NPC-phagy: selective autophagy of the nuclear pore complexes

ABSTRACT

Selective autophagy is critical for the regulation of cellular homeostasis in organisms from yeast to humans. This process is a specific degradation pathway for a wide variety of substrates including unwanted cytosolic components, such as protein aggregates, damaged and/or superfluous organelles, and pathogens. However, it has been less clear as to whether a protein complex or substructure of an organelle can be targeted for removal by selective autophagy. One example of such a substrate is the nuclear pore complex (NPC), a large macromolecular assembly that is present throughout the nuclear envelope. Here, we highlight two recent studies that demonstrate for the first time that NPCs are targeted for vacuolar degradation through selective autophagy.

Abbreviations

AIM: Atg8-interacting motif; NE: nuclear envelope; NPC: nuclear pore complex; Nup: nucleoporin; PMN/micronucleophagy: piecemeal microautophagy of the nucleus

NPCs are large multiprotein channels that penetrate the nuclear envelope (NE) and serve as gateways for macromolecular traffic between the nucleus and cytoplasm. These structures are fundamental components of all eukaryotic cells and have highly conserved functions and architecture. The NPC has an eight-fold symmetry and contains three major substructures: the central core, the cytoplasmic pore filaments and the nuclear basket, which are composed of approximately 30 different proteins termed nucleoporins (Nups) [1]. Tremendous progress has been made in understanding the structure and functions of the NPCs; however, little is known about how NPCs are turned over, which could be vital for their quantity, quality and function.

One possibility is that NPCs are degraded together with part of the nucleus, as they are embedded in the NE. The autophagic digestion and recycling of the nucleus/nuclear components, termed nucleophagy, has been found to occur in at least three manners in the yeast Saccharomyces cerevisiae: receptor Atg39-dependent nucleophagy, Vac8-Nvj1 nuclear-vacuole junction-dependent piecemeal microautophagy of the nucleus (PMN), and late nucleophagy [2]. Because the nucleus is essential for most cells, nucleophagy mostly is just partial degradation; for example, PMN only pinches off part of the NE and nucleus while excluding NPCs and spindle pole bodies [3]. Furthermore, different parts of NPCs have remarkably different turnover rates. These observations raised a question as to whether there is a specific pathway that regulates the turnover of NPCs and Nups. Two teams have answered this question with their recent published studies about selective autophagic degradation of NPCs, termed NPC-phagy (Table 1) [4,5].

Table 1.

Comparison between the two studies.

Common results
NPC/Nups are degraded by autophagy in conditions of TORC1 inactivation (e.g., nitrogen starvation and rapamycin treatment). Autophagic degradation of NPC/Nups is selective, as the degradation involves the Atg11 scaffold. Nup159 interacts with Atg8 via its AIM motif (1078Y-D-K-L1081).
Different results
	Lee et al., [4]	Tomioka et al., [5]
Other pathways for turnover	Proteasomal degradation	Atg39-dependent nucleophagy
Receptor	Nup159	Unknown, but not Nup159
Nucleoporinophagy	N/A	Nup159 is degraded through its AIM-dependent interaction with Atg8
NPC clustering	Enhanced NPC degradation	N/A
Deletion of NUP116 (destablizing the NPC)	Almost abolished the autophagic degradation of the NPC	Enhanced autophagic clearance of the NPC

To study whether and how NPCs are degraded, both teams investigated the processing of GFP-tagged Nups, and found that Nups are degraded after nitrogen starvation or TORC1 inhibition through autophagy, as deletion of genes encoding components of the core autophagy machinery severely impairs their turnover. This process is also selective because it requires an Atg8-interacting motif (AIM)-binding pocket and the scaffold protein Atg11, an essential protein of selective autophagy. One key aspect in understanding NPC degradation is to differentiate it from the turnover of unassembled individual Nups or assembly intermediates. The following three lines of evidence indicate that the whole NPC can be targeted by a phagophore: 1. Nups from different substructures are degraded in a similar manner (Atg8 and Atg11 dependent); 2. Through mass spectrometry analysis of the Atg8 interactome and further co-immunoprecipitation assays, the authors found that Atg8 can be used to affinity isolate almost all of the Nups; 3. Lee and colleagues show that in the vacuoles of vacuolar protease-deficient cells, Nup192, a core scaffold component of the inner ring, colocalizes with Nups from different substructures such as cytoplasmic filament-associated Nup159, and the transmembrane protein Ndc1. However, this does not rule out the possibility of selective degradation of Nup aggregates (nucleoporinophagy).

In the journey of completing the whole picture of NPC-phagy, both teams identified Nup159 as a potential receptor: Nup159 is a putative AIM-containing protein and it belongs to the cytoplasmic filament of NPCs, which makes it accessible to Atg8. The interaction was further found to be dependent on the Nup159 AIM motif ₁₀₇₈Y-D-K-L₁₀₈₁, the Atg8 AIM binding pocket and TORC1 inactivation. However, there are significant contradictory results and conclusions as to whether Nup159 is the cargo receptor for NPC-phagy based on the studies from the two different teams. Tomioka and colleagues found that although the AIM motif in Nup159 is important for its own degradation, it does not affect the degradation of other Nups. Because they also observed cytoplasmic puncta of Nup159-GFP, they concluded that the Atg8 interaction-dependent degradation of Nup159 is a type of nucleoporinophagy, the selective autophagy of Nups but not the whole NPC, and that Nup159 serves as receptor for itself. In contrast, Lee et al. showed that the AIM motif of Nup159 is important for the autophagic clearance of other Nups, including Nup192 and Nup133, both of which are from different substructures. They also showed that the Nup159-Atg8 interaction site colocalizes with the NE; thus, it is unlikely to correspond to unassembled free Nups in the cytosol. Taken together, the paper by Lee et al. drew the conclusion that Nup159 is the receptor mediating autophagic degradation of the NPC. It is possible that other AIM-containing Nups also contribute to the substrate recognition of NPCs.

There are also some other discrepancies between the two studies (listed in Table 1). For example, Lee et al. found that Rpn10-dependent proteasomal degradation accounts for part of the NPC turnover, whereas Tomioka et al. discovered that Atg39-dependent nucleophagy also plays a role in the degradation, but further experiments are needed to determine how these pathways are coordinated. In the end, Lee and colleagues showed that aberrant NPC clustering caused by deletion of the genes encoding certain scaffold Nups will enhance NPC-phagy, whereas a mutation that disassociates cytoplasmic filaments from NPCs abolishes NPC-phagy. This is intriguing, as there need to be pathways for NPC anomaly surveillance and signal transduction to induce NPC-phagy when appropriate.

Despite the discrepancies, the two studies showed for the first time that NPCs are degraded through receptor-mediated selective autophagy. They also laid the foundation for many follow-up questions. For example, is this mechanism conserved in other eukaryotic cells, because NPCs are highly conserved from yeast to mammals? How is NPC-phagy initiated following TORC1 inactivation or other conditions for the purpose of quality control? How is the NPC disassembled from the NE during NPC-phagy, or how do NE-derived vesicles contain NPCs for degradation? An even broader question is, because the NPC is a protein complex embedded in the nuclear envelope and can be selectively degraded, are there any other protein complexes selectively targeted by autophagy even if they are part of other subcellular organelles?

Funding Statement

This work was supported by the National Institute of General Medical Sciences [GM131919].

Disclosure statement

No potential conflict of interest was reported by the authors.