View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector Neuron Previews

Snapin Snaps into the Dynein Complex for Late Endosome-Lysosome Trafficking and Autophagy

Michisuke Yuzaki1,2,* 1Department of Physiology, School of Medicine, Keio University, Tokyo 160-8582, Japan 2CREST, Japan Science and Technology Agency, Saitama 332-0012, Japan *Correspondence: [email protected] DOI 10.1016/j.neuron.2010.09.036

Late endosome-lysosome trafficking plays a key role in regulating cell surface signaling and degradation of intracellular components by autophagy. New work by Cai and coworkers in this issue of Neuron provides evidence that snapin regulates the recruitment of late endosomes to the dynein motor complex for retrograde trafficking along microtubules and maturation of lysosomes.

Plasma membrane endocytosis regulates Klumperman, 2009). Therefore, late endo- complex by directly binding to dynein neuronal signaling processes by control- some-lysosome trafficking plays a key intermediate chain (DIC). Snapin was ling the number of functional receptors role in coordinating the heterophagy and originally identified as a protein that and channels available at the cell surface. autophagy involved in critical neuronal binds to synaptosomal-associated pro- Endosomal cargoes are selected to take signaling pathways and functions, such tein 25 kDa (SNAP-25) and modulates one of two major routes: returning to as synaptic plasticity and cell survival. neurosecretion by stabilizing the release- the plasma membrane in recycling endo- Lysosomes are predominantly localized ready pool of dense-core and synaptic somes, or passing through late endo- in the soma near the microtubule-orga- vesicles and facilitating synchronized somes before delivery to lysosomes for nizing center. To reach the soma, late release of synaptic vesicles (Pan et al., degradation. For example, the number of endosomes and autophagosomes are 2009). Cai et al. (2010) showed that imma- AMPA receptors at the postsynaptic site thought to be transported by the motor ture lysosomes, which did not contain the is thought to regulate synaptic plasticity, protein dynein, which moves along micro- mature form of cathepsin D, accumulated such as long-term potentiation and long- tubules toward their minus-end direction. in Snapin-null neurons (Figure 1). In addi- term depression (LTD). Endocytosed Thus, for highly polarized cells such as tion, retrograde transport of Rab7-posi- AMPA receptors are also recycled to the neurons, transport of endosomes and au- tive late endosomes was significantly cell surface or degraded in lysosomes, tophagosomes from the cell periphery to reduced in snapin null neuronal axons depending on the stimulus conditions, the soma is a challenging task. Indeed, (Figure 1). As a result, degradation of such as activation of NMDA receptors, when dynein functions are disturbed, late endogenous epidermal growth factor subunit composition, and phosphoryla- endosomes were redistributed to the cell (EGF) receptors after EGF-stimulated tion status of AMPA receptor subunits periphery (Burkhardt et al., 1997) and endocytosis was significantly retarded in (Hirling, 2009). In addition, certain cellular retrograde movement of autophagosomes snapin null neurons. Similarly, degrada- signaling events are thought to occur on were severely impaired in neuronal axons tion of exogenous EGF receptor, dextran, endosome membranes. For example, (Katsumata et al., 2010). In addition, muta- or bovine serum albumin was impaired in nerve growth factor and its receptor trkA tions affecting the dynein machinery cause mouse embryonic fibroblasts (MEFs) are endocytosed at nerve terminals and axonal degeneration and impair autopha- derived from snapin null mice. These retrogradely transported along axons in gic clearance of aggregate-prone proteins phenotypes were rescued by reintroduc- vesicles positive for Rab7, a marker for (Ravikumar et al., 2005). Similarly, the tion of wild-type snapin, but not a mutant late endosomes (von Zastrow and Sorkin, accumulation of autophagosomes and snapin (snapin-L99K) that could not bind 2007). In addition to a role in the degrada- autolysosomes has been observed in to DIC, into snapin null neurons. Further- tion of cell surface or exogenous proteins axons during the early stages of many more, late autophagic vacuoles increased (i.e., heterophagy), lysosomes are also neurodegenerative diseases (Wong and in snapin null neurons and MEFs by the responsible for the degradation of intra- Cuervo, 2010), including Alzheimer’s dis- reduced clearance of autolysosomes, re- cellular components by the process of ease, prion disease, Parkinson’s disease, flecting the observed impairment of lyso- autophagy, and newly formed autopha- and Huntington’s disease, and in neurons somal functions (Figure 1). Interestingly, gosomes undergo a stepwise maturation after hypoxia-ischemia and brain injury. overexpression of wild-type snapin, but by fusing with endosomes and lysosomes Nevertheless, how the dynein complex is not snapin-L99K, in wild-type neurons re- (Eskelinen, 2005). Finally, acid hydro- recruited to late endosomes and autopha- sulted in increased formation of elon- lases, enzymes essential for degradation gosomes is not completely clear. gated, tubular lysosomes, suggesting of target substrates in mature lysosomes, In this issue of Neuron, Cai et al. (2010) that snapin-DIC interaction may facilitate are also delivered to lysosomes from the have shown that snapin regulates recruit- fusion between late endosomes and lyso- trans-Golgi via endosomes (Saftig and ment of late endosomes to the dynein somes. Together, these results indicate

4 Neuron 68, October 7, 2010 ª2010 Elsevier Inc. Neuron Previews

that snapin probably plays a protein 2 and have mixed crucial role in heterophagy polarity (Conde and Ca´ ceres, and autophagy by regulating 2009). When overexpressed dynein-based trafficking of in neurons, snapin was previ- late endosomes to lysosomes ously shown to reduce the and by facilitating maturation number of primary dendrites of lysosomes (Figure 1). by binding to cypin and slow- How molecular motors are ing cypin-promoted microtu- specifically targeted to their bule assembly (Chen et al., cargos is a major problem 2005). This phenotype might that remains to be elucidated also be explained by en- in cell biology. Several hanced late endosomal traf- adaptor proteins and the ficking if snapin similarly regulatory molecules are in- modulates late endosomal volved in this process. For trafficking in dendrites. In example, adaptor protein addition, AMPA receptor complex-4 (AP-4) mediates trafficking during LTD might the selective trafficking of be affected in snapin-null AMPA receptors to the soma- neurons. todendritic domain; genetic Finally, snapin is known to disruption of AP-4 results in bind to at least a dozen other the mislocalization of AMPA proteins, such as cypin, receptors, which accumulate SNAP-25, syntaxin 8, and Figure 1. Model for Snapin-Regulated Trafficking of Late in autophagosomes in the Endosomes dysbindin, which are all swollen axons (Matsuda In wild-type neurons (top), snapin links dynein motors to the late endosomes related to a variety of mem- et al., 2008). Many studies and facilitates their retrograde trafficking along microtubules toward the minus brane trafficking and fusion indicate that dynactin helps (–) end. The same pathway is used to transport acid-hydrolase-containing late systems. Importantly, Cai endosomes, which will be fused with immature lysosomes to form mature lyso- to link dynein to its cargo somes. Mature lysosomes are necessary to form autolysosomes. In snapin-null et al. (2010) showed that a (Kardon and Vale, 2009). neurons (bottom), late endosomes cannot be recruited to dynein motors, re- mutant snapin, which cannot Spectrin bIII, which binds sulting in accumulation of standstill late endosomes in axons. Reduced traf- bind to DIC but can presum- ficking of late endosomes also results in accumulation of immature lysosomes to acidic phospholipids on and reduction of autolysosomes. ably still bind to other part- axonal vesicles, has previ- ners, played a dominant- ously been shown to interact negative role in retrograde with the ARP1 filament of dynactin, and anisms by which snapin recognizes late transport of lysosomes, indicating that the Rab7 effector Rab-interacting lyso- endosomes that are ready for transporting binding to DIC is essential for the function somal protein (RILP) is suggested to be along microtubules and releases them at of snapin in regulating late endosomal involved in this step (Kardon and Vale, appropriate sites. Second, the exact trafficking. Nevertheless, it is not com- 2009). In contrast, Cai et al. (2010) mechanisms whereby snapin facilitates pletely clear whether and how other asso- propose that snapin links dynein motors late endosome-lysosome fusion are ciated proteins are involved in the recruit- to late endosomes by directly binding to unclear. Snapin may simply use dynein ment of snapin to late endosomes and DIC. Indeed, binding between the dynein motors to move late endosomes to perinu- their fusion with lysosomes. Furthermore, heavy chain and dynactin was not clear locations where lysosomes are snapin is also reported as a component of affected in snapin-null neurons. There- concentrated and facilitate fusion biogenesis of lysosome-related organ- fore, the identification of a snapin/DIC frequency (Figure 1), but it may also recruit elles complex-1 (BLOC-1), which is complex provides new insights on the other proteins necessary for fusion. Alter- required for the biogenesis of specialized mechanisms responsible for late natively, the transport defects could be lysosome-related organelles, such as endosome-lysosome trafficking in neu- secondary to a fusion step required for melanosomes and platelet-dense gran- rons. Since activation of the autophagic cargo loading and transport. Third, it would ules (Starcevic and Dell’Angelica, 2004). pathway is protective against Hunting- be interesting to know whether the retro- Although mutation in BLOC-1 gene ton’s disease and related conditions grade trafficking of autophagosomes and results in Hermansky-Pudlak syndrome (Wong and Cuervo, 2010), snapin could their fusion competency are also regulated (HPS)-like specific phenotypes character- also be a potential therapeutic target in by snapin. If so, reduced autophagic activ- ized by hypopigmentation and platelet the regulation of autophagy. ities in snapin-null neurons may also be storage pool deficiency, snapin-null There are several issues that remain to affected by these processes. Fourth, it is mice die just after birth and cultured be solved. First, it is unclear how the spec- unclear whether and how snapin regulates neurons display axonal swelling and trin/dynactin-based cargo recognition retrograde transport of late endosomes in reduced survival in vitro, indicating that process is related to the snapin/DIC-based dendrites. Unlike in axons, microtubules phenotypes of snapin-null mice cannot pathway. This question is related to mech- in dendrites bind microtubule-associated be explained by the defective BLOC-1.

Neuron 68, October 7, 2010 ª2010 Elsevier Inc. 5 Neuron Previews

Nevertheless, the role of snapin in regu- Cai, Q., Lu, L., Tian, J.-H., Zhu, Y.-B., Qiao, H., and Pan, P.Y., Tian, J.H., and Sheng, Z.H. (2009). 68 61 lating late endosomal trafficking and lyso- Sheng, Z.-H. (2010). Neuron , this issue, 73–86. Neuron , 412–424. somal fusion could be related to the Chen, M., Lucas, K.G., Akum, B.F., Balasingam, G., function of BLOC-1 and novel forms of Stawicki, T.M., Provost, J.M., Riefler, G.M., Jo¨ rns- Ravikumar, B., Acevedo-Arozena, A., Imarisio, S., HPS. The perinatal lethality of snapin-null ten, R.J., and Firestein, B.L. (2005). Mol. Biol. Cell Berger, Z., Vacher, C., O’Kane, C.J., Brown, S.D., 16, 5103–5114. 37 mice, which may be related to such and Rubinsztein, D.C. (2005). Nat. Genet. , 771–776. a multivalent role of snapin in coordinating Conde, C., and Ca´ ceres, A. (2009). Nat. Rev. Neu- several membrane transport/fusion sys- rosci. 10, 319–332. tems, has hampered elucidation of the Saftig, P., and Klumperman, J. (2009). Nat. Rev. Eskelinen, E.L. (2005). Autophagy 1, 1–10. 10 specific roles of snapin in vivo. Therefore, Mol. Cell Biol. , 623–635. it would be important to clarify the physi- Hirling, H. (2009). Neuroscience 158, 36–44. ological and pathological role of snapin Starcevic, M., and Dell’Angelica, E.C. (2004). Kardon, J.R., and Vale, R.D. (2009). Nat. Rev. Mol. J. Biol. Chem. 279, 28393–28401. in regulating the late endosome-lysosome Cell Biol. 10, 854–865. pathway in vivo by using snapin-L99K knockin mice. Katsumata, K., Nishiyama, J., Inoue, T., Mizush- ima, N., Takeda, J., and Yuzaki, M. (2010). Autoph- von Zastrow, M., and Sorkin, A. (2007). Curr. Opin. 19 agy 6, 378–385. Cell Biol. , 436–445. REFERENCES Matsuda, S., Miura, E., Matsuda, K., Kakegawa, W., Burkhardt, J.K., Echeverri, C.J., Nilsson, T., and Kohda, K., Watanabe, M., and Yuzaki, M. (2008). Wong, E., and Cuervo, A.M. (2010). Nat. Neurosci. Vallee, R.B. (1997). J. Cell Biol. 139, 469–484. Neuron 57, 730–745. 13, 805–811.

VGLUTs—Potential Targets for the Treatment of Seizures?

Reinhard Jahn1,* 1Department of Neurobiology, Max-Planck-Institute for Biophysical Chemistry, D37077 Go¨ ttingen, Germany *Correspondence: [email protected] DOI 10.1016/j.neuron.2010.09.037

Vesicular glutamate transporters (VGLUTs) load glutamate into synaptic vesicles. In this issue of Neuron, Juge et al. report that ketone bodies compete with chloride-dependent activation of VGLUTs, leading to suppression of glutamate release and seizures. These findings provide a surprising explanation for the efficacy of the ketogenic diet in controlling epilepsy.

The functioning of the mammalian CNS parent that transmitter output may also behind because VGLUTs can only be depends on a delicate balance between be controlled by changing the transmitter studied in vesicles that contain an active excitatory and inhibitory synaptic activi- content of synaptic vesicles (for review proton pump, greatly limiting experimental ties. Glutamate is the major excitatory see Ahnert-Hilger et al., 2003; Edwards, flexibility. To deter researchers even neurotransmitter in the brain, and its re- 2007). further, VGLUTs expressed in nonneuro- lease is tightly regulated. At a given syn- Loading of synaptic vesicles with gluta- nal cells display low activities, and once apse, the amount of glutamate released mate is mediated by a small family of incorporated in the plasma membrane per action potential depends on factors vesicular glutamate transporters termed they ‘‘moonlight’’ as Na+-dependent such as past firing patterns, activity of VGLUTs1–3 (Edwards, 2007). Transport transporters for inorganic phosphate neighboring synapses, retrograde sig- is driven by a proton electrochemical (Aihara et al., 2000; Ni et al., 1994). naling, and second-messenger cascades. gradient across the vesicle membrane, Since the early days of measuring Most commonly, the number of synaptic which is generated by a vacuolar ATPase vesicular glutamate uptake (Disbrow vesicles undergoing exocytosis is regu- (Edwards, 2007). In contrast to the well- et al., 1982; Naito and Ueda, 1985), it is lated, and a vast amount of studies have characterized Na+-dependent glutamate known that uptake depends on chloride addressed this mechanism of synaptic transporters in the plasma membrane, ions at millimolar concentrations. How- plasticity. However, it is becoming ap- our knowledge about VGLUTs is lagging ever, it has not been easy to distinguish

6 Neuron 68, October 7, 2010 ª2010 Elsevier Inc.