Sci. Aging Knowl. Environ., 10 September 2003
Vol. 2003, Issue 36, p. pe25
[DOI: 10.1126/sageke.2003.36.pe25]


Autophagy and Aging--When "All You Can Eat" Is Yourself

Ana Maria Cuervo

The author is in the Departments of Anatomy and Structural Biology, Marion Bessin Liver Research Center of the Albert Einstein College of Medicine, Bronx, NY 10461, USA. E-mail: amcuervo{at};2003/36/pe25

Key Words: autophagy • lysosome • vacuole • Caenorhabditis elegans • stress

During the past 10 years, we have witnessed the revitalization of an old organelle, or it would probably be better to say of a whole intricate intracellular process. This organelle is the lysosome and the process is autophagy, or the "self-eating" that we all learned about in our old cell biology texts.

Despite the early descriptions of this process in mammalian cells (1), the main reason for the present "empowering" of autophagy has been the results of genetic screens in yeast for mutations in genes that participate in autophagy and closely related processes. The characterization of these genes (Apg, Aut, and Cvt) is leading to the dissection of the molecular machinery that orchestrates this catabolic process (2-4). Now, another nonmammalian organism, Caenorhabditis elegans, has allowed Levine and colleagues to genetically link macroautophagy, one of the most generalized forms of autophagy, with the aging process (5).


Autophagy refers to any process resulting in the degradation of intracellular components inside lysosomes (or the vacuole, in the case of yeast). Autophagy is conservative in nature, because the constitutive elements of the degraded cellular components are reutilized by the cell (6-8).

The process of autophagy is not only intended for the removal of damaged or abnormal cellular components. Most cellular components are continuously synthesized and degraded. This continuous turnover ensures that the cell will have rapid adaptability to changing extracellular conditions, as well as tight regulation of intracellular homeostasis. The degradation of components that are still functional is probably preventive and protective, because once damaged, their removal might not always be possible (for example, damaged proteins can aggregate to the point where they can no longer be degraded and removed from cells).

Better understanding of the molecular basis of autophagy has arisen from genetic decoding of the major players as well as characterization of this process in species never analyzed in detail before (see below for comments on recent work in flies, amoebas, and plants). These studies are also revealing unexpected new roles for autophagy in physiological and pathological processes as diverse as cell differentiation and development, host-to-pathogen and immune responses, oncogenesis, neurodegeneration, and aging (9-13).

Autophagic Mechanisms

Different pathways converge in this catabolic process with the same result: the complete degradation of cellular components in the cell's lysosomes or vacuole. Some of these pathways (depicted in Fig. 1), such as macroautophagy and microautophagy, are conserved in yeast and mammalian cells, whereas others respond to the particular needs of one or the other. Through macroautophagy, complete regions of the cytosol are sequestered by a double membrane that elongates and seals to form what is known as an autophagosome. Recent genetic studies have revealed that this membrane originates de novo in yeast and from specific regions of the endoplasmic reticulum in mammals (2). Elongation and sealing of the structure are attained through the recruitment or exchange of specific cytosolic components in the forming structure. Two different conjugation systems (protein-protein and protein-lipid) and two kinase complexes modulate this process (2). The autophagosome then fuses with the lysosome or vacuole, which contributes the enzymes required for the degradation of the sequestered components. For a recent review, see (4).

View larger version (23K):
[in this window]
[in a new window]
Fig. 1. Different forms of autophagy. The main autophagic mechanisms that lead to the degradation of intracellular components in the vacuole/lysosome system are depicted. Green boxes denote pathways described only in yeast, orange boxes denote pathways in mammals, and green/orange boxes denote pathways in both organisms. During nutrient deprivation in yeast, macroautophagy can deliver not only substrates but also enzymes to the vacuole, taking over the function of a biogenic pathway. Vid, vacuole import and degradation pathway.

The underlying principle of microautophagy is an analogous translocation of cytosolic components all at once into the lysosomal lumen. However, in this case the lysosomes and vacuole are the cellular structures actively sequestering regions of the cytosol. The molecular machinery mediating this engulfment is also being dissected (14).

Both macro- and microautophagy are considered to be "in bulk" autophagy because of their lack of specificity; any cellular components that are present in the cytosol can be engulfed at the same time. However, there is some specificity for the degradation of intracellular organelles in both processes (14-16). Although currently only the term pexophagy, which describes the selective degradation of peroxisomes by lysosomes, is widely used, names such as nucleophagy [for the selective degradation of nuclear regions (17)] or mitophagy [for selective degradation of mitochondria (18)] have also been proposed. These selective forms of organelle degradation share some of the molecular components of classic macro- and microautophagy but also require specific gene products.

Selective degradation of proteins in the lysosomal compartment can take place in mammalian cells through chaperone-mediated autophagy (19). In this process, specific cytosolic proteins that contain a targeting motif are translocated directly through the lysosomal membrane before their degradation. This translocation is assisted by chaperones on both sides of the lysosomal membrane. Although the presence of this pathway has not been clearly established in yeast, a related two-step process (the vacuole import and degradation pathway, or vid) has been described in which proteins are translocated in a chaperone-dependent manner into small vesicles that then fuse with the vacuole (20).

Although characterization of the intricacies of autophagy is only in the early stages, the complexity of this system and its interconnections with other intracellular processes are already evident. A clear separation between autophagy and heterophagy (the degradation of extracellular components in lysosomes) is no longer tenable in view of the multiple interactions described between vesicular components from both pathways. In addition, autophagy is not just a catabolic process but can be used for anabolic purposes, such as the targeting of functional enzymes into the vacuoles (4).

Why So Many Pathways?

Although it is likely that one autophagic pathway can compensate, at least in part, for the failure of another, these pathways are not redundant in essence. Activation and inactivation of individual pathways appear to be responses to specific cellular requirements (7, 8). Autophagy is both a maintenance- and stress-activated process, but the pathways activated under each of these conditions are different. In most mammalian cells, microautophagy is constitutively activated, whereas stressors such as nutrient scarcity activate macroautophagy and, later on, chaperone-mediated autophagy.

In addition to nutrient deprivation, other typical activators of macroautophagy include specific regulatory amino acids, glucocorticoids, and thyroid hormone. Among the physiological inhibitors are insulin, several growth factors, cyclic adenosine monophosphate (cAMP) and cGMP, and conditions that result in reduced intracellular concentrations of adenosine triphosphate (7, 8).

The signaling pathways that lead to the activation of the different forms of autophagy are currently under intense investigation. A growing number of signaling pathways have been shown to control macroautophagy, including those that involve phosphatidylinositol 3-kinase classes I and III, casein kinases, mitogen-activated protein kinases, and small guanosine triphosphatases [reviewed in (12)]. Most of these pathways converge at a single selective regulator known as TOR (target of rapamycin). TOR is a nutrient sensor kinase that, in the presence of nutrients, arrests several intracellular programs, including macroautophagy. The p70S6 kinase and, recently, the Ras/cAMP pathway have been proposed as links between TOR and the kinase complexes that form part of the molecular machinery of macroautophagy. Although special attention has been given to TOR, evidence exists that supports the participation of TOR-independent (Notch) and even nutrient-independent (for example, during apoptosis) mechanisms of activation of autophagy (12).

Macroautophagy and Aging: The Genetic Proof

We have known for almost 30 years that in all organisms, the rates of total protein breakdown decrease with age (21, 22). The consequences of impaired protein degradation during aging are easily inferred [reviewed in (6)]: a weakened ability to adapt to changing environmental conditions, diminished resistance to stress, intracellular accumulation of damaged products that can no longer be eliminated by degradation (see Gray Review), and an increased susceptibility to physical and biological extracellular aggressions.

Age-related alterations in the lysosomal system are clearly noticeable in the form of a pigmented product (lipofuscin) that accumulates in lysosome-related vesicles as a result of the incomplete digestion of engulfed components [reviewed in (23)]. Lysosomes filled with lipofuscin have a reduced ability to fuse with macroautophagic structures. The molecular basis of the age-related decline in different autophagic pathways is the subject of intense investigation (24, 25).

Although the mechanisms by which different forms of autophagy decrease with age are not yet known, it is becoming clear that preventing this decrease should have a favorable effect on life span. Recent studies in rodents proposed a role for macroautophagy in the beneficial effect of caloric restriction on aging (26) (Fig. 2, top). Even though a decrease in protein degradation with age still occurred in hepatocytes from calorie-restricted rats, macroautophagy was more efficiently activated in response to further nutritional deprivation. By an as-yet-uncharacterized mechanism, a maintained state of semistarvation prepares the organism to be more responsive to nutrient deprivation.

View larger version (31K):
[in this window]
[in a new window]
Fig. 2. Models of the aging-related functions of autophagy in different species. Top (lavender): Rodents. Rates of macroautophagy and chaperone-mediated autophagy decrease with age in rodents. Macroautophagy activity and regulation are well preserved in calorie-restricted rats, which live longer than rats fed ad libitum. Middle (orange): Plants. Autophagy-related genes are up-regulated during leaf senescence. Paradoxically, blockage of autophagy does not prevent, but rather accelerates, leaf senescence. Bottom (yellow): Worms. Intact autophagic function is required to attain maximal life extension in long-lived C. elegans mutants. Autophagy is also necessary during the dauer stage (nonaging) and for the dauer to survive and reenter normal larval development and adulthood.

Recent studies in plants also support the importance of macroautophagy in senescence. Disruption of two different macroautophagy genes in Arabidopsis thaliana made the plant hypersensitive to stress and resulted in premature leaf senescence (27, 28) (Fig. 2, middle). In contrast to animal aging, leaf senescence is an "efficient" aging, in which nutrients from the senescing leaf are reused by other parts of the plant. In that sense, it is not surprising that macroautophagy is up-regulated during normal leaf senescence and is probably a major player in the redistribution of nutrients (27, 28). The inability to carry on this function when macroautophagy is disrupted leads to rapid death, probably by activation of other proteolytic systems.

A more classic model for studies on aging is Caenorhabditis elegans, the experimental system used by Levine and colleagues to present genetic evidence for a role for macroautophagy in life span extension. Their approach was to block the expression of autophagy genes using RNA interference (RNAi) in a mutant strain of C. elegans (daf-2) that exhibits a prolonged life span.

C. elegans develops through four larva stages, but under adverse conditions (scarce nutrients or high population density) enters into a developmentally arrested dispersal stage known as the dauer. The dauer larva is resistant to external insults (oxidation, heat, ultraviolet light, heavy metals, etc.) and has been described as nonaging, because postdauer life span is not reduced no matter how much time is spent in the dauer stage (Fig. 2). Genes that promote dauer entry (daf genes) independently of the environmental conditions have been related to longevity (29). Thus, mutations of a subset of daf genes associated with the insulin-like signaling pathway, such as the daf-2 mutant described in the Melendez et al. paper, significantly increase the life span of adult worms. The daf-2 gene encodes a protein with structural similarity to the mammalian insulin receptor. When mutated, reduced Daf-2 signaling induces changes in many different processes that together result in a prolonged life span.

The daf-2 mutant enters inappropriate constitutive dauer at 25°C, but shows extended adult life span at lower temperatures, allowing Levine and colleagues to separately analyze the role of autophagy both during the dauer stage and in life span extension. In daf-2 mutant worms, using RNAi techniques, Melendez et al. blocked expression of ce-bec-1, a homolog of yeast Apg6/Vsp30, which is known to participate in pre-autophagosome formation. Inhibition of ce-bec-1 expression altered dauer morphogenesis and its ability to resume productive growth, and shortened the prolonged survival typical of the daf-2 adult worm to the median survival values of wild-type worms. A similar effect was found when macroautophagy was disrupted by targeting other autophagy genes. These results imply that autophagy is a downstream process of the insulin-like signaling pathway that is essential both for dauer survival and for the program activated in the daf-2 mutants that leads to life span extension.

Although it had not been demonstrated previously, a role for autophagy in dauer survival was suspected. Entry into dauer is normally induced by the same stimuli that activate autophagy, and the dauer is resistant to environmental insults against which autophagy may have a protective effect. In addition, Daf-2 is an insulin receptor analog, and the insulin pathway has an inhibitory effect on autophagy.

The need for intact autophagy to attain full extension of life span in these mutants is a novel, exciting finding, especially in view of the fact that the disruption of autophagy in the wild-type animal did not shorten its life span significantly. These results suggest that activation of autophagy is of particular importance in the metabolic state reached after disrupting the insulin-like signaling pathway. This metabolic state is probably similar to the dauer and, in some ways, to that of calorie-restricted rodents. It is likely that, as described for calorie-restricted rats, the wild-type worm would suffer from not having a functional autophagic system only under stressful conditions.

In light of the findings of Melendez et al., an important question that arises is how can continuous activation of macroautophagy be maintained throughout life in daf-2 mutants? Considering the lack of specificity of this process, how is unwanted degradation prevented? Would added nutritional stress further activate macroautophagy in daf-2 adults? Although future studies in this ingenious system will elucidate these matters, an important aspect highlighted in the work of Melendez et al. is that activation of macroautophagy both in the dauer stage and in the adult appears to be cell type-specific. Thus, it is not a generalized activation of random degradation likely to end in self-destruction, but instead is a process selectively elicited in specific groups of cells (hypodermal seam cells, a cell type important for the morphological changes of the dauer, some types of neurons, and the vulva) that also show, though to a lesser extent, constitutive macroautophagy in wild-type animals. This cell specificity strongly supports different roles for macroautophagy in different cell types beyond the classic metabolic role.

Future Perspectives: Such a Little Fellow and So Much to Offer

In addition to completing the molecular dissection of different forms of autophagy, studies in yeast will still be indispensable for clarifying the intricate network of connections between autophagy and other catabolic and anabolic pathways. However, the growing interest in defining the physiological role of autophagy and its relationship to pathological conditions will require multicellular organisms, in which tissue specificity, humoral contributions, and organ-related compensatory mechanisms can be addressed. For many years, studies in autophagy have been conducted predominantly in yeast and rodents. This situation is rapidly changing. Recent reports have focused on the role of autophagy in differentiation and infection in the amoeba Dictyostelium discoideum (9) and in the participation of autophagy in cellular growth (30) and death (10) in Drosophila melanogaster, as well as on newly discovered functions of autophagy in plants (27, 28).

Through the work of Melendez et al., C. elegans, already a classic model for studies of the aging process, is returning to prominence in the field of autophagy. In fact, this was one of the first systems in which the decline in protein degradation with age was identified (22). So what is now new? Despite their very short life span and easy genetic manipulation, for a period of time C. elegans was not used to study autophagy because the major participants in this process were unknown and reliable autophagy markers were not well defined. Now we know that a large number of autophagy genes characterized in yeast are conserved, not only in C. elegans but throughout the process of evolution and extending to humans (2-4). Also, the availability of autophagy gene products that are fused to fluorescent proteins, such as the ones used by Melendez et al., draws attention to the important fact that C. elegans is a transparent organism, allowing the visualization of the autophagic structures in vivo.

Many questions will hopefully find answers through studies using simple multicellular organisms. As various age-related genes have been described in C. elegans, it is important to know whether autophagy is a downstream pathway for all of them. Are aging-related autophagy and classic autophagy regulated by the same mechanisms? Are the same components involved? How is massive destruction prevented in cells with constitutive activation of macroautophagy? Do other autophagic forms such as microautophagy also contribute to prolongation of life span? Thus far, studies on aging have focused on individual proteolytic systems. How these systems are coordinated and how changes in this coordination contribute to the aging phenotype are unknown. Do changes in organelle turnover (pexophagy, mitophagy, and nucleophagy) occur during aging? How does activation of autophagy in one tissue affect other tissues? Is there organ specificity for autophagy that could explain the differences observed in the rate at which different organs age? Finally, C. elegans has proven to be a wonderful experimental model for the study of certain age-related diseases, such as several neurodegenerative disorders (Parkinson's disease and Huntington's disease). It is exciting to speculate that this genetic model may be used to decipher the role of autophagy in these diseases. Stay tuned.

September 10, 2003
  1. R. L. Deter, P. Baudhuin, C. De Duve, Participation of lysosomes in cellular autophagy induced in rat liver by glucagon. J. Cell Biol. 35, C11-C16 (1967).[Free Full Text]
  2. Y. Ohsumi, Molecular dissection of autophagy: two ubiquitin-like systems. Nature Rev. Mol. Cell Biol. 2, 211-216 (2001).[CrossRef][Medline]
  3. M. Thumm, Structure and function of the yeast vacuole and its role in autophagy. Microsc. Res. Tech. 51, 563-572 (2000).[CrossRef][Medline]
  4. C. W. Wang, D. J. Klionsky, The molecular mechanism of autophagy. Mol. Med. 9, 65-76 (2003).[Medline]
  5. A. Melendez, Z. Talloczy, M. Scaman, E. L. Eskelinen, D. H. Hall, B. Levine, Essential role of autophagy genes in dauer development and lifespan extension in C. elegans. Science 301, 1387-1391 (2003).[Abstract/Free Full Text]
  6. W. Ward, Protein degradation in the aging organism. Prog. Mol. Subcell. Biol. 29, 35-42 (2002).[Medline]
  7. P. O. Seglen, P. Bohley, Autophagy and other vacuolar protein degradation mechanisms. Experientia 48, 158-172 (1992).[CrossRef][Medline]
  8. A. Meijer, Amino acids as regulators and components of nonproteinogenic pathways. J. Nutr. 133, 2057S-2062S (2003).[Abstract/Free Full Text]
  9. G. Otto, M. Wu, N. Kazgan, O. Anderson, R. Kessin, Macroautophagy is required for multicellular development of the social amoeba Dictyostelium discoideum. J. Biol. Chem. 278, 17636-17645 (2003).[Abstract/Free Full Text]
  10. C. Lee, E. Clough, P. Yellon, T. Teslovich, D. Stephan, E. Baehrecke, Genome-wide analyses of steroid- and radiation-triggered programmed cell death in Drosophila. Curr. Biol. 13, 350-357 (2003).[CrossRef][Medline]
  11. Z. Talloczy, W. Jiang, H.t. Virgin, D. Leib, D. Scheuner, R. Kaufman, E. Eskelinen, B. Levine, Regulation of starvation- and virus-induced autophagy by the eIF2 alpha kinase signaling pathway. Proc. Natl. Acad. Sci. U.S.A. 99, 190-195 (2002).[Abstract/Free Full Text]
  12. E. Ogier-Denis, P. Codogno, Autophagy: a barrier or an adaptive response to cancer. Biochim. Biophys. Acta 1603, 113-128 (2003).[Medline]
  13. K. Kegel, M. Kim, E. Sapp, C. McIntyre, J. Castano, N. Aronin, M. DiFiglia, Huntingtin expression stimulates endosomal-lysosomal activity, endosome tubulation, and autophagy. J. Neurosci. 20, 7268-7278 (2000).[Abstract/Free Full Text]
  14. D. L. Tuttle, W. A. Dunn Jr., Divergent modes of autophagy in the methylotrophic yeast Pichia pastoris. J. Cell Sci. 108, 25-35 (1995).[Abstract/Free Full Text]
  15. Y. Sakai, A. Koller, L. Rangell, G. Keller, S. Subramani, Peroxisome degradation by microautophagy in Pichia pastoris: identification of specific steps and morphological intermediates. J. Cell Biol. 141, 625-636 (1998).[Abstract/Free Full Text]
  16. A. Bellu, J. Kiel, Selective degradation of peroxisomes in yeasts. Microsc. Res. Tech. 61, 161-170 (2003).[CrossRef][Medline]
  17. P. Roberts, S. Moshitch-Moshkovitz, E. Kvam, E. O'Toole, M. Winey, D. Goldfarb, Piecemeal microautophagy of nucleus in Saccharomyces cerevisiae. Mol. Biol. Cell 14, 129-141 (2003).[Abstract/Free Full Text]
  18. J. Lemasters, T. Qian, L. He, J. Kim, S. Elmore, W. Cascio, D. Brenner, Role of mitochondrial inner membrane permeabilization in necrotic cell death, apoptosis, and autophagy. Antioxid. Redox Signal. 4, 769-781 (2002).[CrossRef][Medline]
  19. A. M. Cuervo, J. F. Dice, Lysosomes, a meeting point of proteins, chaperones, and proteases. J. Mol. Med. 76, 6-12 (1998).[CrossRef][Medline]
  20. P. H. Huang, H. L. Chiang, Identification of novel vesicles in the cytosol to vacuole protein degradation pathway. J. Cell Biol. 136, 803-810 (1997).[Abstract/Free Full Text]
  21. S. Goldstein, D. Stotland, R. Cordeiro, Decreased proteolysis and increased amino acid efflux in aging human fibroblasts. Mech. Ageing Dev. 5, 221-233 (1976).[CrossRef][Medline]
  22. A. Reznick, D. Gershon, The effect of age on the protein degradation system in the nematode Turbatrix aceti. Mech. Ageing Dev. 11, 403-415 (1979).[CrossRef][Medline]
  23. U. Brunk, A. Terman, Lipofuscin: mechanisms of age-related accumulation and influence on cell function. Free Radic. Biol. Med. 33, 611-619 (2002).[CrossRef][Medline]
  24. A. Donati, G. Cavallini, C. Paradiso, S. Vittorini, M. Pollera, Z. Gori, E. Bergamini, Age-related changes in the regulation of autophagic proteolysis in rat isolated hepatocytes. J. Gerontol. A Biol. Sci. Med. Sci. 56, B288-B293 (2001).[Abstract/Free Full Text]
  25. A. M. Cuervo, J. F. Dice, Age-related decline in chaperone-mediated autophagy. J. Biol. Chem. 275, 31505-31513 (2000).[Abstract/Free Full Text]
  26. A. Donati, G. Cavallini, C. Paradiso, S. Vittorini, M. Pollera, Z. Gori, E. Bergamini, Age-related changes in the autophagic proteolysis of rat isolated liver cells: effects of antiaging dietary restrictions. J. Gerontol. A Biol. Sci. Med. Sci. 56, B375-B383 (2001).[Abstract/Free Full Text]
  27. J. Doelling, J. Walker, E. Friedman, A. Thompson, R. Vierstra, The APG8/12-activating enzyme APG7 is required for proper nutrient recycling and senescence in Arabidopsis thaliana. J. Biol. Chem. 277, 33105-33114 (2002).[Abstract/Free Full Text]
  28. H. Hanaoka, T. Noda, Y. Shirano, T. Kato, H. Hayashi, D. Shibata, S. Tabata, Y. Ohsumi, Leaf senescence and starvation-induced chlorosis are accelerated by the disruption of an Arabidopsis autophagy gene. Plant Physiol. 129, 1181-1193 (2002).[Abstract/Free Full Text]
  29. L. Guarente, C. Kenyon, Genetic pathways that regulate ageing in model organisms. Nature 408, 255-262 (2000).[CrossRef][Medline]
  30. K. Hennig, T. Neufeld, Inhibition of cellular growth and proliferation by dTOR overexpression in Drosophila. Genesis 34, 107-110 (2002).[CrossRef][Medline]
Citation: A. M. Cuervo, Autophagy and Aging--When "All You Can Eat" Is Yourself. Sci. SAGE KE 2003 (36), pe25 (2003).

Degradation of tau protein by autophagy and proteasomal pathways.
Y. Wang and E. Mandelkow (2012)
Biochm. Soc. Trans. 40, 644-652
   Abstract »    Full Text »    PDF »
Protein quality control during aging involves recruitment of the macroautophagy pathway by BAG3.
M. Gamerdinger, P. Hajieva, A. M. Kaya, U. Wolfrum, F. U. Hartl, and C. Behl (2009)
EMBO J. 28, 889-901
   Abstract »    Full Text »    PDF »
Reactive oxygen species are essential for autophagy and specifically regulate the activity of Atg4.
R. Scherz-Shouval, E. Shvets, E. Fass, H. Shorer, L. Gil, and Z. Elazar (2007)
EMBO J. 26, 1749-1760
   Abstract »    Full Text »    PDF »

Science of Aging Knowledge Environment. ISSN 1539-6150