|Home | About | Journals | Submit | Contact Us | Français|
Pexophagy is a selective autophagy process that degrades damaged and/or superfluous peroxisomes in the yeast vacuole or in mammalian lysosomes. The molecular mechanisms of pexophagy are well studied in yeast. Peroxisomes can be rapidly induced by oleate in the budding yeast, Saccharomyces cerevisiae, and by oleate or methanol in the methylotrophic yeast, Pichia pastoris. A number of peroxisomal matrix enzymes, such as 3-ketoacyl CoA thiolase (thiolase) and alcohol oxidase (AOX), are upregulated correspondingly to meet metabolic demands of the cells. Removal of these peroxisome-inducing carbon sources creates conditions wherein peroxisomes are superfluous and results in pexophagy and the degradation of these peroxisomal matrix enzymes. In this chapter, we discuss different assays to monitor pexophagy in yeast. These assays rely on tracking the localization of the BFP–SKL protein (a peroxisomally targeted version of the blue fluorescent protein) by microscopy, biochemical analysis of the degradation of peroxisomal matrix proteins, thiolase and AOX, and/or measuring the reduction of AOX activity during pexophagy.
Peroxisomes are single-membrane-bound organelles present in most eukaryotic cells. They are involved in essential cellular metabolism, most notably the β-oxidation of fatty acids, as well as the production and degradation of toxic hydrogen peroxide and other reactive oxygen species. The human diseases caused by peroxisome dysfunction underscore the vital importance of this organelle (Ma, Agrawal, & Subramani, 2011).
The number of intracellular peroxisomes can be rapidly adjusted and tightly controlled in response to the changing environment and/or physiological conditions. This can be achieved through the coordination of peroxisome biogenesis and degradation. In rodents, the abundance of peroxisomes increases when the animals are administered with hypolipidemic drugs that are known peroxisome proliferators (Fahimi et al., 1982) or some other chemicals, such as phthalate esters (Yokota, 1986). The number of peroxisomes decreases rapidly and almost returns to the preproliferation state within a week if the drugs are discontinued (Yokota, 1993). Similar dynamic patterns of peroxisome homeostasis (balance between biogenesis and degradation) are also observed in yeast. Peroxisomes are induced when yeast cells are grown in the presence of the specific carbon sources like oleic acid or methanol, whose metabolism in that yeast requires peroxisomal enzymes and metabolism. However, when the favored peroxisome proliferation stimulus is removed, the superfluous and redundant peroxisomes are no longer required and are subjected instead to pexophagy, the selective degradation of peroxisome through autophagy.
Autophagy is a self-eating, dynamic, cellular process involving the sequestration of cytoplasmic constituents into a double-membrane structure called the autophagosome, followed by autophagosome/lysosome fusion and degradation of the cargos in the lysosome or vacuole (Levine & Klionsky, 2004). It is dramatically induced by various extra- and intracellular stresses, such as nutrient or growth factor depletion or mTOR kinase inhibition. Autophagy was initially described as a nonselective bulk degradation process. However, autophagy can also target selective cargos (Farré & Subramani, 2016). Pexophagy involves the general autophagy machinery, as well as specific elements such as the pexophagy receptor, Atg30 in Pichia pastoris (Farre, Manjithaya, Mathewson, & Subramani, 2008), and Atg36 in Saccharomyces cerevisiae (Motley, Nuttall, & Hettema, 2012). These receptors tag peroxisomes for degradation: Atg30 and Atg36 are associated with the peroxisomal membrane proteins (PMPs) and can recruit the core autophagy machinery to peroxisomes by binding the scaffold proteins (Atg11 and Atg17) and the ubiquitin-like protein Atg8 (Farre, Burkenroad, Burnett, & Subramani, 2013; Farre et al., 2008).
P. pastoris serves as an ideal cell model to study pexophagy due to (1) the large size and/or abundance of peroxisomes that can be induced by both methanol and oleate; (2) this process can be easily visualized by both biochemical and microscopy assays; and (3) convenient genetic manipulation of the cells is possible. In this chapter, we will discuss mainly the methods used to monitor pexophagy in P. pastoris. Similar methods can be also applied to S. cerevisiae.
In peroxisome proliferation conditions, such as methanol or oleate medium, a large number of peroxisomal proteins are induced and imported into the peroxisomes utilizing one or more peroxisome-targeting signals (PTSs) present in the amino acid sequences of the proteins. The import of most peroxisomal matrix cargos depends on the PTS1 signal composed of a noncleavable, C-terminal tripeptide, Ser–Lys–Leu (SKL), or its conserved variants (Gould, Keller, Hosken, Wilkinson, & Subramani, 1989). Hence, peroxisomes can be labeled by fusing the BFP (blue fluorescent protein) tag to the SKL sequence (BFP–SKL). Alternatively, PMPs, which are targeted to peroxisomes via an mPTS (membrane PTS) sequence, or peroxisomal matrix proteins may also be fused to GFP or BFP to follow peroxisomes. The fluorescent dye FM4-64 [N-(3-triethylammoniumpropyl)-4-(p-diethylaminophenyl-hexatrienyl) pyridinium dibromide] stains the vacuole membrane through endocytosis and can be used as a marker for the vacuole (Vida & Emr, 1995). BFP–SKL and FM4-64 are used together to follow the localization of peroxisomes relative to the vacuole during pexophagy by fluorescence microscopy.
Peroxisomes are induced in oleate or methanol media, as evidenced by the increased number and size of punctuate BFP–SKL-positive dots compared to that in YPD medium (Fig. 1). Shifting cells from oleate to synthetic dextrose without nitrogen (SD-N) medium results in delivery of the BFP–SKL protein into the vacuole (Fig. 2). Atg30 acts as a selective receptor for pexophagy and its activity depends on the phosphorylation sites S71 and S112 (Farre et al., 2013, 2008). The atg30 deletion mutant strain (atg30Δ) is not able to deliver BFP–SKL to the vacuole and serves as a good negative control; reexpression of Atg30 (wt), but not of Atg30 (S112A), rescues the defect of atg30Δ strain (Fig. 2).
|PPY12||wt P. pastoris strain||his4 arg4||Gould, McCollum, Spong, Heyman, and Subramani (1992)|
|BFP–SKL||Overexpression of BFP–SKL in PPY12||PPY12 PrGAPDH-BFP–SKL::ARG4 his4 arg4||Farre et al. (2008)|
|atg30Δ||atg30 deletion in BFP–SKL-expressing strain||PPY12 atg30Δ::KanMX, PrGAPDH-BFP–SKL::ARG4 his4 arg4||Farre et al. (2008)|
|atg30Δ+Atg30 (wt)||Reexpress Atg30 (wt) in atg30Δ strain||PPY12 atg30Δ::KanMX, PrATG30-Atg30(wt)-HA::HIS4, PrGAPDH-BFP–SKL::ARG4 his4 arg4||Farre et al. (2008)|
|atg30Δ+Atg30 (S112A)||Reexpress Atg30 (S112A) mutant in atg30Δ strain||PPY12 atg30Δ::KanMX, PrATG30-Atg30(S112A)-HA::HIS4, PrGAPDH-BFP–SKL::ARG4 his4 arg4||Farre et al. (2013)|
Peroxisomes are induced rapidly by shifting methylotrophic yeast from glucose to methanol or oleate medium. Correspondingly, a number of enzymes are dramatically upregulated to allow cells to metabolize these carbon sources. Examples are the alcohol oxidase (AOX) for methanol utilization and 3-ketoacyl CoA thiolase (thiolase) for β-oxidation of oleate. Replacing these carbon sources by glucose or ethanol results in the induction of pexophagy and degradation of these enzymes. Hence the protein levels of AOX or thiolase can be used to monitor pexophagy. Below is an example of thiolase detection during pexophagy. In the wt strain, thiolase protein is degraded upon moving cells from oleate to SD-N medium. Consistent with the results based on BFP–SKL localization (Fig. 2), thiolase degradation was blocked in atg30Δ cells and rescued by reexpressing Atg30 (wt) but not the mutant, Atg30 (S112A) (Fig. 3).
The growth media and strains are the same as in Section 2.1.2.
Another way to view thiolase degradation is to detect the cleavage of a thiolase-GFP fusion protein in the vacuole. The GFP tag is fused to the C-terminus of thiolase and its endogenous promoter drives the expression of the fusion protein. During pexophagy, thiolase-GFP protein is delivered to the vacuole, where the GFP moiety is cleaved off by vacuolar hydrolases. GFP is relatively stable in vacuoles compared to the GFP–fusion protein. This allows pexophagy to be monitored through the appearance of free GFP, which would appear and accumulate after shifting the strains from oleate to starvation medium (Fig. 4). The atg8 deletion mutant has a defect in the processing of thiolase-GFP due to blockage of the general autophagy. Hence, the abundance of free GFP reflects the flux of pexophagy and alternatively, the ratio of full-length thiolase-GFP/GFP can give a semiquantitative measure of pexophagy. Our lab used this assay to test the essential phosphorylation sites of Atg30/Atg36 for pexophagy (Farre et al., 2013). In this section, we discuss the methods used to establish stable thiolase-GFP overexpressing strains, as well as detection of processing of thiolase-GFP during pexophagy.
All methylotrophic yeasts possess a common methanol-utilizing pathway. Methanol is first oxidized by AOX in peroxisomes to form formaldehyde and hydrogen peroxide (H2O2), both of which are toxic to cells and are broken down by the enzymes located in the peroxisome (Yurimoto, Oku, & Sakai, 2011). Catalase is a peroxidase that catalyzes and decomposes H2O2 into H2O and oxygen. We can use a similar chemical reaction in vitro to measure the activity of AOX. ABTS (2,2′ -Azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) can be used as a substrate of peroxidase: oxidized by peroxidase in the presence of the oxidizing agent, H2O2. The oxidized ABTS turns green, which is a measure of the activity of AOX and indirectly the abundance of peroxisomes in the AOX plate assay. When cells are shifted from methanol to glucose or ethanol, pexophagy occurs and results in the degradation of AOX protein and disappearance of the green color in the strain patches. Our lab used the assay to demonstrate that Atg37 is required for pexophagy as the atg37 mutant still turns green even under pexophagy-inducing conditions (Nazarko et al., 2014).
In summary, we have described different ways to monitor pexophagy in yeast. Each of them has its own advantages. Biochemical analysis of thiolase or AOX degradation and thiolase–GFP processing is less likely to be biased and can be easily handled for many samples. It can also be quantitative. Microscopic observation of BFP–SKL localization facilitates tracking pexophagy at different stages (Sakai, Koller, Rangell, Keller, & Subramani, 1998). The AOX plate assay allows high-throughput screening of genes required for pexophagy (Stasyk, Nazarko, & Sibirny, 2008). Additionally, however, it is possible that the gene of interest might affect general autophagy and hence pexophagy. Atg8 lipidation and GFP-Atg8 cleavage assays can be employed to exclude this possibility (Cheong & Klionsky, 2008). The above assays are based on different aspects of pexophagy and we recommend using at least two different assays to monitor pexophagy. Similar assays, except the AOX degradation and plate assays, can also be applied to S. cerevisiae because oleate can also induce peroxisomes in this organism.
Since autophagy is a highly conserved cellular process from lower to higher eukaryotes, we assume that this is also true for pexophagy. In fact, the earliest description of pexophagy came from mammalian cells where peroxisomes were detected in the lysosomes (De Duve & Baudhuin, 1966). Though the molecular mechanisms of pexophagy are well established in yeast, little is known in mammalian cells. This is largely due to the challenging detection of the pexophagy process in higher organisms. The pexophagy studies in yeast would definitely help uncover the mysteries of pexophagy in mammalian cells.
We thank Dr. Jean-Claude Farré for careful and critical reading of the manuscript. This work was supported by NIH grant GM 069373 to S.S.