Role of Unc51.1 and its binding partners
in CNS axon outgrowth
Toshifumi Tomoda, Jee Hae Kim, Caixin Zhan, and Mary E. Hatten1
Laboratory of Developmental Neurobiology, The Rockefeller University, New York, New York 10021-6399, USA
Previous studies showed that the serine/threonine kinase Unc51.1 is one of the earliest genes in neuronal
differentiation and is required for granule cell axon formation. To examine the mechanism of Unc51.1
regulation of axon extension, we have identified two direct binding partners. The first, SynGAP, a negative
regulator of Ras, is expressed within axons and growth cones of developing granule cells. Overexpression of
SynGAP blocks neurite outgrowth by a mechanism that involves Ras-like GTPase cascade. The second
binding partner is a PDZ domain-containing scaffolding protein, Syntenin, that binds Rab5 GTPase, the
activity of which is attenuated by SynGAP. Thus, our results demonstrate that the Unc51.1-containing
protein complex governs axon formation via Ras-like GTPase signaling and through regulation of the
Rab5-mediated endocytic pathways within developing axons.
[Keywords: Unc51.1; SynGAP; Syntenin; axon formation; vesicular membrane.]
Received September 5, 2003; revised version accepted February 5, 2004.
The synaptic circuitry of the cerebellar cortex develops
through the coordinated outgrowth of the axons and den-
drites of the two principal classes of neurons, the granule
cell and the Purkinje cell, and the subsequent formation
of synaptic junctions among them. Parallel fibers of the
granule cells form just after the granule cell progenitors
exit the cell cycle, as an early step in the program of
neuronal differentiation (Ramon y Cajal 1995). The cer-
ebellar cortex provides an unique system for studying
neurite outgrowth and connectivity. The granule cell,
the most abundant neuron of the central nervous system
(CNS), has facilitated analyses of the molecular and cel-
lular mechanisms that underlie the basic steps in neu-
ronal differentiation. The development of the cerebellar
granule cell has been studied extensively, including
mechanisms of proliferation, neuronal migration, axon
formation, and synaptogenesis in vivo and in vitro.
To gain insight into initial steps of parallel fiber
formation/axon outgrowth, we previously identified
Unc51.1/Unc51.2, two murine homologs of the Cae-
norhabditis elegans unc-51 gene (Tomoda et al. 1999). In
the worm, unc-51 is important for axon elongation
(Hedgecock et al. 1985; Desai et al. 1988; Siddiqui 1990;
McIntire et al. 1992; Ogura et al. 1994). The unc-51/
Unc51.1/Unc51.2 genes belong to a subfamily of protein
serine/threonine kinases. Unc51.1 is expressed in a
number of neuronal populations, including the cerebel-
lar granule cell, during development. The protein is lo-
calized to both axonal shafts and growth cones of extend-
ing axons. Retroviral infection of granule cell precursors
with a kinase-deficient form of Unc51.1 demonstrates
that Unc51.1 is essential for neurite extension/parallel
fiber formation in cerebellar granule neurons (Tomoda et
al. 1999). Others have recently reported that ULK1, a
human ortholog of Unc51.1, interacts with the Golgi-
associated ATPase enhancer of 16 kD (GATE-16) and
with the ?2 subunit of GABA-A receptor associated pro-
tein (GABARAP; Okazaki et al. 2000). The GATE-16 is
an essential factor for intra-Golgi transport (Sagiv et al.
2000), and the GABARAP has a possible role in the regu-
lation of receptor trafficking (Wang et al. 1999). How-
ever, it is not yet known how these Unc51.1 binding
partners function with Unc51.1 to regulate neurite
To further investigate the molecular basis of Unc51.1/
Unc51.2 function in neurite extension, we searched for
additional proteins that bind Unc51.1/Unc51.2 in yeast
two-hybrid screens. Interacting molecules identified in-
cluded SynGAP, a synaptic GTPase-activating protein
(GAP; Chen et al. 1998; Kim et al. 1998), and Syntenin,
an endocytic vesicular membrane protein with PDZ do-
mains (Grootjans et al. 1997; Fialka et al. 1999). SynGAP
interacts with the PDZ domains of the PSD-95/SAP90
family of proteins via its C?-terminal amino acids,
T/SXV (Kim et al. 1998), and is a major component of
postsynaptic density (PSD), a dense cytoskeletal matrix
found beneath the postsynaptic membrane that is nota-
bly prominent in excitatory synapses (Chen et al. 1998).
SynGAP functions as a brain-specific Ras GTPase-acti-
vating protein and is associated in a large complex with
PSD-95/SAP90, SAP102, and the NMDA receptors at ex-
citatory synapses in cortex and hippocampus. Recent
E-MAIL firstname.lastname@example.org; FAX (212) 327-7145.
Article published online ahead of print. Article and publication date are
GENES & DEVELOPMENT 18:541–558 © 2004 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/04; www.genesdev.org541
studies have shown that SynGAP contributes to neural
plasticity by modulating Ras activity in brain (Komi-
yama et al. 2002; Kim et al. 2003). Syntenin contains two
PDZ domains that bind cytoplasmic domains of various
proteins, including syndecan, neurofascin, glutamate re-
ceptors, and ephrin-B2 (Grootjans et al. 1997; Lin et al.
1999; Koroll et al. 2001; Hirbec et al. 2003). Syntenin
often serves as a scaffolding molecule for synaptic pro-
teins (Hirbec et al. 2003).
In this report, we present evidence that SynGAP is
expressed during early stages of granule neuron develop-
ment and that Unc51.1 and SynGAP function in concert
to regulate axon outgrowth of developing granule neu-
rons. By using bioassays of primary granule neurons ex-
pressing either Unc51.1 or SynGAP or both, we show
that Unc51.1 and SynGAP function cooperatively in
axon formation during brain development. We also show
that Unc51.1 affects the extent of SynGAP modulation
of Ras activity, suggesting a functional link between
Unc51.1 and SynGAP that could lead to long-term
changes in gene expression responsible for axon forma-
tion. Syntenin provides a scaffold for Unc51.1 and for
Rab5 GTPase, an endocytic membrane regulator, and
SynGAP stimulates Rab5 GTPase activity in vitro. In
addition, we show that the SynGAP-induced axon
truncation is restored by overexpressing Ras or Rab5
GTPases, and we provide evidence that the dominant-
negative Unc51.1 and SynGAP expression disorganizes
vesicular membranes within extending axons of primary
granule neuron culture. These studies suggest that
Unc51.1 and its binding partners influence axon forma-
tion via the neuronal endocytic pathway.
Unc51.1/Unc51.2 binds SynGAP
Previous studies suggested the importance of the C?-ter-
minal domains of unc-51/Unc51.1/Unc51.2 in their bio-
logical function (Ogura et al. 1994; Tomoda et al. 1999).
Thus, we used the C?-terminal region of Unc51.1 (amino
acids 653–1051) and Unc51.2 (amino acids 531–1037) as
baits to screen a yeast two-hybrid library prepared from
postnatal day 6 (P6) mouse cerebellum. Screening of 3
million clones with Unc51.2 (amino acids 531–1037) and
subsequent sequencing analyses revealed that one clone
spanning ∼1.7 kb encoded the C?-terminal 500 amino
acids of SynGAP (amino acids 829–1328). To examine
whether SynGAP could also bind Unc51.1, the C?-termi-
nal region of Unc51.1 (amino acids 653–1051) was tested
in a yeast two-hybrid assay, and its association with Syn-
GAP was confirmed (Fig. 1). This suggests that Unc51.1
and Unc51.2 may share similar structural features
within their C?-terminal domains, as is expected from
their primary sequences (70.8% similarity), and that
they may function in overlapping molecular pathways.
SynGAP exists as multiple alternatively spliced forms.
There are at least seven variants in the 3? portion of the
SynGAP mRNA (?1, ?2, ?1, ?2, ?3, ?4, and ?), and they
encode five different protein isoforms (Li et al. 2001).
The SynGAP clone isolated in our screen corresponded
to the ?2 isoform that encodes a protein 35 amino acids
longer than that encoded by the ?1 isoform, an isoform
originally reported (Kim et al. 1998; Chen et al. 1998).
The ?1 isoform ends with TRV, through which it binds
a PDZ domain of PSD-95. The ?2 isoform has one base
insertion within the C?-terminal tail region of the pro-
tein compared with the ?1 isoform (Li et al. 2001), ends
with ADH, and does not bind PSD-95 (data not shown).
Both the ?1 and ?2 isoforms were expressed in P6 cer-
ebellum as confirmed by RT–PCR, and both bound the
Unc51.1/Unc51.2 baits, although the ?2 isoform consis-
tently displayed slightly higher affinity to the baits than
did the ?1 isoform in a yeast two-hybrid assay (data not
shown). We reasoned that the C?-terminal sequence of
SynGAP functions as a tag that determines subcellular
localizations. We therefore generated GFP fusion pro-
teins with the C?-tail domain of SynGAP ?1 or ?2, in-
troduced them retrovirally into granule cells in culture,
and found that the GFP-C?-tail(?2) was localized to ex-
tending axons but the GFP-C?-tail(?1) was only localized
to cell soma (data not shown). This demonstrates that
the ?1 isoform with TRV motif could be the isoform that
partitions into postsynaptic compartments, whereas the
?2 isoform is the one that preferentially goes to axons.
Therefore, we used only the ?2 isoform in subsequent
To determine the minimum region of Unc51.1 that
binds SynGAP, we generated a series of deletion con-
structs of Unc51.1 and tested them with SynGAP (amino
acids 829–1328) in a ?-galactosidase assay in yeast (Fig.
1A). Although the C?-terminal 399 amino acids of
Unc51.1 (amino acids 653–1051), as well as the shorter
C?-terminal domain of Unc51.1 (amino acids 829–1051),
strongly bound SynGAP, the spacer region (amino acids
653–828) and subregions of the C?-terminal domain
(amino acids 829–1001 and amino acids 913–1051) did
not bind SynGAP. This result demonstrates that the C?-
terminal domain of Unc51.1 (amino acids 829–1051) is
the minimum requirement for the binding to SynGAP,
and suggests an important role of this domain as a func-
tional unit. It is noteworthy that this domain is highly
conserved between the C. elegans unc-51 and the two
murine homologs, Unc51.1 and Unc51.2 (Tomoda et al.
1999). To further confirm the binding of Unc51.1 to Syn-
GAP, a glutathione S-transferase (GST) pull-down assay
was performed. The bait regions of Unc51.1 (amino acids
653–1051) and Unc51.2 (amino acids 531–1037) were
fused in-frame with a GST tag and tested for binding to
theC?-terminal region of
829–1328) and its deletion mutants, D1 and D2 (amino
acids 829–1053 and amino acids 1054–1328, respec-
tively). In these assays, GST–Unc51.1 and GST–Unc51.2
specifically bound SynGAP (Fig. 1B). Smaller deletions
within the C? terminus of SynGAP did not bind to the
GST fusion proteins, suggesting that the bulk of the C?-
terminal domain of SynGAP is involved in binding to
Unc51.1/Unc51.2. To determine the minimum region of
SynGAP that binds Unc51.1, a series of deletion con-
structs of SynGAP (D1-D5) were generated and tested
Tomoda et al.
542GENES & DEVELOPMENT
with the bait region of Unc51.1 in a yeast two-hybrid
assay. Although deletion mutants, D1 and D2, did not
bind Unc51.1 as was the case in the GST pull-down as-
say, the central region of SynGAP C?-tail (D5, amino
acids 886–1287) was necessary for the binding (Fig. 1C).
To examine whether the full-length Unc51.1 binds the
full-length SynGAP, a coimmunoprecipitation assay was
done in a mammalian heterologous expression system.
HEK293T cells were transfected with the full-length
SynGAP expression construct with or without a con-
struct expressing myc-tagged Unc51.1. The cell extracts
were immunoprecipitated with a myc antibody, and the
resulting immune complexes were tested for the pres-
ence of SynGAP. The antibody we used recognizes the
GAP domain of SynGAP (Kim et al. 2003). SynGAP spe-
cifically interacted with Unc51.1 in this assay (Fig. 1D).
To confirm interaction of Unc51.1 with SynGAP in
vivo, membrane-enriched fractions were prepared from
P6 cerebellum and immunoprecipitated with an anti-
SynGAP antibody to test for the presence of Unc51.1 in
the resulting immune complex. Unc51.1 was specifically
detected in the immunoprecipitates with SynGAP but
not in the complex immunoprecipitated with the control
normal rabbit IgG protein (Fig. 1E). Taken together, the
results indicate that the Unc51.1 and SynGAP interact
via their C?-terminal domains.
SynGAP expression during cerebellar development
To further characterize SynGAP expression during cer-
ebellar development, we prepared protein extracts from
varying developmental stages of the cerebellum, starting
653–1051) and its deletion mutants were tested for their ability to bind SynGAP ?2 C?-tail (amino acids 829–1328) in a ?-galactosidase
filter assay. Relative binding strength, as judged by darkness of the blue color of representative colonies, is indicated: (+++) very strong;
(++) strong; (+) weak; (−) no binding detected. (B) GST pull-down assay. The C?-tail of SynGAP (amino acids 829–1328) and its deletion
mutants D1 (amino acids 829–1053) and D2 (1054–1328) were in vitro translated, labeled with [35S]-methionine, and applied to the
columns made with GST fusion proteins (GST alone, GST–Unc51.1 C?-tail [amino acids 653–1051], or GST–Unc51.2 C?-tail [amino
acids 531–1037]). Bound proteins were eluted from the columns and analyzed with SDS-PAGE, followed by autoradiography. (C)
Determination of the minimum region of SynGAP that binds Unc51.1. Various deletion mutants (D1–D5) of the C?-tail of SynGAP
were tested for their ability to bind Unc51.1 C?-terminal domain (amino acids 653–1051) in a ?-galactosidase assay. (D) Coprecipitation
assay. A full-length SynGAP expression construct was transfected into HEK293T cells with or without the myc-tagged full-length
Unc51.1 construct. Cell extracts were immunoprecipitated with a myc antibody, and the immune complex was analyzed by immu-
noblot using an anti-SynGAP antibody. (E) Membrane-enriched fractions extracted from P6 cerebellum were immunoprecipitated with
either the anti-SynGAP antibody or the normal rabbit IgG, and the resulting immune complexes were analyzed with SDS-PAGE
followed by immunoblot with an anti-Unc51.1 antibody and an anti-SynGAP antibody.
Unc51.1 binds SynGAP. (A) Summary of the yeast two-hybrid assay. The C?-terminal half of Unc51.1 (amino acids
Unc51.1, SynGAP, Syntenin, and Rab5 in axon formation
GENES & DEVELOPMENT 543
(supernatant). For step-wise sucrose gradient fractionation ex-
periments, the 700 × g supernatant (0.5 mL) was loaded on a
10%–40% sucrose gradient (10%, 20%, 30%, and 40% sucrose,
1 mL each), and centrifuged at 37,000 × g in SW55i rotor (Beck-
man) for 3 h at 4°C. After centrifugation, nine fractions (0.5 mL
each) were recovered from the top of the gradient, and the pro-
tein concentration in each fraction was measured by Bradford
method according to the protocol provided (Pierce). Ten micro-
grams of proteins per each fraction were applied to SDS-PAGE
and analyzed by Western blotting using the following antibod-
ies: anti-Unc51.1 (guinea pig polyclonal, 1:100), anti-SynGAP/
GAP domain (rabbit polyclonal, 1:200), anti-Rab3A (rabbit poly-
clonal, Santa Cruz, 1:200), anti-Rab5 (rabbit polyclonal, Santa
Cruz, 1:200), anti-PSD-95 (mouse monoclonal IgG2a, Upstate
Biotechnology, 1:10,000), anti-LDH (Chemicon, goat poly-
clonal, 1:500), and anti-actin (monoclonal, Amersham-Pharma-
Plasmids for Unc51.1 were as described (Tomoda et al. 1999).
Expression constructs for SynGAP were provided from Richard
Huganir (Kim et al. 1998). The yeast two-hybrid library was
custom-made by GIBCO-BRL by using 15 µg of mRNA ex-
tracted from P6 mouse cerebellum. The cDNAs for Rab3A,
Rab6, and Rab7 were amplified by PCR from P6 mouse cerebel-
lum. Retroviral constructs were made by subcloning full-length
coding regions of cDNAs for mouse Unc51.1, rat SynGAP, and
human Rab5A into EcoR1/Not1 site of the pMSCX–?-actin–
EGFP retroviral vector (a gift from David Solecki, The Rock-
efeller University, New York). For bicistronic expression of
cDNAs with the retroviral expression system, one cDNA was
inserted into the multicloning site of the pMSCX vector fol-
lowed by the human ?-actin promoter plus the second cDNA
coding region N?-terminally tagged with EGFP. The EGFP was
used as a morphological marker for the infected neurons.
Production of recombinant retroviruses
Recombinant ecotropic replication-incompetent retroviruses
were produced as described (Tomoda et al. 1999). In brief, 293
cells were cotransfected with the recombinant retroviral con-
structs together with a pCL-Eco packaging construct (IMGEN).
Twenty-four hours posttransfection, media was replaced with
the granule cell media (Hatten 1985), and the culture superna-
tant containing the retroviruses was harvested 24 and 48 h later.
Virus titers were estimated by infecting NIH3T3 cells with vi-
rus supernatants with serial dilution. Typically, the titers
ranged from 4 × 105to 1 × 107cfu/mL.
Preparation and infection of primary granule cells
Cerebellar granule cells were purified from early postnatal
C57BL/6J mice as described previously (Hatten 1985); 1 to
2 × 106cells were cultured as cellular reaggregates in one well of
a uncoated 16-well Lab-Tek glass chamber slides (Nunc), and
infected with recombinant viruses overnight. After 18–24 h, the
cultures were gently resuspended at 1 × 106cells per milliliter
and plated in Lab-Tek slides coated with poly-L-lysine (Sigma)
and Matrigel (Becton Dickinson).
Antibody staining was as described (Gao et al. 1991). For im-
munoelectron microscopy, the affinity-purified rabbit anti-
Unc51.1 antibody was used (Tomoda et al. 1999). For detection
of Unc51.1 and SynGAP protein expression with cerebellum, P6
mice were perfused with 4% paraformaldehyde/PBS and post-
fixed in the same fixative containing 30% sucrose overnight at
4°C. Sagittal sections (60 µm) were made with a cryotome
(Leica, SM2000R). To detect endogenous Unc51.1 protein, anti-
Unc51.1 serum was generated by immunizing guinea pigs with
the C?-terminal domain of Unc51.1 protein (amino acids 653–
1051) prepared as a GST fusion protein in bacteria, and the
resulting antiserum was affinity-purified by using the same do-
main of Unc51.1 as has been used for immunization. The speci-
ficity of the antibody was confirmed by Western blot using pro-
tein extracts prepared from P6 and adult mouse brain (data not
shown). The dilution of other antibodies used was anti-TAG-1
antibody (monoclonal, IgM; 4D7; 1:2), anti-class III ?-tubulin
antibody (monoclonal, IgG2a?; TUJ1; 1:1000, Covance). Speci-
mens were viewed either with an epifluorescence illumination
(Axiophot 100 microscope, Zeiss) or with a confocal scanning
system Radiance 2000 (Bio-Rad). For detection of infected gran-
ule cells in dissociated culture, cells were fixed in 4% paraform-
aldehyde and probed with the anti-GFP antibody (Molecular
Probes, polyclonal, 1:2000).
We are indebted to Helen Shio for electron microscopy, Yin
Fang and Hirofumi Toda for expert technical assistance, Tom-
oyo Okada and Akihisa Mino for advice on GAP assay, Niels
Adams for help with imaging, and Akira Sawa for advice on
yeast two-hybrid. We also thank the following people for pro-
viding reagents: Richard Huganir for a SynGAP GAP domain
antibody and SynGAP plasmids; Anat Ikin for Rab5 cDNA; Ha-
ruhiko Bito for RhoA, Rac1, and Cdc42 cDNAs; and David
Solecki for a retroviral vector. We are grateful to Nat Heintz,
Richard Huganir, Angus Nairn, Kathy Zimmerman, and David
Solecki for critically reading the manuscript. Supported by
grants from March of Dimes (1-FY99-510) and NIH (NS39991) to
The publication costs of this article were defrayed in part by
payment of page charges. This article must therefore be hereby
marked “advertisement” in accordance with 18 USC section
1734 solely to indicate this fact.
Bar-Sagi, D. and Feramisco, J.R. 1985. Microinjection of the ras
oncogene protein into PC12 cells induces morphological dif-
ferentiation. Cell 42: 841–848.
Callebaut, I., de Gunzburg, J., Goud, B., and Mornon, J.P. 2001.
RUN domains: A new family of domains involved in Ras-
like GTPase signaling. Trends Biochem. Sci. 26: 79–83.
Castellani, V. and Rougon, G. 2002. Control of semaphorin sig-
naling. Curr. Opin. Neurobiol. 12: 532–541.
Chang, L. and Karin, M. 2001. Mammalian MAP kinase signal-
ling cascades. Nature 410: 37–40.
Chen, H.J., Rojas-Soto, M., Oguni, A., and Kennedy, M.B. 1998.
A synaptic Ras-GTPase activating protein (p135 SynGAP)
inhibited by CaM kinase II. Neuron 20: 895–904.
Crump, J.G., Zhen, M., Jin, Y., and Bargmann, C.I. 2001. The
SAD-1 kinase regulates presynaptic vesicle clustering and
axon termination. Neuron 29: 115–129.
Desai, C., Garriga, G., McIntire, S.L., and Horvitz, H.R. 1988. A
genetic pathway for the development of the Caenorhabditis
elegans HSN motor neurons. Nature 336: 638–646.
Dickson, B.J. 2002. Molecular mechanisms of axon guidance.
Science 298: 1959–1964.
Tomoda et al.
556GENES & DEVELOPMENT
Eggert, A., Ikegaki, N., Liu, X., Chou, T.T., Lee, V.M., Tro-
janowski, J.Q., and Brodeur, G.M. 2000. Molecular dissection
of TrkA signal transduction pathways mediating differentia-
tion in human neuroblastoma cells. Oncogene 19: 2043–2051.
Ferro-Novick, S. and Novick, P. 1993. The role of GTP-binding
proteins in transport along the exocytic pathway. Annu.
Rev. Cell Biol. 9: 575–599.
Fialka, I., Steinlein, P., Ahorn, H., Bock, G., Burbelo, P.D.,
Haberfellner, M., Lottspeich, F., Paiha, K., Pasquali, C., and
Huber, L.A. 1999. Identification of syntenin as a protein of
the apical early endocytic compartment in Madin-Darby ca-
nine kidney cells. J. Biol. Chem. 274: 26233–26239.
Fischer von Mollard, G., Stahl, B., Walch-Solimena, C., Takei,
K., Daniels, L., Khoklatchev, A., De Camilli, P., Sudhof,
T.C., and Jahn, R. 1994. Localization of Rab5 to synaptic
vesicles identifies endosomal intermediate in synaptic
vesicle recycling pathway. Eur. J. Cell Biol. 65: 319–326.
Forcet, C., Stein, E., Pays, L., Corset, V., Llambi, F., Tessier-
Lavigne, M., and Mehlen, P. 2002. Netrin-1-mediated axon
outgrowth requires deleted in colorectal cancer-dependent
MAPK activation. Nature 417: 443–447.
Gao, W.-Q., Heintz, N., and Hatten, M.E. 1991. Cerebellar gran-
ule cell neurogenesis is regulated by cell–cell interactions in
vitro. Neuron 6: 705–715.
Georgiou, M. and Tear, G. 2002. Commissureless is required
both in commissural neurones and midline cells for axon
guidance across the midline. Development 129: 2947–2956.
Grootjans, J.J., Zimmermann, P., Reekmans, G., Smets, A., De-
geest, G., Durr, J., and David, G. 1997. Syntenin, a PDZ
protein that binds syndecan cytoplasmic domains. Proc.
Natl. Acad. Sci. 94: 13683–13688.
Harris, B.Z. and Lim, W.A. 2001. Mechanism and role of PDZ
domains in signaling complex assembly. J. Cell Sci.
Hatten, M.E. 1985. Neuronal regulation of astroglial morphol-
ogy and proliferation in vitro. J. Cell Biol. 100: 384–396.
Hedgecock, E.M., Culotti, J.G., Thomson, J.N., and Perkins,
L.A. 1985. Axonal guidance mutants of Caenorhabditis el-
egans identified by filling sensory neurons with fluorescein
dyes. Dev. Biol. 111: 158–170.
Hirbec, H., Francis, J.C., Lauri, S.E., Braithwaite, S.P., Coussen,
F., Mulle, C., Dev, K.K., Coutinho, V., Meyer, G., Isaac, J.T.,
et al. 2003. Rapid and differential regulation of AMPA and
kainate receptors at hippocampal mossy fibre synapses by
PICK1 and GRIP. Neuron 37: 625–638.
Hopkins, C.R., Gibson, A., Shipman, M., and Miller, K. 1990.
Movement of internalized ligand-receptor complexes along a
continuous endosomal reticulum. Nature 346: 335–339.
Jurney, W.M., Gallo, G., Letourneau, P.C., and McLoon, S.C.
2002. Rac1-mediated endocytosis during ephrin-A2- and
semaphorin 3A-induced growth cone collapse. J. Neurosci.
Kamada, Y., Funakoshi, T., Shintani, T., Nagano, K., Ohsumi,
M., and Ohsumi, Y. 2000. Tor-mediated induction of au-
tophagy via an Apg1 protein kinase complex. J. Cell Biol.
Kamiguchi, H. and Lemmon, V. 2000. Recycling of the cell ad-
hesion molecule L1 in axonal growth cones. J. Neurosci.
Keleman, K., Rajagopalan, S., Cleppien, D., Teis, D., Paiha, K.,
Huber, L.A., Technau, G.M., and Dickson, B.J. 2002. Comm
sorts robo to control axon guidance at the Drosophila mid-
line. Cell 110: 415–427.
Kim, J.H., Liao, D., Lau, L.F., and Huganir, R.L. 1998. SynGAP:
A synaptic RasGAP that associates with the PSD-95/SAP90
protein family. Neuron 20: 683–691.
Kim, J.H., Lee, H.K., Takamiya, K., and Huganir, R.L. 2003. The
role of synaptic GTPase-activating protein in neuronal de-
Klionsky, D.J. and Emr, S.D. 2000. Autophagy as a regulated
pathway of cellular degradation. Science 290: 1717–1721.
Komiyama, N.H., Watabe, A.M., Carlisle, H.J., Porter, K.,
Charlesworth, P., Monti, J., Strathdee, D.J., O’Carroll, C.M.,
Martin, S.J., Morris, R.G., et al. 2002. SynGAP regulates
ERK/MAPK signaling, synaptic plasticity, and learning in
the complex with postsynaptic density 95 and NMDA re-
ceptor. J. Neurosci. 22: 9721–9732.
Koroll, M., Rathjen, F.G., and Volkmer, H. 2001. The neural cell
recognition molecule neurofascin interacts with syntenin-1
but not with syntenin-2, both of which reveal self-associat-
ing activity. J. Biol. Chem. 276: 10646–10654.
Laemmli, U.K. 1970. Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227: 680–
Lamaze, C., Chuang, T.H., Terlecky, L.J., Bokoch, G.M., and
Schmid, S.L. 1996. Regulation of receptor-mediated endocy-
tosis by Rho and Rac. Nature 382: 177–179.
Li, W., Okano, A., Tian, Q.B., Nakayama, K., Furihata, T.,
Nawa, H., and Suzuki, T. 2001. Characterization of a novel
synGAP isoform, synGAP-?. J. Biol. Chem. 276: 21417–
Lin, R.C. and Scheller, R.H. 2000. Mechanisms of synaptic
vesicle exocytosis. Annu. Rev. Cell Dev. Biol. 16: 19–49.
Lin, D., Gish, G.D., Songyang, Z., and Pawson, T. 1999. The
carboxyl terminus of B class ephrins constitutes a PDZ do-
main binding motif. J. Biol. Chem. 274: 3726–3733.
Liu, K. and Li, G. 1998. Catalytic domain of the p120 Ras GAP
binds to RAb5 and stimulates its GTPase activity. J. Biol.
Chem. 273: 10087–10090.
Marshall, C.J. 1996. Ras effectors. Curr. Opin. Cell Biol. 8: 197–
Matsuura, A., Tsukada, M., Wada, Y., and Ohsumi, Y. 1997.
Apg1p, a novel protein kinase required for the autophagic
process in Saccharomyces cerevisiae. Gene 192: 245–250.
McIntire, S.L., Garriga, G., White, J., Jacobson, D., and Horvitz,
H.R. 1992. Genes necessary for directed axonal elongation or
fasciculation in C. elegans. Neuron 8: 307–322.
Mueller, B.K. 1999. Growth cone guidance: First steps towards
a deeper understanding. Annu. Rev. Neurosci. 22: 351–
Noda, M., Ko, M., Ogura, A., Liu, D.G., Amano, T., Takano, T.,
and Ikawa, Y. 1985. Sarcoma viruses carrying ras oncogenes
induce differentiation-associated properties in a neuronal
cell line. Nature 318: 73–75.
Ogura, K., Wicky, C., Magnenat, L., Tobler, H., Mori, I., Muller,
F., and Ohshima, Y. 1994. Caenorhabditis elegans unc-51
gene required for axonal elongation encodes a novel serine/
threonine kinase. Genes & Dev. 8: 2389–2400.
Ogura, K., Shirakawa, M., Barnes, T.M., Hekimi, S., and Oh-
shima, Y. 1997. The UNC-14 protein required for axonal
elongation and guidance in Caenorhabditis elegans interacts
with the serine/threonine kinase UNC- 51. Genes & Dev.
Ohsumi, Y. 2001. Molecular dissection of autophagy: Two ubiq-
uitin-like systems. Nat. Rev. Mol. Cell Biol. 2: 211–216.
Okazaki, N., Yan, J., Yuasa, S., Ueno, T., Kominami, E., Ma-
suho, Y., Koga, H., and Muramatsu, M. 2000. Interaction of
the Unc-51-like kinase and microtubule-associated protein
light chain 3 related proteins in the brain: Possible role of
vesicular transport in axonal elongation. Brain Res. Mol.
Brain Res. 85: 1–12.
Unc51.1, SynGAP, Syntenin, and Rab5 in axon formation
GENES & DEVELOPMENT 557
Olson, M.F. and Marais, R. 2000. Ras protein signalling. Semin.
Immunol. 12: 63–73.
Perron, J.C. and Bixby, J.L. 1999. Distinct neurite outgrowth
signaling pathways converge on ERK activation. Mol. Cell
Neurosci. 13: 362–378.
Pfeffer, S.R. 1994. Rab GTPases: Master regulators of membrane
trafficking. Curr. Opin. Cell Biol. 6: 522–526.
Ramon y Cajal, S. 1995. The cerebellum. In Histology of the
nervous system of man and vertebrates. (ed. L.W. Swanson;
trans. N. Swanson), pp. 3–123. Oxford University Press, New
Sagiv, Y., Legesse-Miller, A., Porat, A., and Elazar, Z. 2000.
GATE-16, a membrane transport modulator, interacts with
NSF and the Golgi v-SNARE GOS-28. EMBO J. 19: 1494–
Shimizu, H., Kawamura, S., and Ozaki, K. 2003. An essential
role of Rab5 in uniformity of synaptic vesicle size. J. Cell Sci.
Shio, H. and Lazarow, P.B. 1981. Relationship between peroxi-
somes and endoplasmic reticulum investigated by combined
catalase and glucose-6-phosphatase cytochemistry. J. Histo-
chem. Cytochem. 29: 1263–1272.
Siddiqui, S.S. 1990. Mutations affecting axonal growth and guid-
ance of motor neurons and mechanosensory neurons in the
Somsel Rodman, J. and Wandinger-Ness, A. 2000. Rab GTPases
coordinate endocytosis. J. Cell Sci. 113: Pt 2, 183–192.
Tall, G.G., Barbieri, M.A., Stahl, P.D., and Horazdovsky, B.F.
2001. Ras-activated endocytosis is mediated by the Rab5
guanine nucleotide exchange activity of RIN1. Dev. Cell
Tessier-Lavigne, M. and Goodman, C.S. 1996. The molecular
biology of axon guidance. Science 274: 1123–1133.
Tomoda, T., Bhatt, R.S., Kuroyanagi, H., Shirasawa, T., and Hat-
ten, M.E. 1999. A mouse serine/threonine kinase homolo-
gous to C. elegans UNC51 functions in parallel fiber forma-
tion of cerebellar granule neurons. Neuron 24: 833–846.
Wang, H., Bedford, F.K., Brandon, N.J., Moss, S.J., and Olsen,
R.W. 1999. GABA(A)-receptor-associated
GABA(A) receptors and the cytoskeleton. Nature 397: 69–
Zerial, M. and McBride, H. 2001. Rab proteins as membrane
organizers. Nat. Rev. Mol. Cell Biol. 2: 107–117.
Zimmer, M., Palmer, A., Kohler, J., and Klein, R. 2003. EphB–
ephrinB bi-directional endocytosis terminates adhesion al-
lowing contact mediated repulsion. Nat. Cell Biol. 5: 869–
Tomoda et al.
558 GENES & DEVELOPMENT