JOURNAL OF NEUROCHEMISTRY | 2010 | 113 | 153–165
*Département de pathologie et biologie cellulaire and Groupe de recherche sur le système nerveux central, Université de Montréal,
Montréal, QC, Canada
 Department of Neurology and Neurosurgery, Montreal Neurological Institute, McGill University, Montréal, QC, Canada
àDepartment of Neurology and Neurosurgery, McGill University, and Centre for Research in Neuroscience, Montreal General
Hospital, Montréal, QC, Canada
§The Burnham Institute for Medical Research, La Jolla, CA, USA
Abstract
EphA4, a receptor tyrosine kinase, is expressed in various
pre-, post- and peri-synaptic organelles and implicated in the
regulation of morphological and physiological properties of
synapses. It regulates synaptic plasticity by acting as a
binding partner for glial ephrin-A3 and possibly other pre- or
post-synaptic ephrins. Now, its trafficking mechanisms remain
unknown. In this study, we examine the association of EphA4
with transport, clathrin-coated and synaptic vesicles using cell
fractionation, vesicle immunoisolation and electron micros-
copy. EphA4 was found in highly purified fractions of clathrin-
coated or synaptic vesicles. It was also detected in vesicles
immuno-isolated with antibodies anti-synaptophysin, anti-
vesicular glutamate transporter or anti-vesicular GABA
transporter; demonstrating its presence in synaptic vesicles.
However, it was not detected in immuno-isolated piccolo–
Eph receptors, the largest family of tyrosine kinase receptors,
count 14 members in mammals, divided in two subclasses,
EphAs (A1–8 and A10) and EphBs (B1–4 and B6), based on
their relative binding affinity for ephrin ligands (Himanen
and Nikolov 2003). Ephrin-As (A1–A5) generally stimulate
EphAs while ephrin-Bs (B1–B3) activate EphBs. EphA4,
however, is known to be activated by both forms of ephrins.
The functions of these proteins have been mostly studied in
the CNS development, where they participate in various
processes such as cell migration, segmentation patterning and
axonal growth cone guidance (Klein 2004; Pasquale 2005).
Most Eph receptors are also expressed in the mature CNS
(Goldshmit et al. 2006), where they have been implicated
 2010 The Authors
Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
doi: 10.1111/j.1471-4159.2010.06582.
bassoon transport vesicles. In vivo and in dissociated cultures,
EphA4 was localized by immunoelectron microscopy in
vesicular glutamate transporter 1-positive terminals of hippo-
campal neurons. Remarkably, the cell surface immuno-
fluorescence of EphA4 increased markedly in cultured
hippocampal neurons following KCl depolarization. These
observations indicate that EphA4 is present in subsets of
synaptic vesicles, can be externalized during depolarization,
and internalized within clathrin-coated vesicles. This traffick-
ing itinerary may serve to regulate the levels of EphA4 in the
synaptic plasma membrane and thereby modulate signaling
events that contribute to synaptic plasticity.
Keywords: cell fractionation, electron microscopy, EphA4,
hippocampus, immunocytochemistry, immunoisolation, vesi-
cles.
J. Neurochem. (2010) 113, 153–165.
Received September 14, 2009; revised manuscript received December
31, 2009; accepted January 5, 2010.
Address correspondence and reprint requests to Guy Doucet,
Département de pathologie et biologie cellulaire, Faculté de médecine,
Université de Montréal, C.P. 6128, succ. Centre-ville, Montréal, QC,
Canada H3C 3J7. E-mail: guy.doucet@umontreal.ca
Abbreviations used: BSA, bovine serum albumin; CCVs, clathrin-
coated vesicles; Cdk5, cyclin-dependent kinase-5; CHC, clathrin heavy
chain; DIV15, 15 days in vitro; PAZ, pre-synaptic active zone; PBS,
phosphate-buffered saline; EM, electron microscopy; PFA, paraformal-
dehyde; PLCc, phospholipase-C gamma; PSD, post-synaptic density;
PTVs, Piccolo-bassoon transport vesicles; SVs, synaptic vesicles;
VGAT, vesicular GABA transporter; VGLUT, vesicular glutamate
transporter.
153
154 | D. Bouvier et al.
in synaptogenesis and synaptic plasticity (Murai and
Pasquale 2004; Yamaguchi and Pasquale 2004). Defects
in long-term potentiation or long-term depression have been
reported in EphA4, EphB2, ephrin-A3, ephrin-B2 or ephrin-
B3 null mice (Henderson et al. 2001; Contractor et al.
2002; Grunwald et al. 2004; Armstrong et al. 2006;
Rodenas-Ruano et al. 2006; Bouzioukh et al. 2007; Filosa
et al. 2009; Klein 2009), but the precise roles of these
molecules in synaptic plasticity are still not well under-
stood. The role of Eph receptors in synaptic activity and
plasticity has generally been interpreted with the view of
post-synaptic receptor localization, but EphA4 can be found
in pre-synaptic terminals and perisynaptic astrocytic pro-
cesses, as well as in post-synaptic dendritic spines (Tremb-
lay et al. 2007; Bouvier et al. 2008). Recent reports have
demonstrated that the role of EphA4 in the long-term
potentiation induced in CA1 by theta-burst stimulation
depends on its post-synaptic localization via surface inter-
actions with ephrin-A3 expressed in astrocytes (Carmona
et al. 2009; Filosa et al. 2009); as does its influence on the
morphogenesis of dendritic spines, although through differ-
ent signaling mechanisms (Murai et al. 2003; Zhou et al.
2007). Thus, its roles in pre-synaptic terminals or in
perisynaptic astrocytic leaflets remain obscure. Furthermore,
there is still little knowledge of the mechanisms that control
EphA4 trafficking and levels on the neuronal surface
(Deininger et al. 2008).
We recently demonstrated that EphA4 is located essen-
tially at synapses in the mature cerebral cortex and
hippocampus (Tremblay et al. 2007; Bouvier et al. 2008).
Using cell fractionation and immunoelectron microscopy,
EphA4 was observed at various compartments of pre-
synaptic terminals, particularly the plasma membrane, the
pre-synaptic active zone (PAZ) and subpopulations of
vesicles, as well as the plasma membrane and post-synaptic
density (PSD) of dendritic spines. The association of
EphA4 with vesicles, PAZs and PSDs suggests it may
have roles related to neurotransmitter release or synaptic
plasticity.
To further investigate this possibility, we examined the
association of EphA4 with various types of vesicles includ-
ing clathrin-coated vesicles (CCVs), synaptic vesicles (SVs)
and piccolo-bassoon transport vesicles (PTVs). We used
synaptosomal fractionation, immunoisolation of various
types of vesicles, and immunoelectron microscopy to localize
and characterize the EphA4-containing vesicles. The results
show that EphA4 is associated with SVs and CCVs, as well
as with synaptophysin-, vesicular glutamate transporter
(VGLUT)- and vesicular GABA transporter (VGAT)-posi-
tive SVs in axon terminals and dendrites. Finally, its surface
expression was increased following KCl-induced depolar-
ization of hippocampal and cortical neurons in culture. These
results are consistent with an important role for EphA4 in
synaptic function and plasticity.
 2010 The Authors
Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
Materials and methods
Animals
All procedures were conducted in strict accordance with the Guide
to the Care and Use of Experimental Animals (Ed2) of the Canadian
Council on Animal Care. The experimental protocols were approved
by the Animal Care Committee of the Université de Montreal.
Cell fractionation and immunoisolation experiments were con-
ducted on 14 adult male C57BL/6 mice purchased from Charles
River Laboratories Inc., St-Constant, QC, Canada. These animals
were decapitated under deep CO2 anesthesia.
The electron microscopic analysis was carried out on brain tissue
from four adult male C57BL/6 mice and three adult Sprague–
Dawley rats (Charles River Laboratories). These animals were
deeply anesthetized with sodium pentobarbital (80 mg/kg, i.p.) and
perfused through the aortic arch with a solution of 0.5% acrolein and
4% paraformaldehyde (PFA), in 0.1 M sodium phosphate buffer
(PB) pH 7.4, followed by 4% PFA.
For hippocampal cell culture, 17 litters of E18-19 C57BL/6
mouse fetuses were taken from pregnant females deeply anaesthe-
tized with CO2.
Antibodies
The polyclonal anti-EphA4 antibody (Ab11) has been shown to
recognize EphA4 in rat, mouse and human (Soans et al. 1994;
Martone et al. 1997; Murai et al. 2003). We also tested its
specificity for immunoelectron microscopy, showing no staining
on sections from with EphA4-/- mice (Tremblay et al. 2007). The
monoclonal anti-EphA4 antibody, raised against peptides 279–472
of the extracellular domain of EphA4, was purchased from BD
Transduction Laboratories (BD Biosciences, Mississauga, ON,
Canada) and tested as specific by western blotting from brain
homogenate of EphA4)/) versus wild-type mice. Its specificity for
immunocytochemistry was also tested on brain sections from
EphA4)/) and wild-type littermate mice. Staining was strongly
reduced in the neuropil regions of the hippocampus in KO mice, but
a faint diffuse background remained in the pyramidal and granule
cell layers.
The anti-VGLUT1 and anti-VGAT monoclonal antibodies,
respectively raised against synthetic peptides corresponding to
residues 456–560 of rat VGLUT1, or residues 75–87 of rat VGAT,
were purchased from Synaptic Systems (Goettingen, Germany). The
guinea pig polyclonal anti-VGLUT1 antibody was purchased from
Chemicon/Millipore (Temacula, CA, USA). The polyclonal anti-
body anti-Piccolo, raised against residues 4439–4776 of the rat
protein, and the monoclonal anti-synaptophysin antibody were
purchased from Synaptic Systems. The anti-synaptotagmin antibody
was obtained from Stressgen Biotech (Ann Arbor, MI, USA). The
monoclonal anti-clathrin heavy chain antibody was purchased from
BD Biosciences and the polyclonal anti-clathrin light chain antibody
was previously described (Poupon et al. 2008). Lastly, the
monoclonal anti-Nav1.2 sodium channel was from Antibodies Inc.
(Davis, CA, USA, NeuroMab clone K69/3 against a.a. 1882–2005 –
cytoplasmic C-terminus).
Immunoelectron microscopy
Following perfusion with fixatives, brains were further fixed by
immersion in 4% PFA for 1 h at 4C, and washed in sodium

phosphate-buffered saline (PBS; 0.9% NaCl in 50 mM PB, pH 7.4).
Transverse sections, 50 lm thick, were cut in ice-cooled PBS with a
vibratome. Sections were processed free-floating by pre-embedding
immunoperoxidase and immunogold protocols. The protocol of
double-EphA4 and -VGLUT1 immunolabeling for electron micro-
scopy (EM) was slightly modified from the single labeling protocol
described previously (Tremblay et al. 2007). After simultaneous
incubation with anti-EphA4 (Ab11; 1 : 500 dilution) and with anti-
VGLUT1 (1 : 200 dilution) primary antibodies, for 48 h, at 21C, in
a blocking solution of PBS containing 5% normal goat serum and
0.5% gelatin, VGLUT1 was first detected with immunogold and
then EphA4 with immunoperoxidase. For immunogold labeling,
sections were incubated overnight at 21C in goat anti-mouse IgGs
conjugated to 1 nm colloidal gold particles (AuroProbe One;
Amersham Biosciences, Oakville, ON, Canada) in the same
blocking solution. The diameter of the immunogold particles was
increased with a silver enhancement kit (IntenSE; Amersham
Biosciences) for 12–15 min at 21C. For immunoperoxidase
labeling, the sections were incubated for 2 h at 21C in goat anti-
rabbit IgGs conjugated to biotin (Jackson Immunoresearch, West
Grove, PA, USA) and with streptavidin–horseradish peroxidase
(Jackson Immunoresearch) for 1 h in blocking solution. Afterwards,
labeling was revealed with 0.05% diaminobenzidine and 0.01%
hydrogen peroxide in Tris/HCl-buffered saline (50 mM, pH 7.4).
Sections were post-fixed flat in 1% osmium tetroxide, embedded
in resin (Durcupan ACM; Sigma, St Louis, MO, USA), and
mounted between ACLAR embedding film (EMS, Hatfield, PA,
USA). Small pieces of the latero-ventral hippocampus (CA1
pyramidal cell layer and stratum radiatum) and dentate gyrus
polymorphic layer were excised from the embedding film, re-
embedded at the tip of resin blocks, sectioned (70–80 nm-thick)
with an ultramicrotome (Ultracut S, Leica Canada, St-Laurent,
Québec, Canada), collected on bare square-mesh copper grids,
stained with lead citrate and examined at 60 kV with a Philips
CM100 electron microscope.
For dissociated hippocampal cell cultures, coverslips (see below)
were treated (or untreated) for 5 min with 0.1% Triton X-100 in
PBS, rinsed thrice with PBS, and blocked during 1 h with a
solution of 5% normal donkey serum and 0.5% gelatin in PBS, at
21C. They were incubated for 72 h with or without the
monoclonal anti-EphA4 antibody (BD Bioscience, 1 : 500 dilution
in blocking solution), at 4C. The EphA4 immunoreactivity was
detected using protein-A labeled with 10 nm gold particles
(Bendayan 1984). Cultures were post-fixed in 1% osmium tetroxide
containing 15% K-ferricyanide, followed by uranyl ‘en bloc’
staining. The next day, they were dehydrated in ethanol, embedded
in Epon resin (Mecalab, Montréal, Québec, Canada), and mounted
between ACLAR embedding film. Ultrathin sectioning and further
processing was done as above.
LS1 and LP2 subcellular fractionation of mouse forebrain/
midbrain
Subcellular fractions were prepared from adult mouse forebrain/
midbrain (two for each of three experiments), as described (see
supplementary material in Zhou et al. 2007). Briefly, the forebrain/
midbrain was dissected out on ice, following a transverse cut between
the occipital cortex and cerebellum. All subsequent steps were carried
out at 0–4C. Brain tissues were homogenized by 9 up-and-down
 2010 The Authors
Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
EphA4 in clathrin and synaptic vesicles | 155
strokes at 900 rpm in 20 mL of ice-cold sucrose/HEPES buffer
(0.32 M sucrose, 10 mM HEPES, pH 7.4) containing protease
inhibitors (complete EDTA-free; Roche, Manheim, Germany). The
nuclear material (P1) was discarded following centrifugation at
1000 g during 10 min, and the supernatant (S1) was centrifuged again
at 12 000 g for 15 min. The new supernatant was discarded and the
pellet (P2) was re-suspended in 20 mL of ice-cold sucrose/HEPES
buffer and centrifuged again at 13 000 g for 15 min. The resulting
pellet, constituting the synaptosomal fraction (P2¢) was re-suspended
in approximately 2 mL of sucrose/HEPES buffer.
Two milliliters of this P2¢ fraction were lysed osmotically in
20 mL of 10 mM HEPES buffer containing the protease inhibitors
and then homogenized by three up-and-down strokes at 2000 rpm
and centrifuged at 33 000 g for 20 min. The pellet, representing the
synaptosomal membrane fraction (LP1) was discarded, while the
supernatant (LS1) was used for immunoisolation of vesicles, or
centrifuged again at 260 000 g during 2 h, to obtain the crude
vesicle fraction (LP2). The concentration of EphA4 in fraction LP2
was approximately 30% compared with that in the starting brain
homogenates or the purified synaptosome fraction, P2¢ (see Bouvier
et al. 2008).
Linear sucrose density gradient of crude vesicle fraction
The crude vesicle fraction, LP2, was re-suspended in 3 mL of
sucrose/HEPES buffer (0.04 M sucrose, 10 mM HEPES, pH 7.4).
The fraction was homogenized (three strokes at 900 rpm) followed
by repeated trituration through 22½ and 27½ gauge needles and
layered on top of a 37 mL linear sucrose density gradient ranging
from 10 to 35% of sucrose solution. The gradient was centrifuged at
120 000 g during 4 h using a SW 28 rotor (Beckmann). Sixteen
2.5 mL aliquot fractions were collected from the top (fraction 1) to
the bottom of the tube (fraction 16). Equal volumes of each fraction
(80 lL) were added to 2· loading buffer and loaded on a SDS gel
for western blotting.
Isolation of CCVs
The isolation of CCVs from rat adult brain was performed as
described (Girard et al. 2005a), in two separate experiments. The
preparations were immunoblotted for EphA4, clathrin heavy chain
(CHC) and clathrin light chain proteins and for the sodium channel
Nav1.2.
Isolation of synaptic vesicles
A highly purified fraction of synaptic vesicles was prepared as
described (Huttner et al. 1983). Briefly, a fraction corresponding to
LP2, above, was layered on top of a 50–800 mM continuous sucrose
gradient and centrifuged at 65 000 g for 5 h. The band of material at
the 200–400 mM region, enriched in vesicles, was collected and
constituted the sucrose gradient–vesicular fraction. A further
purification step involved permeation chromatography using glyc-
eryl-coated controlled-pore glass beads (mean pore diameter of
3000 Å), to obtain a controlled-pore glass vesicular fraction (SV) of
pure synaptic vesicles (see Huttner et al. 1983).
Immunoisolation of specific types of synaptic vesicles
The following experiments were repeated thrice. Magnetic beads
(107) with pre-coupled immunoglobulins (Dynabeads M-280
sheep anti-rabbit or anti-mouse IgG; Dynal, Invitrogen, Oslo,
156 | D. Bouvier et al.
Norway), were washed thrice with PBS and incubated overnight
with the primary rabbit polyclonal or monoclonal antibodies
(3 lg of each), or rabbit or mouse control immunoglobulins
(Chromopur rabbit or mouse IgG, Jackson Immunoresearch
Laboratories, West Grove, PA, USA), in PBS containing 1%
bovine serum albumin (BSA). The LS1 supernatant (0.4 lg of
protein per lL) was used as immunoisolation buffer (Ii buffer)
consisting of LS1 fraction, 50 mM Tris HCl at pH 7.4, 50 mM
KCl, 20 mM MgCl2, 0.2 M sucrose and protease inhibitors
(complete EDTA-free; Roche). The antibody-coupled beads were
washed thrice with PBS/BSA 1%. One milliliter of the Ii buffer/
BSA 1% was added to the pre-coupled beads for each
immunoisolation and incubated for 1 h 45 min. The beads were
then washed four times with PBS and re-suspended in 25 lL of
loading buffer (Bio-Rad, Hercules, CA, USA), heated at 95C for
5 min, for standard sodium dodecyl sulfate–polyacrylamide gel
electrophoresis and western blotting. The amount of LS1 that was
loaded in these gels corresponded to 10 lg of protein.
Hippocampal cell culture
Primary neuronal cultures were prepared from the hippocampi of
18–19 days old fetal mice, as described by Banker and Cowan
(1977). For each experiment (n = 6), two litters of C57BL/6 mice
were used. Briefly, the hippocampi were dissociated by treatment
with trypsin (0.25% for 15 min at 37C; Sigma), followed by
trituration with fire-polished Pasteur pipettes. Dissociated cells were
plated at densities of 100 000 to 200 000 on poly-L-lysine-treated
coverslips (Fisher, Nepean, ON, Canada) in minimum essential
medium (Invitrogen) containing 10 mM pyruvic acid, 0.6% glucose
and 10% heat-inactivated horse serum. After 3 h of incubation at
37 ºC, cultures were fed with Neurobasal medium (Invitrogen)
supplemented with 0.5 mM L-glutamine and 2% B27 (Gibco,
Rockville, MD, USA, Invitrogen). Cultures were fixed after 15 days
in vitro (DIV15) with 4% paraformaldehyde.
Immunofluorescence and confocal microscopy
Coverslips were treated for 5 min with 3% Triton X-100 in PBS,
rinsed thrice with PBS, and blocked during 1 h with a solution of
5% BSA in PBS, at 21C. They were incubated overnight with the
primary antibodies (ab11 anti-EphA4 1 : 500, monoclonal or
polyclonal anti-VGLUT1 or anti-VGAT; 1 : 200 dilutions), at
21C, in a humid chamber, in 5% BSA in PBS. After rinses in
PBS, coverslips were incubated for 1 h with FITC-labeled anti-
rabbit and TRITC-labeled anti-mouse antibodies in PBS/BSA. The
coverslips were rinsed with PBS and mounted with Mowiol.
Immunofluorescence was viewed and captured by a Leica DM IRBE
confocal microscope. High-resolution pictures were acquired as
composites of eight confocal optical sections.
KCl depolarization and cell surface EphA4 immunofluorescence of
hippocampal neurons
Hippocampal neurons at DIV15 were incubated for 2 min with
90 mM KCl in Neurobasal medium supplemented with glutamine,
and B27. They were then washed twice with ice-cold PBS, before
fixation in 4% PFA. Triton X-100 was omitted to restrict the
immunodetection of EphA4 to the cell surface by using the anti-
EphA4 antibody (BD Bioscience) directed against the extracellular
domain of the receptor (see Immunoelectron microscopy, above).
 2010 The Authors
Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
Immunofluorescence was viewed and captured by a fluorescence
microscope (Karl Zeiss, Axiophot, Germany) equipped with a
digital CCD camera (Qimaging, Surrey, BC, Canada). In three
experiments, measurement of cell surface EphA4 immunofluores-
cence was carried out with Northern Eclipse Imaging V.6 software
(Empix Imaging, Missisauga, ON, Canada), on 25 frames covering
each 21 000 lm2 for each condition (control vs. KCl treatment).
Unstimulated and KCl-treated cultures were analyzed under the
same image exposure, with the same threshold values. The analysis
was performed in three cultures for each condition. All granular
spots of labeling were treated as individual objects for the following
measurements: object count, selected area, total object area, percent
object area and total gray. Optical density was the result of the gray
value of the spot multiplied by its area and divided by the total area
of the frame.
Results
EphA4 is in multiple subpopulations of vesicles
In adult murine hippocampus, EphA4 is present in 25–40%
of axon terminals (Tremblay et al. 2007), where it localizes
in the plasma membrane, pre-synaptic active zone and
vesicles (Bouvier et al. 2008). It is also frequently found in
dendritic spines, notably in association with PSDs (Tremblay
et al. 2007; Bouvier et al. 2008). During early postnatal
development, EphA4 immunoperoxidase labeling was also
detected in the cell body of pyramidal, granule and astrocytic
cells, in association with vesicles or saccules of the
endoplasmic reticulum, near the Golgi apparatus or in large
dendrites, on transport vesicles of various sizes (Tremblay
et al. 2009). In the present study, using the same antibody
tested on brain sections from EphA4)/) mice (Tremblay
et al. 2007, 2009), electron microscopy disclosed the same
intracellular localizations of EphA4 in the adult mouse
hippocampus, including vesicles on the cis as well as trans
sides of the Golgi apparatus (Fig. S1a,b), consistent with the
synthesis of a glycosylated protein and its subsequent
transport to dendrites and axons. In axon terminals, EphA4
was usually restricted to small subgroups of vesicles, and less
frequently decorated all vesicles of terminal profiles (Fig. 1).
The latter observation suggests that EphA4 is present in
special types of vesicles in the axon terminals. In both
dendrites (Fig. S1c) and nerve terminals (Fig. 1c), EphA4
was also detected in CCVs, identified on the basis of classical
criteria (see figure legends), as described (Zaremba and Keen
1983).
To more specifically characterize the types of EphA4-
positive vesicles, a cell fraction (LP2) enriched in SVs
isolated from lysed synaptosomes was prepared from adult
mouse forebrain/midbrain and further fractionated on a
linear sucrose density gradient (10–35% sucrose, Fig. 2).
The synaptic vesicle markers, synaptophysin, VGLUT1,
VGAT and synaptotagmin, peaked in fractions 4–9. The
vesicular monoamine transporter 2 was distributed in the
 EphA4 in clathrin and synaptic vesicles | 157

Fig. 1 Immunoelectron microscopy of EphA4 in stratum radiatum of

(a)
the CA1 region of adult mouse (a,b) and rat (c) hippocampus showing
that the immunostaining was often restricted to only a subgroup of SVs
(arrows in a,b), or occasional CCVs (arrow in c). The identity of the
CCV is confirmed by the visualization of the triskelia across its diam-
eter, its inner diameter of 30 nm and outer diameter of 55 nm, be-
cause of its coat, and very much in line with a nerve terminal CCV.
t+, immunoreactive axon terminal; s), unlabeled dendritic spine;
d, dendrite. Scale bars: 250 nm.

(b)

Fig. 2 Distribution of EphA4 in a synaptosome-derived vesicle fraction

(LP2) layered on top of a linear 10–35% sucrose density gradient.
EphA4 always showed a bimodal distribution in such density gradi-
ents. Here, a first peak (fractions 6–8) partially overlapped the peak of
SV markers such as synaptophysin, VGLUT1, VGAT, synaptotagmin
and VMAT2 (4–5 to 8–9). The late endosome marker, Rab7, also
peaked in the same range. The second peak of EphA4 occurred with
the higher density particles, but was still accompanied by synapto-
physin and VGLUT1. VGLUT1, VGAT and VMAT2, represent vesic-
(c) ular transporters for glutamate, GABA and monoamines, respectively.

first 6 fractions, while the late endosome marker Rab7 was

also observed in lower amount in the late fractions. In all
experiments, EphA4 had a bimodal distribution, with a
first peak in fractions 6–8 and a second in fractions 12–15.
The first peak of EphA4 partially overlapped the distribu-
tion of vesicles containing synaptotagmin, vesicular mono-
amine transporter 2, and VGAT and follows the
enrichment in synaptophysin and VGLUT1. The second
peak, which was more intense, appeared in fractions of
higher density particles, but still co-existed with synapto-
physin and VGLUT1. All vesicle fractions were negative
for the early endosome marker, EEA1 (not illustrated, see
Fig. 1 in Bouvier et al. 2008), consistent with previous
results and confirming the absence of EphA4 in early
endosomes. This result is consistent with a distribution of
EphA4 in multiple vesicular organelles, including some,
but not all synaptic vesicles.

 2010 The Authors

Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
158 | D. Bouvier et al.
(a)
(b)
Fig. 3 (a) EphA4 in clathrin-coated vesicles (CCVs) isolated by cell
fractionation. EphA4 was present in all fractionation steps where the
clathrin light (CLC, not shown) and heavy chains (CHC) were de-
tected, including the CCV fraction. The purity of the CCV fraction was
confirmed by the absence of a plasma membrane protein, the sodium
channel Nav1.2. H, homogenate; P1, pellet-1; P2, pellet-2 S2;
supernatant-2; SGp, pellet of the sucrose gradient; SGs, supernatant
of the sucrose gradient, as described (Girard et al. 2005a). Results
reproduced in two separate experiments. (b) EphA4 is present in a
highly purified fraction of SVs (Huttner et al. 1983). The absence of the
CHC protein and the enrichment in synaptophysin confirm the purity of
this SV preparation.
EphA4 is associated with clathrin-coated vesicles
The presence of EphA4 was then examined in a preparation
of purified CCVs (Girard et al. 2005a; Allaire et al. 2006).
Clathrin-mediated endocytosis is critical for the reuptake of
synaptic material and the recycling of SVs (Jung and
Haucke 2007). It is also involved in the capture, retrograde
transport and signaling of several tyrosine kinase receptors,
following their activation (Gomes et al. 2006). As shown in
Fig. 3(a), EphA4 was readily detected in the CCV fraction.
EphA4 was not enriched to the same extent as the CHC or
clathrin light chain (not shown) proteins in the CCV
fraction, but its distribution among fractions was similar to
those described previously for synaptotagmin and synapt-
ophysin (Allaire et al. 2006). The purity of the CCV
preparation was confirmed by the absence of the plasma
membrane-specific sodium channel, Nav1.2. This is consis-
tent with EphA4 being a component of SVs that are
undergoing endocytosis.
EphA4 is also associated with purified and
immuno-isolated synaptic vesicles
EphA4 was also detected in the highly purified fraction of
synaptic vesicles (fraction SV, Fig. 3b). The purity of this
fraction of synaptic vesicles (Huttner et al. 1983) was
 2010 The Authors
Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
confirmed by the enrichment of synaptophysin and the
absence of CHC. This result constitutes strong evidence of
the presence of EphA4 in synaptic vesicles.
The above electron microscopy and cell fractionation
results are consistent with the presence of EphA4 in CCVs
and SVs. As it has previously been detected in fractions of
PAZ (Bouvier et al. 2008), EphA4 might be transported to
this specialized compartment of the pre-synaptic terminal
with synaptic vesicle proteins (Nakata et al. 1998) or with
PTVs, which are known to transport the PAZ structural
material (Shapira et al. 2003; Dresbach et al. 2006;
Tao-Cheng 2007). To examine these possibilities, we
immunoisolated subpopulations of SVs and PTVs from
the LS1 supernatant isolated following lysis of synapto-
somes (P2¢ fraction) from adult mouse brain (see Bouvier
et al. 2008). LS1 is known to be enriched in small
vesicles, but devoid of larger membranes (Huttner et al.
1983).
Antibodies directed against the cytoplasmic portion of
synaptophysin or piccolo were used to immunoisolate SVs
and PTVs, respectively (Fig. 4a,d). EphA4 was present in
synaptophysin-immunoisolated vesicles, but absent from
piccolo-containing vesicles. To further characterize the
association of EphA4 with SVs, we immunoisolated two
distinct subpopulations of neurotransmitter-specific
vesicles, using anti-VGLUT1 and anti-VGAT antibodies
(Fig. 4b,c). Both preparations were enriched in synapto-
physin, confirming the successful immunoisolation of SVs.
Control blots of VGLUT1- and VGAT-immuno-isolated
vesicles, immunostained using, respectively, the VGAT or
VGLUT1 antibodies, showed the absence of the other
vesicular transporter in each type of preparations, demon-
strating the specificity of the immunoisolation of SVs
(Fig. 4e). Importantly, EphA4 was present in both
VGLUT1- and VGAT-vesicles. In contrast, EphA4 was
not detected in control immunoisolations carried out with
non-specific IgGs. Moreover, EphA4 did not co-immuno-
precipate with VGLUT1 or VGAT after treatment of
the LS1 fraction with 0.5% Triton X-100, indicating
that EphA4 is associated with VGLUT1 and VGAT SVs
and not likely in a stable physical complex with the
transporters.
EphA4 is present in axon terminals expressing VGLUT1 in
CA1 of adult hippocampus
Double immunoelectron microscopy in region CA1 of the
adult mouse hippocampus showed frequent co-localization of
EphA4 with VGLUT1 in axon terminals (Fig. S2). This
observation shows that EphA4 is present in glutamatergic
axon terminals in vivo. This is consistent with our previous
EM analysis of the adult rat hippocampus, showing EphA4
localized in axon terminals contacting dendritic spines with
asymmetric synapses, which are known to be excitatory
(Tremblay et al. 2007).

(a)
(b)
(c)
(d)
Fig. 4 Immuno-isolation of synaptic and piccolo–bassoon transport
vesicles from the LS1 fraction (0.4 lg/mL of LS1 were used for each
immunoisolation). (a) SVs were immuno-isolated with an anti-synapt-
ophysin antibody (Syn-Ii). In this and the following blots, the amount of
LS1 fraction loaded in the gels corresponded to 10 lg of protein. The
same procedure with control non-specific immunoglobulins (Ctl IgG)
gave no staining for EphA4 or synaptophysin, whereas the immuno-
isolation of vesicles with synaptophysin was positive for both EphA4
and synaptophysin. (b) Immuno-isolation of glutamatergic SVs
(VGLUT1-Ii) and immunoprecipitation of VGLUT1 (VGLUT1-IP) using
an anti-VGLUT1 antibody. The western blot of vesicles treated with the
control IgGs was negative for EphA4 and synaptophysin, but showed
a spot at the level of VGLUT1 (50 KDa) that is likely attributable to
the IgGs. The immuno-isolation with the VGLUT antibody was positive
for EphA4 and synaptophysin (confirming the success of the immu-
noisolation of SVs), while the spot corresponding to VGLUT1 was
strongly increased, also confirming the success of the immuno-isola-
tion. The immunoprecipitation of VGLUT1, following treatment of the
EphA4 colocalizes with VGLUT1 and VGAT in cultured
hippocampal neurons
Following immunofluorescence for EphA4, cultured hippo-
campal neurons at DIV15 displayed a punctate labeling for
EphA4 along their neurites, with a more diffuse staining over
cell bodies (Fig. 5a,d). This staining pattern is consistent
with results from previous studies (Fu et al. 2007). The
immunofluorescence of VGLUT1 or VGAT was also punc-
tate, with granules of various sizes along neurites and cell
bodies, consistent with the known pre-synaptic localization
of these vesicular proteins (Fig. 5b,e). However, the VGAT
 2010 The Authors
Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
EphA4 in clathrin and synaptic vesicles | 159
(e)
vesicles with Triton X-100, was negative for EphA4, indicating that
EphA4 and VGLUT1 are colocalized in the same vesicles, but not in
strong physical association. (c) Immuno-isolation of GABAergic SVs
(VGAT-Ii) and immunoprecipitation of VGAT (VGAT-IP) gave results
similar to VGLUT1, showing that EphA4 is present in these vesicles
but does not interact strongly with VGAT. These results were repeated
in three separate experiments. (d) Piccolo–bassoon transport vesicles
(PTVs) were immuno-isolated with an anti-piccolo antibody (Pic). The
western blot of immuno-isolated PTVs was negative for EphA4. (e)
Control staining of the VGLUT1 and VGAT immuno-isolates. Following
immuno-isolation of vesicles with VGLUT1 (top pictures), the blots are
strongly reactive with the anti-VGLUT1 antibody, but negative with the
anti-VGAT antibody. The immuno-isolation of vesicles with VGAT
(bottom pictures) yielded a fainter signal in VGAT western blotting, and
a comparable detection of the same spot of VGAT with the VGLUT
antibody. Thus, there was no detectable cross-isolation of the vesi-
cles, but the VGLUT1 antibody showed some cross-reactivity with
VGAT in western blotting.
immunostained elements were less abundant, representing
roughly one third of the VGLUT1 immunofluorescent
punctae. Moreover, their distribution differed to some extent,
with VGAT terminals being much more frequently surround-
ing neuronal perikarya, as expected for these inhibitory
terminals (not shown). Spots of EphA4 immunofluorescence
extensively colocalized with those of VGLUT1 (Fig. 5c) or
VGAT (Fig. 5f). Nevertheless, VGLUT1 and VGAT were
detected in distinct nerve terminals, as shown following dual
labeling for VGLUT1 and VGAT (Fig. 5g,h). The polyclonal
anti-VGLUT1 antibody (used for dual immunofluorescence
160 | D. Bouvier et al.
(a) (b) (c)
(a′) (b′) (c′)
(d) (e) (f)
(d′) (e′) (f′)
(g) (h) (i)
(g′) (h′) (i′)
with the monoclonal VGAT antibody) and the monoclonal
anti-VGLUT1 antibody (used for dual immunostaining with
EphA4) gave total co-staining when used together, confirm-
ing the specificity of both anti-VGLUT1 antibodies (not
shown). EphA4 thus colocalized with both VGLUT1 and
VGAT in cultured neonatal hippocampal neurons, consistent
with an association with both subpopulations of SVs in vitro.
Examination of such cultures in immunoelectron micro-
scopy (immunogold with the BD Bioscience monoclonal
antibody recognizing the extracellular domain of EphA4),
following the use of Triton X-100, showed gold particles on
the plasma membrane of perikarya, dendrites and axons, as
well as in nerve terminals, in close association with SVs
(Fig. S3a,b). However, when Triton X-100 was omitted
before incubation with the immunoreagents, the immunogold
labeling was restricted essentially to the plasma membrane of
the cells (Fig. S3c). Quantification of the gold particles
inside and on the surface of the cell profiles in EM gave
1.8 ± 2.0 particles inside versus 13.5 ± 8.0 on the plasma
membrane (n = 63; ratio inside/surface: 0.13) for profiles of
 2010 The Authors
Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
Fig. 5 Double immunofluorescence for
EphA4 (Ab11 antibody against the intra-
cellular domain of EphA4, in the presence
of Triton X-100) (a, a¢, d, d¢) and VGLUT1
(b, b¢) or VGAT (e, e¢) on hippocampal
neurons at DIV15 in culture. Co-localization
of EphA4 with VGLUT1 (c, c¢) or VGAT (f, f¢)
appears in yellow on the overlays. Control
dual immunofluorescence for VGLUT (g, g¢)
and VGAT (h, h¢) shows that the two
vesicular transporters are expressed in
distinct nerve terminals with no overlap.
Scale bars: 5 lm.
cultures immunostained without Triton X-100. For profiles
from cultures immunostained in the presence of Triton
X-100, there was a mean of 27.4 ± 18.6 particles inside
versus 19.3 ± 12.6 on the plasma membrane (n = 36; ratio
inside/surface: 1.42). Moreover, many profiles in Triton-
treated cultures were labeled only on the surface (dendrites,
axons), whereas most of those with many gold particles
inside were identified as axon terminals. These features are
consistent with EphA4 being mostly expressed on the
surface, in the plasma membrane, except for a strong
labeling of synaptic vesicles in axon terminals.
These observations confirm the presence of EphA4 in SVs
of cultured hippocampal neurons as well as the cell surface
restriction of the labeling when Triton X-100 is omitted
during the immunocytochemical procedure (see below).
The cell surface expression of EphA4 is regulated by
neuronal activity in culture
The fractionation and microscopic observations constitute
strong evidence for the presence of EphA4 in SVs and
 EphA4 in clathrin and synaptic vesicles | 161

(a) (b)

Fig. 6 Cell surface expression of EphA4 is

modulated by neuronal activity in DIV15
hippocampal cultures (conditions similar to
Fig. S3). (a–f) Representative pictures of
cell surface EphA4 immunolabeling of
unstimulated neurons (a, c, e) and KCl-
depolarized neurons (b, d, f). The labeling is (c) (d)
distributed in punctates along neurites and
on the cell bodies (arrows). Scale bars:
15 lm. (g–i) Quantitative analysis of the
labeling was performed on 25 frames (each
21 000 lm2) in each condition (Ctl, control;
KCl, 2 min treatment with 90 mM KCl) and (e) (f)
in each of three different experiments.
Labeling on cell bodies was excluded from
the analysis. KCl depolarization increased
the optical density of the EphA4 immuno-
fluorescence of individual puncta (g), their
(g) (h) (i)
percent area relative to the total area of the
cell processes (h) and the number of puncta
(i). Differences between control and KCl
treatment were significant for all three
parameters in each of the three separate
experiments (two-way ANOVA; condition,
***p < 0.001). Comparison between the re-
sults of the three experiments showed no
significant difference (two-way ANOVA,
experiment*condition, b, p = 0.576; c,
p = 0.978; d, p = 0.368; Error bars: SEM).

CCVs. Therefore, we determined if the cell surface expres-
sion of EphA4 was affected by neuronal depolarization. We
used KCl-induced depolarization to mimic synaptic activity,
followed by immunofluorescence on DIV15 hippocampal
cultures, in conditions similar to those used for immunogold
in electron microscopy without Triton X-100 (Fig. S3c,
showing surface labeling). Control neurons and neurons
subjected to 2 min of depolarization with KCl displayed a
punctate cell surface immunofluorescence on perikarya and
neurites (Fig. 6). However, the immunofluorescence was
very faint in control cultures (Fig. 6a,c,e), whereas the
intensity of this immunofluorescence increased markedly
following KCl depolarization (Fig. 6b,d,f). This observation
was repeated in six separate experiments carried out each
time on 2–3 control versus KCl-treated cultures. Quantifica-
tion of the fluorescent puncta on neurites was performed in
three of these six experiments and showed a significant
increase in the mean optical density (fluorescence intensity of
individual puncta), mean surface area and number of puncta,
following KCl stimulation (Fig. 6g–i).
These observations suggest that EphA4 is recruited to
the plasma membrane upon synaptic vesicle fusion with the
plasma membrane, following sustained neuronal depolarization.
Discussion
We previously demonstrated that EphA4 is associated with
pre-synaptic compartments including active zones and ves-
icles, in addition to being present in post-synaptic elements
such as dendritic spines and PSDs (Bouvier et al. 2008). In
the present study, we have further investigated the associa-
tion of EphA4 with SVs, PTVs and CCVs using cell
fractionation, immunoisolation of vesicles, and immunocy-
tochemistry in confocal and electron microscopy. The results
demonstrate the presence of EphA4 in SVs and CCVs, a
novel observation for an Eph receptor.
EphA4 is associated with various types of vesicles
Using electron microscopy and linear sucrose density
gradient, we found that EphA4 is associated with various

 2010 The Authors

Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
162 | D. Bouvier et al.
subpopulations of vesicles. Indeed, the EphA4 immunoper-
oxidase reaction product often decorated only small subpop-
ulations of vesicles in axon terminal profiles, including
occasional CCVs, whereas the distribution of the receptor
among vesicles of increasing density showed a bimodal
pattern different from, although overlapping, that of the
synaptic vesicle markers (VGLUT1, synaptophysin, synapt-
otagmin); indicating that it is also present in other types of
vesicles. This result is consistent with the findings that
EphA4 is not expressed in all nerve terminals of the
forebrain, and that it is also located in several other types
of organelles, including the plasma membrane, trans-Golgi
tubulovesicular organelles, and transport vesicles of various
sizes in dendrites. Further fractionation of these synaptoso-
mal vesicles to obtain highly purified preparations of CCVs
(Girard et al. 2005a) and SVs (Huttner et al. 1983) provided
very strong evidence for the presence of EphA4 in both
subpopulations of vesicles.
According to the immunoisolation results and the immu-
nocytochemistry on cultured hippocampal neurons, at least
two distinct populations of SVs, VGLUT1- or VGAT-
positive, were carrying EphA4. Indeed, the glutamate and
GABA transporters have been demonstrated to reside on
separate vesicles, which are also functionally distinct
subpopulation of SVs (Takamori et al. 2000). The observa-
tion was confirmed for VGLUT1 by immunoelectron
microscopy in CA1 of the adult mouse hippocampus; a
result consistent with previous electron microscopic analyses
of EphA4 in the mature rat hippocampus (Tremblay et al.
2007) and mouse cerebral cortex (Bouvier et al. 2008)
showing EphA4 at excitatory synapses. These results dem-
onstrate the presence of EphA4 in at least these two types of
SVs. It was recently reported, after using similar techniques,
that some axon terminals in layer 5 of the cerebral cortex
contain both types of vesicular transporters (Fattorini et al.
2009). It is therefore possible that such a co-localization went
undetected in the present preparations of vesicles derived
from the whole mouse forebrain–midbrain.
In contrast, the PTVs immunoisolated with an anti-
piccolo antibody were EphA4-negative, suggesting that this
receptor is not transported to the pre-synaptic active zone
in these vesicles. Indeed, PTVs are clearly distinct from the
SVs and have been described as the transport vesicles
carrying the cytomatrix material composed of piccolo,
bassoon, Rim, Munc13, Munc18 and SNAP25 to the active
zone (Shapira et al. 2003). Thus, one hypothesis that
should be examined further is whether EphA4 is trans-
ported to the axon terminals and pre-synaptic active zones
with other synaptic vesicle proteins, through intermediate
vesicular transport organelles of trans-Golgi network origin
(Nakata et al. 1998; Jin and Garner 2008), as appears to be
the case for TrkB, another tyrosine kinase receptor
associated with the development of cortical glutamatergic
synapses (Gomes et al. 2006).
 2010 The Authors
Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
EphA4 and endocytosis
The localization of EphA4 in CCVs is also a novel
observation. EphA4 was detected in CCVs by western blot
of purified CCVs as well as by immunoelectron microscopy
in dendrites and axon terminals in adult rat hippocampus.
This demonstrates that some of the CCVs isolated from
synaptosomes of the forebrain–midbrain and containing
EphA4 were pre-synaptic, participating in SV recycling. In
fact, it has been estimated that approximately 90% of all
CCVs that function in endocytosis in brain are involved in
synaptic vesicle retrieval (Girard et al. 2005b). Thus, this
tyrosine kinase receptor could be recycled together with the
synaptic vesicle material.
Another Eph family member, EphB2, has been implicated
in the modulation of clathrin-dependant endocytosis by
influencing the tyrosine phosphorylation of the phosphatase
synaptojanin-1 (Hopper and O’Connor 2005; Irie et al.
2005). Other receptor tyrosine kinases, such as RET
(Richardson et al. 2006), the insulin-like growth factor
receptor (Monami et al. 2008), the epidermal growth factor
receptor (Carter and Sorkin 1998; Sorkina et al. 2002; Jiang
et al. 2003) and Trk neurotrophin receptors (Zheng et al.
2008) have previously been reported to be internalized via
the clathrin-mediated endocytic pathway. A similar fate is
followed by the p75 neurotrophin receptor (Bronfman et al.
2003), which was recently reported to associate with ephrin-
As as a receptor complex (Lim et al. 2008). Internalized
TrkA, TrkB and p75 have also been shown to remain
functional and capable of signaling within the endocytic
vesicles (Grimes et al. 1996, 1997; Bronfman et al. 2003)
while being transported away from synapses to the cell body
in a phosphorylated state (Watson et al. 1999; McPherson
et al. 2001). By analogy, Eph receptors might also remain
functional within the endocytic vesicles. In support of this
hypothesis, ephrin-A or -B interactions with EphB or EphA4
receptors have been shown to induce a bidirectional trans-
endocytosis of ligand–receptor complexes (Marston et al.
2003; Zimmer et al. 2003; Deininger et al. 2008), and recent
evidence shows that the endocytosis of ephrin-B1 occurs via
a clathrin-mediated pathway (Parker et al. 2004). Internali-
zation of EphA4 has recently been demonstrated to occur on
the dendritic side, following stimulation with ephrin-B3 and
activation of Rin1, a Rab5 guanosine exchange factor
(Rab5GEF) (Deininger et al. 2008). EphA4 could thus
follow this pathway.
As EphA4 can be found in pre-synaptic terminals as
well as in post-synaptic dendritic spines of excitatory
synapses, at least in hippocampal and cerebrocortical areas
(Tremblay et al. 2007; Bouvier et al. 2008), one could
speculate that the clathrin-dependant pathway is bidirec-
tional as well. Indeed, Rab5 has been localized on SVs
(Fischer von Mollard et al. 1994; Star et al. 2005), where
it seems to modulate their size (Shimizu et al. 2003). A
better knowledge of the ultrastructural localization of

ephrin ligands relative to that of EphA4 would help to
clarify the nature of the synapses where such a mechanism
is likely to operate.
Interestingly, EphA4 shares many features with the
neurotrophin TrkB receptor during development and in the
adult cerebral cortex or hippocampus (Gomes et al. 2006).
Similar to TrkB, EphA4 is present in puncta on the surface
and intracellularly in dendrites and axons of cortical neurons
in dissociated cultures; it is colocalized with synaptic vesicle
proteins; present in axonal growth cones and dendritic
filopodia (Tremblay et al. 2009) and gradually becomes
concentrated at glutamatergic synapses (Tremblay et al.
2007, 2009; Bouvier et al. 2008). Therefore, both tyrosine
kinase receptors appear to be involved in the formation and
plasticity of cortical excitatory synapses. It will thus be
important to examine further the interactions between these
two signaling systems in synaptogenesis and synaptic
plasticity, as initiated in recent reports (Lim et al. 2008;
Marler et al. 2008).
EphA4 is associated with synaptic vesicles
The detection of EphA4 in highly purified preparations of
SVs, in immuno-isolated VGLUT1 or VGAT SVs, as well as
its localization in SVs by immunoelectron microscopy,
in vivo and in vitro, provide convincing evidence that EphA4
is associated with these organelles. Other reports have also
shown the localization of an EphA receptor in insulin
secretion granules in pancreas islet cells, regulating insulin
secretion (Konstantinova et al. 2007), as well as in drosoph-
ila neuromuscular junctions, where glutamate secretion was
regulated by an Eph receptor via ephexin-cdc42-Cav1.2
signaling in pre-synaptic homeostatic plasticity (Frank et al.
2009). Thus, the roles of Eph receptors in synaptic vesicles
will need further examination.
Interestingly, the present results suggest that EphA4 cell
surface expression is influenced by neuronal depolarization,
at least in vitro. EphA4 has previously been localized in
PAZs, in addition to pre-synaptic vesicles (Bouvier et al.
2008). Many other adhesion molecules, including cadherins,
neuroligins, neurexins and synaptic cell adhesion molecule
are concentrated in PAZ and are essential for synapse
formation and stabilization (Rosenmund et al. 2003; Dalva
2007). The increase in EphA4 cell surface immunofluores-
cence on cultured hippocampal neurons, following intense
stimulation of neurotransmitter release by KCl, was most
likely associated with synaptic vesicle fusion with the PAZ.
Notably, EphA4 was usually present in only small subgroups
of pre-synaptic vesicles, presumably representing a special
pool of SVs. It will be important to determine whether
EphA4 is externalized only following intense synaptic
stimulation, which might be part of a mechanism of synaptic
plasticity, whereby this receptor is only delivered at highly
active synapses.
 2010 The Authors
Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
EphA4 in clathrin and synaptic vesicles | 163
Acknowledgements
This study was supported by grants from the Natural Sciences and
Engineering Research Council of Canada (NSERC), The Scottish-
Rite Charitable Foundation of Canada, a CIHR grant to P.S.M, and
by an Infrastructure Grant from the Fonds de la recherche
en santé du Québec (FRSQ; grant to the GRSNC). D.B. was
supported by a studentship from the GRSNC, and M.-È. T. was
supported by studentships from the FRSQ, GRSNC and Université de
Montréal. P.S.M. holds the James McGill Chair. The authors thank Dr
Miguel Chagnon for the statistical analyses as well as Drs Johanne
Bertrand and Nicole Leclerc for their expertise in cell culture.
Supporting information
Additional Supporting Information may be found in the online
version of this article:
Figure S1. EphA4 immunoreactive vesicles in the peri-Golgi
region of a pyramidal cell body (a) and dendrites (b,c) from
the pyramidal cell layer in the CA1 region of the adult mouse
(a,b) or polymorphic layer of the dentate gyrus of the rat (c)
hippocampus.
Figure S2. (a,b) Double immunocytochemistry for EphA4
(immunoperoxidase, open arrows) and VGLUT (silver-intensified
immunogold, solid arrows) in the CA1 region of adult mouse
hippocampus show that EphA4 is co-localized with VGLUT1 in
some nerve terminals. Scale bars: 250 nm.
Figure S3. Hippocampal neurons in culture (DIV15) visualized
by immunoelectron microscopy (monoclonal antibody against the
extracellular, or intravesicular, domain of EphA4; protein-A with
10 nm gold particles).
As a service to our authors and readers, this journal provides
supporting information supplied by the authors. Such materials are
peer-reviewed and may be re-organized for online delivery, but are
not copy-edited or typeset. Technical support issues arising from
supporting information (other than missing files) should be
addressed to the authors.
References
Allaire P. D., Ritter B., Thomas S. et al. (2006) Connecdenn, a novel
DENN domain-containing protein of neuronal clathrin-coated
vesicles functioning in synaptic vesicle endocytosis. J. Neurosci.
26, 13202–13212.
Armstrong J. N., Saganich M. J., Xu N. J., Henkemeyer M., Heinemann
S. F. and Contractor A. (2006) B-ephrin reverse signaling is
required for NMDA-independent long-term potentiation of mossy
fibers in the hippocampus. J. Neurosci. 26, 3474–3481.
Banker G. A. and Cowan W. M. (1977) Rat hippocampal neurons in
dispersed cell culture. Brain Res. 126, 397–425.
Bendayan M. (1984) Protein A-gold electron microscopic immunocy-
tochemistry: methods, applications, and limitations. J. Electron
Microsc. Tech. 1, 243–270.
Bouvier D., Corera A. T., Tremblay M. E., Riad M., Chagnon M., Murai
K. K., Pasquale E. B., Fon E. A. and Doucet G. (2008) Pre-syn-
aptic and post-synaptic localization of EphA4 and EphB2 in adult
mouse forebrain. J. Neurochem. 106, 682–695.
Bouzioukh F., Wilkinson G. A., Adelmann G., Frotscher M., Stein V.
and Klein R. (2007) Tyrosine phosphorylation sites in ephrinB2 are
164 | D. Bouvier et al.
required for hippocampal long-term potentiation but not long-term
depression. J. Neurosci. 27, 11279–11288.
Bronfman F. C., Tcherpakov M., Jovin T. M. and Fainzilber M. (2003)
Ligand-induced internalization of the p75 neurotrophin receptor: a
slow route to the signaling endosome. J. Neurosci. 23, 3209–3220.
Carmona M. A., Murai K. K., Wang L., Roberts A. J. and Pasquale E. B.
(2009) Glial ephrin-A3 regulates hippocampal dendritic spine
morphology and glutamate transport. Proc. Natl Acad. Sci. U S A
106, 12524–12529.
Carter R. E. and Sorkin A. (1998) Endocytosis of functional epidermal
growth factor receptor-green fluorescent protein chimera. J. Biol.
Chem. 273, 35000–35007.
Contractor A., Rogers C., Maron C., Henkemeyer M., Swanson G. T.
and Heinemann S. F. (2002) Trans-synaptic Eph receptor-ephrin
signaling in hippocampal mossy fiber LTP. Science 296, 1864–
1869.
Dalva M. B. (2007) There’s more than one way to skin a chimaerin.
Neuron 55, 681–684.
Deininger K., Eder M., Kramer E. R., Zieglgansberger W., Dodt H. U.,
Dornmair K., Colicelli J. and Klein R. (2008) The Rab5 guanylate
exchange factor Rin1 regulates endocytosis of the EphA4 receptor
in mature excitatory neurons. Proc. Natl Acad. Sci. U S A 105,
12539–12544.
Dresbach T., Torres V., Wittenmayer N. et al. (2006) Assembly of active
zone precursor vesicles: obligatory trafficking of presynaptic
cytomatrix proteins Bassoon and Piccolo via a trans-Golgi
compartment. J. Biol. Chem. 281, 6038–6047.
Fattorini G., Verderio C., Melone M., Giovedi S., Benfenati F., Matteoli
M. and Conti F. (2009) VGLUT1 and VGAT are sorted to the same
population of synaptic vesicles in subsets of cortical axon termi-
nals. J. Neurochem. 110, 1538–1546.
Filosa A., Paixao S., Honsek S. D. et al. (2009) Neuron-glia commu-
nication via EphA4/ephrin-A3 modulates LTP through glial glu-
tamate transport. Nat. Neurosci. 12, 1285–1292.
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.
Frank C. A., Pielage J. and Davis G. W. (2009) A presynaptic homeo-
static signaling system composed of the Eph receptor, ephexin,
Cdc42, and CaV2.1 calcium channels. Neuron 61, 556–569.
Fu W. Y., Chen Y., Sahin M. et al. (2007) Cdk5 regulates EphA4-
mediated dendritic spine retraction through an ephexin1-dependent
mechanism. Nat. Neurosci. 10, 67–76.
Girard M., Allaire P. D., Blondeau F. and McPherson P. S. (2005a)
Isolation of clathrin-coated vesicles by differential and density
gradient centrifugation. Curr. Protoc. Cell. Biol. Chapter 3, Unit
3.13. doi: 10.1002/0471143030.cb0313s26.
Girard M., Allaire P. D., McPherson P. S. and Blondeau F. (2005b)
Non-stoichiometric relationship between clathrin heavy and light
chains revealed by quantitative comparative proteomics of clathrin-
coated vesicles from brain and liver. Mol. Cell Proteomics 4, 1145–
1154.
Goldshmit Y., McLenachan S. and Turnley A. (2006) Roles of Eph
receptors and ephrins in the normal and damaged adult CNS. Brain
Res. Rev. 52, 327–345.
Gomes R. A., Hampton C., El-Sabeawy F., Sabo S. L. and McAllister A.
K. (2006) The dynamic distribution of TrkB receptors before,
during, and after synapse formation between cortical neurons.
J. Neurosci. 26, 11487–11500.
Grimes M. L., Zhou J., Beattie E. C. et al. (1996) Endocytosis of acti-
vated TrkA: evidence that nerve growth factor induces formation of
signaling endosomes. J. Neurosci. 16, 7950–7964.
 2010 The Authors
Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
Grimes M. L., Beattie E. and Mobley W. C. (1997) A signaling organelle
containing the nerve growth factor-activated receptor tyrosine
kinase, TrkA. Proc. Natl Acad. Sci. U S A 94, 9909–9914.
Grunwald I. C., Korte M., Adelmann G., Plueck A., Kullander K.,
Adams R. H., Frotscher M., Bonhoeffer T. and Klein R. (2004)
Hippocampal plasticity requires postsynaptic ephrinBs. Nat. Neu-
rosci. 7, 33–40.
Henderson J. T., Georgiou J., Jia Z., Robertson J., Elowe S., Roder J. C.
and Pawson T. (2001) The receptor tyrosine kinase EphB2 regu-
lates NMDA-dependent synaptic function. Neuron 32, 1041–1056.
Himanen J. P. and Nikolov D. B. (2003) Eph receptors and ephrins. Int.
J. Biochem. Cell Biol. 35, 130–134.
Hopper N. A. and O’Connor V. (2005) Ephrin tempers two-faced
synaptojanin 1. Nat. Cell Biol. 7, 454–456.
Huttner W. B., Schiebler W., Greengard P. and De Camilli P. (1983)
Synapsin I (protein I), a nerve terminal-specific phosphoprotein.
III. Its association with synaptic vesicles studied in a highly
purified synaptic vesicle preparation. J. Cell Biol. 96, 1374–
1388.
Irie F., Okuno M., Pasquale E. B. and Yamaguchi Y. (2005) EphrinB-EphB
signalling regulates clathrin-mediated endocytosis through tyrosine
phosphorylation of synaptojanin 1. Nat. Cell Biol. 7, 501–509.
Jiang X., Huang F., Marusyk A. and Sorkin A. (2003) Grb2 regulates
internalization of EGF receptors through clathrin-coated pits.
Mol. Biol. Cell 14, 858–870.
Jin Y. and Garner C. C. (2008) Molecular mechanisms of presynaptic
differentiation. Annu. Rev. Cell Dev. Biol. 24, 237–262.
Jung N. and Haucke V. (2007) Clathrin-mediated endocytosis at syn-
apses. Traffic 8, 1129–1136.
Klein R. (2004) Eph/ephrin signaling in morphogenesis, neural devel-
opment and plasticity. Curr. Opin. Cell Biol. 16, 580–589.
Klein R. (2009) Bidirectional modulation of synaptic functions by Eph/
ephrin signaling. Nat. Neurosci. 12, 15–20.
Konstantinova I., Nikolova G., Ohara-Imaizumi M., Meda P., Kucera T.,
Zarbalis K., Wurst W., Nagamatsu S. and Lammert E. (2007)
EphA-Ephrin-A-mediated beta cell communication regulates
insulin secretion from pancreatic islets. Cell 129, 359–370.
Lim Y. S., McLaughlin T., Sung T. C., Santiago A., Lee K. F. and O’Leary
D. D. (2008) p75(NTR) mediates ephrin-A reverse signaling
required for axon repulsion and mapping. Neuron 59, 746–758.
Marler K. J., Becker-Barroso E., Martinez A. et al. (2008) A TrkB/
EphrinA interaction controls retinal axon branching and synapto-
genesis. J. Neurosci. 28, 12700–12712.
Marston D. J., Dickinson S. and Nobes C. D. (2003) Rac-dependent
trans-endocytosis of ephrinBs regulates Eph-ephrin contact repul-
sion. Nat. Cell Biol. 5, 879–888.
Martone M. E., Holash J. A., Bayardo A., Pasquale E. B. and Ellisman
M. H. (1997) Immunolocalization of the receptor tyrosine kinase
EphA4 in the adult rat central nervous system. Brain Res. 771,
238–250.
McPherson P. S., Kay B. K. and Hussain N. K. (2001) Signaling on the
endocytic pathway. Traffic 2, 375–384.
Monami G., Emiliozzi V. and Morrione A. (2008) Grb10/Nedd4-
mediated multiubiquitination of the insulin-like growth factor
receptor regulates receptor internalization. J. Cell. Physiol. 216,
426–437.
Murai K. K. and Pasquale E. B. (2004) Eph receptors, ephrins, and
synaptic function. Neuroscientist 10, 304–314.
Murai K. K., Nguyen L. N., Irie F., Yamaguchi Y. and Pasquale E. B.
(2003) Control of hippocampal dendritic spine morphology
through ephrin-A3/EphA4 signaling. Nat. Neurosci. 6, 153–160.
Nakata T., Terada S. and Hirokawa N. (1998) Visualization of the
dynamics of synaptic vesicle and plasma membrane proteins in
living axons. J. Cell Biol. 140, 659–674.

Parker M., Roberts R., Enriquez M., Zhao X., Takahashi T., Pat Cerretti
D., Daniel T. and Chen J. (2004) Reverse endocytosis of trans-
membrane ephrin-B ligands via a clathrin-mediated pathway.
Biochem. Biophys. Res. Commun. 323, 17–23.
Pasquale E. B. (2005) Eph receptor signalling casts a wide net on cell
behaviour. Nat. Rev. Mol. Cell Biol. 6, 462–475.
Poupon V., Girard M., Legendre-Guillemin V., Thomas S., Bourbonniere
L., Philie J., Bright N. A. and McPherson P. S. (2008) Clathrin light
chains function in mannose phosphate receptor trafficking via reg-
ulation of actin assembly. Proc. Natl Acad. Sci. U S A 105, 168–173.
Richardson D. S., Lai A. Z. and Mulligan L. M. (2006) RET ligand-
induced internalization and its consequences for downstream sig-
naling. Oncogene 25, 3206–3211.
Rodenas-Ruano A., Perez-Pinzon M. A., Green E. J., Henkemeyer M.
and Liebl D. J. (2006) Distinct roles for ephrinB3 in the formation
and function of hippocampal synapses. Dev. Biol. 292, 34–45.
Rosenmund C., Rettig J. and Brose N. (2003) Molecular mechanisms of
active zone function. Curr. Opin. Neurobiol. 13, 509–519.
Shapira M., Zhai R. G., Dresbach T., Bresler T., Torres V. I., Gundel-
finger E. D., Ziv N. E. and Garner C. C. (2003) Unitary assembly
of presynaptic active zones from Piccolo–Bassoon transport vesi-
cles. Neuron 38, 237–252.
Shimizu H., Kawamura S. and Ozaki K. (2003) An essential role of Rab5
in uniformity of synaptic vesicle size. J. Cell Sci. 116, 3583–3590.
Soans C., Holash J. A. and Pasquale E. B. (1994) Characterization of the
expression of the Cek8 receptor-type tyrosine kinase during
development and in tumor cell lines. Oncogene 9, 3353–3361.
Sorkina T., Huang F., Beguinot L. and Sorkin A. (2002) Effect of
tyrosine kinase inhibitors on clathrin-coated pit recruitment and
internalization of epidermal growth factor receptor. J. Biol. Chem.
277, 27433–27441.
Star E. N., Newton A. J. and Murthy V. N. (2005) Real-time imaging of
Rab3a and Rab5a reveals differential roles in presynaptic function.
J. Physiol. 569, 103–117.
 2010 The Authors
Journal Compilation  2010 International Society for Neurochemistry, J. Neurochem. (2010) 113, 153–165
EphA4 in clathrin and synaptic vesicles | 165
Takamori S., Riedel D. and Jahn R. (2000) Immunoisolation of GABA-
specific synaptic vesicles defines a functionally distinct subset of
synaptic vesicles. J. Neurosci. 20, 4904–4911.
Tao-Cheng J. H. (2007) Ultrastructural localization of active zone and
synaptic vesicle proteins in a preassembled multi-vesicle transport
aggregate. Neuroscience 150, 575–584.
Tremblay M. E., Riad M., Bouvier D., Murai K. K., Pasquale E. B.,
Descarries L. and Doucet G. (2007) Localization of EphA4 in axon
terminals and dendritic spines of adult rat hippocampus. J. Comp.
Neurol. 501, 691–702.
Tremblay M. E., Riad M., Chierzi S., Murai K. K., Pasquale E. B. and
Doucet G. (2009) Developmental course of EphA4 cellular and
subcellular localization in the postnatal rat hippocampus. J. Comp.
Neurol. 512, 798–813.
Watson F. L., Heerssen H. M., Moheban D. B., Lin M. Z., Sauvageot C.
M., Bhattacharyya A., Pomeroy S. L. and Segal R. A. (1999) Rapid
nuclear responses to target-derived neurotrophins require retro-
grade transport of ligand-receptor complex. J. Neurosci. 19, 7889–
7900.
Yamaguchi Y. and Pasquale E. B. (2004) Eph receptors in the adult
brain. Curr. Opin. Neurobiol. 14, 288–296.
Zaremba S. and Keen J. H. (1983) Assembly polypeptides from coated
vesicles mediate reassembly of unique clathrin coats. J. Cell Biol.
97, 1339–1347.
Zheng J., Shen W. H., Lu T. J. et al. (2008) Clathrin-dependent
endocytosis is required for TrkB-dependent Akt-mediated neuronal
protection and dendritic growth. J. Biol. Chem. 283, 13280–13288.
Zhou L., Martinez S. J., Haber M. et al. (2007) EphA4 signaling
regulates phospholipase Cgamma1 activation, cofilin membrane
association, and dendritic spine morphology. J. Neurosci. 27,
5127–5138.
Zimmer M., Palmer A., Kohler J. and Klein R. (2003) EphB-ephrinB
bi-directional endocytosis terminates adhesion allowing contact
mediated repulsion. Nat. Cell Biol. 5, 869–878.