Ubiquitination of RIP1 regulates an NF-kappaB-independent cell-death switch in TNF signaling

Abstract

TNF receptor 1 (TNFR1) can trigger opposing responses within the same cell: a prosurvival response or a cell-death pathway [1, 2]. Cell survival requires NF-kappaB-mediated transcription of prosurvival genes [3-9]; apoptosis occurs if NF-kappaB signaling is blocked [5, 7-9]. Hence, activation of NF-kappaB acts as a cell-death switch during TNF signaling. This study demonstrates that the pathway includes another cell-death switch that is independent of NF-kappaB. We show that lysine 63-linked ubiquitination of RIP1 on lysine 377 inhibits TNF-induced apoptosis first through an NF-kappaB-independent mechanism and, subsequently, through an NF-kappaB-dependent mechanism. In contrast, in the absence of ubiquitination, RIP1 serves as a proapoptotic signaling molecule by engaging CASPASE-8. Therefore, RIP1 is a dual-function molecule that can be either prosurvival or prodeath depending on its ubiquitination state, and this serves as an NF-kappaB-independent cell-death switch early in TNF signaling. These results provide an explanation for the conflicting reports on the role of RIP1 in cell death; this role was previously suggested to be both prosurvival and prodeath [10-12]. Because TRAF2 is the E3 ligase for RIP1 [13], these observations provide an explanation for the NF-kappaB-independent antiapoptotic function previously described for TRAF2 [14-16].

Figure 1. Lysine 377 of RIP1 regulates cell death

(A) RIP1-null cells and RIP1-null cells reconstituted with RIP1-WT or RIP1-K377R were stimulated with the indicated doses of TNF for 5 and 24 hours. Cells were stained with annexin V-PE and analyzed by flow cytometry. The percentage of total cells that were annexin V-reactive is shown. (B) The three cell lines in (A) were stimulated with 10 ng/ml TNF for the indicated times and annexin-V levels were analyzed. (C) RIP1-WT and RIP1-K377R cells were stimulated with 10 ng/ml TNF for different periods of time (left panel) or the indicated doses of TNF for 24 hours (right panel). Cell lysates were blotted with antibodies specific for CASPASE-8, cleaved CASPASE-3 or cleaved PARP. The anti-PARP blots were also overexposed to visualize additional PARP cleavage products. Re-blotting with anti-tubulin and PLCγ was also performed to demonstrate equivalent loading. These western blots are representative of at least three similar experiments.

Figure 2. Lysine 377 of RIP1 provides an early NF-κB-independent and a late NF-κB-dependent anti-apoptotic signal

(A) RIP1-null, RIP1-WT or RIP1-K377R cells were transduced with retroviruses encoding either GST as a negative control (left panels) or IκBαSR (right panels). These cells were stimulated with the indicated doses of TNF for 5 hours (panels I & II) and 24 hours (panels III & IV). Apoptosis was analyzed by annexin V staining and flow cytometry as in Figure 1C. (B) The cell lines in (A) were stimulated with 10 ng/ml TNF and apoptosis analyzed at the indicated times. (C) RIP1-WT and RIP1-K377R cells stably transfected with the myc-tagged IκBαSR were stimulated with the indicated doses of TNF for 24 hours (left panel) or 10 ng/ml TNF for different periods of time (right panel). Cell lysates were blotted as in Figure 1C. The anti-myc blot serves as both a loading control and to show equivalent expression of the IκBαSR in the two cell lines. These western blots are representative of at least three similar experiments.

Figure 3. UBC13 and TRAF2 have a NF-κB-independent anti-apoptotic function

(A) IKKγ/NEMO-deficient Jurkat T cells stably expressing the control protein GST, UBC13DN or TRAF2DN were treated with indicated concentrations of TNF for 5 and 24 hours. Apoptosis was analyzed by annexin V staining. (B) The cell lines in (A) were treated with 2 ng/ml of TNF for the indicated times. 150 ug of cell extracts were analyzed by western blot analysis as in Figure 1C. Anti-ZAP70 blotting was performed to confirm equivalent loading. (C) RIP1-null or RIP1-WT cells were transduced with control GST or TRAF2DN-encoding retroviruses. Stably selected cells were treated with indicated concentrations of TNF for 5 or 24 hours and stained with annexin V-PE. (D) The data from (C) where the cells are stimulated with 5 ng/ml TNF for 24 hours are plotted in a bar graph to highlight the dual functional role of RIP1.

Figure 4. Ubiquitination of RIP1 regulates association with CASPASE-8

(A) RIP1-WT and RIP1-K377R cells were stimulated with 10 ng/ml TNF for the indicated times. FLAG-tagged RIP1 and associated proteins were immunoprecipitated with anti-FLAG beads and eluted with the FLAG peptide. The immune complexes were blotted sequentially with anti-CASPASE-8 and anti-RIP1 as a loading control. The anti-CASPASE-8 blot was also overexposed to visualize the p43/p41 cleaved form. (B) A model for the role of RIP1 in TNF-mediated apoptosis. (I) In the early phase of TNFR1 signaling, TRAF2-mediated ubiquitination of lysine 377 functions as a brake to prevent RIP1 from associating with CASPASE-8. This does not require NF-κB-mediated gene transcription. (II) In the late phase of TNFR1 signaling, a second pro-survival signal provided by the ubiquitination of lysine 377 becomes effective via the up-regulation of NF-κB-dependent anti-apoptotic genes such as c-FLIP. (III) In the absence of K63-linked poly-ubiquitination, RIP1 becomes a death-signaling molecule by forming a complex with CASPASE-8 to initiate apoptosis. Red indicates anti-apoptotic molecules and green indicates pro-apoptotic molecules.