Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

The Fas antigen (CD95, APO-1) belongs to a conserved family of membrane receptors known as the tumor necrosis factor receptor or TNFR family (1). Fas ligand (FasL) exists as a soluble and cell-associated molecule that engages the Fas antigen and initiates programmed cell death (2). Fas and FasL are expressed by mononuclear cells, including T lymphocytes, natural killer T cells, and macrophages. Expression of both factors is essential for continually maintaining peripheral cell populations (3) and critical in downregulating the immune response (4). Imbalance in Fas signaling is also associated with human disease. Insufficient Fas-mediated apoptosis is attributed to lymphoproliferative disorders, (5) whereas exaggerated apoptosis is associated with pathologic cell loss (6).

Parenchymal cells also express Fas antigen, demonstrating that Fas may be critical for maintaining tissue cell populations. In particular, Fas is constitutively expressed on alveolar epithelial type II cells, the putative stem cell of the alveolus (7). During lung development and homeostasis, type II cells divide and differentiate into type I alveolar epithelial cells, cells that comprise the majority of surface area and are intimately involved in gas exchange (8, 9). This dynamic process occurs in a highly regulated fashion to maintain normal tissue architecture and function. Recent evidence suggests that modulation of Fas-FasL interactions are critical in the death and survival of this cell population and influenced in a paracrine manner by inflammatory mediators (7, 10). In support of this, topical administration of FasL to the lung accelerates apoptosis of alveolar epithelial cells and promotes pulmonary fibrosis in mice (11). In a similar model, pulmonary fibrosis does not occur if the Fas-FasL interaction is blocked prior to the insult (12). In humans, Fas-FasL expression is elevated in lung tissue obtained from patients with interstitial pulmonary fibrosis (IPF) (13) and elevation of soluble FasL in bronchoalveolar lavage fluid has recently been identified as a predictor of disease severity in adults with acute respiratory distress syndrome (ARDS) (14). These combined findings strongly suggest an important role for Fas-FasL interactions in the lower airway during dynamic tissue remodeling

The paracrine influence on Fas-FasL signaling by inflammatory mediators is well documented in monocytic cell populations; however, regulation is variable and cell specific (15). By directly affecting Fas-FasL expression and function, we hypothesize that proinflammatory cytokines present in the alveolar microenvironment orchestrate the parenchymal cell response to Fas-FasL and, in turn, affect tissue integrity and function (16). We further propose that the cumulative environmental impact on alveolar epithelial cell Fas regulation and function is critical for homeostatic as well as pathologic remodeling.

In this study, the A549 human lung epithelial cell line was used as a model to examine cytokine influence on Fas-mediated apopotosis. Our results reveal that two proinflammatory cytokines important to the alveolar microenvironment, interleukin-1 (IL-1) and interferon- (IFN-), independently had opposing effects on Fas-mediated apoptosis. IFN- clearly promoted epithelial cell death, whereas IL-1 prevented propagation of the Fas-mediated death signal. Our results establish a biochemical framework to dissect the role of cytokine influence on epithelial cell turnover in the alveolar airway environment and may assist in identifying host factors that distinguish normal from pathologic tissue remodeling.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Cell Culture

A549 cells, a human lung epithelial cell line, were purchased from the American Type Culture Collection (Rockville, MD) (17). Cells were cultured under standard conditions in F-12 nutrient mixture (HAM) (Gibco BRL, Grand Island, NY) supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, Utah), 100 U/ml penicillin and 100 µg/ml streptomycin (Gibco BRL) (complete growth medium). A549 cells were passaged at 80-90% confluence using 0.25% trypsin and 1 mM ethylenediaminetetraacetic acid (EDTA) · 4Na (Gibco BRL).

Treatment with IL-1, IFN-, and Anti-Fas Crosslinking Antibody

A549 cells were plated at 80-90% confluence in either 25 cm² tissue culture flasks (Falcon/Fisher, Pittsburgh, PA) or 8-well glass chamber slides (Fisher). Cells were incubated for 2 d after passage before further manipulation to allow cells to reach full confluence and a native resting state. Medium was removed and replaced alone or containing either IL-1 (50 pg/ml) (Biological Response Modifiers Program, National Cancer Institute, Frederick, MD) or IFN- (250 U/ml) (BioSource International, Camarillo, CA). After 24 h, cells were given a second stimulus of either a Fas-crosslinking antibody (clone CH-11; Kamiya Biomedical Company, Seattle, WA) (10-100 ng/ml) or an isotype control (IgM; Sigma, St. Louis, MO) (10-100 ng/ml). Cells were incubated an additional 4-24 h for immunohistochemistry assays or generation of cell extracts.

Treatment with Caspase and IL-1 Signaling Inhibitors

A549 cells were stimulated as described above; however, 1 h before the addition of the Fas-crosslinking antibody, cells were treated with 100 µM of either Z-LETD-FMK (caspase 8 inhibitor) or Z-DEVD-FMK (caspases 3, 6, 7, 8, and 10 inhibitor) (Enzyme Systems Products, Livermore, CA) (18). A549 cells were also stimulated as described above; however, 30 min before the initial stimulus with IL-1 or IFN-, cells were given either the phosphatidyl inositol 3 kinase inhibitor, LY294002 (10 µM) (CalBiochem, San Diego, CA), or IL-1 receptor antagonist protein, IL-1Ra, (1,000-fold molar excess relative to IL-1 dose) (kind gift from Dr. Daniel Tracey, Pharmacia-UpJohn, Kalamazoo, MI). After 24 h, cells were given a second dose of the above inhibitors at half the original dose, and 30 min later cells were stimulated with the Fas-crosslinking antibody as described above.

Cell Extracts

Following treatment as described above, adherent cells were trypsinized and pooled with nonadherent cells present in the original medium. Cells were then pelleted at 300 × g for 10 min at room temperature, supernatants were discarded, and the cell pellet was resuspended in 200 µl of 1× KPM complete buffer (1× KPM [2× KPM = 100 mM PIPES, pH 7; 100 mM KCl; 20 mM EGTA, and 3.84 mM MgCl₂], 1 mM dithiothreitol (DTT), 0.1 mM PMSF, 10 µg/ml cytochalasin B, 2 µg/ml chymostatin, 2 µg/ml leupeptin, 2 µg/ml antipain, and 2 µg/ml pepstatain A). Resuspended cells were transferred to 0.5 ml Eppendorf tubes and pelleted again (3,000 rpm for 10 min). Cell pellets were resuspended in 15 µl of 1× KPM complete buffer, frozen in liquid nitrogen, placed in a room temperature water bath until cells had thawed, and then vortexed. The freeze-thaw procedure was performed a total of four times. Samples were then centrifuged at 14,000 rpm for 20 min at 4°C. The resulting supernatant or cell extract was removed to a fresh 0.5 ml Eppendorf tube and an aliquot was analyzed for protein concentration by the Bradford method (Bio-Rad, Hercules, CA) frozen in liquid nitrogen, and stored at 70°C.

Enzymatic Caspase Activity Measured with Amino Trifluoromethyl Coumarin

For all amino trifluoromethyl coumarin (AFC) preparations, cells (3 × 10⁶) were collected by centrifugation, washed with KPM buffer, and lysed by four cycles of freeze-thawing as previously described. The presence of active caspases was determined by AFC assay using a specific fluoro-substrate as previously described. Lysates were incubated with DEVD-AFC in a cyto-buffer (10% glycerol, 50 mM PIPES, pH 7.0, 1 mM EDTA) containing 1 mM DTT and 20 µM DEVD-AFC (Enzyme Systems Products). Extracts were also incubated with YVAD-AFC (Enzyme Systems Products) in a YVAD cyto-buffer (10% sucrose, 100 mM Hepes, pH 7.5, 0.1% CHAPS, 10 mM DTT). Specifically, 20 µM YVAD-AFC was added to lysates and incubated for 45 min at room temperature before measurement. Standard recombinant caspase-1 was a gift from Nancy Thornberry, Merck Research Laboratory, Rahway, NJ. In both instances release of free AFC was determined using a Cytofluor 4000 fluorimeter (filters: excitation; 400 nm, emission; 505 nm; Perseptive Company, Framingham, MA).

Western Analysis

Cell extracts (50-100 µg) were denatured by boiling and analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and compared with extracts generated from whole-cell THP1 extracts (ATCC, Rockville, MD) that were activated in a cell-free system and known to contain active caspases as a positive control. All blots were developed using the enhanced chemiluminescence Western detection kit according to manufacturer's guidelines (Amersham/Pharmacia Biotech, Piscataway, NJ). Immunoblots were scanned and densitometry was performed using a Gel-Doc 2000 system and Quantity One software version 4.0.3 (Bio-Rad, Hercules, CA).

Caspase 3, Bcl-x_L, and poly(ADP-ribose) polymeraseExtracts for caspase 3 and Bcl-x_L analysis were separated on 15% SDS-PAGE gels and extracts for poly(ADP-ribose) polymerase (PARP) analysis were separated on 12% SDS-PAGE gels. Proteins were transferred to polyvinyldifluoride membranes. All membranes was blocked for 1 h in Tris-buffered saline with 0.1% Tween-20 (TBS-T) and 5% nonfat milk at room temperature. Membranes were incubated in their respective primary antibody (caspase 3 = a polyclonal rabbit anti-caspase-3, diluted 1:1,000; PharMingen International, San Diego, CA; Bcl-x_L = a polyclonal rabbit anti-Bcl-X, diluted 1:1,000; Transduction Laboratories, San Diego, CA; and PARP = monoclonal mouse anti-PARP, diluted 1:1,000; PharMingen International) diluted in TBS-T containing 2.5% nonfat milk overnight at 4°C. The HRP-labeled anti-rabbit or anti-mouse antibody (Amersham/Pharmacia Biotech) was used at a 1:10,000 dilution in TBS-T with 2.5% nonfat milk and incubated for 1 h at room temperature.

Caspases 6 and 8. Proteins were transferred to nitrocellulose membrane (Amersham Life Science). Primary antibodies against each caspase were purchased from Upstate Biotechnology (Lake Placid, NY), and the Western blotting protocol provided with the antibodies was followed. Specifically, the caspase 6 antibody was used at a 1:500 dilution, and the caspase 8 antibody was used at a 1:1,000 dilution.

Caspase 7. Proteins were separated on a 15% SDS-PAGE gel and transferred to polyvinyldifluoride membrane via standard wet transfer. Membranes were blocked by dipping in methanol and drying in a vacuum oven at 80°C for 30 min. The primary antibody was used at a 1:1,000 dilution in PT-MT buffer (1× PT buffer [20 mM Tris base, pH 7.4, and 150 mM NaCl], 3% nonfat milk, 2% bovine serum albumin, and 0.05% Tween-20). Membranes were incubated in the primary antibody for 1 h at room temperature. Membranes were incubated with the secondary antibody, HRP-labeled anti-mouse at a 1:5000 dilution in PT-MT buffer, for 1 hat room temperature. All washes were done in large volumes of 1× PT buffer plus 0.05% Tween-20.

Immunohistochemistry

A549 cells were stimulated as described above. Cells were then rinsed twice with 1× phosphate-buffered saline (PBS) and fixed in ice-cold pure methanol for 30 min at 20°C. After washing twice with washing buffer (1× PBS and 0.1% Tween 20), wells were blocked with blocking buffer (1× PBS, 1% bovine serum albumin, and 0.1% Tween 20) for 10 min. Cells were then incubated with the M30 CytoDEATH antibody (Boehringer Mannheim, Indianapolis, IN), diluted 1:10 in blocking buffer for 1 h at room temperature. Cells were washed twice with washing buffer and then incubated with 10 µg/ml anti-mouse-Ig-fluorescein (Boehringer Mannheim) for 30 min at room temperature in the dark. Cells were again washed twice with washing buffer before incubation with 0.5 µg/ml 4',6-Diamidine-2'-phenylindole dihyrochloride (DAPI) (Roche Molecular Biochemicals, Indianapolis, IN) for 5 min at room temperature in the dark. Cells were washed three times with washing buffer and allowed to air dry slightly. Antifade (Molecular Probes, Eugene, Oregon) was added to each slide before addition of cover slip. Cells were considered to be in the early stages of apoptosis if caspase cleaved CK-18 appeared in the cytoplasm without concomitant changes in the nucleus and in later stages of apoptosis only if colocalization of cytoplasmic and nuclear changes occurred (19, 20). In addition, an antibody that specifically recognizes native CK-18 was substituted for the M30 antibody and used in the same staining procedure as described above at a concentration of 14 µg/ml. Specificity of the CK5 antibody was established by comparison to an isotype control antibody MOPC21 (Sigma).

Flow Cytometry

A549 cells were plated in 25 cm² tissue culture flasks (3 × 10⁵ cells/flask) in complete growth medium. After 48 h, cells were stimulated with IL-1 (50 pg/ml) or IFN- (250 U/ml) or a combination or both and incubated for another 24 h. Cells were resuspended with nonezymatic cell dissociation solution (Sigma) at 1 × 10⁵ cells/ml and incubated with a monoclonal PE-conjugated anti-human CD95 antibody or matched PE-conjugated isotype control (PharMingen) for 30 min at 4°C. Cells were thoroughly washed three times in PBS, pH 7.4, containing 10% FBS, resuspended at the same density and immediately counted using a FACSCaliber Flow Cytometer (Becton Dickinson, Franklin Lakes, NJ).

Akt In Vitro Kinase Assay

A549 cells were treated as previously described and then lysed in 1 ml ice-cold buffer (50 mM Tris-HCl pH 7.5, 0.1% Triton X-100, 1mM EDTA pH 8.3, 1mM EGTA, pH 8.0, 50 mM NaF, 10 mM B-glycerophosphate, 5 mM sodium pyrophosphate, 10 mg/ml aprotinin, 10 mg/ml leupeptin, 1 mM sodium vanadate, and 14.2 mM  mercaptoethanol) for 15 min. Nuclei were removed by centrifugation and then cytoplasmic extracts were immunoprecipitated with an Akt-specific antibody (Santa Cruz Biotechnology Inc.). Immune complexes were collected with protein G agarose beads (Gibco BRL) at 4°C for 2 h. The immunoprecipitates were then removed from beads in cold lysis buffer and measured for Akt kinase activity using Histone 2B as previously described (21), or glycogen synthase-3 peptide (GSK-3; Cell Signaling Technology, Beverly, MA) as substrate. Proteins and peptides were separated on a 15% SDS-polyacrylamide gel, then transferred onto nitrocellulose membranes. The lower portion of the membrane was immunoblotted with phospho-GSK3 or Histone 2B-specific antibody and then subject to autoradiography. The upper portion of the membrane was immunoblotted with an antibody that recognizes total Akt to confirm equal loading and standardize results.

Statistical Analysis

All data were expressed as mean ± SEM. Paired t tests were used for single comparisons (Microsoft Excel; Microsoft, Redmond, WA). For comparisons that involved multiple variables and observations, two- and three-way analysis of variance (JMP; SAS Institute, Cary, NC) was used. Having passed statistical significance by ANOVA, individual comparisons were made using the contrast method. Statistical significance was defined as a P value < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of IL-1 and IFN- on Fas-Mediated Caspase Activation

We reasoned that Fas-mediated apoptosis of alveolar epithelial cells is regulated by cytokines present in the alveolar microenvironment. We first explored the effect of IL-1 (50 pg/ml), IFN- (250 U/ml), and a Fas-crosslinking antibody (FasAb) (10 ng/ml) on caspase activity in whole-cell lysates prepared from A549 cells, as a function of apoptosis. Substrates for caspase-1-like activity (YVAD-AFC) and caspase-3-like activity (DEVD-AFC) were used to identify modulation of "initiator" and "executioner" caspase activity, respectively (18). Native nonapoptotic-appearing cells had a high constitutive level of caspase-1- and -3-like activity. The level of activity was approximately 10-fold higher than the activity found in extracts taken from native THP-1 cells, a monocyte-like cell line (data not shown). The level of constitutive caspase-1-like activity did not significantly change following treatment with IL-1, IFN-, FasAb, or combinations of these factors (Figure 1A). In contrast, the level of caspase-3-like activity increased over 2-fold following combined treatment with IFN- and FasAb; however, treatment with individual factors, including FasAb alone or a combination of IL-1 and FasAb, had minimal to no effect (Figure 1B).

View larger version (16K): [in this window] [in a new window]   Figure 1.   Caspase activity is modified by cytokines and Fas activation. (A) Caspase-1-like activity was measured in whole cell extracts following treatment with IL-1 (50 pg/ml) or IFN- (250 U/ml). After 24 h cells were given a Fas-crosslinking antibody (100 ng/ml) or isotype control (100 ng/ml). After an additional 24 h, extracts were generated and caspase activity was determined using YVAD-AFC as substrate as described in MATERIALS AND METHODS. (B) The same procedure was repeated with the same extracts using DEVD-AFC as substrate to measure caspase-3-like activity. Data is from a single experiment with triplicate samples and representative of five other experiments (*P < 0.001).

Morphometric Analysis of Apoptosis

Caspase processing of a key structural intermediate filament protein unique to epithelial cells, cytokeratin 18 (CK18), was used to identify cytoplasmic alterations. CK18 has recently been identified as a key substrate for caspases 3, 6, and/or 7 during epithelial cell apoptosis (19, 20). Nuclear condensation and fragmentation, characteristic of apoptosis, was detected concomitantly using conventional DAPI staining. The morphologic alterations during apoptosis followed a consistent pattern. In all cases, detection of caspase cleaved CK18 either preceded (Figure 2B) or occurred in tandem with nuclear condensation (Figure 2C), consistent with cytoplasmic caspase cleaving events being initiated before nuclear alterations. In accord with caspase activity, activation of Fas alone increased the frequency of apoptosis but was substantially less (< 2.85-fold) than the frequency of apoptosis in cells treated with a combination of IFN-/FasAb (Figure 2D). Pretreatment with IL-1 before FasAb decreased the frequency of apoptotic events to background levels. This provided the first evidence demonstrating opposing cytokine effects on Fas-mediated apoptosis. The TUNEL method was also used to verify results obtained from CK-18 immunostaining. Virtually identical results were obtained for each treatment condition, thereby confirming the validity of our previous results (data not shown).

View larger version (141K): [in this window] [in a new window]   Figure 2.   Differential effect of cytokines on Fas-induced apoptotic structural changes. A549 cells were treated as described in Figure 1, fixed, and then concomitantly stained with M30 antibody that only recognizes caspase-cleaved cytokeratin 18 (CK-18) in the cytoplasm, and also with DAPI to identify nuclear fragmentation/condensation. (A) Native untreated cells do not demonstrate CK-18 cleavage in the cytoplasm and present with normal nuclei (in red). (B) Treatment with Fas-crosslinking antibody (FasAb) alone induces morphometric changes consistent with apoptosis. The arrows identify caspase-cleaved CK-18 (in green) in the cytoplasm that precedes significant alteration of nuclear structure. (C) Treatment with a combination of IFN- and FasAb induces a higher degree of apoptosis. The arrows indicate cells in "late stage" apoptosis as identified by concomitant homogenous staining of the cytoplasm for caspase-cleaved CK-18 and fragmentation/condensation of nuclei. (D) The number of "late stage" apoptotic cells were counted in each treatment group and reported in comparison with cells treated with a combination of IFN- and FasAb. Data is from a single experiment and representative of eight different experiments (*P < 0.05).

Bcl-x_L and PARP Cleavage

We next turned our attention to Bcl-x_L and PARP, key intracellular proteins that are susceptible to caspase degradation during the execution phase of Fas-mediated programmed cell death (22, 23). Consistent with previous studies, the active 26-kD Bcl-x_L protein was constitutively present in A549 epithelial cells, whereas Bcl-2, a related family member, was not found (data not shown) (24). Most striking was the > 75% reduction in 26-kD Bcl-x_L expression following treatment with IFN-/FasAb (Figure 3a). Similar to Bcl-x_L expression, a substantial decrease in functional p110 PARP occurred in cells treated with the combination of IFN-/FasAb (Figure 3B). Detection of lower molecular weight fragments indicative of caspase cleavage were not found. Conversely, a modest increase (~ 20%) in Bcl-x_L expression consistently occurred in cells exposed to IL-1 before Fas activation (Figure 3A). This line of investigation provides additional evidence that cytokines differentially influence the expression of key functional proteins involved with cell survival during the Fas-mediated apoptotic cascade.

View larger version (20K): [in this window] [in a new window]   Figure 3.   Modified expression of Bcl-xL and PARP by cytokines during Fas-mediated apoptosis. (A) Cells were exposed to the same treatment conditions and then whole cell extracts were generated and subjected to Western analysis with an antibody that recognizes mature 26 kD Bcl-xL protein (*P < 0.05). (B) Similar experiments were performed and Western analysis was performed with an antibody that recognizes mature 110 kD PARP. Differences in protein expression were determined by densitometric analysis for both proteins and are presented as fold change in comparison with native cells. Data is from a single experiment and representative of seven different experiments.

Effect of IL-1 and IFN- on Caspases Involved in Fas-Mediated Apoptosis

Activation of the apoptotic cascade following Fas-FasL interaction includes activation of the "initiator" caspase-8 followed by activation of "executioner" caspases including 3, 6, and 7 (15). We used Western analysis to record differences in caspase activation as a function of FasAb, IL-1, and IFN-. We reasoned that caspase activation is regulated by IL-1 and IFN- during Fas-mediated cell death and that IL-1 may prevent activation of specific caspases involved in this pathway. As predicted, treatment with FasAb alone induced the activation of caspase-8 and this activity was increased 2-fold by prior treatment with IFN- (Figure 4A). Treatment with IL-1 before FasAb completely prevented the ability to detect active caspase-8, suggesting that IL-1 induces an anti-apoptotic effect early in the Fas-FasL signaling cascade. Activation of the executioner caspases 3 and 7 only occurred in cells treated with a combination of IFN- and FasAb, and surprisingly was never detected in cells treated with FasAb alone (Figure 4B). We were not able to detect activation of caspase 6, suggesting that Fas-mediated death is not caspase-6 dependent in A549 lung epithelial cells. Under all conditions studied, the inactive precursor form of caspases-3, 6, 7, and 8 were present in cell lysates (data not shown).

View larger version (21K): [in this window] [in a new window]   Figure 4.   Detection of active caspase conversion following treatment with cytokines and Fas activation. (A) Cells were treated as previously described and 50 µg of whole cell lysate was analyzed via Western analysis with an antibody that detects the active (and inactive) initiator caspase-8 protein. As a positive control A549 lysates were compared with activated S-100 extracts from the mononuclear cell line, THP-1, known to contain the active caspase protein. (B) Aliquots of the same extracts were individually analyzed via Western analysis with antibodies that detect the active and precursor species for caspases 3, 6, and 7. Only detection of the active intermediates are presented in A and B. (C) A549 cells were treated with combinations of cytokines and FasAb as described above; however, 1 h before the addition of the Fas-crosslinking antibody, cells were treated with 100 µM of either Z-LETD-FMK (caspase 8 inhibitor) or Z-DEVD-FMK (caspase 3, 6, 7, 8 and 10 inhibitor). Twenty-four hours later the cells were fixed and stained with the M30 antibody and DAPI. The number of "late stage" apoptotic cells were counted in each treatment group and reported in comparison with cells treated with a combination of IFN- and FasAb. Data is from a single experiment and representative of three different experiments.

To confirm the functional requirement of caspase 3, 7, and 8 activation during Fas-mediated death in epithelial cells, similar experiments were performed with the irreversible caspase inhibitors Z-LETD-FMK and Z-DEVD-FMK. Inhibition of caspase-8 activity with Z-LETD-FMK resulted in almost complete inhibition of CK-18 cleavage and nuclear condensation in cells receiving IFN-/FasAb (Figure 4c). Inhibition of executioner caspase activity with Z-DEVD-FMK also resulted in a significant decrease in apoptosis.

Fas Expression on A549 Cells

We hypothesized that IFN- and IL-1 influence Fas receptor expression in opposite directions on the surface of lung epithelial cells and that this may explain differences in caspase-8 activation. Consistent with previous investigations, Fas is constitutively expressed on the surface of A549 cells (Figure 5) (7). Treatment with IFN- alone resulted in a significant increase in Fas receptor expression. In contrast, IL-1 treatment had no effect on basal surface Fas expression resulting in expression similar to native cells. Therefore, IL-1 does not prevent activation of caspase-8 through downmodulation of Fas receptor expression. Concomitant treatment with IL-1 and IFN- resulted in an increase in Fas receptor expression comparable to treatment with IFN- alone. Thus IL-1 also does not significantly affect IFN--induced Fas expression.

View larger version (22K): [in this window] [in a new window]   Figure 5.   Modulation of Fas expression by IFN- but not IL-1. Changes in the expression of Fas on the surface of A549 cells was determined by flow cytometry in native cells, cells treated for 24 h with IL-1 (50 pg/ml), IFN- (250 U/ml), or a combination of both. Immunofluorescence was recorded using 1 × 105 cells/ml incubated with a monoclonal PE-conjugated anti-human CD95 antibody (black curve) or matched PE-conjugated isotype control (gray curve) for 30 min at 4°C. Cells were thoroughly washed three times in PBS pH 7.4 containing 10% FBS, resuspended at the same density, and immediately counted. Each histogram of this representative experiment was generated from 10,000 events (ordinate; abscissa = fluorescence intensity; n = 4).

Opposition of IFN-/Fas-Mediated Apoptosis by IL-1

To determine the cumulative effect of Fas-mediated apoptosis in A549 cells exposed concomitantly to cytokines, A549 cells were co-incubated with different sequential combinations of IL-1 and IFN- prior to the addition of FasAb. Cells were then fixed and assayed for CK-18 cleavage and nuclear condensation. Changes in the frequency of apoptotic events were recorded in comparison to cells that had been treated with a combination of IFN-/FasAb that generated the highest degree of apoptosis. Prior exposure to IL-1 reduced the level of Fas-mediated apoptosis to background (Figure 6A). Simultaneous pretreatment with IFN- and IL-1 decreased the frequency of Fas- mediated apoptotic events by 46%. In fact, as little as two hours exposure to IL-1 resulted in a similar decrease in IFN-/FasAb induced apoptosis. However, prolonged exposure to IFN- before the addition of IL-1 prevented the IL-1-induced antiapoptotic effect.

View larger version (18K): [in this window] [in a new window]   Figure 6.   Sequence of cytokine exposure and PI 3-K influence on Fas-mediated cell death. (A) Cells were treated with IFN- and FasAb as previously described and compared with cells treated with different sequential combinations of IL-1 and IFN- before addition of FasAb. The sequence of treatment conditions is indicated numerically along the x axis. The relative frequency of apoptosis was compared with IFN-/FasAb treated cells by counting M30/DAPI positive cells as previously described. The data is from a single experiment and representative of four different experiments. (B) A549 cells were also stimulated as described above; however, 30 min before the initial stimulus with IL-1 or IFN-, cells were given the phosphatidyl inositol 3 kinase inhibitor, LY294002 (10 µM). After 24 h, cells were given a second dose of the above inhibitors at half the original dose, and 30 min later cells were stimulated with the Fas-crosslinking antibody. The relative frequency of apoptosis was compared with IFN-/ FasAb treated cells by counting M30 positive cells (*P < 0.05).

Because IL-1 induces Akt (Protein Kinase B) activity through phosphatidylinositol 3-kinase (PI 3-K) signaling (25), we reasoned that IL-1 protects lung epithelial cells from IFN-/Fas-mediated death via the PI 3-K pathway. To test this, the PI 3-K inhibitor LY294002 was added to cell cultures before the addition of cytokine and FasAb combinations and then the frequency of CK-18 cleavage events was determined. Inhibition of PI 3-K by LY294002 led to a statistically significant increase in the frequency of apoptosis in cells treated with IL-1 before the addition IFN- and FasAb (Figure 6B). Treatment with LY294002 alone minimally increased cell apoptosis above background levels and was not associated with significant cellular toxicity as determined by trypan blue staining (data not shown). Treatment with LY294002 also resulted in a marginal reduction of apoptosis in IFN-/FasAb-treated cells but did not reach statistical significance. Inhibition of IL-1 with a 1,000-fold molar excess of IL-1Ra completely reversed the ability of IL-1 to prevent apoptosis, demonstrating that cell activation through the IL-1 receptor signaling machinery is required in this process (data not shown).

Modulation of Akt Kinase Activity by IL-1, IFN-, and Fas Activation

Based on our results, we predicted that IL-1 promotes survival behavior in A549 cells through PI 3-K dependent activation of the serine/threonine kinase Akt. To determine the individual and combined effect of cytokines and Fas activation on Akt activity, we treated A549 cells as previously described and measured the Akt kinase activity in immunoprecipitates. Treatment with IL-1 alone induced Akt kinase activity ~ 2-fold above baseline values and this was significantly decreased in cells pretreated with LY249002 (10 µM). Consistent with our previous results, IFN- combined with Fas activation led to a substantial decrease in Akt kinase activity, suggesting that the increase in apoptosis was at least in part attributable to inactivation of this key survival pathway.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

In this investigation we demonstrate that IFN- and IL-1 have opposing effects on Fas-mediated apoptosis in a human lung epithelial cell line. Exposure to IFN- before Fas activation significantly increased caspase 3, 7, and 8 activity, caspase processing of CK-18, a key cytoskeletal protein in epithelial cells, and increased the appearance of apoptotic nuclei. IFN- increased Fas-mediated apoptosis 3-fold when compared with Fas activation alone. Pretreatment with IL-1 before Fas activation completely blocked apoptosis. Our results indicate that IFN- and IL-1 influence Fas-mediated apoptosis at multiple checkpoints in the apoptotic casacade. Activation of the key initiator caspase, caspase-8, occurred with Fas activation alone and increased ~ 2-fold with IFN- treatment. This correlated with a substantial induction of surface Fas receptor expression by IFN-. As expected, enhanced activation of caspase-8 by IFN- and FasAb resulted in a higher level of caspase-3 and -7 activity. Perhaps most intriguing, IL-1 prevented the activation of all caspases studied, including caspase-8, thereby rendering an anti-apoptotic effect proximal to or at the location of caspase-8 activation. Inhibition of apoptosis by IL-1 was not caused by changes in Fas expression because the constitutive expression of Fas on the surface of A549 cells was unaffected by IL-1. The differential effects of IL-1 and IFN- on caspase activation correlate well with our analysis of CK-18, Bcl-x_L, and PARP expression. All of these proteins are vital for maintaining epithelial cell function and integrity and all possess consensus cleavage domains that are recognized by activated terminal caspases during cell execution. As we demonstrate, IFN- accelerates the loss and/or proteolysis of each protein during Fas-mediated cell death, whereas IL-1 prevents it. Our results also provide evidence that IL-1 competes with IFN- during Fas-mediated apoptosis to reduce cell death and that the anti-apoptotic effect is at least partially dependent on the activation of PI 3-K and Akt, a key survival pathway. We now hope to confirm our observations in experiments using fully differentiated cultures of primary human lung epithelial cells.

Activation of caspase-8 is a key event in directing the Fas response by IFN- and IL-1. In agreement with previous studies, IFN- increases surface Fas expression (26) and directly correlates with enhanced activation of caspase-8 and cell death. Interestingly, IL-1 had no effect on Fas expression but completely prevented caspase-8 activation. This would imply that IL-1 prevents formation of the Fas death-inducing signaling complex (DISC) (27). We are now directing our investigations to determine if IL-1 can prevent caspase-8 activation by modulating the expression of factors involved with DISC formation and subsequent cell death. In preliminary studies, we have found that IL-1 does not effect the expression of FLICE/ c-FLIP (a.k.a. FADD-like interleukin-1-converting enzyme [FLICE] or cellular FLICE-inhibitory protein), an endogenous inhibitor that prevents DISC formation and caspase-8 activation (data not shown) (28). An alternative mechanism to prevent Fas-mediated apoptosis by IL-1 may occur by the inducing the expression of factors that form an inhibitory complex proximal to caspase-8 activation. Activation of nuclear factor B (NF-B), a transcription factor integral to IL-1 signal transduction, has recently been shown to augment expression of TNFR-associated factor 1 (TRAF1) and TRAF2, along with cIAP-1 and cIAP-2, and coordinately suppress caspase-8 activation and apoptosis (27).

Activation of the caspase cascade during Fas-mediated apoptosis is essential for the initiation and execution of apoptosis but is not entirely responsible for all functions resulting in programmed cell death. The Bcl-2 family is comprised of a number of proteins that play a critical role in determining mitochondrial membrane potential and therefore influence cell viability (29). Certain members of this family promote apoptosis by increasing mitochondrial membrane permeability, whereas others such as Bcl-2 and Bcl-x_L prevent it. This appears to be important because Bcl-x_L can prevent Fas-mediated death despite caspase activation (30). We identified that Bcl-x_L, and not Bcl-2, is constitutively expressed in the A549 cell line. In agreement with morphologic evidence of apoptosis, detection of the functional 26 kD protein was significantly decreased in cells treated with IFN-/FasAb but was consistently increased in cells exposed to IL-1. The latter finding is supported by recent work that has identified an NF-B gene responsive element in the promotor region of Bcl-x_L (31). It is therefore possible that in addition to preventing apoptosis by inhibiting caspase-8 activation, IL-1 promotes lung epithelial cell survival via transcriptional activation of Bcl-x_L expression in an NF-B-dependent manner. However, because the decrease in Bcl-x_L is taking place during excessive cell death, we cannot rule out the possibility that protein loss is a consequence of nonspecific proteolysis.

Recent work has identified that the interaction of PI 3-K with the interleukin-1 type I receptor (IL-1 RI) is one of the early events in IL-1 signaling and that PI 3-K is at least partially involved in activation of NF-B by IL-1 (32, Sizemore, 1999). The serine/threonine protein kinase, Akt, is one of the major early targets of PI 3-K and collaborates in activation of this key survival pathway (33). Most recently, Akt has been identified as the survival factor responsible for epidermal growth factor (EGF)-induced prevention of Fas-mediated apoptosis (34). Taken together, we predicted that activation of PI 3-K and Akt during IL-1 signaling may be a key proximal event in promoting lung epithelial cell survival. Our results indicate that PI 3-K dependent activation of Akt kinase is at least partially responsible for the IL-1-induced prevention of Fas-mediated death through a mechanism that remains to be identified. Perhaps equally important, IFN- enhanced transmission of the death signal by decreasing Akt kinase activity. In this regard, our results demonstrate that IL-1 and IFN- create an intracellular competition between death and survival signals and suggest that the local concentration and timing of cytokine presentation may be instrumental in determining lung epithelial cell outcome during inflammation. The intracellular location(s) where these distinct signaling pathways converge upon the Fas pathway and influence cell outcome remains to be determined.

The concentration of cytokines such as IL-1 and IFN- are elevated in the lower airway environment during inflammation. As our results demonstrate, both cytokines significantly modify transmission of the Fas death signal and influence cell outcome in a human lung epithelial cell line. We believe our observations have fundamental implications to lung diseases associated with acute epithelial cell loss and remodeling. A recent study has identified a causal link between the elevation in alveolar levels of FasL, lung epithelial cell apoptosis, and patient mortality in patients with ARDS (14). In addition, Geiser and colleagues recently reported that pulmonary edema fluid obtained from patients with acute lung injury promotes epithelial cell repair (35). Most striking, IL-1 was identified as the dominant factor involved in mediating lung epithelial cell wound repair. Taken together, cytokines such as IL-1 and IFN- that are expressed in the lower airway microenvironment during the early stages of inflammation direct the parenchymal cell response to Fas pathway activation. We predict that modulation of apoptotic signals by cytokines during the early stages of acute lung injury has a profound influence on tissue repair and may be instrumental in determining the individual host response.