Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

<i>Drosophila</i>α- and β-Spectrin Mutations Disrupt Presynaptic Neurotransmitter Release

Drosophilaα- and β-Spectrin Mutations Disrupt Presynaptic Neurotransmitter Release The Journal of Neuroscience, June 15, 2001, 21(12):4215–4224 Drosophila a- and b-Spectrin Mutations Disrupt Presynaptic Neurotransmitter Release 1 1 2 1 David E. Featherstone, Warren S. Davis, Ronald R. Dubreuil, and Kendal Broadie 1 2 Department of Biology, University of Utah, Salt Lake City, Utah 84112-0840, and Department of Neurobiology, Pharmacology, and Physiology, University of Chicago, Chicago, Illinois 60637 Spectrins are plasma membrane-associated cytoskeletal pro- display severely disrupted neurotransmission without altered teins implicated in several aspects of synaptic development morphological synaptogenesis. Contrary to current models, the and function, including presynaptic vesicle tethering and absence of spectrins does not alter postsynaptic glutamate postsynaptic receptor aggregation. To test these hypotheses, receptor field function or the ultrastructural localization of pre- we characterized Drosophila mutants lacking either a-or synaptic vesicles. However, the subcellular localization of nu- b-spectrin. The Drosophila genome contains only one merous synaptic proteins is disrupted, suggesting that the a-spectrin and one conventional b-spectrin gene, making it an defects in presynaptic neurotransmitter release may be attrib- ideal system to genetically manipulate spectrin levels and ex- utable to inappropriate assembly, transport, or localization of amine the resulting synaptic alterations. Both spectrin proteins proteins required for synaptic function. are strongly expressed in the Drosophila neuromusculature and Key words: spectrin; Drosophila; synapse; neuromuscular highly enriched at the glutamatergic neuromuscular junction. junction; synaptogenesis; cysteine string protein; Discs large; Protein null a- and b-spectrin mutants are embryonic lethal and PSD-95; synaptotagmin; synapsin; syntaxin; glutamate receptor Spectrin was originally discovered in erythrocytes, in which a- synapsin I) of synaptic vesicles near the active zone (Landis, 1988; and b-spectrin heterotetramers form part of a submembrane Goodman et al., 1995; Sikorski et al., 2000); (2) the initiation of meshwork critical for membrane structural integrity (Bennett, SNARE vesicle f usion by “dimpling” the cell membrane (Good- 1990; Bennett and Gilligan, 1993). C losely related spectrin iso- man, 1999); and (3) the anchoring of glutamate and /or acetylcho- forms are found in most other eukaryotic cell types, in which they line receptors within the postsynaptic density (Bloch and Mor- preferentially associate with plasma membranes at sites of cell – row, 1989; Daniels, 1990; Bloch et al., 1997; Wechsler and cell contact (Bennett and Gilligan, 1993). Spectrin (also known as Teichberg, 1998; Hirai and Matsuda, 1999). A recent study using “fodrin”) is particularly abundant in mammalian brain, in which cultured hippocampal cells showed that presynaptic injection of it comprises 2–3% of total protein (Davis and Bennett, 1983; antibodies against the synapsin-binding region of b-spectrin com- Bennett and Gilligan, 1993). However, almost nothing is known pletely blocked synaptic transmission (Sikorski et al., 2000), ar- about the f unction of spectrin in neurons. guing that presynaptic synapsin – spectrin interactions are essen- In neurons, spectrin is often preferentially localized to both tial for synaptic f unction. No other f unctional studies of synaptic central and peripheral synapses (Bloch and Morrow, 1989; spectrin have been done. Daniels, 1990; Masliah et al., 1991; Bewick et al., 1992, 1996; Drosophila i s an attracti ve system in which to test whether Goodman et al., 1995; Kordeli, 2000), suggesting a role for spec- spectrins are requi red for sy naptic development and f unction. trin at the synaptic membrane. C almodulin, sodium channels, Drosophila contain only three members of the highly conser ved munc-13, and synapsin I, which all play important synaptic roles, spectrin family, each encoded by a single gene (Adams et al., have all been shown to bind to spectrins (Srinivasan et al., 1988; 2000), and prev iously identified null mutants ex i st for each Steiner et al., 1989; Dubreuil et al., 1991; Sikorski and Goodman, spectrin subunit (D ubreuil et al., 1998, 2000; T homas et al., 1991; Iga et al., 1997; Sakaguchi et al., 1998; Wood and Slater, 1998). Here we show that, in Drosophila protein null mutants 1998). However, the specific role that spectrins might play at a-or b-spectrin, neuromuscular junction (N MJ) mor- synapses is unknown. Proposed roles for synaptic spectrins in- lacking clude the following: (1) the capture and subsequent tethering (via pholog y i s normal, but neurotransmi ssion i s severely di s- rupted. T he protein null mutants show a reduction in sponta- Received Nov. 6, 2000; revised March 20, 2001; accepted March 22, 2001. neous sy naptic event frequenc y w ith no changes in response to This work was supported by a National Institutes of Health (N IH) National pressure-ejected glutamate or in spontaneous sy naptic event Research Service Award postdoctoral fellowship to D.F., N IH Grant GM49301 to R.R.D., and an EL JB Foundation fellowship, grants from the Muscular Dystrophy amplitude, demonstrating that the neurotransmi ssion defect i s Association, and N IH Grant GM54544 to K .B. We thank L. S. Goldstein for presy naptic. Ultrastructural analysi s reveals no change in pre- b-spectrin antibodies, E. Buchner for synapsin antibodies, K . Z insmaier for C SP antibodies, T. Littleton for synaptotagmin antibodies, H. Bellen for syntaxin 1 sy naptic vesicle di stribution, but immunoc y tochemi str y shows antibodies, and V. Budnik for DLG antibodies. We also thank T. Fergestad for that many classes of sy naptic proteins are dramatically mi slo- confocal assistance and M. Hammarlund, R. Weimer, and C. Rodesch for critical cali z ed or absent in both a- and b-spectrin mutants. We review of this manuscript. Correspondence should be addressed to Kendal Broadie, University of Utah, propose that spectrin in neuronal sy napses i s requi red for Department of Biology, 257 South 1400 East, Salt Lake C ity, UT 84112-0840. capture and tethering of membrane-associated proteins re- E-mail: [email protected]. Copyright © 2001 Society for Neuroscience 0270-6474/01/214215-10$15.00/0 qui red for presy naptic neurotransmitter release. 4216 J. Neurosci., June 15, 2001, 21(12):4215–4224 Featherstone et al. • Spectrins and Neurotransmitter Release represents a measure of both protein localization and antibody quality MATERIALS AND METHODS because poor antibodies might be expected to lower the ratio because of Fl y stock s. Molecularly characterized protein null mutants for a- and high nonspecific immunoreactivity (high background) and /or reduced b-spectrin were used in this study (Lee et al., 1993; Dubreuil et al., 2000). specific immunoreactivity. Antibody quality effects can be eliminated by rg41 em6 a-spec and b-spec do not produce detectable protein, as shown by normalizing the raw ratios to wild type. Normalized ratios (see Fig. 7) em6 immunoblots (Lee et al., 1993; Dubreuil et al., 2000). b-spec produces were calculated by dividing the ratios for wild type and each mutant by a truncated protein product according to immunoblots (Dubreuil et al., these numbers (normalized r 5 R/R ). WT 2000), but b-spectrin protein is immunohistochemically undetectable in Morpholog y. Quantification of NMJ area was performed using the situ (D. E. Featherstone and K . Brodie, unpublished data), presumably public domain Java-based image processing and analysi s program because the truncated protein is rapidly degraded and /or fails to localize. Image/J . Confocal images (see Fig. 2 A) of wild-type and mutant Drosoph- rg41 a-Spectrin null mutant l(3)dre stocks (Lee et al., 1993) were main- ila NMJs, visualized by fluorescently conjugated anti-HRP (which stains all 67c23 tained as heterozygotes using a thi rd chromosome balancer [yw ; nerve membranes), were manually outlined using Image/J. Once given, rg41 rg41 r u l(3)dre st e/In(3LR)TM3, y1Sb Ser]. Homozygous l(3)dre mu- pixel dimensions (recorded automatically by the Z eiss confocal software), tants are rescued to adulthood by transgenic expression of an a-spectrin Image calculated the area of the outlined region (NMJ area). For bouton minigene under a ubiquitin promoter (Lee et al., 1993). b-Spectrin null counting, synaptic varicosities (swellings) were also visualized with fluores- em21 em6 mutants (b-spec and b-spec ) stocks (Dubreuil et al., 2000) were cently conjugated anti-HRP (1:100; Molecular Probes). maintained as heterozygotes using an FM7[Kruppel-GFP] balancer chro- Electron microscopy. Genotyped embryos were prepared for transmis- mosome (C asso et al., 1999). As with a-spectrin mutants, b-spectrin sion electron microscopy (TEM) using standard techniques (Prokop et mutants are rescued to adulthood by transgenic expression of b-spectrin al., 1996; Fergestad et al., 1999). Briefly, mature embryos (22–24 hr after (Dubreuil et al., 2000). Oregon-R (OR) was used for wild-type (W T) egg laying; AEL at 25°C) were manually dechorionated and injected with controls. fixative (5% glutaraldehyde in 0.05 M phosphate buffer). The preparation Embr yo preparation and dissection. Homozygous mutant embryos were was then transferred to 2.5% glutaraldehyde in 0.05 M phosphate buffer selected from siblings based on the absence of balancer chromosome for 30 – 60 min. Specimens were washed in buffer, transferred to 1% markers (green fluorescent protein for b-spectrin mutants, yellow for osmium tetroxide in dH O for 3 hr, washed again in dH O, and stained 2 2 a-spectrin mutants). Homozygous yellow mutants do not show any signif- en bloc in 2% aqueous uranyl acetate for 30 min. Embryos were dehy- icant difference in excitatory junctional current (EJC) amplitude com- drated in an ethanol series, passed through propylene oxide, and trans- pared with Oregon R ( y/y, 1548 6 280 pA; OR, 1476 6 117 pA; n 5 ferred to araldite. Ribbons of thin (;55 nm) sections were obtained and 5–13; p 5 0.78). For electrophysiology and embryonic immunohisto- examined on a Hitachi (Tokyo, Japan) H-7100 TEM. Active zones that chemistry, morphologically and temporally staged [22–24 hr after egg were identified in at least two consecutive sections were imaged and laying (AEL) at 25°C] embryos were dechorionated with bleach and analyzed using N IH Image. Vesicles were considered to be “clustered” if devitellinated manually. For dissection, embryos were glued (Histoacryl they were within 235 nm of the active zone T-bar (Fergestad et al., 1999) Blue; B. Braun Biotech International GmbH, Melsungen, Germany) to and docked if within one-half vesicle diameter of the presynaptic mem- Sylgard (Dow Corning, Midland, M I)-coated coverslips under saline brane (thus allowing for vesicles that may be in contact with the mem- containing (in mM): 135 NaC l, 5 KC l, 4 MgC l , 1.8 CaCl , 72 sucrose, 2 2 brane but were not perfectly bisected in the cross-section). and 5 N-Tris[hydroxy-methyl]methyl-2-aminoethane sulfonic acid (TES), Electrophysiolog y. Electrophysiology and data analysis were performed pH 7.2. A slit was made manually along the dorsal midline using a glass as described previously (Featherstone et al., 2000). Briefly, whole-cell capillary pulled to a sharp point, and the body walls were glued flat to the patch-clamp recordings from embryonic muscle 6 were obtained in an coverslip. If electrophysiology was to be performed on the dissected extracellular solution containing (in mM): 135 NaC l, 5 KC l, 4 MgC l , 1.8 embryos, the exposed muscle sheath was enzymatically removed after CaCl , 72 sucrose, and 5 TES, pH 7.2. For miniature EJC (mEJC) dissection using 1–2 min exposure to 1 mg /ml collagenase (type IV; recordings, calcium was replaced with 5 mM tetrodotoxin (TTX). The Sigma, St. L ouis, MO). patch pipette solution contained (in mM): 120 KC l, 20 KOH, 4 MgC l , Immunohistochemistr y. Dissected embryos or wandering third-instar 0.25 C aC l , 5 EGTA, 4Na ATP, 36 sucrose, and 5 TES. For EJC 2 2 larvae were fixed in 4% paraformaldehyde for 30 – 45 min and processed measurements, the segmental nerve was stimulated by delivering 5–10 V, according to standard techniques (White, 1998; Beumer et al., 1999; 0.1 msec pulses via a glass suction pipette. To assay the glutamate Featherstone et al., 2000). Mouse monoclonal Drosophila a-spectrin receptor field, 1 mM glutamate was pressure ejected (100 msec pulse) antibody (3A9) (Dubreuil et al., 1997) and rabbit polyclonal Drosophila from a small-tipped (;5 mm opening) pipette directly onto the NMJ. b-spectrin antibody (Byers et al., 1989) were used at 1:100. These a- and Data were analyzed using C lampfit 8 or 9a (Axon Instruments, Foster b-spectrin antibodies show no detectable staining in a-or b-spectrin null C ity, CA) and /or Minianalysis 4 (Synaptosoft Inc., Decatur, GA). mutants and /or on immunoblots, confirming antibody specificity. Mouse Statistics. All data are presented as mean 6 SEM. Each n represents a monoclonal synapsin antibody (K lagges et al., 1996) was used at 1:100. different embryo of the stated genotype. Statistics from spontaneous Mouse monoclonal cysteine string protein (C SP) antibody (Z insmaier et EJC s (sEJC s)and mEJC s are derived from at least 5 min of continuous al., 1994) was used at 1:200. Rabbit polyclonal synaptotagmin antibody recording (often 10 –20 min in the case of the low-frequency mEJC s). In (Littleton et al., 1993) was used at 1:500. Mouse monoclonal syntaxin 1A all figures, statistical significance (compared with wild-type controls) is (Schulze et al., 1995) was used at 1:500. Rabbit polyclonal Discs large indicated as *p , 0.05, **p , 0.01, and ***p , 0.001, . Unless otherwise (DLG) (Lahey et al., 1994) was used at 1:1000. Immunoreactivity for stated, statistical significance was determined using Student’s t test. all of these antibodies is abolished in the appropriate null mutants, con- Because spontaneous synaptic event amplitude distributions are skewed firming antibody specificit y. F luorescein i sothioc yanate and tetra- rather than Gaussian, we compared these distributions statistically using methylrhodamine-conjugated secondary antibodies (goat anti-mouse and the Kolmogorov– Smirnov test and do not report variance. goat anti-rabbit; Molecular Probes, Eugene, OR) were used at 1:400. FI TC -conjugated anti-HRP (Molecular Probes) was used at 1:100. Im- RESULTS ages were obtained on a Z eiss (Oberkochen, Germany) L SM510 laser- scanning confocal microscope. Drosophila spectrins and spectrin mutants Synaptic/nonsynaptic immunoreactivit y ratios. Pixel intensity (0 –255) A search of the sequenced Drosophila genome reveals only three for boutons and nearby extrasynaptic regions (muscle for DLG and nerve for all other proteins) was measured in Z eiss Image Browser software members of the highly conserved spectrin family, each encoded by using raw (completely unaltered) confocal fluorescent images. Average a single gene (Adams et al., 2000; Pinder and Baines, 2000). background fluorescence intensity (dark areas beyond–in between mus- Drosophila a-spectrin (GenBank accession number A33733) is cles) was subtracted from these values. To derive the “synaptic /nonsyn- 64% identical at the amino acid level to human brain a-spectrin/ aptic immunoreactivity ratio,” the background-corrected synaptic fluo- fodrin (GenBank accession number A35715). Drosophila b-spectrin rescence intensity was divided by the background-corrected nonsynaptic fluorescence intensity. Thus, the ratio was calculated as follows: r 5 (S 2 (GenBank accession number A46147) is 56% identical to human B)/(N 2 B), where r is the synaptic /nonsynaptic immunoreactivity ratio, b-spectrin (GenBank accession number N P003119). Drosophila S is the fluorescence intensity in synaptic boutons, B is the background b -spectrin/k arst (GenBank accession number CAA37939) is the fluorescence intensity measured from dark nontissue parts of the image, most divergent, with 31% amino acid identity to the human and N is the fluorescence intensity in nonsynaptic tissues (nerve for C SP, synapsin, syntaxin, and synaptotagmin; muscle for DLG). This raw ratio ortholog b V-spectrin (GenBank accession number AAF65317). Featherstone et al. • Spectrins and Neurotransmitter Release J. Neurosci., June 15, 2001, 21(12):4215–4224 4217 Figure 1. a- and b-Spectrin immunoreactiv- ity in the neuromusculature of Drosophila third-instar larvae. A, Confocal fluorescence images of NMJs stained simultaneously with antibodies raised against a-spectrin and b-spectrin. a-Spectrin immunoreacti v it y i s show n in g reen (lef t), b-spectrin immuno- reacti v it y i s show n in red,(middle), and overlapping a/b-spectrin staining appears yellow (merged image, right). Scale bar, 10 mm. B, C onfocal fluorescence images of N MJs stained simultaneously w ith antibod- ies rai sed against C SP and b-spectrin. C SP immunoreacti v it y i s show n in g reen (lef t), and b-spectrin immunoreacti v it y i s show n in red (middle); merged image i s on the right. Scale bar, 10 mm. Drosophila protein null mutants for a- and b-spectrin are embry- b-spectrin and C SP staining is not colocalized, suggesting that onic – early larval lethal, with defects in the structure and f unction the majority of spectrin protein is associated with the periphery of epithelial cells (Lee et al., 1993, 1997; Dubreuil and Grushko, of the presynaptic membrane and /or dense membrane foldings of 1998; Dubreuil et al., 2000). In contrast, null mutants for b - the postsynaptic subsynaptic reticulum. We conclude from this spectrin are semiviable, with mild defects including rough eyes, immunohistochemistry that both a- and b-spectrin are present at disrupted epithelial morphogenesis, tracheal defects, and mis- the wild-type Drosophila NMJ, in both presynaptic and postsyn- shapen wings (Thomas et al., 1998). These results suggest that a- aptic cells. In subsequent experiments, we used the NMJ as a and /or b-spectrin could play vital roles in synaptogenesis and model synapse to examine the role of spectrins in synaptogenesis synaptic f unction, whereas b -spectrin is unessential. Therefore, H and synaptic f unction. we focused our efforts on characterizing the role of a- and b-spectrin subunits in synaptic development and f unction. a- and b-Spectrin null mutants have morphologically For this study, we used previously identified protein null mu- normal neuromuscular junctions tants for a-spectrin (Lee et al., 1993) and b-spectrin (Dubreuil et We examined gross morphology in protein null mutants of both al., 2000). Homozygous a-spectrin null mutants fail to hatch a- and b-spectrin (Lee et al., 1993; Dubreuil et al., 2000). Light (;50%) or die as early first-instar larvae (;50%). Homozygous microscope (4003) examination of several dozen acutely dis- b-spectrin protein null mutants fail to hatch (;90%), and the rest sected mutant embryos reveals that both a- and b-spectrin mu- (;10%) die as early first-instar larvae. Both classes of mutants are tants have normally formed neuromusculature, epidermis, and lethargic and display limited movement, consistent with a neuro- epidermal specializations (e.g., denticles and mouth parts). The physiological or muscular defect. We chose to study these mutants only visible difference is that a-spectrin mutants have slightly at the embryonic NMJ for several reasons. First, this synapse is thinner muscles, and unhatched (but living) a-spectrin embryos accessible in vivo to a variety of cell biological techniques, includ- often have uninflated trachea at normal hatch time (22–24 hr ing patch-clamp electrophysiology, immunohistochemistry, and AEL). The gut phenotype of these mutants has been described electron microscopy. Second, the development, morphology, and previously (Lee et al., 1993; Dubreuil et al., 1998, 2000). f unction of the Drosophila NMJ is well described and relatively To examine NMJ anatomy, we visualized embryonic body wall invariant from animal to animal. Like many synapses in the neuroanatomy with fluorescently labeled anti-HRP, which recog- mammalian C NS, the Drosophila NMJ is glutamatergic. These nizes neural membranes (Fig. 2). We saw no qualitative differ- features make the NMJ an excellent place to detect and quantif y ences in sites of muscle innervation or presynaptic branching any changes resulting from spectrin disruption. pattern. In Figure 2 A, we show confocal fluorescent images of wild-type and mutant embryonic NMJs visualized with fluores- Spectrins are present at the Drosophila NMJ cently conjugated anti-HRP. In each panel, four individual NMJs Using antibodies specific for Drosophila spectrins (Byers et al., are shown. On the lef t is the linear NMJ lying between ventral 1989; Dubreuil et al., 1997), we examined the neuromuscular longitudinal muscles 6 and 7, and on the right are the more lateral localization of both a- and b-spectrin (Fig. 1). As shown in Figure NMJs on muscles 13 and 12. Quantification of morphology at the 1, both a-spectrin (Fig. 1 A, green) and b-spectrin (Fig. 1 A, red) muscle 6/7 NMJ showed that there was no significant difference in are found in presynaptic axons proximal to the NMJ and in the the number of synaptic boutons (Fig. 2 B)(WT,9.9 6 0.6 boutons; periphery of presynaptic boutons. Although a-spectrin staining is rg41 em21 em6 a , 8.4 6 0.6 boutons; b , 9.6 6 0.8 boutons; b , 8.6 6 0.9 typically weaker, most of the a- and b-spectrin staining in the boutons; n 5 6 –10). Because embryonic boutons are often indis- NMJ appears colocalized (Fig. 1 A, right panel, a- and b-spectrin tinct (Fig. 2 A) and therefore difficult to count, we also quantified overlapping expression appears yellow). a- and b-Spectrin immu- NMJ size by measuring muscle 6/7 NMJ area (see Materials and noreactivity is also strong throughout muscle (Fig. 1 A). We independently confirmed the specificity of both a- and b-spectrin Methods). We detected no significant difference in NMJ area antibodies in null mutant backgrounds (see Materials and between wild-type and a-or b-spectrin mutants (Fig. 2C)(WT, 2 rg41 2 em21 2 Methods). 47.0 6 1.9 mm ; a , 44.8 6 4.9 mm ; b , 48.1 6 2.6 mm ; em6 2 In Figure 1 B, we show double-labeling with antibodies against b , 49.1 6 1.9 mm ; n 5 6 –10). We conclude from this quan- b-spectrin and the presynaptic protein C SP. Much of the tification, as well as qualitative observation of several dozen 4218 J. Neurosci., June 15, 2001, 21(12):4215–4224 Featherstone et al. • Spectrins and Neurotransmitter Release Figure 2. Morphology of embryonic NMJs in spectrin mutants is normal. A, Confocal fluores- cence images of NMJs on ventral longitudinal muscles 6/7 and 12 and 13 in a single ventral hemisegment of wild-type and spectrin mutant embryos. NMJ morphology was visualized by staining with fluorescently conjugated anti-HRP antibodies. Scale bar, 10 mm. B, Quantification of synaptic bouton number at the muscle 6/7 NMJ shows no significant difference between the ge- notypes. C, Quantification of muscle 6/7 NMJ area shows no significant difference between the genotypes. embryos, that NMJ morphology is not detectably altered in either Presy naptic defects can be a result of fault y neurotransmitter a-or b-spectrin mutants. release (sy naptic vesicle filling and f usion), vesicle rec ycling defects, or both. Defecti ve sy naptic vesicle c ycling can be Spectrin mutants are defective in revealed when the ner ve i s stimulated at high frequencies neurotransmitter release (Fergestad et al., 1999; Kuromi and K idokoro, 2000). Under conditions of high demand, neurotransmitter release i s re- Because spectrins are present at the NMJ and the morphology of duced because of a reduction in the available pool of spectrin mutant NMJs was normal, we were able to test whether neurotransmitter-filled vesicles (Kuromi and K idokoro, 2000). synaptic f unction was disrupted. To record NMJ f unction, we In a- and b-spectrin mutants, the reduction in sy naptic trans- voltage clamped (260 mV) muscle 6 using standard patch-clamp mi ssion during high-frequenc y stimulation i s slightly, but not techniques. To evoke synaptic activity, we stimulated (0.5 msec, significantly, impai red (Fig. 3C) (normali z ed amplitude at 20 5–15 V) the presynaptic segmental nerve using a suction elec- rg41 em21 H z: W T, 0.59 6 0.06; a , 0.62 6 0.03; b , 0.39 6 0.12; trode. As shown in Figure 3A, evoked EJC s in both a- and em6 b , 0.51 6 0.07; n 5 4 –7). T hese results suggest that short- b-spectrin mutants are reduced to approximately one-quarter term vesicle c ycling in the mutants i s sufficient to maintain the rg41 normal amplitude (W T, 1476 6 117 pA; a , 473 6 92 pA; em6 em21 reduced rate of exoc y tosi s show n in Figure 3A. Because we did b , 453 6 57 pA; b , 334 6 55 pA; n 5 9 –13; p , 0.001 vs not assay endoc y tosi s in the mutants di rectly (e.g., w ith FM1– W T for each allele, using Student’s t test). a- and b-Spectrin 43), we cannot completely rule out defects in endoc y tosi s. mutant EJC amplitudes are statistically indistinguishable from However, because of the relati vely small (and stati stically each other (Fig. 3A). These results demonstrate that, despite insignificant) alteration in mutant responses to high-frequenc y normal morphology, spectrin mutants have severely reduced syn- stimulation, we conclude that the f unctional defect i s primarily aptic transmission. in exoc y tosi s rather than endoc y tosi s or vesicle c ycling. To determine whether the transmission defect in the spectrin Together, the data in Figure 3 suggest that the synaptic trans- mutants was presynaptic or postsynaptic, glutamate (1 mM) was mission defect in a- and b-spectrin mutants is attributable to pressure ejected (5–10 mm tip pipette, 100 msec pulse) directly specific disruption in neurotransmitter release, with no f unctional onto the NMJ of voltage-clamped (260 mV) postsynaptic muscle alteration in the postsynaptic receptors. We confirmed these 6 (Featherstone et al., 2000). If the defect in synaptic transmission conclusions using analysis of spontaneous synaptic currents (Fig. is attributable to an alteration in postsynaptic glutamate receptor 4). A reduction in the probability of presynaptic vesicle f usion is f unction, the resulting glutamate-gated currents should be re- revealed by less frequent spontaneous synaptic events, whereas an duced in spectrin mutants. As shown in Figure 3B, neither a- nor alteration in receptor localization, receptor number, or receptor b-spectrin mutants showed any detectable alteration in glutamate rg41 em6 response (W T, 1805 6 248 pA; a , 1763 6 227 pA; b , biophysics causes changes in the amplitude of spontaneous syn- em21 1755 6 178 pA; b , 1630 6 105 pA; n 5 7–11). Because the aptic events. Figure 4 shows analysis of sEJC s, which are recorded receptor field is f unctionally normal in spectrin mutants yet in the presence of calcium (Fig. 4 A, B, lef t column), and mEJC s, transmission is greatly reduced, the striking transmission defect which are recorded in the absence of extracellular calcium and shown in Figure 3A must be presynaptic. the presence of TTX (Fig. 4 A, B, right column). The frequency of Featherstone et al. • Spectrins and Neurotransmitter Release J. Neurosci., June 15, 2001, 21(12):4215–4224 4219 Figure 3. Patch-clamp electrophysiology from voltage-clamped (260 mV) muscle demonstrates that spectrin mutant NMJs have severely reduced neurotransmitter release, with no f unctional alteration of Figure 4. Analysis of spontaneous synaptic events demonstrates that postsynaptic receptor fields. A, EJC amplitude (evoked by nerve stimula- spectrin null mutants have decreased synaptic vesicle f usion rates but no tion) is significantly reduced in a- and b-spectrin mutants. Representative f unctional alteration of the receptor fields. Currents were recorded in EJC s are shown on the right. B, Currents triggered by pressure ejection of both normal (1.8 mM) calcium saline and saline containing 0 mM calcium 1mM glutamate (100 msec pulse) onto the postsynaptic membrane dem- plus 5 mM TTX (to block endogenous nerve activity). A, Frequency of onstrate that the spectrin mutant glutamate receptor field f unction is not spontaneous synaptic currents in voltage-clamped (260 mV) muscle in significantly different from wild type. Representative glutamate-gated both a-spectrin and b-spectrin mutants is reduced in both high- and currents are shown on the right. C, Stimulation of the NMJ at increasing low-calcium conditions, suggesting disruption of presynaptic vesicle f u- frequencies reveals no significant difference in decrement of EJC ampli- sion. B, Amplitude histograms (composed of data from multiple record- tude between mutants and control. ings) reveal no significant difference (Kolmogorov– Smirnov test) in spec- trin mutant event amplitudes in either normal (1.8 mM) calcium or the both types of event are lowered in the spectrin mutants (Fig. 4 A), absence of endogenous activity (0 calcium plus TTX), suggesting that the spectrin mutants have no f unctional alteration in the postsynaptic gluta- suggesting that a- and b-spectrin mutants share a calcium- mate receptor field. independent deficit in synaptic vesicle f usion (1.8 mM Ca :WT, rg41 em21 11.69 6 1.59 Hz; a , 6.83 6 0.74 Hz; b , 1.69 6 0.41 Hz; em6 21 b , 3.52 6 1.51 Hz; n 5 6 –11; TTX plus 0 mM Ca :WT, rg41 em21 vivo or on immunoblot, possibly because of the small number of 0.13 6 0.03 Hz; a , 0.08 6 0.02 Hz; b , 0.05 6 0.01 Hz; em6 embryonic receptors (100 –200 receptors per NMJ vs tens of b , 0.07 6 0.02 Hz; n 5 5–14). In contrast, the amplitude of thousands of receptors per NMJ in larvae). However, electro- mEJC s is not significantly altered in spectrin mutants compared physiology is arguably the most sensitive (able to detect a single with wild-type controls (Fig. 4 B). These data, like those in Figure f unctional receptor) and most quantitative method of determin- 3B, suggest that the postsynaptic receptor field is f unctionally rg41 ing receptor field integrity. unchanged (mean mEJC amplitudes: W T, 157.9 pA; a , 143.7 em21 em6 Together, the electrophysiological results show that a- and pA; b , 142.6 pA; b , 140.3 pA; n 5 4 –14 embryos, thou- sands of events; p . 0.05 by Kolmogorov– Smirnov test). We are b-spectrin mutants have severely impaired synaptic transmission unable to confirm this finding qualitatively using antibodies raised and that this impairment is attributable specifically to disruption against Drosophila glutamate receptors. Despite success in larvae, of neurotransmitter release, without any f unctional alteration in we have been unable to visualize embryonic receptors either in the postsynaptic receptor field. 4220 J. Neurosci., June 15, 2001, 21(12):4215–4224 Featherstone et al. • Spectrins and Neurotransmitter Release Figure 5. Ultrastructural analysis of embryonic NMJs shows morphologically normal boutons in spectrin mutants, with no alterations in the distribution of active zones or synaptic vesicles. A, TEM cross-sections through embryonic NMJ boutons showing presynaptic active zones with electron-dense T-bars (surrounded by clustered vesicles) in opposition to a postsynaptic density. Active zones are indicated with arrowheads. Scale bar, 250 nm. B, High-magnification images of active zones from each genotype, showing individual T-bars and clustered vesicles. Arrowheads indicate T-bars. C, Quantification of numbers of docked vesicles (within one-half vesicle diameter of the presynaptic membrane), numbers of clustered vesicles (within 235 nm of T-bar), and vesicle density throughout bouton cross-section. [within one-half vesicle diameter of the presynaptic membrane Ultrastructure of spectrin mutants is normal (Broadie et al., 1995)] or clustered [within 235 nm of T-bar Both a- and b-spectrin mutants have normal NMJ morphology rg41 (Fergestad et al., 1999)] vesicles (docked: W T, 1.74 6 0.16; a , but reduced neurotransmitter release, supporting the idea that em21 em6 1.58 6 0.17; b , 1.79 6 0.22; b , 1.42 6 0.76; n 5 19 –27; spectrins may cluster synaptic vesicles at the active zone. This rg41 em21 clustered: W T, 21.3 6 1.1; a , 22.1 6 1.31; b , 19.7 6 1.23; hypothesis, called “casting the line,” suggests that one end of em6 b , 17.95 6 0.73; n 5 19 –27). Similarly, when synaptic vesicle spectrin is anchored to active zones, whereas the other end density throughout the entire bouton cross-section is quantified, captures vesicles via an interaction with synapsin (Landis, 1988; both a- and b-spectrin null mutants are comparable with wild type Goodman et al., 1995; Sikorski et al., 2000). Mislocalization rg41 em21 (vesicle density: W T, 74.4 6 6.9; a , 69.4 6 8.3; b , 96.6 6 and /or absence of synaptic vesicles at the active zone could em6 em6 22.9; b , 49.7 6 9.1; n 5 19 –27). b-spectrin shows a slight explain the spectrin mutant electrophysiological phenotype we (but statistically significant) reduction in clustered vesicles and show in Figures 3 and 4. We tested whether synaptic vesicle vesicle density, but this change is unlikely to explain the synaptic clustering is disrupted in spectrin mutants by examining NMJs transmission defect for two reasons: (1) the change is too small to using electron microscopy (Fig. 5). In a- and b-spectrin mutants, explain the severe decrease in vesicle release, and (2) the change presynaptic and postsynaptic membranes are normally structured rg41 em21 is not shared by either a-spectrin or b-spectrin , which and spaced, internal organelles appear normal, and the distribu- otherwise have identical phenotypes. We conclude that spectrins tion of embryonic T-bars and electron-dense areas associated with do not play a substantial role in synaptic vesicle tethering at active active zones are indistinguishable from wild type (Fig. 5 A, B). zones. Thus, spectrins play no detectable role in the maintenance of gross synaptic morphology. The location of active zones and synaptic vesicles are readily Synaptic protein localization is disrupted in both a- and b-spectrin mutants visible, allowing us to determine whether vesicle clustering is altered in either a-or b-spectrin mutants. We quantified the NMJ morphology in the spectrin mutants is normal by light and number and distribution of synaptic vesicles around each active electron microscopy, yet neurotransmitter release is severely dis- zone, and these results are graphed in Figure 5C. Spectrin mu- rupted. In other (non-neuronal) cell types, spectrins have been tants show no consistent alteration in the number of docked proposed to capture and maintain proteins in distinct membrane- Featherstone et al. • Spectrins and Neurotransmitter Release J. Neurosci., June 15, 2001, 21(12):4215–4224 4221 associated domains, especially at sites of cell – cell interaction (Drubin and Nelson, 1996; Pinder and Baines, 2000). At synapses, proper f unction requires precise assembly and alignment of the molecular machinery required for synaptic vesicle f usion and recycling. If this machinery is mislocalized or incorrectly assem- bled, it would not be surprising to find a synaptic defect such as we observe in a- and b-spectrin mutants. Although there is no method by which we can test whether the in vivo submicrometer assembly of proteins is appropriate in spectrin mutants, we can determine whether synaptic proteins are polarized and properly localized to the NMJ. In epithelial cells, disruption of protein polarization attributable to the absence of spectrin is visible by immunohistochemistry and confocal light microscopy (Dubreuil et al., 2000). We used the same techniques to determine whether spectrins play a similar role in protein compartmentalization at synapses. Figure 6 shows representative staining in wild-type and spec- trin mutant embryos for two of the best Drosophila NMJ markers available: presynaptic anti-C SP and postsynaptic anti-DLG. C SP is present in both vesicular membrane-associated and cytosolic fractions of presynaptic boutons; C SP staining normally appears as tightly localized presynaptic puncta (Z insmaier et al., 1994). DLG is a plasma membrane-associated PDZ [postsynaptic den- sity-95(PSD-95)/ DLG/zona occludens-1] domain protein with 60% homology to PSD-95 that is tightly localized to both presyn- aptic and postsynaptic membranes (Lahey et al., 1994; Budnik et al., 1996). Each panel in Figure 6 shows the body wall neuromus- culature of two to three embryonic hemisegments stained with anti-C SP ( green) and anti-DLG (red). The (out of focus) ventral ganglion (CNS) is visible in the top lef t of each panel, from which segmental nerves (SN ) extend into the body wall musculature on the right. The C NS serves as a positive control for overall image intensity. In wild-type embryos, C SP and DLG staining in the body wall neuromusculature is restricted to tightly defined puncta at the NMJ (Fig. 6, lef t column); little or no staining is visible in either the presynaptic nerve axon or nonsynaptic muscle mem- brane. Thus, neither the segmental nerves nor the majority of muscle tissue is visible in the fluorescence image (Fig. 6, lef t column). In both a- and b-spectrin mutants (Fig. 6, middle and Figure 6. Fluorescent confocal micrographs of embryonic neuromuscula- right columns), the synaptic localization of both presynaptic C SP ture showing distribution of the presynaptic protein C SP and postsynaptic and postsynaptic DLG is dramatically perturbed. The segmental protein DLG. Each panel shows C SP and DLG immunoreactivity in three nerves are now visible (because of C SP immunoreactivity), as are or more hemisegments. In wild-type embryos, DLG (red) and C SP ( green) the muscles (because of DLG immunoreactivity). We conclude tightly associate with NMJ boutons, which appear as immunoreactive that, in both a- and b-spectrin mutants, C SP is distributed abnor- puncta at NMJs in the body wall neuromusculature (lef t column). Note that, in wild-type embryos, anti-C SP and anti-DLG antibodies detect only the mally throughout presynaptic axons, and DLG is distributed NMJ and not the preterminal axon or extrasynaptic regions of the muscle. abnormally throughout muscle cells. Neither protein appears In both a- and b-spectrin null mutants, however, C SP ( green) is abnormally properly polarized and localized to the NMJ boutons in spectrin distributed throughout distal axons (extending horizontally from the C NS mutants. on the lef t into the musculature on the right). In both a- and b-spectrin null mutants, DLG staining (red) is scattered throughout postsynaptic muscles. In addition to C SP and DLG, we examined the staining pat- In each image, a portion of the ventral ganglion (C NS) is shown (out of terns of several other synaptic proteins, including synaptotagmin, focus) as a positive control. Scale bar, 15 mm. synapsin, and syntaxin. Synaptotagmin is a transmembrane pro- tein normally restricted to synaptic vesicles (Littleton et al., 1993; Marqueze et al., 2000). Synapsin is a spectrin-interacting phos- chemically) by comparing staining intensity in NMJ boutons with phoprotein that is associated with the presynaptic actin cytoskel- staining intensity outside the synapse (see Materials and Meth- eton at synaptic boutons (K lagges et al., 1996; Iga et al., 1997; ods). In wild-type embryos, fluorescence intensity from each Hilfiker et al., 1999; T urner et al., 1999). Syntaxin is a transmem- synaptic marker was significantly higher in NMJ boutons than brane protein normally present in presynaptic membrane, includ- elsewhere. Specifically, the synaptic /nonsynaptic fluorescence in- ing both axons and synaptic boutons (Schulze et al., 1995; Gerst, tensity for each marker (in wild-type embryos) was as follows: 1999). All of these proteins, like C SP and DLG, showed severely 30.75 6 6.85 (C SP), 8.56 6 1.65 (DLG), 5.51 6 1.19 (synapsin), disrupted subcellular localization in both a- and b-spectrin mu- 7.46 6 1.88 (synaptotagmin), and 3.41 6 0.74 (syntaxin). In other tant embryos. words, wild-type embryos showed anti-C SP fluorescence that was We quantified protein distribution (measured immunocyto- 30.75 times higher in boutons than in nerve. In contrast, anti- 4222 J. Neurosci., June 15, 2001, 21(12):4215–4224 Featherstone et al. • Spectrins and Neurotransmitter Release 1989; Masliah et al., 1991; Bewick et al., 1992, 1996; Goodman et al., 1995; Gelot et al., 1996; Bloch et al., 1997; Sakaguchi et al., 1998; Wechsler and Teichberg, 1998; Wood and Slater, 1998; Goodman, 1999; Hirai and Matsuda, 1999; Dunaevsky and Con- nor, 2000; Hammarlund et al., 2000; Kordeli, 2000; Sikorski et al., 2000; Sunderland et al., 2000). The role that spectrins might play at synapses has been the subject of intense speculation. The Drosophila genome contains only one a-spectrin and one conven- tional b-spectrin gene, making it an ideal system to genetically manipulate spectrin levels and examine the resulting synaptic alterations. Using protein null mutants for a- and b-spectrin,we tested whether spectrins are required for development and /or f unction of the Drosophila neuromuscular junction. First, we showed that both a- and b-spectrin are present at the Drosophila NMJ (Fig. 1). This observation supports the synaptic localization of spectrins observed in other systems (Bloch and Morrow, 1989; Daniels, 1990; Masliah et al., 1991; Bewick et al., 1992, 1996; Goodman et al., 1995; Kordeli, 2000). Second, we showed that, in a- and b-spectrin mutants, synaptic morphology is normal (Fig. 2). This result contrasts with the severe morpholog- ical defects observed in Caenorhabditis elegans spectrin mutants Figure 7. Spectrin mutants have mislocalized synaptic proteins. Protein (Hammarlund et al., 2000). The normal morphological develop- distribution was quantified by comparing staining intensity in NMJ bou- ment in Drosophila spectrin mutants may be possible because of a tons with staining intensity outside the synapse (see Materials and Meth- ods). In wild-type embryos, fluorescence intensity from each synaptic maternal contribution. Third, because NMJ morphology was nor- marker was much higher in NMJ boutons than elsewhere. In a- and mal, we were able to undertake a detailed electrophysiological b-spectrin mutants, however, immunoreactivity of C SP, DLG, synapto- analysis of synaptic f unction in spectrin mutants (Figs. 3, 4). This tagmin, synapsin, and syntaxin were all reduced in boutons and simulta- analysis showed that both a- and b-spectrin mutants have equal neously increased in nonsynaptic nerve (C SP, synaptotagmin, synapsin, and severe disruptions in synaptic transmission. Using pressure- and syntaxin) or muscle (DLG) membrane. Thus, the relative amount of synaptic protein at the synapse relative to other tissues was significantly ejected glutamate to directly measure postsynaptic glutamate reduced in spectrin mutants. receptor f unction, we were able to rule out the possibility that the transmission defect was attributable to any alteration in glutamate receptor f unction. This conclusion was confirmed using analysis syntaxin fluorescence in wild-type embryos was only 3.41 times of spontaneous synaptic events, which showed reduced probabil- higher in boutons than in nerve. These observations are consis- ity of vesicle f usion yet normal event amplitudes. Thus, we con- tent with the fact that C SP is strongly restricted to synaptic cluded that the source of the synaptic f unction defect in spectrin boutons, whereas syntaxin is present throughout the neuronal mutants was presynaptic. Based on immunohistochemical local- membrane and only weakly polarized to boutons (Schulze et al., ization and biochemistry, spectrins have been proposed to play an 1995). This raw ratio represents a measure of both protein local- important role in development and /or f unction of postsynaptic ization and antibody quality because poor antibodies might be receptor fields (Bloch and Morrow, 1989; Daniels, 1990; Bloch et expected to lower the ratio because of high nonspecific immuno- al., 1997; Wechsler and Teichberg, 1998; Hirai and Matsuda, reactivity (high background) and /or reduced specific immuno- 1999). Our data strongly suggest that this is not true at the reactivity. Drosophila NMJ, although we cannot rule out the possibility that For ease of comparison and to eliminate effects on the ratios maternal spectrin contributes to the initial development of the from antibody quality, we normalized all of these ratios to wild postsynaptic receptor field. type. Normalized ratios for wild-type and spectrin mutant em- What is the cause of the presynaptic defect in spectrin mutants? bryos are shown in Figure 7. The synaptic /nonsynaptic immuno- Spectrins have been proposed to capture and tether (via synapsin reactivity ratios in both a- and b-spectrin mutants for C SP, DLG, I) synaptic vesicles near the active zone (Landis, 1988; Goodman synaptotagmin, synapsin, and syntaxin were each significantly et al., 1995; Sikorski et al., 2000). In support of this hypothesis, it reduced compared with wild type ( p , 0.05; t test). The normal- has been shown that disruption of spectrin – synapsin binding, via ized ratios (W T is 1) for a-spectrin mutants were as follows: antibodies raised against the synapsin binding site of spectrin, 0.04 6 0.004 (C SP), 0.14 6 0.017 (DLG), 0.21 6 0.036 (synapsin), eliminate synaptic transmission in cultured hippocampal cells 0.36 6 0.782 (syntaxin), and 0.47 6 0.060 (synaptotagmin) (n 5 (Sikorski et al., 2000). However, synaptic vesicle localization was 4 –13; mean of 10). The normalized ratios (W T is 1) for b-spectrin never examined in that study. To test whether the defective mutants were as follows: 0.03 6 0.003 (C SP), 0.14 6 0.017 (DLG), neurotransmitter release in Drosophila spectrin mutants was at- 0.23 6 0.051 (synapsin), 0.25 6 0.065 (syntaxin), and 0.12 6 0.006 tributable to altered synaptic vesicle localization, we examined (synaptotagmin) (n 5 4 –13; mean of 10). We conclude from these the ultrastructure of wild-type and spectrin mutant synaptic ter- results that synaptic proteins are improperly polarized and local- minals using electron microscopy (Fig. 5). We found no changes ized in both a- and b-spectrin mutants. in synaptic vesicle distribution in the spectrin mutants. This DISCUSSION observation is in agreement with recent data from C. elegans Spectrins have been known for over a decade to be present at b-spectrin mutants (Hammarlund et al., 2000). Thus, data from both central and peripheral synapses in a variety of organisms both Drosophila and C. elegans spectrin mutants argue that spec- (Lazarides et al., 1984; Bloch and Morrow, 1989; Goodman et al., trins are not required for synaptic vesicle clustering or docking. Featherstone et al. • Spectrins and Neurotransmitter Release J. Neurosci., June 15, 2001, 21(12):4215–4224 4223 Different distributions of dystrophin and related proteins at nerve- We are unable to visualize, in either wild type or mutants, any muscle junctions. NeuroReport 3:857– 860. electron-dense “rods” connecting synaptic vesicles to the active Bewick GS, Young C, Slater CR (1996) Spatial relationships of utrophin, zone. These rods, which are visible in some other preparations, dystrophin, beta-dystroglycan and beta-spectrin to acetylcholine recep- tor clusters during postnatal maturation of the rat neuromuscular have been suggested to be spectrin based on their size (Landis, junction. J Neurocytol 25:367–379. 1988; Goodman et al., 1995). Bloch RJ, Morrow JS (1989) An unusual beta-spectrin associated with In epithelial cells, spectrins are required for polarization and clustered acetylcholine receptors. J C ell Biol 108:481– 493. Bloch RJ, Bezakova G, Ursitti JA, Z hou D, Pumplin DW (1997) A localization of a variety of membrane-associated proteins, espe- membrane skeleton that clusters nicotinic acetylcholine receptors in cially at sites of cell – cell contact (Bennett, 1990; Bennett and muscle. Soc Gen Physiol Ser 52:177–195. Broadie K , Prokop A, Bellen HJ, O’Kane C J, Schulze K L, Sweeney ST Gilligan, 1993; Drubin and Nelson, 1996; Brown and Breton, (1995) Syntaxin and synaptobrevin f unction downstream of vesicle 2000; Dubreuil et al., 2000; Pinder and Baines, 2000). Because docking in Drosophila. Neuron 15:663– 673. synapses are highly polarized sites of cell – cell interaction be- Brown D, Breton S (2000) Sorting proteins to their target membranes. K idney Int 57:816 – 824. tween ectodermally derived cells, it stands to reason that the Budnik V, Koh YH, Guan B, Hartmann B, Hough C, Woods D, Gorczyca f unction of neuronal spectrin might be similar to that shown in M (1996) Regulation of synapse structure and f unction by the Dro- epithelia. We used methods similar to those used in studies of sophila tumor suppressor gene dlg. Neuron 17:627– 640. Byers TJ, Husain-Chishti A, Dubreuil RR, Branton D, Goldstein L S epithelia to show that indeed this is the case; several classes of (1989) Sequence similarity of the amino-terminal domain of Drosoph- synaptic proteins fail to properly polarize and localize in spectrin ila beta spectrin to alpha actinin and dystrophin. J C ell Biol 109:1633–1641. mutants (Figs. 6, 7). Because both a- and b-spectrin are distrib- C asso D, Ramirez-Weber FA, Kornberg TB (1999) GFP-tagged bal- uted widely, spectrin alone cannot be sufficient for organization ancer chromosomes for Drosophila melanogaster. Mech Dev of f unctional synaptic domains. Synaptic spectrins must be “acti- 88:229 –232. Daniels M P (1990) L ocalization of actin, beta-spectrin, 43 x 10(3) Mr vated” via a local synaptic cue or work in conjunction with other and 58 x 10(3) Mr proteins to receptor-enriched domains of newly molecules to capture and accumulate synaptic proteins. In this formed acetylcholine receptor aggregates in isolated myotube mem- regard, neuronal spectrin appears to be different from epithelial branes. J C ell Sci 97:615– 626. Davis J, Bennett V (1983) Brain spectrin. Isolation of subunits and spectrin, which has a polarized distribution that matches its site of formation of hybrids with erythrocyte spectrin subunits. J Biol Chem activity precisely (Dubreuil et al., 2000). Because neurotransmit- 258:7757–7766. Drubin DG, Nelson WJ (1996) Origins of cell polarity. C ell 84:335–344. ter release requires precise organization of presynaptic protein Dubreuil RR, Grushko T (1998) Genetic studies of spectrin: new life for machinery, it is not unreasonable to conclude that the defects in a ghost protein. BioEssays 20:875– 878. synaptic release measured in Drosophila mutants are attributable Dubreuil RR, Brandin E, Reisberg JH, Goldstein L S, Branton D (1991) Structure, calmodulin-binding, and calcium-binding properties of re- to alterations in synaptic protein localization. However, we can- combinant alpha spectrin polypeptides. J Biol Chem 266:7189 –7193. not rule out another, less likely, direct role for spectrin in synaptic Dubreuil RR, Maddux PB, Grushko TA, MacVicar GR (1997) Segrega- vesicle f usion, as has been proposed by Goodman (1999). tion of two spectrin isoforms: polarized membrane-binding sites direct polarized membrane skeleton assembly. Mol Biol C ell 8:1933–1942. In summary, we have shown spectrins are, as in other organ- Dubreuil RR, Frankel J, Wang P, Howrylak J, Kappil M, Grushko TA isms, present in Drosophila synapses. Electrophysiological analy- (1998) Mutations of alpha spectrin and labial block cuprophilic cell differentiation and acid secretion in the middle midgut of Drosophila ses showed that neurotransmitter release in Drosophila a- and larvae. Dev Biol 194:1–11. b-spectrin protein null mutants is severely impaired. However, Dubreuil RR, Wang P, Dahl S, Lee J, Goldstein L S (2000) Drosophila contrary to current models, this synaptic impairment is not at- beta spectrin f unctions independently of alpha spectrin to polarize the Na,K ATPase in epithelial cells. J C ell Biol 149:647– 656. tributable to defects in receptor field f unction or synaptic vesicle Dunaevsky A, Connor EA (2000) F-Actin is concentrated in nonrelease localization. We conclude, based on immunolocalization of sev- domains at frog neuromuscular junctions. J Neurosci 20:6007– 6012. eral classes of synaptic proteins, that proper polarization and Featherstone DE, Rushton EM, Hilderbrand-Chae M, Phillips AM, Jack- son FR, Broadie K (2000) Presynaptic glutamic acid decarboxylase is localization of synaptic proteins does not take place in the ab- required for induction of the postsynaptic receptor field at a glutama- sence of spectrin. We propose, based on these results and the tergic synapse. Neuron 27:71– 84. Fergestad T, Davis WS, Broadie K (1999) The stoned proteins regulate synaptic localization of spectrin, that a spectrin-based scaffold is synaptic vesicle recycling in the presynaptic terminal. J Neurosci formed early in synaptic development, and this scaffold is subse- 19:5847–5860. quently required for proper assembly, transport, or localization of Gelot A, Moreau J, Ben Ari Y, Pollard H (1996) Alpha-brain spectrin mRNA belongs to the population of intradendritically transported synaptic proteins during development. Future work will aim to mRNAs. NeuroReport 8:113–116. understand the time course and mechanisms by which synaptic Gerst JE (1999) SNAREs and SNARE regulators in membrane f usion spectrin is specifically activated and /or localized to capture and and exocytosis. C ell Mol Life Sci 55:707–734. Goodman SR (1999) Discovery of nonerythroid spectrin to the demon- accumulate synaptic proteins. stration of its key role in synaptic transmission. Brain Res Bull 50:345–346. Goodman SR, L opresti L L, Riederer BM, Sikorski A, Z agon IS (1989) REFERENCES Brain spectrin(240/235A): a novel astrocyte specific spectrin isoform. Adams MD, C elniker SE, Holt R A, Evans CA, Gocayne JD, Amanatides Brain Res Bull 23:311–316. PG, Scherer SE, Li PW, Hoskins R A, Galle RF, George R A, Lewis Goodman SR, Z immer W E, C lark MB, Z agon IS, Barker JE, Bloom ML SE, Richards S, Ashburner M, Henderson SN, Sutton GG, Wortman (1995) Brain spectrin: of mice and men. Brain Res Bull 36:593– 606. JR, Yandell MD, Z hang Q, Chen L X, Brandon RC, Rogers YH, Hammarlund M, Davis WS, Jorgensen EM (2000) Mutations in beta- Blazej RG, Champe M, Pfeiffer BD et al. (2000) The genome se- spectrin disrupt axon outgrowth and sarcomere structure. J C ell Biol quence of Drosophila melanogaster. Science 287:2185–2195. 149:931–942. Bennett V (1990) Spectrin: a structural mediator between diverse Hilfiker S, Pieribone VA, C zernik AJ, Kao HT, Augustine GJ, Greengard plasma membrane proteins and the cytoplasm. Curr Opin C ell Biol P (1999) Synapsins as regulators of neurotransmitter release. Philos 2:51–56. Trans R Soc L ond B Biol Sci 354:269 –279. Bennett V, Gilligan DM (1993) The spectrin-based membrane skeleton Hirai H, Matsuda S (1999) Interaction of the C -terminal domain of delta and micron-scale organization of the plasma membrane. Annu Rev glutamate receptor with spectrin in the dendritic spines of cultured C ell Biol 9:27– 66. Purkinje cells. Neurosci Res 34:281–287. Beumer K J, Rohrbough J, Prokop A, Broadie K (1999) A role for PS Iga M, Inui M, Sobue K (1997) Characterization of the interaction integrins in morphological growth and synaptic f unction at the postem- between synapsin I and calspectin (brain spectrin or fodrin). Biochem bryonic neuromuscular junction of Drosophila. Development Biophys Res Commun 231:852– 855. 126:5833–5846. K lagges BR, Heimbeck G, Godenschwege TA, Hof bauer A, Pflugfelder Bewick GS, Nicholson LV, Young C, O’Donnell E, Slater CR (1992) GO, Reifegerste R, Reisch D, Schaupp M, Buchner S, Buchner E 4224 J. Neurosci., June 15, 2001, 21(12):4215–4224 Featherstone et al. • Spectrins and Neurotransmitter Release (1996) Invertebrate synapsins: a single gene codes for several isoforms Schulze K L, Broadie K , Perin MS, Bellen HJ (1995) Genetic and elec- in Drosophila. J Neurosci 16:3154 –3165. trophysiological studies of Drosophila syntaxin-1A demonstrate its role Kordeli E (2000) The spectrin-based skeleton at the postsynaptic mem- in nonneuronal secretion and neurotransmission. C ell 80:311–320. brane of the neuromuscular junction. Microsc Res Tech 49:101–107. Sikorski AF, Goodman SR (1991) The effect of synapsin I phosphoryla- Kuromi H, K idokoro Y (2000) Tetanic stimulation recruits vesicles from tion upon binding of synaptic vesicles to spectrin. Brain Res Bull reserve pool via a cAM P- mediated process in Drosophila synapses. 27:195–198. Neuron 27:133–143. Sikorski AF, Sangerman J, Goodman SR, Critz SD (2000) Spectrin Lahey T, Gorczyca M, Jia X X, Budnik V (1994) The Drosophila tumor (betaSpIIsigma1) is an essential component of synaptic transmission. suppressor gene dlg is required for normal synaptic bouton structure. Brain Res 852:161–166. Neuron 13:823– 835. Srinivasan Y, Elmer L, Davis J, Bennett V, Angelides K (1988) Ankyrin Landis DM (1988) Membrane and cytoplasmic structure at synaptic and spectrin associate with voltage-dependent sodium channels in junctions in the mammalian central nervous system. J Electron Microsc brain. Nature 333:177–180. Tech 10:129 –151. Steiner JP, Walke Jr HT, Bennett V (1989) C alcium /calmodulin inhibits Lazarides E, Nelson WJ, Kasamatsu T (1984) Segregation of two spec- direct binding of spectrin to synaptosomal membranes. J Biol Chem trin forms in the chicken optic system: a mechanism for establishing 264:2783–2791. restricted membrane-cytoskeletal domains in neurons. C ell Sunderland WJ, Son YJ, Miner JH, Sanes JR, C arlson SS (2000) The 36:269 –278. Lee JK , Coyne RS, Dubreuil RR, Goldstein L S, Branton D (1993) C ell presynaptic calcium channel is part of a transmembrane complex link- shape and interaction defects in alpha-spectrin mutants of Drosophila ing a synaptic laminin (alpha4beta2gamma1) with non-erythroid spec- melanogaster. J C ell Biol 123:1797–1809. trin. J Neurosci 20:1009 –1019. Lee JK , Brandin E, Branton D, Goldstein L S (1997) alpha-Spectrin is Thomas GH, Z arnescu DC, Juedes AE, Bales M A, L ondergan A, Korte required for ovarian follicle monolayer integrity in Drosophila melano- CC, K iehart DP (1998) Drosophila betaHeav y-spectrin is essential for gaster. Development 124:353–362. development and contributes to specific cell fates in the eye. Develop- Littleton J T, Bellen HJ, Perin MS (1993) E xpression of synaptotagmin ment 125:2125–2134. in Drosophila reveals transport and localization of synaptic vesicles to T urner K M, Burgoyne RD, Morgan A (1999) Protein phosphorylation the synapse. Development 118:1077–1088. and the regulation of synaptic membrane traffic. Trends Neurosci Marqueze B, Berton F, Seagar M (2000) Synaptotagmins in membrane 22:459 – 464. traffic: which vesicles do the tagmins tag? Biochimie 82:409 – 420. Wechsler A, Teichberg V I (1998) Brain spectrin binding to the NMDA Masliah E, Hansen L, Mallory M, Albright T, Terry RD (1991) Abnor- receptor is regulated by phosphorylation, calcium and calmodulin. mal brain spectrin immunoreactivity in sprouting neurons in Alzhei- EMBO J 17:3931–3939. mer disease. Neurosci Lett 129:1–5. White R AH (1998) Immunolabeling of Drosophila. In: Drosophila:a Pinder JC, Baines AJ (2000) A protein accumulator. Nature practical approach, Ed 2 (Roberts DB, ed), pp 215–240. Oxford: IRL. 406:253–254. Wood SJ, Slater CR (1998) beta-Spectrin is colocalized with both Prokop A, Landgraf M, Rushton E, Broadie K , Bate M (1996) Presyn- voltage-gated sodium channels and ankyrinG at the adult rat neuro- aptic development at the Drosophila neuromuscular junction: assembly muscular junction. J C ell Biol 140:675– 684. and localization of presynaptic active zones. Neuron 17:617– 626. Sakaguchi G, Orita S, Naito A, Maeda M, Igarashi H, Sasaki T, Takai Y Z insmaier K E, Eberle K K , Buchner E, Walter N, Benzer S (1994) (1998) A novel brain-specific isoform of beta spectrin: isolation and its Paralysis and early death in cysteine string protein mutants of Drosoph- interaction with Munc13. Biochem Biophys Res Commun 248:846 – 851. ila. Science 263:977–980. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Journal of Neuroscience Unpaywall

<i>Drosophila</i>α- and β-Spectrin Mutations Disrupt Presynaptic Neurotransmitter Release

Featherstone, David E.; Davis, Warren S.; Dubreuil, Ronald R.; Broadie, Kendal
The Journal of Neuroscience – Jun 15, 2001
Download PDF
10 pages

Loading next page...
 
/lp/unpaywall/i-drosophila-i-and-spectrin-mutations-disrupt-presynaptic-bEkp5OSPjJ

References

References for this paper are not available at this time. We will be adding them shortly, thank you for your patience.

Publisher
Unpaywall
ISSN
0270-6474
DOI
10.1523/jneurosci.21-12-04215.2001
Publisher site
See Article on Publisher Site

Abstract

The Journal of Neuroscience, June 15, 2001, 21(12):4215–4224 Drosophila a- and b-Spectrin Mutations Disrupt Presynaptic Neurotransmitter Release 1 1 2 1 David E. Featherstone, Warren S. Davis, Ronald R. Dubreuil, and Kendal Broadie 1 2 Department of Biology, University of Utah, Salt Lake City, Utah 84112-0840, and Department of Neurobiology, Pharmacology, and Physiology, University of Chicago, Chicago, Illinois 60637 Spectrins are plasma membrane-associated cytoskeletal pro- display severely disrupted neurotransmission without altered teins implicated in several aspects of synaptic development morphological synaptogenesis. Contrary to current models, the and function, including presynaptic vesicle tethering and absence of spectrins does not alter postsynaptic glutamate postsynaptic receptor aggregation. To test these hypotheses, receptor field function or the ultrastructural localization of pre- we characterized Drosophila mutants lacking either a-or synaptic vesicles. However, the subcellular localization of nu- b-spectrin. The Drosophila genome contains only one merous synaptic proteins is disrupted, suggesting that the a-spectrin and one conventional b-spectrin gene, making it an defects in presynaptic neurotransmitter release may be attrib- ideal system to genetically manipulate spectrin levels and ex- utable to inappropriate assembly, transport, or localization of amine the resulting synaptic alterations. Both spectrin proteins proteins required for synaptic function. are strongly expressed in the Drosophila neuromusculature and Key words: spectrin; Drosophila; synapse; neuromuscular highly enriched at the glutamatergic neuromuscular junction. junction; synaptogenesis; cysteine string protein; Discs large; Protein null a- and b-spectrin mutants are embryonic lethal and PSD-95; synaptotagmin; synapsin; syntaxin; glutamate receptor Spectrin was originally discovered in erythrocytes, in which a- synapsin I) of synaptic vesicles near the active zone (Landis, 1988; and b-spectrin heterotetramers form part of a submembrane Goodman et al., 1995; Sikorski et al., 2000); (2) the initiation of meshwork critical for membrane structural integrity (Bennett, SNARE vesicle f usion by “dimpling” the cell membrane (Good- 1990; Bennett and Gilligan, 1993). C losely related spectrin iso- man, 1999); and (3) the anchoring of glutamate and /or acetylcho- forms are found in most other eukaryotic cell types, in which they line receptors within the postsynaptic density (Bloch and Mor- preferentially associate with plasma membranes at sites of cell – row, 1989; Daniels, 1990; Bloch et al., 1997; Wechsler and cell contact (Bennett and Gilligan, 1993). Spectrin (also known as Teichberg, 1998; Hirai and Matsuda, 1999). A recent study using “fodrin”) is particularly abundant in mammalian brain, in which cultured hippocampal cells showed that presynaptic injection of it comprises 2–3% of total protein (Davis and Bennett, 1983; antibodies against the synapsin-binding region of b-spectrin com- Bennett and Gilligan, 1993). However, almost nothing is known pletely blocked synaptic transmission (Sikorski et al., 2000), ar- about the f unction of spectrin in neurons. guing that presynaptic synapsin – spectrin interactions are essen- In neurons, spectrin is often preferentially localized to both tial for synaptic f unction. No other f unctional studies of synaptic central and peripheral synapses (Bloch and Morrow, 1989; spectrin have been done. Daniels, 1990; Masliah et al., 1991; Bewick et al., 1992, 1996; Drosophila i s an attracti ve system in which to test whether Goodman et al., 1995; Kordeli, 2000), suggesting a role for spec- spectrins are requi red for sy naptic development and f unction. trin at the synaptic membrane. C almodulin, sodium channels, Drosophila contain only three members of the highly conser ved munc-13, and synapsin I, which all play important synaptic roles, spectrin family, each encoded by a single gene (Adams et al., have all been shown to bind to spectrins (Srinivasan et al., 1988; 2000), and prev iously identified null mutants ex i st for each Steiner et al., 1989; Dubreuil et al., 1991; Sikorski and Goodman, spectrin subunit (D ubreuil et al., 1998, 2000; T homas et al., 1991; Iga et al., 1997; Sakaguchi et al., 1998; Wood and Slater, 1998). Here we show that, in Drosophila protein null mutants 1998). However, the specific role that spectrins might play at a-or b-spectrin, neuromuscular junction (N MJ) mor- synapses is unknown. Proposed roles for synaptic spectrins in- lacking clude the following: (1) the capture and subsequent tethering (via pholog y i s normal, but neurotransmi ssion i s severely di s- rupted. T he protein null mutants show a reduction in sponta- Received Nov. 6, 2000; revised March 20, 2001; accepted March 22, 2001. neous sy naptic event frequenc y w ith no changes in response to This work was supported by a National Institutes of Health (N IH) National pressure-ejected glutamate or in spontaneous sy naptic event Research Service Award postdoctoral fellowship to D.F., N IH Grant GM49301 to R.R.D., and an EL JB Foundation fellowship, grants from the Muscular Dystrophy amplitude, demonstrating that the neurotransmi ssion defect i s Association, and N IH Grant GM54544 to K .B. We thank L. S. Goldstein for presy naptic. Ultrastructural analysi s reveals no change in pre- b-spectrin antibodies, E. Buchner for synapsin antibodies, K . Z insmaier for C SP antibodies, T. Littleton for synaptotagmin antibodies, H. Bellen for syntaxin 1 sy naptic vesicle di stribution, but immunoc y tochemi str y shows antibodies, and V. Budnik for DLG antibodies. We also thank T. Fergestad for that many classes of sy naptic proteins are dramatically mi slo- confocal assistance and M. Hammarlund, R. Weimer, and C. Rodesch for critical cali z ed or absent in both a- and b-spectrin mutants. We review of this manuscript. Correspondence should be addressed to Kendal Broadie, University of Utah, propose that spectrin in neuronal sy napses i s requi red for Department of Biology, 257 South 1400 East, Salt Lake C ity, UT 84112-0840. capture and tethering of membrane-associated proteins re- E-mail: [email protected]. Copyright © 2001 Society for Neuroscience 0270-6474/01/214215-10$15.00/0 qui red for presy naptic neurotransmitter release. 4216 J. Neurosci., June 15, 2001, 21(12):4215–4224 Featherstone et al. • Spectrins and Neurotransmitter Release represents a measure of both protein localization and antibody quality MATERIALS AND METHODS because poor antibodies might be expected to lower the ratio because of Fl y stock s. Molecularly characterized protein null mutants for a- and high nonspecific immunoreactivity (high background) and /or reduced b-spectrin were used in this study (Lee et al., 1993; Dubreuil et al., 2000). specific immunoreactivity. Antibody quality effects can be eliminated by rg41 em6 a-spec and b-spec do not produce detectable protein, as shown by normalizing the raw ratios to wild type. Normalized ratios (see Fig. 7) em6 immunoblots (Lee et al., 1993; Dubreuil et al., 2000). b-spec produces were calculated by dividing the ratios for wild type and each mutant by a truncated protein product according to immunoblots (Dubreuil et al., these numbers (normalized r 5 R/R ). WT 2000), but b-spectrin protein is immunohistochemically undetectable in Morpholog y. Quantification of NMJ area was performed using the situ (D. E. Featherstone and K . Brodie, unpublished data), presumably public domain Java-based image processing and analysi s program because the truncated protein is rapidly degraded and /or fails to localize. Image/J . Confocal images (see Fig. 2 A) of wild-type and mutant Drosoph- rg41 a-Spectrin null mutant l(3)dre stocks (Lee et al., 1993) were main- ila NMJs, visualized by fluorescently conjugated anti-HRP (which stains all 67c23 tained as heterozygotes using a thi rd chromosome balancer [yw ; nerve membranes), were manually outlined using Image/J. Once given, rg41 rg41 r u l(3)dre st e/In(3LR)TM3, y1Sb Ser]. Homozygous l(3)dre mu- pixel dimensions (recorded automatically by the Z eiss confocal software), tants are rescued to adulthood by transgenic expression of an a-spectrin Image calculated the area of the outlined region (NMJ area). For bouton minigene under a ubiquitin promoter (Lee et al., 1993). b-Spectrin null counting, synaptic varicosities (swellings) were also visualized with fluores- em21 em6 mutants (b-spec and b-spec ) stocks (Dubreuil et al., 2000) were cently conjugated anti-HRP (1:100; Molecular Probes). maintained as heterozygotes using an FM7[Kruppel-GFP] balancer chro- Electron microscopy. Genotyped embryos were prepared for transmis- mosome (C asso et al., 1999). As with a-spectrin mutants, b-spectrin sion electron microscopy (TEM) using standard techniques (Prokop et mutants are rescued to adulthood by transgenic expression of b-spectrin al., 1996; Fergestad et al., 1999). Briefly, mature embryos (22–24 hr after (Dubreuil et al., 2000). Oregon-R (OR) was used for wild-type (W T) egg laying; AEL at 25°C) were manually dechorionated and injected with controls. fixative (5% glutaraldehyde in 0.05 M phosphate buffer). The preparation Embr yo preparation and dissection. Homozygous mutant embryos were was then transferred to 2.5% glutaraldehyde in 0.05 M phosphate buffer selected from siblings based on the absence of balancer chromosome for 30 – 60 min. Specimens were washed in buffer, transferred to 1% markers (green fluorescent protein for b-spectrin mutants, yellow for osmium tetroxide in dH O for 3 hr, washed again in dH O, and stained 2 2 a-spectrin mutants). Homozygous yellow mutants do not show any signif- en bloc in 2% aqueous uranyl acetate for 30 min. Embryos were dehy- icant difference in excitatory junctional current (EJC) amplitude com- drated in an ethanol series, passed through propylene oxide, and trans- pared with Oregon R ( y/y, 1548 6 280 pA; OR, 1476 6 117 pA; n 5 ferred to araldite. Ribbons of thin (;55 nm) sections were obtained and 5–13; p 5 0.78). For electrophysiology and embryonic immunohisto- examined on a Hitachi (Tokyo, Japan) H-7100 TEM. Active zones that chemistry, morphologically and temporally staged [22–24 hr after egg were identified in at least two consecutive sections were imaged and laying (AEL) at 25°C] embryos were dechorionated with bleach and analyzed using N IH Image. Vesicles were considered to be “clustered” if devitellinated manually. For dissection, embryos were glued (Histoacryl they were within 235 nm of the active zone T-bar (Fergestad et al., 1999) Blue; B. Braun Biotech International GmbH, Melsungen, Germany) to and docked if within one-half vesicle diameter of the presynaptic mem- Sylgard (Dow Corning, Midland, M I)-coated coverslips under saline brane (thus allowing for vesicles that may be in contact with the mem- containing (in mM): 135 NaC l, 5 KC l, 4 MgC l , 1.8 CaCl , 72 sucrose, 2 2 brane but were not perfectly bisected in the cross-section). and 5 N-Tris[hydroxy-methyl]methyl-2-aminoethane sulfonic acid (TES), Electrophysiolog y. Electrophysiology and data analysis were performed pH 7.2. A slit was made manually along the dorsal midline using a glass as described previously (Featherstone et al., 2000). Briefly, whole-cell capillary pulled to a sharp point, and the body walls were glued flat to the patch-clamp recordings from embryonic muscle 6 were obtained in an coverslip. If electrophysiology was to be performed on the dissected extracellular solution containing (in mM): 135 NaC l, 5 KC l, 4 MgC l , 1.8 embryos, the exposed muscle sheath was enzymatically removed after CaCl , 72 sucrose, and 5 TES, pH 7.2. For miniature EJC (mEJC) dissection using 1–2 min exposure to 1 mg /ml collagenase (type IV; recordings, calcium was replaced with 5 mM tetrodotoxin (TTX). The Sigma, St. L ouis, MO). patch pipette solution contained (in mM): 120 KC l, 20 KOH, 4 MgC l , Immunohistochemistr y. Dissected embryos or wandering third-instar 0.25 C aC l , 5 EGTA, 4Na ATP, 36 sucrose, and 5 TES. For EJC 2 2 larvae were fixed in 4% paraformaldehyde for 30 – 45 min and processed measurements, the segmental nerve was stimulated by delivering 5–10 V, according to standard techniques (White, 1998; Beumer et al., 1999; 0.1 msec pulses via a glass suction pipette. To assay the glutamate Featherstone et al., 2000). Mouse monoclonal Drosophila a-spectrin receptor field, 1 mM glutamate was pressure ejected (100 msec pulse) antibody (3A9) (Dubreuil et al., 1997) and rabbit polyclonal Drosophila from a small-tipped (;5 mm opening) pipette directly onto the NMJ. b-spectrin antibody (Byers et al., 1989) were used at 1:100. These a- and Data were analyzed using C lampfit 8 or 9a (Axon Instruments, Foster b-spectrin antibodies show no detectable staining in a-or b-spectrin null C ity, CA) and /or Minianalysis 4 (Synaptosoft Inc., Decatur, GA). mutants and /or on immunoblots, confirming antibody specificity. Mouse Statistics. All data are presented as mean 6 SEM. Each n represents a monoclonal synapsin antibody (K lagges et al., 1996) was used at 1:100. different embryo of the stated genotype. Statistics from spontaneous Mouse monoclonal cysteine string protein (C SP) antibody (Z insmaier et EJC s (sEJC s)and mEJC s are derived from at least 5 min of continuous al., 1994) was used at 1:200. Rabbit polyclonal synaptotagmin antibody recording (often 10 –20 min in the case of the low-frequency mEJC s). In (Littleton et al., 1993) was used at 1:500. Mouse monoclonal syntaxin 1A all figures, statistical significance (compared with wild-type controls) is (Schulze et al., 1995) was used at 1:500. Rabbit polyclonal Discs large indicated as *p , 0.05, **p , 0.01, and ***p , 0.001, . Unless otherwise (DLG) (Lahey et al., 1994) was used at 1:1000. Immunoreactivity for stated, statistical significance was determined using Student’s t test. all of these antibodies is abolished in the appropriate null mutants, con- Because spontaneous synaptic event amplitude distributions are skewed firming antibody specificit y. F luorescein i sothioc yanate and tetra- rather than Gaussian, we compared these distributions statistically using methylrhodamine-conjugated secondary antibodies (goat anti-mouse and the Kolmogorov– Smirnov test and do not report variance. goat anti-rabbit; Molecular Probes, Eugene, OR) were used at 1:400. FI TC -conjugated anti-HRP (Molecular Probes) was used at 1:100. Im- RESULTS ages were obtained on a Z eiss (Oberkochen, Germany) L SM510 laser- scanning confocal microscope. Drosophila spectrins and spectrin mutants Synaptic/nonsynaptic immunoreactivit y ratios. Pixel intensity (0 –255) A search of the sequenced Drosophila genome reveals only three for boutons and nearby extrasynaptic regions (muscle for DLG and nerve for all other proteins) was measured in Z eiss Image Browser software members of the highly conserved spectrin family, each encoded by using raw (completely unaltered) confocal fluorescent images. Average a single gene (Adams et al., 2000; Pinder and Baines, 2000). background fluorescence intensity (dark areas beyond–in between mus- Drosophila a-spectrin (GenBank accession number A33733) is cles) was subtracted from these values. To derive the “synaptic /nonsyn- 64% identical at the amino acid level to human brain a-spectrin/ aptic immunoreactivity ratio,” the background-corrected synaptic fluo- fodrin (GenBank accession number A35715). Drosophila b-spectrin rescence intensity was divided by the background-corrected nonsynaptic fluorescence intensity. Thus, the ratio was calculated as follows: r 5 (S 2 (GenBank accession number A46147) is 56% identical to human B)/(N 2 B), where r is the synaptic /nonsynaptic immunoreactivity ratio, b-spectrin (GenBank accession number N P003119). Drosophila S is the fluorescence intensity in synaptic boutons, B is the background b -spectrin/k arst (GenBank accession number CAA37939) is the fluorescence intensity measured from dark nontissue parts of the image, most divergent, with 31% amino acid identity to the human and N is the fluorescence intensity in nonsynaptic tissues (nerve for C SP, synapsin, syntaxin, and synaptotagmin; muscle for DLG). This raw ratio ortholog b V-spectrin (GenBank accession number AAF65317). Featherstone et al. • Spectrins and Neurotransmitter Release J. Neurosci., June 15, 2001, 21(12):4215–4224 4217 Figure 1. a- and b-Spectrin immunoreactiv- ity in the neuromusculature of Drosophila third-instar larvae. A, Confocal fluorescence images of NMJs stained simultaneously with antibodies raised against a-spectrin and b-spectrin. a-Spectrin immunoreacti v it y i s show n in g reen (lef t), b-spectrin immuno- reacti v it y i s show n in red,(middle), and overlapping a/b-spectrin staining appears yellow (merged image, right). Scale bar, 10 mm. B, C onfocal fluorescence images of N MJs stained simultaneously w ith antibod- ies rai sed against C SP and b-spectrin. C SP immunoreacti v it y i s show n in g reen (lef t), and b-spectrin immunoreacti v it y i s show n in red (middle); merged image i s on the right. Scale bar, 10 mm. Drosophila protein null mutants for a- and b-spectrin are embry- b-spectrin and C SP staining is not colocalized, suggesting that onic – early larval lethal, with defects in the structure and f unction the majority of spectrin protein is associated with the periphery of epithelial cells (Lee et al., 1993, 1997; Dubreuil and Grushko, of the presynaptic membrane and /or dense membrane foldings of 1998; Dubreuil et al., 2000). In contrast, null mutants for b - the postsynaptic subsynaptic reticulum. We conclude from this spectrin are semiviable, with mild defects including rough eyes, immunohistochemistry that both a- and b-spectrin are present at disrupted epithelial morphogenesis, tracheal defects, and mis- the wild-type Drosophila NMJ, in both presynaptic and postsyn- shapen wings (Thomas et al., 1998). These results suggest that a- aptic cells. In subsequent experiments, we used the NMJ as a and /or b-spectrin could play vital roles in synaptogenesis and model synapse to examine the role of spectrins in synaptogenesis synaptic f unction, whereas b -spectrin is unessential. Therefore, H and synaptic f unction. we focused our efforts on characterizing the role of a- and b-spectrin subunits in synaptic development and f unction. a- and b-Spectrin null mutants have morphologically For this study, we used previously identified protein null mu- normal neuromuscular junctions tants for a-spectrin (Lee et al., 1993) and b-spectrin (Dubreuil et We examined gross morphology in protein null mutants of both al., 2000). Homozygous a-spectrin null mutants fail to hatch a- and b-spectrin (Lee et al., 1993; Dubreuil et al., 2000). Light (;50%) or die as early first-instar larvae (;50%). Homozygous microscope (4003) examination of several dozen acutely dis- b-spectrin protein null mutants fail to hatch (;90%), and the rest sected mutant embryos reveals that both a- and b-spectrin mu- (;10%) die as early first-instar larvae. Both classes of mutants are tants have normally formed neuromusculature, epidermis, and lethargic and display limited movement, consistent with a neuro- epidermal specializations (e.g., denticles and mouth parts). The physiological or muscular defect. We chose to study these mutants only visible difference is that a-spectrin mutants have slightly at the embryonic NMJ for several reasons. First, this synapse is thinner muscles, and unhatched (but living) a-spectrin embryos accessible in vivo to a variety of cell biological techniques, includ- often have uninflated trachea at normal hatch time (22–24 hr ing patch-clamp electrophysiology, immunohistochemistry, and AEL). The gut phenotype of these mutants has been described electron microscopy. Second, the development, morphology, and previously (Lee et al., 1993; Dubreuil et al., 1998, 2000). f unction of the Drosophila NMJ is well described and relatively To examine NMJ anatomy, we visualized embryonic body wall invariant from animal to animal. Like many synapses in the neuroanatomy with fluorescently labeled anti-HRP, which recog- mammalian C NS, the Drosophila NMJ is glutamatergic. These nizes neural membranes (Fig. 2). We saw no qualitative differ- features make the NMJ an excellent place to detect and quantif y ences in sites of muscle innervation or presynaptic branching any changes resulting from spectrin disruption. pattern. In Figure 2 A, we show confocal fluorescent images of wild-type and mutant embryonic NMJs visualized with fluores- Spectrins are present at the Drosophila NMJ cently conjugated anti-HRP. In each panel, four individual NMJs Using antibodies specific for Drosophila spectrins (Byers et al., are shown. On the lef t is the linear NMJ lying between ventral 1989; Dubreuil et al., 1997), we examined the neuromuscular longitudinal muscles 6 and 7, and on the right are the more lateral localization of both a- and b-spectrin (Fig. 1). As shown in Figure NMJs on muscles 13 and 12. Quantification of morphology at the 1, both a-spectrin (Fig. 1 A, green) and b-spectrin (Fig. 1 A, red) muscle 6/7 NMJ showed that there was no significant difference in are found in presynaptic axons proximal to the NMJ and in the the number of synaptic boutons (Fig. 2 B)(WT,9.9 6 0.6 boutons; periphery of presynaptic boutons. Although a-spectrin staining is rg41 em21 em6 a , 8.4 6 0.6 boutons; b , 9.6 6 0.8 boutons; b , 8.6 6 0.9 typically weaker, most of the a- and b-spectrin staining in the boutons; n 5 6 –10). Because embryonic boutons are often indis- NMJ appears colocalized (Fig. 1 A, right panel, a- and b-spectrin tinct (Fig. 2 A) and therefore difficult to count, we also quantified overlapping expression appears yellow). a- and b-Spectrin immu- NMJ size by measuring muscle 6/7 NMJ area (see Materials and noreactivity is also strong throughout muscle (Fig. 1 A). We independently confirmed the specificity of both a- and b-spectrin Methods). We detected no significant difference in NMJ area antibodies in null mutant backgrounds (see Materials and between wild-type and a-or b-spectrin mutants (Fig. 2C)(WT, 2 rg41 2 em21 2 Methods). 47.0 6 1.9 mm ; a , 44.8 6 4.9 mm ; b , 48.1 6 2.6 mm ; em6 2 In Figure 1 B, we show double-labeling with antibodies against b , 49.1 6 1.9 mm ; n 5 6 –10). We conclude from this quan- b-spectrin and the presynaptic protein C SP. Much of the tification, as well as qualitative observation of several dozen 4218 J. Neurosci., June 15, 2001, 21(12):4215–4224 Featherstone et al. • Spectrins and Neurotransmitter Release Figure 2. Morphology of embryonic NMJs in spectrin mutants is normal. A, Confocal fluores- cence images of NMJs on ventral longitudinal muscles 6/7 and 12 and 13 in a single ventral hemisegment of wild-type and spectrin mutant embryos. NMJ morphology was visualized by staining with fluorescently conjugated anti-HRP antibodies. Scale bar, 10 mm. B, Quantification of synaptic bouton number at the muscle 6/7 NMJ shows no significant difference between the ge- notypes. C, Quantification of muscle 6/7 NMJ area shows no significant difference between the genotypes. embryos, that NMJ morphology is not detectably altered in either Presy naptic defects can be a result of fault y neurotransmitter a-or b-spectrin mutants. release (sy naptic vesicle filling and f usion), vesicle rec ycling defects, or both. Defecti ve sy naptic vesicle c ycling can be Spectrin mutants are defective in revealed when the ner ve i s stimulated at high frequencies neurotransmitter release (Fergestad et al., 1999; Kuromi and K idokoro, 2000). Under conditions of high demand, neurotransmitter release i s re- Because spectrins are present at the NMJ and the morphology of duced because of a reduction in the available pool of spectrin mutant NMJs was normal, we were able to test whether neurotransmitter-filled vesicles (Kuromi and K idokoro, 2000). synaptic f unction was disrupted. To record NMJ f unction, we In a- and b-spectrin mutants, the reduction in sy naptic trans- voltage clamped (260 mV) muscle 6 using standard patch-clamp mi ssion during high-frequenc y stimulation i s slightly, but not techniques. To evoke synaptic activity, we stimulated (0.5 msec, significantly, impai red (Fig. 3C) (normali z ed amplitude at 20 5–15 V) the presynaptic segmental nerve using a suction elec- rg41 em21 H z: W T, 0.59 6 0.06; a , 0.62 6 0.03; b , 0.39 6 0.12; trode. As shown in Figure 3A, evoked EJC s in both a- and em6 b , 0.51 6 0.07; n 5 4 –7). T hese results suggest that short- b-spectrin mutants are reduced to approximately one-quarter term vesicle c ycling in the mutants i s sufficient to maintain the rg41 normal amplitude (W T, 1476 6 117 pA; a , 473 6 92 pA; em6 em21 reduced rate of exoc y tosi s show n in Figure 3A. Because we did b , 453 6 57 pA; b , 334 6 55 pA; n 5 9 –13; p , 0.001 vs not assay endoc y tosi s in the mutants di rectly (e.g., w ith FM1– W T for each allele, using Student’s t test). a- and b-Spectrin 43), we cannot completely rule out defects in endoc y tosi s. mutant EJC amplitudes are statistically indistinguishable from However, because of the relati vely small (and stati stically each other (Fig. 3A). These results demonstrate that, despite insignificant) alteration in mutant responses to high-frequenc y normal morphology, spectrin mutants have severely reduced syn- stimulation, we conclude that the f unctional defect i s primarily aptic transmission. in exoc y tosi s rather than endoc y tosi s or vesicle c ycling. To determine whether the transmission defect in the spectrin Together, the data in Figure 3 suggest that the synaptic trans- mutants was presynaptic or postsynaptic, glutamate (1 mM) was mission defect in a- and b-spectrin mutants is attributable to pressure ejected (5–10 mm tip pipette, 100 msec pulse) directly specific disruption in neurotransmitter release, with no f unctional onto the NMJ of voltage-clamped (260 mV) postsynaptic muscle alteration in the postsynaptic receptors. We confirmed these 6 (Featherstone et al., 2000). If the defect in synaptic transmission conclusions using analysis of spontaneous synaptic currents (Fig. is attributable to an alteration in postsynaptic glutamate receptor 4). A reduction in the probability of presynaptic vesicle f usion is f unction, the resulting glutamate-gated currents should be re- revealed by less frequent spontaneous synaptic events, whereas an duced in spectrin mutants. As shown in Figure 3B, neither a- nor alteration in receptor localization, receptor number, or receptor b-spectrin mutants showed any detectable alteration in glutamate rg41 em6 response (W T, 1805 6 248 pA; a , 1763 6 227 pA; b , biophysics causes changes in the amplitude of spontaneous syn- em21 1755 6 178 pA; b , 1630 6 105 pA; n 5 7–11). Because the aptic events. Figure 4 shows analysis of sEJC s, which are recorded receptor field is f unctionally normal in spectrin mutants yet in the presence of calcium (Fig. 4 A, B, lef t column), and mEJC s, transmission is greatly reduced, the striking transmission defect which are recorded in the absence of extracellular calcium and shown in Figure 3A must be presynaptic. the presence of TTX (Fig. 4 A, B, right column). The frequency of Featherstone et al. • Spectrins and Neurotransmitter Release J. Neurosci., June 15, 2001, 21(12):4215–4224 4219 Figure 3. Patch-clamp electrophysiology from voltage-clamped (260 mV) muscle demonstrates that spectrin mutant NMJs have severely reduced neurotransmitter release, with no f unctional alteration of Figure 4. Analysis of spontaneous synaptic events demonstrates that postsynaptic receptor fields. A, EJC amplitude (evoked by nerve stimula- spectrin null mutants have decreased synaptic vesicle f usion rates but no tion) is significantly reduced in a- and b-spectrin mutants. Representative f unctional alteration of the receptor fields. Currents were recorded in EJC s are shown on the right. B, Currents triggered by pressure ejection of both normal (1.8 mM) calcium saline and saline containing 0 mM calcium 1mM glutamate (100 msec pulse) onto the postsynaptic membrane dem- plus 5 mM TTX (to block endogenous nerve activity). A, Frequency of onstrate that the spectrin mutant glutamate receptor field f unction is not spontaneous synaptic currents in voltage-clamped (260 mV) muscle in significantly different from wild type. Representative glutamate-gated both a-spectrin and b-spectrin mutants is reduced in both high- and currents are shown on the right. C, Stimulation of the NMJ at increasing low-calcium conditions, suggesting disruption of presynaptic vesicle f u- frequencies reveals no significant difference in decrement of EJC ampli- sion. B, Amplitude histograms (composed of data from multiple record- tude between mutants and control. ings) reveal no significant difference (Kolmogorov– Smirnov test) in spec- trin mutant event amplitudes in either normal (1.8 mM) calcium or the both types of event are lowered in the spectrin mutants (Fig. 4 A), absence of endogenous activity (0 calcium plus TTX), suggesting that the spectrin mutants have no f unctional alteration in the postsynaptic gluta- suggesting that a- and b-spectrin mutants share a calcium- mate receptor field. independent deficit in synaptic vesicle f usion (1.8 mM Ca :WT, rg41 em21 11.69 6 1.59 Hz; a , 6.83 6 0.74 Hz; b , 1.69 6 0.41 Hz; em6 21 b , 3.52 6 1.51 Hz; n 5 6 –11; TTX plus 0 mM Ca :WT, rg41 em21 vivo or on immunoblot, possibly because of the small number of 0.13 6 0.03 Hz; a , 0.08 6 0.02 Hz; b , 0.05 6 0.01 Hz; em6 embryonic receptors (100 –200 receptors per NMJ vs tens of b , 0.07 6 0.02 Hz; n 5 5–14). In contrast, the amplitude of thousands of receptors per NMJ in larvae). However, electro- mEJC s is not significantly altered in spectrin mutants compared physiology is arguably the most sensitive (able to detect a single with wild-type controls (Fig. 4 B). These data, like those in Figure f unctional receptor) and most quantitative method of determin- 3B, suggest that the postsynaptic receptor field is f unctionally rg41 ing receptor field integrity. unchanged (mean mEJC amplitudes: W T, 157.9 pA; a , 143.7 em21 em6 Together, the electrophysiological results show that a- and pA; b , 142.6 pA; b , 140.3 pA; n 5 4 –14 embryos, thou- sands of events; p . 0.05 by Kolmogorov– Smirnov test). We are b-spectrin mutants have severely impaired synaptic transmission unable to confirm this finding qualitatively using antibodies raised and that this impairment is attributable specifically to disruption against Drosophila glutamate receptors. Despite success in larvae, of neurotransmitter release, without any f unctional alteration in we have been unable to visualize embryonic receptors either in the postsynaptic receptor field. 4220 J. Neurosci., June 15, 2001, 21(12):4215–4224 Featherstone et al. • Spectrins and Neurotransmitter Release Figure 5. Ultrastructural analysis of embryonic NMJs shows morphologically normal boutons in spectrin mutants, with no alterations in the distribution of active zones or synaptic vesicles. A, TEM cross-sections through embryonic NMJ boutons showing presynaptic active zones with electron-dense T-bars (surrounded by clustered vesicles) in opposition to a postsynaptic density. Active zones are indicated with arrowheads. Scale bar, 250 nm. B, High-magnification images of active zones from each genotype, showing individual T-bars and clustered vesicles. Arrowheads indicate T-bars. C, Quantification of numbers of docked vesicles (within one-half vesicle diameter of the presynaptic membrane), numbers of clustered vesicles (within 235 nm of T-bar), and vesicle density throughout bouton cross-section. [within one-half vesicle diameter of the presynaptic membrane Ultrastructure of spectrin mutants is normal (Broadie et al., 1995)] or clustered [within 235 nm of T-bar Both a- and b-spectrin mutants have normal NMJ morphology rg41 (Fergestad et al., 1999)] vesicles (docked: W T, 1.74 6 0.16; a , but reduced neurotransmitter release, supporting the idea that em21 em6 1.58 6 0.17; b , 1.79 6 0.22; b , 1.42 6 0.76; n 5 19 –27; spectrins may cluster synaptic vesicles at the active zone. This rg41 em21 clustered: W T, 21.3 6 1.1; a , 22.1 6 1.31; b , 19.7 6 1.23; hypothesis, called “casting the line,” suggests that one end of em6 b , 17.95 6 0.73; n 5 19 –27). Similarly, when synaptic vesicle spectrin is anchored to active zones, whereas the other end density throughout the entire bouton cross-section is quantified, captures vesicles via an interaction with synapsin (Landis, 1988; both a- and b-spectrin null mutants are comparable with wild type Goodman et al., 1995; Sikorski et al., 2000). Mislocalization rg41 em21 (vesicle density: W T, 74.4 6 6.9; a , 69.4 6 8.3; b , 96.6 6 and /or absence of synaptic vesicles at the active zone could em6 em6 22.9; b , 49.7 6 9.1; n 5 19 –27). b-spectrin shows a slight explain the spectrin mutant electrophysiological phenotype we (but statistically significant) reduction in clustered vesicles and show in Figures 3 and 4. We tested whether synaptic vesicle vesicle density, but this change is unlikely to explain the synaptic clustering is disrupted in spectrin mutants by examining NMJs transmission defect for two reasons: (1) the change is too small to using electron microscopy (Fig. 5). In a- and b-spectrin mutants, explain the severe decrease in vesicle release, and (2) the change presynaptic and postsynaptic membranes are normally structured rg41 em21 is not shared by either a-spectrin or b-spectrin , which and spaced, internal organelles appear normal, and the distribu- otherwise have identical phenotypes. We conclude that spectrins tion of embryonic T-bars and electron-dense areas associated with do not play a substantial role in synaptic vesicle tethering at active active zones are indistinguishable from wild type (Fig. 5 A, B). zones. Thus, spectrins play no detectable role in the maintenance of gross synaptic morphology. The location of active zones and synaptic vesicles are readily Synaptic protein localization is disrupted in both a- and b-spectrin mutants visible, allowing us to determine whether vesicle clustering is altered in either a-or b-spectrin mutants. We quantified the NMJ morphology in the spectrin mutants is normal by light and number and distribution of synaptic vesicles around each active electron microscopy, yet neurotransmitter release is severely dis- zone, and these results are graphed in Figure 5C. Spectrin mu- rupted. In other (non-neuronal) cell types, spectrins have been tants show no consistent alteration in the number of docked proposed to capture and maintain proteins in distinct membrane- Featherstone et al. • Spectrins and Neurotransmitter Release J. Neurosci., June 15, 2001, 21(12):4215–4224 4221 associated domains, especially at sites of cell – cell interaction (Drubin and Nelson, 1996; Pinder and Baines, 2000). At synapses, proper f unction requires precise assembly and alignment of the molecular machinery required for synaptic vesicle f usion and recycling. If this machinery is mislocalized or incorrectly assem- bled, it would not be surprising to find a synaptic defect such as we observe in a- and b-spectrin mutants. Although there is no method by which we can test whether the in vivo submicrometer assembly of proteins is appropriate in spectrin mutants, we can determine whether synaptic proteins are polarized and properly localized to the NMJ. In epithelial cells, disruption of protein polarization attributable to the absence of spectrin is visible by immunohistochemistry and confocal light microscopy (Dubreuil et al., 2000). We used the same techniques to determine whether spectrins play a similar role in protein compartmentalization at synapses. Figure 6 shows representative staining in wild-type and spec- trin mutant embryos for two of the best Drosophila NMJ markers available: presynaptic anti-C SP and postsynaptic anti-DLG. C SP is present in both vesicular membrane-associated and cytosolic fractions of presynaptic boutons; C SP staining normally appears as tightly localized presynaptic puncta (Z insmaier et al., 1994). DLG is a plasma membrane-associated PDZ [postsynaptic den- sity-95(PSD-95)/ DLG/zona occludens-1] domain protein with 60% homology to PSD-95 that is tightly localized to both presyn- aptic and postsynaptic membranes (Lahey et al., 1994; Budnik et al., 1996). Each panel in Figure 6 shows the body wall neuromus- culature of two to three embryonic hemisegments stained with anti-C SP ( green) and anti-DLG (red). The (out of focus) ventral ganglion (CNS) is visible in the top lef t of each panel, from which segmental nerves (SN ) extend into the body wall musculature on the right. The C NS serves as a positive control for overall image intensity. In wild-type embryos, C SP and DLG staining in the body wall neuromusculature is restricted to tightly defined puncta at the NMJ (Fig. 6, lef t column); little or no staining is visible in either the presynaptic nerve axon or nonsynaptic muscle mem- brane. Thus, neither the segmental nerves nor the majority of muscle tissue is visible in the fluorescence image (Fig. 6, lef t column). In both a- and b-spectrin mutants (Fig. 6, middle and Figure 6. Fluorescent confocal micrographs of embryonic neuromuscula- right columns), the synaptic localization of both presynaptic C SP ture showing distribution of the presynaptic protein C SP and postsynaptic and postsynaptic DLG is dramatically perturbed. The segmental protein DLG. Each panel shows C SP and DLG immunoreactivity in three nerves are now visible (because of C SP immunoreactivity), as are or more hemisegments. In wild-type embryos, DLG (red) and C SP ( green) the muscles (because of DLG immunoreactivity). We conclude tightly associate with NMJ boutons, which appear as immunoreactive that, in both a- and b-spectrin mutants, C SP is distributed abnor- puncta at NMJs in the body wall neuromusculature (lef t column). Note that, in wild-type embryos, anti-C SP and anti-DLG antibodies detect only the mally throughout presynaptic axons, and DLG is distributed NMJ and not the preterminal axon or extrasynaptic regions of the muscle. abnormally throughout muscle cells. Neither protein appears In both a- and b-spectrin null mutants, however, C SP ( green) is abnormally properly polarized and localized to the NMJ boutons in spectrin distributed throughout distal axons (extending horizontally from the C NS mutants. on the lef t into the musculature on the right). In both a- and b-spectrin null mutants, DLG staining (red) is scattered throughout postsynaptic muscles. In addition to C SP and DLG, we examined the staining pat- In each image, a portion of the ventral ganglion (C NS) is shown (out of terns of several other synaptic proteins, including synaptotagmin, focus) as a positive control. Scale bar, 15 mm. synapsin, and syntaxin. Synaptotagmin is a transmembrane pro- tein normally restricted to synaptic vesicles (Littleton et al., 1993; Marqueze et al., 2000). Synapsin is a spectrin-interacting phos- chemically) by comparing staining intensity in NMJ boutons with phoprotein that is associated with the presynaptic actin cytoskel- staining intensity outside the synapse (see Materials and Meth- eton at synaptic boutons (K lagges et al., 1996; Iga et al., 1997; ods). In wild-type embryos, fluorescence intensity from each Hilfiker et al., 1999; T urner et al., 1999). Syntaxin is a transmem- synaptic marker was significantly higher in NMJ boutons than brane protein normally present in presynaptic membrane, includ- elsewhere. Specifically, the synaptic /nonsynaptic fluorescence in- ing both axons and synaptic boutons (Schulze et al., 1995; Gerst, tensity for each marker (in wild-type embryos) was as follows: 1999). All of these proteins, like C SP and DLG, showed severely 30.75 6 6.85 (C SP), 8.56 6 1.65 (DLG), 5.51 6 1.19 (synapsin), disrupted subcellular localization in both a- and b-spectrin mu- 7.46 6 1.88 (synaptotagmin), and 3.41 6 0.74 (syntaxin). In other tant embryos. words, wild-type embryos showed anti-C SP fluorescence that was We quantified protein distribution (measured immunocyto- 30.75 times higher in boutons than in nerve. In contrast, anti- 4222 J. Neurosci., June 15, 2001, 21(12):4215–4224 Featherstone et al. • Spectrins and Neurotransmitter Release 1989; Masliah et al., 1991; Bewick et al., 1992, 1996; Goodman et al., 1995; Gelot et al., 1996; Bloch et al., 1997; Sakaguchi et al., 1998; Wechsler and Teichberg, 1998; Wood and Slater, 1998; Goodman, 1999; Hirai and Matsuda, 1999; Dunaevsky and Con- nor, 2000; Hammarlund et al., 2000; Kordeli, 2000; Sikorski et al., 2000; Sunderland et al., 2000). The role that spectrins might play at synapses has been the subject of intense speculation. The Drosophila genome contains only one a-spectrin and one conven- tional b-spectrin gene, making it an ideal system to genetically manipulate spectrin levels and examine the resulting synaptic alterations. Using protein null mutants for a- and b-spectrin,we tested whether spectrins are required for development and /or f unction of the Drosophila neuromuscular junction. First, we showed that both a- and b-spectrin are present at the Drosophila NMJ (Fig. 1). This observation supports the synaptic localization of spectrins observed in other systems (Bloch and Morrow, 1989; Daniels, 1990; Masliah et al., 1991; Bewick et al., 1992, 1996; Goodman et al., 1995; Kordeli, 2000). Second, we showed that, in a- and b-spectrin mutants, synaptic morphology is normal (Fig. 2). This result contrasts with the severe morpholog- ical defects observed in Caenorhabditis elegans spectrin mutants Figure 7. Spectrin mutants have mislocalized synaptic proteins. Protein (Hammarlund et al., 2000). The normal morphological develop- distribution was quantified by comparing staining intensity in NMJ bou- ment in Drosophila spectrin mutants may be possible because of a tons with staining intensity outside the synapse (see Materials and Meth- ods). In wild-type embryos, fluorescence intensity from each synaptic maternal contribution. Third, because NMJ morphology was nor- marker was much higher in NMJ boutons than elsewhere. In a- and mal, we were able to undertake a detailed electrophysiological b-spectrin mutants, however, immunoreactivity of C SP, DLG, synapto- analysis of synaptic f unction in spectrin mutants (Figs. 3, 4). This tagmin, synapsin, and syntaxin were all reduced in boutons and simulta- analysis showed that both a- and b-spectrin mutants have equal neously increased in nonsynaptic nerve (C SP, synaptotagmin, synapsin, and severe disruptions in synaptic transmission. Using pressure- and syntaxin) or muscle (DLG) membrane. Thus, the relative amount of synaptic protein at the synapse relative to other tissues was significantly ejected glutamate to directly measure postsynaptic glutamate reduced in spectrin mutants. receptor f unction, we were able to rule out the possibility that the transmission defect was attributable to any alteration in glutamate receptor f unction. This conclusion was confirmed using analysis syntaxin fluorescence in wild-type embryos was only 3.41 times of spontaneous synaptic events, which showed reduced probabil- higher in boutons than in nerve. These observations are consis- ity of vesicle f usion yet normal event amplitudes. Thus, we con- tent with the fact that C SP is strongly restricted to synaptic cluded that the source of the synaptic f unction defect in spectrin boutons, whereas syntaxin is present throughout the neuronal mutants was presynaptic. Based on immunohistochemical local- membrane and only weakly polarized to boutons (Schulze et al., ization and biochemistry, spectrins have been proposed to play an 1995). This raw ratio represents a measure of both protein local- important role in development and /or f unction of postsynaptic ization and antibody quality because poor antibodies might be receptor fields (Bloch and Morrow, 1989; Daniels, 1990; Bloch et expected to lower the ratio because of high nonspecific immuno- al., 1997; Wechsler and Teichberg, 1998; Hirai and Matsuda, reactivity (high background) and /or reduced specific immuno- 1999). Our data strongly suggest that this is not true at the reactivity. Drosophila NMJ, although we cannot rule out the possibility that For ease of comparison and to eliminate effects on the ratios maternal spectrin contributes to the initial development of the from antibody quality, we normalized all of these ratios to wild postsynaptic receptor field. type. Normalized ratios for wild-type and spectrin mutant em- What is the cause of the presynaptic defect in spectrin mutants? bryos are shown in Figure 7. The synaptic /nonsynaptic immuno- Spectrins have been proposed to capture and tether (via synapsin reactivity ratios in both a- and b-spectrin mutants for C SP, DLG, I) synaptic vesicles near the active zone (Landis, 1988; Goodman synaptotagmin, synapsin, and syntaxin were each significantly et al., 1995; Sikorski et al., 2000). In support of this hypothesis, it reduced compared with wild type ( p , 0.05; t test). The normal- has been shown that disruption of spectrin – synapsin binding, via ized ratios (W T is 1) for a-spectrin mutants were as follows: antibodies raised against the synapsin binding site of spectrin, 0.04 6 0.004 (C SP), 0.14 6 0.017 (DLG), 0.21 6 0.036 (synapsin), eliminate synaptic transmission in cultured hippocampal cells 0.36 6 0.782 (syntaxin), and 0.47 6 0.060 (synaptotagmin) (n 5 (Sikorski et al., 2000). However, synaptic vesicle localization was 4 –13; mean of 10). The normalized ratios (W T is 1) for b-spectrin never examined in that study. To test whether the defective mutants were as follows: 0.03 6 0.003 (C SP), 0.14 6 0.017 (DLG), neurotransmitter release in Drosophila spectrin mutants was at- 0.23 6 0.051 (synapsin), 0.25 6 0.065 (syntaxin), and 0.12 6 0.006 tributable to altered synaptic vesicle localization, we examined (synaptotagmin) (n 5 4 –13; mean of 10). We conclude from these the ultrastructure of wild-type and spectrin mutant synaptic ter- results that synaptic proteins are improperly polarized and local- minals using electron microscopy (Fig. 5). We found no changes ized in both a- and b-spectrin mutants. in synaptic vesicle distribution in the spectrin mutants. This DISCUSSION observation is in agreement with recent data from C. elegans Spectrins have been known for over a decade to be present at b-spectrin mutants (Hammarlund et al., 2000). Thus, data from both central and peripheral synapses in a variety of organisms both Drosophila and C. elegans spectrin mutants argue that spec- (Lazarides et al., 1984; Bloch and Morrow, 1989; Goodman et al., trins are not required for synaptic vesicle clustering or docking. Featherstone et al. • Spectrins and Neurotransmitter Release J. Neurosci., June 15, 2001, 21(12):4215–4224 4223 Different distributions of dystrophin and related proteins at nerve- We are unable to visualize, in either wild type or mutants, any muscle junctions. NeuroReport 3:857– 860. electron-dense “rods” connecting synaptic vesicles to the active Bewick GS, Young C, Slater CR (1996) Spatial relationships of utrophin, zone. These rods, which are visible in some other preparations, dystrophin, beta-dystroglycan and beta-spectrin to acetylcholine recep- tor clusters during postnatal maturation of the rat neuromuscular have been suggested to be spectrin based on their size (Landis, junction. J Neurocytol 25:367–379. 1988; Goodman et al., 1995). Bloch RJ, Morrow JS (1989) An unusual beta-spectrin associated with In epithelial cells, spectrins are required for polarization and clustered acetylcholine receptors. J C ell Biol 108:481– 493. Bloch RJ, Bezakova G, Ursitti JA, Z hou D, Pumplin DW (1997) A localization of a variety of membrane-associated proteins, espe- membrane skeleton that clusters nicotinic acetylcholine receptors in cially at sites of cell – cell contact (Bennett, 1990; Bennett and muscle. Soc Gen Physiol Ser 52:177–195. Broadie K , Prokop A, Bellen HJ, O’Kane C J, Schulze K L, Sweeney ST Gilligan, 1993; Drubin and Nelson, 1996; Brown and Breton, (1995) Syntaxin and synaptobrevin f unction downstream of vesicle 2000; Dubreuil et al., 2000; Pinder and Baines, 2000). Because docking in Drosophila. Neuron 15:663– 673. synapses are highly polarized sites of cell – cell interaction be- Brown D, Breton S (2000) Sorting proteins to their target membranes. K idney Int 57:816 – 824. tween ectodermally derived cells, it stands to reason that the Budnik V, Koh YH, Guan B, Hartmann B, Hough C, Woods D, Gorczyca f unction of neuronal spectrin might be similar to that shown in M (1996) Regulation of synapse structure and f unction by the Dro- epithelia. We used methods similar to those used in studies of sophila tumor suppressor gene dlg. Neuron 17:627– 640. Byers TJ, Husain-Chishti A, Dubreuil RR, Branton D, Goldstein L S epithelia to show that indeed this is the case; several classes of (1989) Sequence similarity of the amino-terminal domain of Drosoph- synaptic proteins fail to properly polarize and localize in spectrin ila beta spectrin to alpha actinin and dystrophin. J C ell Biol 109:1633–1641. mutants (Figs. 6, 7). Because both a- and b-spectrin are distrib- C asso D, Ramirez-Weber FA, Kornberg TB (1999) GFP-tagged bal- uted widely, spectrin alone cannot be sufficient for organization ancer chromosomes for Drosophila melanogaster. Mech Dev of f unctional synaptic domains. Synaptic spectrins must be “acti- 88:229 –232. Daniels M P (1990) L ocalization of actin, beta-spectrin, 43 x 10(3) Mr vated” via a local synaptic cue or work in conjunction with other and 58 x 10(3) Mr proteins to receptor-enriched domains of newly molecules to capture and accumulate synaptic proteins. In this formed acetylcholine receptor aggregates in isolated myotube mem- regard, neuronal spectrin appears to be different from epithelial branes. J C ell Sci 97:615– 626. Davis J, Bennett V (1983) Brain spectrin. Isolation of subunits and spectrin, which has a polarized distribution that matches its site of formation of hybrids with erythrocyte spectrin subunits. J Biol Chem activity precisely (Dubreuil et al., 2000). Because neurotransmit- 258:7757–7766. Drubin DG, Nelson WJ (1996) Origins of cell polarity. C ell 84:335–344. ter release requires precise organization of presynaptic protein Dubreuil RR, Grushko T (1998) Genetic studies of spectrin: new life for machinery, it is not unreasonable to conclude that the defects in a ghost protein. BioEssays 20:875– 878. synaptic release measured in Drosophila mutants are attributable Dubreuil RR, Brandin E, Reisberg JH, Goldstein L S, Branton D (1991) Structure, calmodulin-binding, and calcium-binding properties of re- to alterations in synaptic protein localization. However, we can- combinant alpha spectrin polypeptides. J Biol Chem 266:7189 –7193. not rule out another, less likely, direct role for spectrin in synaptic Dubreuil RR, Maddux PB, Grushko TA, MacVicar GR (1997) Segrega- vesicle f usion, as has been proposed by Goodman (1999). tion of two spectrin isoforms: polarized membrane-binding sites direct polarized membrane skeleton assembly. Mol Biol C ell 8:1933–1942. In summary, we have shown spectrins are, as in other organ- Dubreuil RR, Frankel J, Wang P, Howrylak J, Kappil M, Grushko TA isms, present in Drosophila synapses. Electrophysiological analy- (1998) Mutations of alpha spectrin and labial block cuprophilic cell differentiation and acid secretion in the middle midgut of Drosophila ses showed that neurotransmitter release in Drosophila a- and larvae. Dev Biol 194:1–11. b-spectrin protein null mutants is severely impaired. However, Dubreuil RR, Wang P, Dahl S, Lee J, Goldstein L S (2000) Drosophila contrary to current models, this synaptic impairment is not at- beta spectrin f unctions independently of alpha spectrin to polarize the Na,K ATPase in epithelial cells. J C ell Biol 149:647– 656. tributable to defects in receptor field f unction or synaptic vesicle Dunaevsky A, Connor EA (2000) F-Actin is concentrated in nonrelease localization. We conclude, based on immunolocalization of sev- domains at frog neuromuscular junctions. J Neurosci 20:6007– 6012. eral classes of synaptic proteins, that proper polarization and Featherstone DE, Rushton EM, Hilderbrand-Chae M, Phillips AM, Jack- son FR, Broadie K (2000) Presynaptic glutamic acid decarboxylase is localization of synaptic proteins does not take place in the ab- required for induction of the postsynaptic receptor field at a glutama- sence of spectrin. We propose, based on these results and the tergic synapse. Neuron 27:71– 84. Fergestad T, Davis WS, Broadie K (1999) The stoned proteins regulate synaptic localization of spectrin, that a spectrin-based scaffold is synaptic vesicle recycling in the presynaptic terminal. J Neurosci formed early in synaptic development, and this scaffold is subse- 19:5847–5860. quently required for proper assembly, transport, or localization of Gelot A, Moreau J, Ben Ari Y, Pollard H (1996) Alpha-brain spectrin mRNA belongs to the population of intradendritically transported synaptic proteins during development. Future work will aim to mRNAs. NeuroReport 8:113–116. understand the time course and mechanisms by which synaptic Gerst JE (1999) SNAREs and SNARE regulators in membrane f usion spectrin is specifically activated and /or localized to capture and and exocytosis. C ell Mol Life Sci 55:707–734. Goodman SR (1999) Discovery of nonerythroid spectrin to the demon- accumulate synaptic proteins. stration of its key role in synaptic transmission. Brain Res Bull 50:345–346. Goodman SR, L opresti L L, Riederer BM, Sikorski A, Z agon IS (1989) REFERENCES Brain spectrin(240/235A): a novel astrocyte specific spectrin isoform. Adams MD, C elniker SE, Holt R A, Evans CA, Gocayne JD, Amanatides Brain Res Bull 23:311–316. PG, Scherer SE, Li PW, Hoskins R A, Galle RF, George R A, Lewis Goodman SR, Z immer W E, C lark MB, Z agon IS, Barker JE, Bloom ML SE, Richards S, Ashburner M, Henderson SN, Sutton GG, Wortman (1995) Brain spectrin: of mice and men. Brain Res Bull 36:593– 606. JR, Yandell MD, Z hang Q, Chen L X, Brandon RC, Rogers YH, Hammarlund M, Davis WS, Jorgensen EM (2000) Mutations in beta- Blazej RG, Champe M, Pfeiffer BD et al. (2000) The genome se- spectrin disrupt axon outgrowth and sarcomere structure. J C ell Biol quence of Drosophila melanogaster. Science 287:2185–2195. 149:931–942. Bennett V (1990) Spectrin: a structural mediator between diverse Hilfiker S, Pieribone VA, C zernik AJ, Kao HT, Augustine GJ, Greengard plasma membrane proteins and the cytoplasm. Curr Opin C ell Biol P (1999) Synapsins as regulators of neurotransmitter release. Philos 2:51–56. Trans R Soc L ond B Biol Sci 354:269 –279. Bennett V, Gilligan DM (1993) The spectrin-based membrane skeleton Hirai H, Matsuda S (1999) Interaction of the C -terminal domain of delta and micron-scale organization of the plasma membrane. Annu Rev glutamate receptor with spectrin in the dendritic spines of cultured C ell Biol 9:27– 66. Purkinje cells. Neurosci Res 34:281–287. Beumer K J, Rohrbough J, Prokop A, Broadie K (1999) A role for PS Iga M, Inui M, Sobue K (1997) Characterization of the interaction integrins in morphological growth and synaptic f unction at the postem- between synapsin I and calspectin (brain spectrin or fodrin). Biochem bryonic neuromuscular junction of Drosophila. Development Biophys Res Commun 231:852– 855. 126:5833–5846. K lagges BR, Heimbeck G, Godenschwege TA, Hof bauer A, Pflugfelder Bewick GS, Nicholson LV, Young C, O’Donnell E, Slater CR (1992) GO, Reifegerste R, Reisch D, Schaupp M, Buchner S, Buchner E 4224 J. Neurosci., June 15, 2001, 21(12):4215–4224 Featherstone et al. • Spectrins and Neurotransmitter Release (1996) Invertebrate synapsins: a single gene codes for several isoforms Schulze K L, Broadie K , Perin MS, Bellen HJ (1995) Genetic and elec- in Drosophila. J Neurosci 16:3154 –3165. trophysiological studies of Drosophila syntaxin-1A demonstrate its role Kordeli E (2000) The spectrin-based skeleton at the postsynaptic mem- in nonneuronal secretion and neurotransmission. C ell 80:311–320. brane of the neuromuscular junction. Microsc Res Tech 49:101–107. Sikorski AF, Goodman SR (1991) The effect of synapsin I phosphoryla- Kuromi H, K idokoro Y (2000) Tetanic stimulation recruits vesicles from tion upon binding of synaptic vesicles to spectrin. Brain Res Bull reserve pool via a cAM P- mediated process in Drosophila synapses. 27:195–198. Neuron 27:133–143. Sikorski AF, Sangerman J, Goodman SR, Critz SD (2000) Spectrin Lahey T, Gorczyca M, Jia X X, Budnik V (1994) The Drosophila tumor (betaSpIIsigma1) is an essential component of synaptic transmission. suppressor gene dlg is required for normal synaptic bouton structure. Brain Res 852:161–166. Neuron 13:823– 835. Srinivasan Y, Elmer L, Davis J, Bennett V, Angelides K (1988) Ankyrin Landis DM (1988) Membrane and cytoplasmic structure at synaptic and spectrin associate with voltage-dependent sodium channels in junctions in the mammalian central nervous system. J Electron Microsc brain. Nature 333:177–180. Tech 10:129 –151. Steiner JP, Walke Jr HT, Bennett V (1989) C alcium /calmodulin inhibits Lazarides E, Nelson WJ, Kasamatsu T (1984) Segregation of two spec- direct binding of spectrin to synaptosomal membranes. J Biol Chem trin forms in the chicken optic system: a mechanism for establishing 264:2783–2791. restricted membrane-cytoskeletal domains in neurons. C ell Sunderland WJ, Son YJ, Miner JH, Sanes JR, C arlson SS (2000) The 36:269 –278. Lee JK , Coyne RS, Dubreuil RR, Goldstein L S, Branton D (1993) C ell presynaptic calcium channel is part of a transmembrane complex link- shape and interaction defects in alpha-spectrin mutants of Drosophila ing a synaptic laminin (alpha4beta2gamma1) with non-erythroid spec- melanogaster. J C ell Biol 123:1797–1809. trin. J Neurosci 20:1009 –1019. Lee JK , Brandin E, Branton D, Goldstein L S (1997) alpha-Spectrin is Thomas GH, Z arnescu DC, Juedes AE, Bales M A, L ondergan A, Korte required for ovarian follicle monolayer integrity in Drosophila melano- CC, K iehart DP (1998) Drosophila betaHeav y-spectrin is essential for gaster. Development 124:353–362. development and contributes to specific cell fates in the eye. Develop- Littleton J T, Bellen HJ, Perin MS (1993) E xpression of synaptotagmin ment 125:2125–2134. in Drosophila reveals transport and localization of synaptic vesicles to T urner K M, Burgoyne RD, Morgan A (1999) Protein phosphorylation the synapse. Development 118:1077–1088. and the regulation of synaptic membrane traffic. Trends Neurosci Marqueze B, Berton F, Seagar M (2000) Synaptotagmins in membrane 22:459 – 464. traffic: which vesicles do the tagmins tag? Biochimie 82:409 – 420. Wechsler A, Teichberg V I (1998) Brain spectrin binding to the NMDA Masliah E, Hansen L, Mallory M, Albright T, Terry RD (1991) Abnor- receptor is regulated by phosphorylation, calcium and calmodulin. mal brain spectrin immunoreactivity in sprouting neurons in Alzhei- EMBO J 17:3931–3939. mer disease. Neurosci Lett 129:1–5. White R AH (1998) Immunolabeling of Drosophila. In: Drosophila:a Pinder JC, Baines AJ (2000) A protein accumulator. Nature practical approach, Ed 2 (Roberts DB, ed), pp 215–240. Oxford: IRL. 406:253–254. Wood SJ, Slater CR (1998) beta-Spectrin is colocalized with both Prokop A, Landgraf M, Rushton E, Broadie K , Bate M (1996) Presyn- voltage-gated sodium channels and ankyrinG at the adult rat neuro- aptic development at the Drosophila neuromuscular junction: assembly muscular junction. J C ell Biol 140:675– 684. and localization of presynaptic active zones. Neuron 17:617– 626. Sakaguchi G, Orita S, Naito A, Maeda M, Igarashi H, Sasaki T, Takai Y Z insmaier K E, Eberle K K , Buchner E, Walter N, Benzer S (1994) (1998) A novel brain-specific isoform of beta spectrin: isolation and its Paralysis and early death in cysteine string protein mutants of Drosoph- interaction with Munc13. Biochem Biophys Res Commun 248:846 – 851. ila. Science 263:977–980.

Journal

The Journal of Neuroscience – Unpaywall

Published: Jun 15, 2001

There are no references for this article.