Dasatinib Reduces Lung Inflammation and Fibrosis in Acute Experimental Silicosis

Silicosis is an occupational lung disease with no effective treatment. We hypothesized that dasatinib, a tyrosine kinase inhibitor, might exhibit therapeutic efficacy in silica-induced pulmonary fibrosis. Silicosis was induced in C57BL/6 mice by a single intratracheal administration of silica particles, whereas the control group received saline. After 14 days, when the disease was already established, animals were randomly assigned to receive DMSO or dasatinib (1 mg/kg) by oral gavage, twice daily, for 14 days. On day 28, lung morphofunction, inflammation, and remodeling were investigated. RAW 264.7 cells (a macrophage cell line) were incubated with silica particles, followed by treatment or not with dasatinib, and evaluated for macrophage polarization. On day 28, dasatinib improved lung mechanics, increased M2 macrophage counts in lung parenchyma and granuloma, and was associated with reduction of fraction area of granuloma, fraction area of collapsed alveoli, protein levels of tumor necrosis factor-α, interleukin-1β, transforming growth factor-β, and reduced neutrophils, M1 macrophages, and collagen fiber content in lung tissue and granuloma in silicotic animals. Additionally, dasatinib reduced expression of iNOS and increased expression of arginase and metalloproteinase-9 in silicotic macrophages. Dasatinib was effective at inducing macrophage polarization toward the M2 phenotype and reducing lung inflammation and fibrosis, thus improving lung mechanics in a murine model of acute silicosis.


Introduction
Silicosis is an occupational disease caused by inhalation of crystalline silica particles, which triggers a persistent inflammatory cascade that leads to progressive lung fibrosis and subsequent respiratory failure due to deterioration of lung function and reduction in gas exchange area [1]. Although therapy for silicosis includes a variety of drugs and non-pharmacological

Animal Preparation and Experimental Protocol
Fifty-two C57BL/6 female mice (weight: 20-25 g, age 8-12 weeks) were assigned to two main groups: control (C) and silicosis (SIL). In group SIL, mice received silica particle suspension (20 mg in 50μL saline, intratracheally [i.t.]), while group C received saline (50 μL, i.t.) using the same protocol. Fourteen days after administration of silica or saline, animals were further randomized to receive dimethyl sulfoxide (DMSO 1% in saline solution, 100 μL, oral gavage) or dasatinib (DAS 1 mg/kg body weight in DMSO 1%, 100 μL, oral gavage) during 14 days ( Fig  1). Additionally, another group of animals was treated with saline (SAL, 100 μL) by oral gavage for 14 days. These animals were subsequently compared to those that received DMSO, aiming to evaluate whether DMSO per se might have any effects on lung morphofunction (Tables A-C in S1 File).

Lung Mechanics
Twenty-four hours after the last dose, the animals were sedated (diazepam 1 mg i.p.), anesthetized (thiopental sodium 20 mg/kg i.p.), tracheotomized, paralyzed (vecuronium bromide, 0.005 mg/kg i.v.), and ventilated with a constant flow ventilator (Samay VR15; Universidad de la Republica, Montevideo, Uruguay) set to the following parameters: respiratory rate 100 breaths/min, tidal volume (V T ) 0.2 mL, and fraction of inspired oxygen (FiO 2 ) 0.21. The anterior chest wall was surgically removed and a positive end-expiratory pressure of 2 cm H 2 O applied. Airflow and tracheal pressure (Ptr) were measured. In an open chest preparation, Ptr reflects transpulmonary pressure (P L ). After a 10-min ventilation period, static lung elastance (Est,L), and lung resistive (ΔP1,L) and viscoelastic/inhomogeneous pressures (ΔP2,L) were measured using the end-inflation occlusion method [17]. All data were analyzed using ANA-DAT software (RHT-InfoData, Inc., Montreal, Quebec, Canada). All experiments lasted less than 15 min.

Histology
Immediately after determination of lung mechanics, a laparotomy was performed and heparin (1000 IU) was injected in the vena cava. The trachea was clamped at end-expiration (PEEP = 2 cmH 2 O), and the vena cava and abdominal aorta were sectioned, leading to a massive hemorrhage and euthanasia by terminal bleeding. The left lung was then removed, quickly frozen by immersion in liquid nitrogen, fixed with Carnoy's solution and paraffin-embedded. Four 4μm-thick slices per lung were cut and stained with hematoxylin-eosin. Lung morphometry analysis was performed with an integrating eyepiece with a coherent system consisting of a grid with 100 points and 50 lines of known length coupled to a conventional light microscope (Olympus BX51, Olympus Latin America-Inc., Sao Paulo, Brazil). The fractional area of lung occupied by collapsed or normal alveoli, or hyperinflated structures (> 120 μm) were determined by the point-counting technique [18] across 10 random, non-coincident microscopic fields at 200× magnification. Briefly, points falling on collapsed, normal pulmonary areas or hyperinflated structures were counted and divided by the total number of points in each microscopic field [18]. Additionally, the fractional area of granuloma was determined in 20 random non-coincident microscopic fields, at 400× magnification [2,19]. Neutrophils, mononuclear cells, and total cells in the alveolar septa and granuloma were evaluated at 1,000× magnification Thirty-four C57BL6 female mice (8-12 weeks, 20-25 g) were divided into two groups: control group (C, n = 16) instilled with sterile saline (50 μL, intratracheally [i.t.]) and silicosis group (SIL, n = 18), instilled with silica particle (20mg in 50 μL saline, i.t.). Fourteen days after disease induction, the animals were randomized to receive a solution of dimethyl sulfoxide (DMSO 1% in saline solution, 100 μL, oral gavage, n = 8/9) or dasatinib (DAS 1 mg/kg body weight in DMSO 1%, 100 μL, oral gavage, n = 8/9). and determined by the point-counting technique. Collagen fibers (Picrosirius polarization method) were computed in the alveolar septa and granuloma at 400× magnification using Image-Pro Plus 6.3 software (Media Cybernetics, Silver Spring, MD, USA) [2,20]. Bronchi and blood vessels were carefully avoided during the measurements. The area occupied by fibers was determined by digital densitometric recognition. The results were expressed as the fractional area occupied by collagen fibers in the alveolar septa or in granuloma.
Paraffin-embedded tissue sections (4 μm) were dewaxed and, after rehydration, subjected to heat-mediated antigen retrieval with citrate buffer 10 mM (pH = 6.0). Endogenous peroxidase activity was inhibited in 70% hydrogen peroxide solution in methanol. Non-specific immunoglobulin binding was blocked by 10% bovine serum albumin in phosphate saline buffer (pH = 7.4), before primary antibody incubation. Antibodies were revealed with biotinylated secondary antibody (Histofine mouse Max PO anti-rat and anti-rabbit, Nichirei Biosciences, Tokyo, Japan). Detection was done with peroxide and the chromogen substrate diaminobenzidine (catalog no. K3468, Dakocytomation). Slides were counterstained with Giemsa.
Analysis was performed in 30 images of high-power fields (400× magnification) per slide, manually selected using a light microscope (Nikon Eclipse 400; Nikon Instruments, Tokyo, Japan), and captured with an Evolution VF Color Cooled 12-bit digital camera (Media Cybernetics, Silver Spring, MD, USA). The areas occupied by nucleated macrophages and cells with positive staining for the phenotype marker in each tissue area were then measured and divided by tissue area using Image-Pro Plus 6.3 software and expressed as fractional area occupied by positive cells.

Enzyme-Linked Immunosorbent Assay (ELISA)
Levels of interleukin (IL)-1β, tumor necrosis factor (TNF)-α and transforming growth factor (TGF)-β were quantified by ELISA in the lung homogenate. Lung tissue was homogenized in lysis buffer (PBS 1×, triton X 0.01%, 1× Roche protease inhibitor cocktail (Roche Diagnostic, Mannheim, Germany]) using a glass Potter homogenizer with Teflon piston. The total amount of cytokines was quantified according to the manufacturer's protocol (Duo Set, R&D Systems, Minneapolis, MN, USA) and normalized to the total protein content quantified by Bradford's reagent (Sigma-Aldrich, St Louis, MO, USA).

In vitro Analysis of Macrophages
RAW 264.7 cells, a mouse peritoneal macrophage cell line, obtained from American Type Culture Collection (Rockville, MD) were maintained in culture, using Dulbecco's Modified Eagle Medium (DMEM)-High Glucose, supplemented with 10% fetal bovine serum, 1,000 U/mL penicillin/streptomycin, 2mM L-glutamine (Invitrogen, Life Technologies Grand Isle, NY). Cells were plated in six-well plates (10 6 cells/well) for 48 hours. The medium was then replaced with fresh medium, and cells were exposed to silica particles (100 μg per mL of medium) for 24 hours [22] or left incubated with regular medium. Supernatant was then removed; cells were washed with 1× PBS, and then incubated with dasatinib (100 ng/mL medium) or regular medium for 24 hours. Once again, supernatant was removed, cells washed with PBS, lifted using 2.5% Trypsin/EDTA (Invitrogen Life Technologies Grand Isle, NY) and pelleted by centrifugation (600 × g for 5 min).
A quantitative real-time reverse transcription (RT) polymerase chain reaction (PCR) was performed to measure mRNA expression of iNOS, arginase, metalloproteinase (MMP)-9, and caspase 3. Cells were lysed for RNA extraction through the RNeasy Plus Mini Kit (Qiagen, Valencia, CA, USA) according to the manufacturer's recommendations. The total RNA concentration was measured by spectrophotometry in Nanodrop ND-1000. First-strand cDNA was synthesized from total RNA using an M-MLV Reverse Transcriptase Kit (Invitrogen). Relative mRNA levels were measured with a SYBR Green detection system using ABI 7500 realtime PCR (Applied Biosystems, Foster City, CA). All samples were measured in triplicate. The relative level of each gene was calculated as the ratio of the study gene to the control gene (acidic ribosomal phosphoprotein P0 [36β4]) and given as the fold change relative to RAW cells incubated with regular medium. The following PCR primers were used: iNOS forward CTTCAGGTATGCGGTATTGG and reverse 5´CAT GGT GAA CAC GTT CTT GG; Arginase-1:

Statistical Analysis
The normality of data (Kolmogorov-Smirnov test with Lilliefors' correction) and the homogeneity of variances (Levene median test) were tested. Parametric data are expressed as mean ± SD. Differences between groups were evaluated by one-way ANOVA followed by Bonferroni's test. Nonparametric data were analyzed using ANOVA on ranks followed by Dunn's post hoc test. All tests were performed using GraphPad Prism v6.0 statistical software package (GraphPad Software, La Jolla, California, USA). Significance was established at p < 0.05.

Results
The mortality rate of animals with acute silicosis was 50% before treatment was initiated. The remaining 18 mice were allocated, with 9 receiving dasatinib and 9 receiving DMSO. No animal died after therapy with DMSO or dasatinib. No significant differences in lung mechanics and inflammation, fraction area of granuloma, and collagen fiber content were observed between the saline and DMSO groups (Tables A-C in S1 File).

Lung Mechanics
Tidal volume and airflow did not differ among groups. Est,L, ΔP1,L, and ΔP2,L were similar in C-DMSO and C-DAS groups. Est,L, ΔP1,L, and ΔP2,L were higher in SIL-DMSO animals than in the C-DMSO group (32%, 995%, and 19%, respectively). In the SIL-DAS group, all mechanical parameters were reduced compared to SIL-DMSO; however, ΔP1,L remained higher than C-DAS group (Fig 2).

Lung Morphometry and Inflammation
In the SIL-DMSO group, we observed granulomatous nodules with infiltration of neutrophils and mononuclear cells, mainly macrophages (Fig 3). Moreover, there were increased areas of alveolar collapse (214%) and cell infiltration (86% neutrophils, 66% mononuclear) in lung parenchyma (Table 1).
Photomicrographs show that granuloma appeared to be disintegrating over time after treatment with dasatinib (Fig 3). The area of collapsed alveoli, the fraction area of granuloma and neutrophils in lung parenchyma and granuloma were reduced in SIL-DAS compared to SIL-DMSO (Table 1; Fig 3).
Macrophages are considered the main cells in the pathophysiology of silicosis. They can be activated by a variety of extracellular signals to polarize into M1 macrophages (associated with antimicrobial response and inflammation) or M2 macrophages (associated with wound healing and resolution of inflammation). The total number of macrophages (F4/80 positive cells) and the M1 (iNOS positive cells) and M2 (arginase-1 positive cells) subpopulations were quantified. The number of macrophages in alveolar septa was higher (1,036%) in the SIL-DMSO group than in the C-DMSO group (Fig 4). Dasatinib did not reduce the number of macrophages either in lung parenchyma or in granuloma (Fig 4, Table D in S1 File). The number of M1 macrophages was increased in lung parenchyma (1,160%) in SIL-DMSO compared to C-DMSO animals, and reduced in SIL-DAS in lung parenchyma (92.3%) and granuloma (96.8%) (  Table E in S1 File). No significant difference was observed in the number of M2 macrophages between the SIL-DMSO and C-DMSO groups. However, dasatinib increased the number of M2 macrophages in lung parenchyma (712.3%) and granuloma (336.5%) in animals with acute silicosis (Fig 6, Table F in S1 File).

Lung Fibrosis (Collagen Fibers and TGF-β)
The amount of collagen fibers and the level of TGF-β were increased in lung parenchyma and granuloma in SIL-DMSO compared to C-DMSO animals. Dasatinib led to reductions in collagen fiber deposition and TGF-β (Fig 7, Table G in S1 File). In Vitro Assays iNOS expression was increased in SIL-DMSO compared to C-DMSO animals. In SIL-DAS, arginase and MMP-9 expressions were increased compared to SIL-DMSO. No significant changes were observed in caspase-3 expression between groups (Fig 8, Table G in S1 File).

Discussion
In the model of acute silicosis used herein, dasatinib improved lung mechanics and led to a reduction of fraction area of granuloma, neutrophils in lung tissue and granuloma, M1 macrophages in lung parenchyma and granuloma, fraction area of collapsed alveoli, collagen fiber content in lung parenchyma, protein levels of IL-1β, TNF-α, and TGF-β, and increased M2 macrophages in lung parenchyma and granuloma. In vitro studies showed that dasatinib led to reduced expression of iNOS and increased expression of arginase and MMP-9. To the best of our knowledge, this was the first study to evaluate the potential therapeutic effects of dasatinib on lung function, inflammation, and remodeling in experimental silicosis. The 1 mg/kg dose of dasatinib was chosen on the basis of pilot studies and studies in experimental endotoxininduced acute lung injury conducted in our lab. The anti-fibrotic effects observed at this dose were similar to those observed at the higher dose (10 mg/kg). However, the higher dose led to further lung damage (data not shown).
The model of acute silicosis induced by a single exposure to crystalline silica used herein led to morphological and functional changes after 2 weeks which resembled human silicosis [2,19,23]. Silicotic mice exhibited higher values of Est,L and ΔP2,L, which were associated with the presence of silicotic granulomas, alveolar collapse, thickening of alveolar septa, and inflammatory cell infiltration. The changes in ΔP1,L may correlate with an increased number of intrabronchial cells associated with lumen obstruction, in accordance with previous studies on silicotic mice [2,19,23,24]. In the SIL-DMSO group, an increase in the number of macrophages and neutrophils, both in lung parenchyma and granuloma, was observed. Macrophages are the main cells involved in the pathophysiology of silicosis. Briefly, silica particles induce macrophage activation and lesion, triggering the release of metalloproteinases, free radicals, proinflammatory mediators such as IL-1β, TNF-α, and TGF-βthrough activation of nuclear factor (NF)-κB pathway [25,26], which is responsible for the recruitment of more inflammatory cells to the site of injury and are involved in lung fibrosis [2]. In order to evaluate whether DMSO might mitigate or induce further lung damage, another group of animals was treated with saline. No significant differences in lung mechanics and morphometry or collagen fiber content were observed between the DMSO and SAL groups (Tables A-C in S1 File); thus, we presented the data obtained from the DMSO group.
Protein tyrosine kinases have been demonstrated to play a crucial role in the inflammatory signaling pathways induced by silica. Evidence suggests that free radicals, generated by silicamediated reactions or macrophage activation, trigger activation of several tyrosine kinases, which subsequently leads to dimerization and nuclear translocation of the pro-inflammatory transcription factor NF-κB [22]. Thus, dasatinib, an ATP-competitive protein tyrosine kinase inhibitor, when used at therapeutic concentrations, inhibits the activity of Abl, Bcr-Abl, Srcfamily kinases and several additional kinases, including Src and Btk family members, c-Kit, PDGFR, and Eph receptors [27]. Besides its effect on malignant cells, dasatinib also blocks certain functions of various hematopoietic cells by inhibiting T lymphocyte activation and proliferation [8], suppressing natural killer cell toxicity [9], blocking allergen-induced release of histamine in blood basophils [10], affecting platelet activation [11] and reducing neutrophil activation and chemotaxis [12,13]. In the present study, dasatinib reduced neutrophils in lung parenchyma and granuloma, which might be explained by the ability of the drug to block neutrophil adhesion and migration through the inhibition of Syk, ERK and the p38 MAPK [12,13], and reduced lung protein levels of the pro-inflammatory and chemoattractant mediators IL-1βand TNF-α. Moreover, these inflammatory changes may be related to the effect of dasatinib in reducing the number of M1 macrophages and increasing the number of M2 macrophages in lung parenchyma and granuloma of silicotic mice. It is well known that activation of NF-κB, which regulates genes controlling several physiological processes including the innate immune responses and inflammation [28], promotes polarization of macrophages toward the M1 phenotype [29]. Therefore, dasatinib might disrupt this pathway signaling, as NF-kB activation is Src tyrosine kinase-dependent in macrophages [22]. The M2 macrophage may help to resolve inflammation through high endocytic clearance abilities and production of trophic factors, as well as reduced pro-inflammatory cytokine secretion [30]. M2 macrophages also generate arginase-1, which suppresses inflammation by inhibiting production of proinflammatory nitric oxide [31]. Furthermore, M2 cells express the IL-1 receptor antagonist, which inhibits the effects of the pro-inflammatory cytokine IL-1, the mannose receptor, and chitinase-like 3 [32].
Regarding remodeling, there is great controversy on the role of M2 macrophages in fibrosis. M2 macrophages are increased in pulmonary fibrosis and are associated with fibrosis development [33,34]. On the other hand, in the context of silicosis, Misson et al. addressed the question of whether lung fibrosis development is associated with M2 macrophages in a murine model of single intratracheal instillation of silica particles. By comparing the phenotype of pulmonary macrophages during the development of silica-induced lung fibrosis in C57BL/6 and BALB/c mice, the authors observed that the amplitude of Arg-1 mRNA up-regulation was not associated with the severity of lung fibrosis. Their data indicate that the establishment of a fibrotic process is not necessarily associated with M2 polarization in a murine silicosis model [35]. Furthermore, new evidence suggests that M2 macrophages may actually contribute to the resolution of fibrosis. Their presence during fibrosis, which has been observed in experiments, may be explained as a failing attempt to clear excess extracellular matrix. Mechanistic studies in models of liver fibrosis showed that M2 macrophages are not required for fibrosis development [36], and that M2 cells are important for fibrosis resolution in the liver [37,38]. Furthermore, it has been reported that a subset of adipose tissue macrophages exhibiting an M2 phenotype produce MMP-9, which in the kidney may contribute to attenuation of fibrotic lesions [39]. This correlates well with findings that uptake of extracellular matrix components appears to be mediated by M2 macrophages, as uptake of these components is mediated by different mannose receptors, which are known as M2 markers [40]. The conflicting roles described in the literature may be the result of difficulties in separating the effects of all existing subtypes of macrophages, since subsets are difficult to distinguish [33]. In our experiments, dasatinib increased the amount of M2 macrophages and reduced collagen deposition in lung parenchyma and granuloma. This phenomenon might be explained by the antifibrotic effects of M2 macrophages, but also by direct pharmacological inhibition of TK associated with profibrotic receptors such as platelet-derived growth factor receptors (PDGFR) α and β, vascular endothelial growth factor receptors (VEGFR) 1, 2, and 3, fibroblast growth factor receptors (FGFR) 1, 2, and 3, and TGF-β.
Tyrosine kinase inhibitors are known to have antifibrotic effects. Nintedanib (formerly known as BIBF 1120) is an intracellular inhibitor that targets multiple tyrosine kinases. A phase 2 trial suggested that treatment with 150 mg of nintedanib twice-daily reduced lungfunction decline and acute exacerbations in patients with idiopathic pulmonary fibrosis. However, several patients discontinued treatment because of adverse effects. Nintedanib was frequently associated with gastrointestinal disturbances such as diarrhea, elevated liver enzyme levels, and adverse events related to cardiac disorders, including ischemic heart disease [41]. Thus, despite some beneficial effects, nintedanib has a considerable adverse event profile. Dasatinib, a second-generation tyrosine kinase inhibitor, was chosen for our study because it is safe, presents potent antifibrotic effects, and has a lower cost [8]. Additionally, a potential therapeutic strategy for the treatment of silicosis would be to administer drugs that could induce the formation of 'regulatory'-like macrophages at sites of inflammation. Dasatinib induces several hallmark features of 'regulatory'-like macrophages. Treatment of macrophages with dasatinib increases production of IL-10 while suppressing production of TNF-α [42].

Conclusions
Tissue fibrosis causes organ failure and death in patients with silicosis, and, to date, there are no clearly effective antifibrotic therapies available. As dasatinib has been demonstrated to reduce lung inflammation, minimize remodeling through inhibition of the pro-inflammatory cascade and stimulation of anti-inflammatory and antifibrotic cells, and improve lung mechanics, it may be an interesting option for the treatment of silicosis.