NF-κB Protects Macrophages from Lipopolysaccharide-induced Cell Death

Macrophages play a pivotal role in the pathogenesis of a variety of diseases. These studies were performed to characterize the mechanisms by which Toll-like receptor 4 (TLR4)-mediated NF-κB activation promotes resistance to cell death in macrophages. When NF-κB activation was inhibited by a super-repressor, IκBα, the TLR4 ligand lipopolysaccharide induced the activation of caspase 8, the loss of mitochondrial transmembrane potential (ΔΨm), and apoptotic cell death in macrophages. The inhibition of caspase 8 activation suppressed DNA fragmentation but failed to protect macrophages against the loss of ΔΨm and resulted in necrotic cell death. In contrast, the reduction of receptor-interacting protein 1 suppressed the loss of ΔΨm and inhibited apoptotic cell death. Further, when caspase 8 activation was suppressed, the knock down of receptor-interacting protein inhibited the loss of ΔΨm and necrotic cell death. These observations demonstrate that following TLR4 ligation by lipopolysaccharide, NF-κB is a critical determinant of macrophage life or death, whereas caspase 8 determines the pathway employed.

Macrophages play a pivotal role in the pathogenesis of a variety of diseases. These studies were performed to characterize the mechanisms by which Toll-like receptor 4 (TLR4)-mediated NF-B activation promotes resistance to cell death in macrophages. When NF-B activation was inhibited by a super-repressor, IB␣, the TLR4 ligand lipopolysaccharide induced the activation of caspase 8, the loss of mitochondrial transmembrane potential (⌬⌿ m ), and apoptotic cell death in macrophages. The inhibition of caspase 8 activation suppressed DNA fragmentation but failed to protect macrophages against the loss of ⌬⌿ m and resulted in necrotic cell death. In contrast, the reduction of receptor-interacting protein 1 suppressed the loss of ⌬⌿ m and inhibited apoptotic cell death. Further, when caspase 8 activation was suppressed, the knock down of receptor-interacting protein inhibited the loss of ⌬⌿ m and necrotic cell death. These observations demonstrate that following TLR4 ligation by lipopolysaccharide, NF-B is a critical determinant of macrophage life or death, whereas caspase 8 determines the pathway employed.
Macrophages play a pivotal role in infection, atherosclerosis, and chronic inflammation such as observed in rheumatoid arthritis. The ligation of Toll-like receptors (TLRs) 3 activates intracellular signal transduction pathways, including NF-B (1). Expressed on macrophages and monocytes, TLR4 is critical for the recognition of lipopolysaccharide (LPS) from Gram-negative bacteria. Recent studies have demonstrated that endogenous TLR4 ligands are highly expressed at sites of chronic inflammation, characterized by the accumulation of macrophages, such as the joints of patients with rheumatoid arthritis (2)(3)(4)(5)(6)(7)(8). Resistance to apoptosis may contribute to the persistence of chronic inflammation (9). Therefore, microbial and endogenous TLR4 ligands may promote resistance to apoptosis through the activation of NF-B.
Apoptosis may be initiated by two different pathways, the death receptor-mediated pathway and the mitochondria-dependent pathway.
Death receptors, such as Fas and TNFR1, possess intracellular death domains that may initiate apoptosis through the recruitment of Fasassociated death domain protein (FADD) and the activation of caspase 8. Activated caspase 8 may directly activate caspase 3, which results in apoptotic cell death (reviewed in Ref. 9). In the mitochondria-dependent pathway, apoptotic signals induce the loss of mitochondrial integrity and the release of pro-apoptotic molecules, including cytochrome c. Once in the cytosol, cytochrome c binds to apoptotic protease activating factor-1, resulting in activation of caspases 9 and 3, thus triggering apoptosis (10). These two pathways may be linked by the pro-apoptotic Bcl-2 family member Bid, which may be cleaved by caspase 8, resulting in truncated Bid (tBid), which may trigger the mitochondrial pathway.
Although TLR4 ligation induces NF-B activation, which may promote cell survival through the induction of anti-apoptotic genes, recent evidence has demonstrated that TLR4 activation may also result in macrophage apoptosis (11)(12)(13)(14)(15). Prior studies have demonstrated that caspase 8 was activated in LPS-induced apoptosis (13,16), but the essential role of caspase 8 in the eventual demise of the macrophage has not been clearly documented. Additionally, receptor-interacting protein 1 (RIP), which is important for TNFR1-or TLR3-induced NF-B activation (17)(18)(19), has been shown to contribute to TNFR1, Fas, and TNFrelated apoptosis-inducing ligand (TRAIL)-mediated cell death (20 -22). Therefore, studies were performed to define the role of caspase 8 and RIP in LPS-induced macrophage cell death when NF-B activation is inhibited.
Employing in vitro monocyte-differentiated macrophages, the current study has demonstrated that, following the inhibition of NF-B activation, LPS induced caspase 8 activation and apoptotic cell death. In contrast to expectations, the inhibition of caspase 8 did not prevent the loss of mitochondrial transmembrane potential (⌬⌿ m ) and macrophage viability was not reduced. Additionally, when caspase 8 activation was suppressed, the mode of cell death was non-apoptotic. Our data suggest that RIP mediated not only apoptotic, but also the non-apoptotic, death of macrophages by different mechanisms. Further, caspase 8 was critical in determining the mode of cell death through its effects on RIP. rent centrifugal elutriation (JE-6B; Beckman Coulter, Palo Alto, CA) in the presence of 10 g/ml polymyxin B sulfate as previously described (23)(24)(25)(26). Isolated monocytes were Ն90% pure as determined by morphology, nonspecific esterase staining, and CD14 expression examined by flow cytometry (data not shown). Monocytes were adhered to plates for 1 h in RPMI and 1 g/ml polymyxin B sulfate. Following adherence, monocytes were differentiated in vitro for 7 days in RPMI containing 20% heat-inactivated FBS, 1 g/ml polymyxin B sulfate, 0.35 mg/ml L-glutamine, 120 units/ml penicillin and streptomycin (20% FBS/RPMI) (23)(24)(25)(26)(27).
Adenovirus Infection of Primary Macrophages-Primary macrophages were infected at a concentration of 50 m.o.i. (multiplicity of infection) with adenoviral vectors expressing a super-repressor IB␣ (Ad IB␣) or a control vector (␤-galactosidase or CMV-blank, Ad Control) for 2 h as previously described (23,25,26,28). Where indicated, macrophages were co-infected with Ad DN FADD (5 m.o.i.), Ad Bcl-x L (6 m.o.i.), AdCrmA (25 m.o.i.) or Ad control (5, 6, or 25 m.o.i., respectively). Within each experiment the total concentration of adenoviral vector was held constant. After infection, 20% FBS/RPMI was added, and the cells were incubated overnight. The macrophages were then washed and incubated in 20% FBS/RPMI for an additional 12 h (total of 32 h from the initiation of the infection). The infected cells were treated with LPS (10 ng/ml) or control medium for 12 or 24 h. The cells were then harvested and analyzed as described.
Cell Transfection-In vitro differentiated macrophages were transfected with either nonspecific or RIP siRNA (final,100 nM; Dharmacon, Inc., Lafayette, CO) employing Lipofectamine 2000 according to the manufacturer's directions (Invitrogen). The cells were then incubated for 72 h prior to analysis by immunoblot assay or adenoviral infection.
Electron Microscopy-In vitro differentiated macrophages cultured in the presence or absence of Ad IB␣, LPS, and IETD were harvested, and cell pellets were fixed overnight at 4°C in a 0.2 M sodium cacodylate buffer (pH 7.4) containing 2% glutaraldehyde. Samples were post-fixed in cacodylate-buffered 1% osmium tetroxide, dehydrated, and embedded in Epon 812 (Nacalai Tesque, Osaka, Japan) for ultrathin sectioning. The samples were stained with uranyl acetate and lead citrate and examined with a JEOL 1220 electron microscope.
Analysis of Mitochondrial Transmembrane Potential and Cell Membrane Integrity-Mitochondrial dysfunction was assessed utilizing the cationic lipophilic green fluorochrome Rh123 as previously described (23,25,26,28,29). Disruption of ⌬⌿ m is associated with a lack of Rh123 retention and a decrease in fluorescence. Cultures were incubated with Rh123 (0.1 g/ml) for 30 min, harvested, and washed by PBS. To assess cell membrane integrity simultaneously, PI (3.3 g/ml) was added to the cells prior to analysis by flow cytometry. For histogram analysis, objects with minimal light scatter representing debris were gated out.
Determination of Subdiploid DNA Content-At the indicated time points, cells were harvested, fixed in 70% ethanol, and stained with PI (50 g/ml) as previously described (30). The apoptotic profile was determined by flow cytometry utilizing a Beckman-Coulter EpicsXL flow cytometer and system 2 software. The subdiploid DNA peak (Ͻ2N DNA) immediately adjacent to the G 0 /G 1 peak (2N DNA) represented apoptotic cells and was quantified by histogram analyses. Objects with minimal light scatter representing debris were excluded as previously described (31), so that quantitation of the subdiploid population would not be inappropriately skewed (23,25,26,28).
Immunoblot Analysis-Whole-cell extracts were prepared from 7-day differentiated macrophages that were treated as indicated. Extracts were electrophoresed on SDS-PAGE 12.5% polyacrylamide gels and transferred to Immobilon-P (Millipore, Bedford, MA) by semi-dry blotting. Membranes were blocked for 1 h at room temperature in PBS/0.2% Tween 20/5% nonfat dry milk (PBS/Tween/milk). The membranes were then incubated overnight at 4°C in PBS/Tween/milk with the indicated antibodies: anti-caspase 3, anti-Bcl-x L , anti-RIP (BD Transduction Laboratories, San Diego, CA), anti-caspase 8 (Cell Signaling Technology, Beverly, MA), or anti-tubulin (Calbiochem, San Diego, CA). Membranes were washed in PBS/Tween/milk and incubated with either donkey anti-rabbit (1:5000) or anti-mouse (1:2000) secondary antibody conjugated to horseradish peroxidase (Amersham Biosciences). Visualization of the protein bands was performed employing the Enhanced Chemiluminescence Plus kit as recommended by the manufacturer (Amersham Biosciences).
Caspase Activity Assay-Seven-day differentiated human macrophages were treated as indicated and harvested. Cell lysates were prepared as instructed by the manufacturer. The lysates were incubated for 1 h at 37°C with the caspase 8 (Ac-IETD-AFC) synthetic fluorogenic substrate. Samples were read on a fluorometer at 400 nm excitation and 505 nm emission.
Caspase Inhibition-Macrophages were infected with control or IB␣-expressing adenoviral vectors as described above. The cells were then pretreated for 1 h with broad specificity caspase inhibitor z-VADfmk (20 M) or the caspase 8 inhibitor IETD-fmk (20 M) (Enzyme System Products, Livermore, CA) or vehicle control Me 2 SO, followed by culture with 10 ng/ml LPS for an additional 12 h. The cells were harvested and the lysates examined as described under "Results." Statistical Analysis-Statistical significances between groups were determined by two-tailed Student's t-test.

LPS Induces Macrophage Apoptosis following NF-B Inhibition-To
characterize the mechanism of LPS-induced apoptosis following NF-B inhibition, in vitro differentiated human macrophages were infected with a control adenoviral vector or one expressing a super-repressor IB␣ (Ad IB␣) that possesses serine-to-alanine mutations that retard its degradation (32). LPS (10 ng/ml)-induced cell death, determined by the loss of cell membrane integrity using the PI exclusion assay (Fig. 1A), was significantly increased at 12 h and further increased at 24 h in macrophages infected with Ad IB␣ compared with those infected with the control adenoviral vector (p Ͻ0.001). The mode of cell death was apoptotic, defined by DNA fragmentation (Ͻ2N DNA) (Fig. 1B), which also increased from 12 to 24 h (p Ͻ0.001). Apoptotic cell death was accompanied by the loss of ⌬⌿ m , determined by the Rh123 retention assay (Fig. 1C). The ⌬⌿ m was greatly reduced by 12 h, with further reduction by 24 h, suggesting an important role for the loss of mitochondrial integrity in LPS-induced cell death. These observations demonstrate that LPS induced the loss of ⌬⌿ m and macrophage apoptotic cell death when NF-B activation was inhibited.
LPS-induced Apoptosis following NF-B Inhibition Is Not Mediated by TNFR1-We previously demonstrated that when NF-B activation was inhibited TNF␣ induced macrophage apoptosis (25). Because LPS induces the production of TNF␣, we next examined whether TNF␣ was involved in LPS-induced macrophage apoptosis. TNF␣ signals through two separate receptors, TNFR1 and TNFR2. TNFR1 contains a death domain that mediates apoptosis through the activation of caspase 8 (33,34). Macrophages were infected with Ad IB␣ or the control adenoviral vector and then incubated with a neutralizing anti-TNFR1 antibody for 1 h before the addition of LPS or TNF␣. Although the neutralizing antibody significantly (p Ͻ0.01) reduced TNF␣-mediated apoptotic cell death (Fig. 2, A-C), it had no impact on LPS-induced cell death ( Fig. 2A), DNA fragmentation (Fig. 2B), and the loss of ⌬⌿ m (Fig. 2C) in macro-phages infected with Ad IB␣. These results demonstrate that LPSinduced apoptosis in macrophages was not mediated by TNFR1.
Caspase 8 Is Necessary for LPS-induced DNA Fragmentation but Not Mitochondrial Dysfunction or Cell Death-Because previous studies demonstrated that caspase 8 was activated by LPS when NF-B activation was suppressed (16), the role of caspase 8 was examined. Macrophages were infected with the control adenoviral vector or Ad IB␣ and then treated with LPS for 12 h. Cells were harvested, and the lysates were analyzed for caspase 8 activation employing a substrate-specific fluorometric assay. Compared with the control-infected macrophages, LPS induced a 10 -15-fold activation of caspase 8-like activity when NF-B activation was inhibited (Fig. 3A). Because the caspase 8 inhibitor IETDfmk suppressed Ͼ90% of the LPS-induced caspase 8-like activity (Fig.  3A), this inhibitor was employed to define the effects of caspase 8 inhibition on apoptotic cell death. The inhibition of caspase 8-like activity with IETD-fmk markedly suppressed LPS-induced DNA fragmentation following NF-B inhibition (Fig. 3B). In contrast, IETD-fmk did not protect macrophages from the loss of ⌬⌿ m (Fig. 3C) or cell death (Fig.  3D), which actually increased. To define the mode of cell death, electron microscopy was performed. Macrophages infected with Ad IB␣ and treated with IETD-fmk plus LPS demonstrated characteristics of necrotic cell death, including vacuolization of the cytoplasm, ruptured cell membrane, and lack of DNA fragmentation (Fig. 3E).
Bcl-x L Protects against Apoptosis, Not Necrosis-To further define the role of the mitochondria in LPS-induced apoptosis, Bcl-x L was ectopically expressed, employing an adenoviral vector (Ad Bcl-x L ) in macrophages co-infected with Ad IB␣ or Ad control. In macrophages infected with Ad IB␣, ectopically expressed Bcl-x L (Fig. 3B) significantly (p Ͻ0.001) suppressed the LPS-induced DNA fragmentation, loss of ⌬⌿ m (Fig. 3C), and cell death (Fig. 3D). Macrophages were also examined to define the effects of Bcl-x L , when caspase 8 activity was also blocked. The combination of IETD-fmk plus Bcl-x L completely prevented DNA fragmentation (Fig. 3B). In contrast, the protection provided by Bcl-x L against the loss of ⌬⌿ m and the induction of cell death was lost when caspase 8 activity was suppressed (Fig. 3, C and D). Therefore, Bcl-x L protected against apoptotic, but not against non-apoptotic or necrotic, cell death. These observations suggest that caspase 8 activation regulates the mode of LPS-induced macrophage cell death when NF-B activation is repressed.
FADD Is Upstream of Caspase 8-An earlier study suggested that TLR4-induced apoptosis was mediated through FADD (12). Therefore, studies were performed to determine the effect of a dominant negative (DN) FADD on caspase 8-like activity and on apoptosis. The adenoviralmediated expression of a DN FADD suppressed caspase 8-like activity (Fig. 4A), suggesting that FADD was upstream of caspase 8 in LPSinduced apoptosis. Consistent with its effects on caspase 8, the expression of the DN FADD also suppressed LPS-induced DNA fragmentation (Fig. 4B). In contrast, the DN FADD failed to prevent either cell death or the loss of ⌬⌿ m (data not shown), consistent with the results observed with IETD-fmk. These observations suggest that FADD is upstream of caspase 8, which contributes to the apoptotic phenotype.
RIP Contributes to LPS-induced Macrophage Apoptotic Cell Death-Because RIP may contribute to TNF␣-induced non-apoptotic cell death (35) and because activated caspase 8 may cleave RIP, the effect of LPS-   DECEMBER 23, 2005 • VOLUME 280 • NUMBER 51

JOURNAL OF BIOLOGICAL CHEMISTRY 41829
induced caspase 8 activation on RIP expression was examined. LPS induced the cleavage of RIP when NF-B activation was inhibited, and this was prevented when caspase 8 activity was suppressed by IETD-fmk (Fig. 5A). To further define the role of RIP in TLR4-mediated apoptosis, the expression of RIP was suppressed employing siRNA (Fig. 5B). Compared with nonspecific control siRNA, the reduction of RIP resulted in protection against LPS-induced cell death (Fig. 5B). Further, the suppression of RIP also protected against DNA fragmentation (Fig. 5C) and  the loss of ⌬⌿ m (Fig. 5D). These observations demonstrate that even though LPS-induced caspase 8 activation was capable of cleaving RIP, RIP contributed to TLR4-mediated apoptotic cell death.
RIP Contributes to Caspase 8 Activation-Because earlier studies suggested that RIP may be upstream of caspase 8 in TLR4 signaling (19), we examined the effects of the reduction of RIP on caspase activation. Macrophages were incubated with geldanamycin, which dissociates RIP from HSP90, resulting in the reduction of RIP by proteasomal degradation (36,37). Geldanamycin pretreatment for 12 h resulted in a marked reduction of RIP (Fig. 6A). When RIP was reduced employing geldanamycin, caspase 8-like activity was suppressed (Fig. 6B). When examined by Western blot, the reduction of RIP was also associated with a marked decrease of activated p18 caspase 8 (Fig. 6C). In contrast, partially processed p41/43 caspase 8 was equivalent in the presence or absence of geldanamycin (Fig. 6C). Consistent with the effects on caspase 8 activation, the processing of Bid, which is downstream of caspase 8, was suppressed in the presence of geldanamycin (Fig. 6C). These observations suggest that RIP mediates TLR4-induced caspase 8 activation in macrophages.
RIP Mediates Non-apoptotic Death in Macrophages-Studies were performed to determine the effects of RIP on non-apoptotic cell death. The reduction of RIP by geldanamycin, as with the RIP siRNA, resulted in suppression (p Ͻ0.05) not only of the apoptotic phenotype (Fig. 7A) but also the loss of ⌬⌿ m (Fig. 7B) and cell death (Fig. 7C). To suppress caspase 8, an adenoviral vector expressing CrmA was employed to co-infect macrophages with the Ad IB␣. Inhibition of caspase 8 with CrmA protected against DNA fragmentation (Fig. 7A) but failed to protect against the loss of ⌬⌿ m (Fig. 7B) or the loss of membrane integrity (Fig. 7C) mediated by LPS, similar to the results observed with IETDfmk or the DN FADD. In contrast, when RIP also was suppressed, CrmA prevented apoptotic cell death (Fig. 7, A and B) and the loss of ⌬⌿ m was further diminished (Fig. 7B). To confirm the effects of RIP on nonapoptotic cell death, the expression of RIP was suppressed employing siRNA and then macrophages were infected with a control vector or AdIB␣. When these macrophages were incubated with LPS plus IEDT-fmk to inhibit caspase 8, the knock down of RIP suppressed the loss of cell membrane integrity (Fig. 7D) and the loss of ⌬⌿ m (Fig. 7E). These observations demonstrate that the non-apoptotic cell death that becomes apparent when caspase 8 activity is inhibited was suppressed by the reduction of RIP.

DISCUSSION
Macrophages play a critical role in a variety of diseases including rheumatoid arthritis, atherosclerosis, infection, and inflammatory bowel disease. NF-B activation is critical not only for the induction of inflammation but also for macrophage survival, because NF-B regulates the expression of both pro-inflammatory and anti-apoptotic proteins (23,38,39). Here we have documented that NF-B activation is essential for macrophage survival following TLR4 ligation with LPS. When NF-B activation is prevented, LPS-induced macrophage apo-  DECEMBER 23, 2005 • VOLUME 280 • NUMBER 51 ptotic cell death proceeds through a RIP-and caspase 8-dependent mechanism, and ectopically expressed Bcl-x L is effective at preventing apoptotic cell death (Fig. 8). When caspase 8 activation is also sup-pressed, the death-inducing pathway is converted to one that is RIP dependent, non-apoptotic, which involves the loss of ⌬⌿ m , and the macrophages are no longer protected by Bcl-x L (Fig. 8). The current  data are consistent with previously published observations that inhibition of NF-B activation with a specific inhibitor, a proteasome inhibitor, or following the inhibition of protein synthesis with cycloheximide sensitized macrophages to LPS-induced apoptosis (15,16,40). Our study has extended these observations, characterizing the mechanisms contributing to the LPS-induced death of macrophages and identifying NF-B activation as the determinant of whether or not the cell dies and caspase 8 as a key mediator determining the mode of cell death.

LPS-induced Macrophage Cell Death
Prior studies suggested that LPS-induced macrophage apoptosis was mediated by the production of TNF␣ (41), and we previously demonstrated that TNF␣ induced macrophage apoptosis following the inhibition of NF-B activation (25). This possibility was examined employing a neutralizing antibody against TNFR1. Although effectively suppressing TNF␣-induced apoptotic cell death, this antibody had no effect on LPS-mediated apoptosis, demonstrating that LPS-induced TNF␣ was not responsible for LPS-induced apoptosis when NF-B activation was prevented. The difference between our data and the previous publication (41) may relate to the fact they employed murine bone marrow macrophages, proliferating in response to M-CSF, and NF-B activation was not inhibited (41). Additionally, NF-B activation is important for the LPS-induced expression of TNF␣; therefore, when NF-B activation is suppressed, TNF␣ secretion is reduced (42,43). Also supporting independent mechanisms, the apoptosis observed in response to LPS is distinct from that induced by TNF␣. In contrast to the protection observed in response to LPS, Bcl-x L failed to protect mitochondrial integrity and cell viability in response to TNF␣ when NF-B activation was suppressed (25). 4 These observations demonstrate that LPS-induced apoptosis is independent of TNF␣ and that the mechanism(s) responsible for macrophage death are different.
Following the inhibition of NF-B, caspase 8 was critical for the induction of LPS-induced apoptosis. The DN FADD, IETD-fmk, CrmA, and z-VAD-fmk (data not shown) each inhibited caspase 8 activation and markedly suppressed LPS-induced DNA fragmentation. These observations are consistent with those employing a macrophage cell line and the proteasome inhibitor MG132 to inhibit NF-B activation (12,13,16). When the activation of caspase 8 was not suppressed, Bcl-x L was effective at preventing DNA fragmentation, the loss of mitochondrial integrity, and cell death. The LPS-induced apoptosis may be mediated through the caspase 8-mediated activation of Bid (Fig. 6A) and caspase 3, which was not examined in this study. Our data suggest further that RIP contributes to apoptosis, because the knock down of RIP protected not only against LPS-induced caspase 8 activation, Bid cleavage, and DNA fragmentation but also against the loss of ⌬⌿ m and the induction of cell death. These observations suggest that RIP is upstream of caspase 8. Supporting this interpretation, cell death induced by the overexpression of RIP was suppressed by the inhibition of caspase 8 or the expression of DN FADD, indicating that the action of RIP to promote apoptosis was upstream of FADD and caspase 8 (19,45). Recent observations employing DN FADD and macrophages deficient in Toll/interleukin-1R domain-containing adapter-inducing IFN-␤ (TRIF) or myeloid differentiation factor 88 demonstrated that TRIF and FADD, but not myeloid differentiation factor 88, were upstream of caspase 8 activation in TLR4-mediated apoptosis (12,13,16) (Fig. 8). Although FADD did not interact directly with TRIF (19), it was capable of interacting with RIP through their death domains (44). Together, these observations suggest that following TLR4 ligation, TRIF may interact with RIP, promoting apoptosis by recruiting FADD and the activation of caspase 8 (Fig. 8).
Another potential explanation for the apoptotic effects of RIP may be that the carboxy-terminal fragment of caspase 8-cleaved RIP (RIPc), which contains the death domain, may continue to promote apoptosis. The overexpression of either RIPc or wild type RIP induced apoptotic cell death (35,46,47). In contrast, expression of the amino-terminal fragment of caspase 8-cleaved RIP, which possesses the kinase domain, failed to induce cell death (35,46). Therefore, caspase 8-cleaved RIP does not prevent, but may promote, the pro-apoptotic effects of RIP.
However, our data suggest that caspase 8-mediated RIP cleavage is associated with the loss of its ability to induce non-apoptotic cell death, suggesting that intact death and kinase domains are required. When caspase 8 activation was suppressed, although the apoptotic phenotype was reduced there was no protection against loss of ⌬⌿ m or cell death. The pattern of death observed was consistent with necrosis (20,37,48). When caspase 8 activation was suppressed, RIP was not cleaved, suggesting that full-length RIP may be necessary for this mode of cell death. The importance of RIP in non-apoptotic cell death was demonstrated by the protection provided by the reduction of RIP induced by geldanamycin or siRNA. The observation that Bcl-x L provided no protection to the mitochondria when the activation of caspase 8 was suppressed is consistent with an alternate, non-caspase-mediated mechanism of macrophage cell death. This interpretation is supported by recent studies showing that full-length, but not caspase 8-cleaved, RIP was responsible for the induction of TNF␣ plus z-VAD-fmk-induced necrosis (37). Our observations are also consistent with the report (20) that RIP-mediated non-apoptotic cell death induced through the Fas, TNFR1, and TNFrelated apoptosis-inducing ligand (TRAIL) pathways was dependent upon RIP kinase activity. Prior studies have demonstrated that overexpressing RIP or RIPc results in apoptosis and the kinase domain is dispensable, similar to the role of RIP in TNFR1-induced NF-B activation (49). In contrast, the induction of RIP-mediated non-apoptotic cell death requires an additional activation signal and the kinase domain is required (20,37). When NF-B activation was not suppressed, caspase 8 was not activated, RIP was not cleaved, and no cell death was observed. Therefore, these observations suggest that LPS-induced NF-B activation protects not only against the activation of caspase 8 but also against the induction of the ability of RIP to induce necrosis (Fig. 8).
In conclusion, our data suggest that in macrophages when NF-B activation is suppressed the inhibition of caspase 8 converts cell death from apoptotic to non-apoptotic cell death, which is mediated through RIP. These two forms of cell death may have different physiological 4 H. Liu, L. Pagliari, and R. M. Pope, unpublished observations. FIGURE 8. Schematic representation of the mechanisms contributing to LPS-induced macrophage cell death following the inhibition of NF-B activation. Upon ligation by LPS, TLR4 recruits adaptor molecules myeloid differentiation factor (MyD88) and TRIF, which leads to the activation of NF-B. Following the inhibition of NF-B activation, LPS induces macrophage apoptosis through two pathways, one of which involves caspase 8 activation, which may be mediated through effects on the mitochondria, and a direct effect on apoptosis. Caspase 8 activation leads to the cleavage of RIP, and the inhibition of caspase 8 activity promotes RIP-mediated non-apoptotic cell death.
consequences (50,51) that will be important to consider for therapeutic interventions involving inhibition of NF-B activation when TLR4 signaling will result either from microbial sources or endogenous TLR ligands that exist locally in chronic inflammatory conditions such as rheumatoid arthritis.