THE JOURNAL OF COMPARATIVE NEUROLOGY 423:500 –511 (2000)
Neurokinin 1 Receptor Distribution in Cholinergic Neurons and Targets of Substance P Terminals in the Rat Nucleus Accumbens VIRGINIA M. PICKEL,1* JENNIFER DOUGLAS,1 JUNE CHAN,1 PATRICK D. GAMP,2 2,3 AND NIGEL W. BUNNETT 1 Department of Neurology and Neuroscience, Weill Medical College of Cornell University, New York, New York 10021 2 Department of Surgery, University of California, San Francisco, California 94143 3 Department of Physiology, University of California, San Francisco, California 94143
ABSTRACT Substance P (SP) is the major endogenous ligand for neurokinin 1 (NK1) receptors and, together with acetylcholine, has an important role in motivated behaviors involving the limbic shell and motor core of the nucleus accumbens (NAc). To determine the functional sites for SP activation of NK-1 receptors and potential interactions with cholinergic neurons in these regions, the authors examined the electron microscopic immunocytochemical localization either of antisera against the NK1 receptor or of the NK1 receptor and either 1) SP or 2) the vesicular acetylcholine transporter (VAchT) in rat NAc. In both the NAc shell and core, NK1 receptor labeling was localized mainly to somatic and dendritic plasma membranes and nearby endosomal organelles in aspiny neurons. In sections through the ventromedial shell that were processed for NK1/SP labeling, 46% of the NK1-immunoreactive dendrites (n ⫽ 603 dendrites) showed symmetric or appositional contacts with SP-containing terminals. These terminals and several others that formed symmetric synapses also occasionally were immunoreactive for NK1 receptors. Analysis of the shell region for NK1/VAchT labeling showed that 61% of the total immunoreactive dendrites (n ⫽ 534 dendrites) contained NK1 receptors without VAchT, 29% contained both products, and 10% contained VAchT only. Many of the labeled somata and dendrites also received synaptic contact from VAchT-containing terminals. These findings suggest that, in the NAc, NK1 receptors are recycled through endosomal compartments and play a role in modulating mainly the postsynaptic responses, but also the presynaptic release, of SP and/or inhibitory neurotransmitters onto aspiny interneurons, some of which are cholinergic. J. Comp. Neurol. 423:500 –511, 2000. © 2000 Wiley-Liss, Inc. Indexing terms: reinforcement; inhibition; vesicular acetylcholine transporter; somatodendritic targeting; electron microscopy
Substance P (SP) and acetylcholine have been implicated in attention and cognitive functions involving distinct populations of forebrain neurons (Bennett et al., 1997; Everitt and Robbins, 1997; Sarter and Bruno, 1997: Schildein et al., 1998; Perry et al., 1999). Throughout the striatum, SP is present in local collaterals of spiny projection neurons (Gerfen, 1988; Yung et al., 1996; Lee et al., 1997), and acetylcholine is contained in many of the large, aspiny interneurons (Satoh et al., 1983; Pickel and Chan, 1991). The role of SP and acetylcholine in striatal function is determined in part by the anatomic connections of the striatal compartments in which these neurons reside. The © 2000 WILEY-LISS, INC.
ventral striatum, and particularly the nucleus accumbens (NAc) shell, receives major input from limbic brain regions
Grant sponsor: National Institutes of Health; Grant numbers: MH00078, MH40342, DA04600, HL18974, and DK39957. *Correspondence to: Virginia M. Pickel, Ph.D., Department of Neurology and Neuroscience, Weill Medical Center College of Cornell University, 411 East 69th Street, KB 410, New York, NY 10021. E-mail:
[email protected] Received 28 February 2000; Revised 29 March 2000; Accepted 29 March 2000
NK-1 RECEPTORS IN NUCLEUS ACCUMBENS (Nirenberg et al., 1997; Kelley, 1999) that have been implicated in attentional dysfunctions and spatial memory (Setlow et al., 2000; Yee, 2000). In contrast, the NAc core has anatomic connections that are more similar to those of the dorsal striatum or caudate-putamen nucleus (CPN) and serves mainly motor functions (Nirenberg et al., 1997; Barrot et al., 1999). The NAc shell and core also contribute to distinct aspects of motivated behaviors (Rada et al., 1993; Ikemoto et al., 1998; Kelley, 1999). Many of the functions ascribed to SP in the striatum are thought to be mediated through neurokinin 1 (NK1) receptors, because the mRNA and binding sites for NK1 receptors are prevalent throughout the CPN and NAc (Lu et al., 1998). NK1 receptors also have higher affinity for SP than for neurokinin A and B, which act mainly through binding to neurokinin 2 and 3 (NK2 and NK3) receptors, respectively (Khawaja and Rogers, 1996; Saria, 1999). In the CPN, NK1 receptor immunolabeling and mRNA are present in striatal interneurons (Gerfen, 1991; Parent et al., 1995), in which the distribution, as revealed under the light microscopic, partially overlaps that of SP immunoreactivity (Kowall et al., 1993). The aspiny interneurons expressing NK1 receptors in the CPN contain choline acetyltransferase, neuronal nitric oxide (NO) synthase (nNOS), and/or somatostatin (Kowall et al., 1993; Li et al., 2000). Furthermore, in this region, SP-immunoreactive terminals form synapses with cholinergic dendrites (Martone et al., 1992), suggesting that the functional sites for NK1 receptor activation are at these synapses. Synaptic transmission is consistent with the known calcium dependence of NK1 receptor-mediated excitation of cholinergic striatal neurons (Bell et al., 1998) that is believed to be largely responsible for the atropine-sensitive, SP-evoked excitation of spiny projection neurons (Galarraga et al., 1999). A recent ultrastructural study has suggested, however, that, in the CPN, SP is released from sites that are remote from NK1 receptors, supporting a paracrine mode of transmission (Li et al., 2000). We examined the electron microscopic immunocytochemical localization of SP and the NK1 receptor in the NAc shell and core of rat brain to determine whether a similar dissociation between ligand and receptor occurs in either of these NAc subdivisions. In each region, we also combined immunogold labeling of the NK1 receptor with immunoperoxidase localization of the vesicular acetylcholine transporter (VAchT; Roghani et al., 1998) to determine the cellular basis for SP modulation of cholinergic interneurons in striatal compartments differentially serving limbic and motor functions. Our results show a prominent extrasynaptic, plasmalemmal distribution of NK1 receptors in perikarya and dendrites of aspiny neurons targeted by SP-immunoreactive terminals and often containing VAchT in the NAc shell and core.
MATERALS AND METHODS Antisera All single labeling and the SP-NK1 receptor dual labeling was done by using a rabbit polyclonal antiserum against amino acids 369 – 407 within the C-terminus of the cloned rat NK1 receptor (Yokota et al., 1989; Mantyh et al., 1996) that was generated previously and has been well characterized (Grady et al., 1996). This antiserum yielded a labeling pattern identical to that seen with a rabbit antiserum against amino acids 393– 407 that was pur-
501 chased from Novus Biologicals Inc. (Littleton, CO) and was used for the dual labeling of VAchT-NK1 receptor. These largely overlapping sequences are present in the long isoform of the NK1 receptor, which predominates in brain (Grady et al., 1996). All labeling was abolished by prior adsorption of the antisera with the corresponding peptides, and the light microscopic distribution of immunoreactivity was similar to that of SP binding sites (Mantyh et al., 1996). Thus, the NK1-like immunoreactivity (NK1-LI) seen by using these antisera most likely is representative of NK1 receptors. A rat monoclonal antibody against SP (Sera Lab, Crawley Down, Sussex, United Kingdom) was obtained commercially from Accurate Chemicals (Westbury, NY). A goat polyclonal antiserum against a C-terminal peptide (amino acids 511–530), as predicted from the cloned rat VAchT (Erickson et al., 1994; Roghani et al., 1994), was obtained from DiaSorin (Stillwater, MN). These antisera have been well characterized and shown to maximally recognize the respective antigens (Milner et al., 1988; Arvidsson et al., 1997).
Animals and fixation procedure Male Sprague-Dawley rats (250 – 400 g) were purchased from Hilltop Laboratory Animals (Scottsdale, PA). The experimental protocol for use of these animals strictly conformed with the National Institutes of Health guidelines for the care and use of laboratory animals and was approved by the Institutional Animal Care and Use Committee of Weill Medical College of Cornell University. The rats were anesthetized with sodium pentobarbital (100 mg/kg, i.p.) and perfused through the aortic arch with 1) 5–10 ml of heparin-saline, 2) 50 ml of 3.8% acrolein in 2% paraformaldehyde, and 3) 200 ml of 2.0% paraformaldehyde. These fixative solutions were prepared in 0.1 M phosphate buffer, pH 7.4 (PB). The brains were removed from the cranium and cut into 2– 4 mm coronal slices of tissue, which were placed in the 2% paraformaldehyde solution for 30 minutes. The slices through the NAc were sectioned at a thickness of 40 m on a Vibratome in PB. These were cut at a level 0.7–1.8 mm anterior to Bregma, according to the rat brain atlas of Paxinos and Watson (1986). To enhance penetration of immunoreagents, the sections were cryoprotected in 25% sucrose and 3.5% glycerol in PB, rapidly frozen in freon followed by liquid nitrogen, and thawed in room temperature PB. These sections were processed for immunocytochemical labeling by using peroxidase and/or gold markers (Leranth and Pickel, 1989).
Immunolabeling For single immunoperoxidase labeling of NK1 receptors, free-floating coronal sections through the NAc were incubated for 48 hours at 4°C in rabbit NK1 receptor antiserum at 1:5,000 dilution in 0.1% bovine serum albumin in Tris-buffered saline. The sections were rinsed and placed for 30 minutes in biotinylated goat anti-rabbit immunoglobulin G (IgG; 1:400 dilution) and avidin-biotin peroxidase complex (ABC; Vector Elite Kit; Vector Laboratories, Burlingame, CA). For peroxidase labeling of SP and immunogold labeling of NK1 receptors, the Vibratome sections were incubated for 48 hours at 4°C in a solution containing rat SP antiserum and rabbit NK1 receptor antisera at dilutions of 1:2,000 and 1:1,000, respectively. After this incubation,
502
V.M. PICKEL ET AL.
the sections of tissue were rinsed and placed for 30 minutes in the biotinylated secondary donkey anti-rat IgG followed by the ABC complex for detection of rat SP antiserum. The bound peroxidase was identified by reaction of the sections for 6 minutes in 3,3⬘-diaminobenzidine (Aldrich Chemicals, Milwaukee, WI) and hydrogen peroxide. After the diaminobenzidine reaction, the sections were rinsed in Tris buffer, pH 7.6, and placed for 2 hours in a 1:50 dilution of goat anti-rabbit IgG with bound 1-nm colloidal gold (Amersham, Arlington, IL) for detection of the rabbit NK1 receptor antiserum. The gold particles were enlarged for microscopic examination by reaction for 6 minutes at room temperature in a silver solution from the IntenS-EM kit (Amersham), as described previously (Chan et al., 1990). A similar dual-labeling protocol was used for immunoperoxidase labeling of VAchT and immunogold labeling of NK1 receptor. In this series, the SP antiserum was replaced by a goat VAchT antiserum at a dilution of 1:16,000, and the donkey anti-rat IgG was replaced by donkey anti-goat IgG as the biotinylated secondary antiserum. Potential artifactual dual labeling attributed to interactions involving donkey anti-goat IgG and goat antirabbit IgG used for detection of the rabbit NK1 receptor antiserum was excluded in control experiments by deleting each of the respective primary antisera. For light microscopy, the single- and dual-labeled sections were mounted on glass slides, dehydrated, coverslipped, and examined on a Nikon microscope (Nikon Inc., Garden City, NY). For electron microscopy, the sections were postfixed in 2% osmium tetroxide in 0.1 M phosphate buffer, dehydrated, and flat embedded between two pieces of Aclar plastic. Ultrathin sections from the outer surface of each Vibratome section in the region of the ventromedial NAc shell and core were collected onto grids by using an LKB ultratome (LKB, Bromma, Sweden). The sections on grids were counterstained with Reynold’s lead citrate and uranyl acetate and were examined by using a Philips CM-10 electron microscope.
Electron microscopic data analysis In the single-labeling study of NK1 receptors, nonoverlapping, ultrathin sections were examined from the NAc shell and core in ten Vibratome sections from three animals. The cellular and subcellular distributions of the receptor were compared visually in the two regions. In the dual-labeling study of SP and NK1 receptors, we semiquantitatively examined the cellular relationships between the respective markers in the ventromedial shell also from ten Vibratome sections through the NAc of three rat brains. In this analysis, all NK1-immunoreactive dendrites that were seen in a single thin section were stored at a magnification of ⫻7,900 by using specimen relocation software on the CM-10 electron microscope. Subsequently, each of these profiles was examined at higher magnification and binned according to 1) their content of SP, 2) the types of labeling seen in afferent axons, and 3) synaptic membrane specializations (absent, symmetric, or asymmetric). All SP-labeled axonal profiles were collected in a similar manner from ultrathin sections from four Vibratome sections in each of two animals. These SP-labeled profiles were subdivided according to the presence or absence of NK1 receptors in the same axons terminals or their targets.
In the dual-labeling study of VAchT and NK1 receptors, we examined the cellular distribution of the antigens in the ventromedial NAc shell from six Vibratome sections taken from four animals. In this analysis, all NK1immunoreactive dendritic profiles in each thin section were stored as described above and subsequently were examined to establish whether they 1) contained VAchT or 2) received synaptic input from VAchT-immunoreactive terminals. In a separate analysis of the same thin sections, all VAchT-containing dendritic profiles were binned and then grouped according to whether or not they contained immunogold labeling for NK1 receptors. Profiles were considered to be immunogold labeled selectively when two or more gold particles were seen in contact with the plasmalemma or limiting membranes of cytoplasmic organelles. This most likely led to an underestimation of immunogold labeling for NK1 receptors in smaller profiles, because the association of even one gold particle with the plasma membrane has been shown to be indicative of selective labeling for surface proteins (Garzon et al., 1999). In the dual-labeling experiments, the frequencies of associations between NK1 receptors and SP- or VAchT-containing profiles also is likely to be under represented mainly due to differences in sensitivities of the peroxidase and gold-labeling methods (Leranth and Pickel, 1989).
Illustrations Photomicrographs that were used for illustrations were scanned on a Power Macintosh 8500/120 Computer (Apple Computer, Inc., Cupertino, CA) with an AGFA Arcus II scanner (Agfa-Gevaert NV., Montsel, Belgium) in combination with FotoLook (Agfa-Gevaert NV) and Photoshop (version 5.0; Adobe Systems Inc., Mountain View, CA) software. QuarkXPress (version 3.32; Quark, Inc., Denver, CO) and Adobe Illustrator (version 7.0; Adobe Systems Inc.) software were then used to prepare and label the composite figures.
RESULTS Under the light microscope, NK1-LI was seen in a dense distribution along the perimeter of somata and thick, straight processes (Fig. 1A). These neurons were dispersed throughout the NAc shell and core. In sections that were processed for NK1 receptor and SP localization, somata and dendrites showing immunogold NK1-LI were seen readily in the neuropil; but peroxidase labeling for SP in varicose processes was resolved poorly by light microscopy (Fig. 1B). Dual localization of NK1 receptors and VAchT showed that many of the larger neurons, but few of the smaller neurons, contained both markers. Electron microscopy confirmed the somatodendritic, plasmalemmal localization of NK1 receptors in NAc neurons that received input from SP-labeled terminals and in aspiny neurons, only some of which contained VAchT. The cellular and subcellular distributions of the antigens were similar qualitatively in the NAc shell and core. Thus, the descriptive results apply equally to each region, and photomicrographs were chosen from each region as representative of the labeling. For simplicity, the quantitative analysis of NK1-immunolabeled profiles was done exclusively in the ventromedial NAc shell.
NK-1 RECEPTORS IN NUCLEUS ACCUMBENS
503
Fig. 1. Light microscopic localization of neurokinin 1 (NK1) receptors and substance P (SP) in the nucleus accumbens (NAc) shell. A: Single immunoperoxidase labeling for NK1 receptors shows a prominent distribution of reaction product along the perimeter and within isolated large (L-NK1) and medium-sized (M-NK1) neuronal somata. The labeling also is seen within thick, straight processes,
presumed to be mainly dendrites (NK1-d). B: Immunogold labeling for NK1 receptors and immunoperoxidase labeling for SP in a section that was processed for dual labeling. The gold particles are associated with profiles similar to those identified in A (L-NK1, M-NK1, and NK1-d), whereas SP labeling is associated with dot-like varicosities (SP-V). Scale bars ⫽ 50 m.
Somatodendritic NK1 receptor distribution
were seen near the Golgi lamellae and the outer nuclear membrane (Fig. 3B). In addition, NK1 receptors were localized to the outer membranes of endosomes and/or multivesicular bodies that were seen near dendritic plasma membranes (Figs. 2C, 4A).
Immunoperoxidase labeling for NK1 receptors was localized intensely to neuronal perikarya (Fig. 2A) that had indented nuclei, which are typical of aspiny striatal neurons (DiFiglia et al., 1980). In somata, NK1-LI was seen mainly on the plasma membrane by using either immunoperoxidase (Fig. 2A) or immunogold (Fig. 3) markers. The immunoreactivity was prevalent beneath selective, unlabeled axon terminals, but it also was seen at extrasynaptic sites on the plasma membrane (Figs. 2A, 3B). Appositional contacts were seen sometimes between pairs of somata, both of which showed plasmalemmal labeling for the NK1 receptor (Fig. 3A). Plasmalemmal targeting of NK1 receptors also was seen in large (Fig. 2B) and small (Fig. 2C) dendrites that usually were without spines. These dendrites frequently showed appositional contacts with unlabeled dendrites and received mainly symmetric synapses from unlabeled terminals. The NK1 receptor labeling in dendrites often was located at or near the junctions formed by the unlabeled terminals (Fig. 2B). In somata and dendrites, cytoplasmic labeling for NK1 receptors was sparse and often was associated with saccules of smooth endoplasmic reticulum or tubulovesicles that were located near immunolabeled portions of the plasma membrane (Fig. 3B). These vesicles, however, also were distributed widely in the cytoplasm and sometimes
SP- and/or NK1-containing terminals SP was localized intensely to axon terminals within the neuropil near somata that contained NK1 receptor immunogold labeling (Fig. 3A). The SP-labeled terminals also occasionally contacted somata that contained NK1 receptors (Fig. 3B). The appositional contacts typically showed no dense membrane specializations, and immunogold particles were located within and along the perimeter of the appositions, sometimes contacting subsurface tubulovesicles that also were immunolabeled for the NK1 receptor (Fig. 3B). Appositions also were seen between SPimmunoreactive terminals and adjacent labeled or unlabeled terminals contacting the same soma (Fig. 3B). Compared with somata, SP-labeled terminals more often formed synapses with dendrites that contained NK1LI. Of 603 dendrites that contained NK1 receptors that were seen in the NAc shell, 46% were apposed to or received synapses from SP-labeled terminals, and 40% were apposed to or received synapses from unlabeled axon terminals. The remaining terminals were without recognizable synaptic input. The synaptic junctions formed by the
504
V.M. PICKEL ET AL.
Fig. 2. Peroxidase labeling for NK1 receptors in somata and dendrites. A: In the NAc core, the peroxidase reaction product (arrowheads) is seen prominently along portions of the plasma membrane contacted by unlabeled terminals (UT). Patches of immunoreactivity also are seen along nonsynaptic portions of the plasma membrane and throughout the cytoplasm. B: In the NAc shell, peroxidase labeling is present in a dendrite (NK1-d) that shows an appositional contact with an unlabeled dendrite (UD) and an unlabeled terminal (UT). The terminal forms a symmetric synapse (open curved arrow) with another unlabeled neuronal profile. The most intense peroxidase reaction product (arrowheads) is seen on the plasma membrane of NK1-d
in contact with a UT. Glial processes (asterisks) appose other portions of the dendritic plasma membrane. C: In the NAc core, immunoreactivity is seen in a diffuse distribution throughout a small dendrite. Somewhat more intense localization of the reaction product (arrowhead) is seen along the limiting membrane of a multivesicular body (MVB) and membranes of nearby tubulovesicles (TV) located near a labeled segment of the extrasynaptic plasma membrane apposing astrocytic processes (asterisks). The dendrite receives an asymmetric synapse (solid curved arrow) from an unlabeled terminal (UT). Go, Golgi lamellae. Scale bars ⫽ 1.0 m in A,B; 0.5 m in C.
SP-labeled terminals as well as the unlabeled terminals were mainly symmetric. In these dendrites, the NK1 receptor immunogold labeling was seen within and near synapses formed by the SP-containing terminals as well as the nearby membranes of endosomes and multivesicu-
lar organelles (Fig. 4A). In contrast with the relatively high proportion of NK1-containing dendrites that received input from SP-labeled terminals, relatively few of the SP-immunoreactive terminals contacted NK1-labeled dendrites. Of 912 SP-immunoreactive terminals that were
NK-1 RECEPTORS IN NUCLEUS ACCUMBENS
Fig. 3. Electron photomicrographs from the ventromedial NAc shell in sections that were processed for immunogold labeling of NK1 receptors and immunoperoxidase localization of SP. A: Gold particles (arrows) are distributed mainly on the plasma membrane of a pair of apposed somata. The nucleus (Nu) of each cell has a deeply indented nuclear membrane. An SP-containing terminal (SP-t) is located nearest to the somatic plasma membrane of one of the neurons, but several other SP-labeled terminals also are seen within the adjacent neuropil. B: Gold particles (arrows) identify NK1 receptors mainly on extrasynaptic portions of the somatic plasma membrane of a large neuron
505
containing a nucleus (Nu), the membrane of which also is indented outside the field of view. The particles on the membrane are seen near an appositional contact from a terminal containing dense peroxidase reaction product for SP (SP-t) as well as at sites contacting other unlabeled neuronal and glial profiles. The gold particles also are localized to tubulovesicles (TV) resembling smooth endoplasmic reticulum beneath the plasma membrane and in the cytoplasm sometimes near trans Golgi lamellae (Go) and the outer nuclear membrane. Scale bars ⫽ 1.0 m.
Fig. 4. Postsynaptic and presynaptic immunogold labeling for NK1 receptors in dendrites receiving input from SP-labeled terminals and in axon terminals with or without SP in the NAc shell. A: Gold particles (solid straight arrows) within the dendrite (NK1-d) are distributed mainly on portions of the plasma membrane near symmetric or appositional contacts (open curved arrows) from axon terminals that contain peroxidase immunoreactivity for SP (SP-t). The gold labeling also is seen near a contact from an unlabeled terminal (UT) and along the outer membrane of a multivesicular body (MVB). B: Immunoperoxidase labeling (arrowheads) for NK1 receptors in an axon terminal (NK1-t) that forms a symmetric synapse
(curved arrow) with an unlabeled dendrite (UD). C: Gold particles (solid straight arrows) are localized on the plasma membrane and cytoplasm of an axon terminal (NK1-t) that forms a symmetric synapse (open curved arrow) with an unlabeled dendrite. D: In the NAc shell, gold particles (solid straight arrows) identifying NK1 receptors are seen on extrasynaptic portions of the plasma membranes of an axon terminal showing peroxidase immunoreactivity for SP (NK1 and SP-t) and an apposed terminal (T) without peroxidase immunoreactivity. The dual-labeled terminal forms an apparent symmetric synapse (open curved arrow) with a small spine of an unlabeled dendrite (UD). Scale bars ⫽ 0.5 m.
NK-1 RECEPTORS IN NUCLEUS ACCUMBENS observed in the ventromedial NAc shell, only 7% were in contact with dendrites containing NK1 receptors, whereas 62% contacted unlabeled dendrites, and the remainder either apposed unlabeled (21%) or SP-labeled (14%) terminals or glia (13%). These percentages constitute 118% due to the fact that some of the terminals contacted more than one neuronal or glial profile within the plane of section. A few axon terminals also contained NK1 receptors and/or SP immunoreactivity (Fig. 4B–D) in the NAc shell and core. Immunoperoxidase labeling for NK1 receptors was distributed diffusely in axon terminals, with a particularly dense distribution on plasma membranes and membranes of clusters of small synaptic vesicles (Fig. 4B). Immunogold NK1 receptor labeling also showed a plasmalemmal and cytoplasmic distribution in selective axon terminals (Fig. 4C). The number of gold particles in individual terminals usually was no more than two or three, but their discrete localization on the plasma membrane provided convincing evidence of their specificity. The NK1-labeled terminals formed symmetric synapses or were without recognizable synaptic junctions. In sections that were processed for dual localization of NK1 receptors and SP, NK1 immunogold labeling also was seen sometimes in axon terminals that contained peroxidase reaction product for SP (Fig. 4D). These formed symmetric axodendritic or axospinous synapses and often apposed other unlabeled axon terminals.
Cholinergic identity of neurons containing NK1 receptors Isolated somata and dendrites contained NK1 receptors and/or VAchT in the NAc shell and core. Of 534 immunoreactive dendrites that were seen in the ventromedial shell from tissue processed for VAchT/ NK1 receptor dual labeling, 61% contained only NK1-LI, 29% contained NK1-LI and VAchT, and 10% contained only VAchT. In these dendrites, VAchT labeling was distributed diffusely in the cytoplasm or was associated with membranes of tubulovesicles, whereas NK1 receptors showed the characteristic extrasynaptic distribution (Fig. 5A). Compared with somata and dendrites, VAchT was localized much more prominently to membranes of small synaptic vesicles in axon terminals. These terminals apposed or formed symmetric synapses with dendrites that contained NK1 receptors and VAchT (Fig. 5A) as well as those that expressed only NK1 receptors (Fig. 5B,C). Of 912 VAchT-immunoreactive terminals that were observed in the ventromedial shell, approximately 5% contacted dendrites that contained NK1 receptors and/or VAchT. The remainder formed symmetric synapses with unlabeled dendrites and/or apposed unlabeled terminals. Appositional contacts with unlabeled terminals also were seen commonly between VAchT-immunoreactive and nonimmunoreactive terminals presynaptic to dendrites that showed NK1 receptor labeling (Fig. 5B). In these dendrites, immunogold labeling for NK1 receptors was seen mainly on nonsynaptic portions of the plasma membrane in contact with astrocytic processes. In a few examples, these astrocytes were apposed to the basal laminae of blood vessels (Fig. 5C). NK1-LI was not seen within the synaptic junctions that were formed by the VAchTimmunoreactive terminals, and it was seen only occasionally within or near the contacts that were formed by unlabeled terminals on the same dendrite.
507
DISCUSSION In the current study, we have shown that, in the NAc shell and core, NK1 receptors are targeted mainly to plasma membranes and are associated more rarely with endosomal organelles in the somata and dendrites of aspiny neurons. Furthermore, NK1 receptors were localized to dendritic plasma membranes within and near symmetric synapses formed by SP-containing terminals, but these terminals more often contacted unlabeled dendrites and occasionally also contained NK1 receptors. Presynaptic NK1 receptors also were seen in a few other axon terminals that formed symmetric synapses but did not contain SP. Together, these results suggest that NK1 receptors in the NAc are primary mediators of the postsynaptic actions of SP at inhibitory-type synapses and also may play a role in the presynaptic release of SP or inhibitory amino acids. The present data also provide morphologic support for the involvement of NK1 receptors in nonsynaptic communication between NAc neurons, including those that contain acetylcholine and receive synaptic input from cholinergic terminals.
Preferential somatodendritic targeting of NK1 receptors in aspiny neurons The observed prominent plasmalemmal and modest endosomal distribution of NK1 receptors in NAc neurons, together with recent evidence that SP induces rapid internalization of striatal plasmalemmal NK1 receptors (Mantyh et al., 1995), suggest that, under resting conditions, relatively small amounts of SP may be available for the activation of NK1 receptors in the NAc. In other systems, however, NK1 receptors are desensitized through phosphorylation that does not require endocytosis, and they become activated through phosphatases within endosomes (Garland et al., 1996). Thus, we cannot exclude the possibility that the immunolabeling on somatodendritic plasma membrane in the present study reflects inactive receptor proteins that are activated within the endosomes (Garland et al., 1996). In either case, the multivesicular bodies or endosomes are likely sites for processing NK1 receptors that are targeted to somatodendritic plasma membranes (Parton et al., 1992; Parton, 1993; Odorizzi et al., 1998). We most frequently saw NK1 receptors along nonsynaptic portions of somatodendritic plasma membranes and either within or near symmetric axodendritic synapses in the NAc. This distribution suggests that NK1 receptors are activated by SP that diffuses from the synaptic cleft of the presynaptic axons or is released by exocytosis from these or other axon terminals within the neuropil (Thureson-Klein et al., 1986). Release of SP from terminals forming symmetric synapses is consistent with the known presence of SP in ␥-aminobutyric acidergic (GABAergic) terminals that are derived from striatal projection neurons that also form symmetric synapses (Bolam and Smith, 1990). We also observed, however, that NK1 receptors are present on synaptic and apposed somatodendritic plasma membranes beneath SP-labeled terminals. These results are consistent with the idea that extrasynaptic NK1 receptors may be mobilized within the postsynaptic plasma membrane to sites of SP release, as shown for GABAA and for several other receptors (Craig et al., 1994; Ciruela et al., 1997; Kneussel et al., 1999; Savtchenko et al., 2000). Furthermore, the present localization of
508
V.M. PICKEL ET AL.
Figure 5
NK-1 RECEPTORS IN NUCLEUS ACCUMBENS
509
NK1 receptors to symmetric synapses comparable to those that display GABAA receptors (Somogyi et al., 1996) suggests that SP and GABA may share some of the same functional sites on target neurons in the NAc. Conceivably, these sites may be ion channels, because NK1 receptor agonists inhibit the inward currents evoked at GABAA receptors (Akasu and Yamada, 1997). In contrast to the prominent dendritic plasmalemmal distribution of NK1 receptors, we saw NK1 receptor labeling in relatively few axon terminals, and, in these terminals, the immunoreactivity was associated more often with cytoplasmic vesicles, suggesting potentially less involvement of NK1 receptors in the presynaptic release of neurotransmitters. The observed NK1-immunoreactive terminals were apposed or formed symmetric synapses that are typical of GABAergic terminals (Ribak et al., 1979). SP also is known to modulate the release of inhibitory amino acids (Maehara et al., 1995). Together, these results suggest that SP activation of NK1 receptors modulates mainly the postsynaptic responses but also can alter the presynaptic release of GABA. This role is consistent with recent evidence showing that NK1 receptor antagonists are anxiolytic in tests of social interactions (File, 1997), a function shared by drugs that potentiate GABA binding to GABAA receptors (Choleris et al., 1998). Furthermore, because SP is present in subpopulations of striatal GABAergic neurons (Gerfen, 1991), our localization of NK1 receptors within SP-containing terminals also largely may reflect involvement of the receptor in the presynaptic release of GABA. Alternatively, these NK1 receptors may serve to autoregulate presynaptically the release of SP, as shown in certain sensory neurons (Malcangio and Bowery, 1999; Von Banchet and Schaible, 1999).
1991). Such appositions are likely sites for communication through gap junctions that may facilitate synchronization of tonically active interneurons (Raz et al., 1996). SP may play a direct role in the adhesion between pairs of striatal neurons by promoting the expression of one or more of the family of adhesion molecules, including limbic systemassociated membrane protein, the distribution of which is dependent on SP innervation (Zhang et al., 1998). This potential function is supported by the known involvement of SP in stimulating the production of adhesion molecules in lymphocytes and endothelial cells in inflammation (Vishwanath and Mukherjee, 1996; Quinlan et al., 1999). A few somata and many dendrites with the features of aspiny neurons contained NK1 receptors but were without detectable VAchT immunoreactivity in the current study. Many of the noncholinergic neurons are likely to contain somatostatin (Kaneko et al., 1993). Striatal neurons that fire low-threshold spikes and that contain both somatostatin and NO synthase, as well as those that have longer lasting afterhyperpolarizations and contain acetylcholine, are excited by SP (Kawaguchi et al., 1997). The possibility that the noncholinergic neurons containing NK1 receptors produce NO is of interest, because internalization of NK1 receptors together with SP may increase the availability of L-arginine, which is an N-terminal metabolite of SP and a substrate for NO synthesis (Goettl and Larson, 1996). This may be particularly important in dendrites that contain NK1 receptors apposing astrocytic processes that contact the basement membrane of blood vessels in the current study, because the activation of these receptors may produce vasodilation through the heightened release of NO (Ralevic et al., 1995).
Coexpression of NK1 receptors and VAchT
Cholinergic input to dendrites containing NK1 receptors
Most neurons that contained VAchT in the current study also expressed NK1 receptors. These results are consistent with light microscopic studies showing that NK1 receptors and their mRNA are present in striatal cholinergic neurons (Gerfen, 1991; Kaneko et al., 1993). The NK1 receptor-immunoreactive neurons also showed somatic and dendritic plasmalemmal appositions that are typical of cholinergic striatal neurons (Pickel and Chan,
Fig. 5: Immunogold labeling for NK1 receptors in dendrites that contain the vesicular acetylcholine transporter (VAchT) or that receive input from VAchT-labeled terminals. A: In the NAc core, immunogold labeling for NK1 (arrows) is seen along the plasma membrane of a large proximal dendrite showing diffuse peroxidase labeling for VAchT. More intense VAchT labeling is associated with membranes of tubulovesicles (TV). VAchT-peroxidase reaction product also is seen in an axon terminal (VAchT-t) that is presynaptic to the dual-labeled dendrite. The dendrite is apposed on most surfaces by small, unmyelinated axons (UA) and unlabeled axon terminals (UT). B,C: In the NAc shell, VAchT peroxidase labeling is localized to membranes of synaptic vesicles in axon terminals (VAchT-t) that form symmetric synapses (open curved arrows) with dendrites (NK1-d) showing immunogold NK1 receptor distribution. The gold particles (solid straight arrows) are seen mainly on portions of the dendritic plasma membrane apposed to astrocytic processes (asterisks) and occasionally near contacts from unlabeled terminals (UT) but not from VAchTlabeled terminals. The astrocytic processes separate NK1-d from the basement membrane (BM) of a blood vessel shown in B. Scale bars ⫽ 1.0 m.
The VAchT-immunoreactive terminals that were observed in the NAc shell and core in the current study were similar morphologically to those that have been described as containing choline acetyltransferase in the dorsal striatum (Contant et al., 1996). These terminals formed symmetric synapses with aspiny neurons, some of which expressed NK1 receptors on their plasma membrane, suggesting that the output from these cells is dually modulated by SP and acetylcholine acting at either nicotinic receptors or muscarinic receptors. There is indirect evidence suggesting that one or both types of cholinergic receptors have functional associations with NK1 receptors. SP shows agonist-induced, high-affinity binding to the nicotinic receptor (Min et al., 1994), and both NK1 and nicotinic receptors play a role in reward associated with the activation of mesolimbic dopaminergic neurons (Tzschentke, 1998). Muscarinic receptors also are known to be present in cholinergic and smaller noncholinergic striatal neurons (Smiley et al., 1999). This distribution is similar to that of NK1 receptors in the present study and in earlier studies (Kaneko et al., 1993). Furthermore, muscarinic receptors are known be postsynaptic autoreceptors (Kitaichi et al., 1999), and we observed input from cholinergic terminals to neurons containing NK1 receptors and/or VAchT. Together, these results suggest that NK1 receptor agonists and muscarinic and/or nicotinic receptor agonists control the activity of interneurons in the NAc.
510
V.M. PICKEL ET AL.
Functional implications The current results suggest that the functions ascribed to NK1 receptors in the NAc are attributed mainly to the modulation of inhibitory postsynaptic responses in cholinergic and noncholinergic, aspiny neurons. These actions, like those ascribed to GABAA, may play a selective role in motivated behaviors as well as social anxiety and reward (Mattioli et al., 1997; Login et al., 1998; Basso and Kelley, 1999; Dib, 1999; Rupniak and Kramer, 1999). The diverse functions also may be influenced by NK1 receptormediated changes in the expression of adhesion molecules that provide synchronization within neuronal clusters and in the release of NO that affects neighboring neurons and blood vessels.
ACKNOWLEDGMENTS This work was supported by National Institutes of Health grants MH00078, MH40342, DA04600, and HL18974 to V.M.P and National Institutes of Health grant DK39957 to N.W.B.
LITERATURE CITED Akasu T, Yamada K. 1997. Neurokinin-1 (NK1) receptors mediate tachykinin-induced depression of GABA current in bullfrog sensory neurons. Kurume Med J 44:33– 41. Arvidsson U, Riedl M, Elde R, Meister B. 1997. Vesicular acetylcholine transporter (VAchT) protein: a novel and unique marker for cholinergic neurons in the central and peripheral nervous systems. J Comp Neurol 378:454 – 467. Barrot M, Marinelli M, Abrous DN, Rouge´-Pont F, Le Moal M, Piazza PV. 1999. Functional heterogeneity in dopamine release and in the expression of Fos-like proteins within the rat striatal complex. Eur J Neurosci 11:1155–1166. Basso AM, Kelley AE. 1999. Feeding induced by GABAA receptor stimulation within the nucleus accumbens shell: regional mapping and characterization of macronutrient and taste preference. Behav Neurosci 113:324 –336. Bell MI, Richardson PJ, Lee K. 1998. Characterization of the mechanism of action of tachykinins in rat striatal cholinergic interneurons. Neuroscience 87:649 – 658. Bennett GW, Ballard TM, Watson CD, Fone KC. 1997. Effect of neuropeptides on cognitive function. Exp Gerontol 32:451– 469. Bolam JP, Smith Y. 1990. The GABA and substance P input to dopaminergic neurons in the substantia nigra of the rat. Brain Res 529:57–78. Chan J, Aoki C, Pickel VM. 1990. Optimization of differential immunogoldsilver and peroxidase labeling with maintenance of ultrastructure in brain sections before plastic embedding. J Neurosci Methods 33:113– 127. Choleris E, Valsecchi P, Wang Y, Ferrari P, Kavaliers M, Mainardi M. 1998. Social learning of a food preference in male and female Mongolian gerbils is facilitated by the anxiolytic, chlordiazepoxide. Pharmacol Biochem Behav 60:575–584. Ciruela F, Saura C, Canela EI, Mallol J, Lluı´s C, Franco R. 1997. Ligandinduced phosphorylation, clustering, and desensitization of A1 adenosine receptors. Mol Pharmacol 52:788 –797. Contant C, Umbriaco D, Garcia S, Watkins KC, Descarries L. 1996. Ultrastructural characterization of the acetylcholine innervation in adult rat neostriatum. Neuroscience 71:937–947. Craig AM, Blackstone CD, Huganir RL, Banker G. 1994. Selective clustering of glutamate and gamma-aminobutyric acid receptors opposite terminals releasing the corresponding neurotransmitters. Proc Natl Acad Sci USA 91:12373–12377. Dib B. 1999. Food and water intake suppression by intracerebroventricular administration of substance P. Brain Res 830:38 – 42. DiFiglia M, Pasik T, Pasik P. 1980. Ultrastructure of Golgi-impregnated and gold-toned spiny and aspiny neurons in the monkey neostriatum. J Neurocytol 9:471– 492. Erickson JD, Varoqui H, Scha¨fer MKH, Modi W, Diebler M-F, Weihe E, Rand J, Eiden LE, Bonner TI, Usdin TB. 1994. Functional identifica-
tion of a vesicular acetylcholine transporter and its expression from a “cholinergic” gene locus. J Biol Chem 269:21929 –21932. Everitt BJ, Robbins TW. 1997. Central cholinergic systems and cognition. Annu Rev Psychol 48:649 – 684. File SE. 1997. Anxiolytic action of a neurokinin1 receptor antagonist in the social interaction test. Pharmacol Biochem Behav 58:747–752. Galarraga E, Herna´ndez-Lo´pez S, Tapia D, Reyes A, Bargas J. 1999. Action of substance P (neurokinin-1) receptor activation on rat neostriatal projection neurons. Synapse 33:26 –35. Garland AM, Grady EF, Lovett M, Vigna SR, Frucht MM, Krause JE, Bunnett NW. 1996. Mechanisms of desensitization and resensitization of G protein-coupled neurokinin1 and neurokinin2 receptors. Mol Pharmacol 49:438 – 446. Garzon M, Vaughan RA, Uhl GR, Kuhar MJ, Pickel VM. 1999. Cholinergic axon terminals in the ventral tegmental area target a subpopulation of neurons expressing low levels of the dopamine transporter. J Comp Neurol 410:197–210. Gerfen CR. 1988. Synaptic organization of the striatum. J Electron Microsc Tech 10:265–281. Gerfen CR. 1991. Substance P (neurokinin-1) receptor mRNA is selectively expressed in cholinergic neurons in the striatum and basal forebrain. Brain Res 556:165–170. Goettl VM, Larson AA. 1996. Nitric oxide mediates long-term hyperalgesic and antinociceptive effects of the N-terminus of substance P in the formalin assay in mice. Pain 67:435– 441. Grady EF, Baluk P, Bo¨hm S, Gamp PD, Wong H, Payan DG, Ansel J, Portbury AL, Furness JB, McDonald DM, Bunnett NW. 1996. Characterization of antisera specific to NK1, NK2, and NK3 neurokinin receptors and their utilization to localize receptors in the rat gastrointestinal tract. J Neurosci 16:6975– 6986. Ikemoto S, Glazier BS, Murphy JM, McBride WJ. 1998. Rats selfadminister carbachol directly into the nucleus accumbens. Physiol Behav 63:811– 814. Kaneko T, Shigemoto R, Nakanishi S, Mizuno N. 1993. Substance P receptor-immunoreactive neurons in the rat neostriatum are segregated into somatostatinergic and cholinergic aspiny neurons. Brain Res 631:297–303. Kawaguchi Y, Aosaki T, Kubota Y. 1997. Cholinergic and GABAergic interneurons in the striatum. Nihon Shinkei Seishin Yakurigaku Zasshi 17:87–90. Kelley AE. 1999. Neural integrative activities of nucleus accumbens subregions in relation to learning and motivation. Psychobiology 27:198 – 213. Khawaja AM, Rogers DF. 1996. Tachykinins: receptor to effector. Int J Biochem Cell Biol 28:721–738. Kitaichi K, Hori T, Srivastava LK, Quirion R. 1999. Antisense oligodeoxynucleotides against the muscarinic m1, but not m4, receptor supports its role as autoreceptors in the rat hippocampus. Brain Res Mol Brain Res 67:98 –106. Kneussel M, Brandsta¨tter JH, Laube B, Stahl S, Mu¨ller U, Betz H. 1999. Loss of postsynaptic GABAA receptor clustering in gephyrin-deficient mice. J Neurosci 19:9289 –9297. Kowall NW, Quigley BJJ, Krause JE, Lu F, Kosofsky BE, Ferrante RJ. 1993. Substance P and substance P receptor histochemistry in human neurodegenerative diseases. Regul Peptides 46:174 –185. Lee T, Kaneko T, Shigemoto R, Nomura S, Mizuno N. 1997. Collateral projections from striatonigral neurons to substance P receptorexpressing intrinsic neurons in the striatum of the rat. J Comp Neurol 388:250 –264. Leranth C, Pickel VM. 1989. Electron microscopic pre-embedding double immunostaining methods. In: Heimer L, Zaborszky L, editors. Neuroanatomical tract-tracing methods 2: recent progress. New York: Plenum Publishing Co. p 129 –172. Login IS, Pal SN, Adams DT, Gold PE. 1998. Muscimol increases acetylcholine release by directly stimulating adult striatal cholinergic interneurons. Brain Res 779:33– 40. Li JL, Wang D, Kaneko T, Shigemoto R, Nomura S, Mizuno N. 2000. Relationship between neurokinin-1 receptor and substance P in the striatum: light and electron microscopic immunohistochemical study in the rat. J Comp Neurol 418:156 –163. Lu XY, Ghasemzadeh MB, Kalivas PW. 1998. Expression of D1 receptor, D2 receptor, substance P and enkephalin messenger RNAs in the neurons projecting from the nucleus accumbens. Neuroscience 82:767– 780. Maehara T, Suzuki H, Yoshioka K, Otsuka M. 1995. Characteristics of
NK-1 RECEPTORS IN NUCLEUS ACCUMBENS substance P-evoked release of amino acids from neonatal rat spinal cord. Neuroscience 68:577–584. Malcangio M, Bowery NG. 1999. Peptide autoreceptors: does an autoreceptor for substance P exist. Trends Pharmacol Sci 20:405– 407. Mantyh PW, Allen CJ, Ghilardi JR, Rogers SD, Mantyh CR, Liu H, Basbaum AI, Vigna SR, Maggio JE. 1995. Rapid endocytosis of a G proteincoupled receptor: substance P evoked internalization of its receptor in the rat striatum in vivo. Proc Natl Acad Sci USA 92:2622–2626. Mantyh PW, Rogers SD, Ghilardi JR, Maggio JE, Mantyh CR, Vigna SR. 1996. Differential expression of two isoforms of the neurokinin-1 (substance P) receptor in vivo. Brain Res 719:8 –13. Martone ME, Armstrong DM, Young SJ, Groves PM. 1992. Ultrastructural examination of enkephalin and substance P input to cholinergic neurons within the rat neostriatum. Brain Res 594:253–262. Mattioli R, Santangelo EM, Costa AC, Vasconcelos L. 1997. Substance P facilitates memory in goldfish in an appetitively motivated learning task. Behav Brain Res 85:117–120. Milner TA, Pickel VM, Abate C, Joh TH, Reis DJ. 1988. Ultrastructural characterization of substance P-like immunoreactive neurons in the rostral ventrolateral medulla in relation to neurons containing catecholamine-synthesizing enzymes. J Comp Neurol 270:402– 405,427– 445. Min CK, Owens J, Weiland GA. 1994. Characterization of the binding of [3H]substance P to the nicotinic acetylcholine receptor of Torpedo electroplaque. Mol Pharmacol 45:221–227. Nirenberg MJ, Chan J, Pohorille A, Vaughan RA, Uhl GR, Kuhar MJ, Pickel VM. 1997. The dopamine transporter: comparative ultrastructure of dopaminergic axons in limbic and motor compartments of the nucleus acccumbens. J Neurosci 17:6899 – 6907. Odorizzi G, Babst M, Emr SD. 1998. Fab1p PtdIns(3)P 5-kinase function essential for protein sorting in the multivesicular body. Cell 95:847– 858. Parent A, Cicchetti F, Beach TG. 1995. Striatal neurones displaying substance P (NK1) receptor immunoreactivity in human and non-human primates. Neuroreport 6:721–724. Parton RG. 1993. Cell biology of neuronal endocytosis. J Neurosci Res 36:1–9. Parton RG, Simons K, Dotti CG. 1992. Axonal and dendritic endocytic pathways in cultured neurons. J Cell Biol 119:123–137. Paxinos G, Watson C. 1986. The rat brain in stereotaxic coordinates. New York: Academic Press. Perry E, Walker M, Grace J, Perry R. 1999. Acetylcholine in mind: a neurotransmitter correlate of consciousness? Trends Neurosci 22:237– 280. Pickel VM, Chan J. 1991. Plasmalemmal appositions between cholinergic and noncholinergic neurons in rat caudate-putamen nuclei. Neuroscience 41:459 – 472. Quinlan KL, Naik SM, Cannon G, Armstrong CA, Bunnett NW, Ansel JC, Caughman SW. 1999. Substance P activates coincident NF-AT- and NF-kappaB-dependent adhesion molecule gene expression in microvascular endothelial cells through intracellular calcium mobilization. J Immunol 163:5656 –5665. Rada PV, Mark GP, Hoebel BG. 1993. In vivo modulation of acetylcholine in the nucleus accumbens of freely moving rats: II. Inhibition by gamma-aminobutyric acid. Brain Res 619:105–110. Ralevic V, Khalil Z, Helme RD, Dusting GJ. 1995. Role of nitric oxide in the actions of substance P and other mediators of inflammation in rat skin microvasculature. Eur J Pharmacol 284:231–239. Raz A, Feingold A, Zelanskaya V, Vaadia E, Bergman H. 1996. Neuronal synchronization of tonically active neurons in the striatum of normal and Parkinsonian primates. J Neurophysiol 76:2083–2088. Ribak CE, Vaughn JE, Roberts E. 1979. The GABA neurons and their axon terminals in rat corpus striatum as demonstrated by GAD immunocytochemistry. J Comp Neurol 187:261–283.
511 Roghani A, Feldman J, Kohan SA, Shirzadi A, Gundersen CB, Brecha N, Edwards RH. 1994. Molecular cloning of a putative vesicular transporter for acetylcholine. Proc Natl Acad Sci USA 91:10620 –10624. Roghani A, Shirzadi A, Butcher LL, Edwards RH. 1998. Distribution of the vesicular transporter for acetylcholine in the rat central nervous system. Neuroscience 82:1195–1212. Rupniak NMJ, Kramer MS. 1999. Discovery of the anti-depressant and anti-emetic efficacy of substance P receptor (NK1) antagonists. Trends Pharmacol Sci 20:485– 490. Saria A. 1999. The tachykinin NK1 receptor in the brain: pharmacology and putative functions. Eur J Pharmacol 375:51– 60. Sarter M, Bruno JP. 1997. Cognitive functions of cortical acetylcholine: toward a unifying hypothesis. Brain Res Brain Res Rev 23:28 – 46. Satoh K, Staines WA, Atmadja S, Fibiger HC. 1983. Ultrastructural observations of the cholinergic neuron in the rat striatum as identified by acetylcholinesterase pharmacohistochemistry. Neuroscience 10:1121– 1123. Savtchenko LP, Korogod SM, Rusakov DA. 2000. Electrodiffusion of synaptic receptors: a mechanism to modify synaptic efficacy? Synapse 35:26 –38. Schildein S, Ågmo A, Huston JP, Schwarting RK. 1998. Intraaccumbens injections of substance P, morphine and amphetamine: effects on conditioned place preference and behavioral activity. Brain Res 790:185– 194. Setlow B, Roozendaal B, McGaugh JL. 2000. Involvement of a basolateral amygdala complex-nucleus accumbens pathway in glucocorticoidinduced modulation of memory consolidation. Eur J Neurosci 12:367– 375. Smiley JF, Levey AI, Mesulam M. 1999. m2 Muscarinic receptor immunolocalization in cholinergic cells of the monkey basal forebrain and striatum. Neuroscience 90:803– 814. Somogyi P, Fritschy JM, Benke D, Roberts JDB, Sieghart W. 1996. The gamma 2 subunit of the GABAA receptor is concentrated in synaptic junctions containing the ␣1 and 2/3 subunits in hippocampus, cerebellum and globus pallidus. Neuropharmacology 35:1425–1444. Thureson-Klein ÅK, KleinRL, Zhu PC. 1986. Exocytosis from large dense cored vesicles as a mechanism for neuropeptide release in the peripheral and central nervous system. Scan Electron Microsc 87:179 – 87. Tzschentke TM. 1998. Measuring reward with the conditioned place preference paradigm: a comprehensive review of drug effects, recent progress and new issues. Progr Neurobiol 56:613– 672. Vishwanath R, Mukherjee R. 1996. Substance P promotes lymphocyteendothelial cell adhesion preferentially via LFA-1/ICAM-1 interactions. J Neuroimmunol 71:163–171. Von Banchet GS, Schaible HG. 1999. Localization of the neurokinin 1 receptor on a subset of substance P-positive and isolectin B4-negative dorsal root ganglion neurons of the rat. Neurosci Lett 274:175–178. Yee BK. 2000. Cytotoxic lesion of the medial prefrontal cortex abolishes the partial reinforcement extinction effect, attenuates prepulse inhibition of the acoustic startle reflex and induces transient hyperlocomotion, while sparing spontaneous object recognition memory in the rat. Neuroscience 95:675– 689. Yokota Y, Sasai Y, Tanaka K, Fujiwara T, Tsuchida K, Shigemoto R, Kakizuka A, Ohkubo H, Nakanishi S. 1989. Molecular characterization of a functional cDNA for rat substance P receptor. J Biol Chem 264: 17649 –17652. Yung KKL, Smith AD, Levey AI, Bolam JP. 1996. Synaptic connections between spiny neurons of the direct and indirect pathways in the neostriatum of the rat: evidence from dopamine receptor and neuropeptide immunostaining. Eur J Neurosci 8:861– 869. Zhang B, Levitt P, Murray M. 1998. Induction of presynaptic reexpression of an adhesion protein in lamina II after dorsal root deafferentation in adult rat spinal cord. Exp Neurol 149:468 – 472.
