Autophagy refers to the group of cellular processes that mediate degradation of intracellular components in lysosomes and the subsequent recycling of their constituents (Mizushima et al., 2008). As such, autophagy shares a common final compartment with  endocytosis and phagocytosis, the lysosome. Continuous turnover of intracellular proteins and organelles allows for their renewal and serves as a mechanism for quality control of newly synthesized cellular structures. Furthermore, the regulated balance between synthesis and degradation of proteins and organelles permits for rapid modification of their cellular levels to accommodate to environmental changes.

Historically, the  discovery of autophagy was temporally close to that of lysosomes, the subcellular compartment with higher content of hydrolases in the cell (Ohsumi, 2014). Protein degradation by autophagy was initially linked mainly to nutritional changes. Starvation or conditions in which nutrients become scarce  often lead to maximal activation of autophagy to facilitate recycling of intracellular components for anabolic purposes. Although nutrient deprivation is one of the best characterized activators of autophagy, better understanding of the molecular basis of the autophagic pathways and the development of tools and markers that permit tracking flux of substrates through the autophagic system have revealed that autophagy is activated in response to a multiplicity of stimuli ranking from cellular stressors (oxidative stress, DNA damage, mitochondria depolarization, etc.) to extracellular cues and inputs (growth factors, cytokines, pathogens, etc.). This better understanding of the conditions associated with autophagy activation, and direct studies analyzing the consequences of upregulating or dowregulating autophagy in whole organisms have unveiled the participation of autophagy in an impressive number of cellular functions: energetic balance quality control, cell and tissue remodeling, cell differentiation, cellular defense, innate immunity, cell death, cellular senescence, among others (Green and Levine, 2014; Kenific and Debnath, 2015; Marino et al., 2014). The multiple physiological roles of autophagy along with the connection of autophagy malfunctioning to common human disorders (neurodegeneration, diabetes, muscle waste, infectious diseases, etc.) have motivated the recent interest in the understanding of the autophagic processes at the molecular level (Jiang and Mizushima, 2014; Schneider and Cuervo, 2014).

Autophagic pathways

Differences in the molecular effectors that contribute to delivery of cargo (material to be degraded) into lysosomes are the bases for the distinction among autophagic pathways. When cargo is sequestered in double membrane vesicles in the cytosol that then acquire the hydrolases required for degradation through lysosomal fusion, autophagy is referred to as macroautophagy (Yang and Klionsky, 2010; Figure 1). In the case of microautophagy, cargo is directly sequestered through invaginations or tubulations of the lysosomal or late endosomal membrane and are released into the lumen of these organelles after pinching off (de Waal et al., 1986; Figure 2). Not all forms of autophagy require vesicles for cargo sequestration; for example, in chaperone-mediated autophagy (CMA) cytosolic proteins reach the lysosomal lumen upon translocation across the lysosomal membrane (Kaushik and  Cuervo,  2012; Figure 3). Different variants have already been described for each of these mechanisms of cargo delivery to lysosomes.

Studies in yeast were key for the identification of a subset of about 35 genes (ATG or autophagy-related genes) that participate in the process of macroautophagy as well as some microautophagy specific genes, whereas the characterization of CMA and some specialized forms of macroand microautophagy was performed in mammalian cells. For simplicity we will use the mammalian nomenclature for all the autophagy-related genes in this entry.

Autophagy step-by-step

Induction of autophagy

Although multiple signaling mechanisms lead to activation of macroautophagy, the better characterized is the mammalian target of rapamycin (mTor) signaling pathway, closely linked to starvation-induced autophagy. The energy-sensing kinase complex target of rapamycin complex 1 (TORC1) displays maximal activity in the presence of nutrients, leading to the phosphorylation and sequestration of ULK1, a kinase required for autophagy initiation (Kanazawa et al., 2004; Hara et al., 2008; Figure 1). Upon deprivation of nutrients, inactivation of mTOR results in release of ULK1 and its localization at the sites of autophagosome formation. The limiting membrane of these organelles forms through combination of proteins and lipids from different cellular compartments such as the endoplasmic reticulum (ER), mitochondria, plasma membrane, Golgi, and sorting endosomes. In all these cases, the area in the membrane from where the limiting membrane or phagophore originates is ‘marked’ through phosphorylation of lipids in these membranes. Besides ULK1, Vps34 is the second lipid kinase commonly recruited to sites of autophagosome biogenesis as part of the Beclin/Vps15/Vps34 complex (Figure 1). Elongation of the membrane occurs through two ubiquitin-like conjugation processes (Ichimura et al., 2000): conjugation of two proteins Atg12 to Atg5 and conjugation of a protein LC3 to the lipid phosphatidylethanolamine. A common ligase, Atg7, is shared by both conjugation events. Tethering of Atg16L-enriched micelles with conjugate LC3 also contributes to growth of the limiting membrane (Ravikumar et al., 2010).

The  first  signaling  mechanism  identified  for  CMA  was through the retinoid-acid receptor alpha that acts as an endogenous repressor of this pathway (Anguiano et al., 2013).

Autophagy 1

Figure 1.  Macroautophagy and selective macroautophagy variants. Left: simplified schematic of the steps of macroautophagy and protein complexes that modulate some of these steps. Right: types of selective macroautophagy.

Autophagy 2

Figure 2.  Yeast and mammalian microautophagy. Left: yeast microautophagy occurs at the vacuole membrane in a multistep process that allows for cargo recognition, formation of the microphagy membrane apparatus (MIPA), and internalization into single-membrane vesicles that pinch of and are subsequently degraded. Right: in mammals, microautophagy has only been described in late endocytic compartments, where cargo can be sequestered in bulk or selectively enriched upon hsc70 binding in vesicles formed in these membranes through assembly of the ESCRT proteins.

Recently, studies in T cells have unveiled activation of CMA by reactive oxygen species (ROS)-dependent signaling through the calcineurin–NFAT (nuclear factor of activated T cells) axis (Valdor et al., 2014). The signaling mechanism that initiates microautophagy remains unknown.

Cargo recognition

Macroautophagy was considered an ‘in bulk’ process in which all the cytosolic components present in the area surrounding the site of autophagosome biogenesis were sequestered at once. However, the selectivity of this process has now been demonstrated after the identification of a subset of proteins, cargo receptors, that have the ability to simultaneously bind to cargo molecules and Atg proteins (Stolz et al., 2014). This dual recruitment facilitates formation of the limiting membrane selectively around mitochondria (mitophagy), peroxisomes (pexophagy), lipid droplets (lipophagy), ribosomes (ribophagy), pathogens (xenophagy), aggregates (agrephagy), etc. (Figure 1). Some of these cargo receptors, p62 or NBr1, are shared among these processes due to their ability to recognize common tagging molecules such as ubiquitin (Novak and Dikic, 2011). A pair of cytosolic chaperones, hsc70 and Bag1, can also serve as cargo recognition molecules for aggregate proteins in a variant of conventional macroautophagy known as chaperone-assisted selective autophagy (CASA) (Arndt et al., 2010).

In yeast microautophagy, sequestration of cytosolic components such as peroxisomes (micropexophagy), nucleus (piecemeal  autophagy),  or  pathogenes   (microxenophagy) by the invaginations at the vacuole membrane occurs at specific contact sides through interaction of cargo and vacuole membrane proteins (Sakai et al., 1998; Roberts et al., 2003). Although microautophagy has not been described in mammalian cells, a similar process of invagination and internalization in single-membrane proteins occurs in the late endocytic compartment/multivesicular bodies and has been named endosomal microautophagy (e-MI) (Sahu et al., 2011). Both in bulk and selective targeting of cytosolic proteins by hsc70 occur through e-MI (Figure 2).

Recognition of the cytosolic proteins that undergo degradation by CMA is mediated by binding of hsc70 to a pentapeptide motif in the protein sequence biochemically related to KFERQ (Dice, 1990). Mutagenesis of this targeting motif abolishes CMA of the protein, and addition of this pentapeptide sequence to proteins that usually do not undergo degradation by CMA reroutes them toward this pathway, supporting that this motif is necessary and sufficient for targeting to CMA. Binding of hsc70 to this region is also required for selective e-MI of cytosolic proteins, but addition of this motif is not sufficient to drive e-MI of a protein (Sahu et al., 2011; Figure 3).

Autophagy 3

Figure 3. Chaperone-mediated autophagy. Schematic of the main steps of CMA. Upon recognition of the targeting motif by hsc70, the protein–chaperone complex is targeted to the lysosomal membrane where it binds to LAMP-2A and induces the multimerization and formation of the translocation complex. After substrate unfolding, the luminal chaperone completes translocation of the substrate.

Lysosomal delivery

After sealing of the limiting membrane to form a double membrane vesicle in a SNARE (soluble N-ethylmaleimidesensitive factor attachment protein receptor)-mediated manner, the autophagosome moves through microtubules tracks, often toward the perinuclear region, to meet the lysosomes (Moreau et al., 2011). Kiss-and-run events infuse part of the lysosomal hydrolases in  autophagosomes forming a  hybrid acid compartment known as autolysosome, where cargo is degraded.

In the case of microautophagy, delivery occurs through vesicles in the surface of the vacuole (forming a structure known as MIPA) in yeast, or through assembly of a series of protein complexes known as endosomal sorting complexes required for transport (ESCRT) proteins in the surface of the late endosome (in mammals). ESCRT proteins are well known because of their ability to form multivesicular bodies in the late endosomes for membrane protein degradation (Figure 2). Whether the two degradative processes, of cytosolic proteins and of plasma membrane proteins, occur in dedicate multivesicular bodies for each purpose or if they occur simultaneously in the same compartments remains unknown.

Once the complex hsc70/cytosolic proteins targeted for CMA reaches the lysosomal membrane they bind to the cytosolic tail of the lysosome-associated membrane protein type 2A (L2A) that is a spliced variant of the LAMP-2 gene. Substrate binding induces multimerization of L2A to form a 700 kDa complex required for protein translocation after unfolding (Cuervo and Dice, 1996; Bandyopadhyay et al., 2008; Figure 3). Translocation also requires a luminal chaperone, a variant of cytosolic hsc70, but whether this chaperone is required for actual pulling of substrates inside lysosomes or just to prevent their retrotranslocation to the cytosol is not known. Translocation complexes are highly dynamic structures that deasemble into monomeric units of L2A right after the substrate crosses the lysosomal membrane, to allow a new cycle of binding and translocation. Chaperones in the cytosolic side of the lysosomal membrane are required for substrate unfolding and they also assist in L2A disassembly, whereas chaperones in the luminal side of the membrane participate in substrate translocation and contribute to maintain L2A stability when transitioning between monomers and multimeric complexes (Bandyopadhyay et al., 2008; Figure 3).

Recycling, an integral part of autophagy

The autophagy process does not end with the degradation of the cytosolic material in the lysosomal lumen; instead, both cargo and functional lysosomal components can be retrieved out of the autolysosome. Permeases dedicated to cytosolic export of amino acids and sugars have been identified in the vacuole membrane in yeast (Sekito et al., 2008). Less is known about the mammalian lysosome permeases but specific amino acid transporters have been known to exist in this membrane for a long time. Recycling of cargo components helps sustain cellular anabolic processes even in the absence of nutrients. Since the three autophagic pathways share the lysosome as a common site of degradation, these permeases and transporters also recycle amino acids from proteins degraded by CMA and microautophagy. In the case of e-MI, the cargo internalized in the multivesicular bodies can be degraded either in these late endosomes, transferred for lysosomal degradation, or exported extracellularly as exosomes upon fusion of lysosomes with the plasma membrane (Figure 2).

Lysosomal components can also be retrieved out of the lysosome into vesicles or tubules that form at the limiting membrane of the lysosome and then pinch off to constitute the basis of a new cargo-free lysosome. Vesiculation, tubulation, and sorting of lysosomal membrane proteins and enzymes into these structures is regulated by a family of lipid kinases that  act in a  coordinate manner with  clathrin and motor proteins that bind to areas of vesicle formation (Yu et al., 2010). This recycling is maximally activated after starvation conditions, when a large fraction of lysosomes has fused to autophagosomes.

Autophagy-related cellular compartments

As described in the previous section, the cellular compartments that participate in macroautophagy are well characterized and include the phagophore or limiting membrane, autophagosomes, autolysosomes, and lysosomes. Autophagososomes can also fuse first with late endosomes, forming an amphisome, where extraand intracellular material converge to ultimately reach the lysosome through endosome/lysosome fusion (Fengsrud et al., 2000). Autophagic flux occurs, to some extent, even in the absence of de novo synthesis, thanks to a reservoir of Atg protein complexes already assembled but that remain in an inactive state through sequestration in different cellular compartments such as acetylated microtubules or plasma membrane connexins, in the case of the Beclin/Vps34 complexes (Geeraert et al., 2010; Bejarano et al., 2014), or lysosomal TORC1, in the case of ULK1 (Kanazawa et al., 2004; Hara et al.,  2008). Whenever autophagy activation occurs, these complexes can be readily released and relocate to sites of autophagosome formation to contribute to  formation  of the limiting membrane. Sustained activation of autophagy requires an expansion of the lysosomal pool to meet the autophagic needs, which is attained in part through the abovementioned recycling of lysosomal components out of autolysosomes, and in part through activation of de novo lysosomal biogenesis. In fact, most of the genes that code for lysosomal proteins and for Atgs are under the control of a family of transcription factors that recognizes a CLEAR motif in the promoter region of these genes. The first and best characterized member of this family is the transcription factor EB (TFEB) that is normally retained in the cytosol through phosphorylation by different kinases, including the TORC1 complex that sits in the surface of lysosomes (Settembre et al., 2011). Amino acids exported from lysosomes maintain TORC1 active, TFEB phosphorylated, and the  autophagic/lysosomal transcriptional program attenuated. However, when rates of degradation through lysosomes decrease, the absence of exiting amino acids leads to TORC1 inactivation and the consequent reduced phosphorylation of TFEB facilitates its nuclear translocation and the activation of the transcriptional program responsible for autophagy and lysosome biogenesis. A newly identified member of this family, TFN3, seems to also undergo lysosomal mTOR phosphorylation and activate the lysosomal/autophagic program through comparable mechanisms (Martina et al., 2014). The most ubiquitous distribution of TFN3 and its presence in organs such as the central nervous system, where levels of TFEB are negligible, support a more generalized regulatory role for this new protein.

Mammalian microautophagy has only been demonstrated so far in late endosomes/multivesicular bodies, compartments shared with endocytosis. However, the impact of expanded or reduced endocytic flux in e-MI remains unknown.

CMA occurs in a specific lysosomal subpopulation that bears in its lumen the lysosomal hsc70 chaperone required for substrate translocation. CMA-active lysosomes are secondary lysosomes that still retain the ability to fuse with autophagic and endocytic vesicles, although less efficiently than other lysosomal subgroups (Cuervo et al., 1997). They correspond to the most acidic group of lysosomes, since lys-hsc70 is more stable at acidic pH and slight reduction in luminal acidification is sufficient to induce its rapid degradation. Rather than biogenically independent lysosomal subgroups, CMA lysosomes and lysosomes engaged in other vesicular processes rapidly convert into each other allowing for expansion or shrinkage of the pool of lysosomes most efficient for each autophagic pathway on demand.

The autophagic Cargo and purpose of its degradation

Activation of macroautophagy during starvation in yeast and mammals was considered necessary to replenish the intracellular pool of amino acids and sustain synthesis of proteins required for adaptation to starvation. However, more recently, macroautophagy has shown the capability to promote degradation of energetically more favorable molecules such as lipids and glycogen. Degradation of lipid droplets (lipophagy) starts with the formation of the limiting membrane in situ, as lipid droplets also contribute lipids for autophagosome biogenesis. This membrane sequesters a whole or a portion of lipid droplets and delivers them to lysosomes through autophagosome/lysosome fusion (Singh et al., 2009). The lipases in the lysosomal lumen break down lipid droplet triglicerides into free fatty acids that, once exported from the lysosome, are oxidized in the mitochondria to generate energy. A similar process mediates degradation of glycogen stores into lysosomes upon autophagic sequestration (Kotoulas et al., 2006).

Macroautophagy can also degrade portions or whole organelles such as mitochondria, ER, peroxisomes, and even lysosomes. Continuous basal degradation of organelles contributes to their renewal. Furthermore, selective macroautophagy can be activated to: (1) eliminate organelles when they are damaged or do no longer function, such as depolarized mitochondria or leaking lysosomes; (2) regulate the size of pool of cellular organelles and modulate their function, such as excess of peroxisomes or mitochondria; and (3) allow for cellular differentiation or cell-type conversion, such as elimination of organelles in reticulocytes as they convert into red cells, or elimination of the spermatozoid mitochondria preovule fecundation (Singh and Cuervo, 2011).

Degradation of cytosolic content by macroautophagy  is often used as a mechanism of cellular defense, such as activation of xenophagy, to eliminate pathogens that reach the cytosol or of aggrephagy to eliminate protein aggregates that could become toxic when accumulating inside cells (Levine, 2005).

Specialized regions of the plasma membrane are also targeted for autophagic degradation, often with regulatory purposes. For example,  macroautophagy  can  have  an  impact in cell-to-cell communication, through degradation of connexins/gap junctions or of neuronal dendritic spines and synaptic components; it can also impact cellular motility through degradation of motile cilia (ciliophagy), environmental sensing through degradation of the primary cilia components, or cell adhesion through degradation of adhesion foci, all present at the plasma membrane.

Yeast microautophagy can also degrade proteins, organelles, and pathogens, but the reasons why this pathway is chosen in specific conditions over macroautophagy remain unknown. Mammalian endosomal microauotphagy has only shown so far to degrade cytosolic proteins, but contrary to CMA, unfolding is not a prerequisite for their degradation through this pathway, making it potentially suitable for the elimination of oligomeric proteins before they  consolidate into large aggregates.

Specific characteristics of cargo delivery by CMA make this pathway only suitable for degradation of single-soluble proteins that can undergo complete unfolding. In principle, all the CMA substrates come from the cytosol in order to bind the cytosolic tail of L2A. Thus, degradation of organelle proteins by CMA only occurs if these proteins relocate to the cytosol or after their de novo synthesis before they reach their final compartment. An example of this type of CMA degradation has been recently observed in a transgenic mouse model unable to perform CMA in the liver, where nuclearly encoded mitochondria enzymes accumulate in part in the cytosol but also inside the mitochondria due to their reduced degradation through CMA (Schneider et al., 2014). This observation opens the intriguing possibility of CMA contributing to regulate the mitochondria enzymatic load. The fact that CMA substrate proteins should bind to the cytosolic tail of the receptor makes this pathway unsuitable for degradation of membrane proteins, at least as far as they are integrated in membranes, since retrotranslocation will not be possible in that case. However, it is possible that the CMA degradation recently proposed for a plasma membrane receptor occurs as quality control before membrane insertion of this protein, or that degradation of this type of proteins by CMA is preceded by a membraneextraction step, comparable to the one proposed for chaperones before the degradation of membrane proteins by AAA proteases. Soluble cytosolic proteins, the canonical CMA cargo, undergo degradation through this pathway both with regulatory purposes or as a mechanism for protein quality control. CMA has been recently demonstrated to contribute to regulation of glucose and lipid metabolic pathways in the liver (Schneider  et  al.,  2014)  and  to  the  proteome  remodeling required for T cell activation, since selective degradation of specific subsets of proteins by CMA reduces their intracellular levels, and subsequently their function (Valdor et al., 2014). The role of CMA in protein quality control has been well demonstrated during oxidative stress where CMA allows targeting of only oxidized forms of the proteins without affecting functional ones.

Pathophysiology of autophagy

The multiplicity of functions attributed to the autophagic process speaks to their already proven physiological relevance. Although studies in unicellular models and mammalian cells in culture have been key to understanding the interplay of autophagy with other cellular processes, importance of proper autophagy functioning at the whole organism level has been inferred from genetically modified mouse models. Animals with whole body blockage of macroautophagy are born, but die few hours after birth, due to their inability to adapt to the starvation period that precedes lactancy (Kuma et al., 2004). Tissue-specific conditional knockout mouse models for specific Atg have offered a better understanding of organ-specific functions of macroautophagy. Thus, complete loss of macroautophagy in the liver leads to accumulation of damaged organelles, lipids (steatosis), and liver dysfunction, whereas blockage of macroauotphagy in  specific areas  of the brain leads to early neurodegeneration with accumulation of intraneuronal ubiquitinated protein aggregates that end killing neuronal cells (Komatsu et al., 2005; Hara et al., 2006). Blockage of autophagy in cardiomyocites, when inflicted early in life, is well tolerated and defective cardiac function can only be elicited upon stress. However, blockage of autophagy in the heart of an adult animal progresses rapidly to heart failure (Nakai et al., 2007). These regulatable models in which autophagy is eliminated at different times in life support the existence of early compensatory mechanisms in response to macroautophagy failure that are lost with aging. Elimination of macroautophagy genes in specific cell types has also connected autophagy to innate and acquired immunity, pancreatic beta-cell homeostasis and insulin secretion, muscle atrophy after immobilization, among others. These models are also revealing Atg functions independent of autophagy as in the case of Atg16L function in the Paneth cells of the intestine (Cadwell et al., 2008). Overexpression of Atg proteins have been a successful approach to increase life span and health span in invertebrates. Although, to date, only one mouse model expressing a copy of Atg5 has been generated, analysis of life span in this model has revealed an encouraging increase (Pyo et al., 2013). Future studies are needed to discriminate the specific functions of macroautophagy responsible for this life-span extension.

Studies on CMA are now also taking advantage of conditional transgenic mouse models with reduced CMA by selectively eliminating the exon that encodes for the L2A variant. Blockage of CMA in the liver leads to major metabolic aberrations and organism problems in metabolism of lipids and glycogen (Schneider et al., 2014). Although none of these macromolecules can be degraded by CMA, many of the enzymes  that  regulate  these  metabolic  pathways  have  been validated as bona fide CMA substrates. Failure to undergo degradation by CMA in this model leads to an increase in their intracellular levels and the subsequent aberrant  increase in function. Mouse models with defective CMA in T cells have shown that this autophagy pathway is required for the proteome changes that lead to the T-cell activation program (Valdor et al., 2014). Failure to degrade inhibitory proteins of this program prevents T cell activation in the CMA-incompetent animals. Interestingly, studies in the T cell model have also revealed that the reduction in CMA with age in these cells is behind their functional failure in old organism (immunosenescence) and that restoration of L2A in old T cells to the levels observed in young animals makes them functional again. Similar expression of an exogenous copy of L2A in mouse liver is sufficient to restore liver homeostasis and function.

Considering this extensive array of physiological functions attributed to autophagy, it was expected that malfunctioning of autophagy underlies the basis of severe human disorders (reviewed in detail elsewhere (Cuervo and Macian, 2014; Jiang and Mizushima, 2014; Lee, 2014; Martin et al., 2015; Schneider and Cuervo, 2014)).

Concluding remarks

The field of autophagy has undergone major expansion during the last two decades that has resulted in important paradigm shifts in this area. Although much remains to be discovered on the basis of selectivity, autophagy is no longer considered an in-bulk nonselective process. Instead, multiple autophagy variants have emerged and selective removal of specific cellular components is being identified. Furthermore, autophagy is no longer a process for destruction of unwanted cellular material, but rather a fine-tuned mechanism that contributes to maintain cellular homeostasis through selective degradation and recycling. Much is left to learn about the regulation of the different types of autophagy, the rules of their cellular coexistence, and about how their activities are coordinated in the context of disease to compensate one for another. It is precisely this connection between autophagy and human disease that has motivated the growing interest in the developing of efficient ways to manipulate the autophagic process with therapeutic purposes.

See also: Organelles: Structure and Function: At the Center of Autophagy: Autophagosomes


Anguiano, J., Garner, T.P., Mahalingam, M., et al., 2013. Chemical modulation of chaperone-mediated autophagy by retinoic acid derivatives. Nature Chemical Biology 9, 374–382.

Arndt, V., Dick, N., Tawo, R., et al., 2010. Chaperone-assisted selective autophagy is essential for muscle maintenance. Current Biology 20, 143–148.

Bandyopadhyay, U., Kaushik, S., Varticovski, L., Cuervo, A.M., 2008. The chaperone-mediated autophagy receptor organizes in dynamic protein complexes at the lysosomal membrane. Molecular and Cellular Biology 28, 5747–5763.

Bejarano, E., Yuste, A., Patel, B., et al., 2014. Connexins modulate autophagosome biogenesis. Nature Cell Biology 16 (5), 401–414.

Cadwell, K., Liu, J.Y., Brown, S.L., et al., 2008. A key role for autophagy and the autophagy gene Atg16l1 in mouse and human intestinal Paneth cells. Nature 456, 259–263.

Cuervo, A.M., Dice, J.F., 1996. A receptor for the selective uptake and degradation of proteins by lysosomes. Science 273, 501–503.

Cuervo, A.M., Dice, J.F., Knecht, E., 1997. A population of rat liver lysosomes responsible for the selective uptake and degradation of cytosolic proteins. Journal of Biological Chemistry 272, 5606–5615.

Cuervo, A.M., Macian, F., 2014. Autophagy and the immune function in aging. Current Opinion in Immunology 29, 97–104.

Dice, J.F., 1990. Peptide sequences that target cytosolic proteins for lysosomal proteolysis. Trends in Biochemical Sciences 15, 305–309.

Fengsrud, M., Erichsen, E.S., Berg, T.O., Raiborg, C., Seglen, P.O., 2000. Ultrastructural characterization of the delimiting membranes of isolated autophagosomes and amphisomes by freeze-fracture electron microscopy. European Journal of Cell Biology. 79, 871–882.

Geeraert, C., Ratier, A., Pfisterer, S.G., et al., 2010. Starvation-induced hyperacetylation of tubulin is required for the stimulation of autophagy by nutrient deprivation. Journal of Biological Chemistry 285, 24184–24194.

Green, D.R., Levine, B., 2014. To be or not to be? How selective autophagy and cell death govern cell fate. Cell 157, 65–75.

Hara, T., Nakamura, K., Matsui, M., et al., 2006. Suppression of basal autophagy in neural cells causes neurodegenerative disease in mice. Nature 441, 885–889.

Hara, T., Takamura, A., Kishi, C., et al., 2008. FIP200, a ULK-interacting protein, is required for autophagosome formation in mammalian cells. Journal of Cell Biology 181, 497–510.

Ichimura, Y., Kirisako, T., Takao, T., et al., 2000. A ubiquitin-like system mediates protein lipidation. Nature 408, 488–492.

Jiang, P., Mizushima, N., 2014. Autophagy and human diseases. Cell Research 24, 69–79.

Kanazawa, T., Taneike, I., Akaishi, R., et al., 2004. Amino acids and insulin control autophagic proteolysis through different signaling pathways in relation to mTOR in isolated rat hepatocytes. Journal of Biological Chemistry 279, 8452–8459.

Kaushik, S., Cuervo, A.M., 2012. Chaperone-mediated autophagy: A unique way to enter the lysosome world. Trends in Cell Biology 22, 407–417.

Keni?c, C.M., Debnath, J., 2015. Cellular and metabolic functions for autophagy in cancer cells. Trends in Cell Biology 25, 37–45.

Komatsu, M., Waguri, S., Ueno, T., et al., 2005. Impairment of starvation-induced and constitutive autophagy in Atg7-de?cient mice. Journal of Cell Biology 169, 425–434.

Kotoulas, O.B., Kalamidas, S.A., Kondomerkos, D.J., 2006. Glycogen autophagy in glucose homeostasis. Pathology, Research and Practice 202, 631–638.

Kuma, A., Hatano, M., Matsui, M., et al., 2004. The role of autophagy during the early neonatal starvation period. Nature 432, 1032–1036.

Lee, M.S., 2014. Role of islet beta cell autophagy in the pathogenesis of diabetes. Trends in Endocrinology & Metabolism 25 (12), 620–627.

Levine, B., 2005. Eating oneself and uninvited guests: Autophagy-related pathways in cellular defense. Cell 120, 159–162.

Marino, G., Niso-Santano, M., Baehrecke, E.H., Kroemer, G., 2014. Selfconsumption: The interplay of autophagy and apoptosis. Nature Reviews Molecular Cell Biology 15, 81–94.

Martin, D.D., Ladha, S., Ehrnhoefer, D.E., Hayden, M.R., 2015. Autophagy in Huntington disease and Huntingtin in autophagy. Trends in Neurosciences 76, 24–36.

Martina, J.A., Diab, H.I., Lishu, L., et al., 2014. The nutrient-responsive transcription factor TFE3 promotes autophagy, lysosomal biogenesis, and clearance of cellular debris. Science Signaling 7 (309), ra9.

Mizushima, N., Levine, B., Cuervo, A.M., Klionsky, D.J., 2008. Autophagy ?ghts disease through cellular self-digestion. Nature 451, 1069–1075.

Moreau, K., Ravikumar, B., Renna, M., Puri, C., Rubinsztein, D.C., 2011. Autophagosome precursor maturation requires homotypic fusion. Cell 146, 303–317.

Nakai, A., Yamaguchi, O., Takeda, T., et al., 2007. The role of autophagy in cardiomyocytes in the basal state and in response to hemodynamic stress. Nature Medicine 13, 619–624.

Novak, I., Dikic, I., 2011. Autophagy receptors in developmental clearance of mitochondria. Autophagy 7, 301–303.

Ohsumi, Y., 2014. Historical landmarks of autophagy research. Cell Research 24, 9–23.

Pyo, J.O., Yoo, S.M., Ahn, H.H., et al., 2013. Overexpression of Atg5 in mice activates autophagy and extends lifespan. Nature Communications 4.2300.

Ravikumar, B., Moreau, K., Jahreiss, L., Puri, C., Rubinsztein, D., 2010. Plasma membrane contributes to the formation of pre-autophagosomal structures. Nature  Cell Biology 12, 747–757.

Roberts, P., Moshitch-Moshkovitz, S., Kvam, E., et al., 2003. Piecemeal  microautophagy of nucleus in Saccharomyces cerevisiae. Molecular Biology of the Cell 14, 129–141.

Sahu, R., Kaushik, S., Cannizzo, E., et al., 2011. Microautophagy of cytosolic proteins by late endosomes. Developmental Cell 20, 131–139.

Sakai, Y., Koller, A., Rangell, L., Keller, G., Subramani, S., 1998. Peroxisome degradation by microautophagy in Pichia pastoris. Identi?cation of speci?c steps and morphological intermediates. Journal of Cell Biology 141, 625–636.

Schneider, J.L., Cuervo, A.M., 2014. Autophagy and human disease: Emerging themes. Current Opinion in Genetics and Development 26, 16–23.

Schneider, J.L., Suh, Y., Cuervo, A.M., 2014. De?cient chaperone-mediated  autophagy in liver leads to metabolic dysregulation. Cell Metabolism 20,  417–432.

Sekito, T., Fujiki, Y., Ohsumi, Y., Kakinuma, Y., 2008. Novel families of vacuolar amino acid transporters. International Union of Biochemistry and Molecular Biology Life 60, 519–525.

Settembre, C., Di Malta, C., Polito, V.A., et al., 2011. TFEB links autophagy to lysosomal biogenesis. Science 332, 1429–1433.

Singh, R., Cuervo, A.M., 2011. Autophagy in the cellular energetic balance. Cell Metabolism 13, 495–504.

Singh, R., Kaushik, S., Wang, Y., et al., 2009. Autophagy regulates lipid metabolism. Nature 458, 1131–1135.

Stolz, A., Ernst, A., Dikic, I., 2014. Cargo recognition and traf?cking in selective autophagy. Nature Cell Biology 16, 495–501.

Valdor, R., Mocholi, E., Botbol, Y., et al., 2014. Chaperone-mediated autophagy regulates T cell responses through targeted degradation of negative regulators of T cell activation. Nature Immunology 15, 1046–1054.

de Waal, E.J., Vreeling-Sindelarova, H., Schellens, J.P., Houtkooper, J.M., James, J., 1986. Quantitative changes in the lysosomal vacuolar system of rat hepatocytes during short-term starvation. A morphometric analysis with special reference to macroand microautophagy. Cell and Tissue Research 243, 641–648.

Yang, Z., Klionsky, D.J., 2010. Mammalian autophagy: Core molecular machinery and signaling regulation. Current Opinion in Cell Biology 22, 124–131.

Yu, L., McPhee, C.K., Zheng, L., et al., 2010. Termination of autophagy and reformation of lysosomes regulated by mTOR. Nature 465, 942–946.