Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Airway inflammation is central to the pathogenesis of asthma and other airway diseases, such as chronic obstructive pulmonary disease. The inflammatory response in the airway is a coordinated effort involving recruitment of leukocytes and activation of granulocytes that communicate with the surrounding respiratory epithelium and smooth muscle. Chemical mediators of inflammation have been isolated from many different cell types in the airway, but the role of smooth-muscle cells has been regarded as passive, responding only through its contractile properties. Recent studies now suggest a further role for airway smooth muscle in inflammation by production of inflammatory mediators (reviewed in Reference 1). An understanding of the contributions of smooth muscle to airway inflammation may therefore be a useful tool in examining the pathogenesis of airway diseases.

Many inflammatory cytokines and chemokines transduce signals via activation of mitogen-activated protein (MAP) kinase pathways. In mammals, three prominent groups of MAP kinases have been identified: the extracellular signal-regulated protein kinases (ERKs), the c-Jun NH2-terminal kinases (JNK), and the p38 MAP kinases. These MAP kinases are activated by conserved protein kinase signaling modules, which include a MAP kinase kinase kinase (MKKK) and a dual-specificity MAP kinase kinase (MKK). The MKKK phosphorylates and activates the MKK, which, in turn, activates the MAP kinases by dual phosphorylation on threonine and tyrosine residues within a Thr-Xaa-Tyr motif located in protein kinase subdomain VIII (2). Activation of ERK MAP kinases is involved in numerous cellular processes, including growth, differentiation, survival, and death (2, 3). In contrast, the p38 MAP kinases and JNK are activated by environmental stresses, such as ultraviolet (UV) radiation, osmotic shock, heat shock, protein synthesis inhibitors, and lipopolysaccharide (LPS) (3). The p38 MAP kinases also are activated by treatment of cells with proinflammatory cytokines, including interleukin (IL)-1 and tumor necrosis factor (TNF)- (4). Further, inhibition of the p38 MAP kinase pathway has been shown to exert anti-inflammatory effects through inhibition of IL-1, IL-6, and TNF expression in monocytes (5). The role of ERK MAP kinases in inflammation is less clear, although some reports demonstrate that both ERK and p38 MAP kinases are necessary for optimal induction of TNF (6, 7) and IL-6 (7).

On the basis of these studies, it has been suggested that MAP kinase signaling pathways are a physiologically important mediator of increased cytokine biosynthesis in response to cellular stress (5). We were interested in determining the role of ERK and p38 MAP kinases on gene expression and cytokine biosynthesis in smooth-muscle cells exposed to inflammatory mediators. We chose a complex inflammatory stimulus consisting of IL-1, TNF-, and interferon (IFN)- to more closely approximate the milieu found in vivo. This cytokine cocktail has previously been shown to induce cyclooxygenase (COX)-2 and monocyte chemotactic proteins (MCPs) in airway myocytes (8, 9). The program of gene expression after stimulation with a cytokine cocktail containing IL-1, TNF-, and IFN- in human bronchial smooth-muscle cells (BSMC) was explored with a complementary DNA (cDNA) array representing 588 human genes. This cDNA array consisted of multiple functional gene classes including transcription factors, oncogenes, cell adhesion molecules, signal transduction mediators, cytokines, and growth factors that were induced by a 20-h exposure to the cytokine cocktail. The genes for IL-1, IL-6, and IL-8 were singled out for further analysis by reverse transcriptase/polymerase chain reaction (RT-PCR), enzyme-linked immunosorbent assay (ELISA), and Western blotting techniques. The results from these studies are among the first to demonstrate that, in human BSMC, both the p38 and ERK MAP kinase pathways regulate expression and secretion of IL-1, IL-6, and IL-8 in response to proinflammatory stimuli. Although it is difficult to assess the contribution of airway myocytes to overall cytokine production levels in the airway, the generation of chemokines and cytokines in response to inflammatory mediators in these studies provides strong evidence that airway smooth-muscle cells have the ability to potentiate the inflammatory response.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Materials

Normal human BSMC, smooth-muscle cell growth medium (SmGm), and all other tissue culture reagents were obtained from Clonetics (San Diego, CA). IFN-, IL-1, and TNF- were purchased from Sigma (St. Louis, MO). SB203580 and PD98059 were purchased from Calbiochem (La Jolla, CA). TRIzol Reagent and Superscript II were purchased from Life Technologies (Rockville, MD). The ATLAS cDNA expression array was purchased from Clontech (Palo Alto, CA). [³²P]adenosine triphosphate (ATP) was purchased from ICN Biomedicals, Inc. (Costa Mesa, CA). PCR reagents and salmon-sperm DNA were purchased from Invitrogen (San Diego, CA). Thermus aquaticus polymerase (Taq) and RNAse H were purchased from Promega Corp. (Madison, WI). Phosphospecific p38 and ERK MAP kinase antibodies were purchased from New England Biolabs (Beverly, MA). The p38 MAP kinase antibody and recombinant activating transcription factor (ATF)-2 were purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The ERK MAP kinase antibody was purchased from Upstate Biotechnologies Inc. (Lake Placid, NY). Quantikine ELISAs for IL-8 and IL-6 and the IL-1 antibody were purchased from R&D Systems (Minneapolis, MN).

Cell Culture

BSMC from passages four through seven were used in these studies and grown in a humidified 5% CO₂ atmosphere at 37°C in SmGm supplemented with 5% fetal bovine serum (FBS), 0.5 ng/ml epidermal growth factor, 5 µg/ml insulin, 2 ng/ml fibroblast growth factor, 50 µg/ml gentamicin, and 50 ng/ml amphotericin B. These cells were screened for human pathogens and for contaminating cell types, such as fibroblasts. Moreover, we determined by Western analysis that these smooth-muscle cells express a panel of contractile proteins (h-caldesmon, calponin, -actin, and tropomyosin) that correlate with the contractile phenotype (W. T. Gerthoffer, unpublished observations). These cultures were grown to confluence and growth-arrested for 24 h in SmGm supplemented with 0.1% FBS, growth factors, gentamicin, and amphotericin B. Cells were then treated with a 0.1% dimethyl sulfoxide (DMSO) vehicle, 25 µM SB203580, or 25 µM PD98059 for 15 min before and during a 20-h treatment with a cytokine cocktail containing 10 ng/ml IL-1, TNF-, and IFN-. In selected experiments, the effects of the individual cytokines were examined at doses of 2, 5, or 10 ng/ml.

RNA Isolation

Total RNA was extracted from human BSMC with TRIzol reagent at 1 ml/10 cm² according to the manufacturer's instructions. After TRIzol extraction, RNA samples were dissolved in nuclease-free water and treated with DNase I (5 U) for 60 min at 37°C. The reaction was stopped by addition of 25 mM ethylenediaminetetraacetic acid and incubation at 65°C for 15 min. RNA samples were then purified by extraction in phenol/chloroform ([pH 5.2] phenol:chloroform:isoamyl alcohol [25:24:1]) followed by an additional chloroform extraction and ethanol precipitation.

cDNA Expression Array

The ATLAS cDNA expression array was used to examine differential gene expression in BSMC stimulated with 10 ng/ml IL-1, TNF-, and IFN- in the presence or absence of 25 µM SB203580. The manufacturer's recommended protocol was followed, and provided reagents were used. Briefly, 4 µg of total RNA was converted into [³²P]-labeled first-strand cDNA using Moloney murine leukemia virus (MMLV) RT and a primer mix provided by the manufacturer. Unincorporated [³²P]-labeled nucleotides were removed by CHROMA SPIN-200 column chromatography. cDNA fractions of highest activity were pooled and hybridized to the ATLAS membranes. The membrane was prehybridized at 68°C with ExpressHyb supplemented with 100 µg/ml of salmon-sperm DNA for 30 min. The probe (total activity 500,000 counts per minute) was added and hybridized overnight. Membranes were washed in 2× saline sodium citrate (SSC) (300 mM NaCl and 30 mM sodium citrate)/1% sodium dodecyl sulfate (SDS) at 68°C followed by washes in 0.1× SSC (15 mM NaCl and 1.5 mM sodium citrate)/0.5% SDS at 68°C. Membranes were exposed to a PhosphorImager screen and analyzed by densitometry. Densitometric data were normalized to hybridization signal from the housekeeping genes visible on each array membrane. The normalized gene expression represents the mean hybridization signal after normalization to three housekeeping genes: ubiquitin, glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and a 23-kD highly basic protein (HBP).

cDNA Synthesis and Relative RT-PCR

First-strand cDNA synthesis was performed at 42°C from 2 µg RNA using 250 ng random hexamers; 50 mM Tris-HCl (pH 8.3); 75 mM KCl, 3 mM MgCl₂; 10 mM dithiothreitol (DTT); 0.125 mM each of deoxyadenosine triphosphate (dATP), deoxythymidine triphosphate (dTTP), deoxyguanidine triphosphate (dGTP), and deoxycytidine triphosphate (dCTP); and 1 U SuperScript II RT. A total of 20 U of RNAse H was then added to remove RNA complementary to the cDNA.

Cytokine genes were amplified using the PCR in a thermal cycler (GeneAmp PCR System 2400; Perkin-Elmer). The reaction mixture contained 60 mM Tris-HCl (pH 8.5); 15 mM (NH₄)SO₄; 1.5 mM MgCl₂; 0.25 mM each of dATP, dCTP, dGTP, and dTTP; 10% DMSO; 20 µM of each primer; 5 µl template cDNA; and 2.5 units Taq. Sequences for IL-8, IL-6, and IL-1 oligonucleotide primers were purchased from Clontech. Oligonucleotides were synthesized by Bio-Synthesis (Lewisville, TX). Amplification of IL-8 took place at 94°C for 30 s and 68°C for 2 min. Amplification of IL-6 and IL-1 took place at 94°C for 30 s and 68°C for 7 min. All PCR amplification took place within the linear range of amplification as discussed later.

Semiquantitative relative PCR was performed using QuantumRNA 18S internal standards according to the manufacturer's protocol (Ambion, Inc., Austin, TX). Total RNA isolation, cDNA synthesis, and PCR amplification took place as described previously with the following exceptions: The linear range of PCR amplification (the period in which amplification efficiency remains constant over a number of cycles) for each gene of interest was determined and quantified by ethidium bromide staining of agarose gels followed by densitometry. Multiplex PCR took place within the linear range of amplification for each gene using an optimized ratio of 18S primers as an endogenous standard along with 18S competimers. These competimers allow modulation of 18S amplification without affecting the performance of the gene-specific PCR targets in the reaction. PCR products were quantified by ethidium bromide staining and analyzed by densitometry. Results were normalized to the amount of 18S RNA present in each sample from four separate RNA preparations.

Immunoblot Analysis

Human BSMC were grown to confluence on six-well plates. After 24 h in 0.1% FBS medium, cells were stimulated with 10 ng/ml IL-1, TNF-, and IFN- for the indicated time points at 37°C in a CO₂ incubator. Where indicated, cells were also pretreated with 25 µM SB203580 or 25 µM PD98059 for 15 min, and the inhibitors remained in the media throughout the cytokine stimulation. The stimulation was stopped by a wash in phosphate-buffered saline, and the cells were immediately lysed with extraction buffer containing 20 mM Tris (pH 7.5), 5 mM [ethylene-bis(oxyethylenenitrilo)]tetraacetic acid, 150 mM NaCl, 1% Nonidet P-40, 0.1 mM Na₃VO₄, 1 mM NaF, 10 mM sodium -glycerophosphate, 0.1 mM phenylmethylsulfonyl fluoride, 1 µg/ml leupeptin, 1 µg/ml Pepstatin A, 10 µg/ml trypsin inhibitor, and 10 µg/ml aprotinin. Cellular extracts were clarified by centrifugation at 10,000 × g for 10 min at 4°C, and the supernatants were used to assay intracellular IL-1 levels, p38, and ERK MAP kinase phosphorylation. Protein concentrations were determined by the bicinchoninic acid method using bovine serum albumin as the standard. To assess IL-1 levels as well as p38 and ERK MAP kinase phosphorylation, total protein extracts (20 µg per lane) were separated by 12% SDS-polyacrylamide gel electrophoresis (SDS-PAGE). Proteins were transferred to nitrocellulose paper in 25 mM Tris, 192 mM glycine, 10% methanol using a Mini-Genie Blotter (Idea Scientific, Minneapolis, MN; 24 V for 2 h, under continuous cooling at 4°C). Blots were blocked with 0.5% gelatin for 1 h. p38 MAP kinase blots were probed with a dual anti-phosphotyrosine-threonine-p38 MAP kinase or p38 MAP kinase primary antibodies (1:500) and goat antirabbit immunoglobulin (Ig) G alkaline phosphatase secondary antibody (1:15,000). ERK MAP kinase blots were probed with a dual anti-phosphotyrosine-threonine-p42/p44 or a p44 primary antibody that cross-reacts with p42 (1:500) and goat antirabbit IgG alkaline phosphatase secondary antibody (1:15,000). Intracellular IL-1 blots were probed with an IL-1 primary antibody (1:500) and antigoat IgG alkaline phosphatase secondary antibody (1:10,000). Images of immunoblots scanned with a UMAX Powerlook flatbed scanner were analyzed by densitometry. Densitometric data were normalized to the unstimulated control cells.

MAP Kinase Activity Assays

A 40-µl kinase reaction contained phosphorylation buffer (25 mM 3- [N-Morpholino]propanesulfonic acid, 25 mM -glycerophosphate, 15 mM MgCl₂, 1 mM ethyleneglycol-bis-[-aminoethyl ether]-N,N'-tetraacetic acid, 0.1 mM NaF, 1 mM Na₃VO₄, and 4 mM DTT; pH 7.2), cellular extracts (20 µg total protein), and 1 µg recombinant ATF-2. The reaction was started by addition of 10 µCi of 250 µM [-³²P]- ATP. After incubation at 30°C for 30 min, the reactions were terminated by diluting them 1:4 with concentrated SDS sample buffer (0.24 M Tris [pH 6.8], 8% SDS, 40% glycerol, and 4 mM DTT). Proteins were resolved by 12% acrylamide SDS-PAGE, and phosphorylated ATF-2 was visualized and quantitated by densitometry.

Cytokine Assays

IL-6 and IL-8 were measured by ELISA using procedures recommended by the manufacturer. The minimum detectable levels of IL-6 and IL-8 with the assay were 0.7 and 10 pg/ml, respectively.

Densitometry and Statistical Analysis

Densitometry was performed using a Bio-Rad Model 525 Molecular Imager and the Volume Analyze feature of Molecular Analyst software (Bio-Rad, Hercules, CA). Statistical analysis was performed by one-way analysis of variance followed by post hoc testing with the Student-Newman-Keuls method or a paired t test using SigmaStat software (Jandel Scientific, San Rafael, CA).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Cytokine-Stimulated Airway Smooth-Muscle Gene Expression

To determine the role of p38 MAP kinase signaling in response to inflammatory mediators, the ATLAS cDNA expression array was used to follow gene expression in human airway myocytes stimulated with a cytokine cocktail. This cDNA array contains 588 human cDNAs, nine housekeeping control cDNAs, and negative controls immobilized in duplicate on a nylon membrane. After a 20-h stimulation with the cytokine cocktail containing 10 ng/ml IL-1, TNF-, and IFN-, hybridization signal was detected for 52 genes, including three housekeeping genes, from cDNA synthesized from total BSMC RNA. No hybridization signal was detected for any negative controls present on the array. A duplicate ATLAS membrane was hybridized with cDNA from BSMC treated for 20 h with the cytokine cocktail containing 10 ng/ml IL-1, TNF-, and IFN- in the presence of 25 µM SB203580, a p38 MAP kinase inhibitor. The hybridization signal obtained for the three visible housekeeping genes, ubquitin, GAPDH, and 23 kD HBP found on both arrays, along with the densitometric analysis of the signal, is shown in Figure 1A. Even though care was taken to ensure that the specific activity of the labeled cDNA samples did not vary when added to the membranes, it was clear that the hybridization signal in the presence of SB203580 was greater. To account for variations in hybridization signals between the two array membranes, densitometric analysis of the hybridization signal was normalized to the signal obtained from each of the three visible housekeeping genes found on each array, and a mean value for gene expression was reported from this analysis. Differential gene expression in the two treatment groups after this normalization is shown in Figure 1B. The line intersecting the graph represents no change in normalized gene expression of the cytokine-stimulated cells in the presence or absence of SB203580. An increase in the expression of 17 genes was seen in the presence of SB203580, as represented by the data points above the line. A decrease in the expression of 32 genes was seen in the presence of SB203580, as shown in the data points falling below the line. The cytokine and chemokine genes that were expressed after stimulation with the cytokine cocktail are shown. A complete list of the genes available on the array can be obtained from the manufacturer. After the analysis of the data obtained from the cDNA array, IL-1, IL-6, and IL-8 were chosen for further study. Figure 1C depicts the mean gene expression for these cytokines in human BSMC after normalization to each of the visible housekeeping genes, as described previously. In the presence of SB203580, expression of IL-8 and IL-6 was decreased, whereas SB203580 did not appear to have an effect on IL-1 expression.

View larger version (23K): [in this window] [in a new window]   Figure 1.   Human BSMC were treated for 20 h with the cytokine cocktail consisting of 10 ng/ml IL-1, TNF-, and IFN- (INF) in the presence of the DMSO vehicle ± 25 µM SB203580. Total RNA was isolated from treated BSMC and labeled cDNA prepared as described. Normalization of gene expression is described in MATERIALS AND METHODS. (A) Hybridization signal spotted in duplicate and densitometric analysis of housekeeping expression. (B) Normalized airway myocyte gene expression. Selected cytokine and chemokine genes are labeled as follows: CSF-1, macrophage-specific colony-stimulating factor-1; IL-1; IL-5R, IL-5 receptor; IL-6; IL-8; MCP-1; MIG, IFN--induced monokine; MIP2, macrophage inflammatory protein 2. (C) Hybridization signal spotted in duplicate and mean normalized gene expression of IL-1, IL-6, and IL-8 expression ± 25 µM SB203580 in the presence of 10 ng/ml cytokine cocktail.

p38 and ERK MAP Kinase Activation and Inhibition in Airway Smooth Muscle

p38 and ERK MAP kinases are activated by upstream kinases MKK3/6 for p38 and MKK1/2 for ERK, by dual phosphorylation of threonine and tyrosine in the regulatory TXY motif (2). Phosphorylation of this motif has been used as an index of MAP kinase activation and can be assayed with an anti-p38 MAP kinase or anti-p44/42 (ERK1/2) phosphotyrosine/threonine-specific antibody recognizing the phosphorylated TXY motif (10). To test the notion that p38 and ERK MAP kinases are activated by the cytokine cocktail stimulation described in Figure 1, BSMC were incubated in 0.1% FBS medium for 24 h and then treated for indicated periods of time with 10 ng/ ml IL-1, TNF-, and IFN-. Time-course experiments, presented in Figures 2A and 2C, demonstrate a transient increase in tyrosine and threonine phosphorylation of p38 and ERK MAP kinases. Stimulation with cytokine cocktail induced an increase in p38 MAP kinase activity after 5 min (Figure 2B). The increase in ERK1 and ERK2 phosphorylation above basal level was slight (Figure 2C). There was a high level of ERK phosphorylation in the untreated cells, which we attribute most likely to other growth factors (see MATERIALS AND METHODS) contained in the culture medium that are known to activate the ERK pathway (2). To inhibit the p38 and ERK MAP kinase pathways, cells were pretreated for 15 min with chemical inhibitors SB203580 and PD98059, respectively (Figures 2B and 2D). These inhibitors are highly selective, and we have determined that 25 µM SB203580 does not interfere with ERK phosphorylation in vivo and ERK kinase activity in vitro, whereas 25 µM PD98059 does not inhibit p38 MAP kinase phosphorylation and activity in vivo (data not shown). Inhibition of ERK MAP kinase phosphorylation by PD98059, which inhibits the upstream MEK1/2 kinases (11), is demonstrated in Figure 2D. Cellular extracts were also used to phosphorylate recombinant ATF-2 in an in vitro assay for p38 activity. The kinase reaction was stopped by addition of concentrated SDS-PAGE sample buffer, and the phosphoproteins were resolved by SDS-PAGE. Radioactive phosphorous incorporation was measured by imaging dried gels with a Bio-Rad Molecular Imager. Results presented in Figure 2B show that p38 MAP kinase activity was blocked in cells that were pretreated with 25 µM SB203580.

View larger version (40K): [in this window] [in a new window]   Figure 2.   (A and C) Cells were stimulated with the 10 ng/ml cytokine cocktail for 0, 1, 5, 15, 30, and 60 min. Cells were lysed and proteins extracted in SDS-PAGE sample buffer (see MATERIALS AND METHODS). Total proteins were resolved by SDS-PAGE and tyrosine/threonine phosphorylation (upper panels) or nonselective immunoreactivity (lower panels) of p38 and ERK MAP kinases were detected by Western blotting with anti-phosphotyrosine/threonine-p38 (p-p38) or ERK MAP kinase (p-ERK) antibody and alkaline phosphatase-conjugated secondary antibody. (B) p38 MAP kinase activity was measured using ATF2 as a substrate. Cells were stimulated with the 10 ng/ml cytokine cocktail for 15 min. Cells were also pretreated with 25 µM SB203580 for 15 min. Cellular extracts were used to phosphorylate 1 µg of ATF-2 in vitro for 30 min. Phosphoproteins were resolved by SDS-PAGE, and radioactive phosphorous incorporation was measured by imaging gels with a Bio-Rad Molecular Imager. (D) Cells were stimulated with the 10 ng/ml cytokine cocktail for 30 min and were pretreated with 25 µM PD98059 for 15 min.

Effects of IL-1, TNF-, and IFN- on 33-kD Intracellular IL-1 Synthesis

The effects of individual components of the cytokine cocktail were examined by stimulating BSMC with 2, 5, or 10 ng/ml IL-1, TNF-, or IFN- for 20 h. Whole-cell lysates were prepared as described and intracellular IL-1 immunoreactivity at 33 kD was measured by immunoblotting. Densitometric anaylsis of this immunoreactivity is shown in Figure 3A. The effect of the cytokine cocktail (10 ng/ml IL-1, TNF-, and IFN- for 20 h) is shown for comparison.

View larger version (27K): [in this window] [in a new window]   Figure 3.   (A) Densitometric analysis of intracellular IL-1 immunoreactivity at 33 kD after a 20-h treatment with 2, 5, or 10 ng/ml IL-1, TNF-, or IFN- (INF), or the 10 ng/ml cytokine cocktail. (B) Western analysis depicting immunoreactivity for intracellular IL-1 after treatment of BSMC with the 10 ng/ml cytokine cocktail at the indicated times.

Effect of MAP Kinase Inhibition on Intracellular IL-1 Expression and Synthesis

Intracellular IL-1 immunoreactivity at 33 kD in BSMC stimulated with the cytokine cocktail was measured over a period of 36 h (Figure 3B). Immunoreactivity was apparent following 4 h of stimulation and continued to be visible for up to 20 h of cytokine cocktail stimulation. Figure 4A depicts the effects of 20 h of cytokine stimulation and MAP kinase inhibition on IL-1 expression as measured by relative RT-PCR. IL-1 expression increased 2-fold with cytokine stimulation compared with unstimulated cultures receiving the DMSO vehicle. The p38 MAP kinase inhibitor SB203580 did not have an effect on IL-1 expression, whereas the ERK MAP kinase inhibitor PD98059 decreased IL-1 expression by 35%. The effect of MAP kinase inhibition on intracellular IL-1 immunoreactivity at 33 kD is shown in Figure 4B. Upon 20 h of cytokine cocktail stimulation, intracellular IL-1 immunoreactivity increased 3.5-fold from DMSO-treated cultures. Treatment with SB203580 did not affect intracellular IL-1, whereas treatment with PD98059 decreased intracellular IL-1 immunoreactivity by 44%.

View larger version (23K): [in this window] [in a new window]   Figure 4.   (A) Relative multiplex RT-PCR from BSMC treated for 20 h with the 10 ng/ml cytokine cocktail in the presence of the DMSO vehicle ± 25 µM SB203580 or 25 µM PD98059. In the top panel, the 18S rRNA PCR product is shown at 324 base pairs (bp), and the IL-1 PCR product is shown at the predicted size of 291 bp from a representative experiment. In the bottom panel, normalized IL-1 expression is shown, n = 4 ± standard error of the mean (SEM). (B) Western analysis demonstrating intracellular IL-1 immunoreactivity from cells treated as described in A, n = 3-5 ± SEM. *Significant difference from DMSO-treated group;  significant difference from cytokine cocktail-treated group, P < 0. 05.

Effect of MAP Kinase Inhibition on IL-6 Expression and Secretion

IL-6 secretion in human BSMC stimulated with the cytokine cocktail was measured by ELISA over a period of 36 h (Figure 5A). Over this time period, IL-6 secretion rose from undetectable levels to 2.5 ng/ml after 8 h of stimulation. IL-6 concentrations continued to increase to 91 ng/ml at the last time point measured. The effects of individual components of the cytokine cocktail on IL-6 secretion are shown in Figure 5C. IL-6 secretion increased from undetectable levels to 12.7 ng/ml upon stimulation of BSMC with 10 ng/ml IL-1. IL-6 secretion was not detected after stimulation with 10 ng/ml TNF- or IFN-. Figure 5B depicts the effect of 20 h of cytokine stimulation and MAP kinase inhibition on IL-6 as measured by relative RT-PCR. IL-6 messenger RNA (mRNA) was significantly increased 3-fold with cytokine stimulation compared with unstimulated cultures receiving the DMSO vehicle. The p38 MAP kinase inhibitor SB203580 decreased IL-6 expression by 29% from cytokine-stimulated cultures, whereas the ERK MAP kinase inhibitor PD98059 decreased IL-6 expression by 41%. The effect of MAP kinase inhibition on IL-6 secretion after stimulation with the cytokine cocktail containing 10 ng/ml IL-1, TNF-, and IFN- is shown in Figure 5C. Upon 20 h of stimulation with the cytokine cocktail, IL-6 secretion increased from undetectable levels to 114.5 ng/ml. Treatment with SB203580 decreased IL-6 secretion by 55% to 51.2 ng/ml and treatment with PD98059 decreased IL-6 secretion by 32% to 72.2 ng/ml. Treatment with both SB203580 and PD98059 decreased IL-6 secretion by 86% to 14.8 ng/ml.

View larger version (14K): [in this window] [in a new window]   Figure 5.   (A) Human BSMC were treated at the indicated times with the 10 ng/ml cytokine cocktail, media were removed, and IL-6 secretion was assayed by ELISA, n = 2 ± SEM. (B) Relative multiplex RT-PCR from BSMC treated for 20 h with the 10 ng/ml cytokine cocktail in the presence of the DMSO vehicle ± 25 µM SB203580 or 25 µM PD98059. In the top panel, the 18S rRNA PCR product is shown at 324 bp, and the IL-6 PCR product is shown at the predicted size of 425 bp from a representative experiment. In the bottom panel, normalized IL-6 expression is shown, n = 4 ± SEM. (C) IL-6 secretion measured by ELISA from cells treated with 10 ng/ml IL-1, TNF-, or IFN- (INF), or the 10 ng/ml cytokine cocktail, n = 2-5 ± SEM. ND indicates IL-6 secretion was not detected. *Significant difference from DMSO-treated group; significant difference from the cytokine cocktail-treated group, P < 0.05.

Effect of MAP Kinase Inhibition on IL-8 Expression and Secretion

IL-8 secretion in human BSMC stimulated with the cytokine cocktail was measured by ELISA over a period of 36 h (Figure 6A). Over this time period, IL-8 secretion rose from undetectable levels to 4.3 ng/ml after 4 h of stimulation. IL-8 concentrations continued to increase to 105.8 ng/ml at the last time point measured. The effects of individual components of the cytokine cocktail on IL-8 secretion are shown in Figure 6C. A total of 10 ng/ml IL-1 and TNF- increased IL-8 secretion from undetectable levels to 115.5 and 53.6 ng/ml, respectively, after 20 h of stimulation. IL-8 secretion was not detected after treatment with 10 ng/ml IFN- for 20 h. Further experiments were performed after 20 h of stimulation with the cytokine cocktail. Figure 6B depicts the effect of cytokine stimulation and MAP kinase inhibition on IL-8 expression at this time point as measured by relative RT-PCR. IL-8 mRNA is significantly increased 2.2-fold compared with unstimulated cultures receiving the DMSO vehicle. The p38 MAP kinase inhibitor SB203580 decreased IL-8 expression by 22% from cytokine-stimulated cultures, whereas the ERK MAP kinase inhibitor PD98059 decreased IL-8 expression by 21%. The effect of MAP kinase inhibition on IL-8 secretion after stimulation with the cytokine cocktail is shown in Figure 6C. Upon 20 h of cytokine stimulation, IL-8 secretion increased from undetectable levels to 131.5 ng/ml. Treatment with SB203580 decreased IL-8 secretion by 51% to 64.4 ng/ml, and treatment with PD98059 decreased IL-8 secretion by 22% to 91.5 ng/ml. Treatment with both SB203580 and PD98059 had an additive effect, decreasing IL-8 secretion by 77% to 26.7 ng/ml.

View larger version (15K): [in this window] [in a new window]   Figure 6.   (A) Human BSMC were treated at the indicated times with the 10 ng/ml cytokine cocktail, media were removed, and IL-8 secretion was assayed by ELISA, n = 2 ± SEM. (B) Relative multiplex RT-PCR from BSMC treated for 20 h with the 10 ng/ml cytokine cocktail in the presence of the DMSO vehicle ± 25 µM SB203580 or 25 µM PD98059. In the top panel, the 18S rRNA PCR product is shown at 324 bp, and the IL-8 PCR product is shown at the predicted size of 188 bp from a representative experiment. In the bottom panel, normalized IL-8 expression is shown, n = 4 ± SEM. (C) IL-8 secretion measured by ELISA from cells treated with 10 ng/ml IL-1, TNF-, or IFN- (INF), or the 10 ng/ml cytokine cocktail, n = 2-5 ± SEM. ND indicates IL-8 secretion was not detected. *Significant difference from DMSO-treated group; significant difference from the cytokine cocktail-treated group, P < 0. 05.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Hyperreactivity of the airway is a characteristic feature of asthma, occurring as an exaggerated response to increases in bronchomotor tone due to hyperplasia and hypertrophy of airway smooth muscle. This leads to a narrowing of the airway opening, increased resistance to airflow, and more work required for breathing (12). These changes in airway reactivity are promoted by increases in epithelial secretions and airway smooth-muscle contraction, along with the presence of chemical mediators of inflammation (reviewed in Reference 13). Cultured human airway smooth-muscle cells have been shown to produce cytokines and chemokines including IL-6; IL-8; IL-1; TNF-; IFN-; regulated on activation, normal T cells expressed and secreted; and granulocyte macrophage colony-stimulating factor in response to inflammatory stimuli (1, 14, 15). An understanding of molecular mechanisms and signaling pathways that modulate smooth-muscle cell gene expression in response to proinflammatory mediators is mostly unknown.

We followed cytokine-induced gene expression in BSMC using a cDNA array. This cDNA array consisted of multiple functional gene classes including transcription factors, oncogenes, cell adhesion molecules, signal transduction mediators, cytokines, and growth factors. The cultures were exposed to a complex inflammatory stimulus consisting of IL-1, TNF-, and IFN- that has previously been shown to induce COX-2 and MCPs in airway myocytes (8, 9). In addition, cytokine-induced gene expression on the array was compared in the presence and absence of a specific inhibitor of the p38 MAP kinase pathway, SB203580. Genes from every functional class on the array were affected by the cytokine stimulation in the presence of SB203580, and three interleukin genesIL-1, IL-6, and IL-8that have previously been shown to be important in inflammation were identified for further evaluation.

IL-1 is a potent inflammatory cytokine that exerts pleiotropic effects on a variety of tissues and plays a central role in inflammatory responses (16). After obtaining results from the cDNA array demonstrating that SB203580 did not have an effect on IL-1 expression, we measured intracellular IL-1 by immunoblotting and IL-1 expression by relative RT-PCR. It was necessary to measure intracellular IL-1 immunblotting rather than measuring secreted IL-1 by ELISA because of the high amount of IL-1 present in the media from the cytokine cocktail stimulus. The antibody used in these experiments detected the 33-kD form of IL-1 that is found intracellularly, thus allowing us to discern intracellular IL-1 from the 17-kD secreted active form of IL-1. Individual components of the cytokine cocktail did increase intracellular IL-1 immunoreactivity in a dose-dependent manner, and the effect of the cytokine cocktail on IL-1 was cumulative. Consequently, the cytokine cocktail was used in further experiments to more closely approximate the proinflammatory milieu found in vivo. Using these methods, intracellular IL-1 was upregulated in BSMC in response to the cytokine cocktail and downregulated when cells were pretreated with an inhibitor of the ERK MAP kinase. However, the p38 MAP kinase pathway does not appear to regulate intracellular IL-1 in cytokine-stimulated smooth-muscle cells using SB203580. This disagrees with previous reports of p38 MAP kinase-mediated IL-1 expression in monocytes stimulated with TNF- and IL-1 (4). These results may reflect differential regulation of IL-1 in smooth-muscle cells or may be a result of the cytokine stimulus used in these experiments. It is also important to note that p38 MAP kinase may regulate the levels of IL-1 expression by affecting mRNA stability.

We also investigated the signaling pathways that regulate the production of IL-6 and IL-8 in smooth-muscle cells in response to inflammatory mediators. IL-6 is a multifunctional cytokine that promotes B-cell growth and differentiation and stimulates acute-phase protein synthesis (17). In contrast, IL-8 is a chemokine that attracts and stimulates leukocytes at sites of inflammation (18). Using relative RT-PCR and measuring protein secretion by ELISA, we demonstrated a dramatic increase in expression and secretion of both IL-6 and IL-8 after a 20-h treatment with the cytokine cocktail containing 10 ng/ml IL-1, TNF-, and IFN-. Expression and secretion of IL-6 and IL-8 were blocked by SB203580 and, to a lesser extent, by PD98059, suggesting a regulatory role for the p38 and ERK MAP kinase pathways. These results agree with other reports of MAP kinase-mediated cytokine expression in nonmuscle cells (19). The effects of individual components of the cytokine cocktail demonstrate that IL-1, but not TNF- or IFN-, results in a small increase in IL-6 secretion. Stimulation with the cytokine cocktail containing IL-1, TNF-, and IFN-, however, greatly potentiates IL-6 secretion in BSMC. In contrast, the effects of the individual components of the cytokine cocktail on IL-8 secretion demonstrate that both IL-1 and TNF-, but not IFN-, increase IL-8 secretion. Stimulation of IL-8 secretion with the cytokine cocktail is similar to the effects seen with IL-1 alone but greater than those seen with TNF-. This result agrees with previously published data reporting that IL-1 and TNF-, but not IFN-, affect IL-8 release in human airway smooth-muscle cells (18). Although it is clear that IL-1 and TNF- stimulate IL-8 expression and release, the contribution of IFN- to the regulation of IL-8 by MAP kinases has not been reported in airway smooth muscle. Further experiments are needed to assess the contribution of IFN- to the regulation of IL-8 secretion or to other inflammatory events that may affect IL-8 function.

The use of the cDNA array to assess gene expression demonstrates the complicated processes involved in analyzing the effects of gene induction on translation of a protein that may be useful to the cell. Any change in the amount of mRNA that is seen on the array is dependent on the rate of RNA transcription, processing, and stability. MAP kinase pathways can regulate gene expression at the levels of transcription, mRNA stability, and translation (4, 20, 22). The p38 MAP kinase pathway has been shown to induce mRNA stabilization via AU-rich sequences that have been identified as crucial cis elements mediating RNA stability (22). The cytokines IL-6 and IL-8 bear AU-rich sequences in their 3'-untranslated regions. Translational control by p38 and ERK pathways may be explained by activation of a common substrate, MAPK signal integrating kinase (MNK1) (23, 24). It has recently been shown that MNK1 interacts with eukaryotic initiation factor 4E and affects binding to the mRNA cap structure (25, 26). Although the results reported here clearly demonstrate regulation of cytokine expression and secretion by ERK and p38 MAP kinase, further studies are required in order to delineate the mechanisms regulating these changes in cytokine production.

Recent studies have led to the identification of substrates for p38 MAP kinase that may be physiologically relevant. These include the transcription factors ATF-2 (10), myocyte enhancing factor (MEF)2C (27), erythroleukemia virus 26-like (ELK-1) (28), and serum response factor accessory protein (SAP)-1 (28), which are phosphorylated and activated by p38 MAP kinase. The p38 MAP kinase also has been reported to phosphorylate and activate several other protein kinases, including MAPKAPK2, MAPKAPK3, MNK1, MNK2, mitogen and stress-activated kinase (MSK1), and p38-regulated/activated protein kinase (PRAK) (23, 24, 29). The use of pyridinyl imidazole derivatives such as SB203580 have provided an understanding of the p38 MAP kinase signaling pathway (4). These drugs have also been shown to exert anti-inflammatory effects in monocytes because they inhibit the expression of cytokines, including IL-1, IL-6, and TNF (4, 5).

The substrates for ERK MAP kinase include cytoskeletal targets, cytoplasmic enzymes, growth factor receptors, and multiple transcription factors including ELK-1, C/EBP (nuclear factor [NF]-IL6), c-Myc, and components of the activator protein-1 complex (reviewed in Reference 33). Although many of these substrates overlap with those for p38 MAP kinase as well as other kinases, it appears that coordinated activation of multiple kinase cascades may be necessary for optimal induction of transcription and synthesis of cytokines (4, 6, 7, 21, 34). For example, stimulation of IL-6 and IL-8 transcription appears to require simultaneous activation of C/EBP (NF-IL6), and NF-B by multiple kinase cascades (21, 35). LPS activation of TNF synthesis also appears to require ERK, p38, and JNK MAP kinase activity (4, 7, 34).

This is the first study that describes the role of p38 and ERK MAP kinase-mediated regulation of gene expression and cytokine production in airway smooth-muscle cells. The contribution of smooth-muscle cells to the pathogenesis of airway diseases, therefore, not only includes changes in bronchomotor tone resulting from hyperplasia and hypertrophy of airway smooth muscle but also includes a role of smooth muscle in mediating the inflammatory response through regulated release of cytokines and chemokines.