Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

Neurotransmitter uptake is believed to be the primary mechanism in removing neuronally released monoaminergic transmitters from the synaptic cleft. The process occurs in the neuron as well as extraneuronally in the surrounding glial cells. In addition, specific NE uptake occurs in peripheral tissues such as smooth muscle, salivary gland, and the heart (Trendelenburg, 1988). Desipramine is an example of a clinically utilized antidepressant compound that selectively inhibits NE uptake. Tomoxetine (Wong et al., 1982) and nisoxetine (Wong et al., 1975) are phenoxyphenyl propylamines that have high affinity for the NE uptake site over the 5HT reuptake site with little affinity for other uptake sites or neurotransmitter receptors. By specifically inhibiting the neuronal uptake of NE, these compounds prolong the duration of responses to both exogenously applied NE as well as neuronally released NE. To date, all NE uptake inhibitors also inhibit the uptake of the metabolically derived transmitter, Epi (Fuller and Hemrick-Luecke, 1983).

Recently, we reported on the discovery of (R)-thionisoxetine (LY368975), an (R)-ortho-methylthio-phenoxy analog of tomoxetine and nisoxetine with improved potency and selectivity for the NE reuptake site (Gehlert et al., 1995). When compared to the effects of tomoxetine on ³H-NE uptake into synaptosomes, LY368975 exhibits approximately a 3-fold improvement in affinity with a K_i of 1.3 nM. Tomoxetine and LY368975 have similar low affinities for ³H-5HT uptake into synaptosomes from cerebral cortex. LY368975 was capable of antagonizing the 6-hydroxydopamine-induced depletion of NE and Epi from rat hypothalamus and metaraminol induced depletion of NE from rat heart and urethra after subcutaneous administration (Gehlert et al., 1995). Hypothalamically administered NE has well documented effects on food intake in animals (Grossman, 1960; Leibowitz, 1988), and a specific NE reuptake inhibitor has been reported to reduce food consumption in rats (Wong et al., 1993). In our study, we have examined the potency of LY368975 in several in vitro and in vivo paradigms designed to evaluate NE reuptake inhibition and explored its acute effects in animal feeding paradigms.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Antagonism of NE depletion in the mouse heart by 6-OHDA. Male standard (ICR) mice (18-22 g) were purchased from Harlan Sprague-Dawley Inc., Cumberland, IN. Animals were acclimated for at least 3 days before testing. Animals were housed at 22°C and given food and water ad libitum. LY368975 HCl was dissolved in 2.25% 2-hydroxypropyl--cyclodextrin in 0.01 N hydrochloric acid (pH adjusted to 6.0 with NaOH after compound dissolution) for oral gavage to mice at doses of 0.1 to 30 mg/kg. In a subsequent study, LY368975 was administered at 10 mg/kg 1 to 25 hr before the administration of 6-OHDA to assess the duration of action. 6-OHDA HBr was purchased from Sigma Chemical Co., St. Louis, MO and dissolved in 0.01N hydrochloric acid for i.p. injection at a dose of 7 mg/kg. Mice were killed 16 hr after 6-OHDA administration. Hearts were quickly dissected, blotted free of blood and frozen on dry ice prior to analysis. NE concentrations were determined in the samples by liquid chromatography with electrochemical detection (Fuller and Perry, 1977). Hearts were sonicated in 0.01 N trichloroacetic acid and centrifuged. The supernatant was placed into tubes containing alumina, the pH was adjusted to 8.6 with 0.5 M tris + 0.01 M EDTA and agitated for 10 min. The liquid was aspirated, 900 µl 0.2 N formic acid were added and the tubes agitated for 10 min. Samples (25 µl) were injected onto a Waters 4.6 × 150 mm Cosmosil 5C18-Ar column at a potential of 700 mV, a sensitivity of 5 nA/V and a flow rate of 1.5 ml/min. The elution buffer was 0.1 M monochloroacetic acid, 1 mM EDTA, 250 mg/liter SOS, 3% acetonitrile, 1% THF, pH 2.8.

Assessment of the effect of LY368975 on 5HT reuptake sites in rat brain. Male Sprague-Dawley rats (160-200 g) were purchased from Charles River Breeding Laboratories, Portage, MI. To determine the effect of LY368975 HCl on 5HT reuptake, LY368975 HCl was dissolved in 2.25% 2-hydroxypropyl--cyclodextrin in 0.01 N HCl acid (pH adjusted to 6.0 with NaOH after compound dissolution) and administered by gavage at 2 ml/kg as a 1 hr pretreatment. PCA HCl was purchased from Regis Chemical Co., Morton Grove, IL and dissolved in distilled H₂O for i.p. administration at 1 ml/kg. Rats were killed 2 hr after PCA injection, brains quickly removed and frozen on dry ice before assay. Brains were homogenized in approximately 12 volumes of 0.01 N trichloroacetic acid and centrifuged. Monoamines in brain supernatant were determined using high pressure liquid chromatography with electrochemical detection (Fuller and Perry, 1989). The supernatant was injected directly onto a Waters 4.6 × 150 mm Cosmosil 5C18-Ar column at a potential of 700 mV, a sensitivity of 5 nA/V and a flow rate of 1.5 ml/min. The elution buffer was 0.1 M monochloroacetic acid, 1 mM EDTA, 250 mg/liter SOS, 4.5% acetonitrile, 1% THF, pH 2.8. Statistical analyses were performed by analysis of variance using Dunnett's test to show significant differences at the P < .05 level.

Inhibition of NE uptake ex vivo. Sprague-Dawley rats (100-150 g; Charles Rivers Laboratories, Portage, MI) were housed in a room with a 12-hr dark/light cycle at 23°C and free access to Purina rat chow pellets and water. Rats, in groups of five, were treated with saline as control or at an indicated dose of LY368975 HCl either by s.c. or p.o. routes of administration. For oral dosing, rats were fasted overnight. At the indicated time interval following administration of LY368975 HCl, rats were killed by decapitation. Brains were quickly removed. For ³H-monoamine uptake, hypothalamus and striatum were dissected. Brain tissues were homogenized in ice-chilled 0.32 M sucrose. Aliquots of homogenates of hypothalamus (equivalent to 1 mg protein) were immediately used for ³H-5HT (Du Pont-NEN, Boston, MA) and ³H-NE (Du Pont-NEN) uptake, although striatal homogenates (equivalent to 0.5 mg protein) were used for ³H-DA (Du Pont-NEN) uptake. For binding studies, frontal cortex was dissected and frozen. Brain tissues were homogenized in ice cold Tris, HCl (50 mM, pH 7.4). Aliquots of cortical homogenate (equivalent to 300 µg protein) were used. Protein was determined by spectrophotometric method (Lowry et al., 1951). Uptake of ³H-monoamines in respective nerve terminals in tissue homogenates was conducted as described (Wong et al., 1993). Briefly, triplicate aliquots of tissue homogenates were introduced into 1 ml of Krebs bicarbonate medium which was saturated with 95% O₂/5% CO₂ and also contained 10 mM glucose, 0.1 mM iproniazid, 1 mM ascorbic acid, 0.17 EDTA and 1 µM ³H-monoamine. The uptake activity was initiated by an incubation at 37°C for 3 min. The reaction mixture was immediately diluted with 2 ml of ice-chilled 0.9% saline and filtered using Whatman GF/B filters, under vacuum with a cell harvester (Brandel, Gaithersburg, MD). Filters were rinsed twice with approximately 5 ml of saline and were transferred to a counting vial containing 10 ml scintillation fluid (Ready Protein+, Beckman Instruments, Inc. Fullerton, CA). Radioactivity was measured by a liquid scintillation spectrometer. Accumulation of radioactivity at 4°C represented the background accumulation and was subtracted from all samples. Binding to the NE and 5HT transporter was accomplished using ³H-tomoxetine (Du Pont-NEN) and ³H-paroxetine (Du Pont-NEN) as previously described (Wong et al., 1993). Statistical analyses were performed by analysis of variance using Dunnett's test to show significant differences at the P < .05 level.

Acute administration of LY368975 to rats deprived of food. Male Sprague-Dawley rats (Harlan Sprague-Dawley Inc., Indianapolis, IN) were individually housed in stainless steel cages in a temperature- and humidity-controlled animal colony room with a 12-hr light-dark cycle (lights on: 0700-1900 hr). Food and water were available ad libitum, and all testing was done in the home cage (0800-1600 hr). Rats were fasted for 18 hr before testing and allowed free access to water. Rats were first assigned to either a treatment or control group (N = 8), then weighed and administered drug or vehicle s.c. and returned to their home cage. Fifteen minutes later, food was made available to the animals. Food + food hopper was weighed at time 0, 1 hr, 2 hr and 4 hr after being made available to the animals. Grams of food consumed (+ spillage) by the treated animals at each time point was compared to food consumed (+ spillage) by the control animals using a one-way analysis of variance, with a Dunnett's post hoc test.

Acute administration of LY368975 to nonfasted rats trained to drink sweetened, condensed milk. Male Sprague-Dawley rats were acclimatized for 1 wk and individually housed in stainless steel cages at 72°C with lights on at 1300 hr and lights off at 0100 hr. Food was available ad libitum, and water was available at all times except during milk drinking sessions. After acclimatization, experiments were conducted between 1300 and 1500 hr. Rats were trained to drink sweetened, condensed milk, diluted 1:3 with water. Vehicle was injected s.c. daily at 1300 hr and milk placed onto home cages at 1400 hr. Rats were allowed to drink milk for 15 min per day. The volume of milk (+ spillage) was measured daily. When each rat drank between 40 and 50 ml/kg/15 min for 3 consecutive days, vehicle or LY368975 was injected s.c. at a volume of 1 ml/kg at doses of 0.1 to 3 mg/kg. Milk consumption and spillage was measured on test day and compared to the average amount of milk consumed on the 2 previous vehicle control days. Means and S.E.M.s for the ratios of milk consumed on test day versus the average of 2 control days are shown for five rats per group. Paired t tests determined significant differences in milk consumption on test day as compared to control days (P < .05), and analysis of variance using a post hoc Dunnett's test determined differences between milk consumption ratios as compared to vehicle control ratios (P < .05).

Effect of LY368975 on locomotor behavior in mice. Male CF-1 BR (Charles River Laboratories, Portage, MI) mice weighing 20 to 30 g at the time of testing were housed in groups of 17 in a large colony room with food and water were available continuously. The lights were on between 0600 and 1800. Studies were conducted between 0800 and 1630 in an adjacent 100m. Locomotor activity was measured with a 20-station Photobeam Activity System (San Diego Instruments, San Diego, CA) with seven photocells per station. Animals were weighed and placed individually in a polypropylene cage (40.6 × 20.3 × 15.2 cm). Thirty minutes after the animals were placed into the test cage they were removed, dosed orally and returned to their cage. Data were then collected for 60 min. Ten mice were used per group. Locomotor activity was recorded as the number of ambulations, where ambulation was defined as the sequential breaking of adjacent photocells. P values comparing the drug-treated groups to control groups were based on an analysis of variance with Dunnett's post hoc test. P < .05 was considered as significant. All computations were done using JMP v3.1 (SAS Institute Inc., Cary, NC).

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Effect of LY368975 on the depletion of NE by 6-OHDA. Figure 1A shows that 16 hr after p.o. injection of 6-OHDA, NE concentrations in mouse heart were significantly depleted by 83%. An hour after oral pretreatment, LY368975 HCl dose-dependently antagonized the 6-OHDA-induced depletion of NE with a 50% effective dose (ED₅₀) of 1.22 mg/kg, p.o. LY368975 HCl alone (open circle) at a dose of 30 mg/kg, p.o. resulted in a slight, yet significant, increase (19%) in NE concentrations 17 hr after gavage. In a comparable experiment, 1 hr after oral pretreatment, tomoxetine dose-dependently antagonized the 6-OHDA-induced depletion of NE with a ED₅₀ value of 7.68 mg/kg, p.o. (data not shown). In this experiment, tomoxetine alone at a dose of 30 mg/kg, p.o. had no significant effect on NE concentrations in mouse heart 17 hr after gavage.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   Effect of orally administered LY368975 HCl on the 6-OHDA-induced depletion of NE concentrations in mouse heart. A, LY368975 HCl or vehicle was administered by gavage at the doses shown 1 hr before administration of 6-OHDA at 7 mg/kg, i.p. Mice were killed 16 hr later. Mean value of control (± S.E.M.) is indicated by the top hatched bar and that of the 6-OHDA-treated group is indicated by the bottom bar. Means and S.E.M.s for six mice per group are shown. * Significantly different from control (P  .05). B, Effect of pretreatment time of orally administered LY368975 HCl on the 6-OHDA-induced depletion of NE concentrations in mouse heart. LY368975 HCl (10 mg/kg, p.o.) was administered by gavage at the times indicated before administration of 6-OHDA at 7 mg/kg, i.p. Mice were killed 16 hr later. Mean value of control is indicated by the top hatched bar and that of the 6-OHDA-treated group is indicated by the bottom hatched bar. Means and S.E.M.s for six mice per group are shown. * Significantly different from control (P  .05).

View larger version (66K): [in this window] [in a new window]   Fig. 2.   Failure of LY368975 HCl to antagonize the PCA-induced depletion of rat brain 5HT concentrations. Rats were pretreated 1 hr with 30 mg/kg, p.o. LY368975 HCl and killed 2 hr after 5 mg/kg, i.p. PCA. Means and S.E.M.s for five rats per group are shown. * Significantly different from control (P  .05).

Inhibition of NE uptake ex vivo. Rats were treated with LY368975 HCl at 0.03, 0.1, 0.3, 1 and 10 mg/kg s.c. One hour after administration, rats were decapitated and ³H-NE uptake into hypothalamic synaptosomes was measured. A significant reduction of ³H-NE uptake was found in groups treated with the three higher doses with an ED₅₀ was calculated to be 0.23 mg/kg s.c. (fig. 3A). To determine oral potency, ³H-NE uptake was measured in rats were treated with LY368975 at 1, 3, 10 or 30 mg/kg, p.o. Significant reduction of ³H-NE uptake in hypothalamic synaptosomes was observed in all treated groups as compared to the saline treated group (control) (fig. 3B). An ED₅₀ of 2.5 mg/kg was calculated, which together with the ED₅₀ of 0.23 mg/kg, s.c. (fig. 3A) yielded a p.o./s.c. ratio of 11 for inhibition of ³H-NE uptake.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Inhibition of 3H-NE uptake ex vivo by LY368975. A, A dose response study using s.c. route of administration. Rats (five per group) were treated with either saline (control) or LY368975 HCl at doses indicated 1 hr before death. Data are presented as mean percent ± S.E.M. of control value of 3H-NE uptake into hypothalamic synaptosomes (2.7 ± 0.2 pmol/mg protein). Significant difference from control is indicated as * P < .001. B, Dose response by oral route of administration.Rats (five per group) were treated with either saline (control) or LY368975 HCl at doses indicated one hour prior to sacrifice. 3H-NE uptake is expressed as mean percent ± S.E.M. of control values (2.3 ± 0.1 pmol/mg protein). * Significantly different from control (P  .001).

Assessment of ³H-tomoxetine and ³H-paroxetine binding ex vivo after oral administration. Rats were administered LY368975 at doses of 0.03, 0.1, 0.3, 1 and 10 mg/kg s.c. with s.c. administration or 1, 3, 10 and 30 mg/kg, p.o. and killed 1 hr after administration. Cerebral cortices were examined for the labeling of ³H-tomoxetine to the NE transporter and ³H-paroxetine to the 5HT transporter ex vivo. After s.c. dosing, a significant inhibition of ³H-tomoxetine binding was observed at 1 and 10 mg/kg with an estimated ED₅₀ of 0.7 mg/kg. In oral studies, doses of 3, 10 and 30 mg/kg produced a maximal inhibition of ³H-tomoxetine binding. An estimated ED₅₀ of 2.7 mg/kg was determined. After a single oral administration of LY368975 HCl at 30 mg/kg, groups of five rats were killed at 0.5, 1, 2, 6 and 16 hr. Binding of ³H-tomoxetine to the NE transporter ex vivo in cortical homogenates was significantly reduced to nearly 9% of control levels within 0.5 hr and the reduction persisted for 2 hr. By 6 hr, binding to the NE transporter returned to control levels (data not shown). Consistent with the relatively low potency of LY368975 HCl to inhibit uptake of 5HT in hypothalamic synaptosomes and ³H-DA uptake in striatal synaptosomes, there were no significant effects on [³H]-paroxetine binding during the entire time-course study after an oral administration of LY368975 HCl at 30 mg/kg (data not shown).

Effect of LY368975 on deprivation-induced feeding in rats. LY368975 significantly decreased feeding in a dose-dependent manner. At 3 and 10 mg/kg of LY368975 feeding was significantly less at 1 and 2 hr than those animals treated with vehicle only (fig. 4). By 4 hr, the 3-mg/kg effect was no longer different from vehicle whereas 10 mg/kg LY368975 still exhibited a suppression of food intake. At 1-hr tomoxetine treatment, feeding was significantly less in those animals treated with 3 and 10 mg/kg tomoxetine, than those animals treated with vehicle. At 2 hr, only the 10-mg/kg tomoxetine dose group was significantly different than vehicle (data not shown).

View larger version (15K): [in this window] [in a new window]   Fig. 4.   Effect of LY368975 on deprivation-induced food consumption. Rats were deprived of food for 18 hr and the effect of various doses of LY368975 (s.c.) administered 1 hr before the feeding period on food consumption assessed. Mean food consumption ± S.E.M. are shown for eight rats per group. * Significantly different from control (P  .05).

Effect of LY368975 on the consumption of sweetened, condensed milk by rats. LY368975 suppressed sweetened, condensed milk drinking in nonfasted rats at doses shown previously to inhibit NE uptake in rat brain ex vivo. One hour after the s.c. administration of LY368975 at 0.3, 1 and 3 mg/kg, s.c. milk drinking was significantly decreased 36, 40 and 44%, respectively (fig. 5). Milk drinking was unaffected by LY368975 at 0.1 mg/kg, s.c., a dose having no effect on NE uptake ex vivo.

View larger version (35K): [in this window] [in a new window]   Fig. 5.   Effect of LY368975 on consumption of sweetened condensed milk by rats. Compound was administered s.c. at the doses indicated 1 hr before presentation of the sweetened, condensed milk. Control animals drank 48.1 ± 4.3 g/kg/15 min. Means for the ratios for five rats per group are shown. * Significantly different from control (P  .05).

Effect of LY368975 on locomotor behavior in mice. In these studies, mice were administered LY368975 orally and the number of ambulations assessed for 60 min thereafter. Under these conditions, LY368975 HCl did not significantly alter locomotion over the dose range of 0.1 to 10 mg/kg p.o. (fig. 6).

View larger version (34K): [in this window] [in a new window]   Fig. 6.   Lack of effect of LY368975 HCl on locomotor activity in mice. Each point represents the mean of 10 animals. The vertical bars represent ± S.E.M.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

LY368975 HCl, 3-ortho-thiomethylphenoxy-N-methyl-3-phenylpropylamine, is an analog of nisoxetine (ortho-methoxy substituted) and tomoxetine (ortho-methyl-substituted). These three ortho-substituted compounds share the common properties of being potent and selective inhibitors of NE uptake in presynaptic terminals of rat brains in vitro and in vivo (Gehlert et al., 1995; Wong and Bymaster, 1976; Wong et al., 1975, 1982). To explore the in vivo potency of LY368975, the compound was administered to mice before the injection of 6-OHDA. The neurotoxin, 6-OHDA, uses the uptake carrier to gain entry into noradrenergic neurons, ultimately leading to neuronal destruction and neurotransmitter depletion. LY368975 HCl inhibited the NE transporter in mice in vivo, as indicated by antagonism of the 6-OHDA induced depletion of NE concentrations in mouse heart. LY368975 HCl was more potent than tomoxetine in blocking the 6-OHDA-induced depletion of heart NE after oral administration (ED₅₀s = 1.2 mg/kg, p.o. and 7.7 mg/kg, p.o., respectively). LY368975 HCl at 10 mg/kg, p.o., a dose completely effective in antagonizing the depletion of heart NE by 6-OHDA, was relatively long-lasting, affording complete protection for the depleting effects of 6-OHDA for up to 6 hr with partial protection at 8 hr. Tomoxetine, administered at 10 mg/kg, p.o., only partially protected against the depletion of heart NE by 6-OHDA at 1 and 2 hr, but not at 4 hr or longer. We had previously shown that LY368975 HCl inhibited the NE transporter in rat brain, as indicated by the antagonism of the depletion of hypothalamic NE by intracerebroventricularly injected 6-OHDA (Gehlert et al., 1995). The ED₅₀ for LY368975 HCl was 0.21 mg/kg, s.c. In those studies, LY368975 HCl also inhibited the transporter on Epi-containing neurons in rat hypothalamus. These findings are not unexpected, because it has been previously shown that Epi depletion, as well as NE depletion, in brain is blocked by tomoxetine (Fuller and Hemrick-Luecke, 1983), and by other uptake inhibitors known to block the NE transporter (Fuller, 1982; Tessel et al., 1978). To evaluate the effects of LY368975 on the 5HT transporter, we used the serotonergic neurotoxin, PCA. PCA gains entry into serotonergic neurons via the 5HT reuptake carrier, destroys the neuron and depletes neurotransmitter levels. 5HT reuptake inhibitors can block the PCA-induced depletion of brain 5HT (Fuller et al., 1974). Using this experimental paradigm, we found that LY368975 HCl was ineffective at doses that block 6-OHDA-induced NE and Epi depletion in a comparable paradigm (Gehlert et al., 1995), suggesting LY368975 is selective in vivo in blocking catecholamine uptake.

LY368975 HCl inhibited NE uptake in hypothalamic homogenates ex vivo with ED₅₀ values of 0.23 mg/kg, s.c. and 2.53 mg/kg, p.o. LY368975 also inhibited the binding of ³H-tomoxetine to the NE transporter in membranes of cerebral cortex with ED₅₀ values of 0.73 mg/kg, s.c. and 2.7 mg/kg, p.o. Thus the p.o./s.c. ratios for the measurement of uptake and transporter of NE are 11 and 3.7, respectively. Previously published data on tomoxetine (Wong et al., 1982) were generated using i.p. route of administration and thus can offer only relative information for comparison. Tomoxetine, however, inhibited NE uptake in hypothalamic homogenates with an ED₅₀ of 3.08 ± 1.01 mg/kg, i.p. Therefore, LY368975 HCl with an ED₅₀ of 2.53 mg/kg, p.o. is comparable in potency if not more potent than tomoxetine as an inhibitor of NE uptake ex vivo in rats. Within 0.5 hr after oral administration of LY368975 HCl at 30 mg/kg, a significant 61% inhibition of ³H-NE uptake was observed and that relative magnitude of inhibition was maintained for 2 hr. At 6 hr, ³H-NE uptake was significantly inhibited by 28%, but recovered to the level of the control group by 16 hr. In total, these studies document that LY368975 HCl is a potent, long-lasting and selective inhibitor of the NE transporter in vivo.

To determine the effects of selective NE reuptake inhibitors on feeding, we tested LY368975 in two acute models of animal consumption. In the first model, rats were deprived of food for 18 hr and the LY368975 was administered s.c. 1 hr before the reintroduction of food. A significant reduction in food intake was observed at the 10 mg/kg dose at 1, 2 and 4 hr thereafter. No effect on food consumption occurred at a 1-mg/kg dose of LY368975. At 3 mg/kg, LY368975 produced a reduction in food consumption at the 1- and 2-hr time points. These doses are consistent with the doses required to produce maximal NE uptake inhibition in vivo and ex vivo. Subsequently, the compound was tested for its ability to affect the consumption of sweetened, condensed milk in nonfasted rats. Doses of LY368795 from 0.3 to 3 mg/kg, s.c. significantly suppressed sweetened condensed milk drinking. These doses are similar to those reported to significantly inhibit NE and Epi uptake in rat brain in vivo (Gehlert et al., 1995). In addition, the response mimics that seen in the ex vivo studies (fig. 3A). Thus, LY368975 suppresses milk consumption in rats at doses that fully inhibit brain NE uptake in vivo. Finally, to evaluate the effects of LY368975 on the general behavior of mice, we observed locomotor behavior in mice after various doses of LY368975. In these studies, no significant increase in locomotion was observed suggesting the compound is not activating like amphetamines. Conversely, no significant decrease in activity was observed suggesting the compound was not producing a reduction in neuromuscular function or compromising the animal's ability to seek food.

When directly injected into the brain, NE can induce eating behavior (Grossman, 1960). At the receptor level, NE has opposing actions at the alpha-1 and alpha-2 receptors. In the paraventricular nucleus of the hypothalamus, NE stimulates feeding through an action at the alpha-2 (Leibowitz, 1988) although the alpha-1 receptor mediates an inhibition of feeding (Wellman and Davies, 1991). The relative balance of the stimulatory and inhibitory actions of NE may be dependent on circulating corticosterone concentrations (Leibowitz, 1988). Only limited results have been published with other selective NE reuptake inhibitors. Nisoxetine, a structurally related selective NE reuptake inhibitor has been reported to reduce consumption after food restriction (Wong et al., 1993). However, nisoxetine administered orally at 30 mg/kg before the dark cycle in rodents, did not produce a reduction in food intake (Jackson et al., 1997). When nisoxetine was administered in combination with the 5HT reuptake inhibitor, fluoxetine, a significant reduction was observed. Because we did not test LY368975 in this paradigm, it is difficult to speculate the reasons for these differences. Because LY368975 has substantially higher potency when compared to nisoxetine (Gehlert et al., 1995), a 30-mg/kg dose of this compound may have more profound effects on the NE transporter than a similar dose of nisoxetine.

In conclusion, LY368975 is a potent and long lasting inhibitor of NE reuptake in vivo. At doses consistent with NE reuptake site inhibition in vivo, LY368975 produces a reduction in food intake using several models of rodent food consumption without a significant effect on locomotor activity. Therefore, selective NE reuptake inhibitors may be useful in the treatment of obesity and eating disorders.