Distribution of the 5-Hydroxytryptamine2CReceptor Protein inAdult Rat Brainand Spinal Cord Determined Usinga Receptor-DirectedAntibody:Effect of 5,7-Dihydroxytryptamine

ANITA SHARMA, TANIYA PUNHANI, AND KEVIN C.F. FONE*Department of Physiology and Pharmacology, Queen’s Medical Centre, Nottingham University,

Nottingham NG7 2UH, England

KEY WORDS 5-hydroxytryptamine2C receptor; 5,7-dihydroxytryptamine; receptor radio-immunoassay; serotonin receptor; choroid plexus; immunocytochemistry

ABSTRACT A synthetic peptide, corresponding to the N-terminal decapeptide(1Y11C12) of the rat 5-hydroxytryptamine2C (5-HT2C) receptor protein was used to producea sheep polyclonal antiserum.Western blot analysis showed that the resultant antibody G241recognised two membrane proteins, one (55 kDa) approximating the molecular mass of the5-HT2C receptor (52 kDa) and a second (63 kDa), which may be a glycosylated form of thereceptor protein. HEK 293 cells transfected with human 5-HT2C cDNAdisplayed intense cellsurface immunoreactivity with the 5-HT2C antiserum, which was completely prevented byincubating the antibody with the synthetic 5-HT2C peptide (10 µM), whilst neither non-immune serum nor untransfected cells displayed any immunoreactivity. A radioimmuno-assay was developed to quantify the regional distribution of 5-HT2C-like immunoreactiv-ity (LI) in the adult rat brain. The choroid plexus contained five-fold higher levels of5-HT2C-LI than any brain region but high levels were found in the frontal cortex, septum,hypothalamus, and striatum, intermediate levels in the thalamus and midbrain, andlower levels in brainstem, cerebellum, and spinal cord. In rat cortical membranes, theBmax value from [3H]-mesulergine binding was ten-fold lower than 5-HT2C-LI levelsdetermined by radioimmunoassay, which may reflect measurement of internalisedreceptor protein by radioimmunoassay which is not detected with conventional 5-HT2C

ligands. Ten days after depletion of 5-HT with the serotonergic neurotoxin 5,7-dihydroxytryptamine (5,7-DHT), there was a significant increase in 5-HT2C-LI in thechoroid plexus and the ventral cervical spinal cord, suggesting that receptors therein arelocated post-synaptic to destroyed serotonergic nerve terminals. In contrast, the signifi-cant reduction in 5-HT2C-LI observed in the midbrain, brainstem, and dorsal thoracicspinal cord following 5,7-DHT implies that 5-HT2C receptors may be located on 5-HTnerve terminals in these regions. Synapse 27:45–56, 1997. r 1997 Wiley-Liss, Inc.

INTRODUCTION

The 5-hydroxytryptamine2C (5-HT2C) receptor wasoriginally described as a 5-HT1 subtype using [3H]-mesulergine binding in pig choroid plexus homogenates(Pazos et al., 1984). During the last decade, however,this receptor has been shown to share a more commonprotein structure, receptor-effector coupling, and phar-macology with the 5-HT2 than the 5-HT1 receptorfamily, culminating in its re-classification (Hoyer et al.,1994; Humphrey et al., 1993) as the 5-HT2C receptor.The original 5-HT2 receptor was consequently renamed5-HT2A and the thirdmember of this subfamily, initially

Abbreviations: BSA, bovine serum albumin; CNS, central nervous system;EDTA, ethylene diamine tetraacetic acid; HPLC, high performance liquidchromatography; i.c.v., intracereboventricular; IgG, immunoglobulin G; KLH,Keyhole Limpet haemocyanin; MBS, m-maleimidobenzoic acid N-hydroxysuccin-imide ester; PAGE, polyacrylamide gel electrophoresis; PBS, phosphate bufferedsaline; PMSF, phenylmethylsulphonyl fluoride; RIA, radioimmunoassay; SDS,sodium dodecylsulphate; 5-HT, 5-hydroxytryptamine (serotonin); 5-HT2C-LI,5-HT2C-like immunoreactivity; 5-HIAA, 5-hydroxyindolacetic acid; 5,7-DHT,5,7-dihydroxytryptamine.

Contract grant sponsor: BBSRC; Contract grant sponsor: SmithKline BeechamPharmaceuticals.

Anita Sharma is now at the Department of Medicine, Queen ElizabethHospital, Edgbaston, Birmingham, B15 2TH, England.

*Correspondence to: Kevin C.F. Fone, Department of Physiology and Pharma-cology, Queen’s Medical Centre, Clifton Boulevard, Nottingham University,Nottingham NG7 2UH, England.

Received 22April 1996; Accepted 14 February 1997

SYNAPSE 27:45–56 (1997)

r 1997 WILEY-LISS, INC.

termed 5-HT2F because of its isolation from the stomachfundus (Kursar et al., 1992; Fognet et al., 1992;Wainscott et al., 1993), was renamed 5-HT2B. As pre-dicted from the 70% amino acid homology in theputative transmembrane spanning domains (contain-ing agonist and antagonist interaction sites) of thethree 5-HT2 subtypes (Julius et al., 1988, 1990; Kursaret al., 1992), they share extremely similar pharmacologi-cal profiles, such that most 5-HT agonists and antago-nists have comparable affinities for each (Baxter et al.,1995; Boess and Martin, 1994; Hoyer, 1988; Kennett,1993; Kursar et al., 1992; Pazos et al., 1984).Despite these similarities, each 5-HT2 receptor sub-

type has a very divergent distribution in the peripheraland central nervous systems, suggesting that theysubserve different functions. The 5-HT2B receptor islocated in the rodent stomach fundus but may also havea restricted distribution in the CNS (Kursar et al.,1992; Loric et al., 1992), low levels being expressed inspecific brain areas in rat (Duxon et al., 1995, 1997),mouse (Loric et al., 1992), and man (Bonhaus et al.,1995). In contrast, the highest 5-HT2A levels occur incortical layers IV and V with intermediate levels inmany other 5-HT terminal areas (Pazos et al., 1985).Highest 5-HT2C receptor levels are located in the cho-roid plexus, a non-neuronal, epithelial tissue lining theventricles (Pazos et al., 1984; Yagaloff and Hartig,1985), where it may regulate cerebrospinal fluid compo-sition and production. However, 5-HT2C mRNA andligand binding sites have a comparable widespreadCNS distribution, being prevalent in all raphe, cranialmotor, and many catecholaminergic nuclei (Hoffmanand Mezey, 1989; Mengod et al., 1990; Molineaux et al.,

1989; Pompeiano et al., 1994). This is consistent withthe proposal that 5-HT2C receptors may participate inseveral central effects of 5-HT. Indeed, 5-HT2C receptorshave been implicated in anxiety, neuroendocrine regula-tion of adrenocorticotrophic hormone (ACTH), prolac-tin, and corticosterone, as well as migraine, locomotorco-ordination, and feeding (Baxter et al., 1995; Fone etal., 1996; Fozard and Gray, 1989; Kennett and Curzon,1988; Kennett, 1993; Lee et al., 1991; Van der Karr andBrownfield, 1993) and possibly epilepsy (Tecott et al.,1995).The current lack of selective antagonists with which

to discriminate between the effects of each 5-HT2receptor subtype is however a major factor impedingtheir full functional characterisation (Baxter et al.,1995; Hoyer et al., 1994). To attempt to further estab-lish the distribution and function of the 5-HT2C recep-tor, we have produced, and characterised, a polyclonalantiserum (G241) directed against amino acids 1-10 inthe predicted N-terminal extracellular tail of the rat5-HT

2Creceptor protein (Fig. 1). In the present report,

antiserum G241 has been used both with immunocyto-chemistry to visualise receptors in a suitable trans-fected cell line and to develop a radioimmunoassay toquantify the regional distribution of 5-HT2C receptorprotein-like immunoreactivity (5-HT2C-LI) in adult ratbrain. In addition, the effect of the 5-HT neurotoxin,5,7-dihydroxytryptamine (5,7-DHT), on 5-HT2C-LI inthe rat brain is also examined to establish the pre- orpost-synaptic relationship of 5-HT2C receptors to sero-tonergic nerve terminals. Preliminary accounts of partof this work have been published elsewhere (Sharma etal., 1992, 1994).

Fig. 1. Diagramatic representation of the rat 5-HT2C receptorprotein indicating the synthetic amino acid sequence used to generatepolyclonal antisera against the N-terminus of the 5-HT2C receptor,together with the corresponding [aligned using GCG database (Kursaret al., 1992)] sequence from the rat 5-HT2A and the 5-HT2B receptors.

Arrows (with position numbers) indicate asparagine residues that arewithin the consensus sequence Asn-X-Ser/Thr thought suitable forN-glycosylation. The equivalent human 5-HT2C N-terminal sequenceas expressed in HEK 293 cell line used in this study is shown forcomparison.

46 A. SHARMA ET AL.

MATERIALS AND METHODSPeptide synthesis

The amino acid sequence corresponding to theN-terminal decapeptide of the rat 5-HT2C receptor[MVNLGNAVRS (Julius et al., 1988)] with additionaltyrosine (Y11) and cysteine (C12) residues at the car-boxyl terminus was synthesised using Fmoc chemistryon a polyamide support (Model 431A peptide synthe-siser, Applied Biosystems, Foster City, CA). This se-quence was chosen on the basis of least homology withotherG-protein coupled receptors (including other 5-HT

2receptors) and high predicted immunogenicity, by usingthe BBSRC Genetics Computer Group software pack-age at Daresbury. The tyrosine residue was included toradiolabel the peptide with [125Iodine] for radioimmuno-assay (RIA) and the cysteine enabled specific conjuga-tion to the carrier peptide, Keyhole Limpet Haemocya-nin (KLH, Sigma Chemical Co., St. Louis, MO). Thepurity of the synthetic peptide was approximately 80%as assessed using HPLC (data not shown).

Antigen preparation and immunisation

The 5-HT2C peptide (10 mg ml21 in 0.1 M phosphatebuffer, pH 6.5) was conjugated to KLH (20 mg ml21 in0.01 M phosphate buffer, pH 7.0) via m-maleimidoben-zoic acid N-hydroxysuccinimide ester (MBS, SigmaChemical Co., St. Louis, MO) by drop-wise addition andconstant stirring for 30 min at room temperature.Undissolved KLH was removed by centrifugation(1,000g for 10min, 4°C), and excessMBSwas separatedfrom the KLH-MBS complex by using a G25 Sephadexcolumn by elution in 0.1 M phosphate buffer (pH 6.5).Synthetic peptide (10 mg ml21 in 0.1 M phosphatebuffer containing 2 mM ethylenediaminetetra-aceticacid (EDTA), pH 6.5) was purified by elution with 0.1 Mphosphate buffer (pH 6.5) through a G25 Sephadexcolumn, before careful addition of the KLH-MBS com-plex at 20 M excess (stirred 4 h, room temperature) andadjustment to pH 7.2. The peptide-KLH conjugate wasdialysed extensively against phosphate buffered saline(PBS, pH 7.2) and stored at 220°C until used.Two adult sheep were inoculated (multisite i.m. and

s.c. injection) with the 5-HT2C peptide-KLH conjugate(3 mg ml21 protein) emulsified in an equal volume ofFreund’s complete adjuvant (2 ml, Difco, Detroit, MI)followed by four further monthly injections (s.c.) inFreund’s Incomplete Adjuvant. Two weeks after thefinal injection, sheep were exsanguinated under so-dium thiopentone anaesthesia (15 mg kg21 i.v.) and theresultant antisera were dialysed extensively againstPBS at 4°C (to remove endogenous 5-HT as confirmedby HPLC-ED) before addition of sodium azide (0.02%w/v) and 1 mM thimerosal and storage at 4°C. Thehighest titre sheep antiserum (G241) from the finalbleed was used in all subsequent experiments.

For the purpose of immunocytochemistry antiserumG241 was precipitated with 33% w/v ammonium sul-phate, dialysed against PBS, and an immunoglobulin(IgG) fraction isolated by size exclusion chromatogra-phy (Sephadex G100 column, elution with 0.3 M phos-phate buffer, pH 7.3) as confirmed by SDS-PAGE (datanot shown).

Detection and characterisation of antibodiesCell culture

Stable transfected HEK 293 cells expressing thehuman 5-HT2C receptor cDNA [20 pmol mg21 protein(Carey et al., 1996)] and untransfected controls werediluted to give approximately 5.0 3105 cells ml21 andaliquoted into 75 mm2 culture flasks. Cells were main-tained in Dulbecco’s modified Eagle’s medium supple-mented with 2 mM L-glutamine and 10% v/v foetal calfserum at 37°C in a water saturated atmosphere of 10%v/v CO2 in air. Penicillin 100 U ml21, streptomycin 100µg ml21, and amphotericin B 0.25 µg ml21 were alsoadded during the first 24 h of culture. The medium waschanged after 1 and 3 days in vitro and thereafter every5 days.

Immunocytochemistry

To characterise antiserum G241 and confirm specific-ity of binding, 5-HT2C receptor transfected and untrans-fected HEK 293 cells were grown to confluence on 10mm glass coverslips. At confluence, cells were washedthree times with PBS, (NaCl, 137; KCl 2.68; KH2PO4,

1.47; Na2HPO4, and 8.10 mM, pH 7.4) and fixed with10% v/v neutral buffered formalin for 10 min at roomtemperature. Following washing, a 20 min incubationwith 0.1% w/v bovine serum albumin (BSA) in PBS wasused to block non-specific antibody binding. Cells werethen incubated overnight at 4°C with 100 µl of eitherSephadex purified (described above) 5-HT2C antiserumor non-immune antiserum (1:10 in blocking solution) inthe presence or absence of synthetic peptide (10 µM).The following day, coverslips were washed (three timesin PBS) and incubated with fluorescein-conjugatedrabbit anti-sheep antiserum (100 µl, DAKO, diluted1:20 in blocking solution) for 60 min at room tempera-ture in the dark. Coverslips were then rinsed in PBS,dried, and mounted in 1:1 glycerol:PBS with 0.1% w/vBSA. Fluorescent micrographs were taken by using aZeiss transmitted-light photomicroscope III.

Western Blot Analysis

Rat brain regions and tissue from cell lines werehomogenised in 15 vol of extraction buffer (50 mM TrisHCl containing 150 mM NaCl, 1 mM EDTA, 10 mMphenylmethylsulphonyl fluoride (PMSF), 1% v/v TritonX-100, and 5% v/v aprotinin, pH 7.5) and precipitatedby centrifugation (1,000g for 10 min, Beckman Mi-

475-HT2C RECEPTOR RADIOIMMUNOASSAY

crofuge 12). The supernatant was then centrifuged(36,000g for 30 min at 4°C Sigma model 3K20) and thepellet resuspended in 10 vol by weight of extractionbuffer. Each sample was solubilised by boiling (1:1 in0.125 M Tris HCl containing 4% v/v SDS, 20% v/vglycerol, 10% v/v 2-mercaptoethanol, and 0.02 % w/vbromophenol blue, pH 6.8) for 5 min. Solubilised mem-branes at equivalent protein concentrations (5 µg) wereseparated by a 12% discontinuous polyacrylamide gel(Bio-Rad mini-protean II) and transferred (using 0.76M glycine containing 0.1% v/v Tris HCl and 10% v/vmethanol) to a nitrocellulose membrane (63 V, 4 h at4°C, Hoefer, confirmed by Ponceau Red staining) aspreviously described (Duxon et al., 1997). Non-specificbinding was prevented with blocking buffer (3% w/vcasein and 0.25% v/v Tween 20 in 4 mM KH2PO4, 18mM Na2HPO4, 115 mM NaCl, pH 7.3) and nitrocellu-lose was incubated with sephadex purified non-immuneor 5-HT2C IgGs (1:50 in blocking buffer for 12 h at roomtemperature), either alone or in combination with 100µM 5-HT2C peptide antigen. Nitrocellulose membraneswere washed with PBS and immunoreactive proteinsdetected using rabbit anti-sheep peroxidase conjugatedsecondary immunoglobulins (DAKO, 1:500 in blockingbuffer for 1 h at room temperature) and visualised withdiaminobenzidine tetrahydrochloride (Sigma ChemicalCo., St. Louis, Mo) with nickel enhancement. TheWestern blot image was captured and the digital imageanalysed by computer (NIH Image version 1.49) toaccurately determine the molecular mass of immu-nopositive bands but original photographs are shown.

Iodination of the 5-HT2C receptor peptidefor radioimmunoassay (RIA)

The 5-HT2C receptor peptide was radiolabelled at theY11 residue using [125Iodine] and a modification of themethod of Bassiri andUtiger (1972). Briefly, peptide (15µg) was incubated with [125I]-sodium iodide (37MBq, 10µl) in 0.05 M phosphate buffer (20 µl, pH 7.4 for 30 sec)before stopping the reaction with 0.01 M sodium meta-bisulphite (250 µl) and 0.06 M potassium iodide (100µl). Unincorporated [125I]-sodium iodide was removedby eluting the acidified mixture through a methanol-activated Sep-pak C18 cartridge (Waters and Associ-ates, Milford, MA) with 0.5 M trifluoroacetic acid untilthe gamma radiation (1 ml fractions) declined to back-ground levels. Iodinated peptide was then eluted as asingle sharp peak (0.5ml fractions) in 0.5M trifluoroace-tic acid in 90% v/v methanol, which was diluted in RIAbuffer (1:60, 0.06 M phosphate buffer, pH 7.4, contain-ing 0.01 M EDTA, 0.5% w/v BSA, and 5% v/v aprotinin)and stored in aliquots at 220°C.

Preparation of rat CNS tissue and developmentof 5-HT2C RIA

Following decapitation, selected brain and spinalcord regions from adult Wistar rats (240–360 g, n 5

8-12) were rapidly isolated on a cool tray (4°C) and anymeningeal membrane and blood carefully removedbefore being frozen in liquid nitrogen and stored at260°C prior to the determination of 5-HT2C receptorprotein-like immunoreactivity (5-HT2C-LI) by RIA(Foneet al., 1996). Tissues were homogenised (in 1 ml of 50mM Tris-HCl, pH 7.5, as for Western blots above) andmembrane preparations (P2 fraction following centrifu-gation twice at 36,000g for 20 min at 4°C) resuspendedin a final volume of RIA buffer to enable each region tobe assayed in triplicate from the linear portion of theRIA curve. Serial dilutions of the synthetic rat 5-HT2Creceptor peptide (10-4,000 pg tube21) or tissue homoge-nates (100 µl) were incubated (4°C for 24 h) withantiserum G241 (50 µl, 1:5,000) and [125I]-5-HT2C pep-tide (50 µl, diluted to give approximately 5,000 cpm inRIA). Free label was precipitated (2,000g for 30 min,4°C, Mistral 6000, Fisons) following incubation (roomtemperature, 30 min) with human plasma (50 µl) andcharcoal (300 µl, 5 g l21 in 0.012 M phosphate buffercontaining 0.154MNaCl and 2.53 1025M dextran) andcounted for 180 sec on an LKB gamma counter (Model1272 Clinigamma). Antiserum G241 showed no cross-reactivity (up to 10 µg tube21) with L-tryptophan, 5-HT,and its major metabolite 5-hydroxyindoleacetic acid(5-HIAA). Importantly, antiserum G241 also failed toshow any cross-reactivity with synthetic peptides(up to 10 µg tube21), corresponding to portions of theN-terminus of the rat 5-HT2A (amino acids 35-50,RDANTSEASNWTIDAE), rat 5-HT2B (amino acids 1-12,MASSYKMSEQST), or an alternative portion of the rat5-HT2C (amino acids 42–56, DGGRLFQFPDGVQNW)receptor, or a variety of neuropeptides co-localised with5-HT in bulbospinal raphe nerve terminals includingsubstance P, thyrotrophin-releasing hormone, and calci-tonin gene-related peptide. In order to calculate levelsof 5-HT2C-LI in tissue samples, a 1 to 1 stoichiometricrelationship between authentic receptor protein andthe synthetic decapeptide for the antibody binding siteswas assumed. Western blot analysis with membranepreparations from the same regions as used for RIAwere performed to identify the molecular mass of theproteins recognised by the current antiserum. In drug-free rats the cortical 5-HT2C receptor levels were com-pared using the current RIA and conventional [3H]-mesulergine binding but comparison in more discretebrain nuclei was impossible due to the lack of sensitiv-ity of ligand binding.

5,7-dihydroxytryptamine treatment

Guide cannulae were implanted above the left lateralventricle [A, 20.85; L, 11.5; D, 22.6 mm from Bregma(Paxinos and Watson, 1986)] in adult male Wistar rats(320–400 g) under sodium methohexitone anaesthesia(60 mg kg21 i.p.). Ten days after cannulation, ratsreceived either 5,7-dihydroxytrytamine (5,7-DHT, 150µg, 5 µl in 0.154 M saline containing 5.68 mM ascorbic

48 A. SHARMA ET AL.

acid, n 5 8) or vehicle (n 5 10) both given twice at 24 hinterval and after pre-treatment with desipramine (15mg kg21 i.p. 230 min) to prevent uptake of 5,7-DHTinto catecholaminergic neurones.

[3H]-mesulergine binding and indoleaminemeasurement

Ten days after the last 5,7-DHT or vehicle injection,or on removal of untreated animals from the home cage,rats were decapitated and selected brain and spinalcord regions isolated on a cool tray (4°C). In 5,7-DHTtreated rats, each region was separated into bilateralhalves to measure 5-HT and 5-HIAA by HPLC withelectrochemical detection and 5-HT2C-LI in total particu-late fractions by RIA (described above). In six untreatedrats [3H]-mesulergine binding (see below) was alsoperformed using the cortex. Rat brain and spinal cordtissue was homogenised in 50 mM Tris-HCl (pH 7.4,4°C) centrifuged (36,000g, 15 min, 4°C) and the super-natant discarded. The pellet was resuspended in theoriginal volume of 50 mM Tris-HCl and centrifugationand resuspension repeated. The resultant particulatefraction was assayed in triplicate using the RIA previ-ously described for 5-HT2C-LI and for protein content(Lowry et al., 1951).In bilateral tissue sections the content of 5-HT and

5-HIAAwas determined following separation by HPLCwith a 10 cm column (2 mm internal diameter) packedwith 3 µmODS Spherisorb by using a phosphatemobilephase (150 mM Na2HPO4 2H2O; 18.8 µM 1-octanesulphonic acid, 730 µM EDTA; 14% v/v methanol, pH3.8) and detected using an LC4 amperometric detector(Bioanalytical Systems, Inc., W. Lafayette, IN) and aglassy carbon electrode (10.65 V).5-HT2C ligand binding was determined in aliquots of

cortical membranes (with frontal cortex removed tominimise 5-HT2A sites) using [3H]-mesulergine, accord-ing to the method of Pranzatelli (1992). In brief, cortexwas homogenized in Tris-HCl (20 ml 50 mM, pH 7.6 at4°C) using a Polytron. Further Tris (20 ml) was addedprior to centrifugation (36,000g, 15 min at 4°C, Sigma3K20), the supernatant discarded, and this procedurerepeated four times (the resuspended pellet being incu-bated at 37°C for 15 min after the third spin to removeendogenous 5-HT) before resuspension in Tris (3.5 ml)and storage at 280°C. Mesulergine binding was as-sayed in a final volume of 250 µl, consisting of [3H]-mesulergine (25 µl) and ketanserin (10 µM, 25 µl) orbuffer blank (50 mM Tris-HCl, pH 7.6 at 4°C), bufferblank (175 µl), and cortical homogenate (25 µl). After 15min incubation at 37°C mixtures were filtered (FP-100Whatman GF/B presoaked for ,6 h in 0.05% polyethyl-enimine to minimise non-specific binding) by using aBrandel Cell Harvester and radioactivity counted byliquid scintillation spectroscopy in 4 ml of LSC-cocktail(LKBWallac, Rackbeta, beta-counter).

Statistical analysis

5-HT and 5-HT2C receptor levels are expressed asmean6 s.e.m. and change in tissue levels with 5,7-DHTwas analysed by using Student’s t-test (P , 0.05 beingconsidered significant).

RESULTSWestern blot

In solubilised membrane preparations from the cho-roid plexus, frontal cortex, hippocampus, septum (Fig.2A), and a transfected cell-line expressing human5-HT2C receptor protein (data not shown), the Sephadex-purified immunoglobulin fraction of the 5-HT2C antise-rum G241 identified two proteins (55 and 63 kDa) ofequal intensity, the lighter in weight being of similarmolecular mass to the 5-HT2C receptor (52 kDa) (Juliuset al., 1988). As would be expected if this staining wasspecific, non-immune antiserum used at the samedilution failed to recognise any proteins in Westernblots of the same membranes. Notably, the intensity ofthe immunopositive bands was at least as great in thechoroid plexus as in any brain region, when run at thesame protein concentration, in agreement with bothour own radioimmunoassay data herein and previousligand binding (Mengod et al., 1990), which show thedensity of 5-HT2C receptors to be greatest in the choroidplexus. Both immunopositive bands were also virtuallyabolished in blots performed with hippocampal andseptal membranes when the 5-HT2C antiserum G241was co-incubated with synthetic 5-HT2C peptide anti-gen (100 µM, Fig. 2B).

Immunocytochemistry

The intense immunofluorescent staining of HEK 293cells (transfected with the human 5-HT2C receptorcDNA) obtained by incubation with the rat N-terminaldirected 5-HT2C antiserum (Fig. 3A) was associatedwith the cell membrane. However, as expected for areceptor protein that is synthesised internally prior toincorporation in the cell membrane, immunofluores-cence was also apparent in the cytoplasm of these cellsbut absent from the nucleus.All the immunofluorescentstaining of transfected HEK 293 cells was abolished byco-incubation of antiserum G241 with the synthetic rat5-HT

2Cpeptide (Fig. 3B), which has 80% amino acid

sequence homology with the corresponding humanN-terminal sequence (Saltzman et al., 1991), thusestablishing the specificity of the staining. In addition,both transfected HEK 293 cells incubated with non-immune antiserum at the same dilution (Fig. 3C) andnon-transfected control cells incubated with 5-HT2Cantiserum (Fig. 3D) failed to demonstrate any cellmembrane or cytoplasmic staining under identical con-ditions.

495-HT2C RECEPTOR RADIOIMMUNOASSAY

Measurement and distribution of 5-HT2C-LI

A sensitive RIA (detecting 23–2,300 fmol 100 µl21

synthetic peptide) was developed using antiserum(G241, 1:5,000 dilution), which showed no cross-reactivity with 5-HT or a variety of neuropeptides andmonoamine neurotransmitters and metabolites (up to10 µg tube21, Fig. 4A). Preliminary experiments alsoconfirmed that the immunoreactivity in tissue extractswas not detectable in the supernatant but only presentin the P2 containing cell membrane fraction (data notshown). The amount of [3H]-mesulergine binding inaliquots of cortical membranes (Bmax 41 6 8 fmol mgprotein21, KD 1.856 0.82 nM) agrees with that reportedpreviously (Pazos et al., 1984) but was ten-fold lowerthan the level of 5-HT2C-LI (450 6 40 fmol mg pro-tein21) measured in this tissue with the current RIA.5-HT2C-LI levels detected by this RIA showed a

marked regional variation (Fig. 4B), being five timeshigher in the choroid plexus (2,1266 174 fmol mg21 wetweight, mean 6 s.e.m., n 5 8) than in any brain orspinal cord region (n 5 8-10). Within the brain, thehighest levels of 5-HT2C-LI were found in the septum(457 6 55 fmol mg21 wet weight), frontal cortex (368 6

69), hypothalamus (348 6 57), striatum (300 6 60), andslightly lower levels in the hippocampus (220 6 60),thalamus (156 6 21), mid brain (136 6 26), and lowestlevels in the cerebellum (98 6 18), brainstem (pons andmedulla, 70 6 15), dorsal (96 6 18), and ventral (60 611) thoraco-lumbar spinal cord. Recovery of 5-HT2C-LIfrom ventral spinal cord homogenates was 84 6 6%(mean 6 s.e.m., n 5 10). The intra-assay coefficient ofvariation measured in cortical homogenates (n 5 15)was 7.5% and the inter-assay coefficient of variation asdetermined from ED50 values in RIA standard curves(n 5 8) was 14.9%.

Effect of 5,7-DHT on 5-HT and 5-HT2C-LI in ratbrain and spinal cord

5,7-DHT (2 3 150 µg i.c.v) produced the expectedsignificant depletion of 5-HT in the choroid plexus (Fig.5, 265% from vehicle treated controls, P 0.05, Student’st-test), frontal cortex (-66%, P 0.05), and dorsal thoracic(-90%, P 0.05) and ventral cervical (-84%, P 0.001)spinal cord regions. Less marked depletion’s were ob-served in the brainstem (-37%, P 0.05) and midbrain(-41%, n.s.). 5-HIAA levels showed a similar pattern of

Fig. 2. A representative example of ten Western blot analyses ofmembrane preparations from choroid plexus (ChP), hippocampus(HIP), septum (SEP), and frontal cortex (F.C.) (left to right-hand lanesas indicated, blot A), in which immunoreactive proteins (55 and 63kDa, arrowed) have been visualised by sequential incubation with aSephadex-purified 5-HT2C receptor-directed sheep antiserum, alkalinephosphatase conjugated rabbit anti-sheep immunoglobulins and DAB,

compared with molecular mass standards (kDa, as indicated againstthe left lane). Non-immune antiserum failed to identify any proteinunder the same conditions (data not shown) and both immunopositivebands were virtually abolished from a blot (B) of hippocampal andseptal membranes when the 5-HT2C receptor-directed antiserum wasco-incubated with 100 µM synthetic 5-HT2C peptide antigen.

50 A. SHARMA ET AL.

Fig. 3. Immunofluorescence (right hand) and corresponding phasecontrast micrographs (left hand side) of HEK 293 cells transfectedwith human 5-HT2C cDNA (A, B, and C) and non-transfected controls(D) incubated with a purified rat 5-HT2C N-terminal directed sheepantiserum G241 (1:20 dilution) in the presence (B) or absence (A andD) of synthetic 5-HT2C peptide (10 µM) or non-immune antiserum

without 5-HT2C peptide (C), visualised by incubation with fluorescein-conjugated rabbit anti-sheep antiserum, by using identical exposuretimes and camera settings. Note the specific immunostaining of5-HT2C-transfected cells (A) is completely prevented by co-incubationwith synthetic 5-HT2C decapeptide (B). Bar 5 5 µM.

Fig. 4. A: Typical radioimmunoassay standard curve (mean 6s.e.m., n54 each concentration) using synthetic 5-HT2C peptide anti-gen (open circles) and the lack of displacement of [125I] 5-HT2C peptideantigen by a variety of indoleamines (5-hydroxytryptamine; 5-HT,5-hydroxyindoleacetic acid; 5-HIAA or tryptophan), neuropeptides(thyrotropin-releasing hormone; TRH and substance P), or the rat5-HT2A receptor N-terminal peptide sequence amino acids 35-50 (as

indicated). B: 5-HT2C-LI levels (fmol mg21 wet weight mean 6 s.e.m.,n58-10) in the choroid plexus (Chp), septum (Sep), frontal cortex (Fc),hypothalamus (Hyp), striatum (Str), hippocampus (Hip), thalamus(Thal) midbrain (Mb), cerebellum (Cer), brainstem (BS, pons andmedulla) dorsal spinal (Dsc), and ventral spinal cord (Vsc) of adultmale Wistar rats.

52 A. SHARMA ET AL.

depletion to that of 5-HT in all regions examined (datanot shown). A significant increase in 5-HT2C-LI accom-panied the depletion of 5-HT both in the choroid plexus(2,724.0 6 438.0 compared with 4,418.5 6 649.8 fmolmg21 protein, P 0.05) and ventral cervical spinal cord(44.3 6 9.4 in vehicles compared with 134.2 6 29.8 fmolmg21 protein, after 5,7-DHT, P 0.001). In contrast,significant reductions in 5-HT2C-LI were observed inmidbrain (-45% from vehicle treated controls, P 0.01),brainstem (-48%, P 0.01), and dorsal thoracic spinalcord (-31.7%, P 0.01). 5-HT2C-LI was also markedlyreduced in the septum (-50% from vehicle), althoughthis just failed to reach significance (data not shown).Interestingly, the significant depletion of 5-HT in thefrontal cortex and several other regions examined,including hippocampus and thalamuswas not accompa-nied by a change in 5-HT2C-LI, demonstrating regionalvariation in the pre- and post-synaptic location of5-HT2C receptor proteins.

DISCUSSION

The present study describes the production andcharacterisation of a polyclonal antiserum (G241) raisedagainst a synthetic peptide (Fig. 1) corresponding to theN-terminal ten amino acids of the rat 5-HT2C receptorprotein (plusY11,C12). The specificity of antiserumG241

for the 5-HT2C receptor protein was confirmed byimmunocytochemical visualisation in HEK 293 cellstransfected with human 5-HT2C receptor cDNA. Al-though cytoplasmic as well as cell membrane stainingwas present, this would be anticipated because someprotein will be in transit from synthesis in the cell body.All immunofluorescence was displaced by co-incubationwith the synthetic rat 5-HT2C peptide as expected, sincethe human N-terminal receptor sequence expressedhas 80% homology (differing only by two, non-consecu-tive amino acids, Fig. 1) with that of the rat (Saltzmanet al., 1991). However, synthetic human 5-HT2C decapep-tide was not available to establish the precise extent ofcross-reactivity withG241 byRIA. Furthermore, 5-HT2C

antiserum staining was absent in untransfected HEK293 cells, confirming the specificity of this antiserum.In homogenates from choroid plexus, frontal cortex,

hippocampus, and septum antiserum, G241 identified aprotein band of the expected molecular mass for the rat5-HT2C receptor, in agreement with the radioimmunoas-say data presented herein. In all tissues, G241 alsoreacted with a heavier 63 kDa protein, which is prob-ably a glycosylated form of the receptor, as recentlydemonstrated in Western blot studies with two other5-HT2C antisera (Abramowski and Staufenbiel, 1995;Backstrom et al., 1995). Both of these bands wereabolished by preabsorption with the synthetic 5-HT2Cpeptide antigen, further confirming the specific natureof the binding.When expressed in oocytes, 5-HT2A receptor mutants

lacking the N-terminal tail show unaltered bindingcompared with the native receptor (Buck et al., 1991),suggesting that this region of 5-HT2 receptors may notbe involved in ligand binding. Instead, a recent reportsuggests that a N-terminal signal sequence (aminoacids 20 to 34) may be cleaved off the 5-HT2C receptor oncell membrane insertion (Abramowski and Staufenbiel,1995). This putative signal sequence contains the re-gion recognised by antiserum G241 (Fig. 1). However,in Western blots, G241 failed to visualise any low (2kDa)molecularmass immunopositive band correspond-ing to the putative signal peptide. The 5-HT2C antibodyused by Abramowski (Abramowski and Staufenbiel,1995) recognised a 38 kDa protein in Western blotsafter deglycosylation of CNS membranes with PNGaseF, which was interpreted as being the receptor devoid ofthe signal peptide, but as only PMSF was present toinhibit proteases during this protocol protein degrada-tion could have generated a polypeptide not present invivo.In RIA, antiserum G241 showed no cross-reactivity

with synthetic peptides corresponding to N-terminalportions of the rat 5-HT2A or 5-HT2B receptors (or analternative 5-HT2C sequence) or a variety of in-doleamines or neuropeptides co-localised with 5-HT incaudal raphe neurones, further demonstrating its speci-ficity. 5-HT2C-LI measured by the current RIA had a

Fig. 5. Effect of intracerebroventricular pretreatment with eitherthe serotonergic neurotoxin 5,7-dihydroxytryptamine (2 3 150 µg5,7-DHT given 30 min after desipramine, 15 mg kg21 i.p, n 5 8) orvehicle (ascorbic acid in 0.154 M saline after desipramine, n510) on5-HT2C-LI (upper histograms: fmols mg21 protein) and 5-HT (lowerhistograms; pmols mg21 protein) levels (mean 6 s.e.m). Abbreviationsas in Figure 4. and (DT) dorsal thoracic and (CV) ventral cervicalspinal cord. Note the significant increase in 5-HT2C-LI in the choroidplexus and ventral cervical spinal cord and yet the reduction in dorsalthoracic spinal cord, brain stem, and midbrain despite significantreductions in 5-HT levels in all these regions. *P , 0.05, **P ,0.01,and ***P ,0.001, Student’s unpaired t-test from vehicle treatedcontrols.

535-HT2C RECEPTOR RADIOIMMUNOASSAY

widespread, heterogeneous CNS distribution which,overall, correlates well with the known pattern ofserotonergic nerve fibres (Steinbush, 1981), 5-HT2Cligand binding sites identified using the conventionalligand [3H]-mesulergine (Pazos and Palacios, 1985),5-HT2C receptor mRNA transcripts measured using insitu hybridisation (Mengod et al., 1990; Pompeiano etal., 1994), and recent immunohistochemical studiesusing 5-HT2C-directed antibodies (Abramowski et al.,1995). The correlation was, however, not complete. No[3H]-mesulergine binding has been identified in theseptum or cerebellum, yet both the present RIA (andWestern blots) and in situ hybridisation (Hoffman andMezey, 1989; Molineaux et al., 1989) detect 5-HT2Creceptor protein and transcript, respectively, in theseregions, probably reflecting the relatively poor sensitiv-ity of mesulergine binding.In the present RIA, highest 5-HT2C-LI was in the

choroid plexus; being five times that in any brainregion. Moderately high levels were found in the frontalcortex and many subcortical areas, including the stria-tum and parts of the limbic system (hippocampus,septum, and hypothalamus), intermediate levels in themidbrain and thalamus, and lowest levels in the rhomb-encephalon and spinal cord. Simple absorption of theantibody or radiolabelled peptide by the membranesused in the current RIA cannot account for the resultsobtained, since the highest levels were detected in theregion having the lowest membrane concentration.However, previous [3H]-mesulergine binding studies(Mengod et al., 1990; Pazos and Palacios, 1985) showsubcortical 5-HT2C receptor levels are approximately 10times lower than those reported in this RIA study, yetthe level of 5-HT2C receptor mRNA in several subcorti-cal areas is equivalent to the choroid plexus (Molineauxet al., 1989). As antiserum G241 did not identify anylowmolecularmass proteins inWestern blots, measure-ment of degradation products is unlikely to account forthe high 5-HT2C levels reported herein by RIA. Thediscrepancy is more likely due to a difference in affinityof G241 IgGs for the intact receptor protein and thesynthetic peptide used as standard to calculate RIAlevels, and/or measurement of internalised receptorsnot detected by ligand binding. Consistent with thelatter proposal, Hamon and co-workers (1991) foundsignificantly higher 5-HT1A levels in most brain areasusing autoradiography and [35S]-receptor antibodiesthan with the 5-HT1A radioligand [3H]-8-OHDPAT.Previous in situ hybridisation work shows dense

5-HT2CmRNA in the basal ganglia and striatum (Men-

god et al., 1990; Molineaux et al., 1989; Pompeiano etal., 1994) consistent with the high striatal 5-HT2C-LIreported herein by RIA. Indeed, activation of striatal5-HT

2Creceptors is thought to produce hypolocomotion

in the rat by modifying dopamine release (Kennett andCurzon, 1988). The serotonergic system also stronglyinfluences numerous hypothalamic functions; includ-

ing neuroendocrine secretions (King et al., 1989; Lee etal., 1991), appetite (Kennett and Curzon, 1991), andthermoregulation (Klodzinski and Chojnacka-Wojcik,1992). Both the current RIA and previous in situhybridisation confirm high levels of 5-HT2C receptorprotein and transcript, respectively, in the hypothala-mus, which may account for these effects.5-HT2C sites have been identified in both dorsal and

ventral spinal horns using [3H]-mesulergine autoradiog-raphy (Pazos and Palacios, 1985) and ligand binding incervical, thoracic, and lumbo-sacral homogenates (Pran-zatelli et al., 1992). Yet only low levels of 5-HT2C-LIwere detected in either the dorsal and ventral thoraco-lumbar spinal cord using the current RIA. However, the5-HT2C receptor transcript appears to be localised in theposterior (lamina V) and intermediate (lamina VII)regions of the dorsal horn (Molineaux et al., 1989) andthe precise lamina distribution was not examinedherein.5,7-DHT-induced destruction of serotonergic neu-

rones has been reported to both elevate (Rocha et al.,1993) and leave 5-HT2C responsiveness unchanged(Ashby et al., 1994) in the rat brain, which may reflectregional variation in receptor adaptation. Indeed, mostgroups report elevated 5-HT2C binding or responsive-ness in the choroid plexus, hippocampus and spinalcord after 5,7-DHT (Conn et al., 1987; Fone et al., 1989;Rocha et al., 1993) but no change in the frontal cortex(Ashby et al., 1994). In the current study, i.c.v. 5,7-DHTproduced a varied pattern of 5-HT depletion and5-HT2C-LI adaptation. Within the choroid plexus, thesignificant increase in 5-HT2C-LI was expected, as 5-HTreleased into the cerebrospinal fluid via the supraepen-dymal raphe terminals (Aghajanian andGallager, 1975;Matsuura et al., 1985; Giordano and Hartig, 1987) is inconcentrations (10–100 nM) capable of activating 5-HT

2Creceptors in the choroid plexus (Linnoila et al., 1986;Volicer et al., 1985). Moreover, concordant with ele-vated 5-HT2C-LI, an identical 5,7-DHT treatment regi-men caused a two-fold increase in 5-HT2C-mediatedphosphoinositide hydrolysis in the choroid plexus (Connet al., 1987). A marked increase in 5-HT2C-LI was alsoobserved in the ventral cervical spinal cord, following asignificant (-80%) depletion of 5-HT by 5,7-DHT, in thecurrent study. A similar up-regulation of spinal [3H]-mesulergine binding was produced by neonatal 5,7-DHT lesions (Pranzatelli, 1990). Furthermore, 5-hy-droxytryptophan-induced flat body posture (involvingspinal 5-HT2C receptors) and m-CPP-induced hypoloco-motion (also 5-HT2Cmediated) are both enhanced follow-ing i.c.v. 5,7-DHT (Colpaert et al., 1989; Lucki et al.,1989). In contrast, definitive demonstration of the pre-or post-synaptic localisation of 5-HT2C receptors withinbrainstem serotonergic nuclei is lacking, yet this regionexpresses 5-HT2C mRNA (Hoffman and Mezey, 1989;Mengod et al., 1990; Molineaux et al., 1989; Pompeianoet al., 1994). Although there is little evidence that

54 A. SHARMA ET AL.

5-HT2C receptors autoregulate 5-HT release, in thepresent 5,7-DHT study, 5-HT2C-LI and 5-HT were bothsignificantly reduced in the thoracic dorsal spinal cord,mid brain and brainstem, implying that 5-HT2C recep-tors may be located on 5-HT nerve terminals in theseregions. Moreover, the reduction in 5-HT release follow-ing incubation of rat lumbar spinal cord slices with5-HT is mimicked by the 5-HT2C agonist m-CPP (Mur-phy and Zemlan, 1989) but its considerable affinity forthe 5-HT1B receptor was thought to account for thispre-synaptic inhibition. The significant decrease inbrainstem 5-HT2C-LI in the current 5,7-DHT study,accompanied by a modest but significant depletion of5-HT, is also consistent with loss of presynaptic 5-HT2C

receptors, which must be on destroyed nerve terminalsrather than surviving cell bodies. Such presynaptic5-HT

2Creceptors could, however, regulate the release of

co-existent neuropeptides in raphe neurones ratherthan affecting 5-HT release per se.The sensitivity of the current receptor RIA should

make it possible to determine 5-HT2C receptor proteinchanges in small brain areas in individual animals andenable these to be correlated with behavioural alter-ations in future studies (Fone et al., 1996), which willimprove our ability to determine 5-HT2C receptor func-tion in the CNS.Where no highly selective conventionalantagonists exist, chronic intracerebroventricular injec-tion of a receptor-directed antiserum to immunoneutra-lise receptor protein (Fone and Sharma, 1993) may alsoprove to be a less toxic alternative to antisense oligo-nucleotide strategies (Le Corre et al., 1996) to deter-mine receptor function in vivo in future studies.

ACKNOWLEDGMENTS

We wish to thank Dr. M. Landon (Department ofBiochemistry, Queen’sMedical Centre, Nottingham) forassisting with the protein synthesis and purification,Ian Topham and Chris Jones for technical assistancewith the animal and RIA experiments, Paul Millns forhelp with tissue culture and immunocytochemistry,and Nottingham Medical School Trust Funds for finan-cial assistance with the antibody production. We arealso indebted to Dr. C. Mannix and Dr. T. P. Blackburn(SmithKline Beecham Pharmaceuticals, Harlow) forthe kind donation of the HEK 293 cell lines. A.S. wassupported by the BBSRC and T.P. (Western blots) isjointly supported by a Nottingham University Student-ship and SmithKline Beecham Pharmaceuticals.

REFERENCESAbramowski, D., Rigo, M., Duc, D., Hoyer, D., and Staufenbiel, M.(1995) Localization of the 5-hydroxytryptamine2C receptor protein inHuman and rat brain using specific antisera. Neuropharmacology,34:1635–1645.

Abramowski, D., and Staufenbiel, M. (1995) Identification of the5-hydroxytryptamine2C receptor as a 60-kDaN-glycosylated proteinin choroid plexus and hippocampus. J. Neurochem., 65:782–790.

Aghajanian, G.K., andGallager, D.W. (1975) Raphe origin of serotoner-gic nerves terminating in the cerebral ventricle. Brain Res., 88:21–231.

Ashby, C.R., Zhang, J.Y., Edwards, E., and Wang, R.Y. (1994) Theinduction of serotonin3-like receptor supersensitivity and dopaminereceptor subsensitivity in the rat medial prefrontal cortex after theintraventricular administration of the neurotoxin 5,7-dihydroxytryp-tamine: Amicroiontophoretic study. Neuroscience, 60:453–462.

Backstrom, J.R., Westphal, R.S., Canton, H., and Sanders-Bush, E.(1995) Identification of rat serotonin 5-HT2C receptors as glycopro-teins containingN-linked oligosaccharides.Mol. Brain. Res., 33:311–318.

Bassiri, R.M., and Utiger, R.D. (1972) The preparation and specificityof antibody to thyrotropin releasing hormone. Endocrinology, 90:722–727.

Baxter, G.S., Kennett, G., Blaney, F., and Blackburn, T.P. (1995) 5-HT2receptor subtypes: A family re-united? Trends Pharmacol. Sci.,16:105–110.

Boess, F.G., and Martin, I.L. (1994) Review: Molecular biology of 5-HTreceptors. Neuropharmacology, 33:275–317.

Bonhaus, D.W., Bach, C., DeSouza, A., Salazar, F.H.R., Matsuoka,B.D., Zuppan, P., Chan, H.W., and Eglen, R.M. (1995) The pharma-cology and distribution of human 5-hydroxytryptamine2B (5-HT2B)receptor gene products: Comparison with 5-HT2A and 5-HT2C recep-tors. Br. J. Pharmacol., 115:622–628.

Buck, F., Meyerhof, W., Werr, H., and Richter, D. (1991) Characteriza-tion of N- and C-terminal deletionmutants of the rat serotonin 5HT2receptor in Xenopuslaevis oocytes. Biochem. Biophys. Res. Com-mun., 178:1421–1428.

Carey, J.E., Wood, M.D., Blackburn, T.P., Browne, M.J., Gale, D.G.,Glen, A., Flanigan, T.P., Hastwell, C., Muir, A., Robinson, J.H., andWilson, S. (1996) Pharmacological characterisation of a recombi-nant human 5HT(2C) receptor expressed in HEK293 cells. Pharma-col. Commun., 7:165–172.

Colpaert, F.C., Koek, W., and Lategan, A. (1989) L-5-hydroxytrypta-mine-induced flat body posture in the rat: Antagonism by ritanserinand potentiation after 5,7-dihydroxytryptamine. Eur. J. Pharmacol.,169:175–178.

Conn, P.J., Janoowsky, A., and Sanders-Bush, E. (1987) Denervationsupersensitivity of 5-HT-1C receptors in rat choroid plexus. BrainRes., 400:396–398.

Conn, P.J., and Sanders-Bush, E. (1986) Agonist-induced phosphoino-sitide hydrolysis in choroid plexus. J. Neurochem., 47:1754–1760.

Duxon, M.S., Reavley, A.C., Flanigan, T.P., Baxter, G., Blackburn, T.P.,and Fone, K.C.F. (1995) Expression of the 5-HT2B receptor protein inthe rat brain. Br. J. Pharmacol., 115:105P.

Duxon, M.S., Flanigan, T.P., Reavley, A.C., Baxter, G., Blackburn, T.P.,and Fone, K.C.F. (1997) Evidence for expression of the 5-hydroxytryp-tamine-2B receptor protein in the rat central nervous system.Neuroscience, 76:323–329.

Foguet, M., Hoyer, D., Pardo, L.A., Parekh,A., Kluxen, F.W., Kalkman,H.O., Stuhmer, W., and Lubbert, H. (1992) Cloning and functionalcharacterization of the rat stomach fundus serotonin receptor.EMBO J., 11:3481–3487.

Fone, K.C.F., Johnson, J.V., Bennett, G.W., and Marsden, C.A. (1989)Involvement of 5-HT2 receptors in the behaviours produced byintrathecal administration of selected 5HT agonists and the TRHanalogue (CG 3509) to rats. Br. J. Pharmacol., 96:599–608.

Fone, K.C.F., and Sharma, A. (1993) Effect of chronic intracerebroven-tricular 5-HT1C receptor antiserum infusion on the 5-HT behav-ioural syndrome and 5-HT turnover. Br. J. Pharmacol., 108:97P.

Fone, K.C.F., Shalders, K., Fox, Z.D., Arthur, R., and Marsden, C.A.(1996) Increased 5-HT2C receptor responsiveness occurs on rearingrats in social isolation. Psychopharmacology, 123:346–352.

Fozard, J.R., and Gray, J.A. (1989) 5-HT1C receptor activation: A keystep in the initiation of migraine? Trends Pharmacol. Sci., 10:307–309.

Giordano, J., and Hartig, P. (1987) Activation of the serotonin receptoron the choroid plexus by CSF-borne serotonin. Soc. Neurosci.Abstracts, 13:1326.

Hamon, M., El Mestikawy, S., Riad, M., Gozlan, H., Daval, G., andVerge, D. (1991) Specific antibodies as new tools for studies ofcentral 5-HT1A receptors. In: Serotonin: Molecular Biology, Recep-tors and Functional Effects. J.R. Fozard and P.R. Saxena, eds.Birkhauser Verlag, Basel. pp. 95–106.

Hoffman, B.J., and Mezey, E. (1989) Distribution of serotonin 5-HT1Creceptor mRNA in adult rat brain. FEBS Lett., 247:453–462.

Hoyer, D. (1988) Functional correlates of serotonin 5-HT1 recognitionsites. J. Recep. Res., 8:59–81.

Hoyer, D., Clarke, D.E., Fozard, J.R., Hartig, P.R., Martin, G.R.,Mylecharrane, E.J., Saxena, P.R., and Humphrey, P.P.A. (1994)International Union of Pharmacology classification of receptors for5-hydroxytryptamine (serotonin). Pharmacol. Rev., 46:157–204.

555-HT2C RECEPTOR RADIOIMMUNOASSAY

Humphrey, P.P.A., Hartig, P., and Hoyer, D. (1993) A re-appraisal of5HT receptor classification. Trends Pharmacol. Sci., 14:233–236.

Julius, D., Haung, K.M., Livelli, T.J., Axel, R., and Jessell, T.M. (1990)The 5HT2 receptor defines a family of structurally distinct butfunctionally conserved serotonin receptors. Proc. Nat. Acad. Sci.U.S.A., 87:928–932.

Julius, D., MacDermott, A.B., Axel, R., and Jessell, T.M. (1988)Molecular characterisation of a functional cDNA encoding theserotonin 1C receptor. Science, 241:558–564.

Kennett, G.A. (1993) 5-HT1C receptors and their therapeutic rel-evance. Curr. Opin. Invest. Drugs, 2:317–362.

Kennett, G.A., and Curzon, G. (1988) Evidence that mCPP may havebehavioural effects mediated by central 5-HT1C receptors. Br. J.Pharmacol., 94:137–147.

Kennett, G.A., and Curzon, G. (1991) Potencies of antagonists indicatethat 5-HT1C receptors mediate 1-2(chlorophenyl)piperazine-inducedhypophagia. Br. J. Pharmacol., 103:2016–2020.

King, B.H., Brazell, C., Dourish, C.T., and Middlemiss, D.N. (1989)MK-212 increases rat plasma ACTH concentration by activation ofthe 5-HT1C receptor subtype. Neurosci. Lett., 105:174–176.

Klodzinski, A., and Chojnacka-Wojcik, E. (1992) Hyperthermia in-duced by m-trifluoromethylphenylpiperazine (TFMPP) or m-chlorophenylpiperazine (m-CPP) in heat-adapted rats. Psychophar-macology, 109:466–472.

Kursar, J.D., Nelson, D.L., Wainscott, D.B., Cohen, M.L., and Baez, M.(1992) Molecular cloning, functional expression, and pharmacologi-cal characterization of a novel serotonin receptor (5-Hydroxytryptamine2F) from rat stomach fundus. Mol. Pharmacol.,42:549–557.

Le Corre, S.M., Burnet, P.W.J., Mellor, R., Sharp, T., and Harrison, P.J.(1997) Critical issues in the antisense inhibition of brain geneexpression in vivo: Experiences targetting the 5-HT1A receptor.Neurochem. Int., in press.

Lee, M.A., Nash, J.F., Barnes, M., and Meltzer, H.V. (1991) Inhibitoryeffect of ritanserin on the 5-hydroxytryptophan-mediated cortisol,ACTH and prolactin secretion in humans. Psychopharmacology,103:258–264.

Linnoila, M., Jacobson, K.A., Marshall, T.H., Miller, T.L., and Kirk,K.L. (1986) Liquid chromatographic assay for cerebrospinal fluidserotonin. Life Sci., 38:687–694.

Loric, S., Launeay, J.-M., and Colas, J.-F. (1992) New mouse 5-HT2-like receptor: expression in brain, heart and intestine. FEBS Lett.,312:203–207.

Lowry, O.H., Rosenbrough, N.J., Farr, A.L., and Randall, R.J. (1951)Protein measurement with the Folin phenol reagent. J. Biol. Chem.,193:265–275.

Lucki, I.,Ward, H.R., and Frazer,A. (1989) Effect of 1-(m-chlorophenyl)-piperazine and 1-(m-trifluoromethylphenyl)piperazine on locomotoractivity. J. Pharmacol. Exp. Thera., 249:155–164.

Matsuura, T., Takenchi, Y., Kojima, M., Ueda, S., Yamada, H., Nojyo-Yushima, K., and Sano, Y. (1985) Immunohistochemical studies ofthe serotonergic supraependymal plexus in themammalian ventricu-lar system, with special reference to the characteristic reticularramification. ActaAnat., 123:207–219.

Mengod, G., Nguyen, H., Le, H., Waeber, C., Lubbert, H., and Palacios,J.M. (1990) The distribution and cellular localisation of theserotonin1C receptor mRNA in the rodent brain examined by in situhybridisation histochemistry: comparison with receptor bindingdistribution. Neuroscience, 35:577–591.

Molineaux, S.M., Jessell, T.M., Axel, R., and Julius, D. (1989) 5-HT1C

receptor is a prominent serotonin receptor subtype in the centralnervous system. Proc. Nat. Acad. Sci. U.S.A., 86:6793–6797.

Murphy, R.M., and Zemlan, F.P. (1989) Functional change in the 5-HTpre-synaptic receptor in spinal cord of aged rats. Neurobiol. Aging,10:95–97.

Paxinos, G., and Watson, C. (1986) The Rat Brain in StereotaxicCo-ordinates. Second Edn. Academic Press, London.

Pazos,A., Cortes, R., and Palacios, J.M. (1985) Quantitative autoradio-graphic mapping of serotonin receptors in the rat brain II: Seroto-nin-2 receptors. Brain Res., 346:231–249.

Pazos, A., Hoyer, D., and Palacios, J.M. (1984) The binding ofserotonergic ligands to the porcine choroid plexus: Characterisationof a new type of serotonin recognition site. Eur. J. Pharmacol.,106:539–546.

Pazos, A., and Palacios, J.M. (1985) Quantitative autoradiographicmapping of serotonin receptors in the rat brain I: Serotonin1receptors. Brain Res., 346:205–230.

Pompeiano, M., Palacios, J.M., and Mengod, G. (1994) Distribution ofthe serotonin 5-HT2 receptor family mRNAs: Comparison between5-HT2A and 5-HT2C receptors. Mol. Brain Res., 23:163–178.

Pranzatelli, M.R. (1990) Neonatal 5,7-DHT lesions upregulate [3H]-mesulergine-labelled spinal 5-HT1C binding sites in the rat. BrainRes. Bull., 25:151–153.

Pranzatelli, M.R., Murthy, J.N., and Pluchino, R.S. (1992) Identifica-tion of spinal 5-HT1C binding sites in the rat: Characterization of[H-3]mesulergine binding. J. Pharmacol. Exp. Therapy, 261:161–165.

Rocha, B., Di Scala, G., Rigo, M., Hoyer, D., and Sandner, G. (1993)Effect of 5,7-dihydroxytryptamine lesion on mianserin-induced con-ditioned place aversion and on 5-hydroxytryptamine1C receptors inthe rat brain. Neuroscience, 56:687–693.

Saltzman,A.G., Morse, B., Whitman., M.M., Ivanschenko, Y., Jaye, M.,and Felder, S. (1991) Cloning of the human serotonin 5-HT2 and5-HT1C receptor subtypes. Biochem. Biophys. Res. Commun., 181:1469–1478.

Sharma,A., Landon, M., and Fone, K.C.F. (1992) Regional distributionof 5-HT1C and 5-HT2 receptor peptides in the rat brain measuredusing polyclonal antisera. Br. J. Pharmacol., 105:219P.

Sharma, A., Topham, I.A., and Fone, K.C.F. (1994) Effect of the5,7-dihydroxytryptamine on 5-HT2C receptor protein-like immunore-activity in the rat brain and spinal cord. Br. J. Pharmacol., 111:150P.

Steinbush, H.W.M. (1981) Distribution of serotonin-immunoreactivityin the central nervous system of the rat-cell bodies and terminals.Neuroscience, 6:557–618.

Tecott, L.H., Sun, L.M., Akana, S.F., Strack, A.M., Lowenstein, D.H.,Dallman, M.F., and Julius, D. (1995) Eating disorder and epilepsy inmice lacking 5-HT2C serotonin receptors. Nature, 374:542–546.

Van der Karr, L.D., and Brownfield, M.S. (1993) Serotonergic neuronsand neuroendocrine functions. News Physiol. Sci., 8:202–207.

Volicer, L., Direnfeld, L.K., Freedman, M., Albert, M.L., Langlias, P.J.,and Bird, E.D. (1985) Serotonin and 5-hydroxyindoleacetic acid:CSF differences in Parkinson’s disease and dementia of the Alzhei-mer’s Type. Arch. Neurol., 42:127–129.

Wainscott, D.B., Cohen, M.L., Schenck, K.W.,Audia, J.E., Nissen, J.S.,Baez, M., Kursar, J.D., Lucaites, V.L., and Nelson, D.L. (1993)Pharmacological characteristics of the newly cloned rat5-hydroxytryptamine2F receptor. Mol. Pharmacol., 43:419–426.

Yagaloff, K.A., and Hartig, P.R. (1985) 125I-LSD binds to a novelserotonergic site on rat choroid plexus epithelial cells. J. Neurosci.,5:3178–3183.

56 A. SHARMA ET AL.