Nitrated fatty acids reverse pulmonary fibrosis by dedifferentiating myofibroblasts and promoting collagen uptake by alveolar macrophages

Patient samples

Human lung tissues were obtained from excess pathological tissue after lung transplantation and organ donation, under a protocol approved by the University of Pittsburgh Institutional Review Board. IPF lung tissues were obtained from explanted lungs of subjects with advanced IPF, and control lungs were donated lungs not suitable for transplantation from the Center for Organ Recovery and Education (CORE). Lung tissues were stored at −80°C until future usage.

Animals

Male C57BL/6 mice were obtained from Jackson Laboratories (Bar Harbor, ME, USA) and were used at 6–8 wk of age (20–25 g). All studies were performed according to protocols reviewed and approved by the Atlanta Veterans Affairs Medical Center Institutional Animal Care and Use Committee.

Fibroblast isolation and culture

Fibroblast cultures were established as described previously (20) from freshly obtained patient lung samples by mincing them in sterile PBS and placing tissue pieces in 100 mm tissue culture dishes containing DMEM supplemented with 10% fetal bovine serum (FBS), 10,000 U/ml penicillin, and 10,000 μg/ml streptomycin (HyClone, Logan, UT, USA) at 37°C in a humidified atmosphere of 5% CO₂-95% air. After cells grew out from the explants, they were trypsinized and plated in supplemented DMEM. Cells were used between passages 6 and 10. Human fetal lung fibroblast (IMR-90) cells were obtained from the Coriell Institute for Medical Research (Camden, NJ, USA), and maintained under the same conditions as patient-derived fibroblasts. Monolayer cultures at 90% confluence were deprived of serum for 24 h before treatment with test compounds as indicated.

To generate PPARγ^−/− fibroblasts, mice carrying a tamoxifen-inducible Cre-recombinase (Cre-ERT) under the control of a fibroblast-specific regulatory sequence from the procollagen type II, α1 (Col2α1) promoter sequence were crossed with mice that were homozygous for the PPARγ flox allele (Jackson Laboratories), which generated Col2α1-cre^+/− PPARγ flox^+/− heterozygote mice. The second cross resulted in Col2α1-cre^+/+ PPARγ flox^+/+ homozygote mice. Mice were then genotyped with specific primers to detect the Cre-ERT and PPARγ flox alleles via a PCR reaction. Mouse lung fibroblasts were isolated as described above, and PPARγ was deleted by treating cells with a stock solution of tamoxifen (4-hydroxitamoxifen; Sigma-Aldrich, St. Louis, MO, USA) as indicated.

Western blotting

Total protein extracts were prepared and Western blotting was performed as described previously (21). Antibodies against PPARγ, α-tubulin, TGFβ receptor 1 (TβR1), p-SMAD2/3, SMAD2/3, and Col1A were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibody against p-Ser was from Cell Signaling Technology (Beverly, MA, USA). Antibody against α-SMA was from Dako North America, Inc. (Carpinteria, CA, USA). Antibody against milk fat globule-EGF factor 8 (MFG-E8) was from R&D Systems (Minneapolis, MN, USA). Following primary antibody reaction, the membrane was washed in Tris-buffered saline with Tween 20 (TBST) and incubated with a 1:5000 dilution of secondary antibodies consisting of donkey anti-mouse IR-680 (red) and goat anti-rabbit IR-780 (green), both from LI-COR Biosciences (Lincoln, NE, USA), for 1 h at room temperature. The infrared signal was detected using an Odyssey infrared imager (LI-COR).

RNA isolation and real-time RT-PCR

RNA was isolated using the RNeasy Mini kit (Qiagen, Valencia, CA, USA), and cDNA was generated from 100 ng of total RNA using MultiScribe reverse transcriptase (Applied Biosystems, Foster City, CA, USA) employing random and oligo-dT primers. Real-time PCR was performed using 100 ng cDNA with 2× SYBR Green Master mix (Applied Biosystems) and specific primers for the genes of interest (Supplemental Table S1). These experiments were performed on an AB 7500 fast thermal cycler (Applied Biosystems) using a 3-step protocol employing the melting curve method. The average of each gene cycle threshold (C_t) was determined for each experiment. Relative cDNA levels (2^−ΔΔCt) for the genes of interest were determined using the comparative C_t method, which generates the ΔΔC_t as the difference between the gene of interest and the housekeeping genes β-actin and 9s rRNA for each sample. Each averaged experimental gene expression sample was compared to the averaged control sample, which was set to 1.

Transcription factor DNA-binding activity assay

Nuclear proteins were extracted using a nuclear extraction kit (Active Motif, Carlsbad, CA, USA), and their concentrations were determined using the BCA Protein Assay kit (Pierce, Rockford, IL, USA). Nuclear extracts were used to quantify DNA-binding activity of PPARγ using an ELISA-based kit (40196; Active Motif).

Transfecting siRNA

Cells were incubated for 8 h with a liposome complex containing siRNA and Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) under serum- and antibiotic-free conditions. The siRNA, either _targeted to the indicated gene (PPARγ, TβR1, or SMAD2/3) or a scrambled control, was obtained from Santa Cruz Biotechnology. After 8 h, fresh medium with 10% FBS was added, and the cells were incubated for a further 16 h. After this 24 h incubation, cells were utilized as described.

Collagen gel contraction assay

Gel contraction assays were performed using a cell contraction assay kit (Cell Biolabs, Inc., San Diego, CA, USA), according to the manufacturer's instructions. Briefly, 0.5 ml of the cell-collagen mixture per well was added into a 24-well plate and incubated for 1 h at 37°C, forming a 3-dimensional collagen gel. The gels were separated from the wells using a sterile pipette tip. The gels were then either untreated or treated with test compounds as indicated. The gels were imaged and weighed after 48 h of treatment. Gel contraction was quantitated as percentage contraction by comparing the weights of different treatment groups with cell-free gels.

Immunoprecipitation

Total protein extracts were prepared and were immunoprecipitated using the Dynabeads Protein G Immunoprecipitation Kit (Invitrogen). Antibodies were bound to Dynabeads Protein G, and the Dynabeads-Ab complex was used to precipitate _target proteins from the total protein extracts. Unbound proteins were washed away and complexes were eluted. All samples (20 μg/lane) were separated by electrophoresis on SDS-polyacrylamide gels and transferred to membranes, and Western blotting was performed.

Chromatin immunoprecipitation (ChIP) assay

The ChIP assay was performed using SimpleChIP enzymatic ChIP kit with magnetic beads (Cell Signaling Technology). Briefly, cellular chromatin was crosslinked with 1% formaldehyde for 10 min at room temperature, the crosslinking was stopped with 0.125 M glycine, and cells were washed twice with ice-cold PBS. Nuclei were pelleted and digested by micrococcal nuclease. Following sonication and centrifugation, equal amounts of sheared chromatin were incubated overnight at 4°C with antibodies, IgG as negative control, and RNA polymerase II as positive control (Cell Signaling Technology). Protein G magnetic beads were then added, and the chromatin was incubated with rotation for 2 h at 4°C. An aliquot of chromatin that was not incubated with any antibody was used as the input control sample. Antibody-bound protein/DNA complexes were eluted and subjected to real-time PCR as described above with specific primers for α-SMA, PPARγ, and α-satellite (Supplemental Table S1).

Transient transfection assay

Transient transfection and luciferase activity assays were performed as described previously (22). Briefly, cells were transiently transfected with plasmids using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions. Following treatment with test compounds, activation was measured using the Dual Luciferase Reporter Assay Kit (Promega, Madison, WI, USA) according to the manufacturer's instructions.

Immunofluorescence staining and confocal imaging

Immunofluorescence staining of cultured cells followed by confocal imaging was carried out as described previously (23).

OA-NO₂ and bleomycin sulfate administration

Mice were anesthetized with 90 mg/kg ketamine and 10 mg/kg xylazine, administered via intraperitoneal injection, and a tracheotomy was performed. Mice were intratracheally instilled with bleomycin sulfate (Bleo; Sigma-Aldrich) as a single dose of 0.025 U in 50 μl saline. Control mice received 50 μl saline. Following Bleo administration, mice were intratracheally administered, using a small animal intubation system (Kent Scientific Corp., Torrington, CT, USA), 25 μg OA-NO₂ (Cayman Chemical, Ann Arbor, MI, USA) in 50 μl of 10% DMSO or received vehicle, daily as indicated. At 24 h following the last drug delivery, the mice were euthanized, bronchoalveolar lavage (BAL) fluid was collected, and the lungs were excised for histopathological examination and measurement of markers of fibrosis.

Lung histopathology

Lungs were inflated and fixed with 10% neutral formalin overnight at room temperature. Lung tissue was dehydrated with increasing ethanol concentrations and then embedded in paraffin. Paraffin sections (5 μm thick) were stained with hematoxylin and eosin (H&E), Masson's trichrome, or picrosirius red. A scoring system was employed as described previously (24). Scores were assigned without knowledge of the treatment involved, and the slides were presented in a random order for each examination.

Measurement of collagen, hydroxyproline, TGFβ, and matrix metalloproteinase 2 (MMP-2) levels in lung

Lung homogenates from different treatment groups were prepared as described previously (21).

Levels of collagen were determined using Sircol collagen assay (Biocolor, Ltd., Belfast, UK). Levels of hydroxyproline were determined using a Biovision hydroxyproline assay kit (Biovision, Milpitas, CA, USA). TGFβ was measured using an ELISA kit (R&D Systems) according to the manufacturer's instructions. MMP-2 activity was measured using the MMP-2 Biotrak activity assay kit (GE Healthcare, Waukesha, WI, USA) according to the manufacturer's instructions.

MFG-E8 expression and collagen uptake

Collagen uptake assays were performed as described previously (25). Briefly, alveolar macrophages (AMs) were isolated and cultured in DMEM supplemented with 10% FBS. Cells were treated with drugs at 37°C in a humidified atmosphere of 5% CO₂-95% air as described, and MFG-E8 expression was analyzed by Western blotting. Other AMs were incubated for 60 min in RPMI with 0.1% BSA and 10% mouse serum. Cells were then treated with FITC-conjugated type I collagen (Invitrogen) for 30 min, after which they were washed to remove uningested collagen, counterstained with DAPI, and mounted on glass slides. They were then examined by confocal microscopy.

TGFβ monomer formation

To demonstrate TGFβ monomer formation, human recombinant TGFβ1 (R&D Systems) was incubated with test compounds for the indicated time periods and monomer formation was analyzed by running these samples in a gradient SDS-PAGE gel under nonreducing conditions and performing Western blot analysis using antibody to TGFβ (Santa Cruz Biotechnology).

Molecular modeling and computer simulations of binding of OA-NO₂ with TGFβ

Molecular modeling and covalent docking analysis was carried out as described previously (26) using Discovery Studio 2.5 (Accelrys Inc., San Diego, CA, USA).

Statistical analysis

Data are presented as means ± sd. Differences between groups were analyzed using an unpaired t test or ANOVA, followed by a Bonferroni multiple comparison correction. Analyses were performed using GraphPad Prism 5.03 (GraphPad Software, La Jolla, CA, USA). A value of P < 0.05 was considered significant.

PPARγ and TGFβ exhibit negative crosstalk

Because recent evidence suggests beneficial actions of PPARγ agonists in IPF, we assessed the potential pathophysiological role of PPARγ in pulmonary fibrosis. We found PPARγ levels both in tissue samples (Fig. 1A) and fibroblasts, the key pathophysiological _target cell in IPF (Fig. 1B), from lungs of patients with IPF were decreased vs. those seen in samples from normal controls. In lung tissues, PPARγ-encoding mRNA transcripts were decreased by >70% (Fig. 1C), and total PPARγ activity (Fig. 1D) was likewise decreased. These data indicate that IPF is associated with decreased PPARγ levels, especially in fibroblasts, accompanied by increased expression of α-SMA, a marker of myofibroblast transdifferentiation (Fig. 1A, B).

To assess whether such down-regulation of PPARγ expression and activity could lead to up-regulation of TGFβ activity and markers of fibrosis, we used a loss-of-function approach. To suppress PPARγ gene expression, we used PPARγ-_targeted siRNA in human lung IMR-90 fibroblast cells (Fig. 1E), and tamoxifen-induced, fibroblast-specific, PPARγ gene knockout in isolated mouse lung fibroblasts (Fig. 1F). With either technique, a >95% reduction of PPARγ compared with baseline levels was seen. We found that even in the absence of exogenous TGFβ these treatments increased type 1 TβR1 levels and phosphorylation of the key TGFβ signaling molecules SMAD2 and SMAD3. PPARγ reduction also increased levels of collagen and the myofibroblast marker α-SMA (Fig. 1G) and, consistent with this, also enhanced myofibroblast transdifferentiation as assayed by a gel contraction assay reflecting the presence of contractile α-SMA (Fig. 1H). The results suggest that the loss of PPARγ function observed in patients with IPF may promote the profibrotic phenotype.

TGFβ down-regulates PPARγ in vitro

Because TGFβ plays a crucial role in the development of pulmonary fibrosis, we hypothesized that it might elicit the PPARγ down-regulation seen in IPF fibroblasts. To test this idea and investigate the molecular mechanisms involved, we assessed responses to TGFβ in cultured human lung fibroblasts. Exposure to TGFβ up-regulated expression of TβR1 and phosphorylation of SMAD2/3 (Fig. 2A) as well as SMAD transcriptional activity (Fig. 2B) accompanied by time- and concentration-dependent decreases in PPARγ transcription (Fig. 2C). ChIP analysis showed that such decreased PPARγ transcription reflected binding of SMAD to a novel SBE in the PPARγ gene's promoter region (Fig. 2D) and also revealed SMAD binding to the α-SMA promoter. TGFβ exposure also decreased PPARγ protein expression in a time- and concentration-dependent manner while increasing the proportion of PPARγ that had undergone inhibitory phosphorylation (Fig. 2E). Decreased PPARγ protein expression was also seen on confocal microscopy (Fig. 2F). Knockdown of TβR1 or of SMAD2/3 abrogated the effects of TGFβ on PPARγ mRNA (Fig. 2G) and protein expression (Fig. 2H, I). These results, together with those shown in Fig. 1, suggest that down-regulation of PPARγ may be a major mechanism by which the profibrotic activity of TGFβ and its key role in IPF pathophysiology are exerted.

NFAs up-regulate and activate PPARγ and block TGFβ effects in vitro

Because the loss-of-function studies indicated that PPARγ opposes the profibrotic effects of TGFβ, we tested whether NFA-induced up-regulation of PPARγ activity would elicit antifibrotic effects. Treating human lung fibroblasts with either OA-NO₂ or LNO₂ increased PPARγ activity (Fig. 3A) and protein expression (Fig. 3B), and also elevated transcription of the PPARγ-regulated fatty acid binding protein 4 (FABP4) gene (Fig. 3C). Thus, NFAs are not only activators but also inducers of PPARγ. We used OA-NO₂ for all further studies, as it was somewhat more potent than LNO₂.

We found that OA-NO₂ treatment inhibited TGFβ-induced α-SMA and collagen synthesis in a time- and concentration-dependent manner, accompanied by reduced SMAD2/3 activation (Fig. 3D). We also confirmed by confocal imaging that OA-NO₂ treatment increased expression of PPARγ and decreased that of α-SMA (Fig. 3E). OA-NO₂ treatment also concentration dependently inhibited TGFβ-induced myofibroblast transdifferentiation measured by gel contraction assays (Fig. 3F), in which 1 μM OA-NO₂ completely blocked TGFβ effects. Together, these results with OA-NO₂ further support the conclusion that PPARγ and its activating ligands may significantly influence IPF pathophysiology.

NFAs block bleomycin-induced pulmonary fibrosis in vivo

Our finding that NFAs inhibit the profibrotic effects of TGFβ in lung fibroblasts in vitro suggested that they might prevent pulmonary fibrosis in vivo. We tested this hypothesis by treating murine experimental pulmonary fibrosis with OA-NO₂. We induced pulmonary fibrosis in mice by intratracheal administration of bleomycin. Because bleomycin-induced inflammation largely recedes by 11 d posttreatment (27), we initiated intratracheal treatment with OA-NO₂ on d 11–20 postbleomycin in order to assess potential antifibrotic rather than earlier anti-inflammatory effects. Lungs were excised for analysis 24 h after the final OA-NO₂ delivery.

The results showed that OA-NO₂ treatment elicited major reductions in bleomycin-induced histopathological changes in lung architecture and collagen content, seen by staining with H&E, Masson's trichrome, and picrosirius red (Fig. 4A–C), and significant reductions in quantitative fibrosis score (Fig. 4D). OA-NO₂-induced reductions in fibrosis were also corroborated biochemically, by decreased collagen (Fig. 4E) and hydroxyproline (Fig. 4F) content of lung homogenates, not seen in treated nonfibrotic control mice. OA-NO₂ also decreased MMP-2 activity (Supplemental Fig. S2) and TGFβ concentrations in lung homogenates of bleomycin-treated mice (Fig. 4G), suggesting that NFAs may act in part by regulating these important profibrotic molecules.

NFA treatment reverses bleomycin-induced fibrosis

The pronounced reduction in bleomycin-induced fibrosis we saw following even substantially delayed OA-NO₂ treatment suggested that the improvement elicited by NFA treatment might include reversal of fibrosis, in addition to cessation of fibrotic progression. Accordingly, we tested this therapeutically important hypothesis in vitro and in vivo. In vitro, we assessed the ability of NFAs to up-regulate MFG-E8, a glycoprotein factor that _targets collagen for uptake and lysosomal degradation by macrophages (25). In isolated alveolar macrophages, OA-NO₂ treatment up-regulated expression of PPARγ (Fig. 5A) and MFG-E8 (Fig. 5A, B) as well as collagen uptake (Fig. 5C). Collagen uptake and MFG-E8 expression were colocalized within the same cells (Fig. 5D). We also tested the ability of NFAs to induce dedifferentiation of myofibroblasts, using isolated cultured lung fibroblasts of patients with IPF, which expressed high baseline α-SMA levels. In these cells, OA-NO₂ treatment up-regulated PPARγ expression while decreasing myofibroblast differentiation as assessed by α-SMA levels detected by Western blotting (Fig. 5E) and confocal imaging (Fig. 5F).

These findings indicated that NFAs elicit multiple activities potentially capable of reversing established pulmonary fibrosis. To test this idea we assessed whether NFA treatment can reverse fibrosis in vivo by assessing fibrosis markers after administering OA-NO₂ to mice with bleomycin-induced pulmonary fibrosis. Mice were treated for 7 d beginning 21 d postbleomycin. The levels of collagen (Fig. 5G) and α-SMA (Fig. 5H) measured in lungs of OA-NO₂-treated mice were significantly reduced compared with either baseline levels (at treatment initiation) or those in vehicle-treated controls. These results demonstrate that NFAs can reverse collagen deposition and associated myofibroblast transdifferentiation in a pulmonary fibrosis model, both in vitro and in vivo.

NFAs convert TGFβ dimers to inactive monomers in a cell-free system

The active form of TGFβ is a dimer that can be converted to an inactive monomer by thiol disruption of its disulfide bond (7, 28). We hypothesized that as electrophiles, NFAs might be capable of disrupting the disulfide bond of TGFβ by temporary addition of the monomeric thiol to the NFA electrophilic center. If so, such conversion of TGFβ to inactive monomers would represent an additional, novel mechanism of antifibrotic NFA action. We first tested this hypothesis by in silico modeling (Fig. 6A–C), which supported the idea by revealing multiple interactions between OA-NO₂ and TGFβ1 monomer, including covalent bonding to the Cys sulfur atom involved in the disulfide bond. To experimentally confirm this predicted OA-NO₂-induced conversion of TGFβ1 to inactive monomers, we incubated TGFβ1 with OA-NO₂ and/or the thiol glutathione (GSH) for 30 min, then analyzed the mixture by Western blotting under nonreducing conditions. Exposure of TGFβ1 to OA-NO₂ (1 mM) or GSH (10 mM) yielded a dramatic induction of monomeric TGFβ1, and the combined exposure to 0.1 mM OA-NO₂ and 1 mM GSH almost completely depleted the dimer form (Fig. 6D). At a concentration of 0.1 mM, OA-NO₂ alone yielded a small but discernable amount of monomeric TGFβ1. Bacitracin, a dithiol that inhibits disulfide exchange, reduced monomer production by ∼50%, perhaps by competing for OA-NO₂. Effects of 1 mM OA-NO₂ or the 0.1 mM OA-NO₂-1 mM GSH combination were largely or completely blocked by the oxidizing agent diamide. Such conversion of TGFβ to inactive monomers thus constitutes an additional mechanism by which NFAs may act to block its profibrotic effects.