cIAPs block Ripoptosome formation, a RIP1/caspase-8 containing intracellular cell death complex differentially regulated by cFLIP isoforms

Abstract

The intracellular regulation of cell death pathways by cIAPs has been enigmatic. Here we show that loss of cIAPs promotes the spontaneous formation of an intracellular platform that activates either apoptosis or necroptosis. This 2 MDa intracellular complex that we designate "Ripoptosome" is necessary but not sufficient for cell death. It contains RIP1, FADD, caspase-8, caspase-10, and caspase inhibitor cFLIP isoforms. cFLIP(L) prevents Ripoptosome formation, whereas, intriguingly, cFLIP(S) promotes Ripoptosome assembly. When cIAPs are absent, caspase activity is the "rheostat" that is controlled by cFLIP isoforms in the Ripoptosome and decides if cell death occurs by RIP3-dependent necroptosis or caspase-dependent apoptosis. RIP1 is the core component of the complex. As exemplified by our studies for TLR3 activation, our data argue that the Ripoptosome critically influences the outcome of membrane-bound receptor triggering. The differential quality of cell death mediated by the Ripoptosome may cause important pathophysiological consequences during inflammatory responses.

Graphical abstract

Figure 1

IAPs Protect Cells from TLR3-Induced Apoptosis and Necroptosis (A, B, F, and H) Cell viability was analyzed by crystal violet assay; average values of three independent experiments (±SEM) are shown. (A) HaCaT, A5RT3, and MET-1 cells were pretreated with IAP antagonist and subsequently stimulated with 50 μg/ml of poly(I:C). Inset shows analysis of cellular lysates by western blotting of the respective proteins. (B–D and H) Cells were stimulated as described in the Experimental Procedures. (B) In the absence of cIAPs, TLR3 ligation induces cell death in a caspase- and RIP1 kinase-dependent manner. (C) cIAPs suppress HMGB1 release upon TLR3 stimulation. HMGB1 release from HaCaT was detected by western blot analysis of cell-free supernatants and total cellular lysates. (D) cIAPs protect cells from TLR3 ligation-induced apoptosis and necroptosis. Cell death morphology of HaCaT cells was investigated by simultaneous staining with Hoechst-33342 and SYTOX Green, immediately followed by transmission and fluorescence microscopy, respectively (Supplemental Experimental Procedures). (E) TNFR1 knockdown was analyzed by western blot analysis in HaCaT cells transfected with control or TNFR1 siRNA. (F) TNFR1 knockdown does not alter TLR3-induced cell death. (G) Comparison of TLR3-induced caspase-8-containing complexes with TNF complex II. For analysis of caspase-8-containing complexes, caspase-8 coimmunoprecipitation (IP) was performed. Cells were prestimulated with IAP antagonist and subsequently stimulated with either poly(I:C) or TNF (10 μg/ml). (H) Primary human keratinocytes (from five different donors [different symbols; the horizontal bar represents the mean of all five donors examined]) are sensitized to TLR3-induced cell death in the absence of IAP function. Either IAP antagonist (panels 3–4) or TWEAK (panels 5–6) sensitizes to TLR3-induced cell death in a TNF-independent manner (panels 7–12).

Figure 2

Increased Cell Death by Loss of cIAPs Is Regulated at the Ripoptosome; RIP1 Is Critical, whereas RIP3 or Caspase-8 Modulates the Quality of Cell Death (A) Caspase-8 siRNA is functional as determined by western blot analysis of caspase-8 protein. (B) The efficiency of RIP3 knockdown in HaCaT was analyzed by qPCR. The mean of two independent experiments (±SEM) is shown. (C and D) Cells were stimulated as described in the Experimental Procedures. Cell viability was analyzed by crystal violet assay; the mean of three independent experiments (±SEM) is shown. (C) The quality of cell death induced by TLR3 in the absence of cIAPs depends on caspase-8 or/and RIP3 expression. (D) Knockdown of RIP1 protects HaCaT from TLR3-induced cell death in the absence of cIAPs. Knockdown efficiency for RIP1 was analyzed by western blot analysis (inset). (E) RIP1 knockdown suppresses Ripoptosome formation. Caspase-8 immunoprecipitation (IP) was performed as outlined in the Experimental Procedures. Prolonged exposure of the blots reveals either caspase-8/RIP1 association in the unstimulated conditions or alternatively represents the background of the biochemical experiment.

Figure 3

cFLIP Suppresses Ripoptosome Formation and Apoptosis in the Absence of cIAPs (A) cFLIP protects from TLR3-induced cell death. A5RT3 cells were transduced with control shRNA (HRS) or cFLIP shRNA and stimulated as described in the Experimental Procedures. Cell viability was analyzed by crystal violet assay; the mean of three independent experiments (±SEM) is shown. (B) cFLIP suppresses Ripoptosome formation in A5RT3 cells. Ripoptosome formation was studied by caspase-8 IP. Note the dramatic accumulation of caspase-10 upon cFLIP knockdown.

Figure 4

cFLIP Isoforms Differentially Regulate Ripoptosome Formation and Modulate the Quality of Cell Death in the Absence of cIAPs (A) The quality of cell death is differentially regulated by cFLIP isoforms. HaCaT cells were transduced with cFLIP_L, cFLIP_S, or vector alone and stimulated as indicated in the Experimental Procedures. Cell death was characterized by Annexin V/PI double staining, followed by FACS analysis. (B) Ripoptosome formation is modulated by cFLIP isoforms in a differential manner. Ripoptosome was precipitated by caspase-8 Abs. (C) Absence of spontaneous Ripoptosome formation in PK upon IAP antagonist treatment when compared to HaCaT-cFLIPs.

Figure 5

cFLIP_S Promotes Spontaneous Ripoptosome Formation, a 2 MDa Complex, in the Absence of cIAPs; Ripoptosome-Induced Necroptosis Proceeds in a RIP1 Kinase- and RIP3-Dependent Manner Independent from TNF Signaling (A, B, and E) cFLIP_S-HaCaT were transfected with control or TNFR1 siRNA. (A, C, and D) Cell viability was analyzed by crystal violet assay; the mean of three independent experiments (±SEM) is shown. (A) Spontaneous cell death induced by cIAPs depletion in cFLIP_S cells is independent from TNFR1. siRNA-transfected cells were prestimulated with IAP antagonist followed by stimulation with TNF (100 ng/ml). (B) Spontaneous cell death of HaCaT-cFLIP_S in the presence of IAP antagonist is independent from autocrine TNF signaling. Cells were stimulated with IAP antagonist and TNFR2-Fc, respectively. Cell death was determined by Annexin V/PI double staining followed by FACS analysis. (C) Spontaneous cell death induced by cIAPs depletion in HaCaT-cFLIP_S is RIP1 kinase dependent. (D) Spontaneous cell death induced by cIAP depletion in HaCaT-cFLIP_S is RIP3 dependent. HaCaT-cFLIP_S were transfected with RIP3 siRNA or control siRNA and then treated with IAP antagonist. (E) Spontaneous Ripoptosome formation in HaCaT-cFLIP_S in the absence of cIAPs is independent from TNFR1 signaling. Ripoptosome formation was analyzed by caspase-8 IP in the presence of zVAD from 2 × 10⁶ transfected cells 4 hr after IAP antagonist stimulation (200 nM). (F) Inhibition of cIAPs promotes Ripoptosome formation, a 2 MDa complex. HaCaT-cFLIP_S were either prestimulated with zVAD (10 μM) alone or in combination with IAP antagonist (200 nM) for 4 hr. Cellular lysates were used for chromatographic gel filtration analysis. Numbers indicate fractions collected. Fractions 3–26 were analyzed by western blotting (left panels) for the respective proteins. All fractions were subsequently pooled as indicated, and caspase-8 IP was performed as described in the Experimental Procedures. Proteins found in different pooled fractions were analyzed by western blotting.

Figure 6

Caspase Activity in the Ripoptosome Is Differentially Regulated by cFLIP Isoforms and Is Negatively Regulated by Proteasome Function (A and C) Ripoptosome formation was studied by caspase-8 IP. (A) cFLIP isoforms differentially regulate caspase activity in the Ripoptosome. HaCaT-cFLIP_S cells were stimulated with IAP antagonist and zVAD for 1 hr with poly(I:C) or diluent alone. (B) In the absence of cIAPs, cFLIP_S but not cFLIP_L blocks caspase-8 activity in the Ripoptosome. HaCaT cells were transduced with retroviral vectors containing Flag-tagged cFLIP_L or cFLIP_S. Cells were prestimulated with IAP antagonist alone or subsequently with poly(I:C), followed by anti-Flag IP. Asterisks indicate nonspecific binding of Abs. (C) Ripoptosome formation is negatively regulated by proteasome function in HaCaT-cFLIP_S. Cells were prestimulated with MG132 for 10 min followed by stimulation with IAP antagonist for 4 hr. Experiments shown in (A)–(C) were performed in the presence of zVAD (10 μM).

Figure 7

Working Model of Activation and Signaling by the Ripoptosome and Its Regulation by cFLIP Isoforms A fraction of RIP1 is constantly “modified” (likely by cellular stress signals) to an active conformation. cIAPs ubiquitinate “modified” RIP1 targeting it to degradation by the 26S proteasome. The unknown cellular compartment contains not only “modified” RIP1 but also caspase-8, FADD, and cFLIP. Addition of IAP antagonist leads to autoubiquitination of cIAPs and degradation. This is the signal for the formation of the Ripoptosome, a complex of “modified” RIP1, FADD, caspase-8/10, and cFLIP isoforms. The Ripoptosome can be recruited to RHIM-binding modules such as TRIF, thereby linking it to TLR3. Depending on the composition of the Ripoptosome, it activates apoptosis and/or necroptosis. Procaspase-8 homodimers within the Ripoptosome form active caspase-8 and subsequent apoptosis. cFLIP_L-caspase-8 heterodimers within the Ripoptosome have enzymatic activity in the complex insufficient to generate active caspase-8 heterotetramers necessary for apoptosis induction, although sufficient for RIP1 cleavage. Caspase-8-cFLIP_S heterodimer lacks proteolytic activity that is required not only for full caspase-8 activation but also for RIP1 degradation. Thus RIP1 accumulates in the Ripoptosome and initiates necroptosis via RIP3 in a RIP1 kinase-dependent manner.