Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Neutrophils are an important component of the acute inflammatory response in a variety of disease states. Recruited neutrophils possess enhanced functional activity and show prolonged survival at sites of inflammation. Cellular infiltrates in asthma and allergic rhinitis typically involve the eosinophilic granulocyte (1). However, striking infiltration of neutrophils occurs into the upper airway during the common cold and into the lower airway during respiratory viral-induced asthma exacerbation (2, 3). Moreover, sudden-onset fatal asthma is characterized by neutrophils in excess of eosinophils in addition to increased numbers of mucous glands (4, 5). Effects of antiasthma treatment with inhaled glucocorticoids have repeatedly documented unchanged or increased numbers of neutrophils in the airways (6).

In vitro studies of neutrophil survival have identified numerous factors that prolong neutrophil survival. These include granulocyte macrophage colony-stimulating factor (GM-CSF) (9), microbial products (10, 11), interleukin (IL)-2 (12), IL-6 (13), leukotriene B₄ (LTB₄) (14), platelet-activating factor (PAF) (15), IL-1, granulocyte colony-stimulating factor (G-CSF), interferon (IFN)-, and lipopolysaccharide (11). Glucocorticosteroids, a crucial antiallergenic medication, inhibit neutrophil but not eosinophil apoptosis (16). Conflicting studies of the effects on neutrophil-programmed cell death have been reported for tumor necrosis factor (TNF)- and C5a (9, 11, 20). On the other hand, neutrophil apoptosis was induced by proteolytic enzymes (21), IL-10 (22), and phorbol myristate acetate (23). Clearly, there are myriad influences on the cellular infiltrates within the inflamed airways. The aim of the current study was to comprehensively investigate epithelial products that influence neutrophil longevity and also to investigate the effects of glucocorticoid treatment on overall survival of neutrophils.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Reagents

Baxter Diagnostics, Inc. (McGaw Park, IL) supplied 6% dextran 70 in 0.9% normal saline. Ficoll-Paque Plus was purchased from Pharmacia Biotech (Piscataway, NJ). GIBCO-BRL (Grand Island, NY) supplied Earle's balanced salt solution (EBSS), TRIzol reagent, and First Strand Kits. Reverse transcriptase/polymerase chain reaction (RT-PCR) primer sets for IL-1, IL-1, IL-6, IFN-, GM-CSF, G-CSF, TNF-, IL-10, and -actin were purchased from Stratagene (San Diego, CA). Taq polymerase was obtained from Perkin-Elmer (Foster City, CA). Eosin Y and pronase were obtained from Sigma Chemical Co. (St. Louis, MO). RiboQuant Multi-Probe RNase Protection Assay (an In Vitro Transcription Kit, a ribonuclease [RNase] protection assay [RPA] kit, and human cytokine/ chemokine Multi-Probe Template Set-2) was purchased from PharMingen (San Diego, CA). Tris-saturated phenol, chloroform, isoamyl alcohol, ethanol, and Ultrapure RNase-free water were supplied by Sigma. Sequagel Concentrate, Diluent, and Buffer Solutions were obtained from National Diagnostics (Atlanta, GA). [³²P]uridine triphosphate (UTP) (10 mCi/ml) was supplied by Amersham (Arlington Heights, IL). RPMI-1640 media, fetal calf serum (FCS) and N-2-hydroxyethylpiperazine-N'-ethane sulfonic acid (Hepes) were purchased from BioWhittaker (Walkersville, MD). Bronchial/tracheal epithelial cell basal media (BEBM), bronchial/tracheal epithelial cell growth media (BEGM) and dexamethasone (Dex) were purchased from Clonetics (San Diego, CA). Paired antibodies and cytokine standards for anti-cytokine enzyme-linked immunosorbent assays (ELISAs) were purchased from PharMingen and used with Nunc Maxisorb 96-well plates. Recombinant human (rh) GM-CSF, rhIL-1, rhIL-1, rhIL-6, and neutralizing antibodies to rhGM-CSF, rhIL-1, and rhIL-6 were purchased from R&D Systems (Minneapolis, MN). Neutralizing antibodies to rhIL-1 were supplied by Endogen (Woburn, MA). PAF, prostaglandin E₂ (PGE₂), and an enzyme immunoassay (EIA) kit for PGE₂ were purchased from Cayman Chemical (Ann Arbor, MI). An In Situ Cell Death Detection Kit (Fluorescein) was purchased from Boehringer Mannheim (Indianapolis, IN).

Neutrophil Preparation

Neutrophils were isolated from the peripheral blood of healthy human donors by a modified procedure using Ficoll-Paque Plus. Blood was drawn into syringes containing dextran (8 ml/50 ml blood) and ethylenediaminetetraacetic acid (EDTA) to achieve a final concentration of 10 mM EDTA. After sedimentation for 1 h, leukocyte-rich plasma was recovered into a 50-ml conical tube and centrifuged at 250 × g for 10 min. The pellet was diluted in 35 ml EBSS and carefully layered over 15 ml of Ficoll-Paque in 50-ml conical centrifuge tubes. After centrifugation at 450 × g for 12 min, the top interface containing the mononuclear cell layer was removed and discarded. The neutrophil-containing fraction appeared as a diffuse cell suspension above the cell pellet. The upper third of the suspension was removed, washed, and resuspended in RPMI containing 10% FCS, 10 mM Hepes, and 1% penicillin/streptomycin. A total of 10 µl of this suspension was added to Turk's solution to count the total number of leukocytes and the percentage of neutrophils. Neutrophils were typically  98% pure.

Isolation of Upper and Lower Airway Epithelial Cells for Primary Culture

Upper-airway specimens were provided by the Pathology Department of Scripps Clinic and Research Foundation (La Jolla, CA) and obtained from patients undergoing surgery for chronic sinusitis. Tissues were transported to the laboratory in EBSS containing 25 mM Hepes and 1% penicillin, streptomycin, and amphotericin B. Preparation of epithelial cell monolayers followed a modification of previously described methods (24). After washing with supplemented EBSS, 0.1% pronase was added and tissues were incubated for up to 2 h at 37°C with 100% humidity and 5% CO₂. The cells were dispersed, pelleted, and resuspended in serum-free liquid culture-media based on LHC-9 (BEGM). BEGM is specifically formulated to discourage adherence of cells of fibroblast or endothelial origin. BEGM, 4 ml, containing 0.5 to 1.0 × 10⁵ cells was transferred to the T-25 tissue culture flask and incubated in a humidified atmosphere at 37°C and 5% CO₂. Over the ensuing 24 to 48 h, epithelial cells became adherent while leukocytes and other contaminating cells remained nonadherent or died. After 48 h, and every 2 d thereafter, medium was exchanged until cells reached confluence, between 6 and 10 d after plating. Primary cell cultures in the third to fifth passages were used for generation of supernatants and for RNA isolation studies. The epithelial cells prepared in this fashion were virtually pure ( > 99.5%) epithelial cell cultures, as confirmed by cytochemical staining using antibody to cytokeratin.

Generation of Epithelial Cell Supernatants

At confluence, BEGM was replaced with unsupplemented basal media (BEBM) in the presence or absence of 25 ng/ ml TNF-. After 1 h of stimulation, media were replaced with BEBM to avoid direct effects of TNF- on neutrophil survival and media were collected after an additional 24 h of culture. For studies involving Dex, Dex (0.1 to 10 µm) was added simultaneously with TNF- and replaced with fresh media after 1 h as previously described. Supernatants were collected after an additional 24 h of culture and kept frozen at 20°C until screened by ELISA or used for cell survival assays.

Neutrophil Survival Studies

Supernatants from unstimulated control (Control Supernatant) or TNF- stimulated (TNF Supernatant) nasal epithelial cells were diluted 1:3 (25%) in Dulbecco's modified Eagle's medium (DMEM) containing 10% FCS. PGE₂ was dissolved in ethanol at a stock concentration of 1 mg/ ml and was kept frozen at 20°C. Immediately before the survival assay, PGE₂ was diluted 100,000-fold in RPMI to prepare a working concentration of 10 ng/ml. Cytokines were reconstituted in saline at a stock concentration of 1 µg/ ml and diluted into culture media immediately before the assay. Purified neutrophils (5 × 10⁶/ml) were added in triplicate to 96-well culture plates that contained diluted epithelial supernatants. Control experiments with cells in culture media alone or containing GM-CSF (100 pg/ml) were assayed in parallel. For studies involving neutralizing antibodies, preliminary studies were performed to determine the optimal antibody concentrations for neutralizing the survival activity of the purified cytokine. Neutralizing antibody to GM-CSF (30 µg/ml), IL-6 (30 µg/ml), or IL-1 (combined IL-1 antibody at 40 µg/ml and IL-1 antibody at 20 µg/ml) was added to TNF--stimulated supernatants. Aliquots were removed daily and viability of neutrophils was ascertained, based on propidium iodide staining using a FACSort Flow cytometer (Becton Dickinson, Bedford, MA), and analyzed with Cellquest software.

Platelet Aggregation Assay

Human platelets were isolated from plasma anticoagulated with citrate-phosphate-dextrose (1:10 dilution). Plasma enriched for platelets was removed from the uppermost layer after centrifugation at 50 × g for 5 min. The procedure was repeated until a sufficient yield of platelets was obtained. The purity and concentration were determined by counting the platelets in a hemocytometer using Trypan blue stain. Platelet purity was > 98%, with the contaminating cells being mononuclear cells. Purified platelets were suspended at a concentration of 300 × 10⁶/ml and biologic activity of PAF was determined using a previously described assay for platelet aggregation (25).

ELISA Measurements

Anticytokine ELISA was performed according to the manufacturer's instructions. Absorbance was detected using a Titertek Multiscan ELISA reader. The sensitivity of the ELISA assays was approximately 10 pg/ml.

RT-PCR

Total RNA was isolated from 0.5 to 1 × 10⁷ cells by the modified guanidine isothiocyanate/acid-phenol method described by Chomczynski and Sacchi (26) and was resuspended in 15 µl diethylpyrocarbonate · H₂O. The concentration was measured spectrophotometrically at 260/ 280 nm. Total RNA (1 to 5 µg) was reverse transcribed with Superscript II RT and oligo (Dt)^12-18 according to the manufacturer's protocol. RT-PCR was performed in a 50-µl final volume containing: 5 µl 10× Taq reaction buffer, 29 µl ddH₂O, 3 µl of 25 mM MgCl₂, 1 µl 25 mM deoxynucleotide triphosphates, 0.2 µl Taq polymerase (5 U/ml), 2 µl of the 5' sense and 3' antisense primers (1 µm final concentration), and 10 µl of 1:10 diluted complementary DNA as previously described (27). After an initial denaturation at 94°C for 3 min, followed by 5 min of annealing at 60°C, amplification was conducted in a Thermocycler for 35 cycles under the following conditions: 1 min denaturation at 94°C, 1 min annealing at 60°C, and 2 min extension at 72°C. After amplification, PCR products were analyzed using 1.5% agarose gel electrophoresis and visualized by ethidium bromide staining.

RPA

Total RNA was isolated as described for RT-PCR. Antisense RNA probes were prepared according to the manufacturer's instructions by in vitro transcription with T7 RNA polymerase incorporating a high-specific activity [³²P]UTP using supplier-provided DNA templates. The DNA template set, hCK-2, is a manufacturer-determined set that includes the templates IL-12p35, IL-12p40, IL-10, IL-1, IL-1, IL-1Ra, IL-6, IFN-, L32, and glyceraldehyde-3-phosphate dehydrogenase. The reaction was terminated by the addition of deoxyribonuclease and labeled probe was extracted, precipitated, and diluted to 3 × 10⁶ Cherenkov counts/µl. The amount of 2 µl of the labeled probes was hybridized with 2 µg of target RNA derived from control or TNF--stimulated nasal epithelial cells in a final volume of 10 µl. The sample was heated to 90°C for 5 min, then incubated for 16 h at 56°C. Unhybridized single-stranded RNA was digested with 100 µl of RNase cocktail (2.5 ml RNase buffer, 6 µl of RNase A plus T1 mix) for 45 min at 30°C. The RNase digests were added to a proteinase K cocktail (390 µl of proteinase K buffer, 30 µl proteinase K, 30 µl yeast transfer RNA) for 15 min at 37°C. The undigested RNA was then purified by ethanol precipitation and separated on a 6% acrylamide/urea sequencing gel. The protected bands were visualized and quantified by scanning the gels using a PhosphoImager (Molecular Dynamics, Sunnyvale, CA). All results were expressed as the relative ratios of cytokine to the hL32 housekeeping gene.

Terminal Deoxynucleotidyl Transferase-Mediated Deoxyuridine Triphosphate Nick-End Labeling Technique

Briefly, neutrophils (5 × 10⁶/well) were cultured in 24-well culture plates in the presence or absence of control or TNF--stimulated supernatants, PGE₂ (1 ng/ml), IL-6 (1 ng/ ml), or GM-CSF (100 pg/ml) for 18 h. Cells were washed, fixed, permeabilized, and labeled according to manufacturer's instructions. A FACSort Flow cytometer (Becton Dickinson) was used for determining incorporation of labeled nucleotides and data were analyzed with Cellquest software.

Analysis of Apoptotic Morphology

Purified neutrophils (5 × 10⁶/ml) were added in triplicate to 96-well culture plates that contained medium, diluted epithelial supernatants, GM-CSF (100 pg/ml), IL-6 (1 ng/ ml), or PGE₂ (10 ng/ml) as described for the neutrophil survival studies. Aliquots were assayed in parallel for viability and for scoring of apoptotic morphology on Days 1, 2, and 3 by a previously described method (14). Aliquots were removed into Eppendorf tubes, washed once with EBSS, and resuspended in 6 µl of autologous plasma. A total of 3 µl of the cell suspension was used to prepare duplicate smears on glass slides. After air drying, slides were fixed in a 95% methanol solution, stained with Wright- Giemsa, and examined by light microscopy. A minimum of 400 cells on duplicate slides was scored as apoptotic versus nonapoptotic.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Conditioned Media from Airway Epithelial Cells Promotes Neutrophil Survival In Vitro

As shown in Figure 1, control supernatants from primary cultures of nasal-derived epithelial cells significantly prolonged neutrophil survival in vitro compared with media alone at Days 2 and 3 of culture. Conditioned media generated by transient exposure of epithelial cells to TNF- for 1 h followed by culture for an additional 24 h (TNF Supernatant) led to additional increases in neutrophil survival. The activity in conditioned media from TNF--primed epithelial cells exceeded the survival enhancement induced by rhGM-CSF. Increased survival of neutrophils by epithelial cell supernatants compared with media was significant at Day 2. The greatest differences in survival were observed at Day 3 (all P values < 0.0001) and this time point was used for the remainder of the survival studies.

View larger version (22K): [in this window] [in a new window]   Figure 1.   Conditioned media from cultured airway epithelial cells promotes neutrophil survival. Primary cultures of human epithelial cells were grown and stimulated as described in MATERIALS AND METHODS. Control wells included BEBM diluted 1:3 with DMEM (Media) or with added GM-CSF at the indicated concentration. Neutrophil survival was assessed by flow cytometry based on propidium iodide staining. Results are expressed as the percentage of neutrophils surviving at each time point as compared with the survival immediately after cell isolation (mean ± standard error [SE] for n = 15 experiments). P values were determined using two-tailed P values. Compared with neutrophil survival in media, survival was significant at Day 2 for GM-CSF (*P = 0.0096), for Control Supernatant (**P = 0.0149), and for TNF Supernatant (***P = 0.0004). Survival at Day 3 for neutrophils was highly significant for GM-CSF, Control, and TNF--stimulated supernatants compared with survival in media (P < 0.0001 for all).

Identification of Potential Modulatory Cytokines in Epithelial Cell Supernatants

Neutrophil longevity in inflammatory sites may be influenced by various factors that either promote survival or enhance apoptosis. A number of endogenous cytokines and eicosanoid products have been implicated in increasing neutrophil survival, including GM-CSF (9), IL-2 (12), IL-6 (13), LTB₄ (14), PAF (15), IL-1, G-CSF, and IFN- (11). Conversely, IL-10 (22) and TNF- (9, 11, 20) have been reported to induce neutrophil apoptosis. We therefore screened epithelial cells for these cytokine transcripts by RT-PCR. Of the cytokines that promote neutrophil survival, messenger RNA (mRNA) for GM-CSF, IL-1, IL-1, and IL-6 was detected. No mRNA for IL-2, IFN-, or G-CSF was detected at baseline or after TNF- stimulation. Of the cytokines that promote neutrophil apoptosis, neither IL-10 nor TNF- transcripts were detected in the presence or absence of TNF- (data not shown). Using RPA, we have previously shown that GM-CSF transcripts are inducible in nasal-derived epithelium (unpublished data). RPA was also used to confirm the presence of IL-1, IL-1, and IL-6 transcripts in nasal epithelial cells and to quantify their increases in response to TNF-. As shown in Figure 2, transcripts for IL-1, IL-1, and IL-6 were detectable in unstimulated epithelial cells. Message levels for IL-1 and IL-6 were increased 1.3- and 1.5-fold, respectively, in the presence of TNF-. Despite the addition of TNF-, there was no significant increase in mRNA for IL-1 or for IL-1Ra, a probe that was not the focus of this study.

View larger version (33K): [in this window] [in a new window]   Figure 2.   TNF- induces IL-6 and IL-1 expression in airway epithelial cells. RPA of unstimulated and TNF--stimulated primary cultures of epithelial cells is shown. Total RNA was isolated after 3 h of exposure to TNF- (25 ng/ml). A probe for human (h) L32 that detects a housekeeping gene product was used as an internal reference, and specific hL32, hIL-1, hIL-1, hIL-1 receptor antagonist (hIL-1Ra), and hIL-6 mRNA levels were measured. Protected RNA duplexes formed between the specific probes are shown for unstimulated epithelial cells (lane 1) and TNF--stimulated epithelial cells (lane 2). The specific protected bands are indicated (hL32 = 113 base pairs [bp], hIL-6 = 180 bp, hIL-1Ra = 202 bp, hIL-1 = 227 bp, and hIL-1 = 255 bp). Result shown is a representative specimen from nasal-derived epithelial cells and reveals a 1.5-fold induction of mRNA for IL-6 and a 1.3-fold induction of mRNA for IL-1 by TNF-.

In addition to cytokines, epithelial cells produce a number of other biologically active mediators that include eicosanoid products (28, 29). PAF activity in the epithelial supernatants was not detectable by a platelet aggregation assay (results not shown). As shown in Table 1, primary cultures of nasal epithelial cells constitutively produce IL-1, IL-1, IL-6, GM-CSF, and PGE₂. Except for IL-1 and IL-1, levels of epithelial-derived products were increased with TNF- stimulation. The addition of Dex during TNF- stimulation of epithelial cells suppressed levels of IL-6, GM-CSF, and PGE₂. We tested the neutrophil survival-promoting effects of each of the factors in Table 1 over broad concentration ranges. The optimal concentration for neutrophil survival enhancement for each factor was determined and the results are depicted in Figure 3. Purified recombinant human IL-1 and IL-1 enhanced neutrophil survival only at concentrations well above those present in epithelial supernatants (10 ng/ml for survival activity versus  100 pg/ml measured in supernatants; see Table 1). In contrast, GM-CSF, IL-6, and PGE₂ induced significant increases in neutrophil survival at concentrations similar to those measured in the epithelial supernatants.

View larger version (31K): [in this window] [in a new window]   Figure 3.   GM-CSF, IL-1, IL-1, IL-6, and PGE2 promote neutrophil survival. GM-CSF (100 pg/ml), IL-6 (1 ng/ml), IL-1 (10 ng/ml), IL-1 (10 ng/ml), and PGE2 (10 ng/ml) were diluted in DMEM and neutrophil survival was assessed as described in Figure 1. All significantly enhanced neutrophil survival (*P < 0.0001 for GM-CSF, IL-1, and IL-1 versus media; **P = 0.0082 for IL-6 versus media; ***P = 0.0176 for PGE2 versus media, two-tailed P values). Results shown are means ± SE for a minimum of 11 experiments.

GM-CSF and IL-6 Are the Predominant Mediators of Neutrophil Survival in Epithelial Cell Conditioned Media

In order to determine the relative contributions of GM-CSF, IL-6, IL-1, IL-1, PGE₂, and other potential epithelial cell products to neutrophil survival, we used neutralizing antibodies to GM-CSF, IL-6, and a combination of IL-1 and IL-1 antibodies. Figure 4 shows that survival enhancement was significantly diminished in TNF--stimulated supernatants by neutralizing antibody to human IL-6 or GM-CSF. The combination of IL-6 and GM-CSF antibodies completely neutralized neutrophil survival enhancement by nasal-derived supernatants and argues for a role for both cytokines in promoting neutrophil survival in vitro.

View larger version (24K): [in this window] [in a new window]   Figure 4.   Inhibition of prolonged neutrophil survival by anti-IL-6 and anti-GM-CSF antibodies. Supernatants from primary cultures of epithelial cells of nasal origin were diluted as described in MATERIALS AND METHODS. Neutralizing antibodies to GM-CSF (30 µg/ml), IL-6 (30 µg/ml), or IL-1 (combined IL-1 antibody at 40 µg/ml and IL-1 antibody at 20 µg/ml) were added to TNF- -stimulated supernatants where indicated and the survival was assessed on Day 3. The survival-enhancing activity of the supernatants was significantly decreased by antibodies to IL-6, GM-CSF, or the combination of antibodies to IL-6 and GM-CSF (*P = 0.0308, **P = 0.0011, and ***P = 0.0007, respectively; two-tailed P values). Data shown are means ± SE for five experiments.

Epithelial Supernatants, GM-CSF, and IL-6 Delay Neutrophil Apoptosis

To clarify the mechanism of enhanced neutrophil survival in the presence of epithelial supernatants and the epithelial products GM-CSF, IL-6, and PGE₂, apoptotic cell death was quantified using two different methods. DNA strand breaks induced in neutrophils undergoing apoptosis were detected by enzymatic in situ incorporation of labeled nucleotides using terminal deoxynucleotidyl transferase (TdT). The results of TdT-mediated deoxyuridine triphosphate (dUTP) nick-end labeling (TUNEL) technique for neutrophils under differing culture conditions are shown in Figure 5. Compared with neutrophils cultured in media alone, the inhibition of DNA fragmentation was greatest for control and TNF--stimulated epithelial supernatants, followed by rhGM-CSF and rhIL-6. PGE₂ showed only modest effects on delaying neutrophil apoptosis.

View larger version (31K): [in this window] [in a new window]   Figure 5.   Delay in apoptotic cell death in cultured neutrophils. Purified neutrophils were cultured in 24-well plates for 24 h with the indicated stimulants and DNA strand breaks were assessed by the TUNEL technique as described in MATERIALS AND METHODS. Representative histograms from one of four experiments are shown. Fluorescence intensity reflects incorporation of fluorescent dUTP into neutrophils undergoing apoptosis and is displayed in arbitrary units (AU) on the x axis versus the number of events counted on the y axis. The negative control contains label but lacks TdT and reflects background fluorescence. Freshly isolated neutrophils in media alone (media, 0 h) show 5.2% of cells to be apoptotic, compared with 78.1% of neutrophils cultured in media after 24 h. In contrast, after 24 h of culture, the percentages of apoptotic neutrophils cultured with GM-CSF (100 pg/ml), control supernatants, TNF--stimulated supernatants, IL-6 (1 ng/ ml), and PGE2 (10 ng/ml) were 48.4, 25.4, 23.6, 47.1, and 60.3, respectively.

The morphologic changes of neutrophils undergoing apoptosis in the presence of epithelial supernatants and the epithelial products GM-CSF, IL-6, and PGE₂ were scored and the results are presented in Table 2, with representative photomicrographs depicted in Figure 6. The results in Table 2 confirm the results of the TUNEL assay and reveal that the number of neutrophils exhibiting apoptotic morphology was significantly increased for neutrophils cultured in media alone versus the epithelial supernatants or GM-CSF as early as Day 1. Significant delays in the appearance of morphologic changes of apoptosis were detectable at Day 1 for IL-6 and PGE₂, although the effects were more modest than the effects of GM-CSF or the epithelial supernatants. At Days 2 and 3, epithelial supernatants, GM-CSF, IL-6, and PGE₂ all showed significantly fewer cells with apoptotic morphology than did neutrophils cultured in media alone.

View larger version (96K): [in this window] [in a new window]   Figure 6.   Morphologic features of neutrophils cultured in the presence or absence of epithelial- derived factors. Neutrophils were prepared and stained with Wright- Giemsa (×1,000) as described in MATERIALS AND METHODS after incubation in vitro for up to 72 h. Panels 1-4 show neutrophils cultured in media alone for (1) 0, (2) 24, (3) 48, and (4) 72 h. Panels 5-9 show neutrophils maintained in the presence of (5) GM-CSF (100 pg/ml), (6) control supernatants, (7) TNF--stimulated supernatants, (8) IL-6 (1 ng/ml), and (9) PGE2 (10 ng/ml). Panel 1 (media, 0 h) demonstrates normal neutrophil morphology with polysegmented nuclei. Panel 2 (media, 24 h) reveals that the majority of neutrophils exhibit normal morphology, but some (35%) have features of apoptosis. Panels 3 and 4 (media, 48 and 72 h, respectively) reveal progressive rounding of cells, chromatin condensation, and coalescence of the nuclear lobes. In addition, numerous ghost cells with an absence of nuclear staining are apparent in panel 4. Panels 5 (GM-CSF), 6 (control supernatant), and 7 (TNF--stimulated supernatant) reveal that about half of the neutrophils continue to exhibit normal morphology and the remainder have features of apoptosis similar to panels 3 and 4. Panels 8 and 9 (IL-6 and PGE2, respectively) reveal only a minority of neutrophils with normal morphology and increasing numbers of cells with features of apoptosis.

The photomicrographs in Figure 6 depict neutrophils cultured in media alone, in the presence of epithelial supernatants, or in the presence of the epithelial products GM-CSF, IL-6, and PGE₂ for up to 72 h. Control neutrophils in media alone (Figure 6, panels 1-4) progressively developed reduction in cell volume, condensation of the nucleus, formation of apoptotic bodies, and vacuolation of the cytoplasm, beginning as early as Day 1. In contrast, the morphology of many of the neutrophils maintained in the presence of GM-CSF ( panel 5), control supernatants ( panel 6 ), or TNF--stimulated supernatants ( panel 7 ) failed to show these changes. The effect of PGE₂ ( panel 9) on delaying the development of apoptotic morphology was significant, but modest compared with the effects of the epithelial supernatants or GM-CSF, whereas the effect of IL-6 ( panel 8) was intermediate between the effect of PGE₂ and that of the supernatants or GM-CSF.

Regulatory Effects of Glucocorticoids on Epithelial Cell Enhanced Neutrophil Survival

Treatment of eosinophil-predominant disorders such as allergic rhinitis and asthma with topical or systemic glucocorticoids is extremely effective. It has been shown that glucocorticoids counteract the survival effects of eosinophil growth factors (30) while promoting neutrophil survival (16). In addition, glucocorticoids regulate epithelial cell cytokine production. As depicted in Table 1, the addition of the potent glucocorticoid Dex to TNF--stimulated epithelial cells led to a dose-dependent decline in IL-6, GM-CSF, and PGE₂ levels, but no change in the levels of IL-1 and IL-1.

In vivo Dex may alter neutrophil survival by at least two mechanisms: (1) effects on the neutrophil directly, and (2) indirect effects related to suppression of cytokine and eicosanoid growth factor production by epithelial cells. To determine the contribution of Dex to neutrophil survival in the airways, we first examined the direct effects of Dex on neutrophil survival. As depicted in Figure 7A, the survival rates of neutrophils cultured for 3 d in the presence of 10 and 1 µm Dex were similar (30.4 and 28.6%, respectively) and were significantly greater than in media alone (16.7%). The direct effects of Dex on neutrophil survival were significantly less than the effect of conditioned media from TNF--primed epithelium (30.4 versus 41.0%, respectively). Combined treatment of neutrophils with both epithelial cell supernatants and Dex did not increase survival beyond levels induced by epithelial cell supernatants alone (41.0 versus 42.3%, respectively); hence, the combined effects of cytokine-mediated and glucocorticoid-mediated neutrophil survival were not additive.

View larger version (20K): [in this window] [in a new window]   View larger version (24K): [in this window] [in a new window]   Figure 7.   (A) Dex enhances neutrophil survival. Indicated concentrations of Dex were diluted in media consisting of 1 part BEBM to 3 parts DMEM. All epithelial cell cultures (Supernatant) were stimulated with TNF- and diluted 1:3 in DMEM. Dex alone, at either 1 or 10 µm, enhanced neutrophil survival compared with survival in media (*P = 0.0013, two-tailed P values). The effect of 10 µm Dex on survival was significantly less than that of epithelial cell supernatants generated in the presence of TNF- (**P = 0.0312 for 10 µm Dex versus Supernatant, two-tailed P values). The combination of 10 µm Dex and TNF--stimulated epithelial cell supernatants (Supernatant plus 10 µm Dex) failed to enhance neutrophil survival compared with TNF--stimulated epithelial cell supernatants alone (P = 0.7038 for Supernatant plus 10 µm Dex versus Supernatant, ns), but significantly enhanced neutrophil survival compared with the effect of 10 µm Dex alone (***P = 0.0133 for Supernatant plus 10 µm Dex versus 10 µm Dex, two-tailed P values). Data are means ± SE for nine experiments. (B) Dex inhibits epithelial prolongation of neutrophil survival. Epithelial cell supernatants were generated following transient TNF- stimulation in the presence or absence of Dex. Dex (10 µm) had no effect on neutrophil survival when added after harvesting the TNF--stimulated epithelial cell supernatants (Supernatant plus 10 µm Dex), but showed a dose-dependent inhibition of epithelial cell-mediated survival when present during TNF- stimulation (Dex-treated Epi). A significant decrease in neutrophil survival compared with survival in TNF--stimulated supernatants was observed only at the concentration of 10 µm Dex (*P = 0.0043 for Dex-treated Epi, 10 µm, versus Supernatant, two-tailed P values). Data are means ± SE for nine experiments.

In contrast to the direct survival-enhancing effects of Dex on neutrophils, Dex inhibits the generation of survival-enhancing factors by stimulated epithelial cells. We examined neutrophil survival in conditioned media from epithelial cell cultures obtained by adding TNF- and varying concentrations of Dex simultaneously to the medium. These results are depicted in Figure 7B. The presence of Dex during generation of epithelial cell supernatants diminished neutrophil survival in a dose-dependent manner compared with survival in conditioned media generated in the absence of Dex (28.8 versus 42.3%). The simultaneous addition of TNF- and 10 µm Dex led to epithelial supernatants with significantly reduced neutrophil survival activity. However, even in this latter supernatant, neutrophil survival activity was comparable to that of 10 µm Dex alone (28.8 and 30.4%, respectively).

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This study demonstrates that bioactive products that promote neutrophil longevity in vitro are present in supernatants derived from primary cultures of epithelial origin (Figures 1-6, Tables 1 and 2). We also demonstrated (Figures 2 and 3 and Table 1) that significant amounts of GM-CSF, IL-6, and PGE₂ are produced in epithelial cell cultures. TNF- increased both the transcription and release of biologically active GM-CSF, IL-6, and PGE₂ from cultured airway epithelial cells. The increased levels of GM-CSF, IL-6, and PGE₂ in TNF--stimulated supernatants correlated with increased neutrophil survival as compared with survival of neutrophils in unstimulated supernatants (Tables 1 and 2 and Figures 1-6). Further, neutralizing antibodies to IL-6, GM-CSF, or the combination of IL-6 and GM-CSF significantly reduced neutrophil survival in nasal-derived epithelial supernatants (Figure 4).

Importantly, TNF- and IL-10, factors previously identified as signals contributing to neutrophil apoptosis, were not produced by epithelial cells. Analysis of apoptotic morphology and of DNA strand breaks that characterize apoptosis confirmed that epithelial cell products inhibited neutrophil apoptosis and that cell death was not simply an artifact of necrosis under these culture conditions (Figures 5 and 6 and Table 2). Therefore, these results confirm previous reports that neutrophil survival is prolonged in the presence of epithelial cell supernatants (31) and that epithelial cell production of GM-CSF has a major role in inhibiting neutrophil apoptosis. Our findings extend previous observations by demonstrating that IL-6 and PGE₂ also delay neutrophil apoptosis (Figures 3-6 and Table 2) and that concentrations of PGE₂ capable of promoting neutrophil survival are present in epithelial cell conditioned media (Table 1 and Figure 3).

This study also confirms the time course of apoptotic events in neutrophils in relation to neutrophil viability, as has been elegantly described in previous reports (14, 18). The biochemical events that occurred in cells undergoing apoptosis preceded the loss of viability as detected by vital stains such as propidium iodide. Differences in apoptotic cell death were detectable in neutrophils at the earliest time point examined, even though the differences detected in neutrophil survival were not statistically significant among any groups at Day 1 (Figures 1, 5, and 6 and Table 2). Thus, the various methods utilized have differing sensitivities for detecting the stages of neutrophil apoptosis, with dye exclusion being the least sensitive indicator of the ability of epithelial-derived factors to delay apoptosis.

The clinical effectiveness of glucocorticoids in the treatment of eosinophil-predominant disorders such as allergic rhinitis and asthma is well established. The predominance of neutrophils in asthmatic airways despite glucocorticoid treatment or in viral respiratory infections of allergic patients treated with corticosteroids has not been well explained. Glucocorticoids have been shown to inhibit transcriptional activation through effects on transcription factors such as nuclear factor B (NFB) (32, 33). TNF- and other stimuli upregulate NFB and glucocorticoids inhibit this upregulation by inducing the cytoplasmic inhibitor of NFB (IB). Binding of NFB to IB abolishes transcriptional activation induced by inflammatory stimuli (34). Consequently, glucocorticoids suppress stimulated epithelial cell cytokine production as shown in Table 1. Alternatively, glucocorticoids have themselves been shown to delay neutrophil apoptosis (16). Thus, in vivo Dex may alter neutrophil survival by exerting direct effects on the neutrophil, and by suppression of cytokine and eicosanoid growth factor production by epithelial cells.

Our results depicted in Figures 7A and 7B suggest that effects on neutrophil survival mediated by cytokines and by glucocorticoids were not additive. Further, the suppressive effects of Dex on epithelial generation of neutrophil survival factors became significant only at higher doses of Dex. Although glucocorticoid-induced downregulation of GM-CSF, IL-6, and PGE₂ production by TNF--stimulated epithelium resulted in increased neutrophil death rates, neutrophil survival remained enhanced compared with survival in media alone. Therefore, the persistence of neutrophils in airway epithelium despite glucocorticoid treatment does not necessarily imply that inflamed epithelium was resistant to the actions of glucocorticoids. Neutrophil survival may be attributable to the direct effects of glucocorticoids on the neutrophils themselves, rather than to the failure of glucocorticoids to suppress epithelial cytokine production.

In summary, our in vitro model of airway inflammation suggests that epithelial products may significantly regulate neutrophil survival in the airways. Epithelial-derived IL-6 delayed neutrophil apoptosis in a manner similar to GM-CSF. Importantly, high-dose Dex treatment of epithelial cells profoundly suppressed the production of GM-CSF, IL-6, and PGE₂. We report a novel observation that despite glucocorticoid suppression of epithelial-derived survival factors, neutrophil survival was prolonged. The enhanced survival despite suppression of epithelial products was attributable to a direct effect of Dex on neutrophil apoptosis. The balance between the opposing effects of glucocorticoid downregulation of epithelial-derived neutrophil growth factors and the direct effects of delaying neutrophil apoptosis may be crucial in regulating neutrophil survival and ongoing airway inflammation.