Altered expression of membrane-bound and soluble CD95/Fas contributes to the resistance of fibrotic lung fibroblasts to FasL induced apoptosis

Tissue samples from patients with (n = 5) and without lung fibrosis (n = 6) were obtained from diagnostic open lung biopsies (fibrotic samples) and from healthy tissue areas during pneumonectomy for tumor resection (non-fibrotic samples). The fibrotic samples comprised the following diseases: usual interstitial pneumonia (UIP, two patients), non-specific interstitial pneumonia (NSIP, one patient), bronchiolitis obliterans-organizing pneumonia (BOOP, two patients) [3]. Although these diseases are different in clinical behavior and outcome, they are characterized by increased amounts of activated fibroblasts and increased matrix deposition. The histology of these tissue samples was recently partially described [9]. The lung fibroblasts were compared to synovial fibroblasts (n = 5), which were obtained from patients with osteoarthritis during joint replacement surgery. An experienced surgical pathologist (C.R.) examined the tissue specimens. All tissue samples were obtained immediately after surgery and used for the isolation of fibroblasts. The tissue sampling was approved by the local ethical committee.

Fibroblasts were trypsinized. For extracellular staining, fibroblast-specific antibodies [clones AS02 (anti-Thy-1) and D7-Fib; Dianova, Hamburg, Germany], a macrophage-specific anti-CD68 (clone KP1; Signet Laboratories, Inc., Dedham, MA) and a pan-leukocyte anti-CD45 antibody were used.

Cells were incubated with the primary antibodies for 30 min and with a fluorescein isothiocyanate (FITC)-labeled goat-anti-mouse IgG for 20 min. In addition, intracellular fluorescence staining was performed with anti-prolyl-4-hydroxylase antibodies (clone 5B5; DPC Biermann, Bad Nauheim, Germany) using the Fix and Perm reagent (Dianova) according to the instructions of the manufacturer. For the analysis, a FACSCalibur (Becton Dickinson, Heidelberg, Germany) flow cytometer was used.

Fibroblasts were obtained by mincing freshly excised lung parenchyma into ~1 mm³ pieces, followed by digestion with collagenase IV (1 mg/ml, Sigma, Deissenhofen, Germany) for 30 min at 37 °C. Fibroblasts were cultured in a 75-ml tissue culture flask containing Iscove's modified Dulbecco's medium with 10% (w/v) fetal calf serum (FCS), 10^-3 M glutamine and antibiotics, at 37°C and 5% (v/v) CO₂ until they reached confluence. Only fibroblasts between passages 3 and 8 were used for the experiments.

Collagen secretion and deposition into the ECM was assessed by proline incorporation assays originally developed by Peterovsky and Diegelmann [10] and described in detail earlier [11, 12]. All assays were performed in triplicate. Briefly, 5 × 10⁴ fibroblasts were seeded into 24-well plates (Falcon, Heidelberg, Germany) in culture medium containing 10 % FCS. After 16 h, the medium was changed to low serum medium (Dulbecco's modified Eagle's medium supplemented with 0.1% FCS, 100 μg/ml L-ascorbic acid) containing [2,3,4,5-³H]-lL-proline (2 μCi/ml, NEN, Boston, MA). When indicated, E64d was added (10 μM). After 72 h, the culture medium was removed and the remaining fibroblasts were lysed with distilled water (10 min, room temperature). The ECM was ethanol fixed (70% ethanol, 15 min, RT). Half of the wells were incubated with 30 U/ml collagenase (Clostridium histolyticum, Sigma, Deissenhofen, Germany) in collagenase assay buffer (50 mM Tris-HCL, pH 7.5, 5 mM CaCl₂, 2.5 mM N-ethylmaleimide) for 4 h at 37°C. The remaining wells were incubated with assay buffer. The supernatants were removed and residual ECM was solubilized by overnight incubation in 0.3 M NaOH-1% SDS. Equal numbers of aliquots of supernatants obtained after collagenase digestion and supernatants containing the residual ECM were subjected to liquid scintillation counting. The counts measured in supernatants after collagenase treatment represent the collagen content. The amount of [³H]proline measured after solubilization of the remaining ECM represents non-collagenous ECM. The total of both counts was equal to the counts from solubilized ECM without collagenase treatment and represents the total proline incorporation. Relative ECM synthesis can be calculated by the established formula [12]: ECM = CPM in collagen + (5.4 × CPM in non-collagen ECM). The formula contains the factor 5.4 to correct for the 5.4-fold higher proline or hydroxyproline content of collagens compared with that of other proteins.

Fibroblasts were stimulated with 100 ng/ml recombinant human FasL for 16 h as described [13]. When indicated, cells were preincubated with TNF-α or cycloheximide (100 μg/ml) for 24 h. Subsequently, apoptosis was determined using a histone fragmentation assay (Cell Death Detection ELISAPlus, Roche Diagnostics, Mannheim, Germany) according to the manufacturer's instructions. This assay is based on a quantitative sandwich-enzyme-immunoassay using mouse monoclonal antibodies against DNA and histones that allow for the specific, quantitative determination of cytoplasmatic histone-associated-DNA-fragments (mono- and oligonucleosomes) in the cell lysates. The ELISA plates were read at 405 nm (490 nm reference). We have shown before that the results obtained using this assay correlate to the amount of apoptotic cells found after TUNEL staining [14].

Additionally, apoptosis was measured in lung fibroblasts using TUNEL staining (ApoBrdU kit, Pharmingen, Heidelberg, Germany). Briefly, following induction of apoptosis, cells were fixed in 1% paraformaldehyde and incubated with Br-dUTP in the presence of TdT enzyme, which results in the incorporation of Br-dUTP into exposed 3-OH DNA ends. Br-dUTP sites were then labeled with FITC-conjugated anti-Br-dUTP antibodies. The number of apoptotic cells was measured using flow cytometry (FACS Calibur, Becton Dickinson), and labeling with Br-dUTP was compared with that of unstimulated controls.

Expression levels of Fas/CD95 mRNA were analyzed by quantitative real time PCR using a fluorogenic 5'-nuclease assay (TaqMan©, Applied Biosystems, Weiterstadt, Germany) on a ABI Prism 7900 HT Sequence Detection system. For each experiment, total RNA was extracted from 10⁵ cells using the RNeasy system (Qiagen, Hilden, Germany). Total RNA was reverse transcribed using random hexamer primers. For quantitative PCR, the appropriate primers and FAM-TAMRA labeled probes were purchased as 20-fold concentrated predeveloped assays from Applied Biosystems and used according to the instructions of the manufacturer. 18S rRNA gene was co-amplified as an internal standard. Data were calculated with the ΔΔCt method as described [15].

For the detection of sFas in the cell culture supernatants of the fibroblasts, a commercially available ELISA (Quantikine Assays, R&D Systems, Wiesbaden, Germany) was used according to the manufacturer's instructions.

The amount of cell-surface-bound Fas was measured by flow cytometry after staining the fibroblasts with FITC-labelled anti-Fas/CD95 antibodies (Becton Dickinson).

The fibroblasts isolated from lung tissue specimens were characterized with respect to the expression of lineage-specific marker proteins (Fig. 1A). The majority of the cells stained with antibodies directed against Thy-1, an antigen that is specific for fibroblasts. These cells also expressed the fibroblast specific antigen D7-Fib and the enzyme prolyl-4-hydroxylase, which is involved in collagen synthesis. Neither CD68, a marker of monocytes/macrophages, nor CD45, a leukocyte membrane protein, were detected. We found no significant differences in the phenotypic characteristics of fibroblasts which were derived from different patient groups.

The matrix production of fibroblasts isolated from fibrotic (fibrotic fibroblasts, f-fibs) and non-fibrotic lung tissues (normal fibroblasts, n-fibs) was also analyzed: F-fibs produced significantly more ECM proteins, including collagen, than n-fibs (Fig. 1B) independent from the underlying disease.

Recently we showed that fibroblasts derived from patients with different inflammatory joint diseases display different susceptibilities to FasL-induced apoptosis [14]. Comparing fibroblasts derived from human lung tissues, we applied the same conditions for induction of apoptosis and found that lung fibroblasts are generally more resistant to FasL-induced apoptosis than synovial fibroblasts (Fig. 1C). Similar results were found after staurosporin treatment of fibroblasts (Fig. 1C) and after induction of apoptosis by anti-Fas antibodies (not shown). The percentage of apoptotic cells, as determined by TUNEL staining, was at the detection limit. Therefore this method could not be used for the comparison of apoptosis in lung fibroblasts. On the other hand, we found low but measurable apoptosis after quantification of the amount of histone-associated DNA fragments in the cellular supernatants. Comparison of f-fibs and n-fibs confirmed that f-fibs were more resistant to FasL-induced apoptosis. Incubation of the cells with TNF-α slightly increased the susceptibility of n-fibs to apoptosis, but it had no affect on apoptosis in f-fibs (Fig. 1D).

Recently Tanaka et al. have shown that the resistance to anti-Fas-induced apoptosis in lung fibroblasts is mediated by the overexpression of the specific inhibitors of apoptosis X-chromosome-linked inhibitor of apoptosis (ILP) and FLICE-like inhibitor protein (FLIP) [16].

Those authors found that suppression of protein synthesis using cycloheximide decreased the concentration of these short-lived inhibitory proteins and led to increased susceptibility to Fas-mediated apoptosis. We used cycloheximide to analyse whether the differences in sensitivity to FasL-induced apoptosis depended on different expression levels of short-lived inhibitory proteins. As expected, preincubation of fibroblasts with cycloheximide increased apoptosis as determined by TUNEL-staining. However, the percentage of apoptotic cells among f-fibs was still lower than among n-fibs (Fig. 2), which suggests that short-lived inhibitory proteins do not contribute to the difference in apoptosis resistance between n-fibs and f-fibs. Therefore, other mechanisms are involved in the regulation of the resistance to FasL-induced apoptosis in f-fibs.

Expression of Fas plays a crucial role in FasL-induced apoptosis. Therefore, we investigated Fas-mRNA expression using quantitative RT-PCR analysis. Fas-mRNA levels were normalized to 18S-ribosomal RNA. Surprisingly, we found increased expression of Fas-mRNA in f-fibs (Fig. 3A). To determine whether increased Fas mRNA in f-fibs translates into increased levels of cell surface Fas, we used flow cytometry to analyze the expression of Fas on the cell surface of lung fibroblasts. We found that 65 ± 3% (mean fluorescence intensity 41 ± 5%) of n-fibs and 41 ± 5% of f-fibs (mean fluorescence intensity 24 ± 2%) expressed Fas at the cell surface (Fig. 3B, C). Based on these data, we then determined the concentration of soluble Fas in the culture supernatant and found an increased concentration of soluble Fas in the supernatant of f-fibs (Fig. 3D).