwell co-culture assay to evaluate the profibrogenic capability of macrophages from LysM + Fbxw7 fl/fl mice and Fbxw7 fl/fl littermates after IL-4 treatment in vitro ( Figure 4F), the results consistently showed that myeloid Fbxw7 deficiency resulted in higher secretion levels of TGF-b in cocultured macrophages after IL-4 treatment ( Figure 4G), Fbxw7 deficient macrophages treated with IL-4 mediates a higher activation of co-cultured fibroblasts by the examination of the expression levels of collagen-I and collagen-III through western blot ( Figure 4H). To further confirm that the enhanced profibrogenic capability of macrophages from LysM + Fbxw7 fl/fl mice is TGF-b-dependent, transwell co-culture assay was performed after IL-4 treatment with or without TGF-b neutralizing antibody. Neutralization of TGF-b abrogated the difference of profibrogenic ability between LysM + Fbxw7 fl/fl macrophages and Fbxw7 fl/ fl macrophages ( Figure 4I). Collectively, these data further support that Fbxw7 deletion in myeloid cells can exacerbate bleomycin-induced production of TGF-b and promote the development of pulmonary fibrosis.

Fbxw7 Regulates TGF-b Expression by Mediating the Ubiquitination and Degradation of c-Jun
Transcription factors including early growth response protein 1 (EGR-1) (52, 53) and NF-kB (54) have been reported to be involved in the pathogenesis of fibrosis via enhancing TGF-b dependent signaling. To further elucidate the mechanisms how Fbxw7 regulates TGF-b production in profibrotic macrophages, we investigated the effect of Fbxw7 deletion on the protein levels of EGR-1 and NF-kB by western blot. The protein levels of EGR1 and NF-kB in alveolar macrophages after bleomycin stimulation were not significantly changed between Fbxw7 fl/fl and LysM + Fbxw7 fl/fl mice, but the level of c-Jun was significantly higher in LysM + Fbxw7 fl/fl mice than that in Fbxw7 fl/fl mice ( Figure 5A). Subsequently, we found significantly higher protein expression of c-Jun in IL-4-stimulated BMDMs after Fbxw7 deletion ( Figure 5B). We then overexpressed c-Jun in RAW264.7 cells (Supplemental Figure 5A) and analyzed the expression of TGF-b by qRT-PCR and ELISA, the results showed that the expression level of TGF-b was significantly increased after c-Jun overexpression (Supplemental Figure 5B), indicating that c-Jun expression is sufficient for TGF-b in macrophages. SP600125 is a specific JNK inhibitor that competes with ATP to inhibit the phosphorylation of c-Jun (55). qRT-PCR analysis and ELISA results showed that the mRNA and protein expression of TGF-b increased significantly after the deletion of Fbxw7 ( Figures 4C, D), but this effect was abrogated after inhibition of c-Jun phosphorylation by SP600125 ( Figure 5C). The MEF cells, co-cultured with peritoneal macrophage from LysM + Fbxw7 fl/fl mice or Fbxw7 fl/fl littermates after inhibiting the phosphorylation of c-Jun, showed similar protein levels of collagen-I and collagen-III, which were detected by western blot (Supplemental Figures 5C, D). Taken together, Fbxw7 deletion leads to the upregulation of TGF-b expression and severity of pulmonary fibrosis through the regulation of c-Jun. It has been demonstrated that the transcription factor c-Jun can be degraded by the ubiquitin-proteasome system as a substrate for Fbxw7 (56). As expected, western blot analysis showed that the degradation of c-Jun protein could be inhibited by protease inhibitor MG132 ( Figure 5D). A cycloheximide chase assay showed that Fbxw7 deletion extended the half-life of endogenous c-Jun protein in BMDMs ( Figure 5E). Immunoprecipitation assay results showed that Fbxw7 can physically bind to c-Jun protein ( Figure 5F) and induce K48linked polyubiquitination of c-Jun ( Figure 5G). Moreover, immunoprecipitation assay was performed by using K48-Ub to pull-down c-Jun and the peritoneal macrophages from LysM + Fbxw7 f/f mice showed weaker ability of mediating c-Jun ubiquitination and proteasomal degradation compared with cells from Fbxw7 f/f littermates ( Figure 5H). This result further revealed that the myeloid Fbxw7 deficiency leads to a significantly accumulation of c-Jun by abolishing its K48 ubiquitination. Collectively, these data confirmed that Fbxw7 promotes TGF-b expression by inducing K48-linked polyubiquitination and proteasome degradation of c-Jun in macrophages.

DISCUSSION
The E3 ubiquitin ligase Fbxw7 has been well characterized as a critical tumor suppressor that can mediate the degradation of various oncoproteins by the ubiquitin-proteasome system (57). Our studies and others have presented compelling evidence that Fbxw7 plays a crucial role in inflammation and antivirus immune response through regulating the function of macrophages (35,36,58). In this study, we demonstrated that the deletion of Fbxw7 in the myeloid lineages aggravates bleomycin-induced lung injury, collagen deposition, and monocyte recruitment, and eventually leads to the exacerbation of pulmonary fibrosis in mice. We revealed its underlying mechanism that Fbxw7 deletion will upregulate TGFb expression by inhibiting the degradation of c-Jun in macrophages. These findings will be beneficial for understanding the role of innate immune response in the pathogenesis of IPF.
Macrophages are the most abundant immune cell population in the lung environment and are involved in tissue repair and lung fibrosis progression, it has also been recognized to play a significant role in driving the fibrogenesis and matrix remodeling. Given the high plasticity of lung macrophages, the relative roles of tissue-resident alveolar macrophages, interstitial macrophages and monocyte-derived infiltrating macrophages in pulmonary fibrosis is complicated (59). High numbers of both interstitial and alveolar macrophages were detected in pulmonary fibrosis (24), the depletion of macrophages, including alveolar and infiltrating macrophage, reduces fibrogenesis (25,60). Experimental fibrosis leads to the recruitment of bone marrow-derived AMs that are more profibrogenic than their disrupted resident counterparts (61). Selectively targeting alveolar macrophage differentiation without affecting global monocytes or tissue-resident alveolar macrophage may ameliorate fibrosis depletion, while the depletion of tissue-resident AMs cannot alter fibrosis severity (61). In addition, IMs are also thought to arise from monocytes, and highly expression of monocyte-related genes, such as Cd14, Cd163, and Csfr1 (62), activated IMs have been reported to induce myofibroblastic activation and ECM production (24). Furthermore, those more profibrogenic macrophages which play a vital role in airway remodeling in pulmonary fibrosis biases alternatively activated macrophages, so we used IL-4-stimulated BMDMs for in vitro experiments to mimic profibrogenic macrophages in vivo. Our data showed that more monocytes are recruited into the lung tissues and BALF of Fbxw7-KO mice, these monocytes further differentiate into profibrogenic macrophages and then participate in the development of pulmonary fibrosis.
The abundant expression of Fbxw7 mRNA and protein can be detected in most human and mice tissues. The decreased level of Fbxw7 mRNA in IPF may be regulated by many factors, including DNA or RNA silencing. The expression of microRNA-155 (miR-155) has been reported to be correlated to the damage degree of IPF (63), and miR-155 reduces the levels of Fbxw7 mRNA (64). The Fbxw7 promoter can be regulated through epigenetic mechanism during lung fibrosis, including histone modifications. Aberrant elevation of DNA methyltransferase 1 (Dnmt1) can mediate the hypermethylation of Fbxw7 promoter and result in decreased gene level of Fbxw7 (65), subsequently contributes significantly to the development of pulmonary fibrosis (66). Moreover, the expression of enhancer of zeste homolog 2 (Ezh2) is increased in IPF patients, increased Ezh2 recruitment and hypermethylated H3K27 at the promoter of fibrotic related genes have been demonstrated, such as Cox-2 (67) and Cxcl10 (68). We speculate the function of Ezh2 in inducing the increase of H3K27me3 at Fbxw7 promotor and silencing of Fbxw7 gene (34) might also be involved in regulating pulmonary fibrosis. IL-4 has been shown to induce TGF-b expression in profibrotic macrophage to activate fibroblasts and promotes extensive tissue fibrosis (50). In our study, we found IL-4 was highly expressed in lung fibrosis tissue of bleomycin-induced mice, and the expression of Fbxw7 was significantly decreased in bleomycin-induced mice lung issue and in BMDMs after IL-4 stimulation as shown in Figures 1, S4B and 5G. These findings suggest that IL-4 in the profibrotic microenvironment might downregulate Fbxw7 expression and block c-Jun degradation.
TGF-b can be produced by a wide variety of cell types, including lung macrophages, alveolar epithelial cells, endothelial cells, fibroblasts and myofibroblasts. However, lung macrophages are revealed to be the main source of TGF-b (6,69). In bleomycin induced pulmonary fibrosis, macrophages induce fibroblast differentiation and proliferation by producing TGF-b, thereby exacerbating pulmonary fibrosis (70,71). The widespread accumulation of collagen in lungs, which was observed in WT mice, disappeared in the myeloid TGF-b deficient mice, which showed similar lung architecture to saline controls (6). Several factors have previously been implicated in the transcriptional induction of Tgfb1, bacterial induction of Tgfb1 mRNA expression in human intestinal epithelial cells is butyrate and SP1-dependent (72). P38/MAPK signal, MEK-1 and JNK are involved in Tgfb1 transcription (73). The higher mRNA expression of Tgfb1 was detected in Egr-1 overexpressed mesangial cell compared with control cells (74). Moreover, p65 and c-Jun were recruited to the promoter of Tgfb1 in A549 cells upon IL-1b stimulation, this result suggesting the role of NF-kB and AP-1 in regulating Tgfb1 expression (75).
C-Jun protein is a component of the dimeric transcription factor activator protein 1 (AP-1) (76), it has been shown to play essential roles in most fibrotic conditions, which can activate multiple signaling pathways, resulting in the expression of many fibrogenesis-associated genes (77). C-Jun N-terminal kinase JNK1 has been reported to promote fibrosis in the lung (78), inhibition of c-Jun phosphorylation in vivo reduced fibrosis and mRNA levels of Tgfb1 in the unilateral ischemia reperfusion injury model (79). C-Jun expression in non-parenchymal liver cells (NPLCs) promotes the fibrogenesis and was particularly correlated with liver fibrosis during non-alcoholic steatohepatitis (80). Moreover, studies have shown that c-Jun can be degraded by the ubiquitin-proteasome system as the substrate of Fbxw7 (81,82), which was also confirmed by our study. Our data show that deletion of Fbxw7 in macrophages does not affect the levels of P65, EGR-1 and their phosphorylation, but significantly increases the level of AP-1 transcription factor subunit c-Jun.
Notably, the published data from Wang et al. (83) showed that FBXW7 deficiency prevents fibrogenesis by increasing telomere capping, which is contrary to the conclusion of our finding. Several potential possibilities exist for this discrepancy. Wang et al. used a mouse model of Fbw7 conditional knockout in pulmonary alveolar epithelial type 2 (AEC2) stem cells, while we used a model of Fbxw7 conditional knockout in myeloid cells. Those reverse effects of Fbxw7 in AEC2 stem cells and myeloid cells in regulating lung fibrosis indicated that the pathogenesis of IPF is complicated, it involves the abnormal repair of lung tissue injury by alveolar epithelial cells, the disordered differentiation and proliferation of fibroblasts, or the hyperactivated recruitment and activation of innate immune cells. The dysfunction in any cell type can lead to different outcomes of IPF. Moreover, subtle genetic background differences in the strains of the mice used and microbiome differences between the mouse colonies may also contribute to these discrepancies.
In conclusion, our study revealed that Fbxw7 plays a crucial role in regulating pulmonary fibrosis progression. Myeloidspecific deletion of Fbxw7 exacerbates pulmonary fibrosis by inhibiting c-Jun degradation and upregulating TGF-b expression in macrophages.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The animal study was reviewed and approved by Medical Experimental Animal Care Commission of Zhejiang University.