MicroRNA‐205‐5p targets E2F1 to promote autophagy and inhibit pulmonary fibrosis in silicosis through impairing SKP2‐mediated Beclin1 ubiquitination

Abstract Silicosis is an occupational disease characterized by extensive pulmonary fibrosis, and the underlying pathological process remains uncertain. Herein, we explored the molecular mechanism by which microRNA‐205‐5p (miR‐205‐5p) affects the autophagy of alveolar macrophages (AMs) and pulmonary fibrosis in mice with silicosis through the E2F transcription factor 1 (E2F1)/S‐phase kinase‐associated protein 2 (SKP2)/Beclin1 axis. Alveolar macrophages (MH‐S cells) were exposed to crystalline silica (CS) to develop an in vitro model, and mice were treated with CS to establish an in vivo model. Decreased Beclin1 and increased SKP2 and E2F1 were identified in mice with silicosis. We silenced or overexpressed miR‐205‐5p, E2F1, SKP2 and Beclin1 to investigate their potential roles in pulmonary fibrosis in vivo and autophagy in vitro. Recombinant adenovirus mRFP‐GFP‐LC3 was transduced into the MH‐S cells to assay autophagic flow. Knocking down Beclin1 promoted pulmonary fibrosis and suppressed the autophagy. Co‐immunoprecipitation and ubiquitination assays suggested that SKP2 induced K48‐linked ubiquitination of Beclin1. Furthermore, chromatin immunoprecipitation‐PCR revealed the site where E2F1 bound to the SKP2 promoter between 1638 bp and 1645 bp. As shown by dual‐luciferase reporter gene assay, the transfection with miR‐205‐5p mimic inhibited the luciferase activity of the wild‐type E2F1 3′untranslated region, suggesting that miR‐205‐5p targeted E2F1. Additionally, miR‐205‐5p overexpression increased autophagy and reduced the pulmonary fibrosis, while overexpression of E2F1 or SKP2 or inhibition of Beclin1 could annul this effect. The current study elucidated that miR‐205‐5p targeted E2F1, thereby inhibiting SKP2‐mediated Beclin1 ubiquitination to promote macrophage autophagy and inhibit pulmonary fibrosis in mice with silicosis.


| INTRODUC TI ON
Silicosis is one of the most common occupational respiratory diseases that can manifest in respiratory failure or fatality in severe circumstances. 1 The fundamental cause of the disease is the inhalation of crystalline silica (CS), which progresses to fibrosis in lung parenchyma. 2 Fibrosis can evidently impair and affect the normal function of lung tissues for essential gas exchange and oxygen supply. 3 Although effective prevention has been impactful, silicosis still persists as a global health concern, especially in China. In 2016, the statistics reported that a total of 27992 people suffered from pneumoconiosis, accounting for 88.06% of occupational diseases, with a detrimental increase of over 1000 cases than in 2015. 4 Currently, in addition to the clinically adopted potential lung transplantation, extracellular vesicles and lung spheroid cells have been reported to serve as therapeutic modalities for lung regeneration, yet the clinical translation remains challenging currently, highlighting the urgent need for the development of other effective treatments. [5][6][7] On account of these literature reports, the identification of vital molecules involved in silicosis is urgent and necessitated for improving the clinical outcome.
Autophagy is a highly conservative catabolic process, instrumental for tissue homeostasis, with vital functionality in regulating inflammation and apoptosis. 8 Essentially, autophagy is a protective mechanism against different diseases, such as cancer, infectious diseases and lung injury. [9][10][11] Additionally, existing evidence has ascertained the significance of autophagy in the development of silicosis. 12 However, whether and how autophagy is involved in the progress of silicosis remain uncertain.
Beclin1 is the mammalian orthologue of yeast autophagy-related gene-6, regarded for involvement in the autophagy process. 13 Its concentration increases under cellular stress, interacts with several cofactors and induces autophagy. 14 Beclin1 was reported to affect the silicosis progression through regulation of autophagy of alveolar macrophages (AMs). 15 Accordingly, the manifestation of reduced autophagic flux in alveolar epithelial cells was regarded to be crucial for CS-induced silicosis, leading to apoptosis and pulmonary fibrosis. 16 Increasing evidence has ascertained the deregulation of mi-croRNAs (miRNAs) in various pathological processes such as pulmonary fibrosis and heart failure. 17,18 Changes in miRNA expression may be consequent for alteration of transcriptional activity of genes in immune-mediated lung disease and are therefore closely associated with respiratory diseases such as silicosis. 19 An existing study reported that miR-1224-5p affects CS-induced pulmonary fibrosis by targeting Beclin1. 20 After bioinformatics analysis, we identified that miR-205-5p might be an upstream miRNA regulating E2F transcription factor 1 (E2F1), which could potentially be the transcription factor of S-phase kinase-associated protein 2 (SKP2). As SKP2 has been reported to mediate the ubiquitination of Beclin1, 21 we speculated the involvement of miR-205-5p in autophagy and pulmonary fibrosis through mediating the E2F1/ SKP2/Beclin1 axis.
To testify this hypothesis, we performed in vitro and in vivo experiments in MH-S cells and mice with silicosis. Our findings revealed that miR-205-5p impaired pulmonary fibrosis and E2F1/ SKP2/Beclin1 axis was responsible for the role of miR-205-5p in such process. Therefore, miR-205-5p may be a promising target for the development of treatment protocols to inhibit silicosis progression.

| Ethical statement
The study involving human beings was conducted with approval of the Ethics Committee of Tangshan People's Hospital. The informed consents were obtained from all the patients and donors or their legal guardians prior to our investigation. The procedures were con-

| Bioinformatics analysis
The pneumosilicosis fibrosis-related gene expression microarray (GSE11 0711) data were obtained from the Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/gds), after which the 'edgeR' software package in R language was adopted for differential analysis. In microarray data set GSE11 0711, 3 normal untreated mouse samples and 3 silicosis model mouse samples were screened for the significant differential genes with |logFoldChange| >1, and p < 0.05 set as the threshold. The target genes of miR-205 were predicted using a combination of the database starBase (http://starb ase.sysu.edu.cn/index.php),

| Clinical samples
Clinical samples including the diseased tissues from silicosis patients and normal tissues from healthy organ donors (n = 10) were collected from the Tangshan People's Hospital. All diagnoses were confirmed by histopathological assessment with haematoxylin-eosin (HE) and Masson staining. 22
From each section, five high power fields of vision were randomly selected with 100 cells from each field. Each experiment was conducted three times independently. To evaluate the alveolar fibrosis, mouse lung tissue specimens were sectioned, fixed on glass slides and subjected to HE staining. To further assess the degree of fibrosis, the tissues were sliced into 6µm-thick sections and stained with the Sirius red staining kit (G1470, Beijing Solarbio Science & Technology Co. Ltd.).

| RT-qPCR
The total RNA content was extracted using the TRIzol reagent Ltd.) containing the universal PCR primers and universal U6 primers was used to obtain a cDNA library of the miRNA containing PolyA tail. The synthesized cDNA was subjected to treatment with the Fast SYBR Green PCR kit (Applied Biosystems, CA, USA) and ABI PRISM 7300 RT-PCR system (Applied Biosystems) for qPCR detection, and 3 replicates were set for each well. The relative expression of miR-NAs and mRNAs was calculated based on the 2 −ΔΔCt method with U6 or β-actin serving as the internal reference. ΔΔCt = (Ct target gene -Ct internal reference ) experimental group -(Ct target gene -Ct internal reference ) control group . U6 and miRNA primers were purchased from Ribobio. Other primer designs are shown in Table S1.

| Western blot analysis
The cells were collected by trypsinization and lysed with the enhanced radioimmunoprecipitation assay (RIPA) buffer containing specific protease inhibitors (BOSTER Biological Technology Co., Ltd.) followed by protein concentration determination using the bicinchoninic acid protein quantification kit (BOSTER).
Proteins were separated by 10% sodium dodecyl sulphate (SDS)polyacrylamide gel electrophoresis, after which the separated proteins were electro-transferred onto polyvinylidene fluoride membranes. The membrane was subjected to a blockade using 5% bovine serum albumin at room temperature for 2 h and probed with the primary antibodies at 4°C overnight and HRP-conjugated goat anti-rabbit or goat anti-mouse secondary antibody for 1 h. After antibody incubation, the membranes were developed using the enhanced chemiluminescence reagents (Millipore, Massachusetts, USA) followed by X-film exposure. Quantification of the bands was performed using the ImageJ software and normalized to βactin. All detailed information regarding the antibodies is shown in Table S2.

| Immunoprecipitation (IP)
After trypsinization, the cells (10 7 -10 8 ) were collected, lysed by RIPA buffer and centrifuged at 4°C, 7000 g for 20 min followed by collection of the supernatant. Then, 100 µl supernatant was regarded as Input, while the remaining supernatant was incubated with Protein A + G Agarose at 4°C for 1 h for pre-clearance and centrifuged for transfer of equal volumes of the supernatant to two EP tubes followed by incubation with protein antibody prepared for testing (Table S2) or the corresponding isotype-matched IgG antibodies, respectively, at 4°C overnight. In the ubiquitination assay, the corresponding plasmids were added in compliance with the above-mentioned steps with the flag-labelled anti-Beclin1/ anti-recombinant Beclin1 antibody in different groups. Then, each tube was incubated with Protein A + G Agarose at 4°C for 1-2 h and centrifuged. The precipitate was rinsed with IP Wash Buffer.
The precipitate and Input were added with the corresponding amount of 2 × SDS Buffer, boiled and subjected to Western blot assay.

| Chromatin immunoprecipitation (ChIP) assay
Chromatin immunoprecipitation assays were performed using the ChIP kit (Millipore). Briefly, after attaining 70%-80% confluence, the MH-S cells were fixed with 1% formaldehyde for 10 min and subjected to sonication to isolate the chromatin fragments.
Fragmented chromatin was immunoprecipitated overnight with the anti-RNA polymerase Ⅱ positive control, anti-human IgG (as NC) or rabbit anti-E2F1 (Table S2). Protein Agarose/Sepharose was added to precipitate the endogenous DNA-protein complex.
The cross-linking was reversed overnight at 65°C. RT-qPCR was performed using these DNA fragments as templates to determine the expression pattern of SKP2. The primer sequences are shown in Table S3.

| Detection of autophagic flux
MH-S cells were treated with the recombinant adenovirus mRFP-GFP-LC3 (Hanbio Biotechnology Co., Ltd.) 23 and then treated with CS or control. Afterwards, the cells were fixed, stained using 4′,6-diamidino-2-phenylindole and observed under a confocal microscope (AMAFD2000, Thermo Fisher Scientific Inc.). The autophagic flux was assessed by counting the numbers of red dots (autolysosomes) and yellow dots (autophagosomes), respectively.

| Silicosis mouse model
A total of 92 C57BL/6 mice aged 6-8 weeks purchased from Tangshan People's Hospital Laboratory Animal Center were separately housed in cages at (22)(23)(24)(25)°C with 60%-65% humidity under a 12-h light/dark cycle with free access to food and water. All the mice were fed adaptively for 1 week, and their health status was visually approved prior to the experiment. With 20 mice used as control, silicosis model was established in 72 mice as described previously. 24 After anaesthesia by an intraperitoneal injection with 50 mg/kg of sodium pentobarbital, 50 μl of CS suspension was injected into mice through the trachea for model establishment. Next, the modelled mice were injected with siBeclin1, miR-205-5p agomir (overexpressed miR-205-5p), Lenti-E2F1 (overexpressed E2F1) or Lenti-SKP2 (overexpressed SKP2) via the tail vein. Furthermore, the mice injected with sterile saline were used as controls.

| Statistical analysis
The SPSS 21.0 software (IBM Corporation, Armonk) was adopted for statistical analysis. Measurement data are denoted as mean ± standard deviation. Two groups were compared with the unpaired t tests.
Multiple groups were compared with one-way analysis of variance (ANOVA). Comparisons between multiple groups at different time points were performed using two-way ANOVA. Bonferroni test was adopted for post hoc analysis. Pearson correlation analysis was performed to analyse the correlation between parameters. A value of p < 0.05 was considered as statistically significant.

| Inhibiting Beclin1 ameliorates pulmonary fibrosis in mice with silicosis
The involvement of Beclin1 as an autophagy-related gene in regulating the progression of pulmonary fibrosis has been previously reported, 23 but its upstream regulatory mechanism remains uncertain.
In order to determine the role of Beclin1 in fibrosis after silicosis, we first established a mouse model of silicosis and then determined the severity of fibrosis in alveolar tissues of these model mice by HE and Sirius red staining ( Figure 1A). The results showed that CS treatment resulted in inflammatory cell infiltration and alveolar fibrosis. The expression patterns of the collagen genes collagen type I alpha 1 (Col1a1) and Col3a1 in the alveolar tissues were determined by RT-qPCR, and the Col1a1 and Col3a1 mRNA levels were significantly increased in the alveolar tissues of CS-treated mice ( Figure 1B). The preceding results indicated that the mouse model of silicosis was successfully established.
Next, Beclin1 expression in the alveolus tissues of early-stage silicosis (day 3 after CS treatment) was observed to be higher than that in the advanced stage silicosis (56 days after CS treatment), and alveolus tissues of silicosis at both stages presented with upregulated Beclin1 expression relative to normal alveolar tissues ( Figure 1C).
Further, through immunohistochemical staining of the Beclin1 protein in the alveolar tissues on the 56th day after CS treatment, Beclin1-positive cells were detected to be increased in the silicosis alveolar tissues compared to the normal alveolar tissues ( Figure 1D).
Consistently, immunohistochemical analysis of clinical samples also indicated that Beclin1-positive cells of alveolar tissues in stage III silicosis were significantly elevated ( Figure 1E). Taken together, these results revealed the upregulated expression of Beclin1 in silicosis, which may be attributed to the lung self-protective mechanism.  As SKP2 might promote the ubiquitination of Beclin1, 21 we initially analysed the microarray data set GSE11 0711 of mice with silicosis in the GEO database by R language and identified that the SKP2 gene was highly expressed in the setting of pulmonary fibrosis ( Figure 3A). Immunohistochemical staining suggested that the SKP2-positive cells were significantly increased in the alveolar tissues of mice with early silicosis compared to the normal alveolar tissues ( Figure 3B). To further verify whether SKP2 was elevated in early silicosis, we analysed the SKP2 mRNA levels in the afore-mentioned tissues by RT-qPCR and confirmed the presence of a significantly increased SKP2 expression pattern in the alveolar tissues of early silicosis ( Figure 3C). We then identified a decreased SKP2 expression pattern in the MH-S cells transfected with siSKP2 ( Figure 3D). After treatment with the protein synthesis inhibitor CHX, the stability of Beclin1 protein was increased

| E2F1 binds to the SKP2 promoter and upregulates its expression
To further explore the upstream mechanism of SKP2, we initially analysed the microarray data set GSE11 0711 in the GEO database using R language and screened out 3246 differentially expressed genes in pulmonary fibrosis ( Figure 4A). Additionally, 1623 mouse transcription factors were provided by AnimalTFDB. By intersecting them and the upregulated genes obtained by microarray  Figure 4B).
Next, we screened E2F1 as a potential transcription factor for SKP2 through the hTFtarget database. R analysis showed that E2F1 gene was highly expressed in pulmonary fibrosis in the microarray data set GSE11 0711 ( Figure 4C). As shown by IHC, E2F1-positive cells were increased notably in the alveolar tissues of mice with silicosis ( Figure 4D). RT-qPCR analysis revealed a significantly increased E2F1 expression pattern in silicosis, which was highly positively correlated with the expression pattern of SKP2 ( Figure 4E). RT-qPCR showed that the E2F1 overexpression significantly increased the mRNA level of SKP2, whereas E2F1 silencing reduced the mRNA level of SKP2 ( Figure 4F). Next, we designed 5 pairs of amplicons for E2F1 potential binding site in the SKP2 promoter based on MotifMap. ChIP-PCR analysis of the MH-S cell line identified that the SKP2 promoter sequence TTCCCGG, spanning from 1638 to 1645, was the potential binding site of E2F1. Moreover, siE2F1 could significantly reduce the enrichment of E2F1 in the SKP2 promoter region ( Figure 4G).
To validate this finding, dual-luciferase reporter gene assay was conducted in the 293T and MH-S cells, which revealed that overexpression of E2F1 significantly promoted the luciferase activity of the wild-type SKP2 promoter sequence, whereas E2F1 inhibition was indicative of a conflicting effect. Furthermore, either promoting or inhibiting E2F1 expression had no significant effect on the luciferase activity of the mutant SKP2 promoter sequence ( Figure 4H). Based on these results, it was rationale to conclude that E2F1 could potentially bind to the SKP2 promoter and upregulate the SKP2 expression.  Figure 5C). To demonstrate that miR-205-5p directly binds to E2F1 UTR, dual-luciferase reporter gene assay was performed.

| miR-205-5p targets E2F1 and inhibits its expression
The results revealed that miR-205-5p mimic could significantly inhibit the luciferase activity of the wild-type E2F1 3'UTR, whereas the miR-205-5p inhibitor led to an opposite effect ( Figure 5D).

| miR-205-5p promotes AM autophagy via E2F1/SKP2/Beclin1 axis
To identify the regulatory role of the miR-205-5p/E2F1/SKP2/Beclin1 axis in autophagy of AMs, the MH-S cells were treated with CS and combined with rapamycin or 3MA. The obtained results revealed that the enhanced autophagy was supported by increased miR-205-5p Rel. SKP2 levels level as well as reduced E2F1 and SKP2 levels ( Figure 6A,B). In addition, miR-205-5p mimic treatment led to decreased SKP2 and increased Beclin1 in MH-S cells, while miR-205-5p inhibitor led to elevated SKP2 and reduced Beclin1 at the mRNA and protein levels ( Figure 6C). Next, we co-transfected oeE2F1/oeSKP2 vector or siBe-clin1 in MH-S cells with miR-205-5p mimic, respectively. The findings revealed that miR-205-5p mimic reduced the expression patterns of E2F1 and SKP2 and increased the expression pattern of Beclin1, while overexpression of E2F1/SKP2 could reverse the increase in Beclin1 expression pattern induced by miR-205-5p ( Figure 6D).
Moreover, increased LC3 and ATG5 were evident in MH-S cells with miR-205-5p mimic, while overexpressing E2F1 or SKP2 or inhibiting Beclin1 reversed the effect of miR-205-5p mimic ( Figure 6E). Finally, after treatment of the above cells with the recombinant adenovirus mRFP-GFP-LC3 and CS, we identified that the miR-205-5p mimic significantly elevated the numbers of LC3 puncta and mRFP fluorescent spots, which could be reversed by overexpressing E2F1 or SKP2 or inhibiting Beclin1 ( Figure 6F). Concisely, E2F1/SKP2/Beclin1 axis was the mediator for miR-205-5p to promote AM autophagy. Recently, autophagy was identified as a novel defence mechanism for lung injury induced by different factors, including CS. 27,28 A prior study reported the ability of enhancement of autophagy of AMs to reduce CS-induced lung inflammation and fibrosis. 23 Beclin1, the initially discovered mammalian autophagy-related gene, could evidently inhibit pulmonary fibrosis in silicosis. 20 In this study, our findings unravelled enhanced autophagy of AMs in response to silencing Beclin1.

| miR-205-5p inhibits pulmonary fibrosis via E2F1/SKP2/Beclin1 axis
Accumulating evidence has elicited the ability of miRNA to bind to transcription factors and regulate gene expression, coherently mediating cell function, thus establishing significance in various diseases. 29 The current study demonstrated that miR-205-5p was poorly expressed in mice with silicosis, which was negatively correlated with the E2F1 expression pattern. Additionally, miR-205-5p suppressed pulmonary fibrosis by inducing the autophagy of AMs. Similarly, it has been previously documented that miR-205 is poorly expressed in pulmonary fibrosis and its upregulation can delay pulmonary fibrosis. 30 E2F1, an established transcription factor in cell cycle regulation, exerts tumour-suppressive activity and anti-proliferative properties. 31 By targeting E2F1, miR-205 inhibits melanoma cell proliferation and induces senescence 32 or enhances the cisplatin sensitivity of glioma cells. 33 miR-205-5p and miR-342-3p synergistically inhibit the transcription factor E2F1 to reduce anti-cancer chemotherapy resistance. 34 On the basis of these findings, we identified the direct binding relationship of miR-205-5p to E2F1 3'UTR.
E2F1 is a member of the E2F transcription factor family, and this factor has been extensively studied due to its involvement in cell cycle regulation. 35 Recently, SKP2 was reported as a potential molecular target for human pulmonary fibrosis as its inhibition reduces the pulmonary fibrosis in silicosis. 52 Additionally, SKP2 negatively regulates autophagy in hypertrophic cardiomyocytes. 53 Notably, K48-linked polyubiquitination of Beclin1 by SKP2 suppresses the autophagic flow. 21 The regulation of Beclin1 through post-translational modification has been well documented, with involvement of phosphorylation and ubiquitination. 54 In addition to the K48-linked poly-ubiquitination of Beclin1 at K402 by SKP2, Beclin1 also undergoes ubiquitination by additional E3 ligases, in different locations and different types of poly-ubiquitination. [55][56][57] According to reports, modified Beclin1 protein in the fibroblasts during idiopathic pulmonary fibrosis is associated with autophagy dysfunction. 58 Here, we identified that SKP2 could evidently facilitate K48-linked poly-ubiquitination of Beclin1.
Finally, the cell and murine models suggested that miR-205-5p impeded SKP2-mediated Beclin1 ubiquitination to promote Beclin1mediated autophagy and suppress pulmonary fibrosis.
In conclusion, our findings revealed the pro-autophagic and antifibrotic effects of miR-205-5p in silicosis. Our research hence suggests that miR-205-5p can be regarded as a promising target for the treatment of silicosis in the future. However, more investigations are warranted before clinical application.

CO N FLI C T S O F I NTE R E S T
All authors declare that they have no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data included in this study are available upon request by contact with the corresponding author.