The first report demonstrating the presence of mitochondria within an autophagosome was made in 1957. Since then, it has been thought that mitochondria are nonselectively engulfed by autophagosomes. Recent studies, however, reveal that autophagic degradation of mitochondria (mitophagy) is a selective pathway. Mitochondria are the organelle that carry out a number of important metabolic processes such as fatty acid oxidation, the citric acid cycle and oxidative phosphorylation. Mitochondrial oxidative phosphorylation supplies a large amount of energy that contributes to a range of cellular activities. On the other hand, mitochondria are a major source of reactive oxygen species that cause oxidative damage to mitochondrial DNA, protein and lipid. Accordingly, proper repair of damaged molecules or elimination of dysfunctional mitochondria is important to maintain optimal cellular homeostasis. Recent evidence from yeast to mammal suggests that mitophagy plays an important role in mitochondria quality control. However, the molecular mechanism of mitophagy is poorly understood.
To figure out the molecular mechanism of mitophagy, we recently established the Om45-GFP processing assay to monitor this process. Om45 is a mitochondrial outer membrane protein. Following the uptake of mitochondria into the vacuole, Om45-GFP is degraded, resulting in the release of the intact form of GFP, which is detected by immunoblotting, or the appearance of green fluorescence that accumulates in the vacuole lumen. With this method, we screened a yeast knockout library for strains that are deficient in mitophagy. Among 4,667 strains, we find 32 strains that show a complete or partial block of mitophagy, in addition to the ATG gene knockout strains. Among 32 genes for which the corresponding knockout strain shows defective mitophagy, nine of them are related to membrane trafficking pathways. It is widely thought that defects in membrane trafficking affect the lipid supply that is needed for extension of the phagophore, the initial sequestering compartment that generates the autophagosome. Indeed, the knockout strains of these nine genes display blocks in the Cvt pathway and macroautophagy. Accordingly, we further focused on the remaining 23 genes. We observed the Cvt pathway and macroautophagy in each gene knockout strain and observed the GFP tagged gene product localization. All of the remaining 23 genes are not required for the Cvt pathway. Ten of them are not required for macroautophagy, suggesting that these are candidates for mitophagy-specific genes, although 13 of them are partially required for macroautophagy. These gene products’ function, localization and requirement for each type of autophagy are summarized in . The functions of these proteins are involved in diverse cellular processes; in most cases it is difficult to explain their involvement in mitophagy based on their reported functions. Atg32 and Dnm1, however, have functions that provide a known or reasonable explanation for their role in mitophagy. Atg32 is recently identified as a mitophagy-specific protein that is a mitochondrial receptor required for mitochondria recognition as a cargo. Dnm1 is a mitochondrial dynamin-related GTPase required for mitochondrial fission. The requirement of Dnm1 for mitophagy is in agreement with previous reports that the fragmentation of mitochondria is a prerequisite for mitophagy in mammalian cells, and the dnm1Δ strain inhibits the mitophagy induced by mdm38 conditional depletion in yeast.
Genes required for mitophagy
The mitophagy-related genes that we found from the screen include eight genes of unknown function. Among them, we decided to initially focus on YLR356W, because the gene product, Ylr356w, is reported to localize on mitochondria. Using a biochemical approach, we confirmed that Ylr356w predominantly localizes to the mitochondrial outer membrane. Although macroautophagy, the Cvt pathway and pexophagy are normal in the ylr356wΔ strain, mitophagy is substantially defective. Accordingly, we conclude that YLR356W is a mitophagy-specific gene and named this gene as ATG33. Interestingly, deletion of ATG33 partially inhibits (36% decrease compared with wild type) mitophagy during starvation, whereas there is almost complete inhibition (more that 80% decrease compared with wild type) at post-log phase in YPL medium (with lactic acid as the primary carbon source). The specific requirement of Atg33 for mitophagy primarily at the post-log phase suggests that there is some difference(s) between the pathways of mitophagy induced during starvation versus post-log phase growth. Because only a fraction of the total mitochondria pool is degraded by mitophagy at post-log phase, it is reasonable to think that aged or dysfunctional mitochondria are selected for degradation. Based on this idea, we hypothesize that Atg33 contributes to the detection or presentation of aged or dysfunctional mitochondria for mitophagy when cells reach the post-log phase.
We demonstrated a mitophagy defect in the majority of the ATG gene knockout strains independent of the genome-wide screen discussed above. ATG genes that play a fundamental role in autophagy (ATG1 to 10, 12 to 16 and 18) are essential for mitophagy, suggesting that the core autophagic machinery is used in common between mitophagy and other types of autophagy. Notably, ATG11, ATG20 and ATG24, which are required for both the Cvt pathway and pexophagy but not for macroautophagy, are required for mitophagy, suggesting that selective autophagy uses certain components in common that are in part different from those used in macroautophagy, although the function of ATG20 and ATG24 is poorly understood. In addition to these genes encoding selective components, ATG17, ATG29 and ATG31, which are required for macroautophagy but not for the Cvt pathway, are also required for mitophagy.
In this screen, we identified 23 mitophagy-related genes. However, the requirement of these genes for mitophagy is poorly understood. So far, only three genes are identified, ATG11, ATG32 and ATG33, which are specifically required as part of the mitochondria degradation machinery. Unfortunately, none of these genes have homologs that are identified in higher eukaryotes. The identification of homologs or counterparts of these genes should be the next step that can allow us to understand the physiological role of mitophagy.