Molecular Determinants of Kinase Pathway Activation by Apo2 Ligand ...

THE JOURNAL OF BIOLOGICAL CHEMISTRY VOL. 280, NO. 49, pp. 40599 –40608, December 9, 2005 © 2005 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in the U.S.A.

Molecular Determinants of Kinase Pathway Activation by Apo2 Ligand/Tumor Necrosis Factor-related Apoptosis-inducing Ligand*□ S

Received for publication, August 30, 2005, and in revised form, October 14, 2005 Published, JBC Papers in Press, October 15, 2005, DOI 10.1074/jbc.M509560200

Eugene Varfolomeev, Heather Maecker, Darcie Sharp, David Lawrence, Mark Renz, Domagoj Vucic, and Avi Ashkenazi1 From the Department of Molecular Oncology, Genentech, Inc., South San Francisco, California 94080 Apo2 ligand/tumor necrosis factor (TNF)-related apoptosis-inducing ligand (Apo2L/TRAIL) mainly activates programmed cell death through caspases. By contrast, TNF primarily induces gene transcription through the inhibitor of ␬B kinase (IKK), c-Jun N-terminal kinase (JNK), and p38 mitogen-activated protein kinase pathways. Apo2L/TRAIL also can stimulate these kinases, albeit less strongly; however, the underlying mechanisms of this stimulation and its relation to apoptosis are not well understood. Here we show that Apo2L/TRAIL activates kinase pathways by promoting the association of a secondary signaling complex, subsequent to assembly of a primary, death-inducing signaling complex (DISC). The secondary complex retained the DISC components FADD and caspase-8, but recruited several factors involved in kinase activation by TNF, namely, RIP1, TRAF2, and NEMO/IKK␥. Secondary complex formation required Fas-associated death domain (FADD), as well as caspase-8 activity. Apo2L/TRAIL stimulation of JNK and p38 further depended on RIP1 and TRAF2, whereas IKK activation required NEMO. Apo2L/TRAIL induced secretion of interleukin-8 and monocyte chemoattractant protein-1, augmenting macrophage migration. Thus, Apo2L/TRAIL and TNF organize common molecular determinants in distinct signaling complexes to stimulate similar kinase pathways. One function of kinase stimulation by Apo2L/ TRAIL may be to promote phagocytic engulfment of apoptotic cells.

Members of the tumor necrosis factor (TNF)2 superfamily regulate many physiological and pathological aspects of immune system development and function (reviewed in Refs. 1–5). The TNF ligand superfamily is defined by structural similarity and ability to recognize corresponding TNF receptor (TNFR) superfamily members, which share homology in their extracellular, cysteine-rich domains. The intracellular domains of TNFRs are devoid of enzymatic activity: The majority

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. □ S The on-line version of this article (available at http://www.jbc.org) contains supplemental Figs. S1–S5. 1 To whom correspondence should be addressed: Dept. of Molecular Oncology, Genentech, Inc., 1 DNA Way, South San Francisco, CA 94080. Tel.: 650-225-1853; Fax: 650467-8195; E-mail: [email protected]. 2 The abbreviations used are: TNF, tumor necrosis factor; TNFR, TNF receptor; Ab, antibody; Apo2L, Apo2L ligand; cIAP, cellular inhibitor of apoptosis protein; DISC, deathinducing signaling complex; FADD, Fas-associated protein with death domain; FasL, Fas ligand; FBS, fetal bovine serum; GST, glutathione S-transferase; FLIP, cellular FLICE inhibitory protein; I␬B, inhibitor of nuclear factor-␬B; JNK, c-Jun amino-terminal kinase; NF-␬B, nuclear factor-␬B; RPMI, Roswell Park Memorial Institute; siRNA, small interfering RNA; RIP, receptor interacting protein; TRADD, TNF receptor-associated death domain; TRAF, TNF receptor-associated factor; TRAIL, tumor necrosis factorrelated apoptosis-inducing ligand; zVAD, benzyloxycarbonyl-Val-Ala-Asp (OMe)-fluoromethyl ketone; IKK, inhibitor of ␬B kinase; MAPK, mitogen-activated protein kinase; MEF, mouse embryo fibroblast; SAPK, stress-activated protein kinase; CHI, ; IL-8, interleukin-8; IB, immunoblot; MCP-1, monocyte chemoattractant protein-1.

DECEMBER 9, 2005 • VOLUME 280 • NUMBER 49

contain an amino acid sequence motif that mediates interaction with TNFR-associated factors (TRAFs). Conversely, a minority of TNFRs share an intracellular “death domain” motif, which interacts with adaptor molecules that contain related death domains. The signaling activities and biological functions of the death-domain-containing TNFR subgroup are diverse. A key signaling activity of Fas, which binds to Fas ligand (FasL), and of DR4 and DR5, which bind to Apo2L/TRAIL, is apoptosis induction. On the other hand, TNFR1, which binds TNF and LT-␣, primarily regulates transcription of pro-inflammatory and immunomodulatory genes and contributes to cell death signaling only in unique situations. The apoptosis-signaling events that the death-inducing ligands FasL and Apo2L/TRAIL trigger are relatively well defined (reviewed in Refs. 6 – 8). Binding of Fas, DR4, or DR5 by cognate ligand leads to conformational changes that promote interaction between the death domain of the receptor and the homologous region of the adaptor molecule Fas-associated death domain (FADD). FADD also contains a “death effector domain,” which mediates recruitment of two death effector domain-containing apical proteases: caspase-8 and caspase-10. This primary signaling complex, consisting of ligand, receptor, FADD, and apical caspase(s), is dubbed the death-inducing signaling complex (DISC) (9). Caspase-8 and -10 belong to the cysteine protease family, which plays a crucial role in programmed cell death. In living cells, caspase-8 and -10 are expressed as inactive precursors; in the DISC, these caspases become activated by oligomerization. Once activated, caspase-8 and -10 undergo proteolytic self-processing, in turn cleaving and activating downstream effector caspases, such as caspase-3, -6, and -7, which mediate the execution phase of apoptotic cell death (10, 11). In certain cell types, death receptor stimulation of caspase-8 and -10 triggers sufficient effector caspase activation to commit the cell to apoptotic death. In others, death requires further amplification of the apoptosis signal. This augmentation can be initiated by caspase-8-mediated cleavage of the proximal pro-apoptotic Bcl-2 family member Bid. The truncated form of Bid activates the distal pro-apoptotic Bcl-2 family members Bax and Bak, which trigger mitochondrial release of factors that promote activation of the apical protease caspase-9. In turn, caspase-9 stimulates further effector caspase activity, ensuring completion of the cell death program. A cellular factor called FLICE inhibitory protein (FLIP), which bears structural similarity to caspase-8 and -10 but lacks caspase activity, can inhibit death receptor-mediated apoptosis signaling (12). FLIP occurs in both short (FLIPS) and long (FLIPL) alternative mRNA splicing forms. Both forms can compete for apical caspase recruitment to the DISC, whereas FLIPL also can inhibit the full processing of caspase-8 and -10. In contrast to the apoptosis-initiating events triggered by binding of FasL and Apo2L/TRAIL to their corresponding death receptors, TNF binding to TNFR1 leads to activation of the IKK/NF-␬B, JNK, and p38 mitogen-activated protein kinase (MAPK) signaling pathways (13). Upon ligation, TNFR1 binds through its death domain to the adaptor

JOURNAL OF BIOLOGICAL CHEMISTRY

40599

Kinase Pathway Activation by Apo2L/TRAIL molecule TNFR-associated death domain (TRADD). TRADD in turn binds two other adpators: RIP1, which also contains a death domain and mediates activation of the IKK and p38 pathways, and TRAF2, which supports activation of IKK and JNK. TNFR1 activates the NF-␬B pathway by inducing TRAF2-dependent formation of the signalosome complex, in which the scaffold protein NEMO/IKK␥ and two kinases, IKK␣ and IKK␤, are key catalytic components (14). IKK signalosome activation results in phosphorylation of the inhibitor of NF-␬B (I␬B), which in resting cells binds to NF-␬B subunits and keeps them in the cytoplasm. Phosphorylation triggers proteosomal degradation of I␬B, liberating NF-␬B to move to the nucleus, where it can promote gene transcription. TRAF2 recruits two additional molecules, cellular inhibitor of apoptosis 1 (cIAP1) and cIAP2, to the TNFR1 complex, but the role of these cIAPs in TNFR1 signaling remains unclear. In most cellular contexts, TNF activates the IKK, JNK, and p38 kinase pathways but not apoptotic caspase pathways. Induction of several NF-␬Bdependent anti-apoptotic genes by TNF, including FLIP, cIAP1, cIAP2, XIAP, Bcl-XL, and A1, prevents apoptosis activation; however, under certain conditions, such as general inhibition of protein synthesis or specific blockade of NF-␬B activation, TNF can stimulate a strong pro-apoptotic signal (reviewed in Refs. 15 and 16). Apoptosis initiation by TNF relies on the formation of a secondary intracellular signaling complex, composed of TRADD, RIP1, and TRAF2, as well as FADD and caspase-8 (17). Although the strongest signaling activity of FasL and Apo2L/TRAIL is apoptosis induction in susceptible cells, there is accumulating evidence for the ability of these ligands to activate the IKK, JNK, and p38 pathways (reviewed in Refs. 18 –23). NF-␬B stimulation by FasL and Apo2L/TRAIL is of particular interest, because of the anti-apoptotic activity of this transcription factor. Indeed, some tumors display a correlation between NF-␬B activity and resistance to death-inducing ligands. Moreover, in some cancer cell lines, inhibition of NF-␬B activity by various molecular approaches reverts resistance to cell death signaling by FasL and Apo2L/TRAIL (reviewed in Refs. 15, 24). However, the biological role and significance of kinase pathway activation in the context of apoptosis signaling by FasL and Apo2L/TRAIL are not well understood. Furthermore, in contrast to the well-defined events mediating caspase stimulation by these ligands, the molecular determinants underlying their kinase-activating function have been controversial. Published studies with FADD or caspase-8 null mice focused on apoptosis signaling and did not examine kinase pathway activation in detail (25, 26). Ectopic expression of dominant-negative FADD or caspase-8 mutants did not interfere with activation of NF-␬B or JNK by TNF, FasL, or Apo2L/TRAIL (27). In contrast, detailed examination of mutant Jurkat T cell lines with deficiency in FADD or caspase-8 revealed an absolute necessity of these proteins for apoptosis as well as kinase pathway activation by death receptor ligands (28 –31). More recent experiments utilizing siRNA-based gene silencing techniques have confirmed the requirement of FADD and caspase-8 for kinase pathway activation by FasL, although Apo2L/TRAIL was not investigated (32). Additional studies yielded conflicting data supporting or excluding a direct involvement of TRAF2 and RIP1 in activation of JNK and NF-␬B by FasL and Apo2L/TRAIL (31–34). Although important insights into kinase pathway activation by FasL and Apo2L/TRAIL have been achieved to date, we aimed in this present study to attain further elucidation of the molecular components and events that mediate this function, specifically as it pertains to Apo2L/ TRAIL. Kinase pathway adaptors such as TRADD and RIP1 were reported to interact with the Apo2L/TRAIL receptors DR4 and DR5 in overexpression model systems (reviewed in Ref. 2). In DR4/5-transfected HEK 293 and HeLa cells, RIP1 is associated with unstimulated

40600 JOURNAL OF BIOLOGICAL CHEMISTRY

DR4 and DR5, increases in abundance after ligand stimulation, and undergoes caspase-dependent cleavage (35). However, studies that examined associations between endogenous cellular components did not confirm recruitment of TRADD, RIP1, or TRAF2 to the Apo2L/ TRAIL DISC (22, 36 –38). Here, we demonstrate the formation of a secondary signaling complex subsequent to assembly of the Apo2L/ TRAIL DISC and identify several components of this complex that are crucial for kinase pathway activation. In addition, we provide evidence that kinase stimulation by Apo2L/TRAIL is associated with increased chemokine production and macrophage migration, suggesting that one role of kinase activation by this pro-apoptotic ligand may be to promote the recruitment of phagocytes to the site of apoptosis.

MATERIALS AND METHODS Cell Lines and Reagents—HT1080 human fibrosarcoma cells, A549 human lung carcinoma cells, SK-MES-1 human lung squamous carcinoma cells, human melanoma UACC62 cells, human colorectal adenocarcinoma HT-29 and HCT15 cells, and human hystiocytic carcinoma HeLa cells were obtained from the ATCC. TRAF2-deficient and matched wild-type mouse embryo fibroblasts were kindly provided by Dr. David Goeddel. Jurkat wild-type (A3 clone), FADD-deficient (E1), and caspase-8-deficient (I9.3) cells were kindly provided by Dr. John Blenis. Adherent cell lines were grown in 50:50 Dulbecco’s modified Eagle’s and FK12 medium supplemented with 10% FBS, penicillin, and streptomycin. Jurkat cell lines were grown in RPMI medium. Human recombinant soluble Apo2L/TRAIL in non-tagged or FLAG-tagged versions, FLAG-tagged FasL, and anti-DR5 (3H3) biotinylated monoclonal antibodies were prepared as described (Sharp et al. (39)). Human recombinant soluble TNF was from Genentech Inc. For immunoprecipitation experiments the following Abs were used: anti-FLAG (M2, Sigma); anti-FADD (no. 06-711) and anti-TRADD (3E11, no. 05-473) from Upstate Biotechnology, NY; anti-RIP1 (G322-2, no. 551042) from BD Bioscience; anti-NEMO (FL-419, no. sc-8330), anti-TRAF2 (C-20, no. sc-876), and anti-JNK1 (C-17, no. sc-474) from Santa Cruz Biotechnology, Inc., Santa Cruz, CA. For immunoblot analysis the following antibodies were used: anti-FADD (no. 610399), anti-RIP1 (no. 610458), and anti-JNK1 (G151-333, no. 554286) from BD Bioscience; anticaspase-8 (for DISC experiments, 5F7, no. IM3148) from Immunotech, France; anti-Caspase-8 (1C12, no. 9746), anti-phospho-SAPK/JNK (G9, no. 9255), anti-SAPK/JNK (no. 9258), anti-phospho-p38 MAPK (no. 9211), anti-p38 MAPK (no. 9212), anti-phospho-I␬B␣ (no. 9241), antiI␬B␣ (no. 9242) from Cell Signaling Technology, Inc.; anti-JNK2 (no. AF1846) from R&D Systems, Inc; anti-TRAF2 (H-249, no. sc-7187) from Santa Cruz Biotechnology, Inc., CA; and anti-Actin (C4, no. 69100) from ICN Biomedicals, Inc., OH. As the secondary reagents the following horseradish peroxidase-conjugated Abs were used: anti-mouse-IgG1 (no. 559626) from BD Bioscience; anti-mouse-IgG2a and anti-mouse-IgG2b (no. 190-05) from Southern Biotechnology Associates; anti-rabbit-IgG (no. 711-056-152) from Jackson; horseradish peroxidase-conjugated streptavidin from Amersham Biosciences. Recombinant c-Jun fusion protein (no. 6093) was purchased from Cell Signaling Technology, Inc. Immunoprecipitations, DISC, and Secondary Signaling Complex Analysis—Cells (5–10 ⫻ 107 cells/time point) were treated with FLAGApo2L/TRAIL cross-linked by anti-FLAG M2 Abs, collected, and lysed, and the DISC was analyzed as described (37). For the analysis of the Apo2L/TRAIL-dependent secondary signaling complex, after immunoprecipitation of the DISC, DISC-depleted cell lysates were subjected to second round of immunoprecipitation using anti-FADD, -RIP1, -TRAF2, or -NEMO Abs (2 ␮g/ml), followed by IB analysis of co-immunoprecipitated FADD, caspase-8, or RIP1.

VOLUME 280 • NUMBER 49 • DECEMBER 9, 2005

Kinase Pathway Activation by Apo2L/TRAIL Gene Silencing Experiments—Sequences of siRNA oligonucleotides were designed by using the Dharmacon siDESIGN Center (Dharmacon Research Inc., Lafayette, CO) software and synthesized at Genentech Inc. The following siRNA pairs were used for gene knockdown experiments: FADD, 5⬘-CACAGAGAAGGAGAACGCA and 5⬘-UGCGUUCUCCUUCUCUGUG (pair no. 4); RIP1, 5⬘-CAGCCUGGUUCACUGCACA and 5⬘-UGUGCAGUGAACCAGGCUG (no. 16), 5⬘-CCACUAGUCUGACGGAUAA and UUAUCCGUCAGACUAGUGG (no. 70), 5⬘-GAAAGAGUAUUCAAACGAA and 5⬘-UUCGUUUGAAUACUCUUUC (no. 71), 5⬘-GAAAGAGUAUUCAAACGAA and 5⬘-UUCGUUUGAAUACUCUUUC (no. 178); caspase-8, 5⬘-UGAAGAUAAUCAACGACUA and 5⬘-UAGUCGUUGAUUAUCUUCA (no. 86), 5⬘GUAUACCUGUUGAGACUGA and 5⬘-UCAGUCUCAACAGGUAUAC (no. 87), 5⬘-GAUUUAUCAUCACCUCAAA and 5⬘-UUUGAGGUGAUGAUAAAUC (no. 88); TRAF2, 5⬘-GUGUCGAGUCCCUUGCAGA and 5⬘-UCUGCAAGGGACUCGACAC (no. 11), 5⬘-GGUCUUGGAGAUGGAGGCA and 5⬘-UGCCUCCAUCUCCAAGACC (no. 12); JNK1, 5⬘-UGAUGUGUCUUCAAUGUCA and 5⬘-UGACAUUGAAGACACAUCA (no. 110), 5⬘-UAAGCCGACCAUUUCAGAA and 5⬘-UUCUGAAAUGGUCGGCUUA (no. 111); JNK2, 5⬘-CCAAACUCAUGCAAAGAGA and 5⬘-UCUCUUUGCAUGAGUUUGG (no. 118). Cells were transfected as described previously (39). Activation of JNK, p38 MAPK, and IKK/NF-␬B—Cells were seeded into 10-cm dishes and in the case of gene silencing experiments were transfected as described above. Twelve hours before treatment cells were washed once with phosphate-buffered saline, and the growth media were replaced by media containing 2% heat-inactivated FBS. On the next day, cells were detached from the plates with non-enzymatic solution (Sigma) and placed into 24-well dishes (0.5–2 ⫻ 106 cells/well) in 0.25 ml of media containing 2% heat-inactivated FBS. In the gene silencing experiments, cells were treated with 0.25 ml of media containing 200 ng/ml Apo2L/TRAIL or 30 ng/ml TNF. In other experiments cells were treated with the ligand concentrations indicated in the figure legends. Upon treatment cells were lysed with a kinase lysis buffer (20 mM Tris-HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton, 1⫻ phosphatase inhibitor mixture II (no. P-5726, Sigma)) and analyzed by SDS-PAGE followed by immunoblot analysis, using phospho-specific anti-p38, -JNK, -I␬B␣, or total I␬B␣ antibodies and ECL kit (Amersham Biosciences). In Vitro JNK1 Assay—Cells were prepared and treated similarly to the immunoblot analysis. Upon lysis, JNK1 was immunoprecipitated, and in vitro kinase reaction was performed using recombinant GST-c-Jun substrate as described before (40). 32P incorporation into GST-c-Jun was determined by SDS-PAGE analysis followed by autoradiography. Cytotoxicity Assays—Cells (1–1.5 ⫻ 104 per well) were seeded into 96-well dishes in media containing 5% of heat-inactivated FBS. 8 –12 h later the media was changed, and cells were treated with human Apo2L/ TRAIL, FasL, or TNF (cycloheximide was used in the latter case at 0.5–2 ␮g/ml, as indicated in the figure legends) for 18 –24 h. Cell viability was measured by neutral red uptake as described (41). Examination of cIAP2 mRNA Expression—Total RNA was extracted from Apo2L/TRAIL (100 ng/ml)-, FasL (100 ng/ml)-, or TNF (100 ng/ml)-treated UACC62 cells by RNeasy mini kit (Qiagen). Real-time PCR (Taqman) was carried out in a 50-␮l reaction containing 50 ng of total RNA, 0.6 ␮M each of gene-specific forward and reverse primers, and 0.2 ␮M of gene-specific fluorescence probe. cIAP2-specific primers (forward, GGACAGGAGTTCATCCGTCAA; reverse, GGGCTGTCTGATGTGGATAGC; and probe, TCAAGCCAGTTACCCTCATCTACTTGAACAGC). Gene-specific PCR products were measured using an ABI PRISM 7700 Sequence Detection System following the manufacturer’s instructions (PE Corp.). The relative levels of cIAP2


mRNA were normalized to RPL19 mRNA levels measured in the same sample. The changes of cIAP2 mRNA expression upon stimulation were calculated as a ratio of relative cIAP2 mRNA expression at the indicated time points (Fig. 1D) to the relative cIAP2 mRNA level at 0 time point. Cytometric Bead Array Immunoassay—Cells were detached by nonenzymatic solution (Sigma) and seeded into 6-well dishes (0.5 ⫻ 106 cells/well in 2 ml of media). Treatment of the cells with Apo2L/TRAIL (100 ng/ml) or FasL (100 ng/ml) or TNF (10 ng/ml) in 1% FBS media was initiated 1 h later. After 8 h of incubation the cells were spun down, aliquots of conditioned media were collected from the supernatant, and the concentrations of cytokines IL-8 and CCL2/MCP-1 were determined by using a Human Inflammation Kit and a Human Chemokine Kit II (BD Bioscience, nos. 551811 and 558015, respectively). Macrophage Migration Assay—Human macrophages were isolated from a donor blood using Miltenyi CD11b⫹ beads. Cells were rested in serum-free medium for 12 h and then subjected (2 ⫻ 105 cells per upper chamber well) to migration assay using BD BioCoatTM Tumor Invasion System (no. 354165, BD Biosciences). 6 h later, the number of macrophages migrated to the bottom chambers filled with 5% fetal calf serum (as a positive control), or with the condition media samples collected from HT1080 or HT29 cells incubated in the serum-free media or treated for 8 h with Apo2L/TRAIL (100 ng/ml) in serum-free media were calculated according to the manufacturer’s instructions.

RESULTS Kinase Pathway Activation by Death-inducing Ligands—We reasoned that cells with moderate apoptosis susceptibility to Apo2L/ TRAIL might facilitate investigation of its kinase activation mechanisms. To this end, we tested several cancer cell lines for sensitivity to death induction by Apo2L/TRAIL, FasL, or TNF plus the protein translation inhibitor cycloheximide. The HT1080 fibrosarcoma cell line displayed intermediate sensitivity to each of the three ligands, whereas other cell lines tested displayed differential sensitivity (Fig. 1A). Taking these data together with the extensive previous characterization of HT1080 cells as a model system for TNF signaling (31, 42), we chose this cell line for further experiments, while verifying key observations in other cell lines. We began our investigation by testing the kinetics and potency of kinase pathway activation by death receptor ligands (Fig. 1B). We determined JNK pathway activation by immunoprecipitation of JNK1 and analysis of its ability to phosphorylate c-Jun; p38 MAPK activation by immunoblot (IB) detection of cellular phospho-p38 levels; and IKK/ NF-␬B pathway activation by IB detection of phospho-I␬B and total I␬B. Apo2L/TRAIL or FasL treatment of HT1080 cells induced a biphasic kinetic profile of JNK activation, with c-Jun phosphorylating activity peaking first at 15 min and then again at 120 or 60 min, respectively (Fig. 1B). TNF caused a more rapid, monophasic activation of JNK, reaching maximum by 5 min, persisting for about 60 min, and declining by 120 min (Fig. 1B). Apo2L/TRAIL and FasL also activated p38 phosphorylation, reaching significant levels by 30 or 60 min, respectively, whereas TNF induced more robust p38 activation already by 10 min (Fig. 1B). Apo2L/TRAIL and FasL also stimulated NF-␬B: I␬B phosphorylation peaked at 30 min and I␬B protein levels gradually declined, indicating I␬B degradation (Fig. 1B). By contrast, TNF induced biphasic I␬B phosphorylation, peaking at 10 min and again at 60 min, with rapid degradation of I␬B protein followed by its reappearance at 120 min, presumably through new synthesis (Fig. 1B). Dose titration analysis showed that equivalent stimulation of p38 phosphorylation and I␬B depletion required up to 100 times more Apo2L/TRAIL or FasL as compared with TNF (Fig. 1C). Analysis of JNK activation indicated a similar difference in dose-dependence (data not shown). As a more quantitative measure


40601

Kinase Pathway Activation by Apo2L/TRAIL

FIGURE 1. Apo2L/TRAIL and FasL induce cell death and activate the JNK, p38 MAPK, and IKK/NF-␬B pathways. A, cell death induction by Apo2L/TRAIL, FasL, or TNF. The graphs depict the proportion of cells remaining viable upon incubation with the indicated ligands for 18 h. B and C, phosphorylation of recombinant c-Jun by immunoprecipitated JNK, phosphorylation of p38 MAPK and of I␬B, and degradation of I␬B, in response to Apo2L/TRAIL, FasL, or TNF. HT1080 cells (1 ⫻ 106 per time point) were treated for the indicated periods of time with 100 ng/ml (B) or the indicated serial dilutions (C) of each ligand. Cells were collected, lysed, and subjected to immunoblot analysis or in vitro JNK1 activity assay as described under “Materials and Methods.” D, Apo2L/TRAIL, FasL, and TNF induce cIAP2 mRNA expression. Quantitative real-time PCR analysis of cIAP2 mRNA expression in UACC62 cells treated by the indicated ligands over various periods of time was done as described under “Materials and Methods.” All values were normalized to an RPL19 mRNA internal control.

of kinase activation, we analyzed the induction of mRNA encoding the NF-␬B-controlled gene c-IAP2 (Fig. 1D). Apo2L/TRAIL induced ⬃60fold up-regulation of c-IAP2 mRNA, as compared with induction of ⬃950-fold by TNF (Fig. 1D). Together, these results show that Apo2L/ TRAIL is indeed capable of stimulating the JNK, p38, and IKK kinase pathways, albeit less rapidly and much less potently than TNF. Apo2L/TRAIL Induces Formation of a Secondary Signaling Complex— Recent work by Tschopp and colleagues (17) demonstrates that, after forming a primary complex that supports kinase pathway activation, TNF stimulates a secondary signaling complex that mediates apoptotic caspase activation. We hypothesized that Apo2L/TRAIL may act in converse fashion, first inducing a DISC that activates caspases, and then a secondary complex that activates specific kinase pathways. To examine the possibility that Apo2L/TRAIL induces assembly of a secondary signaling complex, we performed an initial immunoprecipitation of the DISC from lysates of ligand-stimulated cells through the ligand itself, and then subjected these DISC-depleted lysates to an additional immunoprecipitation, this time using antibodies to FADD, RIP1, TRAF2, or NEMO. Consistent with earlier work in non-transfected cells (36 –38), within 5 min of addition to HT1080 cells, Apo2L/TRAIL stimulated the assembly of a DISC containing DR5 (DR4 is not expressed in these cells), FADD, and caspase-8, but not RIP1 (Fig. 2). Like RIP1, TRAF2 and NEMO also were absent from the DISC (data not shown). IB analysis of the second immunoprecipitation with DR5 antibody confirmed the absence of DR5, verifying complete DISC depletion in the first immunoprecipitation (Fig. 2). FADD immunoprecipitation of DISC-depleted lysates revealed association of FADD with caspase-8 and RIP1; RIP1 immunoprecipitation indicated binding of RIP1 to


FADD and caspase-8; TRAF2 immunoprecipitation suggested interaction of TRAF2 with FADD, caspase-8, and weak association with RIP1; and NEMO immunoprecipitation revealed weak NEMO interaction with FADD and stronger association with caspase-8 and RIP1 (Fig. 2). Most of these interactions occurred within 5 min of Apo2L/TRAIL stimulation, except for TRAF2 association with FADD, which appeared a bit later, at 15 min. The interaction of RIP1 with FADD was transient, peaking at 5 min, whereas the association of RIP1 with NEMO seemed to increase over time. (Because of high background, immunoprecipitation of TRADD, or clear IB detection of TRADD, TRAF2, or NEMO were not possible.) These results indicate that Apo2L/TRAIL induces the formation of a secondary signaling complex (or complexes), which lacks DR5, but retains FADD and caspase-8, and at minimum recruits RIP1, TRAF2, and/or NEMO. Involvement of FADD, RIP1, and TRAF2 in Kinase Pathway Activation by Apo2L/TRAIL—To explore the importance of the secondary Apo2L/TRAIL-induced complex for kinase pathway activation, we used siRNA technology to knock down individual signaling components. FADD siRNA knockdown attenuated the induction of JNK, p38 and I␬B phosphorylation and of I␬B degradation by Apo2L/TRAIL, but not by TNF (Fig. 3, A and B). Analysis of I␬B depletion in a previously characterized mutant variant of the Jurkat T cell line that does not express FADD (29) further confirmed the importance of FADD for NF-␬B activation by Apo2L/TRAIL but not by TNF (Fig. S1A). In addition, FADD siRNA knockdown inhibited Apo2L/TRAIL-, but not TNF-induced cIAP2 mRNA expression (Fig. S1B). FADD siRNA transfection protected HT1080 cells from induction of cell death by Apo2L/TRAIL or



FIGURE 2. Apo2L/TRAIL induces primary and secondary signaling complexes. HT1080 cells were stimulated with FLAG-epitope-tagged Apo2L/TRAIL (1 ␮g/ml) for the indicated periods of time, and lysed. The receptor-associated DISC was immunoprecipitated with anti-FLAG antibody and analyzed by immunoblot (first immunoprecipitation, left panel). Secondary signaling complexes were immunoprecipitated with the indicated antibodies from DISC-depleted lysates, and analyzed by immunoblot (second immunoprecipitation, right panels) as described under “Materials and Methods.”

FIGURE 3. Analysis of Apo2L/TRAIL signaling in FADD-, RIP1-, and TRAF2-depleted cells. A and B, induction of JNK, p38, or I␬B, phosphorylation and of I␬B degradation by Apo2L/TRAIL (A) or TNF (B) in cells transfected with siRNA against FADD (F), RIP1 (R), or TRAF2 (T), or control siRNA (C). Cells were transfected with siRNA and, 48 h later, treated with Apo2L/TRAIL or TNF for the indicated additional time, lysed, and subjected to immunoblot analysis of phosphorylated JNK, p38, or I␬B, or total I␬B and actin. Expression levels of FADD, RIP1, or TRAF2 in cells transfected with gene-specific (F, R, or T) or control (C) siRNA are shown at the bottom. C, effect of FADD (F), RIP1 (R), and TRAF2 (T) siRNA knockdown on the cytotoxic activity of Apo2L/TRAIL, TNF, or etoposide against HT1080 cells. The graphs depict the proportion of viable cells after incubation with the indicated ligands for 24 h. D, TRAF2 knockdown augments processing of caspase-8 and -3 by Apo2L/TRAIL. Cells were treated with Apo2L/TRAIL for the indicated periods of time and lysed. Processing of caspase-8 (left panels) and caspase-3 (right panels) in cells transfected by TRAF2-specific (top) or control (bottom) siRNA were analyzed by immunoblot.

TNF plus cycloheximide, but not by the DNA-damaging agent etoposide (Fig. 3C), verifying an effective and selective FADD knockdown. TRADD serves as a universal adaptor for TNFR1-mediated kinase pathway activation by TNF (43). TRADD is capable of associating with TNFR1, as well as with RIP1, TRAF2, and FADD, which cannot bind directly to TNFR1. Our attempts to knock down TRADD expression in HT1080 and other cell lines by siRNA failed (we tested 16 different siRNA duplexes). Because TRADD knock-out mouse cells were not available at the time of this work, we were unable to investigate the potential involvement


of this adaptor in kinase activation by Apo2L/TRAIL. We therefore turned to studying the roles of RIP1 and TRAF2 in more detail. RIP1 is a death domain containing kinase that binds to TNFR1 in a TRADD-dependent manner and is essential for TNFR1 activation of the NF-␬B and p38 MAPK pathways (44, 45). Recent work suggests that RIP1 contributes also to full JNK activation by TNF (44). There is also some evidence implicating RIP1 in kinase pathway activation by Apo2L/ TRAIL or FasL (31, 34). RIP1 siRNA knockdown in HT1080 cells did not attenuate Apo2L/TRAIL stimulation of JNK, but it substantially inhib-


40603


FIGURE 4. Caspase-8 is required for effective cell death induction and activation of JNK, p38, and NF-␬B by Apo2L/TRAIL. A, effect of caspase-8 knockdown on phosphorylation of JNK, p38, and I␬B, and degradation of I␬B induced by Apo2L/TRAIL or TNF. HT1080 cells were transfected with caspase-8-specific (C-8) or control (C) siRNA duplexes. After 48 h, the cells were treated with Apo2L/TRAIL or TNF for the indicated periods of time, lysed, and subjected to immunoblot analysis as in Fig. 3. The level of caspase-8 in cells transfected with siRNA against caspase-8 compared with control siRNA is shown at the bottom. B, effect of caspase-8 knockdown on the cytotoxic activity of Apo2L/TRAIL, TNF, or etoposide toward HT1080 cells. The graphs depict the proportion of cells, transfected with caspase-8 (black bars) or control (white bars) siRNA, remaining viable after incubation with the indicated ligands for 24 h. C, effect of zVAD on induction of JNK, p38, and I␬B phosphorylation by Apo2L/TRAIL or TNF in HT1080 cells. Cells were preincubated for 1 h with zVAD (20 ␮M), or Me2SO vehicle control, treated with Apo2L/TRAIL or TNF, lysed, and analyzed as above. The effect of zVAD on caspase-8 activation by Apo2L/TRAIL or TNF treatment of HT1080 cells is shown in the middle panel. D, effect of zVAD on primary (right panel) and secondary (left panel) Apo2L/TRAIL signaling complexes. HT1080 cells were pretreated for 1 h with zVAD (20 ␮M) or a Me2SO control, then treated with Apo2L/TRAIL, and signaling complexes were analyzed as above.

ited Apo2L/TRAIL-induced phosphorylation of p38 and I␬B, and I␬B degradation (Fig. 3A). RIP1 depletion inhibited kinase pathway activation by TNF, particularly phosphorylation of p38 and I␬B and degradation of I␬B (Fig. 3B). RIP1 siRNA knockdown strongly attenuated cIAP2 mRNA induction by Apo2L/TRAIL and TNF (Fig. S1B). RIP1 knockdown also augmented cell death induction by Apo2L/TRAIL or TNF, but not by etoposide (Fig. 3C). Thus, RIP1 is important for p38 and IKK activation, but dispensable for JNK stimulation by Apo2L/TRAIL. TRAF2 binds to TNFR1 through TRADD and serves as a critical adaptor for JNK activation by TNF (46). TRAF2 is important also for NF-␬B stimulation by TNF, redundantly with TRAF5 (47). As compared with wild-type mouse embryo fibroblasts (MEFs), TRAF2-deficient MEFs showed impaired JNK activation (measured by c-Jun phosphorylation) by Apo2L/TRAIL, FasL, or TNF (Fig. S2A). In contrast to this defect, siRNA knockdown of TRAF2 surprisingly led to higher phosphorylation levels of JNK1 and JNK2 in unstimulated cells and in Apo2L/TRAIL- or TNF-stimulated HT1080 cells (Fig. 3, A and B; the


high and low molecular weight bands in the p-JNK IBs represent predominantly JNK2 and JNK1, respectively). TRAF2-knock-out MEFs showed enhanced ligand-stimulated JNK1/2 phosphorylation (Fig. S2B), as did A549 or SK-MES-1 cells transfected with TRAF2-specific siRNA (data not shown). Furthermore, TRAF2-knock-out MEFs or TRAF2-knockdown HT1080 cells showed markedly enhanced cell death in response to Apo2L/ TRAIL, FasL, or TNF (Figs. 3C, S2C, and S2D). Consistent with the augmentation in cell death, TRAF2-depleted cells showed a significant increase in Apo2L/TRAIL activation of caspase-8 and caspase-3 (Fig. 3D). JNK signaling plays an important contextual role in the control of TNF-induced cell death (16). Recent evidence indicates that JNK1 has a dominant-positive signaling function, whereas JNK2 plays an inhibitory role in apoptosis activation by TNF (48, 49). Consistently with these findings, siRNA knockdown of JNK1 in HT1080 cells diminished TNFinduced cell death, whereas knockdown of JNK2 did not (Fig. S3). In contrast, siRNA knockdown of either JNK1 or JNK2 or both did not influence Apo2L/TRAIL-induced cell death (Fig. S3). Thus, whereas



FIGURE 5. FLIPL knockdown accelerates Apo2L/ TRAIL-induced phosphorylation and degradation of I␬B. A, effect of FLIPL knockdown on activation of caspase-8. Cells were transfected with FLIPL (FL)-specific or control (C) siRNA duplexes. After 48 h, the cells were treated with Apo2L/TRAIL or TNF for the indicated periods of time and lysed, and caspase-8 processing was analyzed by immunoblot. B, effect of FLIPL knockdown on I␬B phosphorylation and degradation. Cells transfected with FLIPL or control siRNA duplexes were treated as in A, lysed, and analyzed by immunoblot for levels of phosphorylated and total I␬B. The levels of FLIPL and FLIPS in cells transfected with gene-specific or (FL) control (C) siRNA are shown at the bottom.

JNK activity modulates cell death induction by TNF but not by Apo2L/ TRAIL, TRAF2 seems to modulate the cytotoxicity of both ligands. Of note, knockdown of JNK1 did not reverse the sensitization of TRAF2depleted HT1080 cells to death induction by either TNF or Apo2L/ TRAIL (Fig. S3), indicating that cell death modulation by TRAF2 may be independent of and dominant over regulation by JNK activity. Caspase-8 Activity Is Important for Effective Kinase Pathway Activation by Apo2L/TRAIL—In agreement with published data (26, 28, 38, 50, 51), siRNA knockdown of caspase-8 protected cells against death induction by Apo2L/TRAIL or TNF, but did not influence TNF-induced kinase activation (Fig. 4, A and B). However, caspase-8-depleted HT1080 cells showed markedly diminished Apo2L/TRAIL-induced phosphorylation of JNK, p38, and I␬B and degradation of I␬B (Fig. 4A). Although caspase-8 knockdown was less efficient in SK-MES-1 or A549 cells, these cell lines showed a similar, though less complete, attenuation or delay in kinase activation (Fig. S4). Additionally, a caspase-8-deficient Jurkat T cell line variant showed little or no I␬B degradation in response to Apo2L/TRAIL, whereas wild-type Jurkat T cells showed significant I␬B depletion (Fig. S1). Caspase-8 siRNA knockdown in HT1080 cells inhibited cIAP2 mRNA induction by Apo2L/TRAIL (Fig. S1B). These results generally agree with recent data implicating a structural and/or enzymatic caspase-8 requirement for effective activation of JNK and NF-␬B by FasL and Apo2L/TRAIL (31). To examine further the potential importance of enzymatic caspase-8 activity for kinase stimulation by Apo2L/TRAIL, we used the pan caspase inhibitor Z-Val-Ala-Asp (OMe)-fluoromethyl ketone (zVAD) (Figs. 4C and S5). Pretreatment of HT1080 cells by zVAD blocked the activation of caspase-8 by Apo2L/ TRAIL (Fig. 4C, bottom panel; depletion of pro-caspase-8 p55/53 bands indicates stimulation). Pretreatment with zVAD substantially abrogated Apo2L/TRAIL activation of JNK and p38 while leaving TNF stimulation of these kinases intact (Figs. 4C and S5). Additionally, zVAD delayed (although it did not abolish) I␬B phosphorylation in response to Apo2L/TRAIL and left TNF induction of these events unimpeded (Figs. 4C and S5). Analysis of the effect of zVAD on Apo2L/TRAIL-induced DISC assembly showed enhanced and prolonged association between DR5


and FADD (Fig. 4D, left panels), indicating that inhibition of caspase activity by zVAD stabilized the DISC. In contrast to its effect on the DISC, zVAD delayed and attenuated formation of the secondary Apo2L/TRAIL-induced complex (Fig. 4D, right panels). These results suggest that caspase-8 activity is important for second complex formation and kinase activation by Apo2L/TRAIL. To investigate further the importance of caspase-8 activity for second complex formation, we took advantage of the fact that the caspase-related molecule FLIPL functions as an inhibitor of caspase-8 activation by Apo2L/TRAIL (39). We therefore examined the effect of siRNA knockdown of FLIPL on ligand activation of the IKK/NF-␬B pathway in HT1080 cells. Consistent with earlier data, FLIPL knockdown augmented Apo2L/TRAIL induction of DISC recruitment and processing of caspase-8 in the Apo2L/TRAIL DISC (Fig. 5A), potentiating consequent cell death (data not shown). FLIPL knockdown also enhanced the induction of I␬B phosphorylation and degradation by Apo2L/TRAIL, but had little effect on stimulation of these events by TNF (Fig. 5B). Furthermore, siRNA knockdown of FLIPL potentiated cIAP2 mRNA induction by Apo2L/TRAIL (Fig. S1B). These data lend further support to the notion that enzymatic caspase-8 activity (which is stimulated better in the absence of FLIPL) contributes to the formation of a secondary signaling complex that mediates kinase activation by Apo2L/TRAIL. A Potential Biological Function for Kinase Pathway Activation by Apo2L/TRAIL—An important outcome of NF-␬B activation by TNF is the induction of anti-apoptotic genes that can override the apoptosisinducing activity of TNF, enabling the dominance of TNF pro-inflammatory function (16). We were interested in asking whether NF-␬B activation by Apo2L/TRAIL affects the pro-apoptotic activity of this ligand; however, the importance of caspase-8 for Apo2L/TRAIL stimulation of kinase pathways precluded us from direct study of this question. To circumvent this limitation, we pretreated HT1080 cells with TNF to stimulate kinase pathways and subsequently examined the ability of Apo2L/TRAIL to trigger cell death. Surprisingly, TNF pretreatment provided little or no protection against induction of cell death by Apo2L/TRAIL in HT1080, A549, or HeLa cells, and if anything, it augmented Apo2L/TRAIL cytotoxicity against SK-MES-1 and HCT15 cells (Fig. 6). These results suggest that cell death induction by Apo2L/


40605


FIGURE 6. TNF pretreatment does not protect cells against Apo2L/TRAIL-induced death. The indicated cell types were pretreated with TNF for 24 h or left untreated. The cells were then rinsed once with phosphate-buffered saline and treated for another 18 h with increasing amounts of non-tagged soluble Apo2L/TRAIL. The graphs depict the proportion of cells remaining viable after incubation with Apo2L/TRAIL.

TRAIL overrides the potential anti-apoptotic effects of kinase pathway activation by TNF. Given these observations, we reasoned that the caspase-8-dependent stimulation of kinase pathways by Apo2L/TRAIL might serve to support rather than to inhibit the apoptotic program initiated by this ligand. Consistent with this notion, Apo2L/TRAIL induced significant release of the chemokines IL-8 and MCP-1 in HT1080 cells; in contrast, it had little effect on production of these chemokines in HT29 cells, which are resistant to apoptosis induction by trimeric Apo2L/TRAIL (Fig. 7A). TNF induced release of higher levels of both chemokines in both cell lines, with particularly high levels of IL-8 in HT1080 cells (Fig. 7A). To assess further the biological relevance of the chemokine levels induced by Apo2L/TRAIL, we tested the activity of conditioned media from ligand-treated cells in a Transwell cell migration assay. Conditioned media from Apo2L/TRAIL-treated HT1080 cells induced a substantial increase in the migration of human peripheral blood macrophages as compared with media from buffer-treated cells, whereas media from similarly treated HT29 cells did not alter macrophage migration (Fig. 7B). These results suggest that kinase activation by Apo2L/TRAIL increases chemokine secretion and macrophage attraction.

DISCUSSION Earlier studies have provided evidence for the ability of the deathinducing ligands FasL and Apo2L/TRAIL to stimulate the IKK/NF-␬B, JNK, and p38 MAPK pathways. However, most of the information about this activity was obtained under non-physiological conditions, namely, in the context of death receptor overexpression or cell death blockade (3, 52–54). In the present study, we used HT1080 cells as a primary model system, because these cells are only partially sensitive to Apo2L/TRAIL-induced apoptosis, and hence more amenable to physiological interrogation of kinase pathway activation. Apo2L/TRAIL


treatment of HT1080 cells induced not only cell death but also activation of JNK, p38, and NF-␬B, as it did in several other cell lines. Kinase pathway activation by Apo2L/TRAIL was significantly slower and weaker than stimulation by TNF, suggesting either that the molecular determinants of kinase activation by these ligands differ, or alternatively, that the two ligands use similar components, which they organize differentially in distinct signaling complexes. Consistent with the latter possibility, we found that Apo2L/TRAIL assembles not only a receptorassociated DISC, but also a secondary intracellular signaling complex (or complexes) containing molecules that are involved also in kinase pathway activation by TNF. Ligation of DR5 by Apo2L/TRAIL led to rapid recruitment of FADD and caspase-8 to the receptor, forming a DISC. The Apo2L/TRAIL DISC did not contain detectable RIP1, TRAF2, or NEMO. However, examination of secondary Apo2L/ TRAIL-induced intracellular interactions by removal of the primary DISC revealed rapid organization of a complex that lacked the ligand and receptor, but contained FADD, caspase-8, RIP1, TRAF2, and NEMO. TRADD also may be involved in this secondary complex, but we were unable to verify this because of technical immunoblot and siRNA limitations. Thus, TNF and Apo2L/TRAIL appear to trigger two conversely related sets of intracellular interactions. TNF assembles a primary complex that signals kinase pathway activation and a secondary intracellular complex containing FADD and caspase-8 that under specific permissive conditions triggers apoptosis (17). Apo2L/TRAIL, on the other hand, assembles a primary complex that signals apoptosis and a secondary intracellular complex that stimulates kinase pathway activation. In each case, association of the secondary complex depends on formation of the primary complex and presumably requires dissociation of the primary complex from the receptor. The secondary complex assembled by Apo2L/TRAIL may represent multiple complexes containing overlapping and/or distinct components. The differences in



FIGURE 7. Apo2L/TRAIL induces chemokine secretion and macrophage migration. A, induction of cell secretion of IL-8 and MCP-1 by Apo2L/TRAIL or TNF. HT1080 or HT29 cells were treated by the indicated ligand for 8 h. Conditioned media were collected, and the concentrations of IL-8 and MCP-1 were determined by cytometric bead array immunoassay as described under “Materials and Methods.” B, induction of macrophage migration by cell-conditioned media. The ability of conditioned media from cells pretreated with Apo2L/TRAIL or TNF to induce Transwell migration of human peripheral blood macrophages was analyzed as described under “Materials and Methods.” Means ⫾ S.D. for triplicate determinations of migrating cell numbers are shown.

kinetics and strength of the associations between FADD, RIP1, and NEMO in response to Apo2L/TRAIL support the potential existence of multiple complexes, but this warrants further study. Our functional investigation revealed that Apo2L/TRAIL requires FADD both for cell death induction and for kinase stimulation. Activation of p38 and IKK by Apo2L/TRAIL depended on RIP1, whereas stimulation of JNK did not. RIP1 depletion was associated with a modest sensitization to Apo2L/TRAIL-triggered cell death. One potential inference from this latter observation is that RIP1-dependent activation of p38 and IKK by Apo2L/TRAIL negatively regulates cell death induction. However, TNF, which strongly activates p38 and IKK signaling, did not inhibit cell death induction by Apo2L/TRAIL, suggesting that RIP1 inhibits Apo2L/TRAIL-induced apoptosis independently of its involvement in kinase activation. RIP1 may influence apoptosis indirectly, by modulating the secondary Apo2L/TRAIL signaling complex, perhaps through interaction with TRAF2 (55). Indeed, TRAF2 depletion enhanced caspase-8 activation by Apo2L/TRAIL (see below). Apo2L/TRAIL stimulation of p38 and IKK did not require TRAF2. In contrast to p38 and IKK activity, JNK enzymatic activity was significantly reduced in TRAF2-null MEFs. Despite this reduction in JNK activity, both the basal and ligand-induced levels of JNK1 and JNK2 phosphorylation were higher in the absence of TRAF2 in all of the cell lines examined. Recent studies suggest that JNK1 promotes c-Jun phosphorylation and activation of AP-1 and ATF2, whereas JNK2 inhibits these effects by attenuating JNK1 activity and destabilizing c-Jun (48, 49). Thus, the reduction in JNK1 activity in TRAF2-deficient cells may result from dominant inhibition of phosphorylated JNK1 by phosphorylated JNK2.


TRAF2 depletion surprisingly sensitized cells to death induction by Apo2L/TRAIL, FasL, as well as TNF. This finding conflicts with earlier data (obtained in overexpression studies and in the presence of high cycloheximide concentrations), which suggested that TRAF2 is not involved in modulation of apoptosis signaling by Apo2L/TRAIL and FasL (34). How might depletion of TRAF2 promote apoptosis signaling? It is well established that sustained JNK1 activation by TNF supports the ability of TNF to induce cell death (56, 57). Consistent with this, and with the inhibitory effect of JNK2 on JNK1 activity, JNK1 knockdown prevented TNF-induced cell death, whereas JNK2 knockdown potentiated this response. However, knockdown of JNK1 or JNK2 did not reverse the sensitization to TNF cytotoxicity by TRAF2 depletion, suggesting that the modulation of death signaling by TRAF2 does not involve JNK. Unlike TNF, there is no published evidence supporting a significant contribution of JNK to apoptosis induction by Apo2L/ TRAIL and FasL. Indeed, siRNA knockdown of JNK1, JNK2, or both did not alter Apo2L/TRAIL-induced cell death, nor did it reverse the sensitization to this ligand by TRAF2 knockdown. Taken together, these results suggest that TRAF2 depletion causes sensitization to death ligand-induced apoptosis independently of JNK. Moreover, while JNK activation modulates TNF induction of cell death, it does not significantly impact the cytotoxic activity of Apo2L/TRAIL. In agreement with previous studies (28), we found that Apo2L/ TRAIL requires caspase-8 not only for induction of cell death, but also for kinase pathway activation. Does a mere scaffolding function or the actual enzymatic activity of caspase-8 contribute to kinase stimulation? Reconstitution of caspase-8-deficient Jurkat T cells with an enzymatically inactive caspase-8 mutant was sufficient to restore NF-␬B induc-


40607

Kinase Pathway Activation by Apo2L/TRAIL tion by FasL (32), suggesting a lack of requirement for caspase-8 enzymatic activity. Conversely, a catalytically inactive caspase-8 mutant blocked NF-␬B induction by transiently transfected death receptors, FADD, or wild-type caspase-8 (58), supporting a requirement for enzymatic activity. We found that exposure of cells to zVAD, which blocks various caspases, including caspase-8, stabilizes the Apo2L/TRAIL DISC, and decelerates assembly of the secondary complex. Furthermore, zVAD treatment inhibited or delayed kinase pathway activation by Apo2L/TRAIL. In addition, depletion of FLIPL, a molecule that inhibits recruitment and activation of caspase-8 at the DISC (39), augmented kinase activation by Apo2L/TRAIL. These results indicate that the enzymatic activity of caspase-8 is important not only for apoptosis induction but also for kinase pathway activation by Apo2L/TRAIL. These findings do not rule out the possibility that some scaffolding role of caspase-8 also may contribute to kinase pathway stimulation. Given that FADD is retained in the secondary complex, it is plausible that caspase-8 activity facilitates dissociation of FADD from the receptor, but this requires further study. Caspase stimulation by the TNF secondary signaling complex requires inhibition of NF-␬B activation by the TNF primary signaling complex (17). Our results show by contrast that caspase activation by the primary signaling complex of Apo2L/TRAIL promotes the formation of a secondary complex, which in turn leads to kinase pathway activation. One function of the secondary complex may be to curtail the level of caspase activation by Apo2L/TRAIL through an unidentified activity of TRAF2. Nonetheless, kinase pathway activation is not a dominant inhibitory determinant for, and in some cell lines it may in fact facilitate apoptosis induction by Apo2L/TRAIL. In conclusion, our studies show that Apo2L/TRAIL induces kinase pathway activation downstream of DISC assembly and caspase-8 stimulation. Kinase pathway activation by Apo2L/TRAIL is associated with increased production of the chemokines IL-8 and MCP-1 and with enhanced macrophage migration. These results raise the possibility that, within a tissue context, Apo2L/TRAIL activates specific kinase pathways in conjunction with caspase activation to promote chemokine-supported phagocytosis of apoptotic cells (59). Acknowledgments—We thank Scot Masters for technical assistance and Mark Vasser for the oligonucleotide synthesis.

REFERENCES 1. Smith, C. A., Farrah, T., and Goodwin, R. G. (1994) Cell 76, 959 –962 2. Ashkenazi, A., and Dixit, V. M. (1998) Science 281, 1305–1308 3. Wallach, D., Varfolomeev, E. E., Malinin, N. L., Goltsev, Y. V., Kovalenko, A. V., and Boldin, M. P. (1999) Annu. Rev. Immunol. 17, 331–367 4. Idriss, H. T., and Naismith, J. H. (2000) Microsc. Res. Tech. 50, 184 –195 5. Locksley, R. M., Killeen, N., and Lenardo, M. J. (2001) Cell 104, 487–501 6. Ashkenazi, A. (2002) Nat. Rev. Cancer 2, 420 – 430 7. Peter, M. E., and Krammer, P. H. (2003) Cell Death Differ. 10, 26 –35 8. LeBlanc, H. N., and Ashkenazi, A. (2003) Cell Death Differ. 10, 66 –75 9. Kischkel, F. C., Hellbardt, S., Behrmann, I., Germer, M., Pawlita, M., Krammer, P. H., and Peter, M. E. (1995) EMBO J. 14, 5579 –5588 10. Slee, E. A., Harte, M. T., Kluck, R. M., Wolf, B. B., Casiano, C. A., Newmeyer, D. D., Wang, H. G., Reed, J. C., Nicholson, D. W., Alnemri, E. S., Green, D. R., and Martin, S. J. (1999) J. Cell Biol. 144, 281–292 11. Cowling, V., and Downward, J. (2002) Cell Death Differ. 9, 1046 –1056 12. Tschopp, J., Irmler, M., and Thome, M. (1998) Curr. Opin. Immunol. 10, 552–558 13. Chen, G., and Goeddel, D. V. (2002) Science 296, 1634 –1635 14. Zhang, S. Q., Kovalenko, A., Cantarella, G., and Wallach, D. (2000) Immunity 12, 301–311 15. Kucharczak, J., Simmons, M. J., Fan, Y., and Gelinas, C. (2003) Oncogene 22, 8961– 8982 16. Varfolomeev, E. E., and Ashkenazi, A. (2004) Cell 116, 491– 497 17. Micheau, O., and Tschopp, J. (2003) Cell 114, 181–190


18. 19. 20. 21. 22. 23. 24. 25.

26.

27. 28. 29. 30. 31.

32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46.

47.

48. 49. 50. 51. 52.

53. 54. 55. 56. 57. 58. 59.

Krammer, P. H. (2000) Nature 407, 789 –795 Almasan, A., and Ashkenazi, A. (2003) Cytokine Growth Factor Rev. 14, 337–348 Li-Weber, M., and Krammer, P. H. (2003) Semin. Immunol. 15, 145–157 Kelley, S. K., and Ashkenazi, A. (2004) Curr. Opin. Pharmacol. 4, 333–339 Kimberley, F. C., and Screaton, G. R. (2004) Cell Res. 14, 359 –372 Diehl, G. E., Yue, H. H., Hsieh, K., Kuang, A. A., Ho, M., Morici, L. A., Lenz, L. L., Cado, D., Riley, L. W., and Winoto, A. (2004) Immunity 21, 877– 889 Greten, F. R., and Karin, M. (2004) Cancer Lett. 206, 193–199 Yeh, W. C., de la Pompa, J. L., McCurrach, M. E., Shu, H. B., Elia, A. J., Shahinian, A., Ng, M., Wakeham, A., Khoo, W., Mitchel, K., El-Deiry, W. S., Lowe, S. W., Goeddel, D. V., and Mak, T. W. (1998) Science 279, 1954 –1958 Varfolomeev, E. E., Schuchmann, M., Luria, V., Chiannlkulchai, N., Beckmann, J. S., Mett, I. L., Rebrikov, D., Brodianski, V. M., Kemper, O. C., Kollet, O., Lapidot, T., Soffer, D., Sobe, T., Avraham, K. B., Goncharov, T., Holtmann, H., Lonai, P., and Wallach, D. (1998) Immunity 9, 267–276 Wajant, H., Johannes, F. J., Haas, E., Siemienski, K., Schwenzer, R., Schubert, G., Weiss, T., Grell, M., and Scheurich, P. (1998) Curr. Biol. 8, 113–116 Juo, P., Kuo, C. J., Yuan, J., and Blenis, J. (1998) Curr. Biol. 8, 1001–1008 Juo, P., Woo, M. S., Kuo, C. J., Signorelli, P., Biemann, H. P., Hannun, Y. A., and Blenis, J. (1999) Cell Growth & Differ. 10, 797– 804 Wajant, H., Haas, E., Schwenzer, R., Muhlenbeck, F., Kreuz, S., Schubert, G., Grell, M., Smith, C., and Scheurich, P. (2000) J. Biol. Chem. 275, 24357–24366 Kreuz, S., Siegmund, D., Rumpf, J. J., Samel, D., Leverkus, M., Janssen, O., Hacker, G., Dittrich-Breiholz, O., Kracht, M., Scheurich, P., and Wajant, H. (2004) J. Cell Biol. 166, 369 –380 Imamura, R., Konaka, K., Matsumoto, N., Hasegawa, M., Fukui, M., Mukaida, N., Kinoshita, T., and Suda, T. (2004) J. Biol. Chem. 279, 46415– 46423 Hu, W. H., Johnson, H., and Shu, H. B. (1999) J. Biol. Chem. 274, 30603–30610 Lin, Y., Devin, A., Cook, A., Keane, M. M., Kelliher, M., Lipkowitz, S., and Liu, Z. G. (2000) Mol. Cell Biol. 20, 6638 – 6645 Harper, N., Farrow, S. N., Kaptein, A., Cohen, G. M., and MacFarlane, M. (2001) J. Biol. Chem. 276, 34743–34752 Bodmer, J. L., Holler, N., Reynard, S., Vinciguerra, P., Schneider, P., Juo, P., Blenis, J., and Tschopp, J. (2000) Nat. Cell Biol. 2, 241–243 Kischkel, F. C., Lawrence, D. A., Chuntharapai, A., Schow, P., Kim, K. J., and Ashkenazi, A. (2000) Immunity 12, 611– 620 Sprick, M. R., Weigand, M. A., Rieser, E., Rausch, C. T., Juo, P., Blenis, J., Krammer, P. H., and Walczak, H. (2000) Immunity 12, 599 – 609 Sharp, D. A., Lawrence, D. A., and Ashkenazi, A. (2005) J. Biol. Chem 280, 19401–19409. Kerkhoff, E., and Rapp, U. R. (1997) Mol. Cell Biol. 17, 2576 –2586 Johnston, M. D., Finter, N. B., and Young, P. A. (1981) Methods Enzymol. 78, 394 –399 Deng, Y., Ren, X., Yang, L., Lin, Y., and Wu, X. (2003) Cell 115, 61–70 Hsu, H., Xiong, J., and Goeddel, D. V. (1995) Cell 81, 495–504 Devin, A., Lin, Y., and Liu, Z. G. (2003) EMBO Rep. 4, 623– 627 Lee, T. H., Huang, Q., Oikemus, S., Shank, J., Ventura, J. J., Cusson, N., Vaillancourt, R. R., Su, B., Davis, R. J., and Kelliher, M. A. (2003) Mol. Cell Biol. 23, 8377– 8385 Yeh, W. C., Shahinian, A., Speiser, D., Kraunus, J., Billia, F., Wakeham, A., de la Pompa, J. L., Ferrick, D., Hum, B., Iscove, N., Ohashi, P., Rothe, M., Goeddel, D. V., and Mak, T. W. (1997) Immunity 7, 715–725 Tada, K., Okazaki, T., Sakon, S., Kobarai, T., Kurosawa, K., Yamaoka, S., Hashimoto, H., Mak, T. W., Yagita, H., Okumura, K., Yeh, W. C., and Nakano, H. (2001) J. Biol. Chem. 276, 36530 –36534 Sabapathy, K., Hochedlinger, K., Nam, S. Y., Bauer, A., Karin, M., and Wagner, E. F. (2004) Mol. Cell 15, 713–725 Liu, J., Minemoto, Y., and Lin, A. (2004) Mol. Cell Biol. 24, 10844 –10856 Kim, I. K., Chung, C. W., Woo, H. N., Hong, G. S., Nagata, S., and Jung, Y. K. (2000) Biochem. Biophys. Res. Commun. 277, 311–316 Sprick, M. R., Rieser, E., Stahl, H., Grosse-Wilde, A., Weigand, M. A., and Walczak, H. (2002) EMBO J. 21, 4520 – 4530 Sheridan, J. P., Marsters, S. A., Pitti, R. M., Gurney, A., Skubatch, M., Baldwin, D., Ramakrishnan, L., Gray, C. L., Baker, K., Wood, W. I., Goddard, A. D., Godowski, P., and Ashkenazi, A. (1997) Science 277, 818 – 821 Schneider, P., Thome, M., Burns, K., Bodmer, J.-L., Hofmann, K., Kataoka, T., Holler, N., and Tschopp, J. (1997) Immunity 7, 831– 836 Wajant, H., Moosmayer, D., Wuest, T., Bartke, T., Gerlach, E., Schonherr, U., Peters, N., Scheurich, P., and Pfizenmaier, K. (2001) Oncogene 20, 4101– 4106 Hsu, H., Huang, J., Shu, H. B., Baichwal, V., and Goeddel, D. V. (1996) Immunity 4, 387–396 Lin, A. (2003) BioEssays 25, 17–24 Maeda, S., Chang, L., Li, Z. W., Luo, J. L., Leffert, H., and Karin, M. (2003) Immunity 19, 725–737 Chaudhary, P. M., Eby, M. T., Jasmin, A., Kumar, A., Liu, L., and Hood, L. (2000) Oncogene 19, 4451– 4460 Ravichandran, K. S. (2003) Cell 113, 817– 820
