Reboxetine: Functional Inhibition of Monoamine Transporters and Nicotinic Acetylcholine Receptors

Materials and Methods

Subjects.

Male Sprague-Dawley rats (200–250 g) were obtained from Harlan (Indianapolis, IN) and were housed two per cage with free access to food and water. Experimental protocols were in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at the University of Kentucky and at Pharmacia Corporation.

Chemicals.

Reboxetine CH₃SOCH (a racemic mixture of R,R- andS,S- ([2-[α[2-ethoxyphenoxy]benzyl]-morpholine sulfate]) was synthesized and provided by Pharmacia Corporation (Kalamazoo, MI). (±)-Bupropion HCl, (−)-cytisine, desipramine HCl, (±)-epibatidine dihydrochloride, fluoxetine HCl, GBR-12909 HCl, methyllycaconitine citrate, nomifensine maleate,S-(−)-nicotine ditartrate, pargyline HCl, tetrodotoxin (TTX), HEPES, bovine serum albumin, and catechol were purchased from Sigma-Aldrich (St. Louis, MO). [³H]Cytisine HCl [hexahydro-1,5-methano-8H-pyrido-[1,2-a][1,5]diazocin-8-one, 3,5-³H(N); specific activity, 35 Ci/mmol], [³H]epibatidine [[5,6-bicycloheptyl-³H]-(±)-epibatidine; specific activity, 48 Ci/mmol], [³H]NE [norepinephrine, levo-[7-³H]; specific activity, 14.4 Ci/mmol], [³H]DA [dihydroxyphenylethylamine 3,4-ethyl-2-[N-³H]; specific activity, 25.6 Ci/mmol], [³H]5-HT [hydroxytryptamine creatine sulfate 5-[1,2-³H(N)]; specific activity, 27.5 Ci/mmol], and ⁸⁶RbCl (specific activity, 55.2 mCi/mmol) were purchased from PerkinElmer Life Sciences (Boston, MA). [³H]MLA [1α,(S),6β,14α,16β]-20-ethyl-1,6,14,16-tetramethoxy-4-[[[2-([3-³H]-[3-³H]-methyl-2,5-dioxo-1-pyrrolidinyl)benzoyl]oxy]methyl]aconitane-7,8-diol; specific activity, 25 Ci/mmol] was purchased from Tocris Cookson (Ballwin, MO). α-d-Glucose was purchased from Aldrich Chemical Co. (Milwaukee, WI), l-ascorbic acid was purchased from AnalaR-BHD Ltd. (Poole, Dorset, UK), potassium phosphate monobasic was purchased from Mallinckrodt (Paris, KY), and TS-2 tissue solubilizer was purchased from Research Products International (Mount Prospect, IL). All other chemicals were purchased from Fisher Scientific (Pittsburgh, PA).

[³H]Neurotransmitter Uptake Assays.

[³H]NE and [³H]5-HT uptake assays were performed using hippocampal synaptosomes, and [³H]DA uptake assays were performed using rat striatal synaptosomes. Striata or hippocampi from an individual rat were homogenized in 20 ml of ice-cold sucrose solution (0.32 M sucrose and 5 mM NaHCO₃, pH 7.4) with 16 passes of a Teflon pestle homogenizer (clearance 0.003 inches). The homogenate was centrifuged (2000g, 10 min, 4°C), and the resulting supernatant was centrifuged (20,000g, 15 min, 4°C). The resulting pellet was resuspended in 2.4 ml of ice-cold assay buffer (125 mM NaCl, 5 mM KCl, 1.5 mM MgSO₄, 1.25 mM CaCl₂, 1.5 mM KH₂PO₄, 10 mM α-d-glucose, 25 mM HEPES, 0.1 mM EDTA, 0.1 mM pargyline, and 0.1 mM ascorbic acid, saturated with 95% O₂/5% CO₂, pH 7.4). The final protein concentration was 400 μg/ml, as determined by protein-dye bonding using bovine serum albumin as a standard (Bradford, 1976). Assays were performed in duplicate in a total volume of 500 μl. Aliquot parts of synaptosomal suspension (50 μl) were added to tubes containing 350 μl of buffer and 50 μl of one of nine concentrations of drug (reboxetine, desipramine, fluoxetine, or GBR-12909) followed by 50 μl of [³H]NE, [³H]5-HT, or [³H]DA (final concentration of 0.1 μM). Accumulation proceeded for 10 min at 34°C. Accumulation was terminated by addition of 3 ml of ice-cold buffer containing catechol (1 mM) and rapid filtration through Whatman GF/B glass fiber filters presoaked with catechol (1 mM). Filters were washed three times with ice-cold buffer, transferred to scintillation vials, cocktail was added, and radioactivity was determined by liquid scintillation spectroscopy (model B1600 TR scintillation counter; Packard Instrument Company, Inc., Downer's Grove, IL).

To determine the effect of repeated reboxetine administration on [³H]neurotransmitter uptake, groups of rats were injected (i.p.) twice daily (9:00 AM and 4:00 PM) with reboxetine (10 mg/kg) or 0.9% (w/v) saline vehicle for 14 days. One day after the last injection of reboxetine or saline, inhibition of [³H]NE, [³H]5-HT, or [³H]DA uptake was assessed using tissue from one rat from each of the reboxetine treatment and control groups.

[³H]MLA, [³H]Cytisine, and [³H]Epibatidine Binding Assays.

[³H]MLA was used to probe the α7* nAChR (Davies et al., 1999) and [³H]cytisine to probe the α4* nAChR (Pabreza et al., 1990). Binding of [³H]epibatidine in the presence of cytisine (30 nM) was used to probe the α3* nAChR subtype (Xiao et al., 1998). Rat whole brain, minus cerebellum, was dissected quickly, weighed, and homogenized in 9 vol/g wet weight of ice-cold 0.32 M sucrose with 10 passes of a rotating Teflon pestle homogenizer (clearance 0.003 inches; setting 50, 099CK4; Glas-Col, Terre Haute, IN). The homogenate was centrifuged (1000g, 10 min, 4°C). The resulting supernatant was centrifuged (20,000g, 20 min, 4°C). The pellet was resuspended to a final protein concentration of 1 to 8 mg/ml. The homogenate (5 ml) was frozen at −80°C until use in the binding assay. On the day of assay, aliquot parts of membrane suspension were thawed at room temperature and diluted with Krebs-HEPES buffer (20 mM HEPES, 4.16 mM NaHCO₃, 0.44 mM KH₂PO₄, 127 mM NaCl, 5.36 mM KCl, 1.26 mM CaCl₂, and 0.98 mM MgCl₂, pH 7.0) to afford 25 to 150 μg of protein/0.4 ml, which was added to each assay tube.

Competition binding studies determinedK_i values for inhibition of [³H]MLA (final concentration, 3 nM), [³H]cytisine (final concentration, 1 nM), and [³H]epibatidine (final concentration, 1 nM) binding, based on K_d values of 3, 1.7, and 0.71 nM, respectively, determined in preliminary experiments. Aliquot parts of membrane homogenate were added to tubes containing 7 to 11 concentrations (1 nM–1 or 100 μM) of reboxetine, bupropion, MLA, cytisine, epibatidine, or nicotine. One of the three radiolabeled ligands was added to the mixture (total volume of 500 μl), which was incubated at 25°C for 60 min. Nonspecific binding was determined in the presence of 1 mM MLA for the [³H]MLA binding assay or 1 mM nicotine for the [³H]cytisine and [³H]epibatidine binding assays. Incubations were terminated by rapid filtration through Whatman GF/B glass filters, presoaked in 0.05% polyethylenimine and 50 mM Tris-HCl, pH 7.0. Filters were rapidly washed two times with 5 ml of cold 0.9% saline. Radioactivity was determined using a model Tri-Carb 2220 CA scintillation counter (Packard Instrument Company, Inc.).

³H Overflow Superfusion Assay.

³H Overflow from rat striatal slices preloaded with [³H]DA or rat hippocampal slices preloaded with [³H]NE was determined using modifications of a previously published method (Dwoskin and Zahniser, 1986). Coronal striatal or hippocampal slices (500 μm; 6–8 mg from striatum; 3–4 mg from hippocampus) were incubated in Krebs' buffer (118 mM NaCl, 4.7 mM KCl, 1.2 mM MgCl₂, 1.0 mM NaH₂PO₄, 1.3 mM CaCl₂, 11.1 mM α-d-glucose, 25 mM NaHCO₃, 0.11 mM ascorbic acid, and 4.0 μM EDTA, saturated with 95% O₂/5% CO₂, pH 7.4) in a metabolic shaker at 34°C for 30 min. Slices were incubated in fresh buffer containing 0.1 μM [³H]DA or 0.1 μM [³H]NE (6–8 slices/3 ml) for an additional 30 min. After rinsing, each slice was transferred to a glass superfusion chamber maintained at 34°C and superfused at 1 ml/min with oxygenated Krebs' buffer containing pargyline (10 μM) and either nomifensine (10 μM) in [³H]DA release experiments or desipramine (10 μM) in [³H]NE release experiments to inhibit metabolism by monoamine oxidase and to inhibit uptake into the respective presynaptic terminals, ensuring that³H overflow represented primarily the parent³H-neurotransmitter rather than metabolites. After 60 min of superfusion, three 5-min samples (5 ml/sample) were collected to determine basal ³H outflow. After collection of the third basal sample, slices from an individual rat were superfused for 30 min in the absence or presence of six to seven concentrations of reboxetine (1 nM–100 μM), which remained in the buffer until the end of the experiment. Intrinsic activity (i.e., the ability of reboxetine to evoke ³H overflow) was assessed during the 30-min period of superfusion. The complete concentration response for reboxetine in the [³H]NE release assay was determined in one series of experiments. However, two series of experiments were performed to obtain the complete concentration response in the [³H]DA release assay; the first series determined the effect of 1 nM to 10 μM, and the second series determined the effect of 10 nM to 100 μM. Subsequently, nicotine (10 μM) was added to the buffer containing reboxetine, and superfusion continued for an additional 60 min. The ability of reboxetine to inhibit nicotine-evoked ³H overflow was determined during the latter 60-min period of superfusion. Each striatal slice was exposed to only one concentration of reboxetine. Each experiment was replicated a minimum of six times. A repeated measures design was used, such that the concentration effect of reboxetine was determined using brain slices from a single rat, and a minimum of six rats. At the end of the experiment, each slice was solubilized with TS-2. The pH and volume of the solubilized tissue samples were adjusted to those of the superfusate samples. Radioactivity in the superfusate and tissue was determined by liquid scintillation spectroscopy (model B1600 TR scintillation counter; Packard Instrument Company, Inc.).

To determine the mechanism of the reboxetine-induced inhibition of nicotine-evoked [³H]NE overflow, a Schild analysis was performed in which hippocampal slices were superfused in the absence or presence of a single concentration of reboxetine (10 nM, 100 nM, or 1 μM), and the concentration effect for nicotine was determined. These experiments were performed as described above, except that slices from an individual rat were superfused for 30 min in the absence or presence of a single concentration of reboxetine (10 nM–1 μM), and subsequently, one of six concentrations of nicotine (1 nM–100 μM) was added to the buffer containing reboxetine. Superfusion continued for an additional 60 min. Each slice from a single animal was exposed to only one concentration of reboxetine and only one concentration of nicotine. These experiments also used a repeated measures design, such that the concentration effect for nicotine was determined in both the absence and presence of reboxetine using hippocampal tissue from a single rat, and a minimum of six rats for each reboxetine concentration. Additionally, one slice in each experiment was superfused in the absence of reboxetine and nicotine, constituting the buffer control condition. Slices and superfusate were processed as described previously.

Fractional release for each superfusate sample was calculated by dividing the tritium collected in each sample by the total tritium present in the tissue at the time of sample collection. Fractional release was expressed as a percentage of total tritium in the tissue at the time of sample collection. Basal ³H outflow was calculated from the average of the tritium collected in the three 5-min samples just before the addition of reboxetine. Reboxetine- and nicotine-evoked ³H overflow was calculated by summing the increases in collected tritium that resulted from exposure to drug and subtracting the basal outflow.

⁸⁶Rb⁺ Efflux Assay.

The effects of reboxetine on ⁸⁶Rb⁺ efflux were determined using previously published methods (Miller et al., 2000). Thalamus was homogenized and centrifuged (1000g, 10 min, 4°C). The supernatant fraction was centrifuged (12,000g, 20 min, 4°C) to obtain the synaptosomal fraction. Synaptosomes were incubated for 30 min in 35 μl of buffer (140 mM NaCl, 1.5 mM KCl, 2.0 mM CaCl₂, 1.0 mM MgSO₄, and 20 mM α-d-glucose, pH 7.5) containing 4 μCi of⁸⁶Rb⁺.⁸⁶Rb⁺ uptake was terminated by filtration of the synaptosomes onto glass fiber filters (6 mm, type A/E; Gelman Instrument Co., Ann Arbor, MI) under gentle vacuum (0.2 atm), followed by three washes with buffer (0.5 ml each). Subsequently, each filter with⁸⁶Rb⁺-loaded synaptosomes (39 ± 4.8 μg of protein/μl) was placed on a 13-mm glass fiber filter (type A/E) mounted on a polypropylene platform.⁸⁶Rb⁺ efflux assay buffer (125 mM NaCl, 5 mM CsCl, 1.5 mM KCl, 2 mM CaCl₂, 1 mM MgSO₄, 25 mM HEPES, 20 mM α-d-glucose, 0.1 μM TTX, and 1.0 g/l bovine serum albumin, pH 7.5) was superfused at a rate of 2.5 ml/min. TTX and CsCl were included in the assay buffer to block voltage-gated Na⁺ and K⁺ channels, respectively, and to reduce the rate of basal⁸⁶Rb⁺ efflux. The ability of reboxetine (1 nM–100 μM) to inhibit⁸⁶Rb⁺ efflux evoked by 1 μM nicotine was determined. In previous experiments, this concentration of nicotine was the lowest concentration that produced maximal ⁸⁶Rb⁺ efflux (Miller et al., 2000). After 8 min of superfusion, basal samples (sample/18 s) were collected for 2 min. Subsequently, synaptosomes were superfused for 3 min in the absence or presence of one of six concentrations (1 nM–100 μM) of reboxetine. Then, nicotine (1 μM) was added to buffer, and superfusion was continued for an additional 3 min, followed by superfusion for 3 min with buffer in the absence of either drug. Each aliquot part of thalamic synaptosomes was exposed to only one concentration of reboxetine. In each experiment, one synaptosomal aliquot part was also exposed to nicotine (1 μM) in the absence of reboxetine, and one synaptosomal aliquot part was superfused in the absence of either drug to determine basal⁸⁶Rb⁺ efflux during the course of the experiment. Samples were analyzed by liquid scintillation spectroscopy (model B1600 TR scintillation counter; Packard Instrument Company, Inc.). To determine the basal rate of⁸⁶Rb⁺ efflux, an exponential decay curve was used to fit the data points proceeding and following superfusion with reboxetine and nicotine (SigmaPlot 2000; SPSS, Inc., Chicago, IL). Drug-evoked increases in⁸⁶Rb⁺ efflux were calculated as the fractional increase above baseline. Increases were summed to obtain total⁸⁶Rb⁺ efflux during the period of superfusion with reboxetine and/or nicotine, and normalized to ⁸⁶Rb⁺ content in the corresponding synaptosomal sample to reduce variability within and between experiments.

Data Analysis.

For the ³H neurotransmitter uptake assays, IC₅₀ values for each individual concentration-response curve were generated using nonlinear regression and a sigmoidal curve fit (GraphPad Prism, version 3.0; GraphPad Software, San Diego, CA). For the³H neurotransmitter uptake assays, comparisons of IC₅₀ values for reboxetine with desipramine, fluoxetine, or GBR-12909 were analyzed by independent groupt tests. To assess the effect of repeated reboxetine treatment on inhibition of ³H neurotransmitter uptake, two-way analysis of variance (ANOVA) was performed on the IC₅₀ values with repeated drug treatment (reboxetine or saline) and inhibitor (reboxetine versus desipramine, fluoxetine, or GBR-12909) as between-group factors (SPSS version 9.0; SSPS, Inc.). Where appropriate, Tukey post hoc tests and simple main effect analyses (p < 0.05) were performed. In competition binding studies, the inhibition constant (K_i) was calculated from the binding isotherms for [³H]MLA, [³H]cytisine, and [³H]epibatidine binding using a nonlinear regression curve fit program according to the Cheng and Prusoff equation. For the ³H overflow assays, intrinsic activity of reboxetine and the ability of reboxetine to inhibit nicotine-evoked ³H overflow were analyzed via one-way repeated measures ANOVA with reboxetine concentration as a within-subject factor. EC₅₀ values for intrinsic activity and IC₅₀ values for inhibition of nicotine-evoked ³H overflow were determined via nonlinear regression, which fit the mean data to a sigmoidal concentration-response curve. The mechanism of inhibition of [³H]NE overflow was determined via Schild analysis and analyzed via three-way repeated measures ANOVA with nicotine concentration and the absence or presence of reboxetine as within-subject factors, and reboxetine concentration as a between-groups factor. The intrinsic activity of reboxetine on⁸⁶Rb⁺ efflux and the ability of reboxetine to inhibit nicotine-evoked⁸⁶Rb⁺ efflux were analyzed via one-way repeated measures ANOVA with reboxetine concentration as a within-subject factor.

Results

Repeated Administration of Reboxetine Does Not Alter Selectivity as a NET Inhibitor.

The concentration of reboxetine that inhibited [³H]NE uptake by 50% (IC₅₀) was 8.5 nM. IC₅₀values for reboxetine inhibition of [³H]DA and [³H]5-HT uptake were 89 and 6.9 μM, respectively (Fig. 1). Thus, reboxetine was 3 to 4 orders of magnitude more potent as an inhibitor of [³H]NE uptake than [³H]DA and [³H]5-HT uptake. Significant differences were not found between reboxetine and desipramine (IC₅₀ value = 17.0 nM) to inhibit [³H]NE uptake [t₍₆₎ = 0.20, p = 0.85; Fig. 1]. GBR-12909 (IC₅₀ value = 1.5 μM) was significantly more potent [t₍₄₎ = 2.62, p < 0.05] than reboxetine in inhibiting [³H]DA uptake, and fluoxetine (IC₅₀ value = 0.4 μM) was significantly more potent [t₍₄₎ = 15.84, p < 0.001] than reboxetine in inhibiting [³H]5-HT uptake (Fig. 1). Thus, reboxetine is a highly potent and selective [³H]NE uptake inhibitor.

In a separate series of experiments, rats were administered reboxetine (10 mg/kg) or saline twice daily for 14 days, and³H neurotransmitter uptake was subsequently determined. Repeated reboxetine injection did not alter reboxetine- or desipramine-induced inhibition of [³H]NE uptake [F_(1,12) = 0.543, p = 0.48; Table 1], reboxetine- or GBR-12909-induced inhibition of [³H]DA uptake [F_(1,16) = 0.041, p = 0.84; Table 1], and reboxetine- or fluoxetine-induced inhibition of [³H]5-HT uptake [F_(1,12) = 0.439, p = 0.52; Table 1]. Thus, repeated reboxetine injection did not alter the potency or selectivity of reboxetine to inhibit [³H]NE uptake.

Reboxetine Does Not Inhibit [³H]MLA, [³H]Cytisine, or [³H]Epibatidine Binding to Rat Whole Brain Membrane Preparations.

To validate the nAChR binding assays, concentration-dependent inhibition of binding was determined using several standard competitors, MLA, nicotine, cytisine, and epibatidine (Table 2). Neither reboxetine nor bupropion significantly inhibited binding of these three nAChR radioligands (Table 2).

Reboxetine Noncompetitively Inhibits Nicotine-Evoked ³H Overflow from Superfused [³H]NE-Preloaded Hippocampal Slices.

In a concentration-dependent manner, reboxetine increased³H overflow from superfused rat hippocampal slices preloaded with [³H]NE (Table3). A significant main effect of reboxetine concentration was found [F_(6,30) = 10.67, p< 0.001], and post hoc tests revealed greater³H overflow after superfusion with 100 μM reboxetine compared with control (0 M) or compared with lower concentrations (1 nM–10 μM) of reboxetine. Furthermore,³H overflow from slices superfused with 1 nM to 10 μM reboxetine was not different from control (Table 3), indicating that these concentrations of reboxetine produced no intrinsic activity.

In a concentration-dependent manner, reboxetine inhibited nicotine (10 μM)-evoked ³H overflow from hippocampal slices preloaded with [³H]NE (Fig.2). An IC₅₀ value of 7.3 nM was obtained. Because 100 μM reboxetine intrinsically evoked ³H overflow, data from slices superfused in the presence of this concentration of reboxetine were not included in the analysis of the inhibition of the effect of nicotine. A significant main effect of reboxetine concentration was found [F_(5,20) = 5.19, p < 0.01], and post hoc tests revealed that 0.1 to 10 μM reboxetine significantly inhibited nicotine-evoked ³H overflow. Evaluation of the time course (Fig. 2, inset) revealed that reboxetine (10 nM–1 μM) did not increase fractional [³H]NE release compared with fractional release from control slices, which were not exposed to reboxetine. Addition of nicotine to the buffer resulted in an immediate increase in fractional release, which returned to basal levels by the end of the superfusion period. The time course also illustrates the concentration-dependent inhibition produced by reboxetine, such that 1 μM reboxetine completely inhibited the stimulation produced by this concentration of nicotine.

To determine the mechanism of reboxetine-induced inhibition of the effect of nicotine, the concentration response for nicotine-evoked (10 nM–1 μM) ³H overflow from [³H]NE-preloaded hippocampal slices was determined in the absence and presence of 10 nM, 100 nM, or 1 μM reboxetine (Fig. 3). Data from this Schild analysis were analyzed via repeated measures ANOVA, which revealed a significant three-way interaction of nicotine concentration, the absence or presence of reboxetine, and reboxetine concentration [F_(12,84) = 4.73, p< 0.001]. Simple main effect analyses and post hoc tests revealed that nicotine produced a concentration-dependent increase in³H overflow [F_(6,102) = 65.51, p< 0.001]. The low concentration of reboxetine (10 nM) significantly inhibited the 10 μM nicotine-evoked ³H overflow [F_(6,30) = 2.61, p < 0.05]; however, this concentration of reboxetine did not inhibit³H overflow evoked by 100 μM nicotine. Reboxetine at 100 nM and 1 μM inhibited ³H overflow evoked by 100 nM to 100 μM nicotine [F_(6,30) = 4.57, p < 0.01 and F_(6,30) = 11.49,p < 0.001, respectively]. Moreover, the inhibition produced by the latter concentrations of reboxetine was not surmounted by increasing concentrations of nicotine. These results indicate that reboxetine noncompetitively inhibited nicotine-evoked [³H]NE overflow from superfused hippocampal slices.

Reboxetine Evokes ³H Overflow from Rat Striatal Slices Preloaded with [³H]DA; However, Reboxetine Does Not Inhibit Nicotine-Evoked ³H Overflow.

Two series of experiments were performed to determine the effect of reboxetine across 6 orders of magnitude concentration. The effect of reboxetine to increase ³H overflow from [³H]DA-preloaded striatal slices is shown in Table 4. Significant main effects of reboxetine concentration were found in the two series of experiments [F_(5,25) = 3.77, p < 0.05 and F_(6,30) = 119.86,p < 0.001, respectively]. Post hoc tests revealed a concentration-dependent increase in ³H overflow after superfusion with 3 to 100 μM reboxetine compared with control; 1 nM to 1 μM reboxetine did not increase ³H overflow compared with control. Concentrations of reboxetine that did not produce intrinsic activity, also did not inhibit nicotine-evoked³H overflow (Table 4;F_(1,11) = 3.00, p = 0.11 and F_(1,11) = 1.84,p = 0.49). Thus, reboxetine did not inhibit nicotine-evoked ³H overflow from superfused striatal slices preloaded with [³H]DA.

Reboxetine Inhibits Nicotine-Evoked ⁸⁶Rb⁺Efflux from Rat Thalamic Synaptosomes.

Only the highest concentration (100 μM) of reboxetine significantly [F_(6,18) = 13.95, p< 0.001] evoked ⁸⁶Rb⁺efflux above basal levels (Table 5). Thus, this high concentration was not included in the analysis of reboxetine-induced inhibition of nicotine-evoked⁸⁶Rb⁺ efflux (Fig.4). Reboxetine significantly inhibited nicotine-evoked ⁸⁶Rb⁺efflux [F_(6,18) = 6.77,p < 0.01] with an IC₅₀ value of 650 nM. Post hoc tests revealed that 1 and 10 μM reboxetine inhibited nicotine-evoked ⁸⁶Rb⁺efflux, compared with control.

Discussion

The present study demonstrates that reboxetine potently and selectively inhibits [³H]NE uptake into rat hippocampal synaptosomes. [³H]NE uptake was inhibited 3 to 4 orders of magnitude more potently than either [³H]DA or [³H]5-HT uptake. The present results demonstrating that reboxetine is a highly selective inhibitor of NET are consistent with previous findings (Montgomery, 1997; Wong et al., 2000). The current results, demonstrating that repeated injection of reboxetine does not alter its potency or selectivity to inhibit [³H]NE uptake, suggest that reboxetine does not produce adaptive modifications of NET, SERT, or DAT function under the current experimental conditions. Furthermore, tolerance did not develop to the reboxetine-induced inhibition of NET after repeated reboxetine administration.

The same reboxetine injection regimen (10 mg/kg/day i.p., twice daily for 14 days) used in the present ³H neurotransmitter uptake study was previously demonstrated to produce characteristic neural adaptive changes, e.g., down-regulation of β-adrenergic receptors (Harkin et al., 2000), indicating that this reboxetine injection regimen was sufficient to induce neural adaptation. Because reboxetine requires 2 to 3 weeks of treatment before its antidepressant effects are observed, direct inhibition of NET, which occurs immediately upon reboxetine treatment, cannot account entirely for its clinical efficacy. Rather, the neural adaptation must be, at least in part, responsible for the antidepressant effect of reboxetine. One possible neural adaptation may be a change in the surface expression of NET; however, an alteration in inhibition of [³H]NE uptake in the current study was not observed. Thus, the present results taken together with previous findings of Harkin et al. (2000) suggest that continued inhibition of NET by reboxetine may be necessary for the neural adaptation to be manifest.

Studies have demonstrated that the antidepressant bupropion has efficacy for smoking cessation and ameliorates some nicotine withdrawal symptoms (Hurt et al., 1997; Jorenby et al., 1999; Shiffman et al., 2000). Recently, bupropion and other antidepressants were reported to be antagonists of nAChRs (Hennings et al., 1997, 1999; Fryer and Lukas, 1999a,b; Slemmer et al., 2000). Bupropion inhibited rat α3β2, α4β2, and α7 nAChRs expressed in Xenopus oocytes (Slemmer et al., 2000). Bupropion, fluoxetine, sertraline, paroxetine, and nefazodone inhibited the α1βγδ subtype expressed in TE671/RD cells, the α3β4α5 ± β2 subtype expressed in SH-SY5Y neuroblastoma cells, and the chick V274T mutant α7 subtype expressed in SH-EP1 epithelial cells (Fryer and Lukas, 1999a,b). Desipramine, nisoxetine, and citalopram inhibited nicotine-evoked [³H]NE overflow from rat hippocampal slices (Hennings et al., 1997, 1999), suggesting an interaction with the α3β4* nAChR subtype. Based on these findings, it seems that different antidepressants may have different inhibitory profiles across the diversity of nAChR subtypes. Importantly, the specific nAChR subtype(s) inhibited may impart the pharmacological properties necessary for the inhibition of the reinforcing effect of nicotine. Thus, selected antidepressants may prove efficacious as smoking cessation pharmacotherapies, in part depending on their pharmacological profile as inhibitors of specific nAChR subtypes.

The present study provides an initial profile for the functional antagonism produced by reboxetine at several nAChR subtypes. Reboxetine potently inhibited (IC₅₀ value = 7.3 nM) nicotine-evoked [³H]NE overflow from superfused hippocampal slices, suggesting antagonism at the α3β4* subtype. Furthermore, the reboxetine-induced inhibition was not surmounted by increasing concentrations of nicotine, indicative of a noncompetitive mechanism of action. It is notable that reboxetine inhibited the function of this nAChR subtype within the same concentration range that it inhibited NET function. Inclusion of desipramine in the superfusion buffer in the [³H]NE release experiments eliminated the contribution of NET function to the observed effect of reboxetine to inhibit nicotine-evoked [³H]NE release. The functional elimination of NET provided the necessary conditions to focus on the contribution of nAChR function to the observed effect of reboxetine. Thus, the present observation that reboxetine potently inhibited nicotine-evoked [³H]NE release in the presence of desipramine strongly suggests the involvement of the α3β4* nAChR. In agreement, other investigators have reported that IC₅₀values for inhibition of nicotine-evoked [³H]NE overflow from rat hippocampal slices are not correlated withK_i values for inhibition of NET function (Hennings et al., 1999). Thus, both the current studies and the work of others suggest that there is no relationship between antidepressant-induced inhibition of NET function and antidepressant-induced inhibition of the effect of nicotine to release NE from its presynaptic terminal in hippocampus.

Furthermore, the present results suggest that reboxetine selectively inhibits the α3β4* nAChR subtype because reboxetine failed to inhibit nicotine-evoked [³H]DA release from superfused rat striatal slices in the presence of nomifensine, indicating a lack of interaction with the α3β2* subtype. Additionally, reboxetine did not inhibit binding of several nAChR radioligands, including [³H]epibatidine and [³H]MLA, indicating that reboxetine does not interact at the ligand binding site on either the α3β2* or α7* nAChR subtypes. Thus, neither functional inhibition nor inhibition of ligand binding induced by reboxetine was observed at the α3β2* nAChR site, suggesting pharmacological selectivity for reboxetine at the α3β4* subtype.

Regarding the α4β2* nAChR subtype, reboxetine also inhibited nicotine-evoked ⁸⁶Rb⁺efflux (IC₅₀ value = 650 nM) from thalamic synaptosomes, a functional assay for the α4β2* nAChR. The reboxetine-induced inhibition of the α4β2* subtype was 90-fold less potent than inhibition of the α3β4* subtype. Importantly, reboxetine did not inhibit [³H]cytisine binding to rat whole brain membranes, indicating that it does not interact at the ligand binding site on the α4β2* nAChR. Taken together, the present results from the⁸⁶Rb⁺ efflux and [³H]cytisine binding experiments suggest that reboxetine interacts with the α4β2* nAChR subtype in a noncompetitive manner, and does not directly interact with the specific site to which [³H]cytisine binds on the α4β2* protein. The present results do not rule out the involvement of NET in the reboxetine-induced inhibition of nicotine-evoked⁸⁶Rb⁺ efflux because desipramine was not included in the superfusion buffer at concentrations maximally inhibiting NET function. However, the concentration required to for inhibition of α4β2* function was 100-fold greater than that which inhibits NET function, mitigating against the role of NET in this effect of reboxetine. Furthermore, the present results obtained using native receptors are consistent with previous results demonstrating that antidepressants noncompetitively inhibit α4β2 nAChRs expressed in Xenopus oocytes in the absence of NET coexpression (Slemmer et al., 2000).

Pharmacological interventions as smoking cessation agents are restricted in number. Currently, the noncompetitive nAChR antagonist mecamylamine, various forms of nicotine replacement therapy, and the nonselective DA/NE transport inhibitor and nAChR antagonist bupropion have been reported to show limited efficacy. Recent studies demonstrate that reboxetine inhibits nicotine self-administration in rats and that tolerance does not develop to the inhibitory properties of reboxetine after repeated administration (A. S. Rauhut, S. N. Mullins, L. P. Dwoskin, and M. T. Bardo, submitted for publication). The latter results suggest that reboxetine may be a good candidate for the treatment of nicotine dependence and may have utility as a smoking cessation agent. Furthermore, the present study suggests that the α3β4* nAChR may be a novel target for therapeutic intervention in smoking cessation, and that the noradrenergic system may play a more important role than previously appreciated in nicotine-induced reward. Thus, given the concordance of smoking behavior and dysphoric mood disorders, reboxetine may provide an alternative effective treatment for tobacco smoking cessation.