Inhibition of CD95 apoptotic signaling by interferon-╨Ю╤Ц in human osteoarthritic chondrocytes is associated with increased expression of FLICE inhibitory protein.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 50, No. 2, February 2004, pp 498–506 DOI 10.1002/art.20008 © 2004, American College of Rheumatology Inhibition of CD95 Apoptotic Signaling by Interferon-␥ in Human Osteoarthritic Chondrocytes Is Associated With Increased Expression of FLICE Inhibitory Protein Francesco Grassi,1 Anna Piacentini,1 Sandra Cristino,1 Stefania Toneguzzi,1 Andrea Facchini,2 and Gina Lisignoli1 Objective. Cartilage homeostasis dysregulation during osteoarthritis (OA) has been linked to an increased rate of apoptosis of chondrocytes, the only cell type resident in the cartilage. In addition, the CD95– CD95 ligand (the Fas system) has emerged as one of the major pathways of cell death in the cartilage. We undertook the present study to investigate the role of interferon-␥ (IFN␥) in the regulation of the Fas system by analyzing the modulation of intracellular signaling molecules (FLICE inhibitory protein [FLIP] and caspases 3 and 8) in primary cultures of human OA chondrocytes. Methods. CD95-induced apoptotic death of human OA chondrocytes was analyzed in the presence or absence of IFN␥ using cell death immunoassay for apoptosis, real-time polymerase chain reaction for FLIP and caspase 8 expression, Western blotting for FLIP, and proteolytic activity for caspases 3 and 8. Results. CD95-induced apoptotic death of human OA chondrocytes was strongly counteracted by IFN␥ treatment, although the surface expression of CD95 was slightly up-regulated by this cytokine. The messenger RNA (mRNA) expression of FLIP and caspase 8, mediators involved in CD95 signaling, revealed that FLIP expression in human OA chondrocytes was significantly up-regulated (2-fold increase) by IFN␥ treatment. Moreover, the FLIP:caspase 8 mRNA ratio increased significantly. FLIP up-regulation by IFN␥ was confirmed at the protein level. Caspase 8 and caspase 3 proteolytic activities, both induced in these cells by stimulation with anti-CD95, were also significantly down-modulated by IFN␥. Conclusion. These findings suggest that IFN␥ impairs CD95-mediated signaling and apoptotic death in human chondrocytes. Its mechanism of action involves down-regulation of caspase 8 and caspase 3 activities and increased expression of the antiapoptotic protein FLIP, suggesting an interesting mechanism for the inhibition of chondrocyte apoptosis. Cartilage homeostasis dysregulation is a main feature of the osteoarthritic (OA) joint. Following repetitive mechanical stress as well as altered genetic factors, cartilage undergoes profound structural lesions mainly due to an increased synthesis of catabolic cytokines and matrix-degrading proteases (1,2). Chondrocytes, the only cell type present in the cartilage, play an essential regulatory function in matrix turnover. However, the rate of cellular turnover is affected during OA progression. Evidence is accumulating that chondrocyte response to mechanical and biochemical insults includes an increased rate of apoptosis (3–5), thus resulting in decreased tissue cellularity of the OA cartilage (6,7), an increased number of empty lacunae, and abnormal calcification of the subchondral bone (5,8). The hypothesis of a pathogenetic role of chondrocyte apoptosis in OA is supported by studies showing in situ a higher level of apoptosis in the OA cartilage compared with normal tissue (4,5) and a linkage between the rate of chondrocyte apoptosis and the severity of cartilage Supported by MURST (60% fund), Ricerca Corrente IOR, and the FP Health Ministry of Italy. 1 Francesco Grassi, PhD, Anna Piacentini, PhD, Sandra Cristino, PhD, Stefania Toneguzzi, BSc, Gina Lisignoli, PhD: Istituti Ortopedici Rizzoli, Bologna, Italy; 2Andrea Facchini, MD: Istituti Ortopedici Rizzoli, Bologna, Italy, and Università degli Studi di Bologna, Bologna, Italy. Address correspondence and reprint requests to Gina Lisignoli, PhD, Laboratorio di Immunologia e Genetica, I.O.R., Via di Barbiano 1/10, 40136 Bologna, Italy. E-mail: firstname.lastname@example.org. Submitted for publication February 19, 2003; accepted in revised form October 13, 2003. 498 INHIBITION OF APOPTOSIS BY IFN␥ IN HUMAN OA CHONDROCYTES damage (4,9). Moreover, a number of mediators that are overexpressed in the synovial fluid of OA patients, such as nitric oxide, interleukin-17, and CD95 ligand (CD95L), have the potential to induce chondrocyte apoptosis (10,11). CD95 (Fas/APO-1) and its ligand, CD95L, constitute the so-called Fas system, which represents a major mechanism in the regulation of the homeostasis of the immune cells and other cell types (i.e., tumor cells) (12). CD95 engagement triggers a variety of signaling molecules, including caspases, phospholipases, and protein kinases. In particular, sequential caspase activation results in the specific proteolytic events typical of programmed cell death. Some caspases act as initiators (caspases 8 and 10), while some others act as executioners (caspases 3, 6, and 7) (13). FLICE inhibitory protein (FLIP), the enzymatically inactive homolog of caspase 8, has been recently regarded as an intriguing regulator of cellular sensitivity to apoptotic cell death, due to its ability to block caspase 8 recruitment and activation at the death-inducing signaling complex (DISC) level (14,15). In particular, the FLIPL (long) isoform seems to be a more potent inhibitor than the FLIPS (short) isoform at comparable expression levels (14). In different cell types, including synovial macrophages and fibroblasts obtained from patients with arthritis, the level of FLIP expression has been correlated with cellular resistance to apoptosis induced by CD95 (16–21). A pivotal regulatory effect of interferon-␥ (IFN␥) on CD95 expression and function has been demonstrated in other cell systems, leading in most cases to an up-regulation of the CD95-dependent apoptotic death (22–24). Among cytokines and growth factors, which are implicated in the regulation of cartilage homeostasis, the role of IFN␥ is still poorly clarified. In vitro studies have demonstrated relevant effects of IFN␥ on chondrocyte metabolism (25,26) and a protective effect against cartilage breakdown (27). However, no data are available regarding the possible regulatory effect of IFN␥ on the CD95-mediated pathway of cell death in cartilage. In the present in vitro study, we investigated the role of IFN␥ in CD95-mediated cell death in human OA chondrocytes. We demonstrated that CD95-induced apoptotic death of human OA chondrocytes is strongly counteracted by IFN␥ treatment. The antiapoptotic effect of IFN␥ is associated with decreased enzymatic activity of both caspase 8 and caspase 3. Finally, we demonstrated an increased expression of FLIP leading to an increased FLIP:caspase 8 messenger RNA 499 (mRNA) ratio in treated chondrocytes, thus suggesting an intriguing role for IFN␥ in cartilage pathophysiology. MATERIALS AND METHODS Reagents. Recombinant human IFN␥ and tumor necrosis factor ␣ (TNF␣) were purchased from Roche Molecular Biochemicals (Mannheim, Germany). Penicillin/streptomycin, L-glutamine, and cycloheximide (CHX) were purchased from Sigma (St. Louis, MO). Anti-CD95 antibodies were purchased from Medical and Biological Laboratories (Nagoya, Japan) (clone CH-11) and Ancell (Bayport, MN), respectively, for functional tests and flow cytometric analysis of surface receptor expression. Anti–type I collagen and anti–type II collagen antibodies were obtained from Chemicon International (Temecula, CA). Dulbecco’s modified Eagle’s medium (DMEM) was obtained from Life Technologies (Gaithersburg, MD), and fetal calf serum (FCS) was purchased from Mascia Brunelli (Milan, Italy). Specimens and cell culture. Human OA cartilage was obtained from 8 patients who were undergoing total joint replacement of the knee. Diagnosis of OA was based on clinical, laboratory, and radiographic evaluation. The study was approved by the Institutional Review Board. Chondrocytes were obtained by sequential enzymatic digestion of minced fragments of cartilage, as previously described (28). Briefly, the isolated chondrocytes were filtered with nylon meshes, seeded in culture flasks, and grown in DMEM supplemented with 10% FCS, L-glutamine, and antibiotics. Cells were used after the first or second passage of culture. Immunocytochemical assessment of the chondrocyte phenotype was performed using monoclonal antibodies (mAb) against type I or type II collagen. For all experiments described in this report, chondrocytes were seeded at a density of 6 ⫻ 104/cm2. Fluorescence-activated cell sorting analysis. Chondrocytes from 6 donors with OA were assessed for CD95 expression by flow cytometry. Chondrocytes were seeded in 24-well plates for 48 hours and then treated or not treated with IFN␥ and TNF␣ (100 units/ml) for 24 hours. Cells were then trypsinized, preincubated with human Ig (2 mg/ml in phosphate buffered saline [PBS] with 2% FCS and 0.1% sodium azide) for 30 minutes at 4°C, washed twice in PBS, and incubated either with anti-CD95 mAb diluted 1:500 or with IgG1 isotype control for 30 minutes at 4°C. Chondrocytes were washed with PBS and incubated with fluorescein isothiocyanate–conjugated rabbit anti-mouse IgG (diluted 1:25; Dako, Glostrup, Denmark) for 30 minutes at 4°C. Finally, cells were fixed in 2% paraformaldehyde and analyzed by FACStar Plus (Becton Dickinson, Sunnyvale, CA). Cell death detection immunoassay. For quantification of nucleosomal fragment enrichment, chondrocytes were seeded in 24-well plates. After 48 hours, IFN␥ and TNF␣ were added to the culture medium at a final concentration of 100 units/ml for 24 hours. In dose-response experiments, IFN␥ was added to the culture medium at final concentrations of 10 units/ml, 100 units/ml, and 500 units/ml for 24 hours. The medium was then replaced, anti-CD95 mAb (clone CH-11) or isotype-matched mouse IgM control mAb was added at a concentration of 0.75 g/ml, and the cells were incubated for a further 14 hours at 37°C. At the end of the incubation time, the 500 cells were lysed, and fragmented nucleosomal DNA was measured using the photometric enzyme immunoassay (Cell Death Detection ELISA Plus; Roche Molecular Biochemicals) according to the manufacturer’s instructions. Apoptosis induction was calculated as follows: [A405 nm (treated) ⫺ A405 nm (untreated)]/A405 nm (untreated), where A ⫽ absorbance. Real-time polymerase chain reaction (PCR) for FLIP and caspase 8 expression. To analyze FLIP and caspase 8 expression at the mRNA level, chondrocytes were seeded in 12-well plates for 48 hours, then treated with IFN␥, anti-CD95, or the combination of both, as indicated above. Total RNA was isolated by the RNAzol B method (Biotecx Laboratories, Houston, TX) and reverse transcribed using Moloney murine leukemia virus reverse transcriptase and oligo(dT) priming according to the manufacturer’s protocol (Perkin-Elmer Cetus, Norwalk, CT). The following primers were designed for complementary DNA (cDNA) amplification: FLIP, forward 5⬘-TGGACCTTGTGGTTGAGTTG-3⬘ and reverse 5⬘TTGGATTGCTGCTTGGAGA-3⬘ (179-bp product, including both FLIPL and FLIPS isoforms; GenBank accession no. U97074); caspase 8, forward 5⬘-AGAGCCTGAGAGAGCGATG-3⬘ and reverse 5⬘-CACCATCAATCAGAAGGGAAG-3⬘ (166-bp product; GenBank accession no. NM 001228). Primers for the housekeeping gene GAPDH, used as an internal control, were as previously described (29). All primers were chosen to lie in different exons or to span exon junctions to prevent amplification of genomic DNA. Real-time PCR was run in a LightCycler Instrument (Roche Molecular Biochemicals) using the QuantiTect SYBR Green PCR kit (Qiagen, Hilden, Germany) with the following protocol: initial activation of HotStar Taq DNA polymerase at 95°C for 15 minutes, 40 cycles of 94°C for 15 seconds, 58°C for 15 seconds, and 72°C for 10 seconds. To determine absolute mRNA copy numbers, standard curves were generated for FLIP, caspase 8, and GAPDH using 10-fold dilution series of gel-purified PCR products (30). Two microliters of either external standards (ranging from 6 ⫻ 106 to 6 ⫻ 101 copies) or diluted cDNA samples (corresponding to 4 ng of total RNA per sample) was amplified in separate tubes for each target gene. The increase in PCR product was monitored for each amplification cycle by measuring the increase in fluorescence caused by the binding of SYBR Green I dye to doublestranded DNA. The crossing point values (i.e., the cycle number at which the detected fluorescence exceeds the threshold value) were determined for each sample, and specificity of the amplicons was confirmed by melting curve analysis. Amplification efficiencies, as calculated from the slopes of log input amounts plotted versus crossing point values, were confirmed to be high (⬎90%) and comparable for both target genes (⬍4% difference with respect to GAPDH). Subsequently, to check for intersample variations (e.g., during RNA extraction and/or reverse transcription), FLIP and caspase 8 mRNA levels were normalized to the housekeeping gene, using the calculated copy numbers. Western blotting. Total cellular protein extracts were obtained by washing the cells twice in PBS and resuspending them in lysis buffer (0.5% Triton X-100, 300 mM NaCl, 50 mM Tris HCl, pH 7.6, containing 1 mM phenylmethylsulfonyl fluoride, 2 g/ml aprotinin, and 10 g/ml leupeptin). Cells GRASSI ET AL were kept on ice for 30 minutes and then centrifuged at 10,000g for 10 minutes. The amount of cellular protein present in the clarified supernatant was evaluated using the bicinchoninic acid protein assay (Pierce, Rockford, IL). Equal amounts of cellular protein (20 g) from each sample were then separated by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred to Immobilon-P membranes (Millipore, Bedford, MA) using standard procedures. Blots were hybridized with rat anti-human FLIP (clone Dave 3, recognizing both FLIPL and FLIPS isoforms; Alexis Biochemicals, Lausen, Switzerland) or mouse anti–human ␤-actin antibody (Sigma), followed by horseradish peroxidase–conjugated antirat IgG (Alexis Biochemicals) or anti-mouse IgG (Amersham Biosciences, Little Chalfont, UK) antibody. Proteins were visualized using SuperSignal West Pico substrate (Pierce). Caspase 8 activity assay. Proteolytic activity of caspase 8 was evaluated using a FLICE/Caspase-8 Fluorometric Protease Assay Kit (Chemicon International) according to the manufacturer’s instructions. For each experimental point, 1.2 ⫻ 106 chondrocytes were seeded for 48 hours and stimulated with IFN␥, anti-CD95, or the combination of both, as described above. At the end of the incubation time, cells were lysed and incubated with the fluorogenic substrate 7-amino-4trifluoromethyl coumarin for 2 hours at 37°C in a buffer containing 5 mM dithiothreitol. Cells were kept under serumfree culture conditions in order to minimize the background level of enzymatic activity. In blocking experiments, the protein synthesis inhibitor CHX was added to cell culture at a final concentration of 10 g/ml 1 hour before IFN␥ treatment and maintained all during the incubation time. Samples were then analyzed using a Spectra Max Fluorometer (Molecular Devices, Sunnyvale, CA) equipped with a 400-nm excitation filter and a 505-nm emission filter. Caspase 3 activity assay. For the determination of caspase 3–specific activity, a fluorometric immunosorbent enzyme assay (Caspase 3 Activity Assay; Roche Molecular Biochemicals) was used according to the manufacturer’s instructions. For each experimental point, 1.2 ⫻ 106 chondrocytes were plated and stimulated with IFN␥, anti-CD95, or the combination of both, as described above. At the end of the incubation time, cells were washed twice with cold PBS, lysed for 1 minute on ice, and spun down for 15 minutes at 1,500g. One hundred microliters of the lysate was then transferred into a microtiter plate coated with anti–caspase 3 antibody and incubated for 1 hour at 37°C. The plate was washed and incubated for 2 hours at 37°C with 100 l of substrate solution containing the caspase 3–specific substrate Acetyl-Asp-GluVal-Asp-7-amido-4-trifluoromethyl-coumarin. Finally, the resulting fluorescence was measured at 505 nm in a Spectra Max Fluorometer (Molecular Devices). Analysis of CD95L in the synovial fluid of OA and rheumatoid arthritis (RA) patients. The concentrations of CD95L in the synovial fluid of 20 OA patients (mean ⫾ SD age 65 ⫾ 10 years, mean ⫾ SD disease duration 6 ⫾ 4 years) and 20 RA patients (mean ⫾ SD age 63 ⫾ 15 years, mean ⫾ SD disease duration 12 ⫾ 8 years) were analyzed by sandwich enzyme-linked immunosorbent assay (MBL, Tokyo, Japan) according to the manufacturer’s instructions. INHIBITION OF APOPTOSIS BY IFN␥ IN HUMAN OA CHONDROCYTES 501 RESULTS Figure 1. Flow cytometry analysis of CD95 expression on human chondrocytes obtained from patients with osteoarthritis (OA). CD95 expression was evaluated under basal conditions (untreated) and after stimulation for 24 hours with 100 units/ml of tumor necrosis factor ␣ (TNF␣) or 100 units/ml of interferon-␥ (IFN␥). a, Flow cytometry histograms from 1 of 5 representative experiments. Open histograms represent isotype control. Shaded histograms represent CD95-specific antibody. x-axis ⫽ channel numbers; y-axis ⫽ number of events. b, Columns represent the mean ⫾ SD fluorescence intensity of 6 OA chondrocyte samples analyzed by flow cytometry under basal conditions and after TNF␣ or IFN␥ treatment as described above. Statistical analysis. The Wilcoxon test was used to compare experimental groups. The analyses were performed using CSS Statistica statistical software (StatSoft, Tulsa, OK). Effect of IFN␥ and TNF␣ on CD95 surface expression. To determine whether IFN␥ and TNF␣ regulate the expression of CD95 in human OA chondrocytes, 6 of 8 OA chondrocyte samples were analyzed by flow cytometry for the surface expression of CD95. As shown in Figures 1a and b, untreated human OA chondrocytes expressed CD95 on their surface membranes. Incubation of cells for 24 hours with IFN␥ or TNF␣ at a concentration of 100 units/ml resulted in a slight up-regulation of CD95 expression. Down-regulation of CD95-mediated apoptosis by IFN␥. To assess the functional relevance of the CD95 modulation, we treated chondrocytes from 8 OA patients with anti-CD95 mAb (clone CH-11). Microscopic phase-contrast studies showed that the addition of antiCD95 induced significant morphologic changes compared with untreated conditions. In particular, several cells acquired smaller size morphology and rounded shape and showed membrane blebbing. Moreover, a small portion of the cells detached from the flask and floated in the culture medium. When IFN␥ was added 24 hours before anti-CD95 treatment, the number of cells undergoing such modification was much smaller. To evaluate the amount of cellular apoptosis under these conditions, we quantified nucleosomal fragment enrichment. As shown in Figure 2a, anti-CD95 treatment largely induced chondrocyte apoptosis. In contrast, when cells were treated with IFN␥, CD95induced apoptosis was significantly inhibited (P ⫽ 0.007). Interestingly, when chondrocytes were treated with TNF␣ (100 units/ml), the rate of apoptosis induction did not change significantly compared with that in the anti-CD95–treated cells. Dose-response experiments revealed that the antiapoptotic effect of IFN␥ was dose dependent (Figure 2b), and near-maximum inhibition was achieved at a concentration of 100 units/ml, while considerable inhibition was also observed at a concentration of 10 units/ml. IFN␥-induced increase in FLIP production by human chondrocytes. FLIP overexpression has been recently correlated with the suppression of Fas signaling in different cell types. Thus, chondrocytes from 6 OA patients were analyzed for total FLIP expression at the mRNA level, using real-time PCR. As shown in Figure 3a, our findings confirmed that human chondrocytes express FLIP mRNA. Absolute quantification revealed that in unstimulated chondrocytes, there was a mean ⫾ SD of 6,400 ⫾ 1,300 FLIP mRNA copies per 4 ng of total RNA. Interestingly, cell stimulation with IFN␥ at 502 GRASSI ET AL Figure 2. Effect of IFN␥ and TNF␣ on CD95-mediated DNA fragmentation. Analysis was performed on human OA chondrocytes after treatment with IFN␥ alone or after treatment with anti-CD95, IFN␥ ⫹ anti-CD95, or TNF␣ ⫹ anti-CD95. Fragmented DNA was measured by cell death detection immunoassay. a, Columns represent the mean ⫾ SD of 8 OA chondrocyte samples analyzed. Data are expressed as fold apoptosis induction calculated as follows: [A405 nm (treated) ⫺ A405 nm (untreated)]/A405 nm (untreated), where A ⫽ absorbance. ⴱ ⫽ P ⫽ 0.007 versus anti-CD95 alone. b, Results of 1 representative dose-response experiment (of 3 performed), showing the effect of different concentrations of IFN␥ (10 units/ml, 100 units/ml, and 500 units/ml) on CD95-mediated DNA fragmentation. Values are the mean and SD. See Figure 1 for definitions. 100 units/ml induced a statistically significant (P ⫽ 0.02) increase of FLIP mRNA expression, both in the presence and in the absence of anti-CD95 mAb. The rate of increase was 2-fold higher than that in unstimulated cells. Caspase 8 mRNA was also slightly increased with IFN␥ treatment (Figure 3b); finally, the FLIP:caspase 8 mRNA ratio underwent a statistically significant increase (Figure 3c) in samples treated with IFN␥, both in the presence (P ⫽ 0.02) and in the absence (P ⫽ 0.04) of anti-CD95 mAb. To determine whether the increased FLIP mRNA expression resulted in higher protein levels, immunoblotting for FLIPL was performed in samples from 3 OA patients. Although FLIPS (75 kd) was undetectable, the level of FLIPL (50 kd) protein ex- Figure 3. Effect of interferon-␥ (IFN␥) on FLICE inhibitory protein (FLIP) and caspase 8 mRNA expression. Real-time polymerase chain reaction experiments for a, FLIP and b, caspase 8 were performed on 6 osteoarthritic (OA) chondrocyte samples under basal conditions (untreated) and after treatment with IFN␥, anti-CD95, or IFN␥ ⫹ anti-CD95. Data are expressed as calculated mean ⫾ SD copy numbers after normalization to the housekeeping gene GAPDH. For FLIP mRNA copies, differences between values for untreated and IFN␥-treated samples and between values for anti-CD95–treated and IFN␥ ⫹ anti-CD95–treated samples were significant (ⴱ ⫽ P ⫽ 0.02). c, Columns represent the mean ⫾ SD FLIP:caspase 8 ratios for the 6 OA chondrocyte samples analyzed under the conditions indicated above. ⴱ ⫽ P ⫽ 0.04 versus untreated samples; ⴱⴱ ⫽ P ⫽ 0.02 versus samples treated with anti-CD95 alone. INHIBITION OF APOPTOSIS BY IFN␥ IN HUMAN OA CHONDROCYTES pressed by OA chondrocytes was consistent with PCR data, being higher in IFN␥-treated samples and very low or undetectable in unstimulated cells (Figure 4). Thus, inhibition of CD95-dependent apoptosis by IFN␥ is associated with increased expression of FLIP, suggesting a role for this caspase 8 antagonist in blocking the CD95 transduction pathway. Inhibition of caspase 8 and caspase 3 proteolytic activity inhibited by IFN␥. The antiapoptotic effect of IFN␥ led us to further investigate the mechanism by which this cytokine may interfere with CD95-mediated apoptosis. Therefore, 6 of 8 OA samples were analyzed for the proteolytic activity of caspase 8 and caspase 3, since a recent study has demonstrated that these proteases are specifically activated after CD95 engagement in human chondrocytes (31). As shown in Figures 5a and b, respectively, both caspase 8 and caspase 3 were significantly activated in samples treated for 14 hours with anti-CD95 mAb (P ⫽ 0.043 for both caspases), while treatment with IFN␥ resulted in a significant down-modulation of these apoptotic proteases (P ⫽ 0.043). Of interest, the effect of IFN␥ treatment on caspase activation resembled that shown in Figure 2a for nucleosomal fragment enrichment. Effect of CHX on IFN␥-mediated inhibition of caspase 8 activity. To determine whether IFN␥-induced inhibition of caspase 8 activity requires new protein synthesis, proteolytic activity of this enzyme was evaluated in isolated OA chondrocytes pretreated with the protein synthesis inhibitor CHX at 10 g/ml. As shown Figure 4. Effect of IFN␥ on FLIP protein. Western blot analysis for FLIP was performed on 3 OA chondrocyte samples under basal conditions (lane 1) and after treatment with IFN␥ (lane 2), anti-CD95 (lane 3), or IFN␥ ⫹ anti-CD95 (lane 4). One of 3 representative immunoblots is shown. FLIP-L ⫽ long isoform of FLIP (see Figure 3 for other definitions). 503 Figure 5. Effect of IFN␥ on caspase 8 (a) and caspase 3 (b) proteolytic activity. The analysis was performed on 6 OA chondrocyte samples under basal conditions (untreated) and after treatment with IFN␥, anti-CD95, or IFN␥ ⫹ anti-CD95. Data are expressed as mean ⫾ SD fluorescence units. ⴱ ⫽ P ⫽ 0.043 versus samples treated with anti-CD95 alone; ⴱⴱ ⫽ P ⫽ 0.043 versus untreated samples. See Figure 3 for definitions. in Figure 6, CHX treatment completely reversed the inhibitory effect exerted by IFN␥, thus restoring the level of caspase 8 activation observed in the presence of anti-CD95 mAb alone. Levels of CD95L in the synovial fluid of OA and RA patients. To establish the physiologic relevance of our in vitro data, we analyzed the concentration of CD95L in the synovial fluid of OA and RA patients. As shown in Figure 7, synovial fluid from OA patients contained significant amounts of CD95L. The levels of CD95L in the synovial fluid of RA patients were significantly higher than the corresponding levels in OA patients, as reported by other investigators (32,33). DISCUSSION The progression of OA disease seems to be closely related to the balance between life and death of 504 Figure 6. Necessary role of protein synthesis for interferon-␥ (IFN␥)– mediated down-regulation of caspase 8 enzymatic activity. Results shown are from 1 representative experiment (of 3 performed), showing the effect of cycloheximide (CHX; 10 g/ml) on caspase 8 proteolytic activity. Data are expressed as mean and SD fluorescence units. chondrocytes, the only cell type resident inside the cartilage. One of the potential mechanisms by which chondrocyte homeostasis is regulated is through the CD95–CD95L pathway. We investigated the effect of IFN␥ on anti-CD95–induced apoptosis in human OA chondrocytes. We found that chondrocytes treated with IFN␥ at 100 units/ml are largely resistant to apoptosis induced by anti-CD95 mAb (CH-11). In accord with previous results from other investigators, we found that human OA chondrocytes at first passage in culture significantly express CD95 on their surfaces. In contrast, chondrocytes do not seem to produce CD95L by themselves (11). However, significant amounts of this molecule were found by us and other investigators (32,33) in the synovial fluid of OA and RA patients, thus leading us to consider the CD95 pathway as a mechanism relevant to cellular death in the cartilage. Despite the fact that surface expression of CD95 was slightly up-regulated in IFN␥-treated cells, CD95dependent apoptosis was largely decreased, thus emphasizing the relevance of the antiapoptotic effect of IFN␥. This novel finding seems to be cell-type specific, since several reports based on other cell systems provide general evidence for a facilitating effect of this cytokine on CD95-triggered apoptosis (23,24,34). To our knowledge, up to now only one report has documented a delaying effect of IFN␥ (when added to cells at 1,000 units/ml) on CD95-dependent apoptosis in mature erythroid colony-forming cells (35). The specificity of our data was also confirmed by the evaluation of 3 different molecules involved in CD95 signaling. FLIP, also known as FLICE inhibitory protein, is GRASSI ET AL a cytoplasmic protein which prevents the processing and release of active caspase 8 from the receptor, thus acting as a dominant-negative inhibitor of caspase 8 (15). Human chondrocytes are already known to constitutively express FLIP mRNA (36). Here we report that chondrocytes treated with IFN␥ display an increased expression of cellular FLIP, at both the mRNA and protein levels. Increased transcription of FLIP mRNA has been shown to correlate well with apoptosis inhibition (20), demonstrating that FLIP is mainly regulated at the transcriptional level. Therefore, to further investigate the relevance of IFN␥-dependent induction of FLIP expression, we have used real-time PCR to calculate the number of mRNA copies for FLIP and caspase 8 in OA chondrocytes, thus focusing on whether IFN␥ affects the ratio of these two molecules. Unstimulated cells expressed a mean ⫾ SD of 6,400 ⫾ 1,300 FLIP mRNA copies per 4 ng of total RNA; this amount should be considered a fairly high level if compared with the results of similar analyses performed in other cell systems (18), and might explain the slow kinetic of CD95-dependent apoptotic death observed in human chondrocytes by us and others. The level of caspase 8 mRNA expression was surprisingly low in our cells (Figure 3b), nearly 10-fold lower than that of FLIP expression (Figure 3a). The amount of caspase 8 mRNA did not change after anti-CD95 stimulation, although the proteolytic activity under the same conditions increased significantly (Figure 5a). This finding may be explained by the evidence that caspase 8 is mostly regulated at the posttranscriptional level, and perhaps the duration of the anti-CD95 stimulation in our model might not be sufficient to induce an increased transcription of the caspase 8 DNA. Figure 7. Concentrations of CD95 ligand (CD95L) in synovial fluid of patients with osteoarthritis (OA) and rheumatoid arthritis (RA). Each data point represents 1 patient. INHIBITION OF APOPTOSIS BY IFN␥ IN HUMAN OA CHONDROCYTES Treatment of chondrocytes with IFN␥ led to a statistically significant increase in the FLIP:caspase 8 mRNA ratio. Although several reports have correlated the increased expression of FLIP to a reduced sensitivity to CD95-mediated apoptosis, the amount of FLIP needed to block this death pathway is not known. Therefore, the increased FLIP:caspase 8 ratio that we have found is likely to be biologically relevant for explaining the reduced sensitivity of chondrocytes to CD95-dependent apoptosis. Although caspase 8 was undetectable at the protein level in our chondrocyte lysates, these findings are supported by our data on FLIP analyzed at the protein level. Immunoblotting of cultured chondrocytes revealed very low or undetectable levels of FLIPL in unstimulated cells; however, protein expression was clearly increased by IFN␥ treatment, even in the presence of anti-CD95 stimulation. Moreover, our data demonstrate that the enzymatic activity of caspase 3 and caspase 8 (also known as FLICE), both involved in CD95 signaling in human chondrocytes (31), was down-modulated in IFN␥treated samples (Figures 5a and b). Caspase 3 is considered an executioner caspase and represents a pivotal step of the CD95-dependent intracellular pathway. Upstream of its activation, CD95 engagement results in the rapid recruitment of the adapter molecule FADD and the caspase 8 proenzyme at the DISC, finally leading to the activation of caspase 8, the most apical caspase. It has been reported that treatment with the protein synthesis inhibitor CHX inhibits FLIP expression within a few hours (37), thus supporting the evidence of a fast turnover rate of this protein; consistent with these data, pretreatment of human OA chondrocytes with CHX at 10 g/ml completely reversed the effect of IFN␥ on caspase 8 proteolytic activity. This finding supports the hypothesis that FLIP-enhanced synthesis is involved in the survival signal provided by IFN␥ in our model. In conclusion, the observation that IFN␥ protects chondrocytes against CD95-mediated apoptosis discloses a potential intriguing role for this cytokine in the regulation of cartilage homeostasis. Based on our results, IFN␥ may contribute to creating an antiapoptotic microenvironment in the cartilage under pathologic conditions of the joint, such as OA, by counteracting the effects of CD95L produced in the synovial fluids of these patients. Our data reveal that IFN␥ exerts its antiapoptotic effect at relatively low concentrations, close to those found in the synovial fluid of patients with rheumatic diseases (38,39). These findings may offer new 505 therapeutic tools to improve chondrocyte survival based on the prevention of the abnormal increase of apoptosis found in the cartilage after mechanical injury (40) or OA disease. Moreover, the up-regulation of FLIP expression and impairment of CD95 signaling demonstrated in our experiments open new insights into the biologic effects of IFN␥, which, at least in our model, constitute a pivotal survival factor for human chondrocytes. ACKNOWLEDGMENTS The authors wish to thank Mrs. Graziella Salmi for assistance with manuscript preparation and Mr. Luciano Pizzi for technical assistance. REFERENCES 1. Sandell LJ, Aigner T. Articular cartilage and changes in arthritis. An introduction: cell biology of osteoarthritis. Arthritis Res 2001; 3:107–13. 2. Pelletier J-P, Martel-Pelletier J, Abramson SB. Osteoarthritis, an inflammatory disease: potential implication for the selection of new therapeutic targets. Arthritis Rheum 2001;44:1237–47. 3. Blanco FJ, Guitian R, Vázquez-Martul E, de Toro FJ, Galdo F. Osteoarthritis chondrocytes die by apoptosis: a possible pathway for osteoarthritis pathology. Arthritis Rheum 1998;41:284–9. 4. Kim HA, Lee YL, Seong S-C, Choe KW, Song YW. Apoptotic chondrocyte death in human osteoarthritis. J Rheumatol 2000;27: 455–62. 5. Kourı̀ JB, Aguilera JM, Reyes J, Abbud Loyoza K, González S. Apoptotic chondrocytes from osteoarthrotic human articular cartilage and abnormal calcification of subchondral bone. J Rheumatol 2000;27:1005–19. 6. Mitrovic D, Quintero M, Stankovic A, Ryckewaert A. Cell density of adult human femoral condylar articular cartilage: joints with normal and fibrillated surfaces. Lab Invest 1983;49:309–16. 7. Vignon E, Arlot M, Patricot LM, Vignon G. The cell density of human femoral head cartilage. Clin Orthop 1976;121:303–8. 8. Hashimoto S, Ochs RL, Rosen F, Quach J, McCabe G, Solan J, et al. Chondrocyte-derived apoptotic bodies and calcification of articular cartilage. Proc Natl Acad Sci U S A 1998;95:3094–9. 9. Hashimoto S, Ochs RL, Komiya S, Lotz M. Linkage of chondrocyte apoptosis and cartilage degradation in human osteoarthritis. Arthritis Rheum 1998;41:1632–8. 10. Blanco FJ, Ochs RL, Schwarz H, Lotz M. Chondrocyte apoptosis induced by nitric oxide. Am J Pathol 1995;146:75–85. 11. Hashimoto S, Setareh M, Ochs RL, Lotz M. Fas/Fas ligand expression and induction of apoptosis in chondrocytes. Arthritis Rheum 1997;40:1749–55. 12. Lynch DH, Ramsdell F, Alderson MR. Fas and FasL in the homeostatic regulation of immune responses. Immunol Today 1995;16:569–74. 13. Wang J, Lenardo MJ. Roles of caspases in apoptosis, development, and cytokine maturation revealed by homozygous gene deficiencies. J Cell Sci 2000;113:753–7. 14. Irmler M, Thome M, Hahne M, Schneider P, Hofmann K, Steiner V, et al. Inhibition of death receptor signals by cellular FLIP. Nature 1997;388:190–5. 15. Krueger A, Schmitz I, Baumann S, Krammer PH, Kirchhoff S. Cellular FLICE-inhibitory protein splice variants inhibit different steps of caspase-8 activation at the CD95 death-inducing signaling complex. J Biol Chem 2001;276:20633–40. 16. Schedel J, Gay RE, Kuenzler P, Seemayer C, Simmen B, Michel 506 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. GRASSI ET AL BA, et al. FLICE-inhibitory protein expression in synovial fibroblasts and at sites of cartilage and bone erosion in rheumatoid arthritis. Arthritis Rheum 2002;46:1512–8. Perlman H, Pagliari LJ, Liu H, Koch AE, Haines GK III, Pope RM. Rheumatoid arthritis synovial macrophages express the Fasassociated death domain–like interleukin-1␤–converting enzyme–inhibitory protein and are refractory to Fas-mediated apoptosis. Arthritis Rheum 2001;44:21–30. Catrina AI, Ulfgren AK, Lindblad S, Grondal L, Klareskog L. Low levels of apoptosis and high FLIP expression in early rheumatoid arthritis synovium. Ann Rheum Dis 2002;61:934–6. Kim H, Whartenby KA, Georgantas RW III, Wingard J, Civin CI. Human CD34⫹ hematopoietic stem/progenitor cells express high levels of FLIP and are resistant to Fas-mediated apoptosis. Stem Cells 2002;20:174–82. Perlman H, Pagliari LJ, Georganas C, Mano T, Walsh K, Pope RM. FLICE-inhibitory protein expression during macrophage differentiation confers resistance to Fas-mediated apoptosis. J Exp Med 1999;190:1679–88. Rescigno M, Piguet V, Valzasina B, Lens S, Zubler R, French L, et al. Fas engagement induces the maturation of dendritic cells (DCs), the release of interleukin (IL)-1␤, and the production of interferon ␥ in the absence of IL-12 during DC-T cell cognate interaction: a new role for Fas ligand in inflammatory responses. J Exp Med 2000;192:1661–8. Horie T, Dobashi K, Iizuka K, Yoshii A, Shimizu Y, Nakazawa T, et al. Interferon-gamma rescues TNF-alpha-induced apoptosis mediated by up-regulation of TNFR2 and EoL-1 cells. Exp Hematol 1999;27:512–9. Riccioli A, Starace D, D’Alessio A, Starace G, Padula F, De Cesaris P, et al. TNF-␣ and IFN-␥ regulate expression and function of the Fas system in the seminiferous epithelium. J Immunol 2000;165:743–9. Spanaus KS, Schlapbach R, Fontana A. TNF-alpha and IFNgamma render microglia sensitive to Fas ligand-induced apoptosis by induction of Fas expression and down-regulation of Bcl-2 and Bcl-xL. Eur J Immunol 1998;28:4398–408. Henrotin YE, Zheng SX, Labasse AH, Deby GP, Crielaard J-MR, Reginster J-YL. Modulation of human chondrocyte metabolism by recombinant human interferon. Osteoarthritis Cartilage 2000;8: 474–82. Goldring MB, Sandell LJ, Stephenson ML, Krane SM. Immune interferon suppresses levels of procollagen mRNA and type II collagen synthesis in cultured human articular and costal chondrocytes. J Biol Chem 1986;261:9049–55. Bunning RAD, Russell RGG. The effect of tumor necrosis factor ␣ and ␥-interferon on the resorption of human articular cartilage and on the production of prostaglandin E and of caseinase activity by human articular chondrocytes. Arthritis Rheum 1989;32:780–4. Pulsatelli L, Dolzani P, Piacentini A, Silvestri T, Ruggeri R, 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. Gualtieri G, et al. Chemokine production by human chondrocytes. J Rheumatol 1999;26:1992–2001. Blanco FJ, Geng Y, Lotz M. Differentiation-dependent effects of IL-1 and TGF-beta on human articular chondrocyte proliferation are related to inducible nitric oxide synthase expression. J Immunol 1995;154:4018–26. Blaschke V, Reich K, Blaschke S, Zipprich S, Neumann C. Rapid quantitation of proinflammatory and chemoattractant cytokine expression in small tissue samples and monocyte-derived dendritic cells: validation of a new real-time RT-PCR technology. J Immunol Methods 2000;246:79–90. Kühn K, Lotz M. Regulation of CD95 (Fas/APO-1)–induced apoptosis in human chondrocytes. Arthritis Rheum 2001;44: 1644–53. Hashimoto H, Tanaka M, Suda T, Tomita T, Hayashida K, Takeuchi E, et al. Soluble Fas ligand in the joints of patients with rheumatoid arthritis and osteoarthritis. Arthritis Rheum 1998;41: 657–62. Matsuno H, Yudoh K, Watanabe Y, Nakazawa F, Aono H, Kimura T. Stromelysin-1 (MMP-13) in synovial fluid of patients with rheumatoid arthritis has potential to cleave membrane bound Fas ligand. J Rheumatol 2001;28:22–8. Ugurel S, Seiter S, Rappl G, Stark A, Tilgen W, Reinhold U. Heterogenous susceptibility to CD95-induced apoptosis in melanoma cells correlates with bcl-2 and bcl-x expression and is sensitive to modulation by interferon-gamma. Int J Cancer 1999; 82:727–36. Choi I, Muta K, Wickrema A, Krantz SB, Nishimura J, Nawata H. Interferon gamma delays apoptosis of mature erythroid progenitor cells in the absence of erythropoietin. Blood 2000;95:3742–9. Masuko-Hongo K, Sakata M, Yuan GH, Onuma H, Nakamura H, Aoki H, et al. Expression of Fas-associated death domain-like interleukin-1beta-converting enzyme (FLICE) inhibitory protein (FLIP) in human articular chondrocytes: possible contribution to the resistance to Fas-mediated death of in vitro cultured human articular chondrocytes. Rheumatol Int 2001;21:112–21. Willems F, Amraoui Z, Vanderheyde N, Verhasselt V, Aksoy E, Scaffidi C, et al. Expression of c-FLIPL and resistance to Cd95mediated apoptosis of monocyte-derived dendritic cells: inhibition by bisindolylmaleimide. Blood 2000;95:3478–82. Kahle P, Saal JG, Schaudt K, Zacher J, Fritz P, Pawelec G. Determination of cytokines in synovial fluids: correlation with diagnosis and histomorphological characteristics of synovial tissue. Ann Rheum Dis 1992;51:731–4. Westacott CI, Whicher JT, Barnes IC, Thompson D, Swan AJ, Dieppe PA. Synovial fluid concentration of five different cytokines in rheumatic diseases. Ann Rheum Dis 1990;49:676–81. D’Lima DD, Hashimoto S, Chen PC, Lotz M, Colwell CW Jr. Prevention of chondrocyte apoptosis. J Bone Joint Surg Am 2001;83-A Suppl 2(Pt 1):25–6.