Introduction

Top Abstract Introduction Materials and Methods Statistical Analysis Results Discussion References

Pleural fibrosis and diffuse malignant mesothelioma are well-recognized sequelae of asbestos-related pleural injury. Indeed, parietal pleural plaques and visceral pleural fibrosis are the most common radiographic manifestations of asbestos-related disease (1, 2). However, there has been considerable debate about the relative potential of different commercial types of asbestos (chrysotile versus amphiboles) to cause pleural injury (3, 4). Moreover, the pathogenesis of asbestos-related pleural injury remains poorly understood. Previous experimental studies in rats have shown that asbestos fibers can translocate to the pleural space after asbestos inhalation (5) or following the intratracheal injection of asbestos fibers (6). Phagocytized asbestos fibers also have been detected within rat pleural mesothelial cells as long as 2 yr after a single intrabronchial instillation of chrysotile asbestos (7).

The precise mechanisms whereby asbestos fibers may induce pleural damage have not been elucidated. Several studies have demonstrated that in vivo asbestos exposure induces the persistent recruitment of macrophages to the pleural space (5, 8). Furthermore, the functional activity of rat pleural macrophages has been shown to be upregulated after asbestos inhalation, as evidenced by increased production of nitric oxide radical (·NO) and the cytokine tumor necrosis factor- (TNF-) in these cells (5). It is conceivable that pleural macrophage-derived cytokines may stimulate pleural mesothelial proliferation and collagen synthesis (11). Also, there is evidence that reactive oxygen species (ROS) may be implicated in the pathogenesis of asbestos-induced pleural injury. Thus, some asbestos-induced effects on pleural mesothelial cells appear to be mediated by the production of ROS from redox reactions catalyzed on the fiber surface (12). Furthermore, exposure of rat pleural mesothelial cells to crocidolite asbestos fibers has been shown to cause total cellular depletion of the antioxidant glutathione (13). The role of ROS is predicated on the concept that the effects of asbestos are mediated by superoxide anion (·O₂)-driven, iron-catalyzed Haber-Weiss (Fenton) reactions that generate the hydroxyl radical (·OH). Although these reactions may account for some of the effects of crocidolite asbestos (chemical composition: Na₂[Fe³⁺]₂[Fe²⁺]₃[Si₈O₂₂][OH]₂), they do not satisfactorily explain the cytotoxic actions of chrysotile asbestos (chemical composition: Mg₃[Si₂O₅][OH]₄), which is the commercial type of asbestos that has been used most extensively in the United States. First, chrysotile contains only 2 to 3% elemental iron, whereas crocidolite contains 27 to 36% elemental iron (14). Second, several studies have shown that the capacity of chrysotile to generate ROS via Fenton reactions and to induce DNA single-strand breaks is noticeably less than that of crocidolite (15). On the other hand, some particulates, such as carbonyl iron, are nonpathogenic but are capable of generating ·OH and inducing DNA strand breaks (19). The in vivo pathogenicity of mineral particulates may therefore not always be explained readily on the basis of Fenton chemistry.

Considerable attention has recently focused on the putative role of ·NO and peroxynitrite anion (ONOO, the reaction product of ·NO with ·O₂) in mediating pulmonary injury (20). It has been shown that rat pleural mesothelial cells can generate ·NO when challenged with a combination of lipopolysaccharide (LPS) and cytokines (24), an effect that was augmented by hydrogen peroxide (25). The present study was undertaken to determine whether asbestos fibers can activate the inducible form of nitric oxide synthase (iNOS) gene and can induce in vitro formation of ·NO and ONOO by rat pleural mesothelial cells. Another aim of the study was to ascertain whether there are differences in this regard with respect to the biologic effects of chrysotile (serpentine asbestos) and crocidolite (an amphibole variety of asbestos). To assess whether any observed effect was unique for asbestos, comparisons also were made with the nonfibrogenic and noncarcinogenic particulate, carbonyl iron.

	Materials and Methods

Top Abstract Introduction Materials and Methods Statistical Analysis Results Discussion References

Mineral Samples and Reagents

Asbestos mineral samples were obtained from the National Institute of Environmental Health Sciences (NIEHS; Research Triangle Park, NC). These samples have been characterized previously (14), and have been shown to be fibrogenic to rats in vivo (26, 27) and to induce a pleural macrophage inflammatory response in a rat inhalation model (5). Carbonyl iron spheres (size range: < 1 to 10 µm; average particle size: 4.5 to 5.2 µm) were purchased from Sigma Chemical Co. (St. Louis, MO). All of the mineral samples comprised a significant respirable fraction. Details of the fiber geometry of the asbestos samples have been published previously (28). Sodium nitrite, aminoguanidine, cytochalasin B, nitrate reductase, -nicotinamide adenine dinucleotide (-NADPH), naphthylethylenediamine dihydrochloride (NEDD), Harris's hematoxylin, and LPS also were purchased from Sigma. Human recombinant interleukin-1 (IL-1), rat recombinant interferon- (IFN-), and rat recombinant TNF- were purchased from Genzyme (Cambridge, MA). Fetal bovine serum (FBS), phenol red-free Dulbecco's modified Eagle's medium (DMEM), Hanks' balanced salt solution (HBSS), L-glutamine, penicillin, streptomycin, and fungizone were obtained from Biofluids, Inc. (Rockville, MD). Cell culture freezing medium-dimethyl sulfoxide (DMSO), trypsin-ethylenediamine tetraacetic acid (EDTA), and recombinant IFN- were procured from GIBCO (Gaithersburg, MD), and sodium pentobarbital was obtained from Abbott Laboratories (North Chicago, IL). The mouse monocyte/macrophage cell line (RAW 264.7) was obtained from American Type Culture Collection (ATCC, Rockville, MD).

Rat Pleural Mesothelial Cell Cultures

Rat parietal pleural mesothelial cell cultures, derived from adult Fischer-344 rats (Charles River Laboratories, Wilmington, MA), were established and maintained as described previously (29). In brief, the rats were anesthetized with sodium pentobarbital (50 mg/kg) and immediately killed by exsanguination. Thereafter, the entire thoracic wall was removed under sterile conditions and immersed in a Petri dish in DMEM containing 10% FBS, whereupon the parietal pleural surface was scraped gently and repeatedly with a cell scraper (Costar, Cambridge, MA). The cells then were seeded into 75 cm² tissue culture flasks (Corning, Wexford, PA). Subsequently, the cultures were maintained in a humidified environment containing 5% CO₂ at 37°C in DMEM supplemented with 10% FBS, penicillin (100 U/ml), streptomycin (100 µg/ml), fungizone (2.5 µg/ml), and 2 mM L-glutamine (supplemented DMEM). For the present study, the cells were utilized from passages 10 to 12. The cultured cells displayed the typical characteristics of mesothelial cells: a polyhedral, "cobblestone" morphologic appearance, delicate surface microvilli and junctional complexes on transmission electron microscopy, and positive immunocytochemical reactivity for 40 to 55 kD cytokeratins and vimentin (29).

Generation of Pleural Mesothelial Cell Conditioned Medium

Prior to use, all particulate samples were autoclaved, suspended in supplemented DMEM, and dispersed by repeated passage through a 22-gauge needle. For nitrite (NO₂) measurements, confluent rat pleural mesothelial cell cultures were washed twice with Ca²⁺- and Mg²⁺-free HBSS and were then trypsinized for 3 min with a solution containing 0.05% trypsin and 0.02% EDTA. Thereafter, the cells were seeded into six-well culture plates (Costar) at a density of 5 × 10⁵ cells/well in 2.5 ml of supplemented DMEM. The cells were cultured until confluence in 5% CO₂ at 37°C in supplemented DMEM, the medium being changed every 2 d. When the cultures attained confluence, they were washed with HBSS and incubated, with or without the presence of added particulates, for 12 to 72 h in supplemented DMEM in 5% CO₂ at 37°C. Cultures with and without added particulates also were incubated for similar periods in the presence or absence of cytokines. For most experiments, recombinant human IL-1 (50 ng/ ml) was employed. However, some experiments utilized recombinant human IL-1 (50 ng/ml) as well as recombinant rat IFN- (500 U/ml) and recombinant rat TNF- (500 U/ml). For dose-response experiments, particulates were added at final concentrations of 1.05 to 8.4 µg/cm² ( 5 to 40 µg/ml). Supernatant-conditioned medium samples collected at various times of culture were centrifuged at 834 × g for 7 min and then assayed for NO₂ content. Cell viability was measured by lactate dehydrogenase (LDH) activity in conditioned medium, using a commercial LDH kit (Sigma).

Measurement of iNOS Gene Expression

For RT-PCR studies, confluent parietal pleural mesothelial cell cultures in 25 cm² flasks (Costar) were incubated, in the presence (4 µg/cm²) or absence of crocidolite or chrysotile asbestos fibers and in the presence (50 ng/ml) or absence of IL-1, for 2 to 12 h at 37°C in DMEM in 5% CO₂. RAW 264.7 cells stimulated with a combination of LPS (10 ng/ml) and IFN- (10 U/ml) for 24 h served as a positive control for assessment of iNOS gene induction. Total cellular RNA was extracted with the acid-guanidinium thiocyanate-phenol-chloroform method (30). One microgram of the RNA from each sample was heated at 65°C for 15 min. After being cooled on ice, messenger RNA (mRNA) transcripts were reverse transcribed, and their complementary DNA (cDNA) products were determined by PCR amplification with the 1st Strand cDNA Synthesis Kit for RT-PCR (Boehringer Mannheim, Indianapolis, IN). Expression of the gene for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was evaluated concurrently with iNOS mRNA as an internal control. PCR primers were designed from published sequences for rat iNOS (31), deposited in the GenBank data base (accession No. M92649), and found by using a primer analysis software program (OLIGO; National Bioscience, Inc., Plymouth, MN). Primers for GAPDH were determined according to those published previously (32). The primers for iNOS and GAPDH were as follows: iNOS: 5'-GGAGATCAATGTGGCTGTGC-3', 5'-AAGGCCAAACACAGCATACC-3'; GAPDH: 5'-CAGGATGCATTGCTGACAATC-3', 5'-GGTCGGTGTGAACGGATTTG-3'.

The PCR product length of iNOS was 631 bp and that of GAPDH was 441 bp, as confirmed by the patterns of fragmentation cutting with restriction enzymes. PCR amplification was conducted through 35 cycles of denaturation at 94°C for 1 min, oligonucleotide annealing at 55°C for 1 min, and extension at 72°C for 2 min. Reactions were electrophoresed in 2% agarose gels containing ethidium bromide in Tris-acetate/EDTA buffer to visualize the PCR products.

Assays for Nitrite and Nitrate Production

Formation of ·NO was determined by measuring its oxidation product, NO₂, through the Griess reaction (33). Briefly, 100 µl of conditioned medium samples were reacted with 100 µl of a freshly-prepared 1:1 (vol/vol) mixture of 0.1% N-(1-naphthyl)ethylene diamine dihydrochloride (NEDD) in water and 1% sulfanilamide in 5% (vol/vol) H₃PO₄. After incubation for 2 min at ambient temperature, the absorbance of the pink-colored reaction product was measured at 540 nm in a Rainbow microplate reader (Phoenix Research Products, Hayward, CA). Results were expressed as micromoles of NO₂, using authentic NO₂ (sodium nitrate) as the standard. To determine whether there was any significant nitrate (NO₃) production by pleural mesothelial cells, it was necessary to reduce NO₃ to NO₂ in conditioned medium before performing the Griess reaction. Accordingly, in some experiments, conditioned medium samples (50 µl) were incubated for 24 h at ambient temperature with 50 µl of a mixture containing 0.1 M phosphate buffer (pH 7.5), 100 µM -NADPH, and 1 U/ml of nitrate reductase. Thereafter, the NO₂ content of the samples was measured through the Griess reaction (33). For those studies, comparisons were made between conditioned medium samples treated with and not treated with nitrate reductase. The amount of NO₃ in conditioned medium was determined by subtracting the NO₂ value before NO₃ reduction from that obtained after NO₃ reduction.

Immunocytochemistry for iNOS Protein and Nitrotyrosine

For immunocytochemistry studies, confluent parietal pleural mesothelial-cell cultures were incubated in four-well slide chambers (LabTek, Napierville, IL), in the presence (4 µg/cm²) or absence of particulates and in the presence (50 ng/ml) or absence of IL-1, for 36 h at 37°C in DMEM in 5% CO₂. Thereafter, the cells were fixed in acetone for 10 min at ambient temperature. For iNOS protein immunochemistry, the cells were immunostained for 60 min at ambient temperature, using a mouse monoclonal IgG_2a to iNOS protein (Transduction Laboratories, Lexington, KY), diluted 1:100 in 0.05 M Tris-HCl/0.15 M NaCl buffer at a pH of 7.4. For nitrotyrosine immunocytochemistry, the cells were immunostained for 60 min at ambient temperature, using a rabbit polyclonal IgG antibody to nitrotyrosine (Upstate Biotechnology, Lake Placid, NY), diluted 1:200 in 0.05 M Tris-HCl/0.15 M NaCl buffer at a pH of 7.4. For both iNOS protein and nitrotyrosine immunocytochemistry studies, the cells were labeled using a DAKO LSBA 2 Kit (containing an appropriate, biotinylated secondary antibody and streptavidin-conjugated alkaline phosphatase), and colored with the DAKO New Fuchsin Substrate System. The cells were counterstained with Harris's hematoxylin. For assessment of iNOS protein expression, the proportions of iNOS-positive cells were determined by counting the number of immunoreactive cells in 10 randomized microscopic fields at a magnification of ×400 in each sample. To confirm the specificity of the primary antibody for nitrotyrosine, the antibody was incubated with 10 mM nitrotyrosine in 0.05 M Tris-HCl buffer immediately before addition of the antibody to the rat pleural mesothelial cells. As a negative control, the mesothelial cells were incubated with 10% normal rabbit serum in Tris-HCl buffer instead of the antinitrotyrosine antibody.

	Statistical Analysis

Top Abstract Introduction Materials and Methods Statistical Analysis Results Discussion References

The results of the NO₂ assays were expressed as the mean ± SEM of at least three experiments per category, with each experiment performed in triplicate. Statistical comparisons were made with Student's unpaired t test.

	Results

Top Abstract Introduction Materials and Methods Statistical Analysis Results Discussion References

Effect of Asbestos Fibers on iNOS Gene Expression

The effects of IL-1 and asbestos fibers on rat pleural mesothelial cell iNOS mRNA expression are illustrated in Figure 1. When either crocidolite or chrysotile fibers (4 µg/cm²) were added to pleural mesothelial cell cultures in the absence of IL-1, no induction of iNOS mRNA expression occurred. However, when the mesothelial cells were incubated with IL-1 (50 ng/ml), in or not in the presence of added asbestos fibers (4 µg/cm²), iNOS mRNA was detectable after 2 h in culture, and increased progressively for up to 12 h in culture. As expected, the positive-control RAW 264.7 cells stimulated with LPS (10 ng/ml) + IFN- (10 U/ ml) for 24 h showed strong iNOS mRNA expression. The level of GAPDH mRNA expression was similar in all of the samples tested.

View larger version (46K): [in this window] [in a new window]   Figure 1.   Time course of iNOS gene expression in cultured rat pleural mesothelial cells after crocidolite and chrysotile asbestos exposure. Shown are RT-PCR products for iNOS and GAPDH on ethidium bromide-stained gel. Positive control: RAW 264.7 cells stimulated with LPS (10 ng/ml) + IFN- (10 U/ml) for 24 h. (A) Cells cultured in the absence of IL-1. (B) Cells cultured in the presence of IL-1.

Effects of Asbestos Fibers on ·NO Formation

To determine the optimal culture time for the generation of ·NO (measured as NO₂), an initial series of experiments was performed with rat pleural mesothelial cells cultured in the presence of IL-1 (50 ng/ml) with or without added asbestos fibers (4.2 µg/cm²) for periods ranging from 12 to 72 h. As shown in Figure 2, only negligible amounts of NO₂ were produced in 12-h cultures. Thereafter, the amount of NO₂ detected in conditioned medium samples increased progressively with time in all IL-1-containing cultures, being maximal after 72 h. However, distinct differences were observed between asbestos-containing cultures and cultures devoid of asbestos fibers. Thus, 48 h and 72 h after exposure to either chrysotile or crocidolite fibers, significantly greater amounts of NO₂ were generated than in cultures stimulated with IL-1 alone. In this regard, chrysotile upregulated ·NO formation to a greater extent than did crocidolite, especially in 48-h cultures. Therefore, in all subsequent experiments, rat pleural mesothelial cells were cultured for 48 h. When asbestos fibers were added to cell cultures in the absence of IL-1, no NO₂ was generated at any time point (results not shown).

View larger version (10K): [in this window] [in a new window]   Figure 2.   Time course of induction of NO2 formation in rat pleural mesothelial cells cultured in the presence of IL-1 (50 ng/ ml) with or without added crocidolite or chrysotile asbestos fibers (4.2 µg/cm2). Cells cultured with IL-1 + chrysotile (open circles); cells cultured with IL-1 + crocidolite (closed circles); and cells cultured with IL-1 alone (open squares). *P < 0.005 in comparison with cells cultured in the presence of IL-1 alone. Mean ± SEM of three experiments per category.

To determine whether asbestos interacted with cytokines other than IL-1 to upregulate ·NO production in rat pleural mesothelial cells, cultures were incubated for 48 h with either IFN- (500 U/ml) or TNF- in the presence of added particulates (8.4 µg/cm²). No NO₂ formation was detectable in conditioned medium from cultures stimulated with TNF- alone, and only minimal NO₂ production was evident in cultures containing TNF- + crocidolite (0.92 + 0.12 µM), chrysotile (1.24 ± 0.55 µM), or carbonyl iron (0.73 ± 0.05 µM). The effects of stimulating rat pleural mesothelial cells with IFN- alone (3.30 ± 0.15 µM) or with IFN- + particulates (6.73 ± 0.23 µM for crocidolite, 8.58 ± 0.43 µM for chrysotile, and 3.20 ± 0.05 µM for carbonyl iron, respectively) were slightly greater than those seen with TNF-, but were considerably smaller than those observed with IL-1 stimulation (Figure 3).

View larger version (12K): [in this window] [in a new window]   Figure 3.   Effects of increasing doses of particulates on NO2 formation in rat pleural mesothelial cells cultured for 48 h in the presence of IL-1 (50 ng/ml). Cells cultured with IL-1 + chrysotile (open circles); cells cultured with IL-1 + crocidolite (closed circles); cells cultured with IL-1 + carbonyl iron (closed squares); and cells cultured with IL-1 alone (open square on y-axis). *P < 0.0001 in comparison with cells cultured in the presence of IL-1 alone, and P < 0.005 in comparison with cells cultured in the presence of IL-1 alone + equivalent dose of carbonyl iron; and **P < 0.005 in comparison with cells cultured in the presence of IL-1 alone. Mean ± SEM of four to seven experiments per category.

Additional studies were done to evaluate whether there was a dose-response relationship between asbestos exposure and the induction of ·NO formation by rat pleural mesothelial cells. For this purpose, either chrysotile or crocidolite asbestos fibers were added to IL-1-containing cultures at concentrations ranging from 1.05 to 8.4 µg/cm². As shown in Figure 3, a dose-response relationship was demonstrated for both crocidolite and chrysotile. However, at all concentrations tested, chrysotile was more effective at inducing NO₂ production in rat pleural mesothelial cells than was crocidolite. Moreover, significantly greater amounts of NO₂ were generated at all concentrations in chrysotile-stimulated cultures and in the high-dose (4.2 and 8.4 µg/cm²) crocidolite-stimulated cultures than were detected in cultures containing IL-1 alone.

Another set of experiments was done to assess whether the ability to upregulate ·NO formation in IL-1-containing cultures was unique for asbestos fibers. Accordingly, the nonfibrogenic (and noncarcinogenic) particulate, carbonyl iron, was added to IL-1-containing cultures at doses ranging from 1.05 to 8.4 µg/cm². As shown in Figure 3, carbonyl iron particles did not augment the capacity of IL-1 to induce NO₂ production in rat pleural mesothelial cells at any of the concentrations tested. When the effects of asbestos and carbonyl iron were compared, significantly greater amounts of NO₂ were detected in conditioned medium samples after exposure to concentrations of 2.1 to 8.4 µg/ cm² of chrysotile or 4.2 to 8.4 µg/cm² of crocidolite than were observed after exposure to comparable doses of carbonyl iron.

Because NO₂ can be generated from sources other than ·NO (34), the specificity of the ·NO-derived reaction product in conditioned medium samples was determined through use of the selective iNOS inhibitor aminoguanidine (35). This was done in parallel experiments in which rat pleural mesothelial cells were cultured with or without aminoguanidine in the presence of IL-1 (50 ng/ml) and either crocidolite or chrysotile asbestos fibers (8.4 µg/cm²). The results are illustrated in Figure 4, and indicate that aminoguanidine significantly abrogated the production of NO₂ by asbestos-containing cultures. These findings indicate that most of the NO₂ generated was due to induction of iNOS in rat pleural mesothelial cells.

View larger version (14K): [in this window] [in a new window]   Figure 4.   Effects of aminoguanidine and cytochalasin B on NO2 formation in rat pleural mesothelial cells cultured in the presence of IL-1 (50 ng/ml) with or without added crocidolite or chrysotile asbestos fibers (8.4 µg/cm2). A: Aminoguanidine (300 µM); B: cytochalasin B (10 µg/ml). *P < 0.005 in comparison with cells cultured similarly in the absence of an inhibitor; **P < 0.01 in comparison with cells cultured similarly in the absence of an inhibitor. Mean ± SEM of three experiments per category.

Effects of Particulates on LDH Release

To assess whether ·NO formation might be associated with mesothelial cellular injury, we conducted one set of experiments in which rat pleural mesothelial cells were cultured for 48 h in the presence or absence of IL-1 (50 ng/ml) with or without added particulates (8.4 µg/cm²), with the cultures analyzed both for NO₂ production and for LDH release. As illustrated in Table 1, the presence of IL-1 alone in cultures did not induce significant LDH release. However, the addition of either crocidolite or chrysotile fibers (chrysotile > crocidolite) to IL-1-containing cultures induced significantly greater LDH release from rat pleural mesothelial cells than did the same concentration of carbonyl iron particles. Furthermore, there was generally a good correlation between the ability of each particulate to induce LDH release and to stimulate NO₂ formation in IL-1-containing cultures (Table 1). These findings suggest that the generation of reactive nitrogen species (RNS) may play a role in the induction of cellular injury by asbestos fibers.

Effect of Cytochalasin B on Asbestos-induced ·NO Formation

To assess whether the phagocytic uptake of asbestos fibers was necessary for upregulating ·NO formation by rat pleural mesothelial cells, we performed additional studies with cytochalasin B, an inhibitor of actin filament assembly and phagocytosis (36). For this purpose, parallel experiments were performed in which rat pleural mesothelial cells were cultured with or without cytochalasin B (10 µg/ml) in the presence of IL-1 (50 ng/ml) and asbestos fibers (8.4 µg/ cm²). When cytochalasin B was added to cultures containing IL-1 with either crocidolite or chrysotile asbestos fibers, the amount of NO₂ detected in conditioned medium samples was significantly reduced, by approximately 45% to 49%, and was comparable with that observed in cultures stimulated with IL-1 in the absence of asbestos fibers (Figure 4).

Comparisons of NO₂ and NO₃ Generation by Rat Pleural Mesothelial Cells

Certain RNS, such as ONOO, have been shown to nitrate tyrosine residues in proteins, and ONOO can self-decompose to yield NO₃ as a product (37). Because the Griess reaction measures NO₂ and not NO₃, the detection of NO₃ in conditioned medium samples through the Griess reaction requires the prior reduction of NO₃ to NO₂. Accordingly, to determine the NO₃ content of conditioned medium, we assayed individual samples with the Griess reaction before and after NO₃ reduction. The results are shown in Figure 5. The proportions of NO₃ in conditioned medium from cultures treated with IL-1 alone (50 ng/ml) and in conditioned medium from cultures stimulated with IL-1 + particulates (4.2 µg/cm²) were similar (approximately 53% to 58% of total measurable NO₂ reaction product). Nevertheless, significantly greater NO₃ was generated in cultures stimulated with IL-1 + asbestos fibers than in cultures treated with IL-1 + carbonyl-iron particles or with IL-1 in the absence of added particulates. Notably, chrysotile-exposed cultures contained the highest NO₃ content, which was significantly greater than that recorded after crocidolite exposure.

View larger version (15K): [in this window] [in a new window]   Figure 5.   Effect of nitrate reductase treatment on NO2 content of conditioned medium from rat pleural mesothelial cells cultured in the presence of IL-1 (50 ng/ml) with or without added particulates (4.2 µg/cm2). A: Before nitrate reductase treatment; B: after nitrate reductase treatment. *P < 0.0001 in comparison with conditioned medium treated with nitrate reductase from cultures containing IL-1 alone or IL-1 + carbonyl-iron particles; **P < 0.0001 in comparison with conditioned medium treated with nitrate reductase from cultures containing IL-1 alone or IL-1 + carbonyl iron, and P < 0.005 in comparison with conditioned medium treated with nitrate reductase from cultures containing IL-1 + crocidolite. Mean ± SEM of three experiments per category.

Immunohistochemical Studies for iNOS Protein and Nitrotyrosine

In one set of experiments, the presence of iNOS protein was assessed immunohistochemically in rat pleural mesothelial cells after stimulation for 36 h with crocidolite, chrysotile, or carbonyl iron (4 µg/cm²) in the presence or absence of IL-1 (50 ng/ml). None of the particulates induced iNOS protein expression in the cultured cells in the absence of IL-1 (results not shown). However, cytoplasmic iNOS protein expression was induced in all IL-1-stimulated cultures (Table 2). Nevertheless, significantly greater numbers of cells exhibited immunoreactivity for iNOS after exposure to IL-1 + particulates than after stimulation with IL-1 alone (Table 2). Following particulate challenge, maximal iNOS expression was seen after chrysotile exposure (Figure 6 and Table 2), and the least immunoreactivity for iNOS was seen after carbonyl iron exposure (Table 2).

View larger version (133K): [in this window] [in a new window]   Figure 6.   Cytoplasmic immunolocalization of iNOS protein with alkaline phosphatase technique in majority of rat pleural mesothelial cells after stimulation with IL-1 (50 ng/ml) + chrysotile asbestos fibers (4 µg/cm2) for 36 h. No immunoreactivity was detected in cells stained without primary antibody (not shown). Original magnification: ×100.

Nitration of tyrosine residues by ONOO and other RNS has been detected by immunohistochemical evaluation of nitrotyrosine in tissues (22, 23, 40, 41). This technique was used to assess the presence of nitrotyrosine in cultured rat pleural mesothelial cells treated with particulates for 36 h in the presence or absence of IL-1. Immunoreactivity for nitrotyrosine was not detected within mesothelial cells stimulated with IL-1 alone or when the cells were cultured with crocidolite, chrysotile, or carbonyl iron in the absence of IL-1. However, strong cytoplasmic immunoreactivity for nitrotyrosine was observed within more than 90% of cells challenged with chrysotile asbestos fibers + IL-1 (Figure 7). Weaker cytoplasmic staining was evident in cells exposed to crocidolite asbestos fibers + IL-1 (results not shown). Carbonyl iron particles + IL-1 did not induce detectable nitrotyrosine within cultured mesothelial cells. Immunospecificity of the primary antibody for nitrotyrosine was confirmed when immunoreactivity of the cells was abolished by absorption of antinitrotyrosine activity with soluble nitrotyrosine (Figure 7C). Likewise, no immunoreactivity within mesothelial cells was detectable when normal rabbit serum was used instead of the primary antibody (Figure 7B).

View larger version (143K): [in this window] [in a new window]   Figure 7.   Immunoreactivity of rat pleural mesothelial cells for nitrotyrosine after stimulation with IL-1 (50 ng/ml) + chrysotile asbestos fibers (4 µg/cm2) for 36 h. Cytoplasmic immunolocalization of nitrotyrosine with alkaline phosphatase technique is seen in almost all cells (A). No immunoreactivity was detected in cells incubated with 10% normal rabbit serum instead of antinitrotyrosine antibody (B). Immunoreactivity of rat pleural mesothelial cells was abolished when the cells were coincubated with 10 mM nitrotyrosine + antinitrotyrosine antibody (C). Original magnification: ×150.

	Discussion

Top Abstract Introduction Materials and Methods Statistical Analysis Results Discussion References

The purpose of this study was to determine whether asbestos fibers can induce the formation of RNS in rat pleural mesothelial cells, and to assess whether there are differences in this regard in the relative potential of different commercial types of asbestos. Because other investigators have shown previously that IL-1 (especially when used in combination with other cytokines) can upregulate ·NO synthesis by rat pleural mesothelial cells (24, 25), we evaluated the effects of asbestos on mesothelial cells in the present study in the context of costimulation with IL-1. We found that asbestos fibers per se did not activate iNOS mRNA expression or induce rat pleural mesothelial cell production of RNS. In contrast, iNOS mRNA expression was induced after 2 h, and progressively increased over 12 h, in IL-1-containing cultures. Although no obvious synergism between IL-1 and asbestos fibers was apparent with respect to the induction of iNOS mRNA expression in mesothelial cell cultures, subtle quantitative differences would not have been detected with the RT-PCR technique used. Nevertheless, the ability of IL-1 to stimulate rat pleural mesothelial cell NO₂ formation was significantly enhanced by both amphibole and serpentine asbestos fibers. The propensity of asbestos fibers to upregulate NO₂ production in IL-1-containing cultures was found to increase in a time-dependent fashion, being maximal after 72 h. Moreover, when crocidolite or chrysotile fibers were added to IL-1-containing cultures at concentrations of 1.05 to 8.4 µg/ cm², a dose-response relationship was observed. When the mesothelial cultures were stimulated with other cytokines, such as TNF- or IFN-, considerably smaller amounts of NO₂ were generated whether or not particulates were added. Thus, the ability to upregulate ·NO production in rat pleural mesothelial cells was unique for IL-1. Similar observations have been noted previously by other investigators (24).

The addition of 300 µM aminoguanidine to the cultures significantly abrogated the generation of NO₂ induced by crocidolite and chrysotile in rat pleural mesothelial cells. Because aminoguanidine has been shown to be a selective iNOS inhibitor (35), this indicates that the majority of NO₂ generated was due to asbestos-induced iNOS activation in the pleural mesothelial cells. The fact that both commercial types of asbestos (chrysotile > crocidolite) stimulated iNOS protein expression after 36 h in IL-1-containing cultures provides further evidence of this.

We have previously shown that rat parietal pleural mesothelial cells can actively phagocytize mineral particulates such as asbestos fibers and carbonyl iron particles (28). It is therefore noteworthy that the addition of 10 µg/ ml of cytochalasin B, an agent that inhibits actin filament assembly and phagocytosis (36), to cultures containing asbestos fibers + IL-1 significantly inhibited the amount of NO₂ generated. This suggests that phagocytosis of asbestos fibers is a necessary prerequisite for upregulation of ·NO production by rat pleural mesothelial cells.

Because the phagocytic uptake of asbestos fibers was needed for maximal generation of ·NO by rat pleural mesothelial cells, it was important to establish whether the effects observed were unique for asbestos or were simply a nonspecific consequence of the phagocytic stimulus provided by asbestos fibers. Accordingly, comparisons were made between the effects of the two types of asbestos fibers and those produced by carbonyl iron particles, which are nonfibrogenic and noncarcinogenic. The addition of carbonyl iron particles to IL-1-containing cultures did not enhance the formation of NO₂, and the proportion of mesothelial cells expressing iNOS protein was significantly smaller in cultures stimulated with IL-1 + carbonyl iron than in cultures challenged with Il-1 + either crocidolite or chrysotile. Collectively, these findings suggest that properties peculiar to asbestos fibers were responsible for the effects observed in this study.

There is evidence that chrysotile and crocidolite fibers can generate ·O₂ and other ROS in pleural mesothelial cells (42, 43). Because the interaction of ·O₂ with ·NO yields ONOO (40), it was important to determine in the present study whether ONOO was produced in cultures stimulated with IL-1 and asbestos fibers. Notably, significantly greater quantities of NO₃ (measured as NO₂ after NO₃ reduction) were detected in conditioned medium samples from cultures treated with IL-1 + asbestos fibers (chrysotile > crocidolite) than were found in samples from cultures stimulated with either IL-1 alone or IL-1 + carbonyl-iron particles. These observations are consistent with asbestos-induced formation of ONOO by rat pleural mesothelial cells, since ONOO (or an intermediate compound formed by the reaction of ONOO with CO₂) has been shown to nitrate tyrosine residues in proteins, and ONOO can self-decompose to yield NO₃ as a product (37, 44). Nitration of tyrosine residues by ONOO and other RNS has been detected by immunohistochemical evaluation of nitrotyrosine in tissues (22, 23, 40, 41). Accordingly, nitrotyrosine has been used as a marker for ONOO-mediated stress. Thus, the immunolocalization of nitrotyrosine in mesothelial cells challenged with IL-1 + asbestos fibers provides additional evidence that asbestos exposure induced the production of ONOO in rat pleural mesothelial cell cultures in the present study.

Although the precise signaling pathways involved in upregulating the formation of ·NO were not addressed in this study, it is likely that complex signaling cascades may be involved, especially the mitogen-activated protein (MAP) kinase, protein kinase C (PKC), and nuclear factor-B (NF-B) pathways. A recent study involving rat pleural mesothelial cells showed that chrysotile and crocidolite asbestos fibers (but not their nonfibrous, chemically similar analogues, riebeckite and antigorite) induced protracted activation of the extracellular signal-regulated kinases (ERK), ERK1, and ERK2 of the MAP kinase cascade after the fibers had induced phosphorylation of the epidermal growth factor (EGF) receptor (45). These findings may have relevance to the present study, since ERK1/ERK2 activation has been shown to be necessary for IL-1-induced iNOS activation in rat cardiac myocytes and microvascular endothelial cells (46). There is also evidence that some asbestos-related effects appear to be mediated via PKC activation, since this pathway has been implicated in the induction of cytokine expression and ROS formation by asbestos fibers (47, 48). It is conceivable that PKC activation also may play a role in asbestos-induced RNS formation, because several studies have shown that the PKC pathway can modulate iNOS expression in a variety of rodent cell types (46, 49).

Although it is unlikely that the NF-B pathway represents the primary signaling pathway for asbestos-induced iNOS activation, this transcription factor may still play an important role in this effect, since NF-B has been shown to be involved in the activation of iNOS by cytokines (52, 53). Moreover, exposure to crocidolite asbestos has been shown to induce significant increases in nuclear protein complexes that bind the NF-B consensus DNA sequence in both rat type II epithelial cells and rat pleural mesothelial cells (54). In contrast, nonpathogenic particles, such as riebeckite and glass beads, did not enhance nuclear NF-B complexes. Whatever mechanisms may be operative in asbestos-induced iNOS activation, our findings suggest that a posttranscriptional regulatory event may be involved, since asbestos exposure enhanced iNOS protein expression without any detectable change in iNOS mRNA expression.

It is of interest that both varieties of asbestos examined in our study (chrysotile > crocidolite) induced significantly greater amounts of LDH release from IL-1-containing cultures than did similar concentrations of carbonyl iron particles. Moreover, there was good correlation between LDH release and ·NO formation in IL-1-containing cultures. These findings suggest that the generation of RNS may play a role in the induction of cellular injury by asbestos fibers. By generating RNS in pleural mesothelial cells, asbestos fibers may mediate pleural injury through several mechanisms. First, ·NO may exacerbate asbestos-induced inflammation of the pleural space by upregulating the release of inflammatory mediators in a paracrine or autocrine fashion from pleural macrophages and pleural mesothelial cells. In this regard, ·NO has been shown to enhance the release of TNF- and IL-1 from mouse peritoneal macrophages (55). Also, by mobilizing iron from crocidolite fibers or from asbestos bodies, ·NO may induce DNA strand breakages (16) or produce DNA base modification with resultant G to T transversions (56). Furthermore, rat pleural mesothelial cells stimulated with multiple cytokines and LPS are capable of generating potentially carcinogenic N-nitrosating agents from ·NO (57).

The induction of ONOO formation by asbestos fibers in pleural mesothelial cells also may play a role in pleural fibrogenesis and carcinogenesis. ONOO is a potent oxidant that can initiate cell injury via lipid peroxidation and oxidation of protein sulfhydryl moieties (58), as well as by decreasing oxygen consumption and sodium uptake through amiloride-sensitive sodium channels (59, 60). Additionally, a recent study has shown that ONOO can activate the enzyme poly(ADP-ribose) polymerase (PARP) in human pulmonary epithelial cells, thereby inducing depletion of cellular energy, inhibition of mitochondrial respiration, and increased cell permeability (61). In another study, both chrysotile and crocidolite asbestos fibers were shown to activate PARP in rat pleural mesothelial cells (62). However, the role of ONOO in PARP activation was not addressed in that study. It has also been shown that ONOO can mediate apoptosis in diverse inflammatory disorders, such as necrotizing enterocolitis (63) and cardiac allograft rejection (64). It is of interest that two recent studies have documented that asbestos fibers can induce apoptosis in pleural mesothelial cells (43, 65), and in one of these studies apoptosis was abrogated by a PARP inhibitor (43).

In conclusion, we have shown that both chrysotile and crocidolite asbestos fibers can induce the formation of RNS in cultured rat pleural mesothelial cells costimulated with IL-1. That the effects of chrysotile exposure on iNOS protein expression, NO₃ production, LDH release, and nitrotyrosine formation were noticeably greater than those of crocidolite is of special interest, because it has previously been suggested that amphibole asbestos has a greater potential than does chrysotile asbestos to cause pleural injury in vivo (3, 12). However, further studies are needed to determine the in vivo significance of these observations.