- Email: [email protected]
Autophagosome Formation by Endophilin Keeps Synapses in Shape
Neuron
Previews Autophagosome Formation by Endophilin Keeps Synapses in Shape Marijn Kuijpers1 and Volker Haucke1,*
1Leibniz-Institut fu €r Molekulare Pharmakologie, 13125 Berlin, Germany *Correspondence: [email protected] http://dx.doi.org/10.1016/j.neuron.2016.11.016
Soukup et al. (2016), in this issue of Neuron, and Murdoch et al. (2016), in Cell Reports, reveal an unexpected function for the endocytic protein endophilin in autophagosome formation at synapses: preventing neurodegeneration and ataxia. Neurons and the synapses formed between them have to cope with special challenges owed to their high metabolic activity and the reuse of proteins and organelles over extended periods of time. Autophagy (also termed macroautophagy), a process that allows the sequestration and turnover of defective organelles or protein aggregates via a double membrane system, has long been thought to be crucial for the maintenance of synaptic function and for preventing neurodegeneration. How autophagy at neuronal synapses is initiated is poorly understood. Two papers (Murdoch et al., 2016; Soukup et al., 2016) now show that the synapse-enriched endocytic adaptor protein endophilin-A is critical for the formation of autophagosomes and for balancing neuronal protein homeostasis via an endocytosisindependent mechanism controlled by the Parkinson-related protein kinase LRRK2. Synapse structure, function, and plasticity are regulated by altering the abundance of synaptic proteins in a spatially confined manner via local translation as well as turnover of synaptic components. As synapses are often located far away from the soma, where most protein synthesis and turnover occur, they have evolved synapse-specific mechanisms that allow them to control their protein content (Cajigas et al., 2010). Among the synaptic pathways that mediate local protein turnover are the ubiquitin proteasome system (Cajigas et al., 2010); endosomal microautophagy (Uytterhoeven et al., 2015), a pathway that targets cytosolic proteins for turnover in late endosomes, and (macro)autophagy, which is initiated in axons (Maday and Holzbaur,
2014) and is thought to counteract neurodegeneration (Rubinsztein et al., 2012). Autophagy is an evolutionary conserved cellular process that serves to provide nutrients during starvation and to eliminate defective proteins and organelles via lysosomal degradation. During autophagy, portions of the cytoplasm are sequestered within double-membraned or multi-membraned vesicles termed autophagosomes. These undergo subsequent maturation steps before being delivered to lysosomes. Autophagosome formation is initiated by early-acting proteins such as the autophagy-related gene (ATG) 1/ ULK1 kinase complex and ATG3, an enzyme required for lipid-conjugation of microtubule-associated protein 1 light chain 3 (LC3, termed ATG8 in invertebrates and yeast), a central component of autophagosomes. Accumulation of autophagosomes is a hallmark of neurodegenerative disorders, and knockout (KO) of key ATG proteins in animal models causes neurodegeneration (Rubinsztein et al., 2012). Although the coordinated recruitment of ATG proteins during autophagosome formation is comparably well understood, it remains unknown how autophagosome formation at synapses is initiated and locally controlled. Soukup et al. (2016) tackled this problem by tracing autophagosome formation at the fly larval neuromuscular junction (NMJ) in vivo. They demonstrated that autophagosome formation is induced not only by starvation but also by neuronal activity. Using correlative light and electron microscopy, they further showed that LC3-positive autophagosomes indeed reside within synaptic boutons. So, how is their formation coupled to neuronal activity? One candi-
date they followed in depth was the protein endophilin-A, a protein previously characterized for its role in endocytosis at synapses and in non-neuronal cells (Kjaerulff et al., 2011). Endophilin-A is encoded by a single gene in D. melanogaster, while three isoforms (A1–A3) exist in mammals. Endophilin-A harbors a membrane-deforming bin-amphiphysin-rvs (BAR) domain thought to be the main business end of the molecule and a carboxy-terminal src homology 3 (SH3) domain, through which it associates with and recruits other endocytic factors such as synaptojanin and dynamin (Figure 1). Loss of endophilin-A in various models has been shown to result in severe endocytic defects and impaired neurotransmission (Milosevic et al., 2011; Verstreken et al., 2002). Interestingly, endophilin-A is reversibly phosphorylated by the kinase LRRK2, encoded by a gene frequently mutated in familial cases of Parkinson’s disease (PD), thereby regulating endophilin’s membrane remodeling and endocytic activities (Matta et al., 2012). As LRRK2 has been associated with autophagic turnover of mitochondria and neurodegeneration, Soukup et al. (2016) reasoned that endophilin-A might have another secret life as a component of synaptic autophagosome formation. Consistent with this hypothesis, they found endophilin-A to be present on preautophagosomal membranes. Moreover, loss of endophilin-A due either to genetic or to acute light-induced inactivation, interfered with stimulation-induced autophagosome formation. Further in vitro studies using giant liposomes revealed that endophilin-A phosphorylation at S75 by LRRK2 alters endophilin’s membrane-deforming activity to result in
Neuron 92, November 23, 2016 ª 2016 Elsevier Inc. 675
Neuron
Previews
Figure 1. EndophilinA Functions in Synaptic Vesicle Endocytosis and Autophagosome Formation Synapse-enriched endophilin-A can bind high-curvature membranes such as the neck of endocytic pits and recruits other endocytic factors like synaptojanin (involved in clathrin uncoating) and dynamin involved in (fission) to endocytic sites. Additionally, endophilin-A localizes on pre-autophagosomal membranes, where a right balance of endophilin-A phosphorylation/dephosphorylation is critical for the formation of synaptic autophagosomes.
formation of short tubules and high curvature intermediates that allow recruitment of the early-acting autophagy enzyme ATG3 to these sites. Phosphorylation of endophilin-A, as well as LRRK2 itself, was also required for synaptic autophagosome induction in vivo. As LRRK2 is commonly mutated in PD, the authors hypothesized that endophilin-A might counteract neurodegeneration in PD. Indeed, flies, in which endophilin-A was removed specifically in the eye, showed degeneration of photoreceptor synaptic terminals and a progressive decline in light-stimulated photoreceptor depolarization. The expression of neither non-phosphorylatable nor phosphomimetic forms of endophilin-A could rescue this phenotype, suggesting that the right balance between LRRK2 activity, endophilin-A phosphorylation/dephosphorylation, and autophagosome formation is required for synaptic function and neuronal survival. These data thus suggest a model according to which LRRK2 couples synaptic activity to endophilin-mediated autophagosome formation at neuronal synapses. So how does dysfunction of autophagy in the absence of endophilin-A affect the gene and protein networks that control neuronal protein turnover, and how does this relate to PD? To provide some initial answers to at least some of these important questions, Murdoch et al. (2016) 676 Neuron 92, November 23, 2016
studied KO mice lacking multiple copies of the genes that encode the three isoforms of endophilin-A in mice. They were intrigued by the fact that endophilin-Adeficient mice display signs of neurodegeneration together with the upregulation of Parkin, an E3 ubiquitin ligase genetically implicated in PD and clearance of defective mitochondria by autophagy (Cao et al., 2014). As Parkin depletion was insufficient to rescue neurodegeneration in endophilin-A KO mice, they reasoned that additional components might underlie the observed defects. To unravel such components, material from endophilin-A double (DKO) and triple KO (TKO) mice was subjected to RNA sequencing, resulting in the identification of gene changes in various pathways including synaptic transmission and protein homeostasis. Among the most striking and consistent changes was the upregulation of the E3-ubiquitin ligase FBXO32 and the transcription factor FOXO3A that controls FBXO32 expression. Target protein ubiquitination by FBXO32 has been linked to the ubiquitin-proteasome system and to autophagy in muscle. Consistent with this, Murdoch et al. observed increased levels of K48-ubiquitinated substrates destined for proteasomal degradation, while key autophagosomal proteins such as LC3B and ATG5 as well as morphologically
identifyable autophagosomes were reduced in endophilin-A DKO and TKO neurons. These data suggest a crosstalk between endophilin-A-mediated autophagosome formation and the ubiquitin proteasome system for protein turnover. Interestingly, FBXO32 and endophilin-A also interact directly and transiently colocalize on autophagosomes in cellular models. Loss of FBXO32 in fly NMJs phenocopied endophilin-A loss with respect to defective autophagosome formation. Conversely, overexpression of FBXO32 led to apoptotic cell death that could be rescued by co-overexpression of endophilin-A by preventing nuclear translocation of FBXO32. Together, these results indicate that endophilin-A via association with FBXO32 serves as a critical determinant for controlling neuronal health and viability. Disrupting the intricate balance between both proteins leads to defects in autophagy and a concomitant increased load on the proteasomal pathway that may eventually result in accumulation of defective proteins and neurodegeneration. Endophilin-A likely is not the only endocytic protein with a function in autophagy. Recent data suggest that the clathrin adaptor complex AP-2, another key endocytic protein at synapses, may target internalized amyloid precursor protein (APP) for autophagic turnover by complex formation with the LC3 component of autophagosomes (Tian et al., 2013). Defects in autophagy have previously been linked to endocytosis as autophagosome formation requires endocytic trafficking of the transmembrane protein ATG9 to endosomes, while ATG16L, a component of the ligase complex that modifies LC3 protein, associates with clathrin (Rubinsztein et al., 2012). Murdoch et al. and Soukup et al. argue and present several lines of evidence that defective endocytosis does not underlie the observed defects in autophagosome formation in absence of endophilin-A. First, endophilin-A phosphomimetic or non-phosphorylatable mutants display similarly mild defects in endocytosis but differ dramatically with respect to autophagosome formation: endophilin-A S75A suppresses autophagy, while autophagosome formation is constitutively active in the presence of phosphomimetic endophilin-A S75D. Second, conditional inactivation of dynamin in
Neuron
Previews shibirets mutant flies does not affect autophagosome formation, in spite of the complete block of endocytosis (Soukup et al., 2016). Third, loss of endophilin-A in mice has no overt effect on the levels of ATG9 or ATG16L proteins (Murdoch et al., 2016). Collectively, these results support a direct function of endophilin-A in autophagosome formation independent of its established role in endocytosis (Figure 1). As with all exciting new works, a number of open questions remain: How exactly is the function of endophilin-A regulated, and how do LRRK2 and other factors such as FBXO32/ FOXO3A sense neuronal activity? What physiological mechanisms define the switch between the functions of endophilin-A in endocytosis and in autophagy? Which partner proteins associate with endophilin-A in autophagy, and which membranes are the substrates for autophagosome formation at synapses? Finally, we know little
about the actual substrates of neuronal and synaptic autophagy, and we do not know the precise fate of synaptic autophagosomes that undergo retrograde transport to the neuronal soma (Maday and Holzbaur, 2014). Future studies will need to provide answers to these exciting questions.
Milosevic, I., Giovedi, S., Lou, X., Raimondi, A., Collesi, C., Shen, H., Paradise, S., O’Toole, E., Ferguson, S., Cremona, O., and De Camilli, P. (2011). Neuron 72, 587–601.
REFERENCES
Soukup, S.F., Kuenen, S., Vanhauwaert, R., Manetsberger, J., Herna´ndez-Dı´az, S., Swerts, J., Schoovaerts, N., Vilain, S., Gounko, N.V., Vints, K., et al. (2016). Neuron 92, this issue, 829–844.
Cajigas, I.J., Will, T., and Schuman, E.M. (2010). EMBO J. 29, 2746–2752. Cao, M., Milosevic, I., Giovedi, S., and De Camilli, P. (2014). J. Neurosci. 34, 16544–16549. Kjaerulff, O., Brodin, L., and Jung, A. (2011). Cell Biochem. Biophys. 60, 137–154. Maday, S., and Holzbaur, E.L. (2014). Dev. Cell 30, 71–85. Matta, S., Van Kolen, K., da Cunha, R., van den Bogaart, G., Mandemakers, W., Miskiewicz, K., De Bock, P.J., Morais, V.A., Vilain, S., Haddad, D., et al. (2012). Neuron 75, 1008–1021.
Murdoch, J.D., Rostosky, C.M., Gowrisankaran, S., Arora, A.S., Soukup, S.F., Vidal, R., Capece, V., Freytag, S., Fischer, A., Verstreken, P., et al. (2016). Cell Rep. 17, 1071–1086. Rubinsztein, D.C., Shpilka, T., and Elazar, Z. (2012). Curr. Biol. 22, R29–R34.
Tian, Y., Chang, J.C., Fan, E.Y., Flajolet, M., and Greengard, P. (2013). Proc. Natl. Acad. Sci. USA 110, 17071–17076. Uytterhoeven, V., Lauwers, E., Maes, I., Miskiewicz, K., Melo, M.N., Swerts, J., Kuenen, S., Wittocx, R., Corthout, N., Marrink, S.J., et al. (2015). Neuron 88, 735–748. Verstreken, P., Kjaerulff, O., Lloyd, T.E., Atkinson, R., Zhou, Y., Meinertzhagen, I.A., and Bellen, H.J. (2002). Cell 109, 101–112.
Cut Your Losses: Spastin Mediates Branch-Specific Axon Loss Hagar Meltzer1 and Oren Schuldiner1,* 1Department of Molecular Cell Biology, Weizmann Institute of Sciences, Rehovot 7610001, Israel *Correspondence: [email protected] http://dx.doi.org/10.1016/j.neuron.2016.11.004
In this issue of Neuron, Brill et al. (2016) demonstrate that, during synapse elimination in the developing neuromuscular junction, branch-specific microtubule destabilization results in arrested axonal transport and induces axon branch loss. This process is mediated in part by the neurodegeneration-associated, microtubule-severing protein spastin. Developmental neuronal remodeling is crucial for sculpting the mature nervous system. In mammals, neuronal remodeling is largely timed to postnatal development and serves to refine neural circuits that were formed during the embryonic period, often by eliminating exuberant connections. This may include pruning of individual axonal projections while the cell body, and in some cases sister axonal projections, remains intact. Axon pruning is a widespread phenomenon and has been demonstrated throughout the cen-
tral and peripheral nervous systems of both vertebrates and invertebrates. In addition to its developmental significance, understanding the molecular basis of axon pruning may shed light on the resembling process of axon degeneration during certain ‘‘dying-back’’ neurodegenerative diseases (Luo and O’Leary, 2005; Yaron and Schuldiner, 2016). One of the first neural structures in which axon pruning was demonstrated to be essential for establishing proper wiring is the neuromuscular junction (NMJ;
Purves and Lichtman, 1980). When the mammalian NMJ initially forms, several axon branches originating from different motor neurons converge to a single synaptic site on the muscle, which is at this stage poly-innervated. However, during the early postnatal period, extensive axon pruning takes place until only a single axon branch remains. This synaptic elimination is based on ongoing competition between axon branches, in which the ‘‘winning’’ branch gradually expands its territory at the expanse of the ‘‘losing’’
Neuron 92, November 23, 2016 ª 2016 Elsevier Inc. 677
Previews Autophagosome Formation by Endophilin Keeps Synapses in Shape Marijn Kuijpers1 and Volker Haucke1,*
1Leibniz-Institut fu €r Molekulare Pharmakologie, 13125 Berlin, Germany *Correspondence: [email protected] http://dx.doi.org/10.1016/j.neuron.2016.11.016
Soukup et al. (2016), in this issue of Neuron, and Murdoch et al. (2016), in Cell Reports, reveal an unexpected function for the endocytic protein endophilin in autophagosome formation at synapses: preventing neurodegeneration and ataxia. Neurons and the synapses formed between them have to cope with special challenges owed to their high metabolic activity and the reuse of proteins and organelles over extended periods of time. Autophagy (also termed macroautophagy), a process that allows the sequestration and turnover of defective organelles or protein aggregates via a double membrane system, has long been thought to be crucial for the maintenance of synaptic function and for preventing neurodegeneration. How autophagy at neuronal synapses is initiated is poorly understood. Two papers (Murdoch et al., 2016; Soukup et al., 2016) now show that the synapse-enriched endocytic adaptor protein endophilin-A is critical for the formation of autophagosomes and for balancing neuronal protein homeostasis via an endocytosisindependent mechanism controlled by the Parkinson-related protein kinase LRRK2. Synapse structure, function, and plasticity are regulated by altering the abundance of synaptic proteins in a spatially confined manner via local translation as well as turnover of synaptic components. As synapses are often located far away from the soma, where most protein synthesis and turnover occur, they have evolved synapse-specific mechanisms that allow them to control their protein content (Cajigas et al., 2010). Among the synaptic pathways that mediate local protein turnover are the ubiquitin proteasome system (Cajigas et al., 2010); endosomal microautophagy (Uytterhoeven et al., 2015), a pathway that targets cytosolic proteins for turnover in late endosomes, and (macro)autophagy, which is initiated in axons (Maday and Holzbaur,
2014) and is thought to counteract neurodegeneration (Rubinsztein et al., 2012). Autophagy is an evolutionary conserved cellular process that serves to provide nutrients during starvation and to eliminate defective proteins and organelles via lysosomal degradation. During autophagy, portions of the cytoplasm are sequestered within double-membraned or multi-membraned vesicles termed autophagosomes. These undergo subsequent maturation steps before being delivered to lysosomes. Autophagosome formation is initiated by early-acting proteins such as the autophagy-related gene (ATG) 1/ ULK1 kinase complex and ATG3, an enzyme required for lipid-conjugation of microtubule-associated protein 1 light chain 3 (LC3, termed ATG8 in invertebrates and yeast), a central component of autophagosomes. Accumulation of autophagosomes is a hallmark of neurodegenerative disorders, and knockout (KO) of key ATG proteins in animal models causes neurodegeneration (Rubinsztein et al., 2012). Although the coordinated recruitment of ATG proteins during autophagosome formation is comparably well understood, it remains unknown how autophagosome formation at synapses is initiated and locally controlled. Soukup et al. (2016) tackled this problem by tracing autophagosome formation at the fly larval neuromuscular junction (NMJ) in vivo. They demonstrated that autophagosome formation is induced not only by starvation but also by neuronal activity. Using correlative light and electron microscopy, they further showed that LC3-positive autophagosomes indeed reside within synaptic boutons. So, how is their formation coupled to neuronal activity? One candi-
date they followed in depth was the protein endophilin-A, a protein previously characterized for its role in endocytosis at synapses and in non-neuronal cells (Kjaerulff et al., 2011). Endophilin-A is encoded by a single gene in D. melanogaster, while three isoforms (A1–A3) exist in mammals. Endophilin-A harbors a membrane-deforming bin-amphiphysin-rvs (BAR) domain thought to be the main business end of the molecule and a carboxy-terminal src homology 3 (SH3) domain, through which it associates with and recruits other endocytic factors such as synaptojanin and dynamin (Figure 1). Loss of endophilin-A in various models has been shown to result in severe endocytic defects and impaired neurotransmission (Milosevic et al., 2011; Verstreken et al., 2002). Interestingly, endophilin-A is reversibly phosphorylated by the kinase LRRK2, encoded by a gene frequently mutated in familial cases of Parkinson’s disease (PD), thereby regulating endophilin’s membrane remodeling and endocytic activities (Matta et al., 2012). As LRRK2 has been associated with autophagic turnover of mitochondria and neurodegeneration, Soukup et al. (2016) reasoned that endophilin-A might have another secret life as a component of synaptic autophagosome formation. Consistent with this hypothesis, they found endophilin-A to be present on preautophagosomal membranes. Moreover, loss of endophilin-A due either to genetic or to acute light-induced inactivation, interfered with stimulation-induced autophagosome formation. Further in vitro studies using giant liposomes revealed that endophilin-A phosphorylation at S75 by LRRK2 alters endophilin’s membrane-deforming activity to result in
Neuron 92, November 23, 2016 ª 2016 Elsevier Inc. 675
Neuron
Previews
Figure 1. EndophilinA Functions in Synaptic Vesicle Endocytosis and Autophagosome Formation Synapse-enriched endophilin-A can bind high-curvature membranes such as the neck of endocytic pits and recruits other endocytic factors like synaptojanin (involved in clathrin uncoating) and dynamin involved in (fission) to endocytic sites. Additionally, endophilin-A localizes on pre-autophagosomal membranes, where a right balance of endophilin-A phosphorylation/dephosphorylation is critical for the formation of synaptic autophagosomes.
formation of short tubules and high curvature intermediates that allow recruitment of the early-acting autophagy enzyme ATG3 to these sites. Phosphorylation of endophilin-A, as well as LRRK2 itself, was also required for synaptic autophagosome induction in vivo. As LRRK2 is commonly mutated in PD, the authors hypothesized that endophilin-A might counteract neurodegeneration in PD. Indeed, flies, in which endophilin-A was removed specifically in the eye, showed degeneration of photoreceptor synaptic terminals and a progressive decline in light-stimulated photoreceptor depolarization. The expression of neither non-phosphorylatable nor phosphomimetic forms of endophilin-A could rescue this phenotype, suggesting that the right balance between LRRK2 activity, endophilin-A phosphorylation/dephosphorylation, and autophagosome formation is required for synaptic function and neuronal survival. These data thus suggest a model according to which LRRK2 couples synaptic activity to endophilin-mediated autophagosome formation at neuronal synapses. So how does dysfunction of autophagy in the absence of endophilin-A affect the gene and protein networks that control neuronal protein turnover, and how does this relate to PD? To provide some initial answers to at least some of these important questions, Murdoch et al. (2016) 676 Neuron 92, November 23, 2016
studied KO mice lacking multiple copies of the genes that encode the three isoforms of endophilin-A in mice. They were intrigued by the fact that endophilin-Adeficient mice display signs of neurodegeneration together with the upregulation of Parkin, an E3 ubiquitin ligase genetically implicated in PD and clearance of defective mitochondria by autophagy (Cao et al., 2014). As Parkin depletion was insufficient to rescue neurodegeneration in endophilin-A KO mice, they reasoned that additional components might underlie the observed defects. To unravel such components, material from endophilin-A double (DKO) and triple KO (TKO) mice was subjected to RNA sequencing, resulting in the identification of gene changes in various pathways including synaptic transmission and protein homeostasis. Among the most striking and consistent changes was the upregulation of the E3-ubiquitin ligase FBXO32 and the transcription factor FOXO3A that controls FBXO32 expression. Target protein ubiquitination by FBXO32 has been linked to the ubiquitin-proteasome system and to autophagy in muscle. Consistent with this, Murdoch et al. observed increased levels of K48-ubiquitinated substrates destined for proteasomal degradation, while key autophagosomal proteins such as LC3B and ATG5 as well as morphologically
identifyable autophagosomes were reduced in endophilin-A DKO and TKO neurons. These data suggest a crosstalk between endophilin-A-mediated autophagosome formation and the ubiquitin proteasome system for protein turnover. Interestingly, FBXO32 and endophilin-A also interact directly and transiently colocalize on autophagosomes in cellular models. Loss of FBXO32 in fly NMJs phenocopied endophilin-A loss with respect to defective autophagosome formation. Conversely, overexpression of FBXO32 led to apoptotic cell death that could be rescued by co-overexpression of endophilin-A by preventing nuclear translocation of FBXO32. Together, these results indicate that endophilin-A via association with FBXO32 serves as a critical determinant for controlling neuronal health and viability. Disrupting the intricate balance between both proteins leads to defects in autophagy and a concomitant increased load on the proteasomal pathway that may eventually result in accumulation of defective proteins and neurodegeneration. Endophilin-A likely is not the only endocytic protein with a function in autophagy. Recent data suggest that the clathrin adaptor complex AP-2, another key endocytic protein at synapses, may target internalized amyloid precursor protein (APP) for autophagic turnover by complex formation with the LC3 component of autophagosomes (Tian et al., 2013). Defects in autophagy have previously been linked to endocytosis as autophagosome formation requires endocytic trafficking of the transmembrane protein ATG9 to endosomes, while ATG16L, a component of the ligase complex that modifies LC3 protein, associates with clathrin (Rubinsztein et al., 2012). Murdoch et al. and Soukup et al. argue and present several lines of evidence that defective endocytosis does not underlie the observed defects in autophagosome formation in absence of endophilin-A. First, endophilin-A phosphomimetic or non-phosphorylatable mutants display similarly mild defects in endocytosis but differ dramatically with respect to autophagosome formation: endophilin-A S75A suppresses autophagy, while autophagosome formation is constitutively active in the presence of phosphomimetic endophilin-A S75D. Second, conditional inactivation of dynamin in
Neuron
Previews shibirets mutant flies does not affect autophagosome formation, in spite of the complete block of endocytosis (Soukup et al., 2016). Third, loss of endophilin-A in mice has no overt effect on the levels of ATG9 or ATG16L proteins (Murdoch et al., 2016). Collectively, these results support a direct function of endophilin-A in autophagosome formation independent of its established role in endocytosis (Figure 1). As with all exciting new works, a number of open questions remain: How exactly is the function of endophilin-A regulated, and how do LRRK2 and other factors such as FBXO32/ FOXO3A sense neuronal activity? What physiological mechanisms define the switch between the functions of endophilin-A in endocytosis and in autophagy? Which partner proteins associate with endophilin-A in autophagy, and which membranes are the substrates for autophagosome formation at synapses? Finally, we know little
about the actual substrates of neuronal and synaptic autophagy, and we do not know the precise fate of synaptic autophagosomes that undergo retrograde transport to the neuronal soma (Maday and Holzbaur, 2014). Future studies will need to provide answers to these exciting questions.
Milosevic, I., Giovedi, S., Lou, X., Raimondi, A., Collesi, C., Shen, H., Paradise, S., O’Toole, E., Ferguson, S., Cremona, O., and De Camilli, P. (2011). Neuron 72, 587–601.
REFERENCES
Soukup, S.F., Kuenen, S., Vanhauwaert, R., Manetsberger, J., Herna´ndez-Dı´az, S., Swerts, J., Schoovaerts, N., Vilain, S., Gounko, N.V., Vints, K., et al. (2016). Neuron 92, this issue, 829–844.
Cajigas, I.J., Will, T., and Schuman, E.M. (2010). EMBO J. 29, 2746–2752. Cao, M., Milosevic, I., Giovedi, S., and De Camilli, P. (2014). J. Neurosci. 34, 16544–16549. Kjaerulff, O., Brodin, L., and Jung, A. (2011). Cell Biochem. Biophys. 60, 137–154. Maday, S., and Holzbaur, E.L. (2014). Dev. Cell 30, 71–85. Matta, S., Van Kolen, K., da Cunha, R., van den Bogaart, G., Mandemakers, W., Miskiewicz, K., De Bock, P.J., Morais, V.A., Vilain, S., Haddad, D., et al. (2012). Neuron 75, 1008–1021.
Murdoch, J.D., Rostosky, C.M., Gowrisankaran, S., Arora, A.S., Soukup, S.F., Vidal, R., Capece, V., Freytag, S., Fischer, A., Verstreken, P., et al. (2016). Cell Rep. 17, 1071–1086. Rubinsztein, D.C., Shpilka, T., and Elazar, Z. (2012). Curr. Biol. 22, R29–R34.
Tian, Y., Chang, J.C., Fan, E.Y., Flajolet, M., and Greengard, P. (2013). Proc. Natl. Acad. Sci. USA 110, 17071–17076. Uytterhoeven, V., Lauwers, E., Maes, I., Miskiewicz, K., Melo, M.N., Swerts, J., Kuenen, S., Wittocx, R., Corthout, N., Marrink, S.J., et al. (2015). Neuron 88, 735–748. Verstreken, P., Kjaerulff, O., Lloyd, T.E., Atkinson, R., Zhou, Y., Meinertzhagen, I.A., and Bellen, H.J. (2002). Cell 109, 101–112.
Cut Your Losses: Spastin Mediates Branch-Specific Axon Loss Hagar Meltzer1 and Oren Schuldiner1,* 1Department of Molecular Cell Biology, Weizmann Institute of Sciences, Rehovot 7610001, Israel *Correspondence: [email protected] http://dx.doi.org/10.1016/j.neuron.2016.11.004
In this issue of Neuron, Brill et al. (2016) demonstrate that, during synapse elimination in the developing neuromuscular junction, branch-specific microtubule destabilization results in arrested axonal transport and induces axon branch loss. This process is mediated in part by the neurodegeneration-associated, microtubule-severing protein spastin. Developmental neuronal remodeling is crucial for sculpting the mature nervous system. In mammals, neuronal remodeling is largely timed to postnatal development and serves to refine neural circuits that were formed during the embryonic period, often by eliminating exuberant connections. This may include pruning of individual axonal projections while the cell body, and in some cases sister axonal projections, remains intact. Axon pruning is a widespread phenomenon and has been demonstrated throughout the cen-
tral and peripheral nervous systems of both vertebrates and invertebrates. In addition to its developmental significance, understanding the molecular basis of axon pruning may shed light on the resembling process of axon degeneration during certain ‘‘dying-back’’ neurodegenerative diseases (Luo and O’Leary, 2005; Yaron and Schuldiner, 2016). One of the first neural structures in which axon pruning was demonstrated to be essential for establishing proper wiring is the neuromuscular junction (NMJ;
Purves and Lichtman, 1980). When the mammalian NMJ initially forms, several axon branches originating from different motor neurons converge to a single synaptic site on the muscle, which is at this stage poly-innervated. However, during the early postnatal period, extensive axon pruning takes place until only a single axon branch remains. This synaptic elimination is based on ongoing competition between axon branches, in which the ‘‘winning’’ branch gradually expands its territory at the expanse of the ‘‘losing’’
Neuron 92, November 23, 2016 ª 2016 Elsevier Inc. 677