Molecular Biology of the Cell
Vol. 9, 1549–1563, June 1998
Endobrevin, a Novel Synaptobrevin/VAMP-Like
Protein Preferentially Associated with the Early
Siew Heng Wong,* Tao Zhang,* Yue Xu,* V. Nathan Subramaniam,*
Gareth Griffiths,†and Wanjin Hong*‡
*Membrane Biology Laboratory, Institute of Molecular and Cell Biology, Singapore 117609, Singapore;
and†European Molecular Biology Laboratory, 69 Heidelberg, Federal Republic of Germany
Submitted May 27, 1997; Accepted March 4, 1998
Monitoring Editor: Ari Helenius
Synaptobrevins/vesicle-associated membrane proteins (VAMPs) together with syntaxins
and a synaptosome-associated protein of 25 kDa (SNAP-25) are the main components of
a protein complex involved in the docking and/or fusion of synaptic vesicles with the
presynaptic membrane. We report here the molecular, biochemical, and cell biological
characterization of a novel member of the synaptobrevin/VAMP family. The amino acid
sequence of endobrevin has 32, 33, and 31% identity to those of synaptobrevin/VAMP-1,
synaptobrevin/VAMP-2, and cellubrevin, respectively. Membrane fractionation studies
demonstrate that endobrevin is enriched in membrane fractions that are also enriched in
the asialoglycoprotein receptor. Indirect immunofluorescence microscopy establishes
that endobrevin is primarily associated with the perinuclear vesicular structures of the
early endocytic compartment. The preferential association of endobrevin with the early
endosome was further established by electron microscopy (EM) immunogold labeling. In
vitro binding assays show that endobrevin interacts with immobilized recombinant
?-SNAP fused to glutathione S-transferase (GST). Our results highlight the general
importance of members of the synaptobrevin/VAMP protein family in membrane traffic
and provide new avenues for future functional and mechanistic studies of this protein as
well as the endocytotic pathway.
Protein trafficking along the exocytotic and endocy-
totic pathways is a vital and fundamental cellular
process. Proteins destined for the exocytotic pathway
are initially targeted to the endoplasmic reticulum and
transported through the Golgi apparatus. At the trans-
Golgi network (TGN), proteins are sorted to distinct
post-Golgi structures such as the plasma membrane
(or its subdomains) and the endosomal and lysosomal
compartments (Palade, 1975; Mellman and Simons,
1992; Hong and Tang, 1993; Rothman and Wieland,
1996; Schekman and Orci, 1996). The endosomal com-
partment plays a central role in cellular physiology
(Gruenberg and Maxfield, 1995; Mellman, 1996; Rob-
inson et al., 1996). Endocytosed proteins are internal-
ized from the plasma membrane via coated vesicles
and then delivered to the early endosomal compart-
ment, from which proteins can be either recycled to
the plasma membrane or delivered to the late endo-
somal compartment and subsequently to the lysosome
or the TGN.
Intracellular trafficking is primarily mediated by
various types of transport vesicles that bud from a
donor membrane and then fuse with a specific cognate
target membrane. To achieve such specificity, the sol-
uble N-ethylmaleimide-sensitive factor attachment
protein receptor (SNARE) hypothesis proposes that
specific docking and fusion of vesicles with the cog-
nate membrane compartment is mediated by specific
interaction between the vesicle-associated SNAREs
with the cognate target SNAREs on the target mem-
‡Corresponding author. E-mail address: firstname.lastname@example.org.
© 1998 by The American Society for Cell Biology1549
brane (Rothman and Warren, 1994; Bennett, 1995;
Whiteheart and Kubalek, 1995; Pfeffer, 1996). The ma-
jority of the known SNAREs (except for a synapto-
some-associated protein of 25 kDa [SNAP-25] and
Ykt6p) are anchored to their respective membranes by
their C-terminal hydrophobic domain (Sogaard et al.,
1994; Pfeffer, 1996; McNew et al., 1997). Synaptobre-
vins/vesicle-associated membrane proteins (VAMPs)
are vesicle-associated SNAREs associated with the
synaptic vesicles, whereas syntaxin 1 and SNAP-25
are target SNAREs associated with the presynaptic
membrane. The specific pairing of synaptobrevins/
VAMPs with the syntaxin 1–SNAP-25 complex plays a
key role in the docking and fusion of synaptic vesicles
with the presynaptic membrane (Sollner et al., 1993;
Rothman and Warren, 1994; Scheller, 1995; Su ¨dhof,
Three members of the synaptobrevin/VAMP family
and cellubrevin) have been identified and character-
ized in mammalian cells (McMahon et al., 1993; Galli et
al., 1994; Jahn and Sudhof, 1994). Synaptobrevin/
VAMP-1 and synaptobrevin/VAMP-2 are highly ho-
mologous proteins that are expressed in neurons and
endocrine cells, whereas cellubrevin is ubiquitously
expressed and associated with the endosomal com-
partment. In this report, we describe the molecular,
biochemical, and cell biological characterization of a
novel mammalian protein that shares significant
amino acid identity to synaptobrevin/VAMPs and
cellubrevin. Antibodies raised against the recombi-
nant protein recognize a 15-kDa protein that is pref-
erentially localized to the early endosomal compart-
MATERIALS AND METHODS
NRL (normal rat liver), A431 (human epidermoid carcinoma),
NIH3T3 (mouse embryonic fibroblast), VERO (African green mon-
key kidney), C6 (rat glial), CV1 (monkey kidney), HeLa, and OKT9
and HB21 (mouse hybridomas expressing monoclonal antibodies
against the human transferrin receptor) cells were obtained from
American Type Culture Collection (Rockville, MD). MDCK II (Ma-
din-Darby canine kidney strain II) was a generous gift from Dr. Kai
Simons (European Molecular Biology Laboratory, Heidelberg, Ger-
many). Synthetic oligonucleotides were from Oligos Etc (Wilson-
ville, OR). The Pyrococcus furiosus DNA polymerase was a product
of Stratagene (La Jolla, CA). The Taq DNA polymerase and Hybond
C-extra nitrocellulose filters were obtained from Amersham (Little
Chalford, Buckinghamshire, United Kingdom). Glutathione Sepha-
rose 4B was from Pharmacia (Upsala, Sweden). Fluorescein isothio-
cyanate-conjugated goat anti-mouse IgG, rhodamine-conjugated
goat anti-rabbit IgG, and restriction enzymes were from Boehringer
Mannheim (Mannheim, Germany). Brefeldin A (BFA) was obtained
from Epicentre Technologies (Madison, WI). Wortmannin was pur-
chased from Sigma (St. Louis, MO). Local New Zealand White
rabbits were purchased from the Sembawang Laboratory Animals
Centre (Singapore). Freund’s adjuvants (complete and incomplete)
were from Life Technologies–BRL (Bethesda, MD).
cDNA Cloning and Sequencing
Three human expressed sequence tag (EST) clones (accession num-
bers T49805, R01789, and T63214) encoding overlapping sequences
similar to those of known members of the VAMP family were
identified with the use of the BLAST program. The complete coding
sequence of endobrevin was confirmed by sequencing EST clone
Expression of Recombinant Proteins in Bacteria
For the production of glutathione S-transferase (GST) fusion pro-
teins, a DNA fragment encoding residues 1–75 derived from PCR
with the use of primer 1 (5?-GGGAATTCTAACCATGGAGGAAG-
CCAGTGAAGGTG) and primer 2 (5?-GGGTCTAGATCACTTCAC-
GTTCTTCCACCAGAATT) was digested with EcoRI and XbaI re-
striction enzymes and subcloned into the EcoRI and XbaI sites of the
bacterial expression vector pGEX-KG (Guan and Dixon, 1991). The
ligated DNA was transformed into DH5? cells, and ampicillin-
resistant colonies expressing the GST fusion proteins were screened
as described (Sambrook et al., 1989). For the production and purifi-
cation of GST fusion proteins, 100 ml of Lurı ´a-Bertanı ´ (LB) broth
containing 100 ?g/ml ampicillin were inoculated with 100 ?l of an
overnight culture of a expressing clone and grown at 37°C with
shaking. The next day, 25 ml of the culture were used to inoculate
1000 ml of LB broth containing 100 ?g/ml ampicillin in a 5000 ml
flask and were incubated at 37°C with shaking. When the OD 600 of
the culture reached 0.4, IPTG was added to a final concentration of
1 mM, and growing continued for another 4 h. Cells were pelleted
and resuspended in 50 ml of lysis buffer (phosphate buffered-saline
[PBS] containing 50 mM Tris [pH 8], 0.1% Triton X-100, 0.5 mM
MgCl2, 1 mg/ml lysozyme, 5 mM DTT, and a cocktail of protease
inhibitors [0.5 mM phenylmethylsulfonyl fluoride and antipain,
aprotinin, and leupeptin at 10 ?g/ml]). Lysis was performed by
incubation on ice for 1 h followed by sonication. Lysates were
clarified by centrifugation and applied to a glutathione Sepharose
4B column (Pharmacia). After the column was washed with GST
purification buffer (PBS containing 50 mM Tris [pH 8.0], 0.1% Triton
X-100, 0.5 mM MgCl2), GST fusion protein was eluted with 15 mM
reduced glutathione in Tris (pH 8.0) (containing 5% glycerol) and
stored at ?20°C. For the preparation of hexahistidine (HisX6)-
tagged cellubrevin, 1000 ml of LB broth (with 100 ?g/ml ampicillin)
was inoculated with 20 ml of an overnight culture [BL21(DE3) cells
containing the HisX6-tagged cellubrevin in the pET23d vector from
Novagen (Madison, WI)]. Cells were grown until an OD 600 of 0.8,
and IPTG was added to a final concentration of 1 mM. After grow-
ing overnight at room temperature, the cells were harvested by
centrifugation at 4000 ? g for 20 min, and the pellet was resus-
pended in cracking buffer (100 mM HEPES [pH 7.3], 500 mM KCl,
5 mM MgCl2, 2 mM ?-mercaptoethanol, 1 mM phenylmethylsulfo-
nyl fluoride, 0.1% Triton X-100) at four to five volumes per gram of
wet weight. Lysozyme was added to a final concentration of 1
mg/ml, and the cells were sonicated on ice (1 min burst) for a total
of 3 min. Lysate was then centrifuged at 10,000 ? g for 30 min at
4°C, and the supernatant was added to 8 ml of a 50% slurry of
Ni-NTA resin [preequilibrated with buffer A (20 mM HEPES [pH
7.3], 200 mM KCl, 10% glycerol, 2 mM ?-mercaptoethanol) contain-
ing 25 mM imidazole] and then incubated at 4°C for 1 h with
agitation. Resin was then loaded into a column and washed with 10
column volumes of buffer A containing 50 mM imidazole before
eluting with 15 ml of buffer A containing 250 mM imidazole. Frac-
tions of 1 ml each were collected and analyzed by SDS-PAGE.
Fractions containing proteins were pooled and dialyzed against
PBS. For the preparation of HisX6-N-ethylmaleimide-sensitive fac-
tor (NSF) fusion protein, 0.2 mM ATP was added to all the buffers
to maintain the active structure of NSF.
Preparation of Polyclonal Antibodies
GST–endobrevin (300 ?g) emulsified in Freund’s adjuvant was
injected subcutaneously into local New Zealand rabbits. The subse-
S.H. Wong et al.
Molecular Biology of the Cell1550
quent injections (boosters) containing a similar amount of antigen
were then performed every 2 wk. Rabbit serums were collected 10 d
after the second and subsequent booster injections. For affinity
purification, serum was diluted twice with PBS and then sequen-
tially incubated with cyanogen bromide (CNBr)-activated sepha-
rose beads coupled with GST and then GST–endobrevin (3 mg/ml
CNBr sepharose beads) for 2 h at RT. The GST–endobrevin-coupled
beads were washed extensively with 10 volumes of PBS and then
eluted with 10 ml of immunopure IgG elution buffer (Pierce, Rock-
Immunofluorescence microscopy was performed as described pre-
viously (Wong et al., 1992, 1998; Wong and Hong, 1993; Lowe et al.,
1996; Xu et al., 1997). Briefly, cells grown on coverslips were washed
twice with PBS with 1 mM CaCl2and 1 mM MgCl2(PBSCM) and
then fixed with 3% paraformaldehyde. After extensive washing,
cells were permeabilized with PBSCM with 0.2% saponin (PBSCMS)
for 20 min and then sequentially incubated with the primary (rabbit)
and secondary (rhodamine-conjugated goat anti-rabbit IgG) anti-
bodies for 1 h at room temperature. Cells were then washed exten-
sively with PBSCMS, mounted in a drop of Vectastain (Vector
Laboratories, Burlingame, CA), observed with the axiophot micro-
scope (Carl Zeiss, Thornwood, NY), and then photographed with
Kodak Tri-X 400 film. For the treatment of cells with BFA or wort-
mannin, A431 cells grown on coverslips were incubated with OKT9
and HB21 monoclonal antibodies against human transferrin recep-
tor for 30 min, washed with DMEM media (with 10% FBS), and then
treated with BFA (10 ?g/ml) or wortmannin (500 nM). After incu-
bating for 60 min at 37°C, cells were washed twice with PBSCM,
fixed in 3% paraformaldehyde, permeabilized, and then incubated
with antibodies against endobrevin (polyclonal) for 60 min at room
temperature. After extensive washing with PBSCMS, cells were
incubated with rhodamine-conjugated goat anti-rabbit IgG (10 ?g/
ml) and FITC-conjugated sheep anti-mouse IgG (10 ?g/ml) for 60
min at room temperature. Coverslips were then mounted as de-
scribed above after extensive washing with PBSCMS.
Cryosections and electron microscopy (EM) immunogold double
labeling were performed as described previously (Slot et al., 1991;
Griffiths, 1993; Griffiths et al., 1994).
Preparation of Golgi-Enriched Membranes
Preparation and subfractionation of membranes were performed as
described previously (Subramaniam et al., 1992; Wong et al., 1998).
Briefly, livers from Harlan Sprague Dawley (Indianapolis, IN) rats
were homogenized in three volumes (g/ml) of the homogenization
buffer (25 mM HEPES [pH 7.3], 5 mM MgCl2, 1 mM PMSF) con-
taining 0.25 M sucrose with a teflon pestle and centrifuged at
10,000 ? g for 10 min to remove unbroken cells, nuclei, and mito-
chondria. The supernatants were then recentrifuged at 100,000 ? g
in a Beckman Ty45Ti rotor for 1 h. The supernatant of this centrif-
ugation that consists mainly of cytosol was collected, and the total
membrane pellet was resuspended in a minimal volume of homog-
enization buffer containing 0.25 M sucrose. The membrane suspen-
sion was then adjusted to a final concentration of 1.25 M sucrose,
overlaid with step gradients of 10 ml of 1.1 M sucrose, 10 ml of 1.0
M sucrose, and 5.0 ml of 0.5 M sucrose in homogenization buffer,
and then centrifuged at 28,000 rpm for 3 h in a Beckman SW 28
rotor. The G1 (at the 0.5 M/1.0 M sucrose interphase), G2 (at the 1.0
M/1.1 M sucrose interphase), and the pellet that are enriched in
Golgi, endosomes/Golgi, and microsome membranes, respectively,
were collected and used for the subsequent experiments.
Western Blotting (Immunoblotting) Analysis
Proteins separated by SDS-PAGE were transferred to a Hybond
C-extra nitrocellulose filter and blocked and incubated sequentially
with primary antibodies (10 ?g/ml) and125I-protein A (0.1 ?Ci/ml).
Incubation of the filter with primary antibodies and125I-protein A
and washing of the filter was done in blocking buffer (PBS contain-
ing 5% skim milk and 0.05% Tween 20). Filters were washed with
PBS and PBS containing 0.05% Tween 20 and then processed for
autoradiography. Some immunoblots were analyzed by using the
Supersignal Chemiluminescent Kit (Pierce) according to the proto-
col recommended by the manufacturer.
Treatments of Membranes with Salts and Detergents
Preparation and subfractionation of membranes were performed as
described previously (Subramaniam et al., 1992; Wong et al., 1998),
Endobrevin-enriched membrane (500 ?g) was extracted on ice for
1 h in 100 ?l of either PBS, 2.5 M urea, 0.15 M sodium bicarbonate
(pH 11.0), 2 M KCl, 1% Triton X-100 or 1% NP-40 and then centri-
fuged at 100,000 ? g for 1 h at 4°C. The supernatant was collected,
and the pellet was resuspended in 100 ?l of 1? SDS sample buffer.
Aliquots (20 ?l) from both the supernatant as well as the pellet were
separated by SDS-PAGE and analyzed by immunoblotting.
In Vitro Binding Assays
In vitro binding assays were performed as described previously
(Wong et al., 1998). Endobrevin-enriched membranes (3 mg) were
extracted for 1 h (at 4°C) in 500 ?l of incubation buffer (100 mM KCl,
20 mM HEPES [pH 7.3], 2 mM EDTA, 2 mM DTT, 0.2 mM ATP)
containing 1% Triton X-100. The extracted membranes were diluted
with 500 ?l of incubation buffer without Triton X-100 and then
centrifuged at 100,000 ? g at 4°C for 1 h. The extracted proteins in
the supernatant were used for the subsequent binding assays. GST–
?-SNAP beads (2–5 ?g) were washed twice with incubation buffer
containing 0.5% Triton X-100 (1 ml each) and then incubated with
the different amounts of membrane extract (0, 10, 20, 40, 80, 100, 150,
and 200 ?g) in a total volume of 100 ?l at 4°C for 3 h with agitation.
GST–?-SNAP beads were then washed twice with incubation buffer
containing 0.5% Triton X-100, once with incubation buffer contain-
ing 0.1% Triton X-100, and twice with incubation buffer without
Triton X-100 before separation on SDS-PAGE and immunoblotting
In the complex dissociation experiment, the incubation of mem-
brane extract (200 ?g) with GST–?-SNAP was performed in the
presence of increasing indicated amounts of NSF (0, 1, 2.5, 5, and 10
?g) under buffer conditions that either promote complex assembly
(incubation buffer with 1 mM ATP) or disassembly (incubation
buffer with 1 mM ATP and MgCl2).
Endobrevin Is a New Member of the Synaptobrevin/
VAMP Protein Family
Three human EST clones (accession numbers T49805,
R01789, and T63214) encoding overlapping protein
sequences similar to those of known members of the
synaptobrevin/VAMP family were identified during
database searches. Compilation of these sequences
yielded the full-length nucleotide sequence for endo-
brevin. One of the EST clones (R01789) was sequenced
completely from both ends to confirm the assembled
sequence. Within the 393 bp DNA sequence, a single
open reading frame from nucleotide number 25 to 327
was identified that predicts a protein of ?12 kDa
A Novel Protein of the Early Endosome
Vol. 9, June 19981551
(Figure 1A). This open reading frame is flanked in-
frame with a strong initiation Met codon (Kozak, 1984)
at the 5?-end and a stop codon at the 3?-end. In addi-
tion, there is an in-frame stop codon upstream from
the initiation Met at nucleotide 19. Alignment of the
deduced 100 amino acid sequence of this protein with
several members of the synaptobrevin/VAMP family
is shown in Figure 1B. This protein is ?32, 33, and 31%
identical to synaptobrevin/VAMP-1, synaptobrevin/
VAMP-2, and cellubrevin, respectively, and because it
is localized to the endosomes (see below), it was thus
named endobrevin (endosome-associated synaptobre-
vin-like protein; Figure 1B). The amino acid homology
of endobrevin to these proteins spans the entire
polypeptide. However, in contrast to the high degree
of homology between cellubrevin and synaptobrevin/
VAMPs (70–75%), endobrevin is more distantly re-
lated to synaptobrevins/VAMPs and cellubrevin.
Endobrevin Is a 15-kDa Protein Present in the Total
Membrane Fraction of MDCK II, Normal Rat
Kidney (NRK), and HeLa Cells
The predicted cytoplasmic domain (residues 1–75)
was expressed as a fusion protein to GST (GST–endo-
brevin) and used to raise polyclonal antibodies against
endobrevin. Immunoblot analysis was used to detect
endobrevin in total membrane preparations from
three different cell lines (MDCK II, NRK, and HeLa).
As shown in Figure 2 (lanes 1–3), endobrevin-specific
antibodies detected a 15-kDa polypeptide in all three
cell lines. Detection of this polypeptide by endobrevin
antibodies was blocked by preincubation of antibodies
with GST–endobrevin (lanes 7–9) but not by a mixture
of HisX6-tagged cellubrevin and GST (lanes 4–6),
demonstrating the specificity of the antibodies.
Endobrevin Is an Integral Membrane Protein
Enriched in Membrane Fractions that Are Also
Enriched for the Asialoglycoprotein Receptor
Rat liver membrane preparations were fractionated by
a discontinuous sucrose gradient, and the fractions
were analyzed for the presence of endobrevin by im-
munoblot. As shown in Figure 3A, endobrevin was
found to be enriched in G1 (membrane fraction at the
0.5–1.0 M sucrose interface) and even more in G2
(membrane fraction at the 1.0–1.1 M sucrose interface)
membrane fractions. The G1 and G2 membrane frac-
domain is boxed. (B) The alignment of amino acid sequences of endobrevin, synaptobrevins/VAMPs, and cellubrevin. The identical residues
are shaded. The complete nucleotide sequence of endobrevin cDNA has been submitted to GenBank under the accession number AF053233.
(A) The nucleotide (upper line) and deduced amino acid (lower line) sequence of endobrevin. The C-terminal hydrophobic
S.H. Wong et al.
Molecular Biology of the Cell1552
tions were also enriched in the asialoglycoprotein re-
ceptor subunits R1 (46 kDa) and R2/3 (50 and 58 kDa)
that are localized to the early endosomal compartment
and the plasma membrane (Graeve et al., 1990; Spies,
1990; Lodish, 1991) (Figure 3, B and C). Similar to
endobrevin, the asialoglycoprotein receptor subunits
are enriched more in the G2 membrane fraction. Un-
der similar conditions, the Golgi marker ?-2,6-sialyl-
transferase is enriched more in the G1 than in the G2
fraction (Wong, Zhang, Xu, Subramaniam, Griffiths,
and Hong, unpublished observations). The enrich-
ment of endobrevin in these membrane fractions and
the deduced amino acid sequence suggest that endo-
brevin is an integral membrane protein. To confirm
this, G2 membrane fractions were extracted with PBS,
2.5 M urea, 0.15 M sodium bicarbonate (pH 11.0), 2 M
KCl, 1% Triton X-100, and 1% NP-40. Figure 4 shows
that endobrevin is not solubilized in PBS, in 2 M KCl,
in 2.5 M urea, and in 0.15 M sodium bicarbonate (pH
11.0) but is solubilized effectively by 1% Triton X-100
and by 1% NP-40, confirming that endobrevin is in-
deed an integral membrane protein.
HeLa (lanes 3, 6, and 9) cells by Western blot analysis using affinity-purified antibodies against endobrevin. A 15-kDa protein was specifically
detected (lanes 1–3), and the detection of this protein was blocked by preincubation of antibody with GST–endobrevin (lanes 7–9) but not
by a mixture of HisX6-cellubrevin and GST (lanes 4–6).
Detection of endobrevin in the total membrane fraction prepared from MDCK II (lanes 1, 4, and 7), NRK (lanes 2, 5, and 8), and
tor subunits R1- and R2/3-containing membrane fractions. Cytosol
(C), total membrane (TM), microsome-enriched membrane fraction
(M), G1 membrane fraction, and G2 membrane fraction (100 ?g
each) were used in a Western blot analysis using (A) endobrevin
antibodies, (B) polyclonal antibodies against asialoglycoprotein re-
ceptor subunit R1, and (C) antibodies against the asialoglycoprotein
receptor subunits R2/3.
Endobrevin is enriched in the asialoglycoprotein recep-
brane fraction was extracted with a range of different reagents as
indicated and separated by high-speed centrifugation into pellet (P)
and supernatant (S) fractions. Aliquots (100 ?g of proteins) of these
fractions were analyzed by SDS-PAGE and immunoblot using en-
Endobrevin is an integral membrane protein. G2 mem-
A Novel Protein of the Early Endosome
Vol. 9, June 19981553
Endobrevin Interacts with ?-SNAP
An in vitro binding assay was performed to determine
if endobrevin interacts with the general docking and
fusion component ?-SNAP. ?-SNAP was expressed in
E. coli bacteria as a GST fusion protein (the entire
polypeptide of the ?-SNAP protein fused to the C
terminus of the GST protein) and immobilized on
glutathione-agarose beads. Fixed amounts (3 ?g) of
GST–?-SNAP and GST proteins coupled to beads
were incubated with increasing amounts of soluble
membrane extracts of the G2 fraction. Endobrevin
bound to the beads was detected by immunoblotting.
As shown in Figure 5A, endobrevin binds to ?-SNAP
in a dose-dependent manner. Saturation of endobre-
vin binding could be seen after incubation with 80 ?g
or more of membrane extracts (lanes 5–8). Under the
same conditions, endobrevin did not bind GST-cou-
pled beads (Figure 5B). These results suggest that
endobrevin in the G2 membrane extract can interact
with ?-SNAP and that endobrevin is a SNARE (SNAP
receptor). McMahon and Sudhof (1995) have shown
previously that the binding of synaptobrevin/VAMP
to syntaxin is essential for efficient ?-SNAP binding.
This binding was achieved either by forming a com-
posite receptor surface for ?-SNAP or by inducing a
conformational change in syntaxin that results in high-
affinity binding. Therefore, the interaction of endobre-
vin with ?-SNAP is most likely mediated by an endo-
brevin-containing SNARE complex. To demonstrate
this point, we incubated a fixed amount of membrane
extract (stripped by KCl) with GST–?-SNAP beads in
the presence of increasing amounts of recombinant
NSF under buffer conditions that either promote
SNARE-complex assembly or disassembly. As shown
in Figure 5B, under conditions that promote SNARE-
complex assembly (lanes 2–6), substantial amounts of
endobrevin were retained on the GST–?-SNAP beads.
As a positive control, GS28 (a Golgi SNARE) was
similarly retained (Subramaniam et al., 1996, 1997).
However, the association of endobrevin, but not GS28,
with the immobilized GST–?-SNAP is essentially
abolished by NSF under conditions that promote
SNARE-complex disassembly (lanes 8–11). In both the
SNARE-complex assembly and disassembly condi-
tions, only background levels of endobrevin and GS28
were retained by the GST-coupled beads (lanes 12 and
13). These results show that the dissociated GS28 (Sub-
ramaniam et al., 1997) but not endobrevin remains
capable of interacting with ?-SNAP under conditions
that promote SNARE-complex disassembly. Thus,
these results not only demonstrate that the interaction
of endobrevin with ?-SNAP is specific but also further
suggest that endobrevin does not interact with
?-SNAP directly but rather through an endobrevin-
containing SNARE complex.
Endobrevin Is Associated with Vesicular Structures
in Several Cell Lines
As a first step to study the function of endobrevin, we
used polyclonal antibodies against endobrevin to in-
vestigate the subcellular localization of endobrevin in
several mammalian cell lines. Indirect immunofluo-
rescence microscopy revealed labeling of perinuclear
and punctate vesicular structures (throughout the cy-
toplasm) of endobrevin in six different cell lines (CV1,
NIH3T3, C6, NRL, VERO, and MDCK II cells) derived
from five different species (Figure 6A). These vesicular
structures were more concentrated at the perinuclear
region. Figure 6B shows that the labeling of these
structures (panels A, D, and G) was selectively abol-
ished by preincubating the anti-endobrevin antibodies
with GST–endobrevin (panels B, E, and H) but not
with immobilized ?-SNAP. GST–?-SNAP (3 ?g) immobilized on
beads was incubated with increasing amounts of G2 membrane
extract as indicated. After the beads were washed extensively, the
amounts of endobrevin retained by the beads were detected by
immunoblot. (B) The interaction between endobrevin and ?-SNAP
is mediated by an endobrevin-containing SNARE complex. Mem-
brane extracts (200 ?g) were incubated with GST–?-SNAP-coupled
beads (3 ?g) in the presence of the indicated amounts of NSF in
either the SNARE-complex assembly (lanes 2–6) or the SNARE-
complex disassembly (lanes 7–11) buffer. After extensive washing,
the amounts of endobrevin (upper) and GS28 (lower) retained by
the beads were detected by immunoblot. As a binding control,
membrane extracts were incubated with GST-coupled beads in
SNARE-complex assembly (A, lane 12) and disassembly (D, lane 13)
buffer. Lane 1 contains 50 ?g of membrane extract.
(A) Endobrevin in the membrane extracts can interact
S.H. Wong et al.
Molecular Biology of the Cell1554
with a mixture of HisX6-cellubrevin and GST (panels
C, F, and I) in VERO (panels A-C), NRL (panels D-F),
and A431 (G-I) cells. The selective abolishment of en-
dobrevin labeling in these cell lines by GST-endobre-
vin but not by a mixture of HisX6-cellubrevin and GST
further demonstrates the specificity of the antibodies.
Endobrevin Is Associated with the Endocytotic
Compartment Marked by Cell Surface-Internalized
The enrichment of endobrevin in the asialoglycopro-
tein receptor-enriched membrane fractions indicates
that endobrevin may be localized to the endocytic
compartments in conjunction with its intracellular ve-
sicular localization. To pinpoint the exact subcellular
localization of endobrevin, we performed colocaliza-
tion studies using surface-internalized monoclonal an-
tibody against transferrin receptor to mark the early
endosomes in A431 cells. As shown in Figure 7, the
labeling of endobrevin colocalized well with the trans-
ferrin receptor, particularly in the perinuclear vesicu-
lar structures (Figure 7, A and B). After treatment with
BFA for 60 min, endobrevin was observed to colocal-
ize with internalized transferrin receptor in a tubular
network (Figure 7, C and D). Prolonged treatment
with BFA (120 min) caused the tubular network to
concentrate to the perinuclear region that is likely to
be the microtubule-organizing center (MTOC) (Figure
7, E and F). The redistribution of internalized trans-
ferrin receptor by BFA to the tubular network and
subsequent structures around the MTOC have been
reported previously (Klausner et al., 1992; Wood and
Brown, 1992; Whitney et al., 1995). The similar re-
sponse of endobrevin to BFA suggests that endobrevin
is associated with the early endosome.
To define the precise location of endobrevin further,
EM immunogold labeling was performed in cryosec-
tions of J774 cells. As shown in Figure 8, A and B,
endobrevin is indeed enriched in the tubular-vesicular
structures representing the early endosomes marked
by the presence of cell surface-internalized BSA-gold
particles in J774 cells. The labeling of endobrevin to a
much lesser extent was also observed in the plasma
membrane (Figure 8, A and B, indicated “P”) and the
late endosomes (Wong, Zhang, Xu, Subramaniam,
Griffiths, and Hong, unpublished observations). These
results clearly establish that endobrevin is preferen-
tially associated with the early endosome.
Endobrevin Is Localized to the Later
Subcompartment of the Early Endosome
Colocalization of endobrevin with internalized trans-
ferrin receptor in control and BFA-treated cells, in
conjunctionwith the electron
strongly suggests that endobrevin is present primarily
in the early endosomes. To determine which part of
the early endosomal compartment does endobrevin
colocalize with the internalized transferrin receptor,
we determined the kinetics of the cell surface-internal-
ized transferrin receptor through the endocytotic
pathway. Monoclonal antibody (OKT9) against the
ectodomain of the human transferrin receptor was
added to the A431 cells and incubated on ice for 30
min. Cells were then washed twice with DMEM media
and then incubated at 37°C for different periods of
time. At the end of each time point, cells were fixed,
labeled with rabbit antibodies against endobrevin and
then with secondary antibodies (FITC-conjugated an-
ti-mouse IgG and rhodamine-conjugated anti-rabbit
IgG) before viewing under the immunofluorescence
microscope. As shown in Figure 9, surface-bound
monoclonal antibody was only detected on the surface
(B) and then internalized into small peripheral vesic-
ular structures (D). These peripheral structures be-
come concentrated at the perinuclear region and colo-
calize with endobrevin (Figure 9, F and H). These
results suggest that internalized transferrin receptor is
associated initially with peripheral structures that are
slightly enriched for endobrevin. After longer periods
of time, the internalized transferrin receptor moves to
the endobrevin-enriched perinuclear structures, sug-
gesting that endobrevin is enriched in the later sub-
compartment of the early endosomes.
Wortmannin Redistributes Endobrevin to Vacuole-
It has been suggested previously that wortmannin
does not affect the peripheral early endosomes but
causes the later recycling early endosomes to distrib-
ute to the swollen vacuole-like structures (Reaves et
al., 1996). Because endobrevin is localized to the later
perinuclear early endosome, we examined the effect of
wortmannin on the distribution of endobrevin and
internalized transferrin receptor in A431 cells. After
treatment of A431 cells with wortmannin for 30 min, a
tubular network is clearly visible that is positive for
both endobrevin and internalized transferrin receptor
(Figure 10, A and B). A longer treatment (1 h) with
wortmannin caused endobrevin and the internalized
transferrin receptor to redistribute to the swollen vac-
uole-like structures (Figure 10, C and D). Taken to-
gether, these results further suggest that endobrevin is
associated with a later subcompartment of the early
Searching the EST database led to the identification of
endobrevin that is ?31–33% identical to synaptobre-
vin/VAMP-1, synaptobrevin/VAMP-2, and cellubre-
vin. This identity is relatively low compared with the
identities observed among synaptobrevin/VAMP-1,
A Novel Protein of the Early Endosome
Vol. 9, June 19981555
synaptobrevin/VAMP-2, and cellubrevin that are in
the range of 70–75%. Endobrevin is thus a distantly
related member of the synaptobrevin/VAMP family.
In addition, the size of endobrevin is smaller (calcu-
lated size is 12 kDa with an apparent size of 15 kDa),
compared with 18 kDa or more for synaptobrevin/
VAMP-1, synaptobrevin/VAMP-2, and cellubrevin.
The Clostridium tetani and Clostridium botulinum bacte-
slips were processed for indirect immunofluorosence microscopy, as described in detail in Materials and Methods, to detect endobrevin. (B)
The labeling of endobrevin (panels A, D, and G) was abolished by preincubating the antibodies with GST–endobrevin (panels B, E, and H)
but not with a mixture of His-tagged cellubrevin and GST (panels C, F, and I) in VERO (panels A–C), NRL (panels D–F), and A431 (panels
G–I). Bar, 10 ?m.
(A) Perinuclear and punctate labeling for endobrevin is detected in all of the cell lines tested. Cells (as indicated) grown on cover
S.H. Wong et al.
Molecular Biology of the Cell1556
ria produce several neurotoxins that are known to be
potent inhibitors of the exocytotic release of neuro-
transmitters from synaptic vesicles at nerve terminals.
To date, seven serologically distinct botulinal neuro-
toxins (BoNT/A, B, C1, D, E, F, and G) are known.
These neurotoxins act on peripheral motoneurons in
which they cause a blockade of acetylcholine release
and thus produce the clinical manifestation of botu-
lism. In the case of tetanus toxin (TeTx), it blocks the
release of inhibitory neurotransmitters in the nervous
system and thus results in the clinical manifestation of
tetanus. There are specific cleavage sites for these tox-
ins in syntaxins, synaptobrevins/VAMPs (including
cellubrevin), and SNAP-25 (Link et al., 1992; Schiavo et
al., 1992, 1993; Blasi et al., 1993a,b; McMahon et al.,
1993; Binz et al., 1994; Galli et al., 1994). Consensus
sequences of toxin cleavage sites LERDQKLSE for
BoNT/F and BoNT/D and S(Q/V)F for TeTx were
revealed from the compilation of cleavage sites of
synaptobrevins/VAMPs and cellubrevin (Yamasaki et
al., 1994). Interestingly, endobrevin does not contain
these conserved toxin cleavage sites that are present in
the other members of the synaptobrevin/VAMP fam-
ily. Galli et al. (1994) has attempted previously to
determine the function of cellubrevin in the endocy-
totic pathway in Chinese hamster ovary cells by cleav-
A Novel Protein of the Early Endosome
Vol. 9, June 19981557
ing the endogenous cellubrevin by TeTx toxin. How-
ever, the impairment of the recycling of internalized
transferrin receptor back to the cell surface was only
?50% after treatment with TeTx. The partial effect of
TeTx may be due to the existence of other synaptobre-
vin/VAMP-like proteins (with similar functions) that
are insensitive to TeTx cleavage. Endobrevin could be
one of these TeTx-insensitive proteins.
Cell surface transferrin receptor binds to iron-satu-
rated transferrin and then internalizes via coated ves-
transferrin receptor) were either untreated (A and B) or treated with 10 ?g/ml BFA for 60 min (C and D) and 120 min (E and F) before
processing for indirect immunofluorescence microscopy. Cells were double labeled for endobrevin (A, C, and E) and transferrin receptor (B,
D, and F). Bar, 10 ?m.
Endobrevin is associated with the endocytotic compartment. A431 cells (after internalized monoclonal antibodies against
S.H. Wong et al.
Molecular Biology of the Cell 1558
5 min from the cell surface were processed for EM immunogold labeling to detect endobrevin (10 nm gold). Endobrevin (small arrowheads)
is enriched in the early endosome (E) marked by the internalized BSA-gold (arrows). Much lesser amounts of endobrevin can also be detected
on the plasma membrane (P) (large arrowheads in A). Bar, 100 nm.
Endobrevin is enriched in the early endosome. (A and B) Cryosections of J774 cells after internalizing BSA-gold (5 nm gold) for
A Novel Protein of the Early Endosome
Vol. 9, June 19981559
against transferrin receptor on ice, shifted to 37°C, and incubated for different periods of time at 37°C before processing for indirect immunoflu-
orescence microscopy. Cells were double labeled for endobrevin (A, C, E, and G) and transferrin receptor (B, D, F, and H). Bar, 10 ?m.
Endobrevin is localized mainly to the later compartment of the early endosomes. A431 cells were incubated with monoclonal antibodies
S.H. Wong et al.
Molecular Biology of the Cell 1560
icles before delivery to the early endosome. In this
early endosomal compartment, iron is released, and
the receptor with the iron-free transferrin is primarily
returned to the cell surface via recycling from the
later-recycling early endosomal subcompartment. It
has been reported previously that the early endo-
somes fuse with the TGN after treatment of cells with
BFA (Lippincott-Schwartz et al., 1991; Wood et al.,
1991; Whitney et al., 1995). In addition, during the
course of the BFA treatment, a tubular network is
formed that becomes concentrated in the MTOC. In
this study, antibodies bound to the transferrin recep-
tor on the surface of A431 cells were internalized for
30 min to label the early endosomal compartment
including the recycling early endosomes. Under this
condition, endobrevin colocalizes well with the inter-
nalized transferrin receptor especially at the perinu-
clear region. When cells were treated with BFA, both
the transferrin receptor and endobrevin were detected
on the tubular network (after BFA treatment for 60
min) that then collapsed around the MTOC (after BFA
treatment for 2 h). Therefore, these data, in conjunc-
tion with the EM immunogold labeling in J774 cells
strongly suggest that endobrevin is primarily local-
ized to the early endosome. Kinetics of the transferrin
receptor internalization reveals that endobrevin is as-
sociated poorly with early peripheral early endo-
somes. The majority of the endobrevin is associated
with the later perinuclear early endosome marked by
its colocalization with the cell surface-internalized
transferrin receptor (30 and 60 min internalizations).
Colocalization of endobrevin with the internalized
transferrin receptor to the vacuole-like structures after
treatment with wortmannin, a phosphatidylinositol
3-kinase inhibitor, further supports the conclusion that
endobrevin is associated with the later subcompart-
ment of early endosomes. It has been reported previ-
ously that wortmannin inhibits several membrane
traffic pathways. The membrane-trafficking pathways
that are inhibited by wortmannin include the recy-
transferrin receptor antibodies for 30 min at 37°C, A431 cells were treated with 500 nM wortmannin for 30 min (A and B) and 60 min (C and
D) before processing for indirect immunofluorescence microscopy. Bar, 10 ?m.
The distribution of endobrevin is affected by wortmannin, a phosphatidylinositol 3-kinase inhibitor. After internalization of
A Novel Protein of the Early Endosome
Vol. 9, June 19981561
cling from the mannose-6-phosphate receptor-positive
late endosomes to the TGN (Reaves et al., 1996), recy-
cling to the lysosome from the Igp120-positive late
endosomal compartment (Reaves et al., 1996), trans-
port of activated platelet-derived growth factor recep-
tor to the degradative compartment of the endocytic
pathway (Joly et al., 1994), fluid phase endocytosis and
early endosome fusion (Clague et al., 1995; Jones and
Clague, 1995; Li et al., 1995), insulin-stimulated glu-
cose transporter GLUT4 translocation (Kanai et al.,
1993), transcytosis (Hansen et al., 1995), and the re-
cruitment of transferrin receptor to the plasma mem-
brane in 3T3-L1 adipocytes after stimulation by insu-
lin (Shepherd et al., 1995). In the case of transferrin
receptor, it has been reported that its steady-state dis-
tribution was altered to swollen vacuole-like struc-
tures that are partially colocalized with the mannose-
6-phosphate receptor in the presence of wortmannin
(Reaves et al., 1996). Interestingly, it has also been
suggested that wortmannin’s point of inhibition on
the distribution of transferrin receptor is at the recy-
cling compartment but not the peripheral early endo-
somes (Mayor et al., 1993; Reaves et al., 1996). Because
wortmannin causes complete distribution of endobre-
vin to the swollen vacuole-like compartment, we sug-
gest that endobrevin is most likely to reside in the
later-recycling compartment of the early endosome.
The existence of a similar EST clone (accession no.
AA049140) encoding endobrevin was also noticed by
Scheller and colleagues (Bock and Scheller, 1997).
They have named this protein VAMP-8, entirely based
on the partial amino acid sequence and its homology
to synaptobrevins/VAMPs and cellubrevin. Our
present studies provide more detailed molecular, bio-
chemical, and cell biological characterization of this
protein. We feel that the name endobrevin is more
descriptive and would like to propose that this protein
be termed endobrevin.
We thank Anje Habermann for technical assistance, members of
Hong Wanjin’s laboratory for critical reading of the manuscript, and
Dr. Y.H. Tan for his support. This work was funded by the Institute
of Molecular and Cell Biology (to W.H.).
Bennett, M.K. (1995). SNAREs and the specificity of transport ves-
icle targeting. Curr. Opin. Cell Biol. 7, 582–586.
Binz, T., Blasi, J., Yamasaki, S., Baumeister, A., Link, E., Sudhof, T.C.,
Jahn, R., and Niemann, H. (1994). Proteolysis of SNAP-25 by types
E and A botulinal neurotoxins. J. Biol. Chem. 269, 1617–1620.
Blasi, J., Chapman, E.R., Link, E., Binz, T., Yamasaki, S., De Camilli,
P., Sudhof, T.C., Niemann, H., and Jahn, R. (1993a). Botulinum
neurotoxin A selectively cleaves the synaptic protein SNAP-25.
Nature 365, 160–163.
Blasi, J., Chapman, E.R., Yamasaki, S., Binz, T., Niemann, H., and Jahn,
R. (1993b). Botulinum neurotoxin C1 blocks neurotransmitter release
by means of cleaving HPC-1/syntaxin. EMBO J. 12, 4821–4828.
Bock, J.B., and Scheller, R.H. (1997). Protein transport. A fusion of
new ideas. Nature 387, 133–135.
Clague, M., Thorpe, C., and Jones, A.T. (1995). Phosphatidylinositol
3-kinase regulation of fluid phase endocytosis. FEBS Lett. 367, 272–274.
Galli, T., Chilcote, T., Mundigl, O., Binz, T., Niemann, H., and
Camilli, P.D. (1994). Tetanus toxin-mediated cleavage of cellubrevin
impairs exocytosis of transferrin receptor-containing vesicles in
CHO cells. J. Cell Biol. 125, 1015–1024.
Graeve, L., Patzak, A., Drickamer, K., and Rodriguez-Boulan, E.
(1990). Polarized expression of functional rat liver asialoglycopro-
tein receptor in transfected Madin-Darby canine kidney cells. J. Biol.
Chem. 265, 1216–1224.
Griffiths, G. (1993). Cryo and replica techniques for immunolabel-
ling. In: Fine Structure Immunocytochemistry, Berlin: Springer Ver-
Griffiths, G., Ericsson, M., Krijnse-Locker, J., Nilsson, M., Goud, B.,
Soling, H., Tang, B.L., Wong, S.H., and Hong, W. (1994). Localiza-
tion of the Lys, Asp, Glu, Leu tetrapeptide receptor to the Golgi
complex and the intermediate compartment in mammalian cells.
J. Cell Biol. 127, 1557–1547.
Gruenberg, J., and Maxfield, F.R. (1995). Membrane transport in the
endocytic pathway. Curr. Opin. Cell Biol. 7, 552–563.
Guan, K., and Dixon, J.E. (1991). Eukaryotic proteins expressed in
Escherichia coli: an improved thrombin cleavage and purification
procedure of fusion proteins with glutathione S-transferase. Anal.
Biochem. 192, 262–267.
Hansen, S.H., Olsson, A., and Casanova, J.E. (1995). Wortmannin, an
inhibitor of phosphoinositide 3-kinase, inhibits transcytosis in po-
larized epithelial cells. J. Biol. Chem. 270, 28425–28432.
Hong, W., and Tang, B.L. (1993). Protein trafficking along the exo-
cytotic pathway. Bioessays 15, 231–238.
Jahn, R., and Sudhof, T.C. (1994). Synaptic vesicles and exocytosis.
Annu. Rev. Neurosci. 17, 219–246.
Joly, M., Kazlauskas, A., Fay, F.S., and Corvera, S. (1994). Disruption
of PDFG receptor trafficking by mutation of its PI-3 kinase binding
sites. Science 263, 684–687.
Jones, A.T., and Clague, M.J. (1995). Phosphatidylinositol 3-kinase
activity is required for early endosome fusion. Biochem. J. 311,
Kanai, F., Ito, K., Todaka, M., Hayashi, H., Kamohara, S., Ishi, K.,
Okada, T., Hazeki, O., Ui, M., and Ebina, Y. (1993). Insulin-stimu-
lated GLUT4 translocation is relevant to the phosphorylation of
IRS-1 and the activity of PI-3-kinase. Biochem. Biophys. Res. Com-
mun. 195, 762–768.
Klausner, R.D., Donaldson, J.G., and Lippincott-Schwartz, J. (1992).
Brefeldin A: insights into the control of membrane traffic and or-
ganelle structure. J. Cell Biol. 116, 1071–1080.
Kozak, M. (1984). Compilation and analysis of sequences upstream
from the translational start site in eukaryotic mRNAs. Nucleic Acids
Res. 12, 857–872.
Li, G., D’Souza-Schorey, C., Barbieri, M.A., Roberts, R.L., Klippel,
sitol 3-kinase as a regulator of endocytosis via activation of Rab5.
Proc. Natl. Acad. Sci. USA 92, 10207–10211.
Link, E., Edelmann, L., Chou, J.H., Binz, T., Yamasaki, S., Eisel, U.,
Baumert, M., Sudhof, T.C., Niemann, H., and Jahn, R. (1992). Teta-
nus toxin action: inhibition of neurotransmitter release linked to
S.H. Wong et al.
Molecular Biology of the Cell 1562
synaptobrevin proteolysis. Biochem. Biophys. Res. Commun. 189,
Lippincott-Schwartz, J., Yuan, L.C., Tipper, C., Amherdt, M., Orci,
L., and Klausner, R.D. (1991). Brefeldin A’s effects on endosomes,
lysosomes, and the TGN suggest a general mechanism for regulat-
ing organelle structure and membrane traffic. Cell 67, 601–617.
Lodish, H.F. (1991). Recognition of complex oligosaccharides by the
multisubunit asialoglycoprotein receptor. Trends Biochem. Sci. 16,
Lowe, S.L., Wong, S.H., and Hong, W. (1996). The mammalian
ARF-like protein 1 (Arl1) is associated with the Golgi complex.
J. Cell Sci. 109, 209–220.
Mayor, S., Presley, J.F., and Maxfield, F.R. (1993). Sorting of membrane
components from endosomes and subsequent recycling to the cell
surface occurs by a bulk flow process. J. Cell Biol. 121, 1257–1269.
McMahon, H.T., and Sudhof, T.C. (1995). Synaptic core complex of
synaptobrevin, syntaxin, and SNAP25 forms high affinity ?-SNAP
binding site. J. Biol. Chem. 270, 2213–2217.
McMahon, H.T., Ushkaryov, Y.A., Edelman, L., Link, E., Binz, T.,
Niemann, H., Jahn, R., and Sudhof, T.C. (1993). Cellubrevin is a
ubiquitous tetanus-toxin substrate homologous to a putative syn-
aptic vesicle fusion protein. Nature 364, 346–349.
McNew, J.A., Sogaard, M., Lampen, N.M., Machida, S., Ye, R.R.,
Lacomis, L., Tempst, P., Rothman, J.E., and Sollner, T.H. (1997).
Ykt6p, a prenylated SNARE essential for endoplasma reticulum-
Golgi transport. J. Biol. Chem. 272, 17776–17783.
Mellman, I. (1996). Endocytosis and molecular sorting. Annu. Rev.
Cell Dev. Biol. 12, 575–625.
Mellman, I., and Simons, K. (1992). The Golgi complex: in vitro
veritas. Cell 68, 829–840.
Palade, G. (1975). Intracellular aspects of the process of protein
synthesis. Science 189, 347–358.
Pfeffer, S.R. (1996). Transport vesicle docking: SNAREs and associ-
ates. Annu. Rev. Cell Biol. 12, 441–461.
Reaves, B.J., Bright, N.A., Mullock, B.M., and Luzio, J.P. (1996). The
effect of wortmannin on the localization of lysosomal type I integral
membrane glycoproteins suggests a role for phosphoinositide 3-ki-
nase activity in regulating membrane traffic late in the endocytic
pathway. J. Cell Sci. 109, 749–762.
Robinson, M.S., Watts, C., and Zerial, M. (1996). Membrane dynam-
ics in endocytosis. Cell 84, 13–21.
Rothman, J.E., and Warren, G. (1994). Implications of the SNARE
hypothesis for intracellular membrane topology and dynamics.
Curr. Biol. 4, 220–233.
Rothman, J.E., and Wieland, F.T. (1996). Protein sorting by transport
vesicles. Science 272, 227–234.
Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular Clon-
ing: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring
Harbor Laboratory Press.
Schekman, R., and Orci, L. (1996). Coat proteins and vesicle bud-
ding. Science 271, 1526–1532.
Scheller, R.H. (1995). Membrane trafficking in the presynaptic nerve
terminal. Neuron 14, 893–897.
Schiavo, G., Benfenati, F., Poulain, B., Rossetto, O., Polverino de
Laureto, P., DasGupta, B.R., and Montecucco, C. (1992). Tetanus and
botulinum-B neurotoxins block neurotransmitter release by proteo-
lytic cleavage of synaptobrevin. Nature 359, 832–835.
Schiavo, G., Santucci, A., DasGupta, B.R., Mehta, B.B., Jontes, J.,
Benfenati, F., Wilson, M.C., and Montecucco, C. (1993). Botulinum
neurotoxins serotypes A and E cleave SNAP-25 at distinct COOH-
terminal peptide bonds. FEBS Lett. 335, 99–103.
Shepherd, P.R., Soos, M.A., and Siddle, K. (1995). Inhibitors of
phosphoinositide 3-kinase block exocytosis but not endocytosis of
transferrin receptors in 3T3–L1 adipocytes. Biochem. Biophys. Res.
Commun. 211, 535–539.
Slot, J.W., Geuze, H.J., Gigengack, S., Lienhard, G.E., and James,
D.E. (1991). Immuno-localization of the insulin regulatable glucose
transporter in brown adipose tissue of the rat. J. Cell Biol. 113,
Sogaard, M., Tani, K., Ye, R.R., Geromanos, S., Tempst, P., Kirch-
hausen, T., Rothman, J.E., and Sollner, T. (1994). A rab protein is
required for the assembly of SNARE complexes in the docking of
transport vesicles. Cell 78, 937–948.
Sollner, T., Whiteheart, S.W., Brunner, M., Erdjument-Bromage, H.,
Geromanos, S., Tempst, P., and Rothman, J.E. (1993). Snap receptors
implicated in vesicle targeting and fusion. Nature 362, 318–324.
Spies, M. (1990). The asialoglycoprotein receptor: a model for endo-
cytic transport receptors. Biochemistry 29, 10018–10022.
Subramaniam, V.N., Loh, E., and Hong, W. (1997). N-Ethylmaleim-
ide-sensitive factor (NSF) and ?-soluble NSF attachment proteins
(SNAP) mediated dissociation of GS28-syntaxin 5 Golgi SNAP re-
ceptors (SNARE) complex. J. Biol. Chem. 272, 25441–25444.
Subramaniam, V.N., Peter, F., Philip, R., Wong, S.H., and Hong, W.
(1996). GS28, a 28-kilodalton Golgi SNARE that participates in ER-
Golgi transport. Science 272, 1161–1163.
Subramaniam, V.N., Yusoff, A.R.B.M., Wong S.H., Lim, G.B., Chew,
M., and Hong W. (1992). Biochemical fractionations and character-
ization of proteins from Golgi-enriched membranes. J. Biol. Chem.
Su ¨dhof, T.C. (1995). The synaptic vesicle cycle: a cascade of protein-
protein interactions. Nature 375, 645–653.
Whiteheart, S.W., and Kubalek, E.W. (1995). SNAPs and NSF: gen-
eral members of the fusion apparatus. Trends Cell Biol. 5, 64–69.
Whitney, J.A., Gomez, M., Sheff, D., Kreis, T.E., and Mellman, I.
(1995). Cytoplasmic coat proteins involved in endosomes function.
Cell 83, 703–713.
Wong, S.H., and Hong, W. (1993). The SXYQRL sequence in the
cytoplasmic domain of TGN38 plays a major role in trans-Golgi
network localization. J. Biol. Chem. 268, 22853–22862.
Wong, S.H., Low, S.H., and Hong, W. (1992). The 17-residue trans-
membrane domain of ?-galactoside ?2,6-sialyltransferase is suffi-
cient for Golgi retention. J. Cell Biol. 117, 245–258.
Wong, S.H., Xu, Y., Zhang, T., and Hong, W. (1998). Syntaxin 7, a
novel syntaxin member associated with the early endosomal com-
partment. J. Biol. Chem. 273, 375–380.
Wood, S.A., and Brown, W.J. (1992). The morphology but not the
function of endosomes and lysosomes is altered by brefeldin A. J.
Cell Biol. 119, 273–285.
Wood, S.A., Park, J.E., and Brown, W.J. (1991). Brefeldin A causes a
microtubule-mediated fusion of the trans Golgi network and early
endosomes. Cell 67, 591–600.
Xu, Y., Wong, S.H., Zhang, T., Subramaniam, V.N., and Hong, W.
(1997). GS15, a 15-kilodalton Golgi soluble N-ethylmaleimide-sen-
sitive factor attachment protein receptor (SNARE) homologous to
rbet1. J. Biol. Chem. 272, 20162–20166.
Yamasaki, S., Baumeister, A., Binz, T., Blasi, J., Link, E., Cornille, F.,
Roques, B., Fykse, E.M., Sudhof, T.C., Jahn, R., and Niemann, H.
(1994). Cleavage of members of the synaptobrevin/VAMP family
by types D and F botulinal neurotoxins and tetanus toxin. J. Biol.
Chem. 269, 12764–12772.
A Novel Protein of the Early Endosome
Vol. 9, June 19981563