European Journal of Neuroscience, Vol. 25, pp. 2619–2628, 2007
doi:10.1111/j.1460-9568.2007.05523.x
EphB1 null mice exhibit neuronal loss in substantia nigra pars reticulata and spontaneous locomotor hyperactivity A. Brent Richards,1,* Tracy A. Scheel,1 Kan Wang,1 Mark Henkemeyer 2 and Lawrence F. Kromer1 1 2
Department of Neuroscience, Georgetown University, Washington, DC 20007, USA Center for Developmental Biology, University of Texas Southwestern Medical Center, Dallas, TX 75390, USA
Keywords: basal ganglia, development, dopamine, ephrin, mouse, striatum
Abstract The molecular mechanisms that regulate basal ganglia development are largely unknown. Eph receptor tyrosine kinases are potential participants in this process as they regulate development of other CNS regions and are expressed in basal ganglia nuclei, such as the substantia nigra (SN) and striatum. To address the role of Eph receptors in the development of these nuclei, we analysed anatomical changes in the SN and striatum of mice with null mutations for EphB1. These mice express b-galactosidase as a marker for cells normally expressing EphB1. In situ hybridization data and a direct comparison of SN neurons expressing tyrosine hydroxylase (TH) and ⁄ or the b-gal marker for EphB1 revealed that EphB1 is not expressed in TH+ neurons of pars compacta (SNc), but is restricted to neurons in pars reticulata (SNr). Consistent with this, we find that EphB1 null mice exhibit a significant decrease in the volume and number of neurons (40% decrease) in SNr, whereas the volume and number of TH+ neurons in SNc is not significantly affected nor are there changes in the distribution of nigrostriatal dopamine neurons. Although EphB1 is expressed in the striatum, EphB1– ⁄ – mice exhibit no significant changes in striatal volume and TH fiber density, and have no obvious alterations in striatal patch ⁄ matrix organization. Behavioral evaluation of EphB1 null mice in an open-field environment revealed that these mice exhibited spontaneous locomotor hyperactivity. These results suggest that EphB1 is necessary for the proper formation of SNr, and that neuronal loss in SNr is associated with altered locomotor functions.
Introduction The mammalian basal ganglia system is important for integrating motor, sensory and cognitive functions. In addition to its involvement in neurodegenerative disorders, such as Parkinson’s disease and Huntington’s chorea, human functional imaging studies implicate this brain region in several developmental disorders, psychological diseases and addictions. These include Tourette’s syndrome (Nomura & Segawa, 2003), schizophrenia (Winterer & Weinberger, 2004), obsessive-compulsive disorders (Aouizerate et al., 2004), attentiondeficit ⁄ hyperactivity disorder (Castellanos et al., 2003; Madras et al., 2005; Seidman et al., 2005; Spencer et al., 2005) and drug addiction (Graybiel & Rauch, 2000; Gerdeman et al., 2003; Volkow & Li, 2004). An understanding of the molecular interactions that regulate basal ganglia development may lead to clues about the etiology of these disorders. Eph receptors and their ligands, the ephrins, are among the repertoire of axon guidance molecules that could help regulate basal ganglia development as they are developmentally expressed in various basal ganglia nuclei (Maisonpierre et al., 1993; Ciossek et al., 1995; Mori et al., 1995a,b; Martone et al., 1997; Zhang et al., 1997; Halladay et al., 2000; Liebl et al., 2003). For example, EphA4 and EphA7 are expressed within the matrix compartment of the striatum (Janis et al., 1999), and EphB1 is expressed in the substantia nigra Correspondence: Dr L.F. Kromer, as above. E-mail:
[email protected] * Present address: MiNDS Unit, National Institute of Mental Health, Bethesda, MD 20892, USA.
Received 14 October 2006, revised 6 February 2007, accepted 7 March 2007
(SN; Yue et al., 1999b). However, few studies have attempted to determine what role these molecules might play in basal ganglia development. For example, Yue et al. (1999b) observed that ephrin-B2 exhibits a high ventral-to-low dorsal expression gradient in the striatum, and that dopamine (DA) neurons from the SN exhibited a lower survival rate and develop shorter neurites when exposed to soluble ephrin-B2 in vitro. Based on this, they hypothesized that high levels of ephrin-B2 in the ventral striatum repel dopaminergic nigral axons resulting in their preferential innervation of the dorsal striatum. Because EphA5 receptors are also expressed on DA neurons, Sieber et al. (2004) generated transgenic mice that expressed soluble EphA5-Fc chimeric proteins to disrupt EphA ⁄ ephrin-A signaling. In these transgenic mice, the SN pars compacta (SNc) was smaller in volume, although the number of DA neurons in the SNc and the level of DA in the striatum did not differ from controls. However, fewer SNc neurons sent projections to the striatum, and these mice exhibited abnormal responses to the DA agonist, amphetamine (Sieber et al., 2004). Based on the above observations, the present study was designed to evaluate the consequences of EphB1 gene deletions in mice on the anatomical organization of the SN and striatum, and to identify if these mice exhibited a behavioral phenotype. We initially hypothesized that deletion of EphB1 would result in a disruption in the organization of the nigrostriatal DA pathway. In contrast, we did not observe any significant alteration in this projection but observed a significant decrease (40%) in the number of neurons within the SN pars reticulata (SNr) that correlated with an increase in spontaneous locomotor activity. Contrary to prior reports, our anatomical data further demonstrated that neurons within SNr and not DA neurons in SNc express EphB1 receptors.
ª The Authors (2007). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd
2620 A. Brent Richards et al.
Materials and methods Animals All animals were used according to Georgetown University Animal Care and Use Committee policies and in accordance with NIH guidelines. Mice with EphB1 gene deletions were generated by replacing exon 3 of the EphB1 gene with the bacterial lac Z cassette, which codes for b-galactosidase (b-gal). This mutation resulted in a frame shift, which prevents expression of the EphB1 protein (see Williams et al., 2004 for details). All mice were back-crossed onto a CD-1 background (Charles River Laboratories, Wilmington, MA, USA). Mice processed for histological analysis received an overdose of sodium pentobarbital (50 mg ⁄ mL) prior to transcardial perfusion with 4% paraformaldehyde in 0.1 m phosphate buffer (pH 7.3). Brains were rapidly dissected, postfixed overnight at 4 C and infiltrated with 20% sucrose in 0.1 m phosphate buffer for 48 h at 4 C for cryoprotection. Brains were frozen in dry ice and stored at )80 C. For fresh tissue, anesthetized mice were decapitated and their brains were rapidly dissected on an ice-cold plate. The dissected brain regions were frozen in dry ice and stored at )80 C until processed for Western blotting.
Immunolabeling Brain sections were cut on a cryostat at 20 lm and dried on charged glass slides (Fisher Scientific, Hampton, NH, USA). Slides were stored at )20 C until processed for histology. For immunohistochemical staining, sections were blocked for 1 h at room temperature in 10% normal serum (either goat or rabbit depending on secondary antibody) and 2% bovine serum albumin along with 0.2% Triton. Primary antibody incubation occurred overnight at 4 C in a 0.5 · blocking solution. The antibodies used included: rabbit anti-tyrosine hydroxylase (TH, 1 : 200, Imgenex, San Diego, CA, USA), goat anti-b-gal (1 : 500, Biogenesis, Kingston, NH, USA), mouse anti-NeuN (1 : 200, Chemicon, Temecula, CA, USA) and rabbit anti-l opioid receptor (1 : 10 000, Chemicon). The tissue was then incubated in the appropriate secondary antibody (Chemicon) for 1 h at room temperature. All rinses and solutions were made in Tris-buffered saline (TBS). For fluorescent-labeled secondary antibodies, the slides were immediately coverslipped with aqueous mounting media (Molecular Probes, Carlsbad, CA, USA). For immunoperoxidase histochemistry, the slides were further incubated in an avidin ⁄ biotin-horseradish peroxidase solution (Vector Laboratories, Burlingame, CA, USA) for 1 h. The staining was visualized with diaminobenzidine (DAB, 0.5 mg ⁄ mL), nickel chloride (0.04%) and hydrogen peroxide (0.01%). After staining, the tissue was cleared and dehydrated in alcohols and xylenes, and then coverslipped with Permount media (Fisher Scientific). Sections stained for NeuN immunofluorescence underwent antigen retrieval to achieve optimum signal. After sectioning, slides were dried at 55 C for at least 1 h. Sections were defatted in three 5-min rinses in Histoclear followed by 10 min each in 100%, 95% and 70% ethanol. After rinsing in TBS, slides were heated in distilled water to 95 C before being transferred to a solution of 1.5 mm EDTA preheated to 95 C. The slides were incubated for 10 min and then cooled in distilled water for 15 min. Antibody staining was performed as above, except that a fluorescein tyramide amplification kit (Perkin-Elmer, Wellesley, MA, USA) was used to visualize the signal. In situ hybridization To detect EphB1 mRNA we used a truncated form (1029 bp) of the riboprobe used by Yue et al. (1999b). The methods for this procedure
were similar to that used by Janis et al. (1999) with a few modifications. Instead of radiolabeled probes, non-radioactive digoxigenin (DIG) labeling was utilized. After hybridization and washing, sections were blocked for 1 h in blocking reagent (Roche, Indianapolis, IN, USA). Sections were incubated with an alkaline phosphatase-conjugated sheep anti-DIG antibody (Roche) overnight at 4 C. This was followed by incubation in a solution containing 5-bromo-4chloro-3¢-indolyphosphate p-toluidine salt (BCIP) and nitro-blue tetrazolium chloride (NBT; Roche) as substrates to visualize alkaline phosphatase. To determine whether TH+ DA neurons contained EphB1 mRNA, additional sections through the mesencephalon were further processed for TH immunoperoxidase histochemistry. Slides were dehydrated and coverslipped as above.
b-Gal histochemistry To label the b-gal marker in EphB1+ ⁄ – and EphB1– ⁄ – mice, brains were prepared as above for immunolabeling. Sections were rinsed in phosphate-buffered saline and then reacted overnight at 37 C in a 0.1 m sodium phosphate buffer solution (pH 7.3) containing 1 mg ⁄ mL X-Gal (Promega, Madison, WI, USA), 2 mm MgCl2, 5 mm potassium ferrocyanide and 5 mm potassium ferricyanide. The slides were dehydrated and coverslipped as above.
Protein determinations Procedures for determining temporal changes in EphB1 protein levels and phosphorylation states during development were identical to those described previously (Bundesen et al., 2003). For the present study, EphB1 was concentrated from 1 mg total protein ⁄ mL tissue lysate either by using a goat antibody directed against the intracellular domain of EphB1 (Santa Cruz, Santa Cruz, CA, USA) or by using wheat germ agglutinin linked to agarose (50 lL of a 50% slurry; 5 mg ⁄ mL binding capacity; Sigma) to pull-down EphB1 proteins that are glycosylated. The specificity of the EphB1 antibody was confirmed as no protein bands were detected on Western blots from mice with EphB1 gene deletions (Fig. 3C). Both procedures produced identical results. To detect endogenous activation of EphB1 receptors in the striatum, blots containing immunoprecipitated EphB1 protein were probed with a monoclonal antibody specific for phosphorylated tyrosine (1 lg ⁄ mL 4G10; Upstate Biotechnology, Chicago, IL, USA). Protein bands on the Western blots were detected using SuperSignal West Pico chemiluminescence reagents and secondary antibodies (Pierce, Rockford, IL, USA).
Fluorogold injections Mice were anesthetized with a mixture (0.04 mL ⁄ 10 g) of ketamine (25 mg ⁄ mL), xylazine (1.3 mg ⁄ mL) and acepromazine (325 lg ⁄ mL), and then mounted in a mouse stereotaxic unit (Kopf, Tujunga, CA, USA). A hole was drilled in the skull at +3.1 mm anterior and ± 1.5 mm lateral to bregma. To avoid the dorsal striatum, injections were made at a 25 angle anterior to vertical. A 4% solution of Fluorogold (Fluorochrome LLC, Denver, CO, USA) in sterile saline was injected (80 nL) into the ventral striatum with a 1-lL Hamilton syringe fitted with a 30-gage needle. The coordinates of the injection site were +1.0 mm anterior, ± 1.5 mm lateral and )4.5 mm ventral relative to bregma. After injection, animals survived for 4 days before being processed for histology as described above. Sections through the injection site and the entire ventral mesencephalon were analysed using an Olympus BX51 microscope equipped with epifluorescence
ª The Authors (2007). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 25, 2619–2628
Role of EphB1 in the basal ganglia and an 110004v2 filter set (Chroma, Rockingham, VT, USA) to detect Fluorogold. The number of Fluorogold-positive cells in the ventral mesencephalon and their distance from the midline were measured from digital images using Metamorph software (Molecular Devices, Sunnyvale, CA, USA).
Image analysis and morphometry For all morphometry and cell count measurements, the observer was blind to the genotype of the specimens. All measurements were made on digital micrographs using MetaMorph imaging software. To measure the volumes of the striatum, SNc, SNr, and ventral tegmental area (VTA) sections stained for TH alone or TH and NeuN doubleimmunofluorescence were digitally photographed as above, except that 14-lm sections were used. The regions were defined according to The Mouse Brain in Stereotaxic Coordinates (Franklin & Paxinos, 1997). To measure volumes, the region of interest was manually outlined in each section. Metamorph software was used to calculate the cross-sectional area. The total volume was calculated by adding the regional area from each section and multiplying the sum by 84 lm (the distance between each section). For cell counts, neurons were counted as part of the SNc if they were positive for both NeuN and TH, and were located lateral to the VTA. SNr neurons were identified by their position ventral to the SNc and dorsal to the cerebral peduncle (Fig. 3), and by their staining for NeuN but not TH. The physical dissector method (Sterio, 1984; Coggeshall & Lekan, 1996; Oorschot, 1996) was used to correct for the presence of partial cell bodies in each section. Thus, total cell profile counts were multiplied by a correction factor of 0.6 to obtain the actual number of neurons present in a region. The correction factor was obtained as follows: at least two sections (the look-up sections) adjacent to those used for the original counts (the reference sections) from at least three mouse brains of each genotype were aligned with their corresponding reference sections using Adobe Photoshop. A square region of uniform area was outlined on the overlay using Metamorph, and the cell profiles within this area that appeared in both the reference and look-up sections were counted. The average ratio of the true number of cells in the reference section (those that did not also appear in the look-up section) divided by the total number of cells in the reference section was found to be 0.6, and was consistent among all genotypes and brain regions analysed. To study dopaminergic fiber density, TH immunohistochemistry was used to label dopamine fibers in 20-lm brain slices. Fiber number was estimated by counting pixels occupied by TH immunoreactivity in high-power (40 ·) digital monochromatic micrographs. A total of four images from the striatum were captured for each mouse. Two observers independently determined the intensity threshold for including pixels as ‘TH-positive’. This count was then divided by the total number of pixels occupied by striatal gray matter in each image. The gray matter area was measured by subtracting the area occupied by white matter fiber bundles in each image from the total image area. One-way anova with Tukey’s post hoc test was used to check for statistical significance.
Locomotor activity monitoring Mice were monitored using a Tru Scan photobeam system (Coulbourn Instruments, Allentown, PA, USA). This consisted of a square acrylic arena (10 in · 10 in) surrounded by two stacked rings of infrared detectors. Each ring contained 36 detectors spaced 0.3 in apart. Movement in the arenas was detected as breaks in adjacent beams and
2621
was recorded by a computer that scanned for changes in position every 500 ms. Recording occurred in a dimly lit, sound-controlled room during the afternoon. Mice were moved by hand directly from their home cages to the testing arenas, and the testing equipment was immediately activated. All testing and animal handling was done by the same investigator. Locomotor behavior was measured as total distance moved in the horizontal plane during the testing period. Three groups of adult mice were tested: wild-type CD-1 (n ¼ 6), EphB1+ ⁄ – (n ¼ 5) and EphB1– ⁄ – (n ¼ 7). The results were tested for significance with a one-way anova with Tukey’s post hoc test to compare genotypes.
Results Developmental expression of EphB1 protein in the mouse basal ganglia To identify the temporal periods when EphB1 could influence basal ganglia development, we measured EphB1 protein levels and phosphorylation states in the ventral mesencephalon and striatum by Western blot analysis. In the ventral mesencephalon, EphB1 protein levels were highest during late embryonic and early postnatal development (E18–P10), and subsequently declined into adulthood (Fig. 1A). A similar temporal pattern for EphB1 protein expression was also detected in the striatum with protein levels declining after the second postnatal week (Fig. 1B). Evaluation of endogenous phosphorylation levels for EphB1 protein in the striatum demonstrated that
Fig. 1. EphB1 protein expression and phosphorylation levels in the mouse ventral mesencephalon and striatum. Representative Western blots illustrating the developmental time course of EphB1 protein expression in the mouse ventral mesencephalon (A) and striatum (B). In both regions, EphB1 protein levels are downregulated during the second postnatal week. The endogenous activation (phosphorylation) of EphB1 also rapidly decreases after birth (B). (C) EphB1 protein is not expressed in EphB1– ⁄ – mice. To detect total EphB1 protein, wheat germ agglutinin (WGA) was used to precipitate glycosylated proteins from tissue lysates. Western blots were then probed with antibody against EphB1 (aEphB1). Phosphorylated EphB1 was detected by immunoprecipitation with the EphB1-specific antibody followed by probing the blot with a phosphotyrosine antibody (apTyr).
ª The Authors (2007). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 25, 2619–2628
2622 A. Brent Richards et al.
Fig. 2.
Fig. 3. ª The Authors (2007). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 25, 2619–2628
Role of EphB1 in the basal ganglia there was extensive activation of EphB1 during late embryonic and early postnatal development (Fig. 1B). EphB1 protein was not detected in EphB1– ⁄ – mice (Fig. 1C).
Expression pattern of EphB1 in the ventral mesencephalon Cellular EphB1 mRNA expression was studied in normal CD-1 mouse brains from E18–P10 using in situ hybridization procedures, and b-gal histochemistry was used to localize cells expressing EphB1 in heterozygous EphB1lacz mutant mice. These developmental times were chosen for analysis because there is extensive growth of DA axons into the striatum during this perinatal period. In the ventral mesencephalon, EphB1 mRNA hybridization signal and bgal cellular histochemical staining were strongest in the ventrolateral quadrants of the mesencephalon corresponding to the region of the SN. To determine whether neurons expressing EphB1 were localized to SNc or SNr, coronal sections were first processed for in situ hybridization to detect EphB1 mRNA expression with a purple-blue alkaline phosphatase reaction and then processed for immunohistochemistry with the DAB procedure to localize DA neurons containing TH by their brown reaction product. As illustrated in Fig. 2A and B, neurons within the SNr exhibited dense reaction product with the in situ procedure. In contrast, the TH+ neurons within SNc did not contain detectable levels of EphB1 mRNA, even though low levels of EphB1 message were detected in the mesencephalon immediately dorsal to the SNc. As expected from prior studies, we also detected some scattered ventral tier DA neurons that were displaced into the SNr (Fig. 2B and C). These scattered TH+ neurons within the SNr also lacked EphB1 expression. Conversely, an occasional TH– and EphB1+ neuron was observed in the SNc. Because the double in situ and immunohistochemical procedure does not provide optimal immunostaining for TH due to the necessity to treat the tissue with proteinase K, we also doubleimmunostained single sections from heterozygous EphB1 mutants using antibodies against b-gal to mark EphB1-expressing neurons and TH to identify DA neurons in the SNc (Fig. 2D and E). Using this procedure we detected extensive b-gal immunostaining within neurons of the SNr, but could not clearly detect b-gal immunofluorescence within TH+ neurons located in either the SNc or the adjacent VTA. However, a few scattered cells within the SNc were positive for b-gal but negative for TH (Fig. 2E). Thus, both the double TH immunostaining ⁄ EphB1 in situ and the double TH immunostaining ⁄ EphB1-b-gal immunostaining provided identical results, indicating that DA neurons within the SN do not express detectable levels of EphB1 during the period when there is extensive growth of nigral DA axons into the striatum.
2623
Anatomical changes in the ventral mesencephalon of EphB1-deficient mice Brain sections from CD-1 wild-type, EphB1+ ⁄ – and EphB1– ⁄ – mice were double-immunostained with TH and NeuN to label neurons in the SNc, SNr and VTA. This permitted identification of DA neurons in the SNc by co-labeling for TH and NeuN, while SNr neurons were identified by NeuN labeling and the absence of TH. Microscopic evaluation of a series of coronal sections through the rostrocaudal extent of the SN revealed an apparent decrease in size of the SNr in EphB1– ⁄ – mice compared with heterozygous and wild-type controls (Fig. 3). This was quantified by measuring the number of neurons and volume of the SNr and SNc using the unbiased physical dissector method for cell counting by an observer blind to the specimen genotype. Statistical analysis of the data demonstrated that there was a significant decrease (approximately 47%; P < 0.001) in the volume of the SNr in EphB1– ⁄ – mice (0.75 ± 0.03 mm3), as compared with wild-type (1.43 ± 0.01 mm3) or EphB1+ ⁄ – (1.41 ± 0.01 mm3) mice (Fig. 4A). In contrast, the volume of the SNc was not significantly different across the three genotypes. Quantification of neuron number in the SNr demonstrated that there were approximately 40% fewer neurons in the SNr of EphB1– ⁄ – mice (12 900 ± 500), which represented a significant difference (P < 0.001) from the number of neurons in EphB1+ ⁄ – (21 100 ± 400) or wild-type (21 300 ± 400) mice (Fig. 4B). Even though there was a slight decrease in the number of TH+ neurons in the SNc of EphB1– ⁄ – mice (10 500 ± 500), this was not significantly different from that of EphB1+ ⁄ – (12 400 ± 400) or wild-type mice (12 500 ± 700). Although our histological results demonstrated that EphB1 was not expressed by SNc neurons, EphB1 was detected in the striatum where it could act as a guidance cue for nigrostriatal axons. Thus, we studied the nigrostriatal projection in EphB1 mutant mice to determine whether there was a shift in the distribution of axonal projections from DA neurons in the SNc. For this study the retrograde tracer Fluorogold was injected into the ventral striatum of EphB1+ ⁄ – and EphB1– ⁄ – mice, and the medial-to-lateral distribution of Fluorogold-labeled neurons in the ventral mesencephalon was quantified at 4 days postinjection in a series of coronal brain sections through the rostral-caudal extent of the SNc (Fig. 5). No significant difference between homozygous and heterozygous mutant mice was observed in either the number of labeled neurons or their medial-to-lateral distribution. EphB1 in the mouse striatum Previous studies have demonstrated that cells within the striatum express EphB1 mRNA (Yue et al., 1999b; Halladay et al., 2000); however, it is not known whether these cells are localized to a distinct
Fig. 2. EphB1 expression in the mouse SN. (A) In situ hybridization for EphB1 mRNA combined with TH immunohistochemistry in a P3 wild-type (+ ⁄ +) CD-1 mouse. Neurons present in the substantia nigra pars reticulata (SNr) exhibit high expression levels of EphB1 mRNA, whereas EphB1 message is not detected in TH+ neurons located in the substantia nigra pars compacta (SNc). However, low levels of EphB1 mRNA are detected in mesencephalic neurons (asterisks) located dorsal to the SNc. (B) Higher magnification of the box in (A), illustrating a clear boundary between EphB1+ neurons in the SNr and TH+ neurons in the SNc. However, a few ventrally displaced TH+ neurons that lack EphB1 expression (arrows) can be observed within the SNr (B and C). (D) Double-immunostaining for the EphB1 expression marker b-gal (green) and TH (red) in a P6 EphB1 heterozygous mouse (+ ⁄ –) demonstrates that the b-gal marker is not expressed by DA neurons of the SNc. (E) Higher-power magnification of the box in (D), illustrating a b-gal+ and TH– neuron (arrow) located among TH+ neurons (asterisks) within the SNc. Scale bar: 200 lm. VTA, ventral tegmental area.
Fig. 3. EphB1 deletion leads to anatomical changes in the SN. Coronal sections from adult EphB1+ ⁄ + (A–C) and EphB1– ⁄ – (D–F) mice stained for TH (red) and NeuN (green) immunofluorescence. Representative series of coronal sections from rostral (A, D), intermediate (B, E) and caudal (C, F) segments of the SN show a large decrease in cross-sectional area of the substantia nigra pars reticulata (SNr) in EphB1– ⁄ – mice. Abbreviations: ml, medial lemniscus; MT, medial terminal nucleus of the accessory optic tract; RMC, red nucleus, magnocellular part; SNc, substantia nigra pars compacta; VTA, ventral tegmental area. Scale bar: 200 lm. ª The Authors (2007). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 25, 2619–2628
2624 A. Brent Richards et al.
Fig. 5. The topography of the nigrostriatal projection is not altered by EphB1 deletion. Fluorogold was injected into the ventral striatum of adult EphB1+ ⁄ – and EphB1– ⁄ – mice (n ¼ 4). In the ventral mesencephalon, Fluorogold-labeled neurons were counted and their distance from the midline was measured. Each point represents the mean number of cells (± SEM) in distance bins of 120 lm as a fraction of the total number of cells counted.
Fig. 4. Quantification of anatomical changes observed in EphB1 null mice. Bar graphs show the volume (A) and number of neurons (B) in the substantia nigra pars compacta (SNc) and the substantia nigra pars reticulata (SNr) from EphB1+ ⁄ +, EphB1+ ⁄ – and EphB1– ⁄ – adult mice. No significant differences were detected in the volume or number of TH+ neurons in the SNc among the three genotypes. There were also no differences in the volume or neuron number in the SNr between wild-type and EphB1+ ⁄ – mice. In contrast, EphB1– ⁄ – mice exhibited a significant 47% decrease in volume (A) and a 40% decrease in neuronal number (B) in the SNr compared with both wild-type and EphB1+ ⁄ – mice. Bars represent mean ± SEM; n ¼ 4. **P < 0.001; one-way anova with Tukey’s post hoc test to compare genotypes.
striatal subcompartment, such as the well-characterized patches and matrix. Thus, we used in situ hybridization (Fig. 6A) and b-gal histochemistry (Fig. 6B) in EphB1+ ⁄ – mice to identify if EphB1expressing neurons formed a distinct subcompartment within the striatum. Both procedures revealed high reactivity in the matrix compartment for EphB1 message and the b-gal marker, with very weak signal in the patches and subcallosal streak. This was confirmed in P6 mice by double-immunofluorescent staining for b-gal and TH, which marks the dense clusters of DA fibers that are concentrated in the patches and subcallosal streak at this age (Fig. 6C). Because EphB1 is expressed by matrix neurons in the developing striatum, we hypothesized that the anatomy of the striatum might be altered in mice lacking EphB1. However, no gross morphological abnormalities were visible in the striatum of adult or developing EphB1– ⁄ – mice. In addition, the formation of the patch ⁄ matrix organization within the striatum did not appear to be affected by EphB1 deletion. This was confirmed by immunolabeling for several patch ⁄ matrix markers including TH, DARPP-32, the DA transporter and l-opioid receptors (Fig. 7A). We also determined if the density of
the dopaminergic input to the striatum was altered in EphB1– ⁄ – mice. Fiber density was determined by measuring the area occupied by THimmunolabeled axon profiles in digital micrographs and expressing this as a percentage of the total area of the striatal neuropil. These data indicated that there was no significant difference in the density of TH fibers in the striatum between wild-type (n ¼ 4), EphB1+ ⁄ – (n ¼ 4) and EphB1– ⁄ – (n ¼ 3) mice. Moreover, measurement of the striatal volume in specimens stained for TH (Fig. 7B and C) demonstrated that there was no significant difference in striatal volume between wild-type (18.7 ± 1.5 mm3), EphB1+ ⁄ – (18.7 ± 1.1 mm3) and EphB1– ⁄ – (18.2 ± 2.2 mm3) mice. Western blot analysis of lysed proteins from the striatum supported these anatomical findings and confirmed that there were no differences in TH protein levels between EphB1– ⁄ – and EphB1+ ⁄ – mice (Fig. 7D).
Effects of EphB1 deletion on locomotor activity The SNr is the primary output nucleus of the basal ganglia in mice; therefore, we hypothesized that the reduced number of SNr neurons in EphB1 mutant mice would lead to abnormalities in basal ganglia function. To study the behavioral effects of EphB1 deletion, baseline locomotor activity was recorded in both wild-type and mutant mice using an automated monitoring system to quantify total movement distance during a 1-h recording period. Although mutant mice appeared grossly normal when compared qualitatively with wild-type mice, electronic monitoring of locomotor activity revealed that EphB1– ⁄ – mice are hyperactive and exhibited an increase of approximately 60% (P < 0.05) in movement distance during the 1-h monitoring period when compared with wild-type and EphB1+ ⁄ – mice (Fig. 8). No significant differences in movement activity were detected between wild-type and EphB1+ ⁄ – mice.
Discussion The results of this study reveal that EphB1 receptors play a role in the development and function of the basal ganglia. Using Western blot analysis, in situ hybridization and b-gal as a marker for EphB1 expression in heterozygous EphB1lacz transgenic mice, we evaluated
ª The Authors (2007). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 25, 2619–2628
Role of EphB1 in the basal ganglia
2625
Fig. 6. EphB1 expression in the mouse striatum at P6 (all sections in the coronal plane). (A) In situ hybridization for EphB1 mRNA in a control CD-1 mouse (+ ⁄ +) demonstrates a mosaic expression pattern in the striatum with patches of cells (arrowheads) that contain low levels of message. (B) The pattern of b-gal histochemical staining in EphB1+ ⁄ – mutants mimics the EphB1 in situ pattern of stronger staining in the matrix compartment than in the patches (arrowheads) or subcallosal streak (arrows). (C and D) Double-immunofluorescence for TH (C) and b-gal (D) in an EphB1+ ⁄ – mouse confirms that neurons expressing increased levels of EphB1 ⁄ b-gal are present in the matrix and not the dopamine-rich patches (arrowheads) and subcallosal streak (arrows). Scale bar: 200 lm. Abbreviations: ac, anterior commissure; CPu, caudate-putamen; NAc, nucleus accumbens.
the developmental expression pattern for EphB1 in two major integrative centers of the mouse basal ganglia, the SN and striatum (caudate-putamen). During development, EphB1 expression is greatest from the prenatal period to the first postnatal week in both the SN and striatum. Moreover, EphB1 receptors exhibit extensive endogenous activation during this period. Our data also demonstrate that EphB1 is not expressed by dopaminergic neurons of the SNc, as previously proposed (Yue et al., 1999b), and that deletion of this receptor does not affect the targeting of nigrostriatal axons. However, a significant finding in the present study is that deletion of EphB1 does result in gross morphological changes in the SNr, which include a large decrease in volume (47%) and
neuronal number (40%). Moreover, we show that mice with EphB1 deletions exhibit altered locomotor behavior leading to spontaneous hyperactivity. General striatal organization is not altered by EphB1 deletion Although cells in the developing striatum express high levels of EphB1, especially neurons located in the matrix compartment, we did not detect any gross morphological defects within the striatum in EphB1– ⁄ – mice. This includes changes in the overall patch ⁄ matrix organization, striatal volume or the density of TH+ nigrostriatal fibers. Because other Eph receptors are present in the striatum, such as
ª The Authors (2007). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 25, 2619–2628
2626 A. Brent Richards et al.
Fig. 7. EphB1 deletion does not alter striatal anatomy or DA innervation. (A) EphB1– ⁄ – mutant immunostained for the patch marker, l-opioid receptor, at P6 demonstrates that the patch compartments (arrowheads) and subcallosal streak (arrows) develop in the absence of EphB1. (B and C) TH immunohistochemistry in adult (B) EphB1+ ⁄ + and (C) EphB1– ⁄ – mouse striatum. Striatal volume and TH fiber density were not significantly different between wild-type and mutant mice. Scale bar: 200 lm (A–C). (D) Western blot of EphB1+ ⁄ – and EphB1– ⁄ – P5 mouse striatum probed for TH protein. No differences in protein levels were detected between the two groups. Abbreviations: ac, anterior commissure; CPu, caudate-putamen; NAc, nucleus accumbens; OT, olfactory tubercle.
organization of striatal afferent or efferent projections or in the striatal cytoarchitecture could exist that were not detected in the present study. For example, because EphB1 has been implicated in dendritic spine formation in the hippocampus (Henkemeyer et al., 2003; Moeller et al., 2006), dendritic spine abnormalities might be present on the medium spiny striatal neurons of EphB1 mutant mice. Additional studies using the Golgi stain method or intracellular dyes will be required to resolve this possibility.
Fig. 8. EphB1 deletion results in hyperactivity. Locomotor activity was electronically recorded for 1 h. Bars represent the mean distance moved ± SEM. Adult EphB1– ⁄ – (n ¼ 7) mice moved a significantly greater distance than wild-type (n ¼ 6) or EphB1+ ⁄ – (n ¼ 5) mice. *P < 0.05; one-way anova with Tukey’s post hoc test to compare genotypes.
EphA4, which is expressed in a similar pattern to that of EphB1 (Janis et al., 1999), this receptor could potentially compensate for the loss of EphB1 in the caudate-putamen. However, subtle effects on the
EphB1 mutants exhibit anatomical abnormalities in the SNr but not SNc The most significant and surprising anatomical findings in this study were the 40% loss of neurons within the SNr and the striking lack of a change in the volume or number of DA neurons in the SNc or VTA in homozygous EphB1 mice. Three possible mechanisms could account for the decreased number of neurons in the SNr of mice lacking EphB1: (1) decreased generation of SNr neurons from progenitor cells during development; (2) increased apoptosis of SNr neurons during naturally occurring developmental cell death; and ⁄ or (3) errors in the migration of neuroblasts to the SNr. Both in vivo and in vitro studies have implicated Eph receptors and ephrins in regulating neural progenitor populations (Conover et al., 2000; Depaepe et al., 2005; Holmberg et al., 2005; Katakowski et al., 2005) and neuronal survival (Magal et al., 1996; Yue et al., 1999a,b; Gao et al., 2000). In general, these studies suggest that activation of Eph receptors or ephrins negatively
ª The Authors (2007). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 25, 2619–2628
Role of EphB1 in the basal ganglia
2627
Fig. 9. Hypothetical model for increased locomotor activity in EphB1– ⁄ – mice. As the primary output nucleus of the mouse basal ganglia, the substantia nigra pars reticulata (SNr) sends inhibitory projections to the thalamus and locomotor centers of the brainstem. In EphB1– ⁄ – mice, SNr output should be reduced due to the loss of neurons in this region. This would produce decreased inhibition of neurons in the thalamus and brainstem. Disinhibition of the thalamus would then lead to increased activity in the excitatory thalamocortical projection and thus increased output from motor areas of the cerebral cortex. This, coupled with disinhibition of the brainstem locomotor centers, could account for the hyperactivity observed in EphB1– ⁄ – mice. Abbreviations: GABA, c-aminobutyric acid; Glu, glutamate.
regulates cell survival. Moreover, deletion of EphA7 actually results in a reduction in cortical progenitor apoptosis (Depaepe et al., 2005), and interactions between EphA7 and ephrin-A2 negatively regulate adult progenitor cell proliferation (Holmberg et al., 2005). To date, no studies have reported the loss of a specific neuronal population as the result of deletion of a single Eph receptor. Therefore, it is difficult to explain how the deletion of EphB1 results in the loss of SNr neurons without an extensive series of developmental experiments to determine whether SNr neurons die prematurely or are generated in reduced numbers from their progenitor population. Alternatively, EphB1 might regulate the migration of EphB1-expressing neuroblasts to the ventrolateral mesencephalon. Those neurons that do not migrate to their proper location may fail to make appropriate afferent and efferent connections that are essential for providing trophic support during the period of developmental cell death. Although our data support prior observations that neurons in the SN express EphB1 receptors (Yue et al., 1999b), we failed to detect neurons that co-expressed EphB1 mRNA and TH protein in wildtype mice, or neurons that co-localized TH and the b-gal marker for EphB1 expression in heterozygous mutant mice. Based on our ability to co-localize these markers within individual neurons in the same tissue section, we conclude that the majority of DA neurons in the SNc do not express EphB1, which is in contrast to the conclusion of Yue et al. (1999b). In this prior study, the authors based their conclusion that DA neurons in SNc express EphB1 on observations from adjacent sections processed for in situ hybridization to detect message for EphB1 and immunohistochemistry to detect neurons containing TH. The use of both direct (in situ hybridization) and indirect (b-gal immunofluorescence) methods to co-localize EphB1 and TH immunostaining in the same tissue section provides a more accurate method for determining that mesencephalic DA neurons do not express EphB1. Moreover, we feel that our inability to detect EphB1 mRNA in the SNc neurons is not due to the sensitivity of our in situ procedures, which clearly demonstrated an intense hybridization reaction product in SNr neurons and low levels of hybridization within mesencephalic neurons dorsal to the SNc in the same tissue section. Although it is possible that DA neurons in SNc could express EphB1 prior to E18 (the earliest time examined), our observations that homozygous EphB1 mice exhibit no change in the number of DA neurons in the SNc or in the topography and density
of their innervation of the striatum are consistent with the lack of EphB1 receptors on these neurons. EphB1 is required for normal locomotor behavior Results from our locomotor tests complement the anatomical results and further highlight the importance of EphB1 for regulating basal ganglia function. Our data indicate that EphB1– ⁄ – mice are significantly hyperactive as compared with wild-type and EphB1 heterozygous mice. The locomotor hyperactivity observed in EphB1– ⁄ –, but not in EphB1+ ⁄ –, mice is consistent with our anatomical finding that there is a 40% reduction of neurons in the SNr in homozygous but not heterozygous EphB1 mutant mice. Although the cerebral cortex initiates and directly controls locomotor activity, it is also heavily influenced by basal ganglia activity. Therefore, it is not surprising that cell loss in the main output nucleus of the mouse basal ganglia, the SNr, would lead to changes in locomotor behavior. The neurons of the SNr use the inhibitory neurotransmitter c-aminobutyric acid (GABA), and their primary efferent targets are in the thalamus, superior colliculus and brainstem locomotor control centers (Takakusaki et al., 2004). Both the basal ganglia–thalamo-cortical loop and the basal ganglia– brainstem loop function to provide a general inhibitory control over spontaneous locomotor activity (Takakusaki et al., 2004). Therefore, a loss of neurons in the SNr would likely lead to a disinhibition of thalamocortical neurons resulting in increased activity in the motor areas of the cortex. There also would be decreased inhibition of brainstem locomotor control centers. As illustrated in Fig. 9, both of these effects would result in the increased basal locomotor activity observed in EphB1– ⁄ – mice. However, further studies are required to establish a definitive link between the loss of SNr neurons and the change in locomotor behavior observed in EphB1– ⁄ – mice, and to determine whether the behavioral responses observed in these mice are due solely to alterations in the organization of the extrapyramidal basal ganglia motor pathways.
Acknowledgements We wish to thank Miranda Anderson for her valuable technical assistance in preparing the anatomical material, Dr Benjamin Walker with help in setting up the behavioral tasks, and Dr Linda MacArthur for her thoughtful discussions and editorial comments. This research was supported by NIH grant NS27054 (A.B.R.), NIMH grant NS66332 (M.H.) and NINDS grant NS38266 (L.F.K.).
ª The Authors (2007). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 25, 2619–2628
2628 A. Brent Richards et al.
Abbreviations b-gal, b-galactosidase; DA, dopamine; DAB, diaminobenzidine; DIG, digoxigenin; SN, substantia nigra; SNc, substantia nigra pars compacta; SNr, substantia nigra pars reticulata; TBS, Tris-buffered saline; TH, tyrosine hydroxylase; VTA, ventral tegmental area.
References Aouizerate, B., Guehl, D., Cuny, E., Rougier, A., Bioulac, B., Tignol, J. & Burbaud, P. (2004) Pathophysiology of obsessive-compulsive disorder: a necessary link between phenomenology, neuropsychology, imagery and physiology. Prog. Neurobiol., 72, 195–221. Bundesen, L.Q., Scheel, T.A., Bregman, B.S. & Kromer, L.F. (2003) Ephrin-B2 and EphB2 regulation of astrocyte–meningeal fibroblast interactions in response to spinal cord lesions in adult rats. J. Neurosci., 23, 7789– 7800. Castellanos, F.X., Sharp, W.S., Gottesman, R.F., Greenstein, D.K., Giedd, J.N. & Rapoport, J.L. (2003) Anatomic brain abnormalities in monozygotic twins discordant for attention deficit hyperactivity disorder. Am. J. Psychiatry, 160, 1693–1696. Ciossek, T., Lerch, M.M. & Ullrich, A. (1995) Cloning, characterization, and differential expression of MDK2 and MDK5, two novel receptor tyrosine kinases of the eck ⁄ eph family. Oncogene, 11, 2085–2095. Coggeshall, R.E. & Lekan, H.A. (1996) Methods for determining numbers of cells and synapses: a case for more uniform standards of review. J. Comp. Neurol., 364, 6–15. Conover, J.C., Doetsch, F., Garcia-Verdugo, J.M., Gale, N.W., Yancopoulos, G.D. & Alvarez-Buylla, A. (2000) Disruption of Eph ⁄ ephrin signaling affects migration and proliferation in the adult subventricular zone. Nat. Neurosci., 3, 1091–1097. Depaepe, V., Suarez-Gonzalez, N., Dufour, A., Passante, L., Gorski, J.A., Jones, K.R., Ledent, C. & Vanderhaeghen, P. (2005) Ephrin signalling controls brain size by regulating apoptosis of neural progenitors. Nature, 435, 1244–1250. Franklin, K.B.J. & Paxinos, G. (1997) The Mouse Brain in Stereotaxic Coordinates. Academic Press, San Diego. Gao, P.P., Sun, C.H., Zhou, X.F., DiCicco-Bloom, E. & Zhou, R. (2000) Ephrins stimulate or inhibit neurite outgrowth and survival as a function of neuronal cell type. J. Neurosci. Res., 60, 427–436. Gerdeman, G.L., Partridge, J.G., Lupica, C.R. & Lovinger, D.M. (2003) It could be habit forming: drugs of abuse and striatal synaptic plasticity. Trends Neurosci., 26, 184–192. Graybiel, A.M. & Rauch, S.L. (2000) Toward a neurobiology of obsessivecompulsive disorder. Neuron, 28, 343–347. Halladay, A.K., Yue, Y., Michna, L., Widmer, D.A., Wagner, G.C. & Zhou, R. (2000) Regulation of EphB1 expression by dopamine signaling. Brain. Res. Mol. Brain Res., 85, 171–178. Henkemeyer, M., Itkis, O.S., Ngo, M., Hickmott, P.W. & Ethell, I.M. (2003) Multiple EphB receptor tyrosine kinases shape dendritic spines in the hippocampus. J. Cell. Biol., 163, 1313–1326. Holmberg, J., Armulik, A., Senti, K.A., Edoff, K., Spalding, K., Momma, S., Cassidy, R., Flanagan, J.G. & Frisen, J. (2005) Ephrin-A2 reverse signaling negatively regulates neural progenitor proliferation and neurogenesis. Genes Dev., 19, 462–471. Janis, L.S., Cassidy, R.M. & Kromer, L.F. (1999) Ephrin-A binding and EphA receptor expression delineate the matrix compartment of the striatum. J. Neurosci., 19, 4962–4971. Katakowski, M., Zhang, Z., Decarvalho, A.C. & Chopp, M. (2005) EphB2 induces proliferation and promotes a neuronal fate in adult subventricular neural precursor cells. Neurosci. Lett., 385, 204–209. Liebl, D.J., Morris, C.J., Henkemeyer, M. & Parada, L.F. (2003) mRNA expression of ephrins and Eph receptor tyrosine kinases in the neonatal and adult mouse central nervous system. J. Neurosci. Res., 71, 7–22.
Madras, B.K., Miller, G.M. & Fischman, A.J. (2005) The dopamine transporter and attention-deficit ⁄ hyperactivity disorder. Biol. Psychiatry, 57, 1397–1409. Magal, E., Holash, J.A., Toso, R.J., Chang, D., Lindberg, R.A. & Pasquale, E.B. (1996) B61, a ligand for the Eck receptor protein-tyrosine kinase, exhibits neurotrophic activity in cultures of rat spinal cord neurons. J. Neurosci. Res., 43, 735–744. Maisonpierre, P.C., Barrezueta, N.X. & Yancopoulos, G.D. (1993) Ehk-1 and Ehk-2: two novel members of the Eph receptor-like tyrosine kinase family with distinctive structures and neuronal expression. Oncogene, 8, 3277–3288. Martone, M.E., Holash, J.A., Bayardo, A., Pasquale, E.B. & Ellisman, M.H. (1997) Immunolocalization of the receptor tyrosine kinase EphA4 in the adult rat central nervous system. Brain Res., 771, 238–250. Moeller, M.L., Shi, Y., Reichardt, L.F. & Ethell, I.M. (2006) EphB receptors regulate dendritic spine morphogenesis through the recruitment ⁄ phosphorylation of focal adhesion kinase and RhoA activation. J. Biol. Chem., 281, 1587–1598. Mori, T., Wanaka, A., Taguchi, A., Matsumoto, K. & Tohyama, M. (1995a) Differential expressions of the eph family of receptor tyrosine kinase genes (sek, elk, eck) in the developing nervous system of the mouse. Brain Res. Mol. Brain Res., 29, 325–335. Mori, T., Wanaka, A., Taguchi, A., Matsumoto, K. & Tohyama, M. (1995b) Localization of novel receptor tyrosine kinase genes of the eph family, MDK1 and its splicing variant, in the developing mouse nervous system. Brain Res. Mol. Brain Res., 34, 154–160. Nomura, Y. & Segawa, M. (2003) Neurology of Tourette’s syndrome (TS) as a developmental dopamine disorder: a hypothesis. Brain Dev., 25 (Suppl. 1), S37–S42. Oorschot, D.E. (1996) Total number of neurons in the neostriatal, pallidal, subthalamic, and substantia nigral nuclei of the rat basal ganglia: a stereological study using the cavalieri and optical disector methods. J. Comp. Neurol., 366, 580–599. Seidman, L.J., Valera, E.M. & Makris, N. (2005) Structural brain imaging of attention-deficit ⁄ hyperactivity disorder. Biol. Psychiatry, 57, 1263–1272. Sieber, B.A., Kuzmin, A., Canals, J.M., Danielsson, A., Paratcha, G., Arenas, E., Alberch, J., Ogren, S.O. & Ibanez, C.F. (2004) Disruption of EphA ⁄ ephrin-A signaling in the nigrostriatal system reduces dopaminergic innervation and dissociates behavioral responses to amphetamine and cocaine. Mol. Cell. Neurosci., 26, 418–428. Spencer, T.J., Biederman, J., Madras, B.K., Faraone, S.V., Dougherty, D.D., Bonab, A.A. & Fischman, A.J. (2005) In vivo neuroreceptor imaging in attention-deficit ⁄ hyperactivity disorder: a focus on the dopamine transporter. Biol. Psychiatry, 57, 1293–1300. Sterio, D.C. (1984) The unbiased estimation of number and sizes of arbitrary particles using the disector. J. Microsc., 134, 127–136. Takakusaki, K., Saitoh, K., Harada, H. & Kashiwayanagi, M. (2004) Role of basal ganglia-brainstem pathways in the control of motor behaviors. Neurosci. Res., 50, 137–151. Volkow, N.D. & Li, T.K. (2004) Drug addiction: the neurobiology of behaviour gone awry. Nat. Rev. Neurosci., 5, 963–970. Williams, S.E., Mason, C.A. & Herrera, E. (2004) The optic chiasm as a midline choice point. Curr. Opin. Neurobiol., 14, 51–60. Winterer, G. & Weinberger, D.R. (2004) Genes, dopamine and cortical signalto-noise ratio in schizophrenia. Trends Neurosci., 27, 683–690. Yue, Y., Su, J., Cerretti, D.P., Fox, G.M., Jing, S. & Zhou, R. (1999a) Selective inhibition of spinal cord neurite outgrowth and cell survival by the Eph family ligand ephrin-A5. J. Neurosci., 19, 10026–10035. Yue, Y., Widmer, D.A., Halladay, A.K., Cerretti, D.P., Wagner, G.C., Dreyer, J.L. & Zhou, R. (1999b) Specification of distinct dopaminergic neural pathways: roles of the Eph family receptor EphB1 and ligand ephrin-B2. J. Neurosci., 19, 2090–2101. Zhang, J.H., Pimenta, A.F., Levitt, P. & Zhou, R. (1997) Dynamic expression suggests multiple roles of the eph family receptor brain-specific kinase (Bsk) during mouse neurogenesis. Brain Res. Mol. Brain Res., 47, 202–214.
ª The Authors (2007). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd European Journal of Neuroscience, 25, 2619–2628
