Expression of P2x5 Receptors in the Mouse Cns

Neuroscience 156 (2008) 673–692

EXPRESSION OF P2X5 RECEPTORS IN THE MOUSE CNS

W. GUO,a1 X. XU,a1 X. GAO,a G. BURNSTOCK,b C. HEa et al., 1996; Collo et al., 1996; Seguela et al., 1996; Soto AND Z. XIANGa* et al., 1996; Lê et al., 1998b; Kanjhan). etHet- al., 1999 aDepartment of Neurobiology, Second Military Medical Universityeromultimeric assemblies of P2X receptor subtypes have 200433 Shanghai, PR China been described (seeBurnstock, 2007b). bAutonomic Neuroscience Centre, Royal Free and University College All seven subtypes of P2X receptor mRNA and protein Medical School, Rowland Hill Street, London NW3 2PF, UK have been detected in various nuclei in the CNS by reverse transcriptase–polymerase chain reaction (RT-PCR) Abstract—P2X receptors are ATP-gated cationic channels (Kidd et al., 1995; Shibuya et al., 1999; Vorobjev et al., 2003), in situhybridization Kidd( et al., 1995; Shibuya et composed of seven known subunits1-7 ) (P2Xwhich are involved in different functions in neural tissue. The present study al.,inves- 1999) and immunocytochemistryVulchanova ( et al., tigates the P2X5 receptor expression pattern in the mouse CNS1996; Xiang et al., 1998; Loesch et al., 1999; Loesch and using immunohistochemistry andin situhybridization histo- Burnstock, 2001; Yao et al., 2001; Atkinson ).et al., 2004 chemistry. The specificity of the immunostaining has been Therever- has, however, been some controversy about the ified by pre-absorption, Western blotin andsituhybridization existence of P2Xreceptors in the CNS. Hybridization methods. Heavy P2Xreceptor immunostaining was observed 5 5 experiments showed that there was noreceptor P2X in the mitral cells of the olfactory bulb; cerebral cortex; globus 5 pallidum, anterior cortical amygdaloid nucleus, amygdalohip-mRNA in the CNS except for the mesencephalic trigeminal pocampal area of subcortical telencephalon; anterior nuclei,nucleus and spinal cordCollo ( et al., ),1996 and RT-PCR anteroventral nucleus, ventrolateral nucleus of thalamus; su-also showed no 5 P2XmRNA in the rat supraoptic nucleus. praoptic nucleus, ventromedial nucleus, arcuate nucleus of However,hy- other data showed that5 receptor P2X protein pothalamus; substantia nigra of midbrain; pontine nuclei, mes-and mRNA distribute in some regions of the CNS. Single encephalic trigeminal nucleus, motor trigeminal nucleus, am-cell RT-PCR showed that about 35% of neurons ex- biguous nucleus, inferior olive, hypoglossal nucleus, dorsalpressed P2Xreceptor mRNA in the tuberomamillary nu- motor vagus nucleus, area postrema of hindbrain; Purkinje 5 cleus of rat hypothalamus. Pharmacological data indicated cells of cerebellum; and spinal cord. The identification of exten- that functional heteromeric 2/5P2Xreceptors might be sive P2X5 receptor immunoreactivity and mRNA distribution within the CNS of the mouse demonstrated here is consistentpresent in the neurons of hypothalamusVorobjev ( et al., with a role for extracellular ATP acting as a 2003fast) and homomeric 5P2Xreceptors presented in the neurotransmitter. © 2008 IBRO. Published by Elsevier Ltd.cerebellum All Brockhaus( et al., ).2004 Immunohistochemi- rights reserved. cal data showed that5 receptor-immunoreactivityP2X (-ir) was present in rat rostral ventrolateral Thomasmedulla et( Key words: P2Xreceptor, immunohistochemistry,in situ 5 al., 2001), solitary tract nucleusYao (et al., ),2001 cere- hybridization, CNS. bellum Xiang( et al., 2005a), choroid plexusXiang ( et al., 2005b), compact division of the nucleus ambiguous Extracellular nucleotide receptors belong to the P2X(Brosenitsch et al., ),2005 hypothalamus Xiang( et al., ligand-gated cationic channels or P2Y G protein-coupled 2006r e) -and paraventricular nucleusCham ( et al., ).2006 ceptors Abbracchio( and Burnstock, 1994; Ralevic andTogether these data imply that 5thereceptor P2X is dis- Burnstock, 1998; North, ).2002 Both receptor types aretributed widely in the CNS. However, at present there is no widely distributed in the CNS and exhibit various effectssystematic on study of the distribution pattern of5 the P2X both neuronal and glial Burnstock,cells ( 2007a). P2X re- receptor subunit in the whole CNS. Thus, in the present study, detailed information about the distribution pattern of ceptors form a family of seven subunits1-7). Neuronal(P2X P2X receptors in the CNS appear to belong mostlyP2X to5 receptorsthe at both protein and mRNA levels in the CNS of the mouse has been obtained using immunocyto- P2X2, P2X4 or P2X4/P2X6 subtypes Bo( et al., 1995; Buell chemistry andin situhybridization methods. 1 These two authors contributed equally to this work. *Corresponding authors. Tel: ϩ86-21-25074545-8; fax: ϩ86-21- 65492132 (Z. Xiang), Tel: ϩ86-21-65515200; fax: ϩ86-21-65492132 EXPERIMENTAL PROCEDURES (C. He). E-mail address: [email protected] or [email protected] (Z.Tissue preparation Xiang), [email protected] (C. He). Abbreviations: BSA, bovine serum albumin; CA1, field CA1 of hip- All experimental procedures were approved by the Institutional pocampus; CA2, field CA2 of hippocampus; CA3, field CA3 of hip- Animal Care and Use Committee at Second Military Medical Uni- pocampus; CA4, field CA4 of hippocampus; ir, immunoreactivity; LTP, long-term potentiation; NHS, normal horse serum; NMDA, N-methyl- versity and conformed to the UK Animals (Scientific Procedures) D-aspartate; PBS, phosphate-buffered saline; PPADS, pyridoxal phos- Act 1986 and associated guidelines on the ethical use of animals. phate-6-azophenyl-2-4-disulfonic acid; RT-PCR, reverse transcripta- Twelve adult mice (25–35 g) were used. The number of animals se–polymerase chain reaction. used and their suffering in this study were minimized. The mice 0306-4522/08 © 2008 IBRO. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.neuroscience.2008.07.062

673 674 W. Guo et al. / Neuroscience 156 (2008) 673–692

ϫ were killed by asphyxiation with CO2 and perfused through the 50% formamide, 10% dextran sulfate, 0.3 mol/l NaCl, 1 Den- aorta with 0.9% NaCl solution and 4% paraformaldehyde in 0.1 hardt’s solution, 0.05 mol/l Tris–HCl (pH 8.0), 1 mmol/l EDTA and mol/l phosphate buffer pH 7.4. The brains were dissected out 250 ␮g/ml E. coli tRNA (RNase-free). Hybridization was carried immediately and immersed in 4% paraformaldehyde in 0.1 M out for 16 h at 56 °C in a hybridization oven. The sections were phosphate-buffered saline (PBS, pH 7.2) for 2–4 h. The brains washed in 4ϫ SSC for 20 min at 37 °C, followed by incubation in were then transferred to 25% sucrose in PBS and kept in the 2ϫ SSC containing 20 mg/ml RNase A (Sigma) for 30 min at 37 °C solution until they sank to the bottom. Thereafter, the brains were to digest the RNA probes that did not hybridize with the targeted rapidly frozen by immersion in isopentane at Ϫ70 °C for 2 min. RNA. The sections were further washed in 1ϫ SSC and 0.2ϫ SSC Coronal sections (20 ␮m) of the brains were cut with a Leica at 37 °C for 20 min, respectively. The following protocol was used cryostat (CM1900) (Nussloch, Germany) and floated in PBS. to detect the hybridization signals. Briefly, the sections were first incubated in the blocking buffer containing 5% bovine serum Immunohistochemistry albumin and 0.4% Triton X-100 in PBS at room temperature for 30 min, and then with anti-digoxigenin sheep IgG Fab fragments Immunohistochemistry for localization of P2X5 receptors was per- conjugated to alkaline phosphatase (Roche Boehringer Mann- formed using rabbit polyclonal antibody against a unique peptide heim) diluted 1:1000 in the blocking buffer for4hatroom tem- sequence of P2X5 receptor provided by Roche Palo Alto (CA, USA). perature. The sections were washed with PBS for 4ϫ5 min, fol- The immunogens used for the production of the polyclonal P2X5 lowed by washing in TSM1 (0.1 mol/l Tris–HCl buffer, pH 8.0, receptor antibody were synthetic peptides corresponding to the car- containing 0.1 mol/l NaCl and 0.01 mol/l MgCl), then equilibrated boxyl terminal of the cloned rat P2X5 receptor, covalently linked to in TSM2 (0.1 mol/l Tris–HCl 2 buffer, pH 9.5, 0.1 mol/l NaCl and keyhole limpet hemocyanin. The peptide sequences of the P2X5 0.05 mol/l MgCl). The color development was performed with receptor are of amino acid sequence 437–452 (RENAIVNVKQS- 400 ␮g/ml Nitro Blue Tetrazolium, 200 ␮g/ml 5-bromo-4-chloro-3- QILH). The polyclonal antibody was raised by multiple monthly injec- indolyl phosphate and 100 mg/ml levamisole in TSM2 in the dark tions of New Zealand White rabbits with the corresponding peptides at room temperature for 2 h. The sections were rinsed in PBS 3ϫ5 (prepared by Research Genetics, Huntsville, AL, USA). The P2X5 to stop the color development, then dry the sections at 37 °C in an receptor antiserum used in this study has previously been shown to oven for 3 h and mounted with DPX (Sigma). detect the P2X receptor subunit, but not the remaining subunits, 5 The density of P2X5 receptor protein immunostaining and when these subunits were expressed in cell lines (Oglesby et al., mRNA hybridization signal was scored as: absent (Ϫ), weak (ϩ), 1999). As previously reported, no cross-reactivity is observed with moderate (ϩϩ), heavy (ϩϩϩ)(Collo et al., 1996; Braissant et al., other P2X receptor antisera (Oglesby et al., 1999). 2001; Pollio et al., 2005). Endogenous peroxidase was blocked by 3% H2O2 in PBS for 30 min. The sections were pre-incubated in 10% normal horse Control experiments serum (NHS), 0.2% Triton X-100 in PBS for 30 min followed by

incubation with P2X5 receptor antibody, diluted 1:500 in anti- Control experiments were carried out with P2X5 antiserum pre-absorbed ␮ body dilution solution (10% NHS, 0.2% Triton X-100 and 0.4% with P2X5 receptor peptide at a concentration of 25 g/ml. The amino sodium azide in PBS) overnight. Subsequently, the sections acid sequence for this peptide is: 437–452 (RENAIVNVKQSQILH), were incubated with biotinylated donkey anti-rabbit IgG synthesized by Roche Palo Alto. Sense digoxigenin-labeled cRNA (Jackson ImmunoResearch Laboratory, West Grove, PA, USA) at probe was used as a negative control of in situ hybridization. No a dilution of 1:500 in PBS containing 1% NHS for 1 h. The sections staining was observed in those specimens incubated with the were then incubated in ExtrAvidin peroxidase (Sigma Chemical antibody solutions pre-absorbed with P2X5 receptor peptides and

Co., Poole, UK) diluted 1:1000 in PBS for 1 h. The P2X5 receptor sense digoxigenin-labeled cRNA probe (Fig. 1B, D). ir was visualized by a freshly prepared color reaction mixture solution containing 0.05% 3,3=-diaminobenzidine, 0.05 M sodium Combination of immunohistochemistry and in situ

phosphate, 0.004% NH4Cl, 0.2% glucose, 0.04% nickel ammo- hybridization nium sulfate and 0.1% glucose oxidase. All incubations were held at room temperature and separated by 3ϫ5 min washes in PBS. In order to further confirm the specificity of the P2X5 receptor Some sections were counterstained with 1% Neutral Red, which antibody, the combination method of immunohistochemistry and in situ hybridization histochemistry was used. The following is the was used for calculating the percentage of P2X5 receptor-ir neurons in some regions. Finally, the sections were then mounted on protocol for this combination method. In situ hybridization was first slides, dehydrated and cleared in xylene, and embedded. carried out. The procedure of prehybridization treatment, hybridization and washes after hybridization was the same as that In situ hybridization above. After posthybridization wash, endogenous peroxidase was blocked by 3% H2O2 in PBS for 30 min. The sections were Antisense and sense digoxigenin-labeled cRNA probes were syn- pre-incubated in 10% NHS, 0.2% Triton X-100 in PBS for 30 min thesized with a DIG-RNA labeling Kit (Roche Boehringer Mann- followed by incubation with sheep anti-digoxigenin antibody heim) using linearized templates of pGEM-4 inserted by rP2X5 (Roche Boehringer Mannheim), diluted 1:500 in antibody dilution cDNA 1–586 and rP2X5 cDNA 361–785 (GenBank X92069). In solution (10% NHS, 0.2% Triton X-100 and 0.4% sodium azide in situ hybridization was carried out using a protocol used before PBS) overnight at 4 °C. Subsequently, the sections were incu- (Xiang et al., 2001). Briefly, floating mouse brain sections were bated with biotinylated donkey anti-sheep IgG (Jackson) at a washed 3ϫ5 min in 0.1 mol/l PBS, in 0.1 mol/l glycine/PBS and in dilution of 1:500 in PBS containing 1% NHS for 1 h. The sections 0.4% Triton X-100/PBS for 10 min each. The sections were then were then incubated in ExtrAvidin peroxidase (Sigma) diluted ␮ incubated in protease K (1 g/ml) in PBS for 30 min at 37 °C. The 1:1000 in PBS for 30 min at room temperature. The P2X5 mRNA activity of protease K was stopped by fixation in 4% paraformal- hybridization signal was visualized by the TSA (tyramide signal dehyde for 5 min, followed by 2ϫ3 min washes in PBS to remove amplification) Fluorescein system (NEL701, NEN, USA). After

ﬁxative from the sections. The sections were incubated in 0.25% visualization, the sections were incubated with the P2X5 receptor acetic diaminobenanhydride with 0.1 mol/l triethanoloamine (pH antibody diluted 1:500 in the antiserum dilution solution overnight 8.0) for 10 min at room temperature, followed by washing in 0.6 at 4 °C. Subsequently, the sections were incubated with Cy3- mol/l sodium chloride and 0.06 mol/l sodium citrate (2ϫ SSC) for conjugated donkey-anti-rabbit (Jackson) diluted 1:400 in anti- 10 min. Digoxigenin-labeled cRNA (0.5 ␮g/ml) of either antisense serum dilution solution for1hatroom temperature. All the incu- or sense probes were added to hybridization buffer containing bations and reactions were separated by 3ϫ10 min washes in W. Guo et al. / Neuroscience 156 (2008) 673–692 675

PBS. Finally, the sections were then mounted on slides and The number of immunopositive neurons was counted unilater- embedded in 50% glycerol PBS. ally throughout the caudorostral extent of the respective nuclei, as For Western blot analysis, mice were deeply anesthetized by deﬁned by the atlas of Paxinos and Franklin (2001). Data for each of sodium pentobarbital (60 mg/kg) and killed by decapitation. Brains the nuclei analyzed were obtained from each of the mice used. Three were rapidly removed and lysed with 20 mM Tris–HCl buffer, pH to ﬁve sections from each animal were used, and the average per- 8.0, containing 1% NP-40, 150 mM NaCl, 1 mM EDTA, 10% centage of P2X receptor-ir neurons in the individual regions/nuclei ␤ 5 glycerol, 0.1% -mercaptoethanol, 0.5 mM dithiothreitol, and a was calculated. The numbers presented in Table 1 represent the mixture of proteinase and phosphatase inhibitors (Sigma). Protein average percentage of immunopositive cells observed unilaterally in concentration was determined by the BCA protein assay method the individual regions/nuclei. using bovine serum albumin as standard. One hundred micro- grams of protein samples from brain was loaded per lane, separated by SDS-PAGE (12% polyacrylamide gels) and then was RESULTS electrotransferred onto nitrocellulose membranes. The mem-

branes were blocked with 10% nonfat dry milk in Tris-buffered Both P2X5 receptor immunostaining and the in situ hybrid- saline for 1 h and incubated overnight at 4 °C with P2X antibody 5 ization labeling of P2X5 receptor mRNA demonstrated a (Roche Palo Alto) diluted 1:1000 in 2% bovine serum albumin wide-ranging expression pattern for this subunit through- (BSA) in PBS. The membranes were then incubated with alkaline phosphatase–conjugated goat anti-rabbit IgG (Sigma) diluted out the CNS. P2X5 receptor mRNA and P2X5 receptor-ir 1:5000 in 2% BSA in PBS for1hatroom temperature. The color were closely correlated both in distribution and relative ␮ development was performed with 400 g/ml Nitro-Blue Tetrazo- levels. The P2X5 receptor-ir was dense in many regions, lium, 200 ␮g/ml 5-bromo-4-chloro-3-indolyl phosphate and 100 such as olfactory bulb, cerebral cortex, globus pallidum, mg/ml levamisole in TSM2 (0.1 mol/l Tris–HCl buffer, pH 9.5, 0.1 2 hippocampus, thalamus, hypothalamus, cerebellar cortex, mol/l NaCl and 0.05 mol/l MgCl2) in the dark. Bands were scanned using a densitometer (GS-700; Bio-Rad Laboratories). and midbrain and hindbrain nuclei. This staining was blocked in control sections exposed to P2X5 receptor an- Photomicroscopy and data analysis tisera preabsorbed with its speciﬁc peptide (peptide block Images were taken with the Nikon digital camera DXM1200 (Nikon, control; Fig. 1B). Sections, which were incubated in sense Japan) attached to a Nikon Eclipse E600 microscope (Nikon). Images digoxigenin-labeled cRNA probe, showed no hybridization were imported into a graphics package (Adobe Photoshop 5.0, USA). signal (Fig. 1D). Combination methods of immunohisto-

Abbreviations used in the ﬁgures

AM anteromedial thalamic nucleus ot optic tract Amb ambiguus nucleus PaPo paraventricular hypothalamic nucleus, posterior part AP area postrema PDTg posterodorsal tegmental nucleus AV anteroventral thalamic nucleus Pir piriform cortex Ce central amygdaloid nucleus PLco posterolateral cortical amygdaloid nucleus CM central medial thalamic nucleus PMco posteromedial cortical amygdaloid nucleus CPu caudate putamen (striatum) PnC pontine reticular nucleus, caudal part Cu cuneate nucleus Pr5DM principal sensory trigeminal nucleus, dorsomedial part DG dentate gyrus Pr5VL principal sensory trigeminal nucleus, ventrolateral part DpMe deep mesencephalic nucleus PVA paraventricular thalamic nucleus, anterior part D3V dorsal 3rd ventricle py pyramidal tract Ect ectorhinal cortex R red nucleus EPl external plexiform layer of the olfactory bulb Rh rhomboid thalamic nucleus f fornix Rt reticular thalamic nucleus fi fimbria of the hippocampus Sc Schaffer collaterals Gl glomerular layer of the olfactory bulb scp superior cerebellar peduncle GP globus pallidus SN substantia nigra GrO granular cell layer of the olfactory bulb SOL solitary tract nucleus ic internal capsule Sp5 spinal trigeminal nucleus IRt intermediate reticular nucleus st stria terminalis La lateral amygdaloid nucleus LC locus coeruleus Sub submedius thalamic nucleus LD laterodorsal thalamic nucleus SuG superficial gray layer of the superior colliculus LRt lateral reticular nucleus TSA tyramide signal amplification LSO lateral superior olive VA ventral anterior thalamic nucleus MA3 medial accessory oculomotor nucleus VCA ventral cochlear nucleus, anterior part MD mediodorsal thalamic nucleus VL ventrolateral thalamic nucleus MePD medial amygdaloid nucleus, posterodorsal part VPL ventral posteromedial thalamic nucleus Me5 mesencephalic trigeminal nucleus Xi xiphoid thalamic nucleus MG medial geniculate nucleus Zi zona incerta MGP medial globus pallidus 3V 3rd ventricle MHb medial habenular nucleus 7n facial nerve Mi mitral cell layer of the olfactory bulb 10 dorsal motor nucleus of vagus Mo5 motor trigeminal nucleus 12 hypoglossal nucleus mt mammillothalamic tract 676 W. Guo et al. / Neuroscience 156 (2008) 673–692

Fig. 1. Specificity tests. (A, B) P2X5 receptor antibody specificity tests; (C, D) in situ hybridization specificity test. P2X5 receptor immunostaining in sections of the primary somatosensory cortex using P2X5 receptor primary antibody alone (A) or after preabsorption with its peptide antigen (B). Note the absence of immunostaining in B. P2X5 receptor mRNA hybridization signal in sections of the primary somatosensory cortex using antisense (C) or sense (D) digoxigenin-labeled cRNA probe. Note the absence of hybridization signals in D. Scale barsϭ200 ␮m (A–D). (E) Western blotting from brain extracts. M, molecular weight marker. Lanes 1 and 2, P2X5 receptors immunoreactive band is located at 51 kDa. Lanes 3 and 4, preabsorption of the P2X5 receptor antisera with its peptide antigen, which resulted in the absence of the band. chemistry and in situ hybridization histochemistry showed about 51 kDa that corresponded to molecular weight of that the neurons with P2X5 receptor immunostaining were P2X5 receptor (Fig. 1E, lane 1). Preadsorption of the also found to express P2X5 mRNA (Fig. 2A–F). The dis- antiserum with the peptide antigen resulted in the ab- tribution patterns of P2X5 mRNA hybridization signal de- sence of the band (Fig. 1E, lane 2), indicating that the tected by two antisense digoxigenin-labeled cRNA probes antibody detected the appropriate antigen sequence. transcripted from rP2X5 cDNA 1–586 and rP2X5 cDNA 361–785 was the same. Olfactory bulb Western blotting, performed on tissue extracts de- rived from the mouse cerebral cortex (Fig. 1A), as- The main olfactory bulb, accessory olfactory bulb, and pri- sessed the specificity of the polyclonal P2X5 receptor mary olfactory cortex showed P2X5 receptor immunostaining antibody. An immunoreactive band was detected at at high levels (Fig. 3A, B). Olfactory nerve and the glomeruli W. Guo et al. / Neuroscience 156 (2008) 673–692 677

Table 1. Characterization of P2X5 receptor immunostaining and Table 1. continued mRNA levels in the mouse CNS Regions ISH IHC % Regions ISH IHC % Midbrain Olfactory bulb Zona incerta ϩϩNC Mitral cells ϩϩϩ ϩϩϩ NC Lateral geniculate nucleus ϩϩNC Granular layer ϪϪNC Medial geniculate nucleus ϩϩϩ59 Tufted cells ϩϩNC Substantia nigra zona compacta ϩϩ ϩϩϩ 78 Anterior olfactory nucleus ϩϩ ϩϩ 35 Substantia nigra zona reticulate ϩϩ ϩϩϩ 65 Cerebral cortex Supramammillary nucleus ϩϩ ϩ 26 Motor cortex ϩϩ ϩϩϩ 66 Ventral tegmental area ϩϩϩ54 Somatosensory cortex ϩϩ ϩϩϩ 51 Oculomotor nucleus ϩϩ ϩϩ 53 Cingulate cortex ϩϩϩ10 Edinger-Westphal nucleus ϩϩNC Agranular insular cortex ϩϩ ϩϩ 20 Central gray ϩϩNC Piriform cortex ϩϩ ϩϩϩ 78 Superior colliculus ϩϩϩ67 Auditory cortex ϩϩϩ38 Inferior colliculus ϩϩNC Entorhinal cortex ϩϩ ϩϩϩ 65 Interpeduncular nucleus ϩϩNC Subcortical telencephalon Trochlear nucleus ϩϩNC Nucleus diagonal band ϩϩϩ10 Pedunculopontine tegmental nucleus ϩϩϩ62 Bed nucleus stria terminalis ϩϩ 8 Hindbrain Medial septa nucleus ϩϩ55 Pontine nuclei ϩϩ ϩϩϩ 92 Lateral septal nucleus ϩϩ ϩϩ 23 Raphe dorsalis ϩϩNC Islands of Calleja ϪϪ 0 Mesencephalic trigeminal nucleus ϩϩ ϩϩϩ 100 Olfactory tubercle ϩϩϩ74 Dorsomedial tegmental area ϩϩ 56 Accumbens nucleus ϩϩ 6 Locus ceruleus ϩϩNC Caudate Putamen ϩϩϩ3 Subceruleus nucleus ϩϩNC Globus pallidum ϩ ϩϩϩ 90 Barrington’s nucleus ϩϩϩ67 Substantia innominata ϩϩϩ23 Pontine reticular nucleus, ventral part ϩϩ ϩϩ 85 Medial globus pallidum ϩϩ ϩϩ 85 Pontine reticular nucleus, caudal part ϩϩϩ87 Lateral olfactory tract nucleus ϩϩϩ20 Red nucleus ϩϩϩ76 Anterior cortical amygdaloid nucleus ϩ ϩϩϩ 52 Motor trigeminal nucleus ϩϩ ϩϩϩ 79 Basolateral amygdaloid nucleus ϩϩϩ46 Superior olivary nucleus ϩϩϩ86 Central amygdaloid nucleus ϩϩ ϩϩ 63 Principal sensory trigeminal nucleus ϩϩϩ83 Medial amygdaloid nucleus ϩϩ ϩϩ 25 Spinal trigeminal nucleus ϩϩϩ74 Amygdalohippocampal area ϩϩ ϩϩϩ 43 Paragigantocellular nucleus ϩϩ ϩϩ 81 CAl–CA4 ϩϩ ϩϩ 95 Abducens nucleus ϩϩ ϩϩ 87 Dentate gyrus ϩϩ ϩϩ 90 Facial nucleus ϩϩ ϩϩ 82 Epithalamus Accessory facial nucleus ϩϩ ϩϩ 88 Medial habenular nucleus ϩϩϩ35 Ambiguous nucleus ϩϩ ϩϩϩ 93 Lateral habenular nucleus ϩϩ11 Inferior olive ϩϩ ϩϩϩ 91 Thalamus Hypoglossal nucleus ϩϩ ϩϩϩ 90 Anterior nuclei ϩϩ ϩϩϩ 86 Dorsal motor vagus nucleus ϩϩ ϩϩϩ 89 Anteroventral nucleus ϩ ϩ ϩϩϩ 78 Solitary tract nucleus ϩϩϩ69 Laterodorsal nucleus, ϩϩ ϩϩ 87 Lateral reticular nucleus ϩϩϩ85 Mediodorsal nucleus ϩϩϩ87 Curate fasciculus ϩϩ ϩϩ 78 Ventrolateral nucleus ϩ ϩ ϩϩϩ 89 Area postrema ϩϩ ϩϩ NC Ventromedial nucleus ϩϩ78 Cerebellum Reticular nucleus ϩϩ85 Purkinje cells ϩϩϩ ϩϩϩ 100 Paraventricular nucleus ϩϩϩ75 Granular layer ϩϩNC Central medial nucleus ϩϩ56 Cerebellar nucleus ϩϩ ϩϩ 95 Xiphoid nucleus ϪϪ 0 Spinal cord Rhomboid nucleus ϩϩϩ46 Dorsal horn ϩϩ ϩϩ 85 Hypothalamus Ventral horn ϩϩ ϩϩ 75 Suprachiasmatic nucleus ϩϩNC Medial preoptic nucleus ϩϩNC ISH, in situ hybridization; IHC, immunohistochemistry; %, the percentage of P2X receptor-ir neurons against the total number of neu- Medial preoptic area ϩϩϩNC 5 rons stained with 1% Neutral Red in the individual nucleus of the Supraoptic nucleus ϩϩ ϩϩϩ 95 mouse CNS; NC, not counted. The density of P2X receptor protein Paraventricular nucleus ϩϩϩ56 5 immunostaining and mRNA hybridization signal was scored as: absent Ventromedial nucleus ϩ ϩϩϩ 78 (Ϫ), weak (ϩ), moderate (ϩϩ), heavy (ϩϩϩ). Arcuate nucleus ϩϩ ϩϩϩ 80 Tuberomammillary nucleus ϩϩ ϩϩ NC Moderate to heavy immunostaining was detected in the external plexiform layer and internal granular cells respectively. were also stained (Fig. 3A, B). Dendrite-like ﬁbers from ex- No positive cell bodies were demonstrated clearly in these ternal plexiform layer in the glomerular layer were stained. two layers. In the internal granular cell layer the positive ﬁbers 678 W. Guo et al. / Neuroscience 156 (2008) 673–692

Fig. 2. Control experiment using a combination method of immunohistochemistry and in situ hybridization histochemistry. (A) P2X5 receptor immunostaining in the primary somatosensory cortex. (B) Hybridization signal of P2X5 mRNA in the same region of A. (C) The merged image of A and B. Note that all the cell bodies with P2X5 receptor immunostaining were colocalized (yellow) with P2X5 receptor mRNA hybridization signal. However, in the processes, especially a little distance away from the cell bodies, only P2X5 receptor immunostaining was found, but no mRNA hybridization signal. D and E are high power images of the areas indicated by a star in A and B respectively. F is the merged image of D and E. II, III, IV, V, VI are layers II, III, IV, V, VI respectively. Scale barsϭ200 ␮m (A–C); 50 ␮m (D–F). For interpretation of the references to color in this ﬁgure legend, the reader is referred to the Web version of this article. were arranged parallel, but in the external plexiform layer the Cerebral cortex positive ﬁbers were in differential directions. Heavy immuno- Heavy P2X5 receptor immunostaining was observed in the staining was detected in I layer of the piriform cortex (Fig. 3C). motor, somatosensory, piriform and entorhinal cortex. Mod- P2X receptor mRNA positive neurons in the olfactory sys- 5 erate-heavy P2X5 receptor immunostaining was also ob- tem were distributed widely in the main olfactory bulb, the served in the cingulated, agranular insular and auditory cor- mitral cells were heavily labeled, and moderately labeled cells tex. Typically P2X5 receptor immunostaining was mainly lo- were found in the glomerular layer (Fig. 3D, E). In the piriform calized in I, II, III and V layers of the primary somatosensory cortex P2X5 receptor mRNA positive neurons were also cortex, although scattered P2X5 receptor-ir neurons were mainly localized in the Ia layer of the piriform cortex, although also found in the IV and VI layers (Fig. 4A, C, D, E, F). Mainly scattered positive neurons in other layers were observed pyramidal and some nonpyramidal cortical cell bodies and (Fig. 3F). their dendrites were labeled between layers II, III and VI of the W. Guo et al. / Neuroscience 156 (2008) 673–692 679

Fig. 3. Distribution of P2X5 receptor immunostaining (A, C, E) and mRNA signal (B, D, F) in olfactory bulb and primary olfactory cortex. Note that the patterns of cell bodies with P2X5 receptor immunostaining and mRNA hybridization signal matched well (A, B). Immunostaining was mainly conﬁned to dendrite-like processes in the GrO, internal plexiform layer of the olfactory bulb and EPl. Mitral cells are positive, but their outline is not clear (A,

C). (C) High magniﬁcation of the area indicated by a star in A. Heavy hybridization signal for P2X5 receptor mRNA was found in the mitral cell layer, moderate hybridization signal in the tufted cells and periglomerular cells, very weak or no hybridization signal in the granular cell (B, D). (D) High magniﬁcation of the area indicated by a star in B. Heavy immunostaining (E) and hybridization signals (F) of P2X5 receptor were found in the piriform cortex, especially in Ia layer. Scale barsϭ250 ␮m (A, B); 100 ␮m (C, D); 200 ␮m (E, F).

cortex. Pyramidal cells of layer V typically showed the dens- and III were positive for the P2X5 receptor. The neurons est labeling, some P2X5 receptor-ir ascending apical den- positive for P2X5 receptor mRNA were mainly found in II, III drites of pyramidal neurons could be traced as far as layer I and V layers (Fig. 4A, B). where they formed a dense neuropilar staining. Over 85% of Subcortical telencephalon pyramidal cells from layer V were positive for the P2X5 receptor. The neurons that were negative for the P2X5 receptor Moderate immunostaining of neurons was observed in the were mainly non-pyramidal cells in layer V. About 10% and vertical and horizontal parts of diagonal band, lateral septal 5% of neurons from layer VI and IV, respectively, was posi- nucleus, entopeduncular nucleus, substantia innominata tive for the P2X5 receptor. Over 90% of neurons from layer II and lateral olfactory tract nucleus (Fig. 5A, B). Weak im- 680 W. Guo et al. / Neuroscience 156 (2008) 673–692

Fig. 4. Distribution of P2X5 receptor immunostaining (A, C–F) and mRNA signal (B) in the primary somatosensory cortex. The density of immunostaining was higher in layers I, II, III and V than that in layers IV and VI. Dense dendrite-like processes were demonstrated in layers I, II and III and pyramidal neurons in layer V. Scattered neurons and apical dendrites originating from pyramidal neurons in layer V were found in layer IV. Scattered neurons and axon-like processes in the layer VI were clearly demonstrated (F). (C–F) High magnifications of the areas indicated by c, d, e, f in A, respectively. The distribution pattern of P2X5 receptor hybridization signal matched that of cell bodies with P2X5 receptor immunostaining in the primary somatosensory cortex (A, B). Scale barsϭ200 ␮m (A, B); 50 ␮m (C–F). munostaining was also observed in the olfactory tubercle, nuclei exhibited moderate levels of labeling. The anterior, bed nucleus stria terminalis and accumbens nucleus. No basomedial, central and medial amygdaloid nuclei showed immunostaining was observed in the islands of Calleja. moderate P2X5 receptor immunostaining (Fig. 5E). Pyrami- Many small cells in the caudate-putamen, particularly the dal cells of hippocampal fields CA1–CA4 and granule cells of more external regions, showed moderate P2X5 receptor dentate gyrus showed moderate immunostaining for the immunostaining, several larger cells showed darker stain- P2X5 receptor (Fig. 6A, B). Neurons in the hilus of dentate ing. Globus pallidus showed heavy P2X5 receptor immu- gyrus were also labeled. Immunostaining was mainly con- nostaining in fibers and cells (Fig. 5C). Approximately 90% fined to cell bodies and apical dendrites. Almost all the pyra- of neurons in the globus pallidus were positive for the P2X5 midal cells and granule cells of the dentate gyrus were pos- receptor. The posteromedial portion of the amygdalohip- itive for the P2X5 receptor. The synaptic region where CA3 pocampal area and posteromedial portion of cortical amyg- pyramidal cell axon (Schaeffer) collaterals form synapses daloid nuclei showed heavy immunostaining, whereas the onto distal dendrites of CA1 pyramidal cells had particularly posterolateral and anterior portions of cortical amygdaloid high levels of P2X5 receptor-ir (Fig. 6A). The distribution W. Guo et al. / Neuroscience 156 (2008) 673–692 681

Fig. 5. Distribution of P2X5 receptor immunostaining in the nucleus of the horizontal limb of the diagonal band (A), dorsal part of lateral septal nucleus (B), lateral globus pallidus (C) and amygdaloid complexes (E) and mRNA hybridization signals in the amygdaloid complexes (F). (D) High magniﬁcation of the area indicated by a star in the lateral globus pallidus of C. (E=) High magniﬁcation of the Ce in E. The distribution pattern of P2X5 ϭ ␮ receptor hybridization signal matched that of cell bodies with P2X5 receptor immunostaining in the amygdaloid complexes (E, F). Scale bars 200 m (A–C); 50 ␮m (D); 500 ␮m (E); 80 ␮m(E=); 100 ␮m (F).

patterns of P2X5 receptor-ir and mRNA signals in the sub- sal, paraventricular, rhomboid, reticular, laterodorsal and cortical telencephalon matched well (Fig. 5F, Fig. 6C). lateral posterior thalamic nuclei (Fig. 7A). No or weak immunostaining was shown in the central medial, xiphoid Epithalamus and thalamus nuclei. P2X5 receptor immunostaining was mostly conﬁned The medial habenular nucleus showed moderate immuno- to cell bodies in some nuclei, including the habenular, paraventricular, laterodorsal and mediodorsal nuclei (Fig. staining for P2X5 receptors in the cell body of neurons with few processes and the lateral habenular nucleus showed 7B). In some other areas such as the anteroventral, an- weak immunostaining (Fig. 7A). Thalamic nuclei showed teromedial, anterodorsal and ventral–lateral thalamic nu- moderate to heavy immunostaining for P2X5 receptors. clei, P2X5 receptor immunostaining was shown in both cell Heavy immunostaining was observed in anteroventral, an- bodies and ﬁbers (Fig. 7C). In situ hybridization showed teromedial, anterodorsal and ventral–lateral thalamic nu- moderate hybridization signals in those nuclei with P2X5 clei. Moderate immunostaining was observed in mediodor- receptor immunostaining (Fig. 7D, E). 682 W. Guo et al. / Neuroscience 156 (2008) 673–692

Fig. 6. Distribution of P2X5 receptor immunostaining (A, B) and mRNA hybridization signal (C) in CA1–CA4 ﬁelds, DG granule cells, and the Sc of the hippocampus. Note immunostaining in the area where CA3 pyramidal cell axon collaterals (Schaeffer collaterals) form synapses onto CA1 pyramidal cell dendrites (sc). (B) High magniﬁcation of the area indicated by a star in CA1 of A. The distribution pattern of P2X5 receptor hybridization ϭ ␮ ␮ signal matched well that of cell bodies with P2X5 receptor immunostaining in the hippocampus (A, C). Scale bars 180 m (A, C); 60 m (B). W. Guo et al. / Neuroscience 156 (2008) 673–692 683

Fig. 7. Distribution of P2X5 receptor immunostaining (A–C) and mRNA hybridization signal (D, E) in the epithalamus and thalamus. (A) Low-power image of immunostaining in the transverse plane of the thalamus. P2X5 receptor immunostaining was found in MHb, PVA, LD, MD, AV, AM, VA, VL, VPL, Sub, Rt, fi, DG, MGP, mt, PaPo, Xi and Zi. (B) High magnification of an area indicated by a star in the MD of A. (C) High magnification of an area indicated by a star in the VPL of A. (D) Expression pattern of P2X5 receptor hybridization signal in the thalamus. (E) High magnification of an area indicated by a star in D. Scale barsϭ500 ␮m (A); 100 ␮m (B, E); 200 ␮m (D); 50 ␮m (C).

Hypothalamus pothalamus, dorsomedial nucleus and mammillary nuclei (Fig. 8B, C). P2X receptor mRNA was widespread Moderate to heavy P2X receptor immunostaining was 5 5 throughout the hypothalamic nuclei and particularly high in detected throughout the hypothalamus. The ventral and the ventromedial and arcuate hypothalamic nuclei (Fig. medial aspects of hypothalamus including ventromedial 8D, E). and arcuate hypothalamic nuclei had particularly dense staining (Fig. 8A, C). Heavy neuropilar staining was ob- Midbrain served in the medial eminence (Fig. 8C, D). Moderate

P2X5 receptor immunostaining was detected in the preop- In the midbrain neurons of substantia nigra showed heavy tic area, anterior area, paraventricular nucleus, lateral hy- P2X5 receptor immunostaining (Fig. 9A). About 80% and 684 W. Guo et al. / Neuroscience 156 (2008) 673–692

Fig. 8. Distribution of P2X5 receptor immunostaining (A–D) and mRNA hybridization signal (E, F) in the hypothalamus. (A) P2X5 receptor immunostaining is found in the supraoptic nucleus, lateral hypothalamus and nucleus of the horizontal limb of the diagonal band. (B) P2X5 receptor immunostaining is found in the paraventricular nucleus of the hypothalamus. (C) Immunostaining in the arcuate nucleus, ventral medial nucleus, lateral hypothalamus, dorsomedial nucleus and median eminence. (D) The high power image of arcuate nucleus in C. (E) P2X5 receptor hybridization signal in the arcuate nucleus, ventral medial nucleus, lateral hypothalamus and dorsomedial nucleus. (F) The high power image of arcuate nucleus in E. Scale barsϭ100 ␮m (A, B, D, F); 200 ␮m (C, E).

60% of neurons in the compact and reticular part of this Hindbrain nucleus, respectively, were positive for the P2X5 receptor. The medial geniculate nucleus, ventral tegmental area, P2X5 receptor immunostaining was widely distributed in oculomotor nucleus, superior colliculus and pedunculo- the mouse hindbrain and appeared darker than in other pontine tegmental nucleus neurons expressed moderate parts of the brain. Heavy P2X5 receptor immunostaining levels of P2X5 receptor-ir (Fig. 9A, C, D). In the neurons was detected in the pontine nuclei, mesencephalic trigem- of the lateral geniculate nucleus, supramammillary nu- inal nucleus, motor trigeminal nucleus, ambiguous nucleus, Edinger-Westphal nucleus, inferior colliculus, in- cleus, inferior olive, hypoglossal nucleus, dorsal motor terpeduncular nucleus and trochlear nucleus weak im- vagus nucleus, area postrema, dorsal and ventral cochlear munostaining for P2X5 receptors was observed. Central nuclei and vestibular nuclei (Figs. 10A, 11A). Moderate gray neurons showed weak or no immunostaining. In P2X5 receptor immunostaining was detected in the pontine situ hybridization showed moderate to heavy hybridiza- reticular nucleus, red nucleus, Barrington’s nucleus, supe- tion signals in those nuclei with P2X5 receptor immuno- rior olivary nucleus, principal sensory trigeminal nucleus, staining (Fig. 9A, D). spinal trigeminal nucleus, paragigantocellular nucleus, ab- W. Guo et al. / Neuroscience 156 (2008) 673–692 685

Fig. 9. Distribution of P2X5 receptor immunostaining (A, C, D) and mRNA hybridization signal (B) in the midbrain. (A) Heavy immunostaining in the midbrain neurons of SN, moderate immunostaining in the MG, ventral tegmental area, MA3, SuG and R. Central gray neurons showed very weak immunostaining. (B) In situ hybridization histochemistry revealed a similar distribution pattern between P2X5 receptor mRNA and ir in the midbrain (A, B). (C) The high power image of medial geniculate nucleus in A. (D) The high power image of medial accessory oculomotor nucleus and its adjacent region. Scale barsϭ500 ␮m (A, B); 200 ␮m (C, D). ducens nucleus, facial nucleus, accessory facial nucleus, Spinal cord solitary tract nucleus, lateral reticular nucleus and curate P2X receptor immunostaining was observed in neurons fasciculus (Figs. 10A, 11A). Weak P2X receptor immuno- 5 5 and neuropil of the spinal gray matter at the cervical, staining was detected in the dorsomedial tegmental area, thoracic, lumbar and sacral levels (Fig. 13A, C, D). The raphe dorsalis, locus ceruleus and subceruleus nucleus most prominent staining was seen in the dorsal horn (Fig. (Figs. 10A, 11A). In almost all the regions, immunostaining 13C). Immunolabeling was evident in many motoneurons was observed mainly in the cell bodies and dendrites, but in the ventral horn of the spinal cord (Fig. 13D). About 75% in the mesencephalic trigeminal neurons, immunostaining of neurons in the ventral horn of the spinal cord were was observed only in the cell bodies (Fig. 10A, C). In situ positive for P2X5 receptors. In situ hybridization revealed a hybridization showed moderate to heavy hybridization sig- similar distribution pattern of P2X5 receptor mRNA and ir in nals in those nuclei with P2X5 receptor immunostaining the spinal cord (Fig. 13A, B). (Figs. 10, 11). The signiﬁcant correlation in distribution patterns and

levels of P2X5 receptor immunolabeling of native protein Cerebellum and P2X5 receptor mRNA was found in the different nuclei and regions of the whole mouse CNS (Table 1). Addition- The most prominent P2X receptor immunostaining in the 5 ally, the percentage of the P2X receptor-ir neurons cerebellar cortex was in the Purkinje cells, which had 5 against the total number of neurons stained by Neutral Red densely stained cell bodies and dendrites (Fig. 12A, C). All in the different regions/nuclei of the CNS is shown in the the Purkinje cells were positive for P2X5 receptors. Mod- “% ” column. The percentage of P2X5 receptor-ir neurons erate P2X5 receptor-ir was observed in many granule cells in some of regions/nuclei was not analyzed because the of the granular layer (Fig. 12A, C). Deep cerebellar nuclei outline of the P2X5 receptor-ir neuron was not clear. and more ventrally located vestibular nuclei exhibited moderate to heavy P2X receptor immunostaining. Moderate to 5 DISCUSSION heavy levels of P2X5 receptor mRNA were detected in the Purkinje and granule cell layers of the cerebellar cortex This is the ﬁrst extensive study undertaken to investigate and the deep cerebellar nuclei (Fig. 12B, D). the combined distribution of P2X5 receptor protein and 686 W. Guo et al. / Neuroscience 156 (2008) 673–692

Fig. 10. Distribution of P2X5 receptor immunostaining (A, C, E) and mRNA hybridization signal (B, D, F) in the hindbrain at the level of the ventral cochlear nucleus, anterior part. (A) Low power image of immunostaining in the transverse plane. P2X5 receptor immunostaining was found in the LC, PDTg, Me5, Mo5, Pr5DM, Pr5VL, LSO, PnC and VCA. (C) High magnification of the region indicated by a star in Me5 and LC of A. (E) High magnification of the region indicated by a star in Mo5 of A. (D) High magnification of the region indicated by a star in Me5 and LC of B. (F) High magnification of the region indicated by a star in Mo5 of B. The distribution patterns of cell bodies with P2X5 receptor ir and mRNA hybridization signal matched well (A, B). Scale barsϭ500 ␮m (A, B); 100 ␮m (C–F). mRNA in the mouse CNS using immunohistochemistry found to express P2X5 mRNA (Fig. 2A–F) further confirms and in situ hybridization histochemistry. These data pro- the specificity of the antibody and also confirms the spec- vide clear evidence for an extensive expression of P2X5 ificity of in situ hybridization histochemistry for P2X5 recep- receptor in the CNS. The specificity of the antibody for the tor mRNA. detection of the P2X5 receptor protein used in this study The distribution of P2X5 receptor expression within the has previously been validated (Oglesby et al., 1999; Ryten CNS shown here considerably extends previous immunohis- et al., 2002; Ruan and Burnstock, 2005; Xiang and Burn- tochemical studies which reported P2X5 receptor immuno- stock, 2005a, 2005b; Xiang et al., 2006). The significant staining in several populations of neurons including: rat ros- correlation in distribution patterns and levels of P2X5 re- tral ventrolateral medulla (Thomas et al., 2001), solitary tract ceptor immunolabeling of native protein and P2X5 receptor nucleus (Yao et al., 2001), cerebellum (Xiang et al., 2005a), mRNA (Table 1) in the mouse CNS also support the spec- choroid plexus (Xiang et al., 2005b), compact division of the ificity of the P2X5 receptor antibody. The fact that the nucleus ambiguous (Brosenitsch et al., 2005), hypothalamus neurons with P2X5 receptor immunostaining were also (Xiang et al., 2006) and paraventricular nucleus (Cham et al., W. Guo et al. / Neuroscience 156 (2008) 673–692 687

Fig. 11. Distribution of P2X5 receptor immunostaining (A, C–E) and mRNA hybridization signal (B, F) in the hindbrain at area postrema level. (A) Low power image of immunostaining in the transverse plane. P2X5 receptor immunostaining was found in the AP, Cu, 12, 10, Sol, Sp5, Amb, LRt and IRt. (C) High magnification of the region indicated by a star in AP and Sol of A. (D) High magnification of the region indicated by a star in 12 of A. (E) High magnification of the region indicated by a star in IRt of A. (F) High magnification of the region indicated by a star in LRt of B. The distribution patterns ϭ ␮ ␮ of cell bodies with P2X5 receptor ir and mRNA hybridization signals match well (A, B). Scale bars 500 m (A, B); 100 m (C–F). 2006). Although the configuration of subunits required to Little is known about the functional significance of re- assemble into functional ATP-gated ion channel complexes gions of P2X5 receptor expression detected in the present in different regions of the CNS is not known, co-localization of study, but some speculations follow. In the cerebral cortex,

P2X5 receptor protein in the present study in regions reported dense immunolabeling in layers I, III and V, especially in to also express other P2X receptor subunits (P2X2 receptor, layer I (molecular layer), where parallel axons synapse on Vulchanova et al., 1997; Kanjhan et al., 1999; P2X4 receptor, apical dendrites and their branches, suggests a role for Bo et al., 1995; Collo et al., 1996; Seguela et al., 1996; Lê et P2X5 receptors in interconnection of local cortical areas. al., 1998a; P2X5 and P2X6 receptor subunits, Collo et al., Excitatory effects of extracellular ATP and other nucleo- 1996; P2X1 receptor, Florenzano et al., 2008) is compatible tides on cortical neurons, including identiﬁed corticospinal with both homomultimeric and heteromultimeric assembly of cells, were reported (Phillis et al., 1979). Unilateral micro-

P2X5 receptor subunits into non-selective cation channels injection of nucleotides into the rat prepiriform cortex was gated by extracellular ATP in regions such as cerebral cortex, reported to cause a convulsant response that is antago- hypothalamus, cerebellar Purkinje cells, hippocampal pyra- nized by suramin (Weisman et al., 1996). The present midal neurons and spinal cord motor neurons. study showed that cerebral cortex neurons expressed 688 W. Guo et al. / Neuroscience 156 (2008) 673–692

Fig. 12. Distribution of P2X5 receptor immunostaining (A, C) and mRNA hybridization signal (B, D) in the cerebellum. (A) P2X5 receptor immunostaining was found in the Purkinje cells and granular cells. Note the heavy immunostaining in the cell bodies and dendritic trees of Purkinje cells, while no immunostaining was found in the white matter. (B) Heavy hybridization signal in the cell bodies of Purkinje cells and moderate hybridization signals in granular cells was found. No hybridization signal was found in the white matter. The distribution patterns of cell bodies with P2X5 receptor immunostaining and mRNA hybridization signal matched well. (C) High magniﬁcation of the region indicated by a star in A. (D) The distribution of P2X5 receptor hybridization signals in the deep cerebellar nucleus. Scale barsϭ500 ␮m (A); 200 ␮m (B); 50 ␮m (C, D).

plentiful of P2X5 receptors, which of these P2X5 receptor (Wieraszko and Seyfried, 1989). Excitatory postsynaptic might be involved in these activities. currents evoked in CA1 pyramidal cells of rat by stimula- In the central olfactory system including olfactory bulb, tion of Schaeffer collaterals and in CA3 cells by stimulation anterior olfactory nucleus and piriform cortex, the high of mossy ﬁbers are antagonized by either suramin or pyri- levels of both P2X5 receptor immunostaining and hybrid- doxal phosphate-6-azophenyl-2-4-disulfonic acid (PPADS; ization signal were observed there. In these regions P2X5 Motin and Bennett, 1995). Hippocampal pyramidal neurons receptor immunostaining was observed mainly in the den- respond to exogenous ATP in a concentration-dependent drites and cell bodies, where afferent synaptic junctions manner (Wieraszko and Seyfried, 1989). form, conﬁrming a direct involvement of postsynaptic ATP- This study showed that the P2X5 receptor was expressed gated channels in fast excitatory purinergic transmission in extensively in most subnuclei of the thalamus, particularly the central olfactory system. heavy immunostaining was observed in the anteroventral,

Expression of the P2X5 receptor in three major anteromedial, anterodorsal and ventral–lateral thalamic nu- postsynaptic sites in the hippocampus suggests that ATP- clei. Thus ATP as the P2X receptor agonist might modulate gated ion channels contribute to hippocampal excitatory the functions of the neurons via P2X5 receptor in these nuclei. neurotransmission or modulation in Schaeffer collaterals, The first direct evidence for the fast neurotransmitter role of mossy fibers and the perforant pathway. Previous data ATP in the brain was presented by Edwards et al. (1992), showed 10 min application of 10 ␮M ATP induced long- who identified a P2X receptor-mediated synaptic current in term potentiation (LTP) in rat CA1 neurons, which was the rat medial habenula. The source of ATP in the habenula blocked by co-application of the N-methyl-D-aspartate of rat was later confirmed to come from the terminations of (NMDA) receptor antagonist. In ATP-induced LTP, the the triangular septal and septofimbrial nucleus of rat, and its delivery of test synaptic inputs (once every 20 s) to CA1 co-transmitter, if any, is likely to be glutamate (Sperlágh et al., neurons could be replaced by co-application of NMDA 1998). To our knowledge, no further data that ATP via P2X (100 nM) during ATP perfusion (Fujii, 2004). In mice hip- receptor modulates the activities of neurons in these thalamic pocampal slices, high-frequency electrical stimulation of nuclei are available. It would be very interesting to study the

Schaeffer collaterals has been shown to evoke LTP and effect of ATP via P2X5 receptor on the neurons in these simultaneously induce release of ATP at the nM level nuclei. W. Guo et al. / Neuroscience 156 (2008) 673–692 689

Fig. 13. Distribution of P2X5 receptor immunostaining (A, C, D) and mRNA hybridization signal (B) in the thoracic spinal cord. (A) Low power image of P2X5 receptor immunostaining in the spinal cord. Note the heavy immunostaining in the gray matter (A, C, D). Dense neuropil with P2X5 receptor ir was found in the gray matter (C, D). (C) The high magniﬁcation of the dorsal horn indicated by double stars. (D) The high magniﬁcation of the ventral horn indicated by a star. Note that motor neurons and dense neuropil with positive P2X5 receptor immunostaining were demonstrated. The distribution ϭ ␮ ␮ patterns of cell bodies with P2X5 receptor immunostaining and mRNA hybridization signal matched well (A, B). Scale bars 500 m (A, B); 50 m (C, D).

In this study, we found that P2X5 receptor-ir neurons 1996; Xiang et al., 1998; Kanjhan et al., 1999), combined and ﬁbers were distributed widely, but variably, in different with the results of the present study, shows that homo- regions of the mouse hypothalamus, this result of which meric P2X2, P2X5 or even heteromeric P2X2/P2X5 recep- was similar with our previous data from rat (Xiang et al., tors may be involved in these important functions of mouse 2006). In the hypothalamus, ATP was shown to be in- hypothalamus, but further support is needed from studies volved in regulation of body temperature and hormone using speciﬁc P2 receptor antagonists. secretion (Mori et al., 1992; Kapoor and Sladek, 2000; Neurons in the mesencephalic nucleus of the trigeminal

Gurin et al., 2003). Application of ATP to explants of the nerve have been reported to express P2X2, P2X4, P2X5 and hypothalamo-neurohypophysial system was shown to P2X6 receptor subunit mRNA and protein (Collo et al., 1996; evoke an increase in vasopressin release, a response that Kanjhan et al., 1999). Our identiﬁcation of P2X5 receptor was attenuated by the P2 receptor antagonist PPADS expression is supported by electrophysiological evidence (Mori et al., 1992; Kapoor and Sladek, 2000). This ﬁnding showing suramin-sensitive ATP induced neuronal excitation, was supported by evidence demonstrating a direct input where the ATP-activated currents resemble those observed originating from A1 cells located in the caudal ventrolateral from homo-oligomeric expression of either P2X5 or P2X2 medulla that synapse directly on vasopressin-containing receptor subunits (Khakh et al., 1997). neurons in the supraoptic and paraventricular nuclei and The present study shows heavy immunostaining for utilize ATP as a co-transmitter (Day et al., 1992, 1993). P2X5 receptor in the Purkinje cell and granular cell, espe- Recently, Gourine et al. (2002) showed that ATP acting at cially in the cell body and dendritic tree of cerebellar Pur-

P2X receptors played an important role in central mecha- kinje cell. Localization of P2X5 receptor protein on den- nisms of body temperature control at various ambient tem- drites is compatible with extracellular ATP acting as a fast peratures and during fever. These data indicated that, at excitatory neurotransmitter. Previous data have shown the hypothalamic level, ATP-induced activation of P2X that bath application of ATP increased the amplitude and receptors regulates both hormone secretion and body tem- frequency of spontaneous postsynaptic currents (sPSCs) perature of rat. Expression of P2X2 mRNA and receptor almost twofold in the Purkinje cell (Brockhaus et al., 2004). proteins in these nuclei of hypothalamus (Collo et al., These effects were fully suppressed by the P2 receptor 690 W. Guo et al. / Neuroscience 156 (2008) 673–692

␮ antagonist (PPADS; 10 M). Although the seven P2X spinal cord. Weak to moderate P2X5 receptor immuno- receptor subunits were reported to be expressed in the staining was observed in many other nuclei and regions.

cerebellar cortex (Collo et al., 1996; Kanjhan et al., 1999; The identiﬁcation of extensive P2X5 receptor ir and Xiang et al., 2005a), the pharmacological proﬁle indicated mRNA distribution within the CNS demonstrated here is that the ATP effect was mediated by both P2X and P2Y consistent with a role for extracellular ATP acting as a

receptors, most probably of the P2X5- and P2Y(2,4)-like fast neurotransmitter. subunits (Brockhaus et al., 2004). ATP continuously modulates the cerebellar circuit by increasing the activity of Acknowledgments—The authors thank Dr. Gillian E. Knight for her inhibitory input to Purkinje neurons, thus decreasing the excellent editorial assistance. This work was supported by the main cerebellar output activity, which contributes to loco- National Natural Science Foundation of P. R. China (30670639 to motor coordination (Brockhaus et al., 2004). ZH Xiang) and by Program for Changjiang Scholars and Innova- Area postrema and its adjacent regions including tive Research Team in University (IRT0528 to C. He). solitary tract nucleus, dorsal motor nucleus of vagus and

hypoglossal nucleus showed high levels of P2X5 receptor expression by both immunohistochemistry and in situ REFERENCES hybridization. The neurons in area postrema of cat have Abbracchio MP, Burnstock G (1994) Purinoceptors: are there families been reported to be excited by extracellular ATP (Bori- of P2X and P2Y purinoceptors? Pharmacol Ther 64:445–475. son et al., 1975). Dissociated neurons of the solitary Atkinson L, Batten TF, Moores TS, Varoqui H, Erickson JD, Deuchars tract nucleus and dorsal motor nucleus of vagus are J (2004) Differential co-localisation of the P2X7 receptor subunit excited by extracellular ATP (Ueno et al., 1992a,b; Na- with vesicular glutamate transporters VGLUT1 and VGLUT2 in rat bekura et al., 1995). Microinjection of ATP into solitary CNS. Neuroscience 123:761–768. Barraco RA, O’Leary DS, Ergene E, Scislo TJ (1996) Activation of puri- tract nucleus mediates depressor responses antago- nergic receptor subtypes in the nucleus tractus solitarius elicits spe- nized by suramin in vivo (Ergene et al., 1994; Phillis et cific regional vascular response patterns. J Auton Nerv Syst al., 1997). Thus, extracellular ATP may play a functional 59:113–124. role in synaptic mechanisms of blood pressure control Bo X, Zhang Y, Nassar M, Burnstock G, Schoepfer R (1995) A P2X mediated by the solitary tract nucleus. The solitary tract purinoceptor cDNA conferring a novel pharmacological profile. nucleus receives primary afferents from baro- and che- FEBS Lett 375:129–133. Borison HL, Hawken MJ, Hubbard JI, Sirett NE (1975) Unit activity moreceptors via the vagus and glossopharyngeal from cat area postrema influenced by drugs. Brain Res 92: nerves (Housley et al., 1987). ATP may be coreleased 153–156. with other transmitters (e.g. norepinephrine and neu- Braissant O, Gotoh T, Loup M, Mori M, Bachmann C (2001) Differen- ropeptide Y) from primary afferents in the solitary tract tial expression of the cationic amino acid transporter 2(B) in the nucleus as part of the control of blood pressure (Ergene adult rat brain. Brain Res Mol Brain Res 91:189–195. et al., 1994; Barraco et al., 1996). Localization of the Brockhaus J, Dressel D, Herold S, Deitmer JW (2004) Purinergic P2X (Kanjhan et al., 1999) and P2X receptor subunit modulation of synaptic input to Purkinje neurons in rat cerebellar 2 5 brain slices. Eur J Neurosci 19:2221–2230. of the ATP-gated ion channels in the hypoglossal nu- Brosenitsch TA, Adachi T, Lipski J, Housley GD, Funk GD (2005) cleus is consistent with the suramin/PPADS-sensitive Developmental downregulation of P2X3 receptors in motoneurons excitation of rodent hypoglossal motoneurons and nerve of the compact formation of the nucleus ambiguus. Eur J Neurosci output. 22:809–824. Buell G, Lewis C, Collo G, North RA, Surprenant A (1996) An antagonist-insensitive P2X receptor expressed in epithelia and brain. CONCLUSION EMBO J 15:55–62. Burnstock G (2007a) Physiology and pathophysiology of purinergic In summary, this study provides details of the distribu- neurotransmission. Physiol Rev 87:659–797. tion pattern of the P2X5 receptor subunit of ATP-gated Burnstock G (2007b) Purine and pyrimidine receptors. Cell Mol Life Sci ion channels in the adult mouse CNS using P2X5 recep- 64:1471–1483. tor-specific antisera and riboprobe-based in situ hybrid- Cham JL, Owens NC, Barden JA, Lawrence AJ, Badoer E (2006) P2X purinoceptor subtypes on paraventricular nucleus neurones pro- ization. The distribution of P2X5 receptor mRNA expres- jecting to the rostral ventrolateral medulla in the rat. Exp Physiol sion matches well that of P2X5 receptor protein. Heavy P2X receptor immunostaining was observed in the mi- 91:403–411. 5 Collo G, North RA, Kawashima E, Merlo-Pich E, Neidhart S, Sur- tral cells of olfactory bulb, I, III and V layers of cerebral prenant A, Buell G (1996) Cloning OF P2X5 and P2X6 receptors cortex, globus pallidum, anterior cortical amygdaloid nu- and the distribution and properties of an extended family of ATP- cleus, amygdalohippocampal area of subcortical telen- gated ion channels. J Neurosci 16:2495–2507. cephalon, anterior nuclei, anteroventral nucleus, ventro- Day TA, Sibbald JR, Khanna S (1993) ATP mediates an excitatory lateral nucleus of thalamus; supraoptic nucleus, ventro- noradrenergic neuron input to supraoptic vasopressin cells. Brain medial nucleus, arcuate nucleus of hypothalamus, Res 607:341–344. substantia nigra of midbrain, pontine nuclei, mesence- Day TA, Sibbald JR, Smith DW (1992) A1 neurons and excitatory amino acid receptors in rat caudal medulla mediate vagal exci- phalic trigeminal nucleus, motor trigeminal nucleus, am- tation of supraoptic vasopressin cells. Brain Res 594:244–252. biguous nucleus, inferior olive, hypoglossal nucleus, Edwards FA, Gibb AJ, Colquhoun D (1992) ATP receptor-mediated dorsal motor vagus nucleus, area postrema of hindbrain, synaptic currents in the central nervous system. Nature 359: Purkinje cell of cerebellum, lamina I, and lamina II of 144–147. W. Guo et al. / Neuroscience 156 (2008) 673–692 691

Ergene E, Dunbar JC, O’Leary DS, Barraco RA (1994) Activation of Phillis JW, Scislo TJ, O’Leary DS (1997) Purines and the nucleus P2-purinoceptors in the nucleus tractus solitarius mediate depres- tractus solitarius: effects on cardiovascular and respiratory func- sor responses. Neurosci Lett 174:188–192. tion. Clin Exp Pharmacol Physiol 24:738–742. Florenzano F, Carrive P, Viscomi MT, Ferrari F, Latini L, Conversi D, Pollio F, Staibano S, Mansueto G, De Rosa G, Persico F, De Falco M, Cabib S, Bagni C, Molinari M (2008) Cortical and subcortical Di Lieto A (2005) Erythropoietin and erythropoietin receptor system distribution of ionotropic purinergic receptor subunit type 1 in a large uterine myoma of a patient with myomatous erythrocy- (P2X(1)R) immunoreactive neurons in the rat forebrain. Neuro- tosis syndrome: possible relationship with the pathogenesis of science 151:791–801. unusual tumor size. Hum Pathol 36:120–127. Fujii S (2004) ATP- and adenosine-mediated signaling in the central Ralevic V, Burnstock G (1998) Receptors for purines and pyrimidines. nervous system: the role of extracellular ATP in hippocampal long- Pharmacol Rev 50:413–492. term potentiation. J Pharmacol Sci 94:103–106. Ruan HZ, Burnstock G (2005) The distribution of P2X5 purinergic Gourine AV, Melenchuk EV, Poputnikov DM, Gourine VN, Spyer KM receptors in the enteric nervous system of mouse. Cell Tissue Res (2002) Involvement of purinergic signalling in central mechanisms 319:191–200. of body temperature regulation in rats. Br J Pharmacol Ryten M, Dunn PM, Neary JT, Burnstock G (2002) ATP regulates the 135:2047–2055. differentiation of mammalian skeletal muscle by activation of a Gurin VN, Gurin AV, Melenchuk EV, Spyer KM (2003) The effects of P2X5 receptor on satellite cells. J Cell Biol 158:345–355. activation and blockade of central P2X receptors on body temper- Seguela P, Haghighi A, Soghomonian JJ, Cooper E (1996) A novel ature. Neurosci Behav Physiol 33:845–851. neuronal P2x ATP receptor ion channel with widespread distribu- Housley GD, Martin-Body RL, Dawson NJ, Sinclair JD (1987) Brain tion in the brain. J Neurosci 16:448–455. stem projections of the glossopharyngeal nerve and its carotid Shibuya I, Tanaka K, Hattori Y, Uezono Y, Harayama N, Noguchi J, sinus branch in the rat. Neuroscience 22:237–250. Ueta Y, Izumi F, Yamashita H (1999) Evidence that multiple P2X Kanjhan R, Housley GD, Burton LD, Christie DL, Kippenberger A, purinoceptors are functionally expressed in rat supraoptic neu- Thorne PR, Luo L, Ryan AF (1999) Distribution of the P2X2 re- rones. J Physiol 514(Pt 2):351–367. ceptor subunit of the ATP-gated ion channels in the rat central Soto F, Garcia-Guzman M, Gomez-Hernandez JM, Hollmann M, Kar- nervous system. J Comp Neurol 407:11–32. schin C, Stuhmer W (1996) P2X4: an ATP-activated ionotropic Kapoor JR, Sladek CD (2000) Purinergic and adrenergic agonists receptor cloned from rat brain. Proc Natl Acad SciUSA93: synergize in stimulating vasopressin and oxytocin release. J Neu- 3684–3688. rosci 20:8868–8875. Sperlágh B, Magloczky Z, Vizi ES, Freund TF (1998) The triangular Khakh BS, Humphrey PP, Henderson G (1997) ATP-gated cation septal nucleus as the major source of ATP release in the rat channels (P2X purinoceptors) in trigeminal mesencephalic nucleus habenula: a combined neurochemical and morphological study. neurons of the rat. J Physiol 498:709–715. Neuroscience 86:1195–1207. Kidd EJ, Grahames CB, Simon J, Michel AD, Barnard EA, Humphrey Thomas T, Ralevic V, Bardini M, Burnstock G, Spyer KM (2001) PP (1995) Localization of P2X purinoceptor transcripts in the rat Evidence for the involvement of purinergic signalling in the control nervous system. Mol Pharmacol 48:569–573. of respiration. Neuroscience 107:481–490.

Lê KT, Babinski K, Seguela P (1998a) Central P2X4 and P2X6 channel Ueno S, Harata N, Inoue K, Akaike N (1992a) ATP-gated current in subunits coassemble into a novel heteromeric ATP receptor. dissociated rat nucleus solitarii neurons. J Neurophysiol 68:778–785. J Neurosci 18:7152–7159. Ueno S, Ishibashi H, Akaike N (1992b) Perforated-patch method re- Lê KT, Villeneuve P, Ramjaun AR, McPherson PS, Beaudet A, Seg- veals extracellular ATP-induced Kϩ conductance in dissociated rat uela P (1998b) Sensory presynaptic and widespread somatoden- nucleus solitarii neurons. Brain Res 597:176–179. dritic immunolocalization of central ionotropic P2X ATP receptors. Vorobjev VS, Sharonova IN, Haas HL, Sergeeva OA (2003) Expres- Neuroscience 83:177–190. sion and function of P2X purinoceptors in rat histaminergic neu-

Loesch A, Burnstock G (2001) Immunoreactivity to P2X6 receptors in the rat rons. Br J Pharmacol 138:1013–1019. hypothalamo-neurohypophysial system: an ultrastructural study with ex- Vulchanova L, Arvidsson U, Riedl M, Wang J, Buell G, Surprenant A, travidin and colloidal gold-silver labelling. Neuroscience 106:621–631. North RA, Elde R (1996) Differential distribution of two ATP-gated Loesch A, Miah S, Burnstock G (1999) Ultrastructural localisation of channels (P2X receptors) determined by immunocytochemistry.

ATP-gated P2X2 receptor immunoreactivity in the rat hypothalamo- Proc Natl Acad SciUSA93:8063–8067. neurohypophysial system. J Neurocytol 28:495–504. Vulchanova L, Riedl M, Shuster SJ, Buell G, Surprenant A, North RA,

Mori M, Tsushima H, Matsuda T (1992) Antidiuretic effects of purino- Elde R (1997) Immunocytochemical study of the P2X2 and P2X3 ceptor agonists injected into the hypothalamic paraventricular nu- receptor subunits in rat and monkey sensory neurons and their cleus of water-loaded, ethanol-anesthetized rats. Neuropharma- central terminals. Neuropharmacology 36:1229–1242. cology 31:585–592. Weisman GA, Turner JT, Fedan JS (1996) Structure and function of Motin L, Bennett MR (1995) Effect of P2-purinoceptor antagonists on P2 purinocepters. J Pharmacol Exp Ther 277:1–9. glutamatergic transmission in the rat hippocampus. Br J Pharmacol Wieraszko A, Seyfried TN (1989) ATP-induced synaptic potentiation in 115:1276–1280. hippocampal slices. Brain Res 491:356–359. Nabekura J, Ueno S, Ogawa T, Akaike N (1995) Colocalization of ATP Xiang Z, Burnstock G (2005a) Changes in expression of P2X purino- and nicotinic ACh receptors in the identiﬁed vagal preganglionic ceptors in rat cerebellum during postnatal development. Brain Res neurone of rat. J Physiol 489(Pt 2):519–527. Dev Brain Res 156:147–157. North RA (2002) Molecular physiology of P2X receptors. Physiol Rev Xiang Z, Burnstock G (2005b) Expression of P2X receptors in rat 82:1013–1067. choroid plexus. Neuroreport 16:903–907. Oglesby IB, Lachnit WG, Burnstock G, Ford APDW (1999) Subunit Xiang Z, Bo X, Oglesby I, Ford A, Burnstock G (1998) Localization of

speciﬁcity of polyclonal antisera to the carboxy terminal regions of ATP-gated P2X2 receptor immunoreactivity in the rat hypothala-

P2X receptors P2X1 through P2X7. Drug Dev Res 47:189–195. mus. Brain Res 813:390–397. Paxinos G, Franklin KBJ (2001) The mouse brain in stereotaxic coor- Xiang Z, Jiang L, Kang Z (2001) Transient expression of somatostatin dinates. New York: Academic. mRNA in developing ganglion cell layers of rat retina. Brain Res Phillis JW, Edstrom JP, Kostopoulos GK, Kirkpatrick JR (1979) Effects Dev Brain Res 128:25–33. of adenosine and adenine nucleotides on synaptic transmission in Xiang Z, Lv J, Jiang P, Chen C, Jiang B, Burnstock G (2006) the cerebral cortex. Can J Physiol Pharmacol 57:1289–1312. Expression of P2X receptors on immune cells in the rat li- 692 W. Guo et al. / Neuroscience 156 (2008) 673–692

ver during postnatal development. Histochem Cell Biol 126: APPENDIX 453–463. Yao ST, Barden JA, Lawrence AJ (2001) On the immunohistochemical Supplementary data associated with this article can be found, distribution of ionotropic P2X receptors in the nucleus tractus soli- in the online version, at doi: 10.1016/j.neuroscience.2008. tarius of the rat. Neuroscience 108:673–685. 07.062.

(Accepted 26 July 2008) (Available online 7 August 2008)