THE JOURNAL OF COMPARATIVE NEUROLOGY 408:437–448 (1999) Presynaptic Cytomatrix Protein Bassoon Is Localized at Both Excitatory and Inhibitory Synapses of Rat Brain KARIN RICHTER,1 KRISTINA LANGNAESE,1 MICHAEL R. KREUTZ,1 GISELA OLIAS,2 RONG ZHAI,2 HENNING SCHEICH,1 CRAIG C. GARNER,2 AND ECKART D. GUNDELFINGER1* 1Leibniz Institute for Neurobiology (IfN), D-39118 Magdeburg, Germany 2Department of Neurobiology, University of Alabama at Birmingham, Birmingham, Alabama 35213-0021 ABSTRACT Bassoon is a 420-kDa protein specifically localized at the active zone of presynaptic nerve terminals. It is thought to be involved in the structural organization of the neurotransmitter release site. We studied the distribution of Bassoon transcripts and protein in rat brain and assessed which types of presynaptic terminals contain the protein. As shown by in situ hybridization, Bassoon transcripts are widely distributed in the brain and occur primarily in excitatory neurons. In addition, examples of ␥-aminobutyric acid (GABA)-ergic neurons expressing Bassoon are detected. At the light microscopic level, Bassoon immunoreactivity is found in synaptic neuropil regions throughout the brain, with the strongest expression in the hippocampus, the cerebellar cortex, and the olfactory bulb. Immunoelectron microscopy showed that Bassoon immunoreactivity is found in both asymmetric type 1 and symmetric type 2 synapses. Immunopositive asymmetric synapses include mossy fiber boutons and various spine and shaft synapses in the hippocampus and mossy fiber terminals and parallel fiber terminals in the cerebellum. Bassoon-containing symmetric synapses are observed, e.g., between basket and granule cells in the hippocampus, between Golgi cells and granule cells, and between basket cells and Purkinje cells in the cerebellum. Within synaptic terminals, Bassoon appears highly concentrated at sites opposite to postsynaptic densities. In cultured hippocampal neurons, Bassoon was found to colocalize with GABAA and glutamate (GluR1) receptors. These data indicate that Bassoon is a component of the presynaptic apparatus of both excitatory glutamatergic and inhibitory GABAergic synapses. J. Comp. Neurol. 408:437– 448, 1999. r 1999 Wiley-Liss, Inc. Indexing terms: hippocampus; cerebellum; presynaptic terminal; neurotransmitters; GABA; glutamate Presynaptic nerve terminals constitute the principal site of regulated neurotransmitter release from nerve cells. At the active zone, synaptic vesicles dock, fuse with the presynaptic plasma membrane, and release their transmitter contents in a highly coordinated way; thereafter, vesicles are rapidly recycled in a clathrin-mediated process (for review, see Südhof 1995; De Camilli and Takei 1996; Betz and Angleson, 1998; Geppert and Südhof, 1998). A highly specialized apparatus at the active zone composed of the presynaptic membrane and the underlying cytoskeleton is thought to define the site of transmitter release and to spatially organize individual steps of the synaptic vesicle cycle. At the ultrastructural level, a filamentous cytomatrix attached to the presynaptic mem- r 1999 WILEY-LISS, INC. Grant sponsor: Deutsche Forschungsgemeinschaft; Grant numbers: SFB 425/A1 and KR 1879/1-1; Grant sponsor: Fonds der Chemischen Industrie; Grant sponsor: National Institutes of Health; Grant numbers: P50 HD32901, AG 12978-02, and AG 06569-09; Grant sponsor: Keck Foundation; Grant number: 931360. Karin Richter’s current address: Institute of Medical Neurobiology, Faculty of Medicine, Otto von Guericke University, 39120 Magdeburg, Germany. Kristina Langnaese’s current address: Institute for Human Genetics, Faculty of Medicine, Otto von Guericke University, 39120 Magdeburg, Germany. *Correspondence to: E.D. Gundelfinger, Leibniz Institute for Neurobiology, Postfach 1860, D-39008 Magdeburg, Germany. E-mail: [email protected] Received 28 September 1998; Revised 11 January 1999; Accepted 14 January 1999 438 K. RICHTER ET AL. brane and interspersed between synaptic vesicles has been observed (Landis et al., 1988; Hirokawa et al., 1989; Burns and Augustine, 1995; Brodin et al., 1997). The molecular nature of this presynaptic cytomatrix is not well understood. Part of the filamentous meshwork appears to be based on the actin/fodrin (i.e., brain spectrin) cytoskeleton and on synapsin I, which links synaptic vesicles to actin filaments (Landis et al., 1988; Hirokawa et al., 1989; Burns and Augustine, 1995; Pieribone et al., 1995). Other filaments of unknown protein composition and larger in size than actin/spectrin fibers arise from the presynaptic plasmalemma and extend into the proximal axoplasm (Landis et al., 1988; Brodin et al., 1997). It has been postulated that this filamentous cytomatrix may represent the machinery that guides synaptic vesicles from the synapsin-associated pool to the plasma membrane (Brodin et al., 1997). In an effort to identify new synaptic proteins of the mammalian brain (Langnaese et al., 1996), we have characterized two proteins, i.e., Piccolo and Bassoon, that are strong candidates for being components of the presynaptic cytomatrix proximal to the active zone (CasesLanghoff et al., 1996; tom Dieck et al., 1998). Both polypeptides are very large in size (molecular weight ⬎ 400,000), behave biochemically as cytoskeleton-associated proteins, and their distribution within synaptic nerve terminals is restricted to the cytomatrix opposite to postsynaptic densities (Cases-Langhoff et al., 1996; tom Dieck et al., 1998). Piccolo immunoreactivity is detected in neurons of virtually all regions of the rat brain and is found primarily in asymmetric type 1 synapses (CasesLanghoff et al., 1996). We examined the distribution of Bassoon RNA and protein in the rat brain by in situ hybridization and immunohistochemistry, respectively. Moreover, we addressed the question as to which types of synapses contain the protein. Bassoon transcripts are expressed in neurons throughout the brain. Both somata of identifiable excitatory and inhibitory (i.e., ␥-aminobutyric acid [GABA]ergic) neurons contain Bassoon transcripts. Consistent with the in situ hybridization data, Bassoon immunoreactivity is found widely distributed in synaptic neuropil regions of the brain. At the ultrastructural level, brain synapses can be distinguished by the symmetry of their junctional membranes (Gray, 1959). Asymmetric type 1 synapses are characterized by a prominent electron-dense postsynaptic thickening, the postsynaptic density (PSD), and include primarily excitatory synapses that use glutamate as a transmitter. In contrast, symmetric type 2 synapses have a less pronounced postsynaptic thickening; they include GABAergic and glycinergic inhibitory synapses (Gray, 1959; Landis and Reese, 1974; Shepherd and Greer, 1990; Peters et al., 1991; Ziff, 1997). Immunoelectron microscopy showed that Bassoon is present in presynaptic terminals of both asymmetric and symmetric synapses and thus may play a general role in neurotransmitter release and/or synaptic vesicle recycling. MATERIALS AND METHODS Antibodies The production, purification, and characterization of polyclonal and monoclonal Bassoon antibodies have been described previously (tom Dieck et al., 1998). The monoclonal synaptophysin antibody was obtained from DAKO (Hamburg, Germany; clone SY 38), and the rabbit polyclonal anti-glutamate receptor 1 (GluR1) antibody was obtained from Chemicon International Inc. (Temecula, CA; 1:50). GABAA receptors were visualized by using mouse monoclonal antibody against GABAA receptor ␤ chains (clone bd 17, 1:30; Boehringer Mannheim, Mannheim, Germany). Secondary antibodies were purchased from Dianova (Hamburg, Germany; anti-mouse IgG-biotin) or from Boehringer Mannheim (anti-rabbit IgG biotin, F[ab8]2 fragment). Fluorescein-isothiocyanate and Texas Red– conjugated goat anti-mouse or goat anti-rabbit secondary antibodies were obtained from Sigma (St. Louis, MO) and Molecular Probes (Eugene, OR), respectively. In situ hybridization In situ hybridization was performed essentially as described by Seidenbecher et al. (1998) with the following oligodeoxynucleotides: (1) GGAAGACAGGGAGGTGGGTGAGGTGCCAGATGTATAGCTA, (2) ACAGCGGTGTCGTCTTCCTCCAAGTTGTCTTCCTCGGCGC, and (3) TAAGGCTCTCCATCTCCAGCTCAGGCTCCCGGTCTA GGTT as antisense probes. They all produced identical results. Controls included competition with 100-fold excess of unlabeled oligonucleotide, RNAse treatment of sections before hybridization, and hybridization with sense probe complementary to oligodeoxynucleotide 1. No specific signals were observed under any of these conditions (see also tom Dieck et al., 1998). Immunohistochemistry and immunoelectron microscopy Thirty-day-old male rats were used for immunohistochemical studies. Perfusion, fixation, sectioning, and preincubation of 60-µm sections were performed as detailed previously (Richter et al., 1996). After the preincubation procedure, sections were incubated for 2 days at room temperature with 1:1,200 diluted rabbit polyclonal antisera, with polyclonal mouse antiserum diluted 1:800 or with monoclonal Bassoon antibodies diluted 1:10,000 in phosphate buffered saline (PBS) containing 1% (w/v) bovine serum albumin (BSA) and 0.05% (w/v) sodium azide. For comparison, a few sections were incubated with a monoclonal synaptophysin antibody diluted 1:2,000. For light microscopic analysis, 0.1% (v/v) Triton X-100 was added. For visualization of primary antibodies, the biotinavidin-peroxidase complex method was used (ABC Elite kit, Vector, Burlingame, CA). The detection reaction was Fig. 1. Distribution of Bassoon transcripts in rat brain. A: Hybridization signal on a horizontal section of rat brain that was hybridized with a 35S-labeled antisense oligonucleotide. B–D: Hippocampus: Silver grain accumulation over densely packed perikarya of granule cells (Grp) and discrete cell bodies in the hilus region (hi) of the dentate gyrus (DG; B) and over the pyramidal cell body layer (Py) in the CA1 region (C) and CA3 region (D). E: Expression of Bassoon transcript in the cerebellar cortex (Cbx). Dense labeling is detectable over the granule cell bodies in the granule cell layer (gr). In addition, silver grains are found over Purkinje cell bodies (Pu) and cell bodies in the molecular layer (mol), most probably basket cells (arrowheads). F: Silver grain accumulation over perikarya of pyramidal cells (arrowheads mark examples) in the cerebral cortex (Cx) layer V. G: In the caudate-putamen (CPu), the majority of the neuronal cell bodies are labeled. Cb, cerebellum; Cx, cerebral cortex; Hip, hippocampus; sl, stratum lucidum; sr, stratum radiatum; so, stratum oriens; ml, stratum moleculare. Scale bars ⫽ 5 mm in A, 20 µm in B–G. BASSOON IN EXCITATORY AND INHIBITORY NERVE TERMINALS Figure 1 439 440 K. RICHTER ET AL. performed with 0.05% (w/v) diaminobenzidine and 0.01% (v/v) H2O2. Tissue preparation for electron microscopic analysis was done as described by Richter et al. (1996). Ultrathin sections were examined with a Leo912 electron microscope (Leo Elektronenmikroskopie, Oberkochen, Germany) and imaged with a Megascan 2K CCD camera (Gatan, Inc., Pleasanton, CA) using the digital Gatan 2.5 software. To test for nonspecific immunolabeling, sections were incubated exactly as described above but in the absence of the first antibody, with preimmune rabbit serum or with an antibody solution that was preincubated with the fusion protein (2.3 µg/ml). In no case was any nonspecific immunoreactivity observed. Primary cultures of hippocampal neurons and immunofluorescence microscopy Primary cultures of hippocampal neurons were prepared from Sprague-Dawley rat embryos at embryonic day 19 essentially as described by Goslin and Banker (1991), with modifications according to Ye and Sontheimer (1998). Hippocampi were dissected out and stored in HEPES buffered Hank’s balanced salt solution without Ca2⫹ and Mg2⫹, pH 7.3, and dissociated with the papain dissociation system (Worthington, NJ). Cells were plated onto poly-Dlysine-coated glass coverslips at densities from 2 to 5 ⫻ 104 cells/coverslip. Conditioned culture medium was used in plating and maintaining the neurons (Ye and Sontheimer, 1998). For double labeling, immunofluorescence microscopy differentiated cultures (after 23 days in vitro) were fixed in 3.7% formaldehyde in PBS, pH 7.4, for 20–30 minutes, washed twice with PBS, and permeabilized for 5 minutes with 0.25% Triton X-100. Nonspecific binding was blocked for 1 hour with PBS containing 2% BSA, 5% fetal calf serum (FCS), 2% glycine, 50 mM NH4Cl, 0.05% NaN3. Cells were incubated with primary antibodies diluted in 3% FCS in PBS overnight at 4°C. After washing in PBS, cells were incubated with either fluorescein-isothiocyanate or Texas Red–conjugated goat anti-mouse or goat anti-rabbit secondary antibodies. Cultures were rinsed in PBS and mounted with Vectashield mounting medium (Vector). Fluorescent images were taken with a Nikon Diaphot 300 microscope equipped with a Photometrics CH250 CCD camera and converted to Microsoft PowerPoint images. Processing of photomicrographs Light microscopic photographs, taken with a Leica/ LEITZ (Bensheim, Germany) DMRD system, were scanned with an AGFA ARCUS (Orangeburg, NY) scanner (gray scale, 600 ppi), scaled, and adjusted in brightness by using the graphic software Adobe Photoshopy, version 3.0.3, to mount composite displays. Digital electron microscopic and fluorescent images were imported into Adobe Photoshop and processed in the same way. All hard copies were printed with a FUJI Pictography 3000 printer (FUJI Photo Film Europe, Düsseldorf, Germany). Animals All experiments were conducted in accordance with the NIH guidelines for animals in research and with the ethical principles defined by the German Animal Welfare Act. RESULTS Distribution of Bassoon transcripts Bassoon is encoded by an ⬇13-kb transcript that is specifically expressed and widely distributed in the rat brain (tom Dieck et al., 1998). By in situ hybridization, we have assessed the distribution of Bassoon transcripts in the rat brain in detail. Most intense hybridization signals are observed in the cerebellar cortex, the hippocampal formation, and the cerebral cortex; in other brain regions, lower expression levels are detectable (Fig. 1A). At a cellular level, a strong accumulation of silver grains is detected over perikarya of excitatory neurons. These include granule cells in the dentate gyrus (Fig. 1B), pyramidal cells in the Ammon’s horn regions of the hippocampus (Fig. 1C,D), granule cells in the cerebellum (Fig. 1E), and pyramidal cells in the cerebral cortex (Fig. 1F). In addition, inhibitory GABAergic neurons express Bassoon. Examples of these neurons include basket cells in the molecular layer of the cerebellum and cerebellar Purkinje cells (Fig. 1E). Moreover, in the caudate putamen, the vast majority (about 90%) of neurons are GABAergic (Mugnaini and Oertel, 1985; Graybiel, 1990). As shown in Figure 1G, many neurons of this brain region including GABAergic ones are decorated with silver grains. Taken together, the in situ hybridization data indicate that throughout the brain a wide variety of both excitatory and inhibitory neurons express Bassoon. Spatial distribution of Bassoon immunoreactivity The spatial distribution of Bassoon in the brain of 30-day-old rats was studied by immunohistochemistry at light and electron microscopic levels. Rabbit and mouse polyclonal antisera and mouse monoclonal antibody mab7f (tom Dieck et al., 1998) raised against a recombinant fragment of Bassoon produced virtually identical staining patterns. Rabbit and mouse preimmune sera did not react with brain sections, and preincubation of antisera with recombinant Bassoon antigen resulted in an almost complete elimination of immunostaining (not shown). Bassoon immunoreactivity is widely distributed in the brain (Fig. 2A). In general, strong staining is found in synaptic neuropil regions; perikaryal regions are stained to a significantly lesser extent, and white matter areas are essentially devoid of immunoreactivity. The highest levels of Bassoon immunoreactivity are observed in the molecular layer of the cerebellum and in the neuropil regions of the hippocampus, the olfactory tubercle, and the olfactory bulb including the accessory olfactory bulb. Marked staining is also found in the caudate putamen, the cerebral cortex, the thalamic nuclei, inferior colliculus, and superior colliculus (superficial gray layer; Fig. 2A). The globus pallidus is less intensely stained than the caudate putamen (Fig. 2B). In the cerebral cortex, moderate immunolabeling that outlines dendritic and somatic profiles of neurons (Fig. 2C) is found in all cortical layers. In semithin sections, a punctate distribution of the immunoreactivity on neuronal surfaces becomes apparent (Fig. 2D), suggesting a synaptic localization of Bassoon. In the piriform cortex, intense labeling is seen in layer Ia, whereas layers Ib, II, and III are labeled to a lesser extent; the adjacent lateral olfactory tract is not labeled (Fig. 2E). In the main olfactory bulb, high levels of immunoreactivity are observed in the glomeruli and in the external plexiform layer, Fig. 2. Spatial distribution of Bassoon immunoreactivity in rat brain. A: Low magnification photograph of a sagittal section of a 30-day-old rat brain (corresponding to Fig. 81 in the rat stereotaxic atlas by Paxinos and Watson, 1986). High levels of Bassoon immunoreactivity were detected throughout neuropil regions of the whole brain, with highest levels in the hippocampus, the cerebellar cortex, and the olfactory bulb. The irregular light and dark areas of staining in the cerebral cortex are not reproducibly seen on all sections examined. B: Labeling of the caudate putamen (CPu) and globus pallidus (GP); fibers of the internal capsule (ic) are free of reaction product. C: Immunoreactivity in layers I–VI of the cerebral cortex. D: Higher magnification of layer V (semithin section). Note the punctate staining that leaves blank neuronal perikarya (p) and dendrites (d). E: Intense immunostaining of layer Ia and less intense staining of layers Ib–II are observed in the piriform cortex. The lateral olfactory tract (lot) is not stained. F: Main olfactory bulb. The glomeruli (gl) and the external plexiform layer (epl) show strong immunoreactivity, whereas the ganglion cell layer (gcl) shows only moderate staining. AOB, accessory olfactory bulb; Cb, cerebellum; Cx, cerebral cortex; CPu, caudate putamen; SC, superior colliculus; IC, inferior colliculus; Hip, hippocampus; OB, olfactory bulb; Th, thalamic nuclei; Tu, olfactory tubercle. Scale bars ⫽ 2 mm in A, 200 µm in C, 100 µm in B,E,F, 20 µm in D. 442 whereas the ganglion cell layer is only moderately stained (Fig. 2F). Cell-type distribution and ultrastructural localization of Bassoon in hippocampus and cerebellar cortex A detailed analysis of the cell-type distribution and subcellular localization of Bassoon was performed in the hippocampus and the cerebellar cortex because of their well-known circuitry, the relatively small number of constituent neuron types, and the segregation of different classes of synaptic boutons into easily distinguishable cellular layers within these brain structures. Hippocampus. In the neuropil regions of the hippocampus, a very distinct laminated staining pattern that essentially reflects the innervation and intrinsic synaptic circuitry of this brain region (Brown and Zador, 1990; Amaral and Witter, 1995) is observed with Bassoon antisera (Fig. 3A). Both the stratum oriens and stratum radiatum of the CA1 and CA3 regions are strongly immunoreactive. The stratum lacunosum moleculare of CA1 shows significantly less staining. Prominent labeling also occurs in the stratum lucidum of CA3 (see also Fig. 3C) and in the hilus region (polymorphic cell layer). The molecular layer of the dentate gyrus is characterized by a bipartite staining pattern, with moderate Bassoon immunoreactivity in its outer two-thirds and strong reactivity in the proximal one-third (Fig. 3A). Perikarya of CA1–3 pyramidal cells and granule cell bodies of the dentate gyrus appear almost unstained (Fig. 3A,C). Hippocampal sections processed in parallel with anti-synaptophysin antibody show a staining pattern that is very similar to that of Bassoon (Fig. 3B). An analysis of semithin sections through the CA3 region showed a clearly punctate, most likely synaptic, distribution of Bassoon immunoreactivity in the neuropil, e.g., the stratum lucidum and stratum radiatum (Fig. 3C). In the stratum lucidum, punctae are larger and more intensely stained than those in the stratum radiatum (Fig. 3C). In various neuropil regions of the hippocampus, synaptic junctions were analyzed by electron microscopy (Fig. 3D– J). The examination showed a strictly presynaptic localization of Bassoon immunoreactivity. Moreover, immunoreactivity appeared to be most prominent at sites of the presynaptic membrane facing PSDs, as indicated by a strong concentration of peroxidase reaction product at these sites. This is clearly visible in the large mossy fiber terminals in the stratum lucidum of the CA3 region, which are tightly packed with synaptic vesicles (SVs) and form contacts with several postsynaptic compartments (tom Dieck et al., 1998; see Fig. 3D for another example). Via mossy fiber boutons, CA3 pyramidal neurons receive excitatory input from dentate granule cells (Brown and Zador, 1990; Amaral and Witter, 1995). In the molecular layer of the dentate gyrus, the polymorphic cell layer of the hippocampal hilus, the stratum radiatum of the CA1 area, and the stratum lucidum of the CA3 region, Bassoon is present at both shaft and spine synapses (Fig. 3E–J). Most of these synapses are asymmetric and thus may be excitatory (Peters et al., 1991; Ziff, 1997). In addition Bassoon-containing terminals of symmetric synapses are found in the hippocampus. Figure 3E shows an example of a dendrite that is contacted by two synaptic terminals, one of which is clearly asymmetric and the K. RICHTER ET AL. other possibly symmetric. Also, Figure 3G shows Bassoon immunoreactivity in the axon terminal of a presumably symmetric synapse of the molecular layer of the dentate gyrus. Although the symmetric synapses of the transmitter type is unknown in these examples, Figure 3 shows a Bassoon-positive GABAergic nerve terminal of a basket cell making contact with a granule cell body (Kosaka et al., 1984). Cerebellum. At low magnification in the light microscope, the molecular layer of the cerebellum shows an intense uniform immunostaining with Bassoon antibodies, whereas staining in the granule cell layer is patchy. Purkinje cell bodies and white matter are devoid of immunoreactivity (Fig. 4A). In semithin sections, in the stratum moleculare (Fig. 4B), a punctate staining pattern outlining Purkinje cell somata and primary dendrites is visible. In the adjacent granular layer, punctate staining is confined to distinct patches, most likely glomeruli (Fig. 4B). An ultrastructural analysis of the cerebellar cortex confirmed the observations made in the hippocampus (Fig. 4C–G). Both excitatory and inhibitory GABAergic synaptic terminals are found to contain Bassoon. Prominent examples of excitatory synapses are the mossy fiber terminals in glomeruli of the granule cell layer (Fig. 4C), which originate from various spinocerbellar projections (Somogyi et al., 1986; Garthwaite and Brodbelt, 1990; Llinás and Walton, 1990; Voogd, 1995). As in hippocampal mossy fiber boutons, Bassoon immunoreactivity is virtually restricted to regions near synaptic contact sites (Fig. 4C). The molecular layer receives predominantly excitatory input Fig. 3. Localization of Bassoon immunoreactivity in the hippocampus. A: Overview of the hippocampus in a sagittal brain section from a 30-day-old rat immunostained with Bassoon. The neuropil regions stratum oriens (so) and stratum radiatum (sr) and the hilar region (hi) are intensely stained, the stratum lacunosum-moleculare (slm) shows moderate staining, and cell bodies in the pyramidal cell layer of the CA1–3 regions and of the granule cell layer in the dentate gyrus (DG) are essentially devoid of reaction product. The molecular layer (ml) of the DG shows moderate staining in its outer layers and intense staining in the inner part. B: Synaptophysin immunoreactivity in a sagittal section treated in parallel to the section shown in A. The overall distribution pattern of this synaptic vesicle protein in the hippocampus resembles that of Bassoon. C: Semithin section from the Bassoon-immunostained CA3 region. Note the marked punctate staining in the stratum lucidum (sl) versus the fine punctate staining in the stratum radiatum (sr). Pyramidal cells (Py) are unstained. D–I: Electron micrographs of synapses from the sl of the CA3 region (D), the hilus region (polymorphic cell layer, E), the ml (inner part, F,G; outer part, I), and the granule cell layer (J) of the DG. Arrowheads mark postsynaptic densities (PSDs) and asterisks postsynaptic sides of symmetric synapses. D: Mossy fiber bouton (mf) densely filled with vesicles contacts spiny dendritic processes of CA3 pyramidal cells. The peroxidase reaction product is restricted to the presynaptic element and is highly concentrated at regions apposing PSDs. E: Profile of a dendrite (d) that receives en passant contact from two Bassoonimmunopositive terminals; the synaptic junctions are different with respect to the thickness of postsynaptic densities. F: Shaft synapse with apparently separate junctional zones. In the presynaptic element, Bassoon immunoreactivity appears most intense at sites facing PSDs. G: Symmetric junction between a dendritic shaft and a Bassoonimmunopositive presynaptic element filled with pleomorphic vesicles. H: Immunopositive terminal making contact with a dendritic shaft. Immunoreactivity gradually increases toward the plasma membrane facing the PSD. I: Asymmetric synaptic junctions between an immunopositive presynaptic element and a dendritic spine. J: Symmetric synaptic contact between a basket cell axon and a granule cell perikaryon (Grp). Scale bars ⫽ 500 µm in A,B, 50 µm in C, 0.6 µm in D, 0.3 µm in E–G,I,J, 0.2 µm in H. BASSOON IN EXCITATORY AND INHIBITORY NERVE TERMINALS Figure 3 443 444 K. RICHTER ET AL. Figure 4 BASSOON IN EXCITATORY AND INHIBITORY NERVE TERMINALS from cerebellar granule cells via parallel fibers and from the inferior olive via climbing fibers (Palay and ChanPalay, 1974; Somogyi et al., 1986; Llinás and Walton, 1990). Both climbing and parallel fiber terminals forming asymmetric type 1 synapses with Purkinje cell dendrites and interneurons are immunolabeled with Bassoon antibodies (Fig. 4D,E). In addition, axosomatic terminals on Purkinje cells that derive from basket cells (Fig. 4G) and Golgi cell terminals contacting granule cell dendrites in glomeruli (Fig. 4F) are found to contain Bassoon immunoreactivity. Both synapses use GABA as a neurotransmitter (Ottersen and Storm-Mathisen, 1984; Ottersen et al., 1987). Thus, also in the cerebellum Bassoon appears to be present in excitatory and inhibitory GABAergic synapses. Colocalization of Bassoon with marker proteins of glutamatergic and GABAergic synapses The immunoelectron microscopy data indicate the presence of Bassoon in identifiable axon terminals thought to release either excitatory amino acids or GABA. Doublelabeling immunocytochemical studies with antibodies against Bassoon and marker proteins of glutamatergic or GABAergic postsynapses were performed on primary cultures of hippocampal neurons to further assess the association of Bassoon with excitatory and inhibitory synapses, respectively. In neurons cultured for 23 days in vitro, many but clearly not all Bassoon-immunopositive puncta along dendritic profiles were found to colocalize with those immunoreactive for GluR1 subunits of ␣-amino-3-hydroxy5-methyl-4-isoxazole-propionate (AMPA) receptors, suggesting that Bassoon is present at glutamatergic synapses (Fig. 5A,B). Analogous double-labeling experiments with anti-Bassoon and GABAA receptor ␤-subunit antibodies showed that some of the Bassoon-immunopositive puncta are also positive for GABAA receptors (Fig. 5C,D). In essence, this set of data also supports the view that Fig. 4. Localization of Bassoon immunoreactivity in the rat cerebellar cortex. A: Photomicrograph of a sagittal section through the cerebellum showing an overview of staining in the cerebellar layers. The molecular layer (mol) is strongly stained. Significantly less staining is seen in the granule cell layer (gr). Purkinje cell layer (Pc) and white matter (wm) appear unstained. B: Semithin section of the cerebellar cortex showing clear punctate staining in the molecular layer (mol) and a patchy distribution of punctae in the granule cell layer (gr), probably representing glomeruli (arrows). Purkinje cell dendrites and perikarya (Pu) are unlabeled. C–G: Electron micrographs of various cerebellar synaptic junctions (arrowheads mark postsynaptic densities and asterisks mark the postsynaptic elements of symmetric synapses). C: Localization of immunoreactivity in a mossy fiber terminal (mf) of the granule cell layer. Apposing dendritic profiles of granule cells (d) are free of reaction product. The inset shows the concentration of immunoreactivity at the synaptic junction. D: Immunopositive axon terminals or varicosities contacting dendritic spines in the molecular layer. E: Immunoreactivity in a parallel fiber terminal making synaptic contact with the perikaryon (p), probably of a stellate cell. F: Immunoreactivity in a Golgi cell terminal that contacts a dendrite (d) of a granule cell. Note the gradient of increasing amounts of reaction product toward the synaptic cleft. G: Axon, probably of a basket cell (Bax), contacting a Purkinje cell perikaryon (Pu) at two symmetric junctions. Vesicles and immunopositive material colocalize at this presumably GABAergic junction. Grp, granule cell perikaryon. Scale bars ⫽ 200 µm in A, 25 µm in B, 0.7 µm in C, 0.3 µm in D,G, 0.2 µm in E,F, inset in C. 445 Bassoon occurs at the majority of glutamatergic and GABAergic synapses. DISCUSSION Bassoon has been identified recently as a novel protein tightly associated with the cytomatrix at the active zone of presynaptic nerve terminals (tom Dieck et al., 1998). The aims of the present study were to examine the overall distribution of Bassoon and its transcripts in the adult rat brain, to identify types of synapses containing the protein, and to determine the ultrastructural localization of Bassoon within these synapses. Bassoon transcripts are widely expressed in the brain; regions of highest intensity of expression coincide with regions of highest cell density, e.g., the hippocampus, cerebellar cortex, cerebral cortex, or olfactory bulb (Fig. 1; see also tom Dieck et al., 1998). This distribution resembles that of transcripts of the integral synaptic vesicle protein synaptophysin (Marquèze-Pouey et al., 1991) and the vesicle-associated protein synapsin I (Melloni et al., 1993). Synapsin I isoforms occur in the vast majority of presynaptic terminals of the brain regardless of the transmitter type (De Camilli et al., 1983a; Südhof et al., 1989), and synaptophysin has been estimated to be present in more than 90% of clear synaptic vesicles from rat brain (Wiedenmann and Franke, 1985; Floor and Feist, 1989). The widespread distribution of Bassoon transcripts is consistent with a general transmitter type-independent synaptic distribution of the Bassoon protein. Analysis of in situ hybridization data at a cellular level supports this view. In addition to the intense labeling of excitatory neurons, e.g., granule cells in the cerebellum and the dentate gyrus or pyramidal neurons in the hippocampus and cerebral cortex, clear hybridization signals are found over identifiable inhibitory neurons, including Purkinje cells and basket cells in the cerebellar cortex and principal GABAergic neurons in caudate putamen (Mugnaini and Oertel, 1985; Graybiel, 1990). Immunohistochemical experiments with Bassoon antibodies showed a general labeling of synaptic neuropil regions. Most intense staining is observed in regions with high density of synaptic contacts, e.g., the molecular layer of the cerebellum, the various strata of the hippocampus, or plexiform layers of main and accessory olfactory bulbs. The Bassoon staining pattern is very similar to that obtained with synaptophysin antibodies in parallel sections. A virtually complete overlap of the distribution of Bassoon and synaptophysin was also shown in primary hippocampal neurons after 21 days in culture (tom Dieck et al., 1998). Also, these observations on the protein distribution are in agreement with a largely general synaptic occurrence of Bassoon. By immunoelectron microscopy, individual synaptic contacts were analyzed for the presence of Bassoon in their presynaptic compartment. In the two brain regions analyzed at the ultrastructural level, i.e., the hippocampus and the cerebellum, nerve terminals of asymmetric type 1 and symmetric type 2 synapses are found to contain Bassoon. Some of the identified Bassoon-positive synapses have been characterized previously with respect to their neurotransmitter type. They include excitatory synapses, e.g., mossy fiber terminals in the hippocampus and the cerebellum or parallel, climbing fiber synapses in the cerebellar molecular layer, and inhibitory synapses, e.g., 446 K. RICHTER ET AL. Fig. 5. Bassoon codistributes with both GABAA and glutamate receptors (GluR). Double images of primary hippocampal neurons after 23 days in culture fluorescently labeled with antibodies against Bassoon (A) and the GluR1 receptor (B) and against Bassoon (C) and the GABAA receptor ␤ subunit (D). Arrows in A and B mark examples of colocalization; arrowheads mark examples of Bassoon immunoreactivity that does not colocalize with GluR1. Insets in C and D are close-ups of a region particularly rich in GABAergic synapses. Scale bars ⫽ 10 µm. axosomatic synapses of GABAergic basket cells onto Purkinje cells, GABAergic axodendritic synapses between Golgi cells and granule cells in the cerebellum, and GABAergic basket cell synapses onto granule cells of the dentate gyrus. The occurrence of Bassoon in excitatory and inhibitory synapses is further confirmed in cultured hippocampal neurons by the colocalization with AMPA receptors and GABAA receptors, respectively. Moreover, in the rat retina, excitatory ribbon synapses and GABAergic conventional synapses contain Bassoon (J.H. Brandstätter, E.L. Fletcher, C.C. Garner, E.D. Gundelfinger, and H. Wässle, personal communication). Taken together, these data are consistent with the in situ hybridization data and demonstrate that Bassoon is a presynaptic protein found in a wide variety of excitatory and inhibitory synapses throughout the central nervous system. It is worth noting that high levels of Bassoon transcripts are present in dopaminergic cells of the substantia nigra (Richter et al., unpublished observation), suggesting that dopamine release sites also contain Bassoon. Within presynaptic terminals of excitatory synapses of the hippocampal CA3 region, Bassoon has a very restricted distribution, i.e., immunoreactivity is highly concentrated in the vicinity of the active zone (tom Dieck et al., 1998). We have shown that this very specific localization is also found in many synapses of other brain regions and is independent of the transmitter type of a synapse. Particular examples are the large mossy fiber terminals in the cerebellar cortex that, like the mossy fiber boutons in the stratum lucidum of the hippocampal CA3 region, are large fusiform structures making multiple synaptic contacts with postsynaptic neurons. They are filled with excitatory BASSOON IN EXCITATORY AND INHIBITORY NERVE TERMINALS amino acids containing synaptic vesicles (Llinás and Walton, 1990; Amaral and Witter, 1995). Bassoon immunoreactivity appears to be highly concentrated at presynaptic regions facing PSDs. A clear gradient of immunoreactivity with most intense labeling at the neurotransmitter release site is also seen in GABAergic synaptic terminals, e.g., of cerebellar Golgi cells on granule cell dendrites. Frequently, smaller synaptic terminals of both symmetric and asymmetric synapses are filled with the product of peroxidase reaction; this may, however, be due to diffusion of the reaction product rather than to a generally different distribution of Bassoon in these synapses. The highly specific localization of Bassoon within presynaptic nerve terminals is clearly different from that of other integral and associated synaptic vesicle proteins, such as synaptophysin (Wiedenmann and Franke, 1985; Kagotani et al., 1991) or synapsin (De Camilli et al., 1983b; Pieribone et al., 1995; tom Dieck et al., 1998), but resembles that of Piccolo and Rim, two recently described presynaptic proteins (Cases-Langhoff et al., 1996; Wang et al., 1997). All three proteins copurify with marker proteins of the PSD, although they are clearly of presynaptic origin (Cases-Langhoff et al., 1996; Wang et al., 1997; tom Dieck et al., 1998). Rim is a ⬃180-kDa zinc-finger protein thought to be involved in Rab3-dependent regulation of neurotransmitter release (Wang et al., 1997), whereas the functions of Piccolo and Bassoon are unknown. In conclusion, our data show that the distribution of Bassoon is not restricted to synapses of a single transmitter type but is found in most synapses throughout the brain. Therefore, its role may be a general one in the assembly and organization of presynaptic structures. Interestingly, during development Bassoon, like other presynaptic marker proteins such as synapsin or synaptophysin (Fletcher et al., 1991), appears in axons and growth cones before synaptogenesis (G. Olias and C.C. Garner, unpublished observation) and therefore could well be involved in early processes of synapse formation. Based on its biochemical properties, the multidomain structure, and its exquisite distribution within nerve terminals (tom Dieck et al., 1998; present study), we assume that Bassoon is a scaffold protein of the presynaptic cytomatrix that serves to organize the presynaptic active zone, perhaps by clustering the machinery necessary for the release of neurotransmitter and the recycling of synaptic vesicles. ACKNOWLEDGMENTS We are grateful to members of the technical staff of the Departments of Neurochemistry/Molecular Biology and Auditory Plasticity and Speech of the IfN for expert technical assistance and to Constanze Seidenbecher for critically reading the manuscript. This study was supported by grants from the Deutsche Forschungsgemeinschaft (SFB 425/A1, KR 1879/1-1) and the Fonds der Chemischen Industrie to E.D.G. and M.R.K. and by grants from the National Institutes of Health (P50 HD32901, AG 12978-02, AG 06569-09) and the Keck Foundation (931360) to C.C.G. LITERATURE CITED Amaral DG, Witter MP. 1995. Hippocampal formation. In: Paxinos G, editor. The rat nervous system. London: Academic Press. p 443–494. Betz WJ, Angleson JK. 1998. The synaptic vesicle cycle. Annu Rev Physiol 60:347–363. 447 Brodin L, Löw P, Gad H, Gustafsson J, Pieribone VA, Shupliakov O. 1997. Sustained neurotransmitter release: new molecular clues. Eur J Neurosci 9:2503–2511. Brown TH, Zador AM. 1990. Hippocampus. In: Shepherd GM, editor. The synaptic organization of the brain. New York: Oxford University Press Inc. p 346–388. Burns ME, Augustine GJ. 1995. Synaptic structure and function: dynamic organization yields architectural precision. Cell 83:187–194. Cases-Langhoff C, Voss B, Garner AM, Appeltauer U, Takei K, Kindler S, Veh RW, De Camilli P, Gundelfinger ED, Garner CC. 1996. Piccolo, a novel 420 kDa protein associated with the presynaptic cytomatrix. Eur J Cell Biol 69:214–223. De Camilli P, Takei K. 1996. Molecular mechanisms in synaptic vesicle endocytosis and recycling. Neuron 16:481–486. De Camilli P, Cameron R, Greengard P. 1983a. Synapsin I (protein I), a nerve terminal specific phosphoprotein. I. Its general distribution in synapses of the central and peripheral nervous system demonstrated by immunofluorescence in frozen and plastic sections. J Cell Biol 96:1337– 1354. De Camilli P, Harris SM, Huttner WB, Greengard P. 1983b. Synapsin I (protein I), a nerve terminal-specific phosphoprotein. II. Its specific association with synaptic vesicles demonstrated by immunocytochemistry in agarose-embeddeed synaptosomes. J Cell Biol 96:1355–1373. Fletcher TL, Cameron P, De Camilli P, Banker G. 1991. The distribution of synapsin I and synaptophysin in hippocampal neurons developing in culture. J Neurosci 11:1617–1626. Floor E, Feist BE. 1989. Most synaptic vesicles isolated from rat brain carry three membrane proteins, SV2, synaptophysin, and p65. J Neurochem 52:1433–1437. Garthwaite J, Brodbelt AR. 1990. Glutamate as the principal mossy fibre transmitter in rat cerebellum: pharmacological evidence. Eur J Neurosci 2:177–180. Geppert M, Südhof TC. 1998. RAB3 and synaptotagmin: the yin and yang of synaptic membrane fusion. In: Cowan WM, Shooter EM, Stevens CF, Thompson RF, editors. Annual reviews in neuroscience. Volume 21. Palo Alto: Annual Reviews Inc. p 75–95. Goslin K, Banker G. 1991. Rat hippocampal neurons in low-density culture. In: Banker G, Goslin K, editors. Culturing nerve cells. Cambridge (MA): MIT Press. p 251–281. Gray EG. 1959. Axo-somatic and axo-dendritic synapses of the cerebral cortex: an electron microscope study. J Anat 93:420–433. Graybiel AM. 1990. Neurotransmitters and neuromodulators in the basal ganglia. Trends Neurosci 13:244–245. Hirokawa N, Sobue K, Kanda K, Harada A, Yorifuji H. 1989. The cytoskeletal architecture of the presynaptic terminal and molecular structure of synapsin 1. J Cell Biol 108:111–126. Kagotani Y, Picart R, Barret A, Wiedenmann B, Huttner WB, Tixier-Vidal A. 1991. Subcellular localization of secretogranin II and synaptophysin by immunoelectron microscopy in differentiated hypothalamic neurons in culture. J Histochem Cytochem 39:1507–1518. Kosaka T, Hama K, Wu J-Y. 1984. GABAergic synaptic boutons in the granule cell layer of rat dentate gyrus. Brain Res 293:353–359. Landis DMD, Reese TS. 1974. Differences in membrane stucture between excitatory and inhibitory synapses in the cerebellar cortex. J Comp Neurol 155:93–126. Landis DMD, Hall AK, Weinstein LA, Reese TS. 1988. The organization of cytoplasm at the presynaptic active zone of a central nervous system synapse. Neuron 1:201–209. Langnaese K, Seidenbecher C, Wex H, Seidel B, Hartung K, Appeltauer U, Garner A, Voss B, Mueller B, Garner CC, Gundelfinger ED. 1996. Protein components of a rat brain synaptic junctional protein preparation. Mol Brain Res 42:118–122. Llinás RR, Walton KD. 1990. Cerebellum. In: Shepherd GM, editor. The synaptic organization of the brain. New York: Oxford University Press. p 214–245. Marquèze-Pouey B, Wisden W, Malosio ML, Betz H. 1991. Differential expression of synaptophysin and synaptoporin mRNAs in the postnatal rat central nervous system. J Neurosci 11:3388–3397. Melloni JRH, Hemmendinger LM, Hamos JE, DeGennaro LJ. 1993. Synapsin I gene expression in the adult rat brain with comparative analysis of mRNA and protein in the hippocampus. J Comp Neurol 327:507–520. Mugnaini E, Oertel WH. 1985. An atlas of the distribution of GABAergic neurons and terminals in the rat CNS as revealed by GAD immunohistochemistry. In: Björklund A, Hökfelt T, editors. Handbook of chemical 448 neuroanatomy. Volume 4, GABA and neuropeptides in the CNS, part I. Amsterdam: Elsevier Science Publishers BV. p 436–575. Ottersen OP, Storm-Mathisen J. 1984. Glutamate- and GABA-containing neurons in the mouse and rat brain, as demonstrated with a new immunocytochemical technique. J Comp Neurol 229:374–392. Ottersen OP, Davanger S, Storm-Mathisen J. 1987. Glycine-like immunoreactivity in the cerebellum of rat and Senegalese baboon, Papio papio: a comparison with the distribution of GABA-like immunoreactivity and with 3H glycine and 3H GABA uptake. Exp Brain Res 66:211–221. Palay SL, Chan-Palay V. 1974. Cerebellar cortex. Berlin: Springer-Verlag. Paxinos G, Watson C. 1986. The rat brain in stereotaxic coordinates. San Diego: Academic Press. Peters A, Palay SL, Webster H. 1991. The fine structure of the nervous system. Neurons and their supporting cells, 3rd ed. Oxford: Oxford University Press. p 138–211. Pieribone VA, Shupliakov O, Brodin L, Hilfiker-Rothenfluh S, Czernik AJ, Greengard P. 1995. Distinct pools of synaptic vesicles in neurotransmitter release. Nature 375:493–497. Richter K, Hamprecht B, Scheich H. 1996. Ultrastructural localization of glycogen phosphorylase predominantly in astrocytes of the gerbil brain. Glia 17:263–273. Seidenbecher CI, Gundelfinger ED, Böckers TM, Trotter J, Kreutz MR. 1998. Transcripts for secreted and GPI-anchored brevican are differentially distributed in rat brain. Eur J Neurosci 10:1621–1630. Shepherd GM, Greer CA. 1990. Olfactory bulb. In: Shepherd GM, editor. The synaptic organization of the brain. New York: Oxford University Press. p 133–169. K. RICHTER ET AL. Somogyi P, Halasy K, Somogyi J, Storm-Mathisen J, Ottersen OP. 1986. Quantification of immunogold labelling reveals enrichment of glutamate in mossy and parallel fibre terminals in cat cerebellum. Neuroscience 19:1045–1050. Südhof TC. 1995. The synaptic vesicle cycle: a cascade of protein–protein interactions. Nature 375:645–653. Südhof TC, Czernik AJ, Kao HT, Takei K, Johnston PA, Horiuchi A, Kanazir SD, Wagner MA, Perin MS, DeCamilli P, Greengard P. 1989. Synapsins: mosaics of shared and individual domains in a family of synaptic vesicle phosphoproteins. Science 245:1474–1480. tom Dieck S, Sanmartı́-Vila L, Langnaese K, Richter K, Kindler S, Soyke A, Wex H, Smalla K-H, Kämpf U, Fränzer J-T, Stumm M, Garner CC, Gundelfinger ED. 1998. Bassoon, a novel zinc-finger CAG/glutaminerepeat protein selectively localized at the active zone of presynaptic nerve terminals. J Cell Biol 142:499–509. Voogd J. 1995. Cerebellum. In: Paxinos G, editor. The rat nervous system. San Diego: Academic Press. p 309–350. Wang Y, Okamoto M, Schmitz F, Hofmann K, Südhof TC. 1997. Rim is a putative Rab3 effector in regulating synaptic-vesicle fusion. Nature 388:593–598. Wiedenmann B, Franke WW. 1985. Identification and localization of synaptophysin, an integral membrane glycoprotein of Mr 38.000 characteristic of presynaptic vesicles. Cell 41:1017–1028. Ye ZC, Sontheimer H. 1998. Astrocytes protect neurons from neurotoxic injury by serum glutamate. Glia 22:237–248. Ziff EB. 1997. Enlightening the postsynaptic density. Neuron 19:1163– 1174.