FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2167430243> ?p ?o ?g. }

• W2167430243 endingPage "20729" @default.
• W2167430243 startingPage "20722" @default.
• W2167430243 abstract "Autophagic cell death is characterized by the accumulation of vacuoles in physiological and pathological conditions. However, its molecular event is unknown. Here, we show that Atg5, which is known to function in autophagy, contributes to autophagic cell death by interacting with Fas-associated protein with death domain (FADD). Down-regulation of Atg5 expression in HeLa cells suppresses cell death and vacuole formation induced by IFN-γ. Inversely, ectopic expression of Atg5 using adenoviral delivery induces autophagic cell death. Deletion mapping analysis indicates that procell death activity resides in the middle and C-terminal region of Atg5. Cells harboring the accumulated vacuoles triggered by IFN-γ or Atg5 expression become dead, and vacuole formation precedes cell death. 3-Methyladenine or expression of Atg5K130R mutant blocks both cell death and vacuole formation triggered by IFN-γ, whereas benzyloxycarbonyl-VAD-fluoromethyl ketone (Z-VAD-fmk) inhibits only cell death but not vacuole formation. Atg5 interacts with FADD via death domain in vitro and in vivo, and the Atg5-mediated cell death, but not vacuole formation, is blocked in FADD-deficient cells. These results suggest that Atg5 plays a crucial role in IFN-γ-induced autophagic cell death by interacting with FADD. Autophagic cell death is characterized by the accumulation of vacuoles in physiological and pathological conditions. However, its molecular event is unknown. Here, we show that Atg5, which is known to function in autophagy, contributes to autophagic cell death by interacting with Fas-associated protein with death domain (FADD). Down-regulation of Atg5 expression in HeLa cells suppresses cell death and vacuole formation induced by IFN-γ. Inversely, ectopic expression of Atg5 using adenoviral delivery induces autophagic cell death. Deletion mapping analysis indicates that procell death activity resides in the middle and C-terminal region of Atg5. Cells harboring the accumulated vacuoles triggered by IFN-γ or Atg5 expression become dead, and vacuole formation precedes cell death. 3-Methyladenine or expression of Atg5K130R mutant blocks both cell death and vacuole formation triggered by IFN-γ, whereas benzyloxycarbonyl-VAD-fluoromethyl ketone (Z-VAD-fmk) inhibits only cell death but not vacuole formation. Atg5 interacts with FADD via death domain in vitro and in vivo, and the Atg5-mediated cell death, but not vacuole formation, is blocked in FADD-deficient cells. These results suggest that Atg5 plays a crucial role in IFN-γ-induced autophagic cell death by interacting with FADD. Programmed cell death can be classified based on morphological criteria into several categories. The major extensively studied category, apoptosis, is characterized by membrane blebbing, DNA fragmentation, and preservation of organelles (1Kerr J.F. Wyllie A.H. Currie A.R. Br. J. Cancer. 1972; 26: 239-257Crossref PubMed Scopus (12927) Google Scholar). Differing from apoptosis, autophagic cell death or type II programmed cell death exhibits the appearance of vacuoles engulfing bulk cytoplasm and cytoplasmic organelles such as mitochondria and endoplasmic reticulum (2Bursch W. Hochegger K. Torok L. Marian B. Ellinger A. Hermann R.S. J. Cell Sci. 2000; 113: 1189-1198Crossref PubMed Google Scholar, 3Zakeri Z. Bursch W. Tenniswood M. Lockshin R. Cell Death Diff. 1995; 2: 87-96PubMed Google Scholar). Autophagic cell death has been described in IFN-γ 1The abbreviations used are: IFN-γ, interferon-γ; FADD, Fas-associated protein with death domain; DD, death domain; HA, hemagglutinin; LC3, microtubule-associated protein light chain 3; MDC, monodansylcadaverin; z, benzyloxycarbonyl; fmk, fluoromethyl ketone; GST, glutathione S-transferase; GFP, green fluorescent protein; EGFP, enhanced GFP; RFP, red fluorescent protein; 3-MA, 3-methyladenine; AFP, α-feto protein; TNF, tumor necrosis factor; TNFR, TNF receptor. -treated human cervical carcinoma HeLa cells and anti-estrogen tamoxifen-treated human mammary carcinoma MCF-7 cells with the accumulation of autophagic vacuoles (2Bursch W. Hochegger K. Torok L. Marian B. Ellinger A. Hermann R.S. J. Cell Sci. 2000; 113: 1189-1198Crossref PubMed Google Scholar, 4Gozuacik D. Kimchi A. Oncogene. 2004; 23: 2891-2906Crossref PubMed Scopus (1267) Google Scholar). Autophagic cell death is also extensively observed in steroid-triggered cell death during development of Drosophila (5Baehrecke E.H. Nat. Rev. Mol. Cell. Biol. 2002; 10: 779-787Crossref Scopus (335) Google Scholar, 6Lee C.Y. Clough E.A. Yellon P. Teslovich T.M. Stephan D.A. Baehrecke E.H. Curr. Biol. 2003; 13: 350-357Abstract Full Text Full Text PDF PubMed Scopus (176) Google Scholar, 7Lee C.Y. Cooksey B.A. Baehrecke E.H. Dev. Biol. 2002; 250: 101-111Crossref PubMed Scopus (180) Google Scholar). Recently, death-associated protein kinase (DAPk) and death-associated related protein kinase-1 (DRP-1) were shown to play a role in IFN-γ-induced autophagic cell death (8Inbal B. Bialik S. Sabanay I. Shani G. Kimchi A. J. Cell Biol. 2002; 157: 455-468Crossref PubMed Scopus (422) Google Scholar). Also, beclin 1 and Atg7 were reported to be associated with autophagic cell death (9Ogier-Denis E. Codogno P. Biochim. Biophys. Acta. 2003; 1603: 113-128PubMed Google Scholar, 10Yu L. Alva A. Su H. Dutt P. von Freundt E. Welsh S. Baehrecke E.H. Lenardo M.J. Science. 2004; 304: 1500-1502Crossref PubMed Scopus (1108) Google Scholar). However, evidence is not clear yet that autophagic cell death employs autophagy machinery. The physiological and pathological role of the autophagic cell death is emphasized with various evidences, giving a therapeutic value to autophagic cell death. Autophagic cell death was proposed as the action mechanism of some anticancer agents (4Gozuacik D. Kimchi A. Oncogene. 2004; 23: 2891-2906Crossref PubMed Scopus (1267) Google Scholar). Also, Danon disease, cardiomyopathies, and skeletal myopathies with different genetic origins are characterized by the accumulation of autophagic vacuoles (11Nishino I. Fu J. Tanji K. Yamada T. Shimojo S. Koori T. Mora M. Riggs J.E. Oh S.J. Koga Y. Sue C.M. Yamamoto A. Murakami N. Shanske S. Byrne E. Bonilla E. Nonaka I. DiMauro S. Hirano M. Nature. 2000; 406: 906-910Crossref PubMed Scopus (736) Google Scholar, 12Tanaka Y. Guhde G. Suter A. Eskelinen E.L. Hartmann D. Lullmann-Rauch R. Janssen P.M. Blanz J. Figura K. Saftig P. Nature. 2000; 406: 902-906Crossref PubMed Scopus (734) Google Scholar, 13Yamamoto A. Morisawa Y. Verloes A. Murakami N. Hirano M. Nonaka I. Nishino I. Neurology. 2001; 57: 903-905Crossref PubMed Scopus (47) Google Scholar). Likewise, an increase in the number of autophagic vacuoles is a hallmark of various neurodegenerative diseases such as Alzheimer disease, Huntington disease, and Parkinson disease (14Anglade P. Vyas S. Javoy-Agid F. Herrero M.T. Michel P.P. Marquez J. Mouatt-Prigent A. Ruberg M. Hirsch E.C. Agid Y. Histol. Histopathol. 1997; 12: 25-31PubMed Google Scholar, 15Cataldo A.M. Barnett J.L. Berman S.A. Li J. Quarless S. Bursztajn S. Lippa C. Nixon R.A. Neuron. 1995; 3: 671-680Abstract Full Text PDF Scopus (307) Google Scholar, 16Kegel K.B. Kim M. Sapp E. McIntyre C. Castano J.G. Aronin N. Difiglia M. J. Neurosci. 2000; 20: 7268-7278Crossref PubMed Google Scholar, 17Nixon R.A. Cataldo A.M. Mathews P.M. Neurochem. Res. 2000; 9–10: 1161-1172Crossref Scopus (282) Google Scholar). However, most reported pathologies have described the accumulation of autophagic vacuoles without clear evidence for autophagic cell death or its contribution to diseases. Although much information has accumulated over the past decade about genes that mediate the morphological and biochemical changes of apoptosis, much less is known about the molecular mechanisms underlying autophagic cell death. In this study, we show that IFN-γ-induced vacuole-associated cell death comprises two-subsequent steps: vacuole formation and cell death. Atg5 (Apg5) (18Klionsky D.J. Cregg J.M. Dunn W.A. Emr S.D. Sakai Y. Sandoval I.V. Sibirny A. Subramani S. Thumm M. Veenhuis M. Ohsumi Y. Dev. Cell. 2003; 5: 539-545Abstract Full Text Full Text PDF PubMed Scopus (1024) Google Scholar), a key molecule involved in autophagic vacuole formation, is required for the autophagic cell death; Atg5 interacts with death domain of FADD as a necessary step leading to cell death, linking autophagy machinery to type II programmed cell death. Cell Culture, Transfection, and Cell Death Assay—HeLa, MCF-7, and Hep3B cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (Invitrogen). Jurkat, Jurkat (I2-1), and Jurkat (I9-2) cells (19Bodmer J.L. Holler N. Reynard S. Vinciguerra P. Schneider P. Juo P. Blenis J. Tschopp J. Nat. Cell Biol. 2000; 2: 241-243Crossref PubMed Scopus (586) Google Scholar) were maintained with RPMI 1640 containing 10% fetal bovine serum. Transient transfection was performed with Lipofectamine reagent (Invitrogen) following the manufacturer's instructions. Typically, 1 × 105 cells/well in 12-well dishes were transfected with 0.5 μg of plasmids. Cells were exposed with various stimuli, and cell death was then assessed by counting GFP-positive cells showing condensed and fragmented nuclei after staining with Hoechst 33258 (Molecular Probes) or by trypan blue exclusion assay (>400 cells). Staining of autophagic vacuoles with monodansylcadaverin (MDC; Fluka) was described previously (20Biederbick A. Kern H.F. Elsasser H.P. Eur. J. Cell Biol. 1995; 66: 3-14PubMed Google Scholar). Fluorescent microscopy (Axiovert 25; GFP filter-Ex470/40VS495, Em515/30; RFP filter-Ex546/12VS510EmLP590; Zeiss) was performed with an excitation filter range of 470–490 nm or 360–370 nm and a barrier filter of 515 nm or 470–490 nm for GFP, Hoechst 33258, and MDC fluorescence, respectively. DNA and Reagents—The human Atg5, Atg5K130R mutant replacing Lys-130 with Arg (pAtg5K130R-HA), and LC3 (microtubule-associated protein light chain 3) expression plasmids were described previously (21Mizushima N. Yamamoto A. Hatano M. Kobayashi Y. Kabeya Y. Suzuki K. Tokuhisa T. Ohsumi Y. Yoshimori T. J. Cell Biol. 2001; 152: 657-668Crossref PubMed Scopus (1164) Google Scholar, 22Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1297) Google Scholar). Deletion mutants of Atg5 were generated by PCR and cloned into XhoI and XbaI sites of pcDNA3-HA. IFN-γ was obtained from Sigma and LG Biotech. MDC was from Fluka. Etoposide, cis-PLATINUM(II)-diammine dichloride, staurosporine, and 3-methyladenine (3-MA) were purchased from Sigma. Z-VAD-fmk (zVal-Ala-Asp-fluoromethyl ketone) and IETD-fmk (Ile-Glu-Thr-Asp-fluoromethyl ketone) were purchased from Bachem. Yeast Two-hybrid Screen—The death domain region (amino acids 100–179) of FADD was used as bait in the conventional yeast two-hybrid screening using pLexA vector (Clontech) and Jurkat cDNA library (Stratagene). Death domain of FADD was cloned into EcoRI and BamHI sites of pLexA. Antisense Oligonucleotide Treatment—Two kinds of antisense oligonucleotides complementary to human Atg5 mRNA (oligonucleotide 15, 5′-CATCTTTGTCATCTGTCA-3′; oligonucleotide 16, 5′-TTGTCAGTTACCAACGTC-3′) were generated. Comparison of the oligonucleotide sequence with data base detected homology only to the Atg5 sequence. C-2 (5′-GCTACTAGTAGCAGCTAC-3′) scrambled sequences served as a control. HeLa cells were treated with Atg5 antisense or scrambled oligonucleotide in Dulbecco's modified Eagle's medium containing Lipofectamine reagent for 3 h and then in the same mixture containing 10% fetal bovine serum for an additional 72 h. Construction of Recombinant Adenoviruses—Recombinant adenovirus containing Atg5 under the control of hepatoma-specific α-feto protein (AFP) promoter (Rad-AFP-Atg5) was constructed as following. Shuttle vector containing Atg5 under AFP promoter (pΔE1sp1A-AFPe-Atg5-BGH (bovine growth factor)) and adenoviral vector containing the Ad5 genome (vmdl324BstBI) (obtained from Dr. Verca in the University of Fribourgh, Switzerland) were linearized with ScaI and BstBI, respectively, and co-transformed into BJ5183 Escherichia coli (23Chartier C. Degryse E. Gantzer M. Dieterle A. Pavirani A. Mehtali M. J. Virol. 1996; 7: 4805-4810Crossref Google Scholar). To verify the proper homologous recombinants, DNAs purified were digested with HindIII. Rad-AFP-Atg5 was generated after transfection of PacI-digested recombinant into HEK293 cells. Rad-LacZ (β-galactosidase) was hired as a control. All viruses were plaque-purified, propagated, and titered in HEK293 cells using a standard method. Western Blot Analysis—Western blot analysis was performed as described previously (24Song S. Kim S.Y. Hong Y.M. Jo D.G. Lee J.Y. Shim S.M. Chung C.W. Seo S.J. Yoo Y.J. Koh J.Y. Lee M.C. Yates A.J. Ichijo H. Jung Y.K. Mol. Cell. 2003; 12: 553-563Abstract Full Text Full Text PDF PubMed Scopus (138) Google Scholar). Anti-Atg5 (SO4) and anti-LC3 antibodies were described (22Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1297) Google Scholar), and anti-human FADD (clone A66-2) and antihuman-tubulin antibodies (clone B-5-1-2) are from Pharmingen and Sigma, respectively. In Vitro Pull-down Assay—In vitro pull-down assay was carried out as described previously (25Choi Y.H. Kim K.B. Kim H.H. Hong G.S. Kwon Y.K. Chung C.W. Park Y.M. Shen Z.J. Kim B.J. Lee S.Y. Jung Y.K. J. Biol. Chem. 2001; 276: 25073-25077Abstract Full Text Full Text PDF PubMed Scopus (40) Google Scholar). Immunoprecipitation Assay—Cells were lysed in radioimmune precipitation buffer containing 50 mm Tris-HCl (pH 7.4), 1% Nonidet P-40, 150 mm NaCl, 1 mm EDTA, 1 mm phenylmethylsulfonyl fluoride, and 1 μg/ml each aprotinin, leupeptin, and pepstatin. Cell lysates were incubated with anti-human FADD antibody or preimmune IgG after preclearing with protein G-Sepharose (Amersham Biosciences) for 30 min at 4 °C. After adding additional protein G-Sepharose, the mixtures were further rotated for 3 h. Immunocomplexes were collected by centrifugation and examined with Western blotting after separation by SDS-PAGE. Atg5 Is Required for IFN-γ-induced Cell Death and Vacuole Formation—In an attempt to isolate a protein interacting with death domain of FADD, we have isolated Atg5. To define the role of Atg5 in cell death, we directly targeted the expression of Atg5 with antisense oligonucleotides. Interestingly, IFN-γ-induced cell death was inhibited in HeLa cells transfected with Atg5 antisense oligonucleotides but not with scrambled oligonucleotides (Fig. 1A), and the reduced expression of Atg5, but not relevant protein FADD, was confirmed with Western blot analysis (Fig. 1A, insets). Similar effects were confirmed in HeLa cells transfected by Atg5 antisense cDNA (data not shown). However, down-regulation of Atg5 expression did not give any influences on cell death induced by the treatment with etoposide, staurosporine, or cisplatin (Fig. 1B). IFN-γ-induced cell death was previously suggested as an autophagic cell death showing the main characteristics of vacuole formation (8Inbal B. Bialik S. Sabanay I. Shani G. Kimchi A. J. Cell Biol. 2002; 157: 455-468Crossref PubMed Scopus (422) Google Scholar). Thus, autophagic vacuole formation was addressed by using exogenous GFP-fused microtubule-associated protein light chain 3 (GFP-LC3), a specific lysosomal/autophagic biochemical vacuole marker (22Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1297) Google Scholar). Microscopic analysis showed that IFN-γ treatment induced the appearance of LC3 dots in control cells (Fig. 1C, upper panel) but not in HeLa cells transfected with Atg5 antisense oligonucleotides (Fig. 1C, lower panel). IFN-γ-induced vacuole formation was confirmed by MDC staining, too (data not shown). These results indicate that Atg5 is required for IFN-γ-induced cell death and vacuole formation. Biochemical analysis employing Western blotting using anti-LC3 antibody showed that IFN-γ treatment increased the level of LC3-II, the autophagosome-associating form of LC3, as compared with LC3-I (Fig. 1D), indicating that IFN-γ increased the autophagosome membrane association of LC3 for vacuole formation. Interestingly, ectopic expression of Atg5K130R mutant, which is defective in its ability to conjugate with Atg12 during autophagy (22Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1297) Google Scholar), inhibited both cell death and MDC-positive vacuole formation triggered by IFN-γ (Fig. 1E). A vacuole formation assay using GFP-LC3 also showed that the expression of Atg5K130R mutant blocked IFN-γ-induced vacuole formation (Fig. 1F, lower panel). These results suggest that the vacuole formation induced by IFN-γ may share similar machinery as autophagy and be associated with cell death via Atg5. Ectopic Expression of Atg5 Induces Cell Death and Vacuole Formation—Inversely, when Atg5 was transiently overexpressed in HeLa and MCF-7 cells for 48 h, we observed the characteristics of cell death, cell shrinkage, and nuclear condensation as examined with Hoechst dye staining (Fig. 2A and B). As compared with p55/TNFR1 (Fig. 2B, left panel), however, Atg5 expression induced the appearance of condensed nuclei without fragmentation (Fig. 2B, right panel). Also, consistent with its inhibitory role in IFN-γ-induced cell death, the expression of Atg5K130R mutant did not exhibit any pro-cell death activity (Fig. 2A). At an earlier time (24 h), the examination of vacuole formation using GFP-LC3 showed that transient expression of Atg5 in HeLa cells induced the accumulation of LC3 dots without significant cell death, increasing the proportion of cells with autophagic vacuoles to 30–60% (Fig. 2C, left panel). A similar increase of vacuole formation was observed at 24 h in Hep3B human hepatoma cells infected with adenovirus expressing Atg5 under the control of hepatoma-specific α-feto protein promoter (Ade-Atg5) (Fig. 2C, right panel). As shown in Fig. 2D, the conversion of LC3-I to LC3-II was observed by Western blotting of Hep3B cells infected with Ade-Atg5 but not in control cells infected with adenovirus expressing β-galactosidase (Ade-LacZ). We then analyzed the kinetics of vacuole formation and cell death with GFP-LC3 by chasing times in the single cell level (Supplementary Fig. 1). IFN-γ-induced vacuoles began to accumulate at an earlier time (20 h) and sequentially tended to aggregate into bigger spots until 39 h in HeLa cells. At a later time (40 h), most of the cells harboring aggregated vacuoles became shrunken and were eventually dead (Supplementary Fig. 1A). Similarly, vacuole formation was observed at 16 h in the Atg5-transfected cells and apparently followed by the induction of cell death and detachment from the culture plate between 43 and 64.5 h (Supplementary Fig. 1B), linking vacuole accumulation to cell death in the single cell level. These observations demonstrate that IFN-γ- or Atg5-induced vacuole formation precedes cell death and suggest that Atg5 exhibits vacuole formation-associated pro-cell death activity. To map the functional domain of Atg5, we constructed several deletion mutants (Fig. 2E). Subsequently, the ability of each mutant to induce cell death was examined in HeLa cells with transient expression analysis (Fig. 2F). Atg5-(71–170), Atg5-(171–276), and Atg5-(71–276) mutants induced cell death, whereas other mutants including Atg5-(1–70) and Atg5-(1–170) lost their abilities to induce cell death. These results suggest that the C terminus (171–276) of Atg5 contains cell death-inducing activity. Also, the middle region (71–170) of Atg5 is potent to induce cell death, which is inhibited by the N terminus (1–70). A similar expression level of Atg5 mutants was determined with Western blot analysis (Fig. 2F, insets). Identification of Atg5 as a FADD-interacting Molecule—We have isolated Atg5 as a protein interacting with FADD by yeast two-hybrid assay. An in vitro binding assay showed that the radiolabeled Atg5 was pulled-down with purified GST-fused FADD (GST-FADD) and GST-fused death domain of FADD (FADD-DD) but not with GST (Fig. 3A). We determined the domain of Atg5 interacting with FADD after co-transfection into HEK293 cells. Immunoprecipitation assay showed that all deletions except Atg5-(1–70) interacted with FADD (Fig. 3B), indicating that the domain of Atg5 interacting with FADD spans the middle and C-terminal region. FADD-interacting domains generally correspond to cell death-inducing activity of Atg5 deletions except Atg5-(1–170). Then protein-protein interaction of endogenous Atg5 and FADD was examined in HeLa cells (Fig. 3C). Western blot analysis using anti-Atg5 and Atg12 antibodies showed that Atg5 and Atg12 were detected in the immunocomplexes precipitated by anti-FADD antibody and that the amounts of Atg5 detected were almost same in control and IFN-γ-treated cells. These results suggest that Atg5 interacts with FADD and that the interaction is not changed by the treatment with IFN-γ. We could not perform reciprocal immunoprecipitation assay because the molecular weight of FADD overlaps with the light chain of immunoglobulin. Because of the limitation of anti-Atg5 antibody for immunocytochemical analysis, subcellular co-localization of Atg5 with FADD or FADD-DD was examined in HeLa cells after transfection of GFP-fused Atg5 (GFP-Atg5) with either RFP-FADD or RFP-FADD-DD (Fig. 3D). Although GFP-Atg5 alone was detected with diffused pattern in the transfected cells, RFP-FADD and RFP-FADD-DD showed distinct subcellular localizations, aggregated and diffused cytosolic patterns, respectively (Fig. 3D). Co-expression of GFP-Atg5 with RFP-FADD or RFP-FADD-DD induced change in the subcellular localization of GFP-Atg5 to aggregated and diffused cytosolic patterns, overlapping with those of RFP-FADD and RFP-FADD-DD, respectively. These results suggest that Atg5 interacts with FADD via death domain. FADD Is Required for Atg5-mediated Cell Death but Not Vacuole Formation—We then characterized Atg5-mediated cell death by investigating functional significance of the protein-protein interaction between Atg5 and FADD. The role of FADD in Atg5-mediated cell death was addressed by using antisense approaches. Transfection of HeLa cells with FADD antisense cDNA diminished the expression level of endogenous FADD as determined with Western blot analysis (Fig. 4A, insets). Surprisingly, the reduced expression of FADD blocked Atg5-induced cell death to the control level (Fig. 4A). Apoptosis triggered by death receptors, such as p55/TNF receptor (TNFR) 1 and TRAIL receptors that employ FADD as an adaptor molecule, was inhibited in FADD knock-down cells, whereas apoptosis triggered by etoposide, a non-receptor signal, was not suppressed (Fig. 4B). Similarly, Atg5-induced cell death was also blocked in FADD-deficient Jurkat cells (I2-1) and partially in caspase-8-deficient Jurkat cells (I9-2) (Fig. 4C). Further, as shown in Fig. 4D, co-expression of Atg5 with FADD synergistically induced cell death. Expression of Atg5 with FADD-DD was also potent to induce cell death but less potent than FADD. These results indicate that FADD-DD is effective to mediate Atg5-induced cell death. On the contrary, FADD-induced cell death was not affected by the reduced expression of Atg5 in HeLa cells (Fig. 4E) and by the treatment with 3-MA (Fig. 4F), a widely used inhibitor of vacuole formation in autophagy (26Selgen P.O. Gordon P.B. Proc. Natl. Acad. Sci. U. S. A. 1982; 79: 1889-1892Crossref PubMed Scopus (1181) Google Scholar). These results indicate that FADD is required for Atg5-induced cell death. We then investigated the contribution of FADD to vacuole formation and cell death triggered by IFN-γ. Down-regulation of FADD expression significantly reduced IFN-γ-induced cell death (40–25%) (Fig. 5A and B, upper panel) but not the accumulation of vacuoles (Fig. 5, A and B, lower panel). These results indicate that FADD mediates a downstream event of vacuole formation in Atg5-mediated cell death and IFN-γ-induced cell death. Dissection of Autophagic Cell Death into Two Continuous Steps, Vacuole Formation and Cell Death—To further ensure the dissection of Atg5-mediated autophagic cell death into vacuole formation and cell death, 3-MA, Z-VAD-fmk (pan caspase inhibitor), and IETD-fmk (caspase-8 inhibitor) were employed. IFN-γ-induced cell death was almost completely inhibited by the incubation of HeLa cells with Z-VAD-fmk and partially suppressed by 3-MA (Fig. 6A). IETD-fmk was a little effective in reducing IFN-γ-induced cell death. Similar effects of those inhibitors on Atg5-induced cell death were observed (Fig. 6C). It is notable to observe that 3-MA was more potent to suppress cell death induced by the expression of Atg5 as compared with its effects on IFN-γ-induced cell death, implying that IFN-γ induces both 3-MA-sensitive autophagic cell death in which Atg5 plays a role and 3-MA-insensitive apoptosis (Fig. 7). On the contrary, 3-MA completely blocked IFN-γ- or Atg5-induced accumulation of MDC-positive vacuoles to the control levels (Fig. 6, B and D), whereas Z-VAD-fmk and IETD-fmk failed to show such inhibitory effects. Consistently, administration with 3-MA, but not Z-VAD-fmk and IETD-fmk, abolished the appearance of GFP-LC3 dots in HeLa cells exposed to IFN-γ (Fig. 6E, upper panel) or expressing exogenous Atg5 (Fig. 6E, lower panel). Taken together, these results demonstrate that IFN-γ-induced autophagic cell death consists of two continuous but separable processes: 3-MA-inhibitable vacuole formation and Z-VAD-inhibitable, FADD-dependent cell death in which vacuole formation precedes cell death.Fig. 7Proposed roles of Atg5 and FADD in IFN-γ-induced vacuole-associated cell death. IFN-γ (<10,000 units/ml)-induced cell death is mainly contributed by autophagic cell death in which Atg5 mediates IFN-γ-induced vacuole formation and subsequent cell death through interacting with FADD. Although 3-MA suppresses Atg5-mediated vacuole formation and cell death, Z-VAD-fmk inhibits only cell death.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Increasing evidence indicates that autophagic cell death is observed during development and neurodegenerative diseases and mainly characterized by the accumulation of vacuoles in the cytoplasm. Typically, HeLa and MCF-7 cells exposed to IFN-γ and tamoxifen, respectively, were demonstrated to undergo autophagic cell death. By using the autophagic cell death model, we show that Atg5 plays a crucial role in autophagic cell death consisting of two continuous events: accumulation of autophagic vacuoles and cell death (Fig. 7). Moreover, excess accumulation of vacuoles is linked to autophagic cell death as follows. (i) Accumulation of autophagic vacuoles is followed by the incidence of cell death in IFN-γ-treated cells. (ii) Treatment with 3-MA blocks IFN-γ-induced cell death. (iii) Knock-down of Atg5 or expression of Atg5K130R mutant suppresses both vacuole formation and cell death. (iv) Overexpression of Atg5 induces vacuole formation and cell death. (v) Z-VAD-fmk suppresses cell death but not vacuole formation. We observed similar results in the tamoxifen-induced autophagic death model of MCF-7 cells (data not shown). Atg5 may play a dual role in autophagy and autophagic cell death. During autophagy, Atg5 is one of the components in a unique non-ubiquitin conjugation reaction that is required for vacuole formation and conjugated to human Atg12 homologue (21Mizushima N. Yamamoto A. Hatano M. Kobayashi Y. Kabeya Y. Suzuki K. Tokuhisa T. Ohsumi Y. Yoshimori T. J. Cell Biol. 2001; 152: 657-668Crossref PubMed Scopus (1164) Google Scholar, 22Mizushima N. Noda T. Yoshimori T. Tanaka Y. Ishii T. George M.D. Klionsky D.J. Ohsumi M. Ohsumi Y. Nature. 1998; 395: 395-398Crossref PubMed Scopus (1297) Google Scholar). In the presence of cell death pressure, autophagy and autophagic cell death may share common machinery utilizing Atg5. The middle and C-terminal region of Atg5 contains a Lys residue for conjugation with Atg12 and binds to FADD for cell death-inducing activity. Also, the observations that Atg5K130R mutant defective in its ability to conjugate with Atg12 failed to induce vacuole-associated cell death suggest that like autophagy, the conjugation of Atg5 with Atg12 may be required for Atg5-mediated cell death. As observed from immunoprecipitation analysis, Atg5 and Atg12 were detected in the complex associated with FADD although we failed to observe the IFN-dependent formation of such protein complexes. Still, the components and biochemical properties of the protein complexes containing FADD-Atg5 remain to be elucidated to better distinguish autophagic cell death from autophagy. In addition, the inhibitory effects of Atg5K130R mutant on autophagic cell death imply that Atg5K130R mutant may act as a dominant negative in autophagic cell death. During preparation of our report, another link of the basic autophagy to autophagic cell death was recently provided by beclin 1 (10Yu L. Alva A. Su H. Dutt P. von Freundt E. Welsh S. Baehrecke E.H. Lenardo M.J. Science. 2004; 304: 1500-1502Crossref PubMed Scopus (1108) Google Scholar), a tumor suppressor protein that interacts with Bcl-2 and is structurally similar to the yeast autophagic gene Apg6/Vps30p (27Liang X.H. Jackson S. Seaman M. Brown K. Kempkes B. Hibshoosh H. Levine B. Nature. 1999; 402: 672-676Crossref PubMed Scopus (2771) Google Scholar). Until recently, it is believed that autophagic cell death and apoptosis are controlled by different mechanisms, mainly by their dependence on caspase. Autophagic cell death employing IFN-γ is generally known to be caspase-independent (4Gozuacik D. Kimchi A. Oncogene. 2004; 23: 2891-2906Crossref PubMed Scopus (1267) Google Scholar, 8Inbal B. Bialik S. Sabanay I. Shani G. Kimchi A. J. Cell Biol. 2002; 157: 455-468Crossref PubMed Scopus (422) Google Scholar). Doses ranging from 500 to 1,000 unit/ml IFN-γ are frequently employed to induce autophagic death of HeLa cells. However, we consistently observed that Z-VAD-fmk was very potent to suppress autophagic death induced by 1,000 units/ml IFN-γ but not effective to inhibit cell death induced by much higher doses of IFN-γ (>10,000 units/ml) (Supplementary Fig. 2A). On the contrary, Z-VAD-fmk could not suppress vacuole formation induced by various does of IFN-γ (Supplementary Fig. 2B). We have utilized IFN-γ obtained from several different sources with the same results. Thus, we believe that depending on the amounts of death signal applied, IFN-γ may trigger Z-VAD-dependent or -independent autophagic cell death. Studies of Drosophila salivary gland cell destruction show that caspases contribute to the autophagic cell death (6Lee C.Y. Clough E.A. Yellon P. Teslovich T.M. Stephan D.A. Baehrecke E.H. Curr. Biol. 2003; 13: 350-357Abstract Full Text Full Text PDF PubMed Scopus (176) Google Scholar, 7Lee C.Y. Cooksey B.A. Baehrecke E.H. Dev. Biol. 2002; 250: 101-111Crossref PubMed Scopus (180) Google Scholar), and DAPk and DRP-1 control both caspase-dependent and caspase-independent cell death (28Cohen O. Inbal B. Kissil J.L. Raveh T. Berissi H. Spivak-Kroizaman T. Feinstein E. Kimchi A. J. Cell Biol. 1999; 146: 141-148Crossref PubMed Scopus (0) Google Scholar, 29Matsumura H. Shimizu Y. Ohsawa Y. Kawahara A. Uchiyama Y. Nagata S. J. Cell Biol. 2000; 151: 1247-1256Crossref PubMed Scopus (207) Google Scholar). Thus, the molecular events underlying autophagic cell death may depend on the amounts and types of autophagic cell death signals. Even Z-VAD could induce autophagic cell death in L929 cells (10Yu L. Alva A. Su H. Dutt P. von Freundt E. Welsh S. Baehrecke E.H. Lenardo M.J. Science. 2004; 304: 1500-1502Crossref PubMed Scopus (1108) Google Scholar), and whether or not these types of autophagic cell death utilize Atg5 as a death mediator remains to be elucidated. We have identified Atg5 as a protein interacting with death domain of FADD using yeast two-hybrid screening. To our surprise, subsequent analyses exhibit that the pro-cell death activity of Atg5 requires FADD as a downstream mediator. Although overexpression of FADD may activate death effector domain-containing caspases, such as caspase-8, the reduced expression of FADD abrogates Atg5-induced autophagic cell death. FADD is well known as an adaptor molecule mainly mediating receptor-mediated apoptosis. However, many evidences indicate that FADD is a multifunctional protein, displaying diverse activities in non-receptor-mediated apoptosis, necrosis, and cell growth (29Matsumura H. Shimizu Y. Ohsawa Y. Kawahara A. Uchiyama Y. Nagata S. J. Cell Biol. 2000; 151: 1247-1256Crossref PubMed Scopus (207) Google Scholar). Thus, FADD may be a turning point that determines diverse signaling fates. Elucidation of signal-dependent stoichiometry of FADD-containing complexes, such as FADD-Atg5 and FADD-caspase, will be important. At present, we do not think that caspase-8, which interacts with FADD, significantly contributes to Atg5-mediated cell death. We failed to observe the activation of caspase-8 with enzymatic assay using fluorogenic substrate during Atg5- and IFN-γ-mediated autophagic cell death (Supplementary Fig. 3, A and B), consistent with the results showing the occurrence of Atg5-induced cell death in caspase-8-deficient cells. In spite of the weak activation of caspase-3 (Supplementary Fig. 3), we could not observe the detectable appearance of DNA laddering in dying cells triggered by the expression of Atg5 or by the treatment with IFN-γ (data not shown). However, Z-VAD-inhibitable proteases are likely to be activated downstream of FADD to mediate Atg5- and IFN-γ-induced autophagic cell death. Although Z-VAD-fmk is considered as a pan caspase inhibitor, Z-VAD-fmk (about 50 μm) exhibits its inhibitory effects on other proteases, too, as shown in the disruption of the development of the mammalian preimplantation embryo and in the inhibition of the lysosomal protease (30Lockshin R.A. Zakeri Z. Curr. Opin. Cell Biol. 2002; 6: 727-733Crossref Scopus (207) Google Scholar). Effective cell death triggered by FADD-DD and Atg5 indicates that FADD-DD may function to activate downstream mediator of autophagic cell death, whereas FADD-DED (death effector domain) is usually required for cell death. Identification of such downstream mediators deserves further characterization. In summary, we show for the first time that Atg5 plays a crucial role in IFN-γ-induced autophagic cell death by interacting with FADD as a necessary step leading to cell death, establishing a role of the Atg gene in cell death with autophagic features. We thank Dr. J. Blenis (Harvard University, Boston, MA) for Jurkat cells (A3, I2-1, I9-2) and Dr. T. Yoshimori (National Institute of Genetics, Mishima, Japan) for anti-LC3 antibody. Download .pdf (.79 MB) Help with pdf files" @default.
• W2167430243 created "2016-06-24" @default.
• W2167430243 creator A5000598094 @default.
• W2167430243 creator A5003819949 @default.
• W2167430243 creator A5005358526 @default.
• W2167430243 creator A5008205942 @default.
• W2167430243 creator A5023313081 @default.
• W2167430243 creator A5036966057 @default.
• W2167430243 creator A5038112377 @default.
• W2167430243 creator A5041898888 @default.
• W2167430243 creator A5056264201 @default.
• W2167430243 creator A5064563969 @default.
• W2167430243 creator A5065754348 @default.
• W2167430243 creator A5082387951 @default.
• W2167430243 creator A5083469637 @default.
• W2167430243 date "2005-05-01" @default.
• W2167430243 modified "2023-10-06" @default.
• W2167430243 title "Essential Roles of Atg5 and FADD in Autophagic Cell Death" @default.
• W2167430243 cites W149130878 @default.
• W2167430243 cites W1520292475 @default.
• W2167430243 cites W1526993239 @default.
• W2167430243 cites W1534838454 @default.
• W2167430243 cites W1643149493 @default.
• W2167430243 cites W1784425676 @default.
• W2167430243 cites W1940885739 @default.
• W2167430243 cites W1966134982 @default.
• W2167430243 cites W1988022242 @default.
• W2167430243 cites W1989141893 @default.
• W2167430243 cites W1998730944 @default.
• W2167430243 cites W2019136847 @default.
• W2167430243 cites W2021011570 @default.
• W2167430243 cites W2052726485 @default.
• W2167430243 cites W2052853635 @default.
• W2167430243 cites W2056641170 @default.
• W2167430243 cites W2087920955 @default.
• W2167430243 cites W2100495176 @default.
• W2167430243 cites W2109716633 @default.
• W2167430243 cites W2121672863 @default.
• W2167430243 cites W2129712803 @default.
• W2167430243 cites W2149388384 @default.
• W2167430243 cites W2156320420 @default.
• W2167430243 cites W2161234475 @default.
• W2167430243 cites W2167907490 @default.
• W2167430243 doi "https://doi.org/10.1074/jbc.m413934200" @default.
• W2167430243 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/15778222" @default.
• W2167430243 hasPublicationYear "2005" @default.
• W2167430243 type Work @default.
• W2167430243 sameAs 2167430243 @default.
• W2167430243 citedByCount "498" @default.
• W2167430243 countsByYear W21674302432012 @default.
• W2167430243 countsByYear W21674302432013 @default.
• W2167430243 countsByYear W21674302432014 @default.
• W2167430243 countsByYear W21674302432015 @default.
• W2167430243 countsByYear W21674302432016 @default.
• W2167430243 countsByYear W21674302432017 @default.
• W2167430243 countsByYear W21674302432018 @default.
• W2167430243 countsByYear W21674302432019 @default.
• W2167430243 countsByYear W21674302432020 @default.
• W2167430243 countsByYear W21674302432021 @default.
• W2167430243 countsByYear W21674302432022 @default.
• W2167430243 countsByYear W21674302432023 @default.
• W2167430243 crossrefType "journal-article" @default.
• W2167430243 hasAuthorship W2167430243A5000598094 @default.
• W2167430243 hasAuthorship W2167430243A5003819949 @default.
• W2167430243 hasAuthorship W2167430243A5005358526 @default.
• W2167430243 hasAuthorship W2167430243A5008205942 @default.
• W2167430243 hasAuthorship W2167430243A5023313081 @default.
• W2167430243 hasAuthorship W2167430243A5036966057 @default.
• W2167430243 hasAuthorship W2167430243A5038112377 @default.
• W2167430243 hasAuthorship W2167430243A5041898888 @default.
• W2167430243 hasAuthorship W2167430243A5056264201 @default.
• W2167430243 hasAuthorship W2167430243A5064563969 @default.
• W2167430243 hasAuthorship W2167430243A5065754348 @default.
• W2167430243 hasAuthorship W2167430243A5082387951 @default.
• W2167430243 hasAuthorship W2167430243A5083469637 @default.
• W2167430243 hasBestOaLocation W21674302431 @default.
• W2167430243 hasConcept C185592680 @default.
• W2167430243 hasConcept C190283241 @default.
• W2167430243 hasConcept C203522944 @default.
• W2167430243 hasConcept C2776055636 @default.
• W2167430243 hasConcept C2781000418 @default.
• W2167430243 hasConcept C31573885 @default.
• W2167430243 hasConcept C54355233 @default.
• W2167430243 hasConcept C86803240 @default.
• W2167430243 hasConcept C95444343 @default.
• W2167430243 hasConcept C98424977 @default.
• W2167430243 hasConceptScore W2167430243C185592680 @default.
• W2167430243 hasConceptScore W2167430243C190283241 @default.
• W2167430243 hasConceptScore W2167430243C203522944 @default.
• W2167430243 hasConceptScore W2167430243C2776055636 @default.
• W2167430243 hasConceptScore W2167430243C2781000418 @default.
• W2167430243 hasConceptScore W2167430243C31573885 @default.
• W2167430243 hasConceptScore W2167430243C54355233 @default.
• W2167430243 hasConceptScore W2167430243C86803240 @default.
• W2167430243 hasConceptScore W2167430243C95444343 @default.
• W2167430243 hasConceptScore W2167430243C98424977 @default.
• W2167430243 hasIssue "21" @default.
• W2167430243 hasLocation W21674302431 @default.

• next