Development of autotaxin inhibitors: A series of tetrazole cinnamides

Article abstract of DOI:10.1016/j.bmcl.2020.127663

A series of inhibitors of Autotaxin (ATX) have been developed from a high throughput screening hit, 1a, which shows an alternative binding mode to known catalytic site inhibitors. Selectivity over the hERG channel and microsomal clearance were dependent on the lipophilicity of the compounds, and this was optimised by reduction of clogD whilst maintaining high affinity ATX inhibition. Compound 15a shows good oral exposure, and concentration dependent inhibition of formation of LPA in vivo, as shown in pharmacokinetic-pharmacodynamic (PK/PD) experiments.

Full text of DOI:10.1016/j.bmcl.2020.127663

Neuron
Article
ELP3 Controls Active Zone Morphology
by Acetylating the ELKS Family Member Bruchpilot
Katarzyna Miskiewicz,1,2 Liya E. Jose,1,2 Andre Bento-Abreu,3,4,8 Marcus Fislage,5,6,8 Ines Taes,3,4
ꢀ
´
7
Jaros1aw Kasprowicz,1,2 Jef Swerts,1,2 Stephan Sigrist, Wim Versees,5,6 Wim Robberecht,3,4 and Patrik Verstreken1,2,
*
´
1VIB, Department of Molecular and Developmental Genetics, B3000 Leuven, Belgium
2K.U.Leuven, Center for Human Genetics, B3000 Leuven, Belgium
3VIB, Vesalius Research Center, B3000 Leuven, Belgium
4K.U.Leuven, Laboratory for Neurobiology, Experimental Neurology, B3000 Leuven, Belgium
5VIB, Department of Molecular and Cellular Interactions, K.U.Leuven, B1050 Brussels, Belgium
6Vrije Universiteit Brussel, Structural Biology Brussels, B1050 Brussels, Belgium
7
¨
Freie Universitat Berlin, Department of Genetics, Institute for Biology, 14195 Berlin, Germany
8These authors contributed equally to this work
*Correspondence: patrik.verstreken@cme.vib-kuleuven.be
DOI 10.1016/j.neuron.2011.10.010
SUMMARY
et al., 2006). BRP self-assembles in macromolecular entities
where individual BRP strands join at their N-terminal ends near
Elongator protein 3 (ELP3) acetylates histones in the the plasma membrane while sending their C-terminal ends into
the cytoplasm like a parasol (Fouquet et al., 2009; Jiao et al.,
2010). Similar to presynaptic specializations in other species,
nucleus but also plays a role in the cytoplasm. Here,
we report that in Drosophila neurons, ELP3 is neces-
sary and sufficient to acetylate the ELKS family mem-
ber Bruchpilot, an integral component of the presyn-
aptic density where neurotransmitters are released.
We find that in elp3 mutants, presynaptic densities
assemble normally, but they show morphological
BRP is thought to capture synaptic vesicles using its C-terminal
extensions, concentrating synaptic vesicles at active zones and
facilitating synaptic transmission (Hallermann et al., 2010b; Zhai
and Bellen, 2004). Although the abundance of BRP at individual
active zones correlates with the release efficiency (Graf et al.,
2009; Marrus et al., 2004; Schmid et al., 2008), little is known
defects such that their cytoplasmic extensions
cover a larger area, resulting in increased vesicle
tethering as well as a more proficient neurotrans-
about the molecular mechanisms that regulate the function of
presynaptic release sites. Here, we identify Elongator protein 3
(ELP3), a member of the elongator complex as a regulator of
mitter release. We propose a model where ELP3- T bar function and morphology.
ELP3 was originally identified in yeast as a member of the
dependent acetylation of Bruchpilot at synapses
regulates the structure of individual presynaptic
densities and neurotransmitter release efficiency.
nuclear elongator complex (Otero et al., 1999). ELP3 harbors
a Gcn5-related acetyltransferase (GNAT) domain and acetylates
lysines in histone H3 (Han et al., 2008; Winkler et al., 2002) to
modify DNA chromatin structure (Walia et al., 1998). The ELP3
ortholog in plants is largely nuclear, however, in yeast and
several other species, the protein also localizes to the cytoplasm
INTRODUCTION
Synapses are highly specialized structures with tightly apposed where it is thought to take part in tRNA modification and acety-
pre- and postsynaptic elements (Haucke et al., 2011). While the lation of tubulin; however, the mechanistic details are elusive
´
basic building blocks of synapses within a cell may be similar, (Creppe et al., 2009; Solinger et al., 2010; Versees et al., 2010).
synaptic contacts are not invariant, and synaptic efficacy of indi- Interestingly, ELP3 polymorphisms have been associated with
vidual release sites differs (Marrus et al., 2004; Peled and Isacoff, decreased risk for amyotrophic lateral sclerosis (Simpson
2011; Pelkey et al., 2006; Schmid et al., 2008). This heterogeneity et al., 2009), and mutations in ELP1 cause familial dysautonomia
suggests that presynaptic release site function may be locally (Cheishvili et al., 2011; Slaugenhaupt and Gusella, 2002).
regulated (Nicoll and Schmitz, 2005; Pelkey and McBain,
To understand ELP3 function, we have investigated the
2007). Thus, characterization of mechanisms that control the neuronal role for ELP3 in vitro and in vivo. We show that presyn-
function of individual active zones will yield insight into the regu- aptic ELP3 loss of function results in altered morphology and
lation of synaptic plasticity in health and disease.
function of T bars at fruit fly neuromuscular junctions (NMJs),
Synaptic vesicles fuse at active zones, specialized presyn- and this occurs in the absence of defects in tubulin acetylation.
aptic structures directly aligned to the postsynaptic receptor We find that T bars in elp3 mutants change their structure in favor
field (Petersen et al., 1997). In Drosophila, active zones harbor of forming more elaborate cytoplasmic extensions, that more
electron-dense T bars, and Bruchpilot (BRP), a large cytoskel- synaptic vesicles are tethered to these T bars, and that neuro-
etal-like protein that is the ortholog of ELKS in mammals, is an transmitter release becomes more efficient, including a larger
integral part of these structures (Hida and Ohtsuka, 2010; Kittel readily releasable vesicle pool (RRP). Our data indicate that
776 Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc.

Neuron
ELP3 Acetylates BRP
Figure 1. ELP3 Localizes to Synapses
(A and B) elp3 locus, elp3 alleles (green) (A), and
genomic elp3 constructs that express GFP N- and
C-tagged ELP3 (B). elp3D3 and elp3D4 deletion
mutants lack the SAM and GNAT domains, and
elp3D5 removes the start codon.
(C) Rescue of lethality of different elp3 mutants
using one copy of GFP-elp3+ or elp3+-GFP or
using neuronal (nSyb-GAL4) or ubiquitous (Act-
GAL4) expression of help3. Lethality scored on
regular food while assessing if adult males and
females emerged. Expression of help3 using
BG57-GAL4 (muscular expression) did not rescue
lethality (not shown).
(D and E) ELP3 localization using elp3+-GFP in
third-instar larval salivary gland cells and in fat
body cells labeled with anti-GFP. DNA is labeled
using TOTO3. Scale bars, 10 mm.
(F–J) L3 VNCs of w1118 (F, not GFP-expressing
controls), GFP-elp3+/+ (G), elp3+-GFP/+ (H and I),
and elp3D3/D4; elp3+-GFP/+ (J) animals labeled
with anti-GFP and/or TOTO3. Scale bars, 10 mm
(F–H and J) and 5 mm (I).
(K and L) NMJs of elp3+-GFP/+ animals labeled
with anti-GFP and anti-DLG, a pre- and post-
synaptic marker (K) or anti-DYN, a presynaptic
marker (L). Scale bars, 10 mm (K) and 2.5 mm (L).
by mobilizing P{SUP or-P}elp3KG02386
,
a P element inserted in the 50UTR of
elp3. We isolated three different deletions
of the elp3 locus (elp3D3, elp3D4, elp3D5
as well as a precise excision (elp3rev
)
)
that serves as a genetic control (Fig-
ure 1A). These deletions fail to comple-
ment one another, as well as elp31 and
elp32, but not lethal alleles of morgue,
located 50 of elp3. Similar to elp3
null mutants (elp3D3/elp3D4), heteroallelic
combinations of the EMS alleles and the
P element excision alleles die as early
pupae, suggesting that all elp3 alleles
we isolated are severe hypomorphic or
null alleles (Walker et al., 2011).
To determine if the lethality and pheno-
types of the elp3 alleles are solely due to
loss of ELP3 function, we created trans-
genic flies that harbor genomic elp3 res-
ELP3 is necessary and sufficient for BRP acetylation in vitro and cue constructs (Figure 1B) (Venken et al., 2006). The constructs
in vivo, and we propose a model where, similar to acetylation of allow expression of a C- or N-terminally GFP-tagged ELP3 under
histones, acetylation of BRP regulates the cytoplasmic exten- native control (Venken et al., 2008). The presence of elp3+-GFP
sions of T bars, thereby controlling the capture of synaptic vesi- or GFP-elp3+ in several heteroallelic combinations restores
cles at active zones and neurotransmitter release efficiency.
viability (Figure 1C; data not shown) and the cellular phenotypes
in elp3 mutants (see below), indicating that only ELP3 function is
affected in the mutants tested.
RESULTS
Elp3 Is an Essential Gene
ELP3 Localizes to the Cytoplasm in Neurons and
We previously isolated two EMS alleles of elp3 (elp31 and elp32) Enriches at Synapses
that harbor missense mutations in the acetyltransferase domain To determine the subcellular localization of Drosophila ELP3, we
(Simpson et al., 2009) and now created independent null alleles labeled elp3+-GFP and GFP-elp3+ with several markers and
Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc. 777

Neuron
ELP3 Acetylates BRP
Figure 2. Tubulin Acetylation Is Not
Affected upon Loss of ELP3 Function
(A–E) NMJs of y w; nSyb-GAL4/+ (A), y w; UAS-
hdac6/+; nSyb-GAL4/+ (B), control y w; elp3rev (C),
and mutant y w; elp3D3/D4 (D) Drosophila L3 larvae
labeled with anti-acetylated tubulin (ac-TUB)
(red) and anti-HRP (green), a neuronal membrane
marker; and (E) quantification of acetylated tubulin
intensity in bouton: mean ac-TUB intensity relative
to anti-HRP intensity normalized to control (%).
Mean is quantified from six to ten NMJs from
at least four animals. Error bars, SEM. t test,
*p < 0.05. Scale bar, 5 mm. ns, not significant.
Related to Figure S1.
(F) Western blot of brains dissected from control,
y w; elp3rev, y,w; elp3D3/D4, y w; Act-GAL4/+, y w;
UAS-hdac6/+; Act-GAL4/+ (+HDAC6) Drosophila
third-instar larvae labeled with antibodies against
ac-TUB and total tubulin (TUB).
(G) Western blot of zebrafish 30 hpf embryos in-
jected with elp3 (elp3-MO) and control (rev-MO)
morpholinos, and either incubated in DMSO
(control) or in tubastatin A (tubast.). Blots were
probed with anti ac-TUB, TUB, or Actin (ACT).
(H) Quantification of axonal length of 30 hpf ze-
brafish embryos after injection of control (rev-MO)
or elp3 (elp3-MO) morpholinos, treated with DMSO
(control) or tubastatin A (tubast.). t test, *p < 0.05.
(I) Western blot of N2a, HEK, NCS34, and cul-
tured cortical neuron cells (CN) transfected with
elp3siRNA or nontargeting NT-siRNA (controls),
probed with ac-TUB and GAPDH (control) anti-
bodies. Related to Figure S1.
assessed GFP distribution. While control animals not expressing combinations (Figure 1C, and see also below). In contrast,
GFP do not show labeling (Figure 1F; data not shown), in several muscular hELP3 expression using BG57-Gal4 does not restore
cell types of third-instar larvae, including salivary gland cells and viability (data not shown). These data indicate an important
fat body cells, ELP3-GFP as well as GFP-ELP3 label the nucleus role for ELP3 in the nervous system and presynaptically at the
and/or the cytoplasm (Figures 1D and 1E; data not shown). In NMJ and also suggest that the function of ELP3 is evolutionary
contrast, in neurons of the ventral nerve cord (VNC) in third-instar conserved.
larvae, we observe abundant ELP3 that concentrates in the cyto-
plasm, and we do not observe much nuclear labeling overlap- Tubulin Acetylation Is Normal in elp3 Null Mutant Larvae
ping with Toto-3, a DNA marker. Furthermore, our data indicate ELP3 harbors an acetyltransferase domain, and recent evidence
that ELP3 concentrates in the synaptic-rich areas of the VNC and suggests that this function is important to mediate tubulin acet-
overlaps with the synaptic markers anti-Discs Large (DLG) and ylation (Creppe et al., 2009; Solinger et al., 2010). To test if ELP3
anti-Dynamin (DYN; Figures 1G–1J; data not shown). Similarly, plays a role in neuronal tubulin acetylation in vivo, we labeled
also in mouse motor neurons in culture, we observe abundant acetylated tubulin with specific antibodies in controls and elp3
cytoplasmic ELP3 localization, indicating that this feature is null mutant Drosophila larvae. As a control we overexpressed
evolutionary conserved (data not shown). In Drosophila larvae, HDAC6 (nsyb-GAL4), previously shown to act as a tubulin
ELP3-GFP is also present at the presynaptic side of NMJ deacetylase (Hubbert et al., 2002). While neuronal HDAC6
boutons, double labeled with anti-DLG or with anti-DYN (Figures overexpression results in reduced acetylated tubulin labeling in
1K and 1L). Thus, our data suggest a cytoplasmic role for ELP3 in motor neurons (Figures 2A, 2B, and 2E), loss of ELP3 function
motor neurons.
does not result in a difference in labeling intensity (Figures 2C–2E;
To test whether ELP3 plays an important role in the nervous see Figures S1A–S1C available online). Similarly, western blot
system, we generated transgenic animals that harbor a UAS- of acetylated tubulin also does not show a difference between
human ELP3 construct. Driving expression of hELP3 ubiqui- elp3 null mutants and controls (Figure 2F). Furthermore,
tously using Act-Gal4 rescues lethality associated with elp3 kinesin-driven axonal transport of synaptic vesicles, thought to
loss of function (elp31/elp32; elp3D3/elp3D4), and these flies be regulated by microtubule acetylation, is also not affected in
show normal electroretinogram recordings (data not shown) elp3 mutants (Figures S1D–S1F), indicating that acetylation of
(Simpson et al., 2009), indicating that the construct is functional microtubules in Drosophila larvae that lack elp3 is not affected.
(Figure 1C). Driving hELP3 specifically in the nervous system To test if ELP3 is involved in tubulin acetylation in vertebrates,
using nsyb-Gal4 also rescues lethality of elp3 heteroallelic we performed western blots using extracts of zebrafish embryos
778 Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc.

Neuron
ELP3 Acetylates BRP
(30 hr post-fertilization [hpf]) treated with elp3 morpholinos or elp3 mutants, and similarly, we also do not find a difference in
treated with control morpholinos (Figure 2G). As we reported the densities of dots per bouton of other active zone markers
before, elp3 morpholino treatment results in reduced motor including LIP and CAC in elp3 mutants and controls (Figures
axon length at 30 hpf (Simpson et al., 2009) (Figure 2H). 3B, 3C, 3L, and 3M), indicating that elp3 mutations do not affect
However, elp3 knockdown does not show reduced levels of the number of active zones per synaptic area. Furthermore,
acetylated tubulin (Figure 2G). Consistently, treatment of fish compared to controls, we also do not observe altered calcium
with tubastatin, a specific inhibitor of HDAC6-mediated tubulin influx measured using GCaMP3 (Tian et al., 2009) in elp3 mutant
deacetylation, results in the expected increase in acetylated boutons (Figures S2E–S2G), in line with normal calcium channel
tubulin levels, but elp3 morpholino treatment does not coun- clustering and function in the mutants.
teract this effect (Figure 2G). Furthermore, tubastatin treatment
Next, we quantified BRPNC82 dot size (maximum diameter) in
fails to rescue elp3 morpholino-induced reduction in motor controls and elp3 null mutants and found an overall increase in
neuron axon length (Figure 2H), indicating that motor axon the size of individual BRPNC82 dots (Figures 3D and 3N), sug-
extension phenotypes upon elp3 morpholino treatment are not gesting increased immunoreactivity of this antigen at individual
caused by decreased tubulin acetylation. Finally, we also as- active zones. To scrutinize the BRP defect in elp3 mutants in
sessed a role for ELP3 in the acetylation of microtubules in more detail, we also quantified features of BRPN labeling in
N2a, HEK, and NCS34 neuroblastoma cells as well as in mouse controls and elp3 mutants (Figures 3F, 3H, 3K, and 3O). First,
cortical neurons or motor neurons using elp3-siRNA or elp3- we quantified the number of BRPN dots per bouton area but
shRNA but did not observe a decrease in the acetylation status did not find a difference, again indicating that ELP3 does not
of microtubules (Figure 2I; Figures S1G–S1P). Thus, using affect the number of active zones per bouton area (Figure 3K).
different species and cell types, our data indicate that ELP3 is Next, we also quantified BRPN dot size, but in contrast to
not a major acetyltransferase for tubulin, and suggest that BRPNC82 labeling, BRP dot size revealed by BRPN is very similar
ELP3 exerts neuronal functions without affecting tubulin at elp3 mutant boutons and controls (Figure 3O), suggesting
acetylation.
that T bar assembly per se (the number of BRP molecules) is
not affected in elp3 mutants. We further assessed if in elp3
mutants supernumerous ‘‘BRP strands’’ join (Figure 3H), by
BRP Spots Are Enlarged at elp3 Mutant NMJs
Given the cytoplasmic and synaptic localization of ELP3 in also performing western blots of elp3 mutant and control brains
neurons, we assessed the abundance and localization of various probed with different BRP antibodies but found very similar
markers at the Drosophila third-instar larval NMJ. We labeled BRP levels (Figure 3I; data not shown). Thus, the data indicate
control and elp3 mutant synapses with the periactive zone normal assembly of BRP strands at active zones and are
marker anti-FasiclinII (FASII), the synaptic vesicle markers anti- consistent with morphological alterations at the C-terminal of
Cysteine string protein (CSP), anti-synaptobrevin (nSYB), and BRP resulting in a more accessible BRPNC82 epitope in elp3
anti-vesicular glutamate transporter (vGLUT), with the endocytic mutants.
marker anti-DYN and with the active zone markers anti-
BRPNC82, anti-Liprin-a (LIP), and Cacophony-GFP (CAC). While Active Zones Are Malformed in elp3 Mutants
most of the markers tested do not display a quantitative dif- To directly assess active zone morphology in elp3 mutant
ference in labeling intensity or localization in elp3 mutants and controls, we performed transmission electron microscopy
compared to controls, BRPNC82 immunoreactivity is markedly (TEM). Quantification of several synaptic features, including
increased (Figures 3A–3D, 3G; Figures S2A–S2D), and this synaptic vesicle number, synaptic vesicle size, mitochondrial
defect is specific to loss of elp3, as adding a wild-type copy of number, and T bar number, does not reveal major differences
elp3 completely rescues the defect (Figures 3D, 3E, and 3G; between mutants and controls (Figures 4A–4C; Figure S3), indi-
data not shown). BRPNC82 recognizes BRP, an integral member cating that elp3 mutations do not result in widespread synaptic
of the electron-dense T bar within the active zone where synaptic defects or affect synaptic organelle transport. The most pro-
vesicles fuse with the membrane (Figure 3H). However, given minent feature in elp3 mutant boutons is the occurrence of
that other active zone markers including LIP and CAC do not sizable T bars with large protrusions that extend into the cyto-
show differences in labeling (Figures 3B, 3C, and 3G), our data plasm (Figures 4D–4G, arrows). Quantification of T bar top
indicate specific synaptic defects in elp3 mutants, including lengths (platforms) in controls indicates that they never exceed
increased BRPNC82 immunoreactivity at NMJ boutons.
300 nm, while in elp3 mutants we observe more than 20% of
BRP forms macromolecular assemblies that are involved in the T bars with a platform that is larger than 300 nm and up to
shaping the T bar (Fouquet et al., 2009). Several ‘‘BRP strands’’ 400 nm in length (Figures 4D–4G, arrowheads; Figure 4H).
join at their N-terminal ends and contact Cacophony calcium Thus, TEM indicates an increase in T bar size in elp3 mutants,
channels near the presynaptic membrane (Kittel et al., 2006), and these data are consistent with the extensive ‘‘tentacles’’ ex-
while BRP C-terminal ends extend into the cytoplasm (Fig- tending into the cytoplasm in elp3 mutants that we observe in
ure 3H). BRPNC82 antibodies label the BRP C-terminal portion, electron tomograms of elp3 mutant boutons (Figures 4I–4O,
while anti-BRPN antibodies label the N-terminal end of the pro- arrows). In line with these data, we measure a concomitant
tein (Fouquet et al., 2009). To further quantify the defect in increase in the number of synaptic vesicles that are in direct
BRPNC82 labeling, we measured dot number in controls and contact with the T bar (Figure 4P). Hence, the elaboration of
elp3 mutants. As shown in Figure 3J, we do not observe an the dense projections of the T bar in elp3 mutants results in an
increase in the number of BRPNC82 dots per boutonic area in increased number of T bar-tethered vesicles.
Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc. 779

Neuron
ELP3 Acetylates BRP
Figure 3. Bruchpilot BRPNC82 Immunoreac-
tivity Is Expanded in elp3 Mutants, while
Other Synaptic Markers Are Not Affected
(A–F) Labeling of Drosophila third-instar NMJs of
y w; elp3rev (control A–D and F) and y w; elp3D3/D4
(A0–D0 and F0) with anti-HRP (neuronal mem-
branes) or anti-DLG (not shown), and FasiclinII,
a perisynaptic marker (FASII) or the active zones
proteins: Liprin-a (LIP), Cacophony-GFP (CAC),
and Bruchpilot (BRPNC82 and BRPN); and of y w;
elp3rev/+; elp3+-GFP/+ (elp3+-GFP, control) and
y w; elp3D3/D4; elp3+-GFP/+ rescued animals with
BRPNC82 (E and E0). Related to Figure S2. Scale
bar, 5 mm (A–F, in F0).
(G) Quantification of synaptic marker intensity
(genotypes in A–E), relative to anti-HRP intensity
(not shown) and of synaptic vesicle markers Cis-
teine string protein (CSP), Vesicular glutamate
transporter (vGLUT), and Synaptobrevin (nSYB)
and Dynamin (DYN) (Figures S2A–S2D). Data are
normalized to control (%). Mean is quantified from
six to ten NMJs from at least four animals. Error
bars, SEM. t test, *p < 0.05.
(H) Schematic of a T bar indicating the BRP-N
and -C terminals. Double line shows presynaptic
membrane.
(I) Western blot of brains dissected from control,
y w; elp3rev and y,w; elp3D3/D4 L3 larvae labeled
with BRPD2 and nSYB (control).
(J–M) Quantification of the dot number per syn-
aptic area (defined by anti-HRP or anti-DLG)
for BRPNC82 (J), BRPN (K), LIP (L), and CAC-GFP
detected with anti-GFP (CAC, M). Mean is quan-
tified from five to ten NMJs from at least four
animals. Error bars, SEM. t test, p > 0.05. ns, not
significant.
(N and O) Quantification of BRPNC82 dot size in
controls, y w; elp3rev and mutants, y w; elp3D3/D4
as well as in controls, y w; elp3rev/+; elp3+-GFP/+
(elp3+-GFP) and rescued animals, y w; elp3D3/D4
;
elp3+-GFP/+ (N); and quantification of BRPN dot
size in controls and y w; elp3D3/D4 mutants (O).
Mean is quantified for five to ten NMJs from
at least four animals. Error bars, SEM. t test,
***p < 0.0001. ns, not significant.
Loss of elp3 Results in Larger Excitatory Junctional
tically significant. The quantal content (in 0.45 mM calcium)
also trends toward an increase but is not significantly different
Currents and Miniature Excitatory Junctional Currents
To determine functional consequences associated with the loss in controls and mutants (EJC/mEJC; controls, 45.3 3.5 quanta;
of elp3 at the NMJ, we measured synaptic transmission using elp3D3/D4, 54.6 6.7 quanta). Increased mEJC amplitude can
two electrode voltage clamp. The average excitatory junctional be caused by larger synaptic vesicles that harbor more neuro-
current (EJC) amplitude in 0.45 mM calcium is significantly transmitter or by a more elaborate postsynaptic glutamate
increased in elp3 mutants (Figures 5A and 5B), and also current receptor field. Given that synaptic vesicle size distribution in
clamp recordings indicate increased excitatory junctional poten- elp3 mutants is not different from controls, we labeled elp3
tial amplitudes in elp3 mutants compared to controls (Figure S4). mutant NMJs with anti-GluRIIA8B4D2 antibodies and with anti-
To determine quantal content, we measured spontaneous GluRIII/IIC antibodies that each recognize different glutamate
vesicle fusion (mEJC) and quantified the quantal amplitude. As receptor subunits (DiAntonio et al., 1999; Marrus et al., 2004).
shown in Figures 5C–5F, mEJC amplitudes are significantly While we did not observe a difference in GluRIII/IIC labeling
increased in elp3 mutants compared to controls, while the between elp3 mutants and controls (Figures 5G, 5H, and 5M),
mEJC frequency trends toward an increase, but this is not statis- the GluRIIA labeling in elp3 mutants is increased compared
780 Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc.

Neuron
ELP3 Acetylates BRP
Figure 4. T Bars Send Extensive Tentacles
into the Cytoplasm in elp3 Mutants
(A–H) TEM of NMJ boutons of y w; elp3rev control
(A and D) and y w; elp3D3/D4 mutants (B and E–G).
Synaptic vesicle diameter frequency (C) from
>2134 vesicles from >20 boutons from 3 animals
per genotype. (D–G) High-magnification image of
T bars in controls (D) and elp3D3/D4 mutants (E–G)
and distribution of individual T bar top lengths
(arrowheads in D–G) and their averages SEM (H).
Mean is quantified for >17 T bars from 3 animals
per genotype. t test, *p < 0.05. Related to Fig-
ure S3. Scale bars, 0.5 mm (A and B) and 0.1 mm
(D–G).
(I and J) Electron tomogram of a T bar from y w;
elp3rev (I) and from y w; elp3D3/D4 mutant (J). A total
of 40
Z slices are projected. Arrows indicate
tentacles extending in the cytoplasm in elp3
mutants.
(K–O) 3D reconstructions of T bars shown in (I) and
(J) in y w; elp3rev (K) and in y w; elp3D3/D4 mutants
(L–O). Arrows indicate cytoplasmic extensions,
magnified in (M)–(O). T bars are shown frontally
(looking from the cytoplasm to the presynaptic
membrane [light blue]) in their longest dimension,
turned by 30ꢀ along the y axis showing the side
and top.
(P) Quantification of the number of T bar-tethered
synaptic vesicles. Mean is s quantified for >17
T bars from 3 animals per genotype. Error bars,
SEM. t test, ***p < 0.0001.
gDGG 0.97 0.03 nA; not shown). While
in other systems application of gDGG
results in a stronger inhibition of the post-
synaptic response (Foster and Regehr,
2004), our data are in line with previous
results at the Drosophila NMJ (Pawlu
et al., 2004) and indicate that gDGG at
to controls, and this defect is rescued by a genomic fragment least in part prevents postsynaptic receptor saturation in high
that harbors wild-type elp3 (Figures 5I–5M). These data are calcium concentrations. Recordings in the presence of the
consistent with the increased mEJC amplitude in elp3 mutants drug will thus allow us to assess neurotransmitter release while
to be caused by exuberant GluRIIA clustering.
partly suppressing glutamate receptor saturation in controls
and mutants.
We then recorded EJC amplitudes in gDGG and different
Presynaptic Loss of elp3 Results in a Larger RRP
The number of T bar-tethered synaptic vesicles in elp3 is calcium concentrations and extracted quantal parameters from
increased, and we tested whether a larger pool of synaptic vesi- parabolic fits from EJC variance versus EJC mean amplitude
cles is immediately ready for fusion in the mutants. First, we used plots (Figures S5C and S5D) (Foster and Regehr, 2004). Our
fluctuation analysis to estimate the number of release-ready data indicate a larger release-ready pool in elp3 mutants com-
vesicles in controls and elp3 mutants. Given that EJC amplitudes pared to controls (controls, 512.7 35.1 quanta; elp3, 592.2
in elp3 mutants and controls saturate at high calcium (Fig- 41.3 quanta; p < 0.05). Also, we find a similar release probability
ure S5A), we performed this analysis in the presence of a rapidly (Pr) in controls and mutants in low calcium concentrations (Ca2+):
dissociating competitive receptor antagonist g-D-glutamylgly- (0.3 mM) control 0.08 0.001 and elp3 0.13 0.01; (0.4 mM)
cine (g-DGG) that has been used at the Drosophila larval NMJ control 0.20 0.01 and elp3 0.17 0.01; and (0.6 mM) control
before (Pawlu et al., 2004). As shown in Figure S5A, EJC ampli- 0.27 0.02 and elp3 0.24 0.03. Similarly, in 3 mM calcium
tudes recorded in 5 mM external calcium are reduced by 38% our analyses indicate a similar Pr (control, 0.98 0.02; elp3,
when incubated in 10 mM gDGG, and also mEJC amplitudes 0.96 0.02), but under these conditions, postsynaptic receptor
(recorded in 0.5 mM Ca2+) are smaller both in controls (without desaturation by gDGG may be incomplete (Figure S5A), con-
gDGG 1.08 0.05 nA; with gDGG 0.68 0.05 nA; Figure S5B) founding our estimations of the release-ready pool and Pr in
as well as in elp3 mutants (without gDGG 1.29 0.07 nA; with the mutants. Nonetheless, in high calcium, Pr is invariably high,
Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc. 781

Neuron
ELP3 Acetylates BRP
Figure 5. EJC and mEJC Amplitude Are
Increased in elp3 Mutants
(A and B) Sample EJCs (A) from control y w; elp3rev
and mutant y w; elp3D3/D4 animals recorded in
0.45 mM CaCl2 and quantification of the mean
EJC amplitude (B). Error bars, SEM. n, number
of animals tested. t test, *p < 0.05. Related to
Figure S4.
(C–F) mEJCs recorded in 0.5 mM CaCl2 with TTX
in control y w; elp3rev (black) and mutant y w;
elp3D3/D4 (green) animals; quantification of the
mean mEJC frequency (C) and amplitude (D),
sample EJC traces (E), and cumulative probability
histograms of mEJC amplitude (F; the number of
mEJCs included is indicated). Error bars, SEM. n,
number of animals tested. t test, **p < 0.01; ns, not
significant.
(G–L) Labeling of y w; elp3rev (control, G and I), y w;
elp3D3/D4 (H and J), y w; elp3rev/+ ; elp3+-GFP/+
(elp3+-GFP, K), and y w; elp3D3/D4; elp3+-GFP/+
(rescued elp3 mutants, L) L3 Drosophila NMJs
with anti-GluRIII (G and H), anti-GluRIIA8B4D2 (I–L),
and anti-HRP or anti-DLG (not shown). Scale bar,
5 mm for (G)– (L) in (L).
(M) Quantification of GluR intensity relative to anti-
HRP or anti-DLG intensity. Data are normalized to
control (%), and mean is quantified from six to
ten NMJs from more than four animals. Error
bars, SEM. Student’s t test, **p < 0.01. ns, not
significant.
and differences in Pr, if any, between elp3 mutants and controls saturation are not the major cause of the larger number of
remain small.
To independently evaluate presynaptic release properties,
detected quanta in elp3 mutants.
To determine the relative contribution of pre- or postsynaptic
we also measured transmission during a short train of high- loss of elp3 to the defect we observe during high-frequency
frequency stimulation (500 ms, 100 Hz) in 5 mM external calcium, stimulation, we conducted rescue experiments. We expressed
ensuring a high Pr (Figure 6A). This protocol results in the release wild-type ELP3 (hELP3) using nsyb-Gal4 only in neurons (presyn-
of neurotransmitters from vesicles that are ready for fusion aptically at the NMJ) or using BG57-Gal4 only in muscles (post-
during the first stimulations and, subsequently, reveals the rate synaptically at the NMJ) in elp3 null mutants. While neuronal
at which new vesicles are captured and prepared for release expression of ELP3 rescues the increased BRPNC82 immunore-
(30th–50th stimulation) (Hallermann et al., 2010a). While the activity (Figures 6C, 6D, and 6G), but not the GluRIIA8B4D2 defect
rate at which vesicles are refilled into the releasable pool during (Figures 6C0, 6D0, and 6G), muscular expression fails to rescue
this stimulation paradigm is similar in elp3 mutants and controls the BRPNC82 defect (Figures 6E, 6F, and 6H) but rescues the
(p > 0.05), back extrapolation from linear fits of the cumulative GluRIIA8B4D2 defect (Figures 6E0, 6F0, and 6H). Furthermore,
quantal content between the 30th and 50th stimulation reveals muscular expression of ELP3 rescues the increased mEJCs
a larger pool of quanta that readily fuses in elp3 mutants com- seen in elp3 mutants, but neuronal ELP3 expression does not
pared to controls (control, 706.1 36.9 quanta; elp3, 907.5
(Figures 6I–6L). Thus, ELP3 is cell autonomously required in
52.2 quanta; p < 0.05). This effect is likely not caused by a post- neurons to regulate BRP morphology and in muscles to restrict
synaptic change in receptor sensitivity as mEJC amplitude GluRIIA abundance.
distribution in controls and in elp3 mutants before versus imme-
Having established conditions where the postsynaptic GluRIIA
diately following stimulation is similar (Figures S5E and S5F). defect is rescued and the presynaptic defect at the level of BRP
Finally, we also recorded EJCs during a 500 ms 100 Hz train in is not, we evaluated the number of readily released quanta
the presence of gDGG, allowing us to perform recordings where during a 500 ms 100 Hz stimulation train by back extrapolation.
the Pr is high, but postsynaptic receptor saturating is partly We find that expression of ELP3 in the nervous system of elp3
inhibited (Figures S5A–S5C). As shown in Figure 6B, recordings null mutants rescues the increased release seen in elp3 mutants
in gDGG yield very similar results for the sizes of the RRP (nsyb-Gal4/+, 688.1
43.3; elp3 nsyb-Gal4, 620.2
32.6;
compared to recordings in the absence of the drug (control, p > 0.05) (Figures 6A and 6M). Conversely, when we assess
700.8 27.5 quanta; elp3, 909.1 40.7 quanta; p < 0.05). While the number of readily released quanta in elp3 mutants that
gDGG may not completely block receptor saturation in high express ELP3 in muscles, the pool size is still large (BG57/+,
calcium, the data suggest that changes in postsynaptic receptor 716.3 44.7; elp3 BG57, 961.3 18.7; p < 0.05) (Figures 6A
782 Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc.

Neuron
ELP3 Acetylates BRP
Figure 6. Presynaptically, the Size of the
RRP Is Increased in elp3 Mutants
(A and B) The cumulative released quantal content
recorded at 100 Hz in 5 mM CaCl2 without (A) or
with gDGG (B) versus stimulus number in control
(y w; elp3rev, black) and y w; elp3D3/D4 mutant
(green) animals. The y-intercept of the slope of the
trend line (dotted line) at steady state (points
30–50) provides a measure of the average RRP
size (indicated, control: black, elp3: green). Error
bars, SEM; n > 7.
(C–H) Images (C–F) and quantification of labeling
intensity (G and H) in y w; elp3rev/+; nsyb-GAL4/+
(control, C and C0), y w; elp3D3/D4 UAS-help3;
nsyb-GAL4/+ (elp3 mutant with neuronal ELP3
expression;
D y ; BG57-
and D0), w; elp3rev/+
GAL4/+ (control, E and E0), and y w; elp3D3/D4
UAS-help3; BG57-GAL4/+ (elp3 mutant with
muscular ELP3 expression F and F0) with anti-
BRPNC82 (C–F) or anti-GluRIIA8B4D2 (C0–F0). Scale
bar, 5 mm (C)–(F) in (F0). Mean boutonic BRP or
GluRIIA intensity relative to anti-HRP, and data
are normalized to control (%). n, number of NMJs
from at least four animals (indicated in the bars).
Error bars, SEM. t test, *p < 0.05, **p < 0.001. ns,
not significant.
(I–L) mEJC traces of recordings in 0.5 mM
CaCl2 with TTX in y w; elp3rev/+; nsyb-GAL4/+ (I),
y w; elp3D3/D4 UAS-help3; nsyb-GAL4/+ (J), y w;
elp3rev/+; BG57-GAL4/+ (K), and y w; elp3D3/D4
UAS-help3; BG57-GAL4/+ (L). Mean mEJC
amplitudes SEM; n > 6 animals.
(M and N) The cumulative released quantal
content recorded at 100 Hz in 5 mM CaCl2 in
control y w; elp3rev/+; nsyb-GAL4/+ (M, black), y w;
elp3D3/D4 UAS-help3; nsyb-GAL4/+ (M, green),
control y w; elp3rev/+; BG57-GAL4/+ (N, black),
and y w; elp3D3/D4 UAS-help3; BG57-GAL4/+ (N,
green). The y-intercept of the slope of the trend line
(dotted lines) at steady state (points 30–50) is
a measure of the average RRP size (indicated).
Error bars, SEM; n > 6 animals.
(O) Mean relative fluorescence change (DF/Fo) of
SpH expressed in elp3 mutants (y w; elp3D3/D4; nsyb-GAL4/UAS-SpH, green) and in controls (y w; nsyb-GAL4/UAS-SpH, black) prior to, during, and following
a 500 ms 100 Hz stimulation train (black bar). Error bars, SEM; n = 9. ANOVA, p < 0.01. Related to Figure S5.
and 6N). Thus, the data indicate that the larger pool of quanta not shown), and in addition, a potential defect in endocytosis
released under these conditions in elp3 mutants stems from would not be expected to significantly contribute to the increase
a presynaptic defect.
To independently test for a presynaptic defect in vesicle
in fluorescence within this short time period.
release in elp3 mutants, we expressed synaptopHluorin (SpH). ELP3 Is Necessary and Sufficient for BRP Acetylation
SpH is a synaptic vesicle-associated pH sensor. At low vesicular Given that elp3 mutants show morphological defects at the level
pH, SpH GFP is quenched but increases in fluorescence upon of their T bars, we tested whether BRP is a substrate for ELP3-
¨
vesicle fusion (Miesenbock et al., 1998). We monitored SpH fluo- dependent acetylation. First, we expressed Drosophila HIS-
rescence during a 500 ms 100 Hz stimulation paradigm, and ELP3 in E. coli, purified, and refolded the protein (Figures S6A
while the initial baseline fluorescence (Fo) in controls and elp3 and S6B). Acetyltransferases are prone to autoacetylation
mutants is similar (data not shown), GFP fluorescence increases (Choudhary et al., 2009). We therefore incubated ELP3 with
to a much higher level in elp3 mutants compared to controls (Fig- 20 mM Acetyl-CoA for various time periods. Western blots
ure 6O). The data indicate that significantly more synaptic vesi- probed with antibodies against acetylated lysine (Ac-K) indicate
cles in elp3 mutants fuse during such a bout of stimulation. We time-dependent ELP3 autoacetylation (Figure 7A). Next, we
do not believe that the increased fluorescence we observe is tested whether our ELP3 protein can acetylate purified histone
the result of defects in endocytosis in elp3 mutants, as our anal- H3, a well-established target, and tubulin. Our data indicate
yses have not revealed endocytic defects in the mutants (data both concentration- and time-dependent acetylation of histone
Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc. 783

Neuron
ELP3 Acetylates BRP
Figure 7. ELP3 Is Sufficient to Acetylate
BRP
(A) Autoacetylation of ELP3 using an ELP3-
enriched fraction with and without adding Acetyl
CoenzymeA (ac-CoA); western blots probed with
Ac-K. Related to Figure S6.
(B) Time-dependent and concentration-depen-
dent in vitro acetylation of histone H3 with an
ELP3-enriched fraction; western blots were pro-
bed with Ac-K.
(C) Tubulin acetylation assay with an ELP3-
enriched fraction on brain extracts of third-instar
y w; elp3D3/D4 mutant larvae; western blots were
probed with anti-acetylated tubulin (ac-TUB) or
total tubulin (TUB).
(D) In vitro acetylation of IPed BRP (with BRPNC82
)
from w1118 brains with an ELP3-enriched fraction.
Western blots were probed with Ac-K and re-
probed with BRPD2
.
(E and F) Quantification of ELP3-dependent acet-
ylation of tubulin and of BRP normalized to total
tubulin or total BRP levels relative to initial acety-
lation levels (time point ‘‘0’’). Number of indepen-
dent repeats is indicated in the bars.
with Ac-K. BRP immunoprecipitations
(IPs) from control animals show an acety-
lated lysine band that migrates at the
same height of BRP, detected with anti-
bodies against the middle domain of
H3 (Figure 7B), but we did not observe ELP3-dependent acety- BRP, BRPD2 (Figures 8H and 7D). This band is largely absent
lation of tubulin in wild-type fly lysate, or in lysate prepared in western blots of IPs from pharate adult brains that express
from elp3 null mutant animals (Figures 7C and 7E; data not RNAi to brp, indicating that the band is specific to BRP and sug-
shown). Thus, although our ELP3 fraction is active, it does not gesting that at least some BRP is acetylated under basal condi-
support acetylation of tubulin in vitro. Finally, we tested acetyla- tions. Interestingly, in BRP IPs from animals that express RNAi to
tion of BRP in vitro. We immunoprecipitated BRP from fly heads elp3, we are able to clearly detect BRP, but the acetylated lysine
(see also Figure 8H), incubated these BRP-enriched fractions band at the height of BRP is largely absent (Figure 8H). These
with Acetyl-CoA and ELP3, and probed western blots with data corroborate the labeling of acetylated lysines at boutons
Ac-K (Figure 7D). As shown in Figures 7D and 7F, we find and suggest that ELP3 is necessary to maintain the acetylation
obvious time-dependent acetylation of BRP. These data indicate status of the active zone-associated protein BRP.
that ELP3 is sufficient for the acetylation of the active zone-asso-
ciated protein BRP.
To determine if ELP3 acetylates BRP in vivo, we labeled NMJs
DISCUSSION
with Ac-K. Ac-K labels histones in nuclei (data not shown), In this work we provide evidence that ELP3 acetylates the active
microtubules in axons that we marked using the monoclonal zone-associated cytoskeletal-like protein BRP that is increas-
antibody Futsch22C10 (Figures 8A and 8B), as well as several ingly implicated in neuronal diseases (Choi et al., 2010; Zweier
features in synaptic boutons (Figures 8A and 8C). Furthermore, et al., 2009). ELP3-mediated BRP acetylation regulates dense
overexpression of HDAC6 shows a marked reduction in Ac-K body structure, akin to the modification of chromatin structure
that decorates microtubules labeled by Futsch22C10, indicating in the nucleus, and this function is independent of an effect of
that the antibodies are specific (Figure 8E). Interestingly, anti- ELP3 on tubulin acetylation. We suggest that decreased BRP
acetylated lysine labeling that overlaps with BRPNC82 labeling acetylation in elp3 mutants results in expanded cytoplasmic
is much reduced in elp3 mutants compared to controls (Figures specializations that capture synaptic vesicles, and our work
8C, 8D, 8F, and 8G). The reduction in labeling is specific to active points to a model where individual release site morphology
zones because Ac-K labeling overlapping with Futsch22C10 is not and function may be controlled by BRP acetylation.
significantly different in elp3 mutants compared to controls
(Figures 8A, 8B, and 8E). The data suggest that less acetylated Tubulin Acetylation in the Absence of ELP3
lysines are present at active zones in elp3 mutants.
Recent work suggests that besides a role in acetylating histones,
Next, we immunoprecipitated BRP from control and elp3 ELP3 also acetylates tubulin (Creppe et al., 2009; Solinger et al.,
RNAi-expressing pharate adult brains and probed western blots 2010); however, several of our observations using different
784 Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc.

Neuron
ELP3 Acetylates BRP
Figure 8. ELP3 Is Necessary to Acetylate
BRP
(A–G) Double labeling of NMJs in control elp3rev
(A, C, and G) and y,w; elp3D3/D4 mutant L3 larvae
(B, D, and G), stained with Ac-K and Futsch22C10
,
a microtubule-associated protein (A and B), or
BRPNC82 (C, D, and G) and mean Ac-K intensity
within the area marked with Futsch22C10 (E) or
BRPNC82 (F) relative to controls from eight to ten
NMJs from more than four larvae. Error bars, SEM.
t test, *p < 0.05. Scale bars, 5 mm (A–D in D) and
0.5 mm (G). ns, not significant.
(H) Western blot of IPed BRP (using BRPNC82
)
frombrp-RNAi(w DCR2;UAS-brpRNAi/nsyb-Gal4),
control (w DCR2;nsyb-Gal4/+), and elp3-RNAi
(w; UAS-elp3RNAiC8/+; da-Gal4/+), and its control
(w; da-Gal4/+) pupal brains were probed with Ac-K
or with BRPD2
.
species and cell types indicate that microtubules can be acety- coil motifs that were recently shown to be ideal acetylation
lated by a mechanism that does not involve ELP3. Similar to our substrates (Choudhary et al., 2009).
findings, in human neuroblastoma cells or in mouse embryonic
Individual BRP strands organize into parasol-like structures,
fibroblasts, ELP1 knockdown results in a profound reduction of with their N termini facing the plasma membrane, contacting
ELP3 expression, but also this condition did not affect the levels calcium channels, and their C termini extending into the cyto-
of acetylated tubulin (Cheishvili et al., 2011). Although in vitro our plasm capturing synaptic vesicles (Fouquet et al., 2009; Haller-
ELP3 fraction was not able to increase tubulin acetylation, when mann et al., 2010b; Jiao et al., 2010). While mutations that affect
the protein is overexpressed in N2a cells or in Drosophila motor BRP transport to synapses or assembly of T bars at active zones
neurons, we find a mild increase in acetylation of tubulin (data not exist, our data indicate that these processes are not affected in
shown), but our work shows that this activity is limited outside of elp3 mutants. Unlike SRPK79D mutants (Johnson et al., 2009;
overexpression conditions. It is interesting that an alternative Nieratschker et al., 2009), BRPNC82 does not accumulate in
GNAT domain protein, MEC-17, was shown to acetylate tubulin elp3 mutant motor neurons (data not shown), suggesting normal
in different systems, including nematodes, zebrafish, and ciliates axonal transport. In addition, in contrast to rab3 mutants (Graf
(Akella et al., 2010); in addition, an acetyltransferase complex, et al., 2009), the number of T bars per synaptic area is not
ARD1-NAT1, that can acetylate tubulin in vitro has been found different in controls and elp3 mutants.
associated with tubulin in developing dendrites of cultured hip-
Our analyses also identified a postsynaptic role for elp3 in
pocampal neurons and was shown to regulate dendritic out- regulating glutamate receptor subunit IIA abundance in muscles
growth in vitro (Ohkawa et al., 2008). Thus, alternative tubulin at NMJs and, thus, mEJC amplitude; however, unlike ELP3’s
acetyltransferases that regulate neuronal morphology have neuronal function, we show that this role of ELP3 is not critical
been identified.
for viability, as muscular expression of the protein does not
rescue elp3-associated lethality. Nonetheless, by regulating
postsynaptic receptor field size, ELP3 may also modulate
neuronal communication. We present evidence that this defect
ELP3 Is a BRP Acetyltransferase and Controls Active
Zone Morphology
In a search of alternative cytoplasmic ELP3 targets, we identi- is regulated in muscle cells independently of the presynaptic
fied BRP, a large cytoskeletal-like protein that decorates the role of ELP3.
active zone where synaptic vesicles fuse with the membrane.
Using different independent methodologies, we provide
We provide several lines of evidence that ELP3 acts to acetylate evidence that ELP3 regulates presynaptic neurotransmitter
BRP at the Drosophila NMJ. First, ELP3 is present at NMJ bou- release efficiency. We show that during a short high-frequency
tons, localizing the enzyme in close proximity to BRP. Second, stimulation train, elp3 mutants show a stronger increase in
acetylated lysine levels that overlap with BRPNC82 labeling at SpH fluorescence than controls. Furthermore, mutants release
the NMJ are reduced in elp3 mutants. Similarly, BRP-associated more quanta than controls during a short 100 Hz stimulation
acetylated lysine levels detected by western blotting are reduced train, and this is also true in mutant animals that express
in elp3 mutants. Third, immunoprecipitated BRP is efficiently hELP3 in muscles and, thus, do not display increased GluRIIA
acetylated by purified ELP3 in vitro. Without excluding other levels. While these data are consistent with a larger pool of
substrates, our data indicate that ELP3 is necessary and suffi- readily releasable vesicles in the mutants, a larger Pr in elp3
cient to acetylate BRP. BRP is indeed an excellent candidate mutants may also contribute to increased release. Given that
to undergo this modification as it contains numerous coiled- gDGG only partially prevents postsynaptic receptor saturation
Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc. 785

Neuron
ELP3 Acetylates BRP
Biochemistry
at the NMJ, our estimates of Pr in high calcium based on fluctu-
ation analysis are less accurate. However, in 5 mM calcium the
Pr is invariably high, limiting the difference in Pr between controls
and mutants. In addition, an increased Pr but not a larger RRP in
elp3 mutants would alter the time course by which neurotrans-
mitters are released during the 500 ms 100 Hz stimulation para-
digm, but the total number of released quanta would not be
different between elp3 mutants and controls, particularly in
mutants where the postsynaptic defects are rescued, and differ-
ences in receptor abundance are eliminated. Thus, while not
excluding an effect of ELP3 on the Pr in high calcium, our data
are most consistent with an increased RRP in elp3 mutants.
Our work suggests a model where acetylation of BRP reorga-
nizes the cytoplasmic tentacles such that deacetylation leads to
more extensive spreading of the strands, possibly by altering
electrostatic interactions, similar to the regulation of chromatin
structure by histone acetylation (Shogren-Knaak et al., 2006).
At active zones, we speculate that this function regulates vesicle
capturing by the C-terminal end of BRP (Hallermann et al.,
2010b), and transport of vesicles at dense bodies. We present
evidence that the defect in elp3 mutants results in a larger pool
of synaptic vesicles that is ready for immediate release, poten-
tially in part by improved vesicle tethering at T bars. Although
the mechanisms that regulate local ELP3 activity levels at the
synapse (but also those that regulate ELP3 activity in the
nucleus) remain elusive, it will be interesting to identify signaling
pathways that activate ELP3 enzymatic function. The local regu-
lation of ELP3 may enable single active zones to control neuro-
transmitter release and may have important implications for
synaptic transmission regulation in a number of neurological
diseases, including ALS and familial dysautonomia (Simpson
et al., 2009; Slaugenhaupt and Gusella, 2002).
Drosophila, zebrafish, and cellular extracts for westerns were prepared using
standard procedures and probed with the antibodies listed below.
IP of BRP from pupal or adult brain extracts was performed using BRPNC82
diluted 1:5 (Developmental Studies Hybridoma Bank, Iowa City, IA, USA) and
G protein-coupled magnetic beads (BioLabs) using Drosophila head lysate
(Supplemental Experimental Procedures), and IPed protein was either used
for acetylation assays or for westerns.
Drosophila ELP3 was expressed in E. coli Rosetta (DE3) pLysS cells
(Promega). In vitro acetylation was performed by incubating purified ELP3
with substrate in acetylation buffer (50 mM Tris-HCl [pH 8.0], 0.1 mM EDTA,
1 mM DTT, 10 mM Na-butyrate, 10% glycerol) and 20 mM acetyl CoA
(Sigma-Aldrich) at 25ꢀC (Chen and Greene, 2005). Substrates used were puri-
fied Histone H3 (Westburg) and beads with BRP or Drosophila head protein
lysate (for acetylation of tubulin).
Western blotting antibodies were: 1:500 anti-Acetylated lysine (Ac-K)
(rabbit, AB80178; abCAM); 1:1,000 anti-BRPN and 1:1,000 anti-BRPD2
(Fouquet et al., 2009); 1:1,000 anti-neuronal synaptobrevin (nSYBR29) and
1:500 anti-Histone H3 (9715L; Cell Signaling); 1:10,000 anti-Acetylated
a-Tubulin (6-11-B-1; Sigma-Aldrich); 1:1,000 anti-a-Tubulin (B5-12; Sigma-
Aldrich); 1:5,000 anti-b-actin (A5441; Sigma-Aldrich); 1:1,000 anti-BRPNC82
(Developmental Studies Hybridoma Bank); 1:5,000 anti-GAPDH (4300;
Ambion); and 1:1,000 HRP-coupled secondary antibodies (Jackson
ImmunoResearch). Blots were developed with Western Lightning ECL
(PerkinElmer).
Fluorescence Microscopy
Drosophila third-instar larvae were prepared as described (Uytterhoeven et al.,
2011). Primary antibodies were: 1:100 BRPNC82, 1:100 Futsch22C10, 1:25 anti-
GLURIIA8B4D2, 1:50 Anti-DLG4F3, and 1:2 anti-FASII1D4 (all from Develop-
mental Studies Hybridoma Bank); 1:50 anti-CSP49 (Zinsmaier et al., 1994);
1:1,000 anti-HRP (Jackson ImmunoResearch); 1:300 anti-BRPN and 1:1,000
anti-Liprin (Owald et al., 2010); 1:200 anti-DYN (Hudy1; Millipore); 1:500 anti-
nSYBR29 and 1:200 anti-Syndapin (both gifts from H. Bellen, BCM); 1:500
anti-GLURIII (Marrus et al., 2004); 1:10,000 anti-vGLUT (Daniels et al., 2004)
(both gifts from A. Di Antonio, Washington University); 1:200 anti-Acetylated
Lysines (Ac-K) (ab80178; abCAM); 1:1,000 anti-Acetylated a-Tubulin (6-11-
B-1; Sigma-Aldrich); 1:500 anti-GFP (rabbit IgG fraction; Invitrogen) to detect
GFP-ELP3, ELP3-GFP, and Cac-GFP signals; and 1:1,000 secondary Alexa
488, 555, or 645-conjugated antibodies (Invitrogen). Toto3 (Invitrogen) was
used to label DNA and was used at 1:500 in PBS prior to mounting samples.
NMJs were imaged through 633 1.4 NA oil lens on a Zeiss 510 Meta confocal
microscope, and mean fluorescence intensities of labeling per bouton and
background were measured using ImageJ as described (Khuong et al.,
2010; Uytterhoeven et al., 2011). The number of BRP spots per area was
counted following automated thresholding in ImageJ and calculated from
measurements of R 6 type 1b boutons per NMJ on M6/7 in segments A2/3.
EXPERIMENTAL PROCEDURES
Animals and Cells
All Drosophila lines were kept on cornmeal and molasses medium. For
experiments L3 larvae were grown on black currant juice agar plates with fresh
yeast paste. GAL4 > UAS-expressing larvae and controls were raised at 28ꢀC
(rescue and HDAC6) or at 25ꢀC (CAC-GFP, GCaMP3, SpH, and RNAi). elp3D3
elp3D4, and elp3D5 mutants and elp3rev controls were created by P element
dysgenesis using y1w67c23, P{SUP or-P}elp3KG02386
,
.
Adult zebrafish (AB) and embryos were maintained and staged as
described (Westerfield, 2003). The elp3 ATG-morpholino is from Gene Tools,
LLC (Corvallis, OR, USA): 50-TGGCTTTCCCATCTTAGACACAATC-30 (ATG-
MO); reverse control 50-CTAACACAGATTCTACCCTTTCGGT-30 (Ctr-MO).
A total of 2 mM tubastatin A or DMSO treatment was started at 6 hpf. Axonal
defects were evaluated at 30 hpf (Lemmens et al., 2007).
Electron Microscopy
L3 larvae were dissected in HL-3 and prepared for TEM, and bouton profiles in
50 nm sections from M6/7 in segment A2 were visualized on a JEOL TEM100
(Uytterhoeven et al., 2011). At least 17 images from 3 animals were analyzed.
Serial-tilt EM was performed on 300 nm sections, and micrographs were
recorded from ꢁ60ꢀ to 60ꢀ at 2ꢀ intervals. 3D reconstructions were generated
in IMOD (Uytterhoeven et al., 2011).
N2a, HEK293T, and NSC34 cells were grown under standard conditions,
and cortical neurons, motor neurons, and glial feeder layer cells were prepared
as described (Vandenberghe et al., 1998).
Electrophysiology
Constructs
Recordings from L3 M6 in segment A2/3 were performed on an Axoclamp
900A amplifier, filtered at 1 kHz (400 Hz for minis), and stored in pClamp
10.3 in modified HL-3: 110 mM NaCl, 5 mM KCl, 10 mM NaHCO3, 5 mM
HEPES, 30 mM sucrose, 5 mM trehalose, 10 mM MgCl2, CaCl2 (as indicated)
[pH 7.2], using a holding potential of ꢁ70 mV (Uytterhoeven et al., 2011). Where
indicated, 0.5 mM TTX (for mini recordings, not in Figures S6E and S6F) or
10 mM g-DGG (Tocris Bioscience) was added (Pawlu et al., 2004) for 10 min
prior to recordings.
UAS-help3 was created by cloning the human elp3 cDNA (OriGene) into the
EcoRI site of pUAST-attB. Genomic GFP-elp3+ and elp3+-GFP constructs
were generated using recombineering in attB-P(acman)-ApR (Venken et al.,
2006, 2008) using BAC RP98-28K16. These constructs were inserted in
VK31 (62E1) and VK01 (59D3) sites using phiC31-mediated integration
(GenetiVision, Houston).
For protein expression, Drosophila elp3 cDNA (RE35395, BDGP) was cloned
into a pDEST14 expression vector using Gateway technology (Invitrogen) and
includes an N-terminal 6xHIS tag.
Fluctuation analysis was performed as described (Weyhersmu¨ ller et al.,
2011).
786 Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc.

Neuron
ELP3 Acetylates BRP
The RRP size from cumulative EJC plots was determined as described
(Weyhersmu¨ ller et al., 2011; Hallermann et al., 2010a) except that 50 stimuli
were delivered at 100 Hz. EJC amplitudes were measured from peak to the
baseline (lowest level) immediately before the onset of the EJC. Trend lines
were calculated between the 30th and 50th stimulation point and back extrap-
olated to time zero.
DiAntonio, A., Petersen, S.A., Heckmann, M., and Goodman, C.S. (1999).
Glutamate receptor expression regulates quantal size and quantal content at
the Drosophila neuromuscular junction. J. Neurosci. 19, 3023–3032.
Foster, K.A., and Regehr, W.G. (2004). Variance-mean analysis in the presence
of a rapid antagonist indicates vesicle depletion underlies depression at the
climbing fiber synapse. Neuron 43, 119–131.
Fouquet, W., Owald, D., Wichmann, C., Mertel, S., Depner, H., Dyba, M.,
Hallermann, S., Kittel, R.J., Eimer, S., and Sigrist, S.J. (2009). Maturation of
active zone assembly by Drosophila Bruchpilot. J. Cell Biol. 186, 129–145.
Time-Lapse Imaging
GCaMP3 and SpH were expressed with nSyb-Gal4 in elp3rev and elp3
mutants, and imaging was performed as described (Hendel et al., 2008;
Uytterhoeven et al., 2011).
Graf, E.R., Daniels, R.W., Burgess, R.W., Schwarz, T.L., and DiAntonio, A.
(2009). Rab3 dynamically controls protein composition at active zones.
Neuron 64, 663–677.
SUPPLEMENTAL INFORMATION
Hallermann, S., Heckmann, M., and Kittel, R.J. (2010a). Mechanisms of short-
term plasticity at neuromuscular active zones of Drosophila. HFSP J. 4, 72–84.
Supplemental Information includes six figures and Supplemental Experimental
Procedures and can be found with this article online at doi:10.1016/j.neuron.
2011.10.010.
Hallermann, S., Kittel, R.J., Wichmann, C., Weyhersmu¨ ller, A., Fouquet, W.,
¨
Mertel, S., Owald, D., Eimer, S., Depner, H., Schwarzel, M., et al. (2010b).
Naked dense bodies provoke depression. J. Neurosci. 30, 14340–14345.
ACKNOWLEDGMENTS
Han, Q., Lu, J., Duan, J., Su, D., Hou, X., Li, F., Wang, X., and Huang, B. (2008).
Gcn5- and Elp3-induced histone H3 acetylation regulates hsp70 gene tran-
scription in yeast. Biochem. J. 409, 779–788.
We thank the Vienna Drosophila RNAi Center, the Bloomington and Harvard
Drosophila stock centers, and the Developmental Studies Hybridoma bank
as well as Hugo Bellen and Aaron DiAntonio for reagents. We especially thank
Ron Habets as well as Sebastian Munck, Pieter Baatsen, and Jan Slabbaert,
and other members of the P.V., W.R., and W.V. labs for help and comments.
Support was provided by a Marie Curie Excellence grant (MEXT-CT-2006-
042267), an ERC Starting Grant (260678), FWO Grants G094011, G095511,
G074709, and G025909, the Research Fund KU Leuven: BOF-OT and GOA
11/014, Interuniversity attraction Poles (IUAP) program P6/43 of the Belgian
Federal Science Policy Office, the Motor Neuron Disease Association UK
(6046), The European Community’s Health Seventh Framework Programme
(FP7/2007-2013; 259867), a Methusalem grant of the Flemish Government,
the Francqui Foundation, the Hercules Foundation (project AKUL/09/037),
and VIB. W.V. is supported by an FWO postdoctoral grant, M.F. by an IWT
predoctoral grant, L.E.J. by a predoctoral VIB fellowship, and W.R. by
a E. von Behring Chair for Neuromuscular and Neurodegenerative Disorders.
Haucke, V., Neher, E., and Sigrist, S.J. (2011). Protein scaffolds in the coupling
of synaptic exocytosis and endocytosis. Nat. Rev. Neurosci. 12, 127–138.
Hendel, T., Mank, M., Schnell, B., Griesbeck, O., Borst, A., and Reiff, D.F.
(2008). Fluorescence changes of genetic calcium indicators and OGB-1
correlated with neural activity and calcium in vivo and in vitro. J. Neurosci.
28, 7399–7411.
Hida, Y., and Ohtsuka, T. (2010). CAST and ELKS proteins: structural and func-
tional determinants of the presynaptic active zone. J. Biochem. 148, 131–137.
Hubbert, C., Guardiola, A., Shao, R., Kawaguchi, Y., Ito, A., Nixon, A., Yoshida,
M., Wang, X.F., and Yao, T.P. (2002). HDAC6 is a microtubule-associated
deacetylase. Nature 417, 455–458.
´
Jiao, W., Masich, S., Franzen, O., and Shupliakov, O. (2010). Two pools of vesi-
cles associated with the presynaptic cytosolic projection in Drosophila neuro-
muscular junctions. J. Struct. Biol. 172, 389–394.
Accepted: October 5, 2011
Published: December 7, 2011
Johnson, E.L., 3rd, Fetter, R.D., and Davis, G.W. (2009). Negative regulation of
active zone assembly by a newly identified SR protein kinase. PLoS Biol. 7,
e1000193.
REFERENCES
Khuong, T.M., Habets, R.L., Slabbaert, J.R., and Verstreken, P. (2010). WASP
is activated by phosphatidylinositol-4,5-bisphosphate to restrict synapse
growth in a pathway parallel to bone morphogenetic protein signaling. Proc.
Natl. Acad. Sci. USA 107, 17379–17384.
Akella, J.S., Wloga, D., Kim, J., Starostina, N.G., Lyons-Abbott, S.,
Morrissette, N.S., Dougan, S.T., Kipreos, E.T., and Gaertig, J. (2010).
MEC-17 is an alpha-tubulin acetyltransferase. Nature 467, 218–222.
Kittel, R.J., Wichmann, C., Rasse, T.M., Fouquet, W., Schmidt, M., Schmid, A.,
Wagh, D.A., Pawlu, C., Kellner, R.R., Willig, K.I., et al. (2006). Bruchpilot
promotes active zone assembly, Ca2+ channel clustering, and vesicle release.
Science 312, 1051–1054.
Cheishvili, D., Maayan, C., Cohen-Kupiec, R., Lefler, S., Weil, M., Ast, G., and
Razin, A. (2011). IKAP/Elp1 involvement in cytoskeleton regulation and impli-
cation for familial dysautonomia. Hum. Mol. Genet. 20, 1585–1594.
Chen, L.F., and Greene, W.C. (2005). Assessing acetylation of NF-kappaB.
Lemmens, R., Van Hoecke, A., Hersmus, N., Geelen, V., D’Hollander, I., Thijs,
V., Van Den Bosch, L., Carmeliet, P., and Robberecht, W. (2007).
Overexpression of mutant superoxide dismutase 1 causes a motor axonop-
athy in the zebrafish. Hum. Mol. Genet. 16, 2359–2365.
Methods 36, 368–375.
Choi, J.K., Jeon, Y.C., Lee, D.W., Oh, J.M., Lee, H.P., Jeong, B.H., Carp, R.I.,
Koh, Y.H., and Kim, Y.S. (2010). A Drosophila model of GSS syndrome
suggests defects in active zones are responsible for pathogenesis of GSS
syndrome. Hum. Mol. Genet. 19, 4474–4489.
Marrus, S.B., Portman, S.L., Allen, M.J., Moffat, K.G., and DiAntonio, A. (2004).
Differential localization of glutamate receptor subunits at the Drosophila neuro-
muscular junction. J. Neurosci. 24, 1406–1415.
Choudhary, C., Kumar, C., Gnad, F., Nielsen, M.L., Rehman, M., Walther, T.C.,
Olsen, J.V., and Mann, M. (2009). Lysine acetylation targets protein complexes
and co-regulates major cellular functions. Science 325, 834–840.
¨
Miesenbock, G., De Angelis, D.A., and Rothman, J.E. (1998). Visualizing secre-
tion and synaptic transmission with pH-sensitive green fluorescent proteins.
Creppe, C., Malinouskaya, L., Volvert, M.L., Gillard, M., Close, P., Malaise, O.,
Laguesse, S., Cornez, I., Rahmouni, S., Ormenese, S., et al. (2009). Elongator
controls the migration and differentiation of cortical neurons through acetyla-
tion of alpha-tubulin. Cell 136, 551–564.
Nature 394, 192–195.
Nicoll, R.A., and Schmitz, D. (2005). Synaptic plasticity at hippocampal mossy
fibre synapses. Nat. Rev. Neurosci. 6, 863–876.
Daniels, R.W., Collins, C.A., Gelfand, M.V., Dant, J., Brooks, E.S., Krantz, D.E.,
and DiAntonio, A. (2004). Increased expression of the Drosophila vesicular
glutamate transporter leads to excess glutamate release and a compensatory
decrease in quantal content. J. Neurosci. 24, 10466–10474.
Nieratschker, V., Schubert, A., Jauch, M., Bock, N., Bucher, D., Dippacher, S.,
Krohne, G., Asan, E., Buchner, S., and Buchner, E. (2009). Bruchpilot in ribbon-
like axonal agglomerates, behavioral defects, and early death in SRPK79D
kinase mutants of Drosophila. PLoS Genet. 5, e1000700.
Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc. 787

Neuron
ELP3 Acetylates BRP
Ohkawa, N., Sugisaki, S., Tokunaga, E., Fujitani, K., Hayasaka, T., Setou, M.,
and Inokuchi, K. (2008). N-acetyltransferase ARD1-NAT1 regulates neuronal
dendritic development. Genes Cells 13, 1171–1183.
Tian, L., Hires, S.A., Mao, T., Huber, D., Chiappe, M.E., Chalasani, S.H.,
Petreanu, L., Akerboom, J., McKinney, S.A., Schreiter, E.R., et al. (2009).
Imaging neural activity in worms, flies and mice with improved GCaMP calcium
indicators. Nat. Methods 6, 875–881.
Otero, G., Fellows, J., Li, Y., de Bizemont, T., Dirac, A.M., Gustafsson, C.M.,
Erdjument-Bromage, H., Tempst, P., and Svejstrup, J.Q. (1999). Elongator,
a multisubunit component of a novel RNA polymerase II holoenzyme for tran-
scriptional elongation. Mol. Cell 3, 109–118.
Uytterhoeven, V., Kuenen, S., Kasprowicz, J., Miskiewicz, K., and Verstreken,
P. (2011). Loss of skywalker reveals synaptic endosomes as sorting stations
for synaptic vesicle proteins. Cell 145, 117–132.
Vandenberghe, W., Van Den Bosch, L., and Robberecht, W. (1998). Glial cells
potentiate kainate-induced neuronal death in a motoneuron-enriched spinal
coculture system. Brain Res. 807, 1–10.
Owald, D., Fouquet, W., Schmidt, M., Wichmann, C., Mertel, S., Depner, H.,
¨
Christiansen, F., Zube, C., Quentin, C., Korner, J., et al. (2010). A Syd-1 homo-
logue regulates pre- and postsynaptic maturation in Drosophila. J. Cell Biol.
188, 565–579.
Venken, K.J., He, Y., Hoskins, R.A., and Bellen, H.J. (2006). P[acman]: a BAC
transgenic platform for targeted insertion of large DNA fragments in
D. melanogaster. Science 314, 1747–1751.
Pawlu, C., DiAntonio, A., and Heckmann, M. (2004). Postfusional control of
quantal current shape. Neuron 42, 607–618.
Venken, K.J., Kasprowicz, J., Kuenen, S., Yan, J., Hassan, B.A., and
Verstreken, P. (2008). Recombineering-mediated tagging of Drosophila
genomic constructs for in vivo localization and acute protein inactivation.
Nucleic Acids Res. 36, e114.
Peled, E.S., and Isacoff, E.Y. (2011). Optical quantal analysis of synaptic trans-
mission in wild-type and rab3-mutant Drosophila motor axons. Nat. Neurosci.
14, 519–526.
Pelkey, K.A., and McBain, C.J. (2007). Differential regulation at functionally
divergent release sites along a common axon. Curr. Opin. Neurobiol. 17,
366–373.
´
Versees, W., De Groeve, S., and Van Lijsebettens, M. (2010). Elongator,
a conserved multitasking complex? Mol. Microbiol. 76, 1065–1069.
Walia, H., Chen, H.Y., Sun, J.M., Holth, L.T., and Davie, J.R. (1998). Histone
acetylation is required to maintain the unfolded nucleosome structure associ-
ated with transcribing DNA. J. Biol. Chem. 273, 14516–14522.
Pelkey, K.A., Topolnik, L., Lacaille, J.C., and McBain, C.J. (2006).
Compartmentalized Ca(2+) channel regulation at divergent mossy-fiber
release sites underlies target cell-dependent plasticity. Neuron 52, 497–510.
Walker, J., Kwon, S.Y., Badenhorst, P., East, P., McNeill, H., and Svejstrup,
J.Q. (2011). Role of elongator subunit Elp3 in Drosophila melanogaster larval
development and immunity. Genetics 187, 1067–1075.
Petersen, S.A., Fetter, R.D., Noordermeer, J.N., Goodman, C.S., and
DiAntonio, A. (1997). Genetic analysis of glutamate receptors in Drosophila
reveals
a retrograde signal regulating presynaptic transmitter release.
Neuron 19, 1237–1248.
Westerfield, M. (2003). The Zebrafish Book (Eugene, OR: The University of
Oregon Press).
Schmid, A., Hallermann, S., Kittel, R.J., Khorramshahi, O., Fro¨ lich, A.M.,
Quentin, C., Rasse, T.M., Mertel, S., Heckmann, M., and Sigrist, S.J. (2008).
Activity-dependent site-specific changes of glutamate receptor composition
in vivo. Nat. Neurosci. 11, 659–666.
Weyhersmu¨ ller, A., Hallermann, S., Wagner, N., and Eilers, J. (2011). Rapid
active zone remodeling during synaptic plasticity. J. Neurosci. 31, 6041–6052.
Winkler, G.S., Kristjuhan, A., Erdjument-Bromage, H., Tempst, P., and
Svejstrup, J.Q. (2002). Elongator is a histone H3 and H4 acetyltransferase
important for normal histone acetylation levels in vivo. Proc. Natl. Acad. Sci.
USA 99, 3517–3522.
Shogren-Knaak, M., Ishii, H., Sun, J.M., Pazin, M.J., Davie, J.R., and Peterson,
C.L. (2006). Histone H4-K16 acetylation controls chromatin structure and
protein interactions. Science 311, 844–847.
Simpson, C.L., Lemmens, R., Miskiewicz, K., Broom, W.J., Hansen, V.K., van
Vught, P.W., Landers, J.E., Sapp, P., Van Den Bosch, L., Knight, J., et al.
(2009). Variants of the elongator protein 3 (ELP3) gene are associated with
motor neuron degeneration. Hum. Mol. Genet. 18, 472–481.
Zhai, R.G., and Bellen, H.J. (2004). The architecture of the active zone in the
presynaptic nerve terminal. Physiology (Bethesda) 19, 262–270.
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.
Slaugenhaupt, S.A., and Gusella, J.F. (2002). Familial dysautonomia. Curr.
Opin. Genet. Dev. 12, 307–311.
Zweier, C., de Jong, E.K., Zweier, M., Orrico, A., Ousager, L.B., Collins, A.L.,
Bijlsma, E.K., Oortveld, M.A., Ekici, A.B., Reis, A., et al. (2009). CNTNAP2
and NRXN1 are mutated in autosomal-recessive Pitt-Hopkins-like mental
¨
Solinger, J.A., Paolinelli, R., Kloss, H., Scorza, F.B., Marchesi, S., Sauder, U.,
Mitsushima, D., Capuani, F., Stu¨ rzenbaum, S.R., and Cassata, G. (2010). The
Caenorhabditis elegans Elongator complex regulates neuronal alpha-tubulin
acetylation. PLoS Genet. 6, e1000820.
retardation and determine the level of
a common synaptic protein in
Drosophila. Am. J. Hum. Genet. 85, 655–666.
788 Neuron 72, 776–788, December 8, 2011 ª2011 Elsevier Inc.