BMP signaling modulates the probability of neurotransmitter release and readily releasable pools in Drosophila neuromuscular junction synapses

Biochemical and Biophysical Research Communications 479 (2016) 440e446

Contents lists available at ScienceDirect

Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc

BMP signaling modulates the probability of neurotransmitter release and readily releasable pools in Drosophila neuromuscular junction synapses Seung-Hyun Lee 1, Yoon-Jung Kim 1, Se-Young Choi* Department of Physiology, Dental Research Institute, Seoul National University School of Dentistry, Seoul, 110-749, South Korea

a r t i c l e i n f o

a b s t r a c t

Article history: Received 12 August 2016 Accepted 15 September 2016 Available online 23 September 2016

The structure and function of synapses is modulated by the interaction of presynaptic and postsynaptic neurons via cell adhesion molecules or secreted signal molecules. Bone morphogenic protein (BMP) is a secreted molecule mediating retrograde signaling that is involved in the formation and maintenance of synaptic structure throughout many animal species. However, how BMP signaling modulates presynaptic neurotransmitter release is not yet clear. We studied the function of BMP signaling factors in neurotransmitter release in Drosophila neuromuscular synapses using loss-of-function mutants in genes for BMP modulators, Wit, Mad, and Dad. Larvae with mutations in wit and mad commonly showed a decreased synaptic bouton number in neuromuscular synapses. Larvae with dad mutations showed an increased bouton number. The amplitudes of miniature EJC (mEJC) were normal for these mutants. Wit and mad mutants showed decreased evoked EJC (eEJC) amplitude and increased paired pulse facilitation, implying impaired presynaptic neurotransmitter release. We found a reduction in readily releasable neurotransmitters pool sizes in wit and mad mutants. However, dad mutants showed a normal probability of neurotransmitter release and readily releasable pool sizes and normal eEJC amplitude even with clear abnormalities in synaptic structure. These results suggested that BMP signaling was critical for each step of presynaptic neurotransmission. The results also suggested that BMP signaling modulated both synaptic structure and function independently and specifically. © 2016 Elsevier Inc. All rights reserved.

Keywords: Bone morphogenic protein Neurotransmitter release Synaptic vesicle pool Release probability Drosophila Neuromuscular junction

1. Introduction Synaptic transmission is important for neural functions including sensation, perception, decision, learning, and memory. To control synaptic transmission, regulating synaptic structure and function via interaction between presynaptic and postsynaptic neurons is critical. Secreted molecules are core factors mediating cell-to-cell interaction in synapses. In particular, Wnt and bone morphogenic protein (BMP) ligands mediate interactions with specific directions (e.g., pre-to-postsynaptic or post-topresynaptic), unlike diffusible small molecules such as nitric oxide and prostaglandins [1,2]. BMP, which is a TGF-b superfamily member and an important developmental regulator, is a typical

* Corresponding author. Department of Physiology, Seoul National University School of Dentistry, 28 Yeongeon, Jongno, Seoul, 110-749, South Korea. E-mail address: [email protected] (S.-Y. Choi). 1 These authors contributed equally to this work. http://dx.doi.org/10.1016/j.bbrc.2016.09.072 0006-291X/© 2016 Elsevier Inc. All rights reserved.

secreted molecule mediating retrograde signaling in synapses [3]. BMP signaling modulates neuronal differentiation [4,5], axon guidance [5], and dendrite growth [6] in developmental periods. Furthermore, in postmitotic periods, BMP signaling is actively involved in the maintenance of synaptic structure [7] and synaptic plasticity [8,9]. Recently, BMP receptors were reported to be regulated by FMR1 [10] and Ube3a [11], which is closely related to the pathology of fragile X syndrome and Angelman syndrome. Structural modulation mediated by BMP signaling (e.g., synapse formation and maintenance) was elucidated with Drosophila neuromuscular junction (NMJ) synapses [12]. The Drosophila type II BMP receptor, wishful thinking (Wit) is reported to regulate the synaptic growth of NMJs [13e15] and synapse stabilization [16]. Mad, the Drosophila ortholog of Smad, controls the number and size of synaptic boutons, similar to Wit [13,14,17]. In contrast, a loss-offunction mutant of Dad, the Drosophila ortholog of i-smad for inhibiting Smad functions, shows synaptic overgrowth [14,18]. BMP signaling modulates synaptic structure and function. A loss-of-function mutant of wit shows markedly decreased evoked

S.-H. Lee et al. / Biochemical and Biophysical Research Communications 479 (2016) 440e446

441

2. Methods

washed with PBS containing 0.1% Triton X-100 (PBT) and blocked in 5% bovine serum albumin in PBT for 1 h, and incubated with primary antibodies at 4
C overnight. After several washes with PBT, dissections were incubated with secondary antibodies for 1 h at room temperature. Samples were incubated with FITC-conjugated anti-HRP (Jackson ImmunoResearch Laboratories, West Grove, PA) at 1:200 for 2 h at room temperature. Antibodies were diluted as follows: mouse anti-DLG (1:500; Developmental Studies Hybridoma Bank), goat FITC-conjugated anti-HRP (1:200, Jackson ImmunoResearch Laboratories), and donkey Cy3-conjugated secondary antibodies (1:200, Jackson ImmunoResearch Laboratories). Fluorescence images of NMJs labeled with anti-HRP and anti-DLG were acquired with a FV300 laser-scanning confocal microscope (Olympus, Japan) and accompanying FLUOVIEW software using a Plan Apo 40  0.90 NA objective. All quantifications were performed at NMJ 6/7 in the A2 segment. For analysis of NMJ morphology, confocal sections were maximally projected. Quantification of bouton number was performed as previously described [21].

2.1. Fly stocks and maintenance

2.3. Electrophysiology

Drosophila strains were raised on standard yeast, sugar, and agar medium at 25
C. The wild-type strain was w1118 unless otherwise noted. Strains with witA12, witB11, mad237, and dadJ1e4 were from the Bloomington Stock Center. Strains with UAS-Dad were generated as described [19]. The GAL4 lines used in this study were: C155-GAL4 (all-neuron specific) and OK6-GAL4 (motoneuron-specific) from the Bloomington Stock Center.

Two-electrode voltage-clamp (TEVC) recordings of wandering third-instar female Drosophila larvae NMJ were obtained from ventral longitudinal muscle 6 in segment A3-A4 at room temperature as described previously with modifications [22]. All dissections and recordings were performed in HL3.1 saline: 70 mM NaCl, 5 mM KCl, 4 mM MgCl2, 5 mM trehalose, 115 mM sucrose, and 5 mM HEPES [23]. Larval dissection was in Ca2þ-free saline to minimize muscle contraction. TEVC recording was performed in 2 mM Ca2þ. Recording electrodes were filled with 3 M KCl and had resistances of 10e15 MU. Recordings were made from cells with an initial resting membrane potential between 60 and 70 mV at a holding potential of 70 mV with a Geneclamp 500 amplifier (Molecular

excitatory junction potential (EJP) with normal miniature EJP, implying a presynaptic defect [14]. However, except for homeostatic plasticity, the details of the impaired synaptic transmission of wit mutants are not yet fully understood. Even though genetic deletion of BMPR1a/1b in mouse MNTB neurons decreases the probability of neurotransmitter release and the pool size of readily releasable vesicles [7], it is still unclear whether the BMP-mediated modulation of neurotransmitter release is general effect throughout many other synapses, or mouse MNTB neurons specific. At least, the involvement of wit mutants has not been investigated. In addition, whether mad and dad mutants show effects that are similar or opposite to wit mutants is unknown. This study examined the synaptic functions of Drosophila mutants of Wit and its modulators Mad and Dad by characterizing the mechanism of presynaptic neurotransmitter release in the neuromuscular synapses. Using this approach, we elucidated the role of the BMP signaling pathway in synaptic function.

2.2. Immunohistochemistry and imaging of larval NMJs Wandering third-instar larvae were dissected in Ca2þ-free HL3 saline and fixed in Bouin's fixative for 7 min [20]. Larvae were then

Fig. 1. Presynaptic BMP molecules control NMJ structure. Depletion of BMP molecules caused aberrant growth of larval NMJs. (A and C) Representative confocal images of anti-HRPlabeled NMJ 6/7 in BMP signaling mutants and presynaptic-Dad overexpressing mutants. Scale bar, 50 mm. (B and D) Quantification of total bouton number as percentage of control in genotypes: (B) w1118 (wild-type), witA12/witB11, mad237/mad237, dadJ1e4/dadJ1e4, (D) C155-GAL4/þ and C155-GAL4/þ; UAS-dad/þ (dad OE; all-neuron). Number of NMJs examined per genotype is indicated inside bars. Significant differences between control and each genotypes are indicated as ***P < 0.001 by Student's t-test. Error bars are SEM for all figures.

442

S.-H. Lee et al. / Biochemical and Biophysical Research Communications 479 (2016) 440e446

Devices, Sunnyvale, CA, USA). Cut segmental nerves were stimulated with a glass suction electrode at a suprathreshold voltage level (5 V) for 0.2 m. Signals were filtered at 10,000 Hz, acquired with Axoscope 10.2 (Molecular Devices) or Clampex 9.2 (Molecular

Devices) and analyzed with Clampfit 10.2 (Molecular Devices). Readily releasable pool sizes from cumulative EJC plots were determined as described from train stimulation protocols of 100 pulses applied at 60 Hz [24,25]. EJC amplitudes were measured

Fig. 2. Evoked synaptic transmissions of wit and mad mutants decreased whereas dad mutants shows intact evoked synaptic transmissions. (A) Spontaneous mEJCs recorded for controls and BMP molecule mutants. Representative mEJC traces that showing a typical 2 s sample from a 3-min recording session (left), quantification of mean amplitudes of spontaneous mEJC (middle), and cumulative probability histograms of mEJC amplitude (right, n ¼ 10, 9, 14, and 11 larvae) are depicted. (B) Average traces of eEJCs from control and BMP-molecule mutants (left) and quantification of mean EJC amplitudes (right, n ¼ 9, 8, 14, and 11 larvae). (C) Presynaptic dad overexpression with two different neuronal GAL4 drivers. Genotypes were UAS-Dad/þ, C155-GAL4/þ , C155/þ ; UAS-Dad/þ (dad OE; all-neuron), OK6-GAL4/þ, OK6/UAS-Dad (Dad OE; motoneuron). Shown are typical traces (left), quantified mean mEJC amplitudes (middle), and cumulative probability of mEJC amplitudes (right). (D) eEJCs recorded for genotypes UAS-Dad/þ, C155-GAL4/þ, C155/þ; UAS-Dad/þ (Dad OE; all-neuron), OK6-GAL4/þ, OK6/UAS-Dad (Dad OE; motoneuron). Average traces (left) and quantified mean EJC amplitudes (right). Error bars, SEM. Number of animals tested is indicated inside bars. Student t-test, ***P < 0.001; *P < 0.05; ns, not significant.

S.-H. Lee et al. / Biochemical and Biophysical Research Communications 479 (2016) 440e446

from peak to baseline immediately before EJC onset. Trend lines were calculated with last 20 EJC amplitudes (between the 80th and 100th stimulation) and back-extrapolated to time zero. 2.4. Statistical analysis All quantitative data are expressed as mean ± SEM. Results were analyzed for differences using either Student's t-test with direct comparison between an experimental and a control dataset, or one-way ANOVA post hoc tests across multiple datasets, as appropriate. Software was Clampfit 10.2 (Molecular Devices) or SPSS (IBM). The Kolmogorov-Smirnov test was used for analysis of cumulative plot data. Differences were considered to be significant when P < 0.05. 3. Results 3.1. Presynaptic BMP molecules control NMJ structure To test the effect of BMP signaling on synaptic structure, we analyzed total bouton numbers in Drosophila third-instar NMJ synapses of loss-of-function mutants in wit, dad, or mad. The wit and mad mutants showed decreased bouton numbers, whereas dad mutants showed increased bouton numbers (Fig. 1A). We prepared neuron-specific Dad-overexpressing larvae by overexpression of UAS-Dad in the wild-type background using a pan-neuron-specific C155-GAL4 driver. The Dad-overexpressing larvae showed significantly reduced overall bouton numbers compared with controls (P < 0.001, Student's t-test) (Fig. 1C). These results were consistent with previous findings that inhibition of BMP signaling mediates synaptic undergrowth [13,18]. 3.2. Presynaptic BMP signaling molecules differently regulate presynaptic structure and function We tested the basal synaptic transmission of BMP signaling mutants by measuring synaptic currents using two-electrode voltage clamp experiments. Amplitudes for spontaneous mEJCs

A

were normal in wit, mad, and dad mutants (Fig. 2A). The mEJCs were also normal in neuron-specific Dad-overexpressing larvae with the pan-neuron-specific C155-GAL4 driver (Fig. 2C). We obtained the same results with other neuron-specific Dad-overexpressing larvae with a motoneuron-specific OK6-GAL4 driver (Fig. 2C). We monitored the amplitude of eEJCs and found significantly reduced eEJC amplitudes in wit and mad mutant larvae (P < 0.001, Fig. 2B), consistent with previous reports [14,26]. The eEJC amplitudes were normal in dad mutants (Fig. 2B), which had structural phenotypes that were the opposite of wit mutants. The Dad-overexpressing larvae with both the pan-neuronal (C155-GAL4) and motoneuron-specific GAL4 drivers (OK6-GAL4) showed reduced EJC amplitude, similar to wit mutants (Fig. 2D). To address the mechanism for this structure-function mismatch, we analyzed the formation of ghost boutons which are immature synaptic boutons lacking active zones and postsynaptic elements such as Discs-Large (Dlg) [27] in dad mutants. Whereas each WT presynaptic boutons correctly faced a postsynaptic concentration of Dlg, there was an increased frequency of presynaptic boutons that had no opposite postsynaptic Dlg in dad mutants (Fig. S1). These results suggested that dad knockdown resulted in a structural phenotype but not a functional phenotype, meaning a mismatch in structure and function. 3.3. Presynaptic release probability is altered in wit and mad but not dad mutants To further characterize the presynaptic function of BMP signaling, we tested neurotransmitter release probability (Pr) for the mutants' NMJ synapses. The paired pulse ratio (PPR) is the amplitude ratio of the first and second postsynaptic currents evoked by two closely separated stimulations. PPR is useful for analyzing changes in Pr [28e30]. We found that wit and mad mutants showed increased PPR over 10e50 m interstimuli intervals, whereas dad mutants showed a normal PPR compared to wild type animals (witA12/B11, P < 0.001; mad237, P ¼ 0.4; dadJ1E4, P ¼ 0.84, at 25 m inter-stimulus intervals, Fig. 3). These results suggested that

witA12/B11

WT

443

mad237

dadJ1E4

50 nA 10 ms

1.1

witA12/B11 dadJ1E4

1.1

PPR @ 25 ms ISI

Paired pulse ratio

B

WT mad237

1

0.9

0.8 10

25

50

100

** *

1 n.s 0.9

0.8

9

8

14

11

Interstimuli Interval (ms) Fig. 3. Probability of presynaptic vesicle release (Pr) altered by Wit and Mad depletions but not Dad. (A) Typical average paired-pulse traces from indicated genotypes at 25 m interstimulus intervals (ISIs). (B) Quantification of mean paired-pulse ratios at each ISI (10, 25, 50, 100 m each). (C) Bar graph of mean paired-pulse ratio at 25 m ISI. Error bars, SEM. Number of animals tested is indicated inside bars. Student t-test, ***P < 0.001; *P < 0.5; ns, not significant.

444

S.-H. Lee et al. / Biochemical and Biophysical Research Communications 479 (2016) 440e446

(witA12/B11, P ¼ 0.014; mad237, P < 0.001; dadJ1E4, P ¼ 0.247, Fig. 4C). The results showed that BMP signaling was important for maintaining RRP size in presynaptic terminals. The results also confirmed that dad mutants had structural changes without functional changes in NMJ synapses.

wit and mad mutants had reduced Pr, whereas dad mutants had normal Pr. 3.4. The size of readily releasable synaptic vesicle pools (RRP) is decreased in wit and mad but not dad mutants To further characterize presynaptic neurotransmitter release properties in BMP-signaling mutants, we analyzed the size of RRP. The RRP is the secretory vesicle pool that immediately contributes to neurotransmitter release with stimulation and can be calculated by back-extrapolation of high-frequency stimulation-mediated EJC responses [25]. The responses to train stimulation at 60 Hz showed wit and mad mutants with reduced initial peaks and sustained EJC levels, whereas dad mutants showed no difference in EJC amplitude compared to wildtype animals (Fig. 4A and B). Extrapolation of the cumulative quantal contents and time plots showed a reduced RRP size in wit and mad mutants, but normal RRP size in dad mutants

A

4. Discussion This study analyzed neurotransmission in BMP signaling mutants and found a decreased probability of release and decreased readily releasable pool sizes in wit and mad null mutants. We also found that dad mutants showed the opposite structural phenotype of wit and mad mutants, but normal functional neurotransmission phenotypes. In Drosophila NMJ synapses, the most intensely studied BMPmediated modulation of synaptic function is homeostatic plasticity, which is increased presynaptic neurotransmitter release

dadJ1E4

WT

Initial 10 peaks Final 10 peaks

witA12/B11

Initial 10 peaks

Final 10 peaks

mad237 50 nA

B

Evoked EJC amplitude (nA)

10 ms

WT

witA12/B11

100

mad237 dadJ1E4

50

0 0

0.5

1

1.5

D

8000

Readily releasable pool

C

Cumulative quantal contents

Time (sec)

6000 4000 2000 0 0

0.5

1

1.5

2500

n.s .

2000 1500

*

1000

**

500 0

8

7

10

7

Time (sec) Fig. 4. Size of readily releasable pools (RRPs) decreased in wit and mad, but not dad mutants. (A) Representative traces of eEJCs during 60 Hze100 pulse train. Initial and final 10 peaks of indicated genotypes from 100 evoked responses. (B) Quantification of mean eEJC amplitudes during 60 Hze100 pulse train. (C) Average cumulative quantal content in BMP mutants during 100 pulses at 60 Hz train. To estimate RRP sizes, back extrapolation of linear fits to average cumulative quantal content at 200 m yielded estimates for release-ready vesicles of 1512 for WT, 846 for witA12/B11, 674 for mad237, and 1936 for dadJ1E4. (D) Quantification of RRP sizes in indicated genotypes. Error bars, SEM. Number of animals tested is indicated inside bars. Student t-test, *P < 0.05; **P < 0.001; ns, not significant.

S.-H. Lee et al. / Biochemical and Biophysical Research Communications 479 (2016) 440e446

induced by decreased postsynaptic glutamate receptor activity [31]. Many studies support the hypothesis that BMP signaling modulates homeostatic plasticity and maintains normal neurotransmission by compensating for impaired postsynaptic receptor activity [32e35]. However, the modulation mechanisms of homeostatic plasticity and the probability of glutamate release are clearly distinguishable. For instance, Drosophila syntaxin 1A mutants with severely decreased Pr have intact homeostatic plasticity [17]. This study found decreased Pr in wit and mad mutants. We believe a novel feature of BMP signaling-mediated regulation of presynaptic neurotransmission is that it is distinct from homeostatic plasticity. Our findings might not be unexpected because they support the decreased Pr in the calyx of Held MNTN synapses of BMPR1a/1b conditional double-knockout mice [7]. However, confirming these results in Drosophila NMJs where BMP-mediated synaptic modulation has been intensely studied is meaningful to show the modulation is general throughout many other synapses. In addition, we extended the findings downstream of BMP receptors. A notable finding was that mad knockdown showed a similar functional phenotype as wit knockdown but not dad knockdown. We also found a reduced RRP size in BMP-signaling mutants. RRP size is mainly affected by Pr and RRP recovery efficiency, but the molecular mechanisms of the regulations are identical. For instance, loss-of-function mutants of Bruchpilot, which promotes accumulation of Ca2þ channels, neurotransmitter containing vesicles, and postsynaptic GluRII receptors and forms active zones [36,37], have decreased Pr and slowed post-tetanic recovery. However, their RRP sizes, miniature EJPs and eEJPs are normal [38]. Therefore, our results showing that wit and mad mutants have decreased Pr and decreased RRP size provide evidence that BMP signaling modulates multiple steps of presynaptic neurotransmission. Further investigations to identify RRP sizes and molecular Prregulating mechanisms will lead to a deep understanding of BMP signaling as a presynaptic release modulator. In our results, wit and mad mutants showed similar synaptic undergrowth and decreased Pr and RRP size, whereas dad mutants showed synaptic overgrowth but normal Pr and RRP size. For dad mutants, the hypothesis that structural defects cause functional defects is no longer valid. The results are interesting because generally wit/mad and dad mutants are expect to have opposite effects. In fact, a mismatch between the structural and functional phenotypes is not unique to dad mutants. For example, loss-of-function Bruchpilot mutants show decreased eEJC amplitude and increased Pr, but intact synaptic synaptic growth and morphology [38]. Null mutants of Drosophila Rab3-interacting molecule-binding protein, DRBP, show decreased eEJC amplitude and decreased Pr, but normal synaptic growth and morphology [39]. We hypothesized several ideas for the cause of the mismatch of structural (e.g., synaptic growth) and functional phenotypes (e.g., release probability modulation) in dad mutants. The downstream signaling of BMP receptor activation is classified into (1) a nuclear pathway of phosphorylated Mad and its transcriptional modulation with Trio and Rac activation [40], and (2) a local pathway of phosphorylated Mad at synaptic terminals followed by local LIM kinase and cofilin activation and subsequent actin modulation [16,41]. A third pathway different from other known BMP signaling cascades has been suggested [42]. We expected that the structural and functional mismatch in dad mutants might be due to the diversity of BMP signaling pathways such as canonical vs. noncanonical effects or nuclear vs. local effects, or to different sensitivities of BMP signaling. For example, Dad-overexpressing and wit null-mutant larvae are reported to share similar structural phenotypes of activity-dependent ghost bouton budding, however, their synaptic growth is different [41]. These results are explained by differential sensitivity of BMP

445

signaling factors. We found that dad mutant has increased presynaptic boutons without opposite postsynaptic Dlg (Fig. S1). The results suggested that there might be no functional changes in dad mutants because the increased bouton number in dad mutants is mostly with impaired postsynaptic maturation. Although further investigation is necessary to understand the exact mechanisms, our results pose novel questions about structurefunction coupling in synapses. Taken together, we characterized the synaptic functions of Drosophila mutants in canonical BMP signaling factors, Wit, Mad, and Dad. We found that BMP signaling had multiple roles in the functional modulation of presynaptic neurotransmitter release. We believe that our findings will help in our understanding of the contribution of BMP signaling to synaptic functional modulation. Furthermore, the application of BMP signaling-mediated synaptic functional modulation on synapses of the mammalian central nervous system will contribute to better understanding of etiological mechanisms and therapeutic solutions for synaptopathyrelated neurological and psychiatric diseases. Acknowledgements This work was supported by the National Research Foundation of Korea (2016905481, 2016006811). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.bbrc.2016.09.072. Transparency document Transparency document related to this article can be found online at http://dx.doi.org/10.1016/j.bbrc.2016.09.072. References [1] E.M. Dickins, P.C. Salinas, Wnts in action: from synapse formation to synaptic maintenance, Front. Cell. Neurosci. 7 (2013) 162. [2] K. Shen, P. Scheiffele, Genetics and cell biology of building specific synaptic connectivity, Annu. Rev. Neurosci. 33 (2010) 473e507. [3] V. Bayat, M. Jaiswal, H.J. Bellen, The BMP signaling pathway at the Drosophila neuromuscular junction and its links to neurodegenerative diseases, Curr. Opin. Neurobiol. 21 (2011) 182e188. [4] A.A. Teleman, M. Strigini, S.M. Cohen, Shaping morphogen gradients, Cell 105 (2001) 559e562. [5] F. Charron, M. Tessier-Lavigne, Novel brain wiring functions for classical morphogens: a role as graded positional cues in axon guidance, Development 132 (2005) 2251e2262. [6] G.S. Withers, D. Higgins, M. Charette, G. Banker, Bone morphogenetic protein7 enhances dendritic growth and receptivity to innervation in cultured hippocampal neurons, Eur. J. Neurosci. 12 (2000) 106e116. [7] L. Xiao, N. Michalski, E. Kronander, E. Gjoni, C. Genoud, G. Knott, R. Schneggenburger, BMP signaling specifies the development of a large and fast CNS synapse, Nat. Neurosci. 16 (2013) 856e864. [8] M. Sun, M.J. Thomas, R. Herder, M.L. Bofenkamp, S.B. Selleck, M.B. O'Connor, Presynaptic contributions of chordin to hippocampal plasticity and spatial learning, J. Neurosci. 27 (2007) 7740e7750. [9] B. Berke, J. Wittnam, E. McNeill, D.L. Van Vactor, H. Keshishian, Retrograde BMP signaling at the synapse: a permissive signal for synapse maturation and activity-dependent plasticity, J. Neurosci. 33 (2013) 17937e17950. [10] R. Kashima, S. Roy, M. Ascano, V. Martinez-Cerdeno, J. Ariza-Torres, S. Kim, J. Louie, Y. Lu, P. Leyton, K.D. Bloch, T.B. Kornberg, P.J. Hagerman, R. Hagerman, G. Lagna, A. Hata, Augmented noncanonical BMP type II receptor signaling mediates the synaptic abnormality of fragile X syndrome, Sci. Signal 9 (2016) ra58. [11] W. Li, A. Yao, H. Zhi, K. Kaur, Y.C. Zhu, M. Jia, H. Zhao, Q. Wang, S. Jin, G. Zhao, Z.Q. Xiong, Y.Q. Zhang, Angelman syndrome protein Ube3a regulates synaptic growth and endocytosis by inhibiting BMP signaling in Drosophila, PLoS Genet. 12 (2016) e1006062. [12] M. Deshpande, A.A. Rodal, The crossroads of synaptic growth signaling, membrane traffic and neurological disease: insights from Drosophila, Traffic 17 (2016) 87e101.

446

S.-H. Lee et al. / Biochemical and Biophysical Research Communications 479 (2016) 440e446

~es, [13] H. Aberle, A.P. Haghighi, R.D. Fetter, B.D. McCabe, T.R. Magalha C.S. Goodman, Wishful thinking encodes a BMP type II receptor that regulates synaptic growth in Drosophila, Neuron 33 (2002) 545e558. s, H. Bao, T.E. Haerry, M.J. Shimell, P. Duchek, B. Zhang, [14] G. Marque M.B. O'Connor, The Drosophila BMP type II receptor Wishful Thinking regulates neuromuscular synapse morphology and function, Neuron 33 (2002) 529e543. s, A.P. Haghighi, R.D. Fetter, M.L. Crotty, T.E. Haerry, [15] B.D. McCabe, G. Marque C.S. Goodman, M.B. O'Connor, The BMP homolog Gbb provides a retrograde signal that regulates synaptic growth at the Drosophila neuromuscular junction, Neuron 39 (2003) 241e254. [16] B.A. Eaton, G.W. Davis, LIM Kinase1 controls synaptic stability downstream of the type II BMP receptor, Neuron 47 (2005) 695e708. [17] C.P. Gould, G.W. Davis, The BMP ligand Gbb gates the expression of synaptic homeostasis independent of synaptic growth control, Neuron 56 (2007) 109e123. [18] K.M. O'Connor-Giles, L.L. Ho, B. Ganetzky, Nervous wreck interacts with thickveins and the endocytic machinery to attenuate retrograde BMP signaling during synaptic growth, Neuron 58 (2008) 507e518. [19] K. Tsuneizumi, T. Nakayama, Y. Kamoshida, T.B. Kornberg, J.L. Christian, T. Tabata, Daughters against dpp modulates dpp organizing activity in Drosophila wing development, Nature 389 (1997) 627e631. [20] B.A. Stewart, H.L. Atwood, J.J. Renger, J. Wang, C.F. Wu, Improved stability of Drosophila larval neuromuscular preparations in haemolymph-like physiological solutions, J. Comp. Physiol. A 175 (1994) 179e191. [21] M. Nahm, A.A. Long, S.K. Paik, S. Kim, Y.C. Bae, K. Broadie, S. Lee, The Cdc42selective GAP rich regulates postsynaptic development and retrograde BMP transsynaptic signaling, J. Cell. Biol. 191 (2010) 661e675. [22] Y.C. Chen, Y.Q. Lin, S. Banerjee, K. Venken, J. Li, A. Ismat, K. Chen, L. Duraine, H.J. Bellen, M.A. Bhat, Drosophila Neuroligin 2 is required presynaptically and postsynaptically for proper synaptic differentiation and synaptic transmission, J. Neurosci. 32 (2012) 16018e16030. [23] Y. Feng, A. Ueda, C.F. Wu, A modified minimal hemolymph-like solution, HL3.1, for physiological recordings at the neuromuscular junctions of normal and mutant Drosophila larvae, J. Neurogenet. 18 (2004) 377e402. [24] A. Weyhersmüller, S. Hallermann, N. Wagner, J. Eilers, Rapid active zone remodeling during synaptic plasticity, J. Neurosci. 31 (2011) 6041e6052. [25] S. Hallermann, M. Heckmann, R.J. Kittel, Mechanisms of short-term plasticity at neuromuscular active zones of Drosophila, HFSP J. 4 (2010) 72e84. [26] B.D. McCabe, S. Hom, H. Aberle, R.D. Fetter, G. Marques, T.E. Haerry, H. Wan, M.B. O'Connor, C.S. Goodman, A.P. Haghighi, Highwire regulates presynaptic BMP signaling essential for synaptic growth, Neuron 41 (2004) 891e905. [27] B. Ataman, V. Budnik, U. Thomas, Scaffolding proteins at the Drosophila neuromuscular junction, Int. Rev. Neurobiol. 75 (2006) 181e216. [28] T. Manabe, D.J. Wyllie, D.J. Perkel, R.A. Nicoll, Modulation of synaptic transmission and long-term potentiation: effects on paired pulse facilitation and EPSC variance in the CA1 region of the hippocampus, J. Neurophysiol. 70

(1993) 1451e1459. rineau, B.H. G€ [29] D. Debanne, N.C. Gue ahwiler, S.M. Thompson, Paired-pulse facilitation and depression at unitary synapses in rat hippocampus: quantal fluctuation affects subsequent release, J. Physiol. 491 (1996) 163e176. [30] L.E. Dobrunz, C.F. Stevens, Heterogeneity of release probability, facilitation, and depletion at central synapses, Neuron 18 (1997) 995e1008. [31] G.W. Davis, Homeostatic signaling and the stabilization of neural function, Neuron 80 (2013) 718e728. [32] S.A. Petersen, R.D. Fetter, J.N. Noordermeer, C.S. Goodman, A. DiAntonio, Genetic analysis of glutamate receptors in Drosophila reveals a retrograde signal regulating presynaptic transmitter release, Neuron 19 (1997) 1237e1248. [33] G.W. Davis, A. DiAntonio, S.A. Petersen, C.S. Goodman, Postsynaptic PKA controls quantal size and reveals a retrograde signal that regulates presynaptic transmitter release in Drosophila, Neuron 20 (1998) 305e315. [34] G.W. Davis, Homeostatic control of neural activity: from phenomenology to molecular design, Annu. Rev. Neurosci. 29 (2006) 307e323. [35] C.A. Frank, M.J. Kennedy, C.P. Goold, K.W. Marek, G.W. Davis, Mechanisms underlying the rapid induction and sustained expression of synaptic homeostasis, Neuron 52 (2006) 663e677. [36] R.J. Kittel, C. Wichmann, T.M. Rasse, W. Fouquet, M. Schmidt, A. Schmid, D.A. Wagh, C. Pawlu, R.R. Kellner, K.I. Willig, S.W. Hell, E. Buchner, M. Heckmann, S.J. Sigrist, Bruchpilot promotes active zone assembly, Ca2þ channel clustering, and vesicle release, Science 312 (2006) 1051e1054. [37] W. Fouquet, D. Owald, C. Wichmann, S. Mertel, H. Depner, M. Dyba, S. Hallermann, R.J. Kittel, S. Eimer, S.J. Sigrist, Maturation of active zone assembly by Drosophila Bruchpilot, J. Cell. Biol. 186 (2009) 129e145. [38] S. Hallermann, R.J. Kittel, C. Wichmann, A. Weyhersmüller, W. Fouquet, S. Mertel, D. Owald, S. Eimer, H. Depner, M. Schw€ arzel, S.J. Sigrist, M. Heckmann, Naked dense bodies provoke depression, J. Neurosci. 30 (2010) 14340e14345. [39] K.W. Liu, M. Siebert, S. Mertel, E. Knoche, S. Wegener, C. Wichmann, T. Matkovic, K. Muhammad, H. Depner, C. Mettke, J. Bückers, S.W. Hell, M. Müller, G.W. Davis, D. Schmitz, S.J. Sigrist, RIM-binding protein, a central part of the active zone, is essential for neurotransmitter release, Science 334 (2011) 1565e1569. [40] R.W. Ball, M. Warren-Paquin, K. Tsurudome, E.H. Liao, F. Elazzouzi, C. Cavanagh, B.S. An, T.T. Wang, J.H. White, A.P. Haghighi, Retrograde BMP signaling controls synaptic growth at the NMJ by regulating trio expression in motor neurons, Neuron 66 (2010) 536e549. [41] Z.D. Piccioli, J.T. Littleton, Retrograde BMP signaling modulates rapid activitydependent synaptic growth via presynaptic LIM kinase regulation of cofilin, J. Neurosci. 34 (2014) 4371e4381. [42] M.J. Sulkowski, T.H. Han, C. Ott, Q. Wang, E.M. Verheyen, J. LippincottSchwartz, M. Serpe, Novel, noncanonical BMP pathway modulates synapse maturation at the Drosophila neuromuscular junction, PLoS Genet. 12 (2016) e1005810.