Nicotinic Binding in Rat Brain: Autoradiographic Comparison of [3H]Acetylcholine, [3H]Nicotine, and [‘251]-~-Bungarotoxin1

Three radioligands have been commonly used to label putative nicotinic cholinoceptors in the mammalian central nervous system: the agonists [3H]nicotine and [3H]acetylcholine (C3H]ACh-in the presence of atropine to block muscarinic receptors), and the snake venom extract, [1251]-a-bungarotoxin([‘251]BTX), which acts as a nicotinic antagonist at the neuromuscular junction. Binding studies employing brain homogenates indicate that the regional distributions of both [3H]nicotine and [3H]ACh differ from that of [1251]BTX. The possible relationship between brain sites bound by [3H]nicotine and [3H]ACh has not been examined directly. We have used the technique of autoradiography to produce detailed maps of [3H]nicotine, r3H]ACh, and [‘251]BTX labeling; nearadjacent tissue sections were compared at many levels of the rat brain. The maps of high affinity agonist labeling are strikingly concordant, with highest densities in the interpeduncular nucleus, most thalamic nuclei, superior colliculus, medial habenula, presubiculum, cerebral cortex (layers I and III/IV), and the substantia nigra pars compacts/ventral tegmental area. The pattern of [‘251]BTX binding is strikingly different, the only notable overlap with agonist binding being the cerebral cortex (layer I) and superior colliculus. [‘251]BTX binding is also dense in the inferior colliculus, cerebral cortex (layer VI), hypothalamus, and hippocampus, but is virtually absent in thalamus. Various lines of evidence suggest that the high affinity agonist-binding sites in brain correspond to nicotinic cholinergic receptors similar to those found at autonomic ganglia; BTX-binding sites may also serve as receptors for nicotine and are possibly related to neuromuscular nicotinic cholinoceptors. The concept of receptors for nicotine is an old one (Langley, 1905). In the peripheral nervous system, two types of receptor, namely, ganglionic (C-6) and neuromuscular (C-IO), have been distinguished on the basis of antagonist selectivity (Paton and Zaimis, 1951). The existence of functional nicotine receptors in the central nervous system (CNS) is strongly suggested by various experimental approaches-electrophysiological (Krnjevic, 1975) biochemical (Giorguieff et al., 1979) physiological (Armitage and Hall, 1967) and behavioral (Clarke and Kumar, 1983). Received August 3, 1984; Revised November 12, 1984; Accepted November 12, 1984 ’ We gratefully acknowledge the excellent typing skills of June Zuranski. P. B. S. C. is a Fogarty Visiting Fellow. ‘To whom correspondence should be sent, at Btological Psychiatry Branch, National Institute of Mental Health, Building 10, Room 3N212, Bethesda, MD 20205. Putative central nicotinic receptors have been labeled using various ligands with known peripheral actions. d-Tubocurarine, a-bungarotoxin (BTX), and Naja-naja siamensis a-toxin all block nicotiniccholinergic transmission at the neuromuscular junction and bind to rat brain membranes with high affinity (Speth et al., 1977; Nordberg and Larsson, 1980; Schmidt et al., 1980). Binding of [“51]BTX, the most widely used radioligand, is saturable and reversible, and is displaced preferentially by nicotinic agents (Schmidt et al., 1980); hence, the BTX-binding site has been termed “nicotinic,” and the displacement potency of nicotine in vitro suggests that nicotine could act at BTX sites in viva. More recently, the high affinity binding of agonists to putative nicotinic cholinoceptors in rodent brain membranes has also been demonstrated. Studies have employed either [3H]nicotine (Yoshida and Imura, 1979; Roman0 and Goldstein, 1980; Marks and Collins, 1982; Costa and Murphy, 1983) or [3H]acetylcholine ([3H]ACh) in the presence of excess atropine to block muscarinic receptors (Schwartz et al., 1982). Binding of either ligand was potently displaced by nicotinic agonists, and both ligands had an affinity constant in the nanomolar range. There is no clear consensus as to the regional distribution of [3H]nicotine-binding sites in brain (Yoshida and Imura, 1979; Marks and Collins, 1982; Costa and Murphy, 1983). Marks and Collins (1982) found a lack of correlation between the regional distributions of [3H]nicotine and [“‘I]BTX binding in mouse brain. A comparison of nicotinic [3H]ACh binding and [‘*?I BTX binding in rat brain reached the same conclusion (Schwartz et al., 1982). Clearly, nicotinic agonists and BTX do not always label the same molecule in brain. Recently, we have used the autoradiographic method of Herkenham and Pert (1982) to visualize [3H]-DL-nicotine binding to unfixed rat brain sections (Clarke et al., 1984). The binding was displaced by the unlabeled isomers of nicotine in a stereoselective manner; natural L-nicotine was approximately 17 times more potent than Dnicotine. The pharmacological displacement profile and binding affinity constant for [3H]nicotine were similar to those previously found in membrane studies. The pattern of labeling was highly discrete and differed from that previously reported for BTX (Hunt and Schmidt, 1978; Segal et al., 1978). In addition, quantitative autoradiography recently has been used to identify nicotinic binding sites labeled by [3H]ACh in rat brain slices (Rainbow et al., 1984). The kinetic constants and pharmacologic profile were similar to those obtained using brain homogenates (Schwartz et al., 1982). Again, the pattern of labeling was unlike that reported for BTX (Hunt and Schmidt, 1978; Segal et al., 1978). In the study described below, autoradiographic maps of [3H]ACh, [3H]nicotine, and [“51]BTX were obtained from near-adjacent sections of rat brain; the resulting comparisons reveal that [3H]ACh and [3H]nicotine bind with a strikingly similar pattern which shows remarkably little overlap with [‘*?I BTX labeling. Materials and Methods Slide-mounted rat brain sections were prepared as previously described (Herkenham and Pert, 1982). Three male Sprague-Dawley rats (250 gm, 1308 Clarke et al. Vol. 5, No. 5, May 1985 Taconic Farms, NY) were used. Two brains provided coronal sections; the third provided sagittal sections. Following decapitation, the brains were rapidly removed, immersed in isopentane at -35°C and mounted on cryostat chucks. Cutting was performed at -16°C. At intervals through each brain, groups of three adjacent or near-adjacent sections were taken. Such a group never spanned more than 150 pm in the anterior-posterior direction. Each section in a group of three was transferred to a different subbed microscope slide and eventually was incubated with a different ligand. Sections allocated to [3H]nicotine or [3H]ACh were 24 pm thick. For incubation with [‘251]BTX, coronal sections from one rat were cut at 24 pm to permit direct comparisons with the autoradiographs of tritiated ligand binding; the remaining coronal and sagittal secttons were cut at 12 pm in order to improve image definition obtained with the iodinated ligand. Sections were thaw-mounted on precleaned gelatin-coated slides, dried overnight under partial vacuum at 4°C and stored with dessicant at -70°C for at least 24 hr. For autoradiography, the following general method was employed: slides were preincubated, Incubated with radioligand (with or without the appropriate displacing drug to define nonspecific binding), rinsed in several changes of buffer, and rapidly dried under a stream of compressed air. Slides were left at room temperature and pressure in a dessicating jar for 24 hr and then juxtaposed tightly against tritium-sensitive film (Ultrofilm; LKB Instruments, Gaithersburg, MD) and stored at room temperature for several weeks in xray cassettes (Wolf: CGR Scientific, Columbia, MD). Films were processed in Kodak Dl9 developer at 22°C for 5 mm and fixed for 4 min with Kodak Rapid Fixer (wrthout hardener). The tissue sections were then demyelinated (defatted) through aqueous ethanol and xylene, before staining with thionin. Structures were identified by comparing stained sectrons with enlarged photographs of corresponding autoradiographic images. The exact conditions employed were optimized for each ligand. For [3H] nicotine autoradiography, there was no preincubation step, and sectrons were incubated at room temperature for 20 min in 50 mM Tris-HCI (pH 7.4; Sigma Chemical Co., St. Louis, MO), 8 mM CaCl*, and 3.5 nM [3H]pL-nicotine (N-methy/-3H, 71.2 Ci/mmol, New England Nuclear, Boston, MA). Nonspecific binding was assessed in the presence of 10 PM L-nicotine bitartrate (BDH Chemicals, Ltd., Poole, England). Following incubation, the sections were immediately given four rinses of 30 set each in buffer as described above, pH 7.4, at 4°C. Storage in cassettes lasted 12 weeks. For [3H]ACh autoradiography a modification of the procedure recently reported by Rainbow et al. (1984) was used. Sections were preincubated for 15 min at room temperature in 50 mM Tris-HCI buffer (pH 7.4) containing 1.5 PM atropine sulfate, 120 mM NaCI, 5 mM KCI, 1 mM MgCl*, and 2 mM CaCl*. A second preincubation of 10 min duration followed in buffer of the same composition but with 100 FM di-iosopropyl fluorophosphate added. Sections were then incubated for 1 hr at 4°C in buffer containing 10 nM [3H]ACh (80 Ci/mmol, synthesized as described by Schwartz et al., 1982); nonspecrfic binding was assessed in the presence of carbachol (100 PM). Following incubation, the sections were given two rinses, each of 2 min, in 500 ml of buffer at 4°C. Finally, the sections were briefly rinsed in distilled water (2 set) and dried. Storage in cassettes was for 5 weeks. For [‘“?]BTX binding, the sections were preincubated for 30 min at room temperature in 50 mM Tris-HCI (Sigma; pH 7.4) and 1 mg/ml of bovine serum albumin (Sigma). Incubation was 2 hr duration, carried out at room temperature, using the same buffer composition but including [‘251]BTX (50 to 61 Ci/mmol, New England Nuclear, used within 1.5 half-lives of synthesis). Prior to [“‘I]BTX autoradiography, Scatchard and structure-activity analyses were performed using tissue sections which were processed as for autoradiography but which were scraped into tubes for gamma counting (Beckman 4000 counter; 60% efficiency) after drying. In these biochemical experiments, coronal sections (24 pm thick) were taken from a forebrain regron with a minimal longitudinal gradient of binding (interaural 10.5 to 8.5, according to Paxinos and Watson, 1982). The structure-activity study served to verify that the displacement profile was nicotinic in tissue sectrons, as In homogenates; this was considered particularly important, since the pharmacological properties of commercrally supplied BTX can vary according to the source (Costa et al., 1983). In the Scatchard experiment, six concentrations of [‘251]BTX were used, ranging from 0.18 nM to 7.8 nM. The structure-activity and autoradiographic studies utilized a concentration of 1.4 nM. Nonspecific binding was assessed by the addition of L-nicotine bitartrate (1 mM) to the pretncubation and incubatron media. Sections were given six rinses of 30 min each in 500 ml of buffer, pH 7.4, and 4°C. Storage in cassettes was for