INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

THE AIRWAY EPITHELIUM is a first contact site for external stimuli such as infectious agents, exogenous inhaled antigens, and noxious agents. It may serve as a first line of defense against these stimuli by providing mucus secretion and mucociliary clearance to remove the exogenous materials, or it may initiate or amplify the local inflammatory reaction in the airway (21). In the setting of the airway inflammatory reaction, epithelial cells may function as both target and effector cells. As target cells, their defense functions such as mucociliary clearance, mucus secretion, and water and ion transport may be altered by a number of proinflammatory lipid or cytokine mediators. As effector cells, epithelial cells can produce a number of cytokine products that may directly or indirectly participate in the inflammatory cascade via the recruitment, activation, and alteration of the survival of inflammatory cells within the airway. Human airway epithelial cells also synthesize and release a variety of lipid mediators including, primarily, prostaglandin (PG) E₂, 15-hydroxyeicosatetraenoic acid (HETE), PGF₂, and platelet-activating factor (PAF) (3, 13, 15, 35). In the setting of airway inflammation, these same cells are exposed to other lipid mediators produced locally by resident or recruited inflammatory cells. These mediators include leukotriene (LT) B₄, 5-HETE and LTD₄.

Airway epithelial cells produce a variety of cytokine products including - and -chemokines, colony-stimulating factors, lymphocyte chemoattractant factor, and pleiotropic cytokines such as interleukin (IL)-6, IL-11, tumor necrosis factor (TNF)-, and IL-1 (23). Secretion of these cytokines by the airway epithelium may be a primary response to external stimuli, or it may serve as a secondary response to an inflammatory mediator. It is clear that some cytokines, such as TNF- or IL-1, can stimulate epithelial cell cytokine production. Furthermore, lipid mediators such as PAF can stimulate subsequent cellular production of lipid mediators (38). It has been reported that murine alveolar macrophages respond to PAF, LTB₄, and PGE₂ with the production of IL-6 (37). However, less is known of the interaction of these networks of lipid and cytokine mediators in the airway epithelium. Therefore, we studied whether PGE₂ or other lipid mediators present in significant quantities in inflamed airways can modulate cytokine production by a human bronchial epithelial cell line.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Cell culture. BEAS-2B cells, a human bronchial epithelial cell line transformed by an adenovirus 12-SV40 hybrid virus, were supplied by J. E. Lechner (National Cancer Institute, National Institutes of Health, Bethesda, MD) (19). The cells were cultured in the serum-free, hormonally defined culture medium LHC-8 (Biofluids, Rockville, MD) and grown on 175-cm² tissue culture flasks (Falcon, Becton Dickinson, Oxnard, CA) that were coated with a thin layer of type I rat tail collagen (Collaborative Research, Bedford, MA). For passage, cells were detached from the collagen film with 0.02% trypsin, 1% polyvinylpyrrolidine, and 0.02% ethylene glycol bis (E-PET, Biofluids) that was subsequently neutralized with soybean trypsin inhibitor (Biofluids). Replicate cultures were made by passing cells into either 6-well, 35-mm plates precoated with a thin layer of type I rat tail collagen (Collaborative Research) for secretion studies or into 175-cm² tissue culture flasks coated with a thin layer of type I rat tail collagen for gene expression studies. Experiments were performed when the cells were near confluence.

Reagents. PGE₂, PGF₂, 5-HETE, LTB₄, LTD₄, and PAF were purchased from Calbiochem (San Diego, CA). 17-Phenyl trinor PGE₂, a selective EP-1 > EP-3 receptor agonist (2), 11-deoxy-16,16-dimethyl-PGE₂, a relatively selective EP-2 receptor agonist (16, 28, 31), and sulprostone, a selective EP-3 receptor agonist were purchased from Cayman Chemical (Ann Arbor, MI). 8-Bromo-cAMP was purchased from Sigma (St. Louis, Mo).

Experimental design. At the start of each experiment, medium (LHC-8) was replaced by fresh medium. For IL-6 time-course experiments, PGE₂ (10⁷ M), 5-HETE (10⁷ M), LTB₄ (10⁷ M), LTD₄ (10⁷ M), or PAF (10⁷ M) was added to cell cultures in six-well plates, and supernatants were collected 2, 4, 8, or 24 h later. For subsequent dose-response experiments, PGE₂ was added in concentrations ranging from 10⁷ to 10¹⁰ M, and supernatants were collected at 24 h. Similar designs were used for dose-response experiments with PGE₂ receptor-specific agonists, with doses ranging from 10⁶ to 10⁹ M. DNA was measured with benzimidazole (Hoechst 33258, Janssen Chimica, Geel, Belgium) (18), and cytotoxicity was determined by lactate dehydrogenase (LDH) assay (Sigma).

ELISA for secreted IL-6. Immunoreactive levels of IL-6 in culture medium were determined with a sandwich-type ELISA (R&D Systems, Minneapolis, MN). The microtiter plates were coated with specific murine monoclonal antibodies directed against human IL-6. One hundred microliters of cell-free culture medium diluted 1:20 were plated in duplicate for the IL-6 assay. The assay was developed by the addition of horseradish peroxidase-linked goat polyclonal antibody directed against human IL-6.

Ribonuclease protection assay. Total cellular RNA was extracted from BEAS-2B cells in 175-cm² flasks with TRI Reagent (Molecular Research, Cincinnati, OH) for both PGE₂ (10⁷ M)-stimulated and control cultures for 4 and 16 h. One hundred micrograms of total RNA were used for IL-6 assays, and 10 µg of total RNA were used for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) assays. ³²P-labeled riboprobes were made from a plasmid containing an IL-6 cDNA insert labeled to a specific activity ranging from 10⁸ to 10⁹ counts · minute¹ (cpm) · µg¹. GAPDH riboprobes were made from a commercially available template (Ambion, Austin, TX) and were labeled to a specific activity of 10⁷ to 10⁸ cpm/µg. Ribonuclease protection assay (RPA) products were separated under denaturing conditions on 6% Tris-borate-EDTA (TBE)-urea polyacrylamide gels (Novex, San Diego, CA) and detected by autoradiography. Labeled RNA markers (RNA Century Marker template set; Ambion) were employed to determine the size of protected fragments. Both the IL-6 and GAPDH probe concentrations used were in excess of target. The plasmid containing the IL-6 sequence was made by RT-PCR of total RNA from BEAS-2B cells. Total RNA was reverse transcribed into cDNA with Moloney murine leukemia virus reverse transcriptase and random hexamer primers during a 1-h incubation at 42°C (GeneAmp RNA PCR kit, PerkinElmer Cetus, Norwalk, CT). The primer pair for IL-6 amplified a 628-bp product and was composed of the following sequences: 5' primer, ATG AAC TCC TTC TCC ACA AGC GC and 3' primer, GAA GAG CCC TCA GGC TGG ACT G (GenBank accession no. M18403). Denaturation, annealing, and extension temperatures for PCR were 94°C, 55°C, and 72°C, respectively, for 1 min each for 35 cycles. This PCR product was cloned into the pCRII vector with the TA cloning kit (Invitrogen, San Diego, CA). The orientation and sequence of the insert were confirmed by automated fluorescence sequencing. The plasmid was linearized before riboprobe generation with Xba I, resulting in a 147-bp protected sequence, and cRNA transcripts were prepared with Sp6 RNA polymerase and radiolabeled with [³²P]CTP (Lofstrand Laboratories, Gaithersburg, MD). The vector containing the GAPDH sequence was linearized with Hind III and cRNA labeled as outlined above.

RT-PCR amplification of PGE receptor mRNA. PCR was performed with specific primers for the amplification of mRNA for PGE receptor subtypes 1, 2, and 4. A 274-base sequence of EP-1 mRNA was amplified with the 5' primer, 5'-TGTCCAACCTGCTGGCGCTG-3' and the 3' primer, 5'-ACGCCCACGCAGCGCTCCAC-3', corresponding to bases 216-235 and 471-490, respectively, of the published sequence (GenBank accession no. L22647). A 326-bp sequence of the EP-2 receptor subtype mRNA was amplified with a 5' primer, 5'-ATGGGCAATGCCTCCAATGACTCC-3' and a 3' primer, 5'-GCACGCGCGGCTCTCGGGCGCCAG-3' (bases 157-180 and 460-483, respectively, of GenBank accession no. U19487) (31). A 446-bp sequence of EP-2/4 mRNA was amplified with the 5' primer, 5'-CGCTGTCCTCCCGCAGACGA-3' and a 3' primer, 5'-CCACCCCGAAGATGAACATC-3' corresponding to bases 236-255 and 663-682, respectively, of the published sequence (GenBank accession no. L25124) (1). An 838-base sequence of -actin was amplified with a human -actin control amplimer set (CLONTECH Laboratories, Palo Alto, CA). HeLa cells and BEAS-2B cells were grown in culture, and total RNA was extracted with TRI Reagent (Molecular Research Center). Human lung total RNA and human kidney total RNA were obtained from CLONTECH. One microgram of RNA from each source was used as a template for reverse transcriptase (GeneAmp, PerkinElmer, Branchburg, NJ). PCR was performed with 50 ng of cDNA and 1 µM primers with 30 cycles with the following conditions: denaturation at 94°C for 1 min, annealing at 50°C for 30 s, and elongation at 72°C for 30 s. Amplified DNA was separated on a TBE-polyacrylamide gel (Novex) and visualized with SYBR Green I (Molecular Probes, Eugene, OR).

Statistical analysis. Data are shown as means ± SE. Statistical analysis was carried out on a Macintosh G3 with Microsoft Excel. P < 0.05 was considered significant. Increases in IL-6 secretion were analyzed by unpaired t-test. Concentration-dependent effects were evaluated by one-way ANOVA.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effect of PGE₂ on IL-6 secretion by BEAS-2B cells. Cell-free culture supernatants were assayed with an ELISA for IL-6. Unstimulated cells produced detectable amounts of IL-6 in medium at 2 or 4 h of culture. LTB₄, 5-HETE, LTD₄, and PAF, all at 10⁷ M, had no effect on IL-6 release at 2, 4, 8, or 24 h (Fig. 1). Significant increases in IL-6 secretion in response to PGE₂ (10⁷ M) were observed at 2, 4, 8, and 24 h (P < 0.001 for each time point; Fig. 2). Maximal cumulative increases in IL-6 secretion were detected at 24 h. Concentration-dependent increases in PGE₂-induced IL-6 secretion were studied at 24 h after stimulation with PGE₂ (10⁷ to 10¹⁰ M; P < 0.001 by ANOVA; Fig. 3). In a separate experiment, the effect of PGE₂ at 10⁶ M was tested. PGE₂ at 10⁶ M resulted in an IL-6 concentration of 3,345 ± 331 pg/ml at 24 h vs. 1,129 ± 52 pg/ml for control cultures (n = 12; P < 0.001), an effect not greater than the effect of PGE₂ at 10⁷ M. Therefore, treatment with PGE₂ at 10⁷ M appeared to induce a peak effect. No evidence of PGE₂-mediated toxicity was apparent by LDH determination when compared with control cultures at 24 h. In addition, no differences in cell number were present between PGE₂-stimulated and control cultures at 24 h as determined by cellular DNA quantification.

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Effect of the lipid mediators leukotriene (LT) B4, platelet-activating factor (PAF), and 5-hydroxyeicosatetraenoic acid (5-HETE) (A) and LTD4 (B) on interleukin (IL)-6 release into culture medium by BEAS-2B cells (n = 6 experimental points, each determined in duplicate). Cells were treated with each mediator (106 M) for 2, 4, or 8 h. Media were then harvested and assayed (see MATERIALS AND METHODS).

View larger version (10K): [in this window] [in a new window]   Fig. 2.   Time course of IL-6 secretion by BEAS-2B cells both unstimulated and in response to PGE2 (107 M). Significant rises in IL-6 secretion in response to PGE2 were observed at 2, 4, 8, and 24 h. At each point, treated samples differed from control samples (n = 9-21 experimental points, each determined in duplicate). P < 0.001 by unpaired t-test.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Effect of indicated doses of PGE2 on IL-6 secretion from BEAS-2B cells. Cells were stimulated for 24 h with PGE2; n = 6 separate experimental points, each determined in duplicate. The dose effect was significant, P < 0.001, by ANOVA.

2

2

6

2

View larger version (12K): [in this window] [in a new window]   Fig. 4.   A: time course of IL-6 secretion by BEAS-2B cells both unstimulated and in response to PGF2 (106 M). Significant increases in IL-6 secretion in response to PGF2 were noted at each time point; n = 9 experimental points, each determined in duplicate. At each time point, treated samples were different from control samples, P < 0.001 by unpaired t-test. B: effect of indicated doses of PGF2 on IL-6 secretion from BEAS-2B cells. Cells (n = 6 experimental points) were stimulated for 24 h with PGF2. The dose effect was significant, P < 0.001, by ANOVA.

6

6

8

View larger version (27K): [in this window] [in a new window]   Fig. 5.   Concentration-dependent effect on IL-6 secretion in response to relatively selective PGE2 agonists [17-phenyl trinor PGE2, 11-deoxy-16,16-dimethyl (11-deoxy DM)-PGE2, and sulprostone]. Cells were incubated for 24 h with 1 of the PGE receptor agonists (n = 6 experimental points, each determined in duplicate). Concentration-dependent increases in IL-6 secretion were observed with treatment with the EP-1 receptor agonist 17-phenyl trinor PGE2 (P < 0.001 by ANOVA) and the EP-2/4 receptor agonist 11-deoxy-16,16-dimethyl- PGE2.

Effect of 8-bromo-cAMP on IL-6 release. Ligand binding and activation of the EP-2 or EP-4 receptor subtypes resulted in the activation of adenylate cyclase and an increase in intracellular cAMP. Therefore, 8-bromo-cAMP was added to cells to determine if this cAMP analog was capable of inducing changes in IL-6 production. As demonstrated in Fig. 6, the addition of 8-bromo-cAMP to cell cultures induced dose-related increases in IL-6 release, suggesting that the PGE₂ induction of IL-6 release was, at least in part, mediated via EP-2/4 receptor activation.

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Effect of 8-bromo-cAMP on IL-6 release from BEAS-2B cells. Cells (n = 6 experimental points) were exposed to 8-bromo-cAMP at indicated concentrations for 24 h. Cell-free supernatant was assayed for IL-6. 8-Bromo-cAMP induced a dose-related increase in IL-6 production. P < 0.001 by ANOVA.

Detection of PGE₂ receptor subtype mRNAs. Because a PGE₂ receptor agonist with some specificity for the EP-1 receptor and an agonist with some specificity for the EP-2 receptor caused an increase in epithelial cell release of IL-6, studies were performed to determine whether these cells expressed mRNA for these receptors. RT-PCR was performed on total cellular RNA from BEAS-2B cells and, as controls, on total cellular RNA from HeLa cells and RNAs from lung and kidney. As can be seen in Fig. 7, RT-PCR with primers specific for mRNA for the EP-1 receptor subtype revealed an appropriately sized band of RNA from BEAS-2B cells, from lung tissue, and from kidney tissue (Fig. 7A). Similarly RT-PCR with primers specific for the EP-2 receptor subtype produced an appropriately sized band of RNA from BEAS-2B cells and from lung and kidney but not from HeLa cells (Fig. 7B). In addition, RT-PCR with primers specific for the EP-4 receptor subtype resulted in the production of an appropriately sized band (Fig. 7C). Figure 7D presents the results of RT-PCR for -actin and demonstrates undegraded mRNA in all samples. Therefore, BEAS-2B cells express mRNA for the EP-1, the EP-2, and the EP-4 receptor subtypes.

View larger version (33K): [in this window] [in a new window]   Fig. 7.   Detection of mRNA for PGE receptor subtypes by RT-PCR. Total cellular mRNA from BEAS-2B cells, HeLa cells, and lung and kidney was reverse transcribed. The resultant DNA was used as substrate. Oligonucleotide probes for sequences of the coding sequence for EP-1, EP-2, and EP-4 receptor subtypes were used for PCR (see MATERIALS AND METHODS for details). Amplified DNA was separated on a Tris-borate-EDTA polyacrylamide gel and visualized with SYBR Green I. A: results of PCR performed with the EP-1 primers to amplify a 274-bp sequence of EP-1 cDNA. B: results of PCR performed with EP-2 primers to amplify a 326-bp sequence of EP-2 cDNA. C: results of PCR performed with EP-4 primers to amplify a 446-bp sequence of EP-4 cDNA. D: results of PCR with probes specific for -actin to amplify an 838-bp sequence of -actin cDNA (positive control).

Effect of PGE₂ on IL-6 mRNA levels. PGE₂ treatment of these airway epithelial cells resulted in an increase in IL-6 production over a prolonged period of time. Therefore, experiments were performed to determine whether the PGE₂ treatment resulted in an increase in the cellular steady-state levels of IL-6 mRNA during this time. BEAS-2B cells were stimulated with PGE₂ (10⁷ M) for 4 or 16 h. Total cellular RNA was extracted, and RPAs were performed. As demonstrated in Fig. 8, PGE₂ treatment increased the steady-state levels of IL-6 mRNA at both times, suggesting that PGE₂ modulates IL-6 gene expression and therefore augments IL-6 protein synthesis in the BEAS-2B cells.

View larger version (28K): [in this window] [in a new window]   Fig. 8.   Effect of PGE2 on IL-6 mRNA levels. BEAS-2B cells were stimulated with PGE2 (107 M) for 4 or 16 h. Total cellular mRNA was extracted, and ribonuclease protection assays (RPAs) were performed (see MATERIALS AND METHODS). PGE2 treatment increased steady-state levels of IL-6 mRNA at both time points. RPA for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was performed as a control. The RPA shown is representative of 3 different RPAs, each demonstrating the same result.

2

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The airway epithelium has the ability to synthesize or release a variety of inflammatory mediators such as metabolites of arachidonic acid, PAF, and cytokines as well as products that may downregulate the inflammatory response, including eicosanoid metabolites, cytokine receptor antagonists such as IL-1 receptor antagonist, and solubilized cytokine receptors such as soluble TNF receptor (20, 22). These inflammatory mediators can participate in modulating the local inflammatory process in neighboring cells and tissues. Lipid mediators present in the airway may be produced by resident airway cells and by inflammatory cells resident in or recruited to the lung. Lipid mediators such as PAF, LTB₄, and PGE₂ may regulate cytokine expression in inflammatory cells such as alveolar macrophages and peripheral blood mononuclear cells (33, 34, 37).

IL-6 is a multifunctional cytokine important in the production of acute-phase proteins, the immune response to viruses and bacteria, immunoglobulin production, the differentiation of T cells, and, perhaps, for tissue regeneration as well as in a variety of endocrine effects (10, 29, 32). In the airway and alveolar spaces, IL-6 is produced by alveolar macrophages and airway epithelial cells (11). In experimental animals, intratracheal administration of IL-6 induces an infiltration of neutrophils into the interstitial space and into the alveolar space (11). Increased IL-6 has been demonstrated in bronchoalveolar lavage fluid from patients with asthma (25). Furthermore, increased IL-6 production has been reported in airway epithelial cells from patients with asthma (24) and from alveolar macrophages harvested in the setting of the late asthmatic reaction (9).

In this study, we report that PGE₂ treatment of a human bronchial epithelial cell line increased the cellular production of IL-6 over a 2- to 24-h period and that this effect is associated with an increase in steady-state levels of IL-6 mRNA. This effect was not produced by treatment of cells with other lipid mediators such as PAF, LTB₄, and LTD₄. Although PGE₂ may have anti-inflammatory properties including suppression of cellular production of IL-1, inhibition of leukocyte migration, inhibition of superoxide and LT release from polymorphonuclear leukocytes, and inhibition of T lymphocyte proliferation (8), PGE₂ may have important proinflammatory effects as well. These may include vasodilator properties and an increase in vascular permeability in synergy with bradykinin and LTs. In an experimental animal model, carrageenan-induced tissue inflammation, edema, and hyperalgesia were prevented by the administration of an anti-PGE₂ antibody, and this treatment also reduced tissue levels of IL-6, which suggests that PGE₂ can contribute to local inflammatory responses that have been initiated by exogenous stimuli or inflammatory cell products (12, 30). Furthermore, PGE₂ has been reported to stimulate IL-6 release from murine macrophages, murine and rat osteoblasts (17, 26), and human lymphocytes (HSB.2 cells) and astrocytoma cells (7, 40). The IL-6 produced by BEAS-2B cells after PGE₂ stimulation is roughly similar to stimulated IL-6 production by other cells: 500 pg · ml¹ · 3 h¹ from rat osteoblasts (26), 400-600 pg · ml¹ · 24 h¹ from mouse osteoblasts (17), and 200-1,000 pg · ml¹ · 12 h¹ from human lung fibroblasts stimulated by transforming growth factor- (6).

PGE receptors are pharmacologically and molecularly characterized into at least four subtypes (28). Each subtype may have distinct signal transduction pathways. The EP-1 receptor signals through activation of phosopholipase C. The EP-2 receptor signals via activation of adenyl cyclase and increases in cAMP as does the subtype designated EP-4 (1, 14, 27, 28). Activation of the EP-3 subtype signals via a reduction in cAMP (4, 28, 36). The results of the EP receptor agonist studies and the effect of 8-bromo-cAMP on IL-6 release suggest that the effect of PGE₂ on IL-6 release from BEAS-2B cells may be mediated, at least in part, through an EP-2/4 receptor linked to adenylate cyclase.

Given the ability of cytokines, including IL-6, to activate arachidonate metabolism in some tissues (5, 21), these findings suggest that there is a complex regulatory network that may be significant in the modulation of inflammatory responses in the airway mucosa and lumen. These pathways may have an important role in the pathogenesis of inflammatory airway disorders.