Neuron 46, 1–9, January 5, 2006 ª2006 Elsevier Inc.
mitter into the lumen of synaptic vesicles for quantal
release. However, the number of transporters that lo-
calize to each vesicle is not known. In this issue of
Neuron, a study by Daniels et al. using the Drosophila
glutamate transporter suggests that one transporter
may suffice to fill each vesicle.
Some 50 years ago, Bernard Katz and colleagues dem-
onstrated that neurotransmitter is released as packets,
or quanta, and a wealth of subsequent data have shown
that synaptic vesicles (SVs) are the physical correlate of
these quanta (Atwood and Karunanithi, 2002). Vesicular
neurotransmitter transporters are responsible for pack-
aging neurotransmitter into the lumen of the vesicle and
thereby generating the concentration and absolute
quantity of transmitter that makes up each quanta (see
Hediger et al., 2004 and accompanying articles). But
how much neurotransmitter is contained in the lumen of
each vesicle? Is that number absolutely fixed, or might
it vary depending on the activity of the transporter and
the number of transporters that localize to each vesicle?
And finally, how many transporters are required to local-
icle? The molecular-genetic analysis of vesicular trans-
porters has begun to yield answers to some of these
questions (Colliver et al., 2000; Fremeau et al., 2004;
Pothos et al., 2000; Wilson et al., 2005; Wojcik et al.,
2004), and in this issue of Neuron, a paper from the
DiAntonio lab suggests that the number of transporters
required to fill a vesicle may be as few as one (Daniels
et al., 2006).
The success of these experiments and others that
track transmitter storage and release depends on the
ability toaccurately quantitate the amount of transmitter
in a single vesicle. Carbon fiber electrode amperometry
can directly measure the content of monoamine neuro-
transmitters such as dopamine and serotonin from indi-
vidual secretory vesicles (Bruns and Jahn, 1995; Colliver
et al., 2000; Pothos et al., 2000). In contrast, for neuro-
transmitters that are not as easily oxidized, transmitter
release from single SVs must be measured indirectly
by recording the electrophysiologic response of a post-
synaptic cell. The electrophysiologic response to trans-
mitter released by a single vesicle corresponds to the
miniature excitatory postsynaptic potential or ‘‘mini’’
first described by Katz at the frog neuromuscular junc-
tion (NMJ). In flies, as in vertebrates, minis can be reli-
metric for assessing the amount of transmitter in each
vesicle (Atwood and Karunanithi, 2002). Furthermore,
the electrophysiological preparations for recording at
the fly NMJ are extremely robust and have been used
apse (Atwood and Karunanithi, 2002; Karunanithi et al.,
2002). In contrast to mammals and other vertebrates,
rotransmitter released at the fly NMJ is glutamate, and
ble for packaging the transmitter in the fly motoneuron.
In this paper and another recent report, the DiAntonio
group have used the fly NMJ to determine how changes
in VGLUT expression might contribute to the size of
a mini (Daniels et al., 2006; Daniels et al., 2004).
Although three VGLUT genes have been identified in
mammals, the genome of Drosophila is more parsimoni-
ous and contains a single isoform, thus facilitating the
analysis of DVGLUT using classical genetic techniques.
Indeed, the novelty of the study by Daniels et al.
emerges from the interplay between fly genetics and
electrophysiology. Rather than ‘‘knocking out’’ the
dvglut gene and completely eliminating its function as
was reported recently for mouse VGLUT1 (Fremeau
et al., 2004; Wojcik et al., 2004), Daniels et al. generated
an allelic series of dvglut mutants in which the flies show
a graded decrease in the amount of transporter that is
expressed in glutamatergic motoneurons. The use of
these weak or ‘‘hypomorphic’’ alleles reduced but did
not eliminate VGLUT expression and allowed the au-
thors to target a reduced quantity of functional trans-
porters to the SVs at the nerve terminal of the NMJ.
The key to exploring the relationship between trans-
mitter content and transport function is to know how
many transporters are on an individual vesicle. It is pos-
sible that increasingly sophisticated proteomic techni-
ques may someday answer this question directly. In
the absence of that ability, the DiAntonio lab relied on
what amounts to the genetic equivalent of limiting dilu-
tion to force that number to one. They reasoned that if
the number of SVs remains relatively intact in dvglut mu-
tants, then as the number of transporters is decreased,
of a large number of vesicles without a transporter sug-
gests that the SVs that do have a transporter will have
only one copy. Using quantitative electron microscopy,
they show that the number of SVs is not dramatically re-
duced in the dvglut mutants. In contrast, the number of
minis recorded is significantly reduced as the amount
of the DVGLUT expressed at each terminal declines.
These data suggest that a large fraction of the vesicles
are devoid of transmitter and thus unable to produce
a mini. However, this finding in itself does not tell us the
number of transporters that reside on a normal vesicle.
generated by SVs that have only one vesicular trans-
porter. Relative to wild-type flies, the size of the minis
in the dvglut mutants is unchanged. That is, even as
you limit the number of transporters such that each SV
is likely to have one functional unit of DVGLUT, all SVs
fill to the same capacity and fill equivalently to wild-
type vesicles. The authors conclude that a single func-
tional unit of DVGLUT is sufficient to fill an SV (Figure 1).
The idea that one and only one functional transport
unit may reside on an SV is at odds with some
mammalian studies (Wojcik et al., 2004). Therefore, the
authors performed a series of critical controls to rule
out other explanations for the results. One possible rea-
son for the maintenance of SV filling in the face of dimin-
ished DVGLUT expression is that another transporter
compensates for the deficit. If this were true, then this
gene should also compensate for the absence of
DVGLUT in more extreme null dvglut alleles. Since minis
are absent in the null animals, it is unlikely that another
transporter can substitute for DVGLUT at the NMJ. An-
other potential explanation for the maintenance of mini
size is that postsynaptic receptors are upregulated or
more sensitive to lower concentrations oftransmitter re-
lease; however, the authors did not detect a change in
the function of postsynaptic receptors. Finally, the re-
sults could be biased by a change in the release proper-
ties of empty or partially filled vesicles or a change in
vesicle biogenesis in the absence of DVGLUT. VGLUT1
knockout mice do indeed show a 50% change in the
number of vesicles at the synapse, and it is possible
that VGLUT1 knockout mice show defects in at least
a portion of recycling SVs (Fremeau et al., 2004). How-
ever, dvglut hypomorphs show only a minimal change
in vesicle number, and experiments using the lipid dye
FM1-43 indicate that SVs recycle normally in dvglut mu-
These data support the surprising idea that a single
functional unit of DVGLUT is sufficient to fill an SV to
its normalcapacity. Itremainstobedeterminedwhether
monomer. For plasma membrane transporters, abun-
dant biochemical evidence and more recent crystallog-
form stable oligomers (Sonders et al., 2005). Although
there is no clear evidence that vesicular transporters
form similar, stable complexes, it remains possible
that monomers could loosely associate to form an ac-
tive, functional unit.
The possibility of VGLUT oligomers aside, the finding
that a single transporter unit may be necessary and suf-
ficient to fill a single vesicle has a variety of interesting
ular transporters may be exquisitely regulated to main-
tain the appropriate ratio of 1:1. Alternatively, if the syn-
apse is less stringent, there may be a large number of
empty vesicles in the average neuron that do not receive
tain neurotransmitter. The precise matching of trans-
porter to vesicle might be further stressed during pro-
longed stimulation and increased turnover of SVs at
the nerve terminal. In this scenario, the relative number
of empty vesicles might be further increased and per-
haps contribute to the rundown of transmitter release.
These possibilities remain highly speculative. Nonethe-
et al., 2005) highlight the potential importance of vesicu-
lar transporters in regulating synaptic transmission and
their use in unraveling its fundamental properties.
David E. Krantz1
1Department of Psychiatry
and Biobehavioral Sciences
Hatos Center for Neuropharmacology and
Semel Institute for Neuroscience
and Human Behavior
The David Geffen School of Medicine
at the University of California, Los Angeles
Gonda (Goldschmied) Center
for Genetic and Neuroscience Research
695 Charles Young Drive
Los Angeles, California 90095
Atwood, H.L., and Karunanithi, S. (2002). Nat. Rev. Neurosci. 3, 497–
Bruns, D., and Jahn, R. (1995). Nature 377, 62–65.
Colliver, T.L., Pyott, S.J., Achalabun, M., and Ewing, A.G. (2000).
J. Neurosci. 20, 5276–5282.
Daniels, R.W., Collins, C.A., Gelfand, M.V., Dant, J., Brooks, E.S.,
Krantz, D.E., and DiAntonio, A. (2004). J. Neurosci. 24, 10466–10474.
Daniels, R.W., Collins, C.A., Chen, K., Gelfand, M.V., Featherstone,
D.E., and DiAntonio, A. (2006). Neuron 49, this issue, 11–16.
Fremeau, R.T., Jr., Kam, K., Qureshi, T., Johnson, J., Copenhagen,
D.R., Storm-Mathisen, J., Chaudhry, F.A., Nicoll, R.A., and Edwards,
R.H. (2004). Science 304, 1815–1819.
Hediger, M.A., Romero, M.F., Peng, J.B., Rolfs, A., Takanaga, H.,
and Bruford, E.A. (2004). Pflu ¨gers Arch. 447, 465–468.
Karunanithi, S., Marin, L., Wong, K., and Atwood, H.L. (2002). J. Neu-
rosci. 22, 10267–10276.
Pothos, E.N., Larsen, K.E., Krantz, D.E., Liu, Y., Haycock, J.W., Set-
lik, W., Gershon, M.D., Edwards, R.H., and Sulzer, D. (2000). J. Neu-
rosci. 20, 7297–7306.
biol. 15, 296–304.
Wilson, N.R., Kang, J., Hueske, E.V., Leung, T., Varoqui, H., Murnick,
J.G., Erickson, J.D., and Liu, G. (2005). J. Neurosci. 25, 6221–6234.
Wojcik, S.M., Rhee, J.S., Herzog, E., Sigler, A., Jahn, R., Takamori,
S., Brose, N., and Rosenmund, C. (2004). Proc. Natl. Acad. Sci.
USA 101, 7158–7163.
Figure 1. How Many Transporters Does a Vesicle Need?
The number of vesicular transporters required to fill a synaptic ves-
icle with neurotransmitter is not known. Mutations in the dvglut gene
decrease transporter expression such that most vesicles are empty
and do not produce minis. It is therefore likely that vesicles with only
one transporter (blue pyramid) produce the few minis that remain in
the mutant NMJ. dvglut mutant minis are equal in size to wild-type
(see box), suggesting a model in which one transporter normally fills
a synaptic vesicle to capacity.