Atpase–Regulated Form of Cell Death Triggered by Autophagy-Inducing Peptides, Starvation, and Hypoxia–Ischemia

+ + Autosis is a Na ,K -ATPase–regulated form of cell death triggered by autophagy-inducing peptides, starvation, and hypoxia–ischemia

Yang Liua,b, Sanae Shoji-Kawataa,b,1, Rhea M. Sumpter, Jr.a,b, Yongjie Weia,b,c, Vanessa Ginetd, Liying Zhange, Bruce Posnere, Khoa A. Tranf, Douglas R. Greeng, Ramnik J. Xavierh,i,j, Stanley Y. Shawf,j, Peter G. H. Clarked, Julien Puyald,2, and Beth Levinea,b,c,k,2

aCenter for Autophagy Research, Departments of bInternal Medicine, eBiochemistry, and kMicrobiology, and cHoward Hughes Medical Institute, University of Texas Southwestern Medical Center, Dallas, TX 75390; dDepartment of Fundamental Neurosciences, University of Lausanne, CH-1005 Lausanne, Switzerland; fCenter for Systems Biology, hCenter for Computational and Integrative Biology, and iGastrointestinal Unit, Massachusetts General Hospital, Harvard Medical School, Boston, MA 02114; jBroad Institute of Harvard and MIT, Cambridge, MA 02142; and gDepartment of Immunology, St. Jude’s Children’s Research Hospital, Memphis, TN 38105

This contribution is part of the special series of Inaugural Articles by members of the National Academy of Sciences elected in 2013.

Contributed by Beth Levine, October 26, 2013 (sent for review September 16, 2013)

A long-standing controversy is whether autophagy is a bona fide involution following lactation or prostate involution following cause of mammalian cell death. We used a cell-penetrating autoph- castration) require massive cell elimination (4–6). Autophagic agy-inducing peptide, Tat-Beclin 1, derived from the autophagy pro- cell death has also been described in diseased tissues and in cul- tein Beclin 1, to investigate whether high levels of autophagy result in tured mammalian cells treated with chemotherapeutic agents or – cell death by autophagy. Here we show that Tat-Beclin 1 induces other toxic compounds (4 6). “ ” dose-dependent death that is blocked by pharmacological or genetic The term autophagic cell death has been controversial, be- inhibition of autophagy, but not of apoptosis or necroptosis. This cause it has been applied to scenarios where evidence is lacking death, termed “autosis,” has unique morphological features, includ- for a causative role of autophagy in cell death (i.e., there is cell ing increased autophagosomes/autolysosomes and nuclear convolu- death with autophagy but not by autophagy). However, using more stringent criteria to define autophagic cell death, several tion at early stages, and focal swelling of the perinuclear space at late studies in the past decade have shown that autophagy genes are stages. We also observed autotic death in cells during stress condi- essential for cell death in certain contexts. This includes cases of tions, including in a subpopulation of nutrient-starved cells in vitro tissue involution in invertebrate development as well as in cul- and in hippocampal neurons of neonatal rats subjected to cerebral tured mammalian cells lacking intact apoptosis pathways (6, 7). – ∼ hypoxia ischemia in vivo. A chemical screen of 5,000 known bio- In apoptosis-competent cells, high levels of autophagy can also active compounds revealed that cardiac glycosides, antagonists of + + lead to autophagy gene-dependent, caspase-independent cell Na ,K -ATPase, inhibit autotic cell death in vitro and in vivo. Further- – fi + + death (8 10). In neonatal mice, neuron-speci c deletion of Atg7 more, genetic knockdown of the Na ,K -ATPase α1 subunit blocks protects against cerebral hypoxia–ischemia-induced hippocam- peptide and starvation-induced autosis in vitro. Thus, we have iden- pal neuron death (11), and in adult rats, shRNA targeting beclin tified a unique form of autophagy-dependent cell death, a Food and 1 decreases neuronal death in the thalamus that occurs sec- Drug Administration-approved class of compounds that inhibit such ondary to cortical infarction (12). death, and a crucial role for Na+,K+-ATPase in its regulation. These Although such studies provide genetic support for autophagy findings have implications for understanding how cells die during as a bona fide mode of cell death, the nature of autophagic cell certain stress conditions and how such cell death might be prevented. Significance he lysosomal degradation pathway of autophagy plays a cru- Tcial role in enabling eukaryotic cells to adapt to environmental We show that the selective overactivation of autophagy can stress, especially nutrient deprivation (1). The core autophagy cause cell death with unique morphological features distinct machinery was discovered in a genetic screen in yeast for genes from apoptosis or necrosis. This unique type of autophagic cell essential for survival during starvation, and gene knockout or death, termed “autosis,” occurs not only in vitro but also in knockdown studies in diverse model organisms provide strong vivo in cerebral hypoxia–ischemia. Moreover, autosis is inhibi- evidence for a conserved prosurvival function of autophagy during ted both in vitro and in vivo by cardiac glycosides, which are + + starvation (1). This prosurvival function of autophagy results from Na ,K -ATPase antagonists used in clinical medicine. Our its ability to mobilize intracellular energy resources to meet the findings contribute to the basic understanding of cell-death demand for metabolic substrates when external nutrient supplies mechanisms and suggest strategies for protecting cells against are limited. stresses such as hypoxia–ischemia. In contrast to this well-accepted, prosurvival function of autophagy, there has been much debate as to whether autophagy— Author contributions: Y.L., S.S.-K., R.M.S., B.P., D.R.G., R.J.X., S.Y.S., P.G.H.C., J.P., and B.L. — designed research; Y.L., S.S.-K., R.M.S., Y.W., V.G., K.A.T., and J.P. performed research; S.S.-K. especially at high levels also functions as a mode of cell death and K.A.T. contributed new reagents/analytic tools; Y.L., R.M.S., Y.W., V.G., L.Z., B.P., R.J.X., (2). Historically, based on morphological criteria, three types of S.Y.S., P.G.H.C., J.P., and B.L. analyzed data; and Y.L., L.Z., R.J.X., S.Y.S., P.G.H.C., J.P., programmed cell death have been defined: type I apoptotic cell and B.L. wrote the paper. death; type II autophagic cell death; and type III, which includes The authors declare no conflict of interest. necrosis and cytoplasmic cell death (3). Autophagic cell death Freely available online through the PNAS open access option. fi was originally de ned as a type of cell death that occurs without 1Present address: Department of Molecular Immunology and Inflammation, National chromatin condensation and is accompanied by large-scale auto- Center for Global Health and Medicine, Tokyo 162-8655, Japan. phagic vacuolization of the cytoplasm. This form of cell death, 2To whom correspondence may be addressed. E-mail: [email protected] or beth. first described in the 1960s, has been observed ultrastructurally in [email protected]. tissues where developmental programs (e.g., insect metamor- This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. phosis) or homeostatic processes in adulthood (e.g., mammary 1073/pnas.1319661110/-/DCSupplemental.

20364–20371 | PNAS | December 17, 2013 | vol. 110 | no. 51 www.pnas.org/cgi/doi/10.1073/pnas.1319661110 Downloaded by guest on September 25, 2021 death that occurs in mammalian cells and tissues in response to induces cell death in parallel with its ability to induce autophagy

physiological/pathophysiological stimuli remains poorly defined. in a dose- and time-dependent manner. INAUGURAL ARTICLE It is unclear whether cells that die by autophagy have unique We confirmed that Tat-Beclin 1 induced HeLa cell death by morphological features or a unique death machinery. The only detection of cells positive for Sytox Green (a nucleic dye ex- morphological feature that has been linked to autophagic cell cluded by live cells) (Fig. 1C), an increase of propidium iodide death—autophagic vacuolization—may be observed in cells un- (PI)-positive cells (Fig. S1 C and D), and a decline of cellular dergoing apoptotic or necrotic cell death, and no proteins, aside ATP levels (Fig. S1E). In addition, Tat-Beclin 1, but not Tat- from the core autophagy proteins, have been shown to be re- Scrambled, significantly reduced clonogenic survival (Fig. 1D). quired for autophagic cell death. Tat-Beclin 1 also induced cell death in a variety of additional Here we identify a form of autophagic cell death, autosis, which tumor cell lines, in human and rat fibroblasts, and in primary and + has unique morphological features; depends on the cellular Na , E1A/Ras-transformed murine embryonic fibroblasts (MEFs) + K -ATPase; and occurs during treatment with autophagy-inducing (Fig. S1F). peptides, starvation, and cerebral hypoxia–ischemia. We investigated whether Tat-Beclin 1-induced autophagy is mechanistically related to Tat-Beclin 1-induced cell death Results by using pharmacological and genetic approaches to inhibit Autophagy-Inducing Peptides Trigger Autophagy-Dependent Cell Death. autophagy. Treatment with 3-methyladenine (3-MA), an inhibitor Previously, we discovered a potent autophagy-inducing cell of class III PI3K activity and autophagosome formation, partially permeable peptide (13), Tat-Beclin 1, composed of 11 amino acids blocked Tat-Beclin 1-induced HeLa cell death, as measured in- of the HIV Tat protein transduction domain, a diglycine linker, creased cellular levels of ATP (Fig. S1G), a decreased percentage and 18 amino acids (267–284 aa) derived from the autophagy of trypan blue-positive cells (Fig. S1H), and increased clonogenic protein, Beclin 1. This peptide induced autophagy without cy- survival (Fig. 1E). siRNA knockdown of the essential autophagy totoxicity at low doses, but caused cell death at higher doses (13). gene, beclin 1, decreased autophagy in HeLa cells (Fig. S1I), This finding suggested that the Tat-Beclin 1 peptide might induce decreased Tat-Beclin 1-induced cell death (Fig. S1J), and in- a form of autophagy-dependent cell death and serve as a model creased clonogenic survival (Fig. 1F). Furthermore, doxycycline- for defining characteristics of autophagy-dependent cell death inducible shRNA knockdown of ATG13 or ATG14 in U2OS cells that occurs in pathophysiological settings. decreased Tat-Beclin 1-induced autophagy (Fig. S1K), protected We examined the relationship between Tat-Beclin 1-induced against Tat-Beclin 1-induced cell death (Fig. S1L), and increased autophagy and cell death. In HeLa cells, Tat-Beclin 1 led to clonogenic survival (Fig. 1G). Blockade of autophagosomal/

ﬁ CELL BIOLOGY dose-dependent induction of autophagy, as measured by ratios of lysosomal fusion by ba lomycin A1, a vacuolar proton ATPase light chain 3 (LC3)-II/I and degradation of the autophagy sub- inhibitor, did not reduce Tat-Beclin 1-induced cell death (Fig. strate, p62 (Fig. 1A), as well as cell death, as measured by trypan S1M), suggesting that this form of cell death does not require late blue exclusion (Fig. 1B). Increasing durations of exposure to a stages of autophagy. In addition, another autophagy-inducing α ﬁxed concentration of Tat-Beclin 1 resulted in a time-dependent peptide, Tat-vFLIP 2 (which acts by releasing ATG3 from cel- increase in autophagy and cell death (Fig. S1 A and B). No lular FLIP) (14), also induced autophagy (Fig. S1N) which was autophagy induction or cell death was observed after treatment associated with dose- and time-dependent cell death (Fig. S1 O with a control peptide, Tat-Scrambled (13). Thus, Tat-Beclin 1 and P) and reduced by ATG14 knockdown in U2OS cells (Fig. S1Q). Thus, autophagy-inducing peptides trigger cell death that requires the autophagy machinery.

A B 100% C T-S Autophagy Peptide-Induced Death Does Not Require Apoptotic or T-S T-B 80% *** Necropoptotic Machinery. We next asked whether the apoptosis Opti- MEM 20 5 10 20 30 (μM) and/or necroptosis death machinery is involved in Tat-Beclin 1- 60% LC3-I *** LC3-II induced cell death. We found that neither z-VAD, an inhibitor of T-B 40% caspases and apoptosis, nor necrostatin-1 (Nec-1), an inhibitor of p62 *** 20% RIPK1 kinase-mediated necroptosis, rescued Tat-Beclin 1-induced Actin

% Trypan blue-positive cells Trypan % cell death as measured by levels of cellular (Fig. S1G), trypan blue 0% 0 2003020 30 5 10 20 3020 30 (μM) exclusion (Fig. S1H), or clonogenic survival (Fig. S2A). Adeno- T-S T-B virus E1A/Ras-transformed MEFs with null mutations in the two D E -3-MA F NC siRNA G - Dox 120% NS 120% proapoptotic genes, Bax and Bak, were susceptible to Tat-Beclin 120% +3-MA 120% beclin 1 siRNA + Dox 100% 100% 100% 100% 1-induced cell death (Fig. 2A and Fig. S2B) but were resistant to *** 80% 80% *** 80% 80% death induced by the apoptosis-inducing agents, staurosporine 60% 60% 60% 60% ** and etoposide (Fig. S2C). Genetic deletion of two key regulators ** 40% 40% 40% 40% of necroptosis, Ripk1 and Ripk3, failed to protect primary MEFs NS *** 20% 20% 20% 20% from Tat-Beclin 1-induced cell death (Fig. 2B and Fig. S2D). Thus, Relative colony number Relative colony number Relative colony number Relative colony number 0% 0% 0% 0% neither the apoptotic nor necroptotic death machinery is required Opti- T-S T-B T-B - + T-B - + T-B - + - + - + MEM vector sh sh ATG13 ATG14 for Tat-Beclin 1-induced cell death. Additional assays confirmed the lack of apoptosis in Tat- Fig. 1. Tat-Beclin 1 induces autophagy-dependent cell death. (A) Western Beclin 1-induced death. In contrast to staurosporine, Tat-Beclin blot of LC3 and p62 in HeLa cells treated with Tat-Scrambled (T-S) or Tat- 1 did not activate caspase 3, as shown biochemically by the lack Beclin 1 (T-B) peptides for 5 h. (B) Cell death of HeLa cells treated T-S or T-B of cleavage of caspase 3 or its substrate PARP (Fig. 2C) and by for 5 h. (C) Representative images of Sytox Green staining of HeLa cells the lack of immunofluorescence staining for active caspase 3 treated with T-S or T-B (20 μM, 5 h). (Scale bar, 50 μm.) (D) Clonogenic cell μ (Fig. 2D). In addition, minimal pancaspase activity was detected survival of HeLa cells treated with T-S or T-B (20 M, 5 h). (E) Clonogenic in Tat-Beclin 1-treated cells by flow cytometry (Fig. S2 E and F). survival of HeLa cells treated with T-B (20 μM, 4 h) ± 10 mM 3-MA. (F) Clo- nogenic survival of siRNA-transfected HeLa cells treated with T-B (20 μM, Consistent with nonapoptotic cell death, no TUNEL staining 3h). (G) Clonogenic survival of doxycycline (Dox)-inducible U2OS/TR cells (Fig. S2G) or DNA ladder formation (Fig. S2H) was detected in ± μ Tat-Beclin 1-treated cells. We also confirmed that Tat-Beclin 1 stably transfected with empty vector, shATG13,orshATG14 Dox (1 g/mL) fl for 5 d before treatment with T-B (25 μM, 5 h). For B and D–G, error bars [derived from a structurally exible region in the Beclin 1 evo- represent mean ± SEM and similar results were observed in three in- lutionarily conserved domain (15)] did not exhibit a pore-form- dependent experiments. For D–G, the number of colonies in untreated ing ability to release cytochrome c from mitochondria (Fig. S2I), controls was standardized as 100%. NS, not significant; **P < 0.01; ***P < as do certain other amphipathic α-helical peptides (16). Fur- 0.001; t test. See also Fig. S1. thermore, antioxidants that block reactive oxygen species-mediated

Liu et al. PNAS | December 17, 2013 | vol. 110 | no. 51 | 20365 Downloaded by guest on September 25, 2021 Tat-Scrambled Tat-Scrambled internal structure (clumps instead of bands); the ER is dilated A Tat-Beclin 1 B Tat-Beclin 1 C D Anti-cleaved 120% NS caspase 3 NS NS NS and fragmented; and numerous autophagosomes, autolysosomes, 120% Tat-Beclin 1 100% 100% and empty vacuoles are present. In phase 1b, the perinuclear 80% 80% space becomes swollen at discrete regions surrounding the inner Cleaved * 60% 60% caspase 3 nuclear membrane, and these swollen areas contain membrane- Staurosporine 40% 40% Cleaved PARP bound regions with a density and granularity resembling the 20% 20% Actin Relative colony number cytosol. In some cases, the perinuclear space extends through Relative colony number 0% 0% WT Bax-/-; WT Bax-/-; WT Ripk1+/+; Ripk1-/-; -/- -/- substantial distances in the cytoplasm. In phase 2, there is focal BakBak -/- Bak Ripk3-/- Ripk3-/- 20 M 30 M ballooning of the perinuclear space (which appears empty), of- ten associated with a concavity of the nuclear surface. At this late Fig. 2. Tat-Beclin 1-induced cell death does not require the apoptotic or necroptotic machinery. (A) Clonogenic survival of wild-type and Bax−/−;Bak−/− stage, the morphology appears necrotic; mitochondria and other MEFs treated with peptide (5 h). (B) Clonogenic survival of wild-type, Ripk1+/+; organelles are swollen; and autophagosomes, autolysosomes, and − − − − − − Ripk3 / ,andRipk1 / ;Ripk3 / MEFs treated with peptide (20 μM, 5 h). In A ER are rare. There appears to be lysis of the plasma membrane and B, the number of colonies of Tat-Scrambled-treated cells was standardized (which is difficult to discern from EM analyses but is sub- as 100%. (C) Western blot of cleaved caspase 3 and cleaved PARP in HeLa stantiated by PI staining, Sytox Green staining, and live-cell cells treated with 20 μM Tat-Scrambled (T-S), 20 μM Tat-Beclin 1 (T-B), 1 μM imaging of Tat-Beclin 1-treated cells). ± μ staurosporine (STS) 100 M Z-VAD-FMK (z-VAD), or 32 mM H2O2 for 5 h. These ultrastructural changes are distinct from previous clas- The asterisk denotes a cross-reacting band. (D) Representative images of sifications of cell death (3), including type-3B cytoplasmic death cleaved caspase 3 staining in HeLa cells treated with 20 μM Tat-Beclin 1 or (also called “paraptosis”), which also has perinuclear swelling. 1 μM staurosporine for 5 h. Scale bar, 50 μm. For A and B, error bars rep- In type-3B cell death, perinuclear swelling is moderate and uni- resent mean ± SEM of triplicate samples and similar results were observed in fi form around the entire nuclear perimeter, whereas in death of three independent experiments. NS, not signi cant; t test. See also Fig. S2. autophagy-inducing peptide-treated cells, there is a pronounced ballooning in a focal region of the perinuclear space. To avoid “ ” cell death failed to rescue Tat-Beclin 1-induced cell death confusion with terms such as autophagic cell death (which is (Fig. S2J). Similar to Tat-Beclin 1, Tat-vFLIP α2 also failed to sometimes applied to states in which it is not clear that autophagy K is required for cell death and/or in which autophagic features induce caspase 3 or PARP cleavage (Fig. S2 ). Thus, taken “ ” together, our data indicate that autophagy-inducing peptide- coexist with apoptosis or necrosis), we coined the term autosis fi triggered cell death is genetically and biochemically distinct to de ne cell death mediated by autophagy genes and charac- from apoptosis or necroptosis. terized by focal perinuclear swelling. Although several studies have described cell death that is blocked by genetic inhibition of Autophagy Peptide-Induced Cell Death Has Unique Morphological Features. To characterize the nature of this autophagy-dependent, nonapoptotic, and nonnecrotic cell death, we performed live-cell imaging of Tat-Beclin 1-treated cells (Fig. 3A and Movie S1). A B T-S T-B During the initial phase, cells exhibit relatively normal morphol- 0:30 1:00 2:00 3:00 3:50 ogy with increased vacuolar dynamics and perinuclear accumulation of numerous vacuoles. After a few hours, cells undergo an Tom20 Lamin A/C abrupt demise (lasting ∼15–20 min) characterized by the rapid 3:55 4:00 4:05 4:10 4:15 shrinkage of the nucleus with a portion of its surface developing a concave appearance corresponding to a round, vacuole-like PDI Lamin A/C entity, reflecting (based on our EM data; see below) a local sep-

aration of the inner and outer nuclear membranes. This is followed LAMP1 Lamin A/C by focal plasma membrane rupture and extracellular extrusion C Tat-Beclin 1 of cytoplasmic contents. Cells treatedwithTat-Beclin1display Tat-Scrambled Phase 1a Phase 1b Phase 2 ﬁ increased substrate adherence that persists until their nal demise N ﬂ N N (unlike apoptotic or necrotic cells, which generally oat). N N The concave nuclear appearance observed in Tat-Beclin 1- induced cell death is associated with abnormalities in nuclear

lamin-A/C staining (lack of a uniform circular appearance and N N N PNS focal regions of dense staining) (Fig. 3B and Fig. S3A). Tat- INM Beclin 1-treated dying cells also exhibit an abnormal fragmented ONM pattern of Tom20 (mitochondrial marker) and PDI [endoplasmic reticulum (ER) marker] staining, and a striking increase in expression of LAMP1, a marker of late endosomes/autolyosomes (which would be expected in the setting of a robust autophagy response) (Fig. 3B). Tat-vFLIP α2-treated dying cells have a similar concave nuclear appearance and similar abnormalities in N lamin-A/C, Tom20, PDI, and LAMP1 staining (Fig. S3B). We performed ultrastructural analyses to further characterize Fig. 3. Morphological features of Tat-Beclin 1-induced autosis. (A) Repre- the morphology of Tat-Beclin 1-induced death (Fig. 3C and Fig. sentative live-cell imaging of HeLa cells treated with 25 μM Tat-Beclin 1 for S3C). As apparent from live-cell imaging, there were two phases 5h(Movie S1; times shown as hh:mm). The black arrow denotes released of the death process; phase 1 is characterized by a slow phase of intracellular components from a ruptured cell membrane and the white arrow denotes perinuclear space between the inner nuclear membrane and gradual change and phase 2 is characterized by an abrupt phase μ ﬁ cytoplasm at a region of nuclear concavity. (Scale bar, 10 m.) (B) Repre- of nal collapse and cell death. Morphologically, phase 1 can be sentative images of mitochondrial (Tom20), ER (PDI), late endosome/lyso- divided into two stages. In phase 1a, the nucleus becomes con- some (LAMP1), and nuclear lamin-A/C staining in HeLa cells treated with Tat- voluted (but the perinuclear space is normal); chromatin is Scrambled (T-S) or Tat-Beclin 1 (T-B) (20 μM, 5 h). (Scale bar, 20 μm.) (C)EM moderately condensed, forming darker regions in the nucleus analysis of HeLa cells treated with peptide (20 μM, 5 h). White arrows show with borders that are fuzzy (in contrast to clumps of chromatin dilated and fragmented ER; black arrows show regions where the peri- that typically have sharp borders in apoptosis); many of the mi- nuclear space has swollen and contains clumps of cytoplasmic material. tochondria are electron dense and some have an abnormal (Scale bars, 1 μm.) See also Fig. S3.

20366 | www.pnas.org/cgi/doi/10.1073/pnas.1319661110 Liu et al. Downloaded by guest on September 25, 2021 autophagy (6), such studies have not described increased sub- identiﬁed by Sytox-Green staining (Fig. S4A), and also displays

strate adherence and focal perinuclear swelling of dying cells. a marked increase (approximately threefold) in numbers of INAUGURAL ARTICLE Thus, autosis represents a previously undescribed form of cell autophagosomes (GFP-LC3 puncta) compared with numbers in death by autophagy. the majority population of starved cells that float and undergo apoptosis (Fig. 4 B and C). These substrate-adherent cells have Starvation Induces Autosis. We next investigated whether autosis nuclei with concave regions and focal swelling of the perinuclear occurs during physiological stress conditions associated with high space (Fig. 4 A, B, and D and Fig. S4A) and display similar ab- levels of autophagy. Nutrient starvation is a potent physiological normalities in lamin-A/C staining as observed in Tat-Beclin 1 inducer of autophagy in eukaryotic cells, and previous studies and Tat-vFLIP α2 peptide-treated cells (Fig. S4A). Similar to have shown that autophagy delays apoptosis in cells subjected to Tat-Beclin 1 peptide treatment, we observed phase-1 starved starvation, including HeLa cells (1). However, upon subjecting cells with increased autophagosomes/autolysosomes, and regions HeLa cells to amino-acid and serum starvation, we found that, of perinuclear swelling containing organelles and phase-2 cells unlike the vast majority of cells which detach from their substrate with rare autophagosomes/autolysosomes and dilated empty and undergo apoptosis (as evidenced by active caspase 3 stain- regions in the perinuclear space (Fig. 4D). We also observed ing), a small subpopulation (∼1%) of cells become more sub- features of autosis in starved adherent murine bone marrow- strate-adherent and lack evidence of caspase 3 activation (Fig. derived macrophages (BMDMs) and primary MEFs (Fig. 4E), 4A). This population of substrate-adherent, caspase 3-negative suggesting that starvation-induced autosis occurs in primary cells cells undergoes plasma-membrane rupture and cell death, as and is not merely a consequence of mutations that confer re- sistance to apoptosis in continuous cell lines. Moreover, the frequency of autotic cell death was higher (∼5%) in primary cells than in HeLa cells. Normal Starvation Starvation Normal Starvation Starvation fi A media (floating) (adherent) B media (floating) (adherent) To con rm that autophagy is required for starvation-induced death in cells with autotic morphology, we assessed the effects of DIC DIC autophagy gene knockdown on the clonogenic survival of substrate-adherent starved cells. In this assay, floating (apoptotic and necrotic) cells were washed away after 48 h (HeLa cells) or Active caspase 3 GFP- 72 h (U2OS cells and BMDMs) starvation, and the clonogenic LC3 DAPI potential of the remaining adherent cells was assessed (Fig. S4B).

Both ATG7 and beclin 1 siRNA treatment (Fig. S4C) increased CELL BIOLOGY C D colony numbers formed by starved adherent HeLa cells (Fig.

*** 4F), and ATG14 shRNA expression (Fig. S4D) increased colony 120 *** numbers formed by starved adherent U2OS cells (Fig. 4G). Ly-

100 Phase 1 80 sozyme:Cre-mediated deletion of Atg5 (Fig. S4E) also increased ﬂ ﬂ 60 colony numbers formed by starved adherent Atg5 ox/ ox BMDMs 40 *** 20 (Fig. 4H). [ATG7 siRNA, beclin 1 siRNA, ATG14 shRNA and # GFP-LC3 dots/cell 0 Atg5 deletion had minimal effect on the clonogenic survival of PNS – Phase 2 cells cultured in normal media (Fig. S4 C E)]. Thus, autophagy genes are required for starvation-induced autosis. ONM INM

BMDM MEF E F G H *** – 80 400 *** 300 NS Autosis Occurs During Rat Cerebral Hypoxic Ischemic Injury. After *** - Dox + Dox DIC 60 300 establishing ultrastructural criteria for autosis in cultured cells, 200 we evaluated whether autosis occurs in vivo. We performed EM 200 40 ** – 100 analysis of neuronal death following cerebral hypoxia ischemia # colonies # colonies 20 # colonies 100 NS Sytox NS in the brains of neonatal rats (Fig. 5); we focused on dying neurons green 0 0 0 NC ATG7 beclin 1 U2OS/ U2OS/ Mouse 1 2 1 2 in the hippocampus CA3 region because we had previously shown siRNA siRNA siRNA TR shATG14 Atg5fl/fl; Atg5fl/fl; Cre- Cre+ that most of these neurons degenerate with autophagic features (from 6 h after hypoxia–ischemia) without signs of apoptosis or Fig. 4. Starvation induces autosis. (A) Representative images of active cas- necrosis (17). At 24 h after cerebral hypoxia–ischemia, most of pase-3 staining in HeLa cells 48 h after starvation (HBSS). (Center) Active caspase 3-positive floating cells with rounded nuclei. (Right) Active caspase the dying neurons displayed prominent autophagic features, such 3-negative adherent cell with concave nucleus and swollen perinuclear as numerous autophagosomes and autolysosomes, and empty μ vacuoles (phase 1a). Strikingly, some dying neurons displayed space. (Scale bar, 20 m.) (B and C) Representative images (B) and quanti- — tation (C) of GPF-LC3 dots (autophagosomes) in HeLa/GFP-LC3 cells (>50 cells characteristics of phase-1b or the full phase-2 features of autosis analyzed per sample) grown in normal medium or in floating and adherent focal ballooning of the perinuclear space associated with nuclear HeLa/GFL-LC3 cells 6 h after starvation. (Scale bar, 10 μm.) (D)(Upper)EM concavity. Thus, autotic cell death occurs in certain pathophysi- images of phase-1 substrate-adherent HeLa cell 6 h after starvation. (Lower) ological settings in vivo. CLEM images of phase-2 substrate-adherent HeLa cell with concave nucleus and swollen perinuclear space (PNS) (arrow) 8 h after starvation. (Lower Left) A High-Throughput Chemical Screen Identifies Cardiac Glycosides as Phase contrast microscopy; (Lower Center and Lower Right) EM of same cell. Potent Inhibitors of Autosis. To gain insight into the regulation of The black arrow in Right Lower shows outer nuclear membrane (ONM) and autosis, we performed high-throughput compound screening to the white arrow shows inner nuclear membrane (INM). (Scale bars, 1 μm.) (E) identify inhibitors of Tat-Beclin 1-induced cell death, focusing on Representative images of a Sytox Green-positive adherent primary murine compound libraries consisting of bioactive agents with known μ BMDM and MEF 48 h after starvation. (Scale bar, 10 m.) (F) Clonogenic targets. We measured levels of cellular ATP (as a proxy of cel- survival of siRNA-transfected adherent HeLa cells starved for 48 h. NC, lular viability) 5 h after Tat-Beclin 1 treatment of HeLa cells in nontargeting control siRNA. (G) Clonogenic survival of doxycycline (Dox)- ∼ inducible adherent U2OS/TR and U2OS/shATG14 cells ± Dox treatment (1 μg/ the presence of 5,000 Food and Drug Administration (FDA)- mL) for 5 d before starvation for 72 h. (H) Clonogenic survival of adherent approved drugs and bioactive compounds with characterized fl/fl − fl/fl + mechanisms of action (Fig. S5A). We chose for further analysis BMDMs (two mice per genotype; Atg5 ;Lyz-Cre and Atg5 ;Lyz-Cre lit- ≥ termates) starved for 72 h. For C and F–H, error bars represent mean ± SEM the 36 top hits that had z scores 3.0 in the primary screen (Fig. of triplicate samples and similar results were observed in three independent 6A and Dataset S1). These 36 hits were classified into nine fam- experiments. For A, B and E, arrows denote concave nucleus and swollen ilies based on their chemical structures and/or biological functions perinuclear space. NS, not significant; **P < 0.01; ***P < 0.001; t test. See (Dataset S2). Of these 36 hits, eight compounds demonstrated also Fig. S4. >40% rescue of autosis in a repeat ATP assay and were chosen

Liu et al. PNAS | December 17, 2013 | vol. 110 | no. 51 | 20367 Downloaded by guest on September 25, 2021 Sham 24 h Hypoxia-Ischemia 6E). Thus, cardiac glycosides rescue cell death triggered by mul- Phase 1a Phase 1b Phase 2 tiple inducers of autosis. Digoxin partially reversed the majority of morphological ab- PNS normalities in cells undergoing Tat-Beclin 1-induced autosis. N N N N Upon light microscopy analysis, cells treated with Tat-Beclin 1 N and digoxin displayed minimal nuclear abnormalities and lacked N abnormal patterns of mitochondrial (Tom20), ER (PDI), late N N PNS endososome/lysosome (LAMP1), and nuclear lamin-A/C staining (Fig. S6A). Ultrastructurally, the majority of digoxin-rescued

ER N cells had a normal-shaped nuclear membrane without any focal INM M swelling of the perinuclear space, intact ER structure, and the ONM N GA PNS absence of increased numbers of autophagosomes and autolysosomes (Fig. 6F). The only morphological abnormality of autosis not reversed by digoxin in Tat-Beclin 1-treated cells was the pres- Fig. 5. Morphological features of cerebral hypoxia–ischemia-induced autosis. ence of electron-dense mitochondria; however, digoxin alone (in the EM analysis of dying neurons in hippocampal region CA3 in brains of 7-d-old absence of Tat-Beclin 1) resulted in electron-dense mitochondria rats 24 h after exposure to cerebral hypoxia–ischemia. Arrows show regions (Fig. S6B). Together, these data indicate that digoxin reverses where the perinuclear space is swollen and contains clumps of cytoplasmic the morphological changes of autosis except for mitochondrial material. (Scale bars, 1 μm.) GA, Golgi apparatus; INM, inner nuclear mem- abnormalities, but these do not appear to be related to cell death. brane; M, mitochondrion; N, nucleus; ONM, outer nuclear membrane; PNS, We performed Western blot analyses of LC3 and p62 and perinuclear space. quantitation of GFP-LC3 puncta to further evaluate the effects of digoxin on autophagy. Under basal conditions, consistent with for further analysis (Dataset S3). Of these eight compounds, only prior reports (20, 21), we observed a dose-dependent increase in ﬁve, including three cardiac glycosides (digoxin, digitoxigenin, LC3-II conversion and mild reduction in p62 levels (although we and strophanthidin) and two purinergic receptor antagonists did not detect an increase in GFP-LC3 puncta) (Fig. 6G and Fig. (suramin and NF 023) demonstrated more than 80% rescue of Tat-Beclin 1 peptide-induced cell death as measured by Sytox Green staining (Fig. S5B). The purinergic receptor antagonists, - Tat-Beclin 1 A B C Tat-Beclin 1 but not the cardiac glycosides, blocked cellular peptide entry 140% +

10 Cardiac inducers inducers inducers Necrosis *** Apoptosis glycosides Autophagy *** (Fig. S5 C and D) and were not studied further. Thus, our chemical top 120% *** 36 hits screen identiﬁed cardiac glycosides as the only class of agents that 5 100% Z = 3.0 80% inhibited the Tat-Beclin 1-induced cell death without blocking 0 cellular peptide entry. Z-score 60% -5 To assess the signiﬁcance of the effects of cardiac glycosides 40% 20%

-10 Relative colony number on autosis in an unbiased manner, we tested whether the nine Compound library 0% cardiac glycosides present in our library were statistically enriched NES 2.38 -1.71 -1.02 -0.99 -4 among top-scoring compounds. We calculated a weighted Kol- p <10 0.001 0.44 0.46 D E F G 100 mogorov–Smirnov-like statistic, the normalized enrichment score 80 ** 80 - Digoxin ** + Digoxin *** (NES), using compound set enrichment analysis (CSEA). CSEA 80 60 60 *** demonstrated strong, highly significant enrichment for cardiac gly- *** −4 60 *** < 40 40 ** cosides (P 10 )(Fig.6B). In addition to the three cardiac gly- *** 40 # Colonies ≥ fi # Colonies cosides with z scores 3.0 in the primary screen, we con rmed 20 20 20 that other cardiac glycosides in the compound libraries also # GFP-LC3 dots/cell fi 0 0 exhibited a signi cant rescue effect (Dataset S4), as did a cardiac Digoxin - + Neriifolin - + 0 glycoside, neriifolin, which was not in the compound libraries and Bafilomycin A1 - + - + + Normal HBSS Tat-Beclin 1 is known to exert neuroprotective actions (18) (Fig. S5E). We also medium performed CSEA using other previously identified compound sets for autophagy inducers, specific necrosis inducers, and specific Fig. 6. Cardiac glycosides rescue autosis. (A) Ranked distribution of z scores apoptosis inducers (19); none of these sets were enriched among for each compound in primary chemical screen (Dataset S1) for inhibitors of ≥ our top-scoring compounds (Fig. 6B). Although there are previous Tat-Beclin 1-induced cell death. Thirty-six top hits with z 3.0 (Dataset S2) reports that cardiac glycosides may increase basal autophagy (20, were selected for a secondary screen (Dataset S3). (B) Comparison of CSEA of cardiac glycosides in the primary autosis screening with compound sets of 21), an extensive set of autophagy inducers drawn from the lit- specific autophagy, necrosis, or apoptosis inducers. P, permutation P value erature were overrepresented among compounds that enhanced, for the NES compared with a null distribution. Red–blue vertical bars rep- rather than rescued, Tat-Beclin 1-induced cell death. The lack of resent list of screened compounds, ranked according to z score (greatest concordance between compounds that inhibited autophagy, apo- rescue of autosis at top). Each horizontal line indicates where a specific ptosis, or necrosis with the rescue of Tat-Beclin 1-induced autosis compound falls within ranked compound list. (C) Clonogenic survival of is consistent with the latter representing a distinct death process. HeLa cells treated with Tat-Beclin 1 (20 μM, 5 h) + 5 μM digoxin, digitoxi- Consistent with these bioinformatics analyses, the cardiac gly- genin, or strophanthidin. The number of colonies of untreated cells was coside, digoxin, had no effect on apoptotic death induced by standardized as 100%. (D and E) Clonogenic survival of adherent HeLa cells starved for 48 h ± 10 nM digoxin (D) or 1 nM neriifolin (E). (Nanomolar staurosporine or necrotic death induced by H2O2 (Fig. S5 F and G), whereas digoxin rescued Tat-Beclin 1–induced cell concentrations were used as toxicity of digoxin and neriifolin was observed μ during starvation with micromolar concentrations.) (F) Representative EM death with IC50 values below 0.1 M(Fig. S5H). Digoxin also μ + μ rescued Tat-vFLIP α2-induced death in HeLa cells (Fig. S5H) images of a HeLa cell treated with 20 M Tat-Beclin 1 5 M digoxin (5 h). (Scale bar, 1 μm.) (G) Quantitation of GFP-LC3 dots (>100 cells analyzed per and Tat-Beclin 1-induced cell death in U2OS cells (Fig. S5I). We sample) in HeLa/GFP-LC3 cells treated with 20 μM Tat-Beclin 1 or starved in also found that clonogenic survival was rescued in Tat-Beclin 1- HBSS for 2 h ± 0.1 μM digoxin and/or 20 nM bafilomycin A1. For C–E and G, treated HeLa cells by digoxin, digitoxigenin, and strophanthidin error bars represent mean ± SEM of triplicate samples and similar results (Fig. 6C); in Tat-vFLIP α2-treated HeLa cells by digoxin (Fig. were observed in three independent experiments. NS, not significant; S6J); and in the adherent subpopulation of HeLa cells subjected *P < 0.05; **P < 0.01; ***P < 0.001; t test. See also Figs. S5 and S6 and to prolonged starvation by digoxin (Fig. 6D) and neriifolin (Fig. Datasets S1–S4.

20368 | www.pnas.org/cgi/doi/10.1073/pnas.1319661110 Liu et al. Downloaded by guest on September 25, 2021 100 NC siRNA animals 1 wk after cerebral hypoxia–ischemia (Fig. 8 A and B). A NaK- 1 siRNA B C *** 50% ** 250% *** This effect was particularly notable in the hippocampus (Fig. 8B INAUGURAL ARTICLE 80 *** *** *** ** *** *** 40% 200% and Fig. S8A), where significant neuronal pathology, especially in 60 * *** *** 30% 150% the CA3 region, was detected as early as 24 h after hypoxia– 40 NS 20% 100% * ischemia injury (Fig. S8B). The CA3 region of the hippocampus 20 NS * 10% 50% – # GFP-LC3 dots/cell was protected at 24 h and 7 d after neonatal hypoxia ischemia by Relative colony number 0 Relative colony number 0% 0% neriifolin treatment compared with vehicle-treated pups (Fig. S8 A Bafilomycin A1 - + - + - + NC 1 2 3 NC 1 2 3 siRNA Normal NaK- 1 siRNA siRNA NaK- 1 siRNA B HBSS Tat-Beclin 1 and ). In parallel with this neuroprotection, neriifolin prevented medium the increase in autophagy in the CA3 region of the hippocampus + + Fig. 7. Na ,K -ATPase regulates autosis. (A) Quantitation of GFP-LC3 dots that occurred after hypoxia–ischemia injury, as measured by de- (>100 cells analyzed per sample) in HeLa/GFP-LC3 cells 72 h after transfection tection of decreased numbers of endogenous LC3 puncta and with indicated siRNA and treatment with Tat-Beclin 1 (20 μM, 2 h) or star- LAMP1 puncta by immunofluorescence and immunoperoxidase vation (HBSS, 2 h) ± 20 nM bafilomycin A1. (B) Clonogenic survival of HeLa staining (Fig. 8C and Fig. S8C) and decreased levels of LC3-II cells transfected with indicated siRNA for 72 h and then treated with Tat- (Fig. 8 D and E). Strikingly, in contrast to the characteristic μ Scrambled or Tat-Beclin 1 (20 M, 5 h). Shown are the percentage of colonies features of autosis (numerous autophagosomes, autolysosomes, in Tat-Beclin 1– vs. Tat-Scrambled–treated cells for each siRNA. (C) Clono- genic survival of adherent HeLa cells transfected with indicated siRNA for and empty vacuoles; abnormal mitochondria and ER; and focal 72 h, and then starved for 48h. Clonogenic survival of control siRNA trans- separation of the inner and outer nuclear membrane) observed fected cells in starvation conditions relative to normal medium standardized in the CA3 region of vehicle-treated pups 24 h after cerebral as 100%. For A–C, error bars represent mean ± SEM of triplicate samples and hypoxia–ischemia, the CA3-region neurons of neriifolin-treated similar results were observed in three independent experiments. NS, not animals displayed no ultrastructural features associated with significant; *P < 0.05; **P < 0.01; ***P < 0.001; t test. See also Fig. S7. autosis (Fig. 8F). Thus, cardiac glycosides block the increase in autophagy and protect hippocampal neurons against cerebral hypoxia–ischemia-induced autosis in vivo. S6C). In contrast, doses as low as 100 nM resulted in a mild decrease in starvation-induced autophagy and a more dramatic decrease in Tat-Beclin 1-induced autophagy (Fig. 6G and Fig. S6C). The increased p62 accumulation was not due to changes in A vehicle B C LC3-NeuN-Hoechst LAMP1-MAP2-Hoechst

p62 mRNA expression (Fig. S6D). Although cardiac glycosides CELL BIOLOGY have been reported to induce apoptosis (22, 23), we did not 100% vehicle

neriifolin sham observe caspase activation in Tat-Beclin 1-treated cells in the 80% presence of digoxin (Fig. S6E). 60% neriifolin 40% *** *** Na+,K+-ATPase Regulates Autosis. Cardiac glycosides are inhibitors (% contralateral) 20% 0% + + Intact ipsilateral volume Total Hippocampus of Na ,K -ATPase, a plasma membrane pump that generates hemisphere

+ + 24h HI Na and K gradients across the membrane and acts as a versa- +

tile signal transducer (22). We therefore examined whether Na , neriifolin vehicle + K -ATPase regulates autosis using siRNA knockdown of the α1 + + subunit of Na ,K -ATPase. Similar to digoxin treatment, Na,K- F N N ER D 24 h HI N α1–subunit knockdown (Fig. S7 A and B) resulted in a mild sham vehicle neriifolin

sham N M decrease in starvation-induced autophagy and a more dramatic LC3-I N decrease in Tat-Beclin 1-induced autophagy (Fig. 7A and Fig. LC3-II N S7A). The observed increase in p62 accumulation was not due to Tubulin N N N N

changes in p62 mRNA expression (Fig. S7C) or to a block in Phase 1 N peptide delivery into cells (Fig. S7D). In parallel with inhibition N α E of autophagy, three different siRNAs against Na,K- 1 inhibited NS N 300% vehicle N ** * PNS ONM Tat-Beclin 1- and Tat-vFLIP α2-induced death (Fig. S7E). They 250% N INM Phase 2 24 h HI PNS (% sham) 200% also increased clonogenic survival of Tat-Beclin 1-treated cells M 150% (Fig. 7B) and adherent cells subjected to starvation (Fig. 7C). 100% N 50% N ER Na,K-α1 siRNA also exerted a protective effect against autosis in N

LC3-II/tubulin 0% sham vehicle neriifolin F G M human U2OS (Fig. S7 and ) and in mouse NIH 3T3 (Fig. S7 24 h HI neriifolin N H and I) cells. Digoxin did not enhance Na,K-α1 siRNA-mediated protection against autosis triggered by autophagy-inducing – + + Fig. 8. Neonatal hypoxic ischemic brain damage and hippocampal CA3 peptides (Fig. S7J), suggesting that digoxin and Na ,K -ATPase region autophagy and autosis are reduced by treatment with the cardiac inhibition block autosis through the same mechanism. glycoside, neriifolin. (A) Representative Nissl-stained coronal sections through the brain showing the neuroprotective effect of neriifolin (Lower) Cardiac Glycoside-Mediated Protection Against Neuronal Autosis in compared with vehicle (Upper) 1 wk after hypoxia–ischemia (HI). (Scale bar, Rat Cerebral Hypoxia–Ischemia. A previous chemical screen to 1 mm.) (B) Volumes of intact tissue ipsilaterally compared with con- identify compounds that provide neuroprotection in a mouse tralaterally 1 wk after neonatal cerebral hypoxia–ischemia and indicated brain slice-based model for ischemic stroke revealed neriifolin as treatment. Values are mean ± SD (n = 6 for neriifolin and n = 9 for vehicle). a strong hit, and whole-animal studies have shown that neriifolin ***P < 0.001; Welch’s ANOVA test. (C) Representative confocal microscopy images of LC3 dots (red) and LAMP1 dots (green) in CA3 hippocampal and other cardiac glycosides provide neuroprotection in neonatal – models of cerebral hypoxia–ischemia (18, 24, 25). Given these neurons after 24h hypoxia ischemia and indicated treatment or sham op- observations, coupled with our findings described above that rat eration. NeuN (green) and MAP2 (red) are neuronal markers. Hoechst staining (blue) shows cell nuclei. (Scale bars, 20 μm.) (D and E) Representative hippocampal CA3 region neurons die by autosis following hyp- LC3 immunoblots (D) and quantification of LC3-II/tubulin levels (E) from oxia–ischemia, we evaluated whether neriifolin could protect – – immunoblots of hippocampi of rats subjected to hypoxia ischemia. Values neonatal rats against cerebral hypoxia ischemia and reduce are mean ± SD (n = 6 for neriifolin and n = 9 for vehicle). NS, not significant; autosis in the hippocampal region CA3. *P < 0.05; **P < 0.001; Kruskal–Wallis test. (F) EM analysis of neriifolin In agreement with results in mice (18), we found that neriifolin effects in hippocampal region CA3 of 7-d-old rats 24 h after hypoxia–is- was highly neuroprotective in rats; it dramatically increased the chemia. INM, inner nuclear membrane; M, mitochondrion; N, nucleus; ONM, volume of intact tissue in the ipsilateral hemisphere of neonatal outer nuclear membrane; PNS, perinuclear space. (Scale bars, 1 μm.)

Liu et al. PNAS | December 17, 2013 | vol. 110 | no. 51 | 20369 Downloaded by guest on September 25, 2021 Discussion predominantly in brain is sensitive, whereas the rodent-α1 subunit Our findings identify a unique form of autophagic cell death— expressed in many peripheral tissues is resistant (28). autosis—that meets two essential criteria put forth by the No- Although we are not aware of previous reports of similar nu- menclature Committee on Cell Death (2): suppression by in- clear morphological abnormalities in autophagic cell death, the hibition of the autophagic pathway, and lack of features of expression of sterol reductases that are localized to the ER and apoptosis and necrosis. The form of death that we observed in outer nuclear membrane, TM7SF2 and DHCR1, results in mas- adherent cells subjected to starvation and in cells treated with sive ER and perinuclear space expansion resembling that observed autophagy-inducing peptides not only meets these criteria for in autotic cells (29). These observations suggest that disruption autophagic cell death, but also has a distinctive morphological of ER/outer nuclear membrane cholesterol metabolism may and chemical inhibition signature. In addition to the classical produce the phenotype of ER and focal perinuclear space ex- morphological criteria of autophagic cell death (increased auto- pansion. This phenotype is possibly caused by alterations of ER- lysosomes in dying cells lacking features of apoptosis and necrosis), membrane properties, including transport or channel conductance, death induced in starved adherent cells and by autophagy-inducing which would result in osmotic changes and disruption of sig- peptides is accompanied by ER dilation and stereotypic nuclear naling through the nuclear envelope. Given the crucial role of changes, involving an early convoluted appearance, the forma- the ER in autophagosomal biogenesis (30), we speculate that tion of focal concave regions of the nucleus with surrounding stimulation of very high levels of autophagy may perturb normal focal swelling of the perinuclear space, and the accumulation of ER membrane biogenesis/homeostatic mechanisms, leading to structures within this space at early stages of the process. This similar expansions of the ER lumen and perinuclear space. form of cell death, but not apoptosis or necrosis, is also selec- Further studies are needed to investigate the underlying mech- + tively blocked by pharmacological and genetic inhibition of Na , anisms of the morphological abnormalities observed in autosis. + K -ATPase. Cardiac glycosides, a large family of naturally derived steroidal compounds, were first described for the treatment of heart dis- Although the underlying mechanisms of the morphological + + + + eases in 1785 (22). Approximately 50 y ago, Na ,K -ATPase was changes of autosis and the pathway by which Na ,K -ATPase fi mediates autosis remain to be determined, the discovery of this identi ed as the cellular target of cardiac glycosides. This mem- unique morphological and chemical inhibition signature has brane protein uses energy from ATP hydrolysis to facilitate the transport of potassium ions into cells and sodium ions out of important biological implications. Our findings pave the road to + + cells; inhibition of Na ,K -ATPase results in an increase in in- the discovery of physiological and pathophysiological conditions tracellular sodium and calcium ions. Cardiac glycosides also have in which autophagy functions as a death mechanism, and may diverse effects on cellular signaling, proliferation, metabolism, provide a candidate treatment for diseases in which such death survival, gene expression, attachment, and protein trafficking. contributes to pathogenesis. For example, using the morpho- Our chemical screen revealed cardiac glycosides as the most logical criteria we established for autosis in cells subjected to potent inhibitors of autotic cell death, and we found that they starvation or treatment with autophagy-inducing peptides, we inhibited both autophagy and autotic cell death in the setting of identified the presence of autotic death in rat hippocampal – starvation, autophagy-inducing peptide treatment, and neonatal neurons subjected to hypoxic ischemic injury. Moreover, we hippocampal hypoxic–ischemic injury. The mechanism of action — + showed that a class of FDA-approved chemical compounds appears to be inhibition of the target of cardiac glycosides, Na , — + + + cardiac glycosides that inhibited autosis in an in vitro chemical K -ATPase, as we observed similar effects with Na ,K -ATPase α1 compound screen also reduced hippocampal neuronal autosis subunit siRNA knockdown in human and mouse cells. We specu- and conferred neuroprotection in vivo in neonatal rats subjected + + – fi late that the effects of the Na ,K -ATPase on increasing cell at- to cerebral hypoxia ischemia. Thus, by de ning a unique form of tachment (31) may contribute to the increased substrate adherence autophagic cell death and by performing an in vitro chemical of cells undergoing autotic death. In addition, it is possible that fi fi + + screen that identi ed a speci c class of inhibitors of this form of nuclear envelope-associated Na ,K -ATPase activity (32) may alter cell death (e.g., cardiac glycosides), we have been able to es- fi membrane ionic transport and osmolarity and thereby con- tablish a scienti c rationale for the use of cardiac glycosides in tribute to the ER and perinuclear space expansion observed in the treatment of a clinically important disease, neonatal cerebral autotic cells. – fi fi hypoxia ischemia. Based on our identi cation of speci c mor- Previous studies have shown that neriifolin and other cardiac phological criteria for autosis, it should be possible to determine glycosides reduce cerebral infarct size in rodent cerebral hyp- additional pathophysiological settings in which autosis plays a oxia–ischemia models (18, 24, 25); however, their mechanism of role and which may be ameliorated by cardiac glycosides. Con- neuroprotection has been unknown. Our observations suggest versely, mediators in the regulatory network of autosis may serve that inhibition of autophagy and autophagy-dependent death as candidate targets in cancer chemotherapy or other settings pathways may be a central mechanism of cardiac glycoside- where pharmacological induction of cell death may be beneficial. mediated neuroprotection. Several cell-death morphologies have We found that autosis occurs in at least two distinct physio- been identified in different regions of the brain after neonatal logical/pathophysiological conditions: starvation and cerebral hypoxic–ischemic injury, but neuron-specific deletion of Atg7 or hypoxia–ischemia. At present, it is not yet known which, if any, intracerebroventricular treatment with the autophagy inhibitor, previously reported instances of autophagic cell death involve 3-MA, is sufficient to reduce infarct lesion volume, indicating autosis (except for hypoxia–ischemia-induced hippocampal- that autophagy may be upstream of multiple death pathways. region-CA3 death evaluated in this study). It is possible that the This supports our previous recommendation that postischemic unique morphological changes we describe for autosis are present treatment of neonatal cerebral hypoxia–ischemia should target but have been missed in observations of autophagic cell death in autophagy (17, 33). In the present study, we observed inhibition other settings, especially those that lack concurrent features of of autophagy and autotic cell death in hippocampal-CA3-region apoptosis or necrosis and/or in tissues (e.g., heart and kidneys) neurons of rats treated with neriifolin, but the inhibition of autotic where high levels of autophagy are postulated to play a role in cell death in this region of the hippocampus is not sufficient to ischemia–reperfusion injury (26, 27). While the future identifi- explain the dramatic reduction in overall ipsilateral infarct size cation of specific biochemical markers of autosis will facilitate following hypoxia–ischemia. Taken together with previous studies such investigations, it should be possible to determine whether on autophagy, cell death, and neonatal cerebral hypoxia–ischemia, cell death occurs via autosis using the morphological criteria we the most likely explanation for the overall neuroprotection describe, as well as studies examining the inhibitory effect of in neriifolin-treated rats is the blockade of both autophagy- cardiac glycosides. One cautionary note is that certain isoforms dependent autotic death, as well as other death pathways triggered + + of the Na ,K -ATPase in the rodent (but not human) are resistant by high levels of autophagy. Thus, cardiac glycosides and/or other + + to cardiac glycosides; for example, the rodent-α3 subunit expressed agents targeting Na ,K -ATPase may not only ameliorate diseases

20370 | www.pnas.org/cgi/doi/10.1073/pnas.1319661110 Liu et al. Downloaded by guest on September 25, 2021 associated with autotic cell death, but also diseases in which Cell Death Assays. See SI Materials and Methods for details of trypan-blue staining, Sytox Green staining, CellTiter-Glo assays, PI staining, active cas- autophagy is upstream of other death execution pathways. We INAUGURAL ARTICLE cannot definitively rule out indirect effects of neriifolin on neu- pase-3 detection, TUNEL staining, DNA fragmentation assays, and clonogenic survival assays. roprotection, but these seem unlikely in view of our in vitro observations that cardiac glycosides inhibit stress-induced Microscopy Studies. See SI Materials and Methods for details. autophagy and autosis in a cell-autonomous manner. — — It is noteworthy that during cerebral hypoxia or ischemia High-Throughput Chemical Screening. Chemical screening and CSEA were the brain releases an endogenous form of cardiac glycoside performed as previously described (35). Details of screening methodology + + (ouabain or endobain) that inhibits Na ,K -ATPase (34). Thus, and analyses are provided in SI Materials and Methods. + + by releasing its own inhibitor of Na ,K -ATPase in response to hypoxia–ischemia, the neonatal brain may have developed an Autophagy Analyses. See SI Materials and Methods for details. important mechanism to reduce autophagy and cell death by autosis. A broader question is whether basal levels of endogenous Rat Model of Neonatal Cerebral Hypoxia–Ischemia. Neonatal rat cerebral – cardiac glycosides may serve as a naturally occurring “brake” which hypoxia ischemia experiments were performed as described (17). Immedi- ately after carotid artery occlusion, rat pups were injected intraperitoneally functions in multiple mammalian tissues to maintain autophagy with either neriifolin (0.25 mg/kg diluted in 0.5% ethanol/PBS) (Sigma, at physiological levels that promote cell survival, rather than at S961825) or vehicle (0.5%ethanol/PBS). See SI Materials and Methods for pathological levels that promote cell death. details. All experiments were performed in accordance with Swiss laws for the protection of animals and were approved by the Vaud Cantonal Vet- Materials and Methods erinary Office (authorization no. 1745.2). Cell Culture. HeLa cells were obtained from American Type Culture Collection. − − − − − − − − Information on the source of wild-type, Ripk3 / , Ripk1 / ;Ripk3 / ,andBax / ; ACKNOWLEDGMENTS. We thank Noboru Mizushima, Qiong Shi, Herbert − − fl fl fl fl Bak / MEFs, Atg5 ox/ ox and Atg5 ox/ ox-LysM-Cre BMDMs, and U2OSTetR, Virgin, Xiaodong Wang, Qing Zhong, and Sandra Zinkel for providing critical TetR TetR reagents; Shuguang Wei for assistance with high-throughput screening; U2OS /shATG14, and U2OS /shATG13, and other cells used in this study Beatriz Fontoura for helpful discussions; Zhongju Zou for technical support; and culture conditions is provided in SI Materials and Methods. Haley Harrington for assistance with manuscript preparation; the University of Texas (UT) Southwestern Live Cell Imaging Facility; and the Electron Mi- Autophagy-Inducing Peptides. Tat-Scrambled, Tat-Beclin 1, and Tat-vFLIP α2 croscopy Facility of the University of Lausanne. This work was supported by (14) were synthesized and administered to cells as described (13). National Institutes of Health Grants U54 AI057156 and RO1 CA109618 (to B.L.), RO1 A140646 (to D.R.G.); Contract HHSN268201000044C (to S.Y.S.), CELL BIOLOGY PO1 CA95471 (to Dr. Steven McKnight, UT Southwestern Medical Center), Antibodies and siRNAs. See SI Materials and Methods for details of antibodies and P30 CA142543 (to the UT Southwestern Simmons Cancer Center); and by and siRNAs used in this study. the Swiss National Science Foundation Grant 310030-130769 (to J.P.).

1. Levine B, Kroemer G (2008) Autophagy in the pathogenesis of disease. Cell 132(1): 19. Shen S, et al. (2011) Association and dissociation of autophagy, apoptosis and necrosis 27–42. by systematic chemical study. Oncogene 30(45):4544–4556. 2. Galluzzi L, et al. (2012) Molecular definitions of cell death subroutines: Recom- 20. Hundeshagen P, Hamacher-Brady A, Eils R, Brady NR (2011) Concurrent detection of mendations of the Nomenclature Committee on Cell Death 2012. Cell Death Differ autolysosome formation and lysosomal degradation by flow cytometry in a high- 19(1):107–120. content screen for inducers of autophagy. BMC Biol 9:38. 3. Clarke PGH (1990) Developmental cell death: Morphological diversity and multiple 21. Wang Y, et al. (2012) Cardiac glycosides induce autophagy in human non-small cell mechanisms. Anat Embryol (Berl) 181(3):195–213. lung cancer cells through regulation of dual signaling pathways. Int J Biochem Cell 4. Lockshin RA, Zakeri Z (2004) Apoptosis, autophagy, and more. Int J Biochem Cell Biol Biol 44(11):1813–1824. 36(12):2405–2419. 22. Prassas I, Diamandis EP (2008) Novel therapeutic applications of cardiac glycosides. 5. Levine B, Yuan J (2005) Autophagy in cell death: An innocent convict? J Clin Invest Nat Rev Drug Discov 7(11):926–935. 115(10):2679–2688. 23. Newman RA, Yang P, Pawlus AD, Block KI (2008) Cardiac glycosides as novel cancer – 6. Das G, Shravage BV, Baehrecke EH (2012) Regulation and function of autophagy therapeutic agents. Mol Interv 8(1):36 49. during cell survival and cell death. Cold Spring Harb Perspect Biol 4(6):a008813. 24. Kaur S, Rehni AK, Singh N, Jaggi AS (2009) Studies on cerebral protection of digoxin – 7. Lamy L, et al. (2013) Control of autophagic cell death by caspase-10 in multiple my- against ischemia/reperfusion injury in mice. Yakugaku Zasshi 129(4):435 443. eloma. Cancer Cell 23(4):435–449. 25. Dunn DE, et al. (2011) In vitro and in vivo neuroprotective activity of the cardiac 8. Elgendy M, Sheridan C, Brumatti G, Martin SJ (2011) Oncogenic Ras-induced expres- glycoside oleandrin from Nerium oleander in brain slice-based stroke models. J Neurochem 119(4):805–814. sion of Noxa and Beclin-1 promotes autophagic cell death and limits clonogenic 26. Suzuki C, et al. (2008) Participation of autophagy in renal ischemia/reperfusion injury. survival. Mol Cell 42(1):23–35. Biochem Biophys Res Commun 368(1):100–106. 9. Reef S, et al. (2006) A short mitochondrial form of p19ARF induces autophagy and 27. Matsui Y, et al. (2007) Distinct roles of autophagy in the heart during ischemia and caspase-independent cell death. Mol Cell 22(4):463–475. reperfusion: Roles of AMP-activated protein kinase and Beclin 1 in mediating au- 10. Wu WK, et al. (2012) The autophagic paradox in cancer therapy. Oncogene 31(8): tophagy. Circ Res 100(6):914–922. 939–953. 28. Lingrel JB (2010) The physiological significance of the cardiotonic steroid/ouabain- 11. Koike M, et al. (2008) Inhibition of autophagy prevents hippocampal pyramidal binding site of the Na,K-ATPase. Annu Rev Physiol 72:395–412. neuron death after hypoxic-ischemic injury. Am J Pathol 172(2):454–469. 29. Zwerger M, Kolb T, Richter K, Karakesisoglou I, Herrmann H (2010) Induction of 12. Xing S, et al. (2012) Beclin 1 knockdown inhibits autophagic activation and prevents a massive endoplasmic reticulum and perinuclear space expansion by expression of the secondary neurodegenerative damage in the ipsilateral thalamus following focal lamin B receptor mutants and the related sterol reductases TM7SF2 and DHCR7. Mol – cerebral infarction. Autophagy 8(1):63 76. Biol Cell 21(2):354–368. 13. Shoji-Kawata S, et al. (2013) Identification of a candidate therapeutic autophagy- 30. Rubinsztein DC, Shpilka T, Elazar Z (2012) Mechanisms of autophagosome biogenesis. – inducing peptide. Nature 494(7436):201 206. Curr Biol 22(1):R29–R34. 14. Lee JS, et al. (2009) FLIP-mediated autophagy regulation in cell death control. Nat Cell 31. Contreras RG, Shoshani L, Flores-Maldonado C, Lázaro A, Cereijido M (1999) Re- – Biol 11(11):1355 1362. lationship between Na(+),K(+)-ATPase and cell attachment. J Cell Sci 112(Pt 23): 15. Huang W, et al. (2012) Crystal structure and biochemical analyses reveal Beclin 1 as 4223–4232. – a novel membrane binding protein. Cell Res 22(3):473 489. 32. Galva C, Artigas P, Gatto C (2012) Nuclear Na+/K+-ATPase plays an active role in 16. Pfeiffer DR, Gudz TI, Novgorodov SA, Erdahl WL (1995) The peptide mastoparan is nucleoplasmic Ca2+ homeostasis. J Cell Sci 125(Pt 24):6137–6147. a potent facilitator of the mitochondrial permeability transition. J Biol Chem 270(9): 33. Puyal J, Ginet V, Clarke PGH (2013) Multiple interacting cell death mechanisms in the 4923–4932. mediation of excitotoxicity and ischemic brain damage: A challenge for neuro- 17. Ginet V, Puyal J, Clarke PGH, Truttmann AC (2009) Enhancement of autophagic flux protection. Prog Neurobiol 105:24–48. after neonatal cerebral hypoxia-ischemia and its region-specific relationship to apo- 34. De Angelis C, Haupert GT, Jr. (1998) Hypoxia triggers release of an endogenous in- ptotic mechanisms. Am J Pathol 175(5):1962–1974. hibitor of Na(+)-K(+)-ATPase from midbrain and adrenal. Am J Physiol 274(1 Pt 2): 18. Wang JK, et al. (2006) Cardiac glycosides provide neuroprotection against ischemic F182–F188. stroke: Discovery by a brain slice-based compound screening platform. Proc Natl Acad 35. Shaw SY, et al. (2011) Disease allele-dependent small-molecule sensitivities in blood Sci USA 103(27):10461–10466. cells from monogenic diabetes. Proc Natl Acad Sci USA 108(2):492–497.

Liu et al. PNAS | December 17, 2013 | vol. 110 | no. 51 | 20371 Downloaded by guest on September 25, 2021