Localized de novo phospholipid synthesis drives autophagosome biogenesis

ABSTRACT

During (macro)autophagy, cells form transient organelles, termed autophagosomes, to target a broad spectrum of substrates for degradation critical to cellular and organismal health. Driven by rapid membrane assembly, an initially small vesicle (phagophore) elongates into a large cup-shaped structure to engulf substrates within a few minutes in a double-membrane autophagosome. In particular, how autophagic membranes expand has been a longstanding question. Here, we summarize our recent work that delineates a pathway that drives phagophore expansion by localized de novo phospholipid synthesis. Specifically, we found that the conserved acyl-CoA synthetase Faa1 localizes to nucleated phagophores to locally activate fatty acids for de novo phospholipid synthesis in the neighboring ER. These newly synthesized phospholipids are then preferentially incorporated into autophagic membranes and drive the expansion of the phagophore into a functional autophagosome. In summary, our work uncovers molecular principles of how cells coordinate phospholipid synthesis and flux with autophagic membrane formation during autophagy.

Abbreviations: ACS: acyl-CoA synthestases; CoA: coenzyme A; ER: endoplasmic reticulum     

Folding into an autophagosome

Autophagosomes arise in yeast and animals from the sealing of a cup-shaped double-membrane precursor, the phagophore. The concerted action of about 30 evolutionarily conserved autophagy related (ATG) proteins lies at the core of this process. However, the mechanisms allowing phagophore generation and its differentiation into a sealed autophagosome are still not clear in detail, and very little is known in plants. This is due in part to the scarcity of structurally informative, real-time imaging data of ATG proteins at the phagophore site. Among these, the ATG5 complex directs anchoring of ATG8 to the phagophore, an event required for membrane expansion. Detailed real-time and 3D imaging of ATG5, ATG8, and an ER marker at the expanding phagophore allowed us to propose a model for autophagosome formation in plants. This model implies tight connections of the growing phagophore with the outer face of the cortical endoplasmic reticulum and prompts new questions on the mechanism of autophagosome biogenesis.     

Folding into an autophagosome: ATG5 sheds light on how plants do it

Autophagosomes arise in yeast and animals from the sealing of a cup-shaped double-membrane precursor, the phagophore. The concerted action of about 30 evolutionarily conserved autophagy related (ATG) proteins lies at the core of this process. However, the mechanisms allowing phagophore generation and its differentiation into a sealed autophagosome are still not clear in detail, and very little is known in plants. This is due in part to the scarcity of structurally informative, real-time imaging data of ATG proteins at the phagophore site. Among these, the ATG5 complex directs anchoring of ATG8 to the phagophore, an event required for membrane expansion. Detailed real-time and 3D imaging of ATG5, ATG8, and an ER marker at the expanding phagophore allowed us to propose a model for autophagosome formation in plants. This model implies tight connections of the growing phagophore with the outer face of the cortical endoplasmic reticulum and prompts new questions on the mechanism of autophagosome biogenesis.     

The plasma membrane as a control center for autophagy

Autophagosomes may derive membrane from diverse sources, including the plasma membrane, Golgi, endoplasmic reticulum and mitochondria. The plasma membrane contributes membrane to ATG12–ATG5-ATG16L1-positive phagophore precursor vesicles (LC3-negative) by both clathrin-dependent and -independent routes. We recently observed that ARF6 regulates autophagy and that this could be explained, at least in part, by its role in the generation of phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P₂], which influences endocytic uptake of plasma membrane into autophagosome precursors. The subsequent maturation of these small phagophore precursors into phagophores (ATG12–ATG5-ATG16L1-positive and LC3-positive), is assisted by SNARE-mediated homotypic fusion that increase their size and enhance their ability to acquire LC3-II. It appears that a plasma membrane-derived pool of VAMP7 is a key mediator of these fusion events. Thus, events at the plasma membrane may regulate distinct steps in the biogenesis of phagophores.     

The plasma membrane as a control center for autophagy

Autophagosomes may derive membrane from diverse sources, including the plasma membrane, Golgi, endoplasmic reticulum and mitochondria. The plasma membrane contributes membrane to ATG12-ATG5-ATG16L1-positive phagophore precursor vesicles (LC3-negative) by both clathrin-dependent and -independent routes. We recently observed that ARF6 regulates autophagy and that this could be explained, at least in part, by its role in the generation of phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P 2], which influences endocytic uptake of plasma membrane into autophagosome precursors. The subsequent maturation of these small phagophore precursors into phagophores (ATG12-ATG5-ATG16L1-positive and LC3-positive), is assisted by SNARE-mediated homotypic fusion that increase their size and enhance their ability to acquire LC3-II. It appears that a plasma membrane-derived pool of VAMP7 is a key mediator of these fusion events. Thus, events at the plasma membrane may regulate distinct steps in the biogenesis of phagophores.     

Phagophores evolve from recycling endosomes

The membrane origins of autophagosomes have been a key unresolved question in the field. The earliest morphologically recognizable structure in the macroautophagy/autophagy itinerary is the double-membraned cup-shaped phagophore. Newly formed phosphatidylinositol 3-phosphate (PtdIns3P) on the membranes destined to become phagophores recruits WIPI2, which, in turn, binds ATG16L1 to define the sites of autophagosome formation. Here we review our recent study showing that membrane recruitment of WIPI2 requires coincident detection of PtdIns3P and RAB11A, a protein that marks recycling endosomes. We found that multiple core autophagy proteins are more tightly associated with the recycling endosome compartment than with endoplasmic reticulum (ER)-mitochondrial contact sites. Furthermore, biochemical isolation of the recycling endosomes confirmed that they recruit autophagy proteins. Finally, fixed and live-cell imaging data revealed that recycling endosomes engulf autophagic substrates. Indeed, the sequestration of mitochondria after mitophagy stimulation depends on early autophagy regulators. These data suggest that autophagosomes evolve from the RAB11A compartment.     

Electron tomography reveals the endoplasmic reticulum as a membrane source for autophagosome formation

The origin and source of autophagosomal membranes are long-standing questions. By electron microscopy, we show that the endoplasmic reticulum (ER) associates with early autophagic structures called isolation membranes (IM) or phagophores in mammalian culture cells. Overexpression of a mutant of Atg4B, which causes defects in autophagosome formation, caused accumulation of ER-IM complexes. Electron tomography revealed the ER-IM complex as a subdomain of the ER forming a cradle encircling the IM, and showed that both ER and isolation membranes are interconnected.     

Syntaxin 17

The phagophore (also called isolation membrane) elongates and encloses a portion of cytoplasm, resulting in formation of the autophagosome. After completion of autophagosome formation, the outer autophagosomal membrane becomes ready to fuse with the lysosome for degradation of enclosed cytoplasmic materials. However, the molecular mechanism for how the fusion of completed autophagosomes with the lysosome is regulated has not been fully understood. We discovered syntaxin 17 (STX17) as an autophagosomal soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE). STX17 has a hairpin-type structure mediated by two transmembrane domains, each containing glycine zipper motifs. This unique transmembrane structure contributes to its specific localization to completed autophagosomes but not to phagophores. STX17 interacts with SNAP29 and the lysosomal SNARE VAMP8, and all of these proteins are required for autophagosome–lysosome fusion. The late recruitment of STX17 to completed autophagosomes could prevent premature fusion of the lysosome with unclosed phagophores.     

Differing susceptibility to autophagic degradation of two LC3-binding proteins: SQSTM1/p62 and TBC1D25/OATL1

MAP1LC3/LC3 (a mammalian ortholog family of yeast Atg8) is a ubiquitin-like protein that is essential for autophagosome formation. LC3 is conjugated to phosphatidylethanolamine on phagophores and ends up distributed both inside and outside the autophagosome membrane. One of the well-known functions of LC3 is as a binding partner for receptor proteins, which target polyubiquitinated organelles and proteins to the phagophore through direct interaction with LC3 in selective autophagy, and their LC3-binding ability is essential for degradation of the polyubiquitinated substances. Although a number of LC3-binding proteins have been identified, it is unknown whether they are substrates of autophagy or how their interaction with LC3 is regulated. We previously showed that one LC3-binding protein, TBC1D25/OATL1, plays an inhibitory role in the maturation step of autophagosomes and that this function depends on its binding to LC3. Interestingly, TBC1D25 seems not to be a substrate of autophagy, despite being present on the phagophore. In this study we investigated the molecular basis for the escape of TBC1D25 from autophagic degradation by performing a chimeric analysis between TBC1D25 and SQSTM1/p62 (sequestosome 1), and the results showed that mutant TBC1D25 with an intact LC3-binding site can become an autophagic substrate when TBC1D25 is forcibly oligomerized. In addition, an ultrastructural analysis showed that TBC1D25 is mainly localized outside autophagosomes, whereas an oligomerized TBC1D25 mutant rather uniformly resides both inside and outside the autophagosomes. Our findings indicate that oligomerization is a key factor in the degradation of LC3-binding proteins and suggest that lack of oligomerization ability of TBC1D25 results in its asymmetric localization at the outer autophagosome membrane.     

A non-autophagic pathway for diversion of ER secretory proteins to lysosomes

Intracisternal granules (ICG) develop in the rough ER of hyperstimulated thyrotrophs or thyroid hormone-secreting cells of the anterior pituitary gland. To determine the fate of these granules, we carried out morphological and immunocytochemical studies on pituitaries of thyroxine-treated, thyroidectomized rats. Under these conditions the ER of thyrotrophs is dramatically dilated and contains abundant ICG; the latter contain beta subunits of thyrotrophic hormone (TSH-beta). Based on purely morphologic criteria, intermediates were identified that appeared to represent stages in the transformation of a part rough/part smooth ER cisterna into a lysosome. Using immunocytochemical and cytochemical markers, two major types of intermediates were distinguished: type 1 lacked ribosomes but were labeled with antibodies against both ER markers (PDI, KDEL, ER membrane proteins) and a lysosomal membrane marker, lgp120. They also were reactive for the lysosomal enzyme, acid phosphatase, by enzyme cytochemistry. Type 2 intermediates were weakly reactive for ER markers and contained both lgp120 and lysosomal enzymes (cathepsin D, acid phosphatase). Taken together these results suggest that in hyperstimulated thyrotrophs part rough/part smooth ER elements containing ICG lose their ribosomes, their membrane is modified, and they sequentially acquire a lysosome-type membrane and lysosomal enzymes. The findings are compatible with the conclusion that a pathway exists by which under certain conditions, secretory proteins present in the ER as well as ER membrane and content proteins can be degraded by direct conversion of ER cisternae into lysosomes.     

The ULK complex initiates autophagosome formation at phosphatidylinositol synthase-enriched ER subdomains

In our recent paper, we biochemically analyzed autophagosome-related membranes at the initiation stage of macroautophagy/autophagy using atg knockout (KO) cells and demonstrated that the ULK complex is recruited to 2 distinct membranes: the ER membrane and ATG9A-positive autophagosome precursors. We have also identified phosphatidylinositol synthase (PIS)-enriched ER subdomains as the initiation site of autophagosome formation. Based on these findings, we propose that the ULK complex, the PIS-enriched ER subdomain, and ATG9A vesicles together initiate autophagosome formation.     

ER–plasma membrane contact sites contribute to autophagosome biogenesis by regulation of local PI3P synthesis

The double‐membrane‐bound autophagosome is formed by the closure of a structure called the phagophore, origin of which is still unclear. The endoplasmic reticulum (ER) is clearly implicated in autophagosome biogenesis due to the presence of the omegasome subdomain positive for DFCP1, a phosphatidyl‐inositol‐3‐phosphate (PI3P) binding protein. Contribution of other membrane sources, like the plasma membrane (PM), is still difficult to integrate in a global picture. Here we show that ER–plasma membrane contact sites are mobilized for autophagosome biogenesis, by direct implication of the tethering extended synaptotagmins (E‐Syts) proteins. Imaging data revealed that early autophagic markers are recruited to E‐Syt‐containing domains during autophagy and that inhibition of E‐Syts expression leads to a reduction in autophagosome biogenesis. Furthermore, we demonstrate that E‐Syts are essential for autophagy‐associated PI3P synthesis at the cortical ER membrane via the recruitment of VMP1, the stabilizing ER partner of the PI3KC3 complex. These results highlight the contribution of ER–plasma membrane tethers to autophagosome biogenesis regulation and support the importance of membrane contact sites in autophagy.     

Ultrastructural relationship of the phagophore with surrounding organelles

Phagophore nucleates from a subdomain of the endoplasmic reticulum (ER) termed the omegasome and also makes contact with other organelles such as mitochondria, Golgi complex, plasma membrane and recycling endosomes during its formation. We have used serial block face scanning electron microscopy (SB-EM) and electron tomography (ET) to image phagophore biogenesis in 3 dimensions and to determine the relationship between the phagophore and surrounding organelles at high resolution. ET was performed to confirm whether membrane contact sites (MCSs) are evident between the phagophore and those surrounding organelles. In addition to the known contacts with the ER, we identified MCSs between the phagophore and membranes from putative ER exit sites, late endosomes or lysosomes, the Golgi complex and mitochondria. We also show that one phagophore can have simultaneous MCSs with more than one organelle. Future membrane flux experiments are needed to determine whether membrane contacts also signify lipid translocation.     

Mechanisms of autophagosome biogenesis

Autophagy is a unique membrane trafficking process whereby newly formed membranes, termed phagophores, engulf parts of the cytoplasm leading to the production of double-membraned autophagosomes that get delivered to lysosomes for degradation. This catabolic pathway has been linked to numerous physiological and pathological conditions, such as development, programmed cell death, cancer, pathogen infection, neurodegenerative disorders, and myopathies. In this review, we will focus on recent studies in yeast and mammalian systems that have provided insights into two critical areas of autophagosome biogenesis - the source of the autophagosomal membranes, and the mechanisms regulating the fusion of the edges of the double-membraned phagophores to form autophagosomes.     

Live-cell imaging of Aspergillus nidulans autophagy

We exploited the amenability of the fungus Aspergillus nidulans to genetics and live-cell microscopy to investigate autophagy. Upon nitrogen starvation, GFP-Atg8-containing pre-autophagosomal puncta give rise to cup-shaped phagophores and circular (0.9-μm diameter) autophagosomes that disappear in the vicinity of the vacuoles after their shape becomes irregular and their GFP-Atg8 fluorescence decays. This ‘autophagosome cycle’ gives rise to characteristic cone-shaped traces in kymographs. Autophagy does not require endosome maturation or ESCRTs, as autophagosomes fuse with vacuoles directly in a RabS (homolog of Saccharomyces cerevisiae Ypt7 and mammalian RAB7; written hereafter as RabS^RAB7)-HOPS-(homotypic fusion and vacuole protein sorting complex)-dependent manner. However, by removing RabS^RAB7 or Vps41 (a component of the HOPS complex), we show that autophagosomes may still fuse, albeit inefficiently, with the endovacuolar system in a process almost certainly mediated by RabA^RAB5/RabB^RAB5 (yeast Vps21 homologs)-CORVET (class C core vacuole/endosome tethering complex), because acute inactivation of HbrA/Vps33, a key component of HOPS and CORVET, completely precludes access of GFP-Atg8 to vacuoles without affecting autophagosome biogenesis. Using a FYVE₂-GFP probe and endosomal PtdIns3P-depleted cells, we imaged PtdIns3P on autophagic membranes. PtdIns3P present on autophagosomes decays at late stages of the cycle, preceding fusion with the vacuole. Autophagy does not require Golgi traffic, but it is crucially dependent on RabO^RAB1. TRAPPIII-specific factor AN7311 (yeast Trs85) localizes to the phagophore assembly site (PAS) and RabO^RAB1 localizes to phagophores and autophagosomes. The Golgi and autophagy roles of RabO^RAB1 are dissociable by mutation: rabO^A136D hyphae show relatively normal secretion at 28°C but are completely blocked in autophagy. This finding and the lack of Golgi traffic involvement pointed to the ER as one potential source of membranes for autophagy. In agreement, autophagosomes form in close association with ring-shaped omegasome-like ER structures resembling those described in mammalian cells.     

Plasma membrane helps autophagosomes grow

The membrane origin of autophagosomes has long been a mystery and it may involve multiple sources. In this punctum, we discuss our recent finding that the plasma membrane contributes to the formation of pre-autophagic structures via clathrin-mediated endocytosis. Our study suggests that Atg16L1 interacts with clathrin heavy-chain/AP2 and is also localized on vesicles (positive for clathrin or cholera toxin B) close to the plasma membrane. Live-cell imaging studies revealed that the plasma membrane contributes to Atg16L1-positive structures and that this process and autophagosome formation are impaired by knockdowns of genes regulating clathrin-mediated endocytosis.Key words: autophagy, plasma membrane, endocytosis, phagophore, originWhere do autophagosomes get their membrane from? Although the field of autophagy has grown tremendously since its discovery a few decades ago, the origin(s) of the membranes that contribute to autophagosome biogenesis has been a mystery among autophagy researchers until recently. Mammalian autophagosomes are formed randomly throughout the cytoplasm via a process that involves elongation and fusion of phagophores to form double-membraned autophagosomes. This process involves two ubiquitin-like conjugation systems: conjugation of Atg12 to Atg5 that later forms a macromolecular complex with Atg16L1, and conjugation of phosphatidylethanolamine (PE) with Atg8/LC3-I. The Atg12-Atg5-Atg16L1 complex is targeted to the preautophagic structures, which then acquire Atg8. Atg12-Atg5-Atg16L1 dissociates from completed autophagosomes, while LC3-PE (LC3-II) is associated both with pre-autophagic structures and completed autophagosomes.Some recent studies have explored the contribution of membranes from different organelles supporting the general idea that autophagosomes derive membranes from pre-existing organelles. It is quite possible that there may be multiple membrane sources involved. A few groups have revisited the hypothesis that the endoplasmic reticulum (ER) may be one of the membrane donors. High-resolution 2D electron microscopy (EM) and 3D EM-tomography studies have revealed connections between the ER and the growing autophagosomes. Whether the ER contributes to general autophagy or a specific form of autophagy, reticulophagy, remains to be determined. In addition, it has not been shown if ER membrane is required for autophagosome formation. Recently another study has reported that autophagosomes receive lipids from the outer mitochondrial membrane, but only under starvation conditions, again fueling the multiple-membrane source hypothesis.We have now found evidence for plasma membrane contribution to pre-autophagic structures via endocytosis. Unlike the previous studies, which have focused on LC3- positive structures, we looked specifically at the Atg5-, Atg12- and Atg16-positive pre-autophagic structures, an idea that stemmed from our finding that clathrin heavy-chain immunoprecipitates with Atg16L1. We think that this interaction is partly mediated by the adaptor protein AP2, since knockdown of AP2 decreases the clathrin heavy-chain-Atg16L1 interaction. Immunogold EM also shows clathrin localization on Atg16L1-labeled vesicles close to the plasma membrane.These findings led us to test whether knockdown of proteins involved in clathrin-mediated endocytosis affected Atg16L1-positive pre-autophagic structures. Indeed, knockdown of key proteins in the clathrin-mediated endocytic pathway results in a decrease in the formation of Atg16L1-positive structures both under basal or autophagy-induced conditions (starvation or trehalose treatment). This correlates with a decrease in the number of LC3-labeled autophagosomes. When we directly analyzed vesicle fusion by livecell microscopy, we observed that vesicles endocytosed from the plasma membrane fuse to the Atg16L1-positive vesicles close to the plasma membrane. This was confirmed by immuno-EM when we found cholera toxin B-labeling (used to label plasma membrane that is subsequently internalized by endocytosis) on Atg16L1-vesicles. We noticed that overexpression of an Atg16L1 mutant that does not bind clathrin heavy-chain does not form Atg16L1-vesicular structures in the way we see with wild-type Atg16L1, suggesting that the binding of Atg16L1 to AP2/clathrin is required for the subsequent formation of the Atg16L1 vesicles.When we blocked endocytic vesicle scission (using both genetic and chemical inhibitors) we found that Atg16L1 strongly immunoprecipitates with clathrin-heavy chain probably due to the accumulation of clathrin-Atg16L1 structures at the plasma membrane that failed to pinch off. This was strongly supported by our fluorescence microscopy and immuno-EM studies that showed what we predicted—accumulation of Atg16L1 at the plasma membrane. This suggests that Atg16L1 in a complex with AP2/clathrin is targeted to the plasma membrane and subsequently internalized as Atg16L1-positive structures. Thus, our data strongly suggest that plasma membrane contributes to early autophagic precursors that subsequently mature to form phagophores (Fig. 1).Open in a separate windowFigure 1Plasma membrane contributes to the formation of early autophagic precursors. Previous studies show that delivery of fully formed autophagosomes to lysosomes requires fusion of such autophagosomes with early or late endosomes to form amphisomes, which are Atg16L1-negative, LC3-positive and are also positive for endosomal markers. We show that blocking clathrin-mediated endocytosis inhibits formation of Atg16L1-positive structures that mature to form phagophores and later autophagosomes. These Atg16L1-vesicles are positive for other early autophagosomal markers like Atg5 and Atg12, but are negative for early endosomal markers like EEA1, suggesting that they are high up in the autophagosome biogenesis cascade. Inhibition of dynamin with Dynsasore or the use of a dominant negative K44A mutant blocks scission and results in Atg16L1 accumulation on the plasma membrane, suggesting that endosomal scission is critical for this process.Although previous studies suggest that completely formed autophagosomes need to fuse with early or late endosomes in order for subsequent autophagosomelysosome fusion to occur, they did not look at the formation of pre-autophagic structures. Our study shows that active endocytosis is required both for the formation of autophagosomes, when very early endocytic intermediates immediately pinching off the plasma membrane (not early endosomes) fuse with Atg16L1-positive structures to form phagophores, and also for maturation of autophagosomes when early or late endosomes fuse with Atg16L1-negative but LC3-positive autophagosomes to form amphisomes. Since blocking clathrin-mediated endocytosis does not completely abrogate autophagosome formation, we believe that other endocytic pathways may have a similar role. Depending on the cell type or the physiological conditions, the contributions from the different endocytic pathways may vary accordingly. It will be interesting to know if the endocytic pathway continuously delivers membrane for early steps in autophagy as the preautophagic structures grow and mature to form autophagosomes, deriving membrane from other sources.     

A role for Atg8–PE deconjugation in autophagosome biogenesis

Formation of the autophagosome is likely the most complex step of macroautophagy, and indeed it is the morphological and functional hallmark of this process; accordingly, it is critical to understand the corresponding molecular mechanism. Atg8 is the only known autophagy-related (Atg) protein required for autophagosome formation that remains associated with the completed sequestering vesicle. Approximately one-fourth of all of the characterized Atg proteins that participate in autophagosome biogenesis affect Atg8, regulating its conjugation to phosphatidylethanolamine (PE), localization to the phagophore assembly site and/or subsequent deconjugation. An unanswered question in the field regards the physiological role of the deconjugation of Atg8–PE. Using an Atg8 mutant that bypasses the initial Atg4-dependent processing, we demonstrate that Atg8 deconjugation is an important step required to facilitate multiple events during macroautophagy. The inability to deconjugate Atg8–PE results in the mislocalization of this protein to the vacuolar membrane. We also show that the deconjugation of Atg8–PE is required for efficient autophagosome biogenesis, the assembly of Atg9-containing tubulovesicular clusters into phagophores/autophagosomes, and for the disassembly of PAS-associated Atg components.     

A role for Atg8-PE deconjugation in autophagosome biogenesis

Formation of the autophagosome is likely the most complex step of macroautophagy, and indeed it is the morphological and functional hallmark of this process; accordingly, it is critical to understand the corresponding molecular mechanism. Atg8 is the only known autophagy-related (Atg) protein required for autophagosome formation that remains associated with the completed sequestering vesicle. Approximately one-fourth of all of the characterized Atg proteins that participate in autophagosome biogenesis affect Atg8, regulating its conjugation to phosphatidylethanolamine (PE), localization to the phagophore assembly site and/or subsequent deconjugation. An unanswered question in the field regards the physiological role of the deconjugation of Atg8-PE. Using an Atg8 mutant that bypasses the initial Atg4-dependent processing, we demonstrate that Atg8 deconjugation is an important step required to facilitate multiple events during macroautophagy. The inability to deconjugate Atg8-PE results in the mislocalization of this protein to the vacuolar membrane. We also show that the deconjugation of Atg8-PE is required for efficient autophagosome biogenesis, the assembly of Atg9-containing tubulovesicular clusters into phagophores/autophagosomes, and for the disassembly of PAS-associated Atg components.