Pexophagy in plants: a mechanism to remit cells from oxidative damage caused under high-intensity light

ABSTRACT

Light is essential for plant growth, but excessive light energy produces reactive oxygen species (ROS), which can seriously damage cells. Mutants defective in ATG (autophagy related) genes show light intensity-dependent leaf damage and ROS accumulation. We found that autophagy is one of the crucial systems in protecting plants from ROS-induced damage by removing oxidative peroxisomes. Damaged peroxisomes are targeted by the PtdIns3P marker and specifically engulfed by phagophores labeled by ATG18a-GFP. Under high-intensity light, huge peroxisome aggregates are induced and captured by vacuolar membranes. Research provides a deeper understanding of plant stress response to light irradiation.

Plants convert light energy to chemical energy and fix carbon dioxide into organic compounds such as sugars and starches used for plant development. This photosynthesis is assisted by photorespiration, a metabolic process across three organelles: chloroplasts, mitochondria and peroxisomes. While light is essential for plant life, excessive light causes the accumulation of ROS, leading to cellular disturbances such as disrupting the photosynthetic machinery in chloroplasts. Peroxisomes are also an ROS source; peroxisomal oxidases generate hydrogen peroxide during photorespiration. Thus, plants have diverse mechanisms to prevent ROS accumulation. CAT (catalase) in peroxisomes is one of the major antioxidant enzymes and detoxifies hydrogen peroxide. However, CAT is inactivated under a high concentration of hydrogen peroxide; consequently, peroxisomes suffer oxidative damage and become toxic compartments in the cell.

In our previous study, we demonstrated that oxidized peroxisomes are selectively targeted by the autophagosome marker ATG8a and that autophagy-deficient mutants fail to degrade such peroxisomes. We recently isolated and analyzed two Arabidopsis mutants defective in peroxisome degradation, atg2/peup1 and atg7/peup4 [1]. These mutant cells contain an increased number of peroxisomes compared with wild-type cells. These excess peroxisomes generate peroxisome aggregates, which consist of oxidized peroxisomes accumulating a high amount of inactive CAT. Detailed analysis using electron microscopy displayed ER- and autophagosome-like structures adjacent to the peroxisomes in autophagy mutants. In the plant pexophagy study, the location of phosphatidylinositol-3-phosphate (PtdIns3P) synthesis, the origin of the phagophore membrane, and the mechanism of phagophore formation were unknown. Therefore, we used ATG18a-GFP-expressing plants to monitor initiation of phagophore formation. Because ATG18a has a well-conserved FYVE domain that binds to PtdIns3P, we also used the GFP-2×FYVE-expressing plants to monitor cellular PtdIns3P. GFP-2×FYVE and ATG18a-GFP preferentially target peroxisome aggregates in the atg2 and atg7 mutants. Although these GFP markers are observed on peroxisomes in wild-type cells, the number of marker-targeted peroxisomes is significantly restricted; this is probably because targeted peroxisomes are promptly removed by autophagy in the wild type. The ATG18a-GFP and GFP-2×FYVE are mainly observed as dots on the peroxisomes in both the atg2 and atg7 mutants. Some portion of the ATG18a-GFP structures are seen in the shape of a ring surrounding the peroxisome in the atg7 but not in the atg2 mutant. Ring structures are rarely observed in the GFP-2×FYVE line. Electron microscopy analysis showed GFP-2×FYVE signals on the peroxisomal membrane. These results indicate that PtdIns3P is generated adjacent to the peroxisomes and attracts ATG18a before the action of ATG2 and ATG7, and that ATG2 is essential for the phagophore to surround the peroxisomes in the process of macropexophagy (Figure 1).

Figure 1.

A schematic model of pexophagy. ROS produced by the peroxisomal metabolic system gradually oxidize the lumenal proteins (including CAT), which form aggregates (1). ROS level in the peroxisome is recognized by an unknown protein on the peroxisomal membrane and induces the synthesis of PtdIns3P on the peroxisomal or adjacent membrane (2). PtdIns3P is recognized by ATG18, which then recruits a group of ATG factors required for forming the phagophore membrane (PG, 3) to proceed with macropexophagy (4). In the atg2 and atg7 mutants, the respective steps of the process are abrogated, and peroxisomes accumulate as aberrant aggregates (5). Under high-intensity light, excessive damaged peroxisomes form aggregates that are directly engulfed by the vacuole via micropexophagy (6).

Because the activation of pexophagy is associated with the oxidative status of peroxisomes, the effect of ROS from excessive light was subsequently investigated. Illumination with high-intensity light induces bleaching of leaves and chlorophyll degradation in autophagy-deficient mutants. A high amount of CAT accumulates in the inactive form in the mutants. Staining ROS with H2-DCF revealed that peroxisomes contain approximately three times higher amounts of ROS than chloroplasts. The peroxisomes form huge aggregates, which remarkably increase in size and frequency compared to normal light conditions. The ATG18a-GFP and GFP-2×FYVE accumulation on aggregates become prominent compared with that under normal-intensity light and are greater in atg7 than in atg2 mutants. The tonoplast surrounds the part of the huge peroxisome aggregate in atg7, i.e., the aggregate is embedded in a cavity on the vacuole. This phenomenon is reminiscent of chlorophagy, wherein the vacuole captures the entire damaged chloroplast via the microautophagy process. The contact between the huge peroxisome aggregates and the vacuole is an active connection, as the aggregates remain attached to the tonoplast after the vacuole was isolated. Peroxisomes surrounded by tonoplasts are also found in wild-type cells under high-intensity light. These results indicate that excessive light accelerates peroxisome oxidation and the formation of peroxisome aggregates (Figure 1). As the aggregate degradation is markedly interrupted in atg7, ATG7 and its product ATG8–PE may have an important role in the uptake of aggregates into the vacuole during the micropexophagy process.

This study suggests that pexophagy has two possible pathways. The first is macropexophagy observed under normal-light conditions, in which ER-related phagophores envelop oxidized peroxisomes and transport them into the vacuole. The second is micropexophagy induced under high-intensity light conditions, in which the vacuole directly takes up aggregates of oxidized peroxisomes. However, it is still unclear how these two pathways are distinguished and utilized in the cell. The induction of macropexophagy can be connected to ROS, as overexpression of CAT suppresses the increase and accumulation of peroxisomes. The synthesis of PtdIns3P at or adjacent to the peroxisomal membrane corresponds to the top of the autophagy machinery to recruit other related proteins. However, factors that convert the lumenal ROS state into extra-membrane signaling, such as ATM (ATM serine/threonine kinase) in mammalian cells, have not been found in plants, and further exploration for such a factor is indispensable. Furthermore, nothing is known about switching machinery to micropexophagy in plants; is there a specific inductive signal or factor, or does it depend on target size and oxidation level? The current problem is that these two pexophagy pathways are not well distinguished and resolving this is necessary to fully understand the role of pexophagy in plant life.

Funding Statement

This work was supported by a Grant-in-Aid for Scientific Research on Innovative Areas (no. 22120007 to M.N.) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT); a Grants-in-Aid for Scientific Research (no. 17K07467 to Y.H. and K.O., no. 20370024 to M.N., and nos. 26440157 and 20570045 to S.M.) from Japan Society for the Promotion of Science (JSPS); a SONATA Grant (UMO-2019/35/D/NZ3/04500 to S.G.-Y.) from National Science Centre Poland; a TEAM Grant (TEAM/2017-4/41 to K.Y.) from the Foundation for Polish Science; and the Wyeth Foundation to M.N..

Disclosure statement

No potential conflict of interest was reported by the author(s).