Phosphoinositides in cell regulation and
Gilbert Di Paolo1& Pietro De Camilli2
Inositol phospholipids have long been known to have an important regulatory role in cell physiology. The repertoire of
cellular processes known to be directly or indirectly controlled by this class of lipids has now dramatically expanded.
Through interactions mediated by their headgroups, which can be reversibly phosphorylated to generate seven species,
phosphoinositides play a fundamental part in controlling membrane–cytosol interfaces. These lipids mediate acute
responses, but also act as constitutive signals that help define organelle identity. Their functions, besides classical signal
transduction at the cell surface, include regulation of membrane traffic, the cytoskeleton, nuclear events and the
permeability and transport functions of membranes.
turnover of phosphatidylinositol (PtdIns) and its phosphorylated
products, called phosphoinositides1. Studies in this area eventually
focused on the function of phosphatidylinositol-4,5-bisphosphate
(PtdIns(4,5)P2) in signal transduction as a precursor of intracellular
messengers generated by phospholipases2. Unexpected twists came
and of phosphoinositides phosphorylated at the 3 position6–8.
Furthermore, genetic screens in yeast and studies in evolutionarily
highereukaryoticcells demonstrated theinvolvementofphosphoino-
other rare phosphoinositides13,14, phosphoinositide-binding protein
modules15,16and a plethora of phosphoinositide-metabolizing
enzymes17–19have now dramatically advanced this field. Phospho-
The importance of the signalling and constitutive functions of
phosphoinositides is paralleled by that of soluble inositol polyphos-
phates (InsPs). In addition to the fundamental roleof Ins(1,4,5)P3in
intracellular Ca2þsignalling2, many functions have been described
for other InsPs, including gene transcription, RNA editing, nuclear
export and protein phosphorylation. Although InsPs and phospho-
inositides are metabolically and functionally interconnected, this
review will focus selectively on the role of phosphoinositides.
n the early 1950s, Hokin and Hokin first observed changes in the
turnover of membrane phospholipids on stimulation of exocrine
tissues. This so-called ‘phospholipid effect’ was subsequently
detected in a variety of tissues and attributed to changes in the
Metabolism and spatial distribution of phosphoinositides
Inositol phospholipids are concentrated at the cytosolic surface of
membranes. PtdIns, the precursor of phosphoinositides, is synthe-
sized primarily in the endoplasmic reticulum (ER) and is then
delivered to other membranes either by vesicular transport or via
cytosolic PtdIns transfer proteins. Reversible phosphorylation of
its inositol ring at positions 3, 4 and 5 results in the generation of
seven phosphoinositide species (Fig. 1a). PtdIns typically represents
less than 15% of the total phospholipids found in eukaryotic cells
and phosphoinositides are generally less abundant by one order of
magnitude, with PtdIns(4)P and PtdIns(4,5)P2representing the
bulk of these lipids in mammalian cells. Each of the seven phospho-
inositides has a unique subcellular distribution with a predominant
localization in subsets of membranes (Fig. 1b, c; see below). Further-
inositides can be heterogeneous. Pools of phosphoinositides have
also been reported in the nucleus20. The differential intracellular
distribution of phosphoinositides, together with their high turnover,
makes these lipids optimal mediators of signalling events in all
Control of the membrane–cytosol interface
Phosphoinositides achieve direct signalling effects through the bind-
ing of their head groups to cytosolic proteinsorcytosolicdomains of
membrane proteins. Thus, they can regulate the function of integral
membrane proteins, or recruit to the membrane cytoskeletal and
signalling components. This action is reminiscent of that of
phosphotyrosine residues of membrane proteins, which mediate
the recruitment of phosphotyrosine-binding modules (for example,
SH2 domains). Typically, binding of proteins to phosphoinositides
involves electrostatic interactions with the negative charges of the
phosphate(s) on the inositol ring. In some cases, adjacent hydro-
phobic amino acids strengthen the interaction through a partial
penetration into the bilayer15,16. Protein surfaces that interact with
phosphoinositides can consist either of clusters of basic residues
within unstructured regions, such as those found in many actin
regulatory proteins (for example, profilin)5, or of folded modules,
such as the pleckstrin homology domain (Fig. 1d and see
below)15,16,21. The distinction between these two types of interaction
can be blurred. Furthermore, unstructured peptides can undergo
folding on binding to phosphoinositide head groups. The number of
theirdifferential properties greatly amplify the signalling potential of
Coincidence detection code and organelle identity
Generally, the interaction between phosphoinositides and cytosolic
proteins is of low affinity. Higher affinity, and thus more stable,
membrane–protein interactions are produced when phospho-
1Department of Pathology and Cell Biology, The Taub Institute for Research on Alzheimer’s Disease and the Aging Brain, Columbia University Medical Center, New York,
New York 10032, USA.2Departments of Cell Biology and Neurobiology, Program in Cellular Neuroscience, Neurodegeneration and Repair, Howard Hughes Medical Institute and
Kavli Institute for Neuroscience, Yale University School of Medicine, New Haven, Connecticut 06510, USA.
Vol 443|12 October 2006|doi:10.1038/nature05185
© 2006 Nature Publishing Group
phosphoinositides, this dual (or multiple) key-based recognition
mechanism generates a powerful coincidence detection code for the
regulation of membrane–cytosol interactions and thus for the defi-
nition of organelle identity (Fig. 2a)22,23. For example, although the
they are recruited to the Golgi complex or the plasma membrane,
respectively, at least in part owing to the different phosphoinositide
content of these membranes (Fig. 2b)24–27. One advantage of this dual
code is that each component of the code can be independently
regulated (for example, by lipid enzyme activation or co-receptor
phosphorylation), thus offering an opportunity for signal integration.
The spatial restriction and steady-state levels of specific phospho-
inositides are determined primarily by the concerted action of
phosphoinositide kinases and phosphatases, whose localization is
tightly controlled. Phosphoinositide segregation on different mem-
branes helps ensure vectorial membrane traffic, which is critical to
achieve transport of cargo from one compartment to another. For
example, a carrier vesicle must harbour a lipid composition that
promotes the recruitment to its surface of factors needed for fusion
with its acceptor membrane (Fig. 2c). This state must be maintained
until fusion has occurred and terminated thereafter. Conversely, the
acceptor membrane must have properties that make it suitable as a
partner for fusion and as a template for the recruitment of budding
factors needed for the next trafficking step. After budding, these
factors (generally coat proteins) must be shed. Modifications in
phosphoinositide environment after fusion (by a dilution effect
and exposure to a different enzymatic context) and then after
budding (for example, by the selective recruitment of a phospho-
inositide phosphatase onto the vesicle) can efficiently account for
theseglobalchanges becauseanew phosphoinositide wouldgenerate
a new identity code (Fig. 2c).
A phosphoinositide-based code represents a very effective way to
define organelle identity. Phospholipids can rapidly diffuse within,
addition, phosphoinositides can be rapidly interconverted from one
species to another bystrategically localized kinases and phosphatases
as a membrane carrier translocates from one compartment to the
next (Figs1c and 2c).Severalenzymes contain modules that bind the
substrate phosphoinositide (Fig. 1d), thus facilitating recruitment of
enzymes to the appropriate membrane and their retention at this
membrane for catalysis as long as substrates to be consumed are
Cooperation with small GTPases
Another class of molecules that mediate or enable the recruitment of
cytosolic proteins to specific membrane compartments or subcom-
partments is the Ras superfamily of small GTPases. These include
proteins implicated in signalling (for example, the Ras family itself),
actin regulation (for example, the Rho family), organelle and
vesicular transport (for example, the Arf, Arl and Rab families)23.
Small GTPases shuttle between a GDP (inactive)- and a GTP
Figure 1 | Metabolism and subcellular distribution of phosphoinositides.
a, Metabolic reactions leading to the generation of seven phosphoinositide
species from PtdIns. Reactions indicated with dotted arrows have been
shown in vitro, but their importance in living cells remains unclear. DAG,
diacylglycerol. b, Fluorescence micrographs illustrating the predominant
localization of PtdIns(4,5)P2(plasma membrane), PtdIns(4)P (Golgi) and
PtdIns(3)P (endosomes) in Chinese hamster ovary cells as revealed by the
pleckstrin homology (PH) domain of PLCd1, the PH domain of FAPP1 and
the tandem FYVE domain of Hrs (all fused to GFP, green). The nucleus is
shown in blue (DAPI staining). c, Predominant subcellular localization of
phosphoinositide species. PtdIns(4,5)P2and PtdIns(3,4,5)P3are
concentrated at the plasma membrane, possibly enriched in raft-like
structures. PtdIns(3,4)P2is mostly found at the plasma membrane and in
the early endocytic pathway. PtdIns(4)P is enriched at the Golgi complex,
but also present at the plasma membrane. PtdIns(3)P is concentrated in
earlyendosomes and PtdIns(3,5)P2on late compartments of the endosomal
pathway. The location of PtdIns(5)P remains poorly characterized.
Phosphorylation/dephosphorylation events occurring on each subcellular
compartment are illustrated. In the ER, the conversion from phosphatidic
acid (PA) to PtdIns involves more than one reaction. d, Phosphoinositide-
binding modules, their binding preference and some examples of proteins
that contain them.
NATURE|Vol 443|12 October 2006
© 2006 Nature Publishing Group
(active)-bound state. When bound to GTP, they interact with a
variety of effectors. This cycle is regulated by guanine nucleotide
exchange factors (GEFs) and GTPase activating proteins (GAPs),
which stimulate GTP loading and hydrolysis, respectively. Not
phosphoinositide metabolism and these proteins have been reported
(Fig. 3). First, phosphoinositides regulate the recruitment of GAPs
and GEFs to membranes and their activity, given the presence of
phosphoinositide-binding modules in many of these factors (Fig. 3a).
Second, several phosphoinositide-metabolizing enzymes are effectors
asco-receptors with phosphoinositidesintherecruitmentof cytosolic
proteins to specific subcompartments (Fig. 3c). Because mammalian
cells express more than 100 small GTPases, the latter mechanism
dramatically increases (in a combinatorial manner) the repertoire
of identity codes for membranes and membrane domains22,23.
The following sections provide an overview of some of the most
important regulatory roles of phosphoinositides in the major cell
Figure 2 | Coincidence detection in phosphoinositide signalling.
a, Phosphoinositides mediate the recruitment of proteins to membranes
using their phosphorylated head groups. A high specificity is achieved when
other signals act in concert with phosphoinositides. This dual-key
the interactions between cytosolic proteins and membranes. b, To illustrate
theconcept of coincidencedetection,thecaseof theclathrin adaptorAP-2 is
shown.This proteincomplex,whichconsistsof foursubunits(a,b2, m2and
j2), is selectively recruited to the plasma membrane owing, in part, to its
ability to bind PtdIns(4,5)P2. It has been proposed that an initial contact
with this membrane mediated by the PtdIns(4,5)P2binding site present on
the a subunit triggers a conformational change in the m2 subunit. This
change, in turn, leads to the exposure of an additional binding site for
modulated by the phosphorylation of AP-2, cooperate in stabilizing the
association of AP-2 with the plasma membrane. Although cargo proteins
are also present in intracellular membranes, AP-2 is not efficiently recruited
to such membranes owing to the lack of a cognate phosphoinositide26.
c, A schematic drawing illustrating changes in phosphoinositide
composition (represented bychanges in membrane colour) that accompany
fusion and budding reactions. On fusion, a vesicle will acquire the
predominant phosphoinositide composition of the target membrane.
phosphoinositide composition can be rapidly and globally changed.
Figure 3 | Cartoons illustrating examples of functional interplay between
phosphoinositides and small GTPases. a, GTPases (purple) shuttle
between GDP (inactive)- and GTP (active)-bound states, the
(green) regulate the nucleotide cycle of GTPases owing to their ability to
bind and activate both GEFs and GAPs. b, Many small GTPases bind (and
often regulate) phosphoinositide-metabolizing enzymes (here, a
phosphoinositide kinase is depicted), thereby contributing to the
generation of membrane domains enriched in specific phosphoinositides.
c, Phosphoinositides and small GTPases act in concert to recruiteffectors to
specific membrane compartments. These three mechanisms (a, b, c) can
occur simultaneously on the same membrane compartment, thereby giving
rise to complex feedback loops.
NATURE|Vol 443|12 October 2006
© 2006 Nature Publishing Group
Classical signal transduction at the plasma membrane
PtdIns(4,5)P2, which is concentrated in the plasma membrane, par-
ticipates in nearly all events that occur at, or involve, the cell surface
(Fig. 4). In addition, it plays a major part in the transduction of
extracellular signals, either via its metabolites, or through fluctuations
of its own levels.
PtdIns(4,5)P2 is produced by the phosphorylation of either
PtdIns(4)P (by type I PtdInsP kinases) or PtdIns(5)P (by type II
PtdInsP kinases)5,14,19. There is only very little PtdIns(5)P in cellsand
most cell surface PtdIns(4,5)P2is generated from PtdIns(4)P. This
lipid is delivered to the cell surface by membrane carriers derived
from the Golgi complex and from recycling organelles (see below),
hydrolysis is due to both phospholipases and phosphatases with
fundamentally different physiological consequences. Cleavage by
phospholipases, such as phospholipase C (PLC) and phospholipase
A2 (PLA2), gives rise to metabolites that propagate and amplify
signalling,whereasdephosphorylation,primarily by 5-phosphatases,
controls steady-state levels of PtdIns(4,5)P2and turns off its signal-
ling. For example, during endocytosis, dephosphorylation removes
PtdIns(4,5)P2from internalized membranes, thus restricting the
localization of PtdIns(4,5)P2to the cell surface without coupling
the endocytic reaction to the generation of signalling metabolites30.
Another fate of PtdIns(4,5)P2is conversion to PtdIns(3,4,5)P3
primarily by class I phosphoinositide (3)kinases (PI(3) kinases)31–33.
This lipid is present in negligible amounts in resting cells, but can
transiently and dramatically increase in response to growth factor
stimulation. Its regulatory role is of fundamental importance in the
physiology of higher eukaryotes, as it mediates a large variety of
effects, including cell proliferation, migration, chemotaxis, phago-
and macro-pinocytosis, differentiation, survival and metabolic
changes31–33. PtdIns(3,4,5)P3acts by recruiting effectors that activate
that function as regulators of the actin cytoskeleton, and the protein
kinases PDK and AKT/PKB, which cooperate in the activation of
important signalling cascades, such as the Tor pathway31,32.
PtdIns(3,4,5)P3is dephosphorylated by two types of enzymes, with
different outcomes. Dephosphorylation at the 3 position by PTEN is
the ‘off’ signal34. In contrast, dephosphorylation by 5-phosphatases,
such as SHIP-1 and 2, generates PtdIns(3,4)P2, which shares several
actions with PtdIns(3,4,5)P3(in addition to unique signalling
properties) and may prolong the duration of PtdIns(3,4,5)P3signal-
ling17. This has major relevance to medicine, given the importance of
PtdIns(3,4,5)P3metabolism imbalance in diabetes32and cancer31.
PTEN is also a potent tumour suppressor (Table 1)34,35.
Regulation of integral plasma membrane proteins
An emerging theme in phosphoinositide signalling is the involve-
ment of PtdIns(4,5)P2in the regulation of integral membrane
proteins at the plasma membrane. These include a variety of inward
pumps, transient receptor potential channels, epithelial sodium
channels, ion exchangers36,37and probably, many other classes of
proteins, such as receptors for chemical ligands. For some proteins,
presence of binding sites for PtdIns(4,5)P2 may represent a
mechanism for optimal function in their appropriate cellular con-
text, that is, the plasma membrane. In other cases, regulation by
PtdIns(4,5)P2may allow these proteins to function as effectors of
receptors coupled to PtdIns(4,5)P2synthesis or degradation. Regu-
latory actions of PtdIns(4,5)P2are well documented for several
transient receptor potential channels, many of which function in
Plasma membrane–cytoskeleton interactions
PtdIns(4,5)P2and PtdIns(3,4,5)P3, often in concert with small
GTPases, participate in the recruitment and activation of a wide
variety of actin regulatory proteins at the plasma membrane, thereby
controlling cell shape, motility, cytokinesis and a wide variety of
other processes. A well-characterized mechanism governing actin
polymerization and involving PtdIns(4,5)P2is the ARP2/3-mediated
nucleation of actin networks. For example, PtdIns(4,5)P2, in
cooperation with the small GTPase Cdc42, binds N-WASP and
triggers a conformational change that allows its binding to and
activation of the ARP2/3 complex38,39. Other activators of N-WASP
(for example, Toca) are also recruited to the plasma membrane by
the cooperative action of phosphoinositides and Cdc42, thus
providing mechanisms for the amplification of PtdIns(4,5)P2
PtdIns(4,5)P2facilitates actin filament elongation at the plus ends
preferentially occurs) by promoting the dissociation of capping
to profilin, it promotes the dissociation of actin-monomer–profilin
complexes, thus making G-actin available5,38.
that actasadaptors between themembraneand theactin cytoskeleton
atsites of cell–matrixorcell–cell adhesion.These adaptors includethe
band 4.1/ezrin/radixin/moesin (FERM) domain-containing proteins.
Typically, PtdIns(4,5)P2binds to the FERMdomain of these proteins,
talin not only binds PtdIns(4,5)P2, which enhances its affinity for
integrins, but also the PtdIns(4,5)P2-synthesizing enzyme PtdInsP
kinase type 1g and so helps generate this lipid at sites of cell
PtdIns(3,4,5)P3mediates the effect of growth factors on the
formation of the peripheral ruffles implicated in cell migration32,43.
Mechanistically, the actions of PtdIns(3,4,5)P3are similar to those
of PtdIns(4,5)P2, although theyoften involveadistinct set of effector
proteins. Forexample, WAVE, a WASP-related protein, is specifically
recruited and activated by PtdIns(3,4,5)P3(ref. 43). In Dictyostelium
discoideum, the spatial restriction of PtdIns(3,4,5)P3at the leading
edge is responsible for polarized cell migration during chemotaxis44.
not restricted to actin. For instance, PtdIns(4,5)P2-enriched micro-
domains in the plasma membrane participate in the regulation of
microtubule plus-end capture and stabilization, a phenomenon
required for normal polarized motility45. In another example, a
microtubular motor was shown to be anchored to cell organelles
using a phosphoinositide-binding pleckstrin homology domain, thus
implicating phosphoinositides in microtubule-based intracellular
Figure 4 | Examples of processes
regulated by PtdIns(4,5)P2at the
plasma membrane. See text for
explanations. ECM, extracellular
matrix; PI3K, PI(3)kinase.
NATURE|Vol 443|12 October 2006
© 2006 Nature Publishing Group
Plasma membrane PtdIns(4,5)P2is directly implicated in exocytosis
It acts both in cis on plasma membrane proteins (such as CAPS, a
in-trans interactions probably cooperate with SNARE pairing48in
marking the plasma membrane as the appropriate partner for the
fusion of these organelles. Generally, levels of PtdIns(4,5)P2in the
plasma membrane positively correlate with the size of the readily
releasable pool of secretory vesicles (that is the pool of membrane-
docked vesicles that is competent for exocytosis)49–51. PtdIns(4,5)P2
also indirectly affects exocytosis through metabolites produced by
PLC-dependent cleavage: diacylglycerol is a critical ligand for the
priming factor Munc13 (refs 52, 53), whereas Ins(1,4,5)P3mediates
calcium responses that modulate and in some cases trigger secretion.
PtdIns(4,5)P2has an important role in all forms of endocytosis,
because it functions as an important co-receptor for the recruitment
and regulation of endocytic proteins selectively to the plasma
membrane22,27. PtdIns(4,5)P2(in most cases also PtdIns(3,4,5)P3)
binds all the known endocytic clathrin adaptors (for example, AP-2,
AP180/CALM, epsin) and many other endocytic factors, including
dynamin, which controls the fission reaction22,24,27. Other actions of
PtdIns(4,5)P2in endocytosis reflect its effects on the actin cytoskele-
ton, which is implicated in all internalization pathways54.
Specific retention of PtdIns(4,5)P2and PtdIns(3,4,5)P3at the
plasma membrane and rapid dissociation of endocytic factors follow-
ing internalization require the removal of these phosphoinositides
from endocytic membranes. This function is largely fulfilled by
an enzyme implicated in synaptic vesicle recycling11,30. The function
of synaptojanin is tightly interconnected with that of dynamin,
suggesting a mechanism that couples endocytic vesicle fission and
phosphoinositide dephosphorylation11. In budding yeast, conditional
disruption of synaptojanin function results in ectopic accumulation
of PtdIns(4,5)P2on internal membranes55.
Phagocytosis and pathogen entry
Phagocytosis (that is, the engulfment of large particles $0.5mm in
diameter), begins with the activation of cell surface receptors by
ligands that are present on the foreign particle56. This event triggers
the local polymerization of actin, leading to the formation of
cell protrusions (pseudopods) that engulf and ‘seal’ the particle.
This process requires PtdIns(4,5)P2 and local production of
PtdIns(3,4,5)P3, which stimulates actin nucleation. On ‘sealing’
however, catabolism of PtdIns(4,5)P2—mediated at least in part by
PLC—and of PtdIns(3,4,5)P3is critical for the shedding of actin and
fusion of phagosomes with the endosome/lysosome compartment56.
A special form of phagocytosis is the internalization of pathogenic
bacteria, which capitalize in several ways on phosphoinositide-
mediated signalling to invade and thrive in host cells (Table 1)57. In a
notable example of this interrelationship, some bacteria deliver phos-
phoinositide phosphatases into host cell cytoplasm that then help
reprogram cell function to facilitate invasion and/or to determine the
appropriate fate of pathogen-containing vacuoles (Table 1)17,57–61.
Golgi complex function
PtdIns(4)P is the predominant phosphoinositide in the Golgi-
complex region and a deficiency of this lipid affects its structure and
function. In yeast, mutations of the Golgi-localized PtdIns(4)kinase
PIK1result in anaccumulation of abnormal membraneintermediates
(‘Berkeley bodies’) with impairment of transport to the plasma
membrane18. These defects are reminiscent of those produced by
mutations of the GTPase ARF162, which may reflect the property of
PtdIns(4)P to function as a co-receptor with ARF1 in the recruitment
clathrin adaptor AP-1, the GGA proteins and FAPP1/2 (refs 23, 63).
with and is stimulated by ARF1, indicating that this GTPase can
cooperate with PtdIns(4)P by multiple and synergistic mechanisms63
by the knockdown of another PtdIns(4)kinase, type IIa (ref. 25).
Spatial restriction of PtdIns(4)P to the Golgi complex is ensured
by phosphatases, one of which is likely to be Sac1 (ref. 64). This
protein is an ER- and Golgi-complex-associated phosphoinositide
phosphatase, which may thus dephosphorylate PtdIns(4)P when
retrograde traffic occurs from the Golgi complex to the ER64.
Mutant sac1 yeast cells have high levels of PtdIns(4)P, including
of the excessive build-up of PtdIns(4)P in early secretory
The presence of pools of PtdIns(4,5)P2in the Golgi complex has
been suggested by the reported localization of PtdIns(4,5)P2-metabo-
lizing enzymes, primarily the disease-linked inositol(5)phosphatase,
Table 1 | Phosphoinositide-metabolizing enzymes that have been linked to human diseases.
Gene name Chr. EnzymePredominant substrateProduct DiseaseRefs
3Class I PI(3)kinase p110a
Class III PI(3)kinase hVPS34
PtdIns(4)kinase Type IIa
PtdInsP kinase Type IIa
PtdInsP kinase Type III (PIKfyve)
Francois–Neetens fleck cornea dystrophy
MTM1X3-phosphatase myotubularinMyotubular myopathy 70
MTMR2 113-phosphatase myotubularin-related proteinCharcot–Marie–Tooth type 4B171
Type 2 diabetes
Oculocerebrorenal syndrome of Lowe
Bipolar disorder, Down syndrome?
Bacterial geneEnzymePredominant substrateProductPathogenRefs
Refs 74, 75 are additional references for Table 1. Chr., chromosome. Bold, established genetic link; plain, putative link.
NATURE|Vol 443|12 October 2006
© 2006 Nature Publishing Group
OCRL, in this region (Table 1)17,63,66. The functional significance of
PtdIns(4,5)P2in the Golgi complex remains unclear.
PtdIns(3)P and PtdIns(3,5)P2have a crucial role in the biology of
endosomes—a heterogeneous system of membranes that function as
crossroads for traffic from and to the plasma membrane, the Golgi
complex and the lysosomes18. PtdIns(3)P is a major determinant of
early endosome membrane identity, and participates in nearly all
aspects of endosomal function, such as membrane tethering and
fusion, interaction with the cytoskeleton, signalling and motility.
PtdIns(3)P-binding modules, primarily FYVE and PX domains, are
present in a variety of endosomal proteins, such as EEA1, Hrs and
SARA15,16,21,67. These proteins are generally localized to distinct
subdomains of the endosomal system68, suggesting that PtdIns(3)P
binding confers a first layer of specificity to their subcellular targeting
and that other interactions fine-tune their precise localization. Several
PtdIns(3)P-binding proteins also bind Rab5, an early endosome-
associated GTPase, thus providing a powerful and highly specific
to early endosomes. Rab5, in turn, regulates PtdIns(3)P synthesis
through two distinct pathways. The majority of this lipid is produced
from PtdIns by the class III PI(3)kinase VPS34, a Rab5 effector10,18,68.
An additional pool of PtdIns(3)P is generated along the endocytic
The latter pathway, which is upregulated by growth factor stimu-
lation, coordinates the internalization of PtdIns(3,4,5)P3-enriched
membranes with the modification of their phosphoinositide content
to match the property of endosomal membranes29.
PtdIns(3)P can undergo dephosphorylation at the endosomal
surface, degradation within the lumen of multivesicular bodies, or
main function of PtdIns(3,5)P2may be to trigger budding of vesicles
from late endosomes and many of the effects of Fab1p/PIKfyve
disruption (for example, swelling of the yeast vacuole and of late
endosomes in mammalian cells) may be explained by defective
membrane recycling from this compartment. Dephosphorylation
of PtdIns(3)P and PtdIns(3,5)P2is achieved by members of the
myotubularin34and Sac1 families64.
Human genetic diseases have been mapped to genes encoding
myotubularin family members70,71and PIKfyve72, although the precise
role of phosphoinositide metabolism imbalance in the phenotypic
manifestations of these diseases remains poorly understood (Table 1).
Functions at other intracellular locations
Although PtdIns is synthesized in the ER, the presence of its
phosphorylated derivatives in this compartment remains very
much an open question. Phosphoinositide phosphatases located on
this organelle, such as Sac1 and SKIP, may act primarily to dephos-
phorylate phosphoinositides inappropriately delivered to its mem-
branes. Alternatively, they mayactin trans onother membranes. The
potential role of phosphoinositides in mitochondria remains un-
explored, although enzymes, such as a synaptojanin splice variant,
have been reported there19,63. As mentioned above, phosphoinositides
and phosphoinositide-binding proteins have been described in non-
of proteins with hydrophobic pockets that are able to accommodate
their fatty acid tails. Consistent with these observations, major phos-
phoinositide-metabolizing enzymes are found in the nucleus, and
there is evidence for a role of phosphoinositides in pre-messenger-
RNA splicing, chromatin remodelling and gene transcription20.
As is clear from this overview, phosphoinositides occupy a general
and central place in the field of cell signalling and regulation. A
challenge for the future will be to elucidate how signalling effects
elicited by their rapid metabolic changes are coordinated with their
constitutive functions in a variety of cell processes. Surprisingly,
many key phosphoinositide metabolizing enzymes remain poorly
characterized andthe functionofonly a small subsetof them has been
genetically explored, atleast in higherorganisms.Acentral questionis
in some cases as a ‘futile’ cycle) primarily serves to allow for rapid
changes and a fine regulation of their levels, or whether turnover is
intimately connected to their mechanisms of action.
An important issue in phosphoinositide signalling is to what extent
the levels of ‘free’ phosphoinositides are controlled by phospho-
inositide-sequestering proteins (for example, ‘pipmodulins’)45,73. At
buffer PtdIns(4,5)P2and to release it after appropriate stimuli (for
example, protein phosphorylation), thus potentially contributing
another layer of regulation73.
One priority in phosphoinositide research is the further develop-
ment of methods for their efficient detection and for the discrimi-
nation of phosphoinositides with different fatty acid compositions.
Genetically encoded, fluorescent, phosphoinositide-binding modules
are now commonly used to image phosphoinositides in living cells,
but the presence in these modules of additional binding sites for
cellular proteins (and/or other lipids), or the masking of phospho-
inositides by endogenous proteins, may bias the results. The extent to
which specific phosphoinositidepoolsare invisibletotheseprobesisa
highlydebatedissue, because the locations of some phosphoinositide-
metabolizing enzymes seem to be in disagreement with the predo-
minant location of the substrates and/or products revealed by such
probes. Whether pools of phosphoinositides at locations that are not
revealed by these probes do exist and are physiologically important
remains an open question.
Finally, the involvement of several key phosphoinositide metabolic
reactions in disease makes this area of research highly relevant to
medicine (see also Table 1). Further elucidation of the role of phos-
phoinositides in cell biology and organismal physiology is expected to
provide new potential targets for pharmacological interventions in
major human diseases, including cancer and diabetes.
1.Hokin, L. E. Receptors and phosphoinositide-generated second messengers.
Annu. Rev. Biochem. 54, 205– -235 (1985).
Berridge, M. J. & Irvine, R. F. Inositol phosphates and cell signalling. Nature 341,
197– -205 (1989).
Lassing, I. & Lindberg, U. Specific interaction between phosphatidylinositol
4,5-bisphosphate and profilactin. Nature 314, 472– -474 (1985).
Ma, L., Cantley, L. C., Janmey, P. A. & Kirschner, M. W. Corequirement of
specific phosphoinositides and small GTP-binding protein Cdc42 in inducing
actin assembly in Xenopus egg extracts. J. Cell Biol. 140, 1125– -1136 (1998).
Yin, H. L. & Janmey, P. A. Phosphoinositide regulation of the actin cytoskeleton.
Annu. Rev. Physiol. 65, 761– -789 (2003).
Whitman, M., Downes, C. P., Keeler, M., Keller, T. & Cantley, L. Type I
phosphatidylinositol kinase makes a novel inositol phospholipid,
phosphatidylinositol-3-phosphate. Nature 332, 644– -646 (1988).
Traynor-Kaplan, A. E., Harris, A. L., Thompson, B. L., Taylor, P. & Sklar, L. A.
An inositol tetrakisphosphate-containing phospholipid in activated neutrophils.
Nature 334, 353– -356 (1988).
Auger, K. R. et al. PDGF-dependent tyrosine phosphorylation stimulates
production of novel polyphosphoinositides in intact cells. Cell 57, 167– -175 (1989).
Eberhard, D. A., Cooper, C. L., Low, M. G. & Holz, R. W. Evidence that the
inositol phospholipids are necessary for exocytosis. Loss of inositol
phospholipids and inhibition of secretion in permeabilized cells caused by a
bacterial phospholipase C and removal of ATP. Biochem. J. 268, 15– -25 (1990).
10. Schu, P. V. et al. Phosphatidylinositol 3-kinase encoded by yeast VPS34 gene
essential for protein sorting. Science 260, 88– -91 (1993).
11.McPherson, P. S. et al. A presynaptic inositol-5-phosphatase. Nature 379,
353– -357 (1996).
12. Martin, T. F. Phosphoinositide lipids as signaling molecules: common themes
for signal transduction, cytoskeletal regulation, and membrane trafficking.
Annu. Rev. Cell Dev. Biol. 14, 231– -264 (1998).
13. Dove, S. K. et al. Osmotic stress activates phosphatidylinositol-3,5-
bisphosphate synthesis. Nature 390, 187– -192 (1997).
14. Rameh, L. E., Tolias, K. F., Duckworth, B. C. & Cantley, L. C. A new pathway for
synthesis of phosphatidylinositol-4,5-bisphosphate. Nature 390, 192– -196 (1997).
15. Lemmon, M. A. Phosphoinositide recognition domains. Traffic 4, 201– -213
NATURE|Vol 443|12 October 2006
© 2006 Nature Publishing Group
16. Balla, T. Inositol-lipid binding motifs: signal integrators through protein– -lipid Download full-text
and protein– -protein interactions. J. Cell Sci. 118, 2093– -2104 (2005).
17. Majerus, P. W., Kisseleva, M. V. & Norris, F. A. The role of phosphatases in
inositol signaling reactions. J. Biol. Chem. 274, 10669– -10672 (1999).
18. Odorizzi, G., Babst, M. & Emr, S. D. Phosphoinositide signaling and the regulation
of membrane trafficking in yeast. Trends Biochem. Sci. 25, 229– -235 (2000).
19. Roth, M. G. Phosphoinositides in constitutive membrane traffic. Physiol. Rev.
84, 699– -730 (2004).
20. Hammond, G., Thomas, C. L. & Schiavo, G. Nuclear phosphoinositides and their
functions. Curr. Top. Microbiol. Immunol. 282, 177– -206 (2004).
21. Hurley, J. H. & Meyer, T. Subcellular targeting by membrane lipids. Curr. Opin.
Cell Biol. 13, 146– -152 (2001).
22. Wenk, M. R. & De Camilli, P. Protein– -lipid interactions and phosphoinositide
metabolism in membrane traffic: insights from vesicle recycling in nerve
terminals. Proc. Natl Acad. Sci. USA 101, 8262– -8269 (2004).
23. Behnia, R. & Munro, S. Organelle identity and the signposts for membrane
traffic. Nature 438, 597– -604 (2005).
24. Gaidarov, I. & Keen, J. H. Phosphoinositide– -AP-2 interactions required for
targeting to plasma membrane clathrin-coated pits. J. Cell Biol. 146, 755– -764
25. Wang, Y. J. et al. Phosphatidylinositol 4 phosphate regulates targeting of
clathrin adaptor AP-1 complexes to the Golgi. Cell 114, 299– -310 (2003).
26. Honing, S. et al. Phosphatidylinositol-(4,5)-bisphosphate regulates sorting
signal recognition by the clathrin-associated adaptor complex AP2. Mol. Cell
18, 519– -531 (2005).
27. Owen, D. J., Collins, B. M. & Evans, P. R. Adaptors for clathrin coats: structure
and function. Annu. Rev. Cell Dev. Biol. 20, 153– -191 (2004).
28. Honda, A. et al. Phosphatidylinositol 4-phosphate 5-kinase a is a downstream
effector of the small G protein ARF6 in membrane ruffle formation. Cell 99,
521– -532 (1999).
29. Shin, H. W. et al. An enzymatic cascade of Rab5 effectors regulates
phosphoinositide turnover in the endocytic pathway. J. Cell Biol. 170, 607– -618
30. Cremona, O. et al. Essential role of phosphoinositide metabolism in synaptic
vesicle recycling. Cell 99, 179– -188 (1999).
31. Cantley, L. C. The phosphoinositide 3-kinase pathway. Science 296, 1655– -1657
32. Czech, M. P. Dynamics of phosphoinositides in membrane retrieval and
insertion. Annu. Rev. Physiol. 65, 791– -815 (2003).
33. Katso, R. et al. Cellular function of phosphoinositide 3-kinases: implications for
development, homeostasis, and cancer. Annu. Rev. Cell Dev. Biol. 17, 615– -675
34. Wishart, M. J. & Dixon, J. E. PTEN and myotubularin phosphatases: from
3-phosphoinositide dephosphorylation to disease. Trends Cell Biol. 12, 579– -585
35. Li, J. et al. PTEN, a putative protein tyrosine phosphatase gene mutated in
human brain, breast, and prostate cancer. Science 275, 1943– -1947 (1997).
36. Hilgemann, D. W., Feng, S. & Nasuhoglu, C. The complex and intriguing lives of
PIP2with ion channels and transporters. Sci. STKE 2001, RE19, doi:10.1126/
37. Suh, B. C. & Hille, B. Regulation of ion channels by phosphatidylinositol
4,5-bisphosphate. Curr. Opin. Neurobiol. 15, 370– -378 (2005).
38. Pollard, T. D. & Borisy, G. G. Cellular motility driven by assembly and
disassembly of actin filaments. Cell 112, 453– -465 (2003).
39. Rohatgi, R., Ho, H. Y. & Kirschner, M. W. Mechanism of N-WASP activation by
CDC42 and phosphatidylinositol 4,5-bisphosphate. J. Cell Biol. 150, 1299– -1310
40. Ho, H. Y. et al. Toca-1 mediates Cdc42-dependent actin nucleation by
activating the N-WASP– -WIP complex. Cell 118, 203– -216 (2004).
41. Di Paolo, G. et al. Recruitment and regulation of phosphatidylinositol phosphate
kinase type 1 g by the FERM domain of talin. Nature 420, 85– -89 (2002).
42. Ling, K. et al. Type I g phosphatidylinositol phosphate kinase targets and
regulates focal adhesions. Nature 420, 89– -93 (2002).
43. Oikawa, T. et al. PtdIns(3,4,5)P3binding is necessary for WAVE2-induced
formation of lamellipodia. Nature Cell Biol. 6, 420– -426 (2004).
44. Iijima, M., Huang, Y. E. & Devreotes, P. Temporal and spatial regulation of
chemotaxis. Dev. Cell 3, 469– -478 (2002).
45. Golub, T. & Caroni, P. PI(4,5)P2-dependent microdomain assemblies capture
microtubules to promote and control leading edge motility. J. Cell Biol. 169,
151– -165 (2005).
46. Klopfenstein, D. R. & Vale, R. D. The lipid binding pleckstrin homology domain
in UNC-104 kinesin is necessary for synaptic vesicle transport in Caenorhabditis
elegans. Mol. Biol. Cell 15, 3729– -3739 (2004).
47. Bai, J., Tucker, W. C. & Chapman, E. R. PIP2increases the speed of response of
synaptotagmin and steers its membrane-penetration activity toward the
plasma membrane. Nature Struct. Mol. Biol. 11, 36– -44 (2004).
48. Rothman, J. E. The protein machinery of vesicle budding and fusion. Protein Sci.
5, 185– -194 (1996).
49. Milosevic, I. et al. Plasmalemmal phosphatidylinositol-4,5-bisphosphate level
regulates the releasable vesicle pool size in chromaffin cells. J. Neurosci. 25,
2557– -2565 (2005).
50. Gong, L. W. et al. Phosphatidylinositol phosphate kinase type I g regulates
dynamics of large dense-core vesicle fusion. Proc. Natl Acad. Sci. USA 102,
5204– -5209 (2005).
51. Di Paolo, G. et al. Impaired PtdIns(4,5)P2synthesis in nerve terminals produces
defects in synaptic vesicle trafficking. Nature 431, 415– -422 (2004).
52. Lackner, M. R., Nurrish, S. J. & Kaplan, J. M. Facilitation of synaptic
transmission by EGL-30 Gqa and EGL-8 PLCb: DAG binding to UNC-13 is
required to stimulate acetylcholine release. Neuron 24, 335– -346 (1999).
53. Rhee, J. S. et al. b phorbol ester- and diacylglycerol-induced augmentation of
transmitter release is mediated by Munc13s and not by PKCs. Cell 108, 121– -133
54. Engqvist-Goldstein, A. E. & Drubin, D. G. Actin assembly and endocytosis: from
yeast to mammals. Annu. Rev. Cell Dev. Biol. 19, 287– -332 (2003).
55. Stefan, C. J., Audhya, A. & Emr, S. D. The yeast synaptojanin-like proteins
control the cellular distribution of phosphatidylinositol (4,5)-bisphosphate.
Mol. Biol. Cell 13, 542– -557 (2002).
56. Botelho, R. J., Scott, C. C. & Grinstein, S. Phosphoinositide involvement in
phagocytosis and phagosome maturation. Curr. Top. Microbiol. Immunol. 282,
1– -30 (2004).
57. Pizarro-Cerda, J. & Cossart, P. Subversion of phosphoinositide metabolism by
intracellular bacterial pathogens. Nature Cell Biol. 6, 1026– -1033 (2004).
58. Niebuhr, K. et al. Conversion of PtdIns(4,5)P2into PtdIns(5)P by the S. flexneri
effector IpgD reorganizes host cell morphology. EMBO J. 21, 5069– -5078 (2002).
59. Terebiznik, M. R. et al. Elimination of host cell PtdIns(4,5)P2by bacterial SigD
promotes membrane fission during invasion by Salmonella. Nature Cell Biol. 4,
766– -773 (2002).
60. Vergne, I. et al. Mechanism of phagolysosome biogenesis block by viable
Mycobacterium tuberculosis. Proc. Natl Acad. Sci. USA 102, 4033– -4038 (2005).
61. Ungewickell, A. et al. The identification and characterization of two
phosphatidylinositol-4,5-bisphosphate 4-phosphatases. Proc. Natl Acad. Sci.
USA 102, 18854– -18859 (2005).
62. Gaynor, E. C., Chen, C. Y., Emr, S. D. & Graham, T. R. ARF is required for
maintenance of yeast Golgi and endosome structure and function. Mol. Biol.
Cell 9, 653– -670 (1998).
63. De Matteis, M. A., Di Campli, A. & Godi, A. The role of the phosphoinositides
at the Golgi complex. Biochim. Biophys. Acta 1744, 396– -405 (2005).
64. Guo, S., Stolz, L. E., Lemrow, S. M. & York, J. D. SAC1-like domains of yeast
SAC1, INP52, and INP53 and of human synaptojanin encode
polyphosphoinositide phosphatases. J. Biol. Chem. 274, 12990– -12995 (1999).
65. Roy, A. & Levine, T. P. Multiple pools of phosphatidylinositol 4-phosphate
detected using the pleckstrin homology domain of Osh2p. J. Biol. Chem. 279,
44683– -44689 (2004).
66. Attree, O. et al. The Lowe’s oculocerebrorenal syndrome gene encodes a
protein highly homologous to inositol polyphosphate-5-phosphatase. Nature
358, 239– -242 (1992).
67. Birkeland, H. C. & Stenmark, H. Protein targeting to endosomes and
phagosomes via FYVE and PX domains. Curr. Top. Microbiol. Immunol. 282,
89– -115 (2004).
68. Zerial, M. & McBride, H. Rab proteins as membrane organizers. Nature Rev.
Mol. Cell Biol. 2, 107– -117 (2001).
69. Michell, R. H., Heath, V. L., Lemmon, M. A. & Dove, S. K. Phosphatidylinositol
3,5-bisphosphate: metabolism and cellular functions. Trends Biochem. Sci.,
52– -63 (2005).
70. Laporte, J. et al. A gene mutated in X-linked myotubular myopathy defines a
new putative tyrosine phosphatase family conserved in yeast. Nature Genet. 13,
175– -182 (1996).
71. Bolino, A. et al. Charcot-Marie-Tooth type 4B is caused by mutations in the
gene encoding myotubularin-related protein-2. Nature Genet. 25, 17– -19 (2000).
72. Li, S. et al. Mutations in PIP5K3 are associated with Francois-Neetens
mouchetee fleck corneal dystrophy. Am. J. Hum. Genet. 77, 54– -63 (2005).
73. McLaughlin, S. & Murray, D. Plasma membrane phosphoinositide organization
by protein electrostatics. Nature 438, 605– -611 (2005).
74. Bader, A. G., Kang, S., Zhao, L. & Vogt, P. K. Oncogenic PI3K deregulates
transcription and translation. Nature Rev. Cancer 5, 921– -929 (2005).
75. Halstead, J. R., Jalink, K. & Divecha, N. An emerging role for PtdIns(4,5)P2-
mediated signalling in human disease. Trends Pharmacol. Sci. 26, 654– -660
Acknowledgements We thank N. Borgese, V. Haucke and O. Cremona for
critical reading of the manuscript. We also thank B. Chang for providing the
fluorescence images in Fig. 1 and for her comments on the manuscript. DNA
constructs were gifts from H. Stenmark, T. Meyer and A. De Matteis. We
apologize to all the scientists whose original studies and reviews were not
quoted in our manuscript owing to space limitations. G.D.P. is funded by grants
from the National Institute of Health. P.D.C is funded by the HHMI and by
grants from the National Institute of Health, the Yale Center for Genomics and
Proteomics, the Yale/NIDA Neuroproteomics Center and the G. Harold and
Leila Y. Mathers Charitable Foundation.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare no competing financial interests.
Correspondence should be addressed to P.D.C. (email@example.com) and
NATURE|Vol 443|12 October 2006
© 2006 Nature Publishing Group