Presynaptic protein homeostasis and neuronal function

Available online at www.sciencedirect.com

ScienceDirect Presynaptic protein homeostasis and neuronal function Yu-Chun Wang1,2,*, Elsa Lauwers1,2,* and Patrik Verstreken1,2 Proteome integrity is maintained by a coordinated network of molecular chaperones, by protein degradation machineries and by their regulators. Numerous human pathologies are considered as diseases of compromised protein homeostasis (proteostasis), including neurodegeneration. These are characterized by the accumulation of neuronal protein aggregates and by synaptic defects followed by loss of connectivity and cell death. While this suggests that synaptic terminals are particularly sensitive to proteostasis imbalance, our understanding of protein turnover mechanisms and regulation at the synapse remains limited. Recent reports show that different proteolytic pathways act at synapses, including several forms of autophagy. The role of chaperones in controlling the balance between synaptic protein refolding and degradation and how this complex network regulates neuronal function also begins to be unraveled.

nature of neurons and their long lifetimes, potentially for the entire life of the organism, imply that the presynaptic proteome accumulates stress over long periods of time. Given the polarized morphology of neurons, in many respect synapses function as semi-autonomous entities and this is also true in terms of protein quality control. In this manuscript we review the current knowledge on mechanisms that act at the presynapse to prevent protein damage and that have the goal to repair or degrade dysfunctional protein assemblies. In this review we focus on how these processes affect presynaptic activity in healthy neurons, but it is clear that when deregulated, these mechanisms are central to neurodegenerative disease as well [3,4].

Addresses 1 KU Leuven, Department of Neurosciences, Leuven Institute for Neurodegenerative Disease (LIND), Herestraat 49, bus 602, 3000 Leuven, Belgium 2 VIB Center for Brain & Disease Research, Herestraat 49, bus 602, 3000 Leuven, Belgium

Neurotransmission relies on a complex cascade of temporally and spatially controlled protein–protein interactions [5,6] (Figure 1a). During biological activity-induced conformational changes, interactive surfaces of proteins become exposed to the intracellular environment. Molecular chaperones act to prevent or correct unwanted interactions that may occur between these surfaces [7] (Figure 1b). By doing so, chaperones can regulate protein function, avoid or reverse the formation of non-functional protein assemblies and influence the decision to target client proteins for degradation.

Corresponding authors: Lauwers, Elsa ([email protected]), Verstreken, Patrik ([email protected]) * Equal contribution. Current Opinion in Genetics & Development 2017, 44:38–46 This review comes from a themed issue on Molecular and genetic bases of disease Edited by Nancy Bonini, Edward Lee and Wilma Wasco

http://dx.doi.org/10.1016/j.gde.2017.01.015 0959-437X/ã 2017 Elsevier Ltd. All rights reserved.

Introduction Synapses, and in particular the contacts sites between presynaptic and postsynaptic neurons, contain an extremely dense networks of proteins [1]. At presynaptic terminals these proteins function in a highly coordinated and dynamic manner to support neuronal communication, allowing neurotransmitters to be released within 1 ms upon arrival of an action potential and with a frequency that sometimes exceeds 100 Hz. This intense activity places presynaptic proteins at a high risk of misfolding and molecular damage [2]. Moreover the postmitotic Current Opinion in Genetics & Development 2017, 44:38–46

Chaperones watch over the presynaptic proteome

Maintenance of the synaptic vesicle (SV) exocytic machinery requires the action of the Cystein string protein – heat shock cognate 70 – Small glutamine-rich tetratricopeptide repeat-containing protein (CSP-Hsc70-SGT) chaperone complex. This complex binds the plasma membrane R-SNARE protein SNAP-25 in its monomeric form through the Hsc70 chaperone and refolds this client into a SNARE-complex assembly competent state [8] (Figure 1a and c). This activity is indispensable to maintain neurotransmitter release at high frequency during the entire life of the organism, and lack of the CSP co-chaperone leads to presynaptic defects and ultimately neurodegeneration both in Drosophila and in mouse [9–12]. Trans-SNARE complex assembly is also promoted by Alpha-synuclein, and expression of this abundant presynaptic protein can reverse the lethal neurodegeneration observed in Dnajc5 (encoding CSP) knockout mice [13,14]. Alpha-synuclein acts as a SNARE chaperone by simultaneously binding to the vesicular Q-SNARE VAMP-2 and to phospholipids, and this function gains importance with aging and upon increased neuronal www.sciencedirect.com

Presynaptic protein homeostasis Wang, Lauwers and Verstreken 39

Figure 1

(a)

NSF Syntaxin-1A SNAP-25

synaptic cleft VAMP-2

Hsc70 NMNAT

Sgt

CSP

Bruchpilot (ELKS/CAST) active zone

α-synuclein

synaptic vesicle

(b)

(c)

aggregate

disaggregase unfolded

partially folded

increases neurotransmitter release

trans-SNARE complex

native state

foldase

foldase

unfoldase

unfoldase interactive surfaces

α-synuclein Hsc70 CSP

stimulated by neuronal activity

monomeric SNAREs

NSF cis-SNARE complex

Current Opinion in Genetics & Development

Presynaptic chaperones promote neurotransmission by keeping active zone proteins properly folded. (a) Key active zone proteins known to rely on chaperones (dark purple) and co-chaperones (light purple) to maintain their normal function. See the text for details. (b) Chaperones are best known for their ability to act as foldases that help proteins reach their native folding state. Some chaperones also prevent protein aggregation by ‘holding’ unfolded client proteins and in some cases even reverse aggregation. Importantly chaperones can also act as unfoldases, an activity that is critical, for example during cycles of protein–protein binding such as seen during the SNARE assembly/disassembly cycle depicted in (c). All three chaperones activities promote SV exocytosis and counteract use-dependent protein damage. (c) Neurotransmitter release involves the formation of membrane-bridging trans-SNARE complexes between the SV Q-SNARE VAMP2 and the plasma membrane R-SNAREs Syntaxin-1 and SNAP-25. trans-SNARE complexes formation is promoted by the overlapping chaperone activities of the CSP-Hsc70-Sgt complex and of Alpha-synuclein. After membrane fusion and collapse of the SV into the plasma membrane, cis-SNARE complexes must be disassembled to allow endocytic recycling of VAMP-2 and another round of neurotransmitter release. This disassembly is catalyzed by the NSF AAA+ ATPase in an ATP-dependent manner.

activity [15]. After fusion of SVs with the plasma membrane, highly stable cis-SNARE complexes are disassembled by the AAA+ ATPase N-ethylmaleimidesensitive fusion factor (NSF) [16] (Figure 1c).

nicotinamide adenine dinucleotide (NAD) synthesis [20,21]. NMNAT directly binds to Bruchpilot in an activity-dependent manner and protects it against ubiquitylation and proteasomal degradation [18] (Figure 1a).

Other presynaptic proteins beside SNAREs also require chaperones to protect them from activity-dependent damage. Lack of the neuronal maintenance factor NMNAT (nicotinamide mononucleotide adenylyltransferase) [17] leads to reduced levels of several presynaptic markers including CSP itself, Synaptotagmin 1 and the active zone protein Bruchpilot [18]. NMNAT, which was identified in a genetic screen for Drosophila factors involved in synaptic function [19], functions as a bona fide chaperone independently of its enzymatic function in

To date we only know of a few examples where chaperones directly control the activity of presynaptic proteins. Yet the classical dogma that proteins must adopt and retain a particular tridimensional structure to be functional no longer holds true, and protein folding is now instead recognized as an extremely dynamic process [22]. In this context chaperones undoubtedly play a central role in protecting the presynaptic proteome from aggregation and in facilitating changes in folding status during biological activity (Figure 1).

www.sciencedirect.com

Current Opinion in Genetics & Development 2017, 44:38–46

40 Molecular and genetic bases of disease

The ubiquitin-proteasome system turns over misfolded and superfluous presynaptic proteins

of the UPS leads to increased spontaneous and evoked neurotransmitter release in different model synapses, indicating that at least some components of the presynaptic release machinery are normally targeted via this pathway [27–29]. Several presynaptic targets of the UPS have been reported in mature neurons, including the active zone proteins Unc13 [27], Liprin-alpha [30,31], RIM 1 [28] and Bruchpilot [18] (Figure 2A). The E3 ubiquitin ligases (Figure 2B) that modify Unc13 and Bruchpilot remain to be identified. Levels of Liprin-alpha are regulated by the anaphase promoting complex (APC),

The ubiquitin-proteasome system (UPS) is one of the main proteolytic pathways that targets misfolded or unnecessary proteins [23], thereby avoiding their potential cytotoxic effect and recycling their amino acid constituents (Figure 2). While the UPS plays a well-documented role in synapse formation and in postsynaptic plasticity mechanisms [24–26], it is also present in mature presynaptic terminals. Acute pharmacological inhibition Figure 2

(a)

BRE1B Syntaxin-1A SNAP-25 CASK VAMP-2

Liprin-alpha

Synaptophysin

APC Bruchpilot (ELKS/CAST) active zone

SIAH1/2

synaptic cleft

Unc-13 RIM1

synaptic vesicle

SCRAPPER

(b)

Ub E1

increases neurotransmitter release

E1

degradation

E3

E2

binding unfolding deubiquitylation

E2 E3

S E2

S E3

S ubiquitylation

S DUB

S

proteasome

polyubiquitylation

deubiquitylation

stimulated by neuronal activity Current Opinion in Genetics & Development

The ubiquitin-proteasome system targets numerous presynaptic proteins. (a) Active zone proteins known to be ubiquitylated and whose levels are affected by proteasomal activity. E3 ubiquitin ligases that target these proteins are shown in orange. See the text for details. (b) Ubiquitin (Ub, red) is conjugated to substrate proteins by a cascade of activation (E1), conjugation (E2) and ligase (E3) enzymes. Deubiquitylating enzymes (DUBs) can reverse this reaction and release free Ub in the cytosol. Substrates carrying a chain of Ub molecules linked via the lysine 48 of the previous Ub monomer are recognized as substrates by the 26S proteasome. These substrates are deubiquitylated, unfolfed by proteasomal ATPases (see Figure 4), translocated through the proteasome channel and degraded. By removing misfolded or dysfunctional proteins, the UPS contributes to maintain a healthy pool of presynaptic proteins and thus promotes neurotransmitter release. Conversely, DUB activity in the nerve terminal is stimulated by neuronal activity, suggesting that under conditions of intense usage presynaptic proteins are protected against degradation. Current Opinion in Genetics & Development 2017, 44:38–46

www.sciencedirect.com

Presynaptic protein homeostasis Wang, Lauwers and Verstreken 41

a multisubunit E3 ligase that is well known for its role in the degradation of mitotic cyclins [30]. In rat hippocampal neurons, Liprin-alpha-2 is degraded by the UPS and its levels scale with synaptic activity [31]. Liprin-alpha-2 was found to control the size of the readily releasable pool of SVs by recruiting the release machinery components RIM 1 and CASK to the active zone [31]. RIM 1 itself is ubiquitylated by an E3 ligase complex containing the F-box protein SCRAPPER. Hippocampal neurons from Scrapper knockout mice show an increased frequency of spontaneous neurotransmitter release that is dependent on RIM 1 expression, indicating that the effects of this E3 ligase are largely exerted via RIM1 turnover [28]. The turnover of transmembrane and membraneassociated presynaptic proteins can also be regulated by ubiquitylation. The plasma membrane R-SNARE Syntaxin-1A has been reported to be a substrate of the E3 ubiquitin ligase BRE1B (Staring) [32], and ubiquitylated forms of Syntaxin-1A, VAMP-2 and SNAP-25 can be detected in mouse brain extracts [8]. The SV protein Synaptophysin interacts with and is ubiquitylated by the SIAH1and SIAH2 E3 ligases [33]. The stability of SNAP-25 and Synaptophysin has been shown to increase in the presence of proteasome inhibitors [8,33], yet it is still unclear whether these proteins are truly degraded via the UPS or if proteasome inhibition indirectly affects their degradation, for example, by reducing the levels of free ubiquitin and thereby preventing their degradation in the lysosome [34]. The UPS notably targets unfolded or misfolded proteins [8]. Additionally, it appears to have a general role in controlling the stability of presynaptic proteins and thereby in modulating neurotransmission, and UPS activity responds to neuronal activity. In rat brain synaptosomes, the overall level of protein ubiquitylation is reduced upon stimulation by depolarization [35] (Figure 2B). This effect requires calcium entry into the terminals and cannot be explained by accelerated proteasomal degradation of ubiquitylated substrates, as it is already observed after 15 s of depolarization and even in the presence of a proteasome inhibitor. While the underlying mechanism remains to be determined, this work indicates that deubiquitylating enzymes are very active at the presynapse and that protein ubiquitylation is dynamically regulated in response to neuronal activity.

Lysosomal degradation rejuvenates presynaptic protein pools Lysosomal degradation is the other main proteolytic system found in all eukaryotic cells (Figure 3). Lysosomes are membrane-limited organelles, implying that substrates must somehow cross this membrane to become accessible to lysosomal proteases. One of the pathways that allow this is macroautophagy, where substrates are sequestered in a compartment termed phagophore which www.sciencedirect.com

matures into a double-membrane autophagosome that ultimately fuses with a late endosome or lysosome [36] (Figure 3). In this case substrates can be misfolded protein, protein aggregates and dysfunctional organelles, or even non-specific (and therefore not necessarily dysfunctional) portions of the cytoplasm. Autophagosomes form at presynaptic and postsynaptic terminals, and macroautophagy is known to be required during synaptogenesis and to participate in synaptic function [37]. SVs themselves can be targeted to early autophagosomes by the small GTPase Rab26 [38]. This is consistent with the observation that inhibition of the mTOR signaling pathway, the main blocker of autophagy induction [39], leads to a reduced number of SVs and reduces evoked neurotransmitter release in mouse dopaminergic neurons [40]. The current model is thus that autophagy acts presynaptically as a break on neurotransmission. Conversely, increased presynaptic activity, by means of electrical stimulation or activation of the heat-sensitive channel TrpA1, leads to an accumulation of Atg8-positive autophagic structures at the Drosophila neuromuscular junction [41]. There is also evidence that fusion of SVs with the plasma membrane and of autophagosomes with lysosomes might be regulated in a similar manner. Both events indeed require the activity of the same voltagegated calcium channel, of which the pore-forming subunit is encoded by cacophony in fruit flies and by Cacna1a in mouse [42]. In addition to macroautophagy, different types of selective or non-selective autophagy have been described [43]. One of these pathways, termed endosomal microautophagy, was recently reported to take place at presynaptic boutons of the Drosophila neuromuscular junction [44] (Figure 3). This pathway relies on the Hsc70-4 chaperone to recruit cytosolic proteins carrying a KFERQ peptide motif [45] to the endosome, and to deform the endosomal membranes so these client proteins are engulfed into of intraluminal vesicles during the formation of multivesicular bodies [44,46]. The net result is a rejuvenation of the synaptic pool of proteins with a KFERQ motif, which is associated with an increased size of the readily releasable pool of SVs [44]. In mammalian neurons the Hsc704 homologue HSPA8 can also target the KFERQ-containing protein Alpha-synuclein for degradation via chaperone-mediated autophagy (CMA), a pathway where client proteins are directly translocated through the lysosomal membrane by LAMP-2 [47] (Figure 3). Given that about 60% of presynaptic cytosolic proteins carry a KFERQ motif [44], these two endosomal microautophagy and CMA are likely to be important regulators of synaptic strength. Transmembrane proteins can also be delivered to the lysosome by multivesicular bodies. These proteins are sorted into intraluminal vesicles after their delivery to the early endosome either directly from the Golgi complex or Current Opinion in Genetics & Development 2017, 44:38–46

42 Molecular and genetic bases of disease

Figure 3

VGCC active zone

EndoA

synaptic cleft

AP-2 (a) endocytosis

VAMP-2

PICALM synaptic vesicle

stimulated by neuronal activity

TBC1D24 TBC1D5 VAMP-2 ESCRT-0

Rab35 Atg9

Rab5

endosome / MVB Syntaxin-6

Rab26

ESCRT-I ESCRT-II

(c) endosomal

ESCRT-III

microautophagy EndoA

Hsc70

Atg8

ESCRT-0/I/II/III increases neurotransmitter release

autophagosome (b) macroautophagy

VGCC stimulated by neuronal activity

(d)

chaperonemediated autophagy Hsc70

lysosome

LAMP-2 Current Opinion in Genetics & Development

Different pathways converging on the lysosome degrade soluble and membrane-associated presynaptic proteins. (a) SV exocytosis is compensated by the reuptake of SV material from the plasma membrane, mainly via clathrin-mediated endocytosis. Newly formed SVs can either directly integrate the different SV pools or transit via an endosome. This pathway is stimulated by the GTPase Rab35, and inhibited by the GTPase activating protein TBC1D24/Skywalker. From the endosomal membrane, SV proteins can be recycled back or sorted into intraluminal vesicles (ILVs) by the ESCRT machinery. After fusion of the resulting multivesicular body (MVB) with the lysosome, ILVs and their content are degraded by lysosomal hydrolases. (b) Alternatively, SVs can be degraded via macroautophagy. SVs are therefore clustered by the GTPase Rab26 into a preautophagosomal structure (PAS). After PAS elongation and the formation of a double-membrane autophagosome, SVs and cytosol components are delivered to the lysosomal lumen. It is still unclear how this pathway is regulated at the presynapse, but increasing neuronal activity was recently shown to stimulate autophagosome formation. Note that synaptic autophagosomes and MVBs are known to be transported back to the cell soma where they fuse with lysosomes, yet the possibility that local lysosomal degradation also accurs at the nerve terminal cannot be excluded. (c) Soluble presynaptic proteins carrying a KFERQ peptide motif can enter MVBs or (d) directly reach the lysosomal lumen via a LAMP-2 channel. Both pathways require client protein recognition and delivery by the Hsc70/HSPA8 chaperone, which also contributes to endosomal membrane deformation to generate ILVs. Overall, lysosomal degradation results in a rejuvenation of presynaptic protein pools that increases neurotransmitter release.

after their endocytosis from the plasma membrane [48]. Several endosomal markers, including the early endosome GTPase Rab5 and the endosomal SNARE Syntaxin-6 are present on SVs [49], and vesicles from the readily releasable pool were reported to rapidly cycle via endosomes as part of the normal SV cycle [50,51]. This Current Opinion in Genetics & Development 2017, 44:38–46

sorting step may allow plasma membrane proteins to be segregated away from SV components and thus to maintain organelle identity [51] but also to send non-functional SV proteins for lysosomal degradation. In Drosophila the GTPase-activating protein Skywalker (encoded by TBC1D24 in mammals) inhibits Rab35-dependent www.sciencedirect.com

Presynaptic protein homeostasis Wang, Lauwers and Verstreken 43

Figure 4

(a)

LYSOSOMAL DEGRADATION

macroautophagy

endocytosis

endosome LC3 Atg9

AP-2 PICALM PAS

AP-2 PICALM EndoA FBOX32

lysosome

MVB

EndoA

SV Atg9

FBOX32 p62 Ubiquilin

eMiA

NSF

CMA

PRESYNAPTIC PROTEIN HOMEOSTASIS

Ub

native protein substrate unfolding

(b)

UBIQUITINPROTEASOME SYSTEM

Hsc70

NMNAT

Hs

unfolded unprotein

c7

0

CHAPERONES (c)

Current Opinion in Genetics & Development

An integrated network controls presynaptic protein homeostasis. (a) Different pathways leading to lysosomal proteolysis show extensive overlap. This is true in terms of the compartments they use, besides the lysosome which is the final destination in all cases. For example, endocytosis and endosomal microatophagy converge at MVBs, and SVs can cycle via endosomes and/or serve as a source of membrane for autophagosome formation. The endocytic and autophagy machineries also share common factors, like Endophilin A (EndoA) that deforms membrane at forming endocytic vesicles and at the PAS. The endocytic adaptors AP-2 and PICALM also connect endosomal structures with the PAS via LC3/Atg8. (b) Ub not only acts as destruction tag for the proteasome, it is also a sorting signal for transmembrane proteins to enter the MVB pathway. The deubiquitylating activity of the proteasome prevents Ub from being degraded along with substrate proteins, thereby maintaining a sufficient pool of free Ub. Ub is also a signal for some proteins to be degraded via macroautophagy and some Ub receptors such as p62 and Ubiquilin can deliver substrates to both the 26S proteasome and to forming autophagosomes. The balance between these two proteolytic pathways appears to be regulated by EndoA through a mechanism that involves its binding to the FBOX32 E3 Ub ligase. (c) Proteins that fail to reach or maintain proper folding are typically degraded by the UPS, and the Hsc70 chaperone itself delivers some client proteins to the 26S proteasome. Chaperones contribute to unfold substrates so they can be processed by the proteasome, which additionally possess an intrinsic ‘unfoldase’ activity and in some cases also degrade chaperones, as shown in the case of NMNAT. Finally, the Hsc70/HSPA8 is also involved in lysosomal proteolysis through the recruitment of KFERQ-containing proteins for endosomal microautophagy and chaperone-mediated autophagy. We are only starting to unravel the complexity of the presynaptic proteostasis network. As there is increasing interest to target this network for therapeutical intervention against neurodegenerative diseases, we now need to fully understand how it controls presynaptic protein function and how it responds to neuronal activity. www.sciencedirect.com

Current Opinion in Genetics & Development 2017, 44:38–46

44 Molecular and genetic bases of disease

transport of SVs to the endosome [52]. In sky null mutants, turnover of SV proteins is accelerated and this again correlates with a larger readily releasable pool [52,53]. Rab35-dependent degradation of a subset of SV proteins is also observed in rat hippocampal neurons, and this pathway was recently shown to be stimulated in response to neuronal activity [54] (Figure 3). Collectively, these different pathways leading to lysosomal degradation seem to be upregulated in response to presynaptic activity. This likely reflects a need to compensate for a usage-dependent decline in presynaptic protein functionality, and results in a global rejuvenation of the presynaptic pool of proteins leading to a corresponding increase in the size of the readily releasable pool.

Future directions Components of the proteostasis network function in a highly integrated manner. This is well illustrated by chaperones like Hsc70, which play a dual function either in protein refolding [8] or proteolysis via the UPS [55], endosomal microautophagy [44,46], chaperone-mediated autophagy [56] or chaperone-assisted selective autophagy [57]. The levels of chaperones themselves can be regulated by the UPS [58]. There is also an important cross-talk between the UPS and macroautophagy. Several factors that shuttle ubiquitylated client proteins to the proteasome, including Ubiquilin and p62, are also involved in autophagosome-maturation and fusion with lysosomes, and both pathways are under the influence of mTOR signaling [59]. Finally, macroautophagy and endocytosis are also inter-dependent. The endocytic factors AP2 and PICALM, which acts as a complex in clathrin-mediated endocytosis (CME), directly interact with the ubiquitin-like protein LC3 (the mammalian homologue of Atg8) that is required for autophagosome-formation. This interaction facilitates the fusion of autophagosomes with endosomes and by doing so stimulates lysosomal degradation of APP Cterminal fragments that are precursors of Ab peptides [60]. Moreover, while it is still unknown what the source of membrane for the formation of autophagosomes is at the synapse, the plasma membrane could contribute in a way that relies on AP2 [61]. SVs themselves might be involved since the transmembrane protein Atg9 that organizes the phagophore assembly site (PAS) is found in these organelles [62]. During autophagy, Atg9 is rerouted from endosomes to the PAS in a retromer-dependent manner [63], suggesting that SV endosomal cycling might be required for autophagy induction. On the other hand endocytosis does not appear to be acutely required during autophagosome formation, since Atg8-positive punctae still form at presynaptic terminals in a Drosophila temperature-sensitive dynamin mutant (shits1) at the restrictive temperature [41]. Interestingly, one of the factors that connect SV endocytosis and macroautophagy Current Opinion in Genetics & Development 2017, 44:38–46

is Endophilin-A, which upon phosphorylation by the LRRK2 kinase can switch from its well-studied role in CME to a recently described function in recruiting Atg3 to forming autophagosomes at the presynapse [41,64]. In the absence of Endophilin-A ubiquitylated proteins accumulate in mouse brains as a result of proteasome saturation, which might be at least partially explained by an upregulation of the E3 ubiquitin ligase F-box only protein 32 (FBOX32) [64]. Endophilin-A and FBOX32 directly bind to each other. Both protein induce membrane curvature and are needed for autophagosome formation in the mouse brain, suggesting that this complex regulates the balance between macroautophagy and the UPS in mammalian neurons [64]. In order to understand how the presynaptic proteome is maintained in a functional, healthy state, it will be critical to decipher how chaperones, the UPS and lysosomal degradation pathways interconnect and influence each other in a compartment-specific manner specifically at the synapse (Figure 4). This knowledge will be particularly valuable in a context where therapeutical interventions targeting the proteostasis network in neurodegenerative diseases are becoming increasingly popular [65–68].

Acknowledgements Work in the authors’ laboratory is supported by KU Leuven (CREA, GOA), Opening the Future, ERC CoG, FWO, the Hercules Foundation, BELSPO, a Methusalem grant of the Flemish government, and VIB.

References and recommended reading Papers of particular interest, published within the period of review, have been highlighted as:  of special interest  of outstanding interest Wilhelm BG, Mandad S, Truckenbrodt S, Kro¨hnert K, Scha¨fer C, Rammner B, Koo SJ, Claßen GA, Krauss M, Haucke V et al.: Composition of isolated synaptic boutons reveals the amounts of vesicle trafficking proteins. Science 2014, 344:1023-1028. The authors provide a 3D-quantitative model of a ‘model synaptic bouton’, displaying 300,000 proteins with atomic detail. This model illustrates how crowded presynaptic terminals are.

1. 

2.

Ellis RJ: Macromolecular crowding: obvious but underappreciated. Trends Biochem Sci 2001, 26:597-604.

3.

Neefjes J, van der Kant R: Stuck in traffic: an emerging theme in diseases of the nervous system. Trends Neurosci 2014, 37:66-76.

4.

Bayer TA: Proteinopathies, a core concept for understanding and ultimately treating degenerative disorders? Eur Neuropsychopharmacol 2015, 25:713-724.

5.

Jahn R, Fasshauer D: Molecular machines governing exocytosis of synaptic vesicles. Nature 2012, 490:201-207.

6.

Kononenko NL, Puchkov D, Classen GA, Walter AM, Pechstein A, Sawade L, Kaempf N, Trimbuch T, Lorenz D, Rosenmund C et al.: Clathrin/AP-2 mediate synaptic vesicle reformation from endosome-like vacuoles but are not essential for membrane retrieval at central synapses. Neuron 2014, 82:981-988.

7.

Ellis RJ: Protein aggregation: opposing effects of chaperones and crowding. In Folding for the Synapse. Edited by Wyttenbach A, O’Connor V. US: Springer; 2011:9-34. www.sciencedirect.com

Presynaptic protein homeostasis Wang, Lauwers and Verstreken 45

8.

Sharma M, Burre´ J, Su¨dhof TC: CSPa promotes SNAREcomplex assembly by chaperoning SNAP-25 during synaptic activity. Nat. Cell Biol 2011, 13:30-39.

9.

Umbach JA, Zinsmaier KE, Eberle KK, Buchner E, Benzer S, Gundersen CB: Presynaptic dysfunction in drosophila csp mutants. Neuron 1994, 13:899-907.

10. Zinsmaier KE, Eberle KK, Buchner E, Walter N, Benzer S: Paralysis and early death in cystein string protein mutants of Drosophila. Science 1994, 263:977-980. 11. Ferna´ndez-Chaco´n R, Wo¨lfel M, Nishimune H, Tabares L, Schmitz F, Castellano-Mun˜oz M, Rosenmund C, Montesinos ML, Sanes JR, Schneggenburger R et al.: The synaptic vesicle protein CSP alpha prevents presynaptic degeneration. Neuron 2004, 42:237-251. 12. Schmitz F, Tabares L, Khimich D, Strenzke N, de la Villa-Polo P, Castellano-Mun˜oz M, Bulankina A, Moser T, Ferna´ndez-Chaco´n R, Su¨dhof TC: CSPalpha-deficiency causes massive and rapid photoreceptor degeneration. Proc Natl Acad Sci U S A 2006, 103:2926-2931. 13. Bonini NM, Giasson BI: Snaring the function of alpha-synuclein. Cell 2005, 123:359-361. 14. Chandra S, Gallardo G, Ferna´ndez-Chaco´n R, Schlu¨ter OM, Su¨dhof TC: Alpha-synuclein cooperates with CSPalpha in preventing neurodegeneration. Cell 2005, 123:383-396. 15. Burre´ J, Sharma M, Tsetsenis T, Buchman V, Etherton MR, Su¨dhof TC: Alpha-synuclein promotes SNARE-complex assembly in vivo and in vitro. Science 2010, 329:1663-1667. 16. Baker RW, Hughson FM: Chaperoning SNARE assembly and disassembly. Nat Rev Mol Cell Biol 2016, 17:465-479. 17. Ali YO, Li-Kroeger D, Bellen HJ, Zhai RG, Lu H-C: NMNATs, evolutionarily conserved neuronal maintenance factors. Trends Neurosci 2013, 36:632-640. 18. Zang S, Ali YO, Ruan K, Zhai RG: Nicotinamide mononucleotide adenylyltransferase (NMNAT) maintains active zone structure by stabilizing bruchpilot. EMBO Rep 2012, 14:87-94. 19. Zhai RG, Cao Y, Hiesinger PR, Zhou Y, Mehta SQ, Schulze KL, Verstreken P, Bellen HJ: Drosophila NMNAT maintains neural integrity independent of its NAD synthesis activity. PLoS Biol 2006, 4:e416. 20. Zhai RG, Zhang F, Hiesinger PR, Cao Y, Haueter CM, Bellen HJ: NAD synthase NMNAT acts as a chaperone to protect against neurodegeneration. Nature 2008, 452:887-891. 21. Ali YO, Allen HM, Yu L, Li-Kroeger D, Bakhshizadehmahmoudi D, Hatcher A, McCabe C, Xu J, Bjorklund N, Taglialatela G et al.: NMNAT2:HSP90 complex mediates proteostasis in proteinopathies. PLoS Biol 2016, 14:e1002472. 22. Jeng W, Lee S, Sung N, Lee J, Tsai FTF: Molecular chaperones: guardians of the proteome in normal and disease states. F1000Research 2015, 4:1-10. 23. Hershko A, Ciechanover A: The ubiquitin system. Annu Rev Biochem 1998, 67:425-479. 24. Haas KF, Broadie K: Roles of ubiquitination at the synapse. Biochim Biophys Acta Gene Regul Mech 2008, 1779:495-506. 25. Siddoway B, Hou H: Molecular mechanisms of homeostatic synaptic downscaling. Neuropharmacology 2014, 78:38-44. 26. Ehlers MD: Activity level controls postsynaptic composition and signaling via the ubiquitin-proteasome system. Nat Neurosci 2003, 6:231-242. 27. Speese SD, Trotta N, Rodesch CK, Aravamudan B, Broadie K: The ubiquitin proteasome system acutely regulates presynaptic protein turnover and synaptic efficacy. Curr Biol 2003, 13:899-910. 28. Yao I, Takagi H, Ageta H, Kahyo T, Sato S, Hatanaka K, Fukuda Y, Chiba T, Morone N, Yuasa S et al.: SCRAPPER-dependent ubiquitination of active zone protein RIM1 regulates synaptic vesicle release. Cell 2007, 130:943-957. www.sciencedirect.com

29. Willeumier K, Pulst SM, Schweizer FE: Proteasome inhibition triggers activity-dependent increase in the size of the recycling vesicle pool in cultured hippocampal neurons. J Neurosci 2006, 26:11333-11341. 30. Van Roessel P, Elliott DA, Robinson IM, Prokop A, Brand AH: Independent regulation of synaptic size and activity by the anaphase-promoting complex. Cell 2004, 119:707-718. 31. Spangler SA, Schmitz SK, Kevenaar JT, De Graaff E, De Wit H, Demmers J, Toonen RF, Hoogenraad CC: Liprin-alpha2 promotes the presynaptic recruitment and turnover of RIM1/CASK to facilitate synaptic transmission. J. Cell Biol. 2013, 201:915-928. 32. Chin LS, Vavalle JP, Li A: Staring, a novel E3 ubiquitin-protein ligase that targets syntaxin 1 for degradation. J Biol Chem 2002, 277:35071-35079. 33. Wheeler TC, Chin LS, Li Y, Roudabush FL, Li A: Regulation of synaptophysin degradation by mammalian homologues of Seven in Absentia. J Biol Chem 2002, 277:10273-10282. 34. Piper RC, Lehner PJ: Endosomal transport via ubiquitination. Trends Cell Biol 2011, 21:647-655. 35. Chen H, Polo S, Di Fiore PP, De Camilli PV: Rapid Ca2 +-dependent decrease of protein ubiquitination at synapses. Proc Natl Acad Sci U S A 2003, 100:14908-14913. 36. Ariosa AR, Klionsky DJ: Autophagy core machinery: overcoming spatial barriers in neurons. J Mol Med 2016, 94:1217-1227 http://dx.doi.org/10.1007/s00109-016-1461-9. 37. Shen D-N, Zhang L-H, Wei E-Q, Yang Y: Autophagy in synaptic development, function, and pathology. Neurosci Bull 2015, 31:416-426. 38. Binotti B, Pavlos NJ, Riedel D, Wenzel D, Vorbru¨ggen G,  Schalk AM, Ku¨hnel K, Boyken J, Erck C, Martens H et al.: The GTPase Rab26 links synaptic vesicles to the autophagy pathway. Elife 2015, 4:e05597. This article reports a role for the Rab26 GTPase in clustering synaptic vesicles into preautophagosomal structures labelled with the autophagic markers Atg16L1, LC3B and Rab33B. This observation suggests that synaptic vesicles can be degraded by macroautophagy. 39. Neufeld TP: TOR-dependent control of autophagy: biting the hand that feeds. Curr Opin Cell Biol 2011, 22:157-168. 40. Hernandez D, Torres CA, Setlik W, Cebria´n C, Mosharov EV, Tang G, Cheng HC, Kholodilov N, Yarygina O, Burke RE et al.: Regulation of presynaptic neurotransmission by macroautophagy. Neuron 2012, 74:277-284. 41. Soukup S-F, Keunen S, Vanhauwaert R, Herna´ndez-Dı´az S,  Manetsberger J, Swerts J, Schoovaerts N, Vilain S, Gunko N, Vints K et al.: An EndophilinA phosphoswitch is critical for macroautophagy at presynaptic terminals. Neuron 2016, 92:829-844. The authors find that phosphorylated Endophilin-A promotes the formation of highly curved membranes that serve as docking station for autophagic factors, including Atg3, to form autophagosomes at presynaptic terminals. Parkingson’s disease kinase LRRK2 phosphorylates the Endophilin-A BAR domain and switches the function from the classical role of EndoA in synaptic endocytosis to a role in macroautophagy. See also Murdoch et al. 2016—Ref. [64]. 42. Tian X, Gala U, Zhang Y, Shang W, Nagarkar Jaiswal S, di Ronza A,  Jaiswal M, Yamamoto S, Sandoval H, Duraine L et al.: A voltage-gated calcium channel regulates lysosomal fusion with endosomes and autophagosomes and is required for neuronal homeostasis. PLoS Biol 2015, 13:e1002103. The authors show that Drosophila cac mutants, that are affected in the voltage-gated calcium channel (VGCC), accumulate autophagic vacuoles (AV) at photoreceptor terminals. They show that VGCC is not only required for synaptic vesicle fusion with the plasma membrane but also plays a role in autophagy by regulating the fusion of AVs with lysosomes through its calcium channel activity. 43. Jin M, Liu X, Klionsky DJ: SnapShot: selective autophagy. Cell 2013, 152 368–368. e2. 44. Uytterhoeven V, Lauwers E, Maes I, Miskiewicz K, Melo MN,  Swerts J, Kuenen S, Wittocx R, Corthout N, Marrink S-J et al.: Hsc70-4 deforms membranes to promote synaptic protein Current Opinion in Genetics & Development 2017, 44:38–46

46 Molecular and genetic bases of disease

turnover by endosomal microautophagy. Neuron 2015, 88:735-748. This paper identifies a novel role of Hsc70-4 in deforming the endosomal membrane and thereby promotes microautophagy. Gain of microautophagy increases the turnover of several synaptic proteins and stimulates neurotransmission. Sgt, a cochaperone of Hsc70-4, is able to switch the activity of Hsc70-4 from synaptic endosomal microautophagy towards protein refolding. 45. Dice FJ: Peptide sequences that target cytosolic proteins for lysosomal proteolysis. Trends Biochem Sci 1990, 15:305-309. 46. Sahu R, Kaushik S, Clement CC, Cannizzo ES, Scharf B, Follenzi A, Potolicchio I, Nieves E, Cuervo AM, Santambrogio L: Microautophagy of cytosolic proteins by late endosomes. Dev Cell 2011, 20:131-139. 47. Cuervo AM, Stefanis L, Fredenburg R, Lansbury PT, Sulzer D: Impaired degradation of mutant alpha-synuclein by chaperone-mediated autophagy. Science 2004, 305:1292-1295. 48. Henne WM, Buchkovich NJ, Emr SD: The ESCRT pathway. Dev Cell 2011, 21:77-91. 49. Takamori S, Holt M, Stenius K, Lemke EA, Grønborg M, Riedel D, Urlaub H, Schenck S, Bru¨gger B, Ringler P et al.: Molecular anatomy of a trafficking organelle. Cell 2006, 127:831-846. 50. Ja¨hne S, Rizzoli SO, Helm MS: The structure and function of presynaptic endosomes. Exp Cell Res 2015, 335:1-8. 51. Hoopmann P, Punge A, Barysch SV, Westphal V, Bu¨ckers J, Opazo F, Bethani I, Lauterbach M a, Hell SW, Rizzoli SO: Endosomal sorting of readily releasable synaptic vesicles. Proc Natl Acad Sci U S A 2010, 107:19055-19060. 52. Uytterhoeven V, Kuenen S, Kasprowicz J, Miskiewicz K, Verstreken P: Loss of skywalker reveals synaptic endosomes as sorting stations for synaptic vesicle proteins. Cell 2011, 145:117-132. 53. Fernandes AC, Uytterhoeven V, Kuenen S, Wang Y-C,  Slabbaert JR, Swerts J, Kasprowicz J, Aerts S, Verstreken P: Reduced synaptic vesicle protein degradation at lysosomes curbs TBC1D24/sky-induced neurodegeneration. J Cell Biol 2014, 207:453-462. This papers reports the use of fluorescent timers to assess protein turnover in Drosophila motorneurons. It also reveals the importance of endosome-to-lysosome trafficking for presynaptic protein homesotasis. See also Uytterhoeven et al. 2011—Ref. [52]. 54. Sheehan P, Zhu M, Beskow A, Vollmer C, Waites CL:  Activity-dependent degradation of synaptic vesicle proteins requires Rab35 and the ESCRT pathway. J Neurosci 2016, 36:8668-8686. The authors show that the small GTPase Rab35 and the endosomal sorting complex (ESCRT) pathway promote the turnover of subsets of synaptic vesicle proteins in a neuronal activity-dependent manner. This work is much in line with data from fruit flies, see Uytterhoeven et al. 2011—Ref. [52]. 55. McDonough H, Patterson C: CHIP: a link between the chaperone and proteasome systems. Cell Stress Chaperones 2003, 8:303-308. 56. Arias E, Cuervo AM: Chaperone-mediated autophagy in protein quality control. Curr Opin Cell Biol 2011, 23:184-189. 57. Arndt V, Dick N, Tawo R, Dreiseidler M, Wenzel D, Hesse M, Fu¨rst DO, Saftig P, Saint R, Fleischmann BK et al.: Chaperoneassisted selective autophagy is essential for muscle maintenance. Curr Biol 2010, 20:143-148.

Current Opinion in Genetics & Development 2017, 44:38–46

58. Xiong X, Hao Y, Sun K, Li J, Li X, Mishra B, Soppina P, Wu C, Hume RI, Collins CA: The highwire ubiquitin ligase promotes axonal degeneration by tuning levels of nmnat protein. PLoS Biol 2012, 10:1-18. 59. Cohen-Kaplan V, Livneh I, Avni N, Cohen-Rosenzweig C, Ciechanover A: The ubiquitin-proteasome system and autophagy: coordinated and independent activities. Int J Biochem Cell Biol 2016, 79:403-418 http://dx.doi.org/ 10.1016/j.biocel.2016.07.019. 60. Tian Y, Chang JC, Greengard P, Flajolet M: The convergence of endosomal and autophagosomal pathways. Autophagy 2014,  10:694-696. The authors show the existence of a complex between the autophagy marker LC3 and the endocytic factors AP2 and PICALM. AP2/PICALM reroute APP C-terminal fragments (APP-CTF) from the endocytic pathway to macroautophagy. This functional link between endocytosis and macroautophagy might be involved in Ab clearance. 61. Ravikumar B, Moreau K, Jahreiss L, Puri C, Rubinsztein DC: Plasma membrane contributes to the formation of preautophagosomal structures. Cell 2011, 12:747-757. 62. Stavoe AKH, Hill SE, Hall DH, Colo´n-Ramos DA: KIF1A/UNC-104  transports ATG-9 to regulate neurodevelopment and autophagy at synapses. Dev Cell 2016, 38:171-185. The authors found that kinesin KIF1A/UNC104 is required for transport of the integral membrane autophagy protein, Atg9, to the site of autophagosome biogenesis at the presynapse, thus indicating an important role of autophagy at presynaptic terminals. 63. Popovic D, Dikic I: TBC1D5 and the AP2 complex regulate ATG9  trafficking and initiation of autophagy. EMBO Rep 2014, 15:392-401 http://dx.doi.org/10.1002/embr.201337995. The authors found that the RabGAP protein TBC1D5 binds to Atg9 and recruits clathrin and AP2 to the autophagosome formation site. They propose that TBC1D5 and AP2 are novel macroautophagy regulators that reroute Atg9 from vesicular carriers towards preautophagosomes. 64. Murdoch JD, Rostosky CM, Gowrisankaran S, Arora AS,  Soukup S-F, Vidal R, Capece V, Freytag S, Fischer A, Verstreken P et al.: Endophilin-A deficiency induces the Foxo3a-Fbxo32 network in the brain and causes dysregulation of autophagy and the ubiquitin-proteasome system. Cell Rep 2016, 17:1071-1086 http://dx.doi.org/10.1016/j.celrep.2016.09.058. The authors provide evidence that Endophilin-A contributes to autophagosome formation at the presynapse and balances the activities of macroautophagy and of the UPS as part of a FOXO/FBXO32/Endophilin network. See also Soukup et al. 2016—Ref. [41]. 65. Xilouri M, Brekk OR, Landeck N, Pitychoutis PM, Papasilekas T, Papadopoulou-Daifoti Z, Kirik D, Stefanis L: Boosting chaperone-mediated autophagy in vivo mitigates a-synuclein-induced neurodegeneration. Brain 2013, 136:2130-2146. 66. Ciechanover A, Kwon YT: Degradation of misfolded proteins in neurodegenerative diseases: therapeutic targets and strategies. Exp Mol Med 2015, 47:e147. 67. Pratt WB, Gestwicki JE, Osawa Y, Lieberman AP, Arbor A, Arbor A, Francisco S: Targeting proteostasis through the protein quality control function of the Hsp90/Hsp70-based chaperone machinery for treatment of adult onset neurodegenerative diseases. Annu Rev Pharmacol Toxicol 2016, 55:353-371. 68. Bose S, Cho J: Targeting chaperones, heat shock factor-1, and unfolded protein response: promising therapeutic approaches for neurodegenerative disorders. Ageing Res Rev 2016 http://dx.doi.org/10.1016/j.arr.2016.09.004. Oct 1. pii: S1568-1637(16)30092-7.

www.sciencedirect.com