Regulation of neuronal autophagy and the implications in neurodegenerative diseases

Abstract

Neurons are highly polarized and post-mitotic cells with the specific requirements of neurotransmission accompanied by high metabolic demands that create a unique challenge for the maintenance of cellular homeostasis. Thus, neurons rely heavily on autophagy that constitutes a key quality control system by which dysfunctional cytoplasmic components, protein aggregates, and damaged organelles are delivered to the lysosome for degradation. While mature lysosomes are predominantly located in the soma of neurons, the robust, constitutive biogenesis of autophagosomes occurs in the synaptic terminal via a conserved pathway that is required to maintain synaptic integrity and function. Following formation, autophagosomes fuse with late endosomes and then are rapidly and efficiently transported by the microtubule-based cytoplasmic dynein motor along the axon toward the soma for lysosomal clearance. In this review, we highlight the recent knowledge of the roles of autophagy in neuronal health and disease. We summarize the available evidence about the normal functions of autophagy as a protective factor against neurodegeneration and discuss the mechanism underlying neuronal autophagy regulation. Finally, we describe how autophagy function is affected in major neurodegenerative diseases with a special focus on Alzheimer’s disease, Parkinson’s disease, and Amyotrophic Lateral Sclerosis.

1. Introduction

Autophagy is an evolutionarily conserved pathway essential for maintaining neuronal health and preventing neurodegeneration. The term for autophagy was first coined by Christian de Duve (De Duve et al., 1966; Mizushima, 2008) from the Greek αὐτός (the reflexive pronoun, for the “self”) and φαγεῖν (to eat). Macroautophagy, hereafter referred to as autophagy, is a major cytosolic degradative system involving sequestration of damaged cellular components and dysfunctional organelles within autophagosomes for subsequent lysosomal degradation (Ktistakis and Tooze, 2016). Upon autophagy activation, an initial step is to form a pre-autophagosomal membrane called phagophore or isolation membrane that is elongated to facilitate cargo engulfment and finally enclosed to generate autophagosomes. The outer membrane of autophagosomes thus fuses with lysosomes to form autolysosomes so that the engulfed cargoes can be degraded within autolysosomes through the activity of lysosomal hydrolases. This mechanism relies on dedicated autophagy regulators—the autophagy-related proteins—and can be induced by various stress stimuli, including nutrient deprivation, energy loss, redox condition, hypoxia, as well as the presence of protein aggregates and intracellular pathogens. Importantly, loss of core autophagy genes, such as atg5 or atg7, results in aberrant accumulation of ubiquitinated protein aggregates along with late-onset neurodegeneration (Hara et al., 2006; Komatsu et al., 2006), suggesting that this cellular process is crucial for the maintenance of neuronal homeostasis.

Neurons are highly polarized cells and characterized by a complex dendritic arbor and a very long axon that emerges from the soma and bridges vast distances that can extend more than a meter in the human body. The neuronal processes are decorated by an enormous number of synapses in which neurotransmission comes along with high energy demands, enhanced protein turnover rates, and high membrane exchange to support the efficient delivery and constant supply of newly synthesized proteins (Rangaraju, et al., 2014, Harris et al., 2012, Ziv 2018). Therefore, sophisticated mechanisms for the timely and spatially disposal of damaged proteins and organelles at synaptic terminals are particularly important to sustain neuronal activity (Bhukel et al., 2019; Cohen and Ziv 2017; Rinetti and Schweizer, 2010). Autophagy, among others such as chaperone proteins, the ubiquitin-proteasome system (UPS), and endoplasmic reticulum (ER)-associated protein degradation (ERAD), constitutes a key mechanism of protein quality control to ensure the physical integrity of proteins and organelles and has been established to play a pivotal role in maintaining functional synapses and preventing neurodegeneration.

Alterations in the autophagy pathway are widely implicated in protein aggregation and toxicity occurring in neurodegenerative diseases and underlie the earliest events preceding the emergence of neuropathology (Nixon, 2013). These include a progressively increasing number of Alzheimer’s disease (AD), Parkinson’s disease (PD), amyotrophic lateral sclerosis (ALS), frontotemporal dementia (FTD), and Huntington’s disease (HD) (Nixon et al., 2005; Madeo et al., 2009; Ciechanover and Kwon, 2015; Martin et al., 2015; Ramesh and Pandey, 2017; Maurel et al., 2018). Despite etiological differences and neuroanatomical site-specificity, a failure in the cellular clearance systems associated with aggregation and deposition of misfolded, aggregation-prone proteins is appearing as a unifying feature in these diseases (Ciechanover and Kwon, 2015). Examples of specific disease-associated proteins include amyloid-beta (Aβ) and tau in AD, α-synuclein (α-syn) in synucleinopathies such as PD, TAR-DNA-binding protein of 43 kDa (TDP-43) and Cu/Zn superoxide dismutase 1 (SOD1) in ALS, and polyglutamine-expanded huntingtin in HD. Autophagosomal proliferation is a common feature in patient brains with these neurodegenerative diseases, but the underlying mechanisms remain largely unknown. In this review, we summarize the progress that has been made in the elucidation of the mechanisms of neuronal autophagy, and review recent findings of the cellular mechanisms underlying autophagy defects along with impaired turnovers of misfolded proteins and protein aggregates, dysfunctional organelles, and damaged synaptic components and the implications in the pathogenesis of neurodegenerative diseases, including AD, PD, and ALS.

2. Autophagy physiology in neuronal maintenance

The post-mitotic nature of neurons makes it impossible for them to dilute damaged proteins and organelles through divisions. It has been shown that synapses are relatively stable and their integrity and plasticity for an entire lifetime need to be properly maintained (Bishop et al., 2010, Lopez-Otin et al., 2013). Thus, maintaining the functional and structural integrity of neurons and their synapses over decades poses a unique challenge for neurons. Strong evidence was provided more than a decade ago that loss of neuronal autophagy is fatal in the first postnatal months of rodents (Komatsu et al., 2005). The follow-up study of mice lacking atg7 specifically in the central nervous system showed marked neuronal death in the cerebral cortex, coupled with abnormal accumulation of polyubiquitinated proteins as inclusion bodies in neurons deficient in autophagy (Komatsu et al., 2006). Similar defects were observed in mice with selective deletion of atg5 in neurons (Hara et al., 2006). To further delineate the function of autophagy in different neuronal types, targeted knockouts of atg5 and atg7 in Purkinje neurons of the cerebellum (Komatsu et al., 2007; Nishiyama et al., 2007), atg7 in agouti-related peptide (AgRP) neurons of the hypothalamus (Kaushik et al., 2011), as well as atg5 and atg7 in rhodopsin neurons of the retina (Chen et al., 2013; Zhou et al., 2015) have been developed. Deletion of atg5 and atg7 results in cell-autonomous Purkinje neuron degeneration. The earliest signs of homeostatic disruption in these mice were Purkinje axonal swellings, which were observed prior to progressive dystrophy and degeneration of the axon (Komatsu et al., 2007; Nishiyama et al., 2007). These findings collectively support a key role of autophagy in preventing axonal degeneration, an early sign of neuronal dysfunction. Studies from cell culture systems indicate that different neuronal types may have varying capacities to degrade autophagy cargoes, which correlates with the sensitivity of that cell type to the toxic accumulation of aggregate-prone proteins (Tsvetkov et al., 2013). These results suggest that autophagy in neurons operates as a crucial checkpoint for the quality control of proteins and organelles, serving as a homeostatic mechanism required for neuronal maintenance. Notably, autophagy impairment is detrimental, whereas autophagy upregulation appears to be protective. Studies in Caenorhabditis elegans (C. elegans) indicate a clear association between increased longevity and basal autophagy, which is likely mediated through multiple different but overlapping mechanisms, such as nutrient restriction (Hansen et al., 2008), altered mitosis (Ghavidel et al., 2015), and mitochondrial turnover (Palikaras et al., 2015). In line with these data, pan-neuronal overexpression of Atg8a in Drosophila leads to an extended lifespan (Simonsen et al., 2008). Moreover, overexpression of Atg5 in mice induces autophagy and extends the lifespan, which is associated with anti-aging phenotypes including leanness, increased insulin sensitivity, and improved motor function (Pyo et al., 2013). Together, these findings consistently suggest a beneficial effect of enhancing basal autophagy during aging.

3. Autophagy mechanism in neurons

The mechanism underlying autophagosome biogenesis has been elucidated through several decades of studies from yeast and mammalian cells, primarily under stress-induced conditions (Hurley and Young, 2017; Yu et al., 2017; Mizushima and Komatsu, 2011; Mizushima et al., 2008). This pathway consists of induction and nucleation phases involving the activation and recruitment of two key protein complexes—Unc-51 like autophagy activating kinase (ULK1) complex and the Class III phosphatidylinositol (PtIns) 3-kinase (PI3K) complex. This results in the formation of PtdIns(3)-phosphate (PtdIns(3)P) that recruits double FYVE-containing protein 1 (DFCP1) and promotes the formation of the omegasome at the surface of the ER, from which autophagosomes appear to be generated. The E3-like Atg12–Atg5–Atg16L1 complex and the LC3–phosphatidylethanolamine (PE) conjugate are also recruited, which play important roles in the elongation and closure of the phagophore/isolation membrane to generate autophagosome. LC3 (Atg8 homolog) is microtubule-associated protein 1A/1B-light chain 3. A cytosolic form of LC3 (LC3-I) is conjugated to PE and becomes LC3-PE (LC3-II), a structural protein of the autophagosome membrane. The Atg12–Atg5–Atg16L1 complex is also required for the formation of the covalent bond between LC3-I and PE so that LC3 is lipidated and incorporated into the delimiting membranes in the form of LC3-PE (LC3-II) (Fig. 1).

Fig. 1.

Autophagosome biogenesis and maturation in the autophagy pathway. Autophagosome biogenesis is a stepwise process with multiple molecular machineries involved, starting from omegasome formation on the surface of ER membrane. This double membrane omegasome then grows to form an isolation membrane/phagophore that engulfs cargoes such as damaged cellular components, protein aggregates, and defective mitochondria, and is finally enclosed to become an autophagosome. Once autophagosomes are generated, they fuse with lysosomes to mature into autolysosomes in which sequestered cargoes are degraded through the activity of lysosomal hydrolases. Before fusion with lysosomes, autophagosomes usually undergo a progressive maturation process by interacting with multi-vesicular bodies (MVBs)/late endosomes (LEs) to form intermediate/hybrid compartments—amphisomes.

3.1. Spatiotemporal specificity of autophagy in neurons

Autophagosome biogenesis has been proposed to be a process with pronounced temporal and spatial specificity (Fig. 2). The soma of neurons is the primary site of the degradative pathway in which degradative organelles including lysosomes are mainly located. However, robust autophagosome biogenesis can be detected predominantly in distal axons (Maday et al., 2012; Maday et al., 2014; Cheng et al., 2015; Stavoe et al., 2016). Thus, neurons confront unique challenges to efficiently remove these autophagosomes containing sequestered cargoes from distal axons for their clearance within lysosomes in the soma. While the factors specifying this spatial specificity remain to be determined, each biogenesis event at axonal terminals appears to occur within minutes, with components recruited with predictable kinetics, suggesting high temporal specificity (Maday et al., 2012; Maday et al., 2014; Stavoe et al., 2016). Earlier studies revealed the retrograde movement of membranous organelles in the axons of cultured embryonic peripheral neurons, and these organelles appeared as the multilamellar structure at the ultrastructural level by EM analysis, suggesting that they most likely represent autophagic vacuoles (AVs) (Hollenbeck 1993). Furthermore, direct evidence showing dynamic retrograde transport of autophagosomes has been provided lately by several studies using dorsal root ganglion (DRG) neurons from rats (Cheng et al., 2015a), primary mouse embryonic cortical neurons (Lee et al., 2011), as well as DRG neurons cultured from a transgenic mouse expressing the autophagosomal marker LC3 fused to GFP (Maday et al., 2012). Autophagosome biogenesis events can be consistently detected in distal axons, involving recruitment of multi-subunit complexes that coordinate the continuous growth of the initial phagophore membrane and LC3 lipidation (Cheng et al., 2015a; Maday et al., 2014; Wallot-Hieke et al., 2018). Of note, even though the ordered recruitment during autophagosome assembly detected in neurons is very similar to that in non-neuronal cells, autophagosome formation is constitutively active in neurons and remains steady under basal and growth-promoting conditions (Cheng et al., 2015a; Lee et al., 2011; Maday et al., 2012; Maday et al., 2014), which supports the notion that autophagy acts as a baseline homeostatic mechanism in neurons.

Fig. 2.

Autophagy mechanism in neurons. The autophagy pathway in neurons exhibits pronounced temporal and spatial specificity. Robust autophagosomes are generated constitutively at synaptic terminals where damaged synaptic components and dysfunctional mitochondria are sequestered within autophagosomes. Newly formed autophagosomes rapidly fuse with LEs to recruit LE-loaded dynein-Snapin motor-adaptor complex, thereby enabling long-distance retrograde transport motility toward the soma, the primary site of the degradative pathway in which mature lysosomes are mainly located. In addition, PINK1/Parkin-mediated mitophagy predominantly occurs in the soma of neurons.

While studies using in vitro cell culture systems have yielded important information about dynamic autophagosome biogenesis in neurons, recent in vivo work further highlighted some differences in the composition and the rate of autophagy activity and vesicular trafficking in the well-established axonal architecture of mouse brains (Lie et al. 2021; Bagalkot et al. 2021). In mature intact mouse brains, fully mature lysosomes are restricted from myelinated axons in which, besides AVs and late endosomes (LEs), many notably tubulovesicular structures containing both lysosomal hydrolases and the Trans-Golgi network (TGN) protein marker as TGN-derived transport carriers (TCs) are present. These TCs facilitate the delivery of lysosomal constituents to axons. Although TC-mediated delivery of lysosome components is upregulated in the dystrophic axons of disease, the degradation capacity to retrograde organelles in axons is relatively limited and lysosomal proteolysis mainly occurs within the soma of neurons (Lie et al. 2021). Therefore, the vulnerability toward axonal dystrophy associated with neurodegenerative diseases is likely attributed to the restricted entry of mature lysosomes into axons and thus a limited degradation capacity in axons. These findings also support the view that autophagosomes can be quickly eliminated by lysosomes in the soma, whereas a newly formed autophagosome in distal axons has a prolonged lifespan before encountering a lysosome.

3.2. Axonal transport of autophagosomes

Once autophagosomes are produced in distal axons, fusion with LEs/multivesicular bodies (MVBs) to form amphisomes is necessary for nascent autophagosomes to gain retrograde transport motility. Earlier studies documented strong initial experimental evidence of the unusual rapidity by which newly generated autophagosomes acquired the endosomal markers to form amphisomes and thus underwent exclusive retrograde movement in axons (Lee et al., 2011; Cheng et al., 2015a). Such a mechanism allows amphisomes to function as a transporter and return autophagic cargoes to the soma for lysosomal degradation (Lee et al., 2011 JN; Cheng et al., 2015a; Ganesan and Cai, 2021). Multiple lines of evidence indicate that retrograde motility of autophagosomes is driven by cytoplasmic dynein, the major microtubule-dependent and minus-end-directed motor (Jahreiss et al., 2008; Ravikumar et al., 2005). Mutation or inhibition of dynein motor activity limits autophagic clearance (Ravikumar et al., 2005, Katsumata et.al; 2010). Studies have identified multiple adaptor proteins involved in attaching dynein motors to autophagosomes and thereby enabling their retrograde movement. Intriguingly, the function of these dynein motor adaptors is not restricted to the transport of autophagosomes, and they also drive the transport of endosomes, suggesting a common mechanism for retrograde transport of both autophagosomes and endosomes (Zhou et al., 2012, Andres-Alonso et al., 2019, Kononenko et al., 2017, Fu and Holzbaur., 2014, Khobrekar et.al 2020, Wong and Holzbaur 2014a). Our previous studies have established that Snapin functions as a dynein motor adaptor and thus mediates retrograde transport of LEs along axons (Cai et al., 2010, Sheng and Cai 2011). Snapin directly interacts with the dynein intermediate chain (DIC) and this interaction mediates dynein motor recruitment to the membrane of LEs, enabling LE retrograde movement toward the soma. Furthermore, we have demonstrated that nascent autophagosomes in distal axons rapidly fuse with LEs to form amphisomes through which autophagosomes can be loaded with the dynein-Snapin motor-adaptor transport machinery and thus gain retrograde transport motility (Cheng et al., 2015a, 2015b). Therefore, efficient removal of autophagic cargoes from axonal terminals via the dynein-Snapin-mediated retrograde transport is critical for autophagic clearance within lysosomes in the soma, protecting against autophagic stress in distal axons. Besides, AP-2 was also reported to regulate retrograde transport of autophagosomes/amphisomes, involving the association of AP-2α_A, a large brain-specific subunit of AP-2 with LC3 and of AP-2β with the p150^Glued subunit of the dynein motor cofactor—the dynactin complex (Kononenko et al., 2017). Neurons in the absence of AP-2 exhibit impaired retrograde transport of autophagosomes/amphisomes and autophagy defects. Interestingly, the role of AP-2 in autophagosome transport is independent of its established function in endocytosis. It is unclear how the dynein-Snapin and AP-2-dynactin transport machineries are coordinated to modulate retrograde transport of amphisomes in the axon. Finally, several recent studies have identified multiple dynein effectors such as the striatin-interacting phosphatase and kinase (STRIPAK) complex, c-Jun N-terminal kinase–interacting protein 1 (JIP1), Huntingtin, and Htt-associated protein 1 (HAP1) on autophagosomes that can also regulate axonal autophagosome motility in the course of autophagosome maturation (Neisch et al., 2017; Cason et al., 2021; Fu et al.,2013; Fu and Holzbaur 2014; Wong and Holzbaur, 2014a). However, it remains to be determined whether retrograde transport motility of axonal autophagosomes is regulated through a single consolidated, multi-component mechanism, or whether autophagosome transport is differentially regulated by distinct adaptor and effector proteins of dynein motors.

3.3. Autophagy at synapses

The synapse forms the basic unit of communication between neurons and serves as a highly specialized compartment, packed with unique and dedicated molecules in support of neurotransmission. Anatomically, synapses are located far away from the soma and local mechanisms at the synapse for safeguarding proper function could be very important. Emerging evidence suggests autophagy to be a homeostatic mechanism that has adapted to the microenvironment of the synapse. Autophagosome biogenesis at presynaptic terminals has been extensively studied in different neuron types (Hernandez et al., 2012; Soukup et al., 2016; Stavoe et al., 2016; Williamson et al., 2010). Basal autophagy in dopaminergic neurons was shown to contribute to the maintenance of presynaptic structure and function (Hernandez et al., 2010). Loss of autophagy core component Atg7 led to abnormally enlarged presynaptic terminals in dopaminergic neurons, accompanied by an enhanced release of a larger amount of neurotransmitters with a faster presynaptic recovery. These observations indicate that basal autophagy likely acts as a repressor of presynaptic function in dopaminergic neurons. Conversely, the data from autaptic hippocampal cultures failed to show any effect of basal autophagy on neurotransmission (Hoffmann et al., 2019). Together with the earlier studies (Hara et al., 2006; Komatsu et al., 2006), these findings suggest that the significance of basal autophagy might vary at the terminals from different neuron types. Moreover, cargo clearance from synaptic terminals by autophagy is likely important under the conditions when demands for protein turnover are robustly elevated.

At synaptic terminals, autophagy can be activated upon different types of stimuli, including synaptic activity, selective induction of synaptic protein damage, or treatment with the autophagy inducer Rapamycin (Hernandez et al., 2012; Hoffmann et al., 2019, Soukup et al., 2016; Wang et al., 2015; Kulkarni et al., 2021). Activation of PKA or G-protein coupled receptors was found to trigger autophagy at synapses, suggesting that the complex synaptic signaling networks play a role in synaptic autophagy regulation. Importantly, several studies have recently identified some presynaptic proteins with well-established roles in presynaptic physiology as autophagy modulators. Endophilin-A (EndoA) endocytic adaptors, involved in synaptic vesicle recycling, were found to regulate autophagosome biogenesis, which is independent of the roles of EndoA in endocytosis (Murdoch et al., 2016). Another study provided more evidence showing EndoA-mediated modulation of autophagosome formation at Drosophila Neuromuscular Junctions (Soukup et al., 2016). Furthermore, Leucine-rich repeat kinase 2 (LRRK2) was shown to phosphorylate EndoA, which triggers pronounced membrane curvatures and thus enhances the recruitment of autophagy factors such as Atg3. Intriguingly, EndoA can also interact with Synaptojanin1 (Synj1), a lipid phosphatase enriched at presynaptic boutons and localizes to pre-autophagic membranes (Vanhauwaert et al., 2017). Synj1 has two phosphatase domains and the second one, termed SAC1, hydrolyses phosphoinositide phosphates (PIP) such as PI(3)P and PI(3,5)P2 that can be recognized by multiple autophagy proteins required for autophagosome formation. Therefore, these two proteins likely act in concert to regulate the initial stage of autophagosome biogenesis at presynaptic terminals. Finally, Piccolo and Bassoon, two core scaffold proteins of the active zone, were also reported to play a role in the regulation of synaptic autophagy (Okerlund et al., 2017). These two proteins act to restrain presynaptic autophagy, and dysregulated autophagy due to deficiency in these two proteins enhances the degradation of SVs. However, it is not yet clear whether Piccolo and Bassoon specifically regulate autophagy, or autophagy defects are the result of a broader neuronal stress response after the loss of these two crucial synaptic structural determinants. The mechanism underlying Bassoon-regulated synaptic autophagy is discussed in more detail below.

3.4. Autophagic cargoes at synapses

Multiple lines of evidence have demonstrated the presence of autophagosomes at presynaptic terminals as well as a supportive role of autophagy in maintaining the integrity and function of synapses. However, the molecular identity of cargoes and cargo receptors/adaptors of selective autophagy as well as the regulation in response to the changes of synaptic activity remain poorly understood. Synaptic vesicles (SVs) undergo several rounds of recycling at presynaptic boutons, posing a challenge for membrane sorting and high energy demand. Surprisingly, the mechanism underlying the turnover of SVs and SV proteins at synaptic sites is still largely unknown. Several studies have recently provided compelling evidence that SVs are the substrates of autophagy and the Rab26-dependent pathways control the sequestration of SVs into pre-autophagosomal membranes (Hernandez et al., 2012; Binotti et al., 2015; Luningschror et al., 2017; Hoffmann-Conaway et al., 2020). Rab proteins are small guanine triphosphatases (GTPases) that act as key regulators of intracellular membrane trafficking and accomplish their functions by switching between an inactive GDP-bound and an active GTP-bound form, which determines their ability to bind effectors. Rab26 is associated with SVs and is also highly enriched in large membrane-surrounded clusters concentrated with the autophagy machinery such as Atg16L1, Rab33B, and LC3-II (Binotti et al., 2015). The functionality of Rab26 relies on GTPase activator proteins guanosine diphosphate (GDP)/guanosine triphosphate (GTP) exchange factors (GEFs). Motor neurons lacking the GEF Plekhg5 exhibit deficiency in Rab26 activation. More importantly, SV proteins accumulate along with SV enlargement, suggesting the important roles of Rab26 and GEF Plekhg5 in the turnover of SVs through autophagy. Bassoon was recently found to regulate autophagy-mediated SV protein turnover (Hoffmann-Conaway et al., 2020). Loss of Bassoon function in neurons led to a decrease in SV protein levels coupled with a smaller SV pool size and a higher rate of SV protein turnover. Of note, despite the evidence at the ultrastructural levels showing the presence of SVs within autophagosomes (Hoffmann-Conaway et al., 2020), the vast majority of synaptic autophagosomes are devoid of SVs, raising the possibility that the autophagy-independent pathways may exist mediating the turnover of SVs. Indeed, the degradation of SVs and SV proteins via the endolysosomal system has been previously described (Andres-Alonso et al., 2021; Overhoff et al., 2020). SV proteins exhibit high heterogeneity (Cohen et al., 2017), which supports the view that, while the endolysosomal system might be efficient for removal of damaged SV proteins, autophagy-mediated degradation is likely activated under the conditions that a fast and bulk degradation of discrete SVs is on demand.

Mitochondria are the main cellular energy powerhouses that supply most of ATP by oxidative phosphorylation (OXPHOS), essential for neuronal function and survival (Cai and Tammineni, 2016; Chamberlain and Sheng, 2019; Feng et al., 2017; Sheng and Cai, 2012). Mitophagy, a selective form of autophagy, serves as a key quality control mechanism of mitochondria (Cai and Tammineni, 2016; Pickles et al., 2018; Sheng and Cai, 2012; Youle and Narendra, 2011; Cai and Jeong, 2020). PTEN-induced putative kinase protein 1 (PINK1)/Parkin-mediated mitophagy is the most heavily studied and the best-understood mitophagy pathway (Clark et al., 2006; Gautier et al., 2008; Narendra et al., 2008). In brief, loss of mitochondrial membrane potential (Δψ_m) accumulates PINK1 on the outer membrane of mitochondria (OMM) to recruit and activate Parkin, an E3 ubiquitin ligase, through phosphorylation of ubiquitin (Kane et al., 2014; Kawajiri et al., 2010; Kazlauskaite et al., 2014; Kazlauskaite et al., 2021; Koyano et al., 2014; Matsuda et al., 2010; Narendra et al., 2010; Shiba-Fukushima et al., 2012). Parkin then ubiquitinates several OMM proteins so that these ubiquitinated OMM proteins can be degraded via the UPS (Chan et al., 2011; Geisler et al., 2010; Poole et al., 2010; Yoshii et al., 2011; Ziviani et al., 2010). This triggers recruitment of the autophagy machinery to promote the engulfment of damaged mitochondria by phagophore/isolation membranes and thus the formation of mitophagosomes destined for elimination via the lysosomal system. We and others have shown some unique features of PINK1/Parkin-mediated mitophagy in neurons (Cai et al., 2010; Cai et al., 2012; Devireddy et al., 2015; Lee et al., 2019; Maday and Holzbaur, 2016; Sung et al., 2016; Tammineni et al., 2017; Xie et al., 2015; Ye et al., 2015). In particular, Parkin-targeted mitochondria primarily accumulate in the somatodendritic region where they undergo autophagic sequestration for lysosomal degradation. Such spatial aspects of Parkin-dependent mitophagy were also observed in vivo. The PINK1 and Parkin mutant Drosophila exhibit abnormal tubular and reticular mitochondria restricted to the cell body, as well as normal morphology with reduced mitochondrial flux within axons (Devireddy et al., 2015; Sung et al., 2016). In addition, the evidence from the examination of Purkinje neurons in the mito-QC reporter mice indicates that the majority of mitochondrial turnover occurs in the Purkinje somata. This supports the view that damaged mitochondria or mitophagosomes are returned to the soma for clearance (McWilliams et al., 2016). Collectively, these in vitro and in vivo observations suggest that the soma is in the focus of neuronal mitophagy, a selective process with a function to restrict damaged mitochondria to the soma and thus limit the impact of impaired mitochondrial function in distal axons.

Mitochondria-mediated ATP supply and Ca²⁺ buffering sustain various essential functions at synaptic sites (Manji et al., 2012; Sheng and Cai, 2012). Thus, synaptic mitochondria are particularly vulnerable to physiological insults, and defects in synaptic mitochondria have been linked to the early pathophysiology of neurodegenerative diseases. Given that Parkin-mediated mitophagy predominantly occurs in the soma, this raises the possibility that alternative mechanism(s) must function to efficiently remove constantly damaged mitochondria from nerve terminals to maintain synaptic mitochondrial homeostasis. In our recent study, we uncover a new mechanism of mitochondrial quality control in axons and at synaptic terminals, which Small GTPase Ras homolog enriched in brains (Rheb) initiates mitophagy and coordinates with dynein-Snapin-mediated retrograde transport, thereby regulating the integrity of synaptic mitochondria. Rheb facilitates autophagic recruitment of damaged mitochondria to form mitophagosomes in distal axons, whereas dynein-Snapin-mediated retrograde transport removes nascent mitophagosomes and thus reduces mitochondrial stress at synaptic terminals. Importantly, Rheb senses Δψ_m depolarization to initiate mitophagy in distal axons under both physiological and pathophysiological conditions. Activation of Rheb-associated mitophagy within axons requires the BCL-2 homology 3 (BH3)-containing protein NIP3-like X (NIX, also known as BNIP3L), a selective mitophagy receptor that facilitates recruitment of the autophagy machinery to the surface of damaged mitochondria, but is independent of the Parkin-mediated pathway in the soma of neurons (Han et al., 2020a; Han et al., 2020b). Given that PINK1/Parkin-mediated mitophagy predominantly occurs in the soma of neurons, our study provides the first indication that Rheb and Snapin act as key players in mitophagy-mediated quality control of mitochondria at synaptic sites.

4. Autophagy defects in neurodegenerative diseases

Neurodegenerative diseases are characterized by progressive degeneration and loss of neurons, structures, and functions of the nervous system. A prominent feature of these diseases is the abnormal accumulation of misfolded proteins, protein aggregates, or fibrils, suggesting a common mechanism associated with altered protein degradation pathways. Autophagy is physiologically important in neuronal health, which raises the possibility that autophagy dysfunction may play a critical role in neurodegenerative diseases. The autophagy pathway is complex, with multiple steps and modes of regulation. Thus, the absolute contribution of autophagy defects to the disease onset and progression has yet to be established. This neuron-to-neuron transmission of secreted pathologically misfolded proteins or amyloid has been proposed as the molecular basis of propagation of protein malconformation cytopathology and disease progression in neurogenerative diseases. Enormous interest in the autophagy-based unconventional secretion of amyloids from neurons has led to the rapid growth of new findings in this field. Here we summarize the current knowledge of autophagy regulation in major neurodegenerative diseases, including AD, PD, and ALS, and discuss the potential roles of autophagy deficiency in the pathophysiology of these diseases (Fig. 3).

Fig. 3.

Autophagy defects in neurodegenerative diseases. Massive accumulation of autophagosomes is a prominent feature in the patient brains of major neurodegenerative diseases. In Alzheimer’s disease (AD), impaired retrograde transport and lysosomal deficits lead to autophagy dysfunction. Of note, autophagic stress likely promotes autophagy-based unconventional seretion of Aβ and tau through which amphisomes fuse with the plasma membrane and trigger the extracellular release of Aβ and tau in AD and other tauopathy diseases. In Parkinson’s disease (PD), α-synuclein (α-syn) along with PD-causing mutations inhibits autophagosome biogenesis and axonal transport. In addition, autophagy-mediated extracellular secretion of α-syn likely participates in the transmission of α-synucleinopathy. In Amyotrophic Lateral Sclerosis (ALS), disease-linked mutations disrupt autophagosome formation and alter mitophagy, whereas lysosomal deficiency augments autophagy failure in ALS-related motor neurons.

4.1. Alzheimer’s disease

AD is the most common form of neurodegenerative disorder and a leading cause of dementia in aging populations. The disease progression involves cognitive decline, memory loss, and neuronal death in the cerebral cortex and subcortical regions. AD brains are characterized by abnormal accumulation of extracellular amyloid plaque deposits, composed of agglomerated Aβ peptides, as well as intracellular neurofibrillary tangles (NFTs), consisting of hyperphosphorylated tau protein (Long and Holtzman, 2019). The increase in amyloidogenic processing of amyloid precursor protein (APP), which leads to Aβ overproduction, is a key feature underlying the pathogenesis of AD (Haass et al., 2007; 99. Hardy and Selkoe, 2002; Jonsson et al., 2012]. Clear-cut evidence has indicated a link between alterations in the autophagy and endolysosomal systems and early AD pathophysiology (Nixon 2007; Nixon and Yang, 2011; Nixon, 2013). Aβ is a substrate of autophagy and the cellular levels of APP and Aβ were found to be partially regulated by autophagy (Tian et al., 2011). Decreased Beclin 1 levels due to enhanced caspase 3 cleavage was proposed to cause autophagy dysfunction in AD brains (Pickford et al., 2008; Small et al., 2005; Rohn et al., 2011). Notably, autophagy upregulation has been shown to decrease Aβ levels in different model systems (Boland et al., 2008; Spilman et al., 2010; Tian et al., 2011; Vingtdeux et al., 2011). However, using mouse models and cell culture, several studies have demonstrated that Aβ appears to be produced within autophagosomes where APP and Aβ-generating machinery are enriched (Jaegar et al., 2010; Nilsson et al., 2013; Pickford et al., 2008; Yu et al., 2005; Boland et al., 2008; Feng et al., 2017). Furthermore, autophagy deficiency in AD mouse brains with neuronal deletion of atg7 triggers the intracellular accumulation of Aβ, accompanied by a significant reduction in extracellular amyloid plaque burden (Nilsson et al., 2013). The findings support the notion that autophagy is involved in the regulation of Aβ secretion and amyloidogenesis. However, direct evidence showing autophagy-based Aβ secretion in AD neurons is still lacking and whether Aβ is released through autophagosomes or amphisomes remains undefined.

Induced pluripotent stem cells (iPSCs)-derived neurons and post-mortem brains of both familial and sporadic AD patients exhibit massive accumulation of AVs along with endosomes (Nixon, 2007; Nixon and Yang, 2011; Yu et al., 2005; Cataldo et al., 1997; Nixon et al., 2005; Cataldo et al., 2000; Israel et al., 2012), but the cause of such autophagosomal and endosomal proliferations in AD brains remains poorly understood. We and others have demonstrated that AVs are remarkably retained within the dystrophic neurites and synaptic terminals of AD brains (Nixon et al., 2005; Tammineni et al., 2017). Given the fact that retrograde transport mediates proper removal of autophagic cargoes from nerve terminals, this raises a fundamental question as to whether defective retrograde transport interrupts autophagic clearance and thus triggers autophagic stress at AD synapses. Indeed, our work has shown that AV retrograde transport is impaired in AD axons, which is consistent with robust autophagic accumulation at presynaptic terminals in the brains of AD patients and AD-related mutant human APP (hAPP) transgenic (Tg) mice (Tammineni et al., 2017a; Tammineni and Cai, 2017). Importantly, soluble Aβ42 oligomers enriched in axons interact with dynein motors. This interaction disrupts the coupling of the dynein motor with its adaptor Snapin, hampering the attachment of dynein motors to amphisomes. As a result, dynein-Snapin-driven retrograde transport of amphisomes is halted, thereby trapping amphisomes in distal axons and preventing their degradation within lysosomes in the soma of AD neurons (Tammineni et al., 2017a). In agreement with these findings, deletion of snapin in mice phenocopies AD-linked synaptic autophagic stress, whereas overexpression of Snapin decreases autophagic retention at AD synapses by enhancing AV retrograde transport. Furthermore, we found that β-site APP cleaving enzyme 1 (BACE1), a rate-limiting enzyme for APP amyloidogenic processing and Aβ generation, was concentrated within LEs/amphisomes. The dynein-Snapin transport machinery-loaded LEs/amphisomes facilitate BACE1 retrograde trafficking toward the soma for lysosomal degradation (Ye and Cai, 2014; Ye et al., 2017; Feng et al., 2017). Such a mechanism modulates BACE1 turnover and its β secretase activity, thereby controlling BACE1 cleavage of APP and Aβ production. In addition, a recent study reported that AP-2, originally proposed to mediate synaptic vesicle reformation and the retrograde transport of amphisomes containing BDNF/TrkB receptors (Kononenko et al., 2014; Kononenko et al., 2017), also played a role in the regulation of BACE1 trafficking and degradation (Bera et al., 2020). iPSC-derived neurons from patients with late-onset AD displayed a decrease in AP-2 levels. Moreover, mouse brains lacking AP2 showed BACE1 accumulation within LEs/amphisomes coupled with enhanced Aβ generation. Therefore, these studies consistently suggest that defective retrograde transport exacerbates autophagy dysfunction and impedes BACE1 trafficking toward somatic lysosomes for proper degradation, aggravating amyloid pathology in AD. It is conceivable that retrograde transport impairment-induced massive retention of AVs, especially amphisomes, may promote Aβ release at AD synapses.

In addition, previous studies have demonstrated that mutations in the PSEN1 gene encoding Presenilin 1, the most common cause of early-onset familial AD, disrupt lysosomal function and autophagy and markedly accelerate disease onset and neuropathological severity (Lee et al., 2010; Cataldo et al., 2004). Presenilin 1 functions as the catalytic subunit of γ-secretase, an intramembranous protease, which processes APP to generate Aβ following the prior cleavage by β-secretase. Importantly, Presenilin 1 was also reported to regulate lysosomal acidification and thus protease activation by controlling the assembly of lysosomal vATPase (Lee et al., 2010). These findings are the earliest demonstrations linking autophagy impairment in AD directly to AD-causing genes. Accordingly, one of the clearest examples of the rescue of autophagy dysfunction could be the reversal of Presenilin 1-driven lysosomal acidification deficits in affected AD neurons.

Abnormal accumulation of hyperphosphorylated tau proteins in the cytoplasm that leads to the formation of tau aggregates and fibrils is a pathogenic hallmark of tauopathy diseases, including AD. Autophagy stress and the presence of tau within AVs are characteristics in the brains of AD and other tauopathies (Piras et al., 2016). These tau-enriched AVs could be redirected for secretion through direct fusion with the plasma membrane. Importantly, the release of tau is enhanced upon autophagy induction by starvation or pharmacological agents or upon lysosomal deficits, but is reduced under the condition of autophagy inhibition (Chen et al., 2020; Kang et al., 2019; Lonati et al., 2018; Mohamed et al., 2014). Autophagy induction directs the fusion of LEs/MVBs with AVs to generate amphisomes (Fader et al., 2008). This raises the possibility that tau targeted for the extracellular release likely arrives in amphisomes through LEs/MVBs rather than autophagosomes. As for the extracellular vesicle (EV)-mediated secretion, a close link has been established between autophagy and other unconventional secretion pathways (Xu et al., 2018; Ganesan and Cai, 2021). Of note, recent studies highlighted the LC3-Dependent EV Loading and Secretion (LDELS) pathway by which the autophagy machinery participates in EV biogenesis and secretion (Leidal et al., 2020; Leidal and Debnath, 2020). Distinct from classical autophagy, this new secretory pathway employs components of the LC3 conjugation machinery, but not other Atgs essential for autophagosome biogenesis. Given a growing body of evidence showing the presence of misfolded proteins and protein aggregates including tau inside EVs (Ganesan and Cai, 2021), the discovery of LDELS undoubtedly builds an important foundation for rigorously scrutinizing how specific LC3-positive EV populations contribute to the pathologies of neurodegenerative diseases. Therefore, more studies are needed to advance our understanding of how tau proteins utilize AVs or EVs as a carrier in the process of secretion and whether such a mechanism plays a key role in driving pathogenic tau propagation and pathology in tauopathy diseases.

Earlier studies revealed altered mitophagy in AD patient brains, as evidenced by mitochondrial retention within AVs in the soma of vulnerable neurons (Hirai et al., 2001; Moreira et al., 2007a, b). Mitophagy abnormalities along with impaired mitochondrial turnover in AD have been further demonstrated in a number of recent studies (Han et al., 2020; Han et al., 2021; Cummins et al., 2019; Du et al., 2017; Fang et al., 2019; Manczak et al., 2018; Reddy et al., 2018; Ye et al., 2015). Moreover, mitophagy stimulation was shown to mitigate AD pathology and rescue memory impairment in AD models (Du et al., 2017; Fang et al., 2019). Among the mitophagy pathways, PINK1/Parkin-mediated mitophagy has been the focus of current studies in AD. We have shown that the Parkin pathway is robustly activated upon progressive accumulation of Aβ and mitochondrial damage in human patient brains and animal models of AD (Ye et al., 2015). Furthermore, cytosolic Parkin is depleted in AD brains over disease progression, leading to mitophagy failure and augmentation in mitochondrial pathology. In line with our work, another study reported that Parkin was diminished in the AD patient-derived skin fibroblasts and brain biopsies, which is coupled with abnormal PINK1 accumulation (Martin-Maestro et al., 2016). Mitophagy was restored in these cells with overexpression of Parkin, as reflected by decreased PINK1 and the recovery of Δψ_m along with reduced retention of defective mitochondria. Together, these results consistently indicate that impaired mitochondrial function and abnormal accumulation of defective mitochondria could be caused by impaired PINK1/Parkin-mediated mitophagy in AD neurons. In addition, our recent studies revealed that synaptic accumulation of mitophagosomes is a feature in AD-related mutant hAPP mouse brains, which is caused by increased initiation of Rheb-mediated mitophagy coupled with impaired removal of mitophagosomes from AD synapses due to defects in retrograde transport of mitophagosomes. Furthermore, while deficiency in dynein-Snapin-mediated retrograde transport recapitulates synaptic mitophagy stress and induces synaptic degeneration, elevated Snapin expression attenuates mitochondrial defects and ameliorates synapse loss in AD mouse brains (Han et al., 2020a; Han et al., 2020b). Besides, the degradation capacity of lysosomes is required for mitophagic clearance, and lysosomal defects can also trigger mitophagy failure. The lysosomal deficit has been implicated as one of the main cellular defects in AD (Nixon 2007; Nixon and Yang, 2011; Nixon, 2013). We proposed that protease deficit within lysosomes impedes lysosomal proteolysis of damaged mitochondria along with other autophagic cargoes in AD neurons (Tammineni et al., 2017b). Collectively, these findings suggest that defective PINK1/Parkin-mediated mitophagy, impaired retrograde transport, and lysosomal deficit lead to mitophagy deficiency, thereby driving mitochondrial dysfunction and synaptic degeneration in AD. Of note, our recent work further uncovered that mitophagy defects induce metabolic disruption associated with early disease stages of AD (Han et al., 2021). Importantly, lysosomal enhancement in AD neurons reverses impaired metabolic function by promoting the elimination of damaged mitochondria, thus protecting against synapse loss in AD mouse brains. Together, our studies have provided multiple lines of evidence showing mitophagy failure in AD neurons and also suggest how such a defect exacerbates mitochondrial deficits and augments synaptic deterioration associated with AD.

Impaired mitophagy function was also indicated under tauopathy conditions. Recent work found a stable association of an NH₂-htau fragment (a 20-22 kDa NH₂-truncated human tau fragment mapping between 26 and 230 amino acids of the longest human tau isoform) with Parkin and Ubiquitin-C-terminal hydrolase L1 (UCHL-1) in cellular and animal AD models and human AD brains, leading to enhanced mitochondrial recruitment of Parkin and UCHL-1 and thus improper mitochondrial turnover (Corsetti et al., 2015). Mitophagy suppression can restore synaptic mitochondrial density and protect against neuronal death induced by this NH₂-htau. In contrast to this work, human tau (htau) was shown to be inserted into the mitochondrial membrane and thus disrupt mitophagy (Hu et al., 2016). However, in a more recent study, both htau and human P301L mutant tau (htauP301L), a variant of human tau protein that forms neurofibrillary tangles in the brains of patients and mouse models, were shown to impair mitophagy in C. elegans and neuroblastoma cells by reducing Parkin translocation onto mitochondria through a different mechanism. Parkin association with mitochondria was impaired, which was caused by tau-mediated sequestration of Parkin in the cytosol, but not by changes in the Δψ_m or the cytoskeleton (Cummins et al., 2019). Thus, these data suggest different mechanisms leading to alterations in PINK1/Parkin-mediated mitophagy under tauopathy conditions. Finally, a recent study revealed a marked decrease in the basal level of mitophagy in postmortem hippocampal tissues from AD patients, cortical neurons derived from AD-iPSC, as well as AD mouse models (Fang et al., 2019). This study further demonstrated defects in the activation of ULK1 and TANK Binding Kinase 1 (TBK1), the autophagy proteins that mediate autophagy/mitophagy initiation, resulting in impaired mitophagy function. More importantly, pharmacological stimulation of mitophagy mitigates amyloid and tau pathologies, leading to beneficial effects against memory loss in AD mice. Therefore, these observations support the notion that mitophagy failure is likely an early event in AD brains and plays a causative role in the development of AD-linked neuropathology (Fang et al., 2019). Further studies using neurons derived from iPSCs of sporadic AD or other similar models can be very helpful to address whether mitophagy deficiency serves as a key player in driving Aβ/tau proteinopathies.

4.2. Parkinson’s disease

PD is the second most common neurodegenerative disease and is caused by progressive degeneration of dopaminergic neurons in the substantia nigra pars compacta (Lotharius and Brundin 2002, Parkkinen 2011), and by the formation of a major component of abnormal neuronal aggregates known as Lewy bodies (LBs)—α-syn inclusions (Spillantini, et al., 1997; Jucker and Walker 2013). α-syn, an acidic lipid-binding protein, is strongly enriched at presynaptic terminals and localizes to SVs, regulating the SV release (Burré et al., 2010; Bendor et al., 2013). During SV endocytosis, α-syn was proposed to act before dynamin to modulate early steps of SV formation (Ben Gedalya et al., 2009; Bendor et al., 2013; Vargas KJ et al., 2014). The genetic evidence suggests a critical link between elevated α-syn levels and the pathogenesis of both familial and sporadic PD. α-syn can be present in many different conformations, such as monomers, tetramers, high-level oligomers, fibrils, and aggregates. Both oligomers and fibrils of α-syn were proposed to be toxic and can sufficiently induce dopaminergic loss (Wong and Krainc, 2017). Aberrant autophagy function and autophagosome accumulation were observed in PD patient brains (Nixon 2013). α-syn-mediated toxicity induces cellular dysfunction, including defects in the autophagy-lysosomal system. Also, several forms of PD have been linked to genes encoding endocytosis and autophagy, and dysfunctions in both pathways have been implicated in the PD pathophysiology (Abeliovich and Gitler, 2016; Menzies et al., 2015; Vidyadhara et al., 2019). PD-linked genes encoding proteins with the function of neuronal autophagy are represented by LRRK2, SYNJ1, VPS35 (Zavodszky E et al., 2014; Zimprich et al., 2011), atg5 (Chen et al., 2013), PRKN, and PINK1 (Kitada et al., 1998; Valente et al., 2004).

Several studies have shown that α-syn is the substrate of the autophagy pathway and can be degraded through autophagy activity (Vogiatzi et al., 2008; Webb et al., 2003). However, α-syn aggregates have been indicated to suppress autophagy function (Sarkar et al., 2007; Winslow et al., 2010; Tanik et al., 2013). Following incubation of neuronal models with α-syn preformed fibrils (pffs), one study found that autophagosome formation was not affected, but autophagosome maturation and fusion with lysosomes for subsequent degradation were halted, resulting in an impaired turnover of autophagic cargoes (Tanik et al., 2013). α-syn aggregates-induced autophagy failure was also proposed to be caused by impaired axonal transport of autophagosomes. Interestingly, instead of a non-specific blockage of axons, the α-syn inclusions can specifically inhibit the trafficking of autophagic vesicles (Volpicelli-Daley et al., 2014). Moreover, overexpression of α-syn was found to disrupt the autophagic transmembrane protein Atg9 trafficking from ER to Golgi and thus reduces the formation of omegasome, a precursor for autophagosome biogenesis (Winslow et al., 2010). Autophagic clearance relies on lysosomal degradation capacity. Studies from α-syn triplication PD iPSCs revealed reductions in the activity of multiple lysosomal enzymes, including cathepsin B, GCase, β-galactosidase, and hexosaminidase, as a result of impairment in the ER-to-Golgi trafficking (Mazzulli et al., 2011; Mazzulli, et al., 2016; Wong and Krainc, 2016). Thus, lysosomal defects contribute to autophagy dysfunction associated with α-synucleinopathy. Furthermore, α-syn has been documented to undergo autophagy-based unconventional secretion from cultured nerve cells (Alverez-Erviti et al., 2011; Danzer et al., 2012; Ejlerskov et al., 2013; Emmanouilidou et al., 2010 Hazegawa et al., 2011; Jang et al., 2010; Lee et al., 2005). Interneuronal transmission of endogenously produced and secreted α-syn has been demonstrated both in vitro and in vivo (Alverez-Erviti et al., 2011; Desplas et al., 2009; Hansen et al., 2011; Volpicelli-Daley et al., 2011; Li et al., 2008; Luk et al., 2012). Upon inhibition of the autophagy-lysosomal pathway, α-syn can be released and transferred in the forms of monomer and aggregates to the extracellular surroundings via amphisome-like structures (Ejjerskov et al., 2013; Lindersson et al., 2012). However, more work is still needed to elucidate detailed mechanisms underlying autophagy-dependent secretion of α-syn and its relevance to the disease progression in PD.

As discussed above, several classical endocytic proteins have been recently identified as the regulators of synaptic autophagy such as SYNJ1 and EndoA, critical for maintaining synaptic protein homeostasis and functional synapses (Soukup et al., 2018). In SYNJ1 R258Q patient-derived dopaminergic neurons, altered Synj1 function was shown to trigger defects in autophagosome maturation at presynaptic terminals as a result of the retention of Atg18a, an autophagy-related protein (Vanhauwaer et al., 2017). Such a defect led to dopaminergic neuron loss in the SYNJ1 R258Q Drosophila model. Strikingly, mice carrying SYNJ1 R258Q mutation recapitulated PD-like dysfunctions, including defects in motor performance and axonal degeneration of the nigrostriatal pathway (Cao et al., 2017). EndoA was found to play a role in synapse autophagy. In particular, EndoA is involved in the recruitment of Atg3 to autophagic vesicles following LRRK2-dependent phosphorylation (Soukup et al., 2016). LRRK2 mutations are believed to cause PD by interrupting a large range of cellular processes, including the autophagy pathway (Gómez-Suaga et al., 2014; Rivero-Ríos et al., 2019). A recent study revealed that hyperactive LRRK2 kinase activity alters autophagy by disrupting the axonal transport of autophagosomes (Boecker et al., 2021). Additionally, LRRK2 null mice exhibit age-dependent alterations in the autophagy-lysosomal pathway (Tong et al., 2010). Mutations in VPS35, encoding the retromer complex, can cause PD-associated neurodegeneration by impairing autophagy due to its critical role in recruiting the WASH complex to endosomes and sorting Atg9 (Zavodszky et al., 2014). Finally, a heterozygous variant in the promoter region of Atg5, reported to enhance its transcriptional activity, was identified in one patient with sporadic PD (Chen et al., 2013). Whether this allele affects autophagy is currently unknown.

Mitochondrial and metabolic disruptions are a feature in PD neurons (Gao et al., 2017), and such defects have been proposed to be attributed to dysfunctional mitophagy. Many PD-causing genes are related to the mitochondrial phenotypes (Burchell et al., 2013; Chinta et al., 2010; Chung et al., 2016; Clark et al., 2006; Hsieh et al., 2016; Lindstrom et al., 2017; Pukass et al., 2015; Sherer et al., 2002; Zhou et al., 2015) In addition, PD patients have increased rates of mtDNA deletion in the substantia nigra, which further associates defective quality control of mitochondria with PD (Bender et al., 2006). The important role of mitophagy in PD was first identified from an ultrastructural study showing autophagic accumulation of mitochondria in neurons of the patients with PD and Lewy Body Dementia (LBD) (Zhu et al., 2003). Many other studies revealed mitophagy abnormalities in a variety of experimental models representing genetic forms of toxic-environmental PD (Chu et al., 2013; Dagda et al., 2009; Dagda et al., 2008; Osellame and Duchen, 2013; Zhu et al., 2007). Cell-based and mechanistic studies directly link PINK1 and Parkin to mitophagy. However, while loss-of-function mutations in PARK6 (encoding PINK1) and PARK2 (encoding Parkin E3 ubiquitin ligase) genes are linked to familial PD (Chu, 2019), the in vivo role of PINK1/Parkin-mediated mitophagy remains elusive. The PINK1 and Parkin pathway has been extensively studied in Drosophila. Mutant flies show dopaminergic degeneration, reduced lifespan, and locomotive defects (Cha et al., 2005; Clark et al., 2006; Greene et al., 2003; Park et al., 2006; Whitworth et al., 2005). Muscle cells of mutant flies exhibited swollen mitochondria with disrupted cristae, coupled with muscle degeneration (Cornelissen et al., 2018; Greene et al., 2003; Pesah et al., 2004; Yang et al., 2006). PARK6 KO rats showed dopaminergic loss and motor defects (Dave et al., 2014). Importantly, both Drosophila and rat model systems show mitochondrial dysfunction. However, mice with the deletion of PARK6 or PARK2 failed to exhibit robust substantial PD-relevant phenotypes (Akundi et al., 2011; Goldberg et al., 2003; Kitada et al., 2007; Perez and Palmiter, 2005). This might be due to compensations by other mechanisms that are sufficient to maintain neuronal homeostasis under physiological conditions. The accumulation of fragmented mitochondria with abnormal internal structures was recently observed in dopaminergic neurons from aged Parkin KO mice (Noda et al., 2020). Strikingly, when crossing PARK2 KO mice with Mutator mice characterized by accelerated acquisition of mitochondrial DNA (mtDNA) mutations, the resulting phenotypes showed mitochondrial defects as well as dopaminergic neuronal loss (Pickrell et al., 2015). Thus, these findings suggest that Parkin-mutant mice are susceptible to increased mtDNA damage. Given that both impaired mitochondrial function and mitophagy deficit are upstream of neurodegeneration, the lack of robust phenotypes in mice suggest that the PINK1/Parkin pathway might be dispensable under physiological conditions, yet still necessary in response to stress/pathological stimuli for the functional maintenance and survival of PD-related dopaminergic neurons. Other studies suggested additional roles for this pathway including in the regulation of endocytic protein complex formation via Parkin-mediated ubiquitination of Endophilin, Dynamin1, and SYNJ1 (Cao et al., 2014; Trempe et al., 2009), as well as the promotion of the removal of synaptic vesicle proteins by autophagy (Hoffmann-Conaway S et al., 2020). Aside from PINK1/Parkin-associated mitophagy, increased cardiolipin-mediated mitophagy was proposed to play a role in α-syn-induced mitochondrial dysfunction (Ryan et al., 2018).

4.3. Amyotrophic lateral sclerosis

ALS is characterized by progressive degeneration of motor neurons in the brain and the spinal cord, leading to muscle weakness, atrophy, and paralysis (Zarei et al., 2015). ALS is most commonly a sporadic disease, but approximately 5%-10% of disease cases are familial. The pathological hallmark of ALS is the relocalization of TDP43 from the nucleus to the cytoplasm and cytoplasmic neuronal and glial inclusions of TDP-43 are found in 98% of all cases of ALS (sporadic and familial) (Scotter et al., 2015; Brettschneider et al., 2013; Feneberg et al., 2018; Dong and Chen 2018). Mutations in TDP-43 are accountable for about 3% of familial ALS cases and less than 1% of sporadic ALS cases (Zou et al., 2017). Several studies have demonstrated that TDP-43 can be degraded by both autophagy and the UPS. Treatment with either autophagy inhibitor 3-MA or UPS inhibitor MG132 increased the protein levels of TDP-43, whereas administration of the autophagy inducer reduced TDP43 levels and rescued the ALS-related pathology (Wand et al., 2010; Wang et al., 2012; Cheng et al., 2015; Barmada et al., 2014). It should be noted that autophagy was shown to be particularly important for the clearance of oligomeric and aggregated TDP-43 (Scotter et al., 2014; Watabe et al., 2014). In Drosophila and mouse models of TDP-43 proteinopathy, rapamycin-induced autophagy enhanced the clearance of TDP-43 aggregates while lessening ALS-like symptoms (Wang et al., 2012; Cheng et al., 2015). Moreover, TDP-43 and its inclusions can impact autophagy beyond being autophagic cargos. Some mutations have been reported to directly disrupt autophagy initiation. Mutations in the C-terminal domain of TDP-43 inhibited ER-Golgi vesicle transport by depleting the vesicle-trafficking regulator Rab1, a small GTPase required for phagophore formation (Soo et al., 2015; Zoppino et al., 2010). On the other hand, TDP-43 has been indicated to regulate autophagy. The loss of TDP-43 function inhibits the expression of positive regulators of autophagy. TDP-43 ablation was shown to down-regulate the expression of the essential autophagic factor Atg7, thus resulting in autophagy impairment (Bose et al., 2011). Depletion of TDP-43 was reported to decrease Atg4B transcript levels, possibly through cryptic exon inclusion, leading to alterations in the autophagy process (Torres et al., 2018). Conversely, overexpression of both WT and mutant TDP-43 (A315T) in SH-SY5Y cells increased autophagic activity and inhibition of autophagy resulted in increased cell death, particularly in TDP-43 (A315T) cells, indicating that autophagy is cytoprotective in these mutants (Wang et al., 2015). Collectively, it is clear that TDP-43 loss of function triggers a complex array of gene expression dysregulations that lead to various abnormalities of the autophagy pathway. Thus, the initial evidence highlights a critical role for TDP-43 in the regulation of autophagy, supporting the idea that this process, which, among others, is important for the clearance of cytoplasmic aggregates, may be impaired in TDP-43-ALS.

Mutations in SOD1 account for approximately 20% of familial ALS cases and 1% of sporadic ALS cases (Rosen et al., 1993; Kaur et al., 2016; Pasinelli and Brown, 2006). The mechanisms by which mutant SOD1 causes ALS are not fully understood. Many studies suggest that the misfolding or aggregation of mutant SOD1 is a key event in the pathogenic process of ALS. Importantly, inhibition of autophagy was shown to result in the accumulation of insoluble SOD1 aggregates in motor neurons (Kabuta et al., 2006), raising the possibility that autophagy dysregulation is involved in ALS pathophysiology associated with mutant SOD1. Several studies provided direct evidence that mutant SOD1 aggregation affects multiple early steps in autophagy. Mutant SOD1 models have been shown to exhibit either enhanced initiation of autophagy or a failure of late-stage fusion steps with LEs/lysosomes in autophagy (Hetz et al., 2009; Bandyopadhyay et al., 2014; Xie et al., 2015). Given the important role of autophagy in the clearance of misfolded proteins, promoting autophagy in mutant SOD1 models has been thought to be beneficial. To date, however, the assessment of the impact of autophagy enhancement in SOD1G93A ALS mice has revealed conflicting results. While treatment with autophagy inducer lithium carbonate delayed disease progression and prolonged lifespan of this mouse model (Feng et al., 2008; Fornai et al., 2008), a later study showed worsened neuropathology in the SOD1G93A ALS mice administered with lithium (Pizzasegola et al., 2009). Autophagy enhancers trehalose and resveratrol led to delayed disease onset, prolonged survival, and diminished motor neuron loss (Castillo et al., 2013; Zhang et al., 2014; Mancuso et al., 2014). Conversely, treatment with another autophagy enhancer rapamycin was shown to exacerbate motor neuron degeneration, coupled with rapid disease progression and shortened lifespan in the SOD1G93A ALS mice (Zhang et al., 2011). It is possible that overaction of autophagy in SOD1-related ALS may generate toxic fragments of SOD1 or augment lysosomal defects associated with SOD1 mutations (Forloni et al., 2016; Xie et al., 2015). On the other hand, autophagy inhibition by n-butylidenephtalide (n-BP) was reported to yield beneficial effects against motor neuron loss in the SOD1G93A mouse model by preventing the aberrant accumulation of autophagosomes in motor neurons (Hsueh et al., 2016; Zhou et al., 2017). In a recent study, autophagy inhibition in SOD1G93A mice with selective deletion of the essential autophagy gene atg7 in motor neurons induced an earlier disease onset, but slowed disease progression and extended the lifespan of mice (Rudnick et al., 2017), suggesting that autophagy plays a neuroprotective role at the early disease stage but have a deleterious impact at an advanced disease stage. Therefore, autophagy perturbation related to mutant SOD1 varies depending on the phases of disease development, which complicates the association between mutant SOD1 and autophagic dysregulation. Further studies are required to resolve the precise effects of autophagy regulation in mutant SOD1-linked ALS pathology.

Mounting evidence has shown that autophagy can be assisted by specific chaperones and this type of autophagy has been named chaperone-assisted selective autophagy (CASA) (Arndt et al., 2010; Rusmini et al., 2017). This pathway involves a specific chaperone of the family of the small HSPs, known as HSPB8 that dimerizes and forms a stable complex with its co-chapterone BAG3, a nucleotide exchange factor for HSP70s (Rusmini et al., 2017). HSPB8 and BAG3 interact with HSP70 and C terminus of Hsc70-interacting protein (CHIP) and thus the HSPB8-BAG3-HSP70-CHIP complex was termed CASA complex (Arndt et al., 2010). While HSPB8 recognizes misfolded proteins that can be ubiquitinated by CHIP, BAG3 promotes the interaction of the complex with dynein motors for the delivery of misfolded proteins to the microtubule organizing center (MTOC) where the autophagy receptor SQSTM1/p62 sequesters the cargo within autophagosomes (Arndt et al., 2010; Cristofani et al., 2017; Rusmini et al., 2017). Importantly, mutations in HSPB8 or BAG3 have been linked to motor neuron or neuromuscular diseases (Adriaenssens et al., 2017; Bouhy et al., 2018; Ghaoui et al., 2016; Fang et al., 2017; Selcen et al., 2009). A drastic increase of HSPB8 was indicated in the spinal cord motor neurons of ALS patients and mouse models at the end disease stage (Crippa et al., 2010a; Crippa et al., 2010b; Anagnostou et al., 2010). Furthermore, a potent anti-aggregant activity of HSPB8 via the CASA has been described in the cellular models of neurodegenerative diseases involving different neuropathogenic proteins, including Aβ, α-syn, TDP-43, SOD1, and dipeptide repeats coded by chromosome 9 open reading frame 72 (C9ORF72) (Bruinsma et al., 2011; Crippa et al., 2016a; Crippa et al., 2010b, Wilhelmus et al., 2006). Down-regulation of HSPB8 augmented the accumulation of mutant proteins such as SOD1, TDP-43, C9ORF72, supporting its role in the clearance of misfolded proteins (Crippa, et al., 2016b; Crippa et al., 2010b; Cristofani et al., 2017). In addition, increasing the expression of the fly functional ortholog of HSPB8 (HSP67Bc) in two Drosophila models of ALS prevented the mislocalization of a neurotoxic mutant TDP-43 (Ritson et al., 2010). Conversely, HSP67Bc down-regulation led to TDP-43 and polyubiquitinated proteins accumulation and worsened the eye phenotype of mutant TDP-43 flies (Crippa et al., 2016b). Interestingly, functional silencing of HSPB8 in mice failed to display any major phenotypes (Bouhy et al., 2018), suggesting that the activity of HSPB8 via the CASA complex is likely relevant to the pathological conditions associated with abnormal accumulation of misfolded proteins.

Several ALS-associated genes act as autophagy receptors, like SQSTM1/p62 (Fecto et al., 2011), optineurin (OPTN) (Maruyama et al., 2010), and Ubiquilin 2 (Maruyama et al., 2012). These receptors mediate the sequestration of autophagic cargoes within autophagosomes through interaction with LC3. However, ALS-causing mutations in the LC3 binding region of SQSTM1/p62 were shown to halt its recruitment to autophagosomes, resulting in impaired degradation of SOD1 mutant (Gal et al., 2009; Goode et al., 2016). Mutations in OPTN led to reduced autophagy, defects in protein clearance, as well as disrupted OPTN-mediated LC3 recruitment to damaged mitochondria needed for mitophagy activation (Shen et al., 2015; Wong and Holzbaur, 2014b). Notably, one study reported the role of OPTN in the regulation of the trafficking of autophagosomes through binding LC3 and Myosin VI (Tumbarello DA et al., 2012). The majority of ALS-associated mutations in OPTN are located in the myosin VI-binding domain and thus show defects in this binding (Shen et al., 2015; Sundaramoorthy et al., 2015). In addition, mutations in TBK1 were recently linked to ALS (Cirulli et al., 2015; Freischmidt et al., 2015). TBK1 was found to phosphorylate OPTN and thus promote mitophagy (Moore and Holzbaur, 2016). Importantly, silencing of TBK1 or ALS-related TBK1 variants disrupts the recruitment of OPTN and LC3 to defective mitochondria, thereby compromising mitophagy (Moore and Holzbaur, 2016; Richter et al., 2016; Lazarou et al., 2015). Besides, TBK1-mediated phosphorylation of SQSTM1/p62 allows autophagosome maturation (Pilli et al., 2012). Neuronal deletion of TBK1 in mice (driven by the Nestin promoter) disrupts autophagy and triggers multiple neuropathological alterations, including cortical synapse loss and motor dysfunctions (Duan et al., 2019). These findings collectively suggest a critical role of TBK1 in the regulation of neuronal autophagy. Finally, TBK1 was found to regulate microtubule dynamics in non-neuronal cells (Pillai et al., 2015), raising the possibility that this kinase might also play a role in the regulation of autophagosome trafficking in neurons. However, this model has not been investigated yet.

In accord with a central role for ubiquilins in autophagy, Ubiquilin 2 was shown to modulate MTORC1 activity, lysosomal acidification, and, in turn, autophagosome maturation (Senturk et al., 2019; Wu et al., 2020). Rare mutations in Ubiquilin 2 were identified to be associated with familial forms of ALS and ALS-FTD (Deng 2011; Milecamps et al., 2012; Daoud et al., 2012). Overexpression of these ALS-linked Ubiquilin 2 mutants interrupts its interaction with Atg9-Atg16L1 (Osaka et al., 2016), inducing accumulation of autophagy substrates such as SQSTM1 and LC3-II and intracellular inclusions along with motor defects in vitro and in vivo (Ceballos-Diaz et al., 2015; Picher-Martel et al., 2015; Wu et al., 2015). Additionally, a hexanucleotide repeat expansion in the ALS-causing C9ORF72 gene (DeJesus-Hernandez et al., 2011; Renton et al., 2011) was reported to be involved in endocytic vesicle trafficking and autophagy likely through forming a complex with Smith-Magenis syndrome chromosome region, candidate 8 (SMCR8) and WD repeat domain 41 (WDR41) (Farg et al., 2014; Sellier et al., 2016; Sullivan et al., 2016). This complex serves as a GEF to activate Rab8 and Rab39 that regulate autophagosome biogenesis and interact with the autophagy receptors SQSTM1/p62 and OPTN (Pilli et al., 2012; Seto et al., 2013; Sellier et al., 2016). C9ORF72 was also found to interact with ULK1, a key kinase in the control of autophagosome formation, and this interaction mediates the translocation of the ULK1 autophagy initiation complex to the phagophore via Rab1a (Webster et al., 2016). Finally, ALS-associated Alsin (ALS2), a Rab GEF to activate Rab5, has been shown to regulate autophagy (Yang et al., 2001; Topp et al., 2004; Otomo et al., 2011; Ravikumar et al., 2008). A mutant form of ALS2 was found to disrupt Rab5 activation and trigger defects in the formation of amphisomes in this familial form of ALS (Laird et al., 2008). Together, understanding the role of these ALS-associated genes in autophagy is beginning to provide an important link to the mechanism underlying autophagy deficiency and its contributions to the pathologies of ALS.

Impaired mitophagy was proposed to be involved in the denervation of neuromuscular junctions in an ALS mouse model (Rogers et al., 2017). A recent study uncovers that Parkin-mediated mitophagy is activated in the SOD1G93A Tg mouse model of ALS (Palomo et al., 2018). Given that mitophagy induction triggers Parkin-activated and the UPS-mediated degradation of mitochondrial dynamics proteins, Mitofusin-2 (Mfn2) and Mitochondrial Rho GTPase (Miro1) (Bingol et al., 2014; Birsa et al., 2014; Chan et al., 2011; Liu et al., 2012; Wang et al., 2011), increased mitophagy in the spinal cord of the mutant SOD1 mice is coupled with the depletion of Parkin along with mitochondrial dynamics proteins Mfn2 and Miro1. Interestingly, genetic ablation of PARK2 protects against muscle denervation and motor neuron loss and mitigates the depletion of mitochondrial dynamics proteins, which delays disease progression and prolongs life span in mutant SOD1 mice (Palomo et al., 2018). These results suggest that Parkin could be a disease modifier of ALS, and chronic activation of Parkin-dependent mitophagy augments mitochondrial dysfunction by depleting mitochondrial dynamics proteins. In agreement with this study, several other studies also reported a significant reduction of Miro1 in spinal cord tissue of ALS patients and animal models (Zhang et al., 2015). Moreover, ALS-linked mutant SOD1-associated Miro1 decrease is Parkin-dependent (Moller et al., 2017). Miro1 is known as a component of the adaptor-motor complex essential for Kinesin-related protein 5 (KIF5) motors to drive anterograde transport of mitochondria in axons (Devine and Kittler, 2018). Miro1-knockout mice exhibit a marked decrease in mitochondrial distribution within distal axons accompanied by motor neuron degeneration (Nguyen et al., 2014). Thus, ALS-linked mitochondrial trafficking defect is likely attributed to Miro1 deficiency due to enhanced Miro1 turnover caused by robust induction of Parkin-mediated mitophagy (Moller et al., 2017).

As discussed above, some ALS-associated genes, including OPTN, SQSTM1/p62, and TBK1, are known to play a critical role in mitophagy/autophagy induction (Fecto et al., 2011; Freischmidt et al., 2015; Maruyama et al., 2010). However, it remains poorly understood regarding the roles of the mutations in these genes in the pathogenesis of ALS. Given the phosphorylation of OPTN and SQSTM1/p62 by TBK1 to activate autophagy/mitophagy, aberrant accumulation of misfolded proteins and protein aggregates along with impaired mitochondrial turnover may both contribute to ALS-linked mitochondrial dysfunction and motor neuron death. Illuminating the role of these proteins in vivo will be pivotal in dissecting the cellular mechanisms leading to ALS-linked axonal degeneration and motor neuron loss. Additionally, lysosomal defects have been implicated in ALS. Recent studies have provided strong evidence that lysosomal deficits play a critical role in autophagy/mitophagy failure and mitochondrial pathology in SOD1G93A Tg mice (Xie et al., 2015). Impaired lysosomal degradation result in abnormal mitophagosome retention in motor neuron axons of mutant SOD1 mice. More importantly, rescuing lysosomal deficits was shown to enhance mitochondrial turnover, improve motor neuron survival, and ameliorate disease phenotype in this mutant SOD1 mouse model. Given that autophagy/mitophagy is a lysosome-dependent pathway, defective mitophagy/autophagy and mitochondrial dysfunction in ALS are partially attributed to defects in lysosomal proteolysis.

5. Concluding Remarks

Autophagy is a conserved and homeostatic mechanism by which dysfunctional cellular components, misfolded and aggregation-prone proteins, as well as damaged mitochondria at nerve terminals are engulfed and sequestered within autophagosomes. Once an autophagosome is produced distally, it rapidly fuses with an MVB/LE and then undergoes rapid and efficient retrograde movement along the axon toward the soma for lysosomal degradation. Even though we have a broad outline of this pathway, our understanding of the molecular and cellular mechanisms underlying autophagy regulation and its impact on neuronal health needs to be further expanded. Many questions remain: What signals modulate the formation of constitutive autophagosomes at synaptic sites? Does this pathway contribute to the maintenance of synaptic integrity and function? Whether and how neuronal autophagy is affected under the conditions of cellular stress and during aging?

Alterations in the function of autophagy proteins and autophagy deficits at the diverse stages of this pathway have been described in neurodegenerative diseases. Whether these relate to their canonical functions in protein and organelle recycling and degradation in most cases is unclear. Current data suggest complex mechanisms underlying autophagy regulation in neurodegenerative diseases. Some diseases display the disruption of autophagy induction but no significant impact on the clearance aspect of the autophagy pathway (Bordi et al., 2019), whereas autophagy is robustly activated but the clearance is progressively impaired in some other diseases (Xie et al., 2015; Bordi et al., 2016; Tammineni et al., 2017a). However, whether the proliferation of autophagosomes detected in patient brains represents a possible “rescue mechanism” or participates in the development of disease pathologies and whether autophagy defects trigger neurodegeneration are poorly understood. Future studies dealing with the pathophysiology of neurodegenerative diseases should be directed at the understanding of neuron-specific functions of autophagy and testing how precisely genetic alterations in disease-linked genes interfere with autophagy.

Importantly, autophagy is a promising target mechanism from a therapeutic perspective and can promote the removal of the primary toxic entity causing disease, targeting such diseases at their roots. However, the current knowledge remains very limited concerning whether and how autophagy deficits contribute to the pathophysiology of various neurodegenerative diseases and whether autophagy in neurons can be modulated therapeutically. We need to identify the pathways involved in autophagy regulation and determine whether and how they can be manipulated in vivo. Finally, the availability of small-molecule effectors will allow us to further determine whether autophagy can be tuned to make neurons more resilient to the cellular stressors of aging, environmental toxins, and genetic risk factors of diseases, offering some hope for the future. Therefore, further detailed studies to advance our current understanding of autophagy mechanisms will tremendously benefit the development of therapeutic strategies to combat these diseases.

Abbreviations

Aβ

Amyloid β

AD

Alzheimer’s disease

α-syn

α-synuclein

ALS

Amyotrophic lateral sclerosis

ALS2

ALS-associated Alsin

APP

Amyloid precursor protein

AV

autophagic vacuole

BACE1

β-site APP cleaving enzyme 1

BNIP3L

BCL-2 homology 3 (BH3)-containing protein NIP3-like X (NIX)

C9ORF72

chromosome 9 open reading frame 72

CASA

chaperone-assisted selective autophagy

CHIP

C terminus of Hsc70-interacting protein

DFCP1

double FYVE-containing protein 1

DIC

dynein intermediate chain

DRG

dorsal root ganglion

EndoA

endophilin-A

ER

endoplasmic reticulum

ERAD

ER-associated protein degradation

EV

extracellular vesicle

FTD

frontotemporal dementia

GDP

Guanosine diphosphate

GEF

guanine nucleotide exchange factor

GTP

guanosine triphosphate

GTPases

guanine triphosphatases

HAP1

Huntingtin-associated protein 1

hAPP

human amyloid polypeptide

HD

Huntington’s disease

hTau

human tau

iPSCs

induced pluripotent stem cells

JIP1

c-Jun N-terminal kinase–interacting protein 1

KIF5

Kinesin-related protein 5

LB

Lewy body

LBD

Lewy Body Dementia

LC3

microtubule-associated proteins 1A/1B light chain 3

LDELS

LC3-Dependent EV Loading and Secretion

LE

late endosome

LRRK2

leucine-rich repeat kinase 2

Mfn2

Mitofusin-2

Miro1

mitochondrial Rho GTPase

mtDNA

mitochondrial DNA

MTOC

microtubule organizing center

mTOR

mammalian target of rapamycin

MVB

multivesicular body

n-BP

n-butylidenephtalide

OMM

outer membrane of mitochondria

OPTN

optineurin

OXPHOS

oxidative phosphorylation

PD

Parkinson’s disease

PE

phosphatidylethanolamine

Pff

preformed fibril

PI3K

phosphatidlylinositol (PtIns) 3-kinase

PINK1

PTEN-induced putative kinase protein 1

PIP

phosphoinositide phosphates

PKA

protein kinase A

Rheb

Ras homolog enriched in brains

SMCR8

Smith-Magenis syndrome chromosome region, candidate 8

SOD1

superoxide dismutase 1

STRIPAK

striatin-interacting phosphatase and kinase

SV

synaptic vesicle

SYNJ1

synaptojanin1

TBK1

TANK Binding Kinase 1

TC

transport carrier

TDP-43

TAR-DNA-binding protein of 43 kDa

TFEB

transcription factor EB

Tg

transgenic

TGN

Trans-Golgi network

UCHL1

ubiquitin-C-terminal hydrolase L1

ULK1

Unc-51-like autophagy activating kinase 1

UPR

unfolded protein response

UPS

ubiquitin-proteasome system

WDR41

WD repeat domain 41

XBP1

X-Box Binding Protein 1