Download Sequential steps in clathrin-mediated synaptic vesicle endocytosis

312
Sequential steps in clathrin-mediated synaptic vesicle endocytosis
Lennart Brodin*, Peter Löw† and Oleg Shupliakov‡
Synaptic vesicles are recycled with remarkable speed and
precision in nerve terminals. A major recycling pathway
involves clathrin-mediated endocytosis at endocytic zones
located around sites of release. Different ‘accessory’ proteins
linked to this pathway have been shown to alter the shape
and composition of lipid membranes, to modify
membrane–coat protein interactions, and to influence actin
polymerization. These include the GTPase dynamin, the
lysophosphatidic acid acyl transferase endophilin, and the
phosphoinositide phosphatase synaptojanin. Protein
perturbation studies in living nerve terminals are now
beginning to link the actions of these proteins with
morphologically defined steps of endocytosis.
Addresses
The Nobel Institute for Neurophysiology, Department of Neuroscience,
Karolinska Institutet, S-171 77 Stockholm, Sweden;
*e-mail: [email protected]
† e-mail: [email protected]
‡ e-mail: [email protected]
Current Opinion in Neurobiology 2000, 10:312–320
0959-4388/00/$ - see front matter
© 2000 Elsevier Science Ltd. All rights reserved.
Abbreviations
AP
adaptor protein
BARS
brefeldin A-ADP-ribosylated substrate
CtBP
carboxy-terminal-binding protein
Dap
dynamin-associated protein
EH
Eps 15 homology
Eps 15
epidermal growth factor pathway substrate 15
GED
GTPase effector domain
GFP
green fluorescent protein
GTPase
guanosine triphosphate-binding protein
Hsc 70
heat shock cognate protein 70 kDa
LPAAT
lysophosphatidic acid acyl transferase
N-WASP
neural Wiscott-Aldrich syndrome protein
PH
pleckstrin homology
PRD
proline-rich domain
SH3
src homology 3
Introduction
Clathrin-mediated membrane budding is used to transport lipids and proteins from the plasma membrane and
the trans-Golgi network [1]. In nerve terminals, it takes
part in the recycling of synaptic vesicles following neurotransmitter release [2]. The clathrin-mediated
pathway of synaptic vesicle recycling was characterized
in the early 1970s in pioneering work by Heuser and
Reese, but its importance for synaptic function has been
tested only recently. A series of studies have shown that
perturbation of different proteins linked to clathrin
function results in loss or morphological change of
synaptic vesicles, as well as in an impairment of synaptic transmission [3–7,8••]. Other parallel pathways may
also participate in synaptic vesicle recycling, but their
significance is as yet unclear [9,10]. In this review, we
will first give a brief survey of the proteins implicated in
clathrin-mediated endocytosis, and then discuss their
possible actions at morphologically defined stages of
synaptic vesicle membrane retrieval.
Overview of proteins in clathrin-mediated
endocytosis
Clathrin-mediated endocytosis depends on two sets of proteins: those comprising the clathrin coat, and an array of
other proteins often referred to as ‘accessory’ proteins.
Many of these proteins have now been characterized in
considerable detail — work has included studies of their
crystal structure and of their interactions with other proteins and membrane phospholipids [11–15].
The basic building block of the clathrin coat is the threelegged clathrin triskelion (Figure 1), of which each leg
contains a heavy and a light chain of clathrin [1,11]. The
triskelia assemble into a lattice of hexagons and pentagons
(for a live model, see http://www.hms.harvard.edu/
news/clathrin), which attaches to the plasma membrane via
the tetrameric adaptor protein (AP) complex AP2
(Figure 1). The coat also contains a neuron-specific form of
a monomeric adaptor protein, AP180, which can stimulate
clathrin assembly and which appears to be critical for the
generation of synaptic vesicles with an homogenous size.
In Drosophila and C. elegans lacking AP180-like proteins,
nerve terminals still contain synaptic vesicles, but their
average size is larger and the size variability is increased
when compared with that of vesicles in wild-type controls
[5,6]. Moreover, microinjection studies in the squid giant
synapse [7] have shown that a peptide corresponding to
the clathrin assembly domain of AP180 causes a marked
depletion of synaptic vesicles, while the average size of the
remaining vesicles is increased [7].
The formation of an endocytic clathrin coat is thought to
begin with binding of AP2 to the membrane [1,15,16]. One
candidate membrane receptor for AP2 is the synaptic vesicle protein synaptotagmin, which binds AP2 via one of its
cytoplasmic C2 domains (C2B; for references see [2]). Via
a distinct binding site, AP2 also binds tyrosine-based
sequence motifs, which are known as sorting signals in the
context of receptor-mediated endocytosis [16] and which
occur in synaptic vesicle proteins [17••,18••]. Peptides containing a tyrosine-based motif have been shown to
strengthen the synaptotagmin–AP2 interaction [17••], suggesting a functional link between the two binding sites.
The formation of the clathrin coat also involves interactions between coat proteins and membrane phospholipids.
Phosphorylated inositol phospholipids (phosphoinositides)
appear to be particularly important [13,14]. For instance,
the plasma membrane targeting of AP2 depends on its
phosphoinositide binding site [19], and stimulation of
Sequential steps in clathrin-mediated synaptic vesicle endocytosis Brodin, Löw and Shupliakov
313
Figure 1
Clathrin
AP180
AP2
α2
Ear domain
β2
C
Hinge
domain
N
Heavy chain
Clathrin-binding domains
Diagram of proteins implicated in clathrinmediated synaptic vesicle endocytosis. Clathrin
occurs as a triskelion consisting of three heavy
and three light chains of clathrin. Triskelia
assemble together into a lattice of pentagons
and hexagons [1,11], which are connected to
the plasma membrane via the adaptor complex
AP2. The amino-terminal domain of the clathrin
heavy chain (N) binds to the hinge domain in
the AP2 α2 and β2 subunits. The ‘ear’ (or
appendage) domain of AP2 interacts with a
number of proteins including AP180, Eps 15,
amphiphysin and auxilin. AP2 interacts with
synaptotagmin, with tyrosine-based sequence
motifs via the µ2 subunit, and with
phosphoinositides via the α2 subunit (for
further details and references see
[1,2,11,12,19]). AP 180 is a monomeric
adaptor-like protein, which interacts with AP2
and stimulates clathrin assembly [2,5–7]. It also
interacts with phosphoinositides [13,14]. The
remaining proteins (i.e. the accessory proteins)
are usually not detectable in preparations of
clathrin-coated vesicles; this suggests that they
interact transiently with the coat. Only one form
is represented, although each protein is, as a
rule, found in multiple isoforms. Here, each
protein is represented by boxes indicating the
approximate size of the different domains.
Domains with enzymatic activity include the
GTPase domain in dynamin and the 5phosphatase and Sac1 homology domains of
synaptojanin [22,29•,30]. The LPAAT activity of
endophilin is likely to reside in the conserved
amino-terminal region of the protein [8••,31••].
Modular binding domains mediating
protein–protein interactions include SH3
domains which interact with distinct binding
sites within proline-rich domains (PRD), and
EH domains which interact with NPF (Asn-ProPhe) motifs. Multiple NPF repeats occur in the
carboxy-terminal of epsin. NPF repeats are also
σ2
Light chain
µ2
N
Dynamin
Synaptojanin
Amphiphysin
GTPase
PH GED PRD
Sac1 homology
5'-phosphatase
CC
PRD
SH3
Clathrin/AP2-binding
Endophilin
SH3
LPAAT activity
Syndapin
Epsin
Intersectin
Eps 15
SH3
ENTH
EH
DPW
EH
EH EH EH
CC
NPF
SH3
CC
SH3
SH3 SH3
SH3
GEF
PH
C2
DPF
Current Opinion in Neurobiology
present in syndapin and in one isoform of
synaptojanin. ENTH indicates a novel domain
termed the epsin amino-terminal homology
domain. CC indicates coiled-coil domains,
which may be involved in heteromerisation.
Distinct binding sites for AP2 and clathrin are
phosphoinositide synthesis can enhance coat formation [20], while ‘capping’ of phosphoinositides by
overexpressing phosphoinositide-binding proteins has an
inhibitory effect ([21]; see also the section on Uncoating,
below). Clathrin coat formation can also be enhanced by
phospholipase D, which may at least partly involve stimulation of phosphoinositide synthesis [13,14,20].
Phospholipase D was found to further enhance the stimulation of synaptotagmin–AP2 binding by tyrosine-based
motifs [17••], suggesting a possible cooperativity between
lipid signalling and protein–protein interactions.
Among the accessory proteins (Figure 1), dynamin has so
far been the most extensively studied. It was originally
linked to endocytosis through the temperature-sensitive
Drosophila mutant shibire, in which endocytosis is blocked
at the restrictive temperature (for references, see [22]).
Dynamin is a multi-domain protein (Figure 1) with a
GTPase domain, a phospholipid-binding pleckstrin
present in amphiphysin and Eps 15. Alternative
names are as follows; endophilin, SH3p4;
syndapin, pacsin (see [54]); intersectin,
Dap160 [23]; epsin, Ibp (for further details and
references see [1,2,11,12,15,47]).
homology (PH) domain, and a so-called GTPase effector
domain (GED; [22]; see also the section on Fission,
below). The carboxy-terminal consists of a proline-rich
domain (PRD), which interacts with src-homology (SH3)
domains of other accessory proteins including
amphiphysin, endophilin, dap160/intersectin and syndapin/pacsin [2,22,23,24•]. Dynamin has a tendency to
assemble into tetramers, which can polymerize into ringlike structures. When dynamin is incubated with spherical
liposomes these can be turned into narrow tubules surrounded by polymerized dynamin [22,25,26]. The
structure of dynamin polymers is altered by GTP hydrolysis [25,27] which can even cause vesiculation of the lipid
tubules [25]. Recently, amphiphysin was also shown to be
capable of tubulating liposomes — either alone, or together with dynamin [28•]. Two of the accessory proteins,
synaptojanin and endophilin, have been found to act as
lipid-metabolizing enzymes. Synaptojanin is a polyphosphoinositide phosphatase that dephosphorylates
314
Signalling mechanisms
Figure 2
Onset of clathrin-mediated endocytosis in the
endocytic zone of a central synapse.
(a,b) Three-dimensional reconstructions of the
plasma membrane at a synaptic release site in
a lamprey reticulospinal axon. The red area
corresponds to the active zone. The axon was
first stimulated at 20 Hz for 20 min in
physiological solution to partly deplete the
synaptic vesicle pool, and Ca2+ was then
rapidly removed to block endocytosis. The
specimen was maintained for 90 min in Ca2+free solution. (a) shows a synapse fixed
directly after incubation in Ca2+-free solution.
(b) shows a synapse from a part of the
specimen which had subsequently been
incubated for 120 s (at 5°C) in 2.6 mM Ca2+,
leading to the formation of numerous clathrincoated pits (purple) at the plasma membrane.
Synaptic vesicles were not included in the
reconstruction. Scale bar = 1 µm. In
(c), examples are shown of different stages of
clathrin-coated pit formation in the endocytic
zone. They have been placed in order based
on their relative abundance in synapses that
have been fixed at different time points after
addition of Ca2+. Scale bar = 50 nm.
(Reproduced from [37] with permission,
copyright by Cell Press.)
phosphoinositides at positions 3, 4 and 5 of the inositol ring
(see [13]), and it may thus regulate interactions of endocytic proteins with the plasma membrane [2,29•,30].
Endophilin acts as a lysophosphatidic acid acyl transferase
(LPAAT; [31••]). By converting lysophosphatidic acid into
phosphatidic acid, it may alter the biophysical properties of
the lipid bilayer (see below). Some accessory proteins have
been shown to interact with actin-regulating proteins. For
instance, dynamin binds profilin [32], and syndapin/pacsin
binds N-WASP [24•].
Localization of clathrin-mediated endocytosis
at the synapse
The proteins mentioned above are highly concentrated in
nerve terminals. Studies of terminals in Drosophila and
lamprey indicate that, within a terminal, different endocytic proteins are enriched in a plasma membrane region
surrounding the active zone [3,23,33,34]. This region
appears to define an ‘endocytic zone’ which typically
extends about a micron from the edge of an active zone.
Actin polymerization can be induced in this region by activation of GTPases with GTPγS [34]. The rho-type
GTPase-activating protein ‘still life’ has also been localized to this zone [35].
Clathrin-coated pits appear in the endocytic zone (see
below and Figure 2) shortly after exocytosis [2], suggesting
that the synaptic vesicle membrane and its protein components quickly move to this region. The targeting of clathrin
coat formation to the endocytic zone is, however, not
absolute, as coated pits occasionally occur within the active
zone. Their number can increase after manipulations
which cause massive depletion of synaptic vesicles, suggesting that docked synaptic vesicles may limit coat
formation in the active zone (O Shupliakov, P Löw,
L Brodin, unpublished observations). The extent of the
endocytic zone in the plasma membrane does not appear
to be fixed. After certain manipulations that inhibit endocytosis, coated pits can occur in an area extending several
microns from the active zone [4].
Sequential steps in clathrin-mediated synaptic vesicle endocytosis Brodin, Löw and Shupliakov
Figure 3
315
endocytic hot spots identified at the plasma membrane of
non-neuronal cells [36•]
Onset of endocytosis
Clathrin-mediated synaptic vesicle endocytosis is normally
coupled with exocytosis, but the two processes can be
separated experimentally. In the experiment shown in
Figure 2, intense action-potential stimulation was first
applied to deplete synaptic vesicles (not shown in the figure), and endocytosis was then blocked by removal of
extracellular Ca2+ [37]. Under these conditions, coated
structures are virtually absent (Figure 2a). Addition of Ca2+
induces the synchronous formation of coated pits in the
endocytic zone (Figure 2b). The onset of clathrin coat formation thus requires Ca2+ (low micromolar concentrations
are sufficient [37]), but spike-evoked influx is not needed.
It also appears to depend on ATP, as lowering of ATP levels causes vesicle depletion and an appearance of plasma
membrane invaginations ([38]; O Shupliakov, L Brodin,
unpublished observations). In contrast, clathrin coat formation in vitro is not ATP-dependent [26].
Inhibition of synaptic vesicle endocytosis at the stage of shallow
clathrin-coated pits by microinjection of endophilin antibodies.
(a) Electron micrograph of a synapse in a lamprey reticulospinal axon
that has been injected with endophilin antibodies and maintained in
low calcium solution (0.1 mM Ca2+ and 4 mM Mg2+) without
stimulation for 60 min. (b) A synapse in an axon that was stimulated in
a normal Ringer´s solution at 5 Hz for 30 min after the injection of
endophilin antibodies. Note the ‘pocket-like’ membrane expansions
(arrows) at the margin of the synaptic area and the appearance of
numerous shallow coated pits (small arrows). Scale bar = 0.5 µm.
(c,d) The inhibition of the invagination of clathrin-coated pits depends
on the concentration of endophilin antibody. (c1–3) show examples of
clathrin-coated pits from synapses exposed to (1) high, (2)
intermediate, and (3) low antibody concentrations, respectively. The
relative antibody concentration was estimated from the antibodylinked fluorescence in the axon after the injection. Scale bar = 0.2 µm.
(d) Quantitative analysis was made by expressing the ‘curvature
index’ (the actual length of the coated membrane divided by the
distance between the margins of the coated pit) as a function of the
antibody-linked fluorescence. (Reproduced from [8••] with permission,
copyright by Cell Press.)
Clathrin-coated pits can also appear at invaginations of the
plasma membrane that occur after inhibition of endocytosis. In this case their features may differ from those formed
in the intact endocytic zone (see section on Fission,
below), consistent with a functional specialization of the
latter. The synaptic endocytic zone may correspond to the
After addition of Ca2+, sequential stages of coated-pit formation (Figure 2c) can be identified by their relative
abundance at different time points ([37]; see also references in [1,2]). The first stage (Figure 2c1) consists of a
coated membrane patch with slight curvature. The second
(Figure 2c2) is an invaginated coated pit with a broad base.
The third (Figure 2c3) is an invaginated coated pit with a
narrow neck. A ring-like structure is occassionally seen
around the narrow neck which probably defines a fourth
stage (Figure 2c4), although its low abundance makes the
location in the sequence somewhat uncertain. The next
stage is likely to be represented by a free coated vesicle
but, as further discussed below, this stage appears to be
very transient. Data from protein perturbation studies have
shown that each stage (Figure 2c1–4) can be retarded by
interfering with specific proteins, suggesting that these
distinct morphological states correlate with intermediates
in a molecular cascade.
Intermediates in coated-pit formation:
invagination of the coated membrane
Microinjection studies in the lamprey giant synapse suggest
that the proteins of the clathrin coat cannot alone generate
an invaginated coated pit, but that accessory factors, including endophilin, appear to be required. After presynaptic
microinjection of anti-endophilin antibodies [8••], stimulation causes depletion of synaptic vesicles, along with a
massive accumulation of shallow coated pits in the endocytic zone (Figure 3a,b). The invagination process appears to
be inhibited in a concentration-dependent manner, as the
depth of the coated pits decreases with increasing antibody
concentration (Figure 3c,d1–3). The precise mechanism
underlying this effect is not yet clear. In vitro formation of
clathrin coats from brain cytosol is not affected by immunodepletion of endophilin [8••], indicating that endophilin
acts on the membrane, rather than on coat assembly. One
316
Signalling mechanisms
Figure 4
may be indirectly influenced by endophilin, perhaps via the
cytoskeleton. Moreover, binding partners of the endophilin
SH3 domain, including dynamin and synaptojanin, could
also be involved (see below).
Narrowing of the neck region
Clathrin-coated pits trapped at late stages preceeding fission.
(a) Clathrin-coated pit with a long tubular neck surrounded by
electron-dense rings, which was induced by incubating synaptic
membranes with rat-brain cytosol, ATP and GTPγS. Similar structures,
which are connected to the plasma membrane at synapses, can be
induced by presynaptic microinjection of GTPγS [34]. Scale
bar = 30 nm. (Reprinted by permission from [55]. Copyright [1995]
Macmillan Magazines Ltd.) (b) Invaginated clathrin-coated pits at the
plasma membrane adjacent to an active zone in a giant reticulospinal
axon. The axon had been injected with a glutathione-S-transferase
fusion protein containing the SH3 domain of amphiphysin and
subjected to action potential stimulation. Scale bar = 0.1 µm.
(Reproduced from [4] with permission, copyright by AAAS.)
(c) Invaginated clathrin-coated pits at endosome-like plasma
membrane invaginations in a synaptic region of a reticulospinal axon.
The axon had been injected with endophilin antibodies. Note that
shallow coated pits occur at the plasma membrane adjacent to the
active zone (located just to the left of the image field), while
invaginated coated pits are only seen at the invaginated part of the
plasma membrane. Scale bar = 0.2 µm. (Reproduced from [8••] with
permission, copyright by Cell Press.)
possibility is that the conversion of lysophosphatidic acid to
phosphatidic acid by endophilin promotes the generation of
negative membrane curvature at the edges of the coated pit
(Figure 2c1–2), as proposed by Schmidt et al. [31••], although
these authors pointed out that the LPAAT activity may also
induce other effects such as an altered phosphoinositide
metabolism. The general importance of the plasma membrane composition for coated-pit invagination has been
supported by the finding that cholesterol depletion inhibits
this process [39,40]. The surface tension of the plasma
membrane is another potentially important factor which
The process subsequent to invagination — the narrowing
of the neck of the invaginated coated pit
(Figure 1c2–3) — also appears to depend on mechanisms
extrinsic to the clathrin coat. Several factors may be
involved. For instance, perturbation of SH3-domain interactions in permeabilized cells can affect the narrowing of
the neck of the coated pit, as judged from the altered
accessibility to large, but not to small, tracer molecules
[41]. Microinjection of actin toxins can increase the abundance of coated pits with a wide neck at stimulated
synapses (O Shupliakov et al., unpublished data). Actin
disruption has been found to affect the formation of
clathrin-coated pits at the plasma membrane of other cell
types, but the effects are variable [1,2], and the exact role
of actin in coated pit formation remains unclear. A general
involvement of actin in endocytosis has, however, been
supported by imaging studies of GFP-coupled actin in
mast cells, which indicate that actin polymerization occurs
at emerging endocytic membrane invaginations [42•]. The
polymerization of actin in the endocytic zone of synapses
appears to be coupled with synaptic activity, possibly
through signalling via GTPases [34] and phosphoinositides [29•]. Both types of signal can regulate
N-WASP/arp2,3-mediated actin polymerization ([43,44•];
see also Transport of the newly retrieved vesicle, below).
Fission
The fission of the neck of the coated pit is sensitive to a
variety of perturbations. In nerve terminals of the shibire
mutant, invaginated endocytic pits with narrow necks surrounded by an electron-dense ring accumulate at the
restrictive temperature (see [22] for references). The localization of the shibire mutation to dynamin’s GTPase
domain, along with observations that GTP hydrolysis can
alter the structure of dynamin polymers, suggested that
GTP hydrolysis by dynamin may drive fission [22,25–27].
However, recent studies of dynamin argue against this
model. The GED of dynamin appears to mediate the
increased GTPase activity which occurs during oligomerization [45•]. When the assembly-stimulation of the
GTPase activity is perturbed by mutations in the GED,
receptor-mediated endocytosis is found to be enhanced
rather than inhibited [45•]. While this finding does not
explain the role of dynamin’s GTPase activity, it shows
that the GTPase activity is not rate-limiting for endocytosis. It is possible that dynamin, like ‘conventional’
GTPases, acts as a regulator which interacts with downstream effectors in its GTP-bound state [45•].
Another dynamin-related intermediate — a clathrin-coated pit with an elongated neck surrounded by multiple
electron-dense rings (Figure 4a) — can be trapped with
Sequential steps in clathrin-mediated synaptic vesicle endocytosis Brodin, Löw and Shupliakov
the slowly hydrolysable GTP analog GTPγS. This intermediate was first observed in vitro [22], but it also occurs in
stimulated synapses after microinjection of GTPγS [34].
Although the mechanisms underlying the induction of this
intermediate are unclear (other GTPases than dynamin
may be involved), it has provided insight into the composition of the fission machinery. Immunocytochemical
studies suggest that the GTPγS-dependent rings contain
both dynamin, amphiphysin [26,28•], and endophilin
(Figure 4b; [8••]). The interaction between dynamin and
amphiphysin appears to be essential for fission [4,11], as
microinjection of proteins or peptides which inhibit this
interaction causes synaptic vesicle depletion along with a
massive accumulation of invaginated coated pits with narrow necks (Figure 4b). Electron-dense rings are not seen
after this perturbation, indicating that SH3 domain interactions contribute to ring formation [4].
A trapping of similar deeply invaginated coated pits can
occur also after perturbation of endophilin [8••]. In the antibody-injection studies discussed above, plasma membrane
invaginations sometimes extended outside the endocytic
zone. Tracing of these invaginations showed that they
sometimes contain invaginated coated pits with narrow
necks (Figure 4c) differing from the shallow coated pits in
the endocytic zone. These observations appear to converge
with studies of synaptic-like vesicle formation in permeabilized PC12 cells ([31••]; see also [41]). In this assay, both
endophilin and dynamin are required for vesicle formation.
Endophilin is only active when its SH3 domain is intact,
consistent with a role of dynamin–endophilin interactions
in fission. This possibility, however, remains to be tested
experimentally. Vesicle formation in vitro was also found to
be sensitive to treatments interfering with lipid metabolism
catalyzed by the LPAAT activity [31••]. The Golgi trafficking protein CtBP/BARS (carboxy-terminal-binding
protein/brefeldin A-ADP-ribosylated substrate) has also
been shown to possess LPAAT activity. Golgi tubules incubated with increasing concentrations of CtBP/BARS
exhibited an increased occurrence of invaginations which
appeared to proceed to vesiculation [46••].
Thus, recent studies indicate that fission of clathrin-coated
pits depends on a coordinated action of a set of proteins
that includes dynamin, amphiphysin, endophilin, and perhaps others. The exact sequence of events during fission
remains to be elucidated.
Uncoating
It has not yet been possible to track the exact path of the
endocytic vesicle after it has left the plasma membrane in
the endocytic zone. Most likely, a rapid uncoating occurs, as
free coated vesicles are rarely seen in stimulated synapses.
The uncoating reaction involves disassembly of the clathrin
coat by the uncoating ATPase heat shock cognate protein
70 kD (hsc 70) and auxilin [12]. Synaptojanin 1 has also
been proposed to contribute to uncoating [29•]. Using an
in vitro coating assay, cytosol from synaptojanin 1-deficient
317
Figure 5
Established role
AP2
AP180
Clathrin
Possible role
Synaptotagmin
Tyrosine-based motifs
Endophilin
Actin
Intersectin
Dynamin
Amphiphysin
Dynamin
hsc70
Auxilin
Endophilin
Amphiphysin
Synaptojanin
Actin
Current Opinion in Neurobiology
Schematic summary of the established and putative sites of action of
proteins in synaptic vesicle endocytosis. The coat proteins (AP2,
AP180, clathrin) are recruited to the membrane, probably by
interaction with synaptotagmin and possibly also with tyrosine-based
motifs in synaptic vesicle proteins [1,16,17••]. The invagination of the
coated membrane depends on endophilin [8••]. Narrowing of the neck
region may involve several factors, including actin, intersectin, dynamin,
and amphiphysin [25–27,28•,41]. Fission depends on dynamin,
probably in cooperation with other proteins such as amphiphysin and
endophilin [4,8••,22]. Uncoating depends on Hsc 70 and auxilin, and
probably also synaptojanin [12,29•]. Actin may participate in the
transport of the post-endocytic vesicle (see also [42•,49•]).
mice was shown to support coat formation more effectively
than cytosol from wild-type mice, suggesting that synaptojanin 1 can act as a negative regulator of membrane–coat
protein interactions. Electron microscopic analysis showed
318
Signalling mechanisms
that nerve terminals of synaptojanin 1-deficient mice contained synaptic vesicles, and comparatively few coated
vesicles were present. The relative proportion of coated
vesicles versus synaptic vesicles was, however, higher in
knockout mice than in wild-type mice, consistent with a
partially retarded vesicle uncoating [29•].
Transport of the newly retrieved vesicle
The newly uncoated endocytic vesicle may return directly to
the release site as a fully functional synaptic vesicle, or it may
pass through a secondary endosomal fusion-and-budding
step [2]. While evidence for synaptic vesicle formation from
both endosomal and plasma membrane compartments have
been obtained, the single budding step pathway has recently been favored as the main physiological route (for
discussion of this topic see [2,5,15,37,47,48]). Actin-based
transport has been implicated in the transport of endocytic
vesicles in other cell types [42•,49•], and it may well play a
similar role in the transport of synaptic vesicles between the
endocytic zone and the synaptic vesicle cluster.
Conclusions
The results obtained so far in mutation and microinjection
studies have allowed a first glimpse of the sequential actions
of endocytic proteins in living nerve terminals. The continued use of these approaches, and their combination with
high resolution immunolabeling, should help to clarify the
temporal aspects of the endocytic cascade. It is already evident, however, that endocytosis cannot be described as a
simple chain of proteins acting in a strict order (see Figure 5).
For instance, endophilin appears to act at more than one
endocytic stage, and this may well apply to other accessory
proteins such as dynamin and amphiphysin [22,28•,50].
Some proteins, like synaptotagmin, appear to play critical
roles in both exo- and endocytosis [2,17••]. Moreover, the
efficient sorting of synaptic vesicle proteins at each vesicle
cycle is likely to depend on a close functional coupling
between exo- and endocytic proteins [6,15,18••,47,51]. As
well as the clarification of the sequence of protein actions,
fundamental problems such as the spatial and temporal
reorganization of membrane phospholipids during endocytosis remain to be addressed. Another important problem
concerns the regulation of the endocytic molecular
machinery [52], which can now be addressed with dynamic
protein imaging methods using GFP [36•,42•,53•]. These
methods should also allow investigators to test the possible
role of alternative pathways in synaptic vesicle recycling.
Acknowledgements
We are indebted to Dr P De Camilli for discussions, and to Dr O Kjaerulff
for comments on the manuscript.
References and recommended reading
Papers of particular interest, published within the annual period of review,
have been highlighted as:
• of special interest
•• of outstanding interest
1.
Schmid SL: Clathrin-coated vesicle formation and protein sorting:
an integrated process. Annu Rev Biochem 1997, 66:511-548.
2.
Cremona O, De Camilli P: Synaptic vesicle endocytosis. Curr Opin
Neurobiol 1997, 7:323-330.
3.
Gonzalez-Gaitan M, Jackle H: Role of Drosophila alpha-adaptin in
presynaptic vesicle recycling. Cell 1997, 88:767-776.
4.
Shupliakov O, Löw P, Grabs D, Gad H, Chen H, David C, Takei K,
De Camilli P, Brodin L: Synaptic vesicle endocytosis impaired by
disruption of dynamin-SH3 domain interactions. Science 1997,
276:259-263.
5.
Zhang B, Koh YH, Beckstead RB, Budnik V, Ganetzky B, Bellen HJ:
Synaptic vesicle size and number are regulated by a clathrin adaptor
protein required for endocytosis. Neuron 1998, 21:1465-1475.
6.
Nonet ML, Holgado AM, Brewer F, Serpe CJ, Norbeck BA, Holleran J,
Wei LP, Hartwieg E, Jorgensen EM, Alfonso A: UNC-11, a
Caenorhabditis elegans AP180 homologue, regulates the size
and protein composition of synaptic vesicles. Mol Biol Cell 1999,
10:2343-2360.
7.
Morgan JR, Zhao X, Womack M, Prasad K, Augustine GJ, Lafer E: A
role for the clathrin assembly domain of AP180 in synaptic vesicle
endocytosis. J Neurosci 1999, 19:10201-10212.
8.
••
Ringstad N, Gad H, Löw P, Di Paolo G, Brodin L, Shupliakov O,
De Camilli P: Endophilin/SH3P4 is required for the transition from
early to late stages in clathrin-mediated synaptic vesicles
endocytosis. Neuron 1999, 24:143-154.
Presynaptic microinjection of antibodies to endophilin is shown to inhibit
synaptic vesicle recycling at the stage of shallow clathrin-coated pits. In vitro
experiments show that endophilin is not part of the clathrin coat, and suggest that it is a functional binding partner of dynamin. The results indicate
that endophilin is part of a biochemical machinery which acts from early
endocytic stages to fission [see also 31••,46••].
9.
Fesce F, Meldolesi J: Peeping at the vesicle kiss. Nat Cell Biol
1999, 1:E3-E4.
10. Klingauf J, Kavalali ET, Tsien RW: Kinetics and regulation of fast
endocytosis at hippocampal synapses. Nature 1998, 394:581-585.
11. Marsh M, McMahon HT: The structural era of endocytosis. Science
1999, 285:215-220.
12. Ungewickell E: Wrapping the package. Proc Natl Acad Sci USA
1999, 96:8809-8810.
13. Martin TF: Phosphoinositide lipids as signaling molecules: common
themes for signal transduction, cytoskeletal regulation, and
membrane trafficking. Annu Rev Cell Dev Biol 1998, 14:231-264.
14. Corvera S, D’ Arrigo S, Stenmark H: Phosphoinositides in
membrane traffic. Curr Opin Cell Biol 1999, 11:460-465.
15. Kelly RB: Deconstructing membrane traffic. Trends Cell Biol 1999,
9:M29-M32.
16. Kirchhausen T, Bonifacino JS, Riezman H: Linking cargo to vesicle
formation: receptor tail interactions with coat proteins. Curr Opin
Cell Biol 1997, 9:488-495.
17.
••
Haucke V, De Camilli P: AP-2 recruitment to synaptotagmin
stimulated by tyrosine-based endocytic motifs. Science 1999,
285:1268-1271.
Peptides with a tyrosine-based motif (a motif which occurs in the cytoplasmic tail of many transmembrane receptors and also in the synaptic vesicle
protein SV2A) are shown to stimulate binding of AP2 to synaptotagmin and
to enhance AP2 recuitment to membranes. The results suggest a mechanism by which synaptotagmin and proteins containing tyrosine-based motifs
cooperate in the nucleation of the clathrin coat. It is unclear, however,
whether SV2A participates in synaptic vesicle endocytosis [18•].
18. Jan R, Goda Y, Geppert M, Missler M, Sudhof TC: SV2A and SV2B
•• function as redundant Ca2+ regulators in neurotransmitter
release. Neuron 1999, 24:1003-1016.
The study shows that SV2A- and SV2A/SV2B double knockout mice exhibit
seizures and die postnatally, and that synaptic transmission shows sustained
increases that can be reversed by membrane-permeable calcium buffers. As
judged from electron microscopic analysis, however, synapses have a normal
structure. The results indicate that SV2 participates in presynaptic calcium
regulation, but they do not support an essential role of the tyrosine-based
motifs in SV2A in synaptic vesicle endocytosis (see also [17••]).
19. Gaidarov I, Keen JH: Phosphoinositide-AP-2 interactions required
for targeting to plasma membrane clathrin-coated pits. J Cell Biol
1999, 146:755-764.
20. Arneson LS, Kunz J, Anderson RA, Traub LM: Coupled inositide
phosphorylation and phospholipase D activation initiates clathrincoat assembly on lysosomes. J Biol Chem 1999, 274:17794-17805.
Sequential steps in clathrin-mediated synaptic vesicle endocytosis Brodin, Löw and Shupliakov
21. Jost M, Simpson F, Kavran JM, Lemmon MA, Schmid SL:
Phosphatidylinositol [4,5] bisphosphate is required for endocytic
coated vesicle formation. Curr Biol 1998, 8:1399-1402.
22. Schmid SL, McNiven MA, De Camilli P: Dynamin and its partners: a
progress report. Curr Opin Cell Biol 1998, 10:504-512.
23. Roos J, Kelly RB. Dap160, a neural-specific Eps15 homology and
multiple SH3 domain-containing protein that interacts with
Drosophila dynamin. J Biol Chem 1998, 273:9108-9119.
24. Qualmann B, Roos J, DiGregorio PJ, Kelly RB: Syndapin I, a synaptic
•
dynamin-binding protein that associates with the neural WiskottAldrich syndrome protein. Mol Biol Cell 1999, 10:501-513.
This study shows that syndapin (also termed pacsin [54]) binds endocytic
proteins, including dynamin and synaptojanin, and that it also binds the actin
regulator N-WASP. These observations suggest a molecular link between
the endocytic budding machinery and the regulation of the actin cytoskeleton (see also [32,44•]).
25. Sweitzer SM, Hinshaw JE: Dynamin undergoes a GTP-dependent
conformational change causing vesiculation. Cell 1998,
93:1021-1029.
26. Takei K, Haucke V, Slepnev V, Farsad K, Salazar M, Chen H, De Camilli P:
Generation of coated intermediates of clathrin-mediated endocytosis
on protein-free liposomes. Cell 1998, 94:131-141.
27.
Stowell MHB. Marks B, Wigge P, McMahon HT: Nucleotidedependent conformational changes in dynamin: evidence for a
mechanochemical molecular spring. Nat Cell Biol 1999, 1:27-32.
28. Takei K, Slepnev VI, Haucke V, De Camilli P: Functional partnership
•
between amphiphysin and dynamin in clathrin-mediated
endocytosis. Nat Cell Biol 1999, 1:33-39.
This study shows that amphiphysin 1 can bind lipid membranes, that it can
transform spherical liposomes into narrow tubules, and that it can co-assemble with dynamin 1 and enhance the lipid-fragmenting activity of dynamin 1
in the presence of GTP. The results indicate a close functional relationship
between amphiphysin and dynamin (see also [4,11]), and they suggest that
both proteins may contribute to the generation of bilayer curvature during
endocytic membrane budding.
29. Cremona O, Di Paolo G, Wenk MK, Luthi A, Kim W, Takei K, Daniell L,
•
Nemoto Y, Shears SB, Flavell RA et al.: Essential role of
phosphoinositide metabolism in synaptic vesicle recycling. Cell
1999, 99:179-188.
Mice lacking synaptojanin 1 are shown to exhibit neurological deficits and
die shortly after birth. The level of phosphatidylinositol-4,5-bisphosphate is
increased in cultured neurons from knockout mice, and the generation of
clathrin coats on protein-free liposomes (in the presence of ATP and GTPγS)
is enhanced when normal cytosol is replaced with cytosol from knockout
mice. The relative proportion of coated vesicles versus synaptic vesicles in
nerve terminals of knockout mice is higher than in wild-type mice. A model is
proposed in which synaptojanin facilitates the shedding of the clathrin coat
after endocytosis.
30. Guo S, Stolz LE, Lemrow SM, York JD: SAC1-like domains of yeast
SAC1, INP52, and INP53 and of human synaptojanin encode
polyphosphoinositide phosphatases. J Biol Chem 1999,
274:12990-12995.
31. Schmidt A, Wolde M, Thiele C, Fest W, Kratzin H, Podtelejnikov AV,
•• Witke W, Huttner WB, Söling HD: Endophilin I mediates synaptic
vesicle formation by transfer of arachidonate to lysophosphatidic
acid. Nature 1999, 401:133-141.
Endophilin 1 is shown to act as a LPAAT which converts lysophosphatidic
acid to phosphatidic acid, and to be essential for synaptic vesicle-like formation in permeabilized PC12 cells. The LPAAT activity is retained when the
SH3 domain of endophilin is removed, but the capacity to support vesicle
formation is lost. A model is proposed in which endophilin, via its SH3domain interactions with dynamin, is recruited to sites of vesicle formation
where it contributes to the generation of negative membrane curvature via its
LPAAT activity [see also 8••,46••].
32. Witke W, Podtelejnikov AV, Di Nardo A, Sutherland JD, Gurniak CB,
Dotti C, Mann M: In mouse brain profilin I and profilin II associate
with regulators of the endocytic pathway and actin assembly.
EMBO J 1998, 17:967-976.
33. Roos J, Kelly RB: The endocytic machinery in nerve terminals
surrounds sites of exocytosis. Curr Biol 1999, 9:1411-1414.
34. Gustafsson J, Shupliakov O, Takei K, Löw P, De Camilli P, Brodin L:
GTPγS induces an actin matrix associated with coated endocytic
intermediates in presynaptic regions. Soc Neurosci Abstr 1998,
327:19.
319
35. Sone M, Hoshino M, Suzuki E, Kuroda S, Kaibuchi K, Nakagoshi H,
Saigo K, Nabeshima Y, Hama C: Still life, a protein in synaptic
terminals of Drosophila homologous to GDP–GTP exchangers.
Science 1997, 275:543-547.
36. Gaidarov I, Santini F, Warren RA, Keen JH: Spatial control of coated
•
pit dynamics in living cells. Nat Cell Biol 1999, 1:1-7.
A fusion protein consisting of green fluorescent protein and clathrin light
chain is used to study dynamic aspects of endocytosis in COS-1 cells and
other mammalian cell lines. The mobility of invaginating clathrin-coated pits
is shown to be limited, but increases after disruption of the actin cytoskeleton. Coated pits are found to form repeatedly at defined sites, while other
regions are excluded.
37.
Gad H, Löw P, Zotova E, Brodin L, Shupliakov O: Dissociation
between Ca2+-evoked synaptic vesicle exocytosis and clathrinmediated endocytosis at a central vertebrate synapse. Neuron
1998, 21:607-616.
38. Brodin L, Bakeeva L, Shupliakov O: Presynaptic mitochondria and
the temporal pattern of neurotransmitter release. Phil Trans R
Soc Lond B Biol Sci 1999, 354:365-372.
39. Rodal SK, Skretting G, Garred O, Vilhardt F, van Deurs B, Sandvig K:
Extraction of cholesterol with methyl-beta-cyclodextrin perturbs
formation of clathrin-coated endocytic vesicles. Mol Biol Cell
1999, 10:961-974.
40. Subtil A, Gaidarov I, Kobylarz K, Lampson MA, Keen JH, McGraw TE:
Acute cholesterol depletion inhibits clathrin-coated pit budding.
Proc Natl Acad Sci USA 96:6775-6780.
41. Simpson F, Hussain NK, Qualmann B, Kelly RB, Kay BK,
McPherson P, Schmid SL: SH3-domain-containing proteins
function at distinct steps in clathrin-coated vesicle formation. Nat
Cell Biol 1999, 1:119-124.
42. Merrifield CJ, Moss SE, Ballestrem C, Imhof BA, Giese G,
•
Wunderlich I, Almers W: Endocytic vesicles move at the tips of
actin tails in cultured mast cells. Nat Cell Biol 1999, 1:72-74.
A fusion protein consisting of green fluorescent protein and actin is used to
study the dynamics of actin filaments during endocytosis in mast cells. Actin
‘tails’ formed during endocytosis are shown to follow the endocytic vesicle
as it moves into the cell.
43. Machesky LM, Insall RH: Signalling to actin dynamics. J Cell Biol
1999, 146:267-272.
44. Rohatgi R, Ma L, Miki H, Lopez M, Kirchhausen T, Takenawa T,
•
Kirschner MW: The interaction between N-WASP and the Arp2/3
complex links Cdc42-dependent signals to actin assembly. Cell
1999, 97:221-231.
N-WASP, a binding partner of the rho family GTPase Cdc42, is shown to to
stimulate actin polymerization via direct interaction with the actin-nucleating
Arp2/3 complex. The stimulating effect of N-WASP is shown to be greatly
enhanced by Cdc42 and by phosphatidylinositol-4,5-bisphosphate. These
results indicate that N-WASP–Arp2/3 comprise a mechanism that directly
connects GTPase and phosphoinositide signalling to the stimulation of actin
polymerization (see also [24•]).
45. Sever S, Muhlberg AB, Schmid SL: Impairment of dynamin’s GAP
•
domain stimulates receptor-mediated endocytosis. Nature 1999,
398:481-486.
The GTPase effector domain (GED) of dynamin is shown to function as a
GTPase-activating protein (GAP) which coordinates dynamin self-assembly
with increased GTPase activity. Mutations causing a defect in assemblystimulated GTPase activity are shown to stimulate, rather than inhibit, endocytosis. This finding is not compatible with previous models of dynamin
function which suggested that endocytic membrane fission is driven by
dynamin’s GTPase activity. A model is proposed in which dynamin–GTP is
in an active conformation which recruits the fission machinery that is required
for vesicle formation.
46. Welgert R, Silletta MG, Spano S, Turacchio G, Cercola C, Colanzi A,
•• Senatore S, Mancini R, Polischuk EV, Salmona M et al.: CtBP/BARS
induces fission of Golgi membranes by acylating
lysophosphatidic acid. Nature 1999, 402:429-433.
CtBP/BARS is shown to act as a lysophosphatidic acid acyl transferase, and
to be an essential component of the fission machinery operating at Golgi tubular networks. CtBP/BARS-induced fission is shown to be preceeded by the
formation of constricted sites in Golgi tubules which appear to result from local
changes in the membrane lipid composition. The study indicates an important
role for lipid metabolic pathways in membrane fission (see also [8••,30••]).
47.
Huttner WB: Protein and lipid sorting in the secretory and
endocytic pathways - receptors and mechanisms. Semin Cell Dev
Biol 1998, 9:491-492.
320
Signalling mechanisms
48. Murthy VN, Stevens CF: Synaptic vesicles retain their identity
through the endocytic cycle. Nature 1998, 392:497-501.
49. Rozelle AL, Machesky LM, Yamamoto M, Driessens MHE, Insall RH,
•
Roth MG, Luby-Phelps K, Marriott G, Hall A, Yin HL:
Phosphatidylinositol 4,5-bisphosphate induces actin-based
movement of raft-enriched vesicles through WASP-Arp2/3.
Current Biol 2000, 10:311-320.
The study provides evidence for a role of sphingolipid–cholesterol rafts as
preferred platforms for membrane-linked actin polymerization, and for an
involvement of actin tails in the transport of endocytic and Golgi-derived
vesicles (see also [42•]).
52. Slepnev VI, Ochoa GC, Butler MH, Grabs D, Camilli PD: Role of
phosphorylation in regulation of the assembly of endocytic coat
complexes. Science 1998, 281:821-824.
53. Sankaranarayanan S, Ryan TA: Real-time measurement of vesicle
•
SNARE recycling in synapses of the central nervous system. Nat
Cell Biol 2000, 2:197-204.
A pH-sensitive GFP linked to synaptobrevin (also known as vesicle-associated membrane protein, or VAMP) was used to monitor vesicle recycling in
cultured hippocampal neurons. The measured time courses are consistent
with a single, presumably clathrin-mediated, endocytic pathway.
50. Ramjaun A, Phillie J, de Heuvel E, McPherson PS: The N terminus of
amphiphysin II mediates dimerization and plasma membrane
targeting. J Biol Chem 1999, 274:19785-19791.
54. Plohmann M, Lange R, Vopper G, Cremer H, Heinlein VA, Scheff S,
Baldwin SA, Leitges M, Cramer M, Paulsson M, Barthels D: PACSIN,
a brain protein that is upregulated upon differentiation into
neuronal cells. Eur J Biochem 1998, 256:210-211.
51. Okamoto M, Schoch, Südhof T: EHSH1/Intersectin, a protein that
contains EH and SH3 domains and binds to dynamin and
SNAP-25. J Biol Chem 1999, 26:18446-18454.
55. Takei K, McPherson PS, Schmid SL, De Camilli P: Tubular
membrane invaginations coated by dynamin rings are induced by
GTP-gamma S in nerve terminals. Nature 1995, 374:186-190.