Molecular Biology of the Cell
Vol. 18, 4483–4492, November 2007
Phosphatidylinositol 4,5-Bisphosphate Mediates the
Targeting of the Exocyst to the Plasma Membrane for
Exocytosis in Mammalian Cells□
Jianglan Liu,* Xiaofeng Zuo,* Peng Yue, and Wei Guo
Department of Biology, University of Pennsylvania, Philadelphia, PA 19104-6018
Submitted May 17, 2007; Revised August 14, 2007; Accepted August 17, 2007
Monitoring Editor: Tom U. Martin
The exocyst is an evolutionarily conserved octameric protein complex that tethers post-Golgi secretory vesicles at the
plasma membrane for exocytosis. To elucidate the mechanism of vesicle tethering, it is important to understand how the
exocyst physically associates with the plasma membrane (PM). In this study, we report that the mammalian exocyst
subunit Exo70 associates with the PM through its direct interaction with phosphatidylinositol 4,5-bisphosphate
(PI(4,5)P2). Furthermore, we have identified key conserved residues at the C-terminus of Exo70 that are crucial for the
interaction of Exo70 with PI(4,5)P2. Disrupting Exo70-PI(4,5)P2interaction abolished the membrane association of Exo70.
We have also found that wild-type Exo70 but not the PI(4,5)P2-binding–deficient Exo70 mutant is capable of recruiting
other exocyst components to the PM. Using the ts045 vesicular stomatitis virus glycoprotein trafficking assay, we
demonstrate that Exo70-PI(4,5)P2interaction is critical for the docking and fusion of post-Golgi secretory vesicles, but not
for their transport to the PM.
Exocytosis is important for a variety of cellular functions,
ranging from the release of hormones to the incorporation of
membrane proteins for cell growth and morphogenesis. The
late stage of exocytosis is a multistep process that includes
directional transport, tethering, docking, and fusion of post-
Golgi secretory vesicles with the plasma membrane (PM).
The tethering step, defined as the initial contact of secretory
vesicles with the PM before SNARE-mediated docking and
fusion (Pfeffer, 1999; Guo et al., 2000; Waters and Hughson,
2000; Whyte and Munro, 2002), is mediated by the exocyst,
an evolutionarily conserved octameric complex composed of
Sec3, Sec5, Sec6, Sec8, Sec10, Sec15, Exo70, and Exo84 (for
review, see Guo et al., 2000; Hsu et al., 2004; Munson and
Novick, 2006; Wang and Hsu, 2006). In budding yeast, the
exocyst components localize to the growing end of the
daughter cell (“bud”), where active exocytosis and mem-
brane addition take place (TerBush and Novick, 1995; Finger
et al., 1998, Guo et al., 1999). This localization pattern con-
trasts that of the membrane fusion machine, the t-SNAREs,
which are evenly distributed along both the mother and
daughter cell membrane (Brennwald et al., 1994). In mam-
malian cells, the exocyst components were found in the
cytosol, recycling endosomes and trans-Golgi network (Yea-
man et al., 2001; Fo ¨lsch et al., 2003; Ang et al., 2004; Langevin
et al., 2005). However, they are recruited to the PM during a
number of cellular processes. For example, in epithelial cells,
the exocyst is recruited to the adherens junction region upon
cell–cell contact, where it mediates protein and membrane
addition at the basolateral domain (Grindstaff et al., 1998;
Yeaman et al., 2001); in developing neurons, the exocyst is
localized to the growing neurites, where it mediates mem-
brane expansion (Hazuka et al., 1999; Vega and Hsu, 2001);
during cell migration, the exocyst is recruited to the leading
edges of the PM (Rosse et al., 2006; Zuo et al., 2006).
The mechanism by which the exocyst mediates vesicle
tethering to the PM is unclear. One key question yet to be
resolved is how the exocyst itself associates with the PM.
Using fluorescence recovery after photobleaching (FRAP)
analyses and immunoelectron microscopy, Boyd et al. (2004)
have shown that Exo70 is stably localized to the yeast bud
tip membrane and remains polarized even when the actin
cables are disrupted, suggesting that Exo70 is a candidate in
this complex involved in membrane targeting of the exocyst.
In Madin-Darby canine kidney (MDCK) cells, extragenically
expressed GFP-tagged Exo70 is localized to the PM near
cell–cell contacts, suggesting that Exo70 may mediate PM
association independent of the rest exocyst components in
these cells (Matern et al., 2001). Recent structural studies
have revealed that Exo70 contains a number of conserved
basic residues that cluster on a surface patch at the C-
terminal end of the tertiary structure that may directly bind
to the PM (Dong et al., 2005; Hamburger et al., 2006; Moore
et al., 2007). In fact, the C-terminal sequence of Exo70 is the
most evolutionarily conserved region of this protein.
Here, we report that mammalian Exo70 directly interacts
with PI(4,5)P2in the PM via the positively charged residues
at its C-terminus. We have also identified key residues in
Exo70 that are important for this interaction. Finally, using
the ts045 vesicular stomatitis virus glycoprotein (VSV-G)
trafficking assay, we found that the Exo70-lipid interaction is
critical for PM stages of exocytosis, but not for the trafficking
steps through endoplasmic reticulum (ER) and Golgi. Our
This article was published online ahead of print in MBC in Press
on August 29, 2007.
DThe online version of this article contains supplemental material
at MBC Online (http://www.molbiolcell.org).
* These authors contributed equally to this work.
Address correspondence to: Wei Guo (firstname.lastname@example.org).
© 2007 by The American Society for Cell Biology 4483
study revealed a molecular mechanism by which the exocyst
directly interacts with the PM that is critical for vesicle
tethering and exocytosis.
MATERIALS AND METHODS
DNA Plasmid Construction
Wild-type rat Exo70 (rExo70) cDNA and various truncates of Exo70 were
cloned in-frame into pEGFP-C1 for expression as green fluorescent protein
(GFP) fusions or in pEBG for expression as glutathione S-transferase (GST)
fusion in mammalian cells. rExo70 was also cloned into pGEX-KG (a modified
form of pGEX-2T, from Amersham Biosciences, Piscataway, NJ) for expres-
sion as GST fusion in bacteria. The Exo70 mutants were generated using the
QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA). All the
constructs were confirmed by nucleotide sequencing.
Cell Culture and RNA Interference Experiments
HeLa cells were cultured at 37°C in DMEM supplemented with 10% fetal
bovine serum and 100 U/ml penicillin and 100 ?g/ml streptomycin in a 5%
CO2incubator. For RNA interference (RNAi) experiments, cells were grown
to 50% confluence and transfected with small interfering RNA (siRNA) du-
plexes using Oligofectamine (Invitrogen, Carlsbad, CA). The human Exo70
siRNA target sequence is 5?-GGTTAAAGGTGACTGATTA-3?. The control
Luciferase GL2 siRNA target sequence is 5?-AACGTACGCGGAATACT-
TCGA-3?. The efficiency of Exo70 knockdown was determined by Western
Transfected HeLa cells were grown on coverslips, washed with phosphate-
buffered saline (PBS), fixed in 4% paraformaldehyde at room temperature for
12 min, washed, permeabilized for 5 min with PBST (PBS-Tween), and
blocked for 10 min with 2% bovine serum albumin in PBST. The coverslips
were incubated sequentially with primary and secondary antibodies for flu-
orescence observation using the Leica TCS SL laser-scanning confocal micro-
scope (63? objective; Deerfield, IL). Images were processed with Adobe
Photoshop (Adobe Systems, San Jose, CA; version 7.0).
Large Unilamellar Vesicle Sedimentation Assay
Large unilamellar vesicle (LUV) sedimentation assay was performed as pre-
viously described (Hokanson and Ostap, 2006). Phospholipids were pur-
chased from Avanti Polar Lipids (Alabaster, AL). LUVs with a 100-nm diam-
eter were prepared by size extrusion. Various lipids were mixed at different
molar ratios, dried with nitrogen stream, and resuspended at a concentration
of 2 mM in a buffer containing 12 mM HEPES, pH 7.0, and 176 mM sucrose.
The mixed lipids were subjected to five cycles of freeze-thaw and a 1-min bath
sonication before being passed through 100-nm filters using a mini-extruder.
LUVs were dialyzed overnight in the HNa100 buffer (10 mM HEPES, pH 7.0,
100 mM NaCl, 1 mM EGTA, and 1 mM dithiothreitol [DTT]). The percentages
of phosphotidylserine (PS), PI(3)P, PI(4)P, PI(3,5)P2, PI(4,5)P2, and PI(3,4,5)P3
indicated in the text are the molar percentages of total PS and PIPs with the
remainder being phosphatidylcholine (PC). Lipid concentrations are given as
total lipid. The binding of Exo70 to LUVs was determined by sedimentation
assays in 200 ?l total volume using TLA-100 rotor (Beckman Coulter, Fuller-
ton, CA). Sucrose-loaded LUVs were precipitated at 150,000 ? g for 30 min at
25°C. The supernatants and pellets were subjected to 10% SDS-PAGE and
stained with SYPRORed (Invitrogen) for quantification of free and bound
materials with the Image Quant software (Molecular Dynamics, Sunnyvale,
HeLa cells were plated in 10-cm dishes at 1.5 ? 106cells per dish. The next
day cells were transfected with DNA by FuGene6 reagent and incubated at
37°C overnight. Homogenization and subcellular fractionation of the cells to
isolate the PM fraction, cytosol, the low-density microsomal fraction (LDM),
and the high-density microsomal fraction (HDM) were performed basically as
previously described (Weber et al., 1988). All steps were performed at 4°C in
the presence of a protease inhibitor cocktail. The distribution of Exo70 and
Sec8 were detected by monoclonal antibodies (kind gifts of Dr. Shu-Chan
Hsu, Rutgers University).
GST Pulldown Assay
HeLa cells were transfected with GFP-tagged Exo70 or exo70-1, and the cells
were lysed in a buffer containing 20 mM Tris-HCl, pH 7.5, 25 mM KCl, 1 mM
MgCl2, 0.5 mM EGTA, 1 mM DTT, 0.5% Triton X-100, and protease inhibitors.
Cell lysates were incubated overnight with glutathione-Sepharose conjugated
with GST or GST-TC10 (Q75L) at 4°C. After incubation, the beads were
washed five times with the lysis buffer, and the bound proteins were analyzed
by Western blot using an anti-GFP antibody. To detect the interaction of
Exo70 and Sec8 in the cell, HeLa cells were transfected with GST-tagged
Exo70 or exo70-1, and cell lysates were incubated overnight with glutathione-
Sepharose beads at 4°C. After incubation, the beads were washed, and the
bound proteins were detected by Western blot using anti-GST or anti-Sec8
VSV-G Trafficking Assay
HeLa cells were transfected with EXO70 siRNA. luciferase siRNA was used as
the negative control. After 24 h of the siRNA treatment, HeLa cells were
transfected with VSV-G-45ts-GFP mutant and immediately placed at 40°C.
After overnight growth, the cells were shifted to 32°C for 0, 15, 30, 60, and 90
min in the presence of cycloheximide (100 ?g/ml). The cells were then fixed
for GFP observation or immunofluorescence. The 8G5 mAb against the ex-
tracellular domain of VSV-G was kindly provided by Dr. Douglas Lyles
(Wake Forest University). No detergent was used in the immunofluorescence
procedure. Cells with surface VSV-Gs were quantified, and statistical analyses
were performed using Student’s t test. In some cases, HeLa cells were trans-
fected with VSV-G-myc and GST-Exo70 or GST-exo70-1 after 24 h of the
EXO70 siRNA treatment. The cells were divided into two sets based on their
treatments. For Set I, the cells were fixed, permeabilized, and stained with
anti-myc mAb (9E10) and anti-GST polyclonal antibody to test the intracel-
lular traffic of VSV-G and to detect the expression of Exo70 or exo70-1 at 0-,
30-, 60-, and 90-min points. For Set II, cells of the 90-min point group were first
stained with the 8G5 antibody, then permeabilized, and stained with anti-GST
polyclonal antibody. Anti-mouse Alexa488 and anti-rabbit Alexa594 were
used as secondary antibodies for the above experiments. For the quantifica-
tion of surface VSV-G signals at various points, boundary of the cell surface
was outlined, and average fluorescence intensity of surface VSV-G signal was
quantified using ImageJ 1.73v software and then divided by the perimeter of
the cell surface. For the quantification of VSV-G in different membrane
compartments, boundaries of the whole cell, the Golgi, and the cell periphery
were outlined, and VSV-G fluorescence in these areas was then quantified
using ImageJ 1.73v software after subtraction of background outside the cell
using the following equations:
Golgi (%) ? FluorescenceGolgi/Fluorescencetotal
ER ? Cytoplasm (%) ? FluorescenceER ? cytoplasm/Fluorescencetotal
Cell periphery (%) ? (Fluorescencetotal? Fluorescencecytoplasm
Association of Exo70 with the Plasma Membrane in HeLa
We have examined the localization of GFP-tagged rat Exo70
in HeLa cells. As shown in Figure 1A, GFP-Exo70 was
clearly detected at the PM as revealed by serial optical
sectioning from different axis (Figure 1A). In addition, cells
expressing GFP-Exo70 displayed filopodia-like structures as
previously reported (Wang et al., 2004; Xu et al., 2005; Zuo et
al., 2006). We next mapped the region of Exo70 that is
required for its association with the PM by examining the
localization of a series of GFP-tagged Exo70 truncates in
HeLa cells. As shown in Figure 1B, Exo70 with its C-terminal
domain deleted (amino acids 1-408; Exo70-NT) was cytoso-
lic, whereas the C-terminus of Exo70 (amino acids 403-653;
Exo70-CT) was enriched at the PM (Figure 1B), suggesting a
critical role of Exo70 C-terminus in membrane association.
The crystal structure of yeast Exo70 revealed that this pro-
tein is a long rod composed mainly of ?-helices that fold into
four domains (named domains A, B, C, and D; Dong et al.,
2005; Hamburger et al., 2006). Domain D (the C-terminal 114
amino acids) is the most evolutionarily conserved domain in
Exo70 that contains a number of basic residues that cluster
into an electro-positive patch on the surface of the C-termi-
nus. Because in many cases the association of proteins with
the PM involves interactions with the negatively charged
phospholipids in membrane via clusters of basic residues
(for reviews, see McLaughlin et al., 2002; Balla, 2005), it is
likely that those positively charged residues at the C-termi-
nus of mammalian Exo70 are directly involved in membrane
association. We sought to examine whether domain D is
J. Liu et al.
Molecular Biology of the Cell 4484
required for the PM localization of Exo70. We generated
truncations of Exo70, in which the last 95 amino acids (aa
559-653) or the last 45 amino acids (aa 609-653) were deleted.
The resulting Exo70 fragments, Exo70 (aa 1-558) and Exo70
(aa 1-608), were both cytosolic (Figure 1B), indicating that
domain D of mammalian Exo70, especially the last 45 amino
acids, is required for the association of Exo70 with the PM.
We have also tested the localization of Exo70 (aa 403-558)
and Exo70 (aa 403-608), which are C-terminal fragments of
Exo70 with the last 95 amino acids (aa 559-653) or the last 45
amino acids (aa 609-653) deleted, and found that both of the
fragments were cytosolic (Figure 1B). We next investigated
whether the C-terminal fragments of Exo70 containing aa
403-653 or aa 540-653 (domain D) were able to associate with
the PM. As shown in Figure 1B, both fragments were local-
ized to the PM, suggesting that domain D of Exo70 is suffi-
cient for its PM localization. A portion of these fragments
was also detected in the nucleus, probably resulting from
nonspecific retention of the GFP fusion. Collectively, these
data indicate that the C-terminus of Exo70 is both necessary
and sufficient for the PM targeting of Exo70.
Exo70 Directly Interacts with Phospholipids through Its
Based on sequence analysis, domain D of Exo70 contains a
number of basic residues that are well conserved in yeast.
Moreover, based on the crystal structure, the basic residues
cluster onto a surface patch in the folded Exo70 protein.
Therefore, it is likely that Exo70 directly interact with phos-
pholipids through this basic patch. To test this possibility,
we examined the binding between recombinant GST-Exo70
and various phospholipids by cosedimentation assay using
the methods as previously described (Papayannopoulos et
al., 2005; Hokanson and Ostap, 2006). PI(4,5)P2is the major
phosphatidylinositol species, and constitutes 1–5% of the
total lipids in the PM (for review, see McLaughlin et al.,
2002). Binding experiments were performed with LUVs com-
plasma membrane (PM) in HeLa cells. (A)
GFP-Exo70 was detected at the PM. HeLa cells
were transfected with GFP-tagged Exo70,
fixed, and observed using a confocal micro-
scope. Serial optical sections in the x-z and y-z
planes were taken. (B) Various truncates of
Exo70 were tested for their localization at the
PM. HeLa cells were transfected with GFP-
tagged Exo70 and Exo70 truncates as dia-
gramed. The cells were fixed, stained, and ob-
served under microscope. Full-length and the
C-terminal fragments of Exo70 containing do-
main D (aa 558-653) are associated with the
PM. ?, membrane association.
Association of Exo70 with the
PI(4,5)P2Mediates Exocyst Targeting
Vol. 18, November 20074485
posed of the neutral PC and PI(4,5)P2at 5% or the acidic
phospholipid PS in molar percentages of 20, 40, and 60%.
Because the recombinant Exo70 protein could only access the
outer leaflet of the reconstituted LUV vesicles, the effective
PI(4,5)P2and PS in the reaction were only half of the total. As
shown in Figure 2A, Exo70 bound to LUVs containing 5%
PI(4,5)P2, but not to LUVs containing 100% neutral PC. Exo70
also bound to PS; however, the binding was weak unless the
molar ratio of PS in the LUVs was raised to 60%. As a control,
GST did not bind to LUVs with any lipid composition.
We have also measured the affinity of Exo70 for various
lipids (Figure 2B). The bound Exo70 was quantified and
plotted against the lipid concentration with the equation:
B ? BmaxX/[Kd? X], where Kdis the dissociation constant
and X and B represent the concentrations of the free Exo70
and the bound Exo70, respectively. The Kdwas obtained by
nonlinear regression. As shown in Table 1, Exo70 bound to
LUVs composed of 5% PI(4,5)P2with a Kdof 13.9 ? 3.0 ?M
and bound to LUVs composed of 60% PS with a Kdof 46.3 ?
9.7 ?M. The Kdvalue for PI(4,5)P2would be much smaller if
expressed in terms of pure PI(4,5)P2, because the molar ratio
of PI(4,5)P2is only 5% of the total lipids in the reconstituted
LUVs. Overall, these results suggest that Exo70 may associ-
ate with the PM through its direct interaction with PI(4,5)P2.
To determine the selectivity of Exo70-PI(4,5)P2interaction,
we have tested the binding of Exo70 to four additional
phosphoinositides: PI(3)P, PI(4)P, PI(3,5)P2, and PI(3,4,5)P3.
In mammalian cells, PI(4)P and PI(4,5)P2are the two most
abundant phosphoinositides concentrated on the Golgi and
PM, respectively (Di Paolo and De Camilli, 2006). PI(4,5)P2
comprises more than 95% of the bis-phosphoinositides in
mammals (Bonangelino et al., 2002). PI(3)P and PI(3,5)P2are
mainly localized to the endosomal compartments (for re-
views, see Behnia and Munro, 2005; Di Paolo and De Cam-
illi, 2006). PI(3,4,5)P3is present in negligible amount in
mammalian cells at the “resting” state, but is up-regulated at
specific domains of the PM in response to certain stimuli
(Insall and Weiner, 2001). Recently, it was found that
PI(3,4,5)P3is also localized to the basolateral domain of
MDCK cells (Gassama-Diagne et al., 2006). As shown in
Figure 2C, Exo70 barely binds to PI(3)P; the affinity of Exo70
for PI(4,5)P2(Kd? 13.92 ? 3.0 ?M) is ?4–5-fold higher than
PI(3,5)P2(Kd? 58.34 ? 9.2 ?M) and more than 10-fold
higher than PI(4)P (Kd? 173.12 ? 15.5 ?M). The affinity of
Exo70 for PI(3,4,5)P3(Kd? 15.55 ? 4.6 ?M) is comparable to
that of PI(4,5)P2.
Mutations in Domain D Disrupt the Plasma Membrane
Association of Exo70
The direct binding of Exo70 with PI(4,5)P2suggests a critical
role for the domain D basic residues in the PM association of
Exo70. To identify the responsible residues that are essential
for Exo70 to bind to the PM, the conserved basic residues in
domain D of Exo70 (Figure 3A) were targeted for mutagen-
esis and the resulting mutants were tested for their cellular
localization (summarized in Supplementary Table 1).
Among the 10 mutants we tested, 6 failed to associate with
the PM. The other four mutations had no effect on Exo70
localization, suggesting that not all of the basic residues are
involved in PI(4,5)P2binding. We focused on one of the
mutants, named exo70-1, in which residues K632 and K635
have been mutated to alanine. As shown in Figure 3B, GFP-
tagged wild-type Exo70 (GFP-Exo70) was enriched at the
PM, whereas GFP-tagged exo70-1 was distributed diffusely
throughout the intracellular regions. These results indicate
that K632 and K635 are required for the PM localization of
To directly test whether K632 and K635 are essential for
the physical association of Exo70 with the PM, we per-
formed subcellular membrane fractionation assay using
(A) GST-Exo70 purified from bacteria (0.15 ?M) was incubated with
and 60% PS. After centrifugation at 150,000 ? g for 30 min, the proteins
in supernatant (S) and pellet (P) were subjected to SDS-PAGE and
visualized by SYPRORed staining. Exo70 bound to vesicles containing
5% PI(4,5)P2or 60% PS. It also bound weakly to vesicles containing
40% PS. GST did not bind to liposomes with any lipid composition in
the cosedimentation assay. (B and C) The interaction of Exo70 with
phospholipids. Exo70, 0.15 ?M, was incubated with increasing concen-
trations of LUVs composed of (B) 100% PC, 5% PI(4,5)P2, 20% PS, 40%
PS, and 60% PS or (C) PI(3)P, PI(4)P, PI(3,5)P2, PI(4,5)P2, and
PI(3,4,5)P3. The percentage of bound Exo70 was plotted with the in-
creasing liposome concentration with a single rectangular hyperbola
equation (B ? BmaxX/[Kd? X]) using SigmaPlot. Each point is the
average of three measurements. Error bars, SD.
The interaction between Exo70 and phospholipids in vitro.
Table 1. Comparison of the affinities of Exo70 and exo70-1 for
Kd(?M) for Exo70
Kd(?M) for exo70-1
46.30 ? 9.7
13.92 ? 3.0
146.23 ? 14.6
The Kdwas obtained by rectangular hyperbola equation using the
SigmaPlot software. For some of the bindings, the Kdwas ? ?400
because the binding never reached saturation.
J. Liu et al.
Molecular Biology of the Cell4486
HeLa cells transfected with GFP-tagged wild-type Exo70 or
exo70-1. Cell lysates were fractionated, and the presence of
Exo70 or exo70-1 in the PM, cytosol, and other fractions was
examined. As shown in Figure 3C, wild-type Exo70 was
found in the PM fraction, whereas exo70-1 was mostly
present in the cytosol. The amount of Exo70 and exo70-1 in
the HDM (mostly ER membrane) and LDM (mostly Golgi
and endosomal compartments) fractions was almost the
same. During our studies, we have noticed that GFP-tagging
of Exo70 may facilitate its translocation from cytosol to the
PM. It is likely that GFP-tagging causes conformational
changes on Exo70, exposing the lipid-binding site on Exo70;
and mutations on the C-terminus of Exo70 may abolish its
membrane-association (see below). That may explain the
observed clear PM versus cytosol distribution of GFP-Exo70
versus GFP-exo70-1 in Figure 3. The membrane fractionation
data, together with the fluorescence localization results, sug-
gest that K632 and K635 are critical residues in Exo70 that
are involved in the physical association of Exo70 with the
Next, we examined the interaction of the mutant exo70-1
protein with phospholipids in the cosedimentation assay. As
shown in Figure 4 and Table 1, the interaction between
exo70-1 and 5% PI(4,5)P2was almost abolished, indicating
that K632 and K635 are critical residues for the binding of
Exo70 to PI(4,5)P2. Interestingly, the binding of exo70-1 for
PS was only partially affected, suggesting that these two
residues confer certain degree of specificity for PI(4,5)P2.
It was previously shown that Exo70 interacts with the
constitutively activated form of TC10 (Q67L; Inoue et al.,
2003). We then tested whether mutations in exo70-1 affect
this interaction. Cell lysates expressing GFP-tagged Exo70 or
exo70-1 were incubated with GST-TC10 (Q67L) conjugated
to the glutathione-Sepharose for binding reaction. Exo70 or
the exo70 mutant bound to the Sepharose was detected by
immunoblotting using an anti-GFP antibody. The amount of
exo70-1 bound to the TC10 beads was comparable to that of
the wild-type Exo70 (Supplementary Figure 1A). As a con-
trol, neither the wild-type Exo70 nor exo70-1 bound to the
GST beads. The result indicates that mutating the two key
basic residues in exo70-1 does not affect its interaction with
TC10. Therefore, the loss of PM association of the exo70
mutant is unlikely resulted from impaired interaction with
TC10. The binding results strongly suggest that Exo70 asso-
ciates with the PM via its direct interaction with PI(4,5)P2. In
addition to TC10, we have also tested the binding of exo70-1
with another exocyst component Sec8. We found that mu-
tations on exo70-1 did not affect its interaction with Sec8 in
the cell (Supplementary Figure 1B). This result is consistent
with the previous reports in mammalian and yeast cells that
the interaction of Exo70 with the other exocyst components
for complex assembly is mediated by the N-terminal do-
mains of Exo70 (Inoue et al., 2003; Dong et al., 2005).
Exo70 Recruits Sec8 to the Plasma Membrane
We next asked whether Exo70 is involved in recruiting other
members of the exocyst to the membrane. It has been shown
that the exocyst components were mostly located in the
cytoplasm in cultured HeLa cells (Zuo et al., 2006). We then
examined whether an increase of Exo70 at the PM would
recruit Sec8 to the membrane. GFP-Exo70 and GFP-exo70-1
were expressed in HeLa cells, and Sec8 in these cells was
detected by immunofluorescence staining using the anti-
Sec8 mAb 2E12. As shown in Figure 5, both GFP-Exo70 and
Sec8 were detected at the PM. On the contrary, in cells
expressing GFP-exo70-1 that is deficient in binding PI(4,5)P2,
Sec8 was found in the cytoplasm and intracellular mem-
brane structures. This result suggests that the wild type, but
between Exo70 and the PM. (A) Sequence alignment between the
C-termini (domain D) of yeast Exo70 and rat Exo70. The mutated
residues in the exo70-1 mutant were marked in grey. (B) The local-
ization of the exo70 mutant, exo70-1, in HeLa cells. HeLa cells were
transfected with GFP-tagged wild-type Exo70 as a control (left) and
exo70-1 (right). The exo70-1 mutant failed to associate with the PM.
(C) Membrane fractionation was performed to examine the local-
ization of GFP-Exo70 and GFP-exo70-1 in HeLa cells. Equal amounts
of proteins from each fraction of the cells expressing Exo70 and
exo70-1 were loaded on SDS-PAGE. The total proteins in each lane
were detected by SYPRORed staining. Exo70 was detected by West-
ern blot. LDM, low-density microsomal membranes, mostly the
Golgi fraction; HDM, high-density microsomal membranes, mostly
the ER fraction.
Mutations at Exo70 C-terminus abolish the association
tion with phospholipids. (A) GST-tagged Exo70 and exo70-1 (0.15
?M) were used in the lipid cosedimentation assay. The binding of
exo70-1 to 5% PI(4,5)P2was abolished, whereas its interaction with
60% PS was reduced. (B) Binding curves of Exo70 and exo70-1 to
LUVs containing various phospholipids.
Mutations at the C-terminus of Exo70 affect its interac-
PI(4,5)P2Mediates Exocyst Targeting
Vol. 18, November 20074487
not the mutant Exo70, is able to recruit the other exocyst
components to the PM. Because exo70-1 maintains its ability
to interact with Sec8, the loss of PM association of Sec8 in
cells expressing GFP-exo70-1 is unlikely resulted from a
defect in exocyst complex assembly.
Exo70 Is Essential for Exocytosis of VSV-G ts045 at the
Because the exocyst functions to tether post-Golgi secretory
vesicles at the PM, the physical interaction between Exo70
and phospholipids could be important for exocytosis. Here
we examined the effect of Exo70 knockdown on the traffick-
ing of a temperature-sensitive VSV-G mutant (ts045) to the
PM. At the restrictive temperature (40°C), the VSV-G ts045
mutant is misfolded and retained in the ER. Shifting cells to
32°C allows correct folding and trafficking of the protein
through the Golgi apparatus to the PM. We treated HeLa
cells with EXO70 siRNA specific for human Exo70 (hExo70),
and the luciferase siRNA was used as a control. The treat-
ment reduced the level of Exo70 by more than 90% without
affecting the level of other exocyst components such as Sec8
(Figure 6A). Cells treated with EXO70 siRNA or luciferase
siRNA were transfected with GFP-tagged VSV-G mutant
ts045 (VSV-G-GFP). Cells were then kept at 40°C overnight
and shifted to 32°C for various times before being examined
by fluorescence microscopy. The translocation of VSV-G-
GFP inside the cells was detected by GFP fluorescence,
whereas the incorporation of VSV-G-GFP to the PM (at the
90-min point) was evaluated by immunostaining using the
8G5 mAb against the extracellular domain of VSV-G (Le-
francois and Lyles, 1982). As shown in Figure 6B, in luciferase
siRNA-treated cells, VSV-G-GFP was retained in the ER
before the cells were shifted from 40 to 32°C (0 min). After
the shift, the protein was translocated from ER to the Golgi
apparatus around 30 min and then to the cell surface at 90
min. The exocytosis of VSV-G-GFP at the PM at 90 min was
clearly detected by immunostaining of nonpermeabilized
cells with the 8G5 antibody (Figure 6B). In EXO70 siRNA-
treated cells, the translocation of VSV-G-GFP from the ER to
the Golgi apparatus and from the Golgi to the PM followed
time courses similar to that in control siRNA-treated cells.
However, at 90 min, the extracellular exposure of VSV-G-
GFP was barely detectable in nonpermeabilized cells; in-
stead, it could only be detected after 180 min (Figure 6C),
suggesting that the fusion of the exocytic vesicles with the
PM was severely delayed. Quantification of the GFP signal
in different intracellular compartments indicates that the
cellular distributions of VSV-G-GFP at various time points
after the temperature shift were similar in EXO70 siRNA
and luciferase siRNA-treated cells (Figure 6, D and E): at 0
min, a majority of VSV-G-GFP was located at the ER (88%
for luciferase siRNA treatment and 84% for EXO70 siRNA
treatment); at 30 min, most of the VSV-G-GFP was located at
the Golgi apparatus (91% for luciferase siRNA and 92% for
EXO70 siRNA treatment); and at 90 min, VSV-G-GFP was
located at the cell periphery (88% for luciferase siRNA and
74% for EXO70 siRNA treatment). In contrast, quantification
of the surface VSV-G signal revealed a different pattern in
EXO70 siRNA and luciferase siRNA-treated cells. As shown
in Figure 6F, at 0-, 15-, and 30-min points, fluorescence
intensity of exposed VSV-G was negligible both in control
and EXO70 siRNA-treated cells, which suggests that VSV-
G-GFP was translocating in the intracellular compartments;
at 60 and 90 min, fluorescence intensity of exposed VSV-G
was much less in EXO70 siRNA-treated cells than in control
siRNA-treated cells; at 180 min, fluorescence intensity of
surface VSV-G in EXO70 siRNA-treated cells began to ap-
pear at the cell surface. These results suggest that the fusion
of the exocytic VSV-G vesicles with the PM was delayed in
cells treated with EXO70 siRNA. These results demonstrated
that Exo70 is probably not required for the trafficking of
VSV-G from the intracellular compartments to the cell pe-
riphery, but is essential for the efficient tethering or subse-
quent docking/fusion of VSV-G–containing exocytic vesi-
cles with the PM.
The Interaction of Exo70 with PI(4,5)P2Is Important for
the Exocytosis of VSV-G ts045
To investigate the role of the Exo70-PI(4,5)P2binding in
exocytosis, it is necessary to specifically disrupt the interac-
tion between Exo70 and phospholipids in vivo. We therefore
examined the transport of VSV-G in EXO70 siRNA knock-
down cells transfected with the rat exo70 mutant, exo70-1,
that is defective in the PI(4,5)P2binding. EXO70 siRNA-
treated cells expressing wild-type rat Exo70 (rExo70 WT)
were used as control. Rat Exo70 is over 90% identical in
protein sequence to human Exo70, yet on the nucleotide
level, it cannot be targeted by the EXO70 siRNA oligos
used in this study. The level of Exo70 knockdown and the
expression of GST-rExo70 and GST-rexo70-1 are shown in
In EXO70 siRNA-treated cells, only a diminutive amount
of VSV-G-GFP was detected on the cell surface after 90 min
of growth at the permissive temperature (Figure 7B). Expres-
sion of wild-type rat Exo70 restored normal surface incor-
poration of VSV-G (VSV-G-myc) in EXO70 siRNA knock-
down cells (Figure 7C); therefore the rat Exo70 can serve as
a rescue reagent for the RNAi experiment in HeLa cells. In
contrast, in EXO70 siRNA-treated cells expressing the
exo70-1 mutant that is defective in PI(4,5)P2binding, the
surface exposure of VSV-G-myc was diminished, whereas
the translocation of VSV-G-myc from the intracellular com-
partments to the cell periphery was not affected (Figure 7D).
VSV-G-myc rather than VSV-G-GFP was used here so that
different pairs of proteins in the cells could be stained (GST-
rExo70 and myc vs. GST-rExo70 and 8G5). Quantification of
Exo70 and GFP-exo70-1 were transfected into HeLa cells. Sec8 in the
transfected cells was detected by the anti-Sec8 monoclonal antibody
2E12. Sec8 was detected at the PM in cells expressing GFP-Exo70
(top panel), but not in cells expressing GFP-exo70-1 (bottom panel).
Exo70 is required for the PM localization of Sec8. GFP-
J. Liu et al.
Molecular Biology of the Cell 4488
cells with surface VSV-G indicates that expressing wild-type
rat Exo70 in EXO70 siRNA-treated cells restores VSV-G
surface exposure in the majority of the cells (from 18 to 61%),
whereas expressing rat exo70-1 does not have an obvious
effect (Figure 7E). These results suggest that the interaction
of Exo70 with membrane lipids is essential for the exocytosis
of VSV-G at the PM.
To understand the molecular basis of vesicle tethering at the
PM, it is important to elucidate how the tethering complex
itself, the exocyst, is targeted to the PM. Here we identified
a direct interaction between the exocyst component Exo70
and PI(4,5)P2; and demonstrated that this interaction is im-
portant for the recruitment of the exocyst to the PM. Fur-
thermore, disruption of this interaction blocked later stages
of exocytosis of post-Golgi secretory vesicles at the PM.
PI(4,5)P2and PS are the major negatively charged lipids in
the PM. The fact that Exo70 binds to both 5% PI(4,5)P2and
60% PS indicates that the interaction of Exo70 with the
phospholipids is electrostatic in nature. This type of inter-
action has been found in a number of proteins, such as
N-WASP and MARCKS (McLaughlin and Murray, 2005).
Comparing the charges of PI(4,5)P2versus PS at the physi-
ological pH, LUVs composed of 5% PI(4,5)P2have approx-
Luciferase siRNA (as control). Exo70 was knocked down as detected by Western blot. The amount of Sec8 in these cells was not affected. The
anti-Exo70 monoclonal antibody (13F3) and anti-Sec8 monoclonal antibody (2E12) were used in the Western blot analysis. (B) Luciferase
siRNA-treated HeLa cells were transfected with VSV-G-GFP, kept at 40°C overnight, and shifted to 32°C for 0, 15, 30, 60, and 90 min in the
presence of cycloheximide (100 ?g/ml). The cells were then fixed and stained as described in Materials and Methods. VSV-G was transported
from ER, to the Golgi, and to the PM. (C) In EXO70 knockdown cells, VSV-G transport to the PM through the endoplasmic reticulum and
Golgi complex was normal. However, the incorporation of VSV-G (stained by the 8G5 monoclonal antibody recognizing the extracellular
domain of VSV-G) was considerably delayed. Instead of being detected at the surface at 90 min as in control cells, VSV-G was detectable after
180 min from temperature arrest. (D and E) VSV-G association with various intracellular compartments was quantified in cells expressing
luciferase siRNA (D) or EXO70 siRNA (E). For the quantification, boundaries of the whole cell, Golgi region and cell periphery region were
outlined, and VSV-G fluorescence in these areas was then quantified using ImageJ 1.73v software after subtraction of background outside the
cell. (F) Quantification of fluorescence intensity of surface VSV-G signal stained by the 8G5 monoclonal antibody. Boundary of the cell surface
was outlined, and average fluorescence intensity of surface VSV-G signal was quantified using ImageJ 1.73v software and then divided by
the perimeter of the cell surface. Three independent experiments (20 cells each) were carried out. Error bars, SD. p ? 0.01.
Exocytosis of VSV-G ts045 is blocked in EXO70 siRNA knockdown cells. (A) HeLa cells were treated with EXO70 siRNA and
PI(4,5)P2Mediates Exocyst Targeting
Vol. 18, November 20074489
imately the same amount of effective charges as LUVs com-
posed of 15–20% PS. However, the interaction of Exo70 with
LUVs composed of 20% PS is much weaker. Mutations on
exo70-1 that eliminate some of the positive charges in the
Exo70 C-terminus nearly abolished the ability of Exo70 to
bind PI(4,5)P2; however, the ability of this mutant to bind PS
at high concentrations (?60%) was only partially affected.
These data suggest that Exo70 has significant binding spec-
siRNA knockdown cells expressing the exo70-1 mu-
tant. (A) EXO70 siRNA knockdown cells were trans-
fected with GST-tagged rat Exo70 (GST-rExo70) or
rat exo70 mutant (GST-rexo70-1). The amount of en-
dogenous Exo70 in HeLa cells, and the amount of
extragenically expressed GST-tagged rat Exo70 were
detected by anti-Exo70 monoclonal antibody (top
panel). The total proteins in the cell lysates were
detected by SYPRORed staining (bottom panel). (B)
VSV-G-GFP trafficking in HeLa cells transfected
with EXO70 siRNA. The cells were grown at 40°C
overnight after transfection. The cells were then
shifted to 32°C for 0, 30, 60, and 90 min in the
presence of cycloheximide, fixed, and stained as de-
scribed in Materials and Methods. The 8G5 monoclo-
nal antibody was used to detect the surface-incorpo-
rated VSV-G. (C and D) VSV-G-myc trafficking was
examined in EXO70 siRNA knockdown cells ex-
pressing GST-rExo70 (C) or GST-rexo70-1 (D). VSV-
G-myc and GST-tagged wild-type rExo70 (C) or
exo70-1 (D) were cotransfected into EXO70 knock-
down HeLa cells. VSV-G transport to the PM was
rescued in the EXO70 siRNA knockdown cells ex-
pressing wild-type rExo70, whereas VSV-G trans-
port was not rescued in the EXO70 siRNA knock-
down cells expressing exo70-1. (E) Quantification of
the percentage of cells with surface VSV-G staining.
VSV-G ts045 exocytosis defect in EXO70
J. Liu et al.
Molecular Biology of the Cell4490
ificity for PI(4,5)P2over PS. We have also examined the
interaction of Exo70 with other phosphoinositides and ob-
served its selectivity for PI(4,5)P2over the stereoisomeric
PI(3,5)P2and other monophosphorylated phosphoinositi-
des. We have also found that Exo70 binds PI(3,4,5)P3with an
affinity that is comparable to that for PI(4,5)P2. PI(3,4,5)P3
has recently been found to be localized to the basolateral
domain in MDCK cells (Gassama-Diagne et al., 2006), and
the exocyst has been implicated in basolateral vesicle target-
ing (Grindstaff et al., 1998). In other types of mammalian
cells, although the concentration of PI(3,4,5)P3is low at the
PM in resting cells, it can be rapidly up-regulated in re-
sponse to extracellular stimuli. It is therefore possible that
Exo70 binds to PI(3,4,5)P3under certain physiological cir-
cumstances or in certain cell types.
In yeast, we have found that the amount of the exocyst
complex associated with the PM was much lower in the
temperature-sensitive mss4 mutant cells, in which the
PI(4,5)P2level in the PM was reduced (data not shown),
indicating that PI(4,5)P2mediates the membrane targeting of
the exocyst. Moreover, the structural analysis of Exo70 pro-
vided important insights into the potential mechanism of
membrane association of Exo70 (Dong et al., 2005; Ham-
burger et al., 2006). On the basis of the crystal structure
information of yeast Exo70, we have made mutations on the
rat Exo70 residues K632 and K635 (exo70-1), which are pos-
itively charged amino acids well conserved in the yeast
Exo70 sequence. Our in vitro vesicle sedimentation experi-
ments demonstrated that these mutations disrupted the
PI(4,5)P2-binding. Furthermore, the VSV-G trafficking anal-
ysis further demonstrated its functional importance in exo-
cytosis at the PM. While we were preparing this article,
Moore et al. (2007) resolved the crystal structure of mouse
Exo70. Analysis of the structure indicates that the point
mutations on exo70-1 are localized on the loop between
Helix 18 and 19 on the surface of the mouse Exo70, which is
almost identical to that of the yeast Exo70.
Taking advantage of the VSV-G trafficking assay, we were
able to analyze the role of Exo70 in various stages of mem-
brane traffic. More importantly, using the exo70-1 mutant in
the EXO70 RNAi knockdown cells, we were able to specif-
ically examine the functional significance of Exo70- PI(4,5)P2
interaction in VSV-G exocytosis. The exocyst has been found
in various cellular compartments, including Golgi and recy-
cling endosomes, in addition to the PM (Yeaman et al., 2001;
Fo ¨lsch et al., 2003; Ang et al., 2004; Langevin et al., 2005). Here
we found that the Exo70-PI(4,5)P2interaction is not involved
in the early stages of VSV-G trafficking through the endo-
plasmic reticulum and Golgi. Rather, it is critical for the PM
events such as vesicle tethering and fusion. When the Exo70-
PI(4,5)P2interaction was disrupted, the transport of VSV-G
to the PM was barely changed. However, the incorporation
of VSV-G into the PM was significantly affected, as revealed
by the 8G5 antibody specifically recognizing the extracellu-
lar domain of VSV-G. Similarly the exocyst has been impli-
cated in tethering and fusion of Glut4-containing vesicles in
3T3-L1 adipocytes (Inoue et al., 2003; Ewart et al., 2005;
Tsuboi et al., 2005). It is possible that other exocyst compo-
nents also interact with phospholipids (Moskalenko et al.,
2003). However, specific disruption of Exo70-PM interaction
is sufficient to block exocytosis in mammalian cells.
The association of the exocyst with the PM is an important
step in vesicle tethering. When and where this interaction
takes place may regulate the kinetics and location of exocy-
tosis. In the budding yeast S. cerevisiae, the exocyst complex
is specifically localized to the growing tip of the daughter
cell (the “bud tip”), which is the site of active exocytosis and
cell surface expansion. Moreover, Exo70 primarily functions
at the early stages of the yeast cell cycle, suggesting a tem-
poral control of Exo70 function (He et al., 2007). In mamma-
lian cells, growth factor signaling involving small GTPases
may mediate the subunits assembly, translocation of the
exocyst from intracellular compartments to, or activation of
the exocyst at, the PM (Sugihara et al., 2002; Moskalenko et
al., 2002, 2003; Inoue et al., 2003; Takaya et al., 2004; Zuo et al.,
2006). Future work will be focused on the identification and
characterization of proteins that temporally and/or spatially
regulate Exo70 and other exocyst components using differ-
ent eukaryotic systems.
We are grateful to Drs. Margaret Chou, Michael Marks (University of Penn-
sylvania), and Zhaohui Xu (University of Michigan) for their constructive
discussions and Dr. Michael Ostap (University of Pennsylvania) for advice on
lipid binding experiments. We thank Drs. Douglas Lyles (Wake Forest Uni-
versity) and Shu-Chan Hsu (Rutgers University) for the valuable antibodies
used in the experiments. This work is supported by grants from National
Institutes of Health (RO1-GM64690), American Cancer Society, and the Pew
Scholars Program to W.G.
Ang, A. L., Taguchi, T., Francis, S., Fo ¨lsch, H., Murrells, L. J., Pypaert, M.,
Warren, G., and Mellman, I. (2004) Recycling endosomes can serve as inter-
mediates during transport from the Golgi to the plasma membrane of MDCK
cells. J. Cell Biol. 167(3), 531–543.
Balla, T. (2005). Inositol-lipid binding motifs: signal integrators through pro-
tein-lipid and protein-protein interactions. J. Cell Sci. 118, 2093–2104.
Behnia, R., and Munro, S. (2005). Organelle identity and the signposts for
membrane traffic. Nature 438, 597–604.
Bonangelino, C. J., Nau, J. J., Duex, J. E., Brinkman, M., Wurmser, A. E., Gary,
J. D., Emr, S. D., and Weisman, L. S. (2002). Osmotic stress-induced increase
of phosphatidylinositol 3,5-bisphosphate requires Vac14p, an activator of the
lipid kinase Fab1p. J. Cell Biol. 156(6), 1015–1028.
Boyd, C., Hughes, T., Pypaert, M., and Novick, P. (2004). Vesicles carry most
exocyst subunits to exocytic sites marked by the remaining two subunits,
Sec3p and Exo70p. J. Cell Biol. 167(5):889–901.
Brennwald, P., Kearns, B., Champion, K., Keranen, S., Bankaitis, V., and
Novick, P. (1994). Sec9 is a SNAP-25-like component of a yeast SNARE
complex that may be the effector of Sec4 function in exocytosis. Cell 79(2),
Di Paolo, G. P., and De Camilli, P. (2006). Phosphoinositides in cell regulation
and membrane dynamics. Nature 443(7112), 651–657.
Dong, G., Hutagalung, A. H., Fu, C., Novick, P., and Reinisch, K. M. (2005).
The structures of exocyst subunit Exo70p and the Exo84p C-terminal domains
reveal a common motif. Nat. Struct. Mol. Biol. 12, 1094–1100.
Finger, F. P., Hughes, T. E., and Novick, P. (1998). Sec3p is a spatial landmark
for polarized secretion in budding yeast. Cell 92(4), 559–571.
Ewart, M. A., Clarke, M., Kane, S., Chamberlain, L. H., and Gould, G. W.
(2005). Evidence for a role of the exocyst in insulin-stimulated Glut4 traffick-
ing in 3T3–L1 adipocytes. J. Biol. Chem. 280(5), 3812–3816.
Fo ¨lsch, H., Pypaert, M., Maday, S., Pelletier, L., and Mellman, I. (2003). The
AP-1A and AP-1B clathrin adaptor complexes define biochemically and func-
tionally distinct membrane domains. J. Cell Biol. 163, 351–362.
Gassama-Diagne, A., Yu, W., ter Beest, M., Martin-Belmonte, F., Kierbel, A.,
Engel, J., and Mostov, K. (2006). Phosphatidylinositol-3,4,5-trisphosphate reg-
ulates the formation of the basolateral plasma membrane in epithelial cells.
Nat. Cell Biol. 8(9): 963–970.
Grindstaff, K. K., Yeaman, C., Anandasabapathy, N., Hsu, S. C., Rodriguez-
Boulan, E., Scheller, R. H., and Nelson, W. J. (1998). Sec6/8 complex is
recruited to cell-cell contacts and specifies transport vesicle delivery to the
basal-lateral membrane in epithelial cells. Cell 93(5), 731–740.
Guo, W., Roth, D., Walch-Solimena, C., and Novick P. (1999). The exocyst is
an effector for Sec4p targeting secretory vesicles to sites of exocytosis. EMBO
J. 18(4), 1071–1080.
Guo, W., Sacher, M., Barrowman, J., Ferro-Novick, S., and Novick, P. (2000).
Protein complexes in transport vesicle targeting. Trends Cell Biol. 10(6),
PI(4,5)P2Mediates Exocyst Targeting
Vol. 18, November 20074491
Hamburger, Z. A., Hamburger, A. E., West, A. P., Jr., and Weis, W. I. (2006).
Crystal structure of the S. cerevisiae exocyst component Exo70p. J. Mol. Biol.
He, B., Xi, F., Zhang, J., TerBush, D., Zhang, X., and Guo, W. (2007). Exo70p
mediates the secretion of specific exocytic vesicles at early stages of the cell
cycle for polarized cell growth. J. Cell Biol. 176(6), 771–777.
Hokanson, D., and Ostap, M. (2006). Myo1c binds tightly and specifically to
phosphatidylinositol 4,5-bisphosphate and inositol 1,4,5-trisphosphate. Proc.
Natl. Acad. Sci. USA 103(9), 3118–3123.
Hazuka, C. D., Foletti, D. L., Hsu, S. C., Kee, Y., Hopf, F. W., and Scheller,
R. H. (1999). The sec6/8 complex is located at neurite outgrowth and axonal
synapse-assembly domains. J. Neurosci. 19(4), 1324–1334.
Hsu, S. C., TerBush, D., Abraham, M., and Guo, W. (2004). The exocyst
complex in polarized exocytosis. Int. Rev. Cytol. 233, 243–265.
Inoue, M., Chang, L., Hwang, J., Chiang, S. H., and Saltiel, A. R. (2003). The
exocyst complex is required for targeting of Glut4 to the plasma membrane by
insulin. Nature 422(6932), 629–633.
Insall, R. H., and Weiner, O. D. (2001). PIP3, PIP2, and cell movement—
similar messages, different meanings? Dev. Cell 1(6), 743–747.
Langevin, J, Morgan, M. J., Sibarita, J. B., Aresta, S., Murthy, M., Schwarz, T.,
Camonis, J., and Bellaiche, Y. (2005). Drosophila exocyst components Sec5,
Sec6, and Sec15 regulate DE-Cadherin trafficking from recycling endosomes
to the plasma membrane. Dev. Cell 9(3), 355–376.
Lefrancois, L., and Lyles, D. S. (1982). The interaction of antibody with the
major surface glycoprotein of vesicular stomatitis virus. II. Monoclonal anti-
bodies of nonneutralizing and cross-reactive epitopes of Indiana and New
Jersey serotypes. Virology 121(1), 168–174.
Matern, H. T., Yeaman, C., Nelson, W. J., and Scheller, R. H. (2001). The
Sec6/8 complex in mammalian cells: characterization of mammalian Sec3,
subunit interactions, and expression of subunits in polarized cells. Proc. Natl.
Acad. Sci. USA 17, 9648–9653.
McLaughlin, S., Wang, J., Gambhir, A., and Murray, D. (2002). PIP2and
proteins: interactions, organization, and information flow. Annu. Rev. Bio-
phys. Biomol. Struct. 31, 151–175.
McLaughlin, S., and Murray, D. (2005). Plasma membrane phosphoinositide
organization by protein electrostatics. Nature 438(7068), 605–611.
Moore, B. A., Robinson, H. H., and Xu, Z. (2007). The crystal structure of
mouse Exo70 reveals unique features of the mammalian exocyst. J. Mol. Biol.
Moskalenko, S., Henry, D. O., Rosse, C., Mirey, G., Camonis, J. H., and White,
M. A. (2002). The exocyst is a Ral effector complex. Nat. Cell Biol. 1, 66–72.
Moskalenko, S., Tong, C., Rosse, C., Mirey, G, Formstecher, E., Daviet, L.,
Camonis, J., and White, M. A. (2003) Ral GTPases regulate exocyst assembly
through dual subunit interactions. J. Biol. Chem. 278(51), 51743–51748.
Munson, M., and Novick, P. (2006) The exocyst defrocked, a framework of
rods revealed. Nat. Struct. Mol. Biol. 13(7), 577–581.
Papayannopoulos, V., Co, C., Prehoda, K. E., Snapper, S., Taunton, J., and
Lim, W. A. (2005). A polybasic motif allows N-WASP to act as a sensor of
PIP(2) density. Mol. Cell 17(2), 181–191.
Pfeffer, S. R. (1999) Transport-vesicle targeting: tethers before SNAREs. Nat.
Cell Biol. 1(1), E17–E22.
Rosse, C., Hatzoglou, A., Parrini, M. C., White, M. A., Chavrier, P., and
Camonis, J. (2006). RalB mobilizes the exocyst to drive cell migration. Mol.
Cell Biol. 26, 727–734.
Sugihara, K., Asano, S., Tanaka, K., Iwamatsu, A., Okawa, K., and Ohta, Y.
(2002). The exocyst complex binds the small GTPase RalA to mediate filopo-
dia formation. Nat. Cell Biol. 4, 73–78.
Takaya, A., Ohba, Y., Kurokawa, K., and Matsuda, M. (2004). RalA activation
at nascent lamellipodia of epidermal growth factor-stimulated Cos7 cells and
migrating Madin-Darby canine kidney cells. Mol. Biol. Cell 15, 2549–2557.
TerBush, D. R., and Novick, P. (1995). Sec6, Sec8, and Sec15 are components
of a multisubunit complex which localizes to small bud tips in Saccharomyces
cerevisiae. J. Cell Biol. 130(2), 299–312.
Tsuboi, T., Ravier, M. A., Xie, H., Ewart, M. A., Gould, G. W., Baldwin, S. A.,
and Rutter, G. A. (2005). Mammalian exocyst complex is required for the
docking step of insulin vesicle exocytosis. J. Biol. Chem. 280(27), 25565–25570.
Vega, I. E., and Hsu, S. C. (2001). The exocyst complex associates with
microtubules to mediate vesicle targeting and neurite outgrowth. J. Neurosci.
Wang, S., Liu Y., Adamson, C. L., Valdez, G., Guo, W., and Hsu, S.-C. (2004).
The mammalian exocyst, a complex required for exocytosis, inhibits tubulin
polymerization. J. Biol. Chem. 279, 35958–35966.
Wang, S., and Hsu, S. C. (2006) The molecular mechanisms of the mammalian
exocyst complex in exocytosis. Biochem. Soc. Trans. 34(Pt 5):687–690.
Waters, M. G., and Hughson, F. M. (2000) Membrane tethering and fusion in
the secretory and endocytic pathways. Traffic 1(8):588–597.
Weber, T. M., Joost, H., Simpson, I. A., and Cushman, S. W. (1988) Receptor
Biochemistry and Methodology, ed. R. C. Kahn and L. C. Harrison, New York:
Alan R. Liss, 171–187.
Whyte, J. R., and Munro, S. (2002). Vesicle tethering complexes in membrane
traffic. J. Cell Sci. 115(Pt 13), 2627–2637.
Xu, K. F., Shen, X., Li, H., Pacheco-Rodriguez, G., Moss, J., and Vaughan, M.
(2005). Interaction of BIG2, a brefeldin A-inhibited guanine nucleotide-ex-
change protein, with exocyst protein Exo70. Proc. Natl. Acad. Sci. USA 102,
Yeaman, C., Grindstaff, K. K., Wright, J. R., Nelson, W. J. (2001). Sec6/8
complexes on trans-Golgi network and plasma membrane regulate late stages
of exocytosis in mammalian cells. J. Cell Biol. 155(4), 593–604.
Zuo, X., Zhang, J., Zhang, Y., Hsu, S. C., Zhou, D., and Guo, W. (2006). Exo70
interacts with the Arp2/3 complex and regulates cell migration. Nat. Cell Biol.
J. Liu et al.
Molecular Biology of the Cell4492