Human Lung Fibroblasts Express IL-6 in Response to Signaling Following Mast Cell Contact

Asthma is a chronic inflammatory disease of the airways. Mast cell derived cytokines may mediate both airway inflammation as well as remodeling. It has also been shown that fibroblasts can be the source of proinflammatory cytokines. In the human airways, mast cellfibroblast interactions may have pivotal effects on modulating inflammation. To study this further, we cocultured normal human lung fibroblasts (NHLF) with a human mast cell line (HMC-1) and assayed for interleukin-6 (IL-6) production, an important proinflammatory cytokine. When cultured together, NHLF/HMC-1 contact induced IL-6 secretion. Separation of HMC-1 and NHLF cells by a porous membrane inhibited this induction. HMC-1-derived cellular membranes caused an increase in IL-6 production in NHLF. Activation of p38 MAPK was also seen in cocultures by western blot while IL-6 production in cocultures was significantly inhibited be the p38 inhibitor SB203580. IL-6 production in cocultures was minimally inhibited by a chemical inhibitor of NF-κB (Bay11) indicating that NF-κB may have a minimal role in signaling IL-6 production in mast cell/fibroblasts cocultures. Blockade of ICAM-1, TNF-RI, and surface IL-1β with neutralizing antibodies failed to significantly decrease IL-6 production in our coculture indicating that other receptor-ligand associations may be responsible for this activation. These novel studies reveal the importance of cell-cell interactions in the complex milieu of airway inflammation. ABBREVIATIONS NHLF Normal human lung fibroblasts HMC-1 Human mast cells-1 IL-6 Interleukin-6 MAPK Mitogen activated protein kinase NF-κB Nuclear factor-kappa B RT-PCR Reverse transcriptase polymerase chain reaction EMSA Electrophoretic mobility shift assay ICAM Intracellular adhesion molecule TNF Tumor necrosis factor ELISA Enzyme linked immunosorbent assay INTRODUCTION In the asthmatic airway, tissues are highly sensitive to antigenic challenge and respond by mediator production and inflammatory cell infiltration. Mast cells are a major contributor of this response. Following cross linking of high affinity IgE receptors, mast cells can release a host of potent proinflammatory cytokines (1) (2). Histologically, mast cells are in close proximity to fibroblasts and c-kit/stem cell factor interactions can prolong mast cell proliferation and survival. Fibroblasts themselves are capable of releasing cytokines after activation. Therefore, it is likely that mast cells and fibroblasts can mutually activate each other. Mast cell-fibroblast interactions have been shown to be important in many cellular processes. Cairns et al. have demonstrated that tryptase isolated from human tissue caused increases in collagen type I synthesis from the human lung fibroblast cell line MRC-5 (3). Abe et al. showed similar results with human dermal fibroblasts (4). They also found that tryptase increases fibroblast proliferation in a concentration dependent manner. Tryptase induced proliferation of fibroblasts has been associated with protease activated receptor-2 (PAR-2) signaling. PAR-2 is localized more to lung fibroblasts than to dermal fibroblasts and this may potentiate the cascade of fibroblasts hyperplasia in the asthmatic lung (5). Other studies demonstrate the importance of mast cell-fibroblast interactions in collagen gel contraction (6) and growth factor production (7). The reverse has also been demonstrated. Hogaboam et al. have shown that stem cell factor from fibroblasts induced eotaxin production in mast cells (8). Eotaxin being a potent chemoattractant for eosinophils, a major inflammatory cell associated with tissue damage and epithelial denudation. Mutual mast cell-fibroblast interactions leading to enhanced cytokine secretion can lead to increased inflammatory responses and could contribute to late phase airway inflammation. This could partly explain the persistence of asthma even in the absence of antigenic challenge, this process being mediated by novel cell-cell interactions regulated by cell surface molecules. Several investigators have reported on the importance of mitogen activated protein kinases (MAPKs) in fibroblast signaling as well. There are three major MAPKsextracellular signal regulated kinase (ERK), C-Jun-N-terminal kinase (JNK), and p38, all of which have been shown to be expressed in human fibroblasts. Suzuki et al. have shown that in fresh rheumatoid synovial fibroblasts, interleukin (IL)-6 and IL-8 production are dependent on p38 MAPK activation (9). It has also been shown that tissue samples taken from patients with idiopathic pulmonary fibrosis demonstrate p38 MAPK activation in epithelial, endothelial, smooth muscle, and fibroblast cells (10). Another important signaling molecule in fibroblasts is nuclear factorkappaB (NF-κB). NF-κB is a transcription factor specific for proinflammatory gene activation. Located in the cytoplasm in an inactive dimer, NF-κB can be activated by many different stimuli and freed from its dormant state. Once free, NF-κB can translocate to the nucleus where it binds κB domains of proinflammatory genes to initiate transcription. We have previously shown the activation of NF-κB in normal human lung fibroblasts by IL-1β, tumor necrosis factor (TNF)-α, and macrophage contact (11). In this study, we demonstrate a novel cell-cell contact mechanism of fibroblast activation by mast cells mediated by p38 MAPK. In our model, mast cell-fibroblast cocultures induce IL-6 production in a concentration and time dependent manner. IL-6 is a multifunctional cytokine produced by many cell types of the body. It has differentiation and antibody secreting effects on B-cells as well as colony stimulating effects on hematopoietic cells. IL-6 can also induce acute phase protein production from hepatocytes (12) (13). Serum levels of IL-6 have been shown to be elevated in patients with such diseases as arthritis, inflammatory bowel disease, and some cervical cancers (14) (15) (16). In the lung, IL-6 can stimulate the differentiation of B-cells into IgE secreting plasma cells. For its clinical significance as an immunosuppressant and its use as an anti-asthma drug, dexamethasone was used to test inhibition of IL-6 in mast cell-fibroblast cocultures. MATERIALS AND METHODS Tissue Culture HMC-1 cells were grown in RPMI 1640 (Gibco BRL, Frederick, MD) supplemented with 11.1% FBS and 1% 1M Hepes buffer solution (Rockville, MD), 50 mg/mL Gentamycin, 0.05 M ∃-Mercaptoethanol, 1% L-glutamine, 1% sodium bicarbonate as described earlier (17). Normal human lung fibroblasts (NHLF) (Clonetics-BioWhittaker, Walkersville, MD) were grown in fibroblast basal medium (Clonetics-BioWhittaker, Walkersville, MD) at 5% CO2 at 37Ε C. Media was supplemented with 2% fetal bovine serum, human fibroblast growth factor-B (1.0 μg/mL), insulin (5mg/mL), gentamicin and amphotericin B. NHLF were cultured in 24 well culture plates at cell concentrations of 5 x 10 cells per well and incubated overnight. Supernatants were collected after 24 hours and centrifuged to remove cellular debris. Coculturing of NHLF with HMC-1 cells was done in 24 well plates with 5 x 10 NHLF per well and 1 x 10 cells/mL of HMC-1 in 1 mL cultures of NHLF media. Dose dependent coculturing of HMC-1 with NHLF were done at 0.1, 0.25, 0.50, 1, 2, and 3 x 10 HMC-1 per milliliter of media. Time dependent experiments were done with NHLF seeded at 5 x 10 cells per well in a 24 well plate with HMC-1 added at 1 x 10 cells/mL in 1 mL of NHLF media. Supernatants were collected at 12, 24, 48, and 72 hours, centrifuged, and assayed by ELISA. Cell separation experiments were done in 6 well tissue culture plates with NHLF at 1 x 10 cells/well and HMC-1 at 1 x 10/mL with 0.4 μm track-etched membrane inserts to separate HMC-1 from NHLF (Falcon, Becton Dickinson, Franklin Lakes, New Jersey). Cellular derived membranes from NHLF and HMC-1 were cultured with HMC-1 and NHLF respectively. NHLF were plated in 24 well plates at 5 x 10 NHLF per well and incubated with 75 and 100 μg/mL HMC-1-derived membranes. Whole cell HMC-1 cocultures were done for comparison. Likewise, HMC-1 were grown in 24 well plates in 1 mL cultures at 1 x 10 HMC-1/mL and then incubated with 75 and 100 μg/mL NHLFderived membranes. Here, whole NHLF cocultures were used for comparison with membrane treatments. For HMC-1 and coculture conditioned media experiments, HMC-1 were grown in NHLF media and cocultures were incubated in NHLF media for 24 hours before being harvested, centrifuged, and then added to pure NHLF cultures for 24 hours. p38 MAPK inhibition studies were done with SB203580 (Calbiochem, San Diego, CA). NHLF were pretreated with SB203580 (10 μM) for two hours prior to addition of HMC-1. For NF-κB inhibition studies, Bay 11 (1 μM, BAY-11-7082, Biomol Research Laboratories, Plymouth Meeting, PA) was added to NHLF cultures 1 hour prior to addition of HMC-1. Bay 11 inhibits NF-κB by specifically blocking phosphorylation of IκBα, a regulator protein of NF-κB. Blocking experiments were done with anti-intercellular adhesion molecule (ICAM)-1 (BD Bioscience PharMingen, San Diego, CA), anti-tumor necrosis factor receptor-1 (TNF-RI) (R&D; Systems), and anti-IL-1β (NCI Biological Resources Branch). NHLF were pretreated with anti-ICAM-1 (10 μg/mL) and anti TNF-RI (10 μg/mL) for 1 hour and HMC-1 were pretreated for 1 hour with anti-IL-1β (10 μg/mL). Dexamethasone (Sigma-Aldrich, St. Louis, MO) was added to NHLF cultures at 1 μM 48 hrs prior to addition of HMC-1. For all of these experiments, n = 3. Enzyme Linked Immunosorbent Assay (ELISA) IL-6 levels in cell-free supernatants were assayed by enzyme linked immunosorbent assay (ELISA) as previously described (18) (19) (20) (21) (11) using commercially available kits (R&D; Systems, Minneapolis, MN). Values were extrapolated or interpolated from a standard curve. Results were analyzed on an ELISA plate reader (Dynatech MR 5000 with supporting software). IL-6 Gene Expression by RT-PCR Gene expression for IL-6 was assessed using reverse transcriptase-polymerase chain reaction (RT-PCR) as previously described (22) (23) (11). RNA was extracted by an RNAzol technique from cultured cells. Briefly, total cellular RNA was extracted from cultured cells (NHLF at 1 x 10 cells/plate and HMC-1 at 1 x 10/mL) by the addition of 1.0 mL of RNAzol B (Tel-Test, Inc., Friendswood, Texas). The suspension was shook for 1 minute and centrifuged at 12,000 x g for 15 minutes at 4ΕC. The aqueous phase was washed twice with 0.8 ml phenol : chloroform (1:1, v/v), and once with 0.8 mL of chloroform. Each time, the suspension was centrifuged at 12,000 x g for 15 minutes at 4ΕC. An equal volume of isopropanol was added to the aqueous phase, and the preparation refrigerated at -20ΕC overnight. The samples were then centrifugation at 12,000 x g for 30 minutes at 4ΕC and the RNA pellet washed with 1.0 mL 75% ethanol. The RNA pellet was air dried and suspended in 20 μL of DEPC-treated water. RNA was quantitated by optical density readings at 260 nm, equalized to 1000 ng/μL, and electrophored in ethidium bromide-stained agarose gels at 1000 ng per well to determine the integrity of the 28S and 18S RNA bands. First strand cDNA was synthesized in the presence of murine leukemia virus reverse transcriptase (2.5 U/:L), 1 mM each of the nucleotides dATP, dCTP, dGTP and dTTP; RNase inhibitor (1 U/:L), 10X PCR buffer (500 mM KCl, 100 mm TrisHCl, pH 8.3), and MgCl2 (5 mM), using oligo(dT)16 (2.5 μM) as a primer. The preparation was incubated at 42ΕC for 20 minutes in a DNA thermocycler (Perkin-Elmer Corp., Norwalk, CT) for reverse transcription. PCR amplification was done on aliquots of the cDNA in the presence of MgCl2 (1.8 mM), each of dNTPs (0.2 mM), and AmpliTaq polymerase (1 U/50 μL), and paired cytokine-specific primers (0.2 nM of each primer) to a total volume of 50 μl. PCR consisted of 1 cycle of 95 C for 2 min, 45 cycles of 95Ε C for 45 sec, 60Ε C for 45 sec, and 72Ε C for 1min 30 sec, and lastly, 1 cycle of 72Ε C for 10 min. Fourteen microliters of the amplified products were subjected to electrophoresis on a 2% agarose gel stained with ethidium bromide. IL-6 bands were compared to predicted base pair migration distances from a PhiX 174 Hae III DNA marker (Promega, Madison, WI). Cellular Membrane Extraction Cellular membranes were extracted by a previously described method (11,24) . Resting NHLF and HMC-1 cells were collected separately and suspended in 1mL of TKM hypotonic buffer (Tris-HCL 50 mM [pH 7.5], KCl 25 mM, MgCl2 5 mM), incubated in a TKMphenylmethylsulfonyl fluoride (PMSF) solution for 20 minutes on ice, and dounced homogenized. In addition to dounce homogenation, cell membranes were broken with ultrasound on a microultrasonic cell disrupter (Kontes). Cell membranes were separated by high speed centrifugation (120,000 x g at 10 C) in a 73% and 35% STKM sucrose solution. After 1 hour the 35/73% interface was removed, placed in a new tube, and suspended in 1x TKM-PMSF. Finally, membranes were centrifuged at 10 C at 45,000 x g for 30 minutes, re-suspended in 100 μL of PBS, rinsed again in 50 μL of PBS, protein content measured with bicinchonic acid (BCA) assay by Pierce (Pierce Chemical, Rockford, IL), and stored at -80 o C. NHLF and HMC-1 membranes were added to HMC-1 and NHLF cultures respectively at 75 and 100 μg/mL. Since resting cell membranes produced significant increases in IL-6 production, activated cellular membranes were not used. One million HMC-1 yields approximately 100 μg of cellular derived membranes. Extraction of Nuclear Proteins Nuclear proteins were extracted from NHLF by a method previously described with modifications (11) (25). NHLF were trypsinized from 100 x 20 mm plates at 2.0 x 10 cells per plate, washed three times in PBS, and collected in a 1.5 mL microcentrifuge tube. Added to this was 100 μL of ice cold hypotonic buffer: 10 mM HEPES pH 7.9, 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM dithiothreitol (DTT), 0.5 mM phenylmethylsulfonyl fluoride (PMSF), 1 μM aprotinin, 1 μM pepstatin, 14 μM leupeptin, 50 mM NaF, 30 mM β-glycerophosphate, 1 mM Na3VO4, and 20 mM p-nitrophenyl phosphate. Cells were incubated on ice for 30 minutes and vortexed after addition of 6.25 μL of 10 % Nonidet P-40. After 2 minutes of centrifugation at 30,000 x g, supernatants were kept at -80 C while the pellets were collected and vortexed every 20 minutes for three hours in 60 μL of a hypertonic salt solution: 20 mM HEPES pH 7.9, 0.4 M NaCl, 1 mM EDTA, 1 mM EGTA, 12 mM DTT, 1 mM PMSF, 1 μM aprotinin, 1 μM pepstatin, 14 μM leupeptin, 50 mM NaF, 30 mM β-glycerophosphate, 1 mM Na3VO4, and 20 mM pnitrophenyl phosphate. This solution, containing nuclear proteins, was assayed for total protein concentration by the BCA protein assay reagent. Electromobility Shift Assay (EMSA) Nuclear translocation of NF-κB was analyzed by the electrophoretic mobility shift assay (EMSA). Briefly, 7 μg of nuclear protein were added to 2 μL of 10 x binding buffer (50 μg/mL of double stranded poly dI-dC, 10 mM Tris HCl pH 7.5, 50 mM NaCl, 0.5 mM EDTA, 0.5 mM DTT, 1 mM MgCl2, and 10 % glycerol), and 35 fmol of double stranded NF-6B consensus oligonucleotides end labeled with γ-P ATP. The reaction mixture was incubated at room temperature for 20 minutes and analyzed by eletrophoresis on a 5 % nondenaturing polyacrylamide gel. The gel was then dried on a Gel-Drier (Bio-Rad Laboratories, Hercules, CA) and exposed to Kodak X-ray film at -80 C. Cold competition studies were done with 7 μg of nuclear protein from NHLF/HMC-1 cocultures. Seven micrograms of the nuclear protein were aliquoted into three microfuge tubed and hot NF-κB, cold NF-κB, and cold AP-2 consensus oligonucleotides were added. After ten minutes of incubation at room temperature, hot NF-κB was added to the tubes with cold NF-κB and AP-2. The samples were incubated at room temperature for another ten minutes before being run on a 5 % nondenaturing polyacrylamide gel. Western Blot Analysis Cytoplasmic extracts in hypotonic buffer extracted from cocultures at 90 minutes were used to analyze phosphorylated p38 MAPK expression by western blot. Briefly, 10 μg of sample was diluted 1:2 with Laemmli buffer (Bio-Rad laboratories, Hercules, CA) and boiled for 5 minutes in a sand bath. The resultant sample was then run in a Bio-Rad Modular Mini Electrophoresis System (Hercules, CA) on a 10% polyacrylamide gel for 1 hour and then transferred to a 0.2 μm nitrocellulose membrane (Bio-Rad laboratories, Hercules, CA) for 1 hour. The blot was then removed and incubated in blocking buffer (1% BSA, 10 mM Tris pH 7.4, 100 mM NaCl, and 0.1% Tween) for 1 hour at 25C with gentle agitation. Phospho-p38 MAPK (Thr180/Tyr182) rabbit anti-human polyclonal antibody (Calbiochem, San Diego, CA) was diluted 1:1000 in blocking buffer and incubated on the blot overnight at 4C with gentle agitation. The next day, the primary antibody was removed and the blot was washed every 10 minutes for 30 minutes with agitation in wash buffer (10 mM Tris pH 7.4, 100 mM NaCl, and 0.1% Tween). After this, the blot was incubated in horse radish peroxidase conjugate antibody (mouse/human adsorbed anti rabbit, Santa Cruz Biotechnology, Santa Cruz, CA) diluted 1:5000 in blocking buffer. The blot remained in the secondary antibody for 1 hour at 25C. The blot was then washed with wash buffer for 30 minutes and covered with Super Signal West Pico Chemiluminescent Substrate (Pierce, Rockford, IL) for 5 minutes. Fisher brand polyvinyl chloride wrap (Fisher Scientific, Atlanta, GA) was used to cover the blot before exposing it to acetate transparency film (Kodak, Rochester, NY). The blot was exposed for 60 seconds before being developed. Nonphosphorylated p38 MAPK (Abcam, Cambridge, UK) was used as a loading control. Statistical Analysis All experiments were done in triplicate. All values are given as the mean ± standard deviation (SD). Statistical analysis was done using the Students t-test and Statistica version 5 computer software (StatSoft, Inc Tulsa, OK). A p-value of < 0.05 was considered significant. RESULTS NHLF/HMC-1 Cocultures Produce Significant Amounts of IL-6 Coculturing of NHLF with HMC-1 led to a significant increase in IL-6 production. As detected by ELISA in 24 hour cell free supernatants, IL-6 levels in NHLF and HMC-1 controls was 5.05 ± 0.410 pg/mL and 0.402 ± 0.16 pg/mL respectively. While IL-6 levels in NHLF plus HMC-1 (1 x 10 HMC-1/mL) cocultures was 309.5 ± 23.2 pg/mL (p< 0.00003 compared to both controls) (Figure 1A). NHLF/HMC-1 cocultures also demonstrated a dose dependent response dependent on the number of HMC-1 cultured with a constant number of NHLF (Figure 1B). NHLF control produced 33.37 ± 4.61 pg/mL of IL-6 and HMC-1 control produced 0.640 ± .271 pg/mL. IL-6 levels of NHLF/HMC-1 cocultures were 44.77 ± 6.27 pg/mL for 0.1 x 10 HMC-1 cells/mL (A), 67.90 ± 2.78 pg/mL for 0.25 x 10 HMC-1 cells/mL (B)( p< 0.0004 compared to NHLF and HMC-1 control), 102.92 ± 28.0 pg/mL for 0.5 x 10 HMC-1 cells/mL (C) (p< 0.02 compared to NHLF and HMC-1 control), 299.25 ± 16.40 pg/mL for 1 x 10 HMC-1 cells/mL (D) (p< 0.0005 compared to NHLF and HMC-1 control), 326.21 ± 12.97 pg/mL for 2 x 10 HMC-1 cells/mL (E) (p< 0.0005 compared to NHLF and HMC-1 control), and 349.20 ± 47.0 pg/mL IL-6 for 3 x 10 HMC-1 cells/mL (F) (p< 0.0005 compared to NHLF and HMC-1 control). NHLF/HMC-1 cocultures release IL-6 protein in a time dependent manner over 72 hours (Figure 1C). NHLF were cocultured with HMC-1 at 1 x 10 HMC-1/mL and harvested at 12, 24, 48, and 72 hours for assay by ELISA. NHLF alone produced small amounts of IL-6 over the time course: 30.26 ± 0.21 pg/mL at 12 hours, 50.28 ± 2.30 pg/mL at 24 hours, 44.89 ± 0.24 pg/mL at 48 hours, and 98.75 ± 2.82 pg/mL at 72 hours. HMC-1 produced very small amounts of IL-6: 0.355 ± 0.50 pg/mL at 12 hours, 0.695 ± 0.86 pg/mL at 24 hours, 1.56 ± 0.56 pg/mL at 48 hours, and 0.981 ± 1.39 pg/mL at 72 hours. When HMC-1 were cocultured with NHLF, however, IL-6 production significantly rose in as little as 12 hours and began to plateau at 48 to 72 hours (205.6 ± 28.6 pg/mL at 12 hours [p< 0.001 compared to both controls at 12 hrs], 234.42 ± 23.8 pg/mL at 24 hours [p< 0.001 compared to both controls at 24 hrs], 309.2 ± 3.37 pg/mL at 48 hours [p< 0.0001 compared to both controls at 48 hrs], and 310.3 ± 25.4 pg/mL at 72 hours[p< 0.0001 compared to both controls at 72 hrs]). We chose 24 hours for convenience and based on the findings of significant IL-6 induction at this time point. Cell-cell contact is essential for IL-6 production Coculture of NHLF with HMC-1 produced increased amounts of IL-6. To investigate whether direct cell to cell contact is needed for this activation, NHLF and HMC-1 were incubated in the same well but separated from contact by a porous membrane. When the NHLF and HMC1 were separated by a 0.4 μm porous membrane, IL-6 production was decreased to 64.2 ± 10.9 pg/mL as compared to that of the coculture (293.9 ± 9.3 pg/mL, p< 0.00001) (Figure 2A). In order to test which cell is the main producer of this IL-6, cellular derived membranes from resting NHLF and HMC-1 were prepared and incubated in HMC-1 and NHLF cultures respectively. NHLF cultured alone produced 25.7 ± 4.6 pg.mL of IL-6, while HMC-1 alone produced no IL-6 (0.00 ± 0.00 pg/mL). When HMC-1 membranes were added to NHLF cultures at 75 and 100 μg/mL, IL-6 production was increased to 58.6 ± 21.3 pg/mL and 151.7 ± 9.8 pg/mL (p< 0.00004) respectively (Figure 2B). However, when NHLF derived membranes were added to HMC-1 cultures, IL-6 production did not increase over control levels (data not shown). To completely remove HMC-1 cellular influence on NHLF cultures, HMC-1 conditioned media at two dilutions was added to NHLF cultures. Coculture conditioned media was also added to NHLF cultures to test for additional effects on IL-6 production (Figure 2C). HMC-1 conditioned media at 1:10 and 1:100 dilution added to NHLF cultures failed to significantly increase IL-6 production over NHLF control levels (33.3 ± 2.7 and 24.0 ± 0.3 pg/mL for 1:10 and 1:100 dilutions respectively compared to 27.1 ± 3.6 for NHLF control). IL-6 production in NHLF treated with HMC-1 conditioned media was still significantly less than whole cell NHLF/HMC-1 cocultures however (p< 0.0005). Coculture conditioned media had no additional effects on IL-6 production in NHLF (224.2 ± 6.3 pg/mL for NHLF + coculture conditioned media and 222.9 ± 2.0 pg/mL for NHLF + HMC-1). IL-6 gene expression and NF-κB are increased in cocultures NHLF and HMC-1 were incubated separately and together for 6 hours and then harvested for RNA. Analysis of IL-6 gene transcripts by RT-PCR was then performed. Hypoxanthine phophoribosyltransferase (HPRT), an enzyme important for purine synthesis and nervous system function, was used as a house keeping gene. After 6 hours, IL-6 gene transcription was increased markedly in NHLF/HMC-1 cocultures as compared to NHLF or HMC-1 alone (Figure 3). Studies also showed NF-κB activation after two hours of coculture (Figure 4A). Cold competition studies were done to determine the specificity of the NF-κB binding. Here, the NFκB band from the NHLF/HMC-1 cocultures disappears when preincubated with cold NF-κB before hot NF-κB is added. Samples were also preincubated with cold AP-2, a nonspecific nuclear binding protein, before hot NF-κB was added. This band did not disappear indicating that nuclear binding is specific for our NF-κB oligonucleotide (Figure 4B). Role of p38 MAPK in signaling IL-6 production in cocultures Because of its importance in cytokine signaling, phosphorylated p38 MAPK was assayed by western blot. Neither cell type alone, NHLF nor HMC-1, exhibited p38 MAPK activation. When the two cell types were incubated together, however, the presence of phosphorylated p38 MAPK was detectable (Figure 5A). Unphosphorylated (nonactive) p38 MAPK was used as a loading control. To test the functionality of this p38, a specific inhibitor of p38 MAPK activation, SB203580, was used to test the effects on IL-6 protein production in cocultures. After pretreatment for two hours with SB203580, IL-6 production in NHLF/HMC-1 coculture was decreased from 299.3 ± 16.4 pg/mL to 33.0 ± 9.1 pg/mL (p < 0.00002). SB203580 also had an inhibitory effect on IL-6 production in NHLF control (33.4 ± 4.6 pg/mL without the inhibitor versus 7.6 ± 0.9 pg/mL with the inhibitor (Figure 4B) Role of NF-κB Activation in IL-6 production from Cocultures To examine the role NF-κB plays in inducing IL-6 protein production from NHLF/HMC1 cocultures, we inhibited NF-κB with a chemical inhibitor, Bay 11. Pre-incubation of NHLF with Bay 11 (1 μM) for 1 hour inhibited IL-6 production in NHLF/HMC-1 cocultures by 17%. Higher concentrations of Bay 11 had toxic effects on NHLF. NHLF and HMC-1 alone produced little IL-6 (31.5 ± 4.6 pg/mL and 0.604 ± 0.37 pg/mL respectively) while NHLF + HMC-1 produced significantly higher levels of IL-6 (290.0 ± 9.7 pg/mL, p< 0.00002 compared to both controls). Bay 11 inhibited IL-6 production in NHLF control and NHLF/HMC-1 cocultures. NHLF control levels of IL-6 went from 31.5 ± 4.6 pg/mL to 11.7 ± 1.3 pg/mL with Bay 11 (p< 0.002), while that of NHLF/HMC-1 cocultures went from 290.0 ± 9.7 pg/mL to 241.3 ± 1.4 pg/mL with Bay 11 (p< 0.02) (Figure 5). These data indicate that IL-6 production in NHLF/HMC-1 cocultures may be only partially mediated by the NF-kB pathway. Dexamethasone inhibits IL-6 Production Dexamethasone, a drug widely used in the treatment of asthma, was also included to investigate the inhibition of IL-6 production. Synthetic glucocorticoids, like dexamethasone, work by ultimately stabilizing the NF-κB/IκB complex by inducing IκB gene transcription. They may also function through various metabolic pathways not yet understood. Pre-incubation of NHLF cultures with dexamethasone (1 μM) for 48 hours before HMC-1 were added significantly reduced IL-6 production. NHLF alone produced 116.1 ± 11.8 pg/mL of IL-6 and HMC-1 alone produced 1.5 ± 0.2 pg/mL of IL-6. Dexamethasone decreased IL-6 production in NHLF/HMC-1 cocultures from 483.0 ± 8.1 pg/mL to 395.3 ± 22.3 pg/mL (p< 0.05 compared to NHLF + HMC1) (Figure 7). Dexamethasone treatment of NHLF alone reduced IL-6 production but this was not significant from NHLF control. DISCUSSION Inflammation in the airway is orchestrated by a number of different cytokines working together and separately. These cytokines are produced by resident cells of the lungs as well as infiltrating inflammatory cells. The exact mechanisms by which these cytokines work and how they are orchestrated by the host cells of the lung are not completely understood. We do know, however, that cell to cell contact between lung cells and the infiltrating cells of the immune system play a major role in the pathogenesis of inflammation and airway remodeling, the irreversible structural damage caused by repeated inflammatory insult. Among these infiltrating cells are macrophages, eosinophils, and mast cells. All of which are seen in increased numbers in the asthmatic lung (26). As we have previously shown, macrophage-fibroblasts interactions lead to increased granulocyte macrophage-colony stimulating factor (GM-CSF) gene expression and protein production (11). GM-CSF is a potent activator of eosinophils. Here we have shown that mast cell-fibroblast interactions produce significantly higher amounts of IL-6 than control cells. IL-6 is a mediator of many inflammatory processes of the human body including B-cell maturation and differentiation as well as acute phase protein production and hematopoietic effects on stem cells (12) (13). Of interest to the lung, IL-6 can cause class switching of B-cells into IgE producing plasma cells. IgE has been suspect of airway inflammation for a long time. By cross linking high affinity IgE receptors (Fc,R1) on mast cells, IgE attached to antigen can cause a cascade of inflammatory events in the bronchi. Mast cells produce preformed and newly synthesized mediators that are ready to be released within minutes upon activation. These mediators include cytokines, chemokines, luekotrienes, and proteases. We have found that the coculturing of normal human lung fibroblasts with a leukemic mast cell line, HMC-1, increased interleukin-6 secretion into the supernatant after 24 hours (Figure 1A). The IL-6 production in cocultures was positively correlated to the number of HMC1 in NHLF/HMC-1 cocultures and began to plateau at 1 x 10 HMC-1/mL (Figure 1B). A time dependent release of IL-6 was also seen over 72 hours in our study. Twelve hour incubation supernatants from NHLF/HMC-1 cocultures had significant amounts of IL-6 compared to NHLF and HMC-1 alone. This increase of IL-6 production progressed and began to plateau at 48 to 72 hours (Figure 1C). To investigate the role of direct cell to cell contact in our system, we separated the two cell types by seeding the fibroblasts in the bottom of a six well plate and then added the HMC-1 separated by a 0.4 :m porous membrane. When the two cell types were separated from direct contact like this, the amount of IL-6 detected in the supernatant after 24 hours was significantly less than for that of direct cell-cell contact (Figure 2A). The amount of IL-6 in the separated coculture was still significantly greater than NHLF or HMC-1 control levels. This does not completely rule out the role of soluble mediator production in the activation of IL-6 in our model but it does show the importance of direct cell to cell contact for IL-6 production. In order to determine the cell type responsible for the bulk of the IL-6 production, membranes from resting NHLF and HMC-1 were extracted and incubated with whole HMC-1 and NHLF respectively. Although NHLF-derived membranes did not activate HMC-1 to produce IL-6, HMC-1-derived membranes had a dose dependent effect on NHLF IL-6 production (Figure 2B). To completely remove HMC-1 cellular influence from NHLF, HMC-1 conditioned media at two dilutions (1:10 and 1:100) was added to pure NHLF cultures. Neither dilution of HMC-1 conditioned media caused a significant rise in IL-6 production from NHLF (Figure 2C). NHLF/HMC-1 coculture conditioned media was also added to pure NHLF cultures to test additional effects on IL-6 production. IL-6 production with coculture conditioned media was found not to be significantly different from actual NHLF/HMC-1 cocultures (Figure 2C). We also looked at transcription of the IL-6 gene by reverse transcriptase polymerase chain reaction. RNA was extracted from cocultures at six hours and subjected to RT-PCR analysis. Coculturing of NHLF with HMC-1 induced increased expression of the IL-6 gene mRNA (Figure 3). Activation of NF-κB was also assayed in our coculture. Translocation of activated NF-κB to the nucleus of NHLF cocultured with HMC-1 was increased after 2 hours of coculturing (Figure 4A). Cold competition studies with NF-κB were also done to show specificity of NF-κB binding (Figure 4B). For its significance in cytokine signaling, p38 MAPK activity was assayed by western blot. Phosphorylated p38, the activated form of p38 MAPK, was induced in our coculture and was inhibited by a specific inhibitor, SB203580 (Figure 5A). Sano et al. have previously shown that p38 and ERK are involved in angiotensin II-mediated IL-6 production in cardiac fibroblasts, and its induction was independent of NF-κB (27). Others have also shown the activation of p38 MAPK in the induction of IL-6 in fibroblasts or fibroblasts-like synoviocytes (9) (28) (29) (30). In our model, pretreatment of NHLF with SB203580 for two hours dropped IL-6 production to prestimulus levels, indicating that p38 is a major signaling pathway for coculture induced IL-6 production (Figure 5B). It is uncertain whether several signals converge on the activation of p38 or if there is one receptor-mediated signal which activates p38 in our coculture. Further studies are needed to identify this pathway and are out of the scope of this paper. Aan inhibitor of NF-κB, Bay 11, was used to block IL-6 production to see how much of the IL-6 signal may come through NF-κB. ELISA results showed that Bay 11 at 1μM minimally but significantly reduced IL-6 levels in NHLF/HMC-1 cocultures, indicating that IL-6 transcription may be partially regulated by NF-κB (Figure 6). Bay 11 works by inhibiting inhibitory kappa B (IκB) phosphorylation. In the cytoplasm, NF-κB is bound to IκB, and upon phosphorylation of IκB by certain signaling pathways, NF-κB is freed from its complex and is able to translocate to the nucleus where it initiates transcription of inflammatory genes. These data suggests that NF-κB may play a minimal role in IL-6 signaling in NHLF/HMC-1 cocultures. Preliminary experiments looking at blocking IL-6 signaling by using antibodies to specific cell surface molecules like ICAM-1, TNF-RI, and IL-1β were negative (data not shown). Other protein-protein or receptor-ligand interactions in combination or separately may mediate IL-6 activation. Further experiments are needed to delineate this activation cascade and are the focus of future research in our laboratory. For its clinical significance in the treatment of moderate to severe persistent asthma, dexamethasone was used to investigate inhibition of IL-6 protein production in NHLF/HMC-1 cocultures. Corticosteroids like dexamethasone work by inducing IκB gene transcription which will later work to stabilize the NF-κB dimer and prevent avtivation. NHLF pretreated for 48 hours at 1 μM dexamethasone was able to significantly inhibit IL-6 protein production as detected in 24 hour supernatants (Figure 7). These data parallel the Bay 11 data and further suggests that IL-6 production in NHLF/HMC-1 cocultures in minimally dependent on NF-κB. Due to the extended preincubation time with dexamethasone which is needed for its complete effect, base line IL-6 levels in NHLF were higher than previously seen. Dexamethasone, at concentrations used in this study, had no morphological or toxic effects on NHLF. These data give insight into the pathogenesis behind cell-cell mediated inflammation. Resident airway cells along with infiltrating inflammatory cells of the immune system may be one way in which activation of the bronchi occurs. Fibroblasts are known to secrete proinflammatory cytokines including IL-6, IL-8, GM-CSF, and transforming growth factor-beta (TGF-β). Mast cells are known mediators of allergic disease and themselves also secrete a host of vasoand broncho-active cytokines. Mast cells have also been found in increased numbers in the asthmatic lung. Knowing that IL-6 induction in fibroblasts by mast cell contact is mediated through p38 MAPK may lead to further drug targets. ACKNOWLEDGEMENTSThis study was supported by NIH grants AI-43310 and HL-63070, The Rondal Cole Foundation,and the Chair of Excellence in Medicine (State of Tennessee Grant 20233), The CardiovascularResearch Institute, and the Research Development Committee, East Tennessee State University. REFERENCES1. Krishnaswamy, G., Kelley, J., Johnson, D., Youngberg, G., Stone, W., Huang, S. K., Bieber, J.,and Chi, D. S. 2001. The human mast cell: functions in physiology and disease. Front Biosci.6:D1109-D1127. 2. Krishnaswamy, G. 2001. Treatment strategies for bronchial asthma: an update. Hosp.Pract.(OffEd). 36:25-35. 3. Cairns, J. A. and Walls, A. F. 1997. Mast cell tryptase stimulates the synthesis of type Icollagen in human lung fibroblasts. J.Clin.Invest. 99:1313-1321. 4. Abe, M., Kurosawa, M., Ishikawa, O., Miyachi, Y., and Kido, H. 1998. Mast cell tryptasestimulates both human dermal fibroblast proliferation and type I collagen production.Clin.Exp.Allergy. 28:1509-1517. 5. Akers, I. A., Parsons, M., Hill, M. R., Hollenberg, M. D., Sanjar, S., Laurent, G. J., andMcAnulty, R. J. 2000. Mast cell tryptase stimulates human lung fibroblast proliferation viaprotease-activated receptor-2. Am.J.Physiol Lung Cell Mol.Physiol. 278:L193-L201. 6. Skold, C. M., Ohkuni, Y., Liu, X. D., Numerof, R., and Rennard, S. I. 2001. Co-culturedhuman mast cells stimulate fibroblast-mediated contraction of collagen gels. Inflammation. 25:47-51. 7. Artuc, M., Steckelings, U. M., and Henz, B. M. 2002. Mast cell-fibroblast interactions: humanmast cells as source and inducers of fibroblast and epithelial growth factors. J.Invest Dermatol.118:391-395. 8. Hogaboam, C., Kunkel, S. L., Strieter, R. M., Taub, D. D., Lincoln, P., Standiford, T. J., andLukacs, N. W. 1998. Novel role of transmembrane SCF for mast cell activation and eotaxinproduction in mast cell-fibroblast interactions. J.Immunol. 160:6166-6171. 9. Suzuki, M., Tetsuka, T., Yoshida, S., Watanabe, N., Kobayashi, M., Matsui, N., and Okamoto,T. 2000. The role of p38 mitogen-activated protein kinase in IL-6 and IL-8 production from theTNF-alphaor IL-1beta-stimulated rheumatoid synovial fibroblasts. FEBS Lett. 465:23-27. 10. Yoshida, K., Kuwano, K., Hagimoto, N., Watanabe, K., Matsuba, T., Fujita, M., Inoshima, I.,and Hara, N. 2002. MAP kinase activation and apoptosis in lung tissues from patients withidiopathic pulmonary fibrosis. J.Pathol. 198:388-396. 11. Fitzgerald, S. M., Chi, D. S., Hall, H. K., Reynolds, S. A., Aramide, O., Lee, S. A., andKrishnaswamy, G. 2003. GM-CSF Induction in Human Lung Fibroblasts by IL-1beta, TNF-alpha, and Macrophage Contact. J.Interferon Cytokine Res. 23:57-65. 12. Wigmore, S. J., Walsh, T. S., Lee, A., and Ross, J. A. 1998. Pro-inflammatory cytokinerelease and mediation of the acute phase protein response in fulminant hepatic failure. IntensiveCare Med. 24:224-229. 13. O'Riordain, M. G., Ross, J. A., Fearon, K. C., Maingay, J., Farouk, M., Garden, O. J., andCarter, D. C. 1995. Insulin and counterregulatory hormones influence acute-phase proteinproduction in human hepatocytes. Am.J.Physiol. 269:E323-E330. 14. Ishihara, K. and Hirano, T. 2002. IL-6 in autoimmune disease and chronic inflammatoryproliferative disease. Cytokine Growth Factor Rev. 13:357-368. 15. Jego, G., Bataille, R., and Pellat-Deceunynck, C. 2001. Interleukin-6 is a growth factor fornonmalignant human plasmablasts. Blood. 97:1817-1822. 16. Wei, L. H., Kuo, M. L., Chen, C. A., Cheng, W. F., Cheng, S. P., Hsieh, F. J., and Hsieh, C.Y. 2001. Interleukin-6 in cervical cancer: the relationship with vascular endothelial growth factor.Gynecol.Oncol. 82:49-56. 17. Krishnaswamy, G., Martin, R., Walker, E., Li, C., Hossler, F., Hall, K., and Chi, D. S. 2003.Moraxella catarrhalis induces mast cell activation and nuclear factor kappaB-dependent cytokinesynthesis. Front Biosci. 8:A40-A47. 18. Chi, D., Henry, J., Kelley, J., Thorpe, R., Smith, J. K., and Krishnaswamy, G. 2000. Theeffects of HIV infection on endothelial function. Endothelium. 7:223-242. 19. Krishnaswamy, G., Kelley, J., Yerra, L., Smith, J. K., and Chi, D. S. 1999. Human endothelium as a source of multifunctional cytokines: molecular regulation and possible role inhuman disease. J.Interferon Cytokine Res. 19:91-104. 20. Krishnaswamy, G., Smith, J. K., Mukkamala, R., Hall, K., Joyner, W., Yerra, L., and Chi, D.S. 1998. Multifunctional cytokine expression by human coronary endothelium and regulation bymonokines and glucocorticoids. Microvasc.Res. 55:189-200. 21. Krishnaswamy, G., Lakshman, T., Miller, A. R., Srikanth, S., Hall, K., Huang, S. K., Suttles,J., Smith, J. K., and Stout, R. 1997. Multifunctional cytokine expression by human mast cells:regulation by T cell membrane contact and glucocorticoids. J.Interferon Cytokine Res. 17:167-176. 22. Huang, S. K., Essayan, D. M., Krishnaswamy, G., Yi, M., Kumai, M., Su, S. N., Xiao, H. Q.,Lichtenstein, L. M., and Liu, M. C. 1994. Detection of allergenand mitogen-induced humancytokine transcripts using a competitive polymerase chain reaction. J.Immunol.Methods. 168:167-181. 23. Krishnaswamy, G., Liu, M. C., Su, S. N., Kumai, M., Xiao, H. Q., Marsh, D. G., and Huang,S. K. 1993. Analysis of cytokine transcripts in the bronchoalveolar lavage cells of patients withasthma. Am.J.Respir.Cell Mol.Biol. 9:279-286. 24. Noelle, R. J., Daum, J., Bartlett, W. C., McCann, J., and Shepherd, D. M. 1991. Cognateinteractions between helper T cells and B cells. V. Reconstitution of T helper cell function using purified plasma membranes from activated Th1 and Th2 T helper cells and lymphokines.J.Immunol. 146:1118-1124. 25. Li, C., Browder, W., and Kao, R. L. 1999. Early activation of transcription factor NF-kappaBduring ischemia in perfused rat heart. Am.J.Physiol. 276:H543-H552. 26. Kassel, O., de Blay, F., Duvernelle, C., Olgart, C., Israel-Biet, D., Krieger, P., Moreau, L.,Muller, C., Pauli, G., and Frossard, N. 2001. Local increase in the number of mast cells andexpression of nerve growth factor in the bronchus of asthmatic patients after repeated inhalationof allergen at low-dose. Clin.Exp.Allergy. 31:1432-1440. 27. Sano, M., Fukuda, K., Sato, T., Kawaguchi, H., Suematsu, M., Matsuda, S., Koyasu, S.,Matsui, H., Yamauchi-Takihara, K., Harada, M., Saito, Y., and Ogawa, S. 2001. ERK and p38MAPK, but not NF-kappaB, are critically involved in reactive oxygen species-mediated inductionof IL-6 by angiotensin II in cardiac fibroblasts. Circ.Res. 89:661-669. 28. Hayashi, R., Yamashita, N., Matsui, S., Fujita, T., Araya, J., Sassa, K., Arai, N., Yoshida, Y.,Kashii, T., Maruyama, M., Sugiyama, E., and Kobayashi, M. 2000. Bradykinin stimulates IL-6and IL-8 production by human lung fibroblasts through ERKand p38 MAPK-dependentmechanisms. Eur.Respir.J. 16:452-458. 29. Miyazawa, K., Mori, A., Miyata, H., Akahane, M., Ajisawa, Y., and Okudaira, H. 1998.Regulation of interleukin-1beta-induced interleukin-6 gene expression in human fibroblast-like synoviocytes by p38 mitogen-activated protein kinase. J.Biol.Chem. 273:24832-24838. 30. Ridley, S. H., Sarsfield, S. J., Lee, J. C., Bigg, H. F., Cawston, T. E., Taylor, D. J., DeWitt, D.L., and Saklatvala, J. 1997. Actions of IL-1 are selectively controlled by p38 mitogen-activatedprotein kinase: regulation of prostaglandin H synthase-2, metalloproteinases, and IL-6 at differentlevels. J.Immunol. 158:3165-3173. FIGURE LEGENDSFigure 1. A.) NHLF/HMC-1 cocultures induced a significant rise in interleukin-6 levels. Cellcoculture supernatants were harvested after 24 hrs and assayed for IL-6 by ELISA. * p< 0.00003compared to both controls. B.) Increasing numbers of HMC-1 in NHLF/HMC-1 coculturesexhibited a dose dependent response in the amount of IL-6 produced. HMC-1 at A = 0.1 x 10HMC-1/mL, B = 0.25 x 10 HMC-1/mL, C = 0.5 x 10 HMC-1/mL, D = 1 x 10 HMC-1/mL, E =2 x 10 HMC-1/mL, F = 3 x 10 HMC-1/mL were cultured with NHLF for 24 hrs and harvestedfor IL-6 by ELISA. * p< 0.0004, ** p< 0.02, and *** p< 0.0005 compared to NHLF and HMC-1controls. C.) Kinetics of IL-6 production in NHLF/HMC-1 cocultures. NHLF were coculturedwith HMC-1 (1 x 10 cells/mL) for 12, 24, 48, and 72 hours and assayed for IL-6 production byELISA. * p< 0.001 compared to both controls at their respective time point, ** p< 0.0001compared to both controls at their respective time point. Values are means ± SD for triplicatesamples, n = 3.Figure 2. A.) Separation of NHLF and HMC-1 by a porous membrane decreased IL-6 production. NHLF were cultured with HMC-1 in direct contact and separated by a 0.4 μm porousmembrane for 24 hrs and then harvested for IL-6 by ELISA. * p< 0.00001 compared to bothcontrols and ** p< 0.00001 compared to coculture. B.) HMC-1 derived cellular membranesinduced IL-6 production in NHLF in a dose dependent manner. HMC-1 membranes at 75 and 100 μg/mL induced a dose dependent rise in IL-6 production as detected by ELISA in 24 hrsupernatants. * p< 0.000003 compared to NHLF and HMC-1 controls, ** p< 0.00004 comparedto NHLF control. C.) Coculture conditioned and HMC-1 conditioned media do not furtheractivate NHLF to produce IL-6. Cocultures and HMC-1 were incubated for 24 hours in NHLFmedia and then centrifuged and added to pure NHLF cultures. HMC-1 conditioned media was added at 1:10 and 1:100 dilutions. * p< 0.001 compared to NHLF and HMC-1 control, ** p<0.0005 compared to NHLF + HMC-1. Values are means ± SD for triplicate samples, n = 3.Figure 3. A.) NHLF/HMC-1 cocultures increased IL-6 gene expression. NHLF and HMC-1were cultured separate and together for 6 hours and harvested for analysis by RT-PCR. HPRTwas used as a housekeeping gene to ensure equal loading. B.) NHLF/HMC-1 cocultures increase activation of NF-κB. NHLF and HMC-1 were incubated separately and together for 2 hours andassayed for NF-kB translocation by EMSA. Figure 4. A.) NHLF/HMC-1 cocultures increase activation of NF-κB. NHLF and HMC-1 wereincubated separately and together for 2 hours and assayed for NF-kB translocation by EMSA. B.) Cold competition for NF-κB oligonucleotide. Nuclear extracts from NHLF/HMC-1 cocultures were labeled with hot NF-κB oligo, cold NF-κB oligo followed by hot NF-κB oligo, and cold AP2 oligo followed by hot NF-κB oligo. Preincubation with cold NF-κB oligo before hot oligo was added eliminated the NF-κB band. Preincubation with cold AP-2 oligo before hot NF-κB oligo was added did not eliminate the NF-κB band. These results show specificity of the NF-κB oligonucleotide for κB binding domains in our nuclear extracts.Figure 5. A.) Western blot was used to detect the presence of phosphorylated p38 MAPK.Coculturing of NHLF and HMC-1 induced phosphorylation of p38 in NHLF/HMC-1 cocultures.Nonactive, unphosphorylated p38 was used to ensure equal loading. B.) SB203580 decreasedIL-6 production from NHLF-HMC-1 cocultures. NHLF were pretreated for 2 hours withSB203580 before being incubated with HMC-1. Cell supernatants were harvested after 24 hoursand assayed for IL-6 levels by ELISA. * p < 0.00002 compared to NHLF and HMC-1 controls, **p <0.0007 compared to NHLF Control, *** p < 0.00002 compared to NHLF/HMC-1 coculture.Values are means ± SD for triplicate samples, n = 3. Figure 6. Bay 11, an inhibitor of NF-κB, decreased IL-6 production in NHLF/HMC-1 cocultures. Bay 11 at 1 μM added to NHLF 1 hour prior to addition of HMC-1 was able tominimally but significantly decrease the amount of IL-6 detected in 24 hr supernatants perELISA. * p< 0.00002 compared to both controls, ** p< 0.002 compared to NHLF control, *** p<0.02 compared to NHLF/HMC-1 coculture. Values are means ± SD for triplicate samples, n = 3.Figure 7. Dexamethasone decreased IL-6 production in NHLF/HMC-1cocultures.Dexamethasone, a synthetic glucocorticoid, added to NHLF cultures 48 hrs prior toaddition of HMC-1 decreased IL-6 production. Supernatants were harvested after 24 hrs andassayed for IL-6 by ELISA. * p< 0.0008 compared to NHLF and HMC-1 controls ** p< 0.05compared to NHLF/HMC-1 coculture. Values are means ± SD for triplicate samples, n = 3.