Introduction

Nuclear factor-kappa B (NF-B) plays a pivotal role in protection from tumor necrosis factor (TNF)- and genotoxic stress-induced apoptosis. A key mechanism by which NF-B inhibits apoptosis has been considered to enhance transcription of various antiapoptotic genes, including cellular FLICE-inhibitory protein (c-Flip), the members of the Bcl-2 family, and X chromosome-linked inhibitor of apoptosis (Xiap) (Barkett and Gilmore, 1999; Karin and Lin, 2002). In addition to these, recent studies have shown that NF-B suppresses TNF-induced cell death by inhibiting prolonged c-Jun N-terminal kinase (JNK) activation and reactive oxygen species (ROS) accumulation (Sakon et al., 2003; Pham et al., 2004; Ventura et al., 2004; Kamata et al., 2005). Although various signaling intermediates including ferritin heavy chain and manganese-dependent superoxide dismutase to eliminate ROS accumulation have been identified (Pham et al., 2004; Kamata et al., 2005), it remains controversial whether their functions are essential and/or sufficient for eliminating ROS accumulation (Nakano et al., 2006). Moreover, a recent study has shown that activated caspase 3 cleaves the p75 subunit of complex I of the mitochondrial electron transport chain, which would impair the function of complex I, resulting in ROS accumulation (Ricci et al., 2004). In addition, Giorgio et al. (2005) have shown that proapoptotic signals induce the release of p66^Shc from a putative inhibitory complex and the released p66^Shc then oxidizes reduced cytochrome c, thereby generating ROS. Therefore, the molecular mechanism underlying caspase-dependent ROS accumulation during apoptotic cell death is not fully understood.

c-FLIP, also designated as caspase homologue (CASH) or Casper, was first identified as a molecule that interacts with Fas-associated death domain (FADD) or is structurally related to procaspase 8 (Budd et al., 2006). c-Flip encodes two splicing variants, c-FLIP_L and c-FLIP_S. While c-FLIP_S only contains two N-terminal death effector domain (DED)s, c-FLIP_L consists of the N-terminal DEDs and a C-terminal caspase-like domain that does not possess enzymatic activity. Many studies have shown that both c-FLIP_L and c-FLIP_S inhibit death receptor-induced apoptosis by binding to and inhibiting caspase 8 activation via DED–DED interaction (Budd et al., 2006). Consistently, c-Flip^-/- murine embryonic fibroblasts (MEFs) showed an increased susceptibility to TNF- and Fas ligand-induced cell death (Yeh et al., 2000). Very recently, we have shown that TNF-induced ROS accumulation is enhanced in c-Flip^-/- MEFs (Nakajima et al., 2006), indicating that c-FLIP plays a dominant role in suppressing ROS accumulation.

To explore whether downregulation of c-FLIP enhances ROS accumulation in various cell types, we have knocked down endogenous c-FLIP expression by RNA interference (RNAi) in HeLa cells. Consistent with c-Flip^-/- MEFs (Nakajima et al., 2006), TNF and anti-Fas antibody stimulation induced ROS accumulation along with delayed and long-lasting JNK and extracellular signal-regulated kinase (ERK) activation in c-FLIP knockdown HeLa cells. Notably, TNF- and anti-Fas antibody-induced prolonged JNK and ERK, and ROS accumulation were completely inhibited by a caspase inhibitor, suggesting that these events are downstream of the caspase cascade. Similar observations were also observed in c-FLIP knockdown HCT116 and A549 cells. Interestingly, TNF induced caspase-dependent necrotic as well as apoptotic cell death in a similar kinetics. Moreover, TNF and Fas induced the cleavage of mitogen-activated protein kinase (MAPK)/ERK kinase kinase (MEKK)1, resulting in generation of a constitutive active form of MEKK1 leading to JNK activation. Given that accumulated ROS elicits inflammation, resulting in compensatory proliferation of tumor cells, identification of target genes responsible for caspase-dependent ROS accumulation might be crucial to efficiently treat cancers through suppression of caspase-dependent ROS accumulation.

Results

Knockdown of c-FLIP promotes TNF-induced prolonged JNK activation and ROS accumulation in HeLa cells

Our recent study has shown that TNF induces caspase-dependent and -independent ROS accumulation, and prolonged JNK activation in c-Flip^-/- MEFs (Nakajima et al., 2006). To apply this observation to tumor therapy, we first knocked down expression of c-FLIP using small interfering RNA (siRNA) in HeLa cells. While transfection of the control green fluorescent protein (GFP) siRNA oligos did not affect expression of c-FLIP_L and c-FLIP_S, these bands disappeared in cells transfected with two different pairs of siRNA oligos targeting distinct regions of c-FLIP (Figure 1a). Transfection of two c-FLIP siRNA oligos did not affect expression of other antiapoptotic proteins including TNF receptor-associated factor (TRAF)2, receptor-interacting protein (RIP) and XIAP (Figure 1a). Under these experimental conditions, TNF-induced JNK activation was dramatically prolonged and continued at 8 h in both c-FLIP siRNA-, but not control siRNA-transfected HeLa cells (Figure 1b). Moreover, TNF induced ROS accumulation in both c-FLIP siRNA-transfected HeLa cells (Figure 1c). Since we obtained similar results using two different c-FLIP siRNAs as described above, we used c-FLIP(a) siRNA oligos hereafter. We also confirmed that knockdown of c-FLIP did not impair the TNF-induced NF-B activation, since TNF-induced degradation of inhibitor of NF-B (IB) was similarly induced in both control siRNA- and c-FLIP siRNA-transfected HeLa cells (Figure 1d).

Figure 1.

c-FLIP is required for suppression of TNF-induced ROS accumulation and prolonged JNK in HeLa cells. (a) siRNA oligos targeting GFP (control, c) or two different regions of c-FLIP (a and b) were transfected into HeLa cells to knock down expression of endogenous proteins. At 36 h post-transfection, the efficiency and the specificity of c-FLIP RNAi was confirmed by immunoblotting with anti-c-FLIP, anti-TRAF2, anti-RIP and anti-XIAP antibodies. (b) Control siRNA- or c-FLIP siRNA-transfected HeLa cells were stimulated with TNF (10 ng/ml) for the indicated times, and the lysates were immunoblotted with anti-phospho-JNK and anti-total JNK antibodies. (c) The cells were stimulated as in (b), and then stained with CM-H₂DCFDA, and analysed by flow cytometry. ROS levels are expressed as relative fluorescence intensity. (d) TNF-induced degradation of IB is not impaired in c-FLIP knockdown HeLa cells. The cells were stimulated as in (b), and the lysates were immunoblotted with anti-IB and tubulin antibodies. The similar results were obtained with two independent experiments. c-FLIP, cellular FLICE-inhibitory protein; CM-H₂DCFDA, 5-(and-6)-carboxy-2',7'-dichlorodihydrofluorescein diacetate; JNK, c-Jun N-terminal kinase; RIP, receptor-interacting protein; ROS, reactive oxygen species; TNF, tumor necrosis factor-; TRAF, TNF receptor-associated factor; XIAP, X chromosome-linked inhibitor of apoptosis.

Full figure and legend (94K)

TNF-induced ROS accumulation and prolonged JNK activation are caspase dependent

We next investigated the time course of TNF-induced JNK activation in more detail. The early phase of TNF-induced JNK and ERK activation rapidly declined within 1 h in control siRNA- and c-FLIP siRNA-transfected HeLa cells, but the sustained phase of JNK and ERK activation appeared at 2 h and then peaked at 4 h in c-FLIP knockdown HeLa cells (Figure 2a). We could not detect TNF-induced phosphorylation of p38 in HeLa cells under these conditions (data not shown). The processing of procaspase 8 was also induced at 2 h and peaked at 4 h, resulting in generation of the p43, p41 and p18 fragments. Collectively, these results demonstrate that knockdown of c-FLIP expression by RNAi is sufficient for TNF-induced ROS accumulation, and prolonged JNK and ERK activation along with the activation of caspase 8 in HeLa cells.

Figure 2.

TNF and anti-Fas antibody induce prolonged JNK and ERK activation along with the processing of procaspase 8 in c-FLIP knockdown HeLa cells. (a) The cells were stimulated as in Figure 1b, and the lysates were immunoblotted with anti-phospho-JNK, anti-total JNK, anti-phospho-ERK, anti-total ERK, anti-caspase 8 antibodies. C8, Pro, p43/41/18 indicate caspase 8, the proform and the processed form of caspase 8, respectively. (b and c) The cells were stimulated with TNF (10 ng/ml) or anti-Fas antibody (100 ng/ml) in the absence or presence of z-VAD-fmk (50 M) (zV) or BHA (100 M) (B) for the indicated times. The lysates were immunoblotted as in (a). (d and e) The cells were stimulated as in (b or c), and accumulated ROS were analysed as in Figure 1c. The similar results were obtained with three independent experiments. BHA, butylated hydroxyanisole; c-FLIP, cellular FLICE-inhibitory protein; ERK, extracellular signal-regulated kinase; JNK, c-Jun N-terminal kinase; ROS, reactive oxygen species; TNF, tumor necrosis factor-; z-VAD-fmk, benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone.

Full figure and legend (164K)

To determine the contribution of caspases and ROS to JNK and ERK activation in c-FLIP knockdown HeLa cells, we tested the inhibitory effects of a broad caspase inhibitor, benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone (z-VAD-fmk) and an antioxidant, butylated hydroxyanisole (BHA) on TNF-induced ROS accumulation, and JNK and ERK activation. Pretreatment of the cells with z-VAD-fmk inhibited the TNF-induced processing of procaspase 8 in c-FLIP knockdown HeLa cells (Figure 2b). Surprisingly, z-VAD-fmk completely inhibited the TNF-induced prolonged JNK and ERK activation, and ROS accumulation in c-FLIP knockdown HeLa cells (Figures 2b and d). Antioxidants including BHA and N-acetyl cystein (NAC) marginally inhibited JNK and ERK activation at 8 h after TNF stimulation (Figure 2b and data not shown). Similarly, anti-Fas antibody induced ROS accumulation and prolonged JNK activation, and these activations were inhibited by z-VAD-fmk (Figures 2c and e). While we previously showed that BHA almost completely inhibited caspase-independent ROS accumulation (Sakon et al., 2003), the inhibitory effects of BHA and NAC on caspase-dependent ROS accumulation were marginal (Figures 2d and e, and data not shown). The reason for this discrepancy is currently unknown.

TNF- and Fas-stimulation induces ROS accumulation and prolonged JNK activation in c-FLIP knockdown cells

We next investigated whether TNF-induced ROS accumulation and prolonged JNK activation are similarly observed in other cells, in which expression of c-FLIP is knocked down. Knockdown of c-FLIP expression by RNAi induced caspase-dependent prolonged JNK and ERK activation, and ROS accumulation in a colon carcinoma, HCT116 and a lung carcinoma, A549 cells after TNF stimulation (Figure 3). Collectively, TNF and anti-Fas antibody induce prolonged JNK and ERK activation and ROS accumulation in a caspase-dependent manner in various cells when expression of c-FLIP_L is repressed.

Figure 3.

TNF induces prolonged JNK and ERK activation, and ROS accumulation in c-FLIP knockdown HCT116 and A549 cells in a caspase-dependent manner. (a and d) Control siRNA or c-FLIP siRNA were transfected into HCT116 (a) or A549 (d) cells, and the efficiency and the specificity of c-FLIP RNAi were determined by immunoblotting with the indicated antibodies. c-FLIP_S was not detected in HCT116 cells. (b and e) The cells were stimulated with TNF (10 ng/ml) in the absence or presence of z-VAD-fmk (50 M) (zV) or BHA (100 M) (B) for the indicated times. The lysates were immunoblotted with the indicated antibodies. (c and f) The cells were stimulated as in (b), and accumulated ROS were analysed as in Figure 1c. The similar results were obtained with two independent experiments. BHA, butylated hydroxyanisole; c-FLIP, cellular FLICE-inhibitory protein; ERK, extracellular signal-regulated kinase; JNK, c-Jun N-terminal kinase; RNAi, RNA interference; ROS, reactive oxygen species; siRNA, small interfering RNA; TNF, tumor necrosis factor-; z-VAD-fmk, benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone.

Full figure and legend (154K)

Caspase-dependent cleavage of MEKK1 is induced in c-FLIP knockdown cells

Although previous studies have revealed that a critical contribution of ROS to prolonged JNK activation, the kinase(s) responsible for prolonged JNK activation remain to be identified (Sakon et al., 2003; Pham et al., 2004; Ventura et al., 2004). On the other hand, a previous study has shown that full-length MEKK1 is cleaved by caspase 3 to release the N-terminal inhibitory domain, thereby generating a constitutively active form of MEKK1 (MEKK1N) (Widmann et al., 1998). Taken that z-VAD-fmk inhibited prolonged JNK and ERK activation in c-FLIP knockdown HeLa cells, MEKK1 may be a candidate that mediates prolonged JNK and ERK. We first tested whether the proteolytic cleavage of MEKK1 is induced in c-FLIP knockdown HeLa cells after TNF stimulation. The full-length MEKK1 with a molecular mass of 200 kDa gradually decreased at 2 h, and the C-terminal fragment of MEKK1 with a molecular mass of 90 kDa appeared at 4 h and then declined at 8 h (Figure 4a). This kinetics of the processing of MEKK1 was well correlated with those of JNK activation (Figure 2a). Notably, pretreatment with z-VAD-fmk completely inhibited the TNF-induced processing of MEKK1, which was also consistent with the inhibitory effect of z-VAD-fmk on JNK and ERK activation (Figure 1b, Figure 4b). On the other hand, activation of ERK was earlier than the processing of MEKK1, suggesting that a kinase other than MEKK1 might be involved in the caspase-dependent ERK activation. Moreover, anti-Fas antibody stimulation also induced the processing of full-length MEKK1 in a caspase-dependent manner in c-FLIP knockdown HeLa cells (Figure 4c). Similarly, the processing of MEKK1 was induced in c-FLIP knockdown HCT116 and A549 cells after TNF stimulation (Figures 4d and e). Together, the caspase-dependent cleavage of MEKK1 may contribute to prolonged JNK activation in c-FLIP knockdown cells.

Figure 4.

Caspase-dependent processing of MEKK1 in c-FLIP knockdown cells. Control siRNA- or c-FLIP siRNA-transfected HeLa (a–c), HCT116 (d) and A549 (e) cells were stimulated with TNF (10 ng/ml) (a, b, d and e) or anti-Fas antibody (100 ng/ml) (c) in the absence or presence of z-VAD-fmk (50 M) (zV) or BHA (100 M) (B) for the indicated times. The lysates were analysed by immunoblotting with anti-MEKK1 antibody. The arrows indicate full-length (FL) and truncated (N) MEKK1. The asterisks indicate nonspecific bands. The similar results were obtained with two independent experiments. BHA, butylated hydroxyanisole; c-FLIP, cellular FLICE-inhibitory protein; MEKK, mitogen-activated protein kinase/ERK kinase kinase; siRNA, small interfering RNA; TNF, tumor necrosis factor-; z-VAD-fmk, benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone.

Full figure and legend (142K)

TNF-induced prolonged JNK activation is a downstream event of caspase activation and does not play a crucial role in TNF-induced cell death in c-FLIP knockdown cells

To determine whether prolonged JNK activation promotes cell death or is a downstream event of caspase activation, we tested whether a JNK-specific inhibitor, SP600125 inhibits TNF-induced processing of caspase 8 and cell death. Although TNF-induced activation of JNK was completely inhibited in the presence of SP600125, SP600125 did not inhibit TNF-induced processing of procaspase 8 and cell death (Figures 5a and b). Together, these results indicate that prolonged JNK activation is a downstream, but not upstream event of activation of caspase (Figure 5a).

Figure 5.

TNF-induced prolonged JNK activation is a downstream event of the caspase cascade. (a) c-FLIP siRNA-transfected HeLa cells were stimulated with TNF (10 ng/ml) in the absence or presence of z-VAD-fmk (zV), Ac-IETD-CHO or SP600125 (SP) for 4 h. The lysates were immunoblotted as in Figure 2. (b) c-FLIP siRNA-transfected HeLa cells were stimulated as in (a) for 16 h, cell viability was determined by WST assay. The results are presented as means.e.s of triplicate samples. The similar results were obtained with two independent experiments. ^*P<0.05 compared to cells treated with TNF alone. c-FLIP, cellular FLICE-inhibitory protein; JNK, c-Jun N-terminal kinase; siRNA, small interfering RNA; TNF, tumor necrosis factor-; WST, water-soluble tetrazolium salt; z-VAD-fmk, benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone.

Full figure and legend (85K)

Our results clearly showed an essential role for the caspase cascade in JNK activation (Figure 2, Figure 3, Figure 4), prompting us to verify that activation of caspase 8 is required for JNK activation. While z-VAD-fmk completely inhibited JNK activation, a caspase 8-specific inhibitor, Ac-IETD-CHO, even at high concentration (100 M) only weakly inhibited the processing of caspase 8, TNF-induced JNK activation and cell death (Figure 5). Thus, we could not conclude that caspase 8 is essential for TNF-induced JNK activation. Further studies will be required to address this issue.

TNF induces both apoptotic and necrotic cell death in c-FLIP knockdown cells

We and others have previously shown that TNF induces caspase-dependent apoptosis and ROS-dependent necrosis in NF-B activation-deficient cells (Sakon et al., 2003; Ventura et al., 2004; Kamata et al., 2005). To investigate whether c-FLIP knockdown HeLa cells die from apoptosis or necrosis upon TNF stimulation, we examined the morphology of c-FLIP knockdown HeLa cells after TNF stimulation by transmission electron microscopy. TNF stimulation alone did not induce cell death in control HeLa cells (data not shown). c-FLIP knockdown HeLa cells showed nuclear condensation and extensive cytoplasmic blebbing, indicating that these cells died from apoptosis (Figure 6). Interestingly, other cells showed prominent vacuolization in the cytoplasm without affecting nuclear morphology, which was consistent with necrotic cell death (Figure 6). Moreover, z-VAD-fmk almost completely inhibited both apoptotic and necrotic cell death in c-FLIP knockdown HeLa cells (Figure 6). Similarly, TNF induced both apoptotic and necrotic cell death in HCT116 and A549 cells, and both types of cell death were completely inhibited in the presence of a caspase inhibitor (Figure 6). Collectively, TNF induces apoptotic and necrotic cell death in c-FLIP knockdown cells in a caspase-dependent manner.

Figure 6.

TNF induces both apoptotic and necrotic cell death in c-FLIP knockdown cells. c-FLIP siRNA-transfected HeLa, HCT116, and A549 cells were untreated or treated with TNF (10 ng/ml) in the absence or presence of z-VAD-fmk (50 M) for 4 h. The morphological changes were examined by transmission electron microscopy and the representative images are shown. The scale bar represents 2 m. c-FLIP, cellular FLICE-inhibitory protein; siRNA, small interfering RNA; TNF, tumor necrosis factor-; z-VAD-fmk, benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone.

Full figure and legend (183K)

TNF induces apoptotic and necrotic cell death in a similar kinetics in c-FLIP knockdown cells

To assess the contribution of apoptosis and necrosis to TNF-induced cell death more quantitatively, we performed time-course analysis of TNF-induced DNA hypoploidy as an indication of apoptosis using flow cytometry. In parallel, we also performed lactate dehydrogenase (LDH) release assay to detect released LDH in the culture supernatants from ruptured cells, a hallmark of necrosis. As shown in Figure 7a, the populations showing DNA hypoploidy and released LDH increased at 4 h and peaked at 8 h after TNF stimulation in c-FLIP knockdown HeLa cells. Similarly, the populations showing DNA hypoploidy and released LDH increased in a similar kinetics in c-FLIP knockdown HCT116 cells (Figure 7b). Together, these results indicate that necrosis in HeLa and HCT116 cells is not a late stage of apoptotic cell death showing necrotic features, but primary necrosis.

Figure 7.

TNF induces apoptotic and necrotic cell death in a similar kinetics in c-FLIP knockdown cells. Control siRNA- and c-FLIP siRNA-transfected HeLa (a) and HCT116 (b) cells were stimulated with TNF (10 ng/ml) for the indicated times. The populations showing DNA hypoploidy and % LDH release were calculated as in the Materials and methods. The results of % LDH release are presented as means.e.s of triplicate samples. The similar results were obtained with two independent experiments. c-FLIP, cellular FLICE-inhibitory protein; % LDH, percentage of lactate dehydrogenase release; siRNA, small interfering RNA; TNF, tumor necrosis factor-.

Full figure and legend (15K)

Discussion

In the present study, we have shown that knockdown of c-FLIP induces caspase-dependent ROS accumulation and prolonged JNK activation, followed by necrotic and apoptotic cell death.

Consistent with our study using c-Flip^-/- MEFs (Nakajima et al., 2006), knockdown of c-FLIP was sufficient to promote caspase-dependent JNK activation and ROS accumulation. Notably, given that the expression of other antiapoptotic proteins, such as TRAF2, RIP and XIAP, was not reduced in c-FLIP knockdown cells (Figure 1a), these proteins could not compensate the defect of c-FLIP function. Although the expression of many antiapoptotic genes and antioxidant enzyme genes are regulated by NF-B, c-FLIP plays a dominant role in suppression of caspase-dependent JNK activation and ROS accumulation upon TNF stimulation. In contrast to c-Flip^-/-, relA^-/- and traf2^-/- traf5^-/- MEFs (Sakon et al., 2003; Nakajima et al., 2006), TNF did not induce caspase-independent JNK activation and ROS accumulation in c-FLIP knockdown tumor cells, since z-VAD-fmk treatment alone almost completely inhibited these activation (Figure 2, Figure 3). One possible explanation would be that residual expression of c-FLIP_L in c-FLIP knockdown cells might be sufficient to inhibit the caspase-independent pathway(s) leading to JNK activation and ROS accumulation through suppression of the positive feedback loop between JNK and ROS. Alternatively, these differences might come from the different cellular background (MEFs versus epithelial tumor cells).

A recent study has shown that one of the mechanisms by which ROS induce prolonged JNK activation is to inactivate MAP kinase phosphatases by oxidizing cysteine residues critical for their phosphatase activities that would otherwise dephosphorylate and inactivate MAPKs (Kamata et al., 2005). In addition to this mechanism, our present study has demonstrated that activation of the caspase cascade induces the processing of MEKK1, thereby generating a constitutively active form (Figure 4). These results are consistent with our recent findings, in which c-FLIP_L suppresses caspase-dependent JNK activation (Nakajima et al., 2006). It is likely that a kinase other than MEKK1 might also contribute to prolonged JNK and ERK activation. Indeed, several kinases including p21-activated kinase2 and mammalian Sterile20-like kinases have been shown to be cleaved by caspases, and activate the JNK pathway (Lee et al., 1998; Rudel et al., 1998). In addition, given that ROS activate MAPKKKs including apoptosis signal-regulating kinase (ASK)1 (Saitoh et al., 1998), ASK1 might also contribute to prolonged JNK activation in c-FLIP knockdown cells. Moreover, the kinetics of ERK activation was earlier than the cleavage of MEKK1 (Figure 2a, Figure 4a). Collectively, MEKK1 is one of the kinases, but other kinase(s) might also contribute to prolonged JNK and ERK activation in response to TNF in a caspase-dependent manner.

Contribution of the JNK pathway to TNF-induced cell death has been highly debated. Several studies have shown that activation of the JNK pathway, especially prolonged JNK activation is essential for TNF- and stress-induced cell death (Tournier et al., 2000; Tobiume et al., 2001; Deng et al., 2003). In contrast, the JNK-dependent pathway might not play a crucial role in TNF-induced cell death in c-FLIP knockdown cells (Figure 5). Under the conditions, in which c-FLIP_L is downregulated, prolonged JNK activation is a downstream event of activation of the caspase cascade (Figures 2, Figure 3, Figure 5). Therefore, inhibition of the JNK pathway did not suppress TNF-induced caspase activation and cell death in c-FLIP knockdown cells (Figure 5). Taken that prolonged JNK activation enhances activity of an E3 ligase, itchy homolog E3 ubiquitin protein ligase (ITCH), which subsequently degrades c-FLIP_L in primary hepatocytes together (Chang et al., 2006), one possible scenario to explain these discrepancies would be that proapoptotic function of JNK might be largely mediated through degradation of c-FLIP_L. Consistently, a previous study has shown that, in the absence of c-FLIP_L, the TNF-induced signaling complex II containing FADD and caspase 8/10 easily forms, resulting in apoptosis (Micheau and Tschopp, 2003).

While necrosis is characterized by vacuolation of the cytoplasm, cytoplasmic vacuoles are also detected in cells undergoing autophagy. Autophagy is a major intracellular degradation mechanism for longed-lived proteins and organelles, and operates and promotes survival during starvation (Ohsumi, 2001; Levine and Klionsky, 2004). During autophagy, vacuoles engulfed cytoplasmic organelles appear and are called autophagosomes. Recent studies have shown that autophagy is also associated with nonapoptotic cell death (Edinger and Thompson, 2004; Levine and Yuan, 2005), prompting us to test whether TNF-induced vacuoles are associated with autophagy. However, our preliminary experiments could not detect a punctate pattern of LC3 (Light-chain 3), a hallmark of autophagosome, in c-FLIP knockdown HeLa cells (data not shown). Therefore, these vacuoles in c-FLIP knockdown cells might not be associated with autophagy.

The most important findings of our study are that TNF induces caspase-dependent JNK activation and ROS accumulation in several tumor cells followed by necrotic cell death, in which the expression of c-FLIP is downregulated. Upregulation of c-FLIP is frequently observed in various type tumors and might be responsible for escaping from death receptor-induced cell death (Djerbi et al., 1999; Medema et al., 1999; Budd et al., 2006). Therefore, downregulation of c-FLIP by RNAi might be a promising strategy to treat these tumors through promoting caspase-dependent cell death. On the other hand, inflammatory responses to stress-mediated or oncogene-activated necrosis in tumors stimulate angiogenesis and tumor cell proliferation (Balkwill and Coussens, 2004; Vakkila and Lotze, 2004). Moreover, cell death, especially necrotic cell death of tumors promotes inflammatory responses accompanied with compensatory proliferation (Maeda et al., 2005). While the release of a DNA-binding protein of High-mobility group box 1 protein from necrotic cells might play a dominant role in eliciting inflammatory responses in the surrounding tissues, production of ROS also induces inflammation and elicits DNA damage (Vakkila and Lotze, 2004). Collectively, accumulated ROS and/or necrotic cell death associated with tumor therapy might induce DNA damage and inflammation, resulting in compensatory proliferation of tumor cells. In this respect, elucidating the mechanism underlying caspase-dependent ROS accumulation will be crucial to prevent this potential side effect induced by apoptosis promoting tumor therapy.

Materials and methods

Reagents and cell culture

Recombinant human TNF were purchased from BD Biosciences (San Jose, CA, USA). BHA, z-VAD-fmk, Ac-IETD-CHO, 5-(and-6)-carboxy-2',7'-dichlorodihydrofluorescein diacetate (CM-H₂DCFDA) and SP600125 were purchased from Wako (Osaka, Japan), Peptide Institute (Osaka, Japan), Molecular Probes (Eugene, OR, USA), Sigma-Aldrich (St Louis, MO, USA). Anti-c-FLIP (Alexis, Lausanne, Switzerland), anti-RIP (BD Biosciences), anti-TRAF2 (BD Biosciences), anti-XIAP (BD Biosciences), anti-caspase 8 (Cell Signaling Technology, Beverly, MA, USA), anti-Fas (MBL, Nagoya, Japan), anti-MEKK1 and anti-IB (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and anti-tubulin (Sigma-Aldrich) antibodies were purchased from the indicated sources. Antibodies specific for phospho-JNK, phospho-ERK, total JNK and total ERK were purchased from Cell Signaling Technology. HeLa cells were cultured in high-glucose Dulbecco's modified Eagle's medium containing 10% fetal calf serum (FCS). A549 cells were cultured in RPMI1640 containing 10% FCS. HCT116 cells were cultured in McCoy's 5A containing 10% FCS.

Measurement of ROS accumulation

Cells (2–5 10⁵ cells) were stimulated with TNF or anti-Fas antibody for the indicated times, and then labeled with CM-H₂DCFDA (1 M) for 30 min at 37°C. The cells were analysed on a flow cytometer (FACSCalibur, BD Biosciences). Data were processed by using the CellQuest program (BD Biosciences).

Western blotting

Cells (5 10⁵ cells) were lysed in RIPA buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, 0.1% sodium dodecyl sulfate, 25 mM -glycerophosphate, 1 mM sodium orthovanadate, 1 mM sodium fluoride, 1 mM phenylmethylsulfonyl fluoride, 1 g/ml aprotinin and 1 g/ml leupeptin). Western blotting was performed as described previously (Sakon et al., 2003).

Small interfering RNAs

HeLa, HCT116 and A549 cells (2–4 10⁶ cells) were transfected with siRNA oligos using Nucleofector (Amaxa, GmbH, Cologne, Germany). Duplex siRNA oligos with two nucleotides overhang at the 3'-end of the sequence were purchased from Qiagen (Valencia, CA, USA). The target sequences were as follows: human c-FLIP (a), 5'-GGAGCAGGGACAAGUUACA-3'; human c-FLIP (b), 5'-GCAAGGAGAAGAGUUUCUU-3'; GFP, 5'-GGCUACGUCCAGGAGCGCACC-3'. To evaluate the transfection efficiency of siRNA oligos, the cells were transfected with an expression vector for GFP, and GFP-positive cells were analysed by flow cytometry. The transfection efficiency was more than 80% based on GFP-positive cells.

WST assay

HeLa (1 10⁴ cells) were plated in 96-well plates and cultured for 12 h. Then the cells were stimulated with TNF (10 ng/ml) in the presence or absence of various inhibitors for 16 h. Cell viability was determined by water-soluble tetrazolium salt (WST) assay using a Cell Counting kit (Dojindo, Kumamoto, Japan).

Electron microscopy

c-FLIP siRNA-transfected HeLa, HCT116 and A549 cells (4 10⁶ cells) were unstimulated or stimulated with TNF in the presence or absence of z-VAD-fmk for 4 h, and then serially fixed with 2% glutaraldehyde in PBS for 2 h and then with 2% OsO₄ for 2 h before embedding in Epon 812. Thin sections were prepared using an MT-5000 ultramicrotome (Dupont Pharmaceuticals, Wilmington, DE, USA), stained with uranyl acetate followed by lead citrate, and then observed on a JEM1230 electron transmission microscope (JEOL).

Measurement of DNA hypoploidy

DNA hypoploidy was measured as described previously (Nakayama et al., 2002). In brief, control siRNA- or c-FLIP siRNA-transfected HeLa and HCT116 cells (4 10⁵ cells) were untreated or treated with TNF (10 ng/ml) for the indicated times. Then, the cells were harvested and incubated in a buffer containing 50 g/ml propidium iodide, 0.05% NP-40, 4 mM sodium citrate (pH 7.2) and 450 g/ml RNase for 10 min at 4°C, followed by the addition of 1.5 M NaCl. The cells were analysed on a FACSCalibur. Data were processed by using the CellQuest program.

LDH release assay

LDH release assay was performed using cytotoxicity detection kit (LDH) according to the manufacturer's instruction (Roche Diagnostics GmbH, Mannheim, Germany). Briefly, control siRNA- or c-FLIP siRNA-transfected HeLa and HCT116 cells (4 10⁵ cells) were untreated or treated with TNF (10 ng/ml) for the indicated times. Then, the culture supernatants were collected and LDH activities were determined by incubating with LDH substrates and measured by a spectrophotometer (Bio-Rad, Hercules, CA, USA). The percentage of LDH release (% LDH) was calculated by the following formula. % LDH release=(LDH activity at the indicated times–LDH activity at time 0) 100/(maximal LDH activity–LDH activity at time 0). Triton-X 100 was added to a final concentration of 1% for the determination of maximal release of LDH activity.

Statistical analysis

Statistical analysis was performed by Student t-test. P value <0.05 was considered to be significant.

References

1. Balkwill F, Coussens LM. (2004). Cancer: an inflammatory link. Nature 431: 405–406. | Article | PubMed | ISI | ChemPort |
2. Barkett M, Gilmore TD. (1999). Control of apoptosis by Rel/NF-B transcription factors. Oncogene 18: 6910–6924. | Article | PubMed | ISI | ChemPort |
3. Budd RC, Yeh WC, Tschopp J. (2006). c-FLIP regulation of lymphocyte activation and development. Nat Rev Immunol 6: 196–204. | Article | PubMed | ISI | ChemPort |
4. Chang L, Kamata H, Solinas G, Luo JL, Maeda S, Venuprasad K et al. (2006). The E3 Ubiquitin Ligase Itch Couples JNK Activation to TNF-induced Cell Death by Inducing c-FLIP(L) Turnover. Cell 124: 601–613. | Article | PubMed | ISI | ChemPort |
5. Deng Y, Ren X, Yang L, Lin Y, Wu X. (2003). A JNK-dependent pathway is required for TNF-induced apoptosis. Cell 115: 61–70. | Article | PubMed | ISI | ChemPort |
6. Djerbi M, Screpanti V, Catrina AI, Bogen B, Biberfeld P, Grandien A. (1999). The inhibitor of death receptor signaling, FLICE-inhibitory protein defines a new class of tumor progression factors. J Exp Med 190: 1025–1032. | Article | PubMed | ISI | ChemPort |
7. Edinger AL, Thompson CB. (2004). Death by design: apoptosis, necrosis and autophagy. Curr Opin Cell Biol 16: 663–669. | Article | PubMed | ISI | ChemPort |
8. Giorgio M, Migliaccio E, Orsini F, Paolucci D, Moroni M, Contursi C et al. (2005). Electron transfer between cytochrome c and p66Shc generates reactive oxygen species that trigger mitochondrial apoptosis. Cell 122: 221–233. | Article | PubMed | ISI | ChemPort |
9. Kamata H, Honda S, Maeda S, Chang L, Hirata H, Karin M. (2005). Reactive oxygen species promote TNF-induced death and sustained JNK activation by inhibiting MAP kinase phosphatases. Cell 120: 649–661. | Article | PubMed | ISI | ChemPort |
10. Karin M, Lin A. (2002). NF-B at the crossroads of life and death. Nat Immunol 3: 221–227. | Article | PubMed | ISI | ChemPort |
11. Lee KK, Murakawa M, Nishida E, Tsubuki S, Kawashima S, Sakamaki K et al. (1998). Proteolytic activation of MST/Krs, STE20-related protein kinase, by caspase during apoptosis. Oncogene 16: 3029–3037. | Article | PubMed | ISI | ChemPort |
12. Levine B, Klionsky DJ. (2004). Development by self-digestion: molecular mechanisms and biological functions of autophagy. Dev Cell 6: 463–477. | Article | PubMed | ISI | ChemPort |
13. Levine B, Yuan J. (2005). Autophagy in cell death: an innocent convict? J Clin Invest 115: 2679–2688. | Article | PubMed | ISI | ChemPort |
14. Maeda S, Kamata H, Luo JL, Leffert H, Karin M. (2005). IKK couples hepatocyte death to cytokine-driven compensatory proliferation that promotes chemical hepatocarcinogenesis. Cell 121: 977–990. | Article | PubMed | ISI | ChemPort |
15. Medema JP, de Jong J, van Hall T, Melief CJ, Offringa R. (1999). Immune escape of tumors in vivo by expression of cellular FLICE-inhibitory protein. J Exp Med 190: 1033–1038. | Article | PubMed | ISI | ChemPort |
16. Micheau O, Tschopp J. (2003). Induction of TNF receptor I-mediated apoptosis via two sequential signaling complexes. Cell 114: 181–190. | Article | PubMed | ISI | ChemPort |
17. Nakajima A, Komazawa-Sakon S, Takekawa M, Sasazuki T, Yeh WC, Yagita H et al. (2006). An antiapoptotic protein, c-FLIP_L, directly binds to MKK7 and inhibits the JNK pathway. EMBO J 25: 5549–5559. | Article | PubMed | ISI | ChemPort |
18. Nakano H, Nakajima A, Sakon-Komazawa S, Piao JH, Xue X, Okumura K. (2006). Reactive oxygen species mediate crosstalk between NF-B and JNK. Cell Death Differ 13: 730–737. | Article | PubMed | ISI | ChemPort |
19. Nakayama M, Ishidoh K, Kayagaki N, Kojima Y, Yamaguchi N, Nakano H et al. (2002). Multiple pathways of TWEAK-induced cell death. J Immunol 168: 734–743. | PubMed | ISI | ChemPort |
20. Ohsumi Y. (2001). Molecular dissection of autophagy: two ubiquitin-like systems. Nat Rev Mol Cell Biol 2: 211–216. | Article | PubMed | ISI | ChemPort |
21. Pham CG, Bubici C, Zazzeroni F, Papa S, Jones J, Alvarez K et al. (2004). Ferritin heavy chain upregulation by NF-B inhibits TNF-induced apoptosis by suppressing reactive oxygen species. Cell 119: 529–542. | Article | PubMed | ISI | ChemPort |
22. Ricci JE, Munoz-Pinedo C, Fitzgerald P, Bailly-Maitre B, Perkins GA, Yadava N et al. (2004). Disruption of mitochondrial function during apoptosis is mediated by caspase cleavage of the p75 subunit of complex I of the electron transport chain. Cell 117: 773–786. | Article | PubMed | ISI | ChemPort |
23. Rudel T, Zenke FT, Chuang TH, Bokoch GM. (1998). p21-activated kinase (PAK) is required for Fas-induced JNK activation in Jurkat cells. J Immunol 160: 7–11. | PubMed | ISI | ChemPort |
24. Saitoh M, Nishitoh H, Fujii M, Takeda K, Tobiume K, Sawada Y et al. (1998). Mammalian thioredoxin is a direct inhibitor of apoptosis signal-regulating kinase (ASK)1. EMBO J 17: 2596–2606. | Article | PubMed | ISI | ChemPort |
25. Sakon S, Xue X, Takekawa M, Sasazuki T, Okazaki T, Kojima Y et al. (2003). NF-B inhibits TNF-induced accumulation of ROS that mediate prolonged MAPK activation and necrotic cell death. EMBO J 22: 3898–3909. | Article | PubMed | ISI | ChemPort |
26. Tobiume K, Matsuzawa A, Takahashi T, Nishitoh H, Morita K, Takeda K et al. (2001). ASK1 is required for sustained activations of JNK/p38 MAP kinases and apoptosis. EMBO R 2: 222–228. | Article | ISI | ChemPort |
27. Tournier C, Hess P, Yang DD, Xu J, Turner TK, Nimnual A et al. (2000). Requirement of JNK for stress-induced activation of the cytochrome c-mediated death pathway. Science 288: 870–874. | Article | PubMed | ISI | ChemPort |
28. Vakkila J, Lotze MT. (2004). Inflammation and necrosis promote tumour growth. Nat Rev Immunol 4: 641–648. | Article | PubMed | ISI | ChemPort |
29. Ventura JJ, Cogswell P, Flavell RA, Baldwin Jr AS, Davis RJ. (2004). JNK potentiates TNF-stimulated necrosis by increasing the production of cytotoxic reactive oxygen species. Genes Dev 18: 2905–2915. | Article | PubMed | ISI | ChemPort |
30. Widmann C, Gerwins P, Johnson NL, Jarpe MB, Johnson GL. (1998). MEK kinase 1, a substrate for DEVD-directed caspases, is involved in genotoxin-induced apoptosis. Mol Cell Biol 18: 2416–2429. | PubMed | ISI | ChemPort |
31. Yeh WC, Itie A, Elia AJ, Ng M, Shu HB, Wakeham A et al. (2000). Requirement for Casper (c-FLIP) in regulation of death receptor-induced apoptosis and embryonic development. Immunity 12: 633–642. | Article | PubMed | ISI | ChemPort |

Acknowledgements

We thank Y Gotoh, H Nishina, M Takekawa and T Ueno for providing reagents and helpful discussion. This work was supported in part by Grants-in-Aid for 21st Century COE Research and Scientific Research (B) from Japan Society for the Promotion of Science, Japan, a Grant from Human Frontier Science Program (HFSP), and grants from the Takeda Science Foundation, the Tokyo Biochemical Research Foundation, and NOVARTIS Foundation (Japan) for the Promotion of Science.