No retraction, correction or concern found when last checked on Oct 17, 2026.
Crosstalk Between Mammalian Autophagy and the Ubiquitin-Proteasome System
Nur Mehpare Kocaturk, Devrim Gozuacik
Frontiers in Cell and Developmental Biology · 2018 · Review
Abstract
Crosstalk Between Mammalian Autophagy and the Ubiquitin-Proteasome System Nur Mehpare Kocaturk1 and Devrim Gozuacik1,2,3* 1Sabanci University, Faculty of Engineering and Natural Sciences, Molecular Biology, Genetics and Bioengineering Program, Istanbul, 34956, Turkey. 2Sabanci University, Center of Excellence for Functional Surfaces and Interfaces for Nano Diagnostics (EFSUN), Istanbul, 34956, Turkey. 3Sabanci University Nanotechnology and Application Center (SUNUM), Istanbul, 34956, Turkey. * Correspondence: Devrim Gozuacik dgozuacik@sabanciuniv.edu Autophagy and the ubiquitin–proteasome system (UPS) are the two major intracellular quality control and recycling mechanisms that are responsible for cellular homeostasis in eukaryotes. Ubiquitylation is utilized as a degradation signal by both systems, yet, different mechanisms are in play. The UPS is responsible for the degradation of short-lived proteins and soluble misfolded proteins whereas autophagy eliminates long-lived proteins, insoluble protein aggregates and even whole organelles (e.g., mitochondria, peroxisomes) and intracellular parasites (e.g., bacteria). Both the UPS and selective autophagy recognize their targets through their ubiquitin tags. In addition to an indirect connection between the two systems through ubiquitylated proteins, recent data indicate the presence of connections and reciprocal regulation mechanisms between these degradation pathways. In this review, we summarize these direct and indirect interactions and crosstalks between autophagy and the UPS, and their implications for cellular stress responses and homeostasis.
Full text
From PubMed Central, U.S. National Library of Medicine, as current on Oct 17, 2026. Paragraph numbers are ours, so an answer can point at one.
Introduction
The ubiquitin–proteasome system (UPS) and macroautophagy (hereafter referred as autophagy) are two major intracellular protein degradation pathways. Degradation of short-lived proteins through the UPS is initiated by sequential addition of ubiquitin chains to target proteins. Polyubiquitylated proteins are then recognized by the subunits of multicatalytic protease complexes called proteasomes.
Proteasomes are extremely efficient organelles that degrade short-lived proteins and soluble unfolded/misfolded proteins and polypeptides. On the other hand, long-lived proteins, insoluble protein aggregates (usually originating from misfolded proteins, disease-related mutant proteins) and dysfunctional organelles, such as degenerated mitochondria and peroxisomes, are eliminated by the autophagy-lysosome system. Autophagy is characterized by the formation of double-membrane structures termed as autophagosomes, which later on fuse with lysosomes, forming autolysosomes degrading autophagosome contents.
The UPS and autophagy are interconnected, and inhibition of one system was shown to affect the other. There is accumulating evidence in the literature about connections between the UPS and autophagy. In this review article, we will first briefly summarize the two systems, and then discuss in detail various examples of coordination and crosstalk between them. For more detailed discussion on individual systems, the readers are referred to recently published excellent review articles. This review article mainly focuses on the mammalian system and advances in this field. For crosstalk in other systems, such as plants, readers should check other recent and relevant reviews [for example see,].
The Ubiquitin-Proteasome System
Ubiquitylation-dependent degradation is involved in the regulation of several cellular processes, including protein quality control, transcription, cell cycle progression, DNA repair, cell stress response and apoptosis. For example during cell cycle regulation, timely progression of each phase of the cycle rely on sequential transcription and degradation of cell cycle proteins, such as cyclins. During apoptosis, ubiquitylation leading the degradation of survivin depends on ubiquitin ligase XIAP.
Ubiquitylation involves the addition of the small protein ubiquitin to specific lysine residues on the target proteins. Covalent attachment of ubiquitin to protein targets occurs through a three-step mechanism involving E1 (ubiquitin-activating), E2 (ubiquitin-conjugating) and E3 (ubiquitin ligase) enzymes as summarized in Figure 1. At least seven lysine (K) residues in the ubiquitin protein are involved in the polyubiquitin chain formation (K6, K11, K27, K29, K33, K48, or K63). Initially, K48-linked ubiquitin chain formation was introduced as the degradation signal for proteasomal degradation. In contrast, K11 or K63 chains or single ubiquitin moieties (monoubiquitylation) were initially connected to non-proteolytic functions. However, recent reports indicate that K63-linked ubiquitin chains as well as various other chains prime substrates for autophagic elimination.
The 26S proteasome is an ATP-dependent protease complex, consisting of a core complex, the 20S proteasome and a regulatory complex, the 19S proteasome cap. The 20S proteasome forms a barrel-shape structure with two end rings formed by α subunits regulating the entry of unfolded proteins, and two middle rings are composed of β subunits harboring proteolytic activity. Substrates must be unfolded and then guided by α subunits prior to catalytic cleavage. At the end, polypeptides are chopped into 3–25 amino acid long fragments, and further cleavage to single amino acids is carried out by peptidases (Figure 1). By this way, recycling of proteins result in the generation of amino acids that are ultimately reused by cells in the synthesis of new proteins.
The 26S proteasome contains an additional 19S cap structure that further regulates the internalization of ubiquitylated substrates. The central part of the 19S cap consists of six AAA ATPases (Rpt1–Rpt6) forming the Rpt ring that is responsible for substrate binding and unfolding as well as substrate transfer through the channel. Non-ATPase proteins such as Rpn10 and Rpn13 in the 19S cap, possess ubiquitin-binding domains and therefore function as receptors for ubiquitin-labeled substrates.
Recent studies showed that ubiquitylation is a reversible phenomenon. Deubiquitinating enzymes (DUBs) are proteases that remove ubiquitin or ubiquitin-like molecules from substrates and disassemble polyubiquitin chains. DUBs regulate UPS-mediated degradation in different cellular contexts. Moreover, they play an important role in the control of available free ubiquitin pool in cells, allowing recycling and reuse of ubiquitin. Some DUBs are also responsible for processing newly synthesized ubiquitin precursors.
Autophagy
There are three major types of autophagy: Macroautophagy, microautophagy and chaperon-mediated autophagy (CMA). In this review, we chose to focus on macroautophagy (herein autophagy). CMA and microautophagy were discussed in elsewhere.
Autophagy is characterized by the engulfment of cargo molecules by double-membrane vesicles, called autophagosomes. Following closure, autophagosomes are transported by the microtubule system, leading to their fusion with late endosomes and lysosomes, forming autolysosomes. In this new compartment, sequestered cargos are degraded by the action of lysosomal hydrolases. Building blocks that are generated by hydrolysis of macromolecules (e.g., amino acids from protein degradation) are then transferred back to cytosol for reuse (Figure 2). Active at a basal level, autophagy is upregulated following a number of stimuli and stress conditions. Amino acid deprivation, serum starvation and growth factor deprivation, hypoxia, exposure to various chemicals and toxins might be counted among stress conditions activating autophagy.
Most autophagy inducing signals converge at the level of mTOR protein complexes (mTORC1 and mTORC2) that coordinate anabolic and catabolic processes (Figure 2). Cellular energy sensor AMPK directly regulates mTOR and therefore contributes to the regulation of the autophagic activity. Moreover, the ERK/RSK pathway, PI3K/AKT pathway, amino acid sensor RAG system as well as hypoxia are among autophagy-related pathways converging at the level of mTOR. Under normal conditions, mTORC1 limits the autophagic activity through inactivation of the ULK1/2 autophagy complex. mTORC1-dependent phosphorylation of ULK1 and Atg13 result in the inactivation of ULK1/2 complex and down regulation of autophagy. Under stress, mTORC1 is inhibited and ULK1/2 complex dephosphorylated. ULK1/2 then phosphorylates itself, Atg13 and FIP200 and activate autophagy.
A class III phosphatidylinositol 3-kinase (PI3K) complex, including the lipid kinase VPS34 and the regulatory protein Beclin1, controls the membrane nucleation stage and initial phagophore formation. Phosphatidylinositol 3-phosphate (PtdIns3P) that is generated by PI3K activity serves as a landing pad for autophagy-related proteins containing PI3P-binding domains (e.g., FYVE-domains). Among them WIPI1-4 and DFCP1 were involved in the formation of a membrane structure called omegasome or cradle, a structure that creates a platform for the elongation of autophagosome precursor isolation membranes.
Elongation of the isolation membrane depends on two ubiquitin-like conjugation systems. In the first system, autophagy-related gene 12 (ATG12) is covalently conjugated to the ATG5 protein through the action of ATG7 (E1-like) and ATG10 (E2-like) proteins. Then, recruitment of the ATG16L1 protein to ATG12-5 dimer results in the formation of a larger complex. Then forming ATG12-5-16L1 oligomers serve as E3 ligases that conjugate lipid molecules (such as phosphatidylethanolamine) to ATG8 orthologs MAP1LC3, GATE16, GABARAP. Lipid-conjugated ATG8 proteins are required for the elongation, expansion and closure of autophagosome membranes.
In order to acquire lytic capacity, autophagosomes fuse with late endosomes or lysosomes. In mammalian cells, fusion requires lysosomal integral membrane protein LAMP-2, several SNARE proteins (e.g., STX17 and WAMP8) and RAB proteins (e.g., RAB5 and RAB7). Following fusion of the outer membrane of autophagosomes, materials contained in the inner membrane are degraded by the action of lysosomal hydrolases. Building blocks (e.g., amino acids, fatty acids etc.) are then transported back to cytosol for reuse in the metabolic processes of the cells.
Autophagic vesicles engulf targets such as portions of cytoplasm and various cytoplasmic components in a non-selective manner. On the other hand, several selective forms of autophagy have been described. In most cases, ubiquitylation of the cargo constitutes a key step in the chain of events leading to its autophagic removal. Selective targets include mitochondria, peroxisomes, lysosomes, endoplasmic reticulum (ER), ribosomes, cytoplasmic protein aggregates, pathogenic intracellular invaders and even certain free proteins and RNAs were shown to be targets of selective autophagy. By this way, cells control number of the organelles, eliminate dysfunctional components and get rid of potentially harmful aggregates and invaders.
Selectivity is ensured by target-specific autophagy receptors that form a bridge between the ubiquitylated cargo and LC3 component of autophagic membranes. Selective autophagy relies on the recognition and binding capacity of autophagy receptors to various types of cargo, including mitochondria (OPTN, NDP52, Tax1BP1, NIX, FUNDC1), peroxisomes (NBR1), lysosomes (galectin-3), ER (FAM134B, SEC62, RTN3, and CCPG1) and intracellular ubiquitylated aggregates (p62, NBR1, OPTN, TOLLIP receptors), bacterial invaders (p62, OPTN, NDP52 receptors). LC3-interacting region (LIR) is the common motif which allows autophagy receptors to bind lipidated LC3. On the other hand, ubiquitin-associated domain (UBA domain) on autophagy receptors are responsible for the recognition of ubiquitin decorated cargos. Cargos that are wrapped and packed in autophagosomes are then ready for delivery and degradation in lysosomes.
The UPS-Autophagy Connection
The UPS and autophagy are the two major and evolutionarily conserved degradation and recycling systems in eukaryotes. Although their activities are not interdependent, recent studies show that connections and crosstalks exist between the two systems. Mitophagy constitutes a prominent example connecting these two degradative systems, yet several other examples exist. In this section, we will summarize biological events involving autophagy and the UPS, and discuss molecular details of the crosstalk mechanisms.
Compensation Between the Two Degradative Pathways
Initial observations about functional connections between the UPS and autophagy systems revealed that inhibition of one led to a compensatory upregulation of the other system. In order to maintain homeostasis, cellular materials that accumulate following inhibition of one degradative system needs to be cleared, at least in part, by the other system (Figure 3). Here, we will give examples of scenarios where these compensation mechanisms are operational.
Inhibition of the UPS using various compounds (e.g., MG132, bortezomib, lactacystine etc.) or by genetic approaches resulted in the upregulation of the autophagic activity in cells (Figure 3). For example, inhibition of proteasomal activity by the proteasome inhibitor and chemotherapy agent bortezomib led to an increase in the expression of autophagy genes ATG5 and ATG7, and induced autophagy. In fact, autophagy gene upregulation depended on an ER stress-dependent pathway that involved eukaryotic translation initiation factor-2 alpha (eIF2α) phosphorylation. In another study, proteasome inhibition was associated with an increase in p62 and GABARAPL1 levels by Nrf1-dependent and -independent pathways prior to autophagy activation. In other contexts, MG132-mediated proteasome inhibition resulted in a decrease in cell proliferation, cell cycle arrest at G2/M phase and stimulation of autophagy through upregulation of Beclin1 and LC3.
Autophagy induction following proteasome inhibition correlated with AMPK activation as well. A number of studies provided evidence that proteasomal inhibition is sensed by both AMPK and mTORC1, two major regulators of autophagy. For instance, in macrophages, epitelial and endothelial cells, proteasome inhibition using chemicals resulted in the activation of AMPK. In some other cancer cell types, CaMKKβ and glycogen synthase kinase-3β (GSK-3β) were identified as upstream regulators of AMPK activation, proteasome inhibition was linked to a decrease in GSK-3β activity and to the activation of AMPK and autophagy. On the other hand, Torin-1- or rapamycin-mediated inhibition of mTOR stimulated long-lived protein degradation through activation of both UPS and autophagy. In retinal pigment epithelial cells, inhibition of proteasome by lactacystin and epoxomicin was shown to block the AKT-mTOR pathway and induce autophagy. SiRNA-mediated knockdown of Psmb7 gene coding for the proteasome β2 subunit, resulted in enhanced autophagic activity, and it was linked the mTOR activation status of cultured cardiomyocytes.
Similarly, impairment of autophagy correlated with the activation of the UPS. In colon cancer cells, chemical inhibition of autophagy and small RNA mediated knock down of ATG genes resulted in the upregulation of proteasomal subunit levels, including the catalytic proteasome β5 subunit, PSMB5 and led to increased UPS activity. In another study, 3-MA-mediated autophagy inhibition in cultured neonatal rat ventricular myocytes (NRVMs) increased chymotrypsin-like activity of proteasomes.
Since proteasomes were identified as autophagic degradation targets (proteaphagy), enhanced proteasome peptidase activity following autophagy inhibition might be associated with the accumulation of proteasomes. Yet in several cases, autophagy inhibition correlated with the accumulation of ubiquitylated proteins. For instance in independent studies with ATG5 or ATG7 knockout mice, accumulation of ubiquitylated conjugates were observed, especially in the brain and the liver of the animals. Similar results were observed in other animal models such as Drosophila. In line with these data, inhibition of autophagy through siRNA-mediated knockdown of ATG7 and ATG12 in HeLa cells resulted in the impairment of UPS, accumulation of ubiquitylated proteins as well as other important UPS substrates, including p53 and β-catenine. In above-cited papers, autophagy impairment followed by the autophagy receptor p62 accumulation in cells, and played a key role in the observed UPS defects.
Ubiquitylation was proposed to be a common component that directs substrates to the proper degradation system and even contribute to the UPS-autophagy crosstalk. According to this view, proteins that are predominantly linked to K48-based ubiquitin chains are generally directed for degradation through UPS. Conversely, aggregates that are linked to K63-based ubiquitin chains are directed for autophagic degradation. P62 binding capacity was introduced as the critical step in the choice between the UPS and autophagy. Although, p62 is able to attach both K48- and K63-linked ubiquitin chains through its UBA domain, binding affinity of the protein for K63-linked chains seems to be higher. Due to this dual ubiquitin binding ability, p62 might show UPS inhibitory effects in some contexts. A competition between p62 and p97/VCP (a ubiquitin binding ER-associated degradation protein) determined the fate of ubiquitylated proteins in cells. Over expression of p97/VCP protein prevented binding of p62 to ubiquitylated substrates, and directed them for degradation by the UPS. On the other hand, accumulation of p62 following autophagy inhibition led to the sequestration of proteins that were otherwise p97/VCP targets.
In summary, in the case of a defect in one of the two degradation systems, the other system is upregulated in order to eliminate ubiquitylated protein substrates. Yet, compensation does not always work and its success largely depends on cell types, cellular and environmental conditions and target protein load.
Interplay Between the UPS-Autophagy in the Selective Clearance of Cytosolic Proteins
Function of proteins depend on their proper folding and 3D structures. Various insults, including heat shock, organellar stress, oxidative stress etc., might lead to the accumulation of unfolded or misfolded proteins. Moreover several disease-related mutations were associated with folding problems. Failure to refold result in dysfunctional or malfunctional, hence toxic protein accumulations, activation of stress and even cell death pathways. In order to control toxic protein accumulations, an active process of protein aggregate formation comes into play. Additionally some proteins, including mutant proteins are already prone to form aggregates. Selective clearance of most cytosolic proteins require ubiquitylation. Depending on their solubility, ubiquitylated proteins and protein aggregates are then cleared by the UPS or autophagy.
Soluble fractions of proteins with a folding problem are recognized by the chaperone machinery and directed to the UPS for degradation. Hsp70 and Hsp90 chaperone interactor CHIP was identified as one of the E3 ligases that are responsible for K48-linked ubiquitin chain addition to unfolded/misfolded proteins. BAG family proteins, especially BAG1, interact with the Hsp70 complex and induce proteasomal degradation of client proteins.
On the other hand, clearance of insoluble aggregate-prone proteins require formation of aggresomes. Ubiquitylation by a number of different E3 ligases, including CHIP, Parkin, HRD1 and TRIM50 prime aggregate-prone proteins. HDAC6 is another protein that plays a key role in the process of aggresome formation. HDAC6 was shown to provide the link between K63-based ubiquitylated aggregates and microtubule motor protein dynein. Then, dynein-mediated mechanism direct the aggregates toward microtubule organizing centers (MTOCs), resulting in their piling of as aggresomes (Figure 4). Following aggresome formation, direct interaction of adaptor proteins p62 and NBR1 with ubiquitylated aggregates result in their delivery to autophagosomes. Another autophagy-related protein, ALFY, was also identified as a player in the selective autophagy and degradation of aggresomes.
An alternative pathway for aggresome formation require Hsp70 partner proteins BAG3 and CHIP. Similar to HDAC6, BAG3 binds to dynein, and this directs Hsp70 substrates to aggresomes. However, BAG3-dependent aggresome formation was not dependent on the ubiquitylation of substrates as in the case of HDAC6, and CHIP E3 ligase activity was dispensible. Yet, E3 ligases such as CHIP were required for BAG3-dependent aggresome clearance by autophagy.
Proteolytic Degradation of the UPS or Autophagy Components as a Mutual Control Mechanism
Until so far, we focused on the UPS and autophagy as complementary but independent mechanisms. However, there are cases where components of one system were reported to be a proteolytic target of the other system. For example, a number of autophagy proteins were regulated through degradation by the UPS. On the other hand, even the whole proteasomes were shown be selective targets of autophagic degradation. Here, we will give examples of how mutual regulation through proteolysis contributes to the crosstalk and the interplay between the two systems.
Control of the UPS by the Autophagic Activity
Early studies indicated that proteasomes could be degraded in lysosomes. Later on, plant studies revealed that lysosomal degradation of 26S proteasomes occurred by a specific form of selective autophagy, proteaphagy. RPN10 protein was introduced as an ATG8 interacting plant proteaphagy receptor. Unlike the plant protein, yeast and mammalian RPN10 failed to interact with ATG8/LC3. Instead, Cue5 protein in the yeast and its human ortholog TOLLIP, were introduced as selective receptors regulating proteasome clearance by autophagy. Moreover, p62 was also described as another proteaphagy receptor. For example, in mammals, amino acid starvation significantly upregulated ubiquitylation of 19S proteasome cap components RPN1, RPN10, RPN13, and led to their p62-mediated recruitment to autophagosomes (Figure 5). Interestingly during carbon or nitrogen starvation, plant and yeast proteasomes were shown to localize in proteasomal storage granules (PSGs), protecting them from autophagic degradation during stress. Whether similar mechanisms exist in the mammals is currently an open question. These observations underline the importance of selective degradation of proteasome by autophagy in the control of proteasome numbers as well as overall UPS and lytic activity in cells.
Control of Autophagy Components by the UPS
Modulation of the half-life of some proteins in the autophagy pathway by the UPS serves as a means to control cellular autophagic activity. For instance, LC3 protein was shown to be processed in a stepwise manner by the 20S proteasome, a process that was inhibited by p62 binding. On the other hand, E3 ligase NEDD4-mediated K11-linked ubiquitylation of Beclin1 prevented its binding to the lipid kinase VPS34, and led to its degradation. Another E3 ligase, RNF216 ubiquitylated Beclin1 adding K48-linked ubiquitin chains on the protein. Beclin1 ubiquitylation resulted in autophagy blockage in both cases. Conversely, reversal of Beclin1 ubiquitylation by the DUB protein USP19 stabilized the protein under starvation conditions and promoted autophagy. USP10 and USP13 as well as USP9X were characterized as other DUBs that regulated autophagy through control of Beclin1 stability.
Beclin1 is not the only autophagy protein that is targeted by the UPS in a controlled manner. G-protein-coupled receptor (GPCR) ligands and agonists were reported to regulate cellular Atg14L levels, and therefore autophagy, through ZBTB16-mediated ubiquitylation of the protein. Serum starvation increased GSK3β-mediated phosphorylation of ZBTB16, leading to its degradation. Under these conditions, stabilization of Atg14L restored of autophagy. AMBRA1 is another UPS-controlled autophagy protein. Cullin-4 was identified as an E3 ligase that was responsible for the ubiquitylation of AMBRA1, dooming it for degradation under nutrient-rich conditions where autophagy should be inhibited. The PI3K complex subunit p85b is another example. Ubiquitylation of this autophagy signaling component by the E3 ligase SKP1 led to a decrease in its cellular levels and stimulated autophagic activity.
Ubiquitylation of some autophagy proteins did not result in their immediate proteasomal degradation, yet the post-translational modification provided an extra layer of control for the autophagy pathway. For instance, autophagy receptor OPTN was ubiquitylated as a target of the E3 ligase HACE1, and K48-linked ubiquitylation regulated the interaction of the protein with p62. TRAF6, a central E3 ligase of the NF-κB pathway, participated controlled ULK1 activity through K63-linked ubiquitylation. Under nutrient-rich conditions, mTOR phosphorylated AMBRA1 leading to its inactivation. When nutrients were limiting, mTOR inhibition resulted in AMBRA1 dephosphorylation and increased the interaction of the protein with TRAF6. This event facilitated ULK1 ubiquitylation by TRAF6. Ubiquitylation of ULK1 resulted in the stabilization of the protein, controlled its dimerization and regulated its kinase activity. Another ubiquitin-dependent regulation mechanism involved AMBRA1-Cullin-5 interaction in the regulation of mTOR complex component DEPTOR. Above-mentioned AMBRA1-Cullin-4 complex dissociated under autophagy-inducing conditions, allowing AMBRA1 to bind another E3 ligase, Cullin-5. This newly formed complex was shown to stabilize DEPTOR and induce mTOR inactivation, providing a negative feed-back loop in the control of autophagy.
In another study, TLR4 signaling triggered autophagy through Beclin1 ubiquitylation and stabilization. TLR4-associated TRAF6 protein was identified as the E3 ligase responsible for K63-linked ubiquitylation of Beclin1 at its BH3 domain. This modification blocked inhibitory BCL-2 binding to the protein, and free Beclin1 could activate autophagy. On the other hand, the deubiquitinating enzyme A20 reversed TRAF6-mediated ubiquitylation of Beclin1, resulting in autophagy inhibition. Another K63-linked ubiquitylation event on Beclin1 was promoted by AMBRA1 protein. In the same context, the WASH protein interacted with Beclin1, blocked AMBRA1-mediated Beclin1 ubiquitylation, and suppressed autophagy.
LC3 and p62 were also subjected to regulatory ubiquitylation. NEDD4 was identified as the E3 ligase in these reactions. NEDD4 was reported to interact with LC3 and p62, and LC3 binding to NEDD4 stimulated its ubiquitin ligase activity on the p62 protein. Moreover, NEDD4 deficient cells exhibited aberrant p62 containing inclusions, indicating the defect in aggresome clearance. Hence, NEDD4 is important for the regulation of p62 function and autophagy.
Xenophagy: Removal of Intracellular Invaders
Another essential function of autophagy is the clearance of intracellular pathogens. This special form of autophagy, called xenophagy, is a result of a cooperation between the ubiquitylation machinery and the autophagy pathway. Pathogens such as Streptococcus pyogenes, Mycobacterium tuberculosis, Listeria monocytogenes, and Shigella flexneri were identified as autophagy targets. As a form of selective autophagy, xenophagy involves cargo labeling with ubiquitin, followed by the recognition by autophagy receptors (Figure 6). K48- and K63-linked and linear M1-linked ubiquitin chains were shown to mediate recognition of different pathogens by the xenophagy machinery.
Ubiquitylation frequently occurs on various cell penetrating parasites as well as on disrupted endosomes, providing an “eat me” signal for xenophagy. For example, Salmonella enterica serovar Typhimurium was heavily ubiquitylated in mammalian cells, and activation of xenophagy restricted intracellular bacteria numbers. Recent studies showed that, bacterial outer membrane-associated and integral membrane proteins were targets of ubiquitylation. A number of E3 ligases were involved in xenophagy, including Parkin, RNF166, ARIH1, HOIP, and LRSAM1.
For example, both K48- and K63-linked ubiquitylation were observed on Mycobacterium, and Parkin was identified as the E3 ligase catalyzing the K63-linked ubiquitylation. Moreover endosome-free areas on the intracellular Salmonella Typhimurium contained a directly attached ubiquitin coat, and addition of linear M1-linked ubiquitin chains by the E3 ligase HOIP of the LUBAC on these ubiquitins contributed to the autophagy of the intracellular parasite. Xenophagy receptors that were described to date include p62, OPTN, NDP52, and NBR1. These receptors were reported to bind pathogen- and/or endosome-associated ubiquitin, and directing the selective targets to autophagic membranes.
The interplay between ubiquitylation and autophagy achieves the important task of keeping host cells pathogen-free and providing an intracellular innate immune defense mechanism against invaders. In some reports, ubiquitylated bacteria were found to be surrounded by proteasomes as well and proteasomal activity might also be required for efficient killing of intracellular parasites. Whether in the elimination of invading organisms, the crosstalk between the UPS and autophagy systems goes beyond ubiquitylation, needs further consideration. As discussed below, cellular mechanisms controlling commensal-turned ancient intracellular microorganisms, namely mitochondria, indeed rely on the function of both the UPS and autophagy.
Mitophagy: Mitochondrial Turnover
Mitochondria are vital organelles that form an intracellular dynamic network in the cytosol of eukaryotic cells. Through fusion and fission, they are constantly made and destroyed. Under steady state conditions, mitochondria might be eliminated by basal in a non-selective manner. On the other hand, elimination of damaged, dysfunctional or superfluous mitochondria requires a selective form of autophagy called mitophagy. Programmed elimination of mitochondria during development and differentiation (e.g., reticulocyte maturation to erythrocyte, in oocytes after fertilization, during lens formation in the eye) also relies on mitophagy. Recent studies showed that mitophagy is a biological phenomenon that involves both the UPS and autophagy. In this section, we will discuss mechanisms of mitophagy, and analyze connections between the UPS and autophagy in this context.
PINK1/Parkin-Dependent Mitophagy
Depending on the E3 ligase that ubiquitylates proteins on mitochondria, mitophagy can be divided into two major forms: Parkin-dependent and Parkin-independent mitophagy. The E3 ligase Parkin was first characterized as the product of the gene PARK2, mutations of which were linked to early-onset of Parkinson’s Disease. Strikingly, Parkin recruitment to mitochondria was found to be necessary for mitophagy. Further studies showed that Parkin, together with another familiar Parkinson’s Disease-associated gene, PINK1 (PARK7), was responsible for priming mitochondria for autophagic degradation (Figure 7).
Under normal conditions, after being synthesized as precursor in the cytoplasm, PINK1 was imported to mitochondria by its N-terminal mitochondria targeting sequence (MTS). Then, PINK1 was post-translationally modified within mitochondria by resident proteases: MPP and PARL. Cleavage by PARL resulted in destabilization of the protein and its degradation by cytoplasmic proteasomes. Under mitochondrial stress however, PINK1 cleavage did not occur and the protein accumulated on the outer mitochondrial membrane (OMM). Recruitment of cytoplasmic E3 ligase Parkin onto mitochondria required stabilization and the kinase activity of the PINK1 protein. Parkin itself was a substrate of PINK1. Phosphorylation of Parkin by PINK1 resulted in a conformational change overcoming an autoinhibition, and stimulated its E3 ligase activity. Interestingly, PINK1 was shown to phosphorylate ubiquitin molecules on mitochondrial resident proteins as well. Ubiquitin phosphorylation correlated with an increase in the amount of mitochondria-localized Parkin, providing a feed-forward mechanism of Parkin recruitment.
Several proteins on the mitochondrial outer membrane were identified as Parkin ubiquitylation substrates. The list includes VDAC, TOM proteins, mitofusins etc. Following ubiquitylation some of these targets were shown to be degraded by the proteasome (e.g., mitofusins) and some were not (e.g., VDAC). Degradation of proteins related to mitochondrial integrity promoted fission events that facilitate engulfment of mitochondrial portions by autophagosomes, whereas proteins that are not degraded upon ubiquitylation rather contributed to mitochondrial rearrangements (e.g., aggregation).
The UPS activity was a prerequisite in the preparation of mitochondria for autophagy. Ubiquitylation of mitochondrial targets preceeded the recruitment of the autophagic machinery onto mitochondria. Selective autophagy receptors were shown to bind ubiquitin-labeled proteins on mitochondria and recruit ATG8/LC3 proteins for mitophagy. Serial knock out of putative autophagy receptors showed that NDP52, optineurin (OPTN) and TAX1BP1 were functional mitophagy receptors, and a triple knockout of these proteins completely blocked mitophagy. On the other hand, the autophagy receptor p62 was essential for clustering of damaged mitochondria in perinuclear region of the cells, but not for mitophagy.
Ubiquitin modifications on mitochondria might be reversed by the action of DUB proteins. Several DUBs were identified as positive or negative regulators of mitophagy. For example, deubiquitylation of mitochondrial targets by USP15, USP30, and USP35 prevented further progression of mitophagy in a number of cell lines and experimental models. DUB-mediated deubiquitylation of targets decreased Parkin recruitment onto mitochondria as well. USP8-mediated removal of K6-linked ubiquitin chains from Parkin itself affected recruitment of the protein onto mitochondria and therefore mitophagy.
Parkin-Independent Mitophagy
Expression of Parkin is restricted to a few cell types, including dopaminergic neurons. Consequently, Parkin-null animals showed prominent mitophagy defects only in selected brain regions. Therefore in other cell types and tissues, mitophagy has to proceed in a Parkin-independent manner. Alternative E3 ligases were found to play a role in mitophagy in these contexts.
Mulan (MUL1) is an E3 ubiquitin ligase that resided on the OMM, and it was shown to play a role in Parkin-independent mitophagy in different model organisms, including Caenorhabditis elegans, Drosophila and mammals. Mulan stabilized DRP1, led to degradation of MFN2, and interacted with ATG8 family member protein GABARAP. Another E3 ligase that was associated with mitophagy was GP78. Over expression of GP78 induced MFN1 and 2 ubiquitylation and degradation, that was followed by mitochondrial fragmentation and mitophagy in cells lacking Parkin. Synphilin-1-dependent recruitment of the E3 ligase Siah1 to mitochondria resulted in mitochondrial protein ubiquitylation and mitophagy in a PINK1-dependent but Parkin-independent manner. Conversely, another OMM E3 ligase, MITOL (MARCH5), was reported to ubiquitylate FIS1, DRP1 and MFN2, yet inhibited hypoxia-induced and Parkin-independent mitophagy through ubiquitylation and degradation of FUNDC1. All these findings underline the fact that mitophagy might proceed in cells which do not express Parkin. Further studies are required to unravel the molecular mechanisms of Parkin-independent mitophagy in different tissues and cell types, and reveal the details of the crosstalk between the UPS and autophagy under these conditions.
A Special Type of Mitophagy During Reticulocyte Maturation
During differentiation, in order to increase their capacity to load hemoglobin-bound oxygen, reticulocytes lose their organelles, including mitochondria, and become mature red blood cells. During this process, a protein called NIX (also known as BNIP3L) is upregulated. NIX is a C-terminally anchored outer mitochondrial membrane (OMM) protein that contains a LC3-interacting region (LIR) at its cytoplasmic N-terminal part. Through its LIR domain, NIX interacted with LC3, enabling engulfment of mitochondria by autophagosomes in reticulocytes. Characterization of NIX-deficient mice showed that, NIX-deficient Erythrocytes failed to eliminate their mitochondria revealing a critical role for NIX in mitophagy (Figure 7). Although NIX-dependent mitophagy was predominantly studied in reticulocytes, NIX-dependent mitophagy might be important for other cell types as well [for example, see].
A role for the UPS in NIX/BNIP3L-dependent mitophagy was revealed. NIX/BNIP3L was discovered to be ubiquitylated through a PINK1/Parkin-dependent mechanism. Ubiquitylated NIX/BNIP3L colocalized with selective autophagy receptors, and the process was necessary for mitochondrial stress-induced mitophagy. Therefore, the role of NIX/BNIP3L seems to be more general than previously thought and beyond the developmental context, and stress-induced mitochondrial elimination by autophagy might also require NIX/BNIP3L in different cell and organism types.
Pexophagy: Autophagic Removal of Peroxisomes
Autophagy of peroxisomes, pexophagy, is a selective degradation process of peroxisomes during which the UPS and autophagy mechanisms work in collaboration. Peroxisomes are responsible of a number of cellular functions, including fatty acid oxidation, purine metabolism and phospholipid synthesis. Several peroxisomal enzymes are involved in redox regulation due to their dual functions in the generation and scavenging of reactive oxygen and nitrogen species. Therefore, peroxisome biogenesis and degradation must be tightly regulated in order to control peroxisome size, number and function. Moreover under stress conditions such as hypoxia, oxidative stress, starvation or conditions causing UPS defects, pexophagy is upregulated.
During pexophagy, a number of peroxisomal membrane proteins, including peroxins and PMP70 become ubiquitylated. PEX2-PEX10-PEX12 complex serves as an E3 ligase at least for two well studied peroxisome proteins, PEX5 and PMP70. Ubiquitylation of peroxisome proteins result in the recruitment of p62 and/or NBR1 autophagy receptors, priming these organelles for autophagic degradation. For example, PEX2 overexpression or amino acid starvation activated the ubiquitylation of PEX5, and another peroxisomal membrane protein, PMP70, and led to peroxisome degradation. Moreover in response to oxidative stress, ATM was recruited onto peroxisomes through physical interaction with PEX5 and promote its ubiquitylation. Inactivation of mTORC1 in a TSC2-dependent manner and stimulation of ULK1 phosphorylation by ATM, potentiated pexophagy. On the other hand, AAA ATPase complex (PEX1, PEX6, and PEX26) was shown to extract ubiquitylated PEX5 from peroxisomal membranes and regulate pexophagy (Figure 8). Both NBR1 and p62 were shown to be recruited onto peroxisomes during pexophagy. Yet, NBR1 was a major pexophagy receptor in a number of contexts, and p62 increased the efficiency of NBR1-dependent pexophagy through direct interaction with the latter. Altogether, these findings underline the importance of ubiquitylation for the selective degradation of peroxisomes by autophagy.
Autophagic Removal of Ribosomes and Stress Granules
In addition to major cellular organelles, autophagy was implicated in the clearance of ribosomes. Although ribosomes can be degraded in a non-specific manner during non-selective autophagy, a special form of selective autophagy is activated under various stress conditions, and the process is called ribosomal autophagy or ribophagy. On the other hand, mRNA protein complexes that are stalled during translation form stress granules, and their clearance requires both the UPS and autophagy.
Ribophagy was first described in the yeast during nutrient stress, and was shown to involve ubiquitylation of the 60S ribosome protein Rpl25 by the ubiquitin ligase Ltn1/Rkr1. In the mammalian system, in addition to mTOR inhibition, oxidative stress, induction of chromosomal mis-segregation, translation inhibition and stress granule formation were all shown to induce ribophagy. Ubiquitylation of ribosomes was observed under ER stress-inducing conditions. P97/VCP that binds to ubiquitylated proteins and that functions in the delivery of these substrates to proteasome was necessary for ribophagy both in yeast and mammalian cells. Yet, individual ribosomal proteins were indeed shown to be a target of the UPS. NUFIP1-ZNHIT3 proteins were identified as novel ribophagy receptors that directly connected ribosomes to LC3 and autophagy, yet whether ubiquitylation is a prerequisite for ribophagy needs to be clarified by future studies (Figure 9).
Stress granules are composed of actively accumulated non-translating mRNA ribonucleoprotein complexes. Proteins that accumulated in the stress granules, include stalled 40S ribosomal units and various translation initiation factors [e.g., eIF4E, eIF4G, eIF3, eIF2 and poly(A)-binding protein (PABP)] and regulators such as eIF2-α and GCN2. G3BP1 and TIA-1 are also among the proteins that contribute to stress granule formation. Moreover, an interplay between G3BP1 and Caprin1 proteins and the DUB protein USP10 was shown to regulate stress granule formation. HDAC6 protein was a component of stress granules as well.
Accumulating data indicate that both the UPS and autophagy play a role in stress granüle control and elimination, and the p97/VCP protein was a key component in these processes. For example, inhibition of autophagy or p97/VCP deficiency was linked to decreased stress granule removal. Co-factors of p97/VCP determined target selectivity of the protein. In this context, while the association of p97/VCP with the co-factor UFD1L led to the degradation of defective ribosomal products and dysfunctional 60S ribosomes by the UPS, HDAC6 containing p97/VCP and PLAA associated granules were made a target of ribophagy. Therefore depending on the co-factor of choice, p97/VCP has a decisive role in the choice of the degradative pathway through which ribonuclear substrates are eliminated.
Cross Talk Between UPS and Autophagy During Endoplasmic Reticulum Stress
Endoplasmic reticulum (ER) stress is one of the conditions under which both the UPS and autophagy pathways are being activated. Abnormalities in calcium homeostasis, oxidative stress and conditions leading to protein glycosylation or folding defects etc. may result in the accumulation of misfolded and/or unfolded proteins in the ER lumen, a condition known as ER stress. ER stress might be very destructive for cells, therefore ER-specific stress response pathways such as the unfolded protein response (UPR) and the ER-associated degradation (ERAD) pathways were evolved. Both pathways are directly or indirectly connected to the UPS and autophagy.
In mammalian cells, accumulation of unfolded proteins in the lumen of the ER result in the activation of stress responses. Following protein accumulation in the ER, the chaperone protein GRP78/BiP dissociates from the lumen-facing parts of the transmembrane proteins IRE1, ATF-6, and PERK and bind to unfolded proteins in order to assist their refolding. GRP78/BiP release triggers activation of these stress proteins. PERK activation leads to the phosphorylation of the α subunit of the translation initiation factor, eIF2α, which inhibits the assembly of the 80S ribosome and cap-dependent protein synthesis, while allowing cap-independent translation of the stress response genes such as ATF4. Activation of IRE1 and ATF6 promotes transcription of other stress response genes. IRE1-mediated processing generates a splice-form of the XBP1 mRNA, resulting in the production of a transcription factor that upregulates chaperones and other relevant genes. GRP78/BiP dissociation results in the transfer of ATF6 to Golgi where cleavage of the protein by S1P and S2P proteases creates an N-terminal ATF6 fragment possessing a transcriptional activity (Figure 10). Due to a decrease in the protein load in the ER and an increased folding capacity, the UPR facilitates recovery from stress. In case of failure, the UPR sensitizes cells to programmed death mechanisms.
Components of the UPR were subject to active regulation by the UPS. For example, SCF component E3 ligase βTrCP was shown to lead to the ubiquitylation ATF4 following its phosphorylation. On one other hand, persistent ER stress induced transcription of E3 ligase Siah1/2 following PERK-ATF4 and IRE1-XBP1 activation. On the other hand, by targeting prolyl hydroxylase PHD3, Siah1/2 was shown to regulate ATF4 hydroxylation and activity. CHOP stability was regulated by the UPS and p300 and cIAP were responsible for CHOP ubiquitylation and degradation counterbalancing its upregulation during ER stress. Another UPR component, IRE1 was identified as a ubiquitylation target of the E3 ligase CHIP during ER stress. Ubiquitylation IRE1 inhibited its phosphorylation, perturbed its interaction with TRAF2, and attenuating JNK signaling. Under stress conditions, translation of XIAP, an E3 ligase protein and an inhibitor of apoptosis was downregulated in a PERK-eIF2α-dependent manner. In the same context, ATF4 may promote ubiquitylation and degradation of XIAP, leading to sensitization of cells to ER stress-related cell death. Conversely, activation of PERK-eIF2α axis might also show opposing effects through induction of other IAP proteins, cIAP1 and cIAP2, and counter balance cell death induction signals.
Endoplasmic reticulum stress was shown to trigger autophagy, and ER-related stress response mechanisms were involved in the process. PERK-mediated phosphorylation of eIF2α and resulting ATF4 and CHOP activation, were associated with the transcription of genes such as ATG5, ATG12, Beclin1, ATG16L1, LC3, p62 and TSC2 activator, hence mTOR inhibitor REDD1. Moreover, CHOP downregulated BCL2 binding. TRB3, an AKT inhibitor protein, was also described as a target of CHOP. In addition, IRE1 activation resulted in the recruitment of ASK1 by the adaptor TRAF2 and the outcome was the activation of JNK and p38 kinases. BCL2 is one of the targets of JNK, its phosphorylation by the kinase resulted in destabilization the inhibitory BCL2-Beclin1 complex, stimulating autophagy. On the other hand, in its unspliced form, IRE1 splicing target XBP1, in its unspliced form was shown to target the autophagy activator FOXO1 for degradation by the UPS.
Endoplasmic reticulum is a major calcium store in cells, and calcium release to cytosol was observed during ER stress. In addition to problems with SERCA refill pumps and leakiness of membranes during stress, upregulation of ERO1-α by CHOP resulted in an IP3-mediated calcium release. Calcium binding protein calmodulin senses the cytosolic increase in the concentration of the ion, and bind to calmodulin-regulated kinases such as CaMKII and DAPK1, modulating their activity. Activated CaMKII was shown to stimulate autophagy through AMPK phosphorylation and activation. In addition, calmodulin-binding and PP2A-mediated dephosphorylation was necessary for the activation of the autophagy-related kinase DAPK1. DAPK1 could directly phosphorylate Beclin1 on the BH3-domain, resulting in the dissociation of Beclin1 from the BCL2-Beclin1 complex and allowing it to stimulate autophagy.
Proteins that accumulate in the ER are degraded by the ER-associated degradation (ERAD) system. ERAD mediates transport, extraction and ubiquitylation of proteins that cannot be salvaged and target them for degradation in proteasomes. In mammalian cells, ER membrane-resident complexes containing E3 ligases such as HRD1 and GP78, and other regulatory components such as EDEM1, SEL1L, ERManI, and HERP control the ERAD pathway. P97/VCP protein and its co-factors also play a role in the pathway. Unfolded/misfolded proteins are recognized in the lumen of the ER by chaperone proteins, including BiP/GRP78 and EDEM1, and are then subsequently targeted them to the ERAD pathway. During retrotranslocation of client proteins to cytosol, ubiquitylation is followed by a p97/VCP-assisted extraction. P97/VCP also assists in the delivery of proteins to proteasomes for degradation. DUB proteins, including YOD1, USP13, USP19, and Ataxin-3 were implicated in the control of client protein ubiquitylation and ERAD substrate modulation.
ER-associated degradation regulators and therefore ERAD might be controlled by the UPS and autophagy pathways. For example, E3 ligase Smurf1 was found to be downregulated during ER stress, resulting in the accumulation of its direct ubiquitylation target WFS, which is a stabilizer ER-related E3 ligase HRD1. Smurf1 was also involved in selective bacterial autophagy. On the other hand, while the ERAD complex component HERP protein was degraded by the UPS, EDEM1 and ERManI proteins were eliminated by the autophagy machinery. An ER-localized E3 ligase synoviolin protein was shown to ubiquitylate HERP protein and control its degradation by proteasome. Yet, other ERAD-related components, EDEM1 and Derlin2 as well as ubiquitylated EDEM1 proteins colocalized with cytoplasmic aggregates and autophagy receptors p62 and NBR1, they were degraded by selective autophagy. ERManI, a mannosidase that is responsible for priming ER-resident glycosylated proteins for degradation, was described as an accelerator of the ERAD pathway and clearance of clients by the UPS. But, following proteasome inhibition and subsequent ER stress, ERManI colocalized with LC3 and degraded in an autophagy-dependent manner.
All these findings point out to the presence of important junctions and coregulation nodes between the UPS and autophagy in the context of ER stress. Additionally, ERphagy, the autophagy of portions of the ER, was implicated in the recovery from ER stress and control of ER size, but this mechanism was so far described as a ubiquitin-independent process.
Transcriptional Mechanisms Connecting the UPS and Autophagy
Several transcription factors that are regulated by the UPS, including p53, NFκB, HIF1α, and FOXO, have been implicated in the control of autophagy. In general, these factors were shown to directly activate transcription of key autophagy genes under stress conditions. Some autophagy proteins such as LC3 are consumed in the lysosome following delivery, and during prolonged stress, cellular levels of these proteins are sustained by mechanisms, including transcription. On the other hand, regulation of the transcriptional activity NRF2 involves a special crosstalk between the two systems. In this section, we will summarize molecular details of transcription regulation by the UPS and autophagy.
P53, a guardian of the genome, is one of the well-known transcriptonal regulators that has a dual role in autophagy depending on its intracellular localization. In the absence of stress, cellular p53 levels are controlled by the E3 ligase HDM2/MDM2 and the UPS. Under stress conditions, p14/p19/ARF protein binds, sequesters and inactivates HDM2/MDM2, stabilizing p53. Accumulating p53 protein activates transcription of several stress- and death-related genes, including autophagy-related genes PRKAB1, PRKAB2, TSC2, ATG2, ATG4, ATG7, ATG10,ULK1, BNIP3, DRAM1, and SESN2. On the other hand, a cytosolic form of p53 was shown to inhibit AMPK and activate the mTOR pathway. In this context, non-genotoxic stress by autophagy-inducing agents such as rapamycin, tunicamycin and nutrient deprivation favored HMD2/MDM2-dependent p53 degradation by the UPS. Interestingly, HMD2/MDM2 stability and activity were also regulated by E3 ligases SMURF1/2 which in turn affected the stability of p53. SMURF1/2-mediated ubiquitylation was shown to increase MDM2-MDMX heterodimerization, decreasing autoubiquitylation of MDM2, therefore stabilized the protein. Additionally, another E3 ligase, NEDD4-1 was shown to control MDM2 stability and p53 activation. In addition to MDM2, another E3 ligase, PIRH2, was able to ubiquitylate p53 to control its cellular stability.
NF-κB is a well studied transcriptional regulator of autophagy. As a result of its association with IκB, NF-κB is found in an inactive state in the cytosol. In response to agonists, IκB was reported to be ubiquitylated and subsequently degraded by the UPS. Regulation of NF-κB by external signals involved phosphorylation of IκB by upstream kinases of the IKK complex (IKKα, IKKβ, and IKKγ/NEMO). Phosphorylated IκB recruits the E3 ligase SCF-βTRCP, followed by its degradation in the proteasome. After IκB degradation, NF-κB was then free to migrate to the nucleus of the cell, and induce transcription of target genes, including Beclin1 and p62, and induce autophagy.
Another level of regulation involved TNF-α receptor-associated protein complexes. Binding of TNF-α to TNFR1 led to the recruitment of TRADD and RIPK1 to the receptor, promoting TRAF- and cIAP-mediated K63 and/or K11 linked ubiquitylation of the RIPK1. Ubiquitylated RIPK1 could recruit NEMO and TAB-TAK1 complex for IKK activation and hence NF-κB stimulation. Additionally, RIPK1 could also be modified by A20 through addition of K48-linked poly-ubiquitin chains, sending the kinase for proteasomal degradation.
However, in some contexts, TNF-α-induced NF-κB activation was reported to inhibit autophagy. TNF-α-induced activation of IKKα or IKKβ could stimulate phosphorylation of TSC1/2 and activate mTOR, leading to a similar inhibitory outcome. Furthermore in some contexts, RIPK1 silencing activated autophagy under both basal and stress conditions. On the other hand, RIPK1 itself was reported to be a target of p62-mediated selective autophagy. Moreover, autophagy was responsible for the degradation of NF-κB activator NIK and IKK complex subunits, indicating the presence of a tight cross-regulation of the NF-κB pathway by the UPS and autophagy.
Another transcription factor that was controlling the autophagic outcome was HIF1α. Hypoxia induced HIF1α transcriptionally regulated various hypoxia response genes, including GLUT1, NOX2, and PDK1 as well as autophagy genes, including BNIP3, BNIP3L, ATG5, and BECN1 to stimulate autophagy, mitophagy, and pexophagy. HIF1α itself was regulated in a UPS-dependent manner. Under normoxia, hydroxylation of HIF1α specific prolyl hydroxylases (PHDs) hydroxylated HIF1α served as a recognition signal for UbcH5, an E2 enzyme and von Hippel-Lindau protein (the pVHL), E3 ligase complex containing Elongin B and C, Cullin-2, and Rbx1 allowing K48 linked ubiquitination of HIF1α and its proteasomal degradation. In contrast, during hypoxia, PHDs were inhibited and HIF1α stabilized. SCF E3 ligase complex was also a regulator of HIF1α stability in response to GSK3β-mediated phosphorylation of the protein. Another E3 ligase facilitating HIF1α degradation was HAF (also known as SART1800). Unlike pVHL, HAF-mediated ubiquitylation of HIF1α was not depending on the oxygen levels, providing an alternative HIF1α regulation mechanism. Stability of PHD proteins were also controlled by the UPS. For example, SIAH1/2 was shown to direct PHDs for proteasomal degradation under hypoxic stress. Moreover several DUBs were implicated in HIF1α regulation, including USP20, USP28, and USP33.
FOXO family of transcription factors (FOXOs) were associated with various cellular pathways, including autophagy. The activity of FOXOs were regulated by their phosphorylation status and following activation, FOXOs translocated to the nucleus and triggered the expression of a number of genes associated with different stages of the autophagy pathway, including ATG4, ATG12, BECN1, ULK1, PIK3C3, MAP1LC3, and GABARAP. There are several connections between FOXOs and autophagy. Activation of the AKT pathway inhibited FOXO3 activity, led to a decrease in LC3 and BNIP3 expression, therefore blocked autophagy. On the other hand, AMPK activation led to the phosphorylation of FOXO3a and ULK1, inducing MAP1LC3, GABARAP, and BECN1 expression and subsequent autophagy activation. Another FOXO family protein FOXK1/2, a negative regulator of FOXO3, was associated with a decrease in autophagy by removing Sin3A/HDAC complex from histone H4 to diminish its acetylation. In this context, nuclear localization of FOXK1/2 was mTOR-dependent and showed an inhibitory effect on autophagy gene expression under basal conditions. Moreover, JNK deficiency in neurons increased autophagic activity through FOXO1-mediated BNIP3 upregulation and Beclin1 disassociation from BCL-XL. Another example of a link between FOXOs autophagy involved ATG14. Liver specific knockout of FOXOs resulted in the downregulation of ATG14 and this event was associated with high levels of triglycerides in the liver and serum of mice. Additionally, GATA-1 shown to directly regulate FOXO3-mediated activation of LC3 genes to facilitate autophagic activity.
Phosphorylation of FOXO proteins by various protein kinases, including AKT, IKK, and ERK, affected their ubiquitylation by E3 ligases and their stability. For instance, AKT-mediated phosphorylation of FOXO1 provided a signal for its recognition by the SKP protein, an SCF E3 ligase complex component, followed by FOXO1 ubiquitylation and degradation. COP1 was also identified as an E3 ligase that regulated FOXO protein stability. COP1 ubiquitylated FOXO1 and promoted its proteasomal degradation. This type of regulation might be important in the glucose metabolism of hepatocytes, and possibly in autophagy modulation under this conditions. Another FOXO regulating E3 ligase was MDM2 that was reported to be responsible for FOXO1 and FOXO3A ubiquitylation and degradation. MDM2-mediated ubiquitylation was activated by the phosphorylation of FOXOs by AKT. Due to its role in p53 regulation, MDM2 could be part of a more complex regulatory mechanism which might link the UPS, transcriptional regulation and autophagic activity.
NRF2-KEAP1-P62 pathway was defined as another major oxidative stress response mechanism involving an interplay between the UPS and autophagy. NRF2 is a transcription factor, and when activated, is upregulated antioxidant and metabolic enzymes, including TXNRD1, HMOX1, GPX2, GBE1, PHK1, and downregulated proinflammation-related genes such as IL6, IL1B. KEAP1 is an adaptor protein of the E3 ligase Cullin-3 and plays a role in substrate recognition. Under normal conditions, transcription factor NRF2 was found in association with KEAP1-Cullin-3 E3 ligase complex, that catalyzed its ubiquitylation, rendering it a substrate for proteasomal elimination by selective autophagy. Competition resulted in the migration of free NRF2 to the nucleus and transactivation of stress-related cytoprotective genes. Additionally, the NRF2–KEAP1 pathway provides a positive feedback loop for autophagy. P62 was characterized as a direct transcriptional target of activated NRF2. Moreover, KEAP1 regulation by p62 was modulated by the E3 ligase TRIM21. NRF2 activation was negatively affected by TRIM21-mediated K63-linked ubiquitylation of p62.
Autophagy-UPS Crosstalk in Diseases
Crosstalk between autophagy and the UPS may change character under disease conditions, contribute to the pathogenesis of diseases and even affect their outcome. Degenerative diseases and cancer are examples of diseases that illustrate the interplay between the UPS and autophagy in the clearance of misfolded abnormal proteins.
For example, Huntington Disease is caused by poly-glutamine extensions in a protein called Huntingtin (Htt), leading to abnormal organization and eventual aggregation of the protein. Htt protein was shown to be ubiquitylated via K48- or K63-linked ubiquitin chains. Mutant Htt clearance depended on both the UPS and autophagy in different experimental settings. Mutant Htt aggregates were largely cleared by K63-dependent autophagy mechanisms. On the other hand, overexpression of K48-specific E3 ligase Ube3a, resulted in a UPS-dependent degradation of mutant proteins. Yet, cellular levels of E3 ligase was shown to decline in an age-dependent manner. Therefore, in elderly people, accumulation of K63-linked polyubiquitylated proteins might tip the balance toward clearance of protein aggregates by autophagy. A similar UPS switch was also observed in a CHIP-dependent manner.
Another example involves the ERAD protein p97/VCP. Mutant forms of the protein were associated with a rare syndrome that mainly affects muscles, bones and the brain (Inclusion Body Myopathy with the Paget’s Disease of Bone and frontotemporal Dementia, IBMPFD). Moreover, p97/VCP mutations were detected in a fraction of patients suffering from familial forms of Parkinson’s Disease or from Amyotrophic Lateral Sclerosis (ALS). As mentioned in the previous sections, p97/VCP is important for the extraction of misfolded ER proteins as well as their delivery to proteasomes. Moreover, p97/VCP was proposed to play a role in autophagosome maturation and autolysosome formation. We recently showed that some of the disease-related mutations of p97/VCP (namely P137L and G157R) resulted in the aggregation of the protein itself. Mutant p97/VCP proteins formed complexes with wild-type counterparts and led to further accumulation of ubiquitylated proteins upon ER stress, indicating that the ERAD system was negatively affected by the mutant. Indeed, ERAD co-factor and ubiquitin binding capacity of the mutant p97/VCP was decreased. Yet, autophagy was still functional under these conditions, and could significantly eliminate these aggregates. Therefore, preferential elimination of mutant proteins by autophagy might tip the balance in favor of wild-type proteins and restore disease-related loss of cellular functions including UPS-related mechanisms.
The role of the crosstalk between the two systems is also prominent in the cancer context. For example, the P53-regulated and cancer-related protein EI24, was introduced as a critical link between the UPS and autophagy. EI24 controlled the stability of E3 ligases TRIM41, TRIM2, and TRIM28 by the regulation of their autophagic degradation. Cellular levels of other E3 ligases, namely MDM2 and TRAF2, were also regulated by EI24-controlled degradation, modulating p53 and mTOR pathways, respectively, and influencing cancer formation and progression.
Deregulation and/or mutations of proteins that function in the autophagy and/or the UPS were observed in some cancer types, resulting in the modification of individual pathways and possibly affecting the crosstalk between the two systems. Changes include, modulation of levels of E3 ligases such as MDM2, SMURF1, SCF components (e.g., βTrCP), point mutations of NEDD4, COP1, FBXW7, and mutations in autophagy related proteins Beclin1, LKB1, ATG5, ATG4C as well as deletions of genes of proteins, such as Beclin1, AMPK and UVRAG. Under these circumstances, dynamic and complex changes in the regulation of the degradative pathways should have dramatic effects that contribute to cancer-related alterations in the proteomic landscape of cells.
Autophagy-UPS crosstalk emerges as a critical factor that determines the success of disease treatment, chemotherapy is one striking example. For instance, proteasome inhibition by the chemotherapy agent bortezomib resulted in the accumulation of misfolded proteins and induced compensatory autophagy in cancer cells. Under these circumstances, autophagic activity protected cancer cells from bortezomib-induced cell death, and inhibition of autophagy improved the outcome of chemotherapy. These dual autophagy-UPS targeting approaches also gave promising results in clinical trials.
Several companies are now developing drugs that modulate the UPS or autophagy [for example,]. Concepts and data that were discussed above and elsewhere indicate that, depending on the disease type and treatment strategy, the crosstalk between the UPS and autophagy should definitely be taken into account in these efforts.
Conclusion and Perspectives
Autophagy and the ubiquitin proteasome systems are major degradation systems in mammalian cells that allow recycling of cellular contents ranging from soluble proteins to intracellular organelles. Although their mode of action and their requirements for substrate recognition are different, there are several overlaps and interconnections between the UPS and autophagy pathways.
A prominent component of the crosstalk is the ubiquitin protein itself and ubiquitylation. Indeed, ubiquitin is a common signal for both the UPS and autophagy. It was proposed that, ubiquitin chain type could determine the pathway of choice for protein degradation. K48-linked ubiquitylation was introduced to be a signal for the UPS, whereas K63-linked ubiquitylation directed proteins for autophagosomal degradation. Yet, a number of independent studies provided evidence that both ubiquitylation types could lead to autophagic degradation of substrates. Moreover, recent studies underline the importance of ubiquitin phosphorylation as an event that increased the affinity of autophagy receptors for their targets during selective autophagy. Additionally, non-ubiquitin modifications (e.g., acetylation, sumoylation, neddylation etc.) were shown to affect protein degradation as well. Therefore, a barcode of ubiquitin and other modifications seem to prime proteins for one or the other degradation pathway and determine their fate. As another level of regulation, deconjugating enzymes such as DUBs may counteract or redirect proteins for different degradation systems.
E3 ligases emerged as important components of the UPS-autophagy switches. For example, Cullin-3, SMURF1, MDM2 E3 ligases directed proteins to degradation by the UPS, whereas the role of Parkin, LRSAM1, and CHIP in priming proteins for autophagic degradation was observed in several studies. On the other hand, the same E3 ligase that might be able to generate different ubiquitin linkages under different conditions and on different substrates, the switch between degradative pathways being controlled by specific E3 ligase adaptors, post-translational modifications on target proteins as well as other unknown factors. A prominent example is the Parkin protein. During mitophagy although some of the proteins that are ubiquitylated by Parkin are degraded, other ubiquitylated proteins contribute to mitochondrial clustering and recognition by autophagy receptors. To date, factors or modifications that determine the substrate selectivity of Parkin are unknown.
Another example of UPS-autophagy switch involves the p97/VCP protein. While binding of the co-factor PLAA to p97/VCP resulted in the autophagic degradation of ubiquitylated clients of the protein, binding of UFDL1 as a co-factor favored degradation by the UPS. Moreover, p97/VCP was also associated with aggregate formation in collaboration with some autophagy receptors.
Signaling switches involved in the regulated activation of one or the other system was shown to modify cellular responses to stress. For example, NRF2 degradation by the UPS was controlled through p62-mediated KEAP1 elimination by autophagy. Prevention of HIF1α degradation by the UPS, resulted in the expression of stress response genes, including autophagy genes, led to autophagy activation. In another example, the UPS activity was required for NF-κB activation and NF-κB-mediated autophagy gene upregulation. Yet, autophagic degradation of NF-κB activators NIK and IKKs provided a negative feedback loop in the control in this context. Therefore, modification of cellular signaling pathways by degradative systems might modulate upstream signals that control autophagy and/or the UPS, and affect their activation and amplitude.
Degradation of the components or regulators of one system by the other system was also reported. For example, proteasomes were defined as substrates of selective autophagy. Conversely, various autophagy proteins were ubiquitylated and degraded by the UPS in a regulated manner. Therefore, checks and balances between the two systems exist, and these control mechanisms possibly allow remodeling of the cellular proteome under different conditions.
Compensation mechanisms are also operational between the two systems. Inhibition of the UPS generally upregulated autophagy, whereas failures in the autophagy system were associated with increased UPS activity, although inefficient compensation and failure in both systems were also observed under certain conditions. Moreover, alternative protein degradation pathways, such as CMA and microautophagy might come into play under these conditions as well. Nevertheless, depending on the character of the target to be degraded, compensation mechanisms were less or more effective. For example, large aggregates and whole organelles should be cleared by autophagy, but defective ribosomal products that could not be accumulated in stress granules were shown to be directed for proteasomal degradation. Therefore for cellular homeostasis and for proper functioning of cells, ideally both systems should be fully operational.
Data obtained so far demonstrate that crosstalk and communication between autophagy and the UPS generally rely on non-specialized and even indirect links. Yet, there might exist so far unknown specialized proteins providing coordination and co-regulation of the two systems. Furthermore, regulation through direct protein-protein interactions between known system components is another possibility. Therefore, dedicated communication proteins or pathways between the degradation mechanisms may be present, allowing better and faster coordination in case of need. Further studies are required to unveil the nature of these putative proteins, interactions and pathways.
An emerging theme in the regulation and coordination of autophagy and the UPS involves non-coding RNAs and their intricate networks. A growing list of microRNAs as well as long non-coding RNAs were implicated in the control of autophagy as well as the UPS. MicroRNAs have the advantage of affecting the level of multiple proteins at once, and they are able to rapidly reshape cellular signaling mechanisms and pathways. Therefore, non-coding RNA networks possibly contribute to the co-regulation of these degradative systems. Intriguingly, deregulation of non-coding RNA levels contribute to the progression of diseases such as cancer. Future studies on non-coding RNAs will reveal their relevance in the autophagy-UPS crosstalk under physiological and pathological conditions.
Overall, coordination, interconnection and crosstalk mechanisms between the UPS and autophagy exist at various levels. In addition to ubiquitin and ubiquitylation, several proteins and signaling pathways were implicated in the communication and mutual regulation of the two systems. Considering the importance of protein catabolism for cellular and organismal homeostasis and health, a better understanding of individual systems as well as the interconnections and crosstalks between them will be most rewarding from both a basic science perspective and with regards to clinical management of diseases involving protein quality control problems.
Author Contributions
NK and DG wrote the manuscript and did critical reading. NK prepared the illustrations in the manuscript.
Conflict of Interest Statement
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
References 355
- 1.
Aerbajinai W. Giattina M. Lee Y. T. Raffeld M. Miller J. L. (2003 ). The proapoptotic factor Nix is coexpressed with Bcl-xL during terminal erythroid differentiation. Blood 102 712–717.
Not in our library10.1182/blood-2002-11-3324
- 2.
Altieri D. C. (2010 ). Survivin and IAP proteins in cell-death mechanisms. Biochem. J. 430 199–205.
Not in our library10.1042/bj20100814
- 3.
Ambivero C. T. Cilenti L. Main S. Zervos A. S. (2014 ). Mulan E3 ubiquitin ligase interacts with multiple E2 conjugating enzymes and participates in mitophagy by recruiting GABARAP. Cell. Signal. 26 2921–2929.
Not in our library10.1016/j.cellsig.2014.09.004
- 4.
Amodio N. Scrima M. Palaia L. Salman A. N. Quintiero A. Franco R. (2010 ). Oncogenic role of the E3 ubiquitin ligase NEDD4-1, a PTEN negative regulator, in non-small-cell lung carcinomas. Am. J. Pathol. 177 2622–2634.
Not in our library10.2353/ajpath.2010.091075
- 5.
An H. Harper J. W. (2018 ). Systematic analysis of ribophagy in human cells reveals bystander flux during selective autophagy. Nat. Cell Biol. 20 135–143.
Not in our library10.1038/s41556-017-0007-x
- 6.
Anding A. L. Baehrecke E. H. (2017 ). Cleaning house: selective autophagy of organelles. Dev. Cell 41 10–22.
Not in our library10.1016/j.devcel.2017.02.016
- 7.
Antonioli M. Di Rienzo M. Piacentini M. Fimia G. M. Albiero F. Nazio F. (2014 ). AMBRA1 interplay with cullin E3 Ubiquitin ligases regulates autophagy dynamics. Dev. Cell 31 734–746.
Not in our library10.1016/j.devcel.2014.11.013
- 8.
Arora V. Cheung H. H. Plenchette S. Micali O. C. Liston P. Korneluk R. G. (2007 ). Degradation of survivin by the X-linked Inhibitor of Apoptosis (XIAP)-XAF1 complex. J. Biol. Chem. 282 26202–26209.
Not in our library10.1074/jbc.m700776200
- 9.
Banning A. Deubel S. Kluth D. Zhou Z. Brigelius-Flohe R. (2005 ). The GI-GPx gene is a target for Nrf2. Mol. Cell. Biol. 25 4914–4923.
Not in our library10.1128/mcb.25.12.4914-4923.2005
- 10.
Bassik M. C. Scorrano L. Oakes S. A. Pozzan T. Korsmeyer S. J. (2004 ). Phosphorylation of BCL-2 regulates ER Ca2+ homeostasis and apoptosis. EMBO J. 23 1207–1216.
Not in our library10.1038/sj.emboj.7600104
- 11.
Bayraktar O. Oral O. Kocaturk N. M. Akkoc Y. Eberhart K. Kosar A. (2016 ). IBMPFD disease-causing mutant VCP/p97 proteins are targets of autophagic-lysosomal degradation. PLoS One 11:e0164864.
Not in our library10.1371/journal.pone.0164864
- 12.
- 13.
Behrends C. Harper J. W. (2011 ). Constructing and decoding unconventional ubiquitin chains. Nat. Struct. Mol. Biol. 18 520–528.
Not in our library10.1038/nsmb.2066
- 14.
Bellot G. Garcia-Medina R. Gounon P. Chiche J. Roux D. Pouyssegur J. (2009 ). Hypoxia-induced autophagy is mediated through hypoxia-inducible factor induction of BNIP3 and BNIP3L via their BH3 domains. Mol. Cell. Biol. 29 2570–2581.
Not in our library10.1128/mcb.00166-09
- 15.
Benanti J. A. (2012 ). Coordination of cell growth and division by the ubiquitin-proteasome system. Semin. Cell Dev. Biol. 23 492–498.
Not in our library10.1016/j.semcdb.2012.04.005
- 16.
Benyair R. Ogen-Shtern N. Mazkereth N. Shai B. Ehrlich M. Lederkremer G. Z. (2015 ). Mammalian ER mannosidase I resides in quality control vesicles, where it encounters its glycoprotein substrates. Mol. Biol. Cell 26 172–184.
Not in our library10.1091/mbc.e14-06-1152
- 17.
Bernardi K. M. Williams J. M. Inoue T. Schultz A. Tsai B. (2013 ). A deubiquitinase negatively regulates retro-translocation of nonubiquitinated substrates. Mol. Biol. Cell 24 3545–3556.
Not in our library10.1091/mbc.e13-06-0332
- 18.
Bertolotti A. Zhang Y. Hendershot L. M. Harding H. P. Ron D. (2000 ). Dynamic interaction of BiP and ER stress transducers in the unfolded-protein response. Nat. Cell Biol. 2 326–332.
Not in our library10.1038/35014014
- 19.
Bhat K. P. Yan S. Wang C.-E. Li S. Li X.-J. (2014 ). Differential ubiquitination and degradation of huntingtin fragments modulated by ubiquitin-protein ligase E3A. Proc. Natl. Acad. Sci. U.S.A. 111 5706–5711.
Not in our library10.1073/pnas.1402215111
- 20.
Bingol B. Tea J. S. Phu L. Reichelt M. Bakalarski C. E. Song Q. (2014 ). The mitochondrial deubiquitinase USP30 opposes parkin-mediated mitophagy. Nature 510 370–375.
Not in our library10.1038/nature13418
- 21.
Birmingham C. L. Smith A. C. Bakowski M. A. Yoshimori T. Brumell J. H. (2006 ). Autophagy controls Salmonella infection in response to damage to the Salmonella-containing vacuole. J. Biol. Chem. 281 11374–11383.
Not in our library10.1074/jbc.m509157200
- 22.
Bowman C. J. Ayer D. E. Dynlacht B. D. (2014 ). Foxk proteins repress the initiation of starvation-induced atrophy and autophagy programs. Nat. Cell Biol. 16 1202–1214.
Not in our library10.1038/ncb3062
- 23.
Braschi E. Zunino R. McBride H. M. (2009 ). MAPL is a new mitochondrial SUMO E3 ligase that regulates mitochondrial fission. EMBO Rep. 10 748–754.
Not in our library10.1038/embor.2009.86
- 24.
Buchan J. R. Kolaitis R. M. Taylor J. P. Parker R. (2013 ). Eukaryotic stress granules are cleared by autophagy and Cdc48/VCP function. Cell 153 1461–1474.
Not in our library10.1016/j.cell.2013.05.037
- 25.
Budanov A. V. Karin M. (2009 ). The p53-regulated Sestrin gene products inhibit mTOR signaling. Cell 134 451–460.
Not in our library10.1016/j.cell.2008.06.028.the
- 26.
Carvalho A. F. Pinto M. P. Grou C. P. Alencastre I. S. Fransen M. Sá-Miranda C. (2007 ). Ubiquitination of mammalian Pex5p, the peroxisomal import receptor. J. Biol. Chem. 282 31267–31272.
Not in our library10.1074/jbc.m706325200
- 27.
Cassavaugh J. M. Hale S. A. Wellman T. L. Howe A. K. Wong C. Lounsbury K. M. (2011 ). Negative regulation of HIF-1α by an FBW7-mediated degradation pathway during hypoxia. J. Cell. Biochem. 112 3882–3890.
Not in our library10.1002/jcb.23321
- 28.
Chan N. C. Salazar A. M. Pham A. H. Sweredoski M. J. Kolawa N. J. Graham R. L. J. (2011 ). Broad activation of the ubiquitin-proteasome system by Parkin is critical for mitophagy. Hum. Mol. Genet. 20 1726–1737.
Not in our library10.1093/hmg/ddr048
- 29.
Chang H. Liu Y. Wang L. Wang J. Zhao Z. Qu J. (2018 ). MiR-182 promotes cell proliferation by suppressing FBXW7 and FBXW11 in non-small cell lung cancer. Am. J. Transl. Res. 10 1131–1142.
Not in our library
- 30.
Chen C. Pore N. Behrooz A. Ismail-Beigi F. Maity A. (2001 ). Regulation of glut1 mRNA by hypoxia-inducible factor-1: interaction between H-ras and hypoxia. J. Biol. Chem. 276 9519–9525.
Not in our library10.1074/jbc.m010144200
- 31.
Chen M. Chen Z. Wang Y. Tan Z. Zhu C. Li Y. (2016 ). Mitophagy receptor FUNDC1 regulates mitochondrial dynamics and mitophagy. Autophagy 12 689–702.
Not in our library10.1080/15548627.2016.1151580
- 32.
Chen Z. Liu L. Cheng Q. Li Y. Wu H. Zhang W. (2017 ). Mitochondrial E3 ligase MARCH5 regulates FUNDC1 to fine-tune hypoxic mitophagy. EMBO Rep. 18 495–509.
Not in our library10.15252/embr.201643309
- 33.
Christianson J. C. Olzmann J. A. Shaler T. A. Sowa M. E. Bennett E. J. Richter C. M. (2012 ). Defining human ERAD networks through an integrative mapping strategy. Nat. Cell Biol. 14 93–105.
Not in our library10.1038/ncb2383
- 34.
Clague M. J. Heride C. Urbé S. (2015 ). The demographics of the ubiquitin system. Trends Cell Biol. 25 417–426.
Not in our library10.1016/j.tcb.2015.03.002
- 35.
Clausen T. H. Lamark T. Isakson P. Finley K. Larsen K. B. Brech A. (2010 ). p62/SQSTM1 and ALFY interact to facilitate the formation of p62 bodies/ALIS and their degradation by autophagy. Autophagy 6 330–344.
Not in our library10.4161/auto.6.3.11226
- 36.
Cohen-kaplan V. Livneh I. Avni N. Fabre B. Ziv T. Tae Y. (2016 ). p62- and ubiquitin-dependent stress-induced autophagy of the mammalian 26S proteasome. Proc. Natl. Acad. Sci. U.S.A. 113 E7490–E7499.
Not in our library10.1073/pnas.1615455113
- 37.
Collins C. A. De Mazière A. Van Dijk S. Carlsson F. Klumperman J. Brown E. J. (2009 ). Atg5-independent sequestration of ubiquitinated mycobacteria. PLoS Pathog. 5:e1000430.
Not in our library10.1371/journal.ppat.1000430
- 38.
Collins G. A. Goldberg A. L. (2017 ). The logic of the 26S proteasome. Cell 169 792–806.
Not in our library10.1016/j.cell.2017.04.023
- 39.
Copetti T. Bertoli C. Dalla E. Demarchi F. Schneider C. (2009 ). p65/RelA modulates BECN1 transcription and autophagy. Mol. Cell. Biol. 29 2594–2608.
Not in our library10.1128/mcb.01396-08
- 40.
Cornelissen T. Haddad D. Wauters F. Van Humbeeck C. Mandemakers W. Koentjoro B. (2014 ). The deubiquitinase USP15 antagonizes Parkin-mediated mitochondrial ubiquitination and mitophagy. Hum. Mol. Genet. 23 5227–5242.
Not in our library10.1093/hmg/ddu244
Cited by papers we hold 5
- Normal Aging Induces Changes in the Brain and Neurodegeneration Progress: Review of the Structural, Biochemical, Metabolic, Cellular, and Molecular Changes · 2022No notice
- Ubiquitination-Proteasome System (UPS) and Autophagy Two Main Protein Degradation Machineries in Response to Cell Stress · 2022No notice
- Degradation of proteins by PROTACs and other strategies · 2020Not checked
- Mechanisms of Photoreceptor Death in Retinitis Pigmentosa · 2020No notice
- Ubiquitin-mediated regulation of autophagy · 2019No notice
Reads
Read by people on this page
None yet.
Checked or read by AI through naulon
None yet.
A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.
Are you an author?
Is this your paper?
Claim it to add your note and be paid when agents cite it. Free.