The Kinase Activity of Rip1 Is Not Required for Tumor Necrosis Factor- (cid:1) -induced I (cid:2) B Kinase or p38 MAP Kinase Activation or for the Ubiquitination of Rip1 by Traf2*

The death domain kinase Rip1 is recruited to the tumor necrosis factor receptor type 1 and mediates the I (cid:2) B kinase and p38 MAP kinase pathways. In response to tumor necrosis factor- (cid:1) (TNF- (cid:1) ), we find Rip1 phosphorylated and ubiquitinated, suggesting that Rip1 phosphorylation may stimulate its ubiquitination. To address the contribution of the kinase activity of Rip1 to its ubiquitination and to TNF- (cid:1) signaling, we introduced wild type Rip1 and a kinase-inactive form of Rip1, Rip1D138N, into rip1 (cid:3) / (cid:3) murine embryonic fibroblast cells by retroviral infection. TNF- (cid:1) -induced ubiquitination of Rip1 is observed in Rip1D138N cells, supporting the argument that Rip1 autophosphorylation is not required for Rip1 ubiquitination. TNF- (cid:1) -induced Ikk and p38 MAP kinase activation is normal, and the Rip1D138N cells are resistant to TNF- (cid:1) -induced cell death, indicating that the kinase activity of Rip1 is not required to mediate its antiapoptotic functions. In the absence of Traf2, TNF- (cid:1) -induced ubiquitination of Rip1 is impaired, suggesting that Traf2 may be the E3 250 m M NaCl, 0.1% Nonidet P-40, 5 m M EDTA), immunoprecipitated with (cid:1) -RIP Ab (BD Biosciences), and immuno- blotted with (cid:1) -HA Ab. Expression of transfected constructs was con-firmed by immunoblotting with an (cid:1) -HA Ab. Rip1 proteins were de- tected by immunoblotting with an (cid:1) -RIP1 antibody. To determine whether Rip1 ubiquitination occurs at the Tnfr1, wild type cells or cells expressing kinase-inactive Rip1 were left untreated or treated with mTNF- (cid:1) for the time periods indicated. Cells were lysed in 0.5 ml of endogenous lysis buffer (0.5 M Tris (pH 7.6), 0.5 M NaCl, 0.1 M EDTA, 1% Triton X-100, 0.5 M NaF, and 0.5 M sodium pyrophosphate) supplemented with a protease mixture inhibitor (Roche Applied Science), immunoprecipitated with (cid:1) -TNFR1 antibody, and then immunoblotted with (cid:1) -RIP1, (cid:1) -TAK1, or (cid:1) -TRAF2 antibody (Santa Cruz Biotechnol- ogy). In other experiments, cells were lysed in radioimmune precipita-tion assay buffer and immunoprecipitated with (cid:1) -RIP Ab, and ubiquiti- nated Rip1 proteins were detected by immunoblotting with an anti-ubiquitin antibody (Boston Biochem). To determine p38- (cid:1) -MAP kinase activity, 40 (cid:3) g of total protein was immunoblotted with an (cid:1) -phospho-p38 antibody (Cell Signaling Technology). To determine the relative expression levels of p38- (cid:1) MAP kinase or Ikk, wild type rip1 (cid:2) / (cid:2)

The proinflammatory cytokine tumor necrosis factor-␣ (TNF-␣) 1 is a major mediator of apoptosis as well as inflammation and immunity. It has also been implicated in the pathogenesis of several human diseases including inflammatory bowel disease, arthritis, sepsis, diabetes, and cancer (1). TNF-␣ activates the transcription factors NFB and activator protein-1 and exerts its effects by binding to two receptors, the tumor necrosis factor receptor type 1 (Tnfr1 or p55) and the Tnfr type 2 (Tnfr2 or p75) (2). Signaling from the activated Tnfr1 is mediated by the adapter proteins Traf2 and the death domain serine/threonine kinase Rip1 (3)(4)(5); yet, how Rip1 and/or Traf2 mediates activation of the downstream kinases remains unclear.
Ubiquitination has been implicated in the regulation of the NFB pathway as well as in many other biological processes. Ubiquitin is a 76-amino acid protein that is highly conserved and, like phosphorylation, is essential for the degradation of proteins that respond to changes or stresses in the micoenvironment (6). Phosphorylation can stimulate or inhibit ubiquitination by affecting either the target protein or the ubiquitin enzymes. Typically, ubiquitin forms a linkage with proteins via one or more lysine residues. In vivo, Lys-48 polyubiquitin chains lead to the recognition and degradation of proteins by a proteasome. In contrast, Lys-63 linkages are not associated with the proteasomal degradation but have been implicated in other biological processes, including activation of the IB kinase by the E3 ubiquitin ligase Traf6 (7,8).
On TNF-␣ treatment, we found that Rip1 undergoes autophosphorylation and ubiquitination, suggesting that one or both of these posttranslational modifications may be important in Ikk or Mkk-6 activation. Studies in transfected cells suggest that the kinase activity of Rip1 is not required for TNF-␣induced Ikk or p38 MAP kinase activation (9,10); however, transfection of kinase-inactive Rip1 results in its oligomerization with Traf proteins and the subsequent activation of these pathways (5,10,11).
To determine whether Rip1 autophosphorylation triggers its ubiquitination and to test whether Rip1 ubiquitination contributes to MAP kinase activation, we introduced wild type rip1 and a kinase-inactive rip1 into rip1-deficient murine embryonic fibroblasts (MEFs) and examined TNF-␣-induced ubiquitination of Rip1 as well as activation of Ikk and p38 MAP kinase. In response to TNF-␣, both Rip1 and the kinase-inactive form of Rip1 are ubiquitinated, although phosphorylated Rip1 may be a preferred E3 substrate. In addition, TNF-␣-induced activation of Ikk and p38 MAP is not affected in cells expressing an inactive Rip1 kinase, demonstrating that Rip1 autophosphorylation is not required for signaling. Although it is recruited to the Tnfr1, Rip1 ubiquitination is not observed when Traf-deficient cells are stimulated with TNF-␣. Taken together, these studies suggest that TNF-␣-induced Ikk and Mkk-6 activation is mediated by the ubiquitination of Rip1 and the subsequent recruitment of a ubiquitinated Tak1 complex.
Plasmids and Constructs-The full-length RIP-Myc, ⌬KIN-Myc, and ⌬DD-Myc constructs and RIP-K D have been described previously (9,11). The HA-ubiquitin construct was provided by Dr. Zhijan James Chen (University of Texas, Southwestern, Dallas).
In Vitro Kinase Assay-Wild type or rip1Ϫ/Ϫ cells infected with rip1 or rip1D138N retrovirus were plated at 1 ϫ 10 6 cells/10-cm plate. The cells were left untreated or treated with 10 ng/ml TNF-␣ for the time periods indicated. Cells were washed once with cold phosphatebuffered saline, and an in vitro kinase assay was performed as described previously (14).
Interleukin-6 Production Assays-Wild type, rip1Ϫ/Ϫ, or rip1Ϫ/Ϫ cells infected with rip1 or rip1D138N retrovirus were plated at 3 ϫ 10 4 cells/well on 24-well plates. The cells were left untreated or treated with 10 ng/ml TNF-␣ for 20 h. The supernatants were collected, and the levels of IL-6 produced from each cell line were assayed by an OptEIA mouse IL-6 enzyme-linked immunosorbent assay kit (BD Biosciences).
Co-immunoprecipitation and Western Blot Analysis-For all transfection studies, 293T cells were plated at 3 ϫ 10 5 cells/plate on 6-well plates. The transfected cells were lysed in 0.2 ml E1A buffer (50 mM HEPES (pH 7.6), 250 mM NaCl, 0.1% Nonidet P-40, 5 mM EDTA), immunoprecipitated with ␣-RIP Ab (BD Biosciences), and immunoblotted with ␣-HA Ab. Expression of transfected constructs was confirmed by immunoblotting with an ␣-HA Ab. Rip1 proteins were detected by immunoblotting with an ␣-RIP1 antibody. To determine whether Rip1 ubiquitination occurs at the Tnfr1, wild type cells or cells expressing kinase-inactive Rip1 were left untreated or treated with mTNF-␣ for the time periods indicated. Cells were lysed in 0.5 ml of endogenous lysis buffer (0.5 M Tris (pH 7.6), 0.5 M NaCl, 0.1 M EDTA, 1% Triton X-100, 0.5 M NaF, and 0.5 M sodium pyrophosphate) supple-mented with a protease mixture inhibitor (Roche Applied Science), immunoprecipitated with ␣-TNFR1 antibody, and then immunoblotted with ␣-RIP1, ␣-TAK1, or ␣-TRAF2 antibody (Santa Cruz Biotechnology). In other experiments, cells were lysed in radioimmune precipitation assay buffer and immunoprecipitated with ␣-RIP Ab, and ubiquitinated Rip1 proteins were detected by immunoblotting with an antiubiquitin antibody (Boston Biochem). To determine p38-␣-MAP kinase activity, 40 g of total protein was immunoblotted with an ␣-phospho-p38 antibody (Cell Signaling Technology). To determine the relative expression levels of p38-␣ MAP kinase or Ikk, wild type rip1Ϫ/Ϫ or rip1Ϫ/Ϫ infected with rip1 or rip1D138N retroviruses were lysed, and 30 g of total protein was immunoblotted with ␣-p38␣ Ab (Santa Cruz Biotechnology) or with ␣-IKK-␣ antibody (Santa Cruz Biotechnology).
Cell Viability Assays-Briefly, 5 ϫ 10 5 cells were plated overnight on 60-mm tissue culture dishes. Cells were then treated with 10 ng/ml mTNF and 250 ng/ml cycloheximide for 20 h. Cells were then collected and assayed either for cell viability by trypan blue staining or for cell death by annexin V/phosphatidylinositol staining on triplicate cultures.

Death Domain Kinase Rip1 Is Phosphorylated and Ubiquitinated on Tnfr1
Activation-Although biochemical and genetic studies have implicated the Rip1 kinase in TNF-␣ signaling (4,9,10,12,14), the mechanisms by which Rip1 mediates the NFB or p38 MAP kinase responses to TNF remain unclear. Both phosphorylation and, more recently, ubiquitination have been implicated in the activation of the IKK complex. To elucidate the role of Rip1 in the initiation of these pathways, wild type and rip1-deficient MEFs were left unstimulated or treated with TNF-␣ for 2, 5, and 10 min, and Rip1 protein was detected by immunoblotting. In the wild type MEF cells, we observed FIG. 1. Rip1 is phosphorylated and ubiquitinated in response to TNF-␣. A, wild type (wt), rip1Ϫ/Ϫ, or rip1Ϫ/Ϫ(D138N) cells were left untreated or stimulated with mTNF-␣ (10 ng/ml) for the indicated times. Cells were lysed, and 40 g of total protein was immunoblotted with ␣-RIP Ab (BD Biosciences). B, Rip1 is phosphorylated in response to TNF-␣. Wild type cells were left untreated or stimulated with mTNF-␣ (10 ng/ml) (left panel) for 10 min. Cells were then lysed, and 40 g of total protein was treated with 10 units of -phosphatase (-PPase) (New England Biolabs) for 40 min at 30°C. The lysates were then immunoblotted with ␣-RIP Ab. C, Rip1 is also ubiquitinated in response to TNF-␣, and Rip1 ubiquitination is disrupted by pretreatment of cells with methyl-␤-cyclodextrin (MCD), an agent that disrupts lipid rafts. Wild type MEFs were left untreated or treated with TNF-␣ for the time periods indicated. The cells were lysed, immunoprecipitated with ␣-RIP Ab, and immunoblotted with ␣-ubiquitin Ab. Cultures were also pretreated with methyl-␤-cyclodextrin and either left untreated or stimulated with TNF-␣ and analyzed for the presence of ubiquitinated Rip1 proteins. poly-Ub, poly-ubiquitin. the presence of two forms of Rip1. On TNF-␣ stimulation, we observed induction of a slower, migrating form of Rip1. The presence of this form was abolished when the cell lysates were treated with -phosphatase, consistent with the idea that Tnfr1 activation stimulates Rip1 phosphorylation (Fig. 1, A and B). The kinetics of Rip1 phosphorylation with induction at 5 min after TNF-␣ stimulation suggest this is an early event in the TNF-␣ response and may function to initiate signaling.
In addition to phosphorylation and similar to published studies (15)(16)(17), the Rip1 kinase is polyubiquitinated when cells are stimulated with TNF-␣ (Fig. 1C). Moreover, Rip1 ubiquitination in response to TNF-␣ is inhibited by agents such as methyl-␤-cyclodextrin that disrupt lipid rafts (Fig. 1C), suggesting that Rip1 ubiquitination occurs once the activated Tnfr1 is recruited into lipid rafts.

Rip1 Is Not Essential for Ubiquitination or TNF Signaling
demonstrated that TNF-␣-induced Ikk and p38 MAP kinase activation are both impaired in rip1-deficient MEFs (12,14). To further examine the role of the kinase activity of Rip1 in TNF-␣-induced Ikk and p38 MAP kinase activation, we infected rip1Ϫ/Ϫ MEFs with MSCV2.2-IRES-GFP retroviruses that express wild type rip1, a kinase-inactive rip1D138N, or with the MSCV2.2-IRES-GFP retroviral vector alone. Cell lysates were prepared from wild type MEFs and from the infected rip1Ϫ/Ϫ MEFs, and the Rip1 protein expression levels were compared. Importantly, rip1Ϫ/Ϫ MEFs reconstituted with wild type rip1 or kinase-inactive rip1 expressed levels of Rip1 protein similar to those of wild type MEFs (Fig. 2), demonstrating that Rip1 overexpression and oligomerization are not induced in the infected cells.
Wild type, rip1Ϫ/Ϫ, and rip1Ϫ/Ϫ MEFs infected with vector alone, wild type rip1, or with the rip1D138N retrovirus were left untreated or were treated with mTNF-␣ for 10, 30, and 60 min. To measure TNF-␣-induced Ikk activation, cell lysates were immunoprecipitated with anti-IKK␣ antibody, and an in vitro kinase assay was performed using glutathione S-transferase-IB as the substrate. TNF-␣-induced Ikk activation was not observed in cells lacking Rip1 (Fig. 2 and see Ref. 12), and TNF-induced NFB activity was restored in rip1Ϫ/Ϫ cells reconstituted with wild type Rip1 or with a kinase-inactive version of Rip1 ( Fig. 2A).
To measure TNF-␣-induced p38 MAP kinase activity, cell lysates were probed with a phospho-specific p38 MAP kinase antibody. TNF-␣-induced p38 MAP kinase activation was not observed in rip1Ϫ/Ϫ MEFs (as described in Ref. 14); however, TNF-induced p38 MAP kinase activity was observed in TNF-␣-treated wild type MEFs and in rip1Ϫ/Ϫ MEFs infected with the rip1 retrovirus (Fig. 2B, rip1Ϫ/Ϫ(rip1)) as well as the MEFs expressing the kinase-inactive version of Rip1. The NFB and p38 MAP kinase pathways have been implicated in stress-induced cytokine production (18,19), and TNF-␣-induced IL-6 production is impaired in Rip1-deficient cells (14). To test whether the kinase activity of Rip1 contributes to TNF-␣-induced cytokine production, we examined IL-6 production in response to TNF-␣ in the rip1Ϫ/Ϫ(D138N) cells. Treatment of wild type cells with TNF-␣ resulted in increased production of IL-6, whereas little or no TNF-␣-induced IL-6 production was observed in rip1Ϫ/Ϫ cells (Fig. 2C and see Ref. 14). TNF-␣-induced IL-6 production was restored in the rip1Ϫ/Ϫ MEFs infected with wild type rip1 or rip1D138N retroviruses (Fig. 2C). Although the rip1Ϫ/Ϫ(D138N) cells exhibit a higher background of IL-6 production, a 4 -5-fold increase in IL-6 production was observed when wild type MEFs or the rip1D138N-infected MEFs were treated with TNF-␣. Consistent with the p38 MAP kinase and NFB responses to TNF observed in rip1D138N cells, these experiments suggest that the Rip1 kinase activity does not contribute to TNF-␣-induced p38 MAP kinase or Ikk activation.

Rip1 Is Not Essential for Ubiquitination or TNF Signaling
Kinase Activity of Rip1 Does Not Contribute to the Antiapoptotic Response to TNF-␣-Previous transfection studies implicated the kinase activity of Rip1 in apoptosis and FasL-induced necrosis (11,20). To examine whether the Rip1 kinase regulates programmed cell death in mammalian cells, we treated wild type and rip1Ϫ/Ϫ MEFs with TNF-␣ and cycloheximide and quantitated the apoptotic cells. As expected, a rip1 deficiency sensitizes cells to TNF-␣-induced cell death, consistent with the impaired NFB response (Fig. 3B and see Ref. 12). The reintroduction of rip1 into these cells is nearly completely protective, which provides survival to 70% of the cells and correlates with the fact that 72.8% of the cells are GFP-positive and express Rip1 (Fig. 3A). Cells expressing a kinase-inactive Rip1 remain resistant to TNF-␣-induced cell death, presumably because of activation of NFB, indicating that kinase activity of Rip1 does not contribute to the antiapoptotic response to TNF-␣ ( Fig. 2A).
Rip1 Phosphorylation Is Not Required for Its Ubiquitination-In the TNF-␣-induced ubiquitination of the inhibitor IB␣, phosphorylation stimulates recognition by the ␤-transducin repeat-containing protein/Skp1-Cul-F-box complex (21). To test whether Rip1 phosphorylation targets the protein for ubiquitination, cells were left untreated or stimulated with TNF-␣ for 30 or 60 min and immunoprecipitated with an ␣-RIP antibody, and then ubiquitinated Rip1 proteins were detected by immunoblotting with an anti-ubiquitin antibody (Fig. 4). Ubiquitinated Rip1 was induced by TNF-␣ stimulation, and polyubiquitinated Rip1 was also detected in cells expressing a kinase-inactive Rip1D138N only (Fig. 4). TNF-␣-induced ubiquitination of Rip1 and Rip1D138N was inhibited when the cells were pretreated with methyl-␤-cyclodextrin, indicating that Rip1 ubiquitination may occur within lipid rafts (Fig. 4). Thus, consistent with the signaling data (Fig. 2), kinase-inactive Rip1 is polyubiquitinated in response to TNF-␣, revealing that Rip1 phosphorylation is not required for TNF-␣-induced ubiquitination.
The previous experiments suggested that the ubiquitination of kinase-inactive Rip1D138N may be delayed in response to TNF-␣, indicating that kinase-inactive Rip1D138N may be impaired in its ability to be recruited to the Tnfr1. To test this possibility, wild type and kinase-inactive Rip1D138N cells were treated with TNF-␣, immunoprecipitated with an anti-TNFR1 Ab and then immunoblotted with an anti-RIP1 antibody. We observed a transient polyubiquitination of Rip at 10 min, co-incident with the activation of TNF-␣-induced signaling (Fig. 5); however, Rip1 ubiquitination is transient, because polyubiquitinated Rip1 was not detected at the Tnfr1 at longer time points. Consistent with the signaling data (Fig. 2), Rip1 ubiquitination was also observed in TNF-␣-treated cells expressing a kinase-inactive Rip1D138N (Fig. 5), indicating that the kinaseinactive Rip1 is recruited to the Tnfr1 and undergoes ubiquitination in response to TNF-␣. Like IB proteins, Rip1 is phosphorylated in response to TNF-␣, but Rip1 phosphorylation is not required for its ubiquitination at the Tnfr1.
To obtain genetic evidence that Traf proteins ubiquitinate Rip1 in response to TNF-␣, we examined Rip1 ubiquitination in traf2Ϫ/Ϫ and traf2Ϫ/Ϫ/traf5Ϫ/Ϫ MEFs. Three independent traf2-deficient cell lines were left untreated or stimulated with mTNF for 10 or 20 min, and the Tnfr1-associated signaling proteins were immunoprecipitated with an ␣-TNFR1 antibody. Immunoblotting with ␣-RIP antibody detected polyubiquitinated Rip1 in wild type MEFs treated with TNF-␣ for 10 min. By contrast, polyubiquitinated Rip1 was not detected in traf2Ϫ/Ϫ or in traf2Ϫ/Ϫ/traf5Ϫ/Ϫ cells treated with TNF-␣ (Fig. 6A). To demonstrate more directly that polyubiquitinated Rip1 is not detected in Traf2-deficient cells, we stimulated wild type and traf2Ϫ/Ϫ or traf2Ϫ/Ϫ/traf5Ϫ/Ϫ cells with TNF-␣ and immunoprecipitated the samples with the ␣-RIP antibody followed by immunoblotting with an anti-ubiquitin antibody. Polyubiquitinated Rip1 protein was readily detected in TNF-␣stimulated wild type cells but not in TNF-␣-treated traf2Ϫ/Ϫ or traf2Ϫ/Ϫ/traf5Ϫ/Ϫ cells (Fig. 6, B and C). Taken together, these data suggest that Traf2 may be the E3 ubiquitin ligase that initiates TNF-␣ signaling, potentially by ubiquitinating Rip1.
In the Absence of Rip1 a Ubiquitinated Tak1 Complex Is Not Recruited to the Tnfr1-These studies failed to reveal a role for the Rip1 kinase in TNF-␣ signaling, but they suggest that ubiquitinated Rip1 may recruit other ubiquitinated proteins (kinases) capable of activating the Ikk complex or Mkk-6. To test whether ubiquitinated Rip1 recruits the Tak1 kinase complex, known to be activated by ubiquitination, to the Tnfr1 (8), we treated wild type cells and rip1Ϫ/Ϫ cells with TNF-␣ for 5, 10, and 20 min and immunoprecipitated the Tnfr1-associated proteins. We observed the recruitment of a ubiquitinated Tak1 complex in wild type cells stimulated with TNF-␣ for 5 min (Fig. 7A). In contrast, in the absence of Rip1, TNF-␣ stimulation failed to induce recruitment of the Tak1 complex (Fig. 7A), suggesting that Rip1 and potentially ubiquitinated Rip1 are required for Tak1 recruitment to the Tnfr1. Traf2 recruitment, however, is unaffected by an absence of Rip1 and is readily recruited to the Tnfr1 when rip1Ϫ/Ϫ MEFs are stimulated with TNF-␣ (Fig. 7B). DISCUSSION We demonstrate that the death domain kinase Rip1 undergoes autophosphorylation and ubiquitination as early as 2-5 min after TNF-␣ stimulation. Rip1 autophosphorylation occurs within the cytoplasm and appears to precede Rip1 ubiquitination at the Tnfr1. These inducible, posttranslational modifications are reminiscent of those observed for IB␣ and suggest that Rip1 autophosphorylation may trigger its ubiquitination; however, cells that express a kinase-inactive Rip1 undergo TNF-␣-induced ubiquitination and respond normally to TNF-␣. These data indicate that Rip1 autophosphorylation is not required for its recruitment to the activated Tnfr1 or for its ubiquitination in lipid rafts. Rip1 is required for the recruitment of a ubiquitinated Tak1 complex to the activated Tnfr1. Taken together, these studies suggest that Rip1 ubiquitination FIG. 7. The ability to recruit an ubiquitinated Tak1 complex in response to TNF-␣ is impaired in rip1؊/؊ cells. A, Rip1 mediates the recruitment of a ubiquitinated Tak1 (Ub-TAK1) complex to the activated Tnfr1. Wild type (wt) and rip1Ϫ/Ϫ MEFs were left untreated or stimulated with TNF-␣ for the time periods indicated. Cells were then lysed and immunoprecipitated with an ␣-TNFR1 antibody, and associated proteins were detected by immunoblotting with an anti-Tak1 antibody. B, Rip1 is not required for the recruitment of a ubiquitinated Traf2 to the activated Tnfr1. Wild type and rip1Ϫ/Ϫ MEFs were left untreated or stimulated with TNF-␣ for the time periods indicated. Cells were then lysed and immunoprecipitated with an ␣-TNFR1 antibody, and associated proteins were detected by immunoblotting with an ␣-TRAF2 antibody.
rather than its phosphorylation leads to Ikk and Mkk-6 activation.
We provide genetic evidence that the E3 ubiquitin ligase Traf2 may be responsible for the ubiquitination of Rip1. Rip1 ubiquitination in response to TNF-␣ is not observed in cells lacking Traf2 or Traf 5 proteins. Although both proteins contain RING fingers, E3 ubiquitin ligase activity has been demonstrated only for Traf2 and Traf6 (8,22). These findings suggest that Traf2 may regulate TNF-␣ signaling by ubiquitinating Rip1 and potentially other members of the Tnfr1 pathway. TNF-␣-induced ubiquitination of Rip1 is transient, suggesting that Rip1 ubiquitination may be regulated by de-ubiquitinating enzymes such as Cyld (23,24). Cyld overexpression can block NFB activation induced by Rip1 or Traf2 overexpression (24), suggesting that polyubiquitinated Rip1 may be a Cyld substrate in vivo.
Although Rip1 autophosphorylation is induced on TNF stimulation and the Rip1 serine/threonine kinase is recruited to the Tnfr1 on ligand binding, the kinase activity of Rip1 does not appear to participate in TNF-␣ signaling, nor is it required for down-regulation or termination of the TNF-␣ responses. Cells expressing a kinase-inactive Rip1 respond normally to TNF-␣ and terminate the NFB and p38 MAP kinase responses with kinetics similar to TNF-␣-stimulated wild type cells. These data indicate that the kinase activity of Rip1 does not regulate the TNF-␣ pathway but may mediate signals from other receptor pathways. Recently, Rip1 has been implicated in the Trifdependent, Toll-like receptor-3 antiviral response (25). Thus, the kinase activity of Rip1 may mediate Toll-like receptor-3 signals.