In this study, we investigated the regulation and mechanism of IL-8 expression by A549 human lung carcinoma cells treated with neutrophil elastase (NE). NE-treated cells exhibited significantly higher IL-8 protein levels in culture media compared with cells treated with vehicle alone. Blocking of gene transcription with actinomycin D suggested that NE stimulated IL-8 synthesis via increased mRNA expression, which was verified by real-time RT-PCR. NE activated the IL-8 promoter but did not alter the stability of its mRNA, confirming that the protease induced IL-8 synthesis through increased gene transcription. The results from the use of chemical inhibitors and mutant gene constructs against various signal transduction components seem to suggest the linear signaling pathway involving the activation of PKC-δ → dual oxidase 1 → reactive oxygen species → TNF-α-converting enzyme → EGF receptor → p38 → NF-κB for NE-activated IL-8 gene expression. A NF-κB potential binding site, located between nucleotides −82 and −69 of the IL-8 promoter, was identified as necessary for NE-induced IL-8 transcription. We conclude that NE increases IL-8 transcription through p38/NF-κB activation via EGFR transactivation.
- epidermal growth factor receptor
neutrophils are important host components for bacterial killing and clearance from sites of infection. Proteases and oxygen radicals released from activated neutrophils are necessary to kill invading bacteria, but these mediators also damage surrounding tissues. Neutrophil elastase (NE) is a serine protease stored in azurophil granules and released extracellularly after infection (5). Studies using NE knockout (NE−/−) mice showed that NE is required for defense against experimental infection by gram-negative bacteria (61). However, NE−/− mice are not at increased risk of spontaneous infection, and the biological role of NE in normal physiology remains to be established. In addition to its direct antibacterial effects, NE also contributes to immunity at mucosal surfaces by upregulating mucin gene expression (68).
IL-8 was first identified as a neutrophil chemotactic factor produced by human mononuclear cells stimulated with lipopolysaccharide (41, 73). IL-8 is produced by leukocytic cells as well as nonleukocytic somatic cells [e.g., endothelial cells, fibroblasts, and epithelial cells (3, 44, 45, 51)] in response to gram-positive and -negative bacterial infections (2, 16). IL-8 gene expression is regulated by the NF-κB and AP-1 transcription factors (4, 29, 58, 79) via a mechanism involving p38 MAPK (9, 71). Sparkman and Boggaram (62) reported that IL-8 mRNA stability was increased by nitric oxide activation of p38. On the other hand, Yu et al. (74) demonstrated that p38 induced IL-8 synthesis through translational regulation rather than transcript stabilization. Thus multiple mechanisms regulate IL-8 production.
Recently, we reported that NE increased transcription of the MUC1 membrane mucin gene (33). Our studies complemented those of others demonstrating that NE also was responsible for increased production of a major soluble mucin (MUC5AC) and another membrane mucin (MUC4) (19–21, 68). In the case of MUC5AC, Shao and Nadel (60) identified PKC-δ, dual oxidase 1 (Duox1), reactive oxygen species (ROS), TNF-α-converting enzyme (TACE), EGF receptor (EGFR), and MAPK in NE-induced gene expression. Recently, Park et al. (56) also demonstrated that NE stimulates MUC5AC and MUC5B release through PKC-δ. Interestingly, although many of these signaling components were also implicated in IL-8 expression (8, 9, 24, 30, 35, 63, 71, 75, 76), a single pathway linking NE with IL-8 has not been described. On the basis of these studies, we hypothesized that NE activates IL-8 transcription through a PKC-δ → Duox1 → ROS → TACE → EGFR → p38 → NF-κB pathway. In this study, the presence of this signaling pathway in lung epithelial cells was confirmed and a NF-κB potential binding site, located between nucleotides −82 and −69 of the IL-8 promoter, was identified as necessary for NE-induced IL-8 transcription. Thus this is the first report describing a “complete” signaling pathway of NE that is involved in the regulation of IL-8.
MATERIALS AND METHODS
All reagents were from Sigma (St. Louis, MO) unless otherwise indicated. A549, a human lung carcinoma cell line, was from American Type Culture Collection (Manassas, VA). NE was from Elastin Products (Owensville, MO). TAPI-1 was from Peptides International (Louisville, KY). SB202190 and SP600125 were from EMD Biosciences (La Jolla, CA). U0126 was from Cell Signaling (Beverly, MA). SN-50 was from Alexis Biochemicals (Lausen, Switzerland). Duox1 short interfering (si)RNA and its negative control RNA were from Ambion (Austin, TX). The effective concentrations of these inhibitors were chosen on the basis of the published results. A dominant negative PKCδ-pcDNA3.1(+) plasmid (dnPKC-δ) was kindly provided by Dr. S. Atamas (University of Maryland, Baltimore, MD) (53). A mutant IκB-α-pCMV4 plasmid (mIκB-α) was kindly provided by Dr. R. Srivastava (University of Texas, Tyler, TX) (11).
A549 cells were seeded at 5.0 × 104 cells/well in 24-well plates in DMEM (Invitrogen, Carlsbad, CA) containing 10% FBS (Atlanta Biologicals, Lawrenceville, GA), 100 U/ml penicillin, and 100 μg/ml streptomycin (Invitrogen) and incubated at 37°C in a humidified CO2 incubator. In some experiments, A549 cells were seeded at 3.0 × 104 cells/well in 24-well plates for transfection with Duox1 siRNA, dnPKC-δ plasmid, or mIκB-α plasmid. NHBE cells were obtained from Cambrex (Walkersville, MD) and cultured as previously described (6). Briefly, cells were plated in antibiotics-omitted bronchial epithelial growth medium (BEGM BulletKit, Cambrex) at a density of 2.0 × 106 cells/100-mm-diameter tissue culture dish precoated overnight at room temperature with 175 μg/dish of human placental collagen IV. At ∼70–80% confluence, the cells were harvested with 0.25% trypsin-0.38 mg/ml EDTA·4Na (Invitrogen) and seeded in human placental collagen IV-coated (7.5 μg/insert) 12-mm Millicell-CM inserts at 2.5 × 105 cells/insert in 24-well plates using an air-liquid interface (ALI) medium (Cambrex) in the upper and lower chambers. ALI medium was similar to bronchial epithelial growth medium except that a 50:50 mixture of LHC basal medium and DMEM was used as the base, amphotericin and gentamicin were omitted, and the EGF concentration was reduced to 0.5 ng/ml. Upon reaching confluence, medium in the upper chamber was removed, the apical surface of the cells was rinsed with PBS, and medium was replaced only in the bottom compartment. Five days after establishment of the ALI culture, cells were treated with various concentrations of NE for 24 h.
Effect of NE on IL-8 protein release.
Confluent monolayers of A549 cells were washed and starved for 24 h in serum-free DMEM. After starvation, the cells were washed and treated for 24 h with PBS or 100 nM (2.6 U/ml) of NE. NE was prepared as a 20 μM stock solution in PBS containing 50% glycerol and 20 mM sodium acetate to prevent spontaneous degradation of NE in PBS. PBS control also contained 50% glycerol and 20 mM sodium acetate. In some experiments, cells were pretreated for 1 h with chemical inhibitors or transfected for 48 h with mutant gene constructs as described below, cell-conditioned media were collected, 2.0 mM PMSF was added to neutralize NE, and IL-8 protein levels were determined by ELISA.
Ninety-six-well ELISA plates (MaxiSorb, Nalge Nunc, Rochester, NY) were coated overnight at 22°C with 0.5 μg/ml of human IL-8 antibody (MAB208, R&D Systems, Minneapolis, MN), washed with PBS containing 0.05% Tween 20 (PBS-T), blocked for 1 h with PBS containing 1% BSA and 5% sucrose, and washed with PBS-T. A549 cell-conditioned culture media were centrifuged at 10,000 g for 10 min at 4°C, added in triplicate to ELISA plates, incubated for 2 h at 22°C, and washed with PBS-T. The samples were reacted with 20 ng/ml of biotinylated human IL-8 antibody (BAF208, R&D Systems) for 2 h, washed with PBS-T, treated with 2.5 μg/ml of peroxidase-conjugated streptavidin (KPL, Gaithersburg, MD), incubated for 20 min at 22°C, and washed with PBS-T. Bound streptavidin was detected with tetramethylbenzidine substrate (SureBlue, KPL), the reaction was stopped with 1.0 N HCl, and absorbencies at 450 nm were measured by use of a microplate reader (SPECTRAmax 384 Plus, Molecular Devices, Sunnyvale, CA). Standard curves were generated by using human recombinant IL-8 protein (Sigma).
Duox1 siRNA, dnPKC-δ, and mIκB-α plasmid transfections.
At 24 h postseeding, A549 cells were transfected with 400 ng/ml of the pcDNA3.1(+) empty vector, dnPKC-δ plasmid, pCMV4 empty vector, or a mIκB-α plasmid, or 100 nM of Duox1 siRNA (no. 214819, Ambion) or its negative control RNA (negative control no. 1, Ambion) using Lipofectamine2000 (Invitrogen) according to the manufacturer's instructions. Briefly, RNAs or plasmid DNAs were mixed with 1.0 μl of Lipofectamine2000 and diluted with Opti-MEM-I (Invitrogen) to 100 μl. After a 20-min incubation at room temperature, the RNA- or DNA-lipid complexes in a final volume of 400 μl were added to each well and incubated for 48 h. Knockdown of the Duox1 gene was confirmed by RT-PCR using the following primers: forward, 5′-GCAGGACATCAACCCTGCACTCTC-3′; and reverse: 5′-CTGCCATCTACCACACGGATCT GC-3′. PCR cycles (n = 45) consisted of melting at 95°C for 30 s followed by annealing at 62°C for 1 min and extension at 72°C for 1 min.
Effect of NE on IL-8 mRNA expression.
A549 cells were seeded in DMEM plus 10% FBS in 12-well plates at 2 × 105 cells/well, cultured for 24 h at 37°C, starved for 24 h in serum-free DMEM, and treated for 24 h with various concentrations of NE or for various time periods with PBS or 100 nM of NE. The effect of various concentrations of NE on IL-8 mRNA expression was also examined by using NHBE cells grown at an ALI culture. In experiments to determine IL-8 mRNA stability, the cells were treated for 24 h with PBS or 100 nM of NE and chased for 0, 2, 4, 8, or 24 h in the presence of 5.0 μg/ml of actinomycin D. At the end of each chase period, the cells were washed with PBS, total RNA was collected, and IL-8 transcripts were quantified by real-time RT-PCR.
Total RNA was isolated using RNeasy (Qiagen, Valencia, CA) and 0.1 μg was reverse transcribed by use of the iScript cDNA Synthesis kit (Bio-Rad, Hercules, CA) in a total volume of 20 μl. Real-time PCR was carried out using iQ SYBRgreen supermix (Bio-Rad) according to the manufacturer's instructions. Briefly, 2.0 μl of synthesized cDNA or control plasmid were used as a template for amplification in the iQcycler (Bio-Rad) with 200 nM of human IL-8 or GAPDH primers. The primers were designed using Beacon Designer 3.0 software (Biosoft, Palo Alto, CA). The human IL-8 forward primer was 5′-CGGAAGGAACCATCTCACTGTG-3′ and the reverse primer was 5′-AGAAATCAGGAAGGCTGCCAAG-3′. The GAPDH forward primer was 5′-AGCCTCAAGATCATCAGCAATG-3′ and the reverse primer was 5′-GTTGTCATGGAT GACCTTGGC-3′. PCR cycles (n = 40) consisted of melting at 95°C for 30 s followed by annealing at 52°C for 30 s and extension at 72°C for 30 s. The threshold cycle value was defined as the number of PCR cycles required for the specific fluorescence signal to exceed the detection threshold value set by the software installed in the iCycler. Standard curves for the human IL-8 transcripts were generated by serial dilution of a pcDNA3.1(+) vector containing the human IL-8 cDNA(+) (47) and the pBluescriptSK(−) (36) vector containing the GAPDH cDNA. The amounts of IL-8 and GAPDH mRNAs were calculated from the standard curves, and the levels of IL-8 transcripts were normalized to GAPDH transcripts. All reactions were performed in triplicate.
Preparation of IL-8 promoter-luciferase reporter plasmids.
The human IL-8 promoter between nucleotides −1,471 and +44 (numbered relative to the transcription initiation site) was subcloned from a full-length IL-8 promoter cDNA (46, 47) into the pGL3b luciferase plasmid (Promega, Madison, WI) at the NheI and HindIII sites. A short-length IL-8 promoter between −760 and +44 was subcloned from the full-length promoter construct. Mutation of the NF-κB putative binding site at −82/−69 in the full-length promoter was performed by using the Quickchange II XL site-directed mutagenesis kit (Stratagene, La Jolla, CA) according to manufacturer's instructions. The primers for mutagenesis were forward, 5′-GTGCAAATCGTGGAGGTACCTCTACATAATG-3′; and reverse; 5′-CATTATGTCAGAGGTA CCTCCACGATTGCAC-3′ (mutated residues are underlined). PCR cycles (n = 20) consisted of melting at 95°C for 30 s followed by annealing at 55°C for 1 min and extension at 68°C for 7 min. NF-κB putative binding sites were identified using TFSEARCH software (http://mbs.cbrc.jp/research/db/TFSEARCH.html). The fidelity of all plasmid constructs were confirmed by automated DNA sequencing analysis at the University of Maryland Biopolymer Core Facility.
Transient transfection and luciferase assay.
At 24 h postseeding or 48 h after transfection with Duox1 siRNA or dnPKC-δ or mIκB-α plasmids, A549 cells were transfected with 800 ng/well of the pGL3b empty vector or the IL-8 promoter-pGL3b plasmids as described above. After transfection, the cells were washed, starved for 24 h in serum-free DMEM, and treated for 24 h with PBS or 100 nM of NE. In some experiments, the transfected cells were pretreated for 1 h with PBS or 5.0 μM of rottlerin (70), 10 mM of GSH (57), 100 μM of TAPI-1 (18), 10 μM of AG1478(1), 10 μM of SB202190 (52), 10 μM of U0126 (64), 10 μM SP600125 (12, 37), or 50 μg/ml of SN-50 (43) before NE treatment. Luciferase activity was determined by using the dual luciferase assay system (Promega) according to the manufacturer's instructions and a microplate luminometer (Lmax, Molecular Devices). Luciferase activity driven by the IL-8 promoter was normalized to the internal control by calculating the ratio of firefly luciferase activity to Renilla luciferase activity of each sample and expressed as the percentage of IL-8 promoter activity relative to control samples.
A549 cells were seeded at 8 × 105 cells/well into six-well plates and incubated for 24 h in DMEM plus 10% FBS. The cells were washed, starved for 1 h in serum-free DMEM, and treated for 0, 15, 30, or 60 min with PBS or 100 nM of NE. At each time point, the cells were washed with PBS, immediately extracted with lysis buffer (62.5 mM Tris·HCl, pH 6.8, 2.0% SDS, 5.0% glycerol, 1.0% protease inhibitor cocktail, and 2.0 mM PMSF) at 4°C, and equal amounts of proteins were resolved on 10% SDS-polyacrylamide gels. Resolved proteins were transferred to a polyvinylidene difluoride membrane, and the membrane was blocked for 1 h at 22°C in TBS-T (10 mM Tris·HCl, pH 7.5, 136 mM NaCl, 2.0 mM KCl, and 0.1% Tween 20) containing 5% nonfat dry milk. After being blocked, the membrane was washed with TBS-T and reacted with antibodies to phospho-ERK1/2 (diluted 1:2,000 with TBS-T containing 5% BSA, Cell Signaling), phospho-p38 (1:1,000, Cell Signaling), phospho-JNK (1:1,000, Cell Signaling), IκB-α (1:2,000, Santa Cruz), or β-tubulin (1:10,000, Sigma). After primary antibody reactions, the membrane was washed with TBS-T, incubated for 1 h at 22°C with horseradish peroxidase-conjugated anti-mouse IgG (diluted 1:20,000 with TBS-T containing 2% nonfat dry milk, KPL) or anti-rabbit IgG (1:10,000; KPL) and developed with chemiluminescent substrate (Pierce, Rockford, IL). The bands were identified by comigration of prestained protein size markers (Bio-Rad).
Measurement of cytotoxicity.
Cytotoxicity was assessed by light microscopy for possible changes in cell shape and cell floating as well as the lactate dehydrogenase (LDH) assay. Briefly, confluent A549 cells were treated with 5 μM of rottlerin, 10 mM of GSH, 100 μM of TAPI-1, 10 μM of AG1478, 10 μM of U0126, 10 μM of SB202190, 10 μM of SP600125 (JNK inhibitor), or 50 μg/ml of SN-50 for 1 h, the cultures were washed once and then replenished with fresh medium containing the same concentrations of the above inhibitors, and the incubation continued for 24 h before spent media were collected for LDH assay. In case of the transfection experiment, 45–50% confluent A549 cells were transfected with 100 nM of Duox1 siRNA or 400 ng/ml of dnPKC-δ for 48 h as described above before spent media were collected. The LDH activities in both spent media and cell lysates were measured as previously described (33). None of these inhibitors showed significant cytotoxicity on the basis of the above three criteria (data not shown).
Differences between means ± SE values were assessed using the Student's t-test for unpaired samples and considered significant at P < 0.05.
NE increases IL-8 release.
IL-8 protein levels in spent culture media of A549 cells treated for 24 h with 100 nM of NE were measured by ELISA. As shown in Fig. 1, IL-8 levels in NE-treated culture media were greater compared with media from PBS-treated cells. Furthermore, pretreatment of cells with actinomycin D (AcD) blocked the NE-induced increase in IL-8 release, suggesting that NE stimulated IL-8 gene transcription.
NE increases IL-8 mRNA levels.
To determine whether or not NE-mediated IL-8 release was due to greater IL-8 mRNA expression, transcript levels were measured by real-time RT-PCR in NE- and PBS-treated A549 cells. IL-8 mRNA levels were increased in dose- (Fig. 2A) and time-dependent (Fig. 2B) manners in NE-treated cells compared with PBS-treated cells. A virtually identical result was obtained with NHBE cells (Fig. 2C), which supports the validity of using A549 cells for studying the mechanism of NE-induced stimulation of IL-8 transcription.
NE does not increase IL-8 mRNA stability.
Two possible mechanisms were considered to account for increased levels of IL-8 transcripts in NE-treated cells: increased mRNA stability and/or increased gene transcription. To examine IL-8 mRNA stability, A549 cells were treated with NE for 24 h and chased in the presence of AcD for 0, 2, 4, 8, or 24 h. As shown in Fig. 3, the kinetics of IL-8 mRNA levels were not changed by NE treatment compared with PBS treatment. These results suggested that NE increased IL-8 mRNA and protein levels by increased gene transcription and not mRNA stabilization.
NE increases IL-8 gene transcription.
To directly examine the effect of NE on IL-8 gene transcription, a 1,515-bp region (nucleotides −1,471 to +44) containing the transcription initiation site of the IL-8 promoter was subcloned upstream of the firefly luciferase gene in the pGL3b plasmid. A549 cells transfected with the IL-8-pGL3b reporter plasmid and treated with NE showed significantly higher luciferase activity compared with transfected cells treated with PBS (Fig. 4). From these experiments, we concluded that NE increased IL-8 transcription through promoter activation.
NE increases IL-8 transcription and protein release through PKC-δ, Duox1, ROS, TACE, EGFR, p38, and NF-κB.
Cummings et al. (13) showed that PKC-δ induced IL-8 protein secretion in human bronchial epithelial cells, and Lakshminarayanan et al. (34) demonstrated that ROS increased IL-8 production in A549 cells. Because PKC-δ activates Duox1 (7, 77), a member of the NOX family (31) that in turn generates ROS (22, 31), we hypothesized that NE increases IL-8 transcription and protein release through PKC-δ, Duox1, and ROS. To test this hypothesis, A549 cells were treated with rottlerin (a PKC-δ inhibitor), a dnPKC-δ plasmid, Duox1 siRNA, or GSH (an antioxidant), and NE-induced IL-8 promoter activation and protein levels in culture media were measured. Treatment of cells with all four inhibitors blocked the ability of NE to increase IL-8 promoter activation (Fig. 5A) and IL-8 protein secretion (Fig. 5B) compared with PBS-treated cells. Knockdown of PKC-δ and Duox1 gene expression was greater than 95% on the basis of RT-PCR (data not shown). Interestingly, dnPKC-δ suppressed even the basal IL-8 promoter activity, unlike rottlerin (Fig. 5A). Given the apparent lack of cytotoxicity by the former, it is difficult to explain this result at this point. Nonetheless, the results clearly suggest the involvement of both PKC-δ and Duox1 in NE-induced IL-8 expression.
Shao and Nadel (60) showed that NE treatment of lung cells increased the levels of ROS leading to activation of TACE, generation of transforming growth factor-α, and activation of EGFR. Therefore, we next tested the possibility that NE-stimulated IL-8 transcription and protein release were mediated through TACE and EGFR. A549 cells were treated with TAPI-1 (TACE inhibitor) or AG1478 (EGFR inhibitor) and IL-8 promoter activation and protein levels were measured after NE treatment. Both inhibitors blocked the ability of NE to increase IL-8 transcription (Fig. 6A) and protein levels (Fig. 6B) compared with PBS controls.
IL-8 gene expression is increased after activation of p38 and NF-κB nuclear translocation (9, 23, 26, 29, 49, 59, 71). Because PKC-δ induces IL-8 gene expression via NF-κB activation in lung cells (54), and p38 is activated through EGFR (75), we hypothesized that NE stimulates IL-8 transcription and protein release downstream of EGFR through p38 and NF-κB. Initial experiments showed that all three of the canonical MAPKs (ERK, p38, and JNK) were activated in cells treated with NE (Fig. 7). To test our hypothesis, A549 cells were pretreated with PBS, U0126 (ERK inhibitor), SB202190 (p38 inhibitor), or SP600125 (JNK inhibitor) and NE-induced IL-8 promoter activation and protein levels in culture media were measured. Inhibition of p38, but neither ERK nor JNK, blocked NE-mediated IL-8 promoter activation (Fig. 7B) and protein synthesis (Fig. 7C). Next, to analyze the role of NF-κB, A549 cells were treated with NE and cell lysates were examined by immunoblotting with antibodies against IκB-α, the endogenous inhibitor of NF-κB. As shown in Fig. 8A, NE treatment resulted in rapid degradation of IκB-α. Furthermore, transfection of cells with a mIκB-α plasmid that blocked phosphorylation and degradation of wild-type IκB-α, thereby retaining NF-κB in an inactive complex (78), inhibited the ability of NE to increase IL-8 promoter activity (Fig. 8B) and protein synthesis (Fig. 8C). To link p38 activation with NF-κB, the cells were treated with PBS or SB202190 and IκB-α expression was examined by immunoblot analysis. As shown in Fig. 9A, SB202190 inhibited NE-induced degradation of IκB-α. An identical result was seen with use of AG1478, indicating that IκB-α was also downstream ofEGFR. Finally, treatment of cells with SN-50 to directly inhibit NF-κB activity blocked the ability of NE to increase IL-8 promoter activity (Fig. 10A) and protein synthesis (Fig. 10B). Taken together, the forgoing results seem to indicate that NE increased IL-8 gene transcription and protein release through a PKC-δ → Duox1 → ROS → TACE → EGFR → p38 → NF-κB pathway.
Putative NF-κB binding site at nucleotides −82/−69 in the IL-8 promoter is crucial for NE-induced IL-8 transcription.
Two presumed NF-κB binding sites are located in the human IL-8 promoter (Fig. 11A). By a combination of deletion and site-directed mutagenesis, two IL-8 promoter-luciferase plasmids containing either the upstream (−1,363/−1,354) or downstream (−82/−69) NF-κB site were created. As shown in Fig. 11B, deletion of the distal NF-κB site did not impair the ability of NE to stimulate IL-8 transcription, whereas mutation of the proximal site completely blocked NE-induced luciferase activity. These data suggested that the putative NF-κB binding site at −82/−69 in the IL-8 promoter was necessary for NE to stimulate IL-8 transcription. Furthermore, this site was also crucial for basal IL-8 promoter activity. This proximal NF-κB binding site has previously been shown to be crucial for PMA-induced IL-8 transcription in Jurkat T cells (32).
In this study, we demonstrated that IL-8 protein release from A549 lung epithelial cells was dramatically enhanced by NE treatment. This effect was completely blocked by pretreatment of the cells with AcD, leading us to hypothesize that IL-8 release induced by NE was due to increased protein synthesis subsequent to increased IL-8 transcription. Real-time RT-PCR indicated that NE increased IL-8 mRNA levels but did not alter mRNA stability. Additionally, NE increased IL-8 promoter activity in an IL-8-luciferase reporter assay. Thus we concluded that NE stimulates IL-8 release through IL-8 gene transcription. The results obtained from using a combination of chemical inhibitors and mutant genetic constructs targeting various signal transduction molecules suggested that a PKC-δ → Duox1 → ROS → TACE → EGFR → p38 → NF-κB pathway appeared to be responsible for NE-activated IL-8 transcription. Finally, one of two possible NF-κB binding cis-elements in the IL-8 promoter was shown to play a necessary role in NE-induced IL-8 gene expression.
The pathway through which NE increases IL-8 gene transcription documented in this report presents a unified mechanism previously inferred from analysis of individual components of the pathway. For example, numerous studies have shown that IL-8 transcription is stimulated via p38 activation and NF-κB nuclear translocation (4, 10, 23, 29, 48, 49, 55, 58, 59, 67). Chen et al. (9) showed that NE treatment of A549 cells stimulated IL-8 production through p38. Furthermore, Je et al. (28) demonstrated, and we also confirmed that NF-κB is activated via p38 after degradation of IκB-α. As for the upstream part of this pathway, it is well known that TACE activates EGFR through shedding of transforming growth factor-α, a ligand for EGFR (8, 35, 63). Recently, it was shown that NE activated a PKC-δ → Duox1 → ROS → TACE pathway, leading to increased transcription of the mucin MUC5AC gene in lung epithelial cells (60). In the present study, we have been able to link the “upstream” signaling pathway to the “downstream” pathway to describe a “complete” signaling pathway involved in NE-induced IL-8 regulation. Our results seem to suggest that a common signaling mechanism exists despite the different types of airway cells in mediating the expression of diverse proinflammatory genes in response to NE.
Although the cell surface receptor involved in NE-activated IL-8 and MUC5AC gene expression is unknown, two candidates have been suggested, protease-activated receptor-2 (PAR-2) and Toll-like receptor (TLR)4. Uehara et al. (66) showed that NE increased IL-8 production by fibroblasts through PAR-2. In contrast, Dulon et al. (17) reported that NE inactivated PAR-2 in both A549 and 16HBE airway epithelial cells and prevented subsequent activation by trypsin, an agonist for PAR-2. Walsh et al. (69) and Devaney et al. (15) reported that NE stimulated IL-8 gene and protein expression via a TLR4 → MyD88 → IRAK1 → TRAF6 → NF-κB pathway. On the other hand, Dallot et al. (14) showed that PKC-ζ, and not PKC-δ, was essential for LPS-induced NF-κB nuclear translocation. Because LPS is a TLR4 agonist and PKC-δ is essential for NE-induced IL-8 transcription, it seems unlikely that TLR4 is involved in NE-induced IL-8 transcription in A549 cells. Furthermore, a recent report showed that A549 cells do not express TLR4 (65). Ongoing experiments in our laboratory are directed at identifying the cell surface NE receptor leading to IL-8 expression through PKC-δ activation.
Fischer and colleagues (19–21) reported that NE enhanced the stabilities of transcripts encoding the MUC4 transmembrane and MUC5AC soluble mucins while decreasing GAPDH mRNA stability through a ROS-dependent mechanism. Whereas our results showed that NE activated the IL-8 promoter through oxidative stress, increased IL-8 mRNA stability was not a contributing factor. These results corroborated our previous study demonstrating that NE also increased MUC1 mRNA levels by a process independent of transcript stability (33). The mechanism of MUC4 and MUC5AC mRNA stabilization mediated by NE remains to be clarified but likely depends on mRNA primary and secondary structures. In this regard, Meisner et al. (42) reported that human-antigen R (HuR) interacts with an AU-rich region in IL-2 mRNA, thereby enhancing its stability. They also showed that HuR recognizes the sequence motif NNUUNNUUU in the single-stranded conformation. Although there is no evidence that NE induces binding of HuR to mRNAs, it remains possible that the MUC4 and MUC5AC transcripts, unlike those of MUC1 and IL-8, possess HuR recognition domains.
Another query posed by our results relates to how inflammation at sites of bacterial infection is controlled. The fact that NE increases production of IL-8 suggests an autocrine loop that could potentially drive neutrophil hyperaccumulation. At least two mechanisms appear to counterbalance this possible scenario. First, MUC1 protein expression, which is also increased by NE (28), serves to downregulate inflammation (38). Our prior studies demonstrated that both the production of proinflammatory cytokines in the lungs and transepithelial migration of neutrophils into the airway lumen were reduced under conditions promoting MUC1 expression. Second, IL-8 expression in response to NE is biphasic and attenuated at high protease concentrations (>500 nM, unpublished observations). Although the mechanism downregulating IL-8 expression at high NE levels is unclear, the studies of Malago et al. (40) suggest the involvement of heat shock protein 70 (Hsp70). Oxidative stress and proinflammatory cytokines released from monocytes regulate Hsp/heat shock cognate synthesis (27, 39, 50) and Yoo et al. (72) demonstrated that Hsp70 blocked NF-κB activation. Therefore, it is possible that high concentration of NE decreases IL-8 expression through Hsp70. Future studies are needed to clarify the relationship between NE, Hsp70, and IL-8.
This work was supported by National Institutes of Health Grants HL-47125 and HL-63742 (K. C. Kim) and ES-13483 (E. P. Lillehoj).
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society