Abstract
The binding of [3H]epibatidine, an alkaloid isolated from the skin of an Ecuadorean tree frog, was measured both in brain regions dissected from mouse brain and in tissue sections. Binding to each of 12 brain areas was saturable, but apparently monophasic; no indication of multiple binding sites was obtained. However, inhibition of epibatidine binding by nicotine, acetylcholine, methylcarbachol and cytisine in olfactory bulbs revealed a biphasic pattern consistent with the presence of two sites differentially sensitive to inhibition by these nicotinic agonists. Cytisine displayed the greatest difference in inhibitory potency between the two apparent sites. Subsequent analysis of the inhibition of epibatidine binding by cytisine in membranes prepared from 12 brain areas also suggested the presence of two sites in each brain region. The estimated potency of cytisine at each site was similar in each brain region. However, the proportion of [3H]epibatidine binding sites that were more sensitive to inhibtion by cytisine and those sites less sensitive to inhibition by this agonist varied markedly among the brain regions. Quantitative autoradiographic analyses of mouse brain revealed pattern of [3H]epibatidine binding sites less sensitive to inhibition by cytisine that differed markedly from the pattern obtained with [3H]nicotine. Among brain regions demonstrating substantial sites less sensitive to cytisine inhibition were the accessory olfactory nucleus, medial habenula, interpeduncular nucleus, fasciculus retroflexus, superior colliculus, inferior colliculus and the pineal gland. The results indicate that epibatidine binds to at least two distinct nicotinic sites in mouse brain that may represent different nicotinic receptor subtypes, one of which appears to be identical to that measured by the binding of other agonists such as nicotine or cytisine.
Nicotinic cholinergic receptors in the central nervous system can generally be grouped into two large classes: those that are sensitive to inhibition by low concentrations of α-bungarotoxin and those that are resistant to the effects of this toxin (Lindstrom, 1995). Until recently, either radiolabeled α-bungarotoxin (Clarke, 1992; Oswald and Freeman, 1981) or radiolabeled nicotinic agonists [nicotine (Romano and Goldstein, 1980; Marks and Collins, 1982), cytisine (Pabreza et al., 1991), methylcarbachol (Abood and Grassi, 1986; Boska and Quirion, 1987), and acetylcholine (Schwartz et al., 1982)] have been used to measure nicotinic binding sites in brain. α-Bungarotoxin binds to different receptors than do the agonists (Clarke et al., 1985; Marks et al., 1986) although the commonly used agonists appear to label the same sites (Clarke et al., 1985; Marks et al., 1986; Martino-Barrows and Kellar, 1987;Pabreza al., 1991).
The identification and demonstration of differential distribution of the mRNA for nicotinic subunits (Boulter et al., 1986;Goldman et al., 1987; Wada et al., 1988; Boulteret al., 1990; Schoepfer et al., 1990; Couturieret al., 1990; Seguela et al., 1993; Elgoyhenet al., 1994; Deneris et al., 1988, 1989;Duvoisin et al., 1989) implies that many different nicotinic receptors may exist in brain. The alpha-4, beta-2 and alpha-7 appear to be the most widely expressed nicotinic receptor genes in mouse and rat brain (Wada et al., 1989;Seguela et al., 1993; Marks et al., 1992, 1996). Those receptors that bind α-bungarotoxin apparently containalpha-7 receptor subunits (Schoepfer et al., 1990; Seguela et al., 1993), although those receptors that bind agonists with high affinity are predominantlyalpha-4/beta-2 (Whiting and Lindstrom, 1988;Flores et al., 1992). Indeed, beta-2 null mutant mice have no high affinity [3H]nicotine binding and lack functional responses in most thalamic nuclei (Picciottoet al., 1995). However, those receptors containing subunits encoded by the other nicotinic genes probably cannot be easily measured with the commonly used ligands, although [3]nicotine binding has been detected in PC12 cells (Madhok and Sharp, 1992) and [3H]acetylcholine binding has been measured in TE671 cells (Lukas, 1990). Although the relative quantity of mRNA encoding the nicotinic receptor subunits suggests that high affinity agonist binding and α-bungarotoxin binding may label most of the nicotinic receptors in the brain, the unique distribution of the mRNAs encoding the nicotinic receptor subunits suggests that different functional nicotinic receptors may be assembled in discrete locations throughout the brain (Wada et al., 1989, 1990; Deneriset al., 1988, 1989; Duvoisin et al., 1989;Dinely-Miller and Patrick, 1992; Seguela et al., 1993; Markset al., 1992, 1996) and may not be readily detected using classical binding techniques. However, receptors containing α2 and β2 subunits bind nicotine with sufficient affinity to be readily measurable in chick brain (Whiting et al., 1987), but the relative rarity of this subtype in mammalian brain makes detection difficult.
The alkaloid, epibatidine, isolated from the skin of the Ecuadorean tree frog, Epipedobates tricolor (Spande et al., 1992), is a potent nicotinic agonist. Animals administered low doses of epibatidine display mecamylamine-sensitive responses commonly observed for other nicotinic agonists including hypomotility, hypothermia and antinociception (Badio and Daly, 1994; Qian et al., 1993;Sullivan et al., 1994; Bonhaus et al., 1995;Damaj et al., 1994). Epibatidine is also an extremely potent inhibitor of high affinity agonist binding (Badio and Daly, 1994; Qianet al., 1993; Sullivan et al., 1994; Bonhauset al., 1995; Damaj et al., 1994; Gerzanichet al., 1995) and exhibits agonist properties in several different assays for nicotinic receptor function (Fisher et al., 1994; Bonhaus et al., 1995; Alkondon and Albuquerque, 1995; Gerzanich et al., 1995; Sullivan et al., 1994; Marks et al., 1996; Albuquerque et al., 1997). In addition, radiolabeled epibatidine binds to rat brain membranes with high affinity and the curvilinear Scatchard plots observed suggest multiple binding sites (Houghtling et al., 1995). Autoradiographic analysis of [3H]epibatidine binding to rat brain is also consistent with the presence of at least two sites inasmuch as [3H]epibatidine binds to receptor sites in addition to those labeled with [3H]cytisine (Perry and Kellar, 1995). Immunoprecipitation of solubilized nicotinic receptors indicates that epibatidine labels two populations of receptors in rat trigeminal ganglion: one containing α4 and β2 subunits, identical to the sites labeled by high affinity agonist binding in brain, and a second containing α3 and β4 subunits (Flores et al., 1996). These results suggest that epibatidine may be very useful in the measurement of nicotinic receptors that cannot be reliably measured with either other nicotinic agonists or α-bungarotoxin.
In our study [3H]epibatidine binding to mouse brain was evaluated. The results are consistent with the postulate that [3H]epibatidine does, indeed, bind to several different nicotinic receptor subtypes and that differential sensitivity to inhibition by cytisine distinguishes two binding sites in mouse brain.
Materials and Methods
Mice.
Male mice of the strains C57BL/6J and DBA/2JIbg were 60 to 90 days old when used in these studies. Mice were bred at the Institute for Behavioral Genetics and housed five per cage. Animals were permitted free access to food and water. The vivarium was maintained on a 12-hr light/12-hr dark cycle (lights on 7a.m. to 7 p.m.). All procedures used in these studies were approved by the Animal Care and Utilization Committee of the University of Colorado, Boulder.
Materials.
(±)-[3H]Epibatidine (48 Ci/mmol) and l-[3H]nicotine (78 Ci/mmol), were purchased from Du Pont NEN (Boston, MA). α-[125I]Bungarotoxin (Initial specific activity = 220 Ci/mmol) plastic tritium standards and Hyperfilm-3H were purchased from Amersham (Mount Prospect, IL). NaCl, KCl, MgSO4, CaCl2, gelatin, chromium aluminum sulfate, cytisine, acetylcholine and diisopropyfluorophosphate were obtained from Sigma Chemical Co. (St. Louis, MO). Methylcarbachol chloride, (+)-epibatidine tartrate, and (-)-epibatidine tartrate were obtained from RBI (South Natick, MA). Nicotine bitartrate was a product of BDH Chemicals (Poole, England). Glass fiber filters Type A/E were obtained from Gelman Sciences (Ann Arbor, MI) and Type GB from MFS (Dublin, CA). Budget Solve scintillation fluid was obtained from RPI (Arlington Heights, IL).
Tissue preparation.
Each C57BL/6 mouse was killed by cervical dislocation, the brain was removed from the skull and placed on an ice-cold platform. The following 12 brain regions were dissected: olfactory bulbs, cerebellum, hindbrain (pons-medulla), hypothalamus, hippocampus, striatum, cerebral cortex, thalamus, midbrain, interpeduncular nucleus, superior colliculus and inferior colliculus. Samples were homogenized in ice-cold hypotonic buffer (NaCl, 14.4 mM; KCl, 0.2 mM; CaCl2, 0.2 mM; MgSO4, 0.1 mM, HEPES, 2.0 mM; pH = 7.5) using a glass-Teflon tissue grinder. The particulate fraction was obtained by centrifugation at 20,000 × g for 20 min in a Sorvall RC-2B centrifuge. The pellet was resuspended in fresh homogenization buffer, incubated at 37°C for 10 min, and harvested by centrifugation. Each sample was washed twice more by resuspension and centrifugation and stored as a pellet under homogenization buffer at -70°C until use.
[3H]nicotine binding.
The binding of [3H]nicotine was measured using a modification of the method of Marks et al. (1986). Samples (50-200 μg, depending on brain region) were incubated in 96-well polystyrene plates with 20 nM [3H]nicotine at 22°C for 30 min in 100 μl of binding buffer (NaCl, 144 mM; KCl, 1.5 mM, CaCl2, 2 mM; MgSO4, 1 mM; HEPES, 20 mM; pH = 7.5). The binding reaction was terminated by filtration of the samples onto glass fiber filters (MFS GB top, Gelman A/E bottom) that had been soaked in binding buffer containing 0.5% polyethylenimine using an Inotech Cell Harvester (Inotech, East Lansing, MI). Samples were subsequently washed six times with ice-cold binding buffer. Nonspecific binding was determined by including 10 μMl-nicotine in the assay.
α-[125I]bungarotoxin binding.
The binding of α-[125I]bungarotoxin was measured using a modification of the method of Marks et al.(1986). The binding reaction was similar to that used for [3H]nicotine with the following changes: incubation time was 5 hr, samples contained 1 nM α-[125I]bungarotoxin instead of [3H]nicotine and the binding buffer also included .025% bovine serum albumin. Blanks were determined by including 1 mM l-nicotine in the assay.
[3H]epibatidine binding.
The binding of [3H]epibatidine was measured in a method analogous to that of [3H]nicotine with the following changes: incubations were in 1-ml polypropylene tubes in a 96-well format, incubation volume was 500 μl, and [3H]epibatidine rather than [3H]nicotine was used. Nonspecific binding was determined by including 100 μM L-nicotine in the assay. Nonspecific binding at all concentrations of [3H]epibatidine was less than twice background (40 dpm). The following experiments were conducted: construction of curves for inhibition of [3H]epibatidine binding in olfactory bulbs by cytisine, nicotine, acetylcholine (using tissue treated with 10 μM diisopropylflourophosphate during the tissue preparation), methylcarbachol, (+)-epibatidine and (-)-epibatidine (preliminary experiments indicated that inhibition in olfactory bulbs deviated markedly from that expected for a single site); construction of curves for inhibition of [3H]epibatidine binding in 12 brain regions by cytisine; and measurement of the concentration dependence of [3H]epibatidine binding in 12 brain regions. The concentration of [3H]epibatidine used for inhibition curves was about 400 pM (approximately 20 xKd ). This concentration was chosen to maintain ligand binding to the tissue to less than 5% of the total. An incubation time of 60 min was used for these experiments (equilibrium was reached in 20-30 min). For saturation curves, eight [3H]epibatidine concentrations between 6 and 800 pM were used. Incubation time for these experiments was 2 hr (equilibrium was reached by 60 min for all concentrations). In these experiments a significant fraction of the [3H]epibatidine was bound to the tissue, especially at lower ligand concentrations. Free [3H]epibatidine concentration was estimated by correcting for the amount of ligand bound to the tissue at each concentration for every brain region.
Protein.
Protein was measured using the method of Lowryet al. (1951) with bovine serum albumin as the standard.
Quantitative autoradiography of [3H]epibatidine binding.
Quantitative autoradiographic procedures were essentially similar to those described previously (Pauly et al., 1989; Markset al., 1992). Three DBA/2 mice were killed by cervical dislocation, the brains were removed and quickly frozen by immersion in isopentane (-35°C for 10 sec). Frozen brains were stored at -70°C until sectioning. Tissue sections (14-μ thick) were prepared using an IEC Minotome Cryostat refrigerated to -16°C. Sections were thaw mounted on subbed microscope slides (Richard Allen, Richland, MI). Slides were subbed by incubation with gelatin (1%)/chrome alum (0.1%) for 2 min at 37°C, drying overnight at 37°C, incubation for 30 min with .1% poly-l-lysine in 25 mM Tris (pH = 8.0), and drying overnight at 37°C. Six series of slides were collected from each mouse to allow measurement of [3H]epibatidine binding at six concentrations of cytisine. Slides were stored, desiccated, at -70°C until use.
Before incubation with [3H]epibatidine, sections were incubated in binding buffer for 10 min twice. After the initial incubations, samples were incubated with 400 pM [3H]epibatidine for 60 min at 22°C. Separate sets of incubations were conducted using the following six cytisine concentrations: 0, 5, 50, 150, 500 and 5000 nM. These concentrations were chosen because of the results obtained for cytisine inhibition of epibatidine binding in tissue homogenates. After the 60-min incubation, samples were washed as follows (all washes at 0°C): 10 sec in 1 x binding buffer, 10 sec in 1 x binding buffer, 10 sec in 0.1 x binding buffer, 10 sec in 0.1 x binding buffer, 10 sec in 5 mM HEPES, 5 sec in 5 mM HEPES. Sections were dried with a stream of air generated by two 15-cm computer fans. Dried sections were stored at 22°C overnight under vacuum. The slides containing the desiccated sections were apposed to Amersham Hyperfilm-3H in Wolf X-ray cassettes for 3 wk. For calibration, each film was also exposed to plastic tritium standards (Microscales, Amersham). [3H]Nicotine binding was also determined in sections prepared from different DBA/2 mice using the method described above, except that the incubation was conducted in the presence of 20 nM [3H]nicotine instead of 400 pM [3H]epibatidine.
After development of the films that had been exposed to sections to which [3H]epibatidine had been bound, signal intensity in selected brain areas was quantitated using NIH Image 1.55 after image capture using a CCD Imager Camera of sections illuminated with a Northern Light light box. When possible, six independent measurements from different tissue sections were made for each brain area under each incubation condition for each of the three mice. Ligand binding (fmol/mg wet weight) was estimated for each measurement after calculating the intensity of the radioactive signal (nCi/mg) by comparison of sample gray level to the standard curve and using the known specific activity of [3H]epibatidine. Signal intensity for each brain area of each mouse under each experimental condition was obtained by averaging the measurements made for that area. Films exposed to [3H]nicotine were not quantitated.
Calculations.
Results for saturation binding experiments were calculated using the Hill equation: B = Bmax*Ln/(Ln+ Kd n), where B is the binding at free ligand concentration, L, Bmax is the maximum number of binding sites, Kd is the equilibrium dissociation constant, and n is the Hill coefficient. Values of Bmax, Kd and n were calculated using the nonlinear least squares algorithm in Sigma Plot 5 (Jandel Scientific, San Rafael, CA). Results for inhibition of epibatidine binding were calculated using the formulas for either one or two binding sites: B = B0/(1+(I/IC50)) or B = B1/(1+(I/IC50-1)) + B2/(1+(I/IC50-2)), respectively, where B is ligand bound at inhibitor concentration, I, B0 is the binding in the absence of inhibitor, and B1 and B2 are the binding to two sites sensitive to inhibition with IC50-1 and IC50-2. Assuming competitive inhibition: IC50 =Ki x (1 + L/Kd ). Results were also calculated using the Hill equation.
Results
The concentration dependence for epibatidine binding was determined for 12 brain regions. The results of these experiments are summarized in figure 1. Saturable, specific binding was measured in each brain area and all Scatchard plots were linear. Since none of the Hill coefficients differed significantly from 1.0 (average for all 12 regions was 1.0 and the range of values was from 0.85 ± 0.07 in olfactory bulbs to 1.13 ± .05 in thalamus, mean ± S.E.M.),Kd and Bmaxvalues were calculated using the Michaelis-Menten equation. Values for the binding constants are summarized in table 1. Although theKd values ranged from 12.8 pM in hindbrain to 24.1 pM in hippocampus, these apparent differences among brain areas were not statistically significant (one-way analysis of variance, F(11,35) = 0.66, P > .05). Substantial differences in Bmax values were observed. Cerebellum had the fewest sites (12.6 fmol/mg protein) and interpeduncular nucleus the most sites (592 fmol/mg protein).
[3H]epibatidine binding in 12 brain regions. [3H]epibatidine binding to whole particulate fractions prepared from 12 brain regions of C57BL/6 mice indicated was measured using eight ligand concentrations. Each point is the mean ± S.E.M. of three separate experiments. The curves were determined by nonlinear least squares fit of the data to the simple Michaelis-Menten equation. The inset in each panel is the Scatchard plot for the binding data in the main panel. The abcissa for all Scatchard plots is [3H]epibatine bound (fmol/mg protein), although the ordinate is ([3H]epibatidine bound)/([3H]epibatine free) ((fmol/mg protein)/pM).
[3H]Epibatidine binding and cytisine inhibition of [3H]epibatidine binding in 12 brain regions
Inhibition of epibatidine binding in olfactory bulbs by six nicotinic drugs was determined. Four of these compounds (cytisine, nicotine, acetylcholine and methylcarbachol) have been used to measure high affinity agonist binding sites. In addition, the effects of the two isomers of epibatidine were also examined. The inhibition curves obtained with these compounds are illustrated in figure2. Each ligand inhibited [3H]epibatidine binding completely. The inhibition curves for cytisine, nicotine, acetylcholine and methylcarbachol deviated markedly from those expected for a single binding site (compare actual data to the best fit one-site model, dashed lines, in fig. 2). The Hill coefficient for each of these inhibition curves was less than 1.0 (cytisine = 0.51 ± 0.04, nicotine = 0.69 ± 0.05, acetylcholine = 0.67 ± 0.03 and methylcarbachol = 0.61 ± 0.04). Subsequent analysis of the data using a two-site model was performed and the results of these analyses for each agonist are illustrated by the solid lines in figure 2. Approximately 50% of epibatidine binding in olfactory bulbs was inhibited at relatively low concentrations for each agonist. Cytisine was the most potent inhibitor and also displayed the greatest selectivity (approximately 150-fold) between the high affinity and low affinity sites. The heterogeneous inhibition curves suggest that [3H]epibatidine binding to olfactory bulb tissue may occur at two distinct sites, even though epibatidine binding itself appeared monophasic. Monophasic inhibition of [3H]epibatidine binding by both (+)-epibatidine and (-)-epibatidine was also observed (Hill coefficients were 1.06 ± 0.04 and 1.00 ± 0.10, respectively) a result consistent with the observation of monophasic binding. Each isomer inhibited [3H]epibatidine binding completely, but naturally occurring (+)-epibatidine (IC50 = 0.49 ± 0.04 nM) was slightly more potent than (-)-epibatidine (IC50 = 0.76 ± 0.08 nM).
Inhibition of [3H]epibatidine binding in olfactory bulbs by six nicotinic agonists. Whole particulate fractions prepared from olfactory bulbs of C57BL/6 mice were incubated with 400 pM [3H]epibatidine and the indicated concentrations of cytisine, l-nicotine, acetylcholine, methylcarbachol, (+)-epibatidine or (-)-epibatidine. Samples incubated with acetylcholine were treated for 10 min at 37°C with 10 μM diisopropylflourophosphate during the tissue preparation to inhibit acetylcholinesterase. Each point represents the mean ± S.E.M. of three separate determinations. Inhibition curves were calculated by nonlinear least squares method. The dashed curves are the best fits to a single binding site model and the solid curves are best fit to a two-site model. The one-site and two-site fits of the inhibition curves for (+)-epibatidine and (-)-epibatidine were identical.
Because cytisine showed the largest difference in affinity between the two sites of the four agonists tested for their inhibition of epibatidine binding, cytisine inhibition of epibatidine binding was subsequently measured in 12 brain areas. The resulting inhibition curves are shown in figure 3. Each Hill coefficient was significantly less than 1.0 (range 0.48 for olfactory bulbs to 0.81 for striatum), suggesting that the inhibition of epibatidine binding by cytisine was not monophasic in any brain area. The experimental data and the curves obtained from one-site (dashed lines) and two-site (solid lines) fits are shown in figure 3. Although the one site model provided a reasonable fit to the data in several brain regions, the inhibition profiles in most brain regions deviated markedly from those expected for a monophasic process. The inhibition constants estimated for the two site fits and the amount of epibatidine binding corresponding to these two constants are listed in table1. The averageKi values calculated for high and low affinity inhibition were 0.29 and 29.4 nM (corresponding, under the conditions of these experiment (400 nM epibatidine) to IC50 values of 6.1 and 700 nM, respectively). No significant differences in Ki values among brain regions were observed except for the high affinity constant calculated for interpeduncular nucleus, which was significantly higher than the corresponding values in the other eleven brain areas. The fraction of epibatidine binding to sites sensitive to inhibition at low and high cytisine concentrations varied substantially among the brain regions. Although binding sites less sensitive to cytisine inhibition represented only 7% of the total in hippocampus, these sites accounted for 63% of the total in interpeduncular nucleus. The overall correlation between the [3H]epibatidine binding sites more sensitive to cytisine inhibition and those less sensitive to cytisine inhibition in the twelve brain regions was 0.52.
Cytisine inhibition of [3H]epibatidine binding in 12 brain regions. Whole particulate fractions were prepared from each of the 12 brain regions dissected from C57BL/6 mouse brains. Samples were incubated with 400 pM [3H]epibatidine and the indicated concentrations of cytisine. Each point represents the mean ± S.E.M. of three separate experiments. Data were subjected to nonlinear curve fitting. The dashed lines represent the best fit to a one-site model and the solid lines represent the best fit to a two-site model for each brain region.
The regional distribution of epibatidine binding was subsequently compared to the distribution of binding sites measured with nicotine and α-bungarotoxin to determine whether the epibatidine binding sites may correspond to nicotinic receptors measured with these two commonly used ligands (fig. 4). Total epibatidine binding was significantly correlated with nicotine binding (r = 0.95). However, an even closer relationship was observed between that component of epibatidine binding sensitive to inhibition by low cytisine concentrations and nicotine binding (r = 0.99). The slope of this line was not unity, primarily because the concentration of [3H]epibatidine was approximately 20 xKd , and that used for nicotine was approximately 4 x Kd . In contrast, neither total epibatidine binding sites (r = 0.38), nor those inhibited by higher cytisine concentrations were significantly correlated to the number of α-bungarotoxin binding sites (r = 0.45).
Comparison of regional distribution of [3H]epibatidine binding with [3H]nicotine binding and [125I]-α-bungarotoxin binding. In the left panel, regional distribution of [3H]nicotine binding (16 nM) is compared to total [3H]epibatidine binding (340 pM) (solid circles) and to [3H]epibatidine binding inhibited at low cytisine concentrations (open circles). In the right panel, regional distribution of [125I]-α-bungarotoxin binding (1 nM) is compared to total [3H]epibatidine binding (solid squares) and to [3H]epibatidine binding resistant to cytisine inhibition (open squares). Cytisine-sensitive and -resistant binding sites were calculated by nonlinear least squares curves fits for inhibition curves similar to those shown in figure 3. Each point represents the mean ± S.E.M. of at least three determinations. Correlation coefficients have been place near the appropriate lines. The brain areas used for these correlations were the same as those listed in table 1.
Analysis of nicotinic ligand binding in regionally dissected brain areas provides an indication of the distribution of putative nicotinic receptor subtypes labeled with epibatidine, but the anatomical resolution is limited. Therefore, the regional distribution of epibatidine binding was evaluated autoradiographically in tissue samples that were incubated with radioligand and one of the following concentrations of cytisine: 0, 5, 50, 150, 500 and 5000 nM.
The distribution of [3H]nicotine binding, total [3H]epibatidine binding, and [3H]epibatidine binding in the presence of 50 nM cytisine at six levels is shown in figure 5. The distribution of total [3H]epibatidine binding was generally similar to that of [3H]nicotine binding. However, [3H]nicotine binding was noticeably absent in the accessory olfactory bulbs, fasciculus retroflexus, and the pineal gland. Furthermore, [3H]epibatidine binding was observed in the presence of 50 nM cytisine in the medial habenula, dorsal and ventral lateral geniculate nuclei, the interpeduncular nucleus, the superior colliculus and the inferior colliculus.
Comparison of [3H]nicotine binding, total [3H]epibatidine binding and [3H]epibatidine binding in the presence of 50 nM cytisine. Sections (14 μ) prepared from DBA/2 mice were incubated with 20 nM [3H]nicotine (left column) or 400 pM [3H]epibatidine in the presence of 0 nM (center column) or 50 nM (right column) cytisine as described in “Materials and Methods.” Pictures shown are digital images of autoradiograms. The sections for [3H]epibatidine are nearly adjacent, although those for [3H]nicotine are from a different mouse of the same strain. Abbreviations for the brain regions indicated are: AOB, accessory olfactory bulb; AON, accessory olfactory nucleus; Cb, cerebellum; CG, central gray; CP, caudate putamen; DLG, dorsal lateral geniculate nucleus; DTN, dorsal tegmental nucleus; fr, fasciculus retroflexus; FrCx, frontal cortex; HT, hypothalamus; IC, inferior colliculus; IPN, interpeduncular nucleus; LD, laterodorsal thalamic nucleus; MH, medial habenula; NA, nucleus accumbens; OT, olfactory tubercle; Pi, pineal body; SC, superior colliculus; VLG, ventral lateral geniculate nucleus; VP, ventral posterior thalamic nucleus.
Autoradiograms of [3H]epibatidine binding in tissue sections incubated with each of six cytisine concentrations were subsequently quantitated and the results are summarized in table2. As suggested by the autoradiograms in figure 5, the binding of [3H]epibatidine in most of the brain regions analyzed was reduced by at least 80% after incubation with 50 or 150 nM cytisine. Approximately 20% of total epibatidine binding was not inhibited by 150 nM cytisine in several brain regions: nucleus accumbens, caudate putamen, zona incerta, dorsal lateral geniculate, ventral lateral geniculate, medial geniculate, superior colliculus, ventral tegmental nucleus and pons (excluding ventral tegmental nucleus). Approximately 30 to 50% of [3H]epibatidine binding in inferior colliculus was relatively insensitive to cytisine inhibition. Finally, in the accessory olfactory bulb (including the granular cell layer), medial habenula, interpeduncular nucleus, fasciculus retroflexus and pineal gland more than 50% of [3H]epibatidine binding was insensitive to inhibition by cytisine.
Regional [3H]epibatidine binding at six cytisine concentrations
Discussion
The results of this study demonstrate that epibatidine binding in mouse brain occurs to at least two sites. In contrast to rat and human brain (Houghtling et al., 1995), saturation curves for epibatidine binding in mouse brain are essentially monophasic: that is the Scatchard plots are linear and Hill coefficients do not differ from unity. However, the inhibition of epibatidine binding in olfactory bulbs by cytisine, nicotine, acetylcholine or methylcarbachol is multiphasic, indicating that epibatidine binds to at least two sites in mouse brain. The two epibatidine binding sites that are differentially sensitive to cytisine inhibition are differentially distributed throughout mouse brain as indicated both with different patterns of inhibition in regionally dissected tissue as well as the distinct differences in binding observed as cytisine concentration increased. Several lines of evidence suggest that the cytisine-sensitive component of epibatidine binding is identical to the site normally measured by nicotine, cytisine, acetylcholine and methylcarbachol binding: 1) The inhibition constants for the four agonists are comparable to theKd values measured when these compounds are used as ligands; 2) the regional distribution of cytisine-sensitive epibatidine binding determined in regionally dissected tissue is highly correlated with nicotine binding (r = 0.99) and 3) cytisine-sensitive epibatidine binding measured autoradiographically corresponds to nicotine binding. The correspondence between cytisine-sensitive epibatidine binding and nicotine binding suggests that this portion of epibatidine binding occurs to nicotinic receptors of the α4/β2 subtype.
In contrast to the heterogeneous binding of [3H]epibatidine binding observed in rat and human brain (Houghtling et al., 1995), [3H]epibatidine binding in mouse appeared to be monophasic at concentrations up to 800 pM in the 12 brain regions assayed. Heterogeneous binding was only observed when inhibition of [3H]epibatidine binding by other nicotinic agonists was determined. The apparently monophasic binding of epibatidine, itself, suggests that the estimate of the proportion of sites differentially sensitive to cytisine inhibition is relatively independent of epibatidine concentration. The difference in epibatidine affinity between rat and mouse brain suggests that subtle differences in agonist binding sites may exist, particularly for that receptor subtype not normally measured with high affinity agonist binding. However, it is distinctly possible that biphasic [3H]epibatidine binding in mouse brain will be observed with ligand concentrations of more than 800 pM.
The results presented in our study do not directly address the possible molecular composition of the component(s) of [3H]epibatidine binding less sensitive to inhibition by cytisine. However, this component does not appear to be the α7 subunit containing α-bungarotoxin binding site, because the regional distribution of this component does not correspond to that observed either with regionally dissected tissue (this study) or autoradiographically (Pauly et al., 1989). Flores et al. (1996) have demonstrated in rat trigeminal ganglion that the epibatidine binding site that does not bind cytisine with high affinity is primarily an α3/β4 containing receptor. The fact that [3H]epibatidine binding that is less sensitive to inhibition by cytisine is very high in regions that express high levels of α3 mRNA (Marks et al., 1992; Marks MJ, unpublished observations) suggests that this binding site in mouse brain may also contain α3 nicotinic receptor subunits. The mRNA encoding the β4 subunit is also present in many, but not all, of the brain areas expressing α3 mRNA in mouse brain (Marks MJ, unpublished observations). Among those areas expressing α3 but not β4 include substantia nigra, ventral tegmental area, and superior colliculus. It remains to be determined whether [3H]epibatidine binding with low affinity for cytisine occurs to several different nicotinic receptor subtypes, and whether all of these contain the α3 subunit. It should be noted that regional colocalization of receptor subunit and the mRNA encoding that subunit may not occur, especially if the receptor protein is localized axonally or dendritically. Indeed, examples of the location of nicotinic receptors on terminal fields have been reported for the retina-superior colliculus (Swanson et al., 1987), substantia nigra-striatum (Schwartz et al., 1984; Clarke and Pert, 1985) and the medial habenula-interpeduncular nucleus (Clarkeet al., 1986) pathways.
[3H]Epibatidine binding sites relatively insensitive to cytisine inhibition observed in mouse brain are limited to relatively few brain areas. However, the number of sites in several of these areas is very large, suggesting that in these regions the nicotinic receptor insensitive to cytisine inhibition may be of great functional importance. Similarly, the comparison of the binding site densities for cytisine and epibatidine in rat brain also indicated that relatively few brain regions had substantially higher epibatidine binding than cytisine binding and the regions in which epibatidine binding was remarkably higher (medial habenula, fasciculus retroflexus, subiculum, interpeduncular nucleus, superior colliculus and inferior colliculus) (Perry and Kellar, 1995) are similar to the regions displaying significant cytisine-resistant epibatidine binding in mouse brain. Some quantitative differences in receptor distribution between the species is suggested, however. For example, nicotinic binding in inferior colliculus is substantially higher in mouse, while binding in cerebellum is substantially higher in rat. A preliminary report (Arriola et al., 1996) in which nicotine was used to selectively inhibit epibatidine binding in rat brain also presented data substantially similar to those described here. Similarly, another preliminary report comparing the distribution of epibatidine binding to that of cytisine and A-85380, a very high affinity nicotinic agonist, noted that several brain areas (e.g., medial habenula, interpeduncular nucleus and fasciculus retroflexus) were labeled more intensely by epibatidine than by the other two agonists (Pauly et al., 1996). The distribution of [3H]epibatidine binding sites with low affinity for cytisine also shows some similarities to that proportion of [125I]-neuronal bungarotoxin binding in rat brain that is poorly inhibited by α-bungarotoxin, but sensitive to inhibition by nicotine (Schultz et al., 1991), in that the fasciculus retroflexus, medial habenula, inferior colliculus, superior colliculus, dorsal and ventral lateral geniculate nuclei, and the interpeduncular nucleus were regions where this component of toxin binding was observed. The fact that neuronal-bungarotoxin is a potent and relatively selective inhibitor of the α3/β2 nicotinic receptor subtype expressed in Xenopus oocytes (Luetje et al., 1990) suggests that, under the conditions described, [125I]neuronal-bungarotoxin may have been binding to α3-containing receptors.
Although it has not yet been established that [3H]epibatidine binds to only two nicotinic receptor subtypes, there appears to be no question that epibatidine will be extremely useful for the measurement and characterization of nicotinic receptor binding sites unique from those that can be measured with agonists such as nicotine or cytisine or with α-bungarotoxin. The use of selective cytisine inhibition may prove useful in identifying and quantifying the sites that have escaped detection by the ligands commonly used to date.
Footnotes
-
Send reprint requests to: Dr. Allan C. Collins, Institute for Behavioral Genetics, University of Colorado, Campus Box 447, Boulder, CO 80409-0447.
-
↵1 This work was supported by Grants DA-10156 and DA-03194 from the National Institute on Drug Abuse. ACC is supported, in part, by Research Scientist Award DA-00197 from the National Institue on Drug Abuse.
- Abbreviation:
- HEPES
- N-[2-hydroxyethyl]piperazine-N′-[2-ethanesulfonic acid]
- Received August 5, 1997.
- Accepted December 29, 1997.
- The American Society for Pharmacology and Experimental Therapeutics
