Organelle-Specific Initiation of Autophagy - Cell Press

20 downloads 0 Views 3MB Size Report
Aug 20, 2015 - species (RNS) can be detected by a cytoplasmic pool of ATM (a key sensor .... serine/threonine kinase that phosphorylates several substrates,.
Molecular Cell

Review Organelle-Specific Initiation of Autophagy Valentina Sica,1,2,3,4,10 Lorenzo Galluzzi,1,2,3,5,6,10 Jose´ Manuel Bravo-San Pedro,1,2,3,5,6 Valentina Izzo,1,2,3,5,6 Maria Chiara Maiuri,1,2,3,5,6 and Guido Kroemer1,2,5,6,7,8,9,* 1Equipe

11 labellise´e Ligue contre le Cancer, Centre de Recherche des Cordeliers, 75006 Paris, France U1138, 75006 Paris, France 3Gustave Roussy Comprehensive Cancer Institute, 94805 Villejuif, France 4Faculte ´ de Medicine, Universite´ Paris Sud/Paris XI, 94270 Le Kremlin-Biceˆtre, France 5Universite ´ Paris Descartes/Paris V, Sorbonne Paris Cite´, 75006 Paris, France 6Universite ´ Pierre et Marie Curie/Paris VI, 75006 Paris, France 7Metabolomics and Cell Biology Platforms, Gustave Roussy Comprehensive Cancer Institute, 94805 Villejuif, France 8Po ˆ le de Biologie, Hoˆpital Europe´en Georges Pompidou, AP-HP, 75015 Paris, France 9Karolinska Institute, Department of Women’s and Children’s Health, Karolinska University Hospital, 17176 Stockholm, Sweden 10Co-first author *Correspondence: [email protected] http://dx.doi.org/10.1016/j.molcel.2015.07.021 2INSERM,

Autophagy constitutes a prominent mechanism through which eukaryotic cells preserve homeostasis in baseline conditions and in response to perturbations of the intracellular or extracellular microenvironment. Autophagic responses can be relatively non-selective or target a specific subcellular compartment. At least in part, this depends on the balance between the availability of autophagic substrates (‘‘offer’’) and the cellular need of autophagic products or functions for adaptation (‘‘demand’’). Irrespective of cargo specificity, adaptive autophagy relies on a panel of sensors that detect potentially dangerous cues and convert them into signals that are ultimately relayed to the autophagic machinery. Here, we summarize the molecular systems through which specific subcellular compartments—including the nucleus, mitochondria, plasma membrane, reticular apparatus, and cytosol—convert homeostatic perturbations into an increased offer of autophagic substrates or an accrued cellular demand for autophagic products or functions. Introduction Macroautophagy, hereafter called autophagy, is a key catabolic process for the maintenance of cellular (and hence organismal) homeostasis in the eukaryotic kingdom. On the one hand, autophagy operates in virtually all eukaryotic cells at basal levels. Such a constitutive autophagic flux ensures the disposal of intracellular entities that accumulate or degenerate in the course of normal cellular functions (Mizushima et al., 2008). On the other hand, the autophagic machinery is extremely sensitive to a wide panel of perturbations of the intracellular or extracellular microenvironment. Thus, potentially dangerous cues of metabolic, physical, as well as chemical nature generally accelerate autophagic flux, and such an adaptive response is critical for cells to cope with stress (Kroemer et al., 2010). In line with this notion, the pharmacological or genetic inhibition of autophagy most often reduces the resistance of cells to stress (Kroemer et al., 2010). In some instances though, autophagy causally contributes to, rather than counteracts, regulated cell death (RCD) (Galluzzi et al., 2015a). This has been demonstrated not only for programmed instances of RCD that occur during embryonic development (at least in Drosophila melanogaster), but also for the regulated demise of cancer cells exposed to some chemotherapeutic, hypoxia, or specific autophagy-inducing peptides (Galluzzi et al., 2015a). Importantly, adaptive autophagic responses are not isolated, self-standing processes, but occur in the context of (and often occupy a central position within) general mechanisms that aim at the maintenance of both cellular and organismal functions in response to stress (Kroemer et al., 2010). Thus, ongoing (and 522 Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc.

hence potentially successful) adaptive responses actively inhibit RCD to preserve cellular homeostasis. Conversely, when cellular adaptation to stress fails, RCD is initiated as a means to eliminate potential sources of danger for the entire organism (Galluzzi et al., 2014a). This implies that the final outcome of specific perturbations of homeostasis, especially when such perturbations are ‘‘close-to-physiological’’ in intensity and duration, is influenced by a wide range of cell-intrinsic and cell-extrinsic parameters, hence exhibiting a considerable degree of context dependency (Galluzzi et al., 2014d). Autophagy can be viewed as a multi-step process involving (1) the formation of a specialized double-membraned organelle, the so-called ‘‘autophagosome,’’ around intracellular entities destined to degradation (which are generally referred to as ‘‘autophagic substrates’’ or ‘‘autophagic cargo’’) and (2) the fusion of autophagosomes with lysosomes, coupled to the activation of lysosomal hydrolases and cargo degradation (Mizushima et al., 2011). Several proteins are required for the proper execution of so-called ‘‘canonical’’ autophagic programs in mammalian cells, including: (1) unc-51-like autophagy activating kinase 1 (ULK1), which is involved in first steps of autophagosome formation; (2) beclin 1 (BECN1) and phosphatidylinositol 3-kinase, catalytic subunit type 3 (PIK3C3, also known as VPS34), two key components of a multiprotein complex with Class III phosphatidylinositol 3-kinase (PI3K) activity; (3) autophagy-related 5 (ATG5), ATG7, ATG12, and autophagy-related 16-like 1 (ATG16L1), all of which are involved in the elongation and closure of autophagosomes; (4) at least one among various autophagic modifiers (which are responsible for cargo uptake by forming

Molecular Cell

Review Figure 1. Autophagic Offer and Demand Adaptive autophagic responses can be driven either by an increased availability of autophagic substrates (‘‘offer’’) or by an accrued need for autophagic functions (‘‘demand’’). This determines, at least in part, cargo specificity. Situations of increased offer arise in response to specific perturbations of homeostasis, resulting in the autophagic degradation of selected intracellular entities. Conversely, situations of increased demand reflect the establishment of stress conditions (of metabolic, physical, or chemical nature), to which cells adapt by mounting an autophagic response.

autophagosomes), including microtubule-associated protein 1 light chain 3 beta, MAP1LC3B, also known as LC3B), GABA(A) receptor-associated protein (GABARAP), and GABARAP-like 2 (GABARAPL2, best known as GATE.16); and (5) ATG14, which (besides regulating VPS34 activity) is required for the fusion of autophagosomes with lysosomes (Diao et al., 2015; Mizushima et al., 2011). In addition, accumulating evidence points to the existence of alternative, ‘‘non-canonical’’ autophagic responses that do not rely on all these factors (Codogno et al., 2012). A detailed description of the molecular mechanisms that underlie the execution of canonical and non-canonical autophagy exceeds the scope of this review and can be found in Mizushima et al. (2011) and Codogno et al. (2012). Autophagic responses can be relatively non-selective or, on the contrary, highly specific (Okamoto, 2014). The former generally occur in conditions of nutrient deprivation and hence serve metabolic purposes. In this case, cargo selection is not very strict and various intracellular entities can be degraded to provide the cell with metabolic substrates for survival. Conversely, very specific forms of autophagy are activated in response to precise homeostatic needs. Thus, ‘‘mitophagy’’ ensures the selective degradation of damaged mitochondria, ‘‘ribophagy’’ that of ribosomes, and ‘‘pexophagy’’ that of peroxisomes; ‘‘reticulophagy’’ specifically affects portions of the endoplasmic reticulum (ER) and ‘‘nucleophagy’’ parts of the nucleus; ‘‘aggrephagy’’ selectively removes redox-active protein aggregates, ‘‘lipophagy’’ lipid droplets, and ‘‘xenophagy’’ intracellular bacteria that escape endosomes (Okamoto, 2014). The selectivity of autopha-

gic responses is mainly (but not only) dictated by the capacity of the initiating stimulus to cause a precise homeostatic perturbation that increases the availability of specific autophagic substrates (‘‘offer’’) versus a generalized situation of stress that augments the cellular need for autophagic products or functions (‘‘demand’’) (Figure 1). Irrespective of cargo specificity, autophagy-promoting perturbations of homeostasis are detected by molecular sensors that are located in virtually all subcellular compartments and relayed (via one or more signal transducers) to the core autophagic machinery. Here, we discuss how distinct subcellular compartments react to perturbations of homeostasis by increasing the offer of autophagic substrates or exacerbating the cellular demand for autophagic products or functions. Central Regulation of Autophagy Autophagy is regulated by a highly interconnected network of proteins that, in addition, are involved in the control of other key cellular processes, including cell growth, replication, and RCD (Marin˜o et al., 2014). Perhaps the best-characterized upstream regulators of autophagy in mammalian cells are mechanistic target of rapamycin (MTOR) complex 1 (MTORC1) and protein kinase, AMP-activated (PRKA, best known as AMPK). In nutrient-replete, physiological conditions, MTORC1 is hyperactivated as a consequence of growth factor receptor signaling, while the enzymatic activity of AMPK is limited because cells contain elevated amounts of ATP (Mihaylova and Shaw, 2011). In this scenario, autophagy is inhibited owing to the ability of MTORC1 to catalyze the inactivating phosphorylation of ULK1 and ATG13, two critical components of the autophagy-initiating machinery (Hosokawa et al., 2009). Alongside, MTORC1 stimulates cell growth by catalyzing the activating phosphorylation of several other substrates, including eukaryotic translation initiation factor 4E binding protein 1 (EIF4EBP1, also known as 4-EBP1) and ribosomal protein S6 kinase, 70 kDa, polypeptide 1 (RPS6KB1, also known as p70S6K) (Shimobayashi and Hall, 2014). Conversely, in response to nutrient deprivation (and other autophagy-promoting cues), AMPK is activated upon the accumulation of ADP, AMP, and Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc. 523

Molecular Cell

Review cyclic AMP (cAMP) at the expense of ATP. AMPK favors MTORC1 inactivation by catalyzing the activating phosphorylation of the MTORC1 inhibitor tuberous sclerosis 2 (TSC2) (Inoki et al., 2002). Moreover, AMPK can unleash the catalytic activity of ULK1 upon phosphorylation at various serine residues (Egan et al., 2011) as well as that of VPS34 by phosphorylating BECN1 (Kim et al., 2013). Multiple signal transduction pathways that ultimately accelerate or inhibit the autophagic flux do so by impinging of the activation state of MTORC1 and/or AMPK (Mihaylova and Shaw, 2011; Shimobayashi and Hall, 2014). Another important component of the autophagy-regulatory network is the VPS34-containing multiprotein complex that catalyzes the phosphorylation of phosphatidylinositol to generate phosphatidylinositol 3-phosphate (PtdIns3P) (Kim et al., 2013). Besides VPS34 and BECN1, the core of this complex involves ATG14, phosphoinositide-3-kinase, regulatory subunit 4 (PIK3R4, best known as VPS15), UV radiation resistance associated (UVRAG), and autophagy/beclin-1 regulator 1 (AMBRA1), all of which are required for optimal PtdIns3P synthesis in the course of autophagic responses (He and Levine, 2010). In baseline conditions, the VPS34 activator BECN1 is tonically inhibited as it physically interacts with anti-apoptotic B cell CLL/ lymphoma 2 (BCL2)-like proteins, including BCL2 itself and BCL2-like 1 (BCL2L1, best known as BCL-XL) (Pattingre et al., 2005), with GLI pathogenesis-related 2 (GLIPR2, also known as GAPR1) (Shoji-Kawata et al., 2013), as well as with distinct members of the 14-3-3 protein family, such as tyrosine 3-monooxygenase/tryptophan 5-monooxygenase activation protein, epsilon (YWHAE, best known as 14-3-3ε) (Wang et al., 2012). There are at least four physiological mechanisms whereby BECN1 can be displaced from such inhibitory interactions, resulting in the derepression of autophagy: (1) upon the phosphorylation of BCL2 on T69, S70, and S87, a reaction that is catalyzed by mitogen-activated protein kinase 8 (MAPK8, best known as JNK1) (Ravikumar et al., 2010); (2) upon the phosphorylation of BECN1 on T119, a reaction that is catalyzed by deathassociated protein kinase 1 (DAPK1) (Ravikumar et al., 2010); (3) upon the transcriptional or post-translational activation of so-called ‘‘BH3-only’’ proteins, small members of the Bcl-2 protein family that compete with BECN1 for BCL2 and BCL-XL binding (Maiuri et al., 2007); and (4) upon the inhibition of v-akt murine thymoma viral oncogene homolog 1 (AKT1) signaling, resulting in BECN1 dephosphorylation on S295 and decreased affinity for 14-3-3ε (Wang et al., 2012). Furthermore, BECN1 is negatively regulated by E1A binding protein p300 (EP300)-dependent acetylation (Sun et al., 2015). In physiological conditions AMBRA1 is also sequestered by BCL2, and such an inhibitory interaction is lost during autophagic responses (Strappazzon et al., 2011). Moreover, the abundance of AMBRA1 is negatively regulated by two distinct E3 ubiquitin ligases, i.e., cullin 4A (CUL4A) and ring finger protein 2 (RNF2) (Antonioli et al., 2014; Xia et al., 2014). In response to autophagy induction, AMBRA1 is released from CUL4A by an ULK1-dependent mechanism and acquires the capacity to inhibit cullin 5 (CUL5), resulting in the stabilization of the MTOR inhibitor DEP domain containing MTOR-interacting protein (DEPTOR) (Antonioli et al., 2014). Thus, the dynamic interplay between AMBRA1 and CUL family members underlies a feedfor524 Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc.

ward loop for the amplification of autophagy-promoting signals (Antonioli et al., 2014). Several other proteins physically interact with (and hence modulate the catalytic activity of) the VPS34 complex, including (but not limited to) SH3-domain GRB2-like endophilin B1 (SH3GLB1, also known as BIF-1) and KIAA0226 (best known as RUBICON) (He and Levine, 2010). The actual requirement of these factors for optimal autophagic responses, however, has not yet been determined and may exhibit a significant degree of context dependency. Notably, all the core components of the autophagy-regulatory network receive activatory or inhibitory inputs from a wide panel of stress sensors (directly or via signal transduction pathways), as detailed below. Initiation of Autophagy at Mitochondria Functional mitochondria exhibit an elevated mitochondrial transmembrane potential (Dcm), which is required not only for ATP production by the F1FO synthase, but also for protein import and other mitochondrial functions (Galluzzi et al., 2012d). In this setting, PTEN-induced putative kinase 1 (PINK1) is actively imported into mitochondria and degraded by presenilin-associated, rhomboid-like (PARL), a protease of the inner mitochondrial membrane (IMM) (Jin et al., 2010). Several potentially lethal stimuli, including hypoxia and respiratory chain inhibitors like rotenone, antimycin A, and oligomycin, can cause a primary (i.e., not dependent on upstream signal transduction pathways) impairment in mitochondrial functions by provoking the overgeneration of reactive oxygen species (ROS). ROS per se promote autophagy as they oxidize a specific cysteine residue located near the catalytic site of ATG4, a protease involved in the maturation of LC3 (Scherz-Shouval et al., 2007). Besides being genotoxic (which means they can engage nuclear systems of response to stress, see below), ROS as well as reactive nitrogen species (RNS) can be detected by a cytoplasmic pool of ATM (a key sensor of DNA damage), resulting in the activation of serine/ threonine kinase 11 (STK11, best known as LKB1), a positive regulator of AMPK signaling (Alexander et al., 2010; Tripathi et al., 2013). Finally, ROS alter the abundance of various Bcl-2 family proteins, and this favors the so-called mitochondrial permeability transition (MPT), entailing the dissipation of the Dcm (Galluzzi et al., 2012a). Several other stimuli promote mitochondrial depolarization, including the protonophore carbonyl cyanide m-chlorophenyl hydrazine (CCCP), which is widely employed as an experimental inducer of mitophagy (Galluzzi et al., 2012a). Indeed, irrespective of the primary trigger, PINK1 is no longer imported by depolarized mitochondria, accumulates at their surface, and initiates a mitophagic response by phosphorylating several substrates, including itself and ubiquitin (Koyano et al., 2014; Okatsu et al., 2012). This results in the recruitment and activation of parkin RBR E3 ubiquitin protein ligase (PARK2) and valosin containing protein (VCP, also known as p97) (Koyano et al., 2014; Okatsu et al., 2012). Notably, phosphorylated ubiquitin has recently been proposed to constitute the genuine receptor for PARK2 on the outer mitochondrial membrane (OMM) (Okatsu et al., 2015). PARK2 is able to attach ubiquitin moieties to several proteins of the OMM including voltage-dependent anion channel 1 (VDAC1), mitofusin 1 (MFN1), and MFN2 (Geisler et al., 2010;

Molecular Cell

Review Tanaka et al., 2010). Ubiquitinated MFN1 and MFN2 are degraded by the proteasome system in a VCP-dependent manner, promoting mitochondrial fragmentation (Tanaka et al., 2010). Moreover, as ubiquitinated structures are recognized by the autophagic adaptor sequestosome 1 (SQSTM1, best known as p62), PARK2 de facto tags permeabilized mitochondria for recognition by LC3-containing autophagosomal membranes (Narendra et al., 2010). Whether p62 is truly required for the autophagic disposal of permeabilized mitochondria remains a matter of debate (Geisler et al., 2010; Narendra et al., 2010). Thus, the MPT increases the intracellular availability of small mitochondria decorated with ubiquitin moieties, hence stimulating their autophagic removal. This is a good example of an adaptive, organelle-specific autophagic response driven by an increased offer of autophagic substrates. Other homeostatic perturbations that are detected at mitochondria trigger autophagy when cells need autophagic functions or products to cope with stress. For instance, this occurs in the course of viral infection, when accumulating viral nucleic acids promote the secretion of type I interferon (IFN) upon the activation of pattern recognition receptors (PRRs) that are localized at the interface between mitochondria and the ER, the so-called mitochondria-associated ER membranes (MAMs) (Rowland and Voeltz, 2012). One of the PRRs that operates in these conditions is NLR family member X1 (NLRX1), which may promote autophagy by interacting with the ATG5-ATG12 complex via Tu translation elongation factor, mitochondrial (TUFM) (Lei et al., 2012). Although the actual role of NLRX1 in type I IFN responses remains a matter of debate, it is clear that besides triggering xenophagy (see below), viral infection can promote non-selective autophagic responses via NLRX1 (Lei et al., 2012). This is also the case of intracellular bacteria (Chauhan et al., 2015). Indeed, another protein that is localized to MAMs and the OMM, i.e., immunity-related GTPase family, M (IRGM), plays a key role in the initiation of autophagy by bacterial products (Chauhan et al., 2015). Thus, besides promoting the activation of AMPK in response to nutrient deprivation, IRGM operates as a hub for the assembly of autophagy-promoting supramolecular complexes involving ULK1, BECN1, and ATG16L1 upon detection of bacterial muramyl dipeptide by the PRR nucleotide-binding oligomerization domain containing 2 (NOD2) (Chauhan et al., 2015). Both of these scenarios exemplify the activation of adaptive autophagy by an increase in the cellular demand for autophagic functions. Notably, MAMs (and the OMM) have also been suggested to provide lipids for the formation of autophagosomal membranes (Lamb et al., 2013). Interestingly, nutrient deprivation (which is mostly sensed at the cytosolic and lysosomal level, see below) can trigger an adaptive autophagic response that is relatively non-selective, yet preferentially sequesters parts of the mitochondrial network ahead of other intracellular entities, at least in hepatocytes (Kim and Lemasters, 2011). Most probably, this reflects the ability of nutrient deprivation to cause an early (and potentially genotoxic, see below) wave of mitochondrial depolarization (Rodrı´guez-Vargas et al., 2012). In this scenario, cellular viability appears to rely on elongated mitochondria that are spared from autophagic degradation and preserve optimal bioenergetic functions (Gomes et al., 2011). At least in part, this process originates

from the accumulation of cAMP in the course of nutrient deprivation, eventually promoting the activation of protein kinase, cAMP-dependent (PRKA), and the consequent inactivating phosphorylation of dynamin 1-like (DNM1L), a protein that normally mediates mitochondrial fission (Gomes et al., 2011). Accumulating evidence indicates that an active mechanism is in place to preserve perfectly efficient mitochondria as part of an elongated network (which is not degraded by autophagy), at the expense of damaged and suboptimal organelles (which are degraded by autophagy) in the course of adaptive stress responses (Twig et al., 2008). Such a quality control mechanism is regulated by Dcm (which affects the ubiquitination of MFN1 and MFN2), by DNM1L activation state (which is controlled by PRKA), as well as on the levels of optic atrophy 1 (OPA1), another component of the mitochondrial fission machinery (Twig et al., 2008). Altogether, these observations suggest that mitochondria not only ignite adaptive mitophagy as a consequence of primary alterations in their homeostasis, but also play a prominent role in the initiation or execution of non-selective autophagic responses driven by an increased cellular demand for autophagic functions or products (Figure 2). Initiation of Autophagy by the Nucleus A plethora of conditions are potentially toxic for mammalian cells, as they cause DNA damage or perturb/prevent correct DNA replication (Bouwman and Jonkers, 2012). These conditions include not only physical stimuli (e.g., irradiation, hypoxia, nutrient deprivation) and chemicals (e.g., alkylating agents, anthracyclines) that are directly genotoxic or cause mitochondrial ROS overgeneration, but also endogenous states that can arise in the course of (and de facto accelerate) oncogenesis (e.g., the so-called ‘‘DNA replication stress,’’ polyploidization) (Bouwman and Jonkers, 2012). Several sensors are in place to detect alterations in nuclear and genetic homeostasis, and they all operate as they abide to a ‘‘cell first, organism then’’ rule: they initially activate mechanisms for the repair of cellular damage and, if these fail, trigger RCD for the sake of organismal homeostasis (Galluzzi et al., 2014a). Accumulating evidence indicates that virtually all systems that attempt to repair genetic/genomic lesions are activated along with a relatively non-selective autophagic response and that such a cell-wide reaction is required for cells to optimally cope with DNA damage (Galluzzi et al., 2015d). Double-strand DNA breaks are normally detected by ATM, a serine/threonine kinase that phosphorylates several substrates, including the oncosuppressive transcription factor tumor protein p53 (TP53, best known as p53) (Bieging et al., 2014). Initially, this results in the p53-dependent expression of several factors involved in DNA repair, including proteins that temporarily arrest cell-cycle progression like cyclin-dependent kinase inhibitor 1A (CDKN1A) (Bieging et al., 2014). ATM has also been causally involved in the autophagic responses of mammalian cells to various DNA-damaging agents, including (but not limited to) irradiation, alkylating agents like temozolomide, and anthracyclines (Figueiredo et al., 2013; Liang et al., 2013). Two mechanisms have been proposed for explaining the ability of ATM to initiate autophagy in response to DNA damage. First, ATM can catalyze the activating phosphorylation of LKB1, eventually leading to the Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc. 525

Molecular Cell

Review Figure 2. Initiation of Autophagy by Mitochondria (A) Irrespective of the initiating stimulus, depolarized mitochondria are unable to normally import and degrade PTEN induced putative kinase 1 (PINK1), resulting in the PINK1-dependent recruitment of parkin RBR E3 ubiquitin protein ligase (PARK2) and valosin containing protein (VCP) on the outer mitochondrial membrane (OMM). PARK2 attaches ubiquitin (Ub) moieties to several OMM proteins, including voltage-dependent anion channel 1 (VDAC1). This favors the recruitment of LC3-containing autophagosomes to depolarized mitochondria, which (at least in some settings) involves p62. (B) Several microbial constituents (as well as some endogenous danger signals) are specifically detected by pattern recognition receptors (PRRs) located at mitochondria-associated ER membranes (MAMs). Among these PRRs, NLR family member X1 (NLRX1) promotes autophagy by interacting with the ATG5-ATG12 complex via Tu translation elongation factor, mitochondrial (TUFM). Along similar lines, the MAM-resident protein immunity-related GTPase family, M (IRGM) stimulates autophagy in response to bacterial muramyl dipeptide (MDP) bound to the PRR nucleotide-binding oligomerization domain containing 2 (NOD2). (C) Various homeostatic perturbations trigger the overgeneration of reactive oxygen species (ROS) by the mitochondrial respiratory chain, including hypoxia and inhibitors of respiratory complexes. ROS promote autophagy not only by favoring mitochondrial depolarization, but also by modulating the activity of ATG4, by activating a cytoplasmic pool of ATM, as well as by damaging nuclear DNA, hence engaging nuclear stress sensors with pro-autophagic activity. Dcm, mitochondrial transmembrane potential; AMPK (official name PRKA), protein kinase, AMP-activated; ATG16L1, autophagy-related 16-like 1; BECN1, beclin 1; CCCP, carbonyl cyanide m-chlorophenyl hydrazine; IMM, inner mitochondrial membrane; LKB1 (official name STK11), serine/threonine kinase 11; MPT, mitochondrial permeability transition; PARL, presenilin-associated, rhomboid-like; R, ribosome; TIM, translocase of the IMM; TOM, translocase of the OMM; ULK1, unc-51-like autophagy activating kinase 1; VPS34 (official name PIK3C3), phosphatidylinositol 3-kinase, catalytic subunit type 3.

AMPK-dependent inhibition of MTORC1 (see above) (Sapkota et al., 2002). Second, ATM can directly phosphorylate phosphatase and tensin homolog (PTEN), a functional antagonist of Class I PI3Ks, resulting in the nuclear accumulation of PTEN and an autophagic response associated with AMPK activation (Chen et al., 2015). Although cytoplasmic PTEN is known to stimulate autophagy by antagonizing the PI3K/AKT1 signaling axis, the precise mechanism through which nuclear PTEN upregulates the autophagic flux remains to be elucidated. The stabilization of p53 (be it induced by ATM or by other stress sensors, including the DNA damage-responsive kinase ATR and its downstream effector checkpoint kinase 1, CHEK1) stimulates autophagy by various mechanisms (Pietrocola et al., 2013). First, the accumulation of transcriptionally active p53 tetramers in the nucleus entails a depletion in the cytoplasmic p53 pool, and this favors autophagy upon the derepression of RB1inducible coiled-coil 1 (RB1CC1), a critical component of the ULK1 complex (Pietrocola et al., 2013). Second, nuclear p53 transactivates several genes involved in autophagic responses, including (but not limited to) those coding for ULK1, UVRAG, ATG7, LBK1, various BH3-only proteins, and the AMPK activator sestrin 2 (SESN2) (Bieging et al., 2014). Third, nuclear p53 transcriptionally represses at least some of the genes involved in tonic autophagy inhibition, such as those coding for BCL2 and various growth factor receptors (Wang et al., 2010). Thus, similar to other stress-responsive transcription factors like NF-kB and 526 Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc.

signal transducer and activator of transcription 3 (STAT3), p53 appears to regulate autophagy via both transcription-independent (rapid) and transcriptional (delayed) mechanisms (Pietrocola et al., 2013). Another important sensor of DNA damage is poly(ADP-ribose) (PAR) polymerase 1 (PARP1) (Gibson and Kraus, 2012). By catalyzing the addition of PAR moieties to histones, PARP1 recruits several components of the repair machinery to sites of alkylating DNA damage (such as that inflicted by temozolomide) (Gibson and Kraus, 2012). Active PARP1 has also been mechanistically involved in adaptive autophagic responses driven by alkylating agents and irradiation (Ethier et al., 2012). Presumably, this originates from the fact that the reaction catalyzed by PARP1 requires NAD+, implying that PARP1 hyperactivation provokes a drop in intracellular ATP levels that may activate AMPK (Ethier et al., 2012). However, NAD+ is also an obligate co-factor for the deacetylase sirtuin 1 (SIRT1), which robustly promotes autophagy (Lee et al., 2008). These apparently discrepant observations may reflect the fact that AMPK promotes autophagy via post-translational mechanisms (Ethier et al., 2012), whereas SIRT1 does so through transcription factors of the forkhead box O (FOXO) family (Pietrocola et al., 2013). Interestingly, nutrient deprivation has also been shown to cause DNA damage upon the overgeneration of ROS, and PARP1-deficient cells deprived of nutrients mount delayed autophagic responses (Rodrı´guez-Vargas et al., 2012). These observations indicate that

Molecular Cell

Review PARP1 not only initiates autophagy in response to DNA damage, but also exacerbates autophagy induction by mitochondria. Recently, it has been proposed that caspase 2 (CASP2), a cysteine protease that participates in the induction of apoptosis in cells that fail to recover from DNA damage, tonically suppresses autophagy (Tiwari et al., 2014). In line with this notion, the genetic inhibition of CASP2 is associated with the upregulation of autophagy in a wide panel of cells and tissues. Moreover, CASP2-deficient cells exhibit accrued autophagic responses to DNA-damaging agents as compared to their wild-type counterparts (Tiwari et al., 2014). The derepression of autophagy in CASP2-incompetent cells reportedly involves AMPK activation, MTORC1 inhibition, and various other components of the canonical autophagic machinery (Tiwari et al., 2014). However, the precise mechanisms through which CASP2 normally represses autophagy have not yet been clarified. Importantly, the nuclear compartment not only initiates autophagy in response to perturbations of genetic/genomic homeostasis, but also plays a key role in autophagic responses initiated by other organelles. On the one hand, a wide panel of transcription factors that control the synthesis of components or regulators of autophagy, such as p53, NF-kB, and STAT3, respond to signal transduction cascades that are not activated in the nucleus (Pietrocola et al., 2013). A comprehensive list of stress-responsive transcription factors involved in delayed autophagic responses, including transcription factor EB (TFEB) and various members of the FOXO protein family, goes beyond the scope of this review and can be found in Pietrocola et al. (2013). On the other hand, it seems that nutrient deprivationdriven autophagy critically relies on the deacetylation of a nuclear pool of LC3, resulting in its redistribution to the cytoplasm and maturation (Huang et al., 2015). Finally, the nucleus can also be substrate of selective autophagic responses. In yeast, this process has been dubbed ‘‘piecemeal microautophagy of the nucleus’’ or ‘‘nucleophagy,’’ and it occurs when cells are rapidly dividing (perhaps as a consequence of DNA replication stress), especially in nutrientdeprived growth media (Roberts et al., 2003). Often, yeast nucleophagy involves a direct contact between the nucleus and the vacuole (which operates as a lysosome) and therefore does not constitute a bona fide instance of macroautophagy (Farre´ et al., 2009). However, at least in some settings, yeast cells can degrade portions of the nucleus, nucleolus, and perinuclear ER (see below) upon the activation of an Atg39-dependent macroautophagic response (Mochida et al., 2015). In mammalian cells, the autophagic machinery has been involved in the selective degradation of chromatin fragments budding off the nucleus as a consequence of chemical or senescence-associated DNA damage (Ivanov et al., 2013; Lan et al., 2014). Indeed, damaged DNA accumulates in the cytosol of non-phagocytic cells lacking ATG5 or the lysosomal deoxyribonuclease II (DNASE2), as well as in cells treated with bafilomycin A1 (which inhibits the fusion of autophagosomes with lysosomes) (Lan et al., 2014). Moreover, the histone content of human fetal lung IMR90 diploid fibroblasts driven into senescence by oncogene expression progressively decreases, a process that can be blocked with bafilomycin A1 (Ivanov et al., 2013). Interestingly, the autophagic degradation of damaged chromatin fragments

released in the cytosol of (pre-)senescent cells may explain why autophagy appears to be required for the acquisition of a full-blown senescent phenotype (Young et al., 2009). In mammalian cells, autophagy also contributes to the disposal of micronuclei, i.e., small nuclei that arise in cells dividing asymmetrically owing to defects in the mitotic apparatus (Vitale et al., 2011). Similar to damaged chromatin fragments, micronuclei destined to autophagic degradation exhibit signs of an ongoing DNA damage response and partial degeneration of the nuclear envelope, and are decorated by p62 (Rello-Varona et al., 2012). The degree of DNA damage may explain why some micronuclei are specifically destined to autophagic degradation, while others (within the same cells) are not. Taken together, these observations indicate that the nucleus not only initiates non-selective autophagy as a consequence of DNA damage, but also can drive organelle-selective autophagic responses (at least in specific circumstances). Moreover, the nucleus plays a key role in autophagy driven by other cellular compartments, via transcriptional and post-translational mechanisms. Initiation of Autophagy by the Plasma Membrane Owing to its privileged interaction with the extracellular microenvironment, the plasma membrane (PM) initiates adaptive autophagic responses to cell-impermeant inducers of stress. For illustrative purposes, PM-driven autophagy can be subdivided into two main categories: (1) ‘‘signal ON’’ autophagic responses, which are ignited by the binding of specific ligands to PM receptors that are normally inactive, and (2) ‘‘signal OFF’’ autophagic responses, which occur when the availability of ligands for PM that are normally active falls below a specific threshold. ‘‘Signal ON’’ PM-Driven Autophagy Several PM receptors have been ascribed with the ability to respond to their ligands by initiating adaptive autophagic responses that attempt to preserve cellular or organismal homeostasis, including: (1) so-called ‘‘death receptors’’; (2) G protein-coupled receptors (GPCRs); (3) pathogen receptors; and (4) Toll-like receptors (TLRs). Tumor necrosis factor receptor superfamily, member 1A (TNFRSF1A, best known as TNFR1) and TNFRSF10A (best known as TRAILR1) are two prototypical death receptors, i.e., PM receptors that trigger caspase-dependent or -independent forms of RCD upon ligand binding (Galluzzi et al., 2012e). TNFR1 and TRAILR1 also share the ability to promote autophagy in response to ligand binding, a process that has been linked to the activation of JNK1 (He et al., 2012; Keller et al., 2011) or upstream components of the NF-kB signaling machinery (Criollo et al., 2010, 2011). Notably, other death receptors like FAS (also known as CD95) seem unable to trigger autophagy in the presence of their ligands, even though they efficiently activate JNK1 and NF-kB (Latinis and Koretzky, 1996; Ponton et al., 1996). The reasons underlying this apparent discrepancy have not yet been uncovered. Many GPCRs that are normally inactive initiate adaptive autophagy in response to increases in the extracellular availability of their ligands. These include (but are not limited to): (1) free fatty acid receptor 1 (FFAR1), FFAR2, FFAR3, and FFAR4, which Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc. 527

Molecular Cell

Review are activated by free fatty acids; (2) glucagon receptor (GCGR), which responds to glucagon; (3) cholinergic receptor, muscarinic 3 (CHRM3), which is activated by acetylcholine; (4) adrenoceptor beta 2, surface (ADRB2), which is engaged by norepinephrine; and (5) purinergic receptor P2Y, G protein-coupled, 13 (P2RY13), which is responsive to ADP (Wauson et al., 2014). The signal transduction pathways that relay the liganddependent activation of these receptors to the autophagic machinery are heterogeneous but generally involve elevations in the intracellular levels of inositol-1,4,5,-triphosphate and diacylglycerol, or cAMP, which de facto stimulate autophagy via the AMPK/MTORC1 hub (Wauson et al., 2014). Besides operating as a cofactor for the inactivation of the complement cascade, CD46 acts as a receptor for measles virus, human herpesvirus 6 (HHV-6), group A Streptococcus pyogenes, and some variants of Neisseria spp. (Yamamoto et al., 2013). The engagement of CD46 is sufficient to promote a canonical autophagic response dependent on the interaction between the VPS34 complex and the scaffold protein Golgi-associated PDZ and coiled-coil motif containing (GOPC) (Joubert et al., 2009). Chemokine (C-X-C motif) receptor 4 (CXCR4) is not only involved in chemokine signaling, but also constitutes one of the two coreceptors required for lymphotropic human immunodeficiency virus 1 (HIV-1) to infect CD4+ cells (Deng et al., 1996). HIV-1 envelope glycoproteins (i.e., gp120 and gp41) expressed by infected cells can bind CD4/CXCR4 in non-infected bystander cells, hence triggering canonical (BECN1- and ATG7-dependent) autophagy via a mechanism that does not depend on CD4 and CXCR4 signaling (Espert et al., 2006). Interestingly, such a PM-driven autophagic response has been etiologically linked to the demise of non-infected CD4+ T cells exposed to gp41 (Espert et al., 2006), hence constituting a bona fide instance of autophagic cell death (Galluzzi et al., 2015a). TLRs constitute a peculiar class of PRRs expressed at the PM or in the endosomal compartment (Gay et al., 2014). TLRs respond to conserved microbial products or endogenous danger signals (i.e., factors that are exposed or released by cells experiencing stress) by initiating a signal transduction cascade that eventually results in the production of type I IFN and inflammatory cytokines (Gay et al., 2014). In addition, various TLRs have been shown to initiate adaptive autophagic responses upon ligand binding. The human TLRs that are most efficient at inducing autophagy are TLR3 (an endosomal TLR responsive to double-stranded RNA), TLR4 (a TLR expressed at the PM that responds to bacterial lipopolysaccharide as well as to endogenous high mobility group box 1), and TLR7 (an endosomal TLR responsive to single-stranded RNA of microbial origin) (Delgado et al., 2008). This said, virtually all human TLRs have been ascribed with some autophagy-inducing activity (Delgado et al., 2008). The activation of autophagy by TLRs appears to rely on common transducers of TLR-elicited signals, like myeloid differentiation primary response 88 (MYD88) and Toll-like receptor adaptor molecule 1 (TICAM1), possibly owing to their ability to disrupt the BCL2/BECN1 interaction (Shi and Kehrl, 2008). Notably, TLR-driven autophagic responses may serve a dual purpose: (1) they may support the xenophagic removal of invading pathogens, hence favoring the maintenance of cellular homeostasis, and/or (2) they may promote antigen presentation, 528 Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc.

hence favoring pathogen-specific immune responses aimed at the preservation of organismal homeostasis (Ma et al., 2013). ‘‘Signal OFF’’ PM-Driven Autophagy Various PM receptors lose the ability of repressing the autophagic machinery in the absence of their ligands. These include (but perhaps are not limited to): (1) growth factor receptors, (2) some GPCRs, and (3) (at least some) so-called ‘‘dependence receptors.’’ Multiple growth factor receptors expressed at the PM share the ability to initiate a signal transduction cascade involving Class I PI3K that impinges on the activation of AKT1 (Manning and Cantley, 2007). Active AKT1 phosphorylates many substrates including TSC2, eventually resulting in the derepression of MTORC1 (Manning and Cantley, 2007), and BECN1, favoring its sequestration by 14-3-3ε (Wang et al., 2012). Moreover, growth factor signaling is required for the normal expression of various glucose transporters, including (but not limited to) solute carrier family 2 (facilitated glucose transporter), member 1 (SLC2A1, best known as GLUT1), and SLC2A4 (best known as GLUT4) (Leto and Saltiel, 2012). Finally, specific growth factor receptors, such as the epidermal growth factor receptor (EGFR), can directly phosphorylate BECN1, thereby favoring its inactivating interaction by BCL2 and RUBICON (Wei et al., 2013). Thus, in the presence of normal amounts of growth factors, autophagy is actively suppressed as cells receive adequate nutrient supplies, MTORC1 is activated, and BECN1 repressed. This implies that growth factor deprivation stimulates autophagy by (1) limiting the cellular intake of glucose via GLUT1 and GLUT4 (which results in AMPK-dependent MTORC1 inhibition, see below) (Leto and Saltiel, 2012); (2) unleashing the MTORC1-inhibitory activity of TSC2 (Inoki et al., 2002); and (3) allowing for the derepression of BECN1 (Wang et al., 2012; Wei et al., 2013). G protein-coupled receptor, class C, group 6, member A (GPRC6A), calcium-sensing receptor (CASR), gamma-aminobutyric acid B receptor 1 (GABBR1), heteromeric taste receptors, and various metabotropic glutamate receptors, all of which are activated by one or several amino acids, share the ability to repress autophagy in the presence of their ligands (Wauson et al., 2014). The molecular cascades connecting these GPCRs to the autophagic machinery are not completely understood but have been proposed to involve G proteins that provoke a decrease in the intracellular levels of cAMP, resulting in the suppression of AMPK activity (Wauson et al., 2014). In addition, heteromeric taste receptors have been shown to repress autophagy via a Ca2+-dependent mechanism that directly involves amino acid sensing by MTORC1 (Wauson et al., 2012). Patched 1 (PTCH1) and DCC are prototypical dependence receptors, i.e., PM receptors that actively promote RCD in the absence of their ligands (Mehlen and Bredesen, 2011). Both sonic hedgehog (SHH) and netrin 1 (NTN1), the main ligands of PTCH1 and DCC, respectively, actively suppress autophagy, implying that their withdrawal favors the initiation of adaptive autophagic responses (Bouhidel et al., 2015; Jimenez-Sanchez et al., 2012). The molecular mechanisms whereby unoccupied dependence receptors trigger autophagy, however, remain to be elucidated. Moreover, it is not clear whether such autophagic responses attempt to counteract or etiologically contribute to dependence receptor-driven RCD.

Molecular Cell

Review Figure 3. Initiation of Autophagy by the Plasma Membrane (A) Various plasma membrane receptors promote autophagy upon ligand binding. These include (but are not limited to): death receptors that activate JNK1 and the NF-kB signal transduction pathway, such as TNFR1; various Toll-like receptors (TLRs), encompassing TLR3 (which responds to doublestranded RNA, dsRNA), TLR4 (which responds to bacterial lipopolysaccharide, LPS), and TLR7 (which responds to single-stranded RNA, ssRNA); some pathogen receptors, like CD46 and chemokine (C-X-C motif) receptor 4 (CXCR4); and several G protein-coupled receptors (GPCRs). (B) Other plasma membrane receptors tonically suppress autophagy in the presence of their ligands, implying that ligand withdrawal drives autophagic responses. This applies to: multiple growth factor receptors, including the epidermal growth factor (EGR) receptor (EGFR) and the insulin receptor (INSR); dependence receptors, including deleted in colorectal carcinoma (DCC), which responds to netrin 1 (NTN1), and patched 1 (PTCH1), which responds to sonic hedgehog (SHH); and various amino acid-sensing receptors, like T1R1/T1R3 heteromeric taste receptors and G protein-coupled receptor, class C, group 6, member A (GPRC6A). Both ‘‘signal ON’’ (A) and ‘‘signal OFF’’ (B) autophagic responses generally (but not always) involve AMPK activation coupled to mechanistic target of rapamycin (MTOR) complex I (MTORC1) inhibition or the release of beclin 1 (BECN1) from inhibitory interactions with other proteins. ACh, acetylcholine; AKT1, v-akt murine thymoma viral oncogene homolog 1; BCL2, B cell CLL/lymphoma 2; cAMP, cyclic AMP; DAG, diacylglycerol; GCG, glucagon; GLUT1 (official name SLC2A1), solute carrier family 2 (facilitated glucose transporter), member 1; GLUT4 (official name SLC2A4), solute carrier family 2 (facilitated glucose transporter), member 4; HHV-6, human herpesvirus 6; HIV-1, human immunodeficiency virus 1; INS, insulin; MYD88, myeloid differentiation primary response 88; NE, norepinephrine; PI3K, phosphatidylinositol 3-kinase; PtdIns(3,4,5)P3, phosphatidylinositol (3,4,5)-trisphosphate; PtdIns3P, phosphatidylinositol 3-phosphate; TICAM1, Toll-like receptor adaptor molecule 1; TNF, tumor necrosis factor alpha; TSC, tuberous sclerosis complex; VPS34 (official name PIK3C3), phosphatidylinositol 3-kinase, catalytic subunit type 3.

Irrespective of these unknowns, the aforementioned observations suggest that the PM is a privileged site where an increased demand for autophagic functions or products can be translated into a signal for the initiation of autophagy (Figure 3). To the best of our knowledge, no instances of PM-specific autophagy have been reported so far. Initiation of Autophagy by the Reticular System Together, the ER and Golgi apparatus (GA) form a vesicular system that not only is responsible for the synthesis, post-translational modification, and folding of proteins destined to secretion or insertion in the PM, but also plays a key role in the maintenance of intracellular Ca2+ homeostasis and in the detoxification of xenobiotics (Brandizzi and Barlowe, 2013). Herein, we refer to this functional unit as the reticular system. Importantly, the reticular system (encompassing endosomes) is commonly viewed as one of the major sources of autophagic membranes (Lamb et al., 2013). Both the ER and GA are very sensitive to physical and chemical conditions that provoke protein unfolding, including (but not limited to) hyperthermia, oxidative stress, and various molecules, like saturated fatty acids, thapsigargin, an inhibitor of various members of the sarco/endoplasmic reticulum Ca2+-

ATPase (SERCA) family, and tunicamycin, an inhibitor of N-glycosylation (Hetz, 2012). Unfolded proteins accumulating in the ER lumen competitively bind to the chaperone glucoseregulated protein, 78 kDa (GRP78), hence displacing it from inhibitory interactions with eukaryotic initiation factor 2 alpha (EIF2a) kinase 3 (EIF2AK3, best known as PERK), endoplasmic reticulum to nucleus signaling 1 (ERN1, best known as IRE1a), and activating transcription factor 6 (ATF6). The derepression of PERK, IRE1a, and ATF6 collectively sets off the so-called unfolded protein response (UPR) (Hetz, 2012). In the adaptive phase, the UPR attempts to re-establish reticular homeostasis via both post-translational (rapid) and transcriptional (delayed) mechanisms (Hetz, 2012). On the one hand, PERK promotes a generalized arrest in Cap-dependent mRNA translation by phosphorylating EIF2a on S51 (Hetz, 2012). Moreover, IRE1a can catalyze the degradation of specific mRNAs localized at the external surface of the ER, a process known as ‘‘regulated IRE1a-dependent decay’’ (Hetz, 2012). Both of these functions are expected to rapidly relieve (at least to some extent) the intrareticular burden of unfolded proteins, hence favoring the restoration of homeostasis. On the other hand, both spliced X-box binding protein 1 (XBP1s, whose synthesis depends on IRE1a) and ATF6 transactivate GRP78 and Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc. 529

Molecular Cell

Review other genes coding for factors involved in protein (re)folding, hence increasing the ability of the ER to handle unfolded proteins in a delayed manner (Hetz, 2012). Notably, the translation of GRP78-coding mRNAs is not affected by EIF2a phosphorylation, as it involves an internal ribosomal entry site (IRES) (Hetz, 2012). Thus, ATF6 and XBP1s can drive the synthesis of reticular chaperones even upon the establishment of the UPR. Importantly, PERK, IRE1a, and ATF6 are all intimately connected to the autophagic machinery. Cells expressing a nonphosphorylatable mutant of EIF2a (as well as cells depleted of PERK or EIF2AK2, another EIF2a kinase best known as PKR) are unable to respond normally to several inducers of autophagy. These include not only viral infection and oncogene activation (both of which trigger the UPR), but also nutrient deprivation and saturated fatty acids (Hart et al., 2012; Shen et al., 2012; Tallo´czy et al., 2002). At least in part, this reflects the ability of the PERK/EIF2a signaling pathway to selectively promote the synthesis of ATF4 (Hetz, 2012). Indeed, active ATF4 transactivates many genes involved in protein folding, as well as several genes coding for components of the autophagic machinery including ULK1, ATG5, ATG12, LC3, SESN2, and various BH3-only proteins (B’chir et al., 2013). Besides operating as an endonuclease, IRE1a can acquire kinase functions and catalyze the activating phosphorylation of JNK1 (hence promoting autophagy upon the derepression of BECN1) (Hetz, 2012). This signal transduction pathway depends on the adaptor TNFR-associated factor 2 (TRAF2) as well as on mitogen-activated protein kinase kinase kinase 5 (MAP3K5, best known as ASK1) (Hetz, 2012). Finally, ATF6 transactivates several genes other than GRP78, including that coding for the BECN1 activator DAPK1 (Gade et al., 2012). These observations explain why adaptive autophagy near-to-invariably accompanies ER stress responses. The ER also initiates autophagy in response to stimuli that perturb Ca2+ homeostasis. The release of Ca2+ ions into the cytosol causes the activation of several kinases, including (but not limited to) DAPK1 (Cohen et al., 1997) protein kinase C theta (PRKCQ) (Sakaki et al., 2008) and calcium/calmodulin-dependent protein kinase kinase 2 beta (CAMKK2, best known as CAMKKb) (Høyer-Hansen et al., 2007). Whereas the mechanisms whereby DAPK1 and CAMKKb promote autophagy are clear (i.e., by derepressing BECN1 and by activating AMPK, respectively), how PRKCQ does so remains elusive. Notably, ER-localized BCL2 inhibits autophagic responses to nutrient deprivation and other stimuli by limiting the amount of Ca2+ ions that is available for release (Høyer-Hansen et al., 2007). Moreover, reticular BCL2 promotes the ability of inositol 1,4,5trisphosphate receptor, type 1 (ITPR1, an agonist-operated Ca2+ channel of the ER) and ITPR3 to sequester (hence inhibiting) BECN1 (Criollo et al., 2007; Vicencio et al., 2009). The latter mechanism efficiently inhibits nutrient deprivation-driven, but not ER stress-driven, autophagy (Criollo et al., 2007) for hitherto undiscovered reasons. Interestingly, it seems that BECN1 regulates the Ca2+ channel activity of ITPR3 in the course of adaptive responses to ER stress (Decuypere et al., 2011). The links between the autophagy-promoting activity of Ca2+ fluxes, and the autophagy-inhibitory functions of ITPR1 and ITPR3 have not yet been precisely elucidated. 530 Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc.

While saturated fatty acids (like palmitate) promote canonical, BECN1-dependent autophagic responses involving JNK1 and AMPK activation as well as PtdIns3P synthesis (Shen et al., 2012), their unsaturated counterparts (like oleate) fail to do so. Rather, oleate promotes a non-canonical variant of autophagy that does not require BECN1, VPS34, or the generation of PtdIns3P (Niso-Santano et al., 2015). Interestingly, such a non-canonical autophagic response manifests in BECN1-deficient human cells, Becn1+/ mice, Datg6 Saccharomyces cerevisiae, and Caenorhabditis elegans larvae fed with bacteria bearing a construct for the downregulation of BEC-1 (meaning that it is evolutionarily conserved), and requires an intact GA (Niso-Santano et al., 2015). In line with this notion, brefeldin A (a GA poison) inhibits various manifestations of oleate-induced autophagy but does not affect autophagic responses to nutrient deprivation and palmitate (Niso-Santano et al., 2015). The molecular systems that initiate oleate-driven autophagy at the GA remain to be characterized. Interestingly, the GA has also been attributed a prominent role in the tonic suppression of autophagy, reflecting the constitutive inhibitory interaction between BECN1 and the GA protein GLIPR2 (Shoji-Kawata et al., 2013). So far, only an artificial peptide corresponding to the BECN1 domain that binds Nef, an HIV-1 protein with autophagy-inhibitory functions, has been shown to specifically release BECN1 from GLIPR2, causing its redistribution to other subcellular compartments and activation (Shoji-Kawata et al., 2013). Future studies will clarify whether this regulatory node is implicated in physiologically relevant instances of adaptive autophagy. The ER can also be selectively taken up by autophagosomes and degraded, a response that has been dubbed ‘‘reticulophagy’’ or ‘‘ER-phagy.’’ As early as in 1973, Bolender and colleagues reported that hepatocytes treated with phenobarbital exhibit a smooth ER that is significantly enlarged, a phenotype that regresses upon phenobarbital withdrawal along with a massive increase in the number and volume of ER-containing autophagic vacuoles (Bolender and Weibel, 1973). In yeast cells, reticulophagy has been proposed to counteract the expansion of the ER associated with UPRs (Bernales et al., 2006). In many instances, yeast reticulophagy does not require several core components of the autophagic machinery and does not involve autophagosomes (Schuck et al., 2014), de facto constituting an instance of microautophagy. However, recent findings suggest that (at least under some circumstances) yeast cells can specifically deliver portions of the ER to lysosomes via autophagosomes (Mochida et al., 2015). This process involves two ERresident receptors, i.e., Atg39 and Atg40, which specifically tag for degradation the perinuclear and cortical ER, respectively (Mochida et al., 2015). In mammalian cells, reticulophagy mainly relies on family with sequence similarity 134, member B (FAM134B), an Atg40 functional counterpart with LC3- and GABARAP-binding ability (Khaminets et al., 2015). To the best of our knowledge, no cases of GA-specific autophagy have been described so far. Taken together, these observations indicate that the reticular system initiates both selective and non-selective forms of autophagy upon the increased demand of autophagic functions or products that accompanies UPRs (Figure 4).

Molecular Cell

Review Figure 4. Initiation of Autophagy by the Reticular System (A) The accumulation of misfolded proteins in the endoplasmic reticulum (ER) lumen results in the displacement of glucose-regulated protein, 78 kDa (GRP78) from inhibitory interactions with PERK, IRE1a, and activating transcription factor 6 (ATF6), which collectively drive the unfolded protein response (UPR). PERK phosphorylates eukaryotic initiation factor 2 alpha (EIF2a), promoting a generalized arrest in Cap-dependent translation coupled to the enhanced translation of internal ribosome entry site (IRES)-containing mRNAs, such as the ATF4-coding mRNA. ATF4 transactivates several genes involved in autophagic responses. Besides operating as an endonuclease, IRE1a can directly phosphorylate JNK1, hence favoring the derepression of beclin 1 (BECN1) upon the JNK1-mediated phosphorylation of B cell CLL/lymphoma 2 (BCL2). Among various transcriptional targets, ATF6 transactivates the gene encoding death-associated protein kinase 1 (DAPK1). Similar to JNK1, DAPK1 favors the release of BECN1 from inhibitory interactions with BCL2. (B) The transfer of Ca2+ ions from the ER to the cytosol causes the activation of several kinases, including DAPK1, CAMKKb, and protein kinase C theta (PRKCQ). CAMKKb promotes autophagy upon AMPK activation. The precise mechanisms though which PRKCQ stimulates autophagic responses remain elusive. ER-localized BCL2 inhibits autophagy by limiting the amounts of Ca2+ ions available for release and by favoring the sequestration of BECN1 by inositol 1,4,5-trisphosphate receptors (ITPRs). (C) Unsaturated fatty acids drive autophagic responses that require an intact Golgi apparatus (GA). The molecular mechanisms underlying unsaturated fatty aciddriven autophagy, however, have not yet been characterized. Similar to BCL2, GLI pathogenesis-related 2 (GLIPR2), a protein of the GA membrane, sequesters BECN1 in inactive supramolecular complexes. An artificial BECN1-derived peptide can efficiently dissociate these complexes, causing autophagic responses that proceed along with the redistribution of BECN1 to other subcellular compartments. R, ribosome.

Initiation of Autophagy by the Cytosol At least in part, common homeostatic perturbations like glucose deprivation and hypoxia initiate autophagy in the cytosol. In both of these scenarios, the intracellular concentration of ATP drops (either as a consequence of primary substrate deprivation or upon the impairment of mitochondrial functions), correlating with the accumulation of the AMPK activators ADP, AMP, and cAMP and hence with the induction of a canonical, non-selective autophagic response (Inoki et al., 2012). In addition, at least in some cells, a cytosolic system responds to hypoxia by triggering a PINK1- and PARK2-independent mitophagic response (Zhang et al., 2008). This relies on the oxygen-sensing capacity of the heterodimeric transcription factor hypoxia-inducible factor 1 (HIF1). In normoxic conditions, several proline residues of the a subunit of HIF1 are hydroxylated, allowing the E3 ubiquitin ligase von Hippel-Lindau, tumor suppressor (VHL) to target it to proteasomal degradation (Maxwell et al., 1999). Conversely, when oxygen availability decreases, HIF1a accumulates and binds to the constitutively expressed b subunit, hence forming a functionally active heterodimer (Maxwell et al., 1999). Besides transactivating several genes involved in glycolysis and angiogenesis, HIF1 controls the synthesis of two BH3-only proteins, namely BCL2/adenovirus E1B 19 kDa interacting protein 3 (BNIP3) and BNIP3-like (BNIP3L) (Bellot et al., 2009). Both these proteins not only displace BECN1 from inhibitory interactions with Bcl2 family members (Mazure and Pouysse´gur, 2009), but also directly bind to LC3 (Liu et al., 2014). The optimal disposal of mitochondria in hypoxic cells also requires the interaction between LC3 and dephosphorylated FUN14 domain containing 1 (FUNDC1) (Chen et al., 2014). Thus, BNIP3, BNIP3L, and

FUNDC1 operate as autophagic receptors in response to hypoxia. Notably, the synthesis of BNIP3L (and FUNDC1) is also controlled at the post-transcriptional level, owing to a BNIP3Ltargeting (and FUNDC1-targeting) microRNA, i.e., miR-137, that is downregulated by hypoxia (Li et al., 2014). These observations exemplify the mechanisms underlying the activation of nonselective or specific autophagic responses by cytosolic stress sensors. The cytosol is also the site of initiation of autophagic responses that specifically target non-organellar entities. ‘‘Aggrephagy’’ is the term commonly employed to indicate the autophagic removal of (redox-active) aggregates of ubiquitinated proteins. Aggrephagy, which is particularly important for the avoidance of neurodegeneration (Harris and Rubinsztein, 2012) and for the preservation of genomic integrity (Galluzzi et al., 2015d), relies on various autophagic adaptors, including p62, optineurin (OPTN), and the PtdIns3P-binding protein WD repeat and FYVE domain containing 3 (WDFY3, best known as ALFY) (Lamark and Johansen, 2012). ‘‘Lipophagy’’ refers to the selective autophagic degradation of cytosolic lipid droplets (Singh et al., 2009). Besides contributing to hormone-driven lipolysis (Lizaso et al., 2013) as well as lipolysis driven by nutrient deprivation (Singh et al., 2009), lipophagy represents an adaptive response to the potentially toxic accumulation of neutral lipids in the cytosol, as it occurs in the course of hepatic steatosis (Eid et al., 2013). The precise molecular mechanisms whereby lipid droplets are specifically degraded in the course of lipophagy remain unclear. Moreover, recent findings suggest that lipid droplets and autophagy are interconnected in a very complex manner, as (1) neutral lipid droplets have been suggested to Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc. 531

Molecular Cell

Review contribute to autophagosome biogenesis (Dupont et al., 2014) and (2) autophagy has been shown to replenish lipid droplets with fatty acids in starved cells (Rambold et al., 2015). Further experiments are therefore required to clarify the actual role of lipophagy and autophagy in cellular lipid metabolism. ‘‘Xenophagy’’ is the expression coined to indicate the autophagic degradation of intracellular bacteria escaping lysosomes and cytoplasmic viruses (Knodler and Celli, 2011). Xenophagy is a major mechanism of innate immunity, and it relies on various PRRs (e.g., NOD2), as well as on many autophagic adaptors, including p62, OPTN, calcium binding and coiled-coil domain 2 (CALCOCO2, best known as NDP52), and tripartite motif containing 5 (TRIM5) (Chauhan et al., 2015; Mandell et al., 2014). Finally, the term ‘‘ribophagy’’ refers to the specific autophagic disposal of ribosomes triggered by nitrogen deprivation (Kraft et al., 2008). Whereas in yeast this process appears to require a multiprotein complex with ubiquitin protease activity (Kraft et al., 2008), it remains to be clarified whether mammalian instances of ribophagy exist and how they are regulated. Taken together, these observations indicate that several autophagic responses driven by an accrued offer of autophagic substrates are initiated in the cytosol. Initiation of Autophagy by Other Subcellular Compartments Lysosomes Lysosomes are central executioners of autophagy, as they ensure the degradation of autophagosomal cargoes. In line with this notion, specific transcriptional responses such as that orchestrated by TFEB simultaneously control the expression of several genes involved in autophagic responses and lysosomal biogenesis (Settembre et al., 2011). In line with this notion, socalled lysosomotropic toxins (i.e., chemicals that selectively accumulate in lysosomes and cause their generalized disruption) inhibit, rather than stimulate, autophagic responses (Kroemer and Ja¨a¨ttela¨, 2005). However, the disruption of individual lysosomes, which can be achieved with a lysosomotropic photosensitizer coupled to targeted illumination, reportedly drives a lysosome-specific variant of autophagy that efficiently removes permeabilized organelles before they trigger RCD (Hung et al., 2013). This process, which has been dubbed ‘‘lysophagy,’’ involves the ubiquitination of damaged lysosomes, followed by their recognition by p62 and their uptake by forming autophagosomes (Hung et al., 2013). Lysophagy is important for the maintenance of lysosomal homeostasis in vivo (Maejima et al., 2013). In particular, it significantly limits tissue damage in the course of hyperuricemic nephropathy, a type of acute kidney injury associated with lysosomal disruption caused by the precipitation of uric acid and monosodium urate (Maejima et al., 2013). Lysosomes participate in the early detection of nutrient deprivation and in the consequent activation of an autophagic response. In particular, the outer surface of the lysosomal membrane constitutes a privileged site for the detection of amino acid repletion by the so-called ‘‘Ragulator,’’ a multiprotein complex with guanine nucleotide exchange factor (GEF) activity that controls MTORC1 activation by Ras homolog enriched in brain (RHEB) (Bar-Peled et al., 2012). In the presence of amino acids, RAG GTPases are recruited to the lysosomal surface and acti532 Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc.

vated by the Ragulator, resulting in optimal MTORC1 signaling and autophagy inhibition (Sancak et al., 2010). Artificially targeting MTORC1 to the lysosomal surface suffices indeed to activate it irrespective of amino acid (but not RHEB) availability (Sancak et al., 2010). Notably, the pool of amino acids sensed by the Ragulator is not the cytosolic one, but the luminal one (Zoncu et al., 2011). This reflects the ability of the vacuolar H+-ATPase (v-ATPase) to interact with the Ragulator, hence controlling its activity, in an amino acid-sensitive manner (Zoncu et al., 2011). This function of the Ragulator critically depends on a solute carrier family 38, member 9 (SLC38A9), a lysosomal amino acid transporter (Rebsamen et al., 2015; Wang et al., 2015). Recent data indicate that the v-ATPase-Ragulator complex is also required for the optimal activation of AMPK by glucose deprivation (Zhang et al., 2014). In these conditions, the v-ATPaseRagulator complex interacts with an LKB1- and axin 1 (AXIN1)containing complex, resulting not only in AMPK activation, but also in MTORC1 inhibition upon the AXIN1-dependent inactivation of RAG GTPases (Zhang et al., 2014). Thus, the lysosomal surface stands out as a privileged site of autophagy initiation by amino acid depletion, dwindling glucose levels, and (at least theoretically) MTORC1 inhibitors such as rapamycin. Peroxisomes The autophagic degradation of peroxisomes, which is commonly known as ‘‘pexophagy,’’ is among the first instances of organelle-specific autophagy ever described, and it involves both a microautophagic and a macroautophagic component (Farre´ et al., 2008). One of the most specific interventions to trigger pexophagy entails provoking an artificial expansion of the peroxisomal compartment with phthalates, followed by phthalate withdrawal (Iwata et al., 2006). Moreover, several stress conditions have been shown to stimulate pexophagy in yeast or mammalian cells, including hypoxia and nutrient deprivation (Hara-Kuge and Fujiki, 2008; Walter et al., 2014). However, why these general perturbations of homeostasis cause pexophagy in some cells and relatively non-specific instances of autophagy in others remains to be clarified. Irrespective of this incognita, pexophagy has been shown to involve the selective recognition of ubiquitinated peroxisomal proteins by the autophagic adaptors neighbor of BRCA1 gene 1 (NBR1) and ATG30 (Burnett et al., 2015; Deosaran et al., 2013). Other autophagic adaptors like p62 also participate in pexophagic responses, yet are not strictly required for them (Deosaran et al., 2013). Conversely, two distinct peroxisomal membrane proteins are indispensable for optimal pexophagy, namely, peroxisomal biogenesis factor 3 (PEX3) and PEX14 (Hara-Kuge and Fujiki, 2008; Yamashita et al., 2014). On the one hand, PEX3 provides a scaffold for the binding of ATG30, which is requisite for its activating phosphorylation (by a hitherto uncharacterized kinase) (Burnett et al., 2015). On the other hand, PEX14 interacts with lipidated LC3 via microtubules, further facilitating the recruitment of forming autophagosomes to peroxisomes destined to degradation (Hara-Kuge and Fujiki, 2008). Taken together, these observations suggest that lysosomes and peroxisomes can drive organelle-specific autophagic responses when in excess or damaged. Moreover, lysosomes play a key role in adaptive autophagy caused by an increased demand of metabolic intermediates.

Stimulus

Type

Note(s)

References

Mitochondria

hypoxia respiratory chain inhibitors uncouplers

mitophagy

stimulate mitophagy via the MPT/ PINK1/PARK2/VCP pathway

Chen et al., 2014; Geisler et al., 2010; Okatsu et al., 2015; Narendra et al., 2010; Tanaka et al., 2010

MAMPs

non-selective

stimulate autophagy via IRGM, NLRX1, and TUFM

Lei et al., 2012; Rowland and Voeltz, 2012; Chauhan et al., 2015

ROS

mitophagy

promote MPT by altering the levels of Bcl-2 proteins and regulate ATG functions

Galluzzi et al., 2012a; Scherz-Shouval et al., 2007

alkylating agents

non-selective

favor autophagy via PARP1/AMPK signaling

Ethier et al., 2012; Gibson and Kraus, 2012; Rodrı´guezVargas et al., 2012

CASP2 inhibitors

non-selective

involves AMPK activation and MTORC1 inhibition

Tiwari et al., 2014

DNA-damaging agents

non-selective nucleophagy

promotes autophagy via several signaling pathways impinging on AMPK and p53 activation

Alexander et al., 2010; Bieging et al., 2014; Chen et al., 2015; Figueiredo et al., 2013; Lan et al., 2014; Liang et al., 2013; Pietrocola et al., 2013; Sapkota et al., 2002; Wang et al., 2010

mitotic problems

nucleophagy

micronuclei destined to degradation exhibit signs of ongoing DDR and are decorated with p62

Rello-Varona et al., 2012; Vitale et al., 2011

ROS

non-selective

exacerbate DNA damage

Bouwman and Jonkers, 2012

senescence

nucleophagy

CCFs destined to disposal have signs of ongoing DDR and are tagged with p62

Ivanov et al., 2013; Young et al., 2009

death receptor ligands

non-selective

stimulate autophagy via JNK1 and components of the NF-kB signaling pathway

Criollo et al., 2010; Criollo et al., 2011; He et al., 2012; Keller et al., 2011

GCPR agonists

non-selective

involves PtdIns(3,4,5)P3 synthesis and increased cAMP levels

Wauson et al., 2014

pathogen docking

non-selective xenophagy

CD46 and CXCR4 promote canonical autophagic responses

Espert et al., 2006 Joubert et al., 2009

TLR agonists

non-selective xenophagy

promote autophagy via MYD88 or TICAM1

Delgado et al., 2008; Knodler and Celli, 2011; Ma et al., 2013; Shi and Kehrl, 2008

amino acid deprivation

non-selective

several GPCRs repress autophagy in response to extracellular amino acids

Wauson et al., 2012; Wauson et al., 2014

Nucleus

Plasma membrane

Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc. 533

Reticular system

DR ligand withdrawal

non-selective

unknown mechanism

Bouhidel et al., 2015; Jimenez-Sanchez et al., 2012

growth factor deprivation RTK inhibitors

non-selective

promote autophagy by reducing nutrient intake, limiting AKT1 signaling, and derepressing BECN1

Inoki et al., 2002; Leto and Saltiel, 2012; Manning and Cantley, 2007; Wang et al., 2012; Wei et al., 2013

disruption of Ca2+ homeostasis

non-selective

cytosolic Ca2+ promotes autophagy via DAPK1, CAMKKb, and PRKCQ

Cohen et al., 1997; Høyer-Hansen et al., 2007; Sakaki et al., 2008

ER stress

non-selective

trigger autophagy via PERK/EIF2a/ATF4, IRE1a/JNK1, and ATF6/DAPK1 signaling

B’chir et al., 2013; Gade et al., 2012; Hetz, 2012

unsaturated fatty acids

non-selective

non-canonical autophagic responses that require an intact GA

Niso-Santano et al., 2015

UPR responses to xenobiotic

reticulophagy

autophagy regulates the ER size when adaptive response are shut off

Bolender and Weibel, 1973; Bernales et al., 2006; Schuck et al., 2014 (Continued on next page)

Molecular Cell

Detection Site

Review

Table 1. Examples of Non-selective and Targeted Autophagic Responses Initiated at Specific Subcellular Compartments

534 Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc.

Table 1.

Continued

Detection Site

Stimulus

Type

Note(s)

References

Cytosol

bacteria viruses

xenophagy

relies on p62, OPTN, TRIM5, and NDP52

Chauhan et al., 2015; Knodler and Celli, 2011; Mandell et al., 2014

hypoxia

mitophagy

relies on the transactivation of BNIP3 and BNIP3L by HIF1, as well as on FUNDC1

Bellot et al., 2009; Chen et al., 2014 Li et al., 2014; Liu et al., 2014; Maxwell et al., 1999; Mazure and Pouysse´gur, 2009; Zhang et al., 2008

lipid droplets

lipophagy

contributes to hormone- and starvation-driven lipolysis

Eid et al., 2013; Singh et al., 2009 Lamark and Johansen, 2012

Lysosomes

Peroxisomes

protein aggregates

aggrephagy

relies on p62, OPTN, and ALFY

ribosomes

ribophagy

not yet observed in mammalian cells

Kraft et al., 2008

RNS ROS

non-selective

promote autophagy via an ATM/ LKB1 signaling pathway

Alexander et al., 2010; Tripathi et al., 2013

amino acid deprivation

non-selective

stimulates autophagy via Ragulator/ MTORC1 signaling

Bar-Peled et al., 2012; Rebsamen et al., 2015; Sancak et al., 2010; Wang et al., 2015; Zoncu et al., 2011

glucose deprivation

non-selective

stimulates autophagy via Ragulator/ LBK1/AMPK signaling

Zhang et al., 2014

lysosomotropic agents

lysophagy

involves p62

Maejima et al., 2013; Hung et al., 2013

responses to xenobiotic

pexophagy

involves PEX3, PEX14, NBR1, and ATG30

Burnett et al., 2015; Deosaran et al., 2013; Farre´ et al., 2008; Hara-Kuge and Fujiki, 2008; Yamashita et al., 2014

AKT1, v-akt murine thymoma viral oncogene homolog 1; ALFY (official name WDFY3), WD repeat and FYVE domain containing 3; AMPK (official name PRKA), protein kinase, AMP-activated; ATF, activating transcription factor; BECN1, beclin 1; BNIP3, BCL2/adenovirus E1B 19 kDa interacting protein 3; BNIP3L, BNIP3-like; CAMKKb (official name CAMKK2), calcium/calmodulin-dependent protein kinase kinase 2 beta; cAMP, cyclic AMP; CASP2, caspase 2; CCF, condensed chromatin fragment; CXCR4, chemokine (C-X-C motif) receptor 4; DAPK1, death-associated protein kinase 1; DDR, DNA damage response; DR, dependence receptor; EIF2a, eukaryotic initiation factor 2 alpha; ER, endoplasmic reticulum; FUNDC1, FUN14 domain containing 1; GA, Golgi apparatus; GPCR, G protein-coupled receptor; HIF1, hypoxia-inducible factor 1; IRE1a (official name ERN1), endoplasmic reticulum to nucleus signaling 1; IRGM, immunity-related GTPase family, M; JNK1 (official name MAPK8), mitogen-activated protein kinase 8; LKB1 (official name STK11), serine/threonine kinase 11; MAMP, microbe-associated molecular pattern; MPT, mitochondrial permeability transition; MTORC1, mechanistic target of rapamycin complex 1; MYD88, myeloid differentiation primary response 88; NBR1, neighbor of BRCA1 gene 1; NDP52 (official name CALCOCO2), calcium binding and coiled-coil domain 2; NLRX1, NLR family member X1; OPTN, optineurin; PARK2, parkin RBR E3 ubiquitin protein ligase; PARP1, poly(ADP-ribose) polymerase 1; PERK (official name EIF2AK3), eukaryotic initiation factor 2 alpha kinase 3; PEX, peroxisomal biogenesis factor; PINK1, PTEN induced putative kinase 1; PRKCQ, protein kinase C theta; PtdIns(3,4,5)P3, phosphatidylinositol (3,4,5)-trisphosphate; RNS, reactive nitrogen species; ROS, reactive oxygen species; RTK, receptor tyrosine kinase; TICAM1, Toll-like receptor adaptor molecule 1; TLR, Toll-like receptor; TRIM5, tripartite motif containing 5; TUFM, Tu translation elongation factor, mitochondrial; UPR, unfolded protein response; VCP, valosin containing protein.

Molecular Cell

Review

Molecular Cell

Review Conclusions and Perspectives The findings presented above indicate that most (if not all) subcellular compartments are provided with a set of molecular sensors that detect homeostatic perturbations and initiate adaptive programs aimed at coping with stress (Galluzzi et al., 2014a). Often, these programs are accompanied by an autophagic response, and this is essential for the restoration of homeostasis (Kroemer et al., 2010). We propose here that adaptive autophagy can be driven either by an increased offer of autophagic substrates or by an accrued demand for autophagic products or functions (Figure 1), correlating with the degree of selectivity of the response. Thus, relatively non-selective forms of autophagy can be triggered even by specific perturbations of homeostasis, if cells require autophagic products or functions to cope with stress. Conversely, when stress provokes (or corresponds to) an increase in the availability of specific autophagic substrates, these are disposed of via a highly targeted autophagic response (Table 1). Presumably, neither purely non-selective nor purely specific autophagic responses exist. On the one hand, indeed, cells require specific organelles and functions for survival, and these must be spared in the course of non-selective autophagy. This is the case for parts of the mitochondrial network in the course of nutrient deprivation-driven autophagy (Gomes et al., 2011). On the other hand, it is tempting to speculate (yet it remains to be formally demonstrated) that even highly targeted responses driven by the accumulation of autophagic substrates occur in the context of a metabolic rewiring that requires autophagic products and/or functions (Galluzzi et al., 2014d). Further insights into the molecular mechanisms whereby specific subcellular compartments elicit non-selective or specific autophagic responses are required to fully harness the promising therapeutic potential of autophagy modulators for the treatment of neurodegenerative and oncological disorders.

Antonioli, M., Albiero, F., Nazio, F., Vescovo, T., Perdomo, A.B., Corazzari, M., Marsella, C., Piselli, P., Gretzmeier, C., Dengjel, J., et al. (2014). AMBRA1 interplay with cullin E3 ubiquitin ligases regulates autophagy dynamics. Dev. Cell 31, 734–746. B’chir, W., Maurin, A.C., Carraro, V., Averous, J., Jousse, C., Muranishi, Y., Parry, L., Stepien, G., Fafournoux, P., and Bruhat, A. (2013). The eIF2a/ATF4 pathway is essential for stress-induced autophagy gene expression. Nucleic Acids Res. 41, 7683–7699. Bar-Peled, L., Schweitzer, L.D., Zoncu, R., and Sabatini, D.M. (2012). Ragulator is a GEF for the rag GTPases that signal amino acid levels to mTORC1. Cell 150, 1196–1208. Bellot, G., Garcia-Medina, R., Gounon, P., Chiche, J., Roux, D., Pouysse´gur, J., and Mazure, N.M. (2009). Hypoxia-induced autophagy is mediated through hypoxia-inducible factor induction of BNIP3 and BNIP3L via their BH3 domains. Mol. Cell. Biol. 29, 2570–2581. Bernales, S., McDonald, K.L., and Walter, P. (2006). Autophagy counterbalances endoplasmic reticulum expansion during the unfolded protein response. PLoS Biol. 4, e423. Bieging, K.T., Mello, S.S., and Attardi, L.D. (2014). Unravelling mechanisms of p53-mediated tumour suppression. Nat. Rev. Cancer 14, 359–370. Bolender, R.P., and Weibel, E.R. (1973). A morphometric study of the removal of phenobarbital-induced membranes from hepatocytes after cessation of threatment. J. Cell Biol. 56, 746–761. Bouhidel, J.O., Wang, P., Siu, K.L., Li, H., Youn, J.Y., and Cai, H. (2015). Netrin-1 improves post-injury cardiac function in vivo via DCC/NO-dependent preservation of mitochondrial integrity, while attenuating autophagy. Biochim. Biophys. Acta 1852, 277–289. Bouwman, P., and Jonkers, J. (2012). The effects of deregulated DNA damage signalling on cancer chemotherapy response and resistance. Nat. Rev. Cancer 12, 587–598. Brandizzi, F., and Barlowe, C. (2013). Organization of the ER-Golgi interface for membrane traffic control. Nat. Rev. Mol. Cell Biol. 14, 382–392. Burnett, S.F., Farre´, J.C., Nazarko, T.Y., and Subramani, S. (2015). Peroxisomal Pex3 activates selective autophagy of peroxisomes via interaction with the pexophagy receptor Atg30. J. Biol. Chem. 290, 8623–8631. Chauhan, S., Mandell, M.A., and Deretic, V. (2015). IRGM governs the core autophagy machinery to conduct antimicrobial defense. Mol. Cell 58, 507–521.

AUTHOR CONTRIBUTIONS V.S. and L.G. prepared the first draft of the manuscript, integrated inputs from co-authors, and revised the manuscript to address the issues raised by reviewers. V.S. prepared Table 1. J.M.B.-S.P. designed all figures under the supervision of L.G., and J.M.B.-S.P., V.I., and M.C.M. provided corrections and inputs to the manuscript. L.G. and G.K. conceived the manuscript and provided senior guidance to its preparation. ACKNOWLEDGMENTS G.K. is supported by the Ligue contre le Cancer (e´quipe labelise´e); Agence National de la Recherche (ANR) – Projets blancs; ANR under the frame of E-Rare-2, the ERA-Net for Research on Rare Diseases; Association pour la recherche sur le cancer (ARC); Cance´ropoˆle Ile-de-France; Institut National du Cancer (INCa); Fondation Bettencourt-Schueller; Fondation de France; Fondation pour la Recherche Me´dicale (FRM); the European Commission (ArtForce); the European Research Council (ERC); the LabEx Immuno-Oncology; the SIRIC Stratified Oncology Cell DNA Repair and Tumor Immune Elimination (SOCRATE); the SIRIC Cancer Research and Personalized Medicine (CARPEM); and the Paris Alliance of Cancer Research Institutes (PACRI). REFERENCES Alexander, A., Cai, S.L., Kim, J., Nanez, A., Sahin, M., MacLean, K.H., Inoki, K., Guan, K.L., Shen, J., Person, M.D., et al. (2010). ATM signals to TSC2 in the cytoplasm to regulate mTORC1 in response to ROS. Proc. Natl. Acad. Sci. USA 107, 4153–4158.

Chen, G., Han, Z., Feng, D., Chen, Y., Chen, L., Wu, H., Huang, L., Zhou, C., Cai, X., Fu, C., et al. (2014). A regulatory signaling loop comprising the PGAM5 phosphatase and CK2 controls receptor-mediated mitophagy. Mol. Cell 54, 362–377. Chen, J.H., Zhang, P., Chen, W.D., Li, D.D., Wu, X.Q., Deng, R., Jiao, L., Li, X., Ji, J., Feng, G.K., et al. (2015). ATM-mediated PTEN phosphorylation promotes PTEN nuclear translocation and autophagy in response to DNAdamaging agents in cancer cells. Autophagy 11, 239–252. Codogno, P., Mehrpour, M., and Proikas-Cezanne, T. (2012). Canonical and non-canonical autophagy: variations on a common theme of self-eating? Nat. Rev. Mol. Cell Biol. 13, 7–12. Cohen, O., Feinstein, E., and Kimchi, A. (1997). DAP-kinase is a Ca2+/calmodulin-dependent, cytoskeletal-associated protein kinase, with cell deathinducing functions that depend on its catalytic activity. EMBO J. 16, 998–1008. Criollo, A., Maiuri, M.C., Tasdemir, E., Vitale, I., Fiebig, A.A., Andrews, D., Molgo´, J., Dı´az, J., Lavandero, S., Harper, F., et al. (2007). Regulation of autophagy by the inositol trisphosphate receptor. Cell Death Differ. 14, 1029–1039. Criollo, A., Senovilla, L., Authier, H., Maiuri, M.C., Morselli, E., Vitale, I., Kepp, O., Tasdemir, E., Galluzzi, L., Shen, S., et al. (2010). The IKK complex contributes to the induction of autophagy. EMBO J. 29, 619–631. Criollo, A., Niso-Santano, M., Malik, S.A., Michaud, M., Morselli, E., Marin˜o, G., Lachkar, S., Arkhipenko, A.V., Harper, F., Pierron, G., et al. (2011). Inhibition of autophagy by TAB2 and TAB3. EMBO J. 30, 4908–4920. Decuypere, J.P., Welkenhuyzen, K., Luyten, T., Ponsaerts, R., Dewaele, M., Molgo´, J., Agostinis, P., Missiaen, L., De Smedt, H., Parys, J.B., and Bultynck,

Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc. 535

Molecular Cell

Review G. (2011). Ins(1,4,5)P3 receptor-mediated Ca2+ signaling and autophagy induction are interrelated. Autophagy 7, 1472–1489. Delgado, M.A., Elmaoued, R.A., Davis, A.S., Kyei, G., and Deretic, V. (2008). Toll-like receptors control autophagy. EMBO J. 27, 1110–1121. Deng, H., Liu, R., Ellmeier, W., Choe, S., Unutmaz, D., Burkhart, M., Di Marzio, P., Marmon, S., Sutton, R.E., Hill, C.M., et al. (1996). Identification of a major co-receptor for primary isolates of HIV-1. Nature 381, 661–666. Deosaran, E., Larsen, K.B., Hua, R., Sargent, G., Wang, Y., Kim, S., Lamark, T., Jauregui, M., Law, K., Lippincott-Schwartz, J., et al. (2013). NBR1 acts as an autophagy receptor for peroxisomes. J. Cell Sci. 126, 939–952. Diao, J., Liu, R., Rong, Y., Zhao, M., Zhang, J., Lai, Y., Zhou, Q., Wilz, L.M., Li, J., Vivona, S., et al. (2015). ATG14 promotes membrane tethering and fusion of autophagosomes to endolysosomes. Nature 520, 563–566. Dupont, N., Chauhan, S., Arko-Mensah, J., Castillo, E.F., Masedunskas, A., Weigert, R., Robenek, H., Proikas-Cezanne, T., and Deretic, V. (2014). Neutral lipid stores and lipase PNPLA5 contribute to autophagosome biogenesis. Curr. Biol. 24, 609–620. Egan, D.F., Shackelford, D.B., Mihaylova, M.M., Gelino, S., Kohnz, R.A., Mair, W., Vasquez, D.S., Joshi, A., Gwinn, D.M., Taylor, R., et al. (2011). Phosphorylation of ULK1 (hATG1) by AMP-activated protein kinase connects energy sensing to mitophagy. Science 331, 456–461.

Galluzzi, L., Pietrocola, F., Bravo-San Pedro, J.M., Amaravadi, R.K., Baehrecke, E.H., Cecconi, F., Codogno, P., Debnath, J., Gewirtz, D.A., Karantza, V., et al. (2015d). Autophagy in malignant transformation and cancer progression. EMBO J. 34, 856–880. Gay, N.J., Symmons, M.F., Gangloff, M., and Bryant, C.E. (2014). Assembly and localization of Toll-like receptor signalling complexes. Nat. Rev. Immunol. 14, 546–558. Geisler, S., Holmstro¨m, K.M., Skujat, D., Fiesel, F.C., Rothfuss, O.C., Kahle, P.J., and Springer, W. (2010). PINK1/Parkin-mediated mitophagy is dependent on VDAC1 and p62/SQSTM1. Nat. Cell Biol. 12, 119–131. Gibson, B.A., and Kraus, W.L. (2012). New insights into the molecular and cellular functions of poly(ADP-ribose) and PARPs. Nat. Rev. Mol. Cell Biol. 13, 411–424. Gomes, L.C., Di Benedetto, G., and Scorrano, L. (2011). During autophagy mitochondria elongate, are spared from degradation and sustain cell viability. Nat. Cell Biol. 13, 589–598. Hara-Kuge, S., and Fujiki, Y. (2008). The peroxin Pex14p is involved in LC3-dependent degradation of mammalian peroxisomes. Exp. Cell Res. 314, 3531–3541. Harris, H., and Rubinsztein, D.C. (2012). Control of autophagy as a therapy for neurodegenerative disease. Nat. Rev. Neurol. 8, 108–117.

Eid, N., Ito, Y., Maemura, K., and Otsuki, Y. (2013). Elevated autophagic sequestration of mitochondria and lipid droplets in steatotic hepatocytes of chronic ethanol-treated rats: an immunohistochemical and electron microscopic study. J. Mol. Histol. 44, 311–326.

Hart, L.S., Cunningham, J.T., Datta, T., Dey, S., Tameire, F., Lehman, S.L., Qiu, B., Zhang, H., Cerniglia, G., Bi, M., et al. (2012). ER stress-mediated autophagy promotes Myc-dependent transformation and tumor growth. J. Clin. Invest. 122, 4621–4634.

Espert, L., Denizot, M., Grimaldi, M., Robert-Hebmann, V., Gay, B., Varbanov, M., Codogno, P., and Biard-Piechaczyk, M. (2006). Autophagy is involved in T cell death after binding of HIV-1 envelope proteins to CXCR4. J. Clin. Invest. 116, 2161–2172.

He, C., and Levine, B. (2010). The Beclin 1 interactome. Curr. Opin. Cell Biol. 22, 140–149.

Ethier, C., Tardif, M., Arul, L., and Poirier, G.G. (2012). PARP-1 modulation of mTOR signaling in response to a DNA alkylating agent. PLoS ONE 7, e47978. Farre´, J.C., Manjithaya, R., Mathewson, R.D., and Subramani, S. (2008). PpAtg30 tags peroxisomes for turnover by selective autophagy. Dev. Cell 14, 365–376. Farre´, J.C., Krick, R., Subramani, S., and Thumm, M. (2009). Turnover of organelles by autophagy in yeast. Curr. Opin. Cell Biol. 21, 522–530. Figueiredo, N., Chora, A., Raquel, H., Pejanovic, N., Pereira, P., Hartleben, B., Neves-Costa, A., Moita, C., Pedroso, D., Pinto, A., et al. (2013). Anthracyclines induce DNA damage response-mediated protection against severe sepsis. Immunity 39, 874–884. Gade, P., Ramachandran, G., Maachani, U.B., Rizzo, M.A., Okada, T., Prywes, R., Cross, A.S., Mori, K., and Kalvakolanu, D.V. (2012). An IFN-g-stimulated ATF6-C/EBP-b-signaling pathway critical for the expression of Death Associated Protein Kinase 1 and induction of autophagy. Proc. Natl. Acad. Sci. USA 109, 10316–10321. Galluzzi, L., Kepp, O., and Kroemer, G. (2012a). Mitochondria: master regulators of danger signalling. Nat. Rev. Mol. Cell Biol. 13, 780–788. Galluzzi, L., Kepp, O., Trojel-Hansen, C., and Kroemer, G. (2012d). Mitochondrial control of cellular life, stress, and death. Circ. Res. 111, 1198–1207. Galluzzi, L., Vitale, I., Abrams, J.M., Alnemri, E.S., Baehrecke, E.H., Blagosklonny, M.V., Dawson, T.M., Dawson, V.L., El-Deiry, W.S., Fulda, S., et al. (2012e). Molecular definitions of cell death subroutines: recommendations of the Nomenclature Committee on Cell Death 2012. Cell Death Differ. 19, 107–120. Galluzzi, L., Bravo-San Pedro, J.M., and Kroemer, G. (2014a). Organellespecific initiation of cell death. Nat. Cell Biol. 16, 728–736. Galluzzi, L., Pietrocola, F., Levine, B., and Kroemer, G. (2014d). Metabolic control of autophagy. Cell 159, 1263–1276. Galluzzi, L., Bravo-San Pedro, J.M., Vitale, I., Aaronson, S.A., Abrams, J.M., Adam, D., Alnemri, E.S., Altucci, L., Andrews, D., Annicchiarico-Petruzzelli, M., et al. (2015a). Essential versus accessory aspects of cell death: recommendations of the NCCD 2015. Cell Death Differ. 22, 58–73.

536 Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc.

He, W., Wang, Q., Xu, J., Xu, X., Padilla, M.T., Ren, G., Gou, X., and Lin, Y. (2012). Attenuation of TNFSF10/TRAIL-induced apoptosis by an autophagic survival pathway involving TRAF2- and RIPK1/RIP1-mediated MAPK8/JNK activation. Autophagy 8, 1811–1821. Hetz, C. (2012). The unfolded protein response: controlling cell fate decisions under ER stress and beyond. Nat. Rev. Mol. Cell Biol. 13, 89–102. Hosokawa, N., Hara, T., Kaizuka, T., Kishi, C., Takamura, A., Miura, Y., Iemura, S., Natsume, T., Takehana, K., Yamada, N., et al. (2009). Nutrient-dependent mTORC1 association with the ULK1-Atg13-FIP200 complex required for autophagy. Mol. Biol. Cell 20, 1981–1991. Høyer-Hansen, M., Bastholm, L., Szyniarowski, P., Campanella, M., Szabadkai, G., Farkas, T., Bianchi, K., Fehrenbacher, N., Elling, F., Rizzuto, R., et al. (2007). Control of macroautophagy by calcium, calmodulin-dependent kinase kinase-beta, and Bcl-2. Mol. Cell 25, 193–205. Huang, R., Xu, Y., Wan, W., Shou, X., Qian, J., You, Z., Liu, B., Chang, C., Zhou, T., Lippincott-Schwartz, J., and Liu, W. (2015). Deacetylation of nuclear LC3 drives autophagy initiation under starvation. Mol. Cell 57, 456–466. Hung, Y.H., Chen, L.M., Yang, J.Y., and Yang, W.Y. (2013). Spatiotemporally controlled induction of autophagy-mediated lysosome turnover. Nat. Commun. 4, 2111. Inoki, K., Li, Y., Zhu, T., Wu, J., and Guan, K.L. (2002). TSC2 is phosphorylated and inhibited by Akt and suppresses mTOR signalling. Nat. Cell Biol. 4, 648–657. Inoki, K., Kim, J., and Guan, K.L. (2012). AMPK and mTOR in cellular energy homeostasis and drug targets. Annu. Rev. Pharmacol. Toxicol. 52, 381–400. Ivanov, A., Pawlikowski, J., Manoharan, I., van Tuyn, J., Nelson, D.M., Rai, T.S., Shah, P.P., Hewitt, G., Korolchuk, V.I., Passos, J.F., et al. (2013). Lysosome-mediated processing of chromatin in senescence. J. Cell Biol. 202, 129–143. Iwata, J., Ezaki, J., Komatsu, M., Yokota, S., Ueno, T., Tanida, I., Chiba, T., Tanaka, K., and Kominami, E. (2006). Excess peroxisomes are degraded by autophagic machinery in mammals. J. Biol. Chem. 281, 4035–4041. Jimenez-Sanchez, M., Menzies, F.M., Chang, Y.Y., Simecek, N., Neufeld, T.P., and Rubinsztein, D.C. (2012). The Hedgehog signalling pathway regulates autophagy. Nat. Commun. 3, 1200.

Molecular Cell

Review Jin, S.M., Lazarou, M., Wang, C., Kane, L.A., Narendra, D.P., and Youle, R.J. (2010). Mitochondrial membrane potential regulates PINK1 import and proteolytic destabilization by PARL. J. Cell Biol. 191, 933–942.

Lizaso, A., Tan, K.T., and Lee, Y.H. (2013). b-adrenergic receptor-stimulated lipolysis requires the RAB7-mediated autolysosomal lipid degradation. Autophagy 9, 1228–1243.

Joubert, P.E., Meiffren, G., Gre´goire, I.P., Pontini, G., Richetta, C., Flacher, M., Azocar, O., Vidalain, P.O., Vidal, M., Lotteau, V., et al. (2009). Autophagy induction by the pathogen receptor CD46. Cell Host Microbe 6, 354–366.

Ma, Y., Galluzzi, L., Zitvogel, L., and Kroemer, G. (2013). Autophagy and cellular immune responses. Immunity 39, 211–227.

Keller, C.W., Fokken, C., Turville, S.G., Lu¨nemann, A., Schmidt, J., Mu¨nz, C., and Lu¨nemann, J.D. (2011). TNF-alpha induces macroautophagy and regulates MHC class II expression in human skeletal muscle cells. J. Biol. Chem. 286, 3970–3980. Khaminets, A., Heinrich, T., Mari, M., Grumati, P., Huebner, A.K., Akutsu, M., Liebmann, L., Stolz, A., Nietzsche, S., Koch, N., et al. (2015). Regulation of endoplasmic reticulum turnover by selective autophagy. Nature 522, 354–358.

Maejima, I., Takahashi, A., Omori, H., Kimura, T., Takabatake, Y., Saitoh, T., Yamamoto, A., Hamasaki, M., Noda, T., Isaka, Y., and Yoshimori, T. (2013). Autophagy sequesters damaged lysosomes to control lysosomal biogenesis and kidney injury. EMBO J. 32, 2336–2347. Maiuri, M.C., Le Toumelin, G., Criollo, A., Rain, J.C., Gautier, F., Juin, P., Tasdemir, E., Pierron, G., Troulinaki, K., Tavernarakis, N., et al. (2007). Functional and physical interaction between Bcl-X(L) and a BH3-like domain in Beclin-1. EMBO J. 26, 2527–2539.

Kim, I., and Lemasters, J.J. (2011). Mitochondrial degradation by autophagy (mitophagy) in GFP-LC3 transgenic hepatocytes during nutrient deprivation. Am. J. Physiol. Cell Physiol. 300, C308–C317.

Mandell, M.A., Jain, A., Arko-Mensah, J., Chauhan, S., Kimura, T., Dinkins, C., Silvestri, G., Mu¨nch, J., Kirchhoff, F., Simonsen, A., et al. (2014). TRIM proteins regulate autophagy and can target autophagic substrates by direct recognition. Dev. Cell 30, 394–409.

Kim, J., Kim, Y.C., Fang, C., Russell, R.C., Kim, J.H., Fan, W., Liu, R., Zhong, Q., and Guan, K.L. (2013). Differential regulation of distinct Vps34 complexes by AMPK in nutrient stress and autophagy. Cell 152, 290–303.

Manning, B.D., and Cantley, L.C. (2007). AKT/PKB signaling: navigating downstream. Cell 129, 1261–1274.

Knodler, L.A., and Celli, J. (2011). Eating the strangers within: host control of intracellular bacteria via xenophagy. Cell. Microbiol. 13, 1319–1327.

Marin˜o, G., Niso-Santano, M., Baehrecke, E.H., and Kroemer, G. (2014). Self-consumption: the interplay of autophagy and apoptosis. Nat. Rev. Mol. Cell Biol. 15, 81–94.

Koyano, F., Okatsu, K., Kosako, H., Tamura, Y., Go, E., Kimura, M., Kimura, Y., Tsuchiya, H., Yoshihara, H., Hirokawa, T., et al. (2014). Ubiquitin is phosphorylated by PINK1 to activate parkin. Nature 510, 162–166. Kraft, C., Deplazes, A., Sohrmann, M., and Peter, M. (2008). Mature ribosomes are selectively degraded upon starvation by an autophagy pathway requiring the Ubp3p/Bre5p ubiquitin protease. Nat. Cell Biol. 10, 602–610. Kroemer, G., and Ja¨a¨ttela¨, M. (2005). Lysosomes and autophagy in cell death control. Nat. Rev. Cancer 5, 886–897. Kroemer, G., Marin˜o, G., and Levine, B. (2010). Autophagy and the integrated stress response. Mol. Cell 40, 280–293. Lamark, T., and Johansen, T. (2012). Aggrephagy: selective disposal of protein aggregates by macroautophagy. Int. J. Cell Biol. 2012, 736905.

Maxwell, P.H., Wiesener, M.S., Chang, G.W., Clifford, S.C., Vaux, E.C., Cockman, M.E., Wykoff, C.C., Pugh, C.W., Maher, E.R., and Ratcliffe, P.J. (1999). The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 399, 271–275. Mazure, N.M., and Pouysse´gur, J. (2009). Atypical BH3-domains of BNIP3 and BNIP3L lead to autophagy in hypoxia. Autophagy 5, 868–869. Mehlen, P., and Bredesen, D.E. (2011). Dependence receptors: from basic research to drug development. Sci. Signal. 4, mr2. Mihaylova, M.M., and Shaw, R.J. (2011). The AMPK signalling pathway coordinates cell growth, autophagy and metabolism. Nat. Cell Biol. 13, 1016–1023. Mizushima, N., Levine, B., Cuervo, A.M., and Klionsky, D.J. (2008). Autophagy fights disease through cellular self-digestion. Nature 451, 1069–1075.

Lamb, C.A., Yoshimori, T., and Tooze, S.A. (2013). The autophagosome: origins unknown, biogenesis complex. Nat. Rev. Mol. Cell Biol. 14, 759–774.

Mizushima, N., Yoshimori, T., and Ohsumi, Y. (2011). The role of Atg proteins in autophagosome formation. Annu. Rev. Cell Dev. Biol. 27, 107–132.

Lan, Y.Y., London˜o, D., Bouley, R., Rooney, M.S., and Hacohen, N. (2014). Dnase2a deficiency uncovers lysosomal clearance of damaged nuclear DNA via autophagy. Cell Rep. 9, 180–192.

Mochida, K., Oikawa, Y., Kimura, Y., Kirisako, H., Hirano, H., Ohsumi, Y., and Nakatogawa, H. (2015). Receptor-mediated selective autophagy degrades the endoplasmic reticulum and the nucleus. Nature 522, 359–362.

Latinis, K.M., and Koretzky, G.A. (1996). Fas ligation induces apoptosis and Jun kinase activation independently of CD45 and Lck in human T cells. Blood 87, 871–875.

Narendra, D., Kane, L.A., Hauser, D.N., Fearnley, I.M., and Youle, R.J. (2010). p62/SQSTM1 is required for Parkin-induced mitochondrial clustering but not mitophagy; VDAC1 is dispensable for both. Autophagy 6, 1090–1106.

Lee, I.H., Cao, L., Mostoslavsky, R., Lombard, D.B., Liu, J., Bruns, N.E., Tsokos, M., Alt, F.W., and Finkel, T. (2008). A role for the NAD-dependent deacetylase Sirt1 in the regulation of autophagy. Proc. Natl. Acad. Sci. USA 105, 3374–3379.

Niso-Santano, M., Malik, S.A., Pietrocola, F., Bravo-San Pedro, J.M., Marin˜o, G., Cianfanelli, V., Ben-Youne`s, A., Troncoso, R., Markaki, M., Sica, V., et al. (2015). Unsaturated fatty acids induce non-canonical autophagy. EMBO J. 34, 1025–1041.

Lei, Y., Wen, H., Yu, Y., Taxman, D.J., Zhang, L., Widman, D.G., Swanson, K.V., Wen, K.W., Damania, B., Moore, C.B., et al. (2012). The mitochondrial proteins NLRX1 and TUFM form a complex that regulates type I interferon and autophagy. Immunity 36, 933–946.

Okamoto, K. (2014). Organellophagy: eliminating cellular building blocks via selective autophagy. J. Cell Biol. 205, 435–445.

Leto, D., and Saltiel, A.R. (2012). Regulation of glucose transport by insulin: traffic control of GLUT4. Nat. Rev. Mol. Cell Biol. 13, 383–396. Li, W., Zhang, X., Zhuang, H., Chen, H.G., Chen, Y., Tian, W., Wu, W., Li, Y., Wang, S., Zhang, L., et al. (2014). MicroRNA-137 is a novel hypoxia-responsive microRNA that inhibits mitophagy via regulation of two mitophagy receptors FUNDC1 and NIX. J. Biol. Chem. 289, 10691–10701. Liang, N., Jia, L., Liu, Y., Liang, B., Kong, D., Yan, M., Ma, S., and Liu, X. (2013). ATM pathway is essential for ionizing radiation-induced autophagy. Cell. Signal. 25, 2530–2539. Liu, L., Sakakibara, K., Chen, Q., and Okamoto, K. (2014). Receptor-mediated mitophagy in yeast and mammalian systems. Cell Res. 24, 787–795.

Okatsu, K., Oka, T., Iguchi, M., Imamura, K., Kosako, H., Tani, N., Kimura, M., Go, E., Koyano, F., Funayama, M., et al. (2012). PINK1 autophosphorylation upon membrane potential dissipation is essential for Parkin recruitment to damaged mitochondria. Nat. Commun. 3, 1016. Okatsu, K., Koyano, F., Kimura, M., Kosako, H., Saeki, Y., Tanaka, K., and Matsuda, N. (2015). Phosphorylated ubiquitin chain is the genuine Parkin receptor. J. Cell Biol. 209, 111–128. Pattingre, S., Tassa, A., Qu, X., Garuti, R., Liang, X.H., Mizushima, N., Packer, M., Schneider, M.D., and Levine, B. (2005). Bcl-2 antiapoptotic proteins inhibit Beclin 1-dependent autophagy. Cell 122, 927–939. Pietrocola, F., Izzo, V., Niso-Santano, M., Vacchelli, E., Galluzzi, L., Maiuri, M.C., and Kroemer, G. (2013). Regulation of autophagy by stress-responsive transcription factors. Semin. Cancer Biol. 23, 310–322.

Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc. 537

Molecular Cell

Review Ponton, A., Cle´ment, M.V., and Stamenkovic, I. (1996). The CD95 (APO-1/Fas) receptor activates NF-kappaB independently of its cytotoxic function. J. Biol. Chem. 271, 8991–8995.

Strappazzon, F., Vietri-Rudan, M., Campello, S., Nazio, F., Florenzano, F., Fimia, G.M., Piacentini, M., Levine, B., and Cecconi, F. (2011). Mitochondrial BCL-2 inhibits AMBRA1-induced autophagy. EMBO J. 30, 1195–1208.

Rambold, A.S., Cohen, S., and Lippincott-Schwartz, J. (2015). Fatty acid trafficking in starved cells: regulation by lipid droplet lipolysis, autophagy, and mitochondrial fusion dynamics. Dev. Cell 32, 678–692.

Sun, T., Li, X., Zhang, P., Chen, W.D., Zhang, H.L., Li, D.D., Deng, R., Qian, X.J., Jiao, L., Ji, J., et al. (2015). Acetylation of Beclin 1 inhibits autophagosome maturation and promotes tumour growth. Nat. Commun. 6, 7215.

Ravikumar, B., Sarkar, S., Davies, J.E., Futter, M., Garcia-Arencibia, M., Green-Thompson, Z.W., Jimenez-Sanchez, M., Korolchuk, V.I., Lichtenberg, M., Luo, S., et al. (2010). Regulation of mammalian autophagy in physiology and pathophysiology. Physiol. Rev. 90, 1383–1435.

Tallo´czy, Z., Jiang, W., Virgin, H.W., 4th, Leib, D.A., Scheuner, D., Kaufman, R.J., Eskelinen, E.L., and Levine, B. (2002). Regulation of starvation- and virus-induced autophagy by the eIF2alpha kinase signaling pathway. Proc. Natl. Acad. Sci. USA 99, 190–195.

Rebsamen, M., Pochini, L., Stasyk, T., de Arau´jo, M.E., Galluccio, M., Kandasamy, R.K., Snijder, B., Fauster, A., Rudashevskaya, E.L., Bruckner, M., et al. (2015). SLC38A9 is a component of the lysosomal amino acid sensing machinery that controls mTORC1. Nature 519, 477–481.

Tanaka, A., Cleland, M.M., Xu, S., Narendra, D.P., Suen, D.F., Karbowski, M., and Youle, R.J. (2010). Proteasome and p97 mediate mitophagy and degradation of mitofusins induced by Parkin. J. Cell Biol. 191, 1367–1380.

Rello-Varona, S., Lissa, D., Shen, S., Niso-Santano, M., Senovilla, L., Marin˜o, G., Vitale, I., Jemaa´, M., Harper, F., Pierron, G., et al. (2012). Autophagic removal of micronuclei. Cell Cycle 11, 170–176.

Tiwari, M., Sharma, L.K., Vanegas, D., Callaway, D.A., Bai, Y., Lechleiter, J.D., and Herman, B. (2014). A nonapoptotic role for CASP2/caspase 2: modulation of autophagy. Autophagy 10, 1054–1070.

Roberts, P., Moshitch-Moshkovitz, S., Kvam, E., O’Toole, E., Winey, M., and Goldfarb, D.S. (2003). Piecemeal microautophagy of nucleus in Saccharomyces cerevisiae. Mol. Biol. Cell 14, 129–141.

Tripathi, D.N., Chowdhury, R., Trudel, L.J., Tee, A.R., Slack, R.S., Walker, C.L., and Wogan, G.N. (2013). Reactive nitrogen species regulate autophagy through ATM-AMPK-TSC2-mediated suppression of mTORC1. Proc. Natl. Acad. Sci. USA 110, E2950–E2957.

Rodrı´guez-Vargas, J.M., Ruiz-Magan˜a, M.J., Ruiz-Ruiz, C., Majuelos-Melguizo, J., Peralta-Leal, A., Rodrı´guez, M.I., Mun˜oz-Ga´mez, J.A., de Almodo´var, M.R., Siles, E., Rivas, A.L., et al. (2012). ROS-induced DNA damage and PARP-1 are required for optimal induction of starvation-induced autophagy. Cell Res. 22, 1181–1198.

Twig, G., Elorza, A., Molina, A.J., Mohamed, H., Wikstrom, J.D., Walzer, G., Stiles, L., Haigh, S.E., Katz, S., Las, G., et al. (2008). Fission and selective fusion govern mitochondrial segregation and elimination by autophagy. EMBO J. 27, 433–446.

Rowland, A.A., and Voeltz, G.K. (2012). Endoplasmic reticulum-mitochondria contacts: function of the junction. Nat. Rev. Mol. Cell Biol. 13, 607–625.

Vicencio, J.M., Ortiz, C., Criollo, A., Jones, A.W., Kepp, O., Galluzzi, L., Joza, N., Vitale, I., Morselli, E., Tailler, M., et al. (2009). The inositol 1,4,5-trisphosphate receptor regulates autophagy through its interaction with Beclin 1. Cell Death Differ. 16, 1006–1017.

Sakaki, K., Wu, J., and Kaufman, R.J. (2008). Protein kinase Ctheta is required for autophagy in response to stress in the endoplasmic reticulum. J. Biol. Chem. 283, 15370–15380. Sancak, Y., Bar-Peled, L., Zoncu, R., Markhard, A.L., Nada, S., and Sabatini, D.M. (2010). Ragulator-Rag complex targets mTORC1 to the lysosomal surface and is necessary for its activation by amino acids. Cell 141, 290–303. Sapkota, G.P., Deak, M., Kieloch, A., Morrice, N., Goodarzi, A.A., Smythe, C., Shiloh, Y., Lees-Miller, S.P., and Alessi, D.R. (2002). Ionizing radiation induces ataxia telangiectasia mutated kinase (ATM)-mediated phosphorylation of LKB1/STK11 at Thr-366. Biochem. J. 368, 507–516. Scherz-Shouval, R., Shvets, E., Fass, E., Shorer, H., Gil, L., and Elazar, Z. (2007). Reactive oxygen species are essential for autophagy and specifically regulate the activity of Atg4. EMBO J. 26, 1749–1760. Schuck, S., Gallagher, C.M., and Walter, P. (2014). ER-phagy mediates selective degradation of endoplasmic reticulum independently of the core autophagy machinery. J. Cell Sci. 127, 4078–4088. Settembre, C., Di Malta, C., Polito, V.A., Garcia Arencibia, M., Vetrini, F., Erdin, S., Erdin, S.U., Huynh, T., Medina, D., Colella, P., et al. (2011). TFEB links autophagy to lysosomal biogenesis. Science 332, 1429–1433. Shen, S., Niso-Santano, M., Adjemian, S., Takehara, T., Malik, S.A., Minoux, H., Souquere, S., Marin˜o, G., Lachkar, S., Senovilla, L., et al. (2012). Cytoplasmic STAT3 represses autophagy by inhibiting PKR activity. Mol. Cell 48, 667–680. Shi, C.S., and Kehrl, J.H. (2008). MyD88 and Trif target Beclin 1 to trigger autophagy in macrophages. J. Biol. Chem. 283, 33175–33182.

Vitale, I., Galluzzi, L., Castedo, M., and Kroemer, G. (2011). Mitotic catastrophe: a mechanism for avoiding genomic instability. Nat. Rev. Mol. Cell Biol. 12, 385–392. Walter, K.M., Scho¨nenberger, M.J., Tro¨tzmu¨ller, M., Horn, M., Elsa¨sser, H.P., Moser, A.B., Lucas, M.S., Schwarz, T., Gerber, P.A., Faust, P.L., et al. (2014). Hif-2a promotes degradation of mammalian peroxisomes by selective autophagy. Cell Metab. 20, 882–897. Wang, B., Xiao, Z., Ko, H.L., and Ren, E.C. (2010). The p53 response element and transcriptional repression. Cell Cycle 9, 870–879. Wang, R.C., Wei, Y., An, Z., Zou, Z., Xiao, G., Bhagat, G., White, M., Reichelt, J., and Levine, B. (2012). Akt-mediated regulation of autophagy and tumorigenesis through Beclin 1 phosphorylation. Science 338, 956–959. Wang, S., Tsun, Z.Y., Wolfson, R.L., Shen, K., Wyant, G.A., Plovanich, M.E., Yuan, E.D., Jones, T.D., Chantranupong, L., Comb, W., et al. (2015). Metabolism. Lysosomal amino acid transporter SLC38A9 signals arginine sufficiency to mTORC1. Science 347, 188–194. Wauson, E.M., Zaganjor, E., Lee, A.Y., Guerra, M.L., Ghosh, A.B., Bookout, A.L., Chambers, C.P., Jivan, A., McGlynn, K., Hutchison, M.R., et al. (2012). The G protein-coupled taste receptor T1R1/T1R3 regulates mTORC1 and autophagy. Mol. Cell 47, 851–862. Wauson, E.M., Dbouk, H.A., Ghosh, A.B., and Cobb, M.H. (2014). G proteincoupled receptors and the regulation of autophagy. Trends Endocrinol. Metab. 25, 274–282.

Shimobayashi, M., and Hall, M.N. (2014). Making new contacts: the mTOR network in metabolism and signalling crosstalk. Nat. Rev. Mol. Cell Biol. 15, 155–162.

Wei, Y., Zou, Z., Becker, N., Anderson, M., Sumpter, R., Xiao, G., Kinch, L., Koduru, P., Christudass, C.S., Veltri, R.W., et al. (2013). EGFR-mediated Beclin 1 phosphorylation in autophagy suppression, tumor progression, and tumor chemoresistance. Cell 154, 1269–1284.

Shoji-Kawata, S., Sumpter, R., Leveno, M., Campbell, G.R., Zou, Z., Kinch, L., Wilkins, A.D., Sun, Q., Pallauf, K., MacDuff, D., et al. (2013). Identification of a candidate therapeutic autophagy-inducing peptide. Nature 494, 201–206.

Xia, P., Wang, S., Huang, G., Du, Y., Zhu, P., Li, M., and Fan, Z. (2014). RNF2 is recruited by WASH to ubiquitinate AMBRA1 leading to downregulation of autophagy. Cell Res. 24, 943–958.

Singh, R., Kaushik, S., Wang, Y., Xiang, Y., Novak, I., Komatsu, M., Tanaka, K., Cuervo, A.M., and Czaja, M.J. (2009). Autophagy regulates lipid metabolism. Nature 458, 1131–1135.

Yamamoto, H., Fara, A.F., Dasgupta, P., and Kemper, C. (2013). CD46: the ‘multitasker’ of complement proteins. Int. J. Biochem. Cell Biol. 45, 2808– 2820.

538 Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc.

Molecular Cell

Review Yamashita, S., Abe, K., Tatemichi, Y., and Fujiki, Y. (2014). The membrane peroxin PEX3 induces peroxisome-ubiquitination-linked pexophagy. Autophagy 10, 1549–1564. Young, A.R., Narita, M., Ferreira, M., Kirschner, K., Sadaie, M., Darot, J.F., Tavare´, S., Arakawa, S., Shimizu, S., Watt, F.M., and Narita, M. (2009). Autophagy mediates the mitotic senescence transition. Genes Dev. 23, 798–803. Zhang, H., Bosch-Marce, M., Shimoda, L.A., Tan, Y.S., Baek, J.H., Wesley, J.B., Gonzalez, F.J., and Semenza, G.L. (2008). Mitochondrial autophagy is

an HIF-1-dependent adaptive metabolic response to hypoxia. J. Biol. Chem. 283, 10892–10903. Zhang, C.S., Jiang, B., Li, M., Zhu, M., Peng, Y., Zhang, Y.L., Wu, Y.Q., Li, T.Y., Liang, Y., Lu, Z., et al. (2014). The lysosomal v-ATPase-Ragulator complex is a common activator for AMPK and mTORC1, acting as a switch between catabolism and anabolism. Cell Metab. 20, 526–540. Zoncu, R., Bar-Peled, L., Efeyan, A., Wang, S., Sancak, Y., and Sabatini, D.M. (2011). mTORC1 senses lysosomal amino acids through an inside-out mechanism that requires the vacuolar H(+)-ATPase. Science 334, 678–683.

Molecular Cell 59, August 20, 2015 ª2015 Elsevier Inc. 539