The animals used in these experiments were specific pathogen-free Sprague-Dawley rats [HLA:(SD)CVF; Hilltop Laboratories, Scottdale, PA] weighing 250–300 g (~ 8 weeks old at arrival). The animals were housed in an environmentally controlled facility that was accredited by the Association for Assessment and Accreditation of Laborary Animal Care. The rats were monitored to be free of endogenous viral pathogens, parasites, mycoplasmas, Helicobacter, and cilia-associated respiratory bacillus. Rats were acclimated for at least 5 days before use and were housed in ventilated cages, which were provided with HEPA-filtered air and used Alpha-Dri virgin cellulose chips (Shepherd Specialty Papers, Watertown, TN) and hardwood Beta chips (NEPCO, Warrensburg, NY) as bedding. The rats were maintained on 2018S Teklad Global 18% rodent diet (Harlan Teklad, Madison, WI) and tap water, both of which were provided ad libitum.

Rat cytokine kits for IL-6, MCP-1, MIP-2, and TNF-αwere obtained from Biosource (Camarillo, CA). Lactate dehydrogenase (LDH) was measured within 24 hr on refrigerated samples with a COBAS MIRA Plus analyzer (Roche Diagnostics, Indianapolis, IN) using kits from Roche. Lipopolysaccharide B (LPS; from E. coli 026:B6) was obtained from Difco Laboratories (Detroit, MI). The culture medium consisted of Dulbecco’s modified Eagle medium (BioWhittaker, Walkersville, MD), 1 mM glutamine (Sigma, St. Louis, MO), 10 mM N-[2-hydroxyethyl]piperazine-N′-[2-ethane-sulfonic acid] (HEPES; Sigma), 100 U/mL penicillin–streptomycin (GIBCO Life Technologies, Grand Island, NY), 100 μg/mL kanamycin (GIBCO), and 10% (vol/vol) heat-inactivated fetal bovine serum (GIBCO).

We obtained MIN-U-SIL 5 from U.S. Silica (Berkeley Springs, WV). It was examined by proton-induced X-ray emission spectrometry for inorganic contaminants and for desorbable organic compounds by gas chromatography mass spectroscopy. The results of these analyses have been reported elsewhere (Porter et al. 2001). Silica samples were found to be > 99% pure quartz. Mean particle count diameter, determined by scanning electron microscopy, was 2.14 μm, with 99% of the particles < 5 μm. Silica was weighed and dry heated at 170°C for 24 hr to sterilize. Sterile medium was then added to the silica, which was vortexed into suspension before being added to the cell culture.

The animals were anesthetized with pentobarbital sodium (150 mg/kg body weight) and exsanguinated by cutting the abdominal aorta. AMs were obtained by bronchoalveolar lavage (BAL) according to the method of Myrvik et al. (1961). The lungs from each animal were lavaged eight times with 5 mL phosphate-buffered medium (145 mM NaCl, 5 mM KCl, 9.4 mM Na₂HPO₄, and 1.9 mM NaH₂PO₄, pH 7.4) per gram lung weight. The cells were separated from the lavage fluid by centrifugation at 300 × g for 5 min and then washed three times by alternate centrifugation and resuspension in phosphate-buffered medium. The cells were then resuspended in the culture medium for use in all experiments. Cell number was determined by an electronic cell counter (model ZB; Coulter Electronics, Hialeah, FL).

We isolated type II cells as described previously (Miles et al. 1997). Briefly, the procedure involves perfusing the lung to remove blood, removing free AMs by BAL, digestion of lung tissue with elastase, and purification of type II cells by centrifugal elutriation. Cells isolated and purified by this method were > 85% pure type II cells as determined microscopically after staining with phosphine 3R (Jones et al. 1982).

The cells were cultured on collagen gels similar to those described by Lee et al. (1984) for growing hamster tracheal epithelial cells. Collagen gels were prepared from stock solution of collagen type I from rat tail (Sigma-Aldrich, St. Louis, MO) dissolved in 1:1,000 dilution of acetic acid in sterile distilled water overnight at 4°C. A six-well plate was layered with 0.775 mL (each well) of ice-cold collagen gel mixture consisting of 0.5 mL collagen stock, 0.15 mL 10× modified Eagle medium, and 0.125 mL 0.5 N NaOH. The mixture was allowed to polymerize for 4 hr at a humidified atmosphere of 5% CO₂ at 37°C. The polymerized collagen gels were washed with 1 mL epithelial cell growth medium before cells were plated and grown overnight.

We isolated lung fibroblasts as described by Reist et al. (Reist et al. 1991). Briefly, the lungs were perfused with normal saline, lavaged with phosphate-buffered saline (PBS) containing 0.1% glucose, and sectioned four times at 0.5-mm intervals with a McIlwain tissue chopper. The chopped lung tissue from a single rat was digested in 20 mL of HEPES-buffered solution (145 mM NaCl, 5 mM KCl, 1 mM CaCl₂, 505 mM glucose, and 10 mM HEPES, pH 7.4), containing collagenase (0.1%), elastase (40 U/mL), bovine serum albumin (0.5%), and DNAse (0.018%) in a shaker water bath for 30 min at 37°C. The digested mixture was filtered through two layers of sterile gauze that had been washed with culture medium. The cells were sedimented by centrifugation and plated in six-well culture plates. The medium was changed 24 hr later, and the cells were allowed to grow to confluence.

Type II cells cultured overnight on collagen gels in six-well plates (catalog no. 353046; tissue culture treated by vacuum gas plasma, polystyrene, nonpyrogenic; Becton Dickinson, Franklin Lakes, NJ), as described above, were incubated for an additional 4 hr at 37°C in a CO₂ incubator with freshly isolated AMs (1 million cells) with or without silica. Controls were type II cells alone with or without silica. The collagen gels were dissolved in a solution containing 1 mg of Sigma blend collagenase type F made up in 1 mL of type II cell growth medium for each well to be dissolved. The cells were then spun down and used for isolation of total RNA.

Lung fibroblasts were cultured until they became confluent. The cells were trypsinized and 2 × 10⁶ cells were plated in six-well plates. After overnight culture, freshly isolated AMs (2 × 10⁶ cells) were added to the wells and cultured for an additional 4 hr with or without silica. Controls were fibroblasts alone with or without silica. The culture medium was aspirated and spun down, and the supernatant was stored at –80°C. The cells were scraped and combined with the cell pellet from the above step and used for isolation of total RNA.

To measure messenger RNA (mRNA) expression in separated cell populations and to study the interaction of soluble mediators released by cell populations on each other, we conducted experiments in Transwell chambers (CoStar, Corning, NY). For these experiments, cultured lung fibroblasts were trypsinized, and 1 million cells were plated in the outer well of a Transwell plate and cultured for an additional 24 hr. At the end of the 24-hr period, freshly isolated AMs (1 million cells) were placed in the inserts. Silica was added either to the macrophages in the inner wells or to the fibroblasts in the outer well and incubated for 4 hr. Total RNA was isolated from each population separately.

We obtained AM- and polymorphonuclear neutrophil (PMN)–enriched fractions from BAL fluid obtained from rats treated with silica in vivo, as described by Huffman et al. (2003). Briefly, the method consisted of layering BAL cell populations obtained by lavage onto a Histopaque double-density gradient composed of equal amounts of Histopaque 1083 and Histopaque 1119 (Sigma). The gradients were then centrifuged (400 × g, 30 min, room temperature). The AM-enriched fraction localized at the interface between PBS diluent and Histopaque 1083, and the PMN-enriched fraction was located at the bottom as a pellet. This method yields about 60% AMs in the AM-enriched fraction and 90% PMN in the PMN-enriched fraction (Huffman et al. 2003).

We measured the cytokines in culture supernatants after a 24-hr incubation with either 200 μg/mL silica or 1 μg/mL LPS. IL-6, MCP-1, MIP-2, and TNF-α were measured by ELISA kits according to manufacturer instructions (Biosource International, Camarillo, CA). The values were expressed as nanograms or picograms per million cells. For measurement of cytokines in the BAL fluid, lavage fluid from the first wash was collected and spun down to sediment the cellular elements. The supernatant was stored at –80°C for later measurement of cytokine levels by ELISA.

We measured cytokine mRNA levels using a SYBR Green PCR kit with the ABI 5700 Sequence Detector (PE Applied Biosystems, Foster City, CA). Total RNA was isolated using RNAqueous 4PCR kits (Ambion, Austin, TX) from AMs (≈2 million cells) or lung tissue after alveolar lavage (≈50 mg wet tissue). One to two micrograms of the DNAse I–treated RNA was reverse transcribed, using Superscript II (Life Technologies, Gaithersburg, MD). The complementary DNA generated was diluted 1:100, and 15 μL was used to conduct the PCR reaction according to the SYBR Green PCR kit instructions. The comparative ^CT (threshold cycle) method was used to calculate the relative concentrations (User Bulletin no. 2; ABI PRISM 7700 Sequence Detector, PE Applied Biosystems). Briefly, the method involves obtaining the C_T values for the cytokine of interest, normalizing to a housekeeping gene (18S in the present case), and deriving the fold increase compared with the control, unstimulated cells. Table 1 lists the primer sets used for these experiments. In preliminary experiments, the products were analyzed by gel electrophoresis, and a single product was obtained with each primer set. In addition, dissociation curves yielded single peaks.

All experiments were performed on pooled AMs from several animals. AMs were placed in six-well plates, incubated for 2 hr at 37°C, and washed to remove nonadherent cells. Then the cells were incubated with silica (200 μg/mL) or LPS (1 μg/mL) for 4 hr for mRNA measurements, or 24 hr for the measurement of inflammatory cytokines.

Rats were anesthetized with an intraperitoneal injection of 30–40 mg/kg body weight sodium methohexital (Brevital; Eli Lilly and Company, Indianapolis, IN) and were intratracheally instilled using a 20-gauge 4-inch ball-tipped animal feeding needle. Silica (MIN-U-SIL 5) was suspended in endotoxin-free, Ca²⁺/Mg²⁺-free PBS (BioWhittaker, Walkersville, MD), and rats received either 2 mg silica/100 g body weight or an equivalent volume of PBS. The animals were sacrificed 4 hr postexposure, and AMs were isolated as described above. The lavaged lung tissue was used for isolation of total RNA.

A paired t-test was used for in vitro experiments. A t-test assuming unequal variance or a Z-test for means was used to evaluate the in vivo data. The significance was set at < 0.05.

In initial experiments, the effect of silica treatment (200 μg/mL) on cell viability was assessed by measuring the release of LDH into the medium at the end of the 4-hr incubation time. The means ± SEs (U/L) for control versus silica-treated cells (n = 3) were, for fibroblasts, 49 ± 14 versus 49 ± 15; for type II cells, 87 ± 9 versus 91 ± 7; and for macrophages, 101 ± 13 versus 100 ± 7.

AMs were stimulated with either 200 μg/mL silica or 1 μg/mL LPS for 4 hr. The expression of 10 genes, implicated in the induction of an inflammatory response, was measured by real-time RT-PCR. The message levels of only three cytokines (MCP-1, MIP-2, and IL-6) showed a significant increase at 4 hr after in vitro exposure to silica (Figure 1). In contrast, mRNA levels for GM-CSF, ICAM-1, IL-1β, IL-10, iNOS, TGF-β1, and TNF-αwere not significantly elevated after this treatment.

To compare the effect of silica with that of bacterial endotoxin, LPS, we also measured mRNA levels of these inflammatory mediators in AMs stimulated with LPS for 4 hr in vitro (Figure 2). LPS stimulation increased message levels of IL-1β, IL-6, GM-CSF, iNOS, MCP-1, MIP-2, and TNF-αbut not those of ICAM-1, IL-10, and TGF-β1.

Figure 3 shows the mRNA expression in cells isolated from rats 4 hr after intratracheal instillation of silica (2 mg/100 g body weight). The three cytokines that showed an increase at 4 hr in vitro (IL-6, MCP-1, and MIP-2) also showed an increase in vivo. In addition, the expression of four other genes (GM-CSF, IL-1β, IL-10, and iNOS) was increased. Three genes (TGF-β1, TNF-α, and ICAM-1) showed no change either in vitro or in vivo.

We also measured cytokine expression in the lavaged lung tissue 4 hr after the intratracheal instillation of silica (Figure 4). The results were mostly similar to those seen in AMs (Figure 3). There was a significant increase in the message levels of GM-CSF, IL-1β, IL-6, iNOS, MCP-1, MIP-2, and TNF-α. No increase was seen for ICAM-1, IL-10, and TGF-β1.

One major difference between cells obtained by BAL from control rats versus silica-treated rats is the presence of a large number of PMNs in the silica-treated animals. One explanation for the differences seen in gene expression after in vitro and in vivo exposure may be that the neutrophils produce additional cytokines not seen with AMs alone. To determine the role of PMN in mRNA expression after silica treatment, we obtained AM-enriched and PMN-enriched fractions from BAL fluid of animals treated with silica in vivo. The mRNA levels were expressed in relation to the expression levels in relation to AMs. These AMs have been exposed to silica in vivo and express high levels of mRNA, as shown in Figure 3. The mRNA expression in AMs was assigned an arbitrary value of 1 for Figure 5. Figure 5 shows that mRNA expression of the seven cytokines studied was essentially the same in the two fractions, indicating that PMN enrichment is not the cause for differences between in vitro and in vivo treatments.

We measured the levels of four cytokines/chemokines (IL-6, MCP-1, MIP-2, and TNF-α) in the supernatants of AM cultures after 4 hr incubation with either silica or LPS. There was no increase in the protein levels of these mediators with silica, but LPS produced very high levels of these cytokines/ chemokines (Figure 6A–D). There was no increase in these mediators even after 24 hr incubation with silica. In contrast, the cytokine levels of IL-6, MCP-1, and MIP-2 were increased in the alveolar lavage fluid when the animals were exposed to silica for 4 hr in vivo (Figure 7). There was no increase in TNF-α levels 4 hr after exposure to silica either in vitro or in vivo (data not shown).

To determine whether the differences seen in mRNA expression in AMs exposed to silica in vitro and in vivo may be related interaction between AMs and type II cells, we performed coculture experiments. We focused on four genes that were up-regulated only after in vivo exposure. Table 2 shows the expression of these four genes in cocultures of AMs and type II cells. Essentially, there was no difference in the expression of these genes when the cells were cocultured with or without silica. These results indicate that the expression of these inflammatory mediators is not mediated by interaction between AMs and type II cells.

Figure 8 shows the expression of five genes in fibroblasts cultured alone or in the presence of silica. We included IL-6 because lung tissue showed very high levels of IL-6 mRNA levels (Figure 4). The mRNA levels were expressed relative to mRNA levels of AMs alone. It is clear that the mRNA levels for IL-6 and GM-CSF are very high in resting lung fibroblasts. IL-6 protein levels, as determined by ELISA, were 100-fold higher in culture supernatants of lung fibroblasts compared with culture supernatants of AMs (210 ± 75 vs. 2.4 ± 1.1, n = 7), indicating that the message is being translated into protein. In addition, in vitro exposure to silica caused a significant increase in mRNA levels, but this increase in mRNA levels was not reflected in an increase in protein synthesis (210 ± 75 vs. 231 ± 62, control vs. silica, n = 7). This is similar to that seen in AMs. We did not measure the GM-CSF protein levels.

Table 3 shows the relative expression of the five genes in coculture experiments. Although coculture of AMs and lung fibroblasts seems to enhance iNOS and IL-10 mRNA levels seen over and above that seen with each cell type alone, the results were too variable to draw a definitive conclusion concerning a synergistic effect. With regard to IL-6 and GM-CSF, the main source seems to be fibroblasts, but results were too variable to conclude whether cocultures with or without silica enhance the mRNA levels.

The coculture experiments do not allow determination of mRNA expression in individual cell types. Therefore, we conducted Transwell experiments to isolate RNA from each cell type and to study the roles of cell–cell contact versus soluble mediators in these interactions (Table 4). Two conclusions can be drawn from these experiments: First, there is no difference in the mRNA levels of IL-1, IL-10, and iNOS under the different conditions tested; and second, the main sources of IL-6 and GM-CSF are lung fibroblasts. Although AMs seem to enhance IL-6 and GM-CSF mRNA levels in fibroblasts, the extreme variation in the results does not permit a definitive conclusion.

Combining the observations from coculture experiments and Transwell experiments, it appears that factors in addition to cell–cell contact and soluble mediators secreted by these two cell types are involved in regulating the inflammatory mediators in in vivo situations.