Silica Particles Trigger Pulmonary Fibrosis Act Through Cell-To-Cell Delivery of Exosomal Mir-107

Background: Long-term exposure to inhalable silica particles may lead to a serious systemic pulmonary disease called silicosis. However, the role and mechanisms of exosomes in silicosis are not well understood. We previously reported that serum exosomal micro (mi) RNA pro�le was altered in pneumoconiosis patients and silica-exposed macrophages.This study was aimed to explore and verify the role of the exosomal miRNA in lung �brosis when exposed to silica particles. Results: The RT-qPCR result revealed that the levels of the miR-107, miR-122-5p, miR-125a-5p, miR-126-5p, and miR-335-5p were elevated in serous exosomes of silicosis patients. A bioinformatics analysis predicted 5 potential interactions involving these miRNAs, with miR-107–cyclin-dependent kinase (CDK) 6 having the highest score. In a mouse model of silica particle-induced silicosis, miR-107 level in serum exosomes and lung tissue was increased during the development of �brosis, while inhibition of miR-107 reduced pulmonary �brosis. The number of exosomes secreted by macrophages exposed to silica particles was also increased and showed altered cargo composition, and showed a capacity to promote lung �broblast transdifferentiation through a possible mechanism involving the delivery of miR-107 by macrophages to lung �broblasts via exosomes, resulting in targeted inhibition of CDK6, reduced retinoblastoma protein phosphorylation, and inhibition of E2F1 and cell cycle progression. Conclusion: In summary, exosomal miR-107 derived from macrophages exposed to silica particles were transferred to pulmonary �broblasts to trigger their transdifferentiation by targeting CDK6 and arresting cell cycle. These �ndings provide insight into the pathogenesis of silicosis and potential targets for intervention.


Background
Silicon is the second abundant element in the earth crust, mainly in the form of silicate and silica, which is widely found in rocks, gravel, and dust [1,2].As one of prominent compound used in daily life, production and scienti c research, silica exist ubiquitously in nature and human environments, so it may resulted in a variety of adverse health effect especially in industrious conditions.Silicosis, a serious systemic disease, is characterized by irreversible pulmonary brosis caused by inhaled crystalline quartz dust.While preventable, many workers are still affected by silicosis and experience a progressive and fatal disease course because of substandard production processes and inadequate personal protection measures.To date, there is no effective treatment for silicosis.Clarifying the pathologic mechanisms of silicosis at the cellular and molecular levels can provide a basis for the development of effective treatments.
In silicosis, myo broblasts are the main effector cells mediating brogenesis, extracellular matrix deposition, and tissue remodeling.Factors that induce broblast transdifferentiation include signaling molecules such as interleukin (IL)-1, IL-10, and transforming growth factor (TGF)-β [3][4][5]; mechanical factors such as mechanical tension and cell adhesion [6][7][8][9]; and extracellular matrix proteins such as bronectin [10,11].Micro (mi) RNAs are a family of small noncoding RNAs that modulate gene expression by inhibiting mRNA translation or degrading target mRNAs [12].These molecules are widely distributed in vivo.miRNAs circulating in various body uids mainly exist in a free form or are bound by lipoprotein or encapsulated by vesicles including exosomes, which are vesicles with a diameter of 30-150 nm that are secreted by eukaryotic cells and play an important role in intercellular signal communication.Exosomes have certain advantages for miRNA targeting because of their heterogeneity [13,14], speci city [15,16] and stability [17].
miRNAs have been shown to be involved in the regulation of broblast transdifferentiation.For example, miR-21 [18] and miR-145 [19,20] were shown to promote transdifferentiation in pulmonary brosis diseases, whereas miR-9-5p [21], miR-22, miR-27a, miR-29, miR-31 [22], miR-101 [23], miR-1343 [24], and let-7d have an inhibitory role.These miRNAs mostly function by modulating the expression of genes encoding components of the TGF-β signaling pathway.In our previous study, we determined, by highthroughput sequencing, that the serum exosome miRNA pro le of patients with pneumoconiosis was altered compared to that of normal control subjects, alteration also existed in exosome miRNA pro le of macrophages (RAW264.7 cells) with or without silica particles exposure [25,26], suggesting that exosomal miRNAs contribute to the development of silicosis.However, the mechanisms by which this is achieved remain unclear.
To address this issue, we investigated the role of miRNAs in the development of silicosis using clinical specimens and mouse and cell-based models of silicosis.The results showed that miR-107 level was higher in serous exosomes of patients and mice with silicosis.Inhibiting miR-107 alleviated silica particleinduced brosis in mice.We also investigated the mechanism of action of miR-107-including potential targets-in lung broblast transdifferentiation in vitro.Our results provide mechanistic insight into the pathogenesis of silicosis that can aid in the development of effective treatments.

Patients
A total of 50 male patients with silicosis were recruited at Henan Hospital for Occupational Diseases from April 1 to August 31, 2018.All patients were diagnosed with silicosis according to diagnostic criteria for occupational pneumoconiosis (GBZ-70-2015) based on chest X-ray or digital radiography, epidemiologic investigation, and physical examination.Additionally, 50 healthy control subjects who had not been previously exposed to silica were recruited from the First A liated Hospital of Zhengzhou University.Individuals with active tuberculosis, malignant tumor, recent infection, or other disorders that could affect the blood were excluded.A 3 ml sample of venous blood was collected from each subject; the collection tubes were placed vertically for 30 min, followed by centrifugation at 3000 rpm for 10 min to separate the serum.

Mouse model of silicosis
Male C57BL/6N mice (4-6 weeks old) were purchased from Beijing Vital River Laboratory Animal Technology Co. (Beijing, China) and maintained in a speci c pathogen-free environment at the College of Public Health, Zhengzhou University.The mice were exposed to silica particles by intratracheal instillation to establish the silicosis model, or to saline as a control, and sacri ced on days 1, 7, 14, 28, and 56 after exposure for analyses.Peripheral blood and lung tissue samples were collected.Serum was obtained as described above.
In vivo miR-107 antagomir treatment Mice were randomly divided into 4 groups and were cotreated with either miR-107 antagomir (10 mg/kg; GenePharma, Shanghai, China) or a negative control miRNA (miR-NC) and silica suspension via intratracheal instillation, and then injected via the tail vein with miR-107 antagomir or miR-NC (4 mg/kg, GenePharma) or saline, weekly, for the next 3 weeks.Mice exposed to saline only served as the control.On day 28, the mice were sacri ced and peripheral blood and lung tissue samples were collected.
In vitro silicosis model RAW264.7 cells were seeded at 1×10 5 /ml in cell culture dishes (100 mm, Wuxi NEST Biotechnology Co., Wuxi, China) and cultured in complete medium.The following day, the medium was aspirated and the cells were washed 3 times with phosphate-buffered saline (PBS).Serum-free medium containing micronsized silica particles (100 µg/ml; Sigma-Aldrich) or PBS was added to the dishes for 48 h.For exosome treatment, NIH-3T3 cells were seeded at 0.75×10 4 /ml in a 6-well plate with complete medium.The following day, the medium was replaced with free-serum medium as described above.Cells were incubated with exosomes (about 1 µg at the protein level) secreted from macrophages exposed to silica (SiO 2 -exo) or without exposure (c-exo) for 2 days, with cells incubated with PBS serving as the control.

Exosome isolation and characterization
Exosomes were isolated from human serum samples by 5 different methods (Supplementary Materials) and from cell culture medium by ultracentrifugation.For the latter, the medium was centrifuged at 2000×g for 30 min to remove cells and debris.The supernatant was transferred to a new tube and centrifuged at 10,000×g for 45 min to remove large vesicles.The supernatant was passed through a 0.22µm lter and centrifuged at 110,000×g for 120 min (70 Ti rotor, Optima L-100K; Beckman Coulter, Brea, CA, USA).The precipitate was resuspended in PBS and centrifuged at 110,000×g for 70 min to obtain exosomes.The concentration and diameter of the exosomes were determined by nanoparticle tracking analysis (NTA) using a ZetaView instrument (Particle Metrix, Inning am Ammersee, Germany).Brie y, the sample was diluted with PBS to achieve the optimal detectable concentration for the analysis software.The diluted solution was injected into the laser chamber for detection, image output, and data recording.
For morphologic analysis by transmission electron microscopy (TEM), exosome-enriched isolates were added as drops to electron microscope grids and negatively stained with 1% uranyl oxalate.Images were captured with a transmission electron microscope (HT7700; Hitachi High-Tech, Tokyo, Japan).

Target gene prediction
MiRWalk (http://mirwalk.umm.uni-heidelberg.de/)was used to predict miRNA target genes.Brie y, in the miRDB, TargetScan, and miRTarBase databases linked by miRWalk, target genes in the human and mouse genomes were selected based on the criterion of a score > 0.90 in at least 2 of the databases.

Histology
Mouse lung tissue samples were xed in 4% paraformaldehyde, embedded in para n, and cut into sections at a thickness of 6 µm that were stained with hematoxylin and eosin (H&E) to evaluate the in ammation and damage to alveolar structure, or with Masson's trichrome to visualize collagen deposition.

RNA isolation and quantitation
Exosomal miRNA was extracted using an miRNA extraction kit (DP503, Tiangen Biotech, Beijing, China).Brie y, after lysis with MZA buffer, chloroform was added to the samples for RNA extraction.The upper aqueous phase was thoroughly mixed with ethanol and transferred to a column, and miRNAs were puri ed using MRD buffer.Total RNA was extracted using TRIzol reagent (Thermo Fisher Scienti c) and the quality was assessed using a Nanodrop 2000 spectrophotometer (Thermo Fisher Scienti c).
RNA was reverse transcribed into cDNA using miRcute Plus miRNA qPCR Detection Kit (KR211; Tiangen Biotech) or PrimeScript RT Reagent Kit with gDNA Eraser (RR047A; Takara Bio, Otsu, Japan), and quantitated by quantitative real-time (qRT-)PCR using the SYBR Green Kit (Tiangen Biotech; FP411) or TB Green Premix Ex Taq II (Takara Bio; RR820A).Relative levels of miRNAs and mRNAs were normalized to U6 and glyceraldehyde 3-phosphate dehydrogenase (GAPDH).The primers used for PCR ampli cation of miRNA and mRNA were synthesized by Tiangen Biotech and Takara Bio, respectively.All of the steps were carried out according to the manufacturers' recommendations.

Exosome uptake assay
Exosomes were incubated for 20 min at 37℃ in the dark with the cell membrane dye DiO (10 µM) (Beyotime, Shanghai, China).To remove free dye, exosomes were subjected to another round of isolation.NIH-3T3 cells were incubated with the cell membrane dye DiI (10 µM) (Beyotime) for 20 min and washed 3 times with PBS.The labeled exosomes were cocultured with the cells for 4 h.The cells were then xed with 4% paraformaldehyde and the nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) before visualization with a laser scanning confocal microscope (TSC SP-8, Leica, Germany).

miRNA in situ hybridization and immuno uorescence analysis
After baking, depara nization, repair, and prehybridization with 0.5% Triton X-100 in PBS, lung tissue sections were hybridized overnight at 37℃ with digoxigenin-labeled miR-107 probes in hybridization buffer.The sections were incubated overnight at 4℃ with anti-collagen antibody (Boster Bio) and nuclei were stained with DAPI.Signals were detected with a laser scanning confocal microscope.

Cell cycle assay
NIH-3T3 cells transfected with miR-107 inhibitor or inhibitor-NC were harvested and xed overnight at 4℃ with 70% cold ethanol.After washing twice with PBS, the cells were resuspended in propidium iodide/RNase staining solution (Beyotime) followed by incubation for 30 min at room temperature.Cell cycle distribution was analyzed by ow cytometry (Accuri C6; BD Biosciences, Franklin Lakes, NJ, USA) and ModFit software (Verity Software House, Topsham, ME, USA).

Dual luciferase assay
The 3′-untranslated region (UTR) of the CDK6 gene was cloned into the pSI-Check2 vector (Hanbio, Shanghai, China) to obtain the reporter vector for the dual luciferase assay.HEK-293T cells were seeded in a 96-well plate and allowed to attach overnight.Reporter constructs with a mutated or wild-type 3′-UTR of the CDK6 gene were cotransfected into the cells with miR-107 mimic or mimic-NC.After 48 h, luciferase activity was measured using a commercial kit (Promega, Madison, WI, USA).

Statistical analysis
Statistical analyses were performed using SPSS v21.0 software (SPSS Inc, Chicago, IL, USA).Data are expressed as mean ± standard deviation.All statistical tests were 2-sided, and the level of statistical signi cance was set at α = 0.05.

Serum exosomal levels of 5 miRNAs are elevated in patients with silicosis
There was no difference in mean age between silicosis patients and healthy control subjects (Table 1), and no difference in mean working age among patients at different stages of the disease (Table 2).1a).To determine whether the expression of these 5 miRNAs changed over the course of the disease, we compared their levels in serous exosomes of patients at different stages of silicosis and found that they were lower at stage II than at stage I, while miR-122-5p and miR-335-5p levels were higher at stage than at stage (Fig. 1b).
Temporal pro le of miR-107 in serum exosomes and lung tissue of mice In ammatory cell in ltration was observed on day 1 after silica instillation by H&E staining.The alveolar wall was thickened and alveolar structure disrupted by day 7, while in ammatory cell accumulation and cell nodules were observed on day 28.By day 56, the alveolar structure was completely destroyed and collagen deposition was markedly increased, as visualized by Masson's trichrome staining (Fig. 2a,b).These histologic changes were accompanied by the gradual increase in expression of the brosis markers α-SMA, collagen , and bronectin in lung tissue and decrease in that of CDK6 (Fig. 2c).
The temporal pro le of miR-107 expression in serous exosomes and lung tissue was evaluated by qRT-PCR.miR-107 level in serous exosomes was decreased relative to the control group on day 7 after silica exposure but increased gradually starting from day 14 (Fig. 2d).Similarly, miR-107 level in lung tissue was decreased on day 7 but increased gradually starting from day 28 (Fig. 2e).Pearson correlation analysis revealed a positive correlation between miR-107 levels in serum exosomes and lung tissue (r = 0.834, P < 0.001).

Inhibition of miR-107 alleviates silica-induced pulmonary brosis in vivo
To investigate the effect of miR-107 on silica-induced pulmonary brosis in mice, we inhibited miR-107 with an antagomir.miR-107 level was markedly reduced in silicosis mice treated with the antagomir as determined by qRT-PCR, indicating that the inhibition was successful (Fig. 3b).H&E and Masson's trichrome staining showed that this resulted in the disappearance of the large cell nodules and collagen accumulation observed in silica-exposed mice with antagomir treatment (Fig. 3c).Additionally, western blot analysis revealed that the expression of brosis markers was downregulated whereas that of CDK6 was upregulated (Fig. 3d); and in situ hybridization and immuno uorescence analyses demonstrated that miR-107 and collagen levels were reduced (Fig. 3e).These results indicate that miR-107 antagomir attenuates silica-induced pulmonary brosis.

Characterization of exosomes derived from macrophages
Particles isolated from the culture medium of RAW264.7 cells were analyzed by TEM.We observed the cup-shaped vesicles that have been described in other studies [27] (Fig. 4a).NTA showed that although particle size was unaffected, silica treatment increased their concentration (Fig. 4b-d).Western blot analysis revealed that the isolated particles were positive for the exosome markers CD9, CD81, and TSG101, which were upregulated after exposure to silica (Fig. 4e).These results suggest that silica exposure enhances exosome secretion from RAW264.7 cells.
miR-107 levels are elevated in silica-exposed macrophages, their secretory exosomes, and effector cells To examine the uptake of exosomes, the uorescent lipophilic dyes DiO and DiI were used to stain exosomes and broblasts, respectively.After 4 h, NIH-3T3 cells were spindle-shaped and punctate exosomes surrounded the cells or were fused to the cell membrane.There were no signi cant differences in exosome number and distribution between cells with and those without exposure to silica, suggesting comparable uptake of exosomes (Fig. 5a).We examined miR-107 expression in macrophages, their secreted exosomes, and NIH-3T3 cells treated with exosomes and found that the levels in macrophages (Fig. 5b) and exosomes (Fig. 5c) were increased following silica exposure.Notably, miR-107 level was also increased in NIH-3T3 cells incubated with exosomes derived from macrophages exposed to silica particles (SiO 2 -exo) (Fig. 5d).

Exosomes derived from silica-exposed macrophages promote transdifferentiation of pulmonary broblasts
To evaluate the effect of exosomes on broblast effector cells, exosomes derived from macrophages were incubated with NIH-3T3 cells.Western blot analysis indicated that SiO 2 -exo, but not c-exo, increased the expression of the differentiation markers α-SMA, collagen , and bronectin (Fig. 5e).To determine whether miR-107 encapsulated in exosomes can promote broblast transdifferentiation, we transfected macrophages with synthetic miR-107 or mimic-NC.As expected, exosomal miR-107 level was increased (Fig. 5f), which was accompanied by enhanced NIH-3T3 cell transdifferentiation and downregulation of CDK6 (Fig. 5g).

miR-107 potentiates the pro brogenic effect of TGF-β in vitro
To determine whether CDK6 is a target of miR-107 inhibition, we rst predicted the potential binding sites of the miR-107 seed sequence in the human and mouse CDK6 3′-UTR using TargetScan (Fig. 6a).The results of the dual-luciferase reporter assay showed that miR-107 mimic decreased luciferase activity in cells transfected with the wild-type CDK6 3′-UTR sequence but not in those transfected with the mutated sequence (Fig. 6b), suggesting that CDK6 is directly regulated by miR-107.A search of the Kyoto Encyclopedia of Genes and Genomes databases showed that CDK6 is involved in cell cycle-associated signaling downstream of TGF-β; we therefore treated NIH-3T3 cells with different concentrations of TGF-β and found that at 2 ng/ml, miR-107 was upregulated whereas CDK6 was downregulated (Fig. 6c, d), and used this concentration in subsequent experiments.
To further con rm the effect of CDK6 on broblast transdifferentiation, NIH-3T3 cells were transfected with CDK6 overexpression plasmid, which resulted in decreased expression of differentiation markers (Fig. 7a).We also performed a cell recovery experiment by transfecting cells with CDK6 siRNA, with cells transfected with GAPDH siRNA serving as a positive control (Fig. 7b).Of the 3 CDK6 siRNAs that were tested, siRNA-2 had the highest knockdown e ciency (Fig. 7c).Cotransfection of CDK6 siRNA abrogated the anti brotic effect of miR-107 inhibitor (Fig. 7d).

miR-107 modulates broblast transdifferentiation by targetingCDK6
As CDK6 is involved in cell cycle regulation, we transfected NIH-3T3 cells with miR-107 inhibitor and analyzed cell cycle distribution by ow cytometry.Inhibition of miR-107 reduced the size of the G1 fraction and increased the proportion of cells in S phase (Fig. 7e); moreover, it increased the expression of proteins involved in cell cycle signaling including cyclin D, pRb, and E2F1 (Fig. 7f).

Discussion
Long-term silica particle inhalation leads to silicosis.Myo broblasts are terminal effector cells in the development of pulmonary brosis, and the activation and transdifferentiation of pulmonary broblasts is critical for this process.Given that exosomes mediate intercellular signaling in various physiologic and pathologic contexts, the present study examined the role of exosomal miR-107 in the pathogenesis of silicosis.Our results showed that miR-107 levels in lung tissue and serum exosomes increased following in ammation induced by exposure to silica, while inhibition of miR-107 attenuated pulmonary brosis.Exosomal miR-107 derived from silica particle-exposed macrophages promoted the transdifferentiation of lung broblasts, possibly through targeted inhibition of CDK6, pRb, and E2F1.
Exosomes are widely distributed in body uids and can carry nucleic acids, proteins, and lipids, among other biomolecules.These vesicles are internalized by other cells via direct membrane fusion, endocytosis, or cell type-speci c phagocytosis to exert biological effects [28].Exosomal miRNAs are known to regulate gene expression.The species of exosomal miRNAs in tissue or body uids varies for different diseases [29][30][31].We found that the levels of miR-107, miR-122-5p, miR-125a-5p, miR-126-5p, and miR-335-5p were higher in serous exosomes of silicosis patients than in those of healthy control subjects, suggesting that these 5 miRNAs can serve as diagnostic biomarkers or potential therapeutic targets in the treatment of silicosis.The lower levels of these miRNAs in stage as compared to stage I or III patients may be due to the small number of subjects in our study.
We focused on the miR-107-CDK6 interaction in the present study.miR-107 has been implicated in cell cycle regulation in various physiologic as well as pathologic processes such as Alzheimer disease, cancer, and diabetes.For example, miR-107 expression was found to decrease with age, and is thought to play an important role in regulating biological life span [32].As a tumor suppressor, miR-107 negatively regulates cell proliferation, migration, and apoptosis, and was shown to inhibit the proliferation and migration of breast cancer cells by targeting CDK8r [33].miR-107 is downregulated in patients with osteoarthritis and its upregulation enhanced aggrecan and collagen II protein expression in chondrocytes while reducing that of matrix metalloproteinase (MMP)13 and MMP9 [34].However, there have been few studies on the role of miR-107 in brotic diseases.Patients with cystic brosis showed increased levels of miR-103a-3p, miR-103b, and miR-107 [35], but miR-107 was downregulated in chronic graft dysfunction with interstitial brosis and tubular atrophy following renal transplantation [36].In our study, miR-107 levels in serous exosomes of silicosis patients and mouse lung tissue were decreased during in ammation and increased during brosis.Silica particle treatment also increased miR-107 level in macrophages along with their secreted exosomes and effector cells ( broblasts).Overexpression of miR-107 promoted transdifferentiation of RAW264.7 cells and broblasts, whereas miR-107 inhibition had the opposite effect and alleviated silica particle-induced brosis in mice.Thus, miR-107 functions as a regulator of broblast transdifferentiation during brosis.
CDK6 is involved in cell cycle progression and cell differentiation.In eukaryotes, G1 arrest is a prerequisite for cell differentiation [37].Overexpression of CDK6 was shown to inhibit the differentiation of glial precursor cells into astrocytes [38] and suppress BMP-2-induced osteoblast differentiation [39].In the present study, we observed an inverse relationship between CDK6 and miR-107 expression, and con rmed the negative regulatory relationship with the dual luciferase reporter assay.We also found that the expression of transdifferentiation markers in broblasts was negatively regulated by CDK6; this may be related to cell cycle progression, given that inhibition of miR-107 promoted G1-S transition.
To further clarify the mechanism by which inhibition of CDK6 by miR-107 promotes silicosis, we examined the expression of the cell cycle signaling proteins cyclin D, pRb, Rb, and E2F1.Cyclins and CDKs synergistically promote Rb phosphorylation, leading to the dissociation of pRb and its binding partner E2F1; free E2F activates the transcription of DNA replication-related genes.Cyclin D is a regulator of CDK kinase, which forms a complex with CDK4 or CDK6 in which it functions as the regulatory subunit.
Cyclin D activity is necessary for the G1/S transition.Rb is a tumor suppressor that plays an important role in regulating cell cycle progression and DNA replication by binding E2F family transcription factors [40] as well as in the differentiation of eyes, lens, brain, nervous system, epidermis, melanocytes, hair cells, muscle, and liver [41].Rb protein is the main G1 checkpoint for entry into S phase and cell growth, and promotes terminal differentiation by inducing cell cycle exit and tissue-speci c gene expression [42].It may also play a role in cell proliferation following differentiation [43].E2F1 forms a heterodimer with TFDP2 to activate the transcription of cell cycle regulatory genes.In our study, inhibition of miR-107 increased the expression of cyclin D, pRb, and E2F1.Thus, in silicosis, the elevated levels of miR-107 in lung tissue and serum exosomes during silicosis may inhibit the expression of cyclin D and CDK6, leading to decreased phosphorylation of Rb and downregulation E2F1, which block cell cycle progression and promote the transcription of cell differentiation-related genes.However, proliferation and differentiation are not mutually exclusive; the former is essential for expanding the pool of cells that will undergo terminal differentiation, and cells in higher organisms cease to divide only at the end of this process.

Conclusion
In summary, inhibition of miR-107 reduced pulmonary brosis in a mouse model of silicosis.Exosomal miR-107 derived from macrophages exposed to silica particles promoted transdifferentiation of lung broblasts by inhibiting CDK6 and blocking cell cycle progression.These results highlight the role of exosomal miR-107 in the pathogenesis of silicosis and provide potential targets for its treatment.

Figure 5 Role
Figure 5