Biochem. J. (2009) 418, 233–246 (Printed in Great Britain)
233
doi:10.1042/BJ20082105
REVIEW ARTICLE
Phosphatidylinositol metabolism and membrane fusion
Dominic POCCIA*1 and Banafshé LARIJANI†1
*Department of Biology, Amherst College, Amherst, MA, 01002, U.S.A., and †Cell Biophysics Laboratory, Lincoln’s Inn Fields Laboratories, Cancer Research UK,
44 Lincoln’s Inn Fields, London WC2A 3PX, U.K.
Membrane fusion underlies many cellular events, including secretion, exocytosis, endocytosis, organelle reconstitution, transport
from endoplasmic reticulum to Golgi and nuclear envelope
formation. A large number of investigations into membrane
fusion indicate various roles for individual members of the phosphoinositide class of membrane lipids. We first review the
phosphoinositides as membrane recognition sites and their
regulatory functions in membrane fusion. We then consider how
modulation of phosphoinositides and their products may affect
the structure and dynamics of natural membranes facilitating
fusion. These diverse roles underscore the importance of these
phospholipids in the fusion of biological membranes.
INTRODUCTION
is fundamentally an event requiring phospholipid alterations. The
emphasis of this review is on the participation of phospholipid
components derived from PtdIns in intracellular fusions. We will
emphasize only a few examples and refer to more extensive
reviews of individual subjects throughout.
Membrane fusion events
Membrane fusion is at the heart of intracellular membrane traffic
phenomena such as endocytosis, secretion and organelle reconstitution and other interactions such as fertilization or certain viral
infections. Membrane fusion events can be conceived diagrammatically as the reverse of membrane fission events. Fusion
and fission have several distinguishing characteristics, such as
use of different protein players (e.g. dynamin, SNAREs {SNAP
[soluble NSF (N-ethylmaleimide-sensitive factor)-attachment
protein] receptors}), but have other aspects in common, such as
highly curved membrane structures in the constricted membrane
regions formed just prior to the fusion or fission event.
Fusion is normally thought to involve several steps, which
include tethering and docking, priming and bilayer fusion,
although it is often not simple to dissect each [1,2]. The ordering
of events is not certain, but tethering or docking represents the
first stable contact of the membranes closely related to docking,
and priming optimizes complete docking and fusigenic potential.
Tethering or docking establishes specificity of binding of various
membrane compartments. In the docked state, membranes can
receive the signals to proceed with fusion.
The actual fusion event is generally thought to involve the
formation of an intermediate or hemifused state which can
progress to form a fusion pore or regress to the unfused state.
Therefore fusion does not necessarily proceed unidirectionally.
For example, ‘flickering’ or rapid opening and closing of the pore
may release secretory vesicle contents at the plasma membrane
in small amounts [3–8]. Although fusion can take place between
synthetic membranes lacking proteins, natural membrane fusion
is complicated (and regulated) by protein interactions. However, it
Key words: diacylglycerol, membrane curvature, membrane
fusion, nuclear envelope, phosphoinositide-specific phospholipase C (PI-PLC), soluble N-ethylmaleimide-sensitive fusion
protein-attachment protein receptor (SNARE).
Potential roles for phosphoinositides in membrane fusion
A wide range of investigations into membrane fusion indicate
diverse roles for various members of the phosphoinositide class
of membrane lipids. Here we focus on interactions of these phospholipids with proteins that serve as binding sites and/or regulators and more direct roles of phosphoinositides in membrane
structure that could facilitate fusion [9]. Each of these functions
may be altered by changes in the proportions of particular PtdIns
species due to their extensive variety of interconversions and
modifications by enzymes.
Phosphoinositides can have three distinct roles in membrane
fusion events. They can serve as markers or flags aiding the
recruitment of proteins which may influence tethering, docking
or fusion. They can be regulators of the proteins that make up
the fusion machinery, a role related to the former. They can also
have structural roles, altering physical properties, such as fluidity
or curvature of the fusigenic membranes, either globally or at the
site of fusion. The former two roles have received much more
attention in studies of natural membrane systems; the latter has
been studied using synthetic membranes and, to some extent,
natural membranes. Following a brief introduction to some of
the complexities of the family of phosphoinositides and caveats
concerning interpretation of their localization, we survey the
binding and regulatory roles together and then potential structural
ramifications of PtdIns modification.
Abbreviations used: CAPS, calcyphosine; CISK, cytokine-independent survival kinase; DAG, diacylglycerol; DCV, dense core vesicle; DGK, DAG
kinase; EEA1, early endosomal antigen 1; ENTH, epsin N-terminal homology; ER, endoplasmic reticulum; IP3 , myo -inositol 1,4,5-trisphosphate; LUV,
large unilamellar vesicle; NSF, N -ethylmaleimide-sensitive factor; PH domain, pleckstrin homology domain; PI3K, phosphoinositide 3-kinase; PIP,
phosphoinositide phosphate; PIP5K, PtdIns4P 5-kinase; PKC, protein kinase C; PLA, phospholipase A; PLC, phospholipase C; PLD, phospholipase
D; PI-PLC, phosphoinositide-specific PLC; PtdOH, phosphatidic acid; PX domain, Phox homology domain; SARA, Smad anchor for receptor activation;
SNAP, soluble NSF-attachment protein; SNARE, SNAP receptor; SCAMP2, secretory carrier membrane protein 2; VAMP, vesicle-associated membrane
protein.
1
Correspondence may be addressed to either of these authors (email [email protected] or [email protected]).
c The Authors Journal compilation c 2009 Biochemical Society
234
Figure 1
D. Poccia and B. Larijani
Phospholipid metabolic pathways
(A) Structure of PtdIns. The head group can be reversibly phosphorylated at the 3, 4 and 5 hydroxyl group positions. Adapted from van Haesebroeck et al. [10]. (B) Known interconversions of
phosphoinositides by phosphatases, which are counterbalanced by kinases. Adapted from van Haesebroeck et al. [10]. (C) Conversion of DAG into PtdOH by DGK. (D) Positions
of hydrolytic attack of various phospholipases on phospholipids.
PHOSPHOINOSITIDE RECOGNITION AND REGULATION OF
MEMBRANE FUSION
Phosphoinositide metabolism
Phosphoinositides comprise a family of phospholipid molecules
with head groups built upon D-myo-InsP typically esterified to
DAG (diacylglycerol) through a phosphodiester bond at the C-1
position leaving up to five free hydroxyl groups on the inositol
ring (Figure 1A). The inositol of PtdIns may be phosphorylated
in a variety of combinations at the 3, 4 and 5 positions of
the ring, yielding PtdIns3P, PtdIns4P, PtdIns5P, PtdIns(3,4)P2 ,
PtdIns(3,5)P2 , PtdIns(4,5)P2 and PtdIns(3,4,5)P3 [10]. Conversion between these forms is controlled by specific phosphoinositide kinases and phosphatases [11] (Figure 1B). Typically,
phosphoinositides are present in low amounts in membranes
relative to other phospholipids [10], with some notable exceptions
[12].
In addition to changes in phosphorylation state, hydrolysis of
PtdIns(4,5)P2 by PI-PLC [phosphoinositide-specific PLC (phospholipase C)] yields soluble IP3 (myo-inositol 1,4,5-trisphosphate) and membrane-resident DAG. DAG may be converted
into PtdOH (phosphatidic acid) by DGKs (DAG kinases) and
PtdOH converted back into DAG by PtdOH phosphatases [13,14]
(Figure 1C). Both DAG and PtdOH have multiple signalling
roles, including actions downstream on PKC (protein kinase C),
PKD (protein kinase D) and PI5K (phosphoinositide 5-kinase)
[14]. Eukaryotic phospholipases act only on PtdIns(4,5)P2 .
However, PtdIns(4,5)P2 is not only a source of DAG and IP3 ,
but can itself serve as a binding or anchoring site for other
proteins and can activate interacting proteins as well [15]. The
multiplicity of possible interactions of phosphoinositides makes
defining their potential roles in fusion complex.
Phosphoinositide recognition: binding domains in proteins
Although several specific binding domains exist that recognize
PtdIns(4,5)P2 , non-specific interactions through electrostatic
attraction play a role in binding to and sequestering PtdIns(4,5)P2
[16]. For example the MARCKS (myristoylated alanine-rich
C-kinase substrate) protein, an unstructured PKC substrate present in high concentrations in cells, binds to membranes by an
c The Authors Journal compilation c 2009 Biochemical Society
N-terminal myristate and a highly basic stretch of 25 amino acids
[16]. This effector domain binds electrostatically to PtdIns(4,5)P2 ,
sequestering it and inhibiting its hydrolysis by PLC. Proteins
regulating fusion, such as SCAMP2 (secretory carrier membrane
protein 2), also bind primarily electrostatically to PtdIns(4,5)P2
[17]. Such proteins could serve to limit the rapid diffusion
of PtdIns(4,5)P2 from its site of synthesis and contribute to
membrane regions enriched in PtdIns(4,5)P2 [18].
More specific interactions depend on motifs that recognize
PtdIns3P [FYVE or PX (Phox homology) domains], PtdIns4P
[PH (pleckstrin homology)], PtdIns5P (PD) PtdIns(3,4)P2 (PH,
PX), PtdIns(3,5)P2 [ENTH (epsin N-terminal homology)/ANTH
(AP180 N-terminal homology), GRAM], PtdIns(4,5)P2 (PH,
ENTH), PtdIns(3,4,5)P3 (PH) or DAG (C1-domain of PKCδ)
[19,20]. Not all domains bind with high affinity or specifically and
may require additional interactions for their specific localization
[16,21].
Phosphoinositide localization and quantification
Establishing the intracellular locations and quantification of
individual phosphoinositides poses several formidable difficulties [20]. Consequently, estimates of phosphoinositide content
and distribution in cells are subject to considerable uncertainties.
In this respect, agreement of determinations by independent
techniques is more valuable than estimates by single methodologies. The most common determinations of phosphoinositide
content and distribution are either analytical assays on isolated
membrane fractions or through use of probes more or less specific
for individual species.
Since phosphoinositides are often present in very small molar
fractions of total phospholipids, determination by techniques
such as TLC is of limited use unless accompanied by metabolic
radiolabelling, which is not easily achieved for all tissue
samples. Continuing advances in HPLC in combination with electrospray ionization tandem MS allow much more sensitive
determination of phosphoinositides, DAG and PtdOH [12,22–24].
Imaging in fixed or live cells depends on probes that are
constructed from the domains of proteins that bind to phosphoinositides or their metabolic products [25]. Transfection with
fluorescent constructs is useful with the caveat that expression
Phosphatidylinositol metabolism and membrane fusion
may alter normal function of the inositides [19,25,26]. Moreover,
transfection of GFP (green fluorescent protein)–PH domains alone
is not a specific marker for PtdIns(4,5)P2 , since in this context the
domain has a higher affinity for IP3 than PtdIns(4,5)P2 . In addition,
binding specificity of domains can be altered by interactions of
other domains in the proteins from which they are derived, and
the strength of the fluorescent signal may not be proportional
to the concentration of target. Fixed cells may be probed
with the binding domains, but interpretation depends on adequate
fixation of membrane lipids and availability of the lipids, which
can be affected by the in vivo binding of proteins [27]. Thus
in vitro binding to phosphoinositide blots or liposomes may not
fully reflect in vivo binding patterns if in vivo binding sites are
blocked [19]. Additionally, antibodies against PtdIns(4,5)P2 often
also recognize PtdIns(3,4,5)P3 , so that care must be taken in
interpretation of data when these antibodies are used.
Compartmental specificity
In certain cell types, phosphoinositides which react with the
probes containing specific recognition domains indicate major
subcellular localization sites. PtdIns3P and PtdIns(3,5)P2 have
been localized to endosomes and yeast vacuoles [28], PtdIns4P to
Golgi, PtdIns5P to nuclei [29], PtdIns(3,4)P2 and PtdIns(3,4,5)P3
to plasma membranes and PtdIns(4,5)P2 to plasma membranes
and nuclei [20] by microscopy techniques. It is unlikely that these
are either the sole phosphoinositides in these organelles or that any
one of the lipids is specific to a given organelle. MS has mostly
been applied to whole cell extracts, but analyses of individual
isolated membrane fractions suggests multiple phosphorylated
isoforms coexist in these membranes as might be expected for a
metabolically dynamic family of phospholipids [12]. Improved
methods of purification of membranous compartments will be
necessary for better quantification through MS. The notion
that organelle identity can be derived from the distributions of
phosphoinositides, their modifying kinases and phosphatases and
specific Rab GTPases persists [30–32], although these identities
may reflect quantitative rather than qualitative differences or
difficulties of accurate assay.
GTPases responsible for organelle identity are typically
members of the Rab and Arf families. The GTP-bound forms
associate with membranes and return to the cytoplasm in the GDPbound form [32]. Binding is mediated through prenyl groups and
activation is controlled by GAPs (GTPase-activating proteins),
GEFs (guanine-nucleotide-exchange factors) and GDIs (GDPdissociation inhibitors). On the other hand, Arfs are usually
myristoylated and are also regulated by GEFs and GAPs. GTPases
are typically involved in regulation of membrane fusion. In
particular, Rab proteins seem to be involved in regulating several
steps leading to membrane fusion [33]. For example, Rabs are
involved in recruitment of tethering molecules that facilitate
SNARE interactions [34]. Importantly, GTPases can also result in
altered phosphoinositide composition.
SNAREs and phosphoinositides in membrane fusion
The dominant paradigm for membrane fusion has been the
SNARE hypothesis put forth initially by Rothman and co-workers
[35]. Although individual SNAREs were originally hypothesized
to be responsible for specific membrane recognition as well as
fusion, they are now believed to work in asymmetric combinations, participating as components of more complex assemblies
responsible for both recognition and fusion [36]. SNARE function
has been reviewed extensively elsewhere [34,36–42].
235
In addition to the SNARE proteins themselves, which include
syntaxin, synpatobrevin/VAMP (vesicle-associated membrane
protein) and SNAP-25, other proteins also contribute to their
roles in fusion. These include various small GTPases of the
Rab family, sec1/MUNC or SM proteins which bind to syntaxin
[43,44], the Ca2+ -binding synaptotagmins which regulate fusion
[45,46], complexins which either inhibit or facilitate SNARE
action [47,48], and NSF and SNAPs, originally thought to drive
fusion, but which are now known to dissociate the SNARE
complexes after fusion.
There are many examples of interactions of the SNARE machinery with phospholipids [49]. A regulatory role for interaction
of syntaxin1A with phospholipids is suggested by recent experiments in which binding and sequestering of the lipids to fusion
sites depends on syntaxin 1A [50], possibly through its polybasic
unstructured juxtamembrane domain. Bound lipids include phosphoinositides and PtdOH in PC12 cells. Syntaxin 1A mutants that
diminish lipid binding lower exocytosis and alter the dynamics of
the fusion pore, with longer pore durations and smaller pore size.
Synaptotagmin, when it binds Ca2+ , increases its affinity
for phospholipids, which results in pulling membranes closer
together in a tethering or docking step [51]. The effect is mediated through unstructured polybasic regions of the C2 B domain,
facilitating interaction with acidic lipids such as phosphoinositides. Synaptotagmin 1 binds two PtdIns(4,5)P2 molecules
in two different ways. A Ca2+ -independent binding mode allows
penetration of the protein into PtdIns(4,5)P2 -rich membranes with
very rapid kinetics, which has been proposed as a mechanism for
rapid responses to bring vesicle and plasma membrane into close
proximity prior to fusion [52].
Selected examples of regulation of cellular processes and
recognition of cellular compartments
Exocytosis
After packaging of vesicular contents, exocytosis requires fusion
of vesicles with the plasma membrane. The process consists of
both constitutive and regulated exocytosis in a variety of cell
types for a wide range of functions. PtdIns3P, PtdIns(4,5)P2 ,
PtdIns(3,4,5)P3 and DAG all have roles in regulation of
exocytosis.
Roles of PtdIns(4,5)P2 in exocytosis. PtdIns(4,5)P2 is involved
in binding of exocytic vesicles to the plasma membrane. For
example, in budding yeast, polarized exocytosis marshals materials needed for expansion of the cell membrane and construction
of the cell wall at the tip of the bud. Docking of post-Golgi
exocytic vesicles to the plasma membrane involves a complex,
the exocyst, which consists of eight proteins, including Rab and
Rho GTPases (Figure 2). The exocyst may tether vesicles to the
plasma membrane or have a more direct role in SNARE complex
formation [53].
Binding of many of the exocyst components is lipid-mediated.
Sec3 binds to PtdIns(4,5)P2 through a polybasic region in its Nterminus (Figure 2) [54]. Binding to both PtdIns(4,5)P2 and the
Rho GTPase Cdc42 are required for exocytosis (Cdc42 integrates
polarization of actin filaments and exocytosis during budding).
Other proteins, such as Gic1 and Gic2, are effectors of Cdc42,
and deletion of both blocks exocytosis and cell polarization [55].
Moreover, Gic2 binds to Cdc42 through a domain called CRIB
[Cdc42 (cell-division cycle 42)/Rac-interacting binding domain]
[56], and adjacent to CRIB is another polybasic domain that binds
to PtdIns(4,5)P2 . Depletion of PtdIns(4,5)P2 or mutation of the
polybasic domain disrupts exocytosis [56]. In addition to Sec3,
Exo70 has been shown to tether vesicles to the plasma membrane
c The Authors Journal compilation c 2009 Biochemical Society
236
D. Poccia and B. Larijani
Note the out of proportion membrane size to illustrate the role of PtdIns(4,5)P 2 in Sec protein
recruitment. Red lipids denote PtdIns(4,5)P 2 molecules. ++ symbols denote basic residues
on protein. Adapted from Wu et al. [168].
In neuroendocrine chromaffin cells and PC12 cells, polyphosphoinositide synthesis is required for exocytosis of noradrenaline
(norepinephrine). The role of proteins in the regulated secretion
of DCVs (dense core vesicles) in neuroendocrine cells has been
recently reviewed [39]. DCVs are triggered to exocytose by Ca2+ .
Synaptotagmin serves as a Ca2+ sensor in chromaffin cells and is
targeted by binding through its two C2 domains to PtdIns(4,5)P2
[39]. DCV exocytosis is inhibited by the PtdIns(4,5)P2 -binding
PH domain of PLCδ [62] and by depletion of PtdIns(4,5)P2 [63]. It
has been suggested that the Ca2+ bound to synaptotagmin strongly
attracts negatively charged phospholipids to deform membranes
and overcome energy barriers to form a fusion stalk [46].
In the late stages of Ca2+ -dependent exocytosis, SCAMP2
binds through basic electrostatic interactions with PtdIns(4,5)P2 .
Mutation of this binding region interferes with exocytosis by
decreasing the stability of nascent pores and the probability of
their opening in PC12 cells [17].
SCAMP2 also serves as a scaffold for phospholipase D,
Arf and PIP5K. Increasing PtdIns(4,5)P2 levels in the plasma
membrane of chromaffin cells by infusing PtdIns(4,5)P2 or overexpressing PtdIns(4)P 5-kinase increases secretion. Directing the
5-phosphatase domain of synaptojanin to the plasma membrane
decreases secretion [64]. PI3K (phosphoinositide 3-kinase)
inhibition results in a transient increase in PtdIns(4,5)P2 and
secretion. Since fusion kinetics are unaffected, but the size of
the pools of vesicles ready to release increase with PtdIns(4,5)P2
levels, it was suggested that PtdIns(4,5)P2 regulates the number
of vesicles ready for fusion.
At the pre-fusion step of Ca2+ -initiated fusion of chromaffin
granules, a protein known as CAPS (calcyphosine)-1 is required.
Antibody neutralization of CAPS-1 blocks DCV, but not synaptic
vesicle exocytosis. Mutation in Caenorhabditis elegans or
Drosophila of CAPS results in loss of neurosecretion of some
neurotransmitters. CAPS binds specifically to PtdIns(4,5)P2
in vitro [65]. In permeabilized PC12 cells, binding of CAPS to the
plasma membrane is dependent on PtdIns(4,5)P2 and increases
the rate of exocytosis, indicating a role for CAPS and
PtdIns(4,5)P2 at a pre-fusion step in regulating secretion [66].
through the binding of its C-terminal domain to PtdIns(4,5)P2
[57] (see Figure 2).
Exocyst function has also been examined in a variety of animal
cells. An early report of secretory PC12 cells identified three
required cytosolic PEPs (priming exocytosis proteins) which
include a PITP (PtdIns transfer protein) and a PIP5K (PtdIns4P
5-kinase) [58]. Inhibition by antibodies to PtdInsP(4,5)P2 or PLC
indicate a role for PtdIns(4,5)P2 in exocytosis in these cells.
Exocyst proteins are recruited to the plasma membrane during
basal–lateral addition of membrane in epithelial cells, addition of
membrane to the leading edge of migrating cells, and in neurites.
In HeLa cells, Exo70 associates with the plasma membrane by
binding to PtdIns(4,5)P2 . It is essential for docking and fusion,
but not transport to the membrane [59]. Binding is selective
for PtdIns(4,5)P2 over negatively charged phosphoserine, but
an affinity for PtdIns(3,4,5)P3 , which is present in much lower
amounts, offers the additional possibility for recruitment by this
lipid.
PtdIns(4,5)P2 is synthesized upon platelet activation. Addition
of PI-PLC to permeabilized platelets stimulates secretion, as
does addition of recombinant PIP (phosphoinositide phosphate)
kinase [60,61]. Antibodies against type II PIP kinase or
excess PtdIns(4,5)P2 blocks secretion [60], suggesting that the
equilibrium between the production of DAG and formation of
PtdIns (4,5)P2 affects secretion in platelets.
Roles of PtdIns3P and PtdIns(3,4,5)P3 in exocytosis. PI3Ks
play a role in many exocytic events. These include class 1A
PI3Ks in transport of GLUT4 glucose transporters to the plasma
membrane induced by insulin, and class II PI3Ks in neurosecretion [67].
In neurons, synaptic vesicles are loaded with neurotransmitter,
assemble near the synapse and dock to the plasma membrane.
There they are primed to become Ca2+ -sensitive. A nerve impulse
triggers voltage-gated Ca2+ channels, resulting in local increases
of Ca2+ and formation of a fusion pore. Following release of the
contents, membrane can be reclaimed by endocytosis or contents
can be recharged with or without undocking [68]. Fusion pores
required for content release may be transient and rapid or the
entire vesicle may incorporate into the plasma membrane in a
slower process. Rab5 and PI3K are required for neurotransmitter
release and Rab5 helps to recruit PI3K. Rab3, and its effectors
rabphilin and RIM1α/2α, participate in a post-docking step. The
SNAREs synaptobrevin, syntaxin 1 and SNAP-25 and several
regulatory proteins, including MUNC 18-1 and synaptotagmin,
also participate in secretion.
Several synaptic vesicle proteins bind to phosphoinositides,
such as synaptotagmin [69,70], CAPS [65], and the MUNC 18-1interacting Mint proteins [71]. Chemical blocking of synthesis of
phosphoinositides in synaptosomes differentially interferes with
neurotransmitter secretion. Release of noradrenaline is sensitive,
but secretion of GABA (γ -aminobutyric acid) and glutamine is
Figure 2
Exocyst components
c The Authors Journal compilation c 2009 Biochemical Society
Phosphatidylinositol metabolism and membrane fusion
insensitive to phosphoinositide depletion [72], possibly indicating
a difference in requirements for various vesicle types.
PTP-MEG2 is a mammalian tyrosine phosphatase that induces
secretory vesicle fusion. It has a Sec14 homology domain that
binds to PtdIns(3,4,5)P3 and is found on secretory vesicles
containing PtdIns(3,4,5)P3 . Mutations interfering with binding to
PtdIns(3,4,5)P3 or depleting PtdIns(3,4,5)P3 in T-cells lower the
capacity to lead to fusion. The co-localization of several different
phosphoinositide kinases on these secretory vesicles would
provide for regulation of PtdIns(3,4,5)P3 levels and secretion [73].
An early study of histamine-releasing mast cells from rat
peritoneum indicated an increase in DAG content within
1 min of stimulation [74]. Addition of labelled arachidonoylDAG revealed rapid breakdown to arachidonate, suggesting
production and turnover of DAG might be involved in mastcell secretion. PtdIns(4,5)P2 content in the plasma membranes
of these cells appears to decrease transiently, owing to PI-PLC
activity. Hydrolysis occurs after Ca2+ signalling in permeabilized
cells, although the direct effects on exocytosis of changes in levels
of DAG or alterations in PtdIns(4,5)P2 are small [75]. Although
PtdIns metabolism undoubtedly contributes to signalling in these
cells, effects of DAG on membrane curvature have also been
postulated [76]. Mast-cell signalling has been reviewed in [76,77].
Roles of DAG in exocytosis. DGKs, which convert DAG into
PtdOH (Figure 1C), have a wide variety of functions and can
interact with PKCs [14]. DGKs in the T-cell response to Fas
ligand binding regulate both inflammatory response and cell
death. Exosomes are small vesicles that accumulate into late
endosomes or multivesicular bodies [78]. An isoform of DGK,
DGKα, localized on the Golgi and late endosomal compartments,
inhibits secretion of exosomes, which subsequently leads to
apoptosis [79].
Munc13 proteins that have DAG-recognition domains function
in synaptic vesicle priming, being required for release rather
than docking [43,44]. PLC-linked receptors in chromaffin cells
modulate MUNC13-1 by producing DAG [80], resulting in
recruitment of vesicles for exocytosis. A model has been presented
for how Munc13 acts at a post-priming step (i) to facilitate fusion
in which the role of DAG is to upregulate Munc13 catalytic
activity by relieving an intramolecular inhibitory interaction of
its C-terminal ‘catalytic’ domain thus promoting priming, and (ii)
to assist fusion through interaction of Munc13 with SNAREs [81].
Endocytosis
Roles of PtdIns3P in endocytosis. During phagocytosis,
neutrophil or macrophage receptors recognize opsonins on the
target and form a phagosome, which subsequently fuses with
endosomes and, eventually, lysosomes. Early endosomes engage
in both heterotypic (with endocytosing vesicles) and homotypic
fusions. A robust in vitro system has led to one of the most complete models of how phosphoinositides could function in fusion
through assembly of SNARE-interacting factors. PtdIns3P plays
a critical role in early endosomes as a recognition marker for
binding the GTPase Rab5 and the EEA1 (early endosomal antigen
1), a protein with a FYVE domain [67]. Several endocytic
effectors required for homotypic fusion contain FYVE domains
such as EEA1, rabankyrin-5 and rabenosyn-5, all of which bind
to Rab5 as well [67,82]. EEA1 interacts with the SNAREs
syntaxin-6 and syntaxin-13, and rabenosyn-5 interacts with a
regulator of the Sec family. Rab5 recruits PI3K to endosomal
membranes and binds to and stimulates phosphoinositide 5and 4-phosphatases, thus integrating various aspects of PtdIns
metabolism [83]. Specific depletion of PtdIns3P in the endosomal
237
compartment disrupts endosomal traffic and morphology [84]
(see Supplementary Movie 1 at http://www.BiochemJ.org/bj/418/
bj4180233add.htm).
PtdIns3P localization using a tandem FYVE probe results in
labelling of early and multivesicular endosomes in mammalian
fibroblasts and endosomes and vacuoles in yeast, suggesting a
conservation of functional compartmentalization [28]. PtdIns3P
is found in microdomains in early endosomes and internal vesicles
of MLVs (multilamellar vesicles), whereas the primary site
for PtdIns(4,5)P2 is in the plasma membrane of the same cells. The
PtdIns3P-binding proteins Rab5, EEA1, SARA (Smad anchor
for receptor activation) and CISK (cytokine-independent survival
kinase) are restricted to the PtdIns3P-containing microdomains,
whereas the PtdIns3P-binding protein Hrs (hepatocyte growth
factor-regulated tyrosine kinase substrate) is found in the domains
associated with clathrin and lowers levels of PtdIns3P [85].
The in vivo application of a regulated dimerization device
(rapalogue-induced depletion) that locally depletes PtdIns3P
and PtdIns(3,5)P2 from Rab5-positive endosomes indicates that
the normal maturation of the endosomal compartment and the
flux of receptors through it are significantly affected by acute
removal of these two inositides [84]. Moreover, the disruption
of endosomal morphology into a tubularized network delays
degradation of EGFR (epidermal growth factor receptor) and
impairs the recycling of transferrin receptor complex (Figure 3).
Therefore PtdIns3P levels regulate receptor efflux towards the
recycling pathway, the degradation pathway and the trans-Golgi
network. These results indicate both a positive and negative role
for PtdIns3P in defining regulatory domains and a role of assembly
of signalling complexes involved in membrane trafficking.
Other Rabs, such as Rab5 and 7, are involved in formation of
phagosomes by macrophages but are not sufficient for fusion,
which appears to require phosphorylation of PtdIns at the 3
position, perhaps in concert with recruitment of EEA1 or rabynosin [86]. PI3K activity is required for phagocytosis, indicating an
involvement of PtdIns3P or PtdIns(3,4,5)P3 . Inhibition of PI3K
with wortmannin prevents fusion of the phagosome with late
endosomes without blocking recruitment of Rab7 and prolongs
binding of early endosomal Rab5 [86]. A class I PI3K is involved
in formation of the phagosome cup and, after sealing, a second
PI3K is needed for maturation as the phagosome acquires late
endosomal markers and eventually EEA1 [67]. These studies
illustrate an important and differential role of PtdIns3P in related
cellular compartments during endocytosis.
Roles of PtdIns(4,5)P2 in endocytosis. PtdIns(4,5)P2 also
functions in endocytosis and phagocytosis. The first steps in
endocytosis require PtdIns(4,5)P2 at the cell surface, possibly
for recruitment of proteins needed for receptor internalization
[87,88]. Depletion of PtdIns(4,5)P2 at the plasma membrane by
rapalogue-induced translocation of a domain of phosphoinositide
5-phosphatase blocks internalization of the transferrin receptor
[89] in this fission event.
Vacuoles and lysosomes
The vacuole of yeast is similar to the lysosome in higher eukaryotic cells, where a wide range of macromolecular components are
broken down to be recycled. Vacuole fusion in yeast has been
recently reviewed [90].
The vacuole is reconstituted by homotypic fusion of vacuolar
fragments following budding. Several in vitro fusion assays are
available, including a classic one exploiting homotypic fusion,
which utilizes vacuoles isolated from two different mutant strains
of yeast lacking either a phosphatase Pho8 or a protease that
c The Authors Journal compilation c 2009 Biochemical Society
238
Figure 3
D. Poccia and B. Larijani
Model for the role of PtdIns3P in early endosome (EE) morphology and function
(A) Schematic representation of the proposed mechanism through which endosomal PtdIns3P levels regulate early endosome morphology and receptor efflux towards the recycling pathway, the
degradation pathway and the trans -Golgi network (TGN). (B) Effect of endosomal PtdIns3P depletion on the functionality of the early endosomes. Adapted from Fili [84a] with permission. C1-M6PR,
mannose 6-phosphate receptor; PM, plasma membrane; RE, recycling endosome; Snx1, sorting nexin 1.
c The Authors Journal compilation c 2009 Biochemical Society
Phosphatidylinositol metabolism and membrane fusion
239
gene results in fragmented vacuoles and its product phospholipase
C1p, the only PLC in yeast, stimulates fusion.
Since PtdIns(4,5)P2 has been detected on the plasma membrane
but not the vacuoles, the location of the PtdIns(4,5)P2 modifying phosphatases Inp54p, which converts PtdIns(4,5)P2
into PtdIns4P, and Sac1p, which converts PtdIns4P into PtdIns,
were used to alter PtdIns(4,5)P2 levels. Depleting PtdIns(4,5)P2
from the plasma membrane with the 5-phosphatase Ins54p
and the 4-phosphatase Sac1p resulted in fusion defects, with
PtdIns(4,5)P2 appearing on fragmented vacuoles [100]. Using
a tandem PH domain probe, PtdIns(4,5)P2 was seen to be lost
from the plasma membrane and appear in vacuolar fragments
when genes for the two phosphatases were deleted. Since the
mutants were rescued by inactivating the PI4K (PtdIns 4-kinase)
Mss4p, that synthesizes PtdIns(4,5)P2 in the plasma membrane,
it was concluded that the PtdIns(4,5)P2 accumulating in vacuoles
came from the plasma membrane and resulted in fragmentation
or fusion defects. It was suggested that PtdIns(4,5)P2 is carried
by endocytosis to the vacuole, where it promotes fusion.
Figure 4
Diagram of docked yeast vacuoles
Black lines represent membranes of yeast vacuoles. Red areas are the vertices where PtdIns(3)P ,
PtdIns(4,5)P 2 and DAG accumulate. Adapted from Fratti et al. [96].
activates Pho8. Introduction of both wild-type proteins into the
same compartment following fusion reconstitutes the phosphatase
activity, conveniently monitored by a colorimetric reaction. As in
other fusion reactions, vacuolar fusion involves tethering/docking,
priming and fusion steps. Sec18, an AAATPase, helps in
dissociation of the SNARE complexes (cis-SNAREs), activating
them to be re-used in subsequent reactions. Tethering is assisted
by HOPS, a large complex of six proteins which reversibly binds
vacuolar membranes. Docking involves the irreversible binding
of unpaired SNAREs with one another. The steps involved in the
actual fusion event are not well understood.
Roles of PtdIns(4,5)P2 in vacuole fusion. Vacuole fusion in vitro
is blocked by antibodies to PtdIns(4,5)P2 and PI-PLC [91].
PtdIns(4,5)P2 is necessary for priming and also a later step
[91]. PtdIns3P helps to recruit the SNARE Vam7p to the membranes [92,93]. Prior to fusion, a flattened region of membrane
contact called the fusion ring forms. At the periphery of this
structure, a vertex ring forms which is enriched in HOPS, SNAREs
and the small GTPase Ypt7p [94,95].
Fusion begins at the vertices, which also accumulate
diacylglycerol, both 3- and 4-phosphoinositides, and ergosterol
[95,96] (Figure 4). The lipids are required for protein enrichment, and SNAREs and actin are needed for PtdIns3P
enrichment. Overlay and liposome assays indicate the purified
HOPS complex binds to inositides, particularly to PtdIns4P,
PtdIns3P, PtdIns(3,5)P2 and PtdIns(4,5)P2 but not PtdIns, PtdSer
(phosphatidylserine) or PtdOH [97]. Binding of HOPS to
liposomes is attenuated by competition with a tandem-FYVE domain specific for PtdIns3P, a PH domain specific for PtdIns4P,
or an ENTH domain specific for PtdIns(4,5)P2 .
A genomic analysis of deletion mutants indicated that PLC
is essential to vacuolar fusion [98]. Vacuole fusion in yeast is
blocked with sequestration of DAG by the C1b domain from a
mammalian PKCβII or PLC inhibitors [99]. Deletion of the PLC1
Roles of PtdIns3P. Autophagosomes form to deliver damaged
organelles, old proteins and protein aggregates to lysosomes
for degradation. In yeast, the PI3K Vps34 is required, and in
mammals autophagy is inhibited by PI3K inhibitors. Class III
PI3Ks promote and class I PI3Ks inhibit autophagy. Most effector
molecules remain to be defined [67]. One of these is a regulatory
subunit of Vps34, Vps15, which is a protein kinase required for
autophagy in Drosophila. The role of PtdIns3P in autophagy has
been recently reviewed [101].
Summary
Phosphoinositides, particularly PtdIns(4,5)P2 , PtdIns3P and
PtdIns(3,4,5)P3 , play a variety of roles in recognition and regulation of membrane fusion events in exocytosis, endocytosis
and vacuole reconstitution. A large number of PtdIns(4,5)P2 interacting proteins have been found, including exocyst proteins
Sec3, Rho, Gic1 and 2, Cdc42, Exo70, and synaptotagmin,
SCAMP2, CAPs and Mint proteins, all functioning at various
steps in exocytosis. Roles for the tyrosine phosphatase PTP-MEG2, which binds to PtdIns3P and MUNC 13-1 and is regulated by
DAG, confirm the importance of phosphoinositide metabolism
in exocytosis. Binding of HOPS to PtdIns(4,5)P2 in vacuole
fusion, as well as requirements for DAG and PI-PLC, indicate
a similar importance of these lipids. A central role for PtdIns3P in
endocytosis has been well documented in the assembly of Rab5,
EEA1, SARA, CISK, rabankyrin-5 and rabenosyn-5. Various
means of experimental alteration of phosphoinositide levels result
in interference or enhancement of the fusion processes. The
ability of phosphoinositides to serve as markers for the binding
of proteins needed for fusion and their high affinity for, and
capacity to regulate the activity of, some fusion proteins underlie
the mechanisms which control membrane fusion in these natural
membrane systems.
However, one major approach still lacking in this field is the
simultaneous determination of localization and the regulation of
phosphoinositide turnover during cellular processes. To quantitatively evaluate in situ turnover and image phosphoinositide
localization in relation to the modulation of the physicochemical
environment of membrane-associated proteins in various subcellular compartments, mass and optical imaging need to be integrated in a single instrument. Such complete data on phosphoinositide composition and distribution would be unique as a
post-genomic technology and would permit the development of
c The Authors Journal compilation c 2009 Biochemical Society
240
D. Poccia and B. Larijani
quantitative predictive models of cellular processes involving
membrane alterations.
EFFECTS OF PHOSPHOINOSITIDES AND THEIR PRODUCTS ON
MEMBRANE STRUCTURE AND FUSION
Curvature and fusion: theoretical considerations
The role of lipids in membrane fusion has received much
theoretical and experimental attention, in particular concerning
the effects of those phospholipids exhibiting negative curvature
[102–104]. Natural membranes and synthetic lipid bilayers
are sealed in an aqueous environment because of the strong
hydrophobic effect of the lipids. In order to fuse, membranes must
be in close apposition (and thus water removed from the point of
contact) and lipids must be rearranged and proteins cleared (at
least at the site of fusion) to allow the two separate bilayers to
merge into one continuous bilayer.
The most widely accepted model of the fusion intermediate
is the fusion stalk (see [104–109]) (Figure 5). A hemifused intermediate state (Figure 5C) in which the two apposed (proximal)
monolayers fuse but the distal monolayers (which face the
aqueous interior) and the contents of the interiors have not yet
mixed creates a fusion stalk (Figure 5D). Subsequent expansion of this region leads to a fusion pore (Figure 5E), connecting
the interior compartments. Each step requires energy and the energetic considerations have been extensively reviewed elsewhere
[104,105,107].
Formation of the hemifused intermediate is accompanied
by extreme curvature of the proximal leaflets. This can be
facilitated by lipids of negative spontaneous curvature (Figure 5).
Spontaneous curvature is the preferred curvature of a monolayer
and depends on the type of phospholipids making up the
monolayer. Lipids with positive curvature (such as lysophosphatidylcholine) favour bending away from the head groups;
those with negative curvature (such as DAG, cholesterol and phosphatidylethanolamine) favour bending towards the head groups.
In water, phospholipids of these two classes would by themselves
form micelles or inverted micelles respectively rather than
bilayers.
Biophysical studies: liposomes and model membranes
Much experimental evidence shows that synthetic bilayers or
liposomes enriched in negatively curved phospholipids exhibit
spontaneous fusion, whereas incorporation of positively curved
lipids hinders fusion [110]. In natural membranes, proteins
inserted into the bilayer might also contribute to membrane
bending in the region of the stalk [111].
Among lipids that provide negative curvature are phosphatidylethanolamine, cholesterol and DAG, the product of PtdIns(4,5)P2
hydrolysis by PI-PLC [103]. There is also evidence to indicate
that phosphoinositides may themselves promote fusion. Solidstate NMR spectroscopic studies of model membranes that have
high levels of C18:0 /C20:4 DAG and polyunsaturated PtdIns indicate
that they are very fluid, even at low temperatures [112]. Charged
phosphoinositides may also reorganize and phase separately in
membranes [113].
In synthetic vesicle fusion induced by Ca2+ or La3+ , PtdIns(4)P,
PtdIns(4,5)P2 and PtdIns(3,4,5)P3 enhanced fusion even at
concentrations as low as 5 mol % [114]. Unphosphorylated PtdIns
inhibited fusion, which was attributed to steric effects of the large
head group preventing proximity of the opposing membranes.
Other negatively charged lipids did not lead to fusion at the
same concentrations. These experiments were performed in 2 mM
c The Authors Journal compilation c 2009 Biochemical Society
Figure 5
Hemifusion pathway of lipid bilayer fusion
Red denotes lipids with negative curvature (DAG, phosphatidylethanolamine), green denotes
lipids with positive curvature (lysophosphatidylcholine). Black dots show membrane
compartment contents. (A) Two bilayers. (B) A change in localized curvature of membrane protrusions. (C) Hemifusion to form a stalk. (D) Expansion of the stalk to form a hemifusion
diaphragm. (E) Formation of the fusion pore and content mixing. Adapted from Chernomordik
and Kozlov [104]. Reprinted by permission from Macmillan Publishers Ltd: Nat. Struct. Mol.
Biol. (Chernomordik, L. V. and Kozlov, M. M. Mechanics of membrane fusion. 15, 675–683),
copyright (2008).
Ca2+ , which could neutralize the charges on the head-group
phosphate groups and bridge them to allow closer proximity of
the membranes, but the phosphoinositides under these conditions
should still contribute steric effects, a problem which would be
circumvented by DAG.
Using large unilamellar vesicles as substrates, the effect of DAG
in vesicle fusion was demonstrated in several studies. LUVs (large
unilamellar vesicles) with 30– 40 mol % PtdIns will aggregate
and fuse following action of a bacterial PI-PLC (which prefers
PtdIns to PtdIns(4,5)P2 as substrate). Inclusion of 5 mol % DAG
facilitates fusion of liposomes containing up to 5–10 % PtdIns
which otherwise only undergo hemifusion [115,116]. Two effects
of DAG may be at work in this system: production of sufficient
Phosphatidylinositol metabolism and membrane fusion
amounts of DAG to facilitate stalk formation and stimulation of
PI-PLC by increased DAG to increase hydrolytic activity. DAG
may also facilitate lamellar to non-lamellar (including hexagonal)
phase transitions [117,118], an effect experimentally reversed by
lysophophatidylcholine of positive curvature.
Plots of LUV inner monolayer lipid mixing rates against
fractions of DAG in the vesicles support the idea that the
concentration of DAG is the critical variable controlling the rate
of fusion [116]. Similar effects of DAG are produced by PC-PLC
activity on phosphatidylcholine-rich vesicles [119,120]. Lipids
like cholesterol enhance enzyme activity in lamellar phases [121].
PLC was suggested to create local regions of high DAG content
[122]. A lag period before liposome fusion correlates with the
amount of DAG initially present and that subsequently produced
by hydrolysis. Fusion initiation begins at about 10 mol % DAG,
even though LUVs containing 10 % DAG are stable in the absence
of enzyme. It was thus suggested that the enzyme might result in
localized DAG formation to initiate fusion [116].
Although one can achieve fusion of protein-free bilayers, fusion
in vivo undoubtedly depends on both proteins and lipids. SNAREs
reconstituted to liposomes can lead to fusion, although at rates
lower than those observed in vivo [123]. Adjustment of the
concentration of components in the reconstitution assays leads
to more physiological rates [124]. It has been suggested that the
ability of SNAREs to lead to fusion of liposomes depends on
the physical state of the bilayers [125]. In standard reconstitution
assays, SNARE liposomes can be heterogeneous and are most
efficient at high protein densities. It was suggested that slow
fusions occur though stalk mechanisms but preferentially in small
vesicles with high curvature stress, whereas in more homogeneous
larger vesicle populations fusion is minimal.
There is considerable evidence that SNAREs and other fusion
regulatory proteins are affected by, and interact with, specific
phospholipids, in particular phosphoinositides. For example,
negatively charged phospholipids, especially phosphoinositides,
decrease lateral mobility of t-SNAREs in planar model membranes [126]. In addition, SNARE-dependent fusion of liposomes
was found to be dependent on phospholipid composition.
When PtdOH was incorporated into vesicles with syntaxin 4
and PtdIns(4,5)P2 incorporated into vesicles containing VAMP2,
fusion of the two membranes was accelerated, suggesting that the
asymmetric distribution of the these lipids facilitated SNAREdependent fusion [127]. It was hypothesized that PtdOH, of
negative curvature, stabilized expansion of the outer leaflet in
formation of the hemifused intermediate, and PtdIns(4,5)P2 , by
virtue of its large head group, favoured the positive curvature of
the inner leaflet leading to fusion.
Thus there is a variety of results suggesting that negatively curved lipids can facilitate fusion by directly contributing to
curved stalk-like intermediates and/or by interacting with
SNAREs.
Selected examples of structural effects on natural membranes
Experimental evidence for the involvement of lipids exhibiting
negative curvature in fusion reactions has been obtained in
several natural membrane systems. A number of phospholipases
(Figure 1D) are capable of producing such lipids directly or
indirectly, which may have structural as well as signalling
consequences for the membrane in which they are generated.
For example, PLA2 (phospholipase A2 ) forms non-esterified
fatty acids and lysophospholipids, the former exhibiting negative
curvature and the latter positive curvature (for a review, see [128]).
PLD (phospholipase D) proteins release choline, leaving PtdOH
in the membrane, which also exhibits spontaneous negative
241
curvature [129]. Phosphatases convert PtdOH into DAG, which
exerts a negative curvature. Eukaryotic PLCs directly catalyse
production of DAG and IP3 from PtdIns(4,5)P2 . PtdIns(4,5)P2
can also activate PLD.
Each phospholipase also yields products which may have other
consequences, so that it is often difficult to disentangle signalling
and structural roles. For example, polyunsaturated fatty acids released by PLA, such as arachadonic and docosahexaenoic acids,
can affect the SNARE syntaxin and allow SNARE interaction with
MUNCs [130]. These fatty acids directly or indirectly through
their conversion into eicosanoids such as prostaglandins can also
influence fusion [131,132]. DAG can directly modulate MUNC
activity to control release of docked vesicles [80]. IP3 produced
by PLCs causes release from internal stores of Ca2+ which at least
at basal levels is required for fusion.
Membrane curvature by fatty acids may explain the action of
snake venoms that stimulate neurotransmitter release at neuromuscular junctions. Their phospholipase A2 activity produces
lysophospholipids and fatty acids which can mimic the effects of
the venom [133,134]. Lysophospholipids associated with positive
curvature, and fatty acids with negative curvature which promotes
hemifusion, have been proposed to occupy different monolayers
and aid in the bending required for fusion during exocytosis
of neurotransmitters, although the exact mechanism of action
is unclear [135]. PLA2 mechanisms may also operate in sperm
acrosomal exocytosis [136], endosomal fusion, brain neuron
exocytosis [137] and Golgi–ER (endoplasmic reticulum) traffic
[128]. PtdOH, produced by PLD which is involved in chromaffin
cell exocytosis [138], may also directly affect fusion through
structural alterations of the bilayer [139].
For the purposes of the present review, several examples are
discussed below arguing for a role of DAG in membrane fusion.
A structural role for DAG in membrane fusion has received much
less attention than its signalling role and thus many of the studies
are not yet definitive. Additionally, DAG accumulation at the
site of fusion is difficult to study because of its transient nature
and detection problems against the background of other lipids of
the membranes not directly involved in the fusion stalk. Some
progress using cell-free extracts suggest that DAG may have a
direct role in biological membrane fusion.
Plasma-membrane fusion
Early experiments indicated a structural role for DAG in plasmamembrane fusion. Elevated levels of diacylglycerol and PtdOH
were correlated with crenation of red blood cells depleted in
ATP or Ca2+ -triggered, which leads to membrane budding and
fusions creating microvesicles [140]. Loss of PtdIns(4,5)P2 and/or
conversion into DAG were related to shrinkage of the inner monolayer, which could account for the crenation [141].
The Ca2+ -initiated fusion of the plasma membranes of
myoblasts to form myotubes is accompanied by breakdown
of PtdIns(4,5)P2 and DAG and PtdOH synthesis [142,143].
Myoblasts have relatively high polyphosphoinositide contents for
vertebrate cells, which was suggested to lead to plasma-membrane
fusion [142].
Secretion
An early study of histamine-releasing mast cells from rat peritoneum indicated an increase in DAG content within 1 min of
stimulation [74]. Addition of labelled arachidonoyl-DAG revealed
rapid breakdown to arachidonate, suggesting production and
turnover of DAG might be involved in mast cell secretion. Mast
cell signalling has been reviewed in [76,77]. PtdIns(4,5)P2 content
c The Authors Journal compilation c 2009 Biochemical Society
242
D. Poccia and B. Larijani
in the plasma membranes of these cells appears to decrease
transiently due to PI-PLC activity. Hydrolysis occurs after Ca2+
signalling in permeabilized cells, although the direct effects of
DAG or alterations in PtdIns(4,5)P2 levels on exocytosis were
small [75]. Athough PtdIns metabolism undoubtedly contributes
to signalling in these cells, effects of DAG on membrane curvature
has also been suggested [76].
In an early attempt at an in vitro fusion system, polylysineinduced fusion of ghosts of chromaffin granules and plasma membrane targets was monitored by a fluorescence dequenching assay
[144]. Addition of diacylglycerol to the plasma membranes or
model membranes facilitated Ca2+ -mediated fusion.
An important secretory vesicle model is derived from sea urchin
eggs, whose cortical granules contribute to the construction of the
extracellular matrix after fertilization and aid in the prevention of
polyspermic fertilization [145–148]. These granules are docked
and await a Ca2+ wave to trigger exocytosis at fertilization.
Evidence from FRAP (fluorescence recovery after photobleaching) measurements suggests that the cortical granules may be
already in a hemi-fused state with the plasma membrane [149],
similar to that reported for certain synaptic vesicles [150]. Rab3
and SNAREs are associated with these membranes in an immobile
fraction that awaits Ca2+ activation to lead to full fusion.
The lipid composition of the cortical granule membranes and
the plasma membrane are not known in detail, but analyses suggest the granules are enriched in cholesterol and arachadonic
acid [151]. The target plasma membrane is rich in PtdIns (25 %),
phosphatidylethanolamine (60 %) and cholesterol (the latter
approximately equimolar with total phospholipids) [152]. The
role of membrane curvature has been addressed by Churchward
et al. [103] in a cortical granule-plasma membrane in vitro
fusion system. With the caveat that Ca2+ concentrations triggering fusion are higher than those observed in vivo in these
experiments [1], results from these important studies indicate
fusion roles for lipids of negative curvature, including DAG and
cholesterol. Depletion of cholesterol blocks fusion, which can
be restored by DAG and phosphatidylethanolamine [103,153].
In contrast, lysophosphatidylcholine inhibits fusion. Neither the
kinetics of fusion nor its Ca2+ sensitivity, only the ability to
fuse, was affected by the amount of lipids of negative curvature.
Importantly, comparison of different lipids of spontaneous
negative curvature indicated both a quantitative effect for each
lipid and a critical overall curvature, suggesting a threshold of
negative curvature for fusion.
Additionally, although DAG could restore fusion in cholesteroldepleted granules, PtdOH could not, indicating that DAG
could be a central contributor to fusion between non-fusogenic
PtdIns(4,5)P2 and PtdOH, and so the role of PtdOH in vivo would
be upstream of the fusion event. Another important result is that
cholesterol seems to be concentrated in the vesicle population,
suggesting a possible asymmetry in the degree of curvature of the
interacting membranes. The authors suggested that total spontaneous curvature involving cholesterol and other molecules may
be important in this and other systems. These results support a
stalk model of fusion involving contacting monolayers from each
membrane [103].
Another useful in vitro system is derived from alveolar cells.
Alveolar type II cells secrete lung surfactant by fusion of
lamellar bodies with the plasma membrane. The protein annexin
7 promotes Ca2+ -mediated fusion. Vesicles incubated with
DAG or PtdIns(4,5)P2 incorporate these lipids, which facilitates
fusion. Pretreatment of lamellar bodies with PLC to generate
DAG also promotes fusion [154]. These experiments indicate a
potential role for DAG content in annexin 7-mediated membrane
fusion.
c The Authors Journal compilation c 2009 Biochemical Society
Phagocytosis
A striking localization of DAG is seen in phagosome formation in
macrophages. Using fluorescent lipid probes and antibodies, focal
accumulations of PtdIns(4,5)P2 , type 1α PIP kinase and PLC were
detected in the phagosome cup followed by loss of PtdIns(4,5)P2
and formation of DAG [155]. Blocking DAG synthesis blocked
phagocytosis. The authors related these phenomena to regulation
of actin assembly and cytoskeletal remodelling that accompany
this fission event, but it is possible that DAG, which accumulates
maximally just after completion of sealing of the newly formed
phagosome, may have a structural role in the bending required
for its formation. The experiments strikingly illustrate the local
accumulation of phospholipids and modifying enzymes possible
within natural membranes which could equally well apply to
fusion intermediates.
Cytokinesis
Cytokinesis is accompanied by addition of membrane to the cell
surface by fusion of cytoplasmic vesicles. There is evidence for
participation of both Golgi-derived vesicles and recycling endosomal membranes contributing to the cleavage furrow [156].
The SNARE syntaxin is required for cell division in a
variety of species. PtdIns(4,5)P2 and PtdIns4P 5-kinases have
been localized to the cleavage furrow of mammalian cells,
which are also sites of accumulation of the negative curvatureinducing lipid phosphatidylethanolamine [157]. Accumulation of
PtdIns(4,5)P2 by overexpressing a PtdIns4P 5-kinase-deficient
mutant or microinjection of anti-PtdIns(4,5)P2 block cytokinesis.
These experiments again demonstrate domains of altered lipid
composition which could affect membrane curvature directly or
through interaction with the underlying actin cytoskeleton. In
addition, nuclear morphology was altered in these cells injected
with anti-PtdIns(4,5)P2 , indicating a potential additional role in
nuclear structure for this lipid.
Nuclear membrane assembly
The most extensive study of the role of phosphoinositides and
DAG on nuclear membrane fusion reactions has been performed
with a cell-free sea urchin egg preparation [158] that assembles
nuclear envelopes on exogenously added sperm nuclei [159,160].
Two membrane fractions fuse to form the envelope: MV1, derived from vesicles primarily in the egg cortex, and MV2, derived
largely from ER experimentally vesiculated by homogenization
in forming the cell-free extract [161]. MV1 has an extraordinary
phosphoinositide composition: 60 mol % total phospholipid as
PtdIns derivatives, of which two-thirds are polyphosphoinositides
[12,162]. Additionally, MV1 contains PI-PLCγ at > 100-fold
greater concentration than MV2 [12]. PI-PLCγ is activated early
in the GTP-initiated fusion reactions, leading to nuclear envelope
formation by tyrosine phosphorylation [12] catalysed by a Src
family kinase [163].
Exogenous PI-PLC initiates nuclear envelope formation
without GTP hydrolysis or PI3K activity, suggesting a downstream role [24]. A direct role for DAG rather than IP3 -mediated
Ca2+ release is substantiated by a series of experiments. Bacterial
PI-PLC, which hydrolyses PtdIns not PtdIns(4,5)P2 and does
not produce IP3 , results in nuclear envelope formation [24].
Moreover, pre-incubation of MV1 or liposomes mimicking MV1
with eukaryotic PI-PLCγ results in nuclear envelope formation
[164]. Finally, liposomes containing large amounts of DAG
directly initiate fusion and nuclear envelope formation without
PLC activity. These experiments led to the hypothesis that the
Phosphatidylinositol metabolism and membrane fusion
role of phosophoinositide-rich MV1 is to promote fusion of ERderived membranes to form the nuclear envelope by providing
high levels of DAG in at least one of the partner membranes.
As with cortical granule fusion, high levels of negatively curved
lipids may be necessary predominantly in only one of the two
partner membranes fusing.
Formation of the nuclear envelope may share mechanisms
with other cellular fusions. For example, in yeast, three mutated
genes which affect early steps in secretion at the level of the ER
also result in altered nuclear envelope morphology [165]. The
phenotype of syntaxin-4 deletion mutants in C. elegans early
embryos also suggest similarities in nuclear envelope formation
and cytokinesis membrane fusion events [166].
Summary
A large number of biophysical studies support the notion that
lipids associated with negative curvature facilitate synthetic membrane fusion. Among these lipids are phosphatidylethanolamine,
fatty acids and, importantly for purposes of this review, DAG.
Additionally, cholesterol can potentiate their effects. Fewer
studies from natural membranes are available, but several systems
demonstrate effects of DAG content positively correlated with
fusibility. Several cell-free systems exhibiting fusion have been
useful in testing the role of DAG in natural membrane fusion,
including chromaffin granule plasma membrane, cortical granule
plasma membrane, alveolar lamellar body plasma membrane, and
nuclear envelope precursor vesicle ER membrane models. These
have permitted determination of the effects of direct alteration of
phospholipid content on fusion, are generally in agreement with
studies of protein-free liposome fusion and are consistent
with the stalk model. Additionally, the effects of membrane
fluidity of phosphoinositides with polyunsaturated fatty acid
chains on fusion also need to be taken into account in addition to
negative curvature.
243
favouring bending may contribute to fusion, although it is not
clear how such regions would be maintained, given the rapidity of
diffusion of lipids in the bilayer [167]. Improved ability to isolate
and analyse the lipid composition of fusion partner membranes,
development of refined and varied cell-free assay systems and
manipulation of lipid content in specific compartments in vivo
should lead to greater understanding of how such lipids contribute
to the stalk.
In the case of acute depletion of PtdIns3P from early endosomes, extensive endosomal tubularization is observed, which
correlates with the trapping of transferrin receptors within the
tubularized network. A possible explanation is that PtdIns3P
depletion impairs fission of naturally occurring endosomal
tubules, which are the exit platform towards the recycling
pathway [84]. More specifically, absence of the negatively charged
PtdIns3P would be expected to suppress formation of curved
surfaces required for budding. Concomitantly, an excess of generated PtdIns, a lipid that favours membrane fluidity [112], would be
expected to enhance the plasticity of the membrane, thus allowing
its extensive elongation.
Various examples reviewed here, such as localized depletion
of endosomal PtdIns3P, vacuolar DAG dependent-fusion or the
role of DAG in nuclear envelope formation demonstrate that
localized variations of phosphoinositides and their products can
lead to structural alterations in the membrane. These variations in
membrane dynamics facilitate fusion, which is a central element
in many cellular processes. We can therefore suggest that the
phosphoinositides are not only regulators, but also modulators, of
membrane structure and dynamics.
ACKNOWLEDGEMENTS
We thank Veronique Calleja, Richard D. Byrne and Tina Hobday for designing the Figures
and Richard D. Byrne for critically reading the review before its submission.
PERSPECTIVES
FUNDING
The family of phosphoinositides has multiple roles in signalling
affecting a wide variety of cell processes. We have reviewed here
the roles of phosphoinositides and their metabolic products on
membrane fusion, a process that is involved in a wide range of
eukaryotic cellular functions.
Much attention has been paid to the roles of these lipids as
regulatory molecules and as signposts for binding of specific
proteins with roles in fusion to appropriate membranes. In particular, PtdInsP(4,5)P2 , PtdIns3P, PtdIns(3,4,5)P3 and DAG fulfil
such roles, many of which are in prefusion steps such as docking
and priming.
In most cases the paradigm for the fusion event depends on
SNAREs as the major structural machinery, with lipids regulating
directly or indirectly the function of the SNARE complex.
However, it is still not clear to what extent SNARES are sufficient
to effect fusion.
An alternate, but not mutually exclusive, way of understanding
the roles of such lipids, particularly in the generation of DAG from
PtdInsP(4,5)P2 by PI-PLC, is to emphasize the role of DAG as a
lipid characterized by both its negative curvature and an inducer of
non-lamellar structures, which facilitates fusion in both synthetic
and natural membrane systems. Important implications of the few
experiments performed so far on this topic include the possibility
that elevated levels of lipids of negative curvature may only be
needed in one of the two fusion partner membranes [12,103],
and that potentially fusigenic membrane vesicle populations may
contribute for some fusions [12]. An alternative to the latter
idea is that highly localized regions of altered lipid composition
This work was supported by CR-UK (Cancer Research U.K.) core funding and an Amherst
College Faculty Research Award from the H. Axel Schupf ‘57 Fund for Intellectual Life (to
D. L. P.).
REFERENCES
1 Leguia, M. and Wessel, G. (2007) The many faces of egg activation at fertilization.
Signal Transduction 7, 118–141
2 Szule, J. A. and Coorssen, J. R. (2003) Revisiting the role of SNAREs in exocytosis and
membrane fusion. Biochim. Biophys. Acta 1641, 121–135
3 Albillos, A., Dernick, G., Horstmann, H., Almers, W., Alvarez de Toledo, G. and
Lindau, M. (1997) The exocytotic event in chromaffin cells revealed by patch
amperometry. Nature 389, 509–512
4 Gandhi, S. P. and Stevens, C. F. (2003) Three modes of synaptic vesicular recycling
revealed by single-vesicle imaging. Nature 423, 607–613
5 Taraska, J. W. and Almers, W. (2004) Bilayers merge even when exocytosis is transient.
Proc. Natl. Acad. Sci. U.S.A. 101, 8780–8785
6 LoGiudice, L. and Matthews, G. (2006) The synaptic vesicle cycle: is kissing overrated?
Neuron 51, 676–677
7 Toonen, R. F., Kochubey, O., de Wit, H., Gulyas-Kovacs, A., Konijnenburg, B., Sorensen,
J. B., Klingauf, J. and Verhage, M. (2006) Dissecting docking and tethering of secretory
vesicles at the target membrane. EMBO J. 25, 3725–3737
8 Staal, R. G., Mosharov, E. V. and Sulzer, D. (2004) Dopamine neurons release transmitter
via a flickering fusion pore. Nat. Neurosci. 7, 341–346
9 Tamm, L. K., Crane, J. and Kiessling, V. (2003) Membrane fusion: a structural
perspective on the interplay of lipids and proteins. Curr. Opin. Struct. Biol. 13, 453–466
10 Vanhaesebroeck, B., Leevers, S. J., Ahmadi, K., Timms, J., Katso, R., Driscoll, P. C.,
Woscholski, R., Parker, P. J. and Waterfield, M. D. (2001) Synthesis and function of
3-phosphorylated inositol lipids. Annu. Rev. Biochem. 70, 535–602
11 Fruman, D. A., Meyers, R. E. and Cantley, L. C. (1998) Phosphoinositide kinases.
Annu. Rev. Biochem. 67, 481–507
c The Authors Journal compilation c 2009 Biochemical Society
244
D. Poccia and B. Larijani
12 Byrne, R. D., Garnier-Lhomme, M., Han, K., Dowicki, M., Michael, N., Totty, N.,
Zhendre, V., Cho, A., Pettitt, T. R., Wakelam, M. J. et al. (2007) PLCγ is enriched on
poly-phosphoinositide-rich vesicles to control nuclear envelope assembly.
Cell. Signalling 19, 913–922
13 Merida, I., Avila-Flores, A. and Merino, E. (2008) Diacylglycerol kinases: at the hub of
cell signalling. Biochem. J. 409, 1–18
14 Sakane, F., Imai, S., Kai, M., Yasuda, S. and Kanoh, H. (2007) Diacylglycerol kinases:
why so many of them? Biochim. Biophys. Acta 1771, 793–806
15 Hurley, J. H. and Misra, S. (2000) Signaling and subcellular targeting by
membrane-binding domains. Annu. Rev. Biophys. Biomol. Struct. 29, 49–79
16 McLaughlin, S., Wang, J., Gambhir, A. and Murray, D. (2002) PIP2 and proteins:
interactions, organization, and information flow. Annu. Rev. Biophys. Biomol. Struct. 31,
151–175
17 Liao, H., Ellena, J., Liu, L., Szabo, G., Cafiso, D. and Castle, D. (2007) Secretory carrier
membrane protein SCAMP2 and phosphatidylinositol 4,5-bisphosphate interactions in
the regulation of dense core vesicle exocytosis. Biochemistry 46, 10909–10920
18 McLaughlin, S. and Murray, D. (2005) Plasma membrane phosphoinositide organization
by protein electrostatics. Nature 438, 605–611
19 Varnai, P. and Balla, T. (2006) Live cell imaging of phosphoinositide dynamics with
fluorescent protein domains. Biochim. Biophys. Acta 1761, 957–967
20 Rusten, T. E. and Stenmark, H. (2006) Analyzing phosphoinositides and their interacting
proteins. Nat. Methods 3, 251–258
21 Lemmon, M. A. (2003) Phosphoinositide recognition domains. Traffic 4, 201–213
22 Wenk, M. R., Lucast, L., Di Paolo, G., Romanelli, A. J., Suchy, S. F., Nussbaum, R. L.,
Cline, G. W., Shulman, G. I., McMurray, W. and De Camilli, P. (2003) Phosphoinositide
profiling in complex lipid mixtures using electrospray ionization mass spectrometry.
Nat. Biotechnol. 21, 813–817
23 Wakelam, M. J., Pettitt, T. R. and Postle, A. D. (2007) Lipidomic analysis of signaling
pathways. Methods Enzymol. 432, 233–246
24 Byrne, R. D., Barona, T. M., Garnier, M., Koster, G., Katan, M., Poccia, D. L. and
Larijani, B. (2005) Nuclear envelope assembly is promoted by phosphoinositidespecific phospholipase C with selective recruitment of phosphatidylinositol-enriched
membranes. Biochem. J. 387, 393–400
25 Balla, T. and Varnai, P. (2002) Visualizing cellular phosphoinositide pools with
GFP-fused protein-modules. Sci. STKE 2002, PL3
26 Downes, C. P., Gray, A. and Lucocq, J. M. (2005) Probing phosphoinositide functions in
signaling and membrane trafficking. Trends Cell Biol. 15, 259–268
27 Balla, T., Bondeva, T. and Varnai, P. (2000) How accurately can we image inositol lipids
in living cells? Trends Pharmacol. Sci. 21, 238–241
28 Gillooly, D. J., Morrow, I. C., Lindsay, M., Gould, R., Bryant, N. J., Gaullier, J. M.,
Parton, R. G. and Stenmark, H. (2000) Localization of phosphatidylinositol 3-phosphate
in yeast and mammalian cells. EMBO J. 19, 4577–4588
29 Gozani, O., Karuman, P., Jones, D. R., Ivanov, D., Cha, J., Lugovskoy, A. A., Baird, C. L.,
Zhu, H., Field, S. J., Lessnick, S. L. et al. (2003) The PHD finger of the chromatinassociated protein ING2 functions as a nuclear phosphoinositide receptor. Cell 114,
99–111
30 Munro, S. (2004) Organelle identity and the organization of membrane traffic.
Nat. Cell Biol. 6, 469–472
31 De Matteis, M. A. and Godi, A. (2004) PI-loting membrane traffic. Nature Cell Biol. 6,
487–492
32 Behnia, R. and Munro, S. (2005) Organelle identity and the signposts for membrane
traffic. Nature 438, 597–604
33 Grosshans, B. L., Ortiz, D. and Novick, P. (2006) Rabs and their effectors: achieving
specificity in membrane traffic. Proc. Natl. Acad. Sci. U.S.A. 103, 11821–11827
34 Pfeffer, S. R. (2007) Unsolved mysteries in membrane traffic. Annu. Rev. Biochem. 76,
629–645
35 Sollner, T., Bennett, M. K., Whiteheart, S. W., Scheller, R. H. and Rothman, J. E. (1993)
A protein assembly-disassembly pathway in vitro that may correspond to sequential
steps of synaptic vesicle docking, activation, and fusion. Cell 75, 409–418
36 Banfield, D. K. (2001) SNARE complexes–is there sufficient complexity for vesicle
targeting specificity? Trends Biochem. Sci. 26, 67–68
37 Lang, T. (2007) SNARE proteins and ‘membrane rafts’. J. Physiol. 585, 693–698
38 Lipka, V., Kwon, C. and Panstruga, R. (2007) SNARE-ware: the role of SNARE-domain
proteins in plant biology. Annu. Rev. Cell Dev. Biol. 23, 147–174
39 Sugita, S. (2008) Mechanisms of exocytosis. Acta Physiol. 192, 185–193
40 Verhage, M. and Toonen, R. F. (2007) Regulated exocytosis: merging ideas on fusing
membranes. Curr. Opin. Cell Biol. 19, 402–408
41 Jahn, R. and Scheller, R. H. (2006) SNAREs–engines for membrane fusion. Nat. Rev. 7,
631–643
42 Jackson, M. B. and Chapman, E. R. (2006) Fusion pores and fusion machines in
Ca2+ -triggered exocytosis. Annu. Rev. Biophys. Biomol. Struct. 35, 135–160
c The Authors Journal compilation c 2009 Biochemical Society
43 Latham, C. F. and Meunier, F. A. (2007) Munc18a: Munc-y business in mediating
exocytosis. Int. J. Biochem. Cell Biol. 39, 1576–1581
44 Toonen, R. F. and Verhage, M. (2007) Munc18-1 in secretion: lonely Munc joins SNARE
team and takes control. Trends Neurosci. 30, 564–572
45 Rizo, J., Chen, X. and Arac, D. (2006) Unraveling the mechanisms of synaptotagmin and
SNARE function in neurotransmitter release. Trends Cell Biol. 16, 339–350
46 Zimmerberg, J., Akimov, S. A. and Frolov, V. (2006) Synaptotagmin: fusogenic role for
calcium sensor? Nat. Struct. Mol. Biol. 13, 301–303
47 Brose, N. (2008) For better or for worse: complexins regulate SNARE function and
vesicle fusion. Traffic 9, 1403–1413
48 Melia, Jr, T. J. (2007) Putting the clamps on membrane fusion: how complexin sets the
stage for calcium-mediated exocytosis. FEBS Lett. 581, 2131–2139
49 James, D. J., Khodthong, C., Kowalchyk, J. A. and Martin, T. F. (2008)
Phosphatidylinositol 4,5-bisphosphate regulates SNARE-dependent membrane fusion.
J. Cell Biol. 182, 355–366
50 Lam, A. D., Tryoen-Toth, P., Tsai, B., Vitale, N. and Stuenkel, E. L. (2008) SNAREcatalyzed fusion events are regulated by syntaxin1A-lipid interactions. Mol. Biol. Cell
19, 485–497
51 Arac, D., Chen, X., Khant, H. A., Ubach, J., Ludtke, S. J., Kikkawa, M., Johnson, A. E.,
Chiu, W., Sudhof, T. C. and Rizo, J. (2006) Close membrane-membrane proximity
induced by Ca2+ -dependent multivalent binding of synaptotagmin-1 to phospholipids.
Nat. Struct. Mol. Biol. 13, 209–217
52 Bai, J., Tucker, W. C. and Chapman, E. R. (2004) PIP2 increases the speed of response of
synaptotagmin and steers its membrane-penetration activity toward the plasma
membrane. Nat. Struct. Mol. Biol. 11, 36–44
53 Munson, M. and Novick, P. (2006) The exocyst defrocked, a framework of rods revealed.
Nat. Struct. Mol. Biol. 13, 577–581
54 Zhang, X., Orlando, K., He, B., Xi, F., Zhang, J., Zajac, A. and Guo, W. (2008) Membrane
association and functional regulation of Sec3 by phospholipids and Cdc42. J. Cell Biol.
180, 145–158
55 Brown, J. L., Jaquenoud, M., Gulli, M. P., Chant, J. and Peter, M. (1997) Novel
Cdc42-binding proteins Gic1 and Gic2 control cell polarity in yeast. Genes Dev. 11,
2972–2982
56 Orlando, K., Zhang, J., Zhang, X., Yue, P., Chiang, T., Bi, E. and Guo, W. (2008)
Regulation of GIC2 localization and function by phosphatidylinositol 4,5-bisphosphate
during the establishment of cell polarity in budding yeast. J. Biol. Chem. 283,
14205–14212
57 He, B., Xi, F., Zhang, X., Zhang, J. and Guo, W. (2007) Exo70 interacts with
phospholipids and mediates the targeting of the exocyst to the plasma membrane.
EMBO J. 26, 4053–4065
58 Hay, J. C., Fisette, P. L., Jenkins, G. H., Fukami, K., Takenawa, T., Anderson, R. A. and
Martin, T. F. (1995) ATP-dependent inositide phosphorylation required for
Ca2+ -activated secretion. Nature 374, 173–177
59 Liu, J., Zuo, X., Yue, P. and Guo, W. (2007) Phosphatidylinositol 4,5-bisphosphate
mediates the targeting of the exocyst to the plasma membrane for exocytosis in
mammalian cells. Mol. Biol. Cell 18, 4483–4492
60 Rozenvayn, N. and Flaumenhaft, R. (2003) Protein kinase C mediates translocation of
type II phosphatidylinositol 5-phosphate 4-kinase required for platelet α-granule
secretion. J. Biol. Chem. 278, 8126–8134
61 Rozenvayn, N. and Flaumenhaft, R. (2001) Phosphatidylinositol 4,5-bisphosphate
mediates Ca2+ -induced platelet α-granule secretion: evidence for type II
phosphatidylinositol 5-phosphate 4-kinase function. J. Biol. Chem. 276, 22410–22419
62 Holz, R. W., Hlubek, M. D., Sorensen, S. D., Fisher, S. K., Balla, T., Ozaki, S., Prestwich,
G. D., Stuenkel, E. L. and Bittner, M. A. (2000) A pleckstrin homology domain specific
for phosphatidylinositol 4, 5-bisphosphate (PtdIns-4,5-P2 ) and fused to green
fluorescent protein identifies plasma membrane PtdIns-4,5-P2 as being important in
exocytosis. J. Biol. Chem. 275, 17878–17885
63 Aikawa, Y. and Martin, T. F. (2003) ARF6 regulates a plasma membrane pool of
phosphatidylinositol(4,5)bisphosphate required for regulated exocytosis. J. Cell Biol.
162, 647–659
64 Milosevic, I., Sorensen, J. B., Lang, T., Krauss, M., Nagy, G., Haucke, V., Jahn, R. and
Neher, E. (2005) Plasmalemmal phosphatidylinositol-4,5-bisphosphate level regulates
the releasable vesicle pool size in chromaffin cells. J. Neurosci. 25, 2557–2565
65 Loyet, K. M., Kowalchyk, J. A., Chaudhary, A., Chen, J., Prestwich, G. D. and Martin,
T. F. (1998) Specific binding of phosphatidylinositol 4,5-bisphosphate to
calcium-dependent activator protein for secretion (CAPS), a potential phosphoinositide
effector protein for regulated exocytosis. J. Biol. Chem. 273, 8337–8343
66 Grishanin, R. N., Kowalchyk, J. A., Klenchin, V. A., Ann, K., Earles, C. A., Chapman,
E. R., Gerona, R. R. and Martin, T. F. (2004) CAPS acts at a prefusion step in dense-core
vesicle exocytosis as a PIP2 binding protein. Neuron 43, 551–562
67 Lindmo, K. and Stenmark, H. (2006) Regulation of membrane traffic by
phosphoinositide 3-kinases. J. Cell Sci. 119, 605–614
Phosphatidylinositol metabolism and membrane fusion
68 Sudhof, T. C. (2004) The synaptic vesicle cycle. Annu. Rev. Neurosci. 27, 509–547
69 Schiavo, G., Gu, Q. M., Prestwich, G. D., Sollner, T. H. and Rothman, J. E. (1996)
Calcium-dependent switching of the specificity of phosphoinositide binding to
synaptotagmin. Proc. Natl. Acad. Sci. U.S.A. 93, 13327–13332
70 Mehrotra, B., Elliott, J. T., Chen, J., Olszewski, J. D., Profit, A. A., Chaudhary, A.,
Fukuda, M., Mikoshiba, K. and Prestwich, G. D. (1997) Selective photoaffinity labeling
of the inositol polyphosphate binding C2B domains of synaptotagmins. J. Biol. Chem.
272, 4237–4244
71 Okamoto, M. and Sudhof, T. C. (1997) Mints, Munc18-interacting proteins in synaptic
vesicle exocytosis. J. Biol. Chem. 272, 31459–31464
72 Khvotchev, M. and Sudhof, T. C. (1998) Newly synthesized phosphatidylinositol
phosphates are required for synaptic norepinephrine but not glutamate or
γ -aminobutyric acid (GABA) release. J. Biol. Chem. 273, 21451–21454
73 Huynh, H., Wang, X., Li, W., Bottini, N., Williams, S., Nika, K., Ishihara, H., Godzik, A.
and Mustelin, T. (2003) Homotypic secretory vesicle fusion induced by the protein
tyrosine phosphatase MEG2 depends on polyphosphoinositides in T cells. J. Immunol.
171, 6661–6671
74 Kennerly, D. A., Sullivan, T. J., Sylwester, P. and Parker, C. W. (1979) Diacylglycerol
metabolism in mast cells: a potential role in membrane fusion and arachidonic acid
release. J. Exp. Med. 150, 1039–1044
75 Hammond, G. R., Dove, S. K., Nicol, A., Pinxteren, J. A., Zicha, D. and Schiavo, G.
(2006) Elimination of plasma membrane phosphatidylinositol (4,5)-bisphosphate is
required for exocytosis from mast cells. J. Cell Sci. 119, 2084–2094
76 Byrne, R. D. and Larijani, B. (2006) The role of phosphoinositides in mast cell
signalling. Signal Transduction 6, 81–91
77 Sagi-Eisenberg, R. (2007) The mast cell: where endocytosis and regulated exocytosis
meet. Immunol. Rev. 217, 292–303
78 Alonso, R., Mazzeo, C., Merida, I. and Izquierdo, M. (2007) A new role of diacylglycerol
kinase alpha on the secretion of lethal exosomes bearing Fas ligand during
activation-induced cell death of T lymphocytes. Biochimie 89, 213–221
79 Alonso, R., Rodriguez, M. C., Pindado, J., Merino, E., Merida, I. and Izquierdo, M.
(2005) Diacylglycerol kinase alpha regulates the secretion of lethal exosomes bearing
Fas ligand during activation-induced cell death of T lymphocytes. J. Biol. Chem. 280,
28439–28450
80 Bauer, C. S., Woolley, R. J., Teschemacher, A. G. and Seward, E. P. (2007) Potentiation of
exocytosis by phospholipase C-coupled G-protein-coupled receptors requires the
priming protein Munc13-1. J. Neurosci. 27, 212–219
81 Basu, J., Betz, A., Brose, N. and Rosenmund, C. (2007) Munc13-1 C1 domain activation
lowers the energy barrier for synaptic vesicle fusion. J. Neurosci. 27, 1200–1210
82 Ridley, S. H., Ktistakis, N., Davidson, K., Anderson, K. E., Manifava, M., Ellson, C. D.,
Lipp, P., Bootman, M., Coadwell, J., Nazarian, A. et al. (2001) FENS-1 and DFCP1 are
FYVE domain-containing proteins with distinct functions in the endosomal and Golgi
compartments. J. Cell Sci. 114, 3991–4000
83 Shin, H. W., Hayashi, M., Christoforidis, S., Lacas-Gervais, S., Hoepfner, S., Wenk,
M. R., Modregger, J., Uttenweiler-Joseph, S., Wilm, M., Nystuen, A. et al. (2005) An
enzymatic cascade of Rab5 effectors regulates phosphoinositide turnover in the
endocytic pathway. J. Cell Biol. 170, 607–618
84 Fili, N., Calleja, V., Woscholski, R., Parker, P. J. and Larijani, B. (2006) Compartmental
signal modulation: endosomal phosphatidylinositol 3-phosphate controls endosome
morphology and selective cargo sorting. Proc. Natl. Acad. Sci. U.S.A. 103,
15473–15478
84a Fili, N. (2006) Investigation of the role of phosphoinositides in membrane dynamics – a
novel approach. Ph.D. Thesis, Imperial College, University of London, U.K.
85 Gillooly, D. J., Raiborg, C. and Stenmark, H. (2003) Phosphatidylinositol 3-phosphate is
found in microdomains of early endosomes. Histochem. Cell Biol. 120, 445–453
86 Vieira, O. V., Bucci, C., Harrison, R. E., Trimble, W. S., Lanzetti, L., Gruenberg, J.,
Schreiber, A. D., Stahl, P. D. and Grinstein, S. (2003) Modulation of Rab5 and Rab7
recruitment to phagosomes by phosphatidylinositol 3-kinase. Mol. Cell. Biol. 23,
2501–2514
87 Huang, S., Lifshitz, L., Patki-Kamath, V., Tuft, R., Fogarty, K. and Czech, M. P. (2004)
Phosphatidylinositol-4,5-bisphosphate-rich plasma membrane patches organize active
zones of endocytosis and ruffling in cultured adipocytes. Mol. Cell. Biol. 24, 9102–9123
88 Itoh, T., Koshiba, S., Kigawa, T., Kikuchi, A., Yokoyama, S. and Takenawa, T. (2001) Role
of the ENTH domain in phosphatidylinositol-4,5-bisphosphate binding and endocytosis.
Science 291, 1047–1051
89 Varnai, P., Thyagarajan, B., Rohacs, T. and Balla, T. (2006) Rapidly inducible changes in
phosphatidylinositol 4,5-bisphosphate levels influence multiple regulatory functions of
the lipid in intact living cells. J. Cell Biol. 175, 377–382
90 Ostrowicz, C. W., Meiringer, C. T. and Ungermann, C. (2008) Yeast vacuole fusion:
a model system for eukaryotic endomembrane dynamics. Autophagy 4, 5–19
245
91 Mayer, A., Scheglmann, D., Dove, S., Glatz, A., Wickner, W. and Haas, A. (2000)
Phosphatidylinositol 4,5-bisphosphate regulates two steps of homotypic vacuole fusion.
Mol. Biol. Cell 11, 807–817
92 Cheever, M. L., Sato, T. K., de Beer, T., Kutateladze, T. G., Emr, S. D. and Overduin, M.
(2001) Phox domain interaction with PtdIns(3)P targets the Vam7 t-SNARE to vacuole
membranes. Nature Cell Biol. 3, 613–618
93 Lee, S. A., Kovacs, J., Stahelin, R. V., Cheever, M. L., Overduin, M., Setty, T. G., Burd,
C. G., Cho, W. and Kutateladze, T. G. (2006) Molecular mechanism of membrane
docking by the Vam7p PX domain. J. Biol. Chem. 281, 37091–37101
94 Wang, L., Merz, A. J., Collins, K. M. and Wickner, W. (2003) Hierarchy of protein
assembly at the vertex ring domain for yeast vacuole docking and fusion. J. Cell Biol.
160, 365–374
95 Wang, L., Seeley, E. S., Wickner, W. and Merz, A. J. (2002) Vacuole fusion at a ring of
vertex docking sites leaves membrane fragments within the organelle. Cell 108,
357–369
96 Fratti, R. A., Jun, Y., Merz, A. J., Margolis, N. and Wickner, W. (2004) Interdependent
assembly of specific regulatory lipids and membrane fusion proteins into the vertex ring
domain of docked vacuoles. J. Cell Biol. 167, 1087–1098
97 Stroupe, C., Collins, K. M., Fratti, R. A. and Wickner, W. (2006) Purification of active
HOPS complex reveals its affinities for phosphoinositides and the SNARE Vam7p.
EMBO J. 25, 1579–1589
98 Seeley, E. S., Kato, M., Margolis, N., Wickner, W. and Eitzen, G. (2002) Genomic
analysis of homotypic vacuole fusion. Mol. Biol. Cell 13, 782–794
99 Jun, Y., Fratti, R. A. and Wickner, W. (2004) Diacylglycerol and its formation by
phospholipase C regulate Rab- and SNARE-dependent yeast vacuole fusion.
J. Biol. Chem. 279, 53186–53195
100 Wiradjaja, F., Ooms, L. M., Tahirovic, S., Kuhne, E., Devenish, R. J., Munn, A. L., Piper,
R. C., Mayinger, P. and Mitchell, C. A. (2007) Inactivation of the phosphoinositide
phosphatases Sac1p and Inp54p leads to accumulation of phosphatidylinositol
4,5-bisphosphate on vacuole membranes and vacuolar fusion defects. J. Biol. Chem.
282, 16295–16307
101 Axe, E. L., Walker, S. A., Manifava, M., Chandra, P., Roderick, H. L., Habermann, A.,
Griffiths, G. and Ktistakis, N. T. (2008) Autophagosome formation from membrane
compartments enriched in phosphatidylinositol 3-phosphate and dynamically connected
to the endoplasmic reticulum. J. Cell Biol. 182, 685–701
102 Huijbregts, R. P., Topalof, L. and Bankaitis, V. A. (2000) Lipid metabolism and regulation
of membrane trafficking. Traffic 1, 195–202
103 Churchward, M. A., Rogasevskaia, T., Brandman, D. M., Khosravani, H., Nava, P.,
Atkinson, J. K. and Coorssen, J. R. (2008) Specific lipids supply critical negative
spontaneous curvature–an essential component of native Ca2+ -triggered membrane
fusion. Biophys. J. 94, 3976–3986
104 Chernomordik, L. V. and Kozlov, M. M. (2008) Mechanics of membrane fusion.
Nat. Struct. Mol. Biol. 15, 675–683
105 Kozlovsky, Y. and Kozlov, M. M. (2002) Stalk model of membrane fusion: solution of
energy crisis. Biophys. J. 82, 882–895
106 Chernomordik, L. V. and Kozlov, M. M. (2005) Membrane hemifusion: crossing a chasm
in two leaps. Cell 123, 375–382
107 Chernomordik, L. V., Zimmerberg, J. and Kozlov, M. M. (2006) Membranes of the world
unite! J. Cell Biol. 175, 201–207
108 Siegel, D. P. (1999) The modified stalk mechanism of lamellar/inverted phase transitions
and its implications for membrane fusion. Biophys. J. 76, 291–313
109 Siegel, D. P. (1993) Energetics of intermediates in membrane fusion: comparison of stalk
and inverted micellar intermediate mechanisms. Biophys. J. 65, 2124–2140
110 Chernomordik, L. V. and Zimmerberg, J. (1995) Bending membranes to the task:
structural intermediates in bilayer fusion. Curr. Opin. Struct. Biol. 5, 541–547
111 Antonny, B. (2006) Membrane deformation by protein coats. Curr. Opin. Cell Biol. 18,
386–394
112 Larijani, B. and Dufourc, E. (2006) Polyunsaturated phosphatidylinositol and
diacylglycerol substantially modify the fluidity and polymorphism of bio-membranes:
a solid state deuterium NMR study. Lipids 41, 925–932
113 Hermelink, A. and Brezesinski, G. (2008) Do unsaturated phosphoinositides mix with
ordered phosphadidylcholine model membranes? J. Lipid Res. 49, 1918–1925
114 Muller, M., Zschornig, O., Ohki, S. and Arnold, K. (2003) Fusion, leakage and surface
hydrophobicity of vesicles containing phosphoinositides: influence of steric and
electrostatic effects. J. Membr. Biol. 192, 33–43
115 Villar, A. V., Goni, F. M. and Alonso, A. (2001) Diacylglycerol effects on
phosphatidylinositol-specific phospholipase C activity and vesicle fusion. FEBS Lett.
494, 117–120
116 Villar, A. V., Alonso, A. and Goni, F. M. (2000) Leaky vesicle fusion induced by
phosphatidylinositol-specific phospholipase C: observation of mixing of vesicular inner
monolayers. Biochemistry 39, 14012–14018
c The Authors Journal compilation c 2009 Biochemical Society
246
D. Poccia and B. Larijani
117 Nieva, J. L., Alonso, A., Basanez, G., Goni, F. M., Gulik, A., Vargas, R. and Luzzati, V.
(1995) Topological properties of two cubic phases of a phospholipid:cholesterol:
diacylglycerol aqueous system and their possible implications in the phospholipase
C-induced liposome fusion. FEBS Lett. 368, 143–147
118 Basanez, G., Nieva, J. L., Rivas, E., Alonso, A. and Goni, F. M. (1996) Diacylglycerol and
the promotion of lamellar-hexagonal and lamellar-isotropic phase transitions in lipids:
implications for membrane fusion. Biophys. J. 70, 2299–2306
119 Goni, F. M., Nieva, J. L., Basanez, G., Fidelio, G. and Alonso, A. (1994) PhospholipaseC-promoted liposome fusion. Biochem. Soc. Trans. 22, 839–844
120 Nieva, J. L., Goni, F. M. and Alonso, A. (1989) Liposome fusion catalytically induced by
phospholipase C. Biochemistry 28, 7364–7367
121 Ruiz-Arguello, M. B., Goni, F. M. and Alonso, A. (1998) Phospholipase C hydrolysis of
phospholipids in bilayers of mixed lipid compositions. Biochemistry 37, 11621–11628
122 Ruiz-Arguello, M. B., Goni, F. M. and Alonso, A. (1998) Vesicle membrane fusion
induced by the concerted activities of sphingomyelinase and phospholipase C.
J. Biol. Chem. 273, 22977–22982
123 Weber, T., Zemelman, B. V., McNew, J. A., Westermann, B., Gmachl, M., Parlati, F.,
Sollner, T. H. and Rothman, J. E. (1998) SNAREpins: minimal machinery for membrane
fusion. Cell 92, 759–772
124 Pobbati, A. V., Stein, A. and Fasshauer, D. (2006) N- to C-terminal SNARE complex
assembly promotes rapid membrane fusion. Science 313, 673–676
125 Chen, X., Arac, D., Wang, T. M., Gilpin, C. J., Zimmerberg, J. and Rizo, J. (2006)
SNARE-mediated lipid mixing depends on the physical state of the vesicles. Biophys. J.
90, 2062–2074
126 Wagner, M. L. and Tamm, L. K. (2001) Reconstituted syntaxin1a/SNAP25 interacts with
negatively charged lipids as measured by lateral diffusion in planar supported bilayers.
Biophys. J. 81, 266–275
127 Vicogne, J., Vollenweider, D., Smith, J. R., Huang, P., Frohman, M. A. and Pessin, J. E.
(2006) Asymmetric phospholipid distribution drives in vitro reconstituted SNAREdependent membrane fusion. Proc. Natl. Acad. Sci. U.S.A. 103, 14761–14766
128 Brown, W. J., Chambers, K. and Doody, A. (2003) Phospholipase A2 (PLA2) enzymes in
membrane trafficking: mediators of membrane shape and function. Traffic 4, 214–221
129 Kooijman, E. E., Chupin, V., Fuller, N. L., Kozlov, M. M., de Kruijff, B., Burger, K. N. and
Rand, P. R. (2005) Spontaneous curvature of phosphatidic acid and lysophosphatidic
acid. Biochemistry 44, 2097–2102
130 Latham, C. F., Osborne, S. L., Cryle, M. J. and Meunier, F. A. (2007) Arachidonic acid
potentiates exocytosis and allows neuronal SNARE complex to interact with Munc18a.
J. Neurochem. 100, 1543–1554
131 Darios, F., Connell, E. and Davletov, B. (2007) Phospholipases and fatty acid signalling
in exocytosis. J. Physiol. 585, 699–704
132 Davletov, B., Connell, E. and Darios, F. (2007) Regulation of SNARE fusion machinery by
fatty acids. Cell. Mol. Life Sci. 64, 1597–1608
133 Rigoni, M., Caccin, P., Gschmeissner, S., Koster, G., Postle, A. D., Rossetto, O., Schiavo,
G. and Montecucco, C. (2005) Equivalent effects of snake PLA2 neurotoxins and
lysophospholipid-fatty acid mixtures. Science 310, 1678–1680
134 Rigoni, M., Pizzo, P., Schiavo, G., Weston, A. E., Zatti, G., Caccin, P., Rossetto, O.,
Pozzan, T. and Montecucco, C. (2007) Calcium influx and mitochondrial alterations at
synapses exposed to snake neurotoxins or their phospholipid hydrolysis products.
J. Biol. Chem. 282, 11238–11245
135 Zimmerberg, J. and Chernomordik, L. V. (2005) Neuroscience. Synaptic membranes
bend to the will of a neurotoxin. Science 310, 1626–1627
136 Roldan, E. R. and Fragio, C. (1993) Phospholipase A2 activation and subsequent
exocytosis in the Ca2+ /ionophore-induced acrosome reaction of ram spermatozoa.
J. Biol. Chem. 268, 13962–13970
137 Wei, S., Ong, W. Y., Thwin, M. M., Fong, C. W., Farooqui, A. A., Gopalakrishnakone, P.
and Hong, W. (2003) Group IIA secretory phospholipase A2 stimulates exocytosis and
neurotransmitter release in pheochromocytoma-12 cells and cultured rat hippocampal
neurons. Neuroscience 121, 891–898
138 Vitale, N., Mawet, J., Camonis, J., Regazzi, R., Bader, M. F. and Chasserot-Golaz, S.
(2005) The small GTPase RalA controls exocytosis of large dense core secretory
granules by interacting with ARF6-dependent phospholipase D1. J. Biol. Chem. 280,
29921–29928
139 Liscovitch, M. (1996) Phospholipase D: role in signal transduction and membrane
traffic. J. Lipid Mediators Cell Signalling 14, 215–221
140 Allan, D. and Michell, R. H. (1976) A possible role for 1,2-diacylglycerol in fusion of
erythrocytes by Sendai virus. Biochem. Soc. Trans. 4, 253
141 Ferrell, Jr, J. E. and Huestis, W. H. (1984) Phosphoinositide metabolism and the
morphology of human erythrocytes. J. Cell Biol. 98, 1992–1998
Received 17 October 2008/25 November 2008; accepted 26 November 2008
Published on the Internet 11 February 2009, doi:10.1042/BJ20082105
c The Authors Journal compilation c 2009 Biochemical Society
142 Wakelam, M. J. (1983) Inositol phospholipid metabolism and myoblast fusion.
Biochem. J. 214, 77–82
143 Sauro, V. S., Brown, G. A., Hamilton, M. R., Strickland, C. K. and Strickland, K. P. (1988)
Changes in phospholipid metabolism dependent on calcium-regulated myoblast fusion.
Biochem. Cell Biol. 66, 1110–1118
144 Sanchez-Migallon, M. P., Aranda, F. J. and Gomez-Fernandez, J. C. (1994) Role of
phosphatidylserine and diacylglycerol in the fusion of chromaffin granules with target
membranes. Arch. Biochem. Biophys. 314, 205–216
145 Conner, S., Leaf, D. and Wessel, G. (1997) Members of the SNARE hypothesis are
associated with cortical granule exocytosis in the sea urchin egg. Mol. Reprod. Dev. 48,
106–118
146 Conner, S. and Wessel, G. M. (1998) Rab3 mediates cortical granule exocytosis in the
sea urchin egg. Dev. Biol. 203, 334–344
147 Vacquier, V. D. (1975) The isolation of intact cortical granules from sea urchin eggs:
calcium lons trigger granule discharge. Dev. Biol. 43, 62–74
148 Wessel, G. M., Brooks, J. M., Green, E., Haley, S., Voronina, E., Wong, J.,
Zaydfudim, V. and Conner, S. (2001) The biology of cortical granules. Int. Rev. Cytol.
209, 117–206
149 Wong, J. L., Koppel, D. E., Cowan, A. E. and Wessel, G. M. (2007) Membrane
hemifusion is a stable intermediate of exocytosis. Dev. Cell 12, 653–659
150 Zampighi, G. A., Zampighi, L. M., Fain, N., Lanzavecchia, S., Simon, S. A. and Wright,
E. M. (2006) Conical electron tomography of a chemical synapse: vesicles docked to the
active zone are hemi-fused. Biophys. J. 91, 2910–2918
151 Decker, S. J. and Kinsey, W. H. (1983) Characterization of cortical secretory vesicles
from the sea urchin egg. Dev. Biol. 96, 37–45
152 Kinsey, W. H., Decker, G. L. and Lennarz, W. J. (1980) Isolation and partial
characterization of the plasma membrane of the sea urchin egg. J. Cell Biol. 87,
248–254
153 Churchward, M. A., Rogasevskaia, T., Hofgen, J., Bau, J. and Coorssen, J. R. (2005)
Cholesterol facilitates the native mechanism of Ca2+ -triggered membrane fusion.
J. Cell Sci. 118, 4833–4848
154 Chander, A., Chen, X. L. and Naidu, D. G. (2007) A role for diacylglycerol in annexin
A7-mediated fusion of lung lamellar bodies. Biochim. Biophys. Acta 1771, 1308–1318
155 Botelho, R. J., Teruel, M., Dierckman, R., Anderson, R., Wells, A., York, J. D., Meyer, T.
and Grinstein, S. (2000) Localized biphasic changes in phosphatidylinositol4,5-bisphosphate at sites of phagocytosis. J. Cell Biol. 151, 1353–1368
156 Strickland, L. I. and Burgess, D. R. (2004) Pathways for membrane trafficking during
cytokinesis. Trends Cell Biol. 14, 115–118
157 Emoto, K., Inadome, H., Kanaho, Y., Narumiya, S. and Umeda, M. (2005) Local change
in phospholipid composition at the cleavage furrow is essential for completion of
cytokinesis. J. Biol. Chem. 280, 37901–37907
158 Cameron, L. A. and Poccia, D. L. (1994) In vitro development of the sea urchin male
pronucleus. Dev. Biol. 162, 568–578
159 Larijani, B. and Poccia, D. (2007) Protein and lipid signaling in membrane fusion:
nuclear envelope assembly. Signal Transduction 7, 142–153
160 Larijani, B. and Poccia, D. (2009) Biophysics of the formation of the nuclear envelope.
Annu. Rev. Biophys. 38, 107–124
161 Collas, P. and Poccia, D. (1996) Distinct egg cytoplasmic membrane vesicles differing in
binding and fusion properties contribute to sea urchin male pronuclear envelopes
formed in vitro . J. Cell Sci. 109, 1275–1283
162 Larijani, B., Poccia, D. L. and Dickinson, L. C. (2000) Phospholipid identification and
quantification of membrane vesicle subfractions by two-dimensional 31 P-1 H nuclear
magnetic resonance. Lipids 35, 1289–1297
163 Byrne, R., Larijani, B. and Poccia, D. (2009) Tyrosine kinase regulation of nuclear
envelope assembly. Adv. Enzyme Regul., doi:10.1016/j.advenzreg.2008.12.010
164 Barona, T., Byrne, R. D., Pettitt, T. R., Wakelam, M. J., Larijani, B. and Poccia, D. L.
(2005) Diacylglycerol induces fusion of nuclear envelope membrane precursor vesicles.
J. Biol. Chem. 280, 41171–41177
165 Matynia, A., Salus, S. S. and Sazer, S. (2002) Three proteins required for early steps in
the protein secretory pathway also affect nuclear envelope structure and cell cycle
progression in fission yeast. J. Cell Sci. 115, 421–431
166 Jantsch-Plunger, V. and Glotzer, M. (1999) Depletion of syntaxins in the early
Caenorhabditis elegans embryo reveals a role for membrane fusion events in
cytokinesis. Curr. Biol. 9, 738–745
167 Zimmerberg, J. (2001) How can proteolipids be central players in membrane fusion?
Trends Cell Biol. 11, 233–235
168 Wu, H., Rossi, G. and Brennwald, P. (2008) The ghost in the machine: small GTPases
as spatial regulators of exocytosis. Trends Cell Biol. 18, 397–404