Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
TRANSMITTER RELEASE AT THE NEUROMUSCULAR JUNCTION
Thomas L. Schwarz
Program in Neurobiology, Children’s Hospital and Department of NeurobiologyHarvard Medical School, Boston, Massachusetts 02115, USA
I.
INTE
NEUR
DOI:
I
R
1
ntroduction
NATIONAL REVIEW OF 105OBIOLOGY, VOL. 75
Copyright 2006, Elsev
All rights re
0.1016/S0074-7742(06)75006-1 0074-7742/06
II.
P hysiological Properties of Transmitter ReleaseIII.
E xperimental Advantages and Limitations of the Fly NMJA. C
a2þ MeasurementsB. D
ye LoadingC. A
lternative Measures of ExocytosisIV.
H ow Do Vesicles Fuse: Full Fusion or Kiss-and-Run?V.
C ore Machinery of Exocytosis: Syntaxin, VAMP/Synaptobrevin, and SNAP-25A. S
ynaptobrevins in the FlyB. T
etanus Toxin as a Synapse-Silencing ReagentC. S
yntaxin1D. S
NAP-25 and SNAP-24E. S
NARE Proteins Act Late in the Vesicle CycleF. S
ummary of SNARE Proteins in DrosophilaVI.
V esicular ATPase and Membrane FusionVII.
N SF and the Resetting of the SNARE MachineryVIII.
S ynaptotagmin and the Regulation of Transmitter ReleaseA. S
ynaptotagmin and Its Binding PropertiesB. S
ynaptotagmin Phenotypes at the NMJC. O
ther Functions of SynaptotagminD. M
ultiple Synaptotagmins in the FlyE. S
ummary of Synaptotagmin Function at the Fly NMJIX.
E xocyst at the NMJX.
O ther Mutations of Proteins on the Target MembraneA. R
op/Unc-18/n-Sec1B. U
nc-13 and Vesicle PrimingC. C
AST/ERK at the Active ZoneXI.
M utations in Peripheral Synaptic Vesicle ProteinsA. S
ynapsinB. C
ysteine String Protein and Hsc70XII.
S ummaryR
eferencesThe mechanism of transmitter release at the Drosophila neuromuscular junc-
tion (NMJ) continues to be an area of active investigation in many laboratories.
The undertaking has enhanced our knowledge of synaptic transmission and
exocytosis by permitting in vivo tests of the significance of synaptic proteins.
To explore the proteins that mediate exocytosis and synaptic transmission,
ier Inc.
served.
$35.00
106 THOMAS L. SCHWARZ
mutations in nearly two-dozen genes are available. Others that influence trans-
mitter release indirectly, that is, mutations in endocytosis, development, or Kþ
channels, are largely outside the scope of this chapter. From these studies a
detailed picture emerges in which some proteins form a core apparatus for fusion
while others contribute regulation and eYciency to the system. Yet other proteins
may act in docking vesicles at active zones and in regulating their availability for
fusion. Several interesting questions have been addressed. Is exocytosis primarily
by the full fusion of vesicles with the plasma membrane or by ‘‘kiss-and-run’’? Do
the proteins that mediate fusion also account for the specificity of fusion at active
zones? To what extent are the proteins of the synapse specialized and distinct
from those that mediate exocytosis in nonneuronal cells? Are the soluble
N-ethylmaleimide-sensitive factor attachment receptors (SNAREs) suYcient for
synaptic vesicle fusion? How is the release of transmitter coupled to the action
potential and its consequent influx of Ca2þ? In addition, the dissection of the
fusion apparatus has given rise to the development of at least one reagent, tetanus
toxin (TNT) light chain that is used as a transgene in Drosophila to block synapses
and thereby to silence selected circuits. Thus, the investigation of synaptic
mechanisms in Drosophila is helping us to understand synaptic cell biology in
general and also provides tools with which to probe behavioral and developmen-
tal questions. This chapter will begin with a general discussion of the physiologi-
cal properties of transmission at the NMJ and its cell biological underpinnings,
and will then consider the individual proteins and their genetic analysis.
I. Introduction
The Drosophila neuromuscular junction (NMJ) has played a significant role in
our understanding of neurotransmitter release. Genetic strategies have been used
to test hypotheses concerning the function of proteins that had been implicated in
this process by biochemistry or by yeast genetics. These studies have also led to
physiological insights into the nature of the exocytotic event, distinctions among
exocytotic pathways in neurons, and the mechanisms of modulating synaptic
strength. The knowledge of the mechanism of transmitter release is also impor-
tant as a basis for understanding how synaptic function develops and how
synaptic activity can in turn shape development. Moreover, as investigators
increasingly turn their attention to the functions of the central nervous system
(CNS), including the circuits that govern behavior, synaptic proteins have gained
further importance as a means of manipulating neurons within a circuit. What we
know about these proteins in Drosophila is largely derived from studies of the NMJ.
It is in this manner, rather than by the isolation of novel proteins in genetic
screens, that Drosophila has influenced the field.
TRANSMITTER RELEASE AT THE NMJ 107
This chapter will first review our knowledge on transmitter release from a
functional, physiological perspective, and will examine the nature of the exocy-
totic event. Thereafter, the proteins that mediate transmitter release will be
reviewed with regard to their mutant phenotypes, and how the study of these
phenotypes in Drosophila relates to the larger field of membrane traYc in neurons.
II. Physiological Properties of Transmitter Release
The release of neurotransmitter at the Drosophila NMJ is fundamentally akin
to chemical transmission at vertebrate synapses. Neurotransmitter is packaged
into synaptic vesicles and released into the synaptic cleft by the exocytosis of those
vesicles. Single vesicles may fuse spontaneously, giving rise to unitary ‘‘quantal’’
events—miniature excitatory junctional potentials (mEJPs), also called miniature
excitatory postsynaptic potentials (mEPSPs), or most colloquially just called
‘‘minis.’’ The chief distinction between Drosophila and vertebrate NMJs is that
the fly motoneurons use glutamate as their transmitter rather than acetylcholine.
In addition, whereas vertebrate NMJs are specialized to produce large responses,
with each presynaptic action potential triggering an action potential in the
muscle, the Drosophila NMJ gives rise to a subthreshold, graded response. In this
regard, and in the use of glutamate as a transmitter, the fly NMJ is more akin to a
central excitatory synapse in a vertebrate.
In response to an action potential, the opening of voltage-dependent Ca2þ
channels at the presynaptic terminal causes a large increase in the probability of
exocytosis. The consequent release of the contents of numerous vesicles gives rise
to postsynaptic responses that represent the summed responses of individual
quanta. If one is monitoring the voltage of the membrane (i.e., in ‘‘current
clamp’’ mode), the altered voltage at the postsynaptic cell is called an EJP or
sometimes EPSP. If instead, one is recording the current flowing through the
channels (i.e., in voltage-clamp mode), the response is called an excitatory
junctional current (EJC) or excitatory junctional postsynaptic current (EPSC).
The latter mode of recording is generally preferable because the voltage change is
not simply proportional to the number of channels opened or vesicles released,
while the current flux under constant voltage is.
The delay between the arrival of the action potential to the presynaptic
terminal and the fusion of the vesicles at the active zone is too brief to permit
any large-scale movements of vesicles. Therefore, the vesicles are not released
from the general pool at large in the nerve terminal, but only from a subset of
vesicles that are immediately adjacent to the plasma membrane. These vesicles
are often referred to as ‘‘docked’’ and sometimes are said to comprise the ‘‘readily
releasable pool.’’ The concept of a docked vesicle is itself somewhat loose.
108 THOMAS L. SCHWARZ
In some papers it refers only to an anatomical criterion of juxtaposition with the
plasma membrane, while in others it implies a biochemical state in which vesicle
proteins are interacting with proteins of the plasma membrane.
Although the details of Ca2þ influx and its spatial properties have not been
studied in Drosophila, it is likely that, as in mammalian neurons, the release of
neurotransmitter is triggered by microdomains of high Ca2þ that are formed
near the mouth of voltage-dependent Ca2þ channels. From vertebrate studies
it is known that resting Ca2þ in the cytosol is approximately 100 nM and that
Ca2þ concentration ([Ca2þ]) may rise to several hundred micromoles within
approximately 100 nm of the Ca2þ channel, while the channel is open. These
microdomains—flashes of elevated Ca2þ that, as soon as the channel is closed,
dissipate by diVusion into the larger volume of the terminal and by binding to
cytosolic buVers—are formed in the immediate vicinity of docked, fusion-
competent vesicles. Vesicles further away from the open Ca2þ channels will see
much smaller increases in [Ca2þ] and will have low probabilities of fusion.
Which Ca2þ channels mediate the critical influx of Ca2þ in Drosophila
presynaptic terminals? Although several genes coding for Ca2þ channels are
present in the genome, the cacophony (cac) locus appears to be responsible for most,
and perhaps all, of the Ca2þ influx that triggers transmitter release (Kawasaki
et al., 2000, 2004; Rieckhof et al., 2003). These channels are localized at active
zones and hypomorphic cac alleles cause reductions in transmitter release at the
NMJ of third instar larvae. cac null animals die as embryos, presumably due to
severe reductions in transmission. However, it might be unwise to presume that
this gene mediates all Ca2þ influx at this or any synapse in Drosophila, as multiple
Ca2þ channels can be present at mammalian synapses (Reid et al., 2003).
Once inside the nerve terminal, multiple Ca2þ ions act cooperatively to
trigger the fusion of a synaptic vesicle. This has been determined explicitly at
the third instar NMJ by varying extracellular [Ca2þ] and observing the changes
in the magnitude of transmitter release (Jan and Jan, 1976). The relationship
between [Ca2þ] and vesicles released is nonlinear. The slope of this relationship on
a log–log plot is equivalent to the Hill coeYcient used by enzymologists as
a measure of cooperativity in a biochemical reaction. Similarly, the slope of ln
[Ca2þ] versus ln[vesicles released] is an indication of the number of Ca2þ ions
needed to trigger vesicle fusion, or, more accurately, the cooperativity of Ca2þ in
triggering fusion (Dodge and RahamimoV, 1967). Because this method has been
applied frequently, and not always accurately, some technical issues should be
noted here. (1) The slope is related to the cooperativity of Ca2þ only at very low
[Ca2þ], typically under 300 mM. At higher [Ca2þ] two factors can distort the
measurement. Influx through Ca2þ channels may no longer be proportional to
the extracellular [Ca2þ] and therefore the eVective intracellular [Ca2þ] cannotbe inferred from the external [Ca2þ]. Furthermore, the release apparatus itself
may become saturated or partially limiting at high [Ca2þ]. In the extreme, the
TRANSMITTER RELEASE AT THE NMJ 109
quantal content of the response cannot be further increased by raising Ca2þ and
the relationship has a slope of zero. (2) At the necessary low [Ca2þ], the number
of quanta released per impulse can be very low, particularly when examining a
mutation that decreases release. These measurements frequently involve deter-
mining probabilities of release that are much less than one vesicle per action
potential. In this range, it is important to distinguish spontaneous minis that arise
coincidentally after the stimulus, from release evoked by the stimulus itself.
Determining the frequency of spontaneous minis and subtracting their contribu-
tion from the observed responses is therefore a necessary correction. Both
potential errors (saturation in higher [Ca2þ] or the inclusion of spontaneous
events) tend to flatten the measured dependency on Ca2þ and decrease the
measured cooperativity. The best estimates of the cooperativity at wild-type third
instar NMJs falls in the range of 3.5–4.5 ( Jan and Jan, 1976; Okamoto et al.,
2005; Robinson et al., 2002), a value consistent with similarly derived estimates of
the number of Ca2þ ions needed for exocytosis at mammalian synapses. One
further technical note: Ca2þ influx is not only a function of the extracellular [Ca2þ]but it also depends on [Mg2þ] ions, which can compete for access to Ca2þ
channels. Thus, for a given [Ca2þ], influx, and hence transmitter release, will be
lower in a high Mg2þ saline, such as HL3 (Stewart et al., 1994), than in salines
with less Mg2þ such as that of Jan and Jan (1976).
Most of the recordings made at the Drosophila NMJ, either in third instar
larvae or earlier, have been made from muscles 6 and 7 in the abdominal
segments. These muscles are among the largest and most easily accessed if
the larva is opened along the dorsal midline. However, there is at least one
feature of this preparation that is not ideal and that must be kept in mind in
experimental design and analysis. These muscles are innervated by two gluta-
matergic neurons and each contributes to the EJP (Hoang and Chiba, 2001;
Lnenicka and Keshishian, 2000). Damage to a preparation, improper adjustment
of the stimulating voltage, high-frequency stimulation, or a mutant phenotype,
may cause only one of the two axons to be stimulated and this will decrease the
postsynaptic response. When working in high Ca2þ salines with large EJPs, it
may be quite obvious when the amplitude of the response abruptly drops to a
lower value, signaling that stimulation of one of the axons has failed. If one of the
two axons is intermittently activated, the response will fluctuate between two
amplitudes. In other conditions, where the quantal content is low and stimulus to
stimulus variation is therefore large relative to the mean response amplitude,
failure to conduct an action potential can be much more diYcult to spot and may
be confused with a true phenotype or with use-dependent fatigue of the synapse.
Even more problematic is the possibility of branch-point failure—the failure of
an action potential to invade all the terminal branches of the axons onto the
muscle. Should this occur, the amplitude of the resulting change in EJP is
unpredictable. The preference for recording from muscles 6 and 7 is largely
110 THOMAS L. SCHWARZ
historical. In the future, it may be preferable to record from muscle 5, which
receives input from only a single axon (Hoang and Chiba, 2001).
III. Experimental Advantages and Limitations of the Fly NMJ
The Drosophila NMJ has been invaluable for the analysis of mutations in
synaptic proteins. The third instar larval preparation is among the easiest dissec-
tions in any organism. Themuscle fibers are large and easy to impale and the nerve
is easy to pick up and stimulate with a suction electrode.With an appropriate saline
(Stewart et al., 1994) it is possible to record for extended periods. This synapse,
above all, gives an outstanding opportunity to record the detailed physiological
properties of a great many mutations. The embryonic NMJ and first instar larval
NMJ are considerably harder preparations to master. In particular, the dissections
require patience and practice, but only this preparation permits recordings from
strains with lethal mutations that will not develop beyond these stages. In both
third instar and younger preparations, however, the individual quantal events are
easily resolved from the noise and evoked responses can be studied in response to
nerve stimulation. Individualmuscles and their innervation are easy to identify and
their synapses are stereotyped, showing comparatively little variation from animal
to animal. Compared to neurons in the vertebrate CNS, these identified muscles
provide admirable electrical properties and ease of access.
Nevertheless, some shortcomings of the preparation must be confessed, even
while hoping that future work will overcome some of these obstacles.
A. CA2þ MEASUREMENTS
In Drosophila, it is not presently possible to measure directly the [Ca2þ] thatarises in the immediate vicinity of synaptic release sites. Loading terminals with
Ca2þ indicators, however, has permitted some measurements of the bulk rise in
Ca2þ that occurs in the cytosol of the terminal after trains of action potentials
(Dawson-Scully et al., 2000). Methods for photolytic uncaging of Ca2þ in the
cytosol have not been applied for the study of these synapses as a means of
gaining more direct control of instantaneous changes in cytosolic Ca2þ.
B. DYE LOADING
The study of synaptic vesicle release has been restricted to two methods.
The principle method is recordings from the postsynaptic cell whereby the
TRANSMITTER RELEASE AT THE NMJ 111
channels in the postsynaptic membrane serve as sensors for the transmitter
released. The second method is the loading of synaptic vesicles with fluorescent
dyes such as FM1-43 (Cochilla et al., 1999). The uptake of these dyes into
reforming vesicles by endocytosis and the subsequent release of the dye can
monitor vesicle fusion and endocytosis independently of the presence of neuro-
transmitter in those vesicles or the sensitivity of the postsynaptic cell (Chapter 7
by Kidokoro). However, the size of embryonic terminals and the vesicle pool
within them has made it diYcult to apply this method to study mutations with
early lethal periods. Moreover, because the endings of larval motor neuorons are
typically embedded within the muscle cell, there are diVusion barriers for access
to the synapse and the dye cannot be applied and removed as rapidly as it can, for
example, in cultures of mammalian neurons. Therefore, while very precise rates
of dye loading and unloading can be calculated with this method in mammalian
systems, and dyes of diVerent hydrophobicities can be compared (Richards et al.,
2000), at the Drosophila NMJ this method cannot determine rates of endocytosis
or exocytosis. Dye experiments, instead, are limited to the demonstration of the
existence of cycling vesicles and measurements of the size of the cycling pool
(Chapter 7 by Kidokoro).
C. ALTERNATIVE MEASURES OF EXOCYTOSIS
The Drosophila physiologist can also look with envy on some other prepara-
tions that are amenable to additional sophisticated methods for analyzing exocy-
tosis. These include amperometry, capacitance measurements, and total internal reflection
fluorescence (TIRF) microscopy.
Amperometry uses a carbon-fiber electrode to sense certain biogenic amines,
chiefly dopamine, norepinephrine, and serotonin, and thereby measure the
amount of biogenic amine that is released by the fusion of a vesicle independently
of the sensitivity of the postsynaptic membrane (Zhou and Misler, 1995). At its
best, amperometry can resolve partial release of vesicular contents. However, this
method is not appropriate for glutamate and requires direct access to the synaptic
cleft with the sensing electrode.
By patching directly onto endocrine cells or cells with large nerve terminals,
and measuring the capacitance of the cell membrane, physiologists have been able
to detect exocytosis, sometimes of single vesicles, in a manner that is altogether
independent of the transmitter content of the vesicle. This means that the fusion
of empty or partially loaded vesicles can be determined. In addition, capacitance
measurements have been used to determine the resistance of the fusion pore and
the time course of its opening at the onset of vesicle fusion (Beutner et al., 2001).
However, because the Drosophila NMJ is typically very small and also embedded
112 THOMAS L. SCHWARZ
in the muscle, it is not possible to patch directly onto the boutons for capacitance
measurements.
TIRF microscopy employs fluorescent tags on or in synaptic vesicles and, by
imaging vesicles within approximately 100 nM of the coverslip surface, can
resolve the movements and fusions of single vesicles (Steyer et al., 1997). Again
the geometry of the Drosophila NMJ prevents having release sites directly facing a
coverslip, and therefore this technique has not been used in the fly. If it were to be
adapted for Drosophila, it would probably require using cultured neurons and
inducing them to synapse directly onto a coverslip.
IV. How Do Vesicles Fuse: Full Fusion or Kiss-and-Run?
Before discussing the individual proteins that are involved in transmitter
exocytosis, it is necessary to examine the larger question of precisely how the
transmitter leaves the synaptic vesicle and enters the synaptic cleft. This has been
a long-running controversy in cell biology and neuroscience, dating back at least
30 years (Ceccarelli and Hurlbut, 1980; Heuser and Reese, 1973). The issue can
be summarized as follows. Classical exocytosis, as observed in many cell types,
involves the complete fusion of a vesicle with the plasma membrane. The vesicle
flattens out and is incorporated into the plasma membrane. Its contents are
completely disgorged into the extracellular space. At a nerve terminal, where
vesicles must be recycled rapidly for reuse, this exocytosis would be followed by
endocytosis, probably via clathrin, to recover the membrane and its proteins into
a new synaptic vesicle (Chapter 7 by Kidokoro). An alternative model, often
called kiss-and-run, invokes a mechanism in which the vesicular membrane does
not completely merge with and flatten out onto the plasma membrane. Instead, a
brief, partial fusion is envisioned, in which a transient fusion pore is opened
between the vesicle lumen and the extracellular space. The vesicle is reformed by
the reversal of this process and the closing of the fusion pore. If the pore is open
very briefly, it is possible that transmitter will only partially be unloaded.
A hallmark of the kiss-and-run process is that fusion is not complete and that
classical endocytosis is not needed to remake a synaptic vesicle. The literature
from mammalian systems, for and against the existence of a kiss-and-run mecha-
nism, is too extensive to be reviewed here (An and Zenisek, 2004; Jarousse and
Kelly, 2001). On several points the field has not yet reached a consensus, but
some points are generally accepted: (1) classical exocytosis and endocytosis are
likely to exist at most chemical synapses; (2) partial release of transmitter due to
incomplete unloading appears to occur, at least some of the time, in the case of
large dense-cored vesicles released from nonsynaptic sites; and (3) if kiss-and-run
TRANSMITTER RELEASE AT THE NMJ 113
does occur from small clear vesicles at synapses, it may occur only under certain
circumstances or alongside classical exocytotic events.
At the Drosophila NMJ, the issue of kiss-and-run has surfaced and has
provoked some debate (Dickman et al., 2005; Verstreken et al., 2002). There is
no doubt that full fusion and classical endocytosis occur and account for many of
the fusion events at this synapse. The proteins of the classical endocytotic
pathway are present at presynaptic endings and mutations in a clathrin adaptor
protein, Lap-180, can alter the size of synaptic vesicles (Bao et al., 2005; Zhang
et al., 1998a). Mutations in shibire (shi) have been extremely informative. This gene
encodes the protein dynamin, a component of the classical endocytotic pathway
that severs nascent vesicles from the plasma membrane. As discussed elsewhere in
this volume (Chapter 7 by Kidokoro), shi temperature-sensitive mutations, at a
nonpermissive temperature, can cause the run down of a synapse on stimulation.
Vesicles that fuse to the plasma membrane are trapped there until eventually the
releasable pool of vesicles is completely transferred to the plasma membrane
(Koenig and Ikeda, 1989; van de Goor et al., 1995). These observations are
consistent with classical exocytosis serving as the dominant form of transmitter
release at the Drosophila NMJ. The coexistence of kiss-and-run, however, was
reported from an examination of mutations in other proteins in the classical
endocytotic pathway, Endophilin (Endo) and Synaptojanin (Synj) (Verstreken
et al., 2002, 2003). The response of endo and synj mutants to repeated stimulation
at 10 Hz is quite diVerent from that of a shimutant. Whereas the amplitude of the
EJP declines in shi until a negligible response remains, amplitudes in endo and synj
larvae decline only partially and then reach a plateau level that can be sustained,
seemingly indefinitely. During this plateau many more vesicles are released than
were originally present in the terminal, and thus, there must be a means to
recycle and reuse synaptic vesicles even in the absence of these proteins. Arguing
that these mutations block the classical recycling pathway, the authors concluded
that the plateau phase must be maintained by another pathway and in particular
by kiss-and-run. Further analysis of these mutations, however, has demonstrated
that they do not completely block classical endocytosis, but rather diminish the
maximum rate of endocytosis that can be achieved at the NMJ (Dickman et al.,
2005). The residual slowed pathway has the properties of classical endocytosis,
including the ability for vesicles to be loaded with the dye FM1–43 (Chapter 7
by Kidokoro) and a dependency on Dynamin/Shi. The residual slow pathway
is suYcient to account for the ability of synj and endo terminals to sustain a
reduced level of transmitter release even under high-frequency stimulation. Thus,
although it is not possible to completely exclude the existence of kiss-and-run at
this synapse, there is presently no evidence for its occurrence and the full-fusion
pathway is suYcient to explain the physiology of the system to date, including
the absolute requirement for Dynamin/Shi for sustained vesicle cycling.
114 THOMAS L. SCHWARZ
V. Core Machinery of Exocytosis: Syntaxin, VAMP/Synaptobrevin, and SNAP-25
A wealth of data from yeast to mammals has demonstrated the existence of a
core machinery for intracellular membrane fusion. Versions of the proteins that
form this machinery are present at synapses and can also be found on vesicles
that shuttle between the endoplasmic reticulum (ER) and Golgi in yeast, and in
all membrane traYcking compartments of every eukaryotic cell. In the eVort tounderstand the workings of these proteins, the Drosophila NMJ has proven
valuable. It has the advantages, of course, of having the ability to examine
mutations in the genes encoding for these proteins. However, it also provides
the capacity to resolve single vesicle fusions with rapid kinetics by electrophysio-
logical means and to examine the ability of vesicles to find and dock at their
target membranes by electron microscopy. Additionally, the manipulation of one
of these proteins by the expression of tetanus toxin (TNT) is now used extensively
to alter the activity of circuits within the fly brain (Heimbeck et al., 1999; Kaneko
et al., 2000; Martin et al., 2002; Suster et al., 2003; Sweeney et al., 1995). For this
reason, the role of the fusion complex will be examined here in detail.
The proteins of the core complex are often referred to as soluble
N-ethylmaleimide-sensitive factor attachment receptor (SNARE) proteins (Sollner
et al., 1993b) and the prevailing model, compiled from experiments in yeast,
in synaptic preparations, and in vitro fusion assays, is that these proteins, by
binding to one another, can form a bridge between a vesicle and its target
membrane. Moreover, the energy released by the formation of this complex
may provide the driving force for the actual fusion of the two membranes (Chen
and Scheller, 2001). Fusion has been shown to require complementary proteins
on either side of the reaction, that is, on both the vesicle and target membranes.
These are sometimes referred to as v- and t-SNARES, although the distinction as
to who is the vesicle and who the target is sometimes not so clear. At the synapse,
the t-SNARES (found principally on the plasma membrane) are called Syntaxin
and SNAP-25. The v-SNARE (that resides chiefly on the synaptic vesicles) is
called either Synaptobrevin or VAMP. The complex that forms between these
proteins consists of a four-stranded coiled coil. Two of the component strands are
contributed by SNAP-25 and one each by the other proteins. The complex they
form is exceptionally stable and it has been hypothesized that vesicles docked at
the plasma membrane can form a loose complex first, which, at the time of
vesicle fusion, is allowed to ‘‘zipper closed.’’ This zippering drives the two
membranes together (Chen and Scheller, 2001; Lin and Scheller, 2000).
In addition, it was noted that diVerent traYcking compartments within the cell
can have diVerent isoforms of these proteins and that not all isoforms will
function in conjunction with one another (McNew et al., 2000). Thus, it has been
hypothesized that the cognate pairing of v- and t-SNAREs may provide the
TRANSMITTER RELEASE AT THE NMJ 115
specificity that allows a vesicle to fuse selectively with its appropriate target
membrane.
There have been several key questions regarding the function of this core
complex that have been investigated at the fly NMJ. (1) Are these proteins
essential for all types of synaptic vesicle fusion? (2) Is an individual SNARE used
only for synaptic vesicle fusion or is it used for other types of exocytosis? (3) Are
SNAREs necessary for an early step in transmitter release, particularly for the
docking of vesicles at the active zone, or only for the final membrane fusion step?
(4) Do the SNAREs show specificity in their ability to support synaptic vesicle
fusion? Some or all of these questions have been addressed with mutations in the
genes coding for each of the v- and t-SNAREs.
A. SYNAPTOBREVINS IN THE FLY
There are at least two isoforms of Synaptobrevin/VAMP in Drosophila. One,
called Synaptobrevin (Syb), is ubiquitously expressed in the organism (Chin et al.,
1993; DiAntonio et al., 1993a; Sudhof et al., 1989). Mutations in the syb gene are
embryonic lethals, causing arrest of development at what is probably the time
when the maternally contributed syb runs out. Clones of syb null mutants are cell
lethal in the eye (Bhattacharya et al., 2002). Thus, although there have been no
direct assays of membrane traYc in these cells, it seems likely, based on its
phenotype and homology to Synaptobrevins in other species, that Syb mediates
the constitutive traYcking of proteins to the cell surface. Without this traYc, cells
cannot grow or divide.
The second Synaptobrevin isoform is called neuronal-Synaptobrevin (n-Syb)
and it is detected exclusively in the nervous system, where it appears to be highly
concentrated at all synapses (Deitcher et al., 1998; DiAntonio et al., 1993a).
The first analysis of n-syb function in Drosophila was not via mutations in the gene,
but by the expression of a transgene encoding the light chain of TNT (Broadie
et al., 1995; Sweeney et al., 1995). TNT is a product of a Clostridial bacterium,
and the light chain of TNT is a protease that selectively cleaves certain isoforms
of Synaptobrevin (Pellizzari et al., 1999). In Drosophila, it cleaves n-Syb, but not
Syb. Neuronal expression of TNT did not impair the diVerentiation of moto-
neurons or their ability to extend axons and form synapses, but embryos were
paralyzed and could not hatch (Sweeney et al., 1995). Physiological recordings
from embryonic NMJs uncovered a surprising result: EJPs could not be evoked in
the TNT-expressing flies, but spontaneous minis were still observed. The former
finding was, of course, consistent with the hypothesis that n-Syb is the essential
v-SNARE for the fusion of synaptic vesicles, but the persistence of minis was
confounding. This phenomenon has been studied more intensively in null alleles
of n-syb, which circumvent some potential problems of TNT expression—that
116 THOMAS L. SCHWARZ
some protein may persist intact or that the cleaved products may retain some
activity. The phenotype of the n-syb nulls, however, was identical to that observed
on TNT expression. Null mutations lacking n-syb develop normally but are
paralyzed and do not hatch. Even under conditions where evoked EJPs should
be very large (high Ca2þ or the presence of Kþ channel blockers) no EJP was
seen. Minis, however, were normal in size and present at approximately 25% the
frequency of wild type (Deitcher et al., 1998). These findings are novel because
they distinguish the mechanistic requirements for spontaneous and evoked vesicle
fusions. If evoked release were merely the same process as a mini, but with
increased probability and synchronized, the two types of fusion would be equally
dependent on the same SNARE proteins.
Do the remaining minis fuse via a novel mechanism, completely independent
of SNARE proteins? This is unlikely because, as discussed later, minis are
reported to be completely absent in flies lacking the t-SNARE Syntaxin. However,
SNARE-dependent fusion requires a SNARE on the vesicle membrane as
well (Nichols et al., 1997). What then is the mechanism for the release of these
minis? One possibility is that they employ the other v-SNARE, Syb. Another is
that a t-SNARE, perhaps Syntaxin, is partially localized on synaptic vesicles and
can substitute for n-Syb in the formation of the core complex. There is precedent
for fusion proceeding, albeit poorly, by means of t-SNARE–t-SNARE complexes
(Nichols et al., 1997). Whatever this fusion pathway might be, it is not capable of
the fast, synchronous fusions that form an EJP. In a detailed study of these
residual events (Yoshihara et al., 1999), it was determined that their frequency
can be modulated by cytosolic Ca2þ, but the elevated Ca2þ must be sustained.
Thus, in an n-syb null mutant, the frequency of minis could be increased up to
17-fold by a steady state depolarization of the terminal, a tetanic stimulation of
the nerve, or direct Ca2þ influx via an ionophore. Thus, n-syb mutant terminals
retain a Ca2þ-sensor that can couple to the remaining release apparatus.
The inability of Ca2þ influx during a single action potential to have a similar
eVect could indicate that the triggering is too slow to be activated by a transient
change in local Ca2þ, or that the vesicles are no longer in the immediate vicinity
of the open Ca2þ channels.
Another fascinating aspect of these mutant terminals is that the release
apparatus is no longer modulated by cAMP (Yoshihara et al., 1999). In wild-
type larvae, cAMP, via Protein Kinase A, has two actions at the synapse. One
depends on extracellular Ca2þ and likely involves the modulation of Ca2þ
channels. The other is independent of extracellular Ca2þ and likely involves a
modulation of the release apparatus that increases mini frequency. This Ca2þ-independent action was absent from n-syb larvae (Yoshihara et al., 2000) and thus
the pathway downstream of cAMP must pass through this particular v-SNARE.
Another interesting feature of the n-syb phenotype was noted in viable hypo-
morphic alleles (Stewart et al., 2000): a shift in the apparent cooperativity of Ca2þ
TRANSMITTER RELEASE AT THE NMJ 117
ions in triggering release, as measured by systematically altering the extracellular
[Ca2þ]. This finding may reflect the fact that Ca2þ, having bound to its sensor,
must act via the SNARE complex.
Are v-SNAREs specific for particular types of membrane traYc? The speci-
ficity of SNARE pairing is a key element of the hypothesis that SNAREs direct
vesicles to the correct target membrane. Through the availability of null muta-
tions in both syb and n-syb, and of transgenes to drive expression of each, it has
been possible to ask whether they are either interchangeable or selective in their
particular roles, that is, in cell survival and growth versus synaptic transmission
(Bhattacharya et al., 2002). The expression of n-syb in the developing syb null eye
could in fact rescue its development and the expression of syb could partially
restore the EJP at an n-syb null NMJ. Moreover, these two v-SNARES have
behaved identically in biochemical assays of complex formation (Niemeyer and
Schwarz, 2000). Thus, Synaptobrevins may or may not have selectivity for fusion
with the plasma membrane instead of with intracellular organelles. However,
they cannot account for the selective fusion of synaptic vesicles at active zones, as
opposed to the fusion of other vesicles that deliver proteins to other sites on the
cell surface.
B. TETANUS TOXIN AS A SYNAPSE-SILENCING REAGENT
As mentioned earlier, the light chain of TNT can cleave n-Syb and it does
so with great specificity as well as great eYciency: little or no intact n-Syb remains
when TNT expression is driven throughout the nervous system (Sweeney et al.,
1995). There has therefore been great interest in selectively expressing TNT
in subsets of neurons to silence their output (Martin et al., 2002). This method has
been used to establish the functional importance of neurons for such behaviors
as feeding and locomotion (Heimbeck et al., 1999; Kaneko et al., 2000; Suster
et al., 2003; Sweeney et al., 1995). However, some important caveats need to be
considered, because it is not possible in the CNS to determine directly whether
or not TNT expression in the selected neurons is eVective in preventing trans-
mitter release. Instead, there is a presumption, not necessarily accurate, that
those cells will be aVected in the same manner as the well-studied motoneurons.
We do not know, however, that every neuron contains its target, n-Syb, although
it is certainly widespread (DiAntonio et al., 1993a). Moreover, since overexpres-
sion of Syb, which is not cleaved by TNT, can substitute, at least in part,
for n-Syb (Bhattacharya et al., 2002), it is possible that high levels of Syb or a
related SNARE might sustain transmission in subpopulations of CNS synapses.
Similarly, even at the NMJ, TNT does not abolish transmitter release. It only
abolishes the eYcient, synchronous release of transmitter in response to individual
action potentials. Spontaneous minis can therefore be expected to remain
118 THOMAS L. SCHWARZ
in the CNS and any sustained electrical activity, such as high-frequency trains
of action potentials, or sustained depolarizations in nonspiking neurons, such
as photoreceptors, would be expected to increase the release of transmitter
despite the removal of n-Syb. Finally, there are many forms of transmitter release
that may diVer mechanistically from what we have studied in detail at the
glutamatergic motoneuron synapses. The eYcacy of TNT in preventing the
release of aminergic vesicles or dense-cored peptidergic vesicles is only now being
examined. Release of transmitter from nonvesicular pools, for example, from
plasma membrane transporters running ‘‘backward’’ is presumed to be TNT-
insensitive (Yang and Kunes, 2004). Finally, although loss of n-Syb does not cause
gross abnormalities in the development of neurons, the possibility remains that
the eVects of TNT will not be exclusively on the fusion of synaptic vesicles. TNT
may alter the release of signaling molecules, or the surface expression of some
receptors. In summary, although TNT expression is a highly valuable tool for
neurogenetics, it may not silence all forms of synaptic transmission or be devoid
of developmental consequences.
C. SYNTAXIN1
Of the SNAREs that are localized to the plasma membrane, the most
attention has been given to Syntaxin. Although Syntaxin isoforms are present
on many compartments within the cell, a particular Syntaxin, called Syntaxin1
(Syx1) in Drosophila, appears to be the homolog of the Syntaxins found on the
plasma membrane in mammalian cells (Schulze et al., 1995). Genetic analysis of
this protein at the synapse is complicated by the fact that Syx1 is not exclusively
synaptic. Mutations in syx1 are cell lethal, suggesting that it is important in cell
growth and division, like the v-SNARE syb (Schulze et al., 1995; Stowers and
Schwarz, 1999). Maternal germ-line mosaics of null syx1 alleles are infertile due
to arrest of oocyte development. Similarly, mosaics of weaker alleles give rise to
embryos with defects in cellularization, a process that involves a large increase in
membrane surface area (Burgess et al., 1997). These phenotypes attest to the
essential role that Syx1 plays in membrane traYc to the cell surface.
However, due to maternally contributed protein and mRNA, homozygous
null mutant embryos from heterozygous parents can develop until late stages of
embryogenesis before dying (Broadie et al., 1995; Burgess et al., 1997; Schulze
et al., 1995). These embryos have abnormalities indicative of secretory defects in
nonneuronal cells and also mild anatomical abnormalities in their central and
peripheral nervous system. Nevertheless, the maternal contribution appears to be
just barely enough that, at least in most segments, motoneurons are born and can
extend axons to the appropriate muscle targets, but not enough for synapses to
function properly (Schulze et al., 1995). Several papers have appeared in which
TRANSMITTER RELEASE AT THE NMJ 119
the electrophysiological and developmental properties of these synapses are
described (Broadie et al., 1995; Featherstone and Broadie, 2002; Saitoe et al.,
2001, 2002; Schulze et al., 1995), and their findings are not always in agreement.
The diVerences are likely, at least in part, to be due to the fact that these embryonic
synapses are poised at a threshold where the maternally contributed Syx1 is
disappearing. Therefore, the age of the embryo, the position of the muscle, and
other unknown variables, may cause some recordings to be made from synapses
that still contain low levels of Syx1, while others have so little Syx1 that the synapse
never formed or formed and then began to retract. In no case has an EJP been
evoked from a homozygous null syx1 embryo, consistent with a requisite role
for Syx1 in transmitter release (Broadie et al., 1995; Saitoe et al., 2001; Schulze
et al., 1995). It has also been reported that, contrary to initial reports (Broadie et al.,
1995), quanta are not released from the syx terminals by the application of
hypertonic sucrose solutions (Aravamudan et al., 1999). Hypertonic sucrose has
been used in mammalian systems to trigger the release of vesicles from nerve
terminals by an unknown mechanism that bypasses the normal requirement for
Ca2þ influx. This finding again emphasizes the importance of Syntaxin for the
fusion of synaptic vesicles, and further suggests that sucrose triggers release in a
SNARE-dependent manner. However, one group has observed spontaneous
minis at early stages of synapse formation in syx1 mutants (perhaps while some
Syx1 still remains). Subsequently this group observed a large decrease in the
sensitivity of the muscle membrane to glutamate and an absence of glutamate
receptors clusters. These findings were interpreted as showing that minis are
needed to form or preserve glutamate receptors clusters opposite to active zones
(Saitoe et al., 2001, 2002). Others have observed normal levels of glutamate
sensitivity and the absence of minis (Featherstone and Broadie, 2002). These
findings underscore the diYculty of analyzing the function of a protein in trans-
mitter release when that protein is also required for constitutive membrane traYc
and hence for the viability of the cell and the formation of the synapse.
Biochemical studies of the H3 domain of mammalian Syntaxin1 indicate that
it can bind to several proteins other than Synaptobrevin/VAMP and SNAP-25,
the proteins of the core complex. In particular, Syntaxin1 can bind to the protein
known as n-Sec1 or Munc-18 in mammalian systems and encoded by ras opposite
(rop) in Drosophila. This is a very high aYnity interaction, and its function is not yet
clear. However, n-Sec1 is not present in the core complex, and therefore, one
hypothesis is that it holds Syntaxin in a closed state that prevents it from
associating with the other SNAREs (Dulubova et al., 1999). Mutations in rop
are embryonic lethal and, like syb or syx1 mutations, have defects in nonneuronal
structures, suggesting a requirement in constitutive membrane traYcking
(Harrison et al., 1994). This phenotype (and the phenotype of sec1 mutations in
yeast; Novick et al., 1980) is not compatible with n-Sec1 functioning solely as a
negative regulator of release: its phenotype indicates a decrease in secretion,
120 THOMAS L. SCHWARZ
not an increase. Syntaxin also binds to the vesicle protein Synaptotagmin (see in
a later section) and may bind to a region of mammalian Ca2þ channels
(Bezprozvanny et al., 2000; Leveque et al., 1994). Attempts have been made to
understand the physiological significance of these interactions in Drosophila by
using point mutations that selectively alter one or another of the protein–protein
interactions (Fergestad et al., 2001; Wu et al., 1999). Portions of these studies,
however, have been challenged on the basis that the mutations used do not
selectively or suYciently alter the interactions in question (Matos et al., 2000).
D. SNAP-25 AND SNAP-24
The other plasma membrane t-SNARE, SNAP-25, has been particularly
diYcult to study in Drosophila, in part because the gene is located within hetero-
chromatin and is composed of multiple exons that are small and widely spaced
(Risinger et al., 1997). In addition, it is partially redundant with a homolog
protein called SNAP-24 (Niemeyer and Schwarz, 2000; Vilinsky et al., 2002).
The first allele of Snap25 to be isolated was a temperature-sensitive paralytic
mutation that resulted in a single amino acid change in the first amphipathic coil
of SNAP-25 (Rao et al., 2001). Biochemical analysis indicated that this protein
remained capable of forming SNARE complexes. The only temperature-sensitive
defect that was detected in this mutant was in the little-understood phenomenon
of SNARE complex multimerization—the presence of higher molecular weight
complexes of SNARE proteins on SDS gels. These presumed multimers of the
SNARE complex were less stable if SNAP-25 contained the point mutation.
In vivo, the temperature-sensitive allele had the unexpected property of increasing
EJP amplitude at the permissive temperature, perhaps reflecting an enhanced
ability of the mutant protein to proceed toward fusion (Rao et al., 2001).
At elevated temperatures, however, transmission was reduced relative to wild
type, which is the likely cause of the temperature-sensitive paralysis and suggests
that the multimerization of SNAREs may be important to their function in vivo.
The isolation of null alleles of Snap25 oVered an additional surprise:
homozygotes could survive through pupal stages and synaptic transmission at
the larval NMJ was largely unaVected by the absence of the protein (Vilinsky
et al., 2002). This phenomenon is likely to be explained by the presence of the
closely related t-SNARE SNAP-24. SNAP-24 is less abundant than SNAP-25
within the nervous system and is more abundant in nonneuronal tissues, suggest-
ing that it is primarily concerned with constitutive membrane traYc. Instead,
SNAP-25 predominates at nerve terminals (Niemeyer and Schwarz, 2000).
However, in the complete absence of SNAP-25 protein (which is not the case
in the temperature-sensitive Snap25 allele) SNAP-24 can replace SNAP-25
(Vilinsky et al., 2002). In biochemical studies, SNAP-24 and SNAP-25 were
TRANSMITTER RELEASE AT THE NMJ 121
indistinguishable in their shared ability to form SNARE complexes with Syntaxin
and n-Syb, consistent with their interchangeability during transmitter release
(Niemeyer and Schwarz, 2000). In addition, both isoforms could form SNARE
complexes with Syntaxin and Syb, the likely complex involved in constitutive
membrane traYc. Thus, as discussed earlier for n-Syb and Syb, the t-SNAREs
SNAP-24 and SNAP-25 are interchangeable. Specificity of cognate recognition
by SNARE proteins cannot explain the distinctions in membrane targeting
between synaptic vesicles that fuse at active zones and constitutive vesicles that
fuse elsewhere on the surface.
E. SNARE PROTEINS ACT LATE IN THE VESICLE CYCLE
The study of n-Syb (as well as Syntaxin, see in a later section) has been useful
in determining the step at which SNAREs act, that is, whether in the docking of
vesicles at their target membranes, in the priming of them for fusion, or in the
fusion step itself. For these questions of general cell biological importance, the
Drosophila NMJ has several advantages. Not only are there unambiguous null
alleles for SNARE proteins, but there is also a well-defined target membrane that
normally contains a population of docked vesicles that are visible by electron
microscopy. It is not clear just how this pool, defined as ‘‘docked’’ by anatomical
criteria, correlates with the biochemist’s model of ‘‘docked’’ vesicles, which
presumes an association of the vesicle and target membranes via protein–protein
interactions. Nonetheless, in the case of syx1 null alleles or in nerve endings
expressing TNT, there is no detectable loss of synaptic vesicles in close juxtaposi-
tion to the active zone membrane. On the contrary, the number of docked
vesicles appears to be increased (Broadie et al., 1995). These findings indicate
that synaptic vesicles can find nerve terminals, cluster around active zones, and
dock at the active zone membrane in the absence of the SNARE proteins. The
absence of evoked release, in contrast, argues strongly for a block at a late stage in
the process, most likely in fusion itself. The intermediate steps that are hypothe-
sized to occur in order to make a vesicle ready for rapid fusion after it contacts the
plasma membrane (collectively referred to as priming), are more diYcult to
judge. The presence of minis in n-syb null alleles, and their ability to fuse in
response to sustained elevation of Ca2þ, may indicate that some vesicles are
primed for fusion, but their slower rate of fusion indicates that they are not as
competent for fusion as vesicles in wild type.
F. SUMMARY OF SNARE PROTEINS IN DROSOPHILA
In summary, null alleles are available for each of the v- and t-SNAREs
that function at the synapse, syx1, n-syb, and Snap25, as well as for the close
122 THOMAS L. SCHWARZ
homolog syb that principally functions in nonsynaptic traYc to the plasma
membrane. In addition, TNT can selectively cleave n-Syb and thereby prevent
the evoked release of transmitter. By selective expression in particular cell types,
TNT may be a powerful means to silence the output of a cell, although some
caution is needed in extrapolating from the NMJ to the CNS. From studies of the
fly NMJ we have learned that mutations in SNAREs can diVerentially aVectevoked release and minis, suggesting mechanistic diVerences in the nature of
vesicle fusion in these two processes. Because morphologically docked vesicles
persist in these mutants, the SNAREs are likely to act specifically in a late step of
the vesicle cycle, most probably in fusion itself. In addition, although there are
distinctions between the Syb isoforms that normally predominate in constitutive
versus synaptic membrane traYc, those isoforms can substitute for one another in
functional assays. In addition, they show no biochemical specificity in their ability
to complex with one another. Moreover, a single Syx isoform is used in both
pathways. Therefore, SNARE pairing does not provide specificity for targeting
these classes of vesicles to their particular target membrane at the active zone.
VI. Vesicular ATPase and Membrane Fusion
The demonstration that SNARE proteins function in a late stage of exocytosis
raises the question as to whether they are the sole requirement for membrane
fusion. There have been in vitro studies to demonstrate that SNAREs by them-
selves can promote the fusion of lipid vesicles in vitro (Weber et al., 1998), but this
does not necessarily mean that this minimal reconstituted system represents the
normal in vivomechanism. In this context, one of the most intriguing studies using
the Drosophila NMJ (Hiesinger et al., 2005) proposes that a subunit of the vesicular
ATPase, an abundant and required protein in the synaptic vesicle membrane,
also functions at a late step of exocytosis, perhaps by forming a fusion pore
bridging the membranes. The vesicular ATPase is the proton pump responsible
for the acidification of the vesicle lumen at the expense of ATP. As such it
provides, in the form of a proton gradient, the energy that drives the uptake of
transmitter into vesicles. A hypothesis is that at least one of the proteins in this
large complex (the v100–1 subunit of the V0 complex) functions directly in
membrane fusion.
This model has origins in studies of vacuolar fusion in yeast (Bayer et al., 2003;
Peters et al., 2001) wherein it was found that a particular subunit of the vacuolar
ATPase, Vph1p, was necessary in assays of vacuole fusion with one another
in vitro. This ATPase comprises two large complexes, one that hydrolyzes ATP
and another that forms a proteolipid pore through which protons can flux.
The Vph1p subunit is in the latter, the V0 complex. However, the particular
TRANSMITTER RELEASE AT THE NMJ 123
mutations studied did not block proton transport, the canonical function of the
V0 complex. Rather, the defect in the yeast assay was specifically in fusion, at a
step after the requirement for the SNARE proteins. It was hypothesized that the
large Vph1p protein formed a bridge between two vesicles that were linked by
SNAREs and that this bridge could progress into a proteinaceous pore that
subsequently would resolve into full fusion of the membranes.
In this context, it was striking that mutations in the Drosophila homolog of the
same large subunit were isolated in a phototaxis screen designed to identify
mutations of synaptic transmission (Hiesinger et al., 2005). The subunit, called
v100-1, is concentrated at synapses, and null mutations in the v100-1 gene
showed a sevenfold reduction in EJP amplitude at the embryonic NMJ. Minis
were present and of normal amplitude, although reduced in frequency. The
presence of some minis indicates that some transmitter loading and some fusion
can occur in these embryos, perhaps because of a partially redundant second
isoform. But why is the EJP so small? Is it because of a block in fusion or because
a lack of proton transport has generated empty vesicles? The evidence favors the
former because FM1-43 dye loading was below normal levels (Hiesinger et al.,
2005). Moreover, both in yeast and in the fly, the v100-1 subunit can bind to
t-SNAREs (Hiesinger et al., 2005; Peters et al., 2001). Thus, while the precise role
of this ATPase subunit remains uncertain, the hypothesis that it interacts with
t-SNAREs to form a fusion pore, or to initiate membrane fusion is an attractive
possibility.
VII. NSF and the Resetting of the SNARE Machinery
NSF, a factor that support in vitro vesicle traYcking in cell extracts (Block et al.,
1988), is the key element of a multimeric protein that includes six NSF subunits
and six subunits of an associated protein called alpha-SNAP. NSF can bind to
SNARE complexes and, through the catalysis of ATP, can cause them to
dissociate (Sollner et al., 1993a). Originally, it was thought that this ATP-depen-
dent step might represent the fusion reaction itself, but this model has been
rejected because fusion in many systems is ATP-independent, and because
SNAREs in vitro can stimulate membrane fusion without NSF (Weber et al.,
1998). Models (Lin and Scheller, 2000; Schwarz, 1999) place NSF action after
fusion as a means of undoing SNARE complexes so that v- and t-SNAREs can be
separated from one another and recycled to their proper compartments.
The formation of the SNARE complex is so energetically favorable that energy
(provided by the ATPase activity of NSF) is required to pull the SNAREs apart
and free them for another round of vesicle fusion. Separated, the SNAREs are
in a state with high potential energy and that energy can be harnessed by the
124 THOMAS L. SCHWARZ
cell when SNARE complexes subsequently reform and thereby drive membrane
fusion.
The comatose (comt) gene in Drosophila encodes an NSF homolog (Pallanck
et al., 1995). The study of comt permitted a crucial genetic test of NSF function
in vivo. The classic comt alleles are temperature-sensitive paralytics (Siddiqi and
Benzer, 1976). These alleles are hypomorphic mutations of Nsf1, one of two NSF
homologs in the fly (Pallanck et al., 1995). Consistent with the model described
above, when shifted to the nonpermissive temperature, comt flies accumulate
SNARE complexes that reside primarily on the plasma membrane (Tolar and
Pallanck, 1998). Using electron microscopy it is observed that these terminals
accumulate vesicles, including a population that is docked at the active zone
by morphological criteria, suggesting a defect in the priming of vesicles for
transmitter release (Kawasaki et al., 1998). A likely explanation (Littleton et al.,
2001) is that each round of vesicle fusion in a comt mutant creates SNARE
complexes on the plasma membrane that cannot be dissociated and that, over
time, the cycling vesicle pool is depleted of usable, free, SNAREs. At this point,
the vesicles can no longer fuse with the membrane and are therefore trapped in a
docked but fusion-incompetent state.
At adult NMJs on the dorsolateral flight muscles, the electrophysiological
phenotype of comt mutants is wholly consonant with this model. At the permissive
temperature the EJP is normal but, at the nonpermissive temperature, there is
a progressive decline in EJP amplitude. This likely reflects the loss of free
SNAREs and hence the decline in releasable vesicles (Kawasaki et al., 1998).
At the third instar larval NMJ, however, comt does not have a detectable pheno-
type (Golby et al., 2001; Mohtashami et al., 2001). This can be explained by the
fact that the Nsf1 gene that is mutant in comt appears to be predominantly
expressed in the adult nervous system, while NSF2 is likely to play an equivalent
role at the larval NMJ (Stewart et al., 2002). Mutations in Nsf2 are early lethals,
perhaps reflecting the importance of this isoform for membrane traYc in non-
neuronal cells (Golby et al., 2001). To study the physiological consequences of
disrupting NSF2 at the larval NMJ, a dominant negative approach has therefore
been taken (Stewart et al., 2002, 2005). Expression of a dominant negative NSF2
in larval neurons has the expected consequence of causing use-dependent run-
down of the amplitude of the EJP. However, it also causes a substantial over-
growth of the synapse, even at early developmental stages, and some axonal
misrouting. The traYcking defect behind these developmental phenotypes is not
yet known.
The relationship of the two Nsf genes in Drosophila is not yet completely
understood. However, the apparent importance of NSF2 in nonneuronal embry-
onic tissues as well as at larval synapses provides another example of membrane
traYcking proteins that are shared between synaptic and nonsynaptic pathways.
TRANSMITTER RELEASE AT THE NMJ 125
There are no known diVerences in the SNARE proteins at adult versus larval
synapses, and so it seems unlikely that the two NSF isoforms will diVer in their
substrate specificities.
VIII. Synaptotagmin and the Regulation of Transmitter Release
A. SYNAPTOTAGMIN AND ITS BINDING PROPERTIES
Ever since the identification of Synaptotagmin as a major synaptic vesicle
protein and the determination of its sequence (Perin et al., 1990), Synaptotagmin
has been the focus of intense study. The biochemical properties of mammalian
Synaptotagmin have been studied in great detail (Chapman, 2002; Li et al., 1995;
Rizo and Sudhof, 1998; Sudhof and Rizo, 1996), and some of these properties
have been confirmed for Drosophila. The hallmark of the Synaptotagmin protein
family is an N-terminal transmembrane domain, followed by a large cytoplasmic
domain containing two C2 domains in tandem. The C2 domain is a much-
studied motif also present in isoforms of Protein Kinase C and Phospholipase A2.
In these proteins the C2 domain binds Ca2þ and causes a conformational change
in the protein that triggers its translocation to the membrane (Nalefski et al.,
2001). The C2 domains of Synaptotagmin also bind Ca2þ via a set of conserved
negative charges that are clustered at one end of the structure (Shao et al., 1996,
1998). In addition, C2 domains of Synaptotagmin also bind phospholipids and
phosphatidyl inositol (Li et al., 1995; Zhang et al., 1998b). These binding proper-
ties strongly suggest an important role for Synaptotagmin in the regulation of
transmitter release. Synaptotagmin is the best candidate for being the sensor for
cytosolic Ca2þ that triggers the release of vesicles in response to an action
potential.
In addition to these binding properties for small ligands, Synaptotagmin
interacts with SNARE proteins, SNARE complexes, the clathrin adaptor protein
AP-2, as well as with lipid membranes (Li et al., 1995; Zhang et al., 1994). Many of
these interactions have been shown to be modulated by Ca2þ. In considering the
function of Synaptotagmin, it is clear that these interactions could be the means
by which Synaptotagmin regulates the fusion step during transmitter release, the
ability of vesicles to translocate to and dock at the membrane, or the rate of
endocytosis. To resolve the significance of these myriad interactions in vivo genetic
analysis of synaptotagmin (syt) mutants was undertaken, first in the fly (Broadie et al.,
1994; DiAntonio and Schwarz, 1994; DiAntonio et al., 1993b; Littleton et al., 1993,
1994) and Caenorhabditis elegans (Jorgensen et al., 1995; Nonet et al., 1993) and
subsequently in the mouse (Geppert et al., 1994).
126 THOMAS L. SCHWARZ
B. SYNAPTOTAGMIN PHENOTYPES AT THE NMJ
Null mutations of syt, also called syt1 (see in a later section), are poorly viable
and first were believed to be incapable of surviving past the first instar larval stage
(DiAntonio et al., 1993b). However, by segregating homozygotes away from their
more robust heterozygous siblings and by raising them under appropriate
conditions, it is possible for them to survive to the third instar and even to early
adulthood (Loewen et al., 2001). Therefore, data are now available for the
phenotypes of syt null alleles for embryonic, first instar and third instar NMJs,
as well as from weaker alleles. The electrophysiological phenotype of null alleles is
severe—approximately a 10- to 40-fold reduction in the amplitude of the evoked
response—at each of these stages (Broadie et al., 1994; Loewen et al., 2001;
Okamoto et al., 2005; Robinson et al., 2002; Yoshihara and Littleton, 2002).
Yet, as expected from the persistent viability of the flies, transmission is never
abolished. The remaining evoked responses and other physiological parameters
have therefore been examined to determine precisely what is lacking in these
mutants.
One hypothesis has been that Syt serves as a fusion clamp that prevents the
SNARE proteins from interacting in resting Ca2þ, but which releases the
SNAREs and thereby permits vesicle fusion when Ca2þ is bound (Littleton
et al., 1994). Some suggestion of this mechanism might be taken from the
observation that mini frequency is modestly increased at both embryonic and
third instar NMJs that lack Syt (Broadie et al., 1994; DiAntonio and Schwarz,
1994; Littleton et al., 1994; Loewen et al., 2001). In this model, the removal of Syt
would cause the constitutive fusion of synaptic vesicles as soon as they dock at the
active zone, and the evoked response would be reduced because of the inability to
accumulate readily releasable vesicles. This model, however, was rejected
because the number of minis released was not adequate to explain the drastic
reduction in evoked release (DiAntonio and Schwarz, 1994). This finding is also
supported by studies in which Synaptotagmin was acutely inactivated (Marek and
Davis, 2002). For these experiments, the wild-type Syt was replaced with a syt
transgene that included a C-terminus binding site for a fluorescent ligand. After
application of the ligand, laser illumination produces reactive oxygen species
that rapidly destroy the tagged protein (FlAsH-FALI). In this protocol, evoked
transmission could be suppressed within seconds without a concomitant increase
in spontaneous release. Thus, Syt plays a positive role by promoting fusion rather
than by acting as a fusion clamp to prevent spontaneous release.
The dominant hypothesis is that Ca2þ binding to Synaptotagmin is the
necessary and suYcient event to trigger transmitter release in response to an
action potential. However, strong evidence for this hypothesis was not easy to
achieve. The first obstacle has been the persistence of Ca2þ-dependent transmitter
release even in null alleles, a phenomenon also observed in C. elegans and
TRANSMITTER RELEASE AT THE NMJ 127
mice (Geppert et al., 1994; Nonet et al., 1993). Thus, Synaptotagmin can enhance
transmitter release, but is not absolutely necessary. In the absence of a complete
blockade of transmission, two interpretations were available. Synaptotagmin
might not be the Ca2þ-trigger but might instead be a facilitator of release that
increases the availability of vesicles for fusion or the probability of their fusing.
Alternatively, Synaptotagmin might be a trigger for the EJP, but coexists with a
second Ca2þ-sensor that mediates the residual release. The sequencing of the
Drosophila genome revealed additional synaptotagmin genes that might represent
this residual Ca2þ sensor, although none has yet been demonstrated to play this
role (see in a later section). Therefore, attempts to clarify the importance of
Synaptotagmin have centered on examining the properties of exocytosis in
wild-type and syt mutants, in the hope of finding an indication that the residual
release is using a Ca2þ-sensor that is distinct from that in wild-type synapses. The
nonsynaptotagmin Ca2þ-sensor might have a diVerent aYnity for Ca2þ or
involve the binding of a diVerent number of Ca2þ ions. The latter would be
reflected in a diVerent slope for the plot of log[Ca2þ] versus log response
amplitude (see in an earlier section). Particularly, if there were two sensors, both
of which needed to be activated by Ca2þ, the removal of one of them should
decrease the slope of this relationship. Recordings from null embryos, however,
failed to see such a shift (Broadie et al., 1994). Although some alleles were
reported to have a shift in the slope or aYnity (Littleton et al., 1994), most
recordings indicated a predominant eVect on the number of vesicles secreted
(Vmax) rather than on the slope or aYnity. A change in the number of vesicles
secreted could be caused by changes in many aspects of transmission other than a
change in the Ca2þ sensor. These studies are exceptionally diYcult because the
evoked responses are so small in the mutants and because it is necessary to work
in very low Ca2þ to determine the cooperativity of the relationship. The most
thorough study of evoked release in null syt embryos (Okamoto et al., 2005) found
a substantial decrease in the slope, consistent with the hypothesis that Syt is the
major Ca2þ sensor for exocytosis and that the residual release is driven by a
sensor with distinct properties.
Reports of recordings from syt alleles sometimes noted that the time course of
release was altered, with more release occurring slower or asynchronously than in
wild type (Loewen et al., 2001; Yoshihara and Littleton, 2002). This has given rise
to speculation that the terminals contain a fast sensor (Syt) and a slow sensor
(unknown) governing the residual release. However, a loss of synchrony in
transmitter release can be explained by changes in docking, priming, and fusion,
in addition to changes in the Ca2þ sensor, and others have observed low levels of
fast synchronous release to persist in these mutants (Okamoto et al., 2005).
Some of the most compelling evidence that Synaptotagmin is the major
Ca2þ-sensor for triggering fusion has come from studies of the Ca2þ-binding sitesin the C2 domain (Mackler et al., 2002; Okamoto et al., 2005). If Synaptotagmin
128 THOMAS L. SCHWARZ
is the crucial Ca2þ sensor, these studies reasoned, mutation of the aspartate
residues that form the Ca2þ-binding site should cause drastic and measurable
changes in the Ca2þ-dependent properties of transmission. The Ca2þ-bindingsite in the C2 domain had been described in detail from structural studies, and
the significance of each aspartate for the Ca2þ-dependent properties of Synapto-tagmin had been defined for mammalian Synaptotagmin I (Li et al., 1995; Rizo
and Sudhof, 1998; Shao et al., 1996, 1998; Zhang et al., 1998b). These mamma-
lian studies implicated the first of the two C2 domains (C2A) as the more
important. However, when syt null Drosophila mutants were supplied with syt
transgenes carrying mutations that completely disrupted this site (and therewith
the ability of the C2A domain to undergo Ca2þ-dependent binding to Syntaxin
or phospholipids) synaptic transmission was restored in a manner completely
comparable to rescue by a wild-type transgene. Even the slope of the Ca2þ/response relationship was comparable to that of wild type, indicating that Ca2þ
binding by the C2A domain cannot be required for or have a significant influence
on the release of transmitter (Robinson et al., 2002). Mutations in the C2B
domain resulted in dramatically diVerent consequences (Mackler et al., 2002).
Neutralizing the equivalent aspartates in this second C2 domain were suYcient to
render the protein inactive in rescue experiments, and expression of this trans-
gene had a dominant-negative eVect by suppressing synaptic transmission. Simi-
lar dramatic eVect was observed in syt alleles that either remove the C2B domain
or carry a point mutation near the C2B Ca2þ-binding domain (DiAntonio and
Schwarz, 1994; Okamoto et al., 2005; Yoshihara and Littleton, 2002).
Together, the evidence from studies of Ca2þ-dependency in the syt null
and the eYcacy of alterations in the Ca2þ-coordinating aspartates of the C2B
domain, build a strong case for Synaptotagmin being the crucial sensor for
triggering exocytosis. It remains to be determined why some release persists
and also why the neutralization of charges in the C2A domain (highly conserved
through evolution and partaking in many Ca2þ-dependent interactions) appearsto be of so little consequence to synaptic physiology.
C. OTHER FUNCTIONS OF SYNAPTOTAGMIN
The significance of Synaptotagmin for the Ca2þ-dependence of transmitter
release does not preclude additional functions of the protein. Electron micrographic
studies of central synapses as well as the NMJ (Loewen et al., 2006; Reist
et al., 1998) indicate that syt mutant terminals have fewer vesicles, although those
vesicles present continue to cluster in the vicinity of the active zone. This
phenotype suggests a defect in either the biogenesis of synaptic vesicles or in
their retrieval from the membrane after fusion. A similar paucity of vesicles at
the terminals of C. elegans mutants led to the hypothesis that Synaptotagmin
TRANSMITTER RELEASE AT THE NMJ 129
regulates endocytosis (Jorgensen et al., 1995). Such a function is likely to be
mediated by the ability of Synaptotagmin to bind the clathrin adaptor protein
AP-2 and thereby recruit clathrin to vesicles (Zhang et al., 1994). Quantitative
measurements of endocytotic rates have been undertaken in sytmutants by means
of a GFP-based reporter of vesicle fusion (Poskanzer et al., 2003). Because endocy-
tosis is coupled to exocytosis, it had not previously been possible to distinguish any
direct eVect of Synaptotagmin on the former from its clear eVects on the latter.
Particularly, by using FlAsH-FALI (see in an earlier section) to inactivate Synap-
totagmin after a round of exocytosis, it was possible to measure these rates
independently. Loss of Syt slows the rate of endocytosis and particularly limits
the ability of the synapse to increase its rate of endocytosis in response to increased
exocytosis (Poskanzer et al., 2003).
The interactions of Synaptotagmin with phosopholipid membranes and with
the proteins of the plasma membrane also led to the hypothesis that Synapto-
tagmin may promote the docking of vesicles at active zones, akin perhaps to the
function of C2 domains in various enzymes. Electron microscopy of syt mutants
revealed a selective decrease in the pool of vesicles immediately adjacent to the
active zone (Reist et al., 1998). Loss of the pool that presumably correspond to
vesicles primed and ready for fusion may be a major contributor to the syt
phenotype. Mutations of the C2B domain that retain the rest of the protein do
not cause these ultrastructural changes (Loewen et al., 2006). This may hold a
significant clue for understanding the function of the rest of the molecule.
D. MULTIPLE SYNAPTOTAGMINS IN THE FLY
The Drosophila genome predicts six additional isoforms of Synaptotagmin,
but there is little functional data at present concerning any of them (Adolfsen
and Littleton, 2001). Syt4 is most homologous to mammalian Synaptotagmins
4 and 11 and, like Syt1, is abundant in the nervous system where it can be
found enriched at synapses. At the NMJ it is present in muscle cells, appearing
in puncta near the synapse (Adolfsen et al., 2004), but it can also be found
in neurosecretory endings, including peptidergic terminals at the NMJ (Bohm
and Schwarz, unpublished data). Within the CNS it is not known if Syt4 is
predominantly pre- or postsynaptic. Syt� and Syt� (with no immediate relative
in mammals but closest to Synaptotagmin 12) are also reported to be concen-
trated in some neurosecretory endings (Adolfsen et al., 2004), whereas Syt7,
homolog to mammalian Synaptotagmin 7, is found in many tissues including
nerve and muscle, but has not been observed to be concentrated at synapses.
The remaining genes, syt12 and syt14 are the least understood, although low
levels of Syt14 have been observed by in situ hybridization in the embryonic CNS
(Adolfsen et al., 2004).
130 THOMAS L. SCHWARZ
Of these additional Synaptotagmins, only Syt4 has been analyzed genetically.
Null alleles [from the excision of a P-element (Adolfsen et al., 2004), or made
by homologous recombination (Bohm and Schwarz, unpublished data)] are
homozygous viable, fertile, and overtly normal. In NMJ morphology and EJP
amplitude, the third instar syt4 mutants are similarly normal. The analysis of null
alleles does not support earlier suggestions that syt4 could account for residual
transmitter release in syt1 null larvae. My laboratory had reported that syt4
transgenes could rescue synaptic transmission in syt1 mutants, but this was in
error (Robinson et al., 2002), as was an earlier report on overexpression of syt4
that concluded that syt4 regulates transmitter release by interfering with syt1
function (Littleton et al., 1999). However, at the NMJ of third instar larvae,
postsynaptic Syt4 translocates to the plasma membrane in response to activity
and this may represent the release of an unidentified retrograde signal (Yoshihara
et al., 2005). Also, in embryonic synapses at hatching stages, an electrophysio-
logical phenotype has been reported: whereas in wild-type embryos high-
frequency trains of stimuli result in an increase in minifrequency, this was not
observed in the syt4 nulls. The eVect on minis could be restored by the rescue of
syt4 expression in the muscle, consistent with a model in which muscle activity
releases a retrograde messenger that modulates the release of minis from the
presynaptic terminal (Yoshihara et al., 2005).
E. SUMMARY OF SYNAPTOTAGMIN FUNCTION AT THE FLY NMJ
The analysis of syt1 mutations indicates an important role for Synaptotagmin
in promoting the fusion of vesicles in response to an action potential. Because syx1
and n-syb mutants are also required for the EJP, Syt and the SNAREs are part of
the same release pathway. The function of Syt is likely to be closely tied to its
ability to bind Ca2þ ions. The Ca2þ-binding site of the C2A domain, however,
appears rather insignificant, whereas the Ca2þ-binding site in the C2B domain
has an essential role. The residual release of vesicles in response to an action
potential or elevated cytosolic Ca2þ remains a puzzle. If it is due to the presence
of a redundant synaptotagmin gene, the gene has not yet been identified. In
addition to a role in promoting fusion, Syt is likely to promote vesicle recycling
and vesicle docking.
IX. Exocyst at the NMJ
The SNARE complex is not the only protein complex that is implicated
in membrane traYc. The genetic screens for secretory defects in yeast that
uncovered mutations in SNARE proteins also uncovered a complex that is
TRANSMITTER RELEASE AT THE NMJ 131
frequently referred to as the exocyst (EauClaire and Guo, 2003; Lipschutz and
Mostov, 2002). The eight components of this complex are Sec3, Sec5, Sec6, Sec8,
Sec10, Sec15, Exo70, and Exo84 and their phenotypes in yeast include blocking
a late stage in vesicle transport to the plasma membrane. In budding yeast,
vesicles are transported into the growing bud but fail to fuse at their target sites
(Novick et al., 1980). One of the most intriguing aspects of the exocyst is that it
localizes to the site of vesicle fusion, that is, at the bud tip when daughter cells are
growing and at the bud neck when membrane is being added for cytokinesis
(Finger and Novick, 1997; Finger et al., 1998). In contrast, the SNARE proteins
are uniformly distributed within the plasma membranes of yeast. The exocyst
might therefore be responsible for docking vesicles at target membranes.
With so many other mechanistic parallels between membrane traYc in yeast
and at the synapse (Bennett and Scheller, 1993), the question naturally arose as to
whether the exocyst plays a similar role in the nervous system, particularly with
regard to targeting synaptic vesicles to active zones. The exocyst complex was
purified from mammalian brain (Hsu et al., 1996) and antibody studies suggested
an important function in membrane traYc in epithelial cells (Hazuka et al., 1999).
A murine mutation of sec8 was discovered (Friedrich et al., 1997), but it caused
lethality at a very early stage of development and therefore was unsuitable for
more detailed studies of membrane traYc.
The analysis of exocyst mutations and neuronal function therefore
commenced with the identification of sec5 mutations in Drosophila (Murthy
et al., 2003). As previously discussed regarding syx mutations, the analysis is
complicated by the fact that members of this complex have vital functions and
cell-lethal phenotypes. sec5 null alleles are cell lethal when clones are made in the
eye or female germ line (Murthy and Schwarz, 2004; Murthy et al., 2003).
However, as in the case of syx, maternally contributed sec5 mRNA is suYcient
to permit the embryo to develop and, in the case of sec5, larvae hatch out and can
survive for up to 3 days. These larvae, however, do not achieve their expected
size but rather remain at the border of first and second instars. This can be
attributed to the need for sec5 in constitutive membrane addition. Although there
is suYcient maternal Sec5 in homozygous null embryos to support normal
development, by hatching it has fallen to 29% of wild-type levels and by 48 h after
egg laying (AEL) to 11%, at which point no further growth is observed. By 72 h
AEL, Sec5 levels are barely detectable, no more than 3% of wild type. At the
NMJ this is manifest in the arrest of muscle growth and a similar arrest in the
development of the synapse. At 96 h, motoneurons on muscles 6/7 have bouton
counts appropriate for 48 h larvae (Murthy et al., 2003). Therefore, by studying
neurons from larvae that were 72 h AEL or older, the significance of Sec5 could be
assessed in the almost complete absence of the protein. It was found that these
neurons had severe defects in membrane traYc. If neurons were dissociated from
larval brains they were incapable of re-extending an axon. If a reporter gene was
132 THOMAS L. SCHWARZ
turned on at this stage, it could be synthesized but was not eYciently traYcked to
the cell surface (Murthy et al., 2003). These defects not withstanding, the synapse
at the NMJ continued to function and the EJP was slightly larger at 96 h than it
had been at 48 h, perhaps reflecting the limited formation of new release sites.
Although the amplitude of the EJP was well below that of a normal 96 h larva,
presumably due to the small size of the NMJ, release per bouton was quite normal.
Thus, Sec5 is essential in many membrane traYcking steps, including those that
are responsible for the growth of the NMJ and formation of new boutons.
However, it is dispensable for the actual targeting and fusion of synaptic vesicles
once a bouton has formed.
Mutations have been reported in the exocyst components sec6 (Beronja et al.,
2005; Murthy et al., 2005) and sec15 (Mehta et al., 2005). Although these studies
have not closely examined synaptic transmission or the NMJ per se, the mutant
phenotypes reported are consistent with the conclusion that the exocyst is not
needed for the release of neurotransmitter. Synaptic phenotypes at the NMJ have
been examined for mutations in sec8 (Liebl et al., 2005) and through the use of RNAi
to reduce levels of sec10 (Andrews et al., 2002). In each case the release of
transmitter was unaVected. The significance of the exocyst or of individual compo-
nents of it for synaptic development, however, remains an area of great interest
because it has the potential not only to permit but perhaps also to direct the addition
of new boutons presynaptically and the expression of receptors postsynaptically.
To a large extent, synaptic transmission can be thought of as a variation on
exocytosis in all cells, including yeast, but with additional regulatory components,
such as Synaptotagmin, added on. Sec5 and the exocyst as a whole, however, are
a counter example. The exocyst is a protein complex that is essential in yeast and
in the constitutive traYcking pathway for many, perhaps all, membrane proteins,
and yet it is dispensable for transmitter release. One possible explanation is that
the specialized architecture of the active zone provides a unique mechanism for
the docking of synaptic vesicles at release sites. If so, this mechanism may
substitute for the normal role of the exocyst (Murthy et al., 2003).
X. Other Mutations of Proteins on the Target Membrane
A. ROP/UNC-18/N-SEC1
As discussed earlier, neither mutations in SNARE-coding genes nor in genes
encoding components of the exocyst have phenotypes that are adequate to
explain how synaptic vesicles are preferentially clustered near and docked at
the active zone. From studies of mammalian synapses, there are several major
proteins that may contribute to these events, but which presently have received
TRANSMITTER RELEASE AT THE NMJ 133
less attention in Drosophila. One of these is a Syntaxin-binding protein called Sec1
in yeast and n-Sec1 or Unc-13 in mammals and C. elegans, respectively. Mutations
in the Drosophila homolog gene were recovered fortuitously in a screen for ras
mutants and were given the name of rop (Harrison et al., 1994). The phenotype of
rop mutants, including developmental abnormalities in embryos, strongly impli-
cated rop as necessary for both constitutive and synaptic exocytosis, which was
subsequently confirmed by the analysis of point mutations (Wu et al., 1998).
Overexpression of Rop, however, inhibited transmitter release (Schulze et al.,
1994). These findings have led to conflicting interpretations, according to which
Rop is either an inhibitor of transmission or a necessary promoter of vesicle
release. The enhancement of synaptic transmission by mutations in the Rop-
binding region of Syt appeared to favor a model in which this protein restricts the
availability of Syt for SNARE complex formation (Wu et al., 1999). This has
proven controversial due to the diYculty of determining in vivo the extent to
which protein interactions have actually been prevented (Matos et al., 2000; Wu
et al., 2001). From mammalian studies, it is clear that the n-Sec1/Unc-18 protein
is not bound to Syntaxin when Syntaxin is complexed with the other SNARE
proteins. It is therefore most attractive, at present, to envision Rop as a protein
that cycles on and oV Syntaxin, priming it for activity in transmission, but then
releasing it so that Syntaxin could bind to the other SNAREs. Such a cyclical role
could explain both its requirement for secretion and the ability of overexpressed
Rop to interfere with transmitter release.
B. UNC-13 AND VESICLE PRIMING
Unc-13, a protein that has been extensively studied in both C. elegans and the
mouse (Rhee et al., 2002; Richmond et al., 1999), may also regulate the release
apparatus. The prevailing model, from studies in those organisms, is that Unc-13
can regulate the state of Syntaxin and thereby influence its ability to form
SNARE complexes (Richmond et al., 2001). Unc-13 appears to be a crucial focus
for modulating the strength of synaptic transmission in response to changes
in second messengers, particularly diacylglycerol (Rhee et al., 2002). Drosophila
lacking unc-13 have a severe phenotype: a complete loss of both the embryonic
EJP and minis (Aravamudan et al., 1999) and an accumulation of docked vesicles.
In addition, the levels of Unc-13 may be modulated by second messenger systems
(Aravamudan and Broadie, 2003).
C. CAST/ERK AT THE ACTIVE ZONE
The monoclonal antibody nc82 has been a favorite for the analysis of
synaptic structures due to its remarkable specificity for active zones. The gene
134 THOMAS L. SCHWARZ
encoding the nc82 epitope has been cloned and shown to encode the fly homolog
of ELKS/CAST (Wagh et al., 2006), a protein whose localization to active zones
has also been established in vertebrates (Ohtsuka et al., 2002; Wang et al., 2002).
In Drosophila, the locus was given the name bruchpilot (brp). Although the function
of ELKS/CAST is not known, it binds to a protein called RIM-1 that in turn
binds to Unc-13 (see in an earlier section). Although no mutation in the fly
homolog has yet been reported, protein levels were knocked down by means of
RNAi expression (Wagh et al., 2006). Reducing nc82 immunoreactivity produced
a reduction in the amplitude of the EJP, although minis persisted normally.
A striking ultrastructural phenotype confirmed the importance of the protein in
organizing the active zone. In Drosophila, many although not all active zones
possess an electron dense structure, sometimes called a T-bar. This structure may
be akin to the ribbons and similar structures seen at some mammalian synapses.
RNAi to Drosophila CAST/ELKS prevented the formation of these T-bars.
XI. Mutations in Peripheral Synaptic Vesicle Proteins
A. SYNAPSIN
The Synapsin family was one of the first major synaptic proteins to
be identified and it associates with the cytoplasmic surface of synaptic vesicles.
It was recognized early to be the subject of phosphorylation by both cAMP- and
Ca2þ-dependent protein kinases (Greengard et al., 1993). Consequently, there has
been much hope and speculation that it might be a crucial control point for such
synaptic phenomena as modulation by amine transmitters and paired-pulse
facilitation. By and large, these expectations have not been met by the pheno-
types of knockout mice (Rosahl et al., 1993, 1995), although the presence of three
partially redundant genes in the mouse makes this analysis more complicated.
Murine data suggest a role for the Synapsins in regulating the availability of
reserve pool vesicles during prolonged neuronal activity (Chi et al., 2001; Sun
et al., 2006). In Drosophila, the presence of a single synapsin (syn) gene, albeit
an alternatively spliced gene, has assisted its genetic analysis (Godenschwege
et al., 2004). The syn mutant phenotype, however, remains astonishingly mild—
undetectable at the level of cellular analysis at the NMJ. No alteration was
detected in the ultrastructure of these synapses, including the density of synaptic
vesicles in the vicinity of active zones. Thus, no indication was found that
Synapsins were essential for coupling vesicles to the local cytoskeleton. In addi-
tion, the properties of the EJP and minis were unchanged from control.
Even with 2 s of 5 Hz stimulation, no diVerences were observed, with a modest
TRANSMITTER RELEASE AT THE NMJ 135
degree of synaptic depression in both mutant and control. However, the flies have
distinct behavioral abnormalities, including learning defects, in both adult
and larval stages (Godenschwege et al., 2004; Michels et al., 2005). Thus, the
importance of Synapsin to the functioning of the nervous system was confirmed,
but the specific cell biological function of this vesicle protein has not been
identified at fly synapses. More detailed physiological tests may eventually reveal
a subtle electrophysiological phenotype.
B. CYSTEINE STRING PROTEIN AND HSC70
The cysteine string protein (CSP) is another protein associated with the
cytoplasmic surface of synaptic vesicles (Chamberlain and Burgoyne, 2000).
It has homology to DNAJ/Hsc40 proteins and through its J domain can interact
with the chaperone Hsc70. Therefore, studies have focused on a likely role as a
chaperone protein that can assist in protein folding and refolding (Braun and
Scheller, 1995; Braun et al., 1996). Consistent with such a model, Hsc70 mutants
have also been shown to have a phenotype at the fly NMJ. The significance of
such a protein on synaptic vesicles remains unclear. It may involve interactions
with Syntaxin and Synaptotagmin (Bronk et al., 2001, 2005).
The phenotype of Csp mutants in Drosophila has been crucial to determining
its physiological significance. CSP cannot be central to the process of transmitter
release and exocytosis—transmitter release persists in null mutants (Zinsmaier
et al., 1994). Nevertheless, synaptic function is not normal. Particularly at elevated
temperatures synaptic function is impaired causing smaller evoked responses
(Dawson-Scully et al., 2000; Ranjan et al., 1998; Umbach et al., 1994).
This reduction does not appear to be due to a decrease in Ca2þ-channel activ-ity. The synaptic defects are accompanied by a progressive degeneration of
neurons and eventual paralysis and death. The importance of the interaction
with Hsc70 was confirmed by the isolation of mutations in that gene and the
finding that they give rise to phenotypes very similar to those of Csp (Bronk et al.,
2001), although some eVects of Csp may be independent of Hsc70 (Arnold et al.,
2004; Bronk et al., 2005). The prevailing model, at present, is that repeated
cycling of synaptic vesicles, particularly at elevated temperatures, results in the
production of denatured proteins. It is not yet known who the critically vulnera-
ble proteins are. CSP on the vesicles, working with Hsc70 and perhaps, an
additional associated protein (Tobaben et al., 2001), can restore proper
folding of these proteins and prevent the ensuing degeneration and loss of
synaptic strength (Fernandez-Chacon et al., 2004). In this regard, the neurode-
generative phenotype of CSP may become an interesting model for the study of
degenerative disorders.
136 THOMAS L. SCHWARZ
XII. Summary
Nearly two decades of work on the release of transmitter at the Drosophila
NMJ has produced a sizeable collection of mutations and a great deal of
phenotypic analysis. Mutations have been generated in all the central compo-
nents of the release machinery and in many additional regulatory and peripheral
components of the nerve terminal. It is interesting that mutant screens in
Drosophila have not yet identified essential elements of the exocytosis apparatus
that were not previously identified by biochemical means. However, by bringing
electrophysiological and ultrastructural analysis in vivo to the study of synaptic
components, the system has made substantial contributions to our understanding
of synaptic function in all organisms.
The task of elucidating synaptic mechanisms is not over. One of the most
significant unanswered questions centers on the active zone. The prominent
electron density of this structure suggests numerous highly specialized compo-
nents, but few have as yet been identified or subject to genetic analysis. We still
do not understand why synaptic vesicles accumulate in terminals and cluster near
active zones. Nor do we understand how synaptic vesicles are selectively targeted
to these sites rather than elsewhere on the cell surface, while other cargo-carrying
vesicles do not fuse at the active zone. Similarly, the mechanistic relationship of
minis to EJPs remains unclear. Perhaps the most important question for synaptic
function entails understanding the diVerences between synapses. Why does one
synapse has a high probability of release while another has a low probability?
How is the release of peptidergic granules for modulatory and neuroendocrine
function distinct from the release of small clear vesicles? How are synapses
specialized for tonic or phasic release of transmitter? It is likely that the Drosophila
NMJ will continue to figure prominently in the exploration of these issues.
References
Adolfsen, B., and Littleton, J. T. (2001). Genetic and molecular analysis of the synaptotagmin family.
Cell Mol. Life Sci. 58, 393–402.
Adolfsen, B., Saraswati, S., Yoshihara, M., and Littleton, J. T. (2004). Synaptotagmins are traYcked to
distinct subcellular domains including the postsynaptic compartment. J. Cell Biol. 166, 249–260.
An, S., and Zenisek, D. (2004). Regulation of exocytosis in neurons and neuroendocrine cells. Curr.
Opin. Neurobiol. 14, 522–530.
Andrews, H. K., Zhang, Y. Q., Trotta, N., and Broadie, K. (2002). Drosophila sec10 is required for
hormone secretion but not general exocytosis or neurotransmission. TraYc 3, 906–921.
Aravamudan, B., and Broadie, K. (2003). Synaptic Drosophila UNC-13 is regulated by antagonistic
G-protein pathways via a proteasome-dependent degradation mechanism. J. Neurobiol. 54,
417–438.
TRANSMITTER RELEASE AT THE NMJ 137
Aravamudan, B., Fergestad, T., Davis, W. S., Rodesch, C. K., and Broadie, K. (1999). Drosophila
UNC-13 is essential for synaptic transmission. Nat. Neurosci. 2, 965–971.
Arnold, C., Reisch, N., Leibold, C., Becker, S., Prufert, K., Sautter, K., Palm, D., Jatzke, S., Buchner,
S., and Buchner, E. (2004). Structure-function analysis of the cysteine string protein in Drosophila:
Cysteine string, linker and C terminus. J. Exp. Biol. 207, 1323–1334.
Bao, H., Daniels, R. W., MacLeod, G. T., Charlton, M. P., Atwood, H. L., and Zhang, B. (2005).
AP180 maintains the distribution of synaptic and vesicle proteins in the nerve terminal and
indirectly regulates the eYcacy of Ca2þ-triggered exocytosis. J. Neurophysiol. 94, 1888–1903.
Bayer, M. J., Reese, C., Buhler, S., Peters, C., and Mayer, A. (2003). Vacuole membrane fusion:
V0 functions after trans-SNARE pairing and is coupled to the Ca2þ-releasing channel. J. Cell
Biol. 162, 211–222.
Bennett, M. K., and Scheller, R. H. (1993). The molecular machinery for secretion is conserved from
yeast to neurons. Proc. Natl. Acad. Sci. USA 90, 2559–2563.
Beronja, S., Laprise, P., Papoulas, O., Pellikka, M., Sisson, J., and Tepass, U. (2005). Essential
function of Drosophila Sec6 in apical exocytosis of epithelial photoreceptor cells. J. Cell Biol. 169,
635–646.
Beutner, D., Voets, T., Neher, E., and Moser, T. (2001). Calcium dependence of exocytosis and
endocytosis at the cochlear inner hair cell aVerent synapse. Neuron 29, 681–690.
Bezprozvanny, I., Zhong, P., Scheller, R. H., and Tsien, R. W. (2000). Molecular determinants of the
functional interaction between syntaxin and N-type Ca2þ channel gating. Proc. Natl. Acad. Sci.
USA 97, 13943–13948.
Bhattacharya, S., Stewart, B. A., Niemeyer, B. A., Burgess, R. W., McCabe, B. D., Lin, P., Boulianne,
G., O’Kane, C. J., and Schwarz, T. L. (2002). Members of the synaptobrevin/vesicle-associated
membrane protein (VAMP) family in Drosophila are functionally interchangeable in vivo for
neurotransmitter release and cell viability. Proc. Natl. Acad. Sci. USA 99, 13867–13872.
Block, M. R., Glick, B. S., Wilcox, C. A., Wieland, F. T., and Rothman, J. E. (1988). Purification of
an N-ethylmaleimide-sensitive protein catalyzing vesicular transport. Proc. Natl. Acad. Sci. USA 85,
7852–7856.
Braun, J. E., and Scheller, R. H. (1995). Cysteine string protein, a DnaJ family member, is present on
diverse secretory vesicles. Neuropharmacology 34, 1361–1369.
Braun, J. E., Wilbanks, S. M., and Scheller, R. H. (1996). The cysteine string secretory vesicle protein
activates Hsc70 ATPase. J. Biol. Chem. 271, 25989–25993.
Broadie, K., Bellen, H. J., DiAntonio, A., Littleton, J. T., and Schwarz, T. L. (1994). Absence of
synaptotagmin disrupts excitation-secretion coupling during synaptic transmission. Proc. Natl.
Acad. Sci. USA 91, 10727–10731.
Broadie, K., Prokop, A., Bellen, H. J., O’Kane, C. J., Schulze, K. L., and Sweeney, S. T. (1995).
Syntaxin and synaptobrevin function downstream of vesicle docking in Drosophila. Neuron 15,
663–673.
Bronk, P., Wenniger, J. J., Dawson-Scully, K., Guo, X., Hong, S., Atwood, H. L., and Zinsmaier,
K. E. (2001). Drosophila Hsc70-4 is critical for neurotransmitter exocytosis in vivo. Neuron 30,
475–488.
Bronk, P., Nie, Z., Klose, M. K., Dawson-Scully, K., Zhang, J., Robertson, R. M., Atwood, H. L.,
and Zinsmaier, K. E. (2005). The multiple functions of cysteine-string protein analyzed at
Drosophila nerve terminals. J. Neurosci. 25, 2204–2214.
Burgess, R. W., Deitcher, D. L., and Schwarz, T. L. (1997). The synaptic protein syntaxin1 is required
for cellularization of Drosophila embryos. J. Cell Biol. 138, 861–875.
Ceccarelli, B., and Hurlbut, W. P. (1980). Ca2þ-dependent recycling of synaptic vesicles at the frog
neuromuscular junction. J. Cell Biol. 87, 297–303.
Chamberlain, L. H., and Burgoyne, R. D. (2000). Cysteine-string protein: The chaperone at the
synapse. J. Neurochem. 74, 1781–1789.
138 THOMAS L. SCHWARZ
Chapman, E. R. (2002). Synaptotagmin: A Ca(2þ) sensor that triggers exocytosis? Nat. Rev. Mol. Cell.
Biol. 3, 498–508.
Chen, Y. A., and Scheller, R. H. (2001). SNARE-mediated membrane fusion. Nat. Rev. Mol. Cell. Biol.
2, 98–106.
Chi, P., Greengard, P., and Ryan, T. A. (2001). Synapsin dispersion and reclustering during synaptic
activity. Nat. Neurosci. 4, 1187–1193.
Chin, A. C., Burgess, R. W., Wong, B. R., Schwarz, T. L., and Scheller, R. H. (1993). DiVerential
expression of transcripts from syb, a Drosophila melanogaster gene encoding VAMP
(synaptobrevin) that is abundant in non-neuronal cells. Gene 131, 175–181.
Cochilla, A. J., Angleson, J. K., and Betz, W. J. (1999). Monitoring secretory membrane with FM1–43
fluorescence. Annu. Rev. Neurosci. 22, 1–10.
Dawson-Scully, K., Bronk, P., Atwood, H. L., and Zinsmaier, K. E. (2000). Cysteine-string protein
increases the calcium sensitivity of neurotransmitter exocytosis in Drosophila. J. Neurosci. 20,
6039–6047.
Deitcher, D. L., Ueda, A., Stewart, B. A., Burgess, R. W., Kidokoro, Y., and Schwarz, T. L. (1998).
Distinct requirements for evoked and spontaneous release of neurotransmitter are revealed by
mutations in the Drosophila gene neuronal-synaptobrevin. J. Neurosci. 18, 2028–2039.
DiAntonio, A., and Schwarz, T. L. (1994). The eVect on synaptic physiology of synaptotagmin
mutations in Drosophila. Neuron 12, 909–920.
DiAntonio, A., Burgess, R. W., Chin, A. C., Deitcher, D. L., Scheller, R. H., and Schwarz, T. L.
(1993a). Identification and characterization of Drosophila genes for synaptic vesicle proteins.
J. Neurosci. 13, 4924–4935.
DiAntonio, A., Parfitt, K. D., and Schwarz, T. L. (1993b). Synaptic transmission persists in
synaptotagmin mutants of Drosophila. Cell 73, 1281–1290.
Dickman, D. K., Horne, J. A., Meinertzhagen, I. A., and Schwarz, T. L. (2005). A slowed classical
pathway rather than kiss-and-run mediates endocytosis at synapses lacking synaptojanin and
endophilin. Cell 123, 521–533.
Dodge, F. A., Jr., and RahamimoV, R. (1967). Co-operative action of calcium ions in transmitter
release at the neuromuscular junction. J. Physiol. 193, 419–432.
Dulubova, I., Sugita, S., Hill, S., Hosaka, M., Fernandez, I., Sudhof, T. C., and Rizo, J. (1999).
A conformational switch in syntaxin during exocytosis: Role of munc18. EMBO J. 18, 4372–4382.
EauClaire, S., and Guo, W. (2003). Conservation and specialization. The role of the exocyst in
neuronal exocytosis. Neuron 37, 369–370.
Featherstone, D., and Broadie, K. (2002). Response: Meaningless minis? Trends Neurosci. 25, 386–387.
Fergestad, T., Wu, M. N., Schulze, K. L., Lloyd, T. E., Bellen, H. J., and Broadie, K. (2001).
Targeted mutations in the syntaxin H3 domain specifically disrupt SNARE complex function in
synaptic transmission. J. Neurosci. 21, 9142–9150.
Fernandez-Chacon, R., Wolfel, M., Nishimune, H., Tabares, L., Schmitz, F., Castellano-Munoz, M.,
Rosenmund, C., Montesinos, M. L., Sanes, J. R., Schneggenburger, R., and Sudhof, T. C.
(2004). The synaptic vesicle protein CSP alpha prevents presynaptic degeneration. Neuron 42,
237–251.
Finger, F. P., and Novick, P. (1997). Sec3p is involved in secretion and morphogenesis in
Saccharomyces cerevisiae. Mol. Biol. Cell 8, 647–662.
Finger, F. P., Hughes, T. E., and Novick, P. (1998). Sec3p is a spatial landmark for polarized secretion
in budding yeast. Cell 92, 559–571.
Friedrich, G. A., Hildebrand, J. D., and Soriano, P. (1997). The secretory protein Sec8 is required for
paraxial mesoderm formation in the mouse. Dev. Biol. 192, 364–374.
Geppert, M., Goda, Y., Hammer, R. E., Li, C., Rosahl, T. W., Stevens, C. F., and Sudhof, T. C.
(1994). Synaptotagmin I: A major Ca2þ sensor for transmitter release at a central synapse. Cell
79, 717–727.
TRANSMITTER RELEASE AT THE NMJ 139
Godenschwege, T. A., Reisch, D., Diegelmann, S., Eberle, K., Funk, N., Heisenberg, M., Hoppe, V.,
Hoppe, J., Klagges, B. R., Martin, J. R., Nikitina, E. A., Putz, G., et al. (2004). Flies lacking all
synapsins are unexpectedly healthy but are impaired in complex behaviour. Eur. J. Neurosci. 20,
611–622.
Golby, J. A., Tolar, L. A., and Pallanck, L. (2001). Partitioning of N-ethylmaleimide-sensitive fusion
(NSF) protein function in Drosophila melanogaster: dNSF1 is required in the nervous system, and
dNSF2 is required in mesoderm. Genetics 158, 265–278.
Greengard, P., Valtorta, F., Czernik, A. J., and Benfenati, F. (1993). Synaptic vesicle phosphoproteins
and regulation of synaptic function. Science 259, 780–785.
Harrison, S. D., Broadie, K., van de Goor, J., and Rubin, G. M. (1994). Mutations in the
Drosophila Rop gene suggest a function in general secretion and synaptic transmission. Neuron
13, 555–566.
Hazuka, C. D., Foletti, D. L., Hsu, S. C., Kee, Y., Hopf, F. W., and Scheller, R. H. (1999). The sec6/8
complex is located at neurite outgrowth and axonal synapse-assembly domains. J. Neurosci. 19,
1324–1334.
Heimbeck, G., Bugnon, V., Gendre, N., Haberlin, C., and Stocker, R. F. (1999). Smell and taste
perception in Drosophila melanogaster larva: Toxin expression studies in chemosensory neurons.
J. Neurosci. 19, 6599–6609.
Heuser, J. E., and Reese, T. S. (1973). Evidence for recycling of synaptic vesicle membrane during
transmitter release at the frog neuromuscular junction. J. Cell Biol. 57, 315–344.
Hiesinger, P. R., Fayyazuddin, A., Mehta, S. Q., Rosenmund, T., Schulze, K. L., Zhai, R. G.,
Verstreken, P., Cao, Y., Zhou, Y., Kunz, J., and Bellen, H. J. (2005). The v-ATPase V0 subunit
a1 is required for a late step in synaptic vesicle exocytosis in Drosophila. Cell 121, 607–620.
Hoang, B., and Chiba, A. (2001). Single-cell analysis of Drosophila larval neuromuscular synapses. Dev.
Biol. 229, 55–70.
Hsu, S. C., Ting, A. E., Hazuka, C. D., Davanger, S., Kenny, J. W., Kee, Y., and Scheller, R. H.
(1996). The mammalian brain rsec6/8 complex. Neuron 17, 1209–1219.
Jan, L. Y., and Jan, Y. N. (1976). Properties of the larval neuromuscular junction in Drosophila
melanogaster. J. Physiol. 262, 189–214.
Jarousse, N., and Kelly, R. B. (2001). Endocytotic mechanisms in synapses. Curr. Opin. Cell Biol. 13,
461–469.
Jorgensen, E. M., Hartwieg, E., Schuske, K., Nonet, M. L., Jin, Y., and Horvitz, H. R. (1995).
Defective recycling of synaptic vesicles in synaptotagmin mutants of Caenorhabditis elegans. Nature
378, 196–199.
Kaneko, M., Park, J. H., Cheng, Y., Hardin, P. E., and Hall, J. C. (2000). Disruption of synaptic
transmission or clock-gene-product oscillations in circadian pacemaker cells of Drosophila cause
abnormal behavioral rhythms. J. Neurobiol. 43, 207–233.
Kawasaki, F., Mattiuz, A. M., and Ordway, R. W. (1998). Synaptic physiology and ultrastructure in
comatose mutants define an in vivo role for NSF in neurotransmitter release. J. Neurosci. 18,
10241–10249.
Kawasaki, F., Felling, R., and Ordway, R. W. (2000). A temperature-sensitive paralytic mutant
defines a primary synaptic calcium channel in Drosophila. J. Neurosci. 20, 4885–4889.
Kawasaki, F., Zou, B., Xu, X., and Ordway, R. W. (2004). Active zone localization of presynaptic
calcium channels encoded by the cacophony locus of Drosophila. J. Neurosci. 24, 282–285.
Koenig, J. H., and Ikeda, K. (1989). Disappearance and reformation of synaptic vesicle membrane
upon transmitter release observed under reversible blockage of membrane retrieval. J. Neurosci.
9, 3844–3860.
Leveque, C., el Far, O., Martin-Moutot, N., Sato, K., Kato, R., Takahashi, M., and Seagar, M. J.
(1994). Purification of the N-type calcium channel associated with syntaxin and synaptotagmin.
A complex implicated in synaptic vesicle exocytosis. J. Biol. Chem. 269, 6306–6312.
140 THOMAS L. SCHWARZ
Li, C., Ullrich, B., Zhang, J. Z., Anderson, R. G., Brose, N., and Sudhof, T. C. (1995). Ca(2þ)-
dependent and -independent activities of neural and non-neural synaptotagmins. Nature 375,
594–599.
Liebl, F. L., Chen, K., Karr, J., Sheng, Q., and Featherstone, D. E. (2005). Increased synaptic
microtubules and altered synapse development in Drosophila sec8 mutants. BMC Biol. 3, 27.
Lin, R. C., and Scheller, R. H. (2000). Mechanisms of synaptic vesicle exocytosis. Annu. Rev. Cell Dev.
Biol. 16, 19–49.
Lipschutz, J. H., and Mostov, K. E. (2002). Exocytosis: The many masters of the exocyst. Curr. Biol.
12, R212–R214.
Littleton, J. T., Stern, M., Schulze, K., Perin, M., and Bellen, H. J. (1993). Mutational analysis of
Drosophila synaptotagmin demonstrates its essential role in Ca2þ-activated neurotransmitter
release. Cell 74, 1125–1134.
Littleton, J. T., Stern, M., Perin, M., and Bellen, H. J. (1994). Calcium dependence of
neurotransmitter release and rate of spontaneous vesicle fusions are altered in Drosophila
synaptotagmin mutants. Proc. Natl. Acad. Sci. USA 91, 10888–10892.
Littleton, J. T., Serano, T. L., Rubin, G. M., Ganetzky, B., and Chapman, E. R. (1999). Synaptic
function modulated by changes in the ratio of synaptotagmin I and IV. Nature 400, 757–760.
Littleton, J. T., Barnard, R. J., Titus, S. A., Slind, J., Chapman, E. R., and Ganetzky, B. (2001).
SNARE-complex disassembly by NSF follows synaptic-vesicle fusion. Proc. Natl. Acad. Sci. USA
98, 12233–12238.
Lnenicka, G. A., and Keshishian, H. (2000). Identified motor terminals in Drosophila larvae show
distinct diVerences in morphology and physiology. J. Neurobiol. 43, 186–197.
Loewen, C. A., Mackler, J. M., and Reist, N. E. (2001). Drosophila synaptotagmin I null mutants
survive to early adulthood. Genesis 31, 30–36.
Loewen, C. A., Royer, S. M., and Reist, N. E. (2006). Drosophila synaptotagmin I null mutants show
severe alterations in vesicle populations but calcium-binding motif mutants do not. J. Comp.
Neurol. 496, 1–12.
Mackler, J. M., Drummond, J. A., Loewen, C. A., Robinson, I. M., and Reist, N. E. (2002). The C(2)B
Ca(2þ)-binding motif of synaptotagmin is required for synaptic transmission in vivo. Nature 418,
340–344.
Marek, K. W., and Davis, G. W. (2002). Transgenically encoded protein photoinactivation (FlAsH-
FALI): Acute inactivation of synaptotagmin I. Neuron 36, 805–813.
Martin, J. R., Keller, A., and Sweeney, S. T. (2002). Targeted expression of tetanus toxin: A new tool
to study the neurobiology of behavior. Adv. Genet. 47, 1–47.
Matos, M. F., Rizo, J., and Sudhof, T. C. (2000). The relation of protein binding to function: What is
the significance of munc18 and synaptotagmin binding to syntaxin 1, and where are the
corresponding binding sites? Eur. J. Cell Biol. 79, 377–382.
McNew, J. A., Parlati, F., Fukuda, R., Johnston, R. J., Paz, K., Paumet, F., Sollner, T. H., and
Rothman, J. E. (2000). Compartmental specificity of cellular membrane fusion encoded in
SNARE proteins. Nature 407, 153–159.
Mehta, S. Q., Hiesinger, P. R., Beronja, S., Zhai, R. G., Schulze, K. L., Verstreken, P., Cao, Y.,
Zhou, Y., Tepass, U., Crair, M. C., and Bellen, H. J. (2005). Mutations in Drosophila sec15
reveal a function in neuronal targeting for a subset of exocyst components. Neuron 46,
219–232.
Michels, B., Diegelmann, S., Tanimoto, H., Schwenkert, I., Buchner, E., and Gerber, B. (2005).
A role for Synapsin in associative learning: The Drosophila larva as a study case. Learn. Mem. 12,
224–231.
Mohtashami, M., Stewart, B. A., Boulianne, G. L., and Trimble, W. S. (2001). Analysis of the mutant
Drosophila N-ethylmaleimide sensitive fusion-1 protein in comatose reveals molecular correlates of
the behavioural paralysis. J. Neurochem. 77, 1407–1417.
TRANSMITTER RELEASE AT THE NMJ 141
Murthy, M., and Schwarz, T. L. (2004). The exocyst component Sec5 is required for membrane
traYc and polarity in the Drosophila ovary. Development 131, 377–388.
Murthy, M., Garza, D., Scheller, R. H., and Schwarz, T. L. (2003). Mutations in the exocyst
component Sec5 disrupt neuronal membrane traYc, but neurotransmitter release persists. Neuron
37, 433–447.
Murthy, M., Ranjan, R., Denef, N., Higashi, M. E., Schupbach, T., and Schwarz, T. L. (2005). Sec6
mutations and the Drosophila exocyst complex. J. Cell Sci. 118, 1139–1150.
Nalefski, E. A., Wisner, M. A., Chen, J. Z., Sprang, S. R., Fukuda, M., Mikoshiba, K., and Falke, J. J.
(2001). C2 domains from diVerent Ca2þ signaling pathways display functional and mechanistic
diversity. Biochemistry 40, 3089–3100.
Nichols, B. J., Ungermann, C., Pelham, H. R., Wickner, W. T., and Haas, A. (1997). Homotypic
vacuolar fusion mediated by t- and v-SNAREs. Nature 387, 199–202.
Niemeyer, B. A., and Schwarz, T. L. (2000). SNAP-24, a Drosophila SNAP-25 homologue on granule
membranes, is a putative mediator of secretion and granule-granule fusion in salivary glands.
J. Cell Sci. 113(Pt. 22), 4055–4064.
Nonet, M. L., Grundahl, K., Meyer, B. J., and Rand, J. B. (1993). Synaptic function is impaired but
not eliminated in C. elegans mutants lacking synaptotagmin. Cell 73, 1291–1305.
Novick, P., Field, C., and Schekman, R. (1980). Identification of 23 complementation groups required
for post-translational events in the yeast secretory pathway. Cell 21, 205–215.
Ohtsuka, T., Takao-Rikitsu, E., Inoue, E., Inoue, M., Takeuchi, M., Matsubara, K., Deguchi-
Tawarada, M., Satoh, K., Morimoto, K., Nakanishi, H., and Takai, Y. (2002). Cast: A novel
protein of the cytomatrix at the active zone of synapses that forms a ternary complex with RIM1
and munc13–1. J. Cell Biol. 158, 577–590.
Okamoto, T., Tamura, T., Suzuki, K., and Kidokoro, Y. (2005). External Ca2þ dependency of
synaptic transmission in Drosophila synaptotagmin I mutants. J. Neurophysiol. 94, 1574–1586.
Pallanck, L., Ordway, R. W., and Ganetzky, B. (1995). A Drosophila NSF mutant. Nature 376, 25.
Pellizzari, R., Rossetto, O., Schiavo, G., and Montecucco, C. (1999). Tetanus and botulinum
neurotoxins: Mechanism of action and therapeutic uses. Philos. Trans. R. Soc. Lond. B. Biol. Sci.
354, 259–268.
Perin, M. S., Fried, V. A., Mignery, G. A., Jahn, R., and Sudhof, T. C. (1990). Phospholipid binding
by a synaptic vesicle protein homologous to the regulatory region of protein kinase C. Nature 345,
260–263.
Peters, C., Bayer, M. J., Buhler, S., Andersen, J. S., Mann, M., and Mayer, A. (2001). Trans-complex
formation by proteolipid channels in the terminal phase of membrane fusion. Nature 409,
581–588.
Poskanzer, K. E., Marek, K. W., Sweeney, S. T., and Davis, G. W. (2003). Synaptotagmin I is
necessary for compensatory synaptic vesicle endocytosis in vivo. Nature 426, 559–563.
Ranjan, R., Bronk, P., and Zinsmaier, K. E. (1998). Cysteine string protein is required for calcium
secretion coupling of evoked neurotransmission in Drosophila but not for vesicle recycling.
J. Neurosci. 18, 956–964.
Rao, S. S., Stewart, B. A., Rivlin, P. K., Vilinsky, I., Watson, B. O., Lang, C., Boulianne, G., Salpeter,
M. M., and Deitcher, D. L. (2001). Two distinct eVects on neurotransmission in a temperature-
sensitive SNAP-25 mutant. EMBO J. 20, 6761–6771.
Reid, C. A., Bekkers, J. M., and Clements, J. D. (2003). Presynaptic Ca2þ channels: A functional
patchwork. Trends Neurosci. 26, 683–687.
Reist, N. E., Buchanan, J., Li, J., DiAntonio, A., Buxton, E. M., and Schwarz, T. L. (1998).
Morphologically docked synaptic vesicles are reduced in synaptotagmin mutants of Drosophila.
J. Neurosci. 18, 7662–7673.
Rhee, J. S., Betz, A., Pyott, S., Reim, K., Varoqueaux, F., Augustin, I., Hesse, D., Sudhof,
T. C., Takahashi, M., Rosenmund, C., and Brose, N. (2002). Beta phorbol ester- and
142 THOMAS L. SCHWARZ
diacylglycerol-induced augmentation of transmitter release is mediated by Munc13s and not
by PKCs. Cell 108, 121–133.
Richards, D. A., Guatimosim, C., and Betz, W. J. (2000). Two endocytic recycling routes selectively
fill two vesicle pools in frog motor nerve terminals. Neuron 27, 551–559.
Richmond, J. E., Davis, W. S., and Jorgensen, E. M. (1999). UNC-13 is required for synaptic vesicle
fusion in C. elegans. Nat. Neurosci. 2, 959–964.
Richmond, J. E., Weimer, R. M., and Jorgensen, E. M. (2001). An open form of syntaxin bypasses the
requirement for UNC-13 in vesicle priming. Nature 412, 338–341.
Rieckhof, G. E., Yoshihara, M., Guan, Z., and Littleton, J. T. (2003). Presynaptic N-type calcium
channels regulate synaptic growth. J. Biol. Chem. 278, 41099–41108.
Risinger, C., Deitcher, D. L., Lundell, I., Schwarz, T. L., and Larhammar, D. (1997). Complex gene
organization of synaptic protein SNAP-25 in Drosophila melanogaster. Gene 194, 169–177.
Rizo, J., and Sudhof, T. C. (1998). C2-domains, structure and function of a universal Ca2þ-bindingdomain. J. Biol. Chem. 273, 15879–15882.
Robinson, I. M., Ranjan, R., and Schwarz, T. L. (2002). Synaptotagmins I and IV promote
transmitter release independently of Ca(2þ) binding in the C(2)A domain. Nature 418, 336–340.
Rosahl, T. W., Geppert, M., Spillane, D., Herz, J., Hammer, R. E., Malenka, R. C., and Sudhof,
T. C. (1993). Short-term synaptic plasticity is altered in mice lacking synapsin I. Cell 75,
661–670.
Rosahl, T. W., Spillane, D., Missler, M., Herz, J., Selig, D. K., WolV, J. R., Hammer, R. E., Malenka,
R. C., and Sudhof, T. C. (1995). Essential functions of synapsins I and II in synaptic vesicle
regulation. Nature 375, 488–493.
Saitoe, M., Schwarz, T. L., Umbach, J. A., Gundersen, C. B., and Kidokoro, Y. (2001). Absence of
junctional glutamate receptor clusters in Drosophila mutants lacking spontaneous transmitter
release. Science 293, 514–517.
Saitoe, M., Schwarz, T. L., Umbach, J. A., Gundersen, C. B., and Kidokoro, Y. (2002). Response:
Meaningless minis? Trends Neurosci. 25, 385–386.
Schulze, K. L., Littleton, J. T., Salzberg, A., Halachmi, N., Stern, M., Lev, Z., and Bellen, H. J.
(1994). rop, a Drosophila homolog of yeast Sec1 and vertebrate n-Sec1/Munc-18 proteins, is a
negative regulator of neurotransmitter release in vivo. Neuron 13, 1099–1108.
Schulze, K. L., Broadie, K., Perin, M. S., and Bellen, H. J. (1995). Genetic and electrophysiological
studies of Drosophila syntaxin-1A demonstrate its role in nonneuronal secretion and
neurotransmission. Cell 80, 311–320.
Schwarz, T. L. (1999). NSF is up to new tricks. Nat. Cell Biol. 1, E141–E143.
Shao, X., Davletov, B. A., Sutton, R. B., Sudhof, T. C., and Rizo, J. (1996). Bipartite Ca2þ-bindingmotif in C2 domains of synaptotagmin and protein kinase C. Science 273, 248–251.
Shao, X., Fernandez, I., Sudhof, T. C., and Rizo, J. (1998). Solution structures of the Ca2þ-free andCa2þ-bound C2A domain of synaptotagmin I: Does Ca2þ induce a conformational change?
Biochemistry 37, 16106–16115.
Siddiqi, O., and Benzer, S. (1976). Neurophysiological defects in temperature-sensitive paralytic
mutants of Drosophila melanogaster. Proc. Natl. Acad. Sci. USA 73, 3253–3257.
Sollner, T., Bennett, M. K., Whiteheart, S. W., Scheller, R. H., and Rothman, J. E. (1993a). A protein
assembly-disassembly pathway in vitro that may correspond to sequential steps of synaptic vesicle
docking, activation, and fusion. Cell 75, 409–418.
Sollner, T., Whiteheart, S. W., Brunner, M., Erdjument-Bromage, H., Geromanos, S., Tempst, P.,
and Rothman, J. E. (1993b). SNAP receptors implicated in vesicle targeting and fusion. Nature
362, 318–324.
Stewart, B. A., Atwood, H. L., Renger, J. J., Wang, J., and Wu, C. F. (1994). Improved stability of
Drosophila larval neuromuscular preparations in haemolymph-like physiological solutions.
J. Comp. Physiol. [A] 175, 179–191.
TRANSMITTER RELEASE AT THE NMJ 143
Stewart, B. A., Mohtashami, M., Trimble, W. S., and Boulianne, G. L. (2000). SNARE proteins
contribute to calcium cooperativity of synaptic transmission. Proc. Natl. Acad. Sci. USA 97,
13955–13960.
Stewart, B. A., Mohtashami, M., Rivlin, P., Deitcher, D. L., Trimble, W. S., and Boulianne, G. L.
(2002). Dominant-negative NSF2 disrupts the structure and function of Drosophila neuromuscular
synapses. J. Neurobiol. 51, 261–271.
Stewart, B. A., Pearce, J., Bajec, M., and Khorana, R. (2005). Disruption of synaptic development
and ultrastructure by Drosophila NSF2 alleles. J. Comp. Neurol. 488, 101–111.
Steyer, J. A., Horstmann, H., and Almers, W. (1997). Transport, docking and exocytosis of single
secretory granules in live chromaYn cells. Nature 388, 474–478.
Stowers, R. S., and Schwarz, T. L. (1999). A genetic method for generating Drosophila eyes composed
exclusively of mitotic clones of a single genotype. Genetics 152, 1631–1639.
Sudhof, T. C., and Rizo, J. (1996). Synaptotagmins: C2-domain proteins that regulate membrane
traYc. Neuron 17, 379–388.
Sudhof, T. C., Baumert, M., Perin, M. S., and Jahn, R. (1989). A synaptic vesicle membrane protein
is conserved from mammals to Drosophila. Neuron 2, 1475–1481.
Sun, J., Bronk, P., Liu, X., Han, W., and Sudhof, T. C. (2006). Synapsins regulate use-dependent
synaptic plasticity in the calyx of Held by a Ca2þ/calmodulin-dependent pathway. Proc. Natl.
Acad. Sci. USA 103, 2880–2885.
Suster, M. L., Martin, J. R., Sung, C., and Robinow, S. (2003). Targeted expression of tetanus toxin
reveals sets of neurons involved in larval locomotion in Drosophila. J. Neurobiol. 55, 233–246.
Sweeney, S. T., Broadie, K., Keane, J., Niemann, H., and O’Kane, C. J. (1995). Targeted expression
of tetanus toxin light chain in Drosophila specifically eliminates synaptic transmission and causes
behavioral defects. Neuron 14, 341–351.
Tobaben, S., Thakur, P., Fernandez-Chacon, R., Sudhof, T. C., Rettig, J., and Stahl, B. (2001).
A trimeric protein complex functions as a synaptic chaperone machine. Neuron 31, 987–999.
Tolar, L. A., and Pallanck, L. (1998). NSF function in neurotransmitter release involves rearrange-
ment of the SNARE complex downstream of synaptic vesicle docking. J. Neurosci. 18,
10250–10256.
Umbach, J. A., Zinsmaier, K. E., Eberle, K. K., Buchner, E., Benzer, S., and Gundersen, C. B.
(1994). Presynaptic dysfunction in Drosophila csp mutants. Neuron 13, 899–907.
van de Goor, J., Ramaswami, M., and Kelly, R. (1995). Redistribution of synaptic vesicles and their
proteins in temperature-sensitive shibire(ts1) mutant Drosophila. Proc. Natl. Acad. Sci. USA 92,
5739–5743.
Verstreken, P., KjaerulV, O., Lloyd, T. E., Atkinson, R., Zhou, Y., Meinertzhagen, I. A., and Bellen,
H. J. (2002). Endophilin mutations block clathrin-mediated endocytosis but not neurotransmitter
release. Cell 109, 101–112.
Verstreken, P., Koh, T. W., Schulze, K. L., Zhai, R. G., Hiesinger, P. R., Zhou, Y., Mehta, S. Q.,
Cao, Y., Roos, J., and Bellen, H. J. (2003). Synaptojanin is recruited by endophilin to promote
synaptic vesicle uncoating. Neuron 40, 733–748.
Vilinsky, I., Stewart, B. A., Drummond, J., Robinson, I., and Deitcher, D. L. (2002). A Drosophila
SNAP-25 null mutant reveals context-dependent redundancy with SNAP-24 in neurotransmis-
sion. Genetics 162, 259–271.
Wagh, D. A., Rasse, T. M., Asan, E., Hofbauer, A., Schwenkert, I., Durrbeck, H., Buchner, S.,
Dabauvalle, M. C., Schmidt, M., Qin, G., Wichmann, C., Kittel, R., et al. (2006). Bruchpilot, a
protein with homology to ELKS/CAST, is required for structural integrity and function of
synaptic active zones in Drosophila. Neuron 49, 833–844.
Wang, Y., Liu, X., Biederer, T., and Sudhof, T. C. (2002). A family of RIM-binding proteins
regulated by alternative splicing: Implications for the genesis of synaptic active zones. Proc. Natl.
Acad. Sci. USA 99, 14464–14469.
144 THOMAS L. SCHWARZ
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.
Wu, M. N., Littleton, J. T., Bhat, M. A., Prokop, A., and Bellen, H. J. (1998). ROP, the Drosophila
Sec1 homolog, interacts with syntaxin and regulates neurotransmitter release in a dosage-
dependent manner. EMBO J. 17, 127–139.
Wu, M. N., Fergestad, T., Lloyd, T. E., He, Y., Broadie, K., and Bellen, H. J. (1999). Syntaxin 1A
interacts with multiple exocytic proteins to regulate neurotransmitter release in vivo. Neuron 23,
593–605.
Wu, M. N., Schulze, K. L., Lloyd, T. E., and Bellen, H. J. (2001). The ROP-syntaxin interaction
inhibits neurotransmitter release. Eur. J. Cell Biol. 80, 196–199.
Yang, H., and Kunes, S. (2004). Nonvesicular release of acetylcholine is required for axon targeting in
the Drosophila visual system. Proc. Natl. Acad. Sci. USA 101, 15213–15218.
Yoshihara, M., and Littleton, J. T. (2002). Synaptotagmin I functions as a calcium sensor to
synchronize neurotransmitter release. Neuron 36, 897–908.
Yoshihara, M., Ueda, A., Zhang, D., Deitcher, D. L., Schwarz, T. L., and Kidokoro, Y. (1999).
Selective eVects of neuronal-synaptobrevin mutations on transmitter release evoked by sustained
versus transient Ca2þ increases and by cAMP. J. Neurosci. 19, 2432–2441.
Yoshihara, M., Suzuki, K., and Kidokoro, Y. (2000). Two independent pathways mediated by cAMP
and protein kinase A enhance spontaneous transmitter release at Drosophila neuromuscular
junctions. J. Neurosci. 20, 8315–8322.
Yoshihara, M., Adolfsen, B., Galle, K. T., and Littleton, J. T. (2005). Retrograde signaling by Syt 4
induces presynaptic release and synapse-specific growth. Science 310, 858–863.
Zhang, B., Koh, Y. H., Beckstead, R. B., Budnik, V., Ganetzky, B., and Bellen, H. J. (1998a). Synaptic
vesicle size and number are regulated by a clathrin adaptor protein required for endocytosis.
Neuron 21, 1465–1475.
Zhang, J. Z., Davletov, B. A., Sudhof, T. C., and Anderson, R. G. (1994). Synaptotagmin I is a high
aYnity receptor for clathrin AP-2: Implications for membrane recycling. Cell 78, 751–760.
Zhang, X., Rizo, J., and Sudhof, T. C. (1998b). Mechanism of phospholipid binding by the
C2A-domain of synaptotagmin I. Biochemistry 37, 12395–12403.
Zhou, Z., and Misler, S. (1995). Amperometric detection of stimulus-induced quantal release of
catecholamines from cultured superior cervical ganglion neurons. Proc. Natl. Acad. Sci. USA 92,
6938–6942.
Zinsmaier, K. E., Eberle, K. K., Buchner, E., Walter, N., and Benzer, S. (1994). Paralysis and early
death in cysteine string protein mutants of Drosophila. Science 263, 977–980.