Outside-to-inside signaling through transmembrane tumor necrosis factor reverses pathologic interleukin-1 production and deficient apoptosis of rheumatoid arthritis monocytes.код для вставкиСкачать
ARTHRITIS & RHEUMATISM Vol. 60, No. 9, September 2009, pp 2612–2621 DOI 10.1002/art.24778 © 2009, American College of Rheumatology Outside-to-Inside Signaling Through Transmembrane Tumor Necrosis Factor Reverses Pathologic Interleukin-1␤ Production and Deficient Apoptosis of Rheumatoid Arthritis Monocytes Undine Meusch, Manuela Rossol, Christoph Baerwald, Sunna Hauschildt, and Ulf Wagner ing inhibited the constitutive NF-B activation in RA monocytes, suppressed IL-1␤ secretion, and normalized spontaneous in vitro apoptosis. This normalization was reversible by the addition of exogenous IL-1␤. Conclusion. This study demonstrates that outside-to-inside signaling through transmembrane TNF after ligation by infliximab inhibits constitutive NF-B activation and suppresses spontaneous IL-1␤ production by monocytes from patients with RA. Besides the induction of monocyte apoptosis, this inhibition could also contribute to the therapeutic effects observed during treatment with TNF␣ inhibitors. Objective. Monocytes are a major source of proinflammatory cytokines in rheumatoid arthritis (RA), and inhibitors of monocytic cytokines are highly efficient agents for treatment of the disease. The aim of this study was to analyze the effects of a therapeutic anti– tumor necrosis factor ␣ (anti-TNF␣) antibody on monocytes from patients with RA and healthy control subjects. Methods. Peripheral blood monocytes from patients with RA and healthy control subjects were incubated in the presence of anti-TNF␣ antibody or IgG. Annexin V staining, caspase activation, poly(ADP-ribose) polymerase cleavage, and DNA staining with propidium iodide were used to analyze apoptosis. The signaling events elicited in monocytes by infliximab were analyzed by Western blotting and electromobility shift assay. Results. Peripheral blood monocytes from patients with RA were characterized by increased expression of transmembrane TNF␣, spontaneous in vitro production of interleukin-1␤ (IL-1␤), and a decreased rate of spontaneous ex vivo apoptosis. Incubation with infliximab induced significantly increased apoptosis in monocytes from patients with RA but not in monocytes from healthy control subjects. This apoptosis was triggered by reverse signaling of transmembrane TNF after ligation by infliximab and was independent of caspase activation. Instead, transmembrane TNF reverse signal- The cytokine tumor necrosis factor ␣ (TNF␣) plays a pivotal role in the pathogenesis of rheumatoid arthritis (RA). Following de novo synthesis, the cytokine is integrated in the cell membrane as a type II transmembrane protein arranged in stable homotrimers and is released in its soluble form exclusively after proteolytic cleavage from this membrane-integrated form by the metalloproteinase TNF␣-converting enzyme. Transmembrane TNF␣, besides being a precursor of the soluble cytokine, has also been shown to exert an important immunologic function on its own. TNF-binding compounds are now widely used in clinical practice, and several cellular mechanisms for their beneficial effects have been described, including complement-dependent cytotoxicity, antibody-dependent cell-mediated cytotoxicity, and outside-to-inside (reverse) signal transduction through transmembrane TNF (1). Our group previously showed that the cross-talk between transmembrane TNF and its 2 receptors during direct cell contact of monocytes with preactivated T cells results in monocyte activation (2). Reverse signaling of transmembrane TNF on the monocyte cell surface was observed to partially mediate this effect. Other groups of investigators have shown that reverse signaling induces the expression of E-selectin in activated CD4⫹ T cells (3), mediates resis- Supported by grants from the German Ministry for Education and Science (Interdisziplinäres Zentrum für Klinische Forschung Leipzig, Teilprojekt A 21). Undine Meusch, Manuela Rossol, PhD, Christoph Baerwald, MD, Sunna Hauschildt, PhD, Ulf Wagner, MD: University of Leipzig, Leipzig, Germany. Ms Meusch and Dr. Rossol contributed equally to this work. Address correspondence and reprint requests to Ulf Wagner, MD, Division of Rheumatology, Department of Internal Medicine II, University of Leipzig, Liebigstrasse 20, 04103 Leipzig, Germany. E-mail: firstname.lastname@example.org. Submitted for publication December 22, 2008; accepted in revised form June 1, 2009. 2612 EFFECTS OF ANTI-TNF␣ ANTIBODY ON MONOCYTES FROM PATIENTS WITH RA tance to lipopolysaccharide (LPS) in monocytes (4), and influences cytokine production (5,6). However, in other in vitro settings, the main cellular effect of reverse signaling following ligation of transmembrane TNF in monocytes was the induction of apoptosis (6–9). The reported signal transduction pathways include activation of MAPK p38 (9), autocrine production of transforming growth factor ␤1 (9), activation of caspases 8, 9, and 3 (8,9), and cleavage of poly(ADP-ribose) polymerase 1 (PARP-1) (9). In view of those conflicting results, the goal of our study was to investigate the cellular response to and the proapoptotic effects of transmembrane TNF reverse signaling in monocytes from healthy individuals and patients with RA, as well as the underlying signal transduction events. PATIENTS AND METHODS Patients and control subjects. Peripheral blood samples obtained from 48 patients with a diagnosis of RA according to the American College of Rheumatology (formerly, the American Rheumatism Association) 1987 revised criteria for the classification of RA (10) were analyzed. The study design was approved by the ethics committee of the University of Leipzig, and informed consent was obtained from each patient before study enrollment. Eighty-eight percent of the patients had IgM rheumatoid factor–seropositive disease, and 81% of the patients had anti–cyclic citrullinated peptide antibodies. The median age of the patients was 66 years (range 22–91 years), the median age at disease onset was 52 years (range 18–69 years), and the median disease duration was 14 years (range 1–38 years). The patients did not receive any TNF␣inhibiting therapy before study enrollment. Current treatment regimens included conventional disease-modifying antirheumatic drugs, given either as monotherapy or in combination, as well as low-dose prednisolone in 80% of the patients. Fifty age-matched control subjects with no history of inflammatory arthritis were recruited from among healthy blood donors, with ethics committee approval. The median age of the control subjects was 34 years (range 23–65 years). Monocyte isolation, cell culture, and stimulation of human monocytes. Monocytes were isolated as described previously (11). Monocytes (2 ⫻ 105/200 l) were incubated in RPMI 1640 supplemented with 5% human AB serum (heat inactivated) and either 100 g/ml IgG (Intraglobin CP; Biotest, Dreieich, Germany) or infliximab (Centocor, Leiden, The Netherlands) for 16 hours. In some experiments, the soluble TNFR2:Ig construct etanercept (Wyeth-Pharma, Münster, Germany) was used instead of infliximab. For the inhibition of casein kinase 1 (CK-1), cells were preincubated with the CK-1 inhibitor D4476 (150 M; Calbiochem, La Jolla, CA) for 15 minutes. For global inhibition of caspases, the pan-specific inhibitor Z-VAD-FMK (10 M; Calbiochem) was used. To inhibit caspase 3, the cells were incubated with the inhibitor Z-DEVD-FMK (10 M; Calbiochem). Detection of transmembrane TNF by flow cytometry. Monocytes (2 ⫻ 105/50 l) were incubated with 1 mg/ml IgG (Intraglobin CP; Biotest) in phosphate buffered saline (PBS) 2613 supplemented with 2% fetal calf serum (FCS) and 0.1% Na-azide for 30 minutes at 4°C to block Fc receptor binding. Subsequently, phycoerythrin-labeled anti-TNF antibody (clone 6402.31) or IgG1 isotype control (R&D Systems, Minneapolis, MN) was added, and cells were incubated for an additional 30 minutes at 4°C. After the cells were washed with PBS supplemented with 2% FCS and 0.1% Na-azide, they were fixed with 1% formaldehyde. Transmembrane TNF expression was analyzed by flow cytometry (FACSCalibur; BD Biosciences, San Jose, CA), and analysis was performed using CellQuest software (BD Biosciences). Determination of apoptosis. Apoptotic and necrotic cells were stained with 10 l fluorescein isothiocyanate– labeled annexin V (Southern Biotechnology, Birmingham, AL) and 50 g/ml propidium iodide (PI), respectively. Flow cytometry was performed using the FACSCalibur system, and the results were analyzed using CellQuest software. Measurement of DNA degradation. Monocytes (1 ⫻ 106/ml) were washed with PBS supplemented with 0.05% Na-azide and incubated with 4% paraformaldehyde for 15 minutes at 4°C. Cells were washed again with PBS supplemented with 0.05% Na-azide and subsequently incubated in 0.1% saponin and 62.5 g/ml PI for 20 minutes at 4°C. Flow cytometry was performed using the FACSCalibur system, and results were analyzed using CellQuest software. Gel electrophoresis and Western blotting. Gel electrophoresis and Western blotting were performed as described previously (2). Rabbit monoclonal antibodies for the detection of caspase 3, Bcl-xL, and phosphorylated IB␣, mouse monoclonal antibodies for the detection of cleaved PARP and caspase 8, and polyclonal rabbit antibodies for the detection of caspase 9, caspase 1, and interleukin-1␤ (IL-1␤) were purchased from Cell Signaling Technology (Beverly, MA). Rabbit polyclonal anti-human IB␣ antibody and secondary antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Preparation of nuclear extracts. Nuclear extracts were prepared from 6 ⫻ 106 monocytes after incubation with IgG or infliximab for 16 hours. Cells were washed with PBS supplemented with 1 mM EDTA, resuspended in lysis buffer containing 10 mM Tris (pH 7.8), 5 mM MgCl2, 10 mM KCl, 1 mM EGTA (pH 7.0), 10% sucrose, 1 mM dithiothreitol, 10 mM ␤-glycerophosphate (Sigma-Aldrich, St. Louis, MO), 1 mM phenylmethylsulfonyl fluoride (PMSF), and Complete mini protease inhibitors (Roche Diagnostics, Mannheim, Germany), placed on ice for 15 minutes, and then vortexed for 10 seconds after the addition of 25 l of 10% Nonidet P40. After a 5-minute centrifugation (1,700g at 4°C), the pelleted nuclei were gently washed with 200 l lysis buffer without Nonidet P40 and pelleted (1,700g for 1 minute at 4°C). Pelleted nuclei were resuspended in 30 l extraction buffer containing 20 mM Tris (pH 7.8), 5 mM MgCl2, 320 mM KCl, 0.2 mM EGTA, 25% glycerol, 0.1 mM dithiothreitol, 10 mM ␤-glycerophosphate, 1 mM PMSF, and Complete mini protease inhibitors. The nuclear extracts were maintained on ice for 15 minutes, with occasional vortexing. Subsequently, nuclear extracts were cleared (20,800g for 5 minutes at 4°C) and transferred to fresh tubes. Protein concentrations were determined with a detergent-compatible protein assay (Bio-Rad, Hercules, CA). All extracts were stored at –80°C. Electrophoretic mobility shift assay. Single-stranded oligonucleotides containing the sequences corresponding to 2614 MEUSCH ET AL the NF-B consensus site (5⬘-AGTTGAGGGGACTTTCCCAGGC-3⬘ and 3⬘-TCAACTCCCCTGAAAGGGTCCG-5⬘ [the binding site sequence is underlined]) were synthesized by personnel at MWG Biotech (Ebersberg, Germany). To determine whether the observed shifted bands are specific for NF-B, the following mutant oligonucleotide probes were used: for NF-B, 5⬘-AGTTGAGGCGACTTTCCCAGGC-3⬘ and 3⬘-TCAACTCCGCTGAAAGGGTCCG-5⬘ (the binding site sequence is underlined, and the mutated site is shown in bold). Single-stranded oligonucleotide probes were endlabeled with [␥-32P]-ATP using T4 polynucleotide kinase and separated from the unincorporated label using the QIAquick Nucleotide Removal Kit (Qiagen, Hilden, Germany). The purified sense and antisense oligonucleotide probes were annealed for 10 minutes at room temperature to a final concentration of 2 g/ml. The protein–DNA binding reaction was carried out in a volume of 10 l containing binding buffer (10 mM Tris [pH 8.0], 150 mM KCl, 0.5 mM EDTA, 0.1% Triton X-100, 12.5% glycerol, 0.2 mM dithiothreitol), 2 g of poly(dG-dC), 1 g of poly(dA-dT) (both from Sigma-Aldrich), 2 ng of labeled double-stranded oligonucleotide probe, and 5 g of nuclear extract. The binding reaction was allowed to proceed for 30 minutes at room temperature. Samples were subjected to electrophoresis in 5% nondenaturating polyacrylamide gel in a 0.5⫻ Tris–borate–EDTA buffer system for 2.5 hours at 4°C at 150V. After electrophoresis, the gel was transferred to Whatman paper, dried on a gel dryer for 2 hours at 80°C, and visualized using the FLA-3000 Phosphor Imaging System (Fujifilm, Duesseldorf, Germany). Detection of IL-1␤ and IL-1 receptor antagonist (IL1Ra). IL-1␤ and IL-1Ra concentrations were measured by using an IL-1␤ enzyme-linked immunosorbent assay (ELISA) from BD Biosciences and an IL-1Ra ELISA from R&D Systems, according to the manufacturer’s instructions. Statistical analysis. For statistical analysis, SigmaStat software (SPSS, Chicago, IL) was used. Prior to all comparisons, a normality test was performed. To assess statistical significance, Student’s t-test (normal distribution of data) or the Mann-Whitney rank sum test (unequal distribution of data) was used. Correlation of 2 parameters was analyzed with Pearson’s product-moment correlation. RESULTS High-level transmembrane TNF␣ expression by peripheral blood monocytes from patients with RA. Our group has previously demonstrated that ligation of transmembrane TNF on monocytes triggers reverse signaling and induction of TNF␣. Such reverse signaling is a strong stimulus for monocytes during direct cellular contact with preactivated T cells and can be inhibited by anti-TNF antibodies (2). To investigate a potential role for transmembrane TNF in patients with RA, cell surface expression of the molecule was analyzed by flow cytometry. Monocytes from healthy donors and patients with RA expressed transmembrane TNF on the cell surface, Figure 1. Expression of transmembrane tumor necrosis factor (tmTNF) on freshly isolated monocytes from healthy donors (HD; n ⫽ 21) and patients with rheumatoid arthritis (RA; n ⫽ 23), as determined using flow cytometry. A, Expression on monocytes stained with phycoerythrin-labeled IgG1 control antibody (open histogram) or anti-TNF antibody (shaded histogram). B, Dot plot depicting the median fluorescence intensity of transmembrane TNF surface expression. Bars show the median. as shown in representative histograms in Figure 1A. A comparison of the median fluorescence intensity of monocytes from healthy donors and patients with RA showed significantly higher expression of transmembrane TNF on monocytes from patients with RA (P ⫽ 0.012), while monocytes from healthy donors expressed no or only low levels of transmembrane TNF (Figure 1B). Spontaneous in vitro production of IL-1␤ increased in monocytes from RA patients and inhibited by transmembrane TNF reverse signaling. The cytokine IL-1␤ plays a pivotal role in monocyte biology and has both proinflammatory functions and an antiapoptotic effect. Therefore, in vitro IL-1␤ production in culture supernatants was analyzed by ELISA. After 4 hours of in vitro culture, monocytes from patients with RA spontaneously produced significant amounts of IL-1␤, whereas monocytes from healthy donors did not (Figure 2A). Increased IL-1␤ levels were also observed in sera from patients with RA, while no IL-1␤ was detected in sera from healthy donors (Figure 2B). When RA monocytes were incubated in the presence of infliximab, IL-1␤ concentrations were reduced to levels comparable with those in monocytes from healthy donors (Figure 2A). Correspondingly, protein expression of inactive procaspase 1, the IL-1␤–converting enzyme, and of proIL-1␤ as determined by Western blot analysis was significantly increased following the addition of infliximab (Figure 2C). In contrast, spontaneous production of IL-1Ra did not differ between monocytes from healthy donors and those from patients with RA and was not influenced by the addition of infliximab (Figure 2D). EFFECTS OF ANTI-TNF␣ ANTIBODY ON MONOCYTES FROM PATIENTS WITH RA 2615 optosis during in vitro culture revealed that the percentage of apoptotic cells was significantly decreased in monocytes from patients with RA compared with control subjects (Figure 3A). To eliminate the influence of a prolonged in vitro culture period on this result, freshly separated CD14⫹ peripheral blood monocytes were analyzed ex vivo for apoptosis. Again, monocytes from healthy individuals were more frequently annexin V positive than were monocytes from patients with RA (mean ⫾ SEM 14.3 ⫾ 1.3% [n ⫽ 6] and 10.6 ⫾ 0.6% [n ⫽ 7], respectively; P ⫽ 0.019). Increased in vitro apoptosis of monocytes by transmembrane TNF reverse signaling. In previous studies investigating reverse signaling via transmem- Figure 2. Inhibition of interleukin-1␤ (IL-1␤) production in monocytes from patients with RA by transmembrane TNF reverse signaling. A and D, Production of IL-1␤ (A) and IL-1 receptor antagonist (IL-1Ra) (D) by monocytes from patients with RA (n ⫽ 7) and healthy donors (n ⫽ 4). Cells were incubated for 4 hours in the presence of infliximab (inflix.) or IgG. Bars show the mean and SEM. B, IL-1␤ concentration in sera from healthy donors (n ⫽ 31) and patients with RA (n ⫽ 32), as analyzed by enzyme-linked immunosorbent assay. Bars show the median. C, Expression of procaspase 1 and proIL-1␤ in monocytes from patients with RA, as analyzed by Western blotting after 4 hours of incubation with infliximab or IgG. Results are representative of 3 experiments. See Figure 1 for other definitions. In order to assess the net effect of infliximabinduced transmembrane TNF reverse signaling on the IL-1␤/IL-1Ra system, the ratio of the IL-1␤ to IL-1Ra concentration was calculated. The addition of infliximab to cultures of monocytes from patients with RA was shown to decrease this ratio, corresponding to the decreased IL-1␤ production in monocytes from patients with RA following transmembrane TNF reverse signaling (mean ⫾ SEM 0.024 ⫾ 0.006 without infliximab versus 0.008 ⫾ 0.001 with infliximab; P ⫽ 0.029), indicating that the effects of IL-1␤ were indeed inhibited. In addition, concentrations of IL-6 and IL-8 in the culture supernatants were determined. No significant influence of infliximab on the secretion of those cytokines by monocytes from patients with RA and healthy individuals was detected (data not shown). Decreased spontaneous ex vivo apoptosis of monocytes from patients with RA. Apoptosis of monocytes was investigated by double staining with annexin V and the DNA dye PI, with subsequent analysis by flow cytometry. A comparison of spontaneous monocyte ap- Figure 3. Spontaneous and transmembrane TNF reverse signaling induced–apoptosis in monocytes from patients with RA and control subjects. A, Spontaneous apoptosis during in vitro culture, as evaluated by annexin V and propidium iodide staining, in monocytes from healthy donors (n ⫽ 50) and patients with RA (n ⫽ 48) that were incubated for 16 hours in the presence of IgG. B, Representative dot blots of annexin V– and propidium iodide–stained monocytes from healthy donors and patients with RA after incubation with IgG or infliximab. C, Apoptosis, as evaluated by annexin V and propidium iodide staining, in monocytes from healthy donors (n ⫽ 50) and patients with RA (n ⫽ 48) that were incubated for 16 hours in the presence of infliximab or IgG. Data are presented as box plots, where the boxes represent the 25th to 75th percentiles, the lines within the boxes represent the median, and the lines outside the boxes represent the 10th and 90th percentiles. D, Decreased transmembrane TNF reverse signaling–induced apoptosis of monocytes from patients by specific inhibition of casein kinase 1 (CK-1) activity using D4476. Monocytes from patients with RA were incubated in the presence of IgG or infliximab for 16 hours. Monocytes were preincubated with the CK-1 inhibitor D4476 (n ⫽ 14). Bars in A and D show the mean and SEM. See Figure 1 for other definitions. 2616 brane TNF, investigators claimed that induction of apoptosis is the dominant event following transmembrane TNF reverse signaling (6,8,9). Because IL-1␤ is a major survival factor for monocytes (12), the extent of apoptosis triggered in vitro in monocytes from patients with RA by the addition of infliximab and the resulting IL-1␤ withdrawal was investigated. For that purpose, monocytes from healthy donors and patients with RA who had no history of receiving anti-TNF␣ therapy were isolated from peripheral blood and incubated with 100 g/ml infliximab or control IgG for 16 hours. Apoptosis was calculated as a ratio of the percentage of apoptotic cells in infliximab-treated monocytes to that of IgGtreated monocytes. Monocytes from healthy donors showed no increased apoptosis ratio after incubation with infliximab (mean ⫾ SEM 1.06 ⫾ 0.04; n ⫽ 50), as shown in a representative dot blot in Figure 3B. In contrast, transmembrane TNF reverse signaling induced apoptosis in monocytes from patients with RA (mean ⫾ SEM apoptosis ratio 1.55 ⫾ 0.13; n ⫽ 48), as shown in a representative dot blot in Figure 3B. At all earlier time points analyzed (after 4 hours and 8 hours of incubation), no transmembrane TNF reverse signaling– induced apoptosis was detectable (data not shown). The DNA dye PI was used to differentiate between apoptotic cells and necrotic/late apoptotic cells. In monocytes from patients with RA, transmembrane TNF reverse signaling led to an increase in annexin V staining in both PI-negative and PI-positive cells (mean ⫾ SEM apoptosis ratio 1.55 ⫾ 0.13 and 1.25 ⫾ 0.05, respectively), although the increase was much lower in the PI-positive cells. All further apoptosis ratios were calculated by using all annexin V–positive cells (PI-negative and PI-positive cells). When the ratios obtained for healthy donors and patients with RA were compared, a significantly higher ratio of induced apoptosis following transmembrane TNF reverse signaling was evident in monocytes from patients with RA (Figure 3C). This effect was also noticed when etanercept was used instead of infliximab (mean ⫾ SEM apoptosis ratio 1.86 ⫾ 0.28 versus 1.02 ⫾ 0.05 [n ⫽ 11]; P ⫽ 0.005). However, this increase in the rate of apoptosis reflected primarily the lower initial rate of spontaneous apoptosis in monocytes from patients with RA, while the absolute percentages of apoptotic monocytes did not differ between infliximab-induced apoptosis of monocytes from patients with RA and spontaneous apoptosis of monocytes from control subjects (mean ⫾ SEM apoptosis 30.7 ⫾ 1.7% versus 32.44 ⫾ 2.0%; P ⫽ 0.51). MEUSCH ET AL Induction of apoptosis appeared to depend on the expression of transmembrane TNF, because transmembrane TNF expression on monocytes from patients with RA correlated with the degree of infliximab-induced apoptosis (R ⫽ 0.700, P ⫽ 0.016 [n ⫽ 11]). No such correlation was observed for transmembrane TNF expression and apoptosis induction in monocytes from healthy donors (R ⫽ ⫺0.141, P ⫽ 0.542 [n ⫽ 21]). Importance of CK-1 activity for transmembrane TNF reverse signaling–induced apoptosis. The CK-1 enzyme has been reported to be involved in proximal signal transduction events of reverse signaling following transmembrane TNF ligation. It is responsible for the constitutive phosphorylation of at least 1 serine residue in the cytoplasmic domain of transmembrane TNF, and dephosphorylation of this serine residue by a stillunidentified phosphatase leads to calcium influx (13). When monocytes from patients with RA were pretreated with 2 pan-phosphatase inhibitors, sodium vanadate and sodium fluoride, transmembrane TNF reverse signaling–induced apoptosis was markedly reduced (data not shown). Specific inhibition of CK-1 activity by the selective inhibitor D4476 also led to a decrease in transmembrane TNF reverse signaling–induced apoptosis of RA monocytes (Figure 3D). No involvement of caspases in transmembrane TNF reverse signaling–induced apoptosis. In contrast to our findings, previous reports described infliximabinduced apoptosis in monocytes from healthy individuals and a dependence on this process on the activation of caspases 8, 9, and 3. Therefore, caspase activation in monocytes from patients with RA was analyzed. Monocytes from healthy individuals were not included, because no apoptosis was observed. Under the culture conditions used in the present study, caspases 8, 9, and 3 were not activated, as demonstrated by the absence of cleavage of the inactive caspase pro forms (Figure 4A). Accordingly, global inhibition of caspases by Z-VADFMK and inhibition of caspase 3 by Z-DEVD-FMK did not influence transmembrane TNF reverse signaling– induced apoptosis in monocytes (Figure 4B). Other cellular events during apoptosis are cleavage of the DNA repair enzyme PARP-1 and degradation of DNA. As shown in Figures 4C and D, no cleavage of PARP-1 or DNA degradation was observed in transmembrane TNF reverse signaling–induced apoptosis. In addition, expression of the antiapoptotic protein Bcl-xL was not decreased in transmembrane TNF reverse signaling–induced apoptosis (Figure 4C). In summary, apoptosis following transmembrane TNF reverse signal- EFFECTS OF ANTI-TNF␣ ANTIBODY ON MONOCYTES FROM PATIENTS WITH RA Figure 4. Caspase-independent transmembrane tumor necrosis factor (TNF) reverse signaling–induced apoptosis of monocytes from patients with rheumatoid arthritis (RA). Monocytes from patients with RA were incubated in the presence of infliximab (inflix.) or IgG control for 16 hours. A, Activity of caspase 8, caspase 9, and caspase 3, as determined by immunoblot analysis. Arrows show pro forms and cleaved caspases. Results are representative of 10 independent experiments. B, Influence of global inhibition of caspases by Z-VAD-FMK and inhibition of caspase 3 by Z-DEVD-FMK on transmembrane TNF reverse signaling–induced apoptosis in monocytes. Bars show the mean and SEM. C, Expression of cleaved poly(ADP-ribose) polymerase 1 (PARP-1) and Bcl-xL, as determined by immunoblot analysis. Results are representative of 3 independent experiments. D, DNA content, as determined by propidium iodide staining. The marker is set on subdiploid, apoptotic cells. Results are representative of 5 independent experiments. ing appears to be caspase independent and not accompanied by DNA fragmentation. Inhibition of constitutive NF-B activation by transmembrane TNF reverse signaling. One important survival mechanism for monocytes and macrophages is the constitutive activity of the transcription factor NF-B (14–16), of which IL-1␤ is a strong activator (17). To test whether constitutive NF-B activity is influenced by reverse signaling of transmembrane TNF, the expression of the inhibitory protein IB␣ and its phosphorylation status were determined. After 16 hours of culture, monocytes from patients with RA had low levels of IB␣ (Figure 5A), which was highly phosphorylated (Figure 5B) and therefore subjected to ubiquitinylation and degradation. In contrast, when monocytes from patients with RA were cultured in the presence of infliximab, increased levels of IB␣ were observed (see representative Western blot in Figure 5A), which was poorly phosphorylated (Figure 5B). High levels of IB␣ lead to the retention of NF-B in the cytoplasm. To test whether the observed 2617 Figure 5. Transmembrane TNF reverse signaling–induced IB␣ accumulation and NF-B inhibition in monocytes from patients with RA. Monocytes from patients with RA (n ⫽ 3) were incubated in the presence of infliximab or IgG for 4 hours. A and B, Expression and phosphorylation of IB␣, as determined by immunoblot analysis. Bars show the mean and SEM. Immunoblots show 1 representative experiment. C, Results of electromobility shift assay. The gel shift was present only with wild-type oligonucleotide (wt) and was almost absent with mutant oligonucleotide (mt), demonstrating specificity of the shifted band. Results are representative of 7 experiments. See Figure 4 for other definitions. increase in IB␣ expression is indeed associated with diminished translocation of NF-B into the nucleus, nuclear extracts were prepared and subjected to a NF-B electromobility shift assay. As shown in Figure 5C, NF-B was constitutively translocated into the nucleus in IgG-treated monocytes but was almost absent in Figure 6. Prevention of anti-TNF–induced apoptosis of monocytes from patients with RA by the addition of exogenous interleukin-1␤ (IL-1␤). Monocytes from patients with RA (n ⫽ 3) were incubated in the presence of IgG, infliximab, or infliximab plus soluble IL-1␤ (100 pg/ml) for 16 hours. Bars show the mean and SEM. See Figure 4 for other definitions. 2618 MEUSCH ET AL infliximab-treated monocytes. The specificity of the observed NF-B:probe complex was verified by use of a mutant probe that has a disrupted NF-B binding site. Inhibition of transmembrane TNF reverse signaling–induced apoptosis by the addition of IL-1␤. To investigate the potential influence of IL-1␤ withdrawal on the rate of apoptosis, IL-1␤ was added to the infliximab-treated monocyte cultures to simulate the IL-1␤ levels in IgG-treated cultures. As shown in Figure 6, the addition of IL-1␤ completely reversed the increase in transmembrane TNF reverse signaling–induced in vitro apoptosis, again supporting the notion of an IL-1␤ deficit as the underlying mechanism of apoptosis induction in RA monocytes. DISCUSSION In the present study, we provide evidence demonstrating that monocytes from patients with RA express high amounts of transmembrane TNF and show strong spontaneous in vitro IL-1␤ production. Triggering of reverse signaling of transmembrane TNF by ligation with an anti-TNF antibody suppressed constitutive NF-B activation and inhibited this spontaneous, constitutive production of IL-1␤, which in turn led to increased in vitro apoptosis of RA monocytes. Here, we demonstrate for the first time the expression of transmembrane TNF on freshly isolated, resting monocytes from patients with RA. The correlation between the observed transmembrane TNF reverse signaling–triggered apoptosis in RA monocytes and the determined transmembrane TNF expression levels indicates that transmembrane TNF reverse signaling is indeed involved in apoptotic cell death in vitro. Inhibition of CK-1, an important kinase necessary for reverse signaling of transmembrane TNF, led to a decrease in infliximab-induced apoptosis, further pointing to transmembrane TNF reverse signaling as the mechanism mediating infliximab-induced apoptosis. Until now, transmembrane TNF reverse signaling–induced apoptosis of monocytes from RA patients has not been compared with that of monocytes from healthy individuals, while several studies have analyzed monocytes from patients with Crohn’s disease (8). Only one other study investigating infliximabinduced apoptosis in monocytes from patients with RA demonstrated increased annexin V expression after incubation with infliximab or etanercept, but this finding did not reach statistical significance (7). There have been studies on the ability of infliximab to induce apoptosis in vivo, mainly in the synovial membrane of treated RA patients, but results have been conflicting. Recently, Wijbrandts et al did not observe any apoptosis in vivo after infliximab infusion (18), whereas Catrina et al described increased synovial apoptosis, mainly in the monocyte/macrophage compartment (7). The network of proapoptotic and antiapoptotic factors in vivo is certainly more complex than that in in vitro assays, however, and factors that are still unknown might contribute to the observed discrepancies. It has been shown in several studies that incubation of primary human monocytes from healthy donors with infliximab leads to the induction of apoptosis (6,8,9), although the opposite has also been reported (19). We were unable to detect transmembrane TNF reverse signaling–induced apoptosis in monocytes from healthy individuals. Differences in the experimental systems used and different monocyte purification methods (positive versus negative selection) might contribute to the discrepant results. The demonstrated lack of transmembrane TNF expression on the surface of monocytes from healthy control subjects can readily explain the inability of infliximab to induce apoptosis, however. This absence of transmembrane TNF expression on resting monocytes has been reported previously (20,21), although other reports have been conflicting (6,8,22), again possibly due to the different purification methods discussed above. Monocytes undergo spontaneous apoptosis when cultured in vitro (23,24). This spontaneous apoptosis can be prevented by activation of monocytes with LPS or by the addition of proinflammatory cytokines such as TNF and IL-1␤ (12,23). Our results demonstrate that monocytes from patients with RA display a decreased level of spontaneous apoptosis when cultured in vitro. Simultaneously, we observed spontaneous production of IL-1␤, which was not detected in cultures of monocytes from healthy individuals. Defects in macrophage apoptosis have been observed previously in the synovial membrane of patients with RA (25–27), but not in peripheral blood monocytes (25). We observed the most significant difference in monocyte apoptosis between RA patients and control subjects in the spontaneous apoptosis during in vitro culture, which has not been previously reported. A study involving Bim-deficient mice, which have more severe disease, also suggested a possible pathogenetic role for defective apoptosis in the exacerbation of inflammatory arthritis (28). Although an intrinsic defect cannot be excluded as the underlying reason for the decreased apoptosis, it appears equally possible that predifferentiation or preactivation occurs in vivo in EFFECTS OF ANTI-TNF␣ ANTIBODY ON MONOCYTES FROM PATIENTS WITH RA patients with RA, which accounts for the increase in transmembrane TNF and IL-1␤ and the decrease in spontaneous apoptosis. Due to the redundant functions of soluble and transmembrane TNF, and due to the possibility of forward and reverse signaling of transmembrane TNF either via TNFR1/2 or via the cytoplasmic domain of transmembrane TNF itself, we cannot entirely exclude the inhibition of an autocrine effect of secreted TNF␣ by infliximab as the underlying mechanism of apoptosis induction in vitro. However, because CK-1 inhibited the observed induction of apoptosis, which correlated strongly with monocyte transmembrane TNF expression, and because the addition of exogenous IL-1␤ reversed the induction of apoptosis observed in patients with RA, it appears unlikely that this neutralization of autocrine TNF␣ is a major inductor of monocyte apoptosis in vitro. An inhibitory effect of reverse signaling via transmembrane TNF on LPS-induced IL-1␤ production has already been reported for a monocytic cell line (4,29). The published observation of decreased spontaneous ex vivo IL-1␤ production of monocytes from patients with RA 24 hours after infliximab infusion (30) also corroborates our finding that transmembrane TNF reverse signaling inhibits spontaneous IL-1␤ production in RA. The inhibition of apoptosis reported in this study was reversed by addition of IL-1␤ to infliximab-treated monocyte cultures, which again implies inhibition of the survival factor IL-1␤ as a key mechanism in transmembrane TNF reverse signaling–induced apoptosis in RA monocytes. A similar mechanism has been described for glucocorticoid-induced apoptosis in monocytes, with inhibition of IL-1␤ production by glucocorticoids and prevention of glucocorticoid-induced apoptosis achieved by the addition of IL-1␤ (31). Similarly, inhibition of spontaneous apoptosis of neutrophils has been reported to occur after the addition of IL-1␤ (32). Several signal transduction mechanisms for transmembrane TNF reverse signaling–induced apoptosis have been described. These include the activation of caspase 8, caspase 9, and caspase 3, mitochondrial release of cytochrome c, expression of the proapoptotic proteins Bak and Bax, and PARP-1 cleavage in monocytes (6,8,9) and monocytic cell lines (33). To our surprise, we were not able to observe activation of caspase 8, caspase 9, and caspase 3, PARP-1 cleavage, and degradation of DNA after incubation of monocytes with infliximab. Furthermore, inhibition of caspases did not prevent transmembrane TNF reverse 2619 signaling–induced apoptosis in monocytes from patients with RA. Although the activation of caspases is the hallmark of apoptosis, reports on caspase-independent apoptosis have recently gained increasing attention (34– 36). Here, we demonstrate that the mechanism of apoptosis induction following transmembrane TNF reverse signaling is largely caspase independent. We propose that the inhibition of constitutive NF-B activation and subsequent suppression of IL-1␤ secretion is a relevant mechanism of transmembrane TNF reverse signaling–induced apoptosis. The canonical NF-B activation pathway is primarily involved in de novo protein synthesis under stimulatory conditions. However, NF-B is also constitutively active in many cell types, including monocytes and macrophages (14–16), and has antiapoptotic effects due to the induction of genes such as cellular inhibitors of apoptosis, FLICE inhibitory protein, TNFRassociated factor 1 (TRAF1), and TRAF2 (37) in those cells. Accordingly, inhibition of constitutive NF-B activation has been shown to induce apoptosis in a variety of cell types under various conditions, including leukemia cell lines (38–40), prostate cancer cells (41), multiple myeloma cells (42), and mantle cell lymphoma cells (43), but also human T lymphotropic virus type I–infected T cell lines and primary adult T cell leukemia cells (44). Our results demonstrated translocated NF-B in the nucleus of resting monocytes from patients with RA and a decrease in translocated NF-B after incubation of monocytes with infliximab, while expression of IB␣ increased. We hypothesize, therefore, that inhibition of NF-B followed by caspase-independent induction of apoptosis is an important signal transduction mechanism event leading to the apoptotic signal elicited by transmembrane TNF reverse signaling. Interestingly, survival of the monocytes could be restored by addition of IL-1␤, pointing to an autocrine, NF-B–dependent survival signal for monocytes that can be interrupted by transmembrane TNF reverse signaling. Accordingly, this induction of monocyte apoptosis could be an alternative antiinflammatory mechanism of action of TNF inhibitors, with particular relevance at sites of monocyte activation or monocytic cytokine production, such as the rheumatoid synovial membrane. In addition, the observed inhibition of constitutive NF-B activation following transmembrane TNF reverse signaling is likely to down-regulate the expression of NF-B–dependent proinflammatory cytokines or chemokines, which might 2620 MEUSCH ET AL contribute to the observed clinical effects of anti-TNF treatment. 13. AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Wagner had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Meusch, Rossol, Baerwald, Wagner. Acquisition of data. Meusch, Rossol. Analysis and interpretation of data. Meusch, Rossol, Baerwald, Hauschildt, Wagner. 14. 15. 16. 17. REFERENCES 18. 1. Mitoma H, Horiuchi T, Tsukamoto H, Tamimoto Y, Kimoto Y, Uchino A, et al. Mechanisms for cytotoxic effects of anti–tumor necrosis factor agents on transmembrane tumor necrosis factor ␣–expressing cells: comparison among infliximab, etanercept, and adalimumab. Arthritis Rheum 2008;58:1248–57. 2. Rossol M, Meusch U, Pierer M, Kaltenhauser S, Hantzschel H, Hauschildt S, et al. Interaction between transmembrane TNF and TNFR1/2 mediates the activation of monocytes by contact with T cells. J Immunol 2007;179:4239–48. 3. Harashima S, Horiuchi T, Hatta N, Morita C, Higuchi M, Sawabe T, et al. Outside-to-inside signal through the membrane TNF-␣ induces E-selectin (CD62E) expression on activated human CD4⫹ T cells. J Immunol 2001;166:130–6. 4. Eissner G, Kirchner S, Lindner H, Kolch W, Janosch P, Grell M, et al. Reverse signaling through transmembrane TNF confers resistance to lipopolysaccharide in human monocytes and macrophages. J Immunol 2000;164:6193–8. 5. Kirchner S, Holler E, Haffner S, Andreesen R, Eissner G. Effect of different tumor necrosis factor (TNF) reactive agents on reverse signaling of membrane integrated TNF in monocytes. Cytokine 2004;28:67–74. 6. Shen C, Assche GV, Colpaert S, Maerten P, Geboes K, Rutgeerts P, et al. Adalimumab induces apoptosis of human monocytes: a comparative study with infliximab and etanercept. Aliment Pharmacol Ther 2005;21:251–8. 7. Catrina AI, Trollmo C, af Klint E, Engstrom M, Lampa J, Hermansson Y, et al. Evidence that anti–tumor necrosis factor therapy with both etanercept and infliximab induces apoptosis in macrophages, but not lymphocytes, in rheumatoid arthritis joints: extended report. Arthritis Rheum 2005;52:61–72. 8. Lugering A, Schmidt M, Lugering N, Pauels HG, Domschke W, Kucharzik T. Infliximab induces apoptosis in monocytes from patients with chronic active Crohn’s disease by using a caspasedependent pathway. Gastroenterology 2001;121:1145–57. 9. Waetzig GH, Rosenstiel P, Arlt A, Till A, Brautigam K, Schafer H, et al. Soluble tumor necrosis factor (TNF) receptor-1 induces apoptosis via reverse TNF signaling and autocrine transforming growth factor-␤1. FASEB J 2005;19:91–3. 10. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, et al. The American Rheumatism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 1988;31:315–24. 11. Rossol M, Kaltenhauser S, Scholz R, Hantzschel H, Hauschildt S, Wagner U. The contact-mediated response of peripheral-blood monocytes to preactivated T cells is suppressed by serum factors in rheumatoid arthritis. Arthritis Res Ther 2005;7:R1189–99. 12. Mangan DF, Welch GR, Wahl SM. Lipopolysaccharide, tumor necrosis factor-␣, and IL-1␤ prevent programmed cell death 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. (apoptosis) in human peripheral blood monocytes. J Immunol 1991;1461541–6. Watts AD, Hunt NH, Wanigasekara Y, Bloomfield G, Wallach D, Roufogalis BD, et al. A casein kinase I motif present in the cytoplasmic domain of members of the tumour necrosis factor ligand family is implicated in ‘reverse signalling’. EMBO J 1999; 18:2119–26. Farver CF, Raychaudhuri B, Buhrow LT, Connors MJ, Thomassen MJ. Constitutive NF-B levels in human alveolar macrophages from normal volunteers. Cytokine 1998;10:868–71. Frankenberger M, Pforte A, Sternsdorf T, Passlick B, Baeuerle PA, Ziegler-Heitbrock HW. Constitutive nuclear NF-B in cells of the monocyte lineage. Biochem J 1994;304(Pt 1):87–94. Pagliari LJ, Perlman H, Liu H, Pope RM. Macrophages require constitutive NF-B activation to maintain A1 expression and mitochondrial homeostasis. Mol Cell Biol 2000;20:8855–65. Ghosh S, Karin M. Missing pieces in the NF-B puzzle. Cell 2002;109(Suppl):S81–96. Wijbrandts CA, Remans PH, Klarenbeek PL, Wouters D, van den Bergh Weerman MA, Smeets TJ, et al. Analysis of apoptosis in peripheral blood and synovial tissue very early after initiation of infliximab treatment in rheumatoid arthritis patients. Arthritis Rheum 2008;58:3330–9. Ringheanu M, Daum F, Markowitz J, Levine J, Katz S, Lin X, et al. Effects of infliximab on apoptosis and reverse signaling of monocytes from healthy individuals and patients with Crohn’s disease. Inflamm Bowel Dis 2004;10:801–10. Caron G, Delneste Y, Aubry JP, Magistrelli G, Herbault N, Blaecke A, et al. Human NK cells constitutively express membrane TNF-␣ (mTNF␣) and present mTNF␣-dependent cytotoxic activity. Eur J Immunol 1999;29:3588–95. Kriegler M, Perez C, DeFay K, Albert I, Lu SD. A novel form of TNF/cachectin is a cell surface cytotoxic transmembrane protein: ramifications for the complex physiology of TNF. Cell 1988;53: 45–53. Robertshaw HJ, Brennan FM. Release of tumour necrosis factor ␣ (TNF␣) by TNF␣ cleaving enzyme (TACE) in response to septic stimuli in vitro. Br J Anaesth 2005;94:222–8. Mangan DF, Wahl SM. Differential regulation of human monocyte programmed cell death (apoptosis) by chemotactic factors and pro-inflammatory cytokines. J Immunol 1991;147:3408–12. 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. 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 enzymeinhibitory protein and are refractory to Fas-mediated apoptosis. Arthritis Rheum 2001;44:21–30. Ceponis A, Hietanen J, Tamulaitiene M, Partsch G, Patiala H, Konttinen YT. A comparative quantitative morphometric study of cell apoptosis in synovial membranes in psoriatic, reactive and rheumatoid arthritis. Rheumatology (Oxford) 1999;38:431–40. Sugiyama M, Tsukazaki T, Yonekura A, Matsuzaki S, Yamashita S, Iwasaki K. Localisation of apoptosis and expression of apoptosis related proteins in the synovium of patients with rheumatoid arthritis. Ann Rheum Dis 1996;55:442–9. Scatizzi JC, Bickel E, Hutcheson J, Haines GK III, Perlman H. Bim deficiency leads to exacerbation and prolongation of joint inflammation in experimental arthritis. Arthritis Rheum 2006;54: 3182–93. Nesbitt A, Fossati G, Bergin M, Stephens P, Stephens S, Foulkes R, et al. Mechanism of action of certolizumab pegol (CDP870): in vitro comparison with other anti-tumor necrosis factor ␣ agents. Inflamm Bowel Dis 2007;13:1323–32. Schuerwegh AJ, Van Offel JF, Stevens WJ, Bridts CH, De Clerck EFFECTS OF ANTI-TNF␣ ANTIBODY ON MONOCYTES FROM PATIENTS WITH RA 31. 32. 33. 34. 35. 36. 37. 38. LS. Influence of therapy with chimeric monoclonal tumour necrosis factor-␣ antibodies on intracellular cytokine profiles of T lymphocytes and monocytes in rheumatoid arthritis patients. Rheumatology (Oxford) 2003;42:541–8. Schmidt M, Pauels HG, Lugering N, Lugering A, Domschke W, Kucharzik T. Glucocorticoids induce apoptosis in human monocytes: potential role of IL-1␤. J Immunol 1999;163:3484–90. McNamee JP, Bellier PV, Kutzner BC, Wilkins RC. Effect of pro-inflammatory cytokines on spontaneous apoptosis in leukocyte sub-sets within a whole blood culture. Cytokine 2005;31:161–7. Shen C, Maerten P, Geboes K, Van Assche G, Rutgeerts P, Ceuppens JL. Infliximab induces apoptosis of monocytes and T lymphocytes in a human-mouse chimeric model. Clin Immunol 2005;115:250–9. Johansson U, Higginbottom K, Londei M. CD47 ligation induces a rapid caspase-independent apoptosis-like cell death in human monocytes and dendritic cells. Scand J Immunol 2004;59:40–9. Mercalli A, Sordi V, Ponzoni M, Maffi P, De Taddeo F, Gatti G, et al. Rapamycin induces a caspase-independent cell death in human monocytes. Am J Transplant 2006;6:1331–41. Drenou B, Blancheteau V, Burgess DH, Fauchet R, Charron DJ, Mooney NA. A caspase-independent pathway of MHC class II antigen-mediated apoptosis of human B lymphocytes. J Immunol 1999;163:4115–24. Liu ZG, Hsu H, Goeddel DV, Karin M. Dissection of TNF receptor 1 effector functions: JNK activation is not linked to apoptosis while NF-B activation prevents cell death. Cell 1996; 87:565–76. Fabre C, Carvalho G, Tasdemir E, Braun T, Ades L, Grosjean J, 39. 40. 41. 42. 43. 44. 2621 et al. NF-B inhibition sensitizes to starvation-induced cell death in high-risk myelodysplastic syndrome and acute myeloid leukemia. Oncogene 2007;26:4071–83. Pickering BM, de Mel S, Lee M, Howell M, Habens F, Dallman CL, et al. Pharmacological inhibitors of NF-B accelerate apoptosis in chronic lymphocytic leukaemia cells. Oncogene 2007;26: 1166–77. Hansson A, Marin YE, Suh J, Rabson AB, Chen S, Huberman E, et al. Enhancement of TPA-induced growth inhibition and apoptosis in myeloid leukemia cells by BAY 11-7082, an NF-B inhibitor. Int J Oncol 2005;27:941–8. Zheng X, Chang RL, Cui XX, Avila G, Huang MT, Liu Y, et al. Inhibition of NF-B by (E)3-[(4-methylphenyl)-sulfonyl]-2-propenenitrile (BAY11-7082; BAY) is associated with enhanced 12-O-tetradecanoylphorbol-13-acetate-induced growth suppression and apoptosis in human prostate cancer PC-3 cells. Int J Oncol 2008;32:257–64. Dai Y, Pei XY, Rahmani M, Conrad DH, Dent P, Grant S. Interruption of the NF-B pathway by Bay 11-7082 promotes UCN-01-mediated mitochondrial dysfunction and apoptosis in human multiple myeloma cells. Blood 2004;103:2761–70. Pham LV, Tamayo AT, Yoshimura LC, Lo P, Ford RJ. Inhibition of constitutive NF-B activation in mantle cell lymphoma B cells leads to induction of cell cycle arrest and apoptosis. J Immunol 2003;171:88–95. Mori N, Yamada Y, Ikeda S, Yamasaki Y, Tsukasaki K, Tanaka Y, et al. Bay 11-7082 inhibits transcription factor NF-B and induces apoptosis of HTLV-I-infected T cell lines and primary adult T cell leukemia cells. Blood 2002;100:1828–34.