Abstract
To investigate the role of bronchiolar epithelial NF-κB activity in the development of inflammation and fibrogenesis in a murine model of asbestos inhalation, we used transgenic (Tg) mice expressing an IκBα mutant (IκBαsr) resistant to phosphorylation-induced degradation and targeted to bronchial epithelium using the CC10 promoter. Sham and chrysotile asbestos-exposed CC10-IκBαsr Tg+ and Tg− mice were examined for altered epithelial cell proliferation and differentiation, cytokine profiles, lung inflammation, and fibrogenesis at 3, 9, and 40 days. KC, IL-6 and IL-1β were increased (p ≤ 0.05) in bronchoalveolar lavage fluid (BALF) from asbestos-exposed mice, but to a lesser extent (p ≤ 0.05) in Tg+ vs Tg− mice. Asbestos also caused increases in IL-4, MIP-1β, and MCP-1 in BALF that were more elevated (p ≤ 0.05) in Tg+ mice at 9 days. Differential cell counts revealed eosinophils in BALF that increased (p ≤ 0.05) in Tg+ mice at 9 days, a time point corresponding with significantly increased numbers of bronchiolar epithelial cells staining positively for mucus production. At all time points, asbestos caused increased numbers of distal bronchiolar epithelial cells and peribronchiolar cells incorporating the proliferation marker, Ki-67. However, bronchiolar epithelial cell and interstitial cell labeling was diminished at 40 days (p ≤ 0.05) in Tg+ vs Tg− mice. Our findings demonstrate that airway epithelial NF-κB activity plays a role in orchestrating the inflammatory response as well as cell proliferation in response to asbestos.
Asbestos is a family of naturally occurring mineral fibers associated with the development of lung cancers, mesotheliomas, and fibroproliferative diseases, i.e., pleural and diffuse interstitial fibrosis or asbestosis (1, 2). Asbestos-associated lung repair has been compared with models of wound healing wherein early injury to epithelial cells by inhaled fibers leads to compensatory hyperplasia. The mechanisms of asbestos-related diseases are still unclear, but rodent inhalation models have revealed that bronchiolar and alveolar type II (ATII)3 epithelial cells, that when transformed give rise to bronchiolar and peripheral carcinomas, undergo early injury and proliferation and are linked to the initiation of inflammation and fibroproliferation (3, 4, 5).
Recently, the airway epithelium has been recognized as a critical regulator of the innate immune system. For example, examination of bronchial biopsies and isolated airway epithelial cells in culture has revealed that many inflammatory cascades become activated in lung epithelium after various stresses or infection (6). These intracellular signaling cascades promote inflammatory responses that aid in the elimination of infectious agents and/or contribute to the development of respiratory disease.
Of the many signaling cascades activated in airway epithelium in response to stimulation, NF-κB has been implicated as one of the most important in regulation of inflammation. NF-κB is a ubiquitous transcription factor that can be activated by cytokines, reactive oxygen species, growth factors, bacteria and viruses, UV irradiation, airborne particulate matter, and inorganic minerals such as asbestos or silica (6, 7, 8, 9, 10). NF-κB activity is tightly controlled by the inhibitory protein, IκBα, that is normally present in the cytosol complexed to NF-κB dimers, thereby preventing the nuclear localization of NF-κB and ensuring low basal transcriptional activity. Upon stimulation, IκBα becomes phosphorylated at serines 32 and 36 by the activity of the IκB kinase complex, then ubiquinated, and degraded through the 26S proteasome pathway. This exposes the nuclear localization sequence of NF-κB, allowing its retention in the nucleus, thus facilitating DNA binding and the transcriptional up-regulation of genes downstream of the κB motif. The regulation and degradation of NF-κB are topics of contemporary interest, as many NF-κB-inducible genes encode inflammatory chemokines and cytokines, adhesion molecules, growth factors, enzymes and transcription factors (7). For example, IL-6 (11) and IL-8 (12)/MIP-2 (13), two putative mediators of inflammation and fibrogenesis in lung, have NF-κB binding sequences in their promoter regions that are critical to their transcriptional activation. NF-κB is also linked to increased cell survival and may govern proliferative responses after stress (14).
We have shown previously that asbestos fibers cause activation of the NF-κB signaling pathway in tracheal epithelial cells in vitro (8) and in lung epithelium after inhalation of asbestos by rats (15). In vivo, striking increases in nuclear translocation of p65 (Rel-A), the subunit causing transcriptional activation of NF-κB, occur in distal bronchiolar and alveolar epithelial cells after brief exposures to asbestos fibers (15). Thus, the induction of NF-κB in airway epithelium by asbestos may be a critical event promoting asbestos-associated epithelial cell alterations, inflammation and fibrogenesis. To test this hypothesis, we created a transgenic (Tg) mouse expressing an IκBα mutant (also referred to as an IκBα superrepressor) resistant to phosphorylation-induced degradation and under transcriptional control of the CC10 promoter to inhibit NF-κB selectively in airway epithelial cells. The CC10-IκBαsr mice, as previously characterized (14, 16), were then bred into the C57BL/6 background and evaluated in a murine inhalation model of chrysotile asbestos-induced fibrogenesis at time points of peak epithelial cell proliferation, inflammation and fibrogenesis, i.e., 3, 9, and 40 days, respectively (4, 5). Based on results with CC10-IκBαsr mice after intranasal instillation of LPS (16) and in the OVA sensitization and challenge model (14), we hypothesized that acute inflammation in response to asbestos would be curtailed. However, our results show that inhalation of chrysotile asbestos causes lung recruitment of eosinophils that are proportionally greater in Tg+ mice at 9 days and minimal recruitment of neutrophils. Data also reveal candidate cytokines and chemokines that may govern altered immune cell profiles and increased mucus production by bronchiolar epithelial cells in Tg+ mice, in contrast to patterns found in the OVA model (14). In addition, we hypothesized, based on a compendium of data in tumor cells after exposure to chemotherapeutic drugs, that NF-κB might be a survival factor promoting epithelial cell mitogenesis after injury by asbestos. We demonstrate that asbestos-induced bronchiolar epithelial and interstitial cell proliferation is diminished at 40 days in CC10-IκBαsr mice, suggesting that NF-κB may be a survival factor in normal epithelial cells that modulate mitogenesis of adjacent fibroblasts after epithelial cell injury by asbestos fibers.
Materials and Methods
CC10-IκBαsr mice
These mice were originally characterized by Poynter et al. (14, 16) and bred six to eight times into the C57BL/6 background. Eight- to 12-wk-old mice were housed and allowed to acclimate for 1 wk in a HEPA-filtered clean air environment under controlled conditions of temperature, humidity, and light and provided food and water ad libitum before the initiation of inhalation exposures. Before starting the experiments, the presence or absence of the transgene was determined using Southern blot analysis on tail DNA isolated using a DNA Extraction kit (Strategene Cloning Systems), according to the manufacturer’s protocol. Ten micrograms of whole tail DNA was immobilized on a nitrocellulose membrane and hybridized with a 32P-labeled hGH probe. The signal was visualized by exposing the hybridized membrane to X-OMAT AR film (Kodak Scientific Imaging) at −80°C overnight. Mice were housed in the University of Vermont Animal Inhalation Facility during experiments. All studies were approved by the University of Vermont Institutional Animal Care and Use Committee.
Inhalation protocol CC10-IκBαsr Tg+ mice and Tg− (Tg−) littermates (n = 5 per each of four groups (Tg+ sham, Tg+ asbestos, Tg− sham, Tg− asbestos) per time point) were exposed to either ambient air (sham) or National Institute of Environmental Health Sciences reference samples of chrysotile asbestos for 6 h per day, 5 days a wk for a total of 3, 9, or 40 days in duplicate experiments. These time points were selected based on previous work showing that they represented peak times of epithelial cell proliferation, inflammation, and fibrogenesis, respectively, in lungs of chrysotile asbestos-exposed C57BL/6 mice (4, 5, 17). The chemical and physical characteristics of National Institute of Environmental Health Sciences chrysotile asbestos have been previously described (18), and fibers were endotoxin free as determined by the Limulus amebocyte lysate gel clot assay (Endosage; Charles River Laboratories). Asbestos fibers were aerosolized using a modified Timbrell generator to generate a target concentration of 7–10 mg/m3 air, as described previously (19). Aerosol characteristics and concentrations were measured daily using a Sierra cascade impactor.
Mice were euthanized via intraperitoneal injection of sodium pentobarbital and lungs were lavaged as described below. The lungs were then perfused and inflated under pressure with PBS. The left lobes were tied off, excised, fixed in 4% paraformaldehyde, and embedded in paraffin for histology and immunocytochemistry, as described previously (4, 5). The right lobes were excised and frozen at −80°C for isolation of protein or RNA.
Nuclear translocation of Rel-A (p65) in distal bronchiolar epithelium
To determine whether nuclear translocation of Rel-A occurred at initial sites of asbestos deposition and to verify that this did not occur in asbestos-exposed Tg+ mice, Tg+ and Tg− sham and asbestos-exposed mouse lungs were instilled with PBS for 5 min at a pressure of 25 cm H2O, placed into Tissue-Tek OCT Compound (Sakura Finetek), and frozen sections were prepared. Slides were fixed for 5 min in 3% paraformaldehyde in PBS, washed, and permeabilized for 20 min with 1% Triton X-100 in PBS and blocked with 10% goat serum in PBS for 1 h. Slides were then incubated at room temperature with Ab for Rel-A (10 μg/ml, SC-372; Santa Cruz Biotechnology) in 1% BSA/PBS for 3 h. Following 3× PBS washes, slides were incubated for 30 min with goat anti-rabbit Alexa 647-labeled secondary Ab (Molecular Probes) in PBS and counterstained with a 1/1000 dilution of SYTOX Green (Molecular Probes) in PBS to label nuclei. Slides were then washed and coverslipped, and distal bronchioles (defined as those with a <800-μm perimeter at ×400 magnification) were scanned using a Bio-Rad MRC 1024 confocal scanning laser microscope.
Bronchoalveolar lavage (BAL) and differential cell counts
Following euthanasia, tracheas of mice were cannulated with polyethylene tubing. Lungs were then lavaged with sterile calcium and magnesium free PBS in a total volume of 1 ml. Total cells in BAL fluid (BALF) were enumerated, and 2 × 104 cells were centrifuged onto glass slides at 600 rpm. Cytospins were stained using the Hema3 kit (Biochemical Sciences), and differential cell counts were performed on 500 cells/mouse (4).
Bio-Plex analysis of BALF cytokines and chemokines
To quantify cytokine and chemokine levels in BALF, a multiplex suspension protein array was performed using the Bio-Plex Protein Array System and a Mouse Cytokine 22-plex Panel (Bio-Rad) as described previously (5). This method of analysis is based on Luminex technology and simultaneously measures IL-1α, IL-1β, IL-2, IL-4, IL-5, IL-6, IL-9, IL-10, IL-12(p40), IL-12(p70), IL-13, IL-17, TNF-α, RANTES, MIP-1α, MIP-1β, MCP-1, KC, G-CSF, GM-CSF, IFN-γ, and eotaxin protein. Briefly, anti-cytokine/chemokine-Ab conjugated beads were added to individual wells of a 96-well filter plate and adhered using vacuum filtration. After washing, 50 μl of prediluted standards (range between 32,000 and 1.95 pg/ml) or BALF (n = 5 samples/group) were added, and the filter plate shaken at 300 rpm for 30 min at room temperature. Thereafter, the filter plate was washed, and 25 μl of prediluted multiplex biotin-conjugated detection Ab was added for 30 min. After washing, 50 μl of prediluted streptavidin-conjugated PE was added for 10 min followed by an additional wash and the addition of 125 μl of Bio-Plex assay buffer to each well. The filter plate was analyzed using the Bio-Plex Protein Array System, and concentrations of each cytokine and chemokine were determined using Bio-Plex Manager version 3.0 software.
Histopathology to evaluate inflammation and fibrosis
Paraffin sections at 3 and 9 days were stained with H&E and at 40 days by the Masson’s trichrome technique to detect collagen. Sections were evaluated by a board-certified pathologist (K. J. Butnor) using a blind coding system. Inflammation was scored on a scale from 1 through 4 (1, no inflammation; 2 mild inflammation that was rarely peribronchiolar and consisted of primarily lymphocytes; 3, moderate inflammation with peribronchiolar neutrophils, eosinophils, lymphocytes, and abundant macrophages; and 4, severe inflammation). At 40 days, fibrosis was evaluated on a 1–4 scale (1, no fibrosis; 2, focal fibrosis; 3, moderate fibrosis; and 4, severe fibrosis).
Ki-67 immunoperoxidase technique
Ki-67 is a marker of cycling cells undergoing proliferation (17, 20). Lung sections were deparaffinized in xylene, rehydrated through graded ethanols and equilibrated in PBS (17). Ag retrieval was then performed using a 1/10 dilution in PBS of 10× Dako Target Retrieval Solution (DakoCytomation) in a 95°C water bath for 40 min followed by 20 min cooling to room temperature. Sections were then treated with Dako Peroxidase Block for 30 min followed by a 5 min wash in TBS before incubation in Dako Serum Free Protein Block for 30 min. Sections were then immersed in 50 μl of a 1/25 dilution of monoclonal rat anti-mouse Ki-67 primary Ab (DakoCytomation), a 1/600 dilution of biotinylated anti-rat IgG secondary Ab (Vector Laboratories), a 1/25 dilution of rat whole serum (Zymed Laboratories) and 1% BSA in PBS at room temperature for 30 min before blocking of excess secondary Ab with normal rat serum for 1.5 h. Negative controls were incubated in PBS without primary Ab, and all sections were incubated overnight at 4°C in a humidified chamber. Sections were washed three times in TBS, treated for 30 min with HRP-streptavidin (Vector Laboratories), and incubated in 3,3′-diaminobenzidine (DakoCytomation) for 3 min. Sections were then rinsed in double-distilled H2O, counterstained for 30 s in hematoxylin, dehydrated through increasing dilutions of ethanol, and washed in xylene twice for 15 min before coverslips were mounted in Histomount (Zymed Laboratories). Ki-67-positive cells were quantitated in three compartments: 1) the distal bronchiolar epithelium/alveolar duct epithelium; 2) the peribronchiolar compartment of these same bronchioles; and 3) the lung interstitium excluding vessels and bronchioles. Distal bronchioles were restricted to those with less than an 800-μm perimeter as viewed at ×400 magnification. Ki-67 positively stained cells in all compartments presented with distinct brown vs purple nuclei, and the total number of Ki-67-positive and -negative nuclei from all bronchioles on a lung section and their peribronchiolar region were quantitated to obtain an average of the percentage of positively stained cells per animal. For the interstitial compartment, an image of the interstitium of the lung was viewed at ×400 with a grid superimposed. For each image, the percentage of Ki-67-positive cells in five boxes, excluding blood vessels, RBC, and bronchioles was determined to achieve an average (means ± SEM) per animal. Means and SEM were then calculated for all animals in each experimental group.
Alcian blue staining
We have recently reported that inhalation of chrysotile asbestos induces increased mucin production in bronchiolar epithelium as confirmed by both Alcian blue and Alcian blue/periodic acid-Schiff staining, which peaks at 9 days (5). To determine the percentage of mucus producing cells, lung sections from 3 and 9 day mice were deparaffinized, rehydrated, equilibrated in PBS, and stained with Alcian blue (5). Lung sections were then graded by a board certified pathologist (K. J. Butnor) using a blind coding system for the severity of mucous metaplasia after evaluation of >10 distal bronchioles (defined in size as <800 μm perimeter) per mouse. The numbers of Alcian blue-positive epithelial cells were scored in individual bronchioles using a 1–4 scale for severity (1, no positive cells; 2, 1–25% positive cells; 3, 26–50% positive cells; and 4, 51–100% positive cells). The extent of mucin production was expressed as the percentage of bronchioles exhibiting Alcian blue-positive epithelial cells.
RNA isolation and RNase protection assays (RPA)
Total RNA was extracted from lavaged frozen lungs using TRIzol Reagent (Invitrogen Life Technologies), followed by a cleaning procedure using a RNeasy Mini kit (Qiagen). Gene expression was assessed qualitatively and quantitatively using the MultiProbe RPA system (Riboquant; BD Pharmingen) (21). The murine cytokine template set mCK-5c was used on RNA from 9 day animals (the time of peak inflammation) using radiolabeled antisense RNA probes for Ltn, RANTES, mip-α, mip-1β, mip-2, IP-10, mcp-1, tca-3, eotaxin, L32, and gapdh. A fibrosis template was custom made by Riboquant Pharmingen and used on RNA from 40-day exposed animals with probes for IL-1β, collagen 1, 2, and 10, tgf-β1, IL-6, tenascin, L32, and gapdh. In vitro transcription was conducted by incubation of 1 μl of RNAsin (40 U/μl), 1 μl of GACU pool (10 mM ATP, 10 mM CTP, 10 mM GTP, 10 mM UTP), 2 μl of DTT (100 mM), 4 μl of transcription buffer, 1 μl of RPA template, 10 μl of [α-32P]UTP, and 1 μl of T7 RNA polymerase (20 U/μl). The mixture was incubated at 37°C for 60 min and then treated with DNase I at 37°C for 30 min. The mixture was extracted with a combination of phenol and chloroform, and the RNA was precipitated with ethanol at −80°C and collected by centrifugation at 4°C. The RNA was resuspended in 50 μl of hybridization buffer and diluted to 3 × 105 cpm/μl for a total of 2 μl/reaction. The lung RNA samples (5 μg from each mouse) were dried in a vacuum evaporator and resuspended in 8 μl of hybridization buffer. The RNA was denatured at 95°C for 3 min and then annealed to the probe overnight at 56°C. RNase digestion was conducted at 30°C for 45 min to remove all the ssRNA. The protected RNA duplexes were purified by phenol/chloroform extraction and ethanol precipitation at −80°C and collected by centrifugation at 4°C. The pelleted RNA was resuspended in 8 μl gel loading buffer, incubated at 95°C for 3 min, and put on ice until separation on a 5% polyacrylamide/8 M urea gel. Each specific hybridized product migrates according to its size, thereby allowing identification of individual bands that were assigned to specific mRNA products. After gels were dried, autoradiograms were developed and quantitated using a Bio-Rad PhosphorImager (Bio-Rad). Results were normalized to expression of the housekeeping gene, L32.
Statistical analyses
ANOVA using the Student-Newman-Keul’s procedure for adjustment of multiple pair-wise comparisons was performed to identify significant differences between groups at each time point. Analyses of statistical differences in the Alcian blue staining data and differential cell counts were performed using the nonparametric Kruskal-Wallis test. Values of p ≤ 0.05 were considered statistically significant.
Results
Our strategy was to first determine whether abrogation of NF-κB signaling in bronchial epithelial cells affected epithelial cell and fibrogenic responses to asbestos. We then examined cytokine and inflammatory cell profiles in BALF in sham and asbestos-exposed Tg+ and Tg− mice comparatively. Since chrysotile asbestos fibers preferentially deposit after inhalation in the distal bronchioles and alveolar duct regions, we focused specifically on these lung areas in our histopathologic and other analyses.
Distal bronchioles of CC10-IκBαsr Tg+ mice do not show nuclear translocation of Rel-A (p65) in response to asbestos
As shown in Fig. 1,A, distal bronchiolar epithelium from sham Tg− mice showed little nuclear localization of the NF-κB subunit, Rel-A. At 9 days after asbestos inhalation (Fig. 1,C), marked nuclear translocation of Rel-A was observed in some, but not all distal bronchioles, presumably reflecting those bronchioles in which maximum impingement and interaction of fibers with epithelial cells occurred. Sham Tg+ mice also showed little nuclear localization of Rel-A (Fig. 1,C) and no nuclear translocation in response to asbestos (Fig. 1 D).
Distal bronchioles of CC10-IκBαsr Tg+ mice exhibit decreases in chronic epithelial cell proliferation in response to asbestos
We have shown recently that the peak time point of epithelial cell proliferation, as measured by nuclear Ki-67 immunoreactivity, occurs 3 days after exposure of C57BL/6 wild-type mice to chrysotile asbestos (5). In studies here, we also measured cell proliferation in other distal lung compartments at 3, 9, and 40 days postinhalation of asbestos to determine whether they were altered by inhibition of NF-κB in bronchiolar epithelium. Bronchiolar epithelial cell proliferation was increased in asbestos-exposed Tg− and Tg+ mouse lungs at all time points; however, significantly fewer distal bronchiolar epithelial cells in Tg+ mice incorporated Ki-67 at 40 days (Fig. 2,A). Proliferation in the peribronchiolar regions of these bronchioles was also elevated in response to asbestos at all time points (Fig. 2,B). Ki-67 labeling in the lung interstitium was not elevated in response to asbestos at 3 or 9 days, but at 40 days, there was significantly less proliferation in Tg+ mice overall (Tg+ vs Tg− averaged over both sham and asbestos-exposed mice) (Fig. 2 C). No changes in any other end point examined (see below) were observed between sham Tg+ and sham Tg− mouse at any time point.
Mucin production in asbestos-exposed CC10-IκBαsr (Tg+) mice is increased in comparison to Tg− mice
We have recently reported that mucus accumulates in the bronchiolar epithelium of chrysotile asbestos-exposed C57BL/6 mice with a peak response at 9 days. Grading of Alcian blue-stained lung sections showed that Tg+ mice exhibited more mucin-positive bronchiolar epithelial cells at 9 days compared with Tg− mice (Fig. 3, Table I). In addition, the percentage of bronchioles exhibiting mucus production was increased at the earlier 3 day time point in asbestos-exposed Tg+ mice.
Group . | Sham Tg− . | Asbestos Tg− . | Sham Tg+ . | Asbestos Tg+ . |
---|---|---|---|---|
Severityb | ||||
3 days | 1.0 ± 0 | 2.0 ± 0.3 | 1.0 ± 0.3 | 2.0 ± 0.3 |
9 days | 1.0 ± 0 | 3.0 ± 0.0d | 1.3 ± 0.3 | 3.75 ± 0.3de |
Extentc | ||||
3 days | 42 ± 14 | 63 ± 16 | 15 ± 12 | 88 ± 7d |
9 days | 34 ± 10 | 25 ± 18 | 48 ± 18 | 63 ± 7 |
Group . | Sham Tg− . | Asbestos Tg− . | Sham Tg+ . | Asbestos Tg+ . |
---|---|---|---|---|
Severityb | ||||
3 days | 1.0 ± 0 | 2.0 ± 0.3 | 1.0 ± 0.3 | 2.0 ± 0.3 |
9 days | 1.0 ± 0 | 3.0 ± 0.0d | 1.3 ± 0.3 | 3.75 ± 0.3de |
Extentc | ||||
3 days | 42 ± 14 | 63 ± 16 | 15 ± 12 | 88 ± 7d |
9 days | 34 ± 10 | 25 ± 18 | 48 ± 18 | 63 ± 7 |
Values represent means ± SEM (n = 5 mice/group).
Severity was graded on a 1–4 score as described in Materials and Methods.
Extent was scored as the percentage of bronchioles exhibiting Alcian blue-positive epithelial cells.
p ≤ 0.05 compared with respective sham group.
p ≤ 0.05 compared with asbestos-exposed Tg− group.
CC10-IκBαsr mice show altered patterns of asbestos-associated inflammation compared with Tg− mice
Although we have reported increased numbers of polymorphonuclear cells in the lungs and BALF of C57BL/6 mice inhaling chrysotile asbestos (17), we have not characterized the individual cell types in BALF comparatively over time. In this study, we show in comparison to sham mice with a BALF population consisting of alveolar macrophages that both asbestos-exposed Tg+ and Tg− mice exhibit increased total cell numbers in BALF with increased proportions of eosinophils and neutrophils at 3 days and elevated numbers of eosinophils, neutrophils, and lymphocytes at 9 days (Fig. 4). At 40 days, total cell numbers in BALF are largely diminished, but small numbers of lymphocytes are observed and increased in Tg+ mice. In comparison to asbestos-exposed Tg− mice, CC10-IκBαsr Tg+ mice inhaling asbestos also demonstrated significantly more eosinophilia in BALF at 9 days.
As shown in Fig. 5, lung sections from sham Tg− and Tg+ mice exhibited no inflammation, but lungs from asbestos-exposed animals at 3 and 9 days showed mild to moderate peribronchiolar inflammation with no apparent transgene effects (Table II). After 40 days of exposure to asbestos, staining of collagen with the Masson’s trichrome technique showed the development of focal peribronchiolar fibrosis that was not seen in the lungs of sham animals (Table II, Fig. 5,B). Because the development of fibrosis was minimal in chrysotile asbestos-exposed mice as has been previously reported (4, 5, 17) and unlikely to reveal transgene differences, we screened lung homogenates from 40-day mice using a custom-made RPA template for genes linked to fibrogenesis (Fig. 6,A). Significant increases in tgf-β1, collagen 1, and tenascin mRNA levels were seen in asbestos-exposed lungs. Tgf-β1 gene expression was reduced (p ≤ 0.05) in Tg+ vs Tg− mice, and tenascin gene expression was increased only in asbestos-exposed Tg- mice in comparison to sham mice. Increased gene expression of eotaxin and mip-1α were observed in asbestos-exposed mouse lungs at 9 days (Fig. 6 B). Other genes analyzed on RPA did not demonstrate asbestos-induced changes in expression, and results were not graphed.
Group . | Sham Tg− . | Asbestos Tg− . | Sham Tg+ . | Asbestos Tg+ . |
---|---|---|---|---|
Immune cell infiltrationa | ||||
3 days | 1.3 ± 0.3 | 2.75 ± 0.3c | 1.0 ± 0.0 | 3.0 ± 0.0c |
9 days | 1.5 ± 0.5 | 3.0 ± 0.0c | 1.0 ± 0.0 | 3.5 ± 0.3c |
40 days | 1.25 ± 0.3 | 2.75 ± 0.3c | 1.0 ± 0.0 | 2.75 ± 0.3c |
Fibrosis (40 days)b | 1.0 ± 0.0 | 2.3 ± 0.3c | 1.0 ± 0.0 | 2.0 ± 0.0c |
Group . | Sham Tg− . | Asbestos Tg− . | Sham Tg+ . | Asbestos Tg+ . |
---|---|---|---|---|
Immune cell infiltrationa | ||||
3 days | 1.3 ± 0.3 | 2.75 ± 0.3c | 1.0 ± 0.0 | 3.0 ± 0.0c |
9 days | 1.5 ± 0.5 | 3.0 ± 0.0c | 1.0 ± 0.0 | 3.5 ± 0.3c |
40 days | 1.25 ± 0.3 | 2.75 ± 0.3c | 1.0 ± 0.0 | 2.75 ± 0.3c |
Fibrosis (40 days)b | 1.0 ± 0.0 | 2.3 ± 0.3c | 1.0 ± 0.0 | 2.0 ± 0.0c |
Evaluated on H&E stained sections using a 1–4 scoring system where 1, no infiltration; 2, mild infiltration; 3, moderate infiltration; and 4, severe infiltration.
Evaluated on lung sections stained with the Masson’s trichrome method to detect collagen where 1, no fibrosis; 2, focal fibrosis; 3, moderate fibrosis; and 4, severe fibrosis.
p ≤ 0.05 compared with respective sham group.
CC10-IκBαsr Tg+ mice show altered cytokine and chemokine levels in response to asbestos
A number of cytokines were induced by asbestos and significantly diminished or increased selectively in Tg+ vs Tg− mice (Fig. 7). Levels of KC, a potent neutrophil chemoattractant, were significantly higher in BALF after inhalation of asbestos by Tg− but not Tg+ mice at all time points. IL-10 was induced by asbestos at 9 days in Tg− mice, but no asbestos-induced induction of IL-10 was observed in Tg+ mice. Additionally, IL-1β and IL-6 levels were higher in asbestos-exposed Tg− vs Tg+ mice at 40 days (p ≤ 0.05). In contrast, asbestos-associated increases in IL-4, MIP-1β, and MCP-1 were significantly higher in the BALF of 9 day asbestos-exposed Tg+ vs Tg− mice. These data suggest that several asbestos-induced cytokines and chemokines not produced exclusively by bronchiolar epithelial cells are modulated by suppression of NF-κB in the bronchiolar epithelium. Levels of IL-13, IL-5, RANTES, G-CSF, and GM-CSF were significantly increased in asbestos-exposed mice at one or more time points but consistent transgene effects were not observed (data not shown). Of the panel of 22 cytokines/chemokines measured in BALF, eotaxin, MIP-1α, and TNF-α levels were low or highly variable, showing no differences between groups at any time point. Levels of IL-12 (p70), IL-17, IFN-γ, IL-1α, IL-2, IL-3, and IL-9 were detected in BALF, but levels were not significantly different in sham vs asbestos-exposed mice (data not shown).
Discussion
We have shown previously that asbestos fibers cause activation of the NF-κB signaling pathway in airway epithelial cells (8, 15). In this study, we reveal that NF-κB activation in epithelial cells of the lung modulates their differentiation and chronic proliferation by asbestos, as well as chemokines and cytokines, produced by both epithelial and nonepithelial cells. Cell signaling in epithelial cells is integral to recruitment of inflammatory cell types and chemokine/cytokine expression, as has been shown recently in by others (22) and in our previous work using the CC10-IκBαsr mouse in LPS and allergen-induced models of airway inflammation (14, 16). However, there are distinct differences between the LPS, OVA, and asbestos inhalation models and end points examined in these studies. In the LPS model (16), a single bolus of LPS was intranasally instilled into mice, and an acute time study (30 min to 24 h) revealed increased neutrophil influx into BALF and lungs of LPS-exposed Tg− vs Tg+ mice that was maximal at 4 h. ELISA on BALF revealed increased MIP-2 and TNF-α protein in these mice, and RPA on lung mip-2 mRNA levels at this time point also showed selective increases by LPS in Tg− animals. In contrast, in the OVA sensitization and challenge model (14), all analyses were performed at 48 h after a third daily administration of OVA challenge. These acute studies documented: 1) total and differential cell counts in BALF and lung histopathology, both reflecting increased eosinophilia in Tg− mice; 2) eotaxin-1, IL-4, IL-5, IL-13, and IFN-y by ELISA in BALF, some of which were reduced in Tg+ vs Tg− mice; and 3) lung mRNA levels of chemokines (RANTES, IP-10, MCP-1, eotaxin-1, and CCL20) that were increased in OVA-challenged Tg− vs Tg+ mice. Although expression of Gob5 by semiquantitative PCR and periodic acid-Schiff staining appeared less in lungs of OVA-challenged Tg+ mice, data were not examined statistically. Moreover, pulmonary function data did not reveal transgene effects.
In this study, we performed a subchronic kinetic study (3, 9, and 40 days) with CC10-IκBα sr (Tg+) and Tg− mice using physiological administration of inhaled asbestos fibers at airborne concentrations equivalent to historical occupational exposures (1, 2). We characterized novel patterns in differential cell counts in BALF by asbestos over time and profiled cytokines in BALF using a robust Bio-Plex analysis. Because asbestos fibers are inhaled over time and the majority are cleared from the lungs, it was unsurprising that the magnitude of inflammation and cytokine expression in this model was less robust than that seen with LPS (16) or the OVA model (14).
Asbestos causes initial injury to epithelial (23) and mesothelial cells (15) in vitro that is followed by compensatory proliferation. Data in Fig. 1 using incorporation of Ki-67 as a marker of proliferation correlate with earlier studies in rat and murine asbestos inhalation model using labeling with BrdU (24) or proliferating cell nuclear Ag (25). Studies show a robust increase in acute epithelial proliferation in response to asbestos that is followed by less dramatic mitogenesis over time. Data here reveal that inhibition of NF-κB in distal bronchiolar epithelial cells suppresses chronic proliferation (40 days), but not acute mitogenesis, in response to asbestos. Our results are consistent with studies implicating NF-κB as a survival factor after cell injury by chemotherapeutic agents and oxidant stress (26). We have shown that other survival pathways such as extracellular signal-regulated kinases (ERK1/2 and ERK5) are important mediators of early epithelial cell proliferation by asbestos (27) and it is likely that cross-talk between the ERK/AP-1 and NF-κB signaling pathways occur in our model. Our recent studies also reveal increased expression of cyclin D1 in the distal bronchiolar epithelium of mice inhaling chrysotile asbestos at 9 days (17). Cyclin D1 and its partner, cyclin-dependent kinase 4, are critical in promoting the G1-S phase progression via phosphorylation of the retinoblastoma protein, and expression of cyclin D1 is regulated by NF-κB in a myoblast model of proliferation and growth (28). Lack of NF-κB-regulated cyclin D1 expression in bronchiolar epithelial cells from Tg+ mice might explain decreased epithelial cell mitogenesis in response to asbestos at later time points (40 days). We also noted decreased overall trends in interstitial cell proliferation in Tg+ mice at 40 days, which reflected less Ki-67 labeling in thickened interstitial areas containing fibroblasts. These data support the concept that the extent of bronchiolar epithelial cell mitogenesis or survival correlates with fibroblast proliferation that is intrinsic to fibrogenesis. In this regard, a compendium of studies document epithelial cell modulation of fibroblast growth by cytokine production and other mechanisms (29, 30). Although focal peribronchiolar and interstitial fibrosis was noted in asbestos-exposed Tg+ and Tg− mice (Table II), longer times of exposure will be required to demonstrate possible transgene differences.
We recently reported that chrysotile asbestos caused increased expression of Gob5 and mucin production peaking at 9 days in our inhalation model (5). At later time points, mucin production is diminished, presumably because fibers move peripherally within the lung over time and fiber dose to the epithelium diminishes. Novel findings here (Table I) show that suppression of NF-κB in epithelial cells increases the severity and extent of mucin production at 9 days—moreover, increased numbers of mucin-positive epithelial cells are seen at 3 days only in asbestos-exposed Tg+ mice. Increased mucin production might have many functional ramifications in this model such as increased clearance of fibers and epithelial protection from fiber injury. Moreover, several cytokines linked to mucus metaplasia, such as IL-13 (31) and IL-4 (32) were elevated in BALF from mice exposed to chrysotile asbestos, and dramatic increases in IL-4 were seen in Tg+ mice at 9 days, the time point of greatest mucin production. Although the increased eosinophilia and mucin production at 9 days may be linked to IL-4, future inhalation experiments using anti-IL-4 Abs or IL-4 null mice will be necessary to establish a cause and effect relationship. However, it should also be noted that mucin production by CC10-IκBαsr (Tg+) vs Tg− mice in the OVA model appeared decreased, possibly reflecting less IL-13 in BALF in the absence of changes in IL-4 in Tg+ mice (14). These results suggest different mechanisms of regulation of mucin production in these models.
We also reveal here significant increases in eosinophils in the BALF of mice exposed to chrysotile asbestos that are strikingly increased in CC10-IκBαsr Tg+ mice at 9 days (Fig. 4). At this time point, neutrophils in BALF (albeit small in numbers) were significantly lower in Tg+ mice in comparison to Tg− mice. The interplay and balance between neutrophils and eosinophils may dictate the extent of inflammation as well as its general resolution by 40 days in the asbestos model. Levels of IL-5, a chemokine linked to eosinophilporesis, and eosinophil recruitment and survival were increased in both Tg− and Tg+ mice at 3 days. Although lung mRNA levels of eotaxin were increased in asbestos-exposed Tg− and Tg+ mouse lungs at 40 days, no increases in eotaxin, an NF-κB-dependent chemokine that is important for eosinophil recruitment to the lungs and is expressed by airway epithelial cells (33), were detected in the BALF of asbestos-exposed mice. In the OVA model, eotaxin-1 and IL-5 were both decreased in BALF of OVA-exposed Tg+ mice, corresponding with less eosinophilia in BALF (14). A novel observation here is that, KC levels associated with neutrophil recruitment were strikingly decreased in asbestos-exposed CC10-IκBαsr Tg+ mice at all time points.
Several proinflammatory cytokines and chemokines such as IL-1β, MIP-1β, MCP-1 and IL-6 implicated as putative mediators of inflammation and fibrogenic responses to asbestos (34), were elevated in BALF from mice exposed to asbestos, and exhibited transgene differences in expression at later time points. Although we were unable to demonstrate transgene differences in the focal fibrosis occurring in our model, decreased levels of IL-6 and IL-1β in BALF from asbestos exposed Tg+ mice at 40 days correlated with decreases in tgf-β1 and tenascin gene expression, indicators of fibrogenesis and lung remodeling.
Airway epithelial NF-κB activation has been implicated in the initiation and perpetuation of lung inflammatory responses and is, therefore, regarded as a potential target of therapies for a number of diseases. NF-κB in the airway epithelium has been studied experimentally using adenoviral vectors to induce (35) or inhibit (36) its activity and has also been inhibited using genetic approaches (14, 16, 37). These studies demonstrate the capacity of airway epithelial NF-κB to modulate neutrophilic and eosinophilic inflammatory responses to endotoxin and allergens. The data presented herein demonstrate that airway epithelial NF-κB modulates the expression of neutrophilic chemokines. Although airway epithelial NF-κB has been shown to modulate the recruitment of eosinophils in the OVA (14) and LPS (16) models, asbestos-induced eotaxin mRNA levels in lung were similar in both Tg+ and Tg− mice, and levels of eotaxin protein in BALF were comparable. However, the intriguing observation that increased eosinophilia and mucin production both occur at 9 days suggest that they may be linked to IL-4 expression.
In summary, we show in Table III, the different responses to asbestos in Tg+ vs Tg− mice. These results suggest that NF-κΒ activity in the bronchiolar epithelium is not only important in modulation of airway epithelial proliferation and differentiation, but also is involved in the regulation of cytokine production and recruitment of inflammatory cells to the lung. Our studies lend further support to the hypothesis that epithelial cells have an impact on innate and possibly adaptive responses to asbestos.
End Points . | . | . | . | . | . | |||||
---|---|---|---|---|---|---|---|---|---|---|
Days . | Cytokine production . | Differential cell counts . | Mucin production . | Cell proliferation . | Lung gene expression . | |||||
3 | <KC | NDa | ND | ND | NEb | |||||
<KC, >IL-4 | ||||||||||
9 | >MCP-1, >MIP-1β, No>IL-10c | >Eosinophils | Increased | ND | NE | |||||
40 | <KC, <IL-6, <IL-1β | >Lymphocytes | NE | <Bronchiolar epithelial | <TGF-β1 | |||||
<Interstitial | No>tenascin |
End Points . | . | . | . | . | . | |||||
---|---|---|---|---|---|---|---|---|---|---|
Days . | Cytokine production . | Differential cell counts . | Mucin production . | Cell proliferation . | Lung gene expression . | |||||
3 | <KC | NDa | ND | ND | NEb | |||||
<KC, >IL-4 | ||||||||||
9 | >MCP-1, >MIP-1β, No>IL-10c | >Eosinophils | Increased | ND | NE | |||||
40 | <KC, <IL-6, <IL-1β | >Lymphocytes | NE | <Bronchiolar epithelial | <TGF-β1 | |||||
<Interstitial | No>tenascin |
ND, No difference compared with Tg− mice.
NE, Not evaluated.
No increase in IL-10 in contrast to Tg− mice.
Acknowledgments
We thank Dr. David Hemenway and Justin Robbins (Votey Inhalation Facility, University of Vermont) for performing inhalation experiments. Stacie Beuschel provided valuable technical assistance. We also thank Professor Emiel Wouters for his support and guidance.
Disclosures
The authors have no financial conflict of interest.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
This work is supported by Grant P01HL67004 from the National Heart, Lung and Blood Institute.
Abbreviations used in this paper: ATII, alveolar type II; BAL, bronchoalveolar lavage; BALF, BAL fluid; RPA, RNase protection assay; Tg, transgenic.