Caspase-cleaved arrestin-2 and BID cooperatively facilitate cytochrome C release and cell death

Abstract

Apoptosis is programmed cell death triggered by activation of death receptors or cellular stress. Activation of caspases is the hallmark of apoptosis. Arrestins are best known for their role in homologous desensitization of G protein-coupled receptors (GPCRs). Arrestins quench G protein activation by binding to activated phosphorylated GPCRs. Recently, arrestins have been shown to regulate multiple signalling pathways in G protein-independent manner via scaffolding signalling proteins. Here we demonstrate that arrestin-2 isoform is cleaved by caspases during apoptosis induced via death receptor activation or by DNA damage at evolutionarily conserved sites in the C-terminus. Caspase-generated arrestin-2-(1-380) fragment translocates to mitochondria increasing cytochrome C release, which is the key checkpoint in cell death. Cells lacking arrestin-2 are significantly more resistant to apoptosis. The expression of wild-type arrestin-2 or its cleavage product arrestin-2-(1-380), but not of its caspase-resistant mutant, restores cell sensitivity to apoptotic stimuli. Arrestin-2-(1-380) action depends on tBID: at physiological concentrations, arrestin-2-(1-380) directly binds tBID and doubles tBID-induced cytochrome C release from isolated mitochondria. Arrestin-2-(1-380) does not facilitate apoptosis in BID knockout cells, whereas its ability to increase caspase-3 activity and facilitate cytochrome C release is rescued when BID expression is restored. Thus, arrestin-2-(1-380) cooperates with another product of caspase activity, tBID, and their concerted action significantly contributes to cell death.

Figure 1

Arrestin-2 is uniformly cleaved by caspases during apoptosis induced by TNFα or etoposide. (a) A3KO MEFs exposed to 10 ng/ml of TNFα with 10 μg of CHX (left) or 100 μM etoposide (right) for the indicated times were lysed and analyzed by western blotting. Black arrow, FL arrestin-2; open arrowheads, 1-408 and 1-380 fragments. Small black arrowheads on the left show the positions of M_r standards (kDa) or characteristic cleavage fragments. Arrestin-2 cleavage parallels the activation of caspase-3, -6, and -8, and the cleavage of caspase substrates PARP and BID. Here and in other figures, GAPDH serves as loading control. (b) A3KO MEFs were pre-treated for 2 h with indicated concentrations of pan-caspase inhibitor z-VAD-fmk and then exposed to TNFα/CHX for 12 h (left) or etoposide for 36 h (right). Cell lysates were analyzed by western blotting with F4C1 monoclonal antibody: short exposure (upper blot) shows WT arrestin-2 levels (black arrow) and longer exposure (lower blot) reveals cleavage products (open arrowhead, 1-380). The activation of caspase-3, -6, and -8 and PARP cleavage was determined in parallel. (c) Purified arrestin-2 (75 ng) was incubated for 3 h at 37 °C in 50 μl of caspase cleavage buffer (low salt) in the presence of 1 unit of indicated active human recombinant caspases (Millipore). Initiator caspases-2, -8, -9 and -10 were also assayed in the presence of 1.0 M ammonium citrate (high salt). The products were analyzed by western blotting with F4C1 antibody. Indicated forms of purified arrestin-2 were run as standards. (d) Caspase-8 fl/fl and caspase-8 knockout (KO) MEFs were treated with TNFα/CHX (left) or etoposide (right) for the indicated times. TNFα induces robust activation of caspases-8, -6, and -3 in caspase-8 fl/fl MEFs and appearance of arrestin-2-(1-380) (open arrowhead) after 6–9 h. Caspase-8 knockout MEFs do not show caspase activation or generation of 1-380. On etoposide treatment, both caspase-8 fl/fl and knockout MEFs show similar activation of caspase-3 and -6 and generation of 1-380. See also Supplementary Figure S1

Figure 2

Caspase-generated 1-380 translocates to mitochondria and facilitates cytochrome C release in TNFα-treated cells. (a) Control (−) and TNFα/CHX-treated (+) A3KO MEFs were fractionated, as described in the Materials and Methods. 1-380 generated from endogenous arrestin-2 localizes to mitochondria. Black arrow, FL arrestin-2; open arrowhead, 1-380. Western blots for mitochodrial (subunit IV of cytochrome C oxidase, COX-IV) and cytoplasmic (tubulin and caspase-3) markers are shown. (b) DKO MEFs expressing 1-380 and pDS-Red2-Mito were fixed, and 1-380 was detected with arrestin polyclonal primary (F431, 1 : 500) and green secondary antibody. (c) Intracellular distribution of expressed WT arrestin-2 (wt), DblE, and 1-380 in A3KO MEFs treated with CHX or TNFα/CHX for 6 h. Cell lysates and fractions were resolved by SDS-PAGE and immunoblotted for arrestin-2, caspase-3, or COX-IV. Black arrow, FL arrestin-2; open arrowhead, 1-380. (d) Cytosolic cytochrome C in the same cells was detected by western blotting. (e) Quantification of the data in panel (d), shown as normalized optical density (means±S.E.M.). The results of four experiments were analyzed by one-way ANOVA with Protein as the main factor. The effect of Protein was highly significant (F(3,8)=22.2, P=0.0003). **P<0.01 to empty vector (vector); ^cP<0.001; and ^bP<0.01 to 1-380 according to Bonferroni/Dunn post hoc test. (f) Subcellular fractionation of CHX or TNFα/CHX-treated A3KO MEFs expressing increasing amounts of 1-380 (1-380 DNA per 60 mm dish is shown). (g) Cytosolic cytochrome C in these cells was measured by western blotting. (h) Quantification of the data in (g), shown as normalized optical density (means±S.E.M.). The results of four experiments were analyzed by one-way ANOVA with Concentration as the main factor. The effect of Concentration was highly significant (F(3,12)=44, P<0.0001). ***P<0.001, **P<0.01, *P<0.05 to 0 μg of 1-380 DNA; and ^cP<0.001 to 6 μg of 1-380 DNA, according to Bonferroni/Dunn post hoc test

Figure 3

Arrestin-2-(1-380) facilitates cytochrome C release from mitochondria in etoposide-treated Rat1 cells. (a) Control (−) and etoposide-treated (40 μM) (+) Rat-1 cells were fractionated, as described in the Materials and Methods. 1-380 (open arrowhead) generated from endogenous arrestin-2 localizes to mitochondria, whereas WT arrestin-2 (black arrow) remains in the cytosol. Western blots for mitochondrial (COX IV) and cytoplasmic (tubulin and caspase-3) markers are shown. (b) Intracellular distribution of expressed FL arrestin-2 (wt), DblE, and 1-380 in control (−) and etoposide-treated (+) Rat-1 cells. Black arrow, FL arrestin-2; open arrowhead, 1-380. (c) Indicated forms of arrestin-2 in the mitochondrial fraction were quantified by western blotting and expressed as the percentage of the total. (1-380) generated—the fragment generated from expressed WT arrestin-2 during apoptosis and present only in etoposide-treated cells. Its accumulation in the mitochondria is quantitatively similar to that of the expressed 1-380 fragment. The data were analyzed by ANOVA with Protein and Drug (+/− etoposide) as main factors. Both factors were highly significant: Protein, F(2.12)=120.9, P<0.0001; Drug, F(1,12)=21.1, P=0.0006. As the Protein × Drug interaction was significant (F(2,12)=3.9, P=0.049), the amount of arrestins in mitochondria with or without etoposide treatment was analyzed separately. **P<0.01, ***P<0.001 to wild type; ^aP<0.05, ^bP<0.01, ^cP<0.001 to DblE. Comparison of the effect of etoposide treatment for each protein was made by Student's t-test (results shown in brackets). (d) Cytosolic cytochrome C in the same cells was measured by western blotting. (e) Quantification of cytosolic cytochrome C in etoposide-treated cells. Normalized optical density of western blot bands is shown (means±S.E.M.). The results of four experiments were analyzed by one-way ANOVA with Protein as the main factor. The effect of Protein was significant (F(3,6)=29.5, P=0.0001). ***P<0.001; **P<0.01, to empty vector; ^#P<0.05 to wt; ^cP<0.001 to 1-380. (f) Intracellular distribution of 1-380 (open arrowhead) in control (−) and etoposide-treated (+) cells transfected with indicated amounts of 1-380 DNA. (g) Western blotting of cytosolic cytochrome C in these cells. (h) Quantification of cytosolic cytochrome C in etoposide-treated cells as normalized optic density (means±S.E.M.). The results of four experiments were analyzed by one-way ANOVA with Concentration as the main factor. The effect of the factor was significant (F(3,8)=23.1, P=0.0003); ***P<0.001; **P<0.01 to 0 μg; ^bP<0.01 to the 10 μg

Figure 4

Arrestin-2-(1-380) directly binds tBID and isolated mitochondria, facilitating tBID-induced cytochrome C release. (a) Isolated mouse liver mitochondria (20 μg) were incubated with or without 10 nM tBID and 0, 50, or 100 nM purified 1-380 for 20 min at room temperature. Mitochondria were pelleted by centrifugation at 16 000 × g for 10 min at 4 °C. The distribution of cytochrome C (Cyto C), 1-380, and tBID in the pellet and supernatant is shown. (b) The fraction of cytochrome C released by 10 nM tBID without (0 nM) or with 1-380 (50 and 100 nM) is shown as the percentage of the total cytochrome C (released by Triton X-100) (means±S.E.M. of four experiments). **P<0.01; *P<0.05 as compared with 10 nM tBID alone by one-way ANOVA with Concentration as the main factor (F(2,18)=10.1, P=0.0011) followed by Bonferroni/Dunn post hoc comparison. (c) Isolated mouse liver mitochondria (20 μg) were incubated with or without 100 nM BimL and 0, 50, or 100 nM purified 1-380 for 1 h at 37 °C. Mitochondria were pelleted by centrifugation at 16 000 × g for 10 min at 4 °C. The distribution of cytochrome C (Cyto C), 1-380, and BimL in the pellet and supernatant is shown. (d) The fraction of cytochrome C released by 100 nM BimL without (0 nM) or with 1-380 (50 and 100 nM) is shown as the percentage of the total cytochrome C (released by Triton X-100) (means±S.E.M. of four experiments). ^#P<0.05 as compared with 100 BimL plus 50 nM 1-380 by one-way ANOVA with Bonferroni/Dunn post hoc comparison. (e–g) In vitro binding of purified recombinant DblE or 1-380 to His₆-tagged tBID, which retains full biological activity (Supplementary Figure S2C). (e) Coomassie gel showing the purity of DblE, 1-380, and tBID. (f) Bound His₆-tagged tBID (20% of total eluate) was analyzed by SDS-PAGE and Coomassie staining. Similar amounts of tBID were bound to Ni-beads used for pull-down. (g) Bound DblE and 1-380 were measured by western blotting, with known amounts of purified recombinant proteins used as standards. Quantification of DblE and 1-380 bound to tBID (numbers above bars are means±S.E.M.; n=4) is shown. See also Supplementary Figure S2

Figure 5

The generation of 1-380 facilitates apoptotic cell death. (a–c) A3KO and DKO MEFs were treated with TNFα/CHX (10 ng/ml and 10 μg/ml) for the indicated times. Apoptosis was assessed by the percentage of cells positive for active caspase-3 (a) or stained with YO-PRO-1 (b) using FACS in four independent experiments. The data were analyzed by two-way ANOVA with MEF and Time as main factors. In both caspases-3 and YO-PRO experiments, the effects of MEF were highly significant (P<0.0001). ***P<0.001; **P<0.01; *P<0.05 to DKO MEFs according to unpaired Student's t-test for each time point. (c) The activation of caspases in A3KO and DKO MEFs treated with TNFα/CHX analyzed by western blotting. (d) A3KO and DKO MEFs were treated with etoposide (100 μM) for the indicated times. Apoptosis was assessed by the percentage of cells stained with YO-PRO-1. The two-way ANOVA analysis with MEF and Time as main factors yielded highly significant effect of MEF (P=0.0001). **P<0.001; *P<0.005 to DKO MEFs according to unpaired Student's t-test for each time point. (e) The activation of caspase-6 and -3 in A3KO and DKO MEFs treated with etoposide analyzed by western blotting. (f) A3KO or DKO MEFs expressing GFP, wt Arr2+GFP, DblE+GFP, or 1-380+GFP were exposed to TNFα/CHX for the indicated times. Apoptosis was determined as the percentage of cells positive for active caspase-3 in GFP-positive subsets measured by FACS (means±S.E.M. of four experiments). The data were analyzed by one-way ANOVA for each time point with MEF Type as the main factor. The MEF effect was highly significant for 6 and 10 h TNFα/CHX treatment time points. **P<0.001, *P<0.01 to DKO, DKO DblE, and DKO Arr2 wt; ***P<0.001; *P<0.05 to DKO MEFs; ^$P<0.05, ^#P<0.001 to DblE-GFP; ^bP<0.01 to Arr2-GFP according to Bonferroni/Dunn post hoc test. See also Supplementary Figure S3

Figure 6

The action of 1-380 depends on the expression of BID. (a) BID KO MEFs were transduced with virus expressing GFP (control); GFP+1-380; GFP+BID; or GFP+1-380+BID (Supplementary Figure S4A). Expression of arrestin (F4C1), GFP, and BID in BID KO MEFs was determined by western blotting in lysates of control and TNFα/CHX-treated cells. Western blotting for procaspase-3 (Casp-3) shows the cleavage of caspase-3 induced by TNFα/CHX (10 ng/ml and 10 μg/ml). (b) BID KO MEFs expressing GFP, GFP+1-380, GFP+BID, or GFP+1-380+BID were exposed to TNFα/CHX for 6 h. Apoptosis was measured as a fraction of GFP-expressing cells positive for active caspase-3 by FACS (means±S.E.M. of four experiments). The data were analyzed by one-way ANOVA with protein as the main factor. The effect of Protein was highly significant (F(3,16)=19.7, P<0.001). ***P<0.001; *P<0.05 to GFP control, ^aP<0.05, ^cP<0.001 to 1-380; and ^#P<0.05 to BID according to Bonferroni/Dunn post hoc comparison. (c) BID KO MEFs were transduced with virus expressing GFP (control), GFP+1-380, GFP+BID, or GFP+1-380+BID. Cells were pre-sorted for GFP and re-plated before treatment with TNFα/CHX for 6 h. Expression of arrestin (F4C1), GFP, and BID in BID KO MEFs was determined by western blotting. (d) BID KO MEFs expressing GFP, GFP+1-380, GFP+BID, or GFP+1-380+BID were exposed to TNFα/CHX for 6 h. Apoptosis was measured as a fraction of cells with cytosolic cytochrome C among GFP-positive cells by FACS (means±S.E.M. of four experiments). The data were analyzed by one-way ANOVA with protein as the main factor. The effect of Protein was highly significant (F(3,12)=27.9, P<0.001). ***P<0.001 to GFP control; ^cP<0.001 to 1-380; ^#P<0.05 to BID according to Bonferroni/Dunn post hoc comparison. (e) Subcellular localization of 1-380 in BID KO MEFs transduced with retrovirus expressing 1-380 with and without TNFα/CHX treatment for 6 h. (f) BID KO MEFs were transduced with virus expressing GFP (control) or GFP+1-380. Cells were pre-sorted for GFP and re-plated before treatment with etoposide (100 μM) for 36 h. Apoptosis was measured as a fraction of cells with cytosolic cytochrome C among GFP-positive cells by FACS (means±S.E.M. of four experiments). Two-way ANOVA with Protein (GFP versus 1-380) and Treatment (control versus etoposide) yielded significant effect of Treatment (P<0.0001) but no effect of Protein or Protein × Treatment interaction (P>0.05). ***P<0.001 to respective control values according to Student's t-test. See also Supplementary Figure S4

Figure 7

Caspase-cleaved arrestin-2 and BID cooperatively enhance cytochrome C release and cell death. The activation of the death receptor by TNFα results in the assembly of the multi-protein complex II that directly activates caspase-8. Caspase-8 cleaves arrestin-2 and BID, generating 1-380 and tBID. 1-380 directly binds tBID and greatly enhances its ability to induce cytochrome C release from mitochondria. Cytochrome C organizes apoptosome, activating caspase-9, which then activates massive amounts of caspase-3. Therefore, cytochrome C release considered ‘the point of no return' in cell commitment to death is facilitated via interaction of tBID with caspase-generated arrestin-2 fragment 1-380.TNFR, TNFα receptor; RIP, receptor-interacting serine/threonine-protein kinase 1; FADD, Fas-associated death domain protein; TRADD, TNF receptor-associated death domain (TRADD); TRAF, TNF receptor-associated factor; FLIP, FLICE-like inhibitory protein (a.k.a. CFLAR, CASP8, and FADD-like apoptosis regulator); DD, death domain; DED, death effector domain