Localization of 5-HT7 receptors in rat brain by immunocytochemistry, in situ hybridization, and agonist stimulated cFos expression

Journal of Chemical Neuroanatomy 21 (2001) 63 – 73 www.elsevier.com/locate/jchemneu

Localization of 5-HT7 receptors in rat brain by immunocytochemistry, in situ hybridization, and agonist stimulated cFos expression J.F. Neumaier a, T.J. Sexton a, J. Yracheta b, A.M. Diaz b, M. Brownfield b,* a

Department of Psychiatry and Beha6ioral Sciences and Harbor6iew Medical Center Uni6ersity of Washington, Seattle WA 98195, USA b Department of Comparati6e Biosciences and Neurosciences Training Program Uni6ersity of Wisconsin Madison WI 53706, USA Received 15 June 2000; received in revised form 31 August 2000; accepted 11 September 2000

Abstract 5-HT7 receptors are recently identified members of the serotonin receptor family that have moderate to high affinity for several important psychotropic drugs. However, the lack of selective ligands has impeded the study of the brain distribution of these receptors. In this report, we describe the localization of 5-HT7 receptor in rat forebrain by immunocytochemistry, in situ hybridization of 5-HT7 mRNA, and functional stimulation of cFOS expression by 5-HT7 receptor activation. The anatomical localization of 5-HT7 mRNA in situ hybridization signal. Prominent immunostaining was apparent in numerous sites within the cerebral cortex, hippocampal formation, tenia tecta, thalamus and hypothalamus. 5-HT7 receptors were detected in suprachiasmatic nucleus by both immunocytochemistry and in situ hybridization. At a microscopic level, both cell bodies and proximal fibers were strongly stained in these regions, suggesting a somatodendritic subcellular distribution. 5-HT7 receptor-like immunoreactivity was further compared with 5-HT7 mediated biological function by administering 8-OH-DPAT intracerebroventricular injection (icv)with WAY 100135 (to block 5-HT1A receptors) followed by double immunostaining localization of cFos activation and 5-HT7 receptors. In all regions examined, cFos stimulation and 5-HT7-like immunoreactivity colocalized to the same neurons. Furthermore, cFos activation by 8-OH-DPAT was blocked by pimozide — a 5-HT7 antagonist. Therefore, by using multiple strategies, we were able to localize 5-HT7 receptors in rat brain unequivocally. The distribution of these receptors is consistent with their involvement in the control of circadian activity and the action of anti-depressants and atypical neuroleptics. © 2001 Elsevier Science B.V. All rights reserved. Keywords: 5-HT7 receptor; cFos; In situ hybridization; Immunocytochemistry; Serotonin receptor

1. Introduction The field of serotonin receptors has expanded from a few identified receptors to 14 cloned receptors during the last decade (Barnes and Sharp, 1999). These receptors are now being found to have diverse cellular distributions, modes of signal transduction, and biological functions. The recently cloned rat 5-HT7 receptor (Bard et al., 1993; Lovenberg et al., 1993; Ruat et al., 1993; Shen et al., 1993; Tsou et al., 1994), has several alternative mRNA splice variants (Heidmann et al., 1997; Stam et al., 1997), all of which can stimulate * Corresponding author. Tel.: +1-608-2635863; fax: + 1-6082633926. E-mail address: [email protected] (M. Brownfield).

cAMP synthesis by adenylate cyclase, when activated by serotonin. Understanding the biological function of 5-HT7 receptors has been hampered, however, by the lack of highly selective agonists and antagonists to study its in vivo distribution and pharmacology of the receptor (Vanhoenacker et al., 2000). The 5-HT7 receptor is unique in that it shares some pharmacological overlap with the 5-HT1A receptor (high affinity for 8-hydroxy-2-(di-n-propylamino) tetralin (8-OH-DPAT) and 5-carboxamidotryptamine) and with the 5-HT2A receptor (high affinity for clozapine, risperidone, mesulergine, and pimozide). Indeed, the earlier described effects of some of these drugs may actually be mediated by 5-HT7 instead of 5-HT1A or 5-HT2A receptors. A case in point is the recent evidence that 5-HT7 and not 5-HT1A receptors account for 8-OH-DPAT induced

0891-0618/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved. PII: S 0 8 9 1 - 0 6 1 8 ( 0 0 ) 0 0 0 9 2 - 2

64

J.F. Neumaier et al. / Journal of Chemical Neuroanatomy 21 (2001) 63–73

phase shifts in the suprachiasmatic nucleus (SCN) (Lovenberg et al., 1993; Bobrzynska et al., 1996; Ying and Rusak, 1997). However, the presence of 5-HT7 receptors in suprachiasmatic nucleus has been disputed (Moyer and Kennaway, 1999) and there is disagreement as to the distribution of 5-HT7 mRNA within hippocampus and thalamus (Gustafson et al., 1996; Le Corre et al., 1997; Venero et al., 1997; Vizuete et al., 1997; Heidmann et al., 1998). Several attempts have been made to label 5-HT7 binding sites in rat brain using available radioligands to rat brain homogenates (Stowe and Barnes, 1998; Hemedah et al., 1999) or tissue sections (Waeber and Moskowitz, 1995; Gustafson et al., 1996). While these binding strategies were rational, they leave open the possibility of misidentifying at least some proportion of binding sites based on cross-selectivity to other known or yet to be described binding sites. In particular, the autoradiographic studies used a relatively non-selective radioligand [3H]-5-carboxamidotryptamine in the presence of drugs to mask other binding sites. This strategy may underestimate 5-HT7 binding sites due to some occupancy by the masking agents and/or non-recognition of low agonist affinity states. Detection of mRNA by in situ hybridization offers far greater certainty due to the exquisite specificity of nucleotide sequences, but dose not reflect receptor protein levels or post-translational regulation. Information about the subcellular distribution of 5-HT7 receptors is also lacking. For these reasons, we have developed a selective antiserum directed against 5-HT7 peptide fragments to allow immunocytochemical detection of the receptor. The purpose of this study was to characterize this antibody, to compare the distribution of 5-HT7 immunoreactive neurons with 5-HT7 mRNA in the adult rat brain, and to compare 5-HT7 receptor activation of cFos with immunocytochemical localization of 5-HT7 receptors in rat brain using double immunostaining.

A 379 nucleotide fragment corresponding to 862 – 1241 of the rat 5-HT7 splic B variant (Shen et al., 1993) was cloned from the full length cDNA clone into pBluescript and used as template for RNA probe (riboprobe) synthesis. This probe hybridizes to all splice variants of 5-HT7 mRNA and sense strand probe produces no specific hybridization (Heidmann et al., 1998). To synthesize antisense riboprobe, the plasmid was linearized with BamHI, and riboprobe was synthesized using T3 RNA polymerase. The nucleotides included 50% [33P]-UTP (Andotek, Irvine, CA), yielding a specific activity of 134 mCi/pmol probe. The labeled riboprobe was diluted (1 pmol/ml) in a hybridization buffer containing 50% formamide, 10% dextran sulfate, 0.3 M sodium chloride, 10 mM Tris (pH 8.0), 1 mM EDTA, 1×Denhart’s (0.2% each bovine serum albumin, Ficoll, and polyvinylpyrollidine), 0.4 mg/ml yeast tRNA, and 10 mM dithiothreitol. Then, 50 ml of the hybridization mixture was applied to each slide and the sections were covered with silanized coverslips. The slides were incubated in moist, covered trays at 58°C overnight. Following the hybridization reaction, coverslips were removed and the slides were washed in 1X SSC (150 mM NaCl and 15 mM sodium citrate) for 30 min at room temperature and were treated with RNase A (20 mg/ml in 0.01 M Tris, 0.5 M NaCl, 1 mM EDTA, pH 8.0) at 37°C for 30 min, and again rinsed in 1X SSC for 30 min at room temperature. Slides were then washed for 1 h at 55°C in 1X SSC, in 0.1X SSC at room temperature for 1 h, and were then rinsed in distilled water, dehydrated through graded alcohols containing 0.3 M NH4OAc, and air dried. Hybridization signal was visualized by apposing the slides to autoradiographic film (Hyperfilm bmax, Amersham) for 20 days. In preliminary experiments, 1 pmol/ml of riboprobe was found to produce saturating hybridization signal; sense riboprobes produced only background hybridization (data not shown).

2.2. Antibody production 2. Materials and methods

2.1. In situ hybridization histochemistry (ISHH) Male Sprague-Dawley rats (180 – 220 g) were housed and used in accordance with protocols approved at our institutions. Animals were narcotized with CO2 and decapitated. The brains were rapidly removed, frozen on a dry ice slab, and stored at − 70°C. Frontal tissue sections (20 mm) were prepared using a Cryostat and thaw mounted onto silanized glass slides, and processed for ISHH. Briefly, the sections were fixed in cold 4% formaldehyde in phosphate buffered saline (PBS) (pH 7.4), rinsed in PBS, treated with acetic anhydride (0.25% in 0.1 M triethanolamine), dehydrated, delipidated, and air dried.

Polyclonal anti-peptide antibodies were generated in rabbits against rat 5-HT7 residues 8–23 (amino terminal extracellular domain). The production of the antiserum used in this study was described in a preliminary form (Brownfield et al., 1998). The targeted peptide sequence was selected by computer analysis of the full length 5HT7 receptor listed in the Swiss Protein Database (http://www.expasy.ch/cgibin/niceprot.pl?P32305) using GeneRunner (Hastings Software, Hastings on Hudson, NY) and Peptide Structure-Genetics Computer Group (GCG, Madison, WI) software. The peptide was selected based on high antigenicity, surface probablity, and surface topology indices, and is in a region with shared homology among all three 5-HT7 receptor sequences (Heidmann et al., 1998). Cysteine

J.F. Neumaier et al. / Journal of Chemical Neuroanatomy 21 (2001) 63–73

65

was added to the C-terminus as a cross-linking site for the preparation of peptide-protein immunogens and affinity media. The peptide was synthesized commercially (Quality Controlled Biochemicals Inc, Milford MA). Rabbits were injected with rat 5-HT7 [8 –23-C] conjugated to keyhole limpet hemocyanin (initially with 500 mg and subsequently with 50 mg boosts) once per month for 18 months. Rabbits were bled 10 days after each immunogen injection, plasma was collected, affinity purified, and evaluated by ELISA, western blot, and immunocytochemistry. Purified antibody was prepared by affinity column chromatography against immobilized rat 5HT7[8 – 23-C] media prepared using a SulfolinkT kit (Pierce Chemical Company, Rockford IL). Antibodies were eluted from the column with 0.2 M glycine, pH 2.5 and collected into 1.0 M Tris, pH 8.0. Antisera was extensively dialyzed against PBS and made up in 1% bovine serum albumin (BSA) and 0.01% sodium azide.

fixed in 4% paraformaldehyde in PBS, pH 7.4 for 30 min. Slides were stored at − 20°C until needed. Slides were rinsed briefly in PBS pH 7.4, rinsed briefly again in PBS, permeabilized in PBS 0.03% Triton-X 100 for 30 min and then blocked with 5% BSA +1% normal goat serum (NGS) in PBS for 2 h. Sections were incubated overnight at 4°C in the 5-HT7 primary antibody diluted 1:250 in 2.5% BSA, 0.5% goat serum in PBS. They were then rinsed 3×5 min with PBS, incubated with goat anti-rabbit Cy3 conjugated secondary antibody (Jackson ImmunoResearch Laboratories, West Grove PA) at dilution of 1:200 in 2.5% BSA + 0.5% NGS in PBS-0.03% Triton-X 100 for 1 h at 37°C. Slides were then rinsed 3× 5 min with PBS and rinsed briefly in deionized water. The excess water was removed and coverslips mounted with Gel-Mount (Biomeda Corp, Foster City, CA).

2.3. Western blotting

Adult male Sprague-Dawley rats (275 –325 g, Harlan Sprague-Dawley, Madison WI) were deeply anesthetized with pentobarbital (60 mg/kg ip) and perfused intracardially with cold calcium free heparinized Tyrode’s solution with 0.2% procaine followed by 4% p-formaldehyde with 0.1% glutaraldehyde in 0.1 M sodium phosphate buffer, pH 7.5. The forebrain blocks were extensively washed in cold PBS and 30 m sections were taken using a Vibratome (Model 1000; Ted Pella Inc, Redding CA). Sections were processed free floating with mild agitation on an orbital mixer. Aldehydes and endogenous peroxidase were quenched by sequential incubation in 1% sodium borohydride and 1% H2O2 in PBS for 30 min, respectively. Nonspecific staining was blocked by incubation in 5% NGS in ICC buffer (PBS containing 0.3% gelatin, 0.01% thimerosaol, and 0.02% neomycin, pH 7.5). Sections were incubated sequentially with 5-HT7 antibody (1:200 in ICC buffer-NGS) at 4°C overnight, biotinylated goat anti-rabbit IgG (1:400 in ICC-NGS with 0.3% Triton X-100) at room temperature for 45 min, and avidin-biotin-peroxidase complex (ABC Elite; Vector Laboratories, Burlingame, CA; 1:400 in ICC buffer) at room temperature for 45 min. Sections were washed in ICC buffer with 0.3% Triton X-100 (4×5 min) following each immune reagent incubation and with 0.01 M PBS-0.3% Triton X-100, following the ABC incubation. Bound antibody was visualized by incubation with 3,3%-diaminobenzidine –4HCl (0.05%) and H2O2 (0.01%) intensified with nickel ammonium sulfate (0.25%) in 0.01 M PBS for 7–15 min, controlled by visual inspection. Color development was stopped with 0.01 M PBS, and sections were mounted on slides subbed with gelatin-chromium potassium sulfate, dried, and coverslipped. Macro- and microphotographs were obtained digitally using the MCID Image Analysis system (Imaging Research Inc,

HeLa cells expressing 5-HT7 receptor were grown in DMEM with 400 mg/ml geneticin, rinsed in cold PBS and harvested in cold Versene (Sigma, St. Louis, MO). Cells were collected by centrifugation at 500 g for 10 min, and the pellet was resuspended in 100 mM Tris, 1 mM EDTA, 0.1 mM Triton-X 100, 100 mg/ml leupeptin, 10 mg/ml pepstatin A, and 2mM PMSF. Samples were homogenized with a Wheaton douncer, disrupted with a probe sonicator (5 s), and stored at − 20°C. Protein concentrations were determined by BCA assay (Pierce). Protein samples (50 mg per lane) were electrophoresed on a 10% SDS-PAGE gel. The protein was transferred to nitrocellulose membranes by electroblotting overnight (15 mV at 4°C). Membranes were blocked using 5% instant milk in TBS (50 mM Tris in 0.9% saline, pH 7.4) for 2 h and then incubated with the 5-HT7 primary antibody at a dilution of 1:250 in 5% instant milk in TBS overnight at 4°C. The blots were rinsed 3× 10 min in TBS and incubated with horseradish peroxidase conjugated goat anti-rabbit secondary antibody (Amersham ECL kit, Piscataway, NJ) at a dilution of 1:1000 in TBS in 5% instant milk for 2 h at room temperature, rinsed 3 times in TBS for 10 min each, and developed according to the ECL kit protocols. Blots were exposed to Kodak X-Omat film and exposed for 1– 30 min.

2.4. Immunocytochemistry (ICC) of 5 -HT7 expressing cells in 6itro 5-HT7 expressing HeLa cells with a Bmax of approximately 4000 fmol/mg protein (Heidmann et al., 1997) were grown in DMEM with 400 mg/ml geneticin on teflon coated ten-well slides to 95% confluency and then

2.5. Receptor immunocytochemistry in rat brain

66

J.F. Neumaier et al. / Journal of Chemical Neuroanatomy 21 (2001) 63–73

St Catharines, Ont., Canada) control of digital cameras (Sierra Scientific Model MS-4030 Digital Camera and Dage-MIT CCD Model 72S), interfaced with a light box (Northern light) and brightfield-differential interference microscope (Leitz Ortholux II). Photographs were processed, assembled, and printed digitally (Photoshop and Pagemaker software, Adobe Systems Inc, San Jose, CA).

2.6. Correlation of agonist-stimulated cFos and 5 -HT7 ICC Rats under pentobarbital anesthesia were fitted stereotaxically with a single lateral cerebroventricular (coordinates; AP= 1.0 mm, Lat.9 1.5, with respect to Bregma) stainless steel guide cannula (23 gauge). Animals were rested 24 h and on the next 2 succeeding days, were handled to simulate intracerebroinjection (ic6). On the day of the experiment, an injection (28 gauge) was inserted into the guide cannula so that the tip of the injection cannula extended 1 mm beyond the guide cannula tip. Drugs were injected in a volume of 10 ml of normal saline vehicle. Rats were first injected with the selective 5-HT1A antagonist WAY 100135 (25 mg, ic6), the injection cannula was left in place 5 min and then replaced with a second cannula and injected with the 5-HT1A/5-HT7 agonist 8-OH-DPAT (1 mg, ic6). The 8-OH-DPAT dose was chosen to be midrange in an earlier dose response study (Pan and Yang, 1996), while the higher dose of WAY 100135 was chosen to block 5-HT1A receptors in the forebrain maximally. Pharmacological control rats received pimozide (2 mg, icv) with the WAY 100135. Then, 60 min later, rats were anesthetized with pentobarbital and their brains prepared for immunocytochemistry, essentially as described above. Tissue sections were first immunocytochemically stained for cFos (1:30 000 rabbit anti cFos Ab5; Oncogene Sciences, Uniondale, NY) with nickel intensification and then followed with 5-HT7 receptor ICC without nickel intensification. This procedure results in dark blue stained immunoreactive cFos nuclei and brown stained immunopositive 5-HT7 receptor protein containing neuronal perikarya.

Fig. 1. Immunofluorescence demonstration of 5-HT7 receptor protein in HeLa cells (A), and lack of staining in wild type HeLa cells (B). Omission of primary (C) or secondary antibody (D) blocked staining of 5-HT7 expressing HeLa cells.

use required affinity purification against a 5-HT7 peptide-resin column. HeLa cells stably expressing 5-HT7 receptor were used to assess the anti-5-HT7 antiserum. Preimmune antiserum or antibody preabsorbed with peptide antigen produced no labeling (data not shown). Very strong immunoreactivity was detected in 5-HT7 expressing cells (Fig. 1A) while there was minimal signal in wild type cells (Fig. 1B). Exclusion of primary or secondary antiserum produced no labeling (Fig. 1C and D). Western blots of membranes prepared from 5-HT7 expressing HeLa cells showed a single immunoreactive with an apparent size of approximately 70 kDa (Fig. 2), consistent with the predicted molecular weight of nearly 49 kDa without glycosylation; no bands were recognized in wildtype HeLa cells.

3. Results

3.1. Characterization of 5 -HT7 antibody Antibody was successfully generated in all three rabbits and could be detected by ELISA as early as the second month. However, antibodies usable for immunocytochemistry were not developed until 9 – 12 months into the innoculation series. Antibodies used for these studies were collected approximately 1 year after project initiation. Their

Fig. 2. Western blot of immunoreactive 5-HT7 receptor protein extracted from HeLa cells. Protein samples (50 mg) from wild-type (lane 1) and 5-HT7 expressing (lane 2) HeLa cells was electrophoresed, blotted, and immunostained (as described in Section 2). Note that there is a strong band of approximately 70 kDa recognized only in the 5-HT7 transfected cells.

J.F. Neumaier et al. / Journal of Chemical Neuroanatomy 21 (2001) 63–73

3.2. In situ hybridization histochemistry 5-HT7 mRNA was detected in rat brain by ISHH in a distribution that agreed substantially with earlier reports (Ruat et al., 1993; Gustafson et al., 1996; Mengod et al., 1996). The distribution is summarized in Table 1. The most remarkable observation was the heavy hybridization signal for the 5-HT7 receptor in the thalamus, the hippocampal formation, tenia tecta, and parts of the hypothalamus. In particular, there was strong hybridization signal in the islands of Cajal, the indusium griseum (Fig. 3A), the thalamic nuclear complex including paraventricular, paratenial, anterior medial, anterior ventral thalamic nuclei (Fig. 3C, E and G). Moderate signal was associated with the ventral posteromedial and posterolateral thalamic nuclei (Fig. 3G) and the lateral septal nucleus (Fig. 3A). The hippocampal formation was moderately to heavily labeled (Fig. 3G). While all CA fields were labeled, with CA2/3 had the densest hybridization signal, and the dentate gyrus and CA1 being moderately and slightly less densely labeled, respectively. The cerebral cortex was differentially labeled, with heaviest labeling in the retrosplenial (Fig. 3G) and piriform cortex (Fig. 3C and E) and moderate to low labeling throughout the rest of the cerebral cortex. Hypothalamic labeling signal was extensive. The suprachiasmatic (Fig. 3C) and ventromedial nuclei (Fig. 3E) had the highest signals while the anterior hypothalamic area was moderately heavily labeled medially with moderate signal extending laterally through the lateral hypothalamus and into the medial and basal amygdala (Fig. 3C and E).

3.3. 5 -HT7 receptor immunocytochemistry in rat brain ICC mapping of 5-HT7 receptor protein revealed a nearly identical distribution when compared with the receptor mRNA in situ hybridization signal (Fig. 3 and Table 1). For this analysis, similar macroscopic sections were taken at different rostro-caudal locations of the rat forebrain. Immunoreactive 5-HT7 receptor protein was widely distributed throughout the rat forebrain. The microscopic analysis of ICC results confirmed this interpretation. Dense labeling was observed in the tenia tecta (Fig. 3B), superficial cortical lamina (Fig. 3B, D, F and G), the pyramidal cell layer of all hippocampal CA fields. (Fig. 3B and H), the paraventricular thalamic nucleus and the margins of the anteroventral and anteromedial thalamic nuclei (Fig. 3F), and the suprachiasmatic nucleus (Fig. 3F). Core regions of the anteroventral and anteromedial thalamic nuclei (Fig. 3H), the medial and anterior hypothalamic areas (Fig. 3D, F and G), and caudate-putamen (Fig. 3B), were moderately labeled. The globus pallidus, and major fiber bundles, including the corpus callosum and fornix were, respectively, lightly labeled or were immunonega-

67

Table 1 Comparative in situ hybridization and immunocytochemistry of 5HT7 receptors in rat forebrain Brain region

In situ hybridizationa

Immunocytochemistrya

Forebrain Neocortex superficial Middle Deep

+++ + +++

++++ ++ +++

Olfactory complex Olfactory tubercle Tenia tecta Piriform cortex

++ ++++ ++++

+++ ++++ ++++

Septal region Lateral septal nucleus Caudate putamen Globus pallidus Nucleus accumbens Ventral pallidum

++ + + ++ ++

++++ ++ + ++ ++

+++ +++

++++ +++

++ ++++ ++++ +++ ++

+++ ++++ +++ + +++

++ ++ + + +

++ +++ ++ ++ +

++ + ++++ +++ +++

++ + ++++ +++ ++

+ +++ ++

+ ++ ++

++++ +++ ++ +++

++++ +++ +++ +++

++

++

+

++

Hippocampal formation Induseum gresium Septohippocampal nucleus Hippocampus CA1 CA2 CA3 Dentate gyrus and hilus Entorhinal cortex Amygdala Cortical nucleus Central nucleus Lateral nucleus Bastolateral nucleus Basomedial nucleus Thalamus Anterior nuclei Lateral nuclei Medial nuclei Posterior nuclei Lateral geniculate nucleus Zona incerta Habenula, medial Habenula, lateral Hypothalamus Preoptic area-medial Preoptic area-lateral Periventricular Anterior hypothalamic area Lateral hypothalamic area Paraventricular nucleus-

Par6icellular Paraventricular nucleus- ++

++

Magnocellular Supraoptic nucleus Suprachiasmatic Retrochiasmatic area Arcuate Dorsomedial Ventromedial

++ ++++ +++ +++ + ++

++ ++++ ++ ++ + ++

a Levels of detectable labeling — +week; ++ low; +++ moderate; ++++ strong.

68

J.F. Neumaier et al. / Journal of Chemical Neuroanatomy 21 (2001) 63–73

tive. In some regions, the in situ hybridization signal was mottled at low magnification, while immunoreactivity was more densely homogenous in these regions. This was interpreted as indicative of only cell body labeling (ISHH) and cell body with fiber labeling (ICC). The 5-HT7 antiserum was not effective for western blots in rat brain, yielding inconsistent and low levels of protein staining although it is not unusual for polyclonal, antipeptide antisera to work preferentially in ICC applications.

Microscopic analysis of ICC revealed intricate anatomical details of 5-HT7 expressing neurons. Perikarya and their processes were both labeled. In the cerebral cortex (Fig. 4A), numerous densely labeled pyramidal cell bodies in lamina 5 were observed projecting a diffuse fiber network into laminae 1 through 3. Other, smaller oval immunoreactive cells were seen in lamina 2 and 3. In the hippocampus proper, pyramidal cell bodies and fibers were labeled in all CA fields (Fig. 4B). Staining in CA1 was heaviest in the pyramidal

Fig. 3. Comparative in situ hybridization and immunocytochemical localization of 5-HT7 mRNA (left column) and 5-HT7 protein (right column), respectively (10X). There is a close association between the distribution of mRNA and 5-HT7-like immunoreactivity, although in some regions the receptor protein appears stronger in dendritic regions (e.g. hippocampus). Abbreviations; amyg, amygdala; av, anteroventral thalamic nucleus; atn, anterior thalamic nuclei; hip, hippocampus; lsn, lateral septal nuclei; neo, neocortex; mpa, medial preoptic area; pir, piriform cortex; pvn, paraventricular thalamic nuclei; pvp, posterior paraventricular nucleus; rc, retrosplenial cortex; scn, suprachiasmatic nuclei; str, striatum; tt, tenia tecta.

J.F. Neumaier et al. / Journal of Chemical Neuroanatomy 21 (2001) 63–73

69

of the anteromedial and anteroventral thalamic nuclei, with moderate labeling in the core of these nuclei were labeled perikarya were also evident (Fig. 4E). Other thalamic nuclei, including the central medial, mediodorsal, and paratenial thalamic nuclei were moderately labeled. In these nuclei both immunopositive fibers and cells were present, the latter in a patchy distribution. Lateral thalamic nuclei were weakly immunoreactive. Numerous areas of the hypothalamus were immunopositive. The anterior and lateral hypothalamic area was typical of many areas of the hypothalamus in which there are modest to numerous labeled multipolar neurons interspersed among a moderately dense, but weak to moderately intense immunopositive fibrous network. Neurons of the paraventricular and supraoptic magnocellular system were moderately immunopositive. Cells of the parvocellular division of the paraventricular nucleus were also labeled but less intensely. The ventromedial-arcuate region was somewhat more densely stained than the rest of the hypothalamus, but structurally appeared similar. The suprachiasmatic nucleus was prominently stained, containing numerous immunoreactive perikarya against a background of a fine fiber network (Fig. 4F).

3.4. Correlation of 5 -HT7 agonist-stimulated cFos and 5 -HT7 immunoreacti6ity

Fig. 4. Microscopic immunocytochemical localization of 5-HT7 receptor protein in rat brain vibratome sections. Cerebral cortex (4A, 500X, layers 1 – 5 are indicated). Hippocampus, macroscopic view (4B, 50X) shows strong immunoreactivity in both cell body and dendritic layers in CA1– CA3 and dentate gyrus (DG); hippocampus, microscopic view of CA1 (4C, 250X) shows immunoreactivity of pyramidal neurons (py) and proximal dendrites in stratum radiatum (sr) and to a lesser extent in stratum oriens (so). Tenia tecta (4D, 100X), paraventricular thalamic nucleus (4E, 125X, in the medial thalamic nuclear group), and suprachiasmatic nucleus (4F, 150X) also show examples 5-HT7-like immunoreactivity.

layer (Fig. 4C), while in CA3 the oriens layer was most intense (Fig. 4B). Fibers in stratum lacunosum-moleculare of both CA1 and CA3 showed moderately dense immunostaining. Rare cell bodies were observed in the strata radiatum and oriens. In the tenia tecta, densely immunolabeled cell bodies were present in the dorsal peduncular cortex (Fig. 4D). Subcortical regions also demonstrated substantial 5HT7-like immunoreactivity. The lateral septal nucleus revealed the presence of a large number of labeled multipolar neurons and a moderate accumulation of labeled fibers. Labeling of several regions of the thalamus was especially dense. The paraventricular nucleus of the thalamus showed dense fiber labeling (Fig. 4E). Dense fiber labeling was also present along the borders

In order to correlate the distribution of 5-HT7-like immunoreactivity with pharmacologically defined 5HT7 receptors, we compared the distribution of immunoreactive cFos activation by 5-HT7 selective agonist stimulation and anti-5-HT7 ICC by simultaneous dual immunohistochemistry. Rats were injected icv with the partially selective serotonergic agonist 8OH-DPAT in the presence of WAY 100135, a selective antagonist of 5-HT1A receptors (Cliffe et al., 1993) (as described in Section 2). Overall, the general distribution of cFos immunoreactivity corresponded closely with that of immunopositive 5-HT7 in all forebrain regions, and in double labeled sections, cytoplasmic 5-HT7-like immunoreactivity and nuclear cFos immunoreactivity was almost always colocalized in the same cells. Examples of immunoreactivity for cFos and 5-HT7 receptors colocalized in individual neurons include the hippocampus (Fig. 5A), thalamus (Fig. 5C), and piriform cortex (Fig. 5E). When the rats were pretreated with pimozide prior to agonist treatment, cFos labeling was almost entirely blocked (Fig. 5B, D and F). In many brain regions, agonist stimulation appeared to redistribute 5-HT7 immunostaining away from cell bodies into the dendrites (e.g. hippocampus, Fig. 5A vs. B, and piriform cortex, Fig. 5E vs. F). This indicates that the same neurons that display 5-HT7-like immunoreactivity also demonstrate functional 5-HT7 receptors.

70

J.F. Neumaier et al. / Journal of Chemical Neuroanatomy 21 (2001) 63–73

4. Discussion The major contribution of this report is the development of an antiserum that selectively identifies rat 5-HT7 receptors. The antiserum recognizes 5-HT7 receptors in heterologous expressing HeLa cells by ICC and Western blot. The distribution of 5-HT7-like immunoreactivity in rat brain corresponded closely to 5-HT7 mRNA localized by in situ hybridization. Finally, a strategy designed to activate cFos using 5-HT7 selective receptor stimulation produced cFos immunoreactivity that colocalized with the 5-HT7 immunostained neurons. In both of these cases, there were no brain regions that demonstrated a noticeable difference between 5-HT7-like immunoreactivity and the cFos activation or in situ hybridization signal. The use of two distinct correlative approaches reduces substantially the likelihood that cross-reactivity of the 5-HT7

antiserum with any unintended antigenic targets led to misidentification of 5-HT7 expressing cells.

4.1. In 6itro characterization We used HeLa cells expressing 5-HT7 receptors to confirm the specificity of the antiserum. These cells provided an unequivocal comparison between 5-HT7 expressing and non-expressing (wild-type) cells, showing strong immunolabeling only in 5-HT7 transfected cells. These cells have been earlier characterized by radioligand binding and adenylate cyclase activity measurements, confirming that the transfected cells express functional 5-HT7 receptors (Heidmann et al., 1997). This stably transfected cell line was used to show that 5-HT7 receptor protein, 5-HT7 receptors are regulated by agonists and antagonists in a physiological manner (Zhukovskaya and Neumaier, 2000); the high expression levels used (4000 fmol/mg protein) should not alter the immunoreactive properties for the purpose of this study. A broad band of around 70 kDa was detected following Western blot only in 5-HT7 expressing cells. This approximate molecular weight is consistent with glycosylation of the native 49 kDa protein, and confirms that this antiserum recognizes the appropriately sized protein.

4.2. ISHH characterization

Fig. 5. Double immunostaining of 5-HT7 agonist-induced cFos expression and 5-HT7 receptor. WAY-DPAT, cFos expression (intensely stained nuclei) is induced by 5-HT7 receptor stimulation by the agonist (8-OH-DPAT) in the presence of the masking drug WAY 100635 (5-HT1A antagonist). Light 5-HT7 immunostaining can be seen in the cell body regions and proximal dendrites. PIM/WAYDPAT, pimozide (a 5-HT7 antagonist) blocks cFos stimulation by 5-HT7 receptor activation, while 5-HT7 immunoreactivity remains strong. Studies are shown for hippocampus CA1 (A and B, 600X), central medial thalamic nucleus (C and D, 300X), and piriform cortex (E and F, 600X). There appears to be redistribution of 5-HT7-like immunoreactivity from cell bodies to dendrites in hippocampus and piriform cortex presumably due to agonist induced redistribution within neurons.

While the use of ‘pure culture’ cell lines in helpful in characterizing an antibody, it does not rule out the possibility of additional, non-specific immunolabeling of undesired protein targets in brain sections. However, we found the same tissue distribution of 5-HT7-like immunoreactivity and 5-HT7 mRNA hybridization. Since ISSH depends on the specific nucleotide sequence rather than the peptide sequence, it utilizes an entirely different localization strategy yet yielded the same distribution of 5-HT7 expressing cells in rat brain. The primary, albeit subtle, differences that were noted between the ISHH and the ICC studies appeared in the cerebral cortex, the thalamus, and the hippocampus proper. Immunocytochemical labeling was found in fibers in superficial cortical layers eminating from cell bodies in layer 5, whereas 5-HT7 in situ hybridization signal was greatest in a thin layer corresponding to lamina 5. In the thalamus, ICC revealed more differences in staining intensity among the various nuclei than ISHH, especially at their borders, which were more intensely immunostained. In hippocampus, ISHH labeled 5-HT7 mRNA only in pyramidal and dentate granule layers. ICC staining for receptor protein was more widespread, including cell body as well as adjacent lamina. This suggests that 5-HT7 receptors have a somatodendritic in these neurons. While we cannot rule out the presence of 5-HT7 receptors in proximal axons,

J.F. Neumaier et al. / Journal of Chemical Neuroanatomy 21 (2001) 63–73

we did not detect evidence of distal axonal or terminal 5-HT7-like immunoreactivity.

4.3. Pharmacological-immunological correlations Differences in tissue preparation techniques prevented us from colocalizing 5-HT7 mRNA and 5-HT7like immunoreactivity in the same tissue sections. However, we used a pharmacological strategy to compare 5-HT7-mediated changes in gene expression in the same neurons that were detected with 5-HT7 ICC. cFos is an immediate early gene product which is transiently expressed strong neuronal activation such as occurs with receptor stimulated increases in cAMP production. 9-OH-DPAT is a relatively selective 5-HT7 ligand when 5-HT1A receptors are masked by WAY 100135. Since 5-HT7 receptors are positively coupled to adenylate cyclase, we predicted that 5-HT7 agonists would stimulate cFos immunoreactivity in cells bearing functional 5-HT7 receptors. Indeed, neurons throughout the forebrain showed cFos activation following 5-HT7 receptor stimulation, and these cells showed colocalized 5-HT7like immunoreactivity. To our knowledge, this is the first use of pharmacological activation of cFos to confirm the specificity of an antiserum directed against a G-protein coupled receptor. It also demonstrates that the activation of 5-HT7 receptors leads to a measurable biological effect that can be antagonized by 5-HT7 receptor blockers such as pimozide. The activation and blockade of biological effects by exogenous drugs are important elements in proving that 5-HT7 binding sites constitute functional receptors. Mullins et al. (1999) recently used a similar cFos strategy to study 5-HT7 receptor regulation after anti-depressant treatments; our results are compatible with theirs and extend their observations to a number of other brain regions.

4.4. Comparison with radioligand binding Two earlier groups have used binding of [3H]-5-carboxamidotryptamine to rat brain sections in the presence of masking agents. Waeber and Moskowitz (1995) used 8-OH-DPAT and GR127935 (100 nM each, to block 5-HT1A and 5-HT1B/D binding, respectively) as blocking agents, while Gustafson et al. (1996) masked with 30 nM p-aminophenethyl-m-trifluoromethylphenyl piperazine and 1.6 mM ( − )-pindolol. Both groups found a rather low level of binding under these conditions in generally the same distribution, although they interpreted the binding differently. The imperfect selectivity of these compounds and the use of an agonist radioligand (that may not recognize all affinity states of 5-HT7 receptors) are the primary problems with this approach. Our results using ICC shows more detail, and suggests some subtle differences. These are most apparent in the hippocampus, were we found strong

71

5-HT7-like immunoreactivity in CA3 stratum oriens and moderate staining in CA1 and in stratum lacunosum-moleculare of both CA1 and CA3. Waeber and Moskowitz (1995) found a similar distribution of binding but the relative intensities between regions were very different; they also concluded that some residual 5-HT1A binding was not masked under these conditions. Gustafson et al. (1996) did not describe hippocampal subregions in detail, but apparently did not detect as much 5-HT7 receptor in CA1 as we did, based on their figures.

4.5. Functional implications of 5 -HT7 distribution The functions of the tenia tecta is poorly understood. However, this primitive trilaminar cortical region has reciprocal connections with the olfactory and limbic regions (Ottersen, 1982; Luskin and Price, 1983). Since 5-HT7 antiserum robustly labeled neurons in tenia tecta and few if any other 5-HT binding sites appear to be localized in this region by indirect evidence (Waeber and Moskowitz, 1995), this may be an important area in which to study 5-HT7 physiology. Suprachiasmatic nucleus (SCN) is the primary circadian clock in mammalian brain. Earlier, 5-HT1A receptors were believed to be responsible for phase advancing the biological rhythms by inhibiting SCN neurons. However, it now appears that 5-HT7 receptors are responsible for this effect (Lovenberg et al., 1993; Bobrzynska et al., 1996; Ying and Rusak, 1997). 5-HT7 mRNA has been detected in SCN in some (Gustafson et al., 1996; Heidmann et al., 1998) but not all studies (Ruat et al., 1993; Gustafson et al., 1996). One explanation for this discrepency may be that different technical approaches have had variable sensitivity in detecting the several alternatively spliced 5-HT7 isoforms (Heidmann et al., 1997, 1998). We found ISHH, ICC, and pharmacological evidence for 5-HT7 receptors in SCN. If 5-HT7 agonists or antagonists, or drugs that modulate 5-HT7 receptor function in SCN induce phase shifts in the circadian clock, then they may influence sleep and activity in affective or other psychiatric disorders. Indeed, earlier studies have suggested that anti-depressants may downregulate 5-HT7 binding sites in hypothalamic membranes (Sleight et al., 1995) or cFos stimulation in SCN (Mullins et al., 1999). 5-HT7 receptors are densely distributed in several thalamic nuclei. Thalamus plays a broad range of roles in cortical –subcortical interactions, making predictions of the functional implications of 5-HT7 receptor activity in these nuclei difficult. However, since 5-HT innervation is diffuse and probably plays a modulatory role on the activity of many brain regions, perhaps 5-HT7 receptors can alter the sensitivity of thalamic neurons to stimulation by other neurotransmitter systems, changing the input-output relationship for thalamic

72

J.F. Neumaier et al. / Journal of Chemical Neuroanatomy 21 (2001) 63–73

information processing. For example, 5-HT7-induced elevations of cAMP in thalamic neurons might enhance the excitatory effect of other neurotransmitters such as glutamate. Stimulation of adenylate cyclase in hippocampus has recently been suggested to be a ‘final common pathway’ for many anti-depressant drugs, leading to increased cFos, CREB phosphorylation, and BDNF biosynthesis (Duman et al., 1997). This theory has specifically proposed that Gs coupled receptors in hippocampus play an important ‘downstream’ role in the action of antidepressants that modify monoaminergic neurotransmission. For example, serotonin-selective reuptake inhibitor and tricyclic anti-depressants both enhance serotonergic neurotransmission in hippocampus (Blier et al., 1987) by subtly different mechanisms. This would lead to greater stimulation of hippocampal 5-HT7 receptors, leading to some of the adaptive changes described by Duman. However, these adaptations take several weeks to occur, and may be region specific (e.g. anti-depressants downregulate 5-HT7 mediated responses in hypothalamus (Mullins et al., 1999). It might be possible to accelerate anti-depressant drug action if downstream targets are developed, perhaps including 5-HT7 receptors. Finally, 5-HT7 receptors are broadly distributed in cortex. 5-HT7 receptors in frontal cortex were found either in layer 5 pyramidal cells or more superficially located small oval neurons. Again, the functional implications of this distribution cannot yet be determined. However, many atypical anti-psychotic drugs such as clozapine, risperidone, and quetiapine bind to 5-HT7 and 5-HT2A receptors. Careful delineation of the relative contribution of each receptor type to the physiological effects of 5-HT and atypical antipsychotics in cortex may lead to further refinement in the targets for future treatments for schizophrenia. The existence of specific antibodies for the 5-HT7 receptor should play an important role in clarifying the role of this receptor in serotonergic neurotransmission in normal as well as pathological states. Acknowledgements This research was supported by the American Heart Association (MB) and the Stanley Foundation (JN). We thank Dr Mark Hamblin and colleagues for providing the 5-HT7 plasmid and stable cell line. References Bard, J.A., Zgombick, J., Adham, N., Vaysse, P., Branchek, T.A., Weinshank, R.L., 1993. Cloning of a novel human serotonin receptor (5-HT7) positively linked to adenylate cyclase. J. Biol. Chem. 268, 23422– 23426.

Barnes, N.M., Sharp, T., 1999. A review of central 5-HT receptors and their function. Neuropharmacology 38, 1083– 1152. Blier, P., De Montigny, C., Chaput, Y., 1987. Modifications of the serotonin system by antidepressant treatments: implications for the therapeutic response in major depression. J. Clin. Psychopharmacol. 7, 24S– 35S. Bobrzynska, K.J., Godfrey, M.H., Mrosovsky, N., 1996. Serotonergic stimulation and nonphotic phase-shifting in hamsters. Physiol. Behav. 59, 221– 230. Brownfield, M.S., Yracheta, J., Chu, F., Lorenz, D., Diaz, A., 1998. Functional chemical neuroanatomy of serotonergic neurons and their targets: antibody production and immunohistochemistry (1HC) for 5-HT, its precursor (5-HTP) and metabolite (5-HIAA), biosynthetic enzyme (TPH), transporter (SERT), and three receptors (5-HT2A, 5-ht5a, 5-HT7). Ann. NY Acad. Sci. 861, 232–233. Cliffe, I.A., Brightwell, C.I., Fletcher, A., Forster, E.A., Mansell, H.L., Reilly, Y., Routledge, C., White, A.C., 1993. (S)-N-tert-butyl-3-(4-(2-methoxyphenyl)-piperazin-1-yl)-2-phenylpropanamide [(S)-WAY-100135]: a selective antagonist at presynaptic and postsynaptic 5-HT1A receptors. J. Med. Chem. 36, 1509– 1510. Duman, R.S., Heninger, G.R., Nestler, E.J., 1997. A molecular and cellular theory of depression. Arch. Gen. Psychiatry 54, 597–606. Gustafson, E.L., Durkin, M.M., Bard, J.A., Zgombick, J., Branchek, T.A., 1996. A receptor autoradiographic and in situ hybridization analysis of the distribution of the 5-HT7 receptor in rat brain. Br. J. Pharmacol. 117, 657– 666. Heidmann, D.E., Metcalf, M.A., Kohen, R., Hamblin, M.W., 1997. Four-5-hydroxytryptamine7 (5-HT7) receptor isoforms in human and rat produced by alternative splicing: species differences due to altered intronexon organization. J. Neurochem. 68, 1372–1381. Heidmann, D.E., Szot, P., Kohen, R., Hamblin, M.W., 1998. Function and distribution of three rat 5-hydroxytryptamine7 (5-HT7) receptor isoforms produced by alternative splicing. Neuropharmacology 37, 1621– 1632. Hemedah, M., Coupar, I.M., Mitchelson, F.J., 1999. [3H]-Mesulergine labels 5-HT7 sites in rat brain and guinea-pig ileum but not rat jejunum. Br. J. Pharmacol. 126, 179– 188. Le Corre, S., Sharp, T., Young, A.H., Harrison, P.J., 1997. Increase of 5-HT7 (serotonin-7) and 5-HT1A (serotonin-1A) receptor mRNA expression in rat hippocampus after adrenalectomy. Psychopharmacology (Berlin) 130, 368– 374. Lovenberg, T.W., Baron, B.M., de Lecea, L., Miller, J.D., Prosser, R.A., Rea, M.A., Foye, P.E., Racke, M., Slone, A.L., Siegel, B.W., et al., 1993. A novel adenylyl cyclase-activating serotonin receptor (5-HT7) implicated in the regulation of mammalian circadian rhythms. Neuron 11, 449– 458. Luskin, M.B., Price, J.L., 1983. The topographic organization of associational fibers of the olfactory system in the rat, including centrifugal fibers to the olfactory bulb. J. Comp. Neurol. 216, 264– 291. Mengod, G., Vilaro, M.T., Raurich, A., Lopez-Gimenez, J.F., Cortes, R., Palacios, J.M., 1996. 5-HT receptors in mammalian brain: receptor autoradiography and in situ hybridization studies of new ligands and newly identified receptors. Histochem. J. 28, 747–758. Moyer, R.W., Kennaway, D.J., 1999. Immunohistochemical localization of serotonin receptors in the rat suprachiasmatic nucleus. Neurosci. Lett. 271, 147– 150 in process citation. Mullins, U.L., Gianutsos, G., Eison, A.S., 1999. Effects of anti-depressants on 5-HT7 regulation in the rat hypothalamus. Neuropsychopharmacology 21, 352– 367. Ottersen, O.P., 1982. Connections of the amygdala of the rat IV: corticoamygdaloid and intraamygdaloid connections as studied with axonal transport of horseradish peroxidase. J. Comp. Neurol. 205, 30 – 48. Pan, J.T., Yang, I.C., 1996. Central administration of 8-OH-DPAT and mCPP stimulates prolactin secretion in ovariectomized, estrogen-treated rats: lack of an effect on tuberoinfundibular dopaminergic neuron activity. Life Sci. 58, 1189– 1194.

J.F. Neumaier et al. / Journal of Chemical Neuroanatomy 21 (2001) 63–73 Ruat, M., Traiffort, E., Leurs, R., Tardivel Lacombe, J., Diaz, J., Arrang, J.M., Schwartz, J.C., 1993. Molecular cloning, characterization, and localization of a high-affinity serotonin receptor (5-HT7) activating cAMP formation. Proc. Natl. Acad. Sci. USA 90, 8547– 8551. Shen, Y., Monsma, F.J., Jr, Metcalf, M.A., Jose, P.A., Hamblin, M.W., Sibley, D.R., 1993. Molecular cloning and expression of a 5-hydroxytryptamine7 serotonin receptor subtype. J. Biol. Chem. 268, 18200– 18204. Sleight, A.J., Carolo, C., Petit, N., Zwingelstein, C., Bourson, A., 1995. Identification of 5-hydroxytryptamine7 receptor binding sites in rat hypothalamus: sensitivity to chronic antidepressant treatment. Mol. Pharmacol. 47, 99–103. Stam, N.J., Roesink, C., Dijcks, F., Garritsen, A., van Herpen, A., Olijve, W., 1997. Human serotonin 5-HT7 receptor: cloning and pharmacological characterisation of two receptor variants. FEBS Lett. 413, 489– 494. Stowe, R.L., Barnes, N.M., 1998. Selective labelling of 5-HT7 receptor recognition sites in rat brain using [3H]5-carboxamidotryptamine. Neuropharmacology 37, 1611–1619. Tsou, A.P., Kosaka, A., Bach, C., Zuppan, P., Yee, C., Tom, L., Alvarez, R., Ramsey, S., Bonhaus, D.W., Stefanich, E., et al.,

.

73

1994. Cloning and expression of a 5-hydroxytryptamine7 receptor positively coupled to adenylyl cyclase. J. Neurochem. 63, 456– 464. Vanhoenacker, I., Haegeman, I., Leysen, J.E., 2000. 5-HT7 receptors: current knowledge and future prospects. Trends Pharmacol. Sci. 21, 70 – 77 In Process Citation. Venero, J.L., Vizuete, M.L., Machado, A., Cano, J., 1997. Developmental expression of 5-HT7 receptor mRNA in rat brain visual structures after neonatal enucleation. Neuroreport 8, 1531–1535. Vizuete, M.L., Venero, J.L., Traiffort, E., Vargas, C., Machado, A., Cano, J., 1997. Expression of 5-HT7 receptor mRNA in rat brain during postnatal development. Neurosci. Lett. 227, 53 – 56. Waeber, C., Moskowitz, M.A., 1995. Autoradiographic visualisation of [3H]5-carboxamidotryptamine binding sites in the guinea pig and rat brain. Eur. J. Pharmacol. 283, 31 – 46. Ying, S.W., Rusak, B., 1997. 5-HT7 receptors mediate serotonergic effects on light-sensitive suprachiasmatic nucleus neurons. Brain Res. 755, 246– 254. Zhukovskaya, N.L., Neumaier, J.F., 2000. Clozapine downregulates 5-hydroxytryptamine(6) (5-HT(6)) and upregulates 5-HT(7) receptors in HeLa cells. Neurosci. Lett. 288, 236– 240 in process citation.