HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 277, Issue 39, 36233-36243, September 27, 2002

This Article Abstract Full Text (PDF) All Versions of this Article: 277/39/36233    most recent M202586200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sarret, P. Articles by Beaudet, A. Search for Related Content PubMed PubMed Citation Articles by Sarret, P. Articles by Beaudet, A. Social Bookmarking                      What's this?

Pharmacology and Functional Properties of NTS2 Neurotensin Receptors in Cerebellar Granule Cells^*

Philippe Sarret^§, Louis Gendron^¶, Peter Kilian^**, Ha Minh Ky Nguyen, Nicole Gallo-Payet^¶, Marcel-Daniel Payet^**, and Alain Beaudet^§§

From the  Department of Neurology and Neurosurgery, Montreal Neurological Institute, McGill University, Montreal, Quebec H3A 2B4, Canada, the ^¶ Service of Endocrinology, Department of Medicine, Sherbrooke University, Quebec J1H 5N4, Canada, and the ^** Department of Physiology and Biophysics, Faculty of Medicine, Sherbrooke University, Quebec J1H 5N4, Canada

Received for publication, March 18, 2002, and in revised form, June 3, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The binding and signaling properties of neuronal NTS2 neurotensin (NT) receptors were examined in cultured rat cerebellar granule cells. As shown by reverse transcription-PCR, receptor autoradiography, and confocal microscopic localization of fluorescent NT, these cells selectively express the NTS2 receptor subtype. Accordingly, a single apparent class of ¹²⁵I-NT-binding sites, with an affinity of 3.1 nM, was detected in cerebellar granule cell cultures. This binding was competed for with high affinity (IC₅₀ = 5.7 nM) by the NTS2 ligand levocabastine and with low affinity (IC₅₀ = 203 nM) by the NTS1 antagonist SR48692. Hypertonic acid stripping of surface-bound ligand and hyperosmolar sucrose treatment revealed that 64% of specifically bound ¹²⁵I-NT was internalized at equilibrium via a clathrin-dependent pathway. In cells loaded with the Ca²⁺-sensitive fluorescent dye Fluo4, SR48692, but neither NT nor levocabastine, triggered a marked increase in cytosolic [Ca²⁺]_i. By contrast, both NT and levocabastine, but not SR48692, induced a sustained (>60 min) activation of the mitogen-activated protein kinases, p42/p44, indicating functional coupling of NTS2 receptors. Complementary experiments carried out on synaptosomes from adult rat cerebellum demonstrated the presence of presynaptic NTS2 receptors. However, in contrast to perikaryal NTS2 sites, these presynaptic receptors did not internalize in response to NT stimulation. Taken together, the present results demonstrate that NTS2 receptors are present both presynaptically and postsynaptically in central neurons and that NT and levocabastine act as agonists on these receptors.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Neurotensin (NT)¹ is a tridecapeptide documented to act as a neurotransmitter/neuromodulator in the central nervous system and as a local hormone in the periphery (for review see Ref. 1). In the central nervous system, NT has been shown to modulate dopamine transmission in nigrostrial and mesolimbic pathways (2) and to play a role in the regulation of pain, temperature, appetite, and pituitary hormone secretion (for review see Ref. 3).

NT exerts its central and peripheral effects through interaction with specific membrane receptors. Three different subtypes of NT receptors, referred to as NTS1, NTS2, and NTS3, have been cloned (for review see Ref. 4). Although most of the documented effects of NT appear to be exerted through the high affinity, NTS1 receptor, recent studies suggest that the levocabastine-sensitive, low affinity NTS2 subtype may be responsible for the mediation of the antinociceptive actions of NT (5).

Originally referred to as an acceptor site (6), NTS2 has since been demonstrated to correspond to a bona fide, 7-transmembrane domain, G protein-coupled receptor (7). Autoradiographic (6, 8, 9) and in situ hybridization (10, 11) studies revealed that NTS2 is widely expressed throughout the rodent central nervous system, although it appears late during ontogeny (~1 month after birth in the rat) (10, 12, 13). NTS2 mRNA was detected in both neurons (10) and reactive astrocytes (14). However, whereas the binding and internalization properties of NTS2 have been characterized in astrocytes (14), nothing is known of its pharmacological and functional properties in neurons.

NT, levocabastine, and the NTS1 antagonist SR48692 were all found to induce an inward calcium-activated chloride current in Xenopus oocytes transfected with cDNA encoding the mouse NTS2 receptor (7, 15). By contrast, SR48692, but neither NT nor levocabastine, was capable of activating classical second messenger systems, including Ca²⁺ mobilization, inositol phosphate (InsPs) production, or MAPK phosphorylation in mammalian cells transfected with human NTS2 (16). Furthermore, the effects of SR48692 were blocked by NT and levocabastine, suggesting that these drugs could act as competitive antagonists at NTS2 sites (16). Yet mouse NTS2 receptors expressed in transfected epithelial cells were reported to internalize upon NT binding (17), a property usually associated with activation by agonists.

The aim of the present study was to characterize the binding, internalization, and signaling properties of NTS2 receptors in central neurons and to determine whether NT is agonist or antagonist at these sites. These properties were tested on rat cerebellar granule cells, because these cells express among the highest concentrations of NTS2 mRNA in the brain (10) and can be maintained for several weeks in culture (18), a feature required by the late ontogenetic appearance of the receptor. We also examined the pharmacological properties of NTS2 receptors in synaptosomes from adult rat cerebellum to discriminate putative presynaptic from post-synaptic features.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

All of the animal-related procedures were approved by the McGill and Sherbrooke University Animal Care Committees and carried out according to the regulations of the Canadian Council on animal care.

Autoradiographic Localization of NTS2-binding Sites-- Adult male Sprague-Dawley rats (220-250 g; Charles Rivers, St. Constant, Canada) were killed by decapitation, and their brains were rapidly frozen by immersion in liquid isopentane at 40 °C for 30 s. The brains were then deposited on dry ice for 10 min and stored at 80 °C until use. Sagittal sections (20 µm thick) were cut on a cryostat at 18 °C, mounted onto polylysine-coated slides, and stored at 20 °C for at least 24 h before further processing. For ¹²⁵I-Tyr³-neurotensin (¹²⁵I-NT) labeling, the sections were equilibrated at 4 °C for 5 min in ice-cold binding buffer (50 mM Tris-HCl, 5 mM MgCl₂, pH 7.4, and 0.2% BSA). They were then incubated with 8 nM ¹²⁵I-NT (100 Ci/mmol) (kindly provided by Dr. Jean Mazella, Sophia-Antipolis, France), with or without 1 µM levocabastine, for 60 min at 4 °C in 125 µl of binding buffer supplemented with 0.8 mM 1,10-phenanthroline. Additional sections were incubated in the presence of 10⁵ M unlabeled NT for the determination of nonspecific binding. After incubation, the sections were washed twice for 2 min at 4 °C in 50 mM Tris-HCl, pH 7.4, containing 0.2% BSA and again twice for 2 min at 4 °C in 50 mM Tris-HCl, pH 7.4. Autoradiographs were obtained by apposition of ¹²⁵I-labeled sections to max hyperfilm (Amersham Biosciences) for 2 weeks at room temperature in a light-proof x-ray cassette.

Films were scanned using an AGFA Duoscan T1200 scanner at 1200 pixels/inch resolution. The resulting TIFF files were processed using Adobe Photoshop 6.0 and Deneba Canvas 7 imaging software on an Apple PowerBook G3.

Cerebellar Granule Cell Cultures-- Primary cultures of mixed cerebellar cells were prepared as described by Cambray-Deakin (19), with the following modifications. Cerebelli (4-6 per culture) from 7-9-day-old Long Evans rats were isolated and minced with fine scissors. The tissue was transferred to buffer A (0.25 mg/ml trypsin and 200 µl of solution 4 (BSA fraction V (150 mg/ml), 0.7 M glucose, 75 mM MgSO₄·7H₂O) in 10 ml of Earle's balanced salt solution) and incubated for 15 min in a shaking water bath at 37 °C. Buffer B (0.4 mg/ml soybean trypsin inhibitor, 200 µl of solution 4, 25 units of DNase I, and 6 mM MgSO₄·7H₂O in 10 ml of Earle's balanced salt solution) was then added to the tissue mixture, and the suspension was centrifuged for 10 s at 180 × g to pellet the tissue chunks. The cells were then dispersed mechanically by repeated gentle pipetting through a sterile siliconized Pasteur pipette in 1.5 ml of buffer C (0.4 mg/ml soybean trypsin inhibitor, 200 µl of solution 4, 75 units of DNase I, and 6 mM MgSO₄·7H₂O in 10 ml of Earle's balanced salt solution). After 1-2 min of settling time, the supernatant was passed through a 4% BSA gradient and then centrifuged at 180 × g for 5 min at room temperature. The cell pellet was suspended in culture medium (minimum essential medium containing 10% horse serum, 0.4 mM L-glutamine, 1.2 mg/ml glucose, 0.36 mg/ml KCl, and 0.25% penicillin-streptomycin (Invitrogen)) and seeded into Petri dishes precoated with poly-D-Lysine (0.1 mg/ml) at an initial plating density of 1000 cells/mm². The cells were grown in a humidified atmosphere of 95% air, 5% CO₂, at 37 °C and used for experiments after 20 days in culture.

RNA Extraction and Reverse Transcription-Polymerase Chain Reaction Analysis-- Total RNAs were extracted from cultured cerebellar granule cells as well as from adult rat brain using the method of Chomczynski and Sacchi (20). These total RNAs (2 µg) were then reverse-transcribed at 42 °C for 1 h using 1 µg of oligo(dt)₁₅ primer (reverse transcription system kit; Promega) and 30 units of avian myeloblastosis virus reverse transcriptase in a total volume of 20 µl of the supplied buffer. The reaction was stopped by heating at 95 °C for 5 min. First strand cDNAs were subjected to 35 cycles of PCR in 25 µl of a final reaction volume containing 50 mM KCl, 10 mM Tris, pH 9, 1.5 mM MgCl₂, 0.1% Triton X-100, 0.02% BSA, 200 µM dNTPs, 0.5 unit of Taq DNA polymerase, and 100 ng of either one of the following three pairs of sense and antisense primers. The first pair (5'-ACACCCATTGTGGACACAGCC-3' and 5'-TTCATCCGAGATATAGCAGAA-3') allowed the amplification of a fragment of NTS1 receptor cDNA from bases 676-1011 in the sequence reported previously (21). The predicted size of the amplified fragment was 335 bp. The second pair (5'-GAATGTGCTGGTGTCCTTCGC-3' and 5'-ACTTGTATTTCTCCCAGGCTG-3') allowed amplification of a fragment of NTS2 receptor cDNA from bases 667-1287 in the sequence reported previously (22). The predicted sizes of the amplified fragments were 620 bp for NTS2 and 439 bp for the splice variant form of this receptor (23). The third pair (5'-TCCCGAGAACTCTGGAAAGGT-3' and 5'-CACAGAGGCGAAGAGGAAACG-3') allowed amplification of a fragment of NTS3 receptor cDNA from bases 255-681 in the sequence reported previously (24). The predicted size of the amplified fragment was 426 bp. Amplification was carried out with a first cycle at 94 °C for 3 min, 52 °C for 2 min, 72 °C for 1.5 min, followed by 34 cycles at 94 °C for 45 s, 52 °C for 40 s, 72 °C for 1 min, and a final extension step at 72 °C for 8 min. PCR products were analyzed on a 2% agarose gel. In all of the reverse transcription experiments, two types of controls were performed: 1) each total RNA sample was subjected to reverse transcription (RT) in the absence of enzyme to control for intrinsic contamination by genomic DNA, and 2) the reaction was performed on the RT mixture without RNA added to control for contamination during the experiment.

Binding of ¹²⁵I-NT to Cerebellar Granule Cells-- To determine the binding parameters of ¹²⁵I-NT to cultured granule cells and to investigate whether specifically bound ¹²⁵I-NT was internalized in these cells, three sets of binding experiments were carried out.

For saturation studies, the culture medium was discarded from 12-mm dishes containing 2 × 10⁵ cells, and the cells were equilibrated for 10 min at 37 °C in Earle's buffer (140 mM NaCl, 5 mM KCl, 1.8 mM CaCl₂, 0.9 mM MgCl₂, and 25 mM HEPES, pH 7.4) supplemented with 0.1% glucose and 0.1% BSA. Equilibration buffer was then replaced by 250 µl of Earle's buffer containing 0.1-16 nM of ¹²⁵I-NT isotopically diluted with unlabeled NT in the presence of 0.8 mM 1,10-phenanthroline for 30 min at 37 °C. At the end of the incubation, the cells were washed twice with 1 ml of equilibration buffer and harvested with 1 ml of 0.1 M NaOH. The associated radioactivity was then counted in a  counter. Dissociation constant (K_d) and maximal binding capacity (B_max) were derived from Scatchard analysis of the data.

The competition experiments were carried out on membranes freshly prepared from cerebellar granule cells. For this purpose, the cells were scraped off the culture dishes with ice-cold phosphate-buffered saline and centrifuged in microcentrifuge tubes at 15,000 × g for 5 min at 4 °C. The pellet was homogenized by incubation in hypotonic TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 7.5) and sonicated. The membrane homogenates were then recovered by centrifugation at 15,000 × g for 30 min at 4 °C. The cell membranes (50 µg) were incubated with 2 nM ¹²⁵I-NT (100 Ci/mmol) for 30 min at 25 °C in 250 µl of binding buffer (50 mM Tris-HCl, pH 7.5, containing 0.2% BSA and 0.8 mM 1-10 phenanthroline) in the presence of increasing concentrations (from 10¹¹ to 10⁵ M) of nonradioactive NT, N-Bodipy-NT-(2-13), levocabastine (kindly provided by Janssen Research, Beerse, Belgium) (6), or SR48692 (kindly provided by Sanofi Research, Toulouse, France) (25). The binding experiments were terminated by addition of 3 ml of ice-cold buffer followed by filtration through cellulose acetate filters (Sartorius). The filters were then washed twice with 3 ml of ice-cold buffer, and the radioactivity retained in them was counted in a  counter. The IC₅₀ values were determined from inhibition curves as the unlabeled ligand concentration inhibiting 50% of ¹²⁵I-NT-specific binding.

To investigate ¹²⁵I-NT internalization, whole cells were incubated with 0.4 nM ¹²⁵I-NT in Earle's buffer, with or without 0.45 M sucrose, for 45 min at 37 °C. At the end of the incubation, the cells were washed twice either with 1 ml of Earle's buffer or with 1 ml of a hypertonic acid buffer (Earle's buffer containing 0.2 M acetic acid and 0.5 M NaCl, pH 4) for 3 min to strip off surface-bound radioactivity. The cells were then harvested with 1 ml of 0.1 M NaOH, and associated radioactivity was counted in a  counter. Nonspecific binding, as measured in the presence of 1 µM unlabeled NT, represented less than 5% of the total binding. All of the binding/internalization data were calculated and plotted using Prism 3.02 (Graph Pad Software) and represent the means ± S.D. of n determinations (as indicated under "Results").

Binding of N-Bodipy-NT-(2-13) (Fluo-NT) to Cerebellar Granule Cells-- For fluo-NT labeling, the cerebellar granule cells were grown on 12-mm polylysine-treated glass coverslips in 18-mm Petri dishes. The cells were equilibrated for 10 min at 37 °C in Earle's buffer containing 0.2% BSA and 0.1% glucose and then incubated for 10 or 40 min in the same buffer with 20 nM fluo-NT (kindly provided by PerkinElmer Life Sciences) in the presence or absence of 10⁵ M nonfluorescent NT, 10⁵ M levocabastine, or 10⁵ M SR48692.

To determine whether fluo-NT internalization proceeded through clathrin-coated pits, a second set of experiments was performed on whole cells by leaving them untreated or adding 0.45 M sucrose, 10 µM phenylarsine oxide (Sigma) or 1 µM monodansylcadaverine (Sigma) to both the equilibration and binding buffers.

At the end of the incubation, the cells were washed twice either with binding buffer or with hypertonic acid buffer (pH 4) for 3 min. The cells were then either air-dried and mounted on glass slides with Aquamount or further processed for immunocytochemistry as described below.

Double Immunofluorescence Labeling-- To identify cell types and intracellular compartments in which fluo-NT had been internalized, fluo-NT-incubated granule cells were fixed with 4% paraformaldehyde containing the cross-linking agent Bis(sulfosuccinimidyl) suberate (5 mM; Pierce) for 15 min at room temperature. The cells were then washed twice with Earle's buffer, rinsed twice with phosphate-buffered saline (PBS), preincubated with PBS containing 10% normal goat serum, 0.05% Triton X-100, and 2% BSA for 20 min, rinsed again with PBS, and incubated for 60 min at room temperature with one of the following mouse antibodies: anti-glial fibrillary acidic protein (GFAP; 1:200; Sigma), anti-neurofilament (N52; 1:200; Sigma), and anti-syntaxin 6 (3 µg/ml; Transduction Laboratories). All of the primary antibodies were diluted in PBS containing 0.5% normal goat serum and 0.02% Triton X-100. After rinsing three times (5 min each time) with PBS, bound primary antibodies were revealed with goat anti-mouse antibodies conjugated to either fluorescein isothiocyanate (diluted 1:50 in PBS) or to Alexa 488 (Molecular Probes; diluted 1:500 in PBS) for 60 min at room temperature.

After washing, the coverslips were mounted on glass slides using Aquamount and examined with either a Nikon Eclipse TE300 microscope equipped for epifluorescence using a B-1E fluorescein isothiocyanate filter set (Nikon, Mississauga, Canada) or a Zeiss laser scanning microscope (CLSM 410) equipped with an Axiovert 100 inverted microscope and an argon/krypton laser. To determine the proportion of neurons endowed with NTS2 receptors, the percentage of N52-immunoreactive neurons exhibiting fluo-NT labeling was determined in 12 wells from three different experiments (total of 425 neurons counted), and the results were expressed as the means ± S.E.

Intracellular Calcium Measurements-- For intracellular calcium ([Ca²⁺]_i) measurements, the cells were cultured for 21 days on plastic coverslips (Sartstedt, St-Laurent, Canada) and incubated in serum-free Dulbecco's modified Eagle's medium (Invitrogen) supplemented with 5 µM fluo-4/acetoxymethyl ester (Molecular Probes, Eugene, OR) at 37 °C for 30 min. The cells were then washed three times with 0.5% BSA and further incubated in PBS-HEPES (140 mM NaCl, 5.4 mM KCl, 2 mM CaCl₂, 1 mM MgCl₂·6H₂O, 10 mM HEPES, pH 7.35) at 37 °C for 30 min to allow for hydrolysis of the acetoxymethyl ester form. The coverslips were then mounted on the stage of a Nikon Eclipse TE300 inverted microscope.

The cells were maintained at 37 °C throughout the experiments with a heating Peltier element. NT, levocabastine, and SR48692 diluted in 1 ml of fresh PBS-HEPES were added to the incubation medium, individually or in combination, at concentrations of 10⁸-10⁶ M. Images were acquired each 10 s using a CoolSnap f_x digital CCD camera (Roper Scientific, Tucson, AZ) cooled at 35 °C. Band pass filters were used for excitation and emission (450-490 and 520-560 nm, respectively). Average intensity for each cell was measured on-line using the Metafluor software package (Universal Imaging Corp, West Chester, PA). Further analyses were performed with the Metafluor software on images stocked on the hard disc of an IBM-compatible computer. Each Ca²⁺ curve represents the average response (± S.D.) of n cells as indicated in the figure legends. Statistical significance was verified using Student's t test.

Measurement of InsPs Accumulation-- The experiments were performed as described previously (26). Briefly, the cells were grown for 21 days in minimum essential culture medium containing 10% horse serum and labeled for 18 h with fresh medium containing 5 µCi/ml of myo-[³H]inositol. The radioactive medium was then discarded, and the cells were incubated in isotope-free culture medium. After 1 h, the cultures were equilibrated for 15 min in Hanks' balanced salts solution with glucose and LiCl (10 mM). The stimulation was performed for 15 min using NT (10⁷ and 10⁶ M), levocabastine (10⁷ and 10⁶ M), SR48692 (10⁷ and 10⁶ M), or fluoroaluminate. The incubation was stopped by aspiration of the medium and rapid addition of 1 ml of 5% (v/v) HClO₄ and 200 µl of BSA (20 mg/ml). InsPs were separated by ion exchange chromatography on Dowex 1 × 8 columns. The radioactivity incorporated in the InsPs fractions was determined by scintillation counting in gel phase in a Beckman  counter, with a counting efficiency of 18%. The values correspond to the means ± S.E. of three independent experiments, each performed in triplicate.

Western Blotting Analyses of Mitogen-activated Protein Kinases, p42/p44^mapk-- Cultured granule cells (21 days post-plating) were incubated for various time intervals (from 1 to 120 min) at 37 °C in culture medium in the presence of NT (10⁸ and 10⁷ M), levocabastine (10⁶ M), or SR48692 (10⁷ and 10⁶ M). The reaction was stopped by aspiration of the medium and the addition of Hanks' balanced salt solution with glucose containing 0.1 µM staurosporine and 1 mM sodium orthovanadate. The cells were then lysed at 4 °C in 50 mM HEPES, pH 7.8, containing 1% Triton X-100, 0.1 µM staurosporine, 1 mM sodium orthovanadate, and Complete^TM protease inhibitor (Roche Diagnostic Laboratories). The cell extracts were centrifuged at 8,000 × g for 15 min at 4 °C, and the supernatants were stored at 20 °C until use.

Equal amounts of proteins (30 µ g) were separated on 10% SDS-polyacrylamide gels. The Western blots were performed as described previously (27). Briefly, polyvinylidene difluoride membranes (Roche Diagnostic Laboratories) containing proteins were incubated for 2 h at room temperature with anti-phosphorylated p42/p44^mapk (dilution 1:1,000; New England Biolabs, Missisauga, Canada), anti-p42/p44^mapk (dilution 1:1,000; New England Biolabs) antibodies, followed by four washes with Tris-buffered saline/Tween 20 buffer. Detection was accomplished using horseradish peroxidase-conjugated anti-rabbit or anti-mouse antibodies (1:2,000; Amersham Biosciences) and an ECL detection system (Roche Diagnotics Laboratories).

To quantitate the effects of NT and SR48692 on p42/p44^mapk phosphorylation, the ratios of phosphorylated p42/p44^mapk over total p42/p44^mapk levels were determined by densitometry, using Molecular Dynamics Image Quant Imaging software. The statistical significance was verified using Barlett's test, and the p values were obtained from Dunett's tables. The calculations and statistical analyses were performed using Excel 2000 (Microsoft) and Prism 3.02 (Graph Pad Software).

Binding of [³H]NT to Rat Cerebellar Synaptosomes-- To determine whether cerebellar granule cells did target NTS2 receptor proteins presynaptically, synaptosomes were prepared from the cerebelli of rat. The rats were killed by decapitation, and their brains were rapidly removed. The cerebelli from 10 rats were pooled and homogenized with a hand glass homogenizer using a Teflon-coated pestle rotating at 900 rpm in ice-cold HEPES, pH 7.4, containing 0.32 M sucrose and peptidase inhibitors. The homogenate was centrifuged at 750 × g for 5 min at 4 °C, and the resulting supernatant was recuperated and further centrifuged at 12,000 × g for 15 min at 4 °C. The synaptosomal P2 pellet was equilibrated in 0.32 M sucrose and centrifuged at 14,500 × g for 15 min at 4 °C and then resuspended in modified Earle's buffer (150 mM choline chloride, 5 mM KCl, 1.8 mM CaCl₂, and 0.9 mM MgCl₂) for experimental use.

To assess morphological preservation, the synaptosomes were fixed with 2% acrolein-2% paraformaldehyde in 0.1 M phosphate buffer. To test the effects of experimental manipulations on morphological integrity, the synaptosomes were preincubated for 25 min in Earle's buffer, rinsed or not with a hypertonic solution (pH 4) containing 0.5 M NaCl, and fixed as above. The synaptosomes were post-fixed with 2% osmium tetroxide for 30 min and rinsed with 0.1 M phosphate buffer. They were then dehydrated in graded alcohols, embedded in Epon, pelleted in a plastic mold by centrifugation (11,000 × g), and polymerized for 4 days at 60 °C. Ultrathin sections (80 nm) were cut on an ultramicrotome and counterstained with uranyl acetate and lead citrate prior to examination with a JEOL 100CX electron microscope (Peabody, MA).

For saturation binding experiments, 100 µg of synaptosomal proteins were incubated in 250 µl of Earle's buffer containing increasing concentrations (from 0.5 to 10 nM) of [³H]NT (108 Ci/mmol; PerkinElmer Life Sciences) and 0.8 mM 1,10-phenanthroline for 25 min at 37 °C. For determination of nonspecific binding, additional samples were incubated in the presence of 1 µM nonradioactive NT. Binding was terminated by two successive additions of 12 ml of ice-cold Earle's buffer and filtering under vacuum through GF/B filters presoaked for 1-2 h at 4 °C in Earle's buffer containing 0.3% polyethylenimine. After incubation for >5 h with 10 ml of Safe liquid Scintillation mixture (ICN Biomedicals), the radioactivity was measured in a Beckman  counter with a counting efficiency of 30%.

For competition experiments, the synaptosomes were incubated for 20 min at 37 °C in Earle's buffer containing 3 nM [³H]NT and increasing concentrations (from 10¹¹ to 10⁵ M) of nonradioactive NT, levocabastine, or SR48692. The binding was terminated, and the radioactivity counted as described above.

To determine the proportion of internalized [³H]NT, the synaptosomes were incubated at 37 °C in Earle's buffer containing 3 nM [³H]NT for intervals ranging between 2 and 20 min. They were then rinsed for 3 min with either ice-cold Earle's buffer or with a hypertonic acid solution (pH 4), and the radioactivity was counted as above. The binding data were calculated and plotted using Prism 3.02 (Graph Pad Software) and represent the means ± S.D. from two experiments performed in triplicate.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Autoradiographic Distribution of ¹²⁵I-NT-labeled NTS2-binding Sites in Rat Cerebellum-- To determine whether the distribution of NTS2-binding sites correlated with the documented expression (10) of NTS2 mRNA in rodent cerebellar cortex, sagittal sections from the adult rat brain were incubated with 8 nM ¹²⁵I-NT in the presence or absence of an excess of the NTS2 ligand, levocabastine. In the absence of levocabastine, high levels of specific (i.e. NT-displaceable) ¹²⁵I-NT binding were observed throughout the cerebellar cortex (Fig. 1A, n = 2). Cerebellar ¹²⁵I-NT labeling was totally abolished in the presence of levocabastine, indicating that it concerned exclusively NTS2 sites (Fig. 1B).

View larger version (42K): [in this window] [in a new window]   Fig. 1.   Autoradiographic distribution of 125I-NT binding in rat cerebellum. The sagittal sections were incubated with 8 nM 125I-NT (100 Ci/mmol), in the absence (A) or in the presence (B) of 1 µM levocabastine. A, in the absence of levocabastine, labeled 125I-NT-binding sites are evident in both granule cell (Gc) and molecular (ml) layers of the cerebellar cortex. B, this labeling is reduced to background levels in the presence of levocabastine. Scale bar, 2.5 mm.

NT Receptor Expression in Cultured Cerebellar Granule Cells-- After 21 days in culture, cerebellar granule cells exhibited highly refringent, rounded perikarya, and formed a dense network of highly ramified processes superimposed over a monolayer of glial cells (Fig. 2).

View larger version (160K): [in this window] [in a new window]   Fig. 2.   Representative phase contrast micrograph of a rat cerebellar cell culture, 21 days post-plating. Several small, rounded cell bodies of granular cells (arrows) and numerous neuritic processes (arrowheads) are visible, growing on a layer of glial cells. The morphological appearance of granule cells indicate extensive differentiation. Scale bar, 20 µm.

To determine which of the cloned NT receptors were expressed in these cultures, the cell homogenates were subjected to RT-PCR analysis of their mRNA content using three different pairs of primers designed to selectively recognize the cloned rat NTS1 (21), NTS2 (22), and NTS3/gp95 sortilin (24) receptors. As shown in Fig. 3, PCR amplification of mRNA with NTS2 probes yielded two bands of 620 and 439 bp in size, respectively. These two bands were of the same molecular weight as the unspliced and spliced variants of the NTS2 receptor amplified from rat brain extracts (23). Probing amplified mRNA with NTS3 receptor primers yielded a single band of 426 bp, corresponding to the molecular weight of the cloned NTS3 receptor (24). A fragment of identical size was also amplified from mRNA prepared from rat brain extracts. No hybridization signal was detected in cerebellar culture extracts probed with NTS1 primers. However, NTS1 fragments of appropriate size (335 bp) were amplified in rat brain extracts using the same pair of primers, demonstrating the efficacy of the probes. Finally, no signal was detected when transcribed products from cultured cells were amplified with either one of the sense or antisense primers alone (not shown).

View larger version (30K): [in this window] [in a new window]   Fig. 3.   Expression of NT receptor mRNAs in 21-day-old cerebellar granule cells as compared with the adult rat brain. PCR reactions were performed on mRNAs reverse-transcribed using specific primers for rat NTS1, NTS2, or NTS3. Two bands of 620 and 439 bp in size are detected in cerebellar granule cells with NTS2 primers. The large band corresponds to the unspliced form of the rat NTS2 receptor, and the 439-bp form corresponds to the spliced variant (vNTS2). These bands are of the same molecular weight as in rat brain extracts. No band is detected in cerebellar granule cells by using the NTS1 primers, although a signal migrating as 335 bp, consistent with the size of rat NTS1, is present in whole rat brain extracts. Finally, a single band of 426 bp is observed in cerebellar granule cells using NTS3 primers. This band migrates at the same molecular weight mark as in rat brain extracts.

Binding and Internalization of ¹²⁵I-NT in Cerebellar Granule Cell Cultures-- Incubation of 21-day-old granule cell cultures with increasing doses of ¹²⁵I-NT for 30 min at 37 °C revealed the presence of specific (i.e. NT-displaceable) and saturable ¹²⁵I-NT binding (Fig. 4A). Scatchard analysis of the data yielded a linear plot, indicating that the binding involved a single apparent population of sites (Fig. 4A, inset). The deducted dissociation constant (K_d) of ¹²⁵I-NT binding to these sites was 3.1 ± 0.1 nM, and the maximal binding capacity (B_max) was 50.5 ± 1.7 fmol/mg (n = 5).

View larger version (11K): [in this window] [in a new window]   Fig. 4.   Specific binding of 125I-NT to cerebellar granule cells in culture. A, saturation binding experiments performed on whole cells for 30 min at 37 °C with increasing concentrations of labeled NT. Scatchard analysis (inset) indicates binding to a single apparent population of sites, with a Kd of 3.1 ± 0.1 nM and a maximal binding capacity (Bmax) of 50.5 ± 1.7 fmol/mg. The values correspond to the means ± S.D. of five independent experiments performed in duplicate. B, competition inhibition of 125I-NT (2 nM) binding to membranes of cultured granule cells by nonradioactive NT (open squares), N-Bodipy-NT-(2-13) (closed circles), levocabastine (closed squares), or SR48692 (open circles). Each point represents the mean of two separate experiments performed in triplicate (means ± S.D.). The IC50 values are listed in Table I. C, binding kinetics of 125I-NT to whole cerebellar cells at 37 °C. The experiments were performed in the absence (squares) or in the presence (circles) of 0.45 M sucrose. At the indicated times, the cells were either washed twice with 500 µl of Earle's/HEPES/Tris buffer (open symbols) or treated with 500 µl of acid-NaCl buffer (pH 4) for 3 min (closed symbols). The values are the means ± S.D. of four independent experiments carried out in duplicate.

The ability of various unlabeled peptide and nonpeptide compounds to compete with ¹²⁵I-NT binding in membrane preparations from cultured granule cells is illustrated in Fig. 4B and Table I. Levocabastine (IC₅₀ = 5.7 ± 0.5 nM) was almost as potent as unlabeled NT (IC₅₀ = 3.2 ± 0.1 nM) in inhibiting ¹²⁵I-NT binding (Fig. 4B, n = 2). The fluorescent analog N-Bodipy-NT-(2-13) (fluo-NT) also competed with specific ¹²⁵I-NT binding but with less potency than unlabeled NT (IC₅₀ = 9.2 ± 0.3 nM). The nonpeptide NTS1 antagonist, SR48692, was the least potent of all drugs tested with an IC₅₀ value of 203 ± 21 nM (Fig. 4B).

To determine whether ¹²⁵I-NT internalized through NTS2, association kinetics of ¹²⁵I-NT binding were performed on whole cells at 37 °C, and the proportion of sequestered radioactivity was determined after hypertonic acid wash of surface bound molecules. In the absence of any pretreatment, ¹²⁵I-NT bound specifically to intact cultures in a time-dependent manner (Fig. 4C, n = 4). Specific ¹²⁵I-NT binding reached a plateau within 20 min. At that time, removal of surface-bound radioactivity by acid NaCl wash revealed that 64.4 ± 0.7% of total bound ¹²⁵I-NT was intracellular (Fig. 4C). After preincubation and labeling of the cells in the presence of 0.45 M sucrose, which inhibits internalization via clathrin-coated pits (28), the amount of radioactivity specifically associated with the cells at equilibrium was about 75.5 ± 0.9% of that measured in the absence of treatment (Fig. 4C). This bound radioactivity was almost entirely washed off by hypertonic acid buffer treatment, indicating that the binding was confined to the cell surface (Fig. 4C). Taken together, these data suggest that in cerebellar granule cell cultures, NT internalizes in an NTS2-mediated, clathrin-dependent manner.

Binding and Internalization of Fluo-NT in Cerebellar Granule Cell Cultures-- To visualize NT binding and internalization, 21-day-old cerebellar granule cell cultures were incubated at 37 °C with 20 nM fluo-NT. Following 10 min of incubation, punctate fluorescent labeling was evident over both perikarya and processes of neuronal cells (Fig. 5, A and B). This labeling was specific in that it was entirely competed for by an excess of nonfluorescent NT (not shown). Hypertonic acid wash confirmed that the bulk of this fluorescent signal was intracellular (Fig. 5B). In keeping with the results of radioligand binding studies, fluo-NT labeling was totally abolished when the incubation was carried out in the presence of an excess of levocabastine or of SR48692 (not shown). When the incubation was carried out in the presence of 0.45 M sucrose, phenylarsine oxide, or monodansylcadaverine, specifically bound fluo-NT molecules remained clustered on the cell surface (Fig. 5C). This peripheral labeling was exclusively surface-bound because it was entirely strippable by hypertonic acid wash (Fig. 5D).

View larger version (23K): [in this window] [in a new window]   Fig. 5.   Confocal microscopic images of fluo-NT-labeled cerebellar granule cells in culture. A, after 10 min of incubation with 20 nM N-Bodipy-NT-(2-13) at 37 °C, small fluorescent hot spots are visible throughout the cytoplasm. Note the presence of labeling in both the perikaryon and processes (arrowheads). B, hypertonic acid wash of cell surface binding reveals that fluo-NT labeling is mainly intracellular (arrowheads). C, when the incubation is carried out in the presence of 0.45 M sucrose, bound fluorescent molecules remain confined to the periphery of the cell (arrowheads). D, this peripheral labeling disappears after hypertonic acid wash, confirming that it is associated with the cell surface. N, nucleus. Scale bar, 5 µm.

Between 10 and 40 min of incubation, hot spots of internalized fluo-NT hot spots coalesced into a single juxta-nuclear fluorescent cluster (Fig. 6, A and B). Immunostaining of fluo-NT-labeled cells with an antibody against syntaxin-6, which selectively labels the recycling endosome/trans-Golgi network (TGN) complex (29), revealed that this juxta-nuclear compartment corresponded to the TGN (Fig. 6C).

View larger version (30K): [in this window] [in a new window]   Fig. 6.   Time course of fluo-NT binding internalization in cultured cerebellar granule cells. A-C, confocal microscopic images of cells incubated with 20 nM N-Bodipy-NT-(2-13). A, after 10 min of incubation with fluo-NT at 37 °C, the internalized ligand forms numerous hot spots distributed throughout the cytoplasm of the cell (arrows). B, by 40 min, the bulk of intracellular fluorescence is concentrated in the perinuclear region (arrow). C, immunocytochemical labeling of the cell in B with antibodies against syntaxin-6 identifies this juxtanuclear compartment as being part of the TGN/pericentriolar recycling endosome complex. D-I, epifluorescence localization of internalized fluo-NT (D) and of GFAP (E) in doubled-labeled cells; merged images in F demonstrate that cells internalizing the fluorescent ligand are strictly GFAP-negative (F, arrows); by contrast, co-labeling of Fluo-NT (G, arrowheads) with N52 (H, arrowheads) shows that fluo-NT internalizes in N52-positive, i.e. neuronal cells (I, merged image; arrowheads). Scale bar, 10 µm.

To confirm the identity of fluo-NT-labeled cells, fluo-NT labeling was combined with immunocytochemical detection of selective neuronal or glial markers. Combined epifluorescence visualization of internalized fluo-NT (Fig. 6D) with the glial marker GFAP (Fig. 6E) showed a complete lack of fluo-NT uptake by GFAP-positive glial cells (Fig. 6F). In contrast, dual localization of fluo-NT (Fig. 6G) and of the neuronal marker N52 (Fig. 6H) showed that all cells labeled with fluo-NT were N52-positive and thus corresponded to neurons (Fig. 6I). Conversely, 59.6 ± 0.5% of N52-immunoreactive neurons exhibited fluo-NT labeling, suggesting that only a subpopulation of cultured granule cells expresses NTS2 receptors.

Measurements of Intracellular Calcium-- To determine whether NTS2 receptor stimulation resulted in intracellular Ca²⁺ mobilization, cultured cerebellar granule cells were loaded with the calcium-sensitive fluorescent dye, Fluo4 and stimulated with a variety of purported NTS2 agonists. As shown in Fig. 7, neither 1 µM NT (Fig. 7A) nor 1 µM levocabastine (Fig. 7B) were able to stimulate cytosolic [Ca²⁺] mobilization. Yet, intracellular Ca²⁺ stores were readily releasable in unsuccessfully stimulated neurons because application of the drug thapsigargin, a known inhibitor of the sarco(endo)plasmic reticulum Ca²⁺-ATPase pumps, induced a robust Ca²⁺ signal (Fig. 7).

View larger version (10K): [in this window] [in a new window]   Fig. 7.   Intracellular Ca2 mobilization in cultured cerebellar granule cells. A, application of 1 µM NT does not stimulate cytosolic Ca2+ mobilization in fluo4-loaded neurons (n = 18) (successful loading demonstrated by thapsigargin (TG) depletion of intracellular Ca2+ stores). B, similarly, no Ca2+ response is induced by stimulation with 1 µM levocabastine (n = 15). Each curve represents the mean ± S.D. of n responding cells.

In accordance with data on heterologous transfection systems (15, 16, 30), perfusion of cerebellar granule cells with the NTS1 antagonist SR48692 (1 µM) caused a marked elevation of free intracellular calcium (Fig. 8A), with a stimulation ratio of 4-6-fold over basal values, a level similar to that observed with thapsigargin. This increase varied from cell to cell but consistently showed a biphasic profile with rapid ascending and slow descending slopes (Fig. 8A). This stimulatory effect of SR48692 was antagonized neither by prior (Fig. 8B) nor concomitant (Fig. 8C) application of 1 µM NT. Likewise, the SR48692 response was unaffected by concomitant application of 10 µM levocabastine (Fig. 8D). In cells preincubated for 5 min in a Ca²⁺-free medium, stimulation with 1 µM SR48692 induced a transient Ca²⁺ increase (Fig. 8E). Furthermore, the sustained phase of the Ca²⁺ response to SR48692 was markedly shortened by the addition of 5 mM EGTA (a Ca²⁺ chelator) to the extracellular medium (Fig. 8F). These responses revealed that the increase in [Ca²⁺]_i was the consequence of both the release of Ca²⁺ from intracellular stores and the entry of Ca²⁺ from the external medium.

View larger version (21K): [in this window] [in a new window]   Fig. 8.   Intracellular Ca2+ mobilization in fluo4-loaded cultured cerebellar granule cells. A, application of 1 µM of SR48692 induces a transient increase in Ca2+ followed by a plateau (n = 50). B and C, this response is not antagonized by prior (B, n = 42) or concomitant (C, n = 25) application of 1 µM NT. D, similarly, the Ca2+ response to SR48692 is unaffected by concomitant application of 10 µM levocabastine (Levo, n = 18). E and F, Ca2+ response to application of 1 µM SR48692 (SR) in the absence of extracellular Ca2+ (E, n = 8) or in the presence of 5 mM EGTA (F, n = 15). The curves represent the means ± S.D. of n responding cells. The peak values from Ca2+ increase in A-C and are not statistically different from one another at p  0.05 level.

Measurements of Inositol Phosphate Accumulation-- To further investigate the putative involvement of the Ca²⁺/phosphoinositide pathways in the mechanism of action of the NTS2, the capacity of NT, levocabastine, and SR48692 to modulate InsPs accumulation was also assayed. As shown in Fig. 9, neither NT (10⁷ or 10⁶ M) nor levocabastine (10⁷ or 10⁶ M) stimulation significantly altered the basal level of InsPs present in control cells. Similarly, SR48692 (10⁷ or 10⁶ M) application did not modify phosphoinositide turnover. By contrast, after 15 min of stimulation with fluoroaluminate, a nonspecific activator of all heterotrimeric G proteins, there was a 3-fold increase in InsPs accumulation, indicating that granule cells in culture were able to produce this class of second messengers.

View larger version (17K): [in this window] [in a new window]   Fig. 9.   InsPs accumulation in 21-day-old cerebellar granule cells. Irrespective of the concentrations used, NT, levocabastine, and SR48692 (SR) do not modify the intracellular levels of InsPs in cerebellar granule cells. The ability of these cells to produce InsPs is confirmed by stimulation with AlF3. The values corresponded to the means ± S.E. of three independent experiments, each performed in triplicate.

NTS2-induced p42/p44^mapk Phosphorylation-- Stimulation of granule cells in culture with 0.1 µM NT for periods ranging between 1 and 60 min induced a sustained increase in p42/p44^mapk phosphorylation. This increase was first detectable 5 min after NT application and was maintained for over 60 min thereafter (Fig. 10A, n = 4). This effect on MAPK activation was also observed at lower doses of NT (10⁸ M) (Fig. 10B, n = 3). Densitometric analyses of phosphorylated p42/p44^mapk over total p42/p44^mapk ratios indicated that NT induced a robust increase in MAPK phosphorylation that reached a plateau at ~30 min of incubation (Fig. 10F). Similar p42/p44^mapk activation was observed following 10 and 30 min of stimulation with 1 µM levocabastine, confirming the implication of NTS2 in the MAPK activation (Fig. 10C, n = 3). By contrast, under the same conditions, stimulation with 0.1 µM (Fig. 10, D and F, n = 3) or 1 µM (Fig. 10, E and F, n = 2) SR48692 failed to modify the level of p42/p44^mapk phosphorylation.

View larger version (47K): [in this window] [in a new window]   Fig. 10.   p42/p44mapk phosphorylation following activation of NTS2 receptors endogenously expressed in cultured cerebellar granule cells. A, NT (0.1 µM) induces a time-dependent and sustained increase in MAPK phosphorylation (upper panel) that appears after 5 min of incubation and is maintained for over 60 min after NT application. The lower panel corresponds to Western blotting of total p42/p44mapk in the same samples. B, the same effect on MAPK activity is observed at lower doses of NT (108 M). C, similar effects are observed following 10 and 30 min of stimulation with 1 µM levocabastine, suggesting that p42/p44mapk are activated through stimulation of NTS2 receptors. D, under the same conditions, SR48692 (107 M) fails to modify the basal level of MAPK phosphorylation. E, similarly, no modification of the MAPK phosphorylation pattern is observed following stimulation with 1 µM SR48692. F, comparative densitometric analysis of the ratio of phosphorylated p42/p44mapk over total p42/p44mapk levels following incubation with 0.1 µM NT (circles, n = 4), 0.1 µM SR48692 (squares, n = 3), or 1 µM SR48692 (triangles, n = 2). The values are the means ± S.E. of n experiments as indicated above. *, p  0.05; **, p  0.02 as compared with controls (C).

[³H]NT Binding to Cerebellar Synaptosomes-- To determine whether NTS2-binding sites were also associated with the terminal arbor of cerebellar granule cells, synaptosomal preparations from adult rat cerebelli were incubated with increasing concentrations of [³H]NT for 25 min at 37 °C. Electron microscopy confirmed that these synaptosomes were structurally well preserved and were largely comprised of synaptic terminals endowed with small clear vesicles (Fig. 11A). The affinity (K_d = 2.4 ± 0.5 nM; n = 2) and maximal capacity (B_max = 60.1 ± 2.7 fmol/mg) of [³H]NT binding to these synaptosomes were similar to those measured in cultured granule cells (compare Figs. 11B and 4A). Furthermore, competition experiments showed that the NTS2 ligand, levocabastine, displaced this specific [³H]NT binding with an IC₅₀ of 16.5 ± 1.3 nM (Fig. 11C, n = 2), i.e. with the same affinity as in cultured granule cells (Table I). NT and SR48692 also competed for [³H]NT binding with IC₅₀ values (6.5 ± 0.2 nM and 142 ± 7.8 nM, respectively) similar to those obtained on membranes from cultured cerebellar granule cells (Fig. 11C and Table I).

View larger version (49K): [in this window] [in a new window]   Fig. 11.   Pharmacological profile of [3H]NT binding to synaptosomes from adult rat cerebellum. A, electron microscopic assessment of synaptosomal integrity. Three well preserved axon terminals (AT) show numerous clear synaptic vesicles as well as several large dense cored vesicles (arrows). Scale bar, 0.6 µm. B, saturation experiments. The incubations were carried out at 37 °C for 25 min with increasing concentrations of [3H]NT (0.5-10 nM). C, competition of [3H]NT binding to rat cerebellar synaptosomes by unlabeled NT (open squares), levocabastine (closed squares), and SR48692 (open circles). The IC50 values are listed in Table I. D, association kinetics of [3H]NT binding to synaptosomal preparations. The synaptosomes were incubated with 3 nM [3H]NT for increasing time intervals at 37 °C, and nonspecific binding was determined in the presence of an excess of 1 µM unlabeled NT. The proportion of internalized ligand was obtained by comparing association curves before (open squares) and after (closed squares) hypertonic acid treatment. The values corresponded to the means ± S.D. from two independent experiments performed in triplicate.

To determine whether specifically bound [³H]NT was internalized within cerebellar synaptosomes, the synaptosomal preparations were incubated for 2-20 min at 37 °C with 3 nM [³H]NT, and the proportion of internalized [³H]NT was assessed following dissociation of surface-bound ligand by hypertonic acid wash (Fig. 7D, n = 2). Association kinetics of [³H]NT to whole synaptosomes at 37 °C reached a plateau value between 5 and 10 min (Fig. 11D). As opposed to what was observed on cultured granule cells (Fig. 4C), specific [³H]NT binding was entirely strippable by hypertonic acid wash, indicating that the bound molecules were confined to the cell surface (Fig. 11D).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The present study reveals that NTS2 neurotensin receptors are associated with somatodendritic and terminal arbors of rat cerebellar granule cells. It also provides the first demonstration that NTS2 is a functional receptor in neurons and that both NT and levocabastine may act as agonists at this site.

Selective Expression of NTS2 Receptors by Cerebellar Granule Cells-- Receptor autoradiography demonstrated high concentrations of ¹²⁵I-NT-binding sites in sections of rat cerebellum. Virtually no residual ¹²⁵I-NT labeling was observed in the presence of levocabastine, confirming the lack of NTS1 receptors in this structure (6, 8, 31). Radiolabeled NTS2 receptors were most conspicuous in the granule cell layer, in keeping with earlier in situ hybridization data, which demonstrated high expression of NTS2 mRNA in granule cells of mouse cerebellar cortex (10). Relatively dense autoradiographic labeling was also observed in the molecular layer. This labeling most likely corresponds to presynaptic NTS2 receptors associated with granule cell axons (parallel fibers) because neither ¹²⁵I-NT-labeling (present study) nor NTS2 mRNA (10) were detected over Purkinje cells, the dendrites of which form the bulk of the molecular layer. This interpretation is also supported by the demonstration of presynaptic NTS2 receptors in synaptosomes prepared from adult rat cerebellum (see below). Nonetheless, the possibility that a small proportion of NTS2 receptors labeled in this layer might be associated with stellate or basket cells cannot be completely excluded because sparse interneurons were reported to express NTS2 mRNA in mouse cerebellar cortex (10).

RT-PCR analysis of cerebellar granule cell cultures revealed the presence of mRNA for the two NTS2 receptor variants previously detected in whole brain extracts (23). Although the present experiments did not allow us to determine whether NTS2 mRNA was of neuronal or glial origin, the fact that only the larger form of NTS2 receptor mRNA was detected in astrocytes cultured from rat cerebral cortex (14) suggests that the shorter form of the NTS2 receptor may be selectively neuronal. Furthermore, fluorescence imaging experiments using fluo-NT and specific neuronal or glial markers demonstrated that in our culture preparations, NTS2 receptor proteins were only present in neurons.

In keeping with the results of autoradiographic binding data, no NTS1 mRNA was detected by RT-PCR in mixed cerebellar cultures. However, NTS3/sortilin mRNA was detected in this preparation, suggesting that NTS3 receptors are present in cerebellar granule and/or glial cells. The paucity of NTS3 receptor mRNA detected by in situ hybridization over granule cells in sections of adult rat cerebellum² suggests that the NTS3 message detected here by RT-PCR is mainly of glial origin.

Binding experiments carried out on cerebellar granule cell cultures demonstrated saturable ¹²⁵I-NT binding to a single apparent population of sites. The affinity of these sites for the radiolabeled ligand was very close to that reported previously for NTS2 receptors transfected in either COS-7 (7, 22), HEK 293 (17), or Chinese hamster ovary (16) cells. It also conformed to that of levocabastine-sensitive NT binding measured in rat brain (6, 32). Specific ¹²⁵I-NT binding was entirely competed for by the NTS2 ligand, levocabastine, suggesting that if translated in granule cell cultures, the NTS3 receptor was not recognized by the radioactive ligand. This lack of recognition by NTS3 is consistent with the reported low affinity of this site for NT in heterologous transfection systems (40 nM) (33). The IC₅₀ of levocabastine in granule cell cultures (5.7 ± 0.5 nM) was similar to that reported in epithelial cells transfected with either rat (22) or mouse (7, 17) NTS2. However, it was markedly lower than that reported in Chinese hamster ovary cells transfected with human NTS2 (100 nM; 16). ¹²⁵I-NT binding to cerebellar granule cells was also competed for by the NTS1 antagonist SR48692, with an affinity comparable with that observed on recombinant NTS2 receptors (7, 16, 17, 22). Taken together, these results indicate that the sole NT receptor recognized by exogenous NT in mixed cerebellar cultures corresponds to the NTS2 subtype and that this receptor is selectively expressed by granule cells. Not all granule cells appear to express NTS2 receptors, however, because only 60% of cultured neurons, identified using the neuronal marker N52, were found to bind fluo-NT in dual labeling experiments.

Ligand-induced Internalization of Neuronal NTS2 Receptors-- Following incubation of cultured cerebellar granule cells with either ¹²⁵I-NT or fluo-NT, ~65% of specifically bound ligand was resistant to hypertonic acid wash, indicating that it had been internalized. These results are congruent with those obtained in cells transfected with cDNA encoding the mouse NTS2 receptor (17) but differ from those obtained on cortical astrocytes endogeneously expressing the rat receptor (34). NT internalization was blocked by hypertonic sucrose, phenylarsine oxide, and monodansylcadaverine, suggesting that it proceeded via clathrin-coated pits (for review see Ref. 28). Accordingly, confocal microscopy revealed that the internalized ligand was first concentrated within small endosome-like organelles, a pattern consistent with earlier reports of receptor-mediated internalization of fluorescent peptides in neurons (34-36). At longer time intervals, internalized fluo-NT was clustered deeper within the cell, in the juxta-nuclear region. Double labeling experiments identified this late targeting compartment as being syntaxin 6-positive, i.e. as corresponding to the TGN/pericentriolar recycling endosome (29, 37, 38). Although trafficking of internalized ligand to the TGN has previously been demonstrated in cells transfected with the NTS1 receptor subtype (29), the present data provide the first demonstration that an internalized neuropeptide may be targeted to the TGN in neurons. More broadly, the finding that NT internalizes via native NTS2 receptors suggests that NT acts as an agonist on this receptor.

Functional Coupling of Neuronal NTS2 Receptors-- Despite their high affinity for the NTS2 receptor, neither NT nor levocabastine were able to induce Ca²⁺ mobilization in cerebellar granule cells loaded with the Ca²⁺-sensitive fluorescent dye, Fluo-4. A similar lack of Ca²⁺ response had been observed following application of the same drugs onto cells transfected with the human NTS2 receptor (16). By contrast, both NT and levocabastine were reported to elicit Ca²⁺ currents in frog oocytes transfected with the mouse NTS2 receptor or in Chinese hamster ovary cells transfected with the rat NTS2 receptor (7, 15, 30). The present results therefore suggest that endogenously and ectopically expressed NTS2 receptors are differentially coupled to Ca²⁺-activating systems.

As previously reported in cells transfected with either mouse (15), rat (30), or human (16) NTS2 receptors, stimulation with SR48692 induced a dose-dependent Ca²⁺ mobilization in cerebellar granule cells. However, in contrast to what had been reported in heterologous transfection systems, this Ca²⁺ response was blocked by neither NT nor levocabastine, suggesting that in cerebellar granule cells, the effect of SR48692 on Ca²⁺ mobilization may not be mediated through NTS2 receptors. The mechanisms by which SR48692 exerts these Ca²⁺ mobilizing effects remain to be elucidated. Our results demonstrate that they involve Ca²⁺ mobilization from both intracellular stores and extracellular medium and therefore that they implicate more than one effector. The nature of the mobilized intracellular stores is unknown, although the similarity in the Ca²⁺ release profiles elicited by stimulation with thapsigargin and with SR48692 in the absence of extracellular Ca²⁺ suggests that it may correspond to the thapsigargin-sensitive pool described previously in cerebellar granule cells (39, 40). Clearly, intracellularly released Ca²⁺ does not originate from an InsPs-sensitive pool, because none of the NTS2 ligands tested (NT, levocabastine, or SR48692) elicited an increase in the concentration of InsPs in cerebellar granule cells. This result differs from those obtained in epithelial cells transfected with the human NTS2, in which stimulation with SR48692, but not with NT nor levocabastine, induced an increase in inositol phosphate turnover (16).

By contrast, stimulation of cultured cerebellar granule cells with both NT and levocabastine, but not with SR48692, induced a sustained activation of the mitogen-activated kinase p42/p44. Again, this result is at odds with the reported effects of human NTS2 receptor stimulation in transfected Chinese hamster ovary cells, in which stimulation with SR48692, but with neither NT nor levocabastine, was found to activate the p42/p44^mapk cascade (16). These differential effects may be due to differences in the structure of the C-terminal tail of the rat versus the human NTS2 receptor or to differential coupling of endogenously versus ectopically expressed NTS2 receptors. In any event, the present results provide the first demonstration that neuronal NTS2 receptors are functional and, together with our internalization data, indicate that NT is an agonist on these receptors.

Presynaptic Localization of Central NTS2 Receptors-- To determine whether, as suggested by our autoradiographic binding data, cerebellar NTS2 receptors were expressed presynaptically and to characterize the pharmacological properties of these putative presynaptic sites, binding and internalization assays were carried out on synaptosomes freshly prepared from adult rat cerebellar cortex. Binding experiments demonstrated that in these preparations, radiolabeled NT bound to the same apparent population of sites as in cerebellar granule cells in culture. The NT affinity and drug sensitivity of these sites were also similar to those observed in cerebellar granule cells, indicating that they correspond to NTS2 receptors. However, contrary to our observations in cultured granule cells, specific [³H]NT binding was entirely strippable by hypertonic acid wash, indicating that surface-bound ligand was not internalized. This result cannot be attributed to the type of preparation used, because synaptosomes similarly prepared from rat neostriatum were found to internalize [³H]NT in an NTS1-dependent manner (41). Therefore, it appears that presynaptic NTS2 receptors, in contrast to somatodendritic NTS2 ones, do not internalize in response to ligand binding. This difference may be due to interaction of NTS2 receptors with distinct internalization-related proteins (e.g. receptor kinases, -arrestins, amphiphysins, etc.) in somatodendritic versus axonal domains.

In conclusion, the present results demonstrate that, in rat cerebellum, NTS2 receptors are localized in granule cells and are functionally coupled to the p42/p44^mapk cascade and not to the phospholipase C/Ca²⁺ pathway. Recent studies have shown that stimulation of the MAPK pathway could involve binding of -arrestins to the G protein receptor kinase-phosphorylated receptor and hence be linked to ligand-induced receptor internalization (for review see Ref. 42). Further studies will be needed to determine whether the NTS2-induced activation of the MAPK pathway reported here occurs in response to -arrestin binding or as a result of G protein activation. In any event, the protracted time course of the NTS2 response suggests that cerebellar NTS2 receptors are associated with long term metabolic effects rather than with rapid synaptic-like actions.

	ACKNOWLEDGEMENTS

We thank Lyne Bilodeau, Lucie Chouinard, Thomas Stroh, and Mariette Houle for technical assistance and Naomi Takeda for secretarial help in the preparation of this manuscript.

	FOOTNOTES

^* This work was supported by Canadian Institutes of Health Research Grants MT-7366 (to A. B.) and MT-13679 (to M. D. P. and N. G. P.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ Recipient of Ligue Nationale Contre le Cancer Research and Fonds de la Recherche en Santé du Québec fellowships.

Recipient of a studentship from Fonds de la Recherche en Santé du Québec.

Recipient of a studentship from the Ministère de l'Education du Québec.

^§§ To whom correspondence should be addressed: Dept. of Neurology and Neurosurgery, Montreal Neurological Institute, 3801 University St., Montreal, PQ H3A 2B4, Canada. Tel.: 514-398-1913; Fax: 514-398-5871; E-mail: alain.beaudet@mcgill.ca.

Published, JBC Papers in Press, June 25, 2002, DOI 10.1074/jbc.M202586200

² P. Sarret, P. Krzywkowski, M. S. Nielsen, C. M. Petersen, J. Mazella, T. Stroh, and A. Beaudet, unpublished data.

	ABBREVIATIONS

The abbreviations used are: NT, neurotensin; InsPs, inositol phosphate; MAPK, mitogen-activated protein kinase; ¹²⁵I-NT, ¹²⁵I-Tyr³-neurotensin; BSA, bovine serum albumin; RT, reverse transcription; PBS, phosphate-buffered saline; GFAP, glial fibrillary acidic protein; TGN, trans-Golgi network.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				J H Li, F Sicard, M A Salam, M Baek, J LePrince, H Vaudry, K Kim, H B Kwon, and J Y Seong Molecular cloning and functional characterization of a type-I neurotensin receptor (NTR) and a novel NTR from the bullfrog brain J. Mol. Endocrinol., June 1, 2005; 34(3): 793 - 807. [Abstract] [Full Text] [PDF]

				A. Perron, P. Sarret, L. Gendron, T. Stroh, and A. Beaudet Identification and Functional Characterization of a 5-Transmembrane Domain Variant Isoform of the NTS2 Neurotensin Receptor in Rat Central Nervous System J. Biol. Chem., March 18, 2005; 280(11): 10219 - 10227. [Abstract] [Full Text] [PDF]

				L. Gendron, A. Perron, M. D. Payet, N. Gallo-Payet, P. Sarret, and A. Beaudet Low-Affinity Neurotensin Receptor (NTS2) Signaling: Internalization-Dependent Activation of Extracellular Signal-Regulated Kinases 1/2 Mol. Pharmacol., December 1, 2004; 66(6): 1421 - 1430. [Abstract] [Full Text] [PDF]

				E. Y. Tan, M. Mujoomdar, and J. Blay Adenosine Down-Regulates the Surface Expression of Dipeptidyl Peptidase IV on HT-29 Human Colorectal Carcinoma Cells: Implications for Cancer Cell Behavior Am. J. Pathol., July 1, 2004; 165(1): 319 - 330. [Abstract] [Full Text] [PDF]

				F. St-Gelais, M. Legault, M.-J. Bourque, P.-P. Rompre, and L.-E. Trudeau Role of Calcium in Neurotensin-Evoked Enhancement in Firing in Mesencephalic Dopamine Neurons J. Neurosci., March 10, 2004; 24(10): 2566 - 2574. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 277/39/36233    most recent M202586200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sarret, P. Articles by Beaudet, A. Search for Related Content PubMed PubMed Citation Articles by Sarret, P. Articles by Beaudet, A. Social Bookmarking                      What's this?