SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Objective

Anti-obesity drugs have adverse effects which limit their use, creating a need for novel anti-obesity compounds. We studied effects of dopamine (DA) and norepinephrine (NE) reuptake inhibitor bupropion (BUP), alone and after blocking α1- or α2-adrenoceptors (AR), D1/5, D2/3, or D4 receptors, to determine which receptors act downstream of BUP.

Design and Methods

Effects on caloric intake, meal patterning and locomotion were assessed, using an automated weighing system and telemetry in male rats with 18-h access to Western Human style diet.

Results

BUP (30 mg/kg) induced hypophagia by reducing meal size and postponing meal initiation. WB4101 (α1-AR; 2 mg/kg) and imiloxan (α2B-AR; 5 mg/kg) attenuated BUP's effect on meal size, while WB4101 and BRL 44408 (α2A/D-AR; 2 mg/kg) counteracted effect on meal initiation. Atipamezole (α2-AR; 1 mg/kg) and imiloxan further postponed initiation of meals. SKF 83566 (D1/5; 0.3 mg/kg), raclopride (D2/3; 0.5 mg/kg) and to a lesser extent FAUC 213 (D4; 0.5 mg/kg), attenuated BUP-induced hypophagia. BUP stimulated locomotion, which was blocked by all antagonists, except FAUC 213 or BRL 44408.

Conclusions

Alpha1-, α2A/D- and α2B-ARs, and DA receptors underlie BUP's effects on size and initiation of meals, while blocking pre-synaptic α2-ARs enhanced BUP-induced hypophagia. An inverse agonist of (pre-synaptic) α2A-ARs could enhance BUP-induced anorexia and treat eating disorders and obesity.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

The brain monoamines, norepinephrine (NE) and dopamine (DA) are interesting drug targets for the treatment of obesity. NE is known to regulate food intake, and a NE reuptake inhibitor has been reported to reduce cumulative intake of chow or milk in non-deprived and deprived male rats [1]. Adrenoceptors (ARs) mediate food intake differently, such that activation of α1- and β2-ARs decreases caloric intake [2], while activation of α2-ARs increases it [3]. Similarly to a NE reuptake inhibitor, the DA reuptake inhibitor vanoxerine reduces 1-h cumulative feeding of palatable mash in rats [4] and DA receptor activation inhibits food intake in rats [5]. DA is involved in satiety and reinforcing effects of food, and eating induces DA release in the striatum, which increases pleasantness of the meal [6]. NE and DA strongly interact in many brain functions, and rodent data among others [7, 8] suggest that simultaneous targeting of NE and DA may offer a novel approach for regulation of energy balance and body weight.

In line with findings from rodents [8], the DA and NE reuptake inhibitor bupropion has been reported to cause mild weight loss in obese humans [9] and this inhibitor is currently being evaluated for a combination treatment for obesity [10]. In rats, findings from bupropion's effects on food intake are inconsistent, for example, high doses of bupropion have been shown to reduce feeding of chow in meal-fed male rats [11], whereas in females it did not affect feeding [12]. In addition, not much is known of bupropion's effects on meal patterning (meal size and intermeal interval), a mechanism that is important for the regulation of energy intake and body weight. Moreover, next to feeding, other factors seem to contribute to bupropion-induced body weight loss as well, i.e., bupropion has been suggested to increase energy expenditure, as shown by increased locomotor activity [8, 11] and thermogenesis [8, 13].

So far, little is known about the adrenoceptor (AR) and DA receptor subtypes mediating bupropion's effects on energy intake and energy expenditure. We here administered α1-AR, non-selective α2-AR, selective α2A/D-AR (human subtype A that equals to D in rodents), α2B-AR antagonist, and D1/5, D2/3 and D4 receptor antagonist, alone or prior to bupropion, and studied effects on meal patterning, total caloric and water intake and locomotor activity in male Wistar rats. We used an automated 24-h monitoring system which gave detailed, continuous, and synchronous information about meal size and meal patterning, drinking behavior, and locomotion without disturbing the rat in its home cage. The telemetry system allowed monitoring of locomotor activity in home cages thus avoiding exploratory activity in novel environment as a confounding factor. Drugs were administered via an i.p. cannula that allowed administration of a drug without restraining or disturbing the rat and its normal feeding.

The AR and DA receptor antagonists used are known to penetrate the brain, and the doses were selected such that they do not significantly affect total caloric intake on their own. The antagonist doses or dose ranges were first pre-selected from literature with the criteria for being centrally effective in vivo and for being adequately selective for certain receptor subtype [WB4101 1-5 mg/kg [14]; Atipamezole 1 mg/kg (Jukka Sallinen, Orion Pharma, personal communication); BRL 44408 0.5-5.0 mg/kg [15, 16]; imiloxan 5 mg/kg [17], SKF 83566 0.04-0.15 mg/kg and raclopride 0.1-0.4 mg/kg [18], FAUC 213 0.5 mg/kg (our preliminary studies)]. Second, for further dose selection we carried out in-house studies, partially published in Ref. 19, with the criteria that the dose should not markedly alter size of a meal or total caloric intake on its own. Indeed, none of antagonist doses had marked effects on total 18-h caloric intake.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Animals

Single-housed male Wistar rats (220-250 g) (Crl-Wu, Germany) were fed a Western Human Style diet (10% saturated fat blend; 20% protein; 60% carbohydrates; 3.94 kcal/g, AB-Diets, Woerden, the Netherlands) with water available ad libitum. Animals were kept in a temperature- and humidity-controlled room (21 ± 2°C) under a 12-h light/dark cycle (lights on: 04:00 h to 16:00 h). All experimental procedures were approved by the ethical committee for animal experimentation (University of Utrecht) and conducted under Dutch and European law (Wet Dierproeven, 1996; Guideline 86/609/EEC).

Surgical procedure

The experimental protocol was previously described [19]. One week after arrival, rats received i.p. dosing cannulae (Raumedic, Helmbrechts, Germany) and transmitters (TA10TA-F40 Data Sciences International, St. Paul, MN) in the abdominal cavity under fentanyl/fluanisone (1 mL/kg, i.m.; Hypnorm®, Janssen Pharmaceutica, Beerse, Belgium) and midazolam (0.5 mL/kg, i.p.; Dormicum® 50 mg/mL, Hoffman-LaRoche, Mijdrecht, the Netherlands) anesthesia. Cannula were externalized on the head, attached to a connector pedestal (Plastics One, Roanoke, USA), and fixed to the skull. After surgery, rats were given carprofen (0.1 mL/kg, s.c.; Rimadyl® 50 mg/mL, Pfizer Animal Health, Capelle a/d Ijssel, the Netherlands) and saline (6 mL/kg, s.c.), and allowed 2-week recovery.

Experimental setup

Drugs were administered via an i.p. cannula without disturbing the rat and normal feeding. The cannula were kept open with 0.3 mL saline injections every 2-3 days. Our preliminary studies showed that unlike conventional i.p. administration such cannula did not cause visceral stress or disturb food intake. An automated meal-pattern system monitored food intake over 24 h without disturbing rat in its home cage. Two weeks post-surgery, eating and drinking patterns were recorded daily. One week before and throughout the experimental period the rats were under a restricted feeding schedule, with food removed daily at 10:00 h and returned at 16:00 h (i.e., the beginning of the dark period), to limit effects of meals taken in the light phase on first meals in the dark phase (and to avoid that rats take a meal just before dark onset). The rats then had 18-h ad libitum access to food until 10:00 h the next day when food was again removed. Water was continuously available. Food hoppers were automatically weighed, and data were sent to a computer every 12 s using the program “Scales” (Department Biomedical Engineering, UMC Utrecht, the Netherlands). A meal was defined as an episode of food intake with a minimal consumption of 0.5 g of chow and an intermeal interval of 5 min. The telemetry system allowed monitoring of locomotion avoiding novel environment-induced exploratory activity. Telemetry transmitters were switched on by magnetic field induction at two weeks after the surgery. Locomotor activity was automatically and continuously recorded via a transmitter that sent digitized data via radio frequency signals to a nearby receiver. These data were recorded every 10 min using DSI software (DSI, St. Paul, MN).

After the rats had adjusted to the restricted feeding, baseline values of all parameters were assessed on three days after a daily 0.3-mL vehicle injection. Drugs were acutely administered, in random order, at 15:30 h, 30 min before returning the food. Water consumption was measured daily at 10:00 h. Each acute drug administration was followed by a 2- to 5-day washout period before a new drug was administered.

A washout period of 2 days was considered adequate to prevent any long-term effects or drug interactions for most test compounds because either their (or their active metabolites') half-life was between 1 to 3 h or duration of pharmacological action in the brain was a few hours in rats [17, 20-23]. After WB4101, BRL 44408, and SKF 83566 washout period was longer (3-5 days) because their half-lives were not well described in literature or might have been longer than 3 h [15, 16, 24, 25]. A total of 19 rats were used in these experiments.

Drugs

Bupropion HCl (Sigma-Aldrich, Taufkirchen, Germany) was dissolved in physiological saline (0.9% NaCl). WB4101 HCl (Sigma), atipamezole HCl (Orion Pharma, Espoo, Finland), BRL 44408 maleate (Tocris, Bristol, UK), imiloxan HCl (Tocris), SKF 83566 HBr (Tocris), raclopride tartrate (Sigma), and FAUC 213 (Sigma) were dissolved in sterile water, which served as vehicle in baseline measurements. FAUC 213 was dissolved by rapidly sonicating, warming up in water bath and shaking vigorously. All solutions were freshly prepared the day of use and administered in a volume of 1-2 mL/kg. All doses refer to base.

Data analysis

Data are presented as group mean ± SEM, and each rat served as its own control in data analyses. The effects of bupropion and all antagonists on meal patterning were assessed with three-way mixed ANOVA (bupropion × antagonist × number of meal 1-6) (SPSS Statistics 19.0). The effects on cumulative caloric intake, meal frequency, size of meals, and locomotion were assessed with two-way independent ANOVA (bupropion × antagonist). Simple main effects analyses for feeding parameters were conducted between vehicle, bupropion and an antagonist, on its own and in combination with bupropion. If the condition of sphericity was not met, Greenhouse-Geisser correction was used. When appropriate (P < 0.05), the ANOVA was followed by Dunnett's post-hoc test.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

The effects of bupropion and adrenoceptor antagonists on feeding

A two-way ANOVA revealed significant main effects for antagonist pre-treatment and bupropion treatment on the 2-h caloric intake and mean size of a meal [pre-treatment: 2-h intake F(7,104) = 2.02, P = 0.054; mean meal size F(7,104) = 2.08, P = 0.052; treatment: 2-h intake F(1,104) = 43.73, P < 0.001; mean meal size F(1,104) = 19.95, P < 0.001]. There was also a significant pre-treatment × treatment-interaction for the same parameters [2-h intake: F(7,104) = 3.19, P=0.004; mean meal size: F(7,104) = 3.22, P = 0.004]. ANOVA revealed no significant interactions between the effects of antagonists and bupropion on total 18-h caloric intake or meal frequency.

Simple main effects analysis between vehicle, bupropion and an antagonist, on its own and in combination with bupropion, revealed that bupropion (30 mg/kg) significantly reduced caloric intake during the first 2 h to 49%, when compared to vehicle baseline values [Figure 1B; F(3,29) = 9.66, P < 0.001, post-hoc P < 0.01]. On their own, AR antagonists had no marked effects on 2-h caloric intake (Figure 1A). When given prior to bupropion, the α1-AR antagonist WB4101 (2 mg/kg) and the α2B-AR antagonist imiloxan (5 mg/kg) partially counteracted bupropion-induced hypophagia, such that 2-h caloric intake was 70 and 63% and no longer significantly different from baseline value (Figure 1B). Pre-treatment with the α2-AR antagonist atipamezole (1 mg/kg) or the α2A/D-AR antagonist BRL 44408 (2 mg/kg) did not alter bupropion's effect on 2-h caloric intake and the values differed significantly from baseline or the antagonist alone [Figure 1B; ATI+BUP F(3,28)=3.25, P = 0.039, post-hoc P < 0.05 versus VEH/ATI; BRL+BUP F(3,29) = 7.16, P = 0.001, post-hoc P < 0.01 versus VEH/BRL].

image

Figure 1. The relative changes in 2-h caloric intake after the administration of α1- or α2-AR (adrenoceptor) or dopamine (DA) D1/5, D2/3, and D4 receptor antagonists, on their own (A) or prior to bupropion (B; BUP, 30 mg/kg). Below are shown antagonists and their primary target receptors, and BUP treatment (B). AR antagonists were: WB4101 (WB, 2 mg/kg, α1), atipamezole (ATI, 1 mg/kg, α2), BRL 44408 (BR, 2 mg/kg, α2A/D), imiloxan (IMIL, 5 mg/kg, α2B). DA antagonists were: SKF 83566 (SKF, 0.3 mg/kg, D1/5), raclopride (RAC, 0.5 mg/kg, D2/3), and FAUC 213 (FAUC, 0.5 mg/kg, D4). Food was removed for the last 6 h of the light period, and the antagonists were administered via an i.p. cannula 15 min before BUP and 45 min before returning the food in the beginning of the dark period. Feeding was investigated during the first 2 h after returning the food (16:00–18:00) using an automated system (for details see “Methods and Procedures” section). Baseline values (=100%) were measured for each rat after vehicle administration. Data expressed as percentage change from the baseline, mean ± SEM, n = 7–10 for all groups. ANOVA followed by Dunnett's post-hoc test: * P < 0.05, ** P < 0.01 versus vehicle baseline (the first column), # P < 0.05 versus BUP, % P < 0.05, %% P < 0.01 versus antagonist alone in A.

Download figure to PowerPoint

None of the compounds alone markedly altered total 18-h caloric intake, which is likely due to their relatively short half-lives in rat (data not shown; for more details see “Methods and Procedures” section). Bupropion reduced 18-h caloric intake to 83%, and the antagonists had no significant effects on it, although atipamezole and imiloxan further reduced it to 68 and 66%.

The effects of bupropion and dopamine receptor antagonists on feeding

In general, feeding is sensitive to very low doses of DA antagonists, and already 1/10th i.p. doses than those used here (0.0375 mg/kg of SKF 83566 and 0.05 mg/kg of raclopride) have been found to significantly increase chow intake [18]. Consistently, in our preliminary studies, already low doses of DA antagonists tended to increase 2-h caloric intake; SKF 83566 0.1 mg/kg increased it to 114% (±19%, SEM), raclopride 0.2 mg/kg to 143% (±40%) and FAUC 213 0.2 mg/kg to 115% (±20%) (data not shown). In this study, SKF 83566 (D1/5; 0.3 mg/kg), but not the other two DA antagonists, significantly increased 2-h caloric intake when given on its own [Figure 1A; F(3,30) = 5.58, P = 0.004, post-hoc P < 0.05 versus VEH].

When given prior to bupropion, which reduced 2-h caloric intake to 49%, SKF 83566 (D1/5; 0.3 mg/kg) and FAUC 213 (D4; 0.5 mg/kg) partially counteracted bupropion-induced hypophagia and almost doubled 2-h caloric intake to 80%, so that it no longer significantly differed from baseline (Figure 1B). Two-hour caloric intake by SKF 83566+bupropion significantly differed from SKF 83566 alone [Figure 1B; F(3,30) = 5.58, P = 0.004, post-hoc P < 0.01]. When given after the D2/3 antagonist raclopride (0.5 mg/kg), bupropion failed to reduce caloric intake and the combination resulted in hyperphagia that was comparable to that by raclopride alone and significantly different from bupropion alone [Figure. 1B; F(3,29) = 5.09, P = 0.007, post-hoc P < 0.05].

The only compounds that were able to reduce 24-h water intake were SKF 83566 (D1/5) and raclopride (D2/3) which on their own significantly reduced water intake by 24% (data not shown; P < 0.01). Reduction in water intake paralleled to the reduction in meal frequency (longer intermeal intervals), supporting findings that drinking behavior follows each meal in rats [19].

The effects of bupropion and antagonists on size of meals

Figure 2 includes meal patterning that consists of relative sizes (compared to baseline value) of the six meals eaten during the dark phase (16:00 h-04:00 h) and time points when each meal was initiated. Regarding size of meals, three-factor mixed ANOVA revealed a significant main effect for bupropion treatment [F(1,7) = 20.42, P = 0.004] and meal number 1 to 6 [F(5,30) = 5.44, P = 0.001], but only a tendency for antagonist pre-treatment [F(7,42) = 1.92, P = 0.090]. There was a significant pre-treatment × treatment interaction [F(7,42) = 2.40, P = 0.037] and treatment × meal number interaction [F(5,30) = 4.49, P = 0.004], but no significant pre-treatment × meal number interaction. In general, larger first meal was followed by two smaller meals and thereafter by three larger meals, so that meals 5 and 6 were significantly larger than meal 3 (P = 0.001 and 0.044). Bupropion-treated rats consumed significantly smaller meals than vehicle-treated rats (P = 0.004), and this was affected differently by the number of consumed meals such that particularly the first three meals were affected by bupropion (Figure 2). The effects of antagonists on size of meals were different in bupropion-treated than in vehicle-treated rats. There was also a significant pre-treatment × treatment × meal number interaction [F(35,210) = 1.93, P = 0.003], suggesting that the combined effect of bupropion treatment and meal number was different between different antagonist pre-treatments. Indeed, pre-treatment with AR antagonists did not markedly alter bupropion's effect on size of the first three meals, while two dopamine antagonists clearly attenuated it (Figure 2).

image

Figure 2. Meal patterning consisting of a relative meal size of the first six, or all consumed meals if less than six, and mean time point (hours) when a meal was initiated during the dark period. Food was removed for the last 6h of the light period and drugs were administered via an i.p. cannula at 45 min (antagonists) or 30 min (bupropion) before returning the food at dark onset. Adrenoceptor (A; AR) antagonists and their targets were: WB4101 (WB, 2 mg/kg, α1), atipamezole (ATI, 1 mg/kg, α2), BRL 44408 (BRL, 2 mg/kg, α2A/D), imiloxan (IMIL, 5 mg/kg, α2B). Dopamine (B; DA) antagonists and their targets were: SKF 83566 (SKF, 0.3 mg/kg, D1/5), raclopride (RACLO, 0.5 mg/kg, D2/3), and FAUC 213 (FAUC, 0.5 mg/kg, D4). Meal patterning was investigated using an automated system (for details see “Methods and Procedures” section). Baseline values (=100%) were measured for each rat after vehicle (VEH) administration. Data expressed as percentage change from baseline, mean ± SEM, n = 7–10. Meal size *P < 0.05, **P < 0.01 versus corresponding meal size during VEH baseline; # P < 0.05 versus corresponding meal size by BUP; % P < 0.05 versus corresponding meal size by BUP + antagonist; Initiation time of a meal ¤ P < 0.05, ¤¤ P < 0.01, ¤¤¤ P < 0.001 versus corresponding initiation time point during VEH baseline; ‡‡ P < 0.01 versus corresponding initiation time point by BUP; $ P < 0.05, $$ P < 0.01, $$$ P < 0.001 versus Corresponding initiation time point by BUP+antagonist.

Download figure to PowerPoint

In detail, bupropion (30 mg/kg) significantly reduced the size of the first three meals to 44-73% as compared to baseline values [Figure 2A; meal 1 F(3,30) = 8.61, P < 0.001, post-hoc P < 0.01; meal 2 F(3,30) = 3.45, P = 0.030; meal 3 F(3,30) = 5.61, P = 0.004, post-hoc meals 2 and 3 P < 0.05]. Concerning pre-treatment with AR antagonists, WB4101 (α1-AR; 2 mg/kg) and imiloxan (α2B-AR; 5 mg/kg) partially attenuated bupropion's effect on the size of the first meal and increased it from 50% to 84 and 62% (Figure 2A). The effect of WB4101 + bupropion on the size of the first three meals significantly differed from that by WB4101 alone [Figure 2A; meal 1 F(3,30) = 8.61, P < 0.001; meal 2 F(3,30) = 3.45, P = 0.030; meal 3 F(3,30) = 5.61, P = 0.004; post-hoc meals 1, 2, and 3 P < 0.05 vs. WB, meal 3 P < 0.01 vs. VEH]. Atipamezole (α2-AR; 1 mg/kg) and BRL 44408 (α2A/D-AR; 2 mg/kg) did not alter size of meals either alone or in combination with bupropion, and occasional meals remained significantly smaller than corresponding meals during baseline or after an antagonist alone [Figure 2A; ATI+BUP meal 1 F(3,28) = 4.52, P = 0.012, post-hoc P < 0.05 vs. VEH/ATI; meal 5 F(3,26) = 3.62, P = 0.028, post-hoc P < 0.05 vs. VEH; BRL+BUP meal 1 F(3,30) = 6.48, P = 0.002, post-hoc P < 0.05 vs. VEH/BRL; meal 2 F(3,30) = 4.71, P = 0.009, post-hoc P < 0.05 vs. VEH].

Pre-treatment with SKF 83566 (D1/5; 0.3 mg/kg) partially and raclopride (D2/3; 0.5 mg/kg) completely abolished bupropion's effect on size of meals (Figure 2B). The effect of SKF 83566 + bupropion on the size of the first three meals significantly differed from SKF 83566 alone, but not from bupropion or baseline [Figure 2B; meal 1 F(3,31) = 8.03, P = 0.001; meal 2 F(3,31) = 2.96, P = 0.050; meal 3 F(3,31) = 3.76, P = 0.022; post-hoc meals 1, 2, and 3 P < 0.05 vs. SKF]. Raclopride (D2/3; 0.5 mg/kg) had a similar hyperphagic effect on the size of meals alone and in combination with bupropion, so that it abolished bupropion-induced hypophagia and significantly increased size of the second meal, as compared to bupropion alone [Figure 2B; F(3,30) = 2.90, P = 0.050, post-hoc P < 0.05]. Pre-treatment with FAUC 213 (D4; 0.5 mg/kg) only modestly counteract bupropion's effect on the size of the first three meals, which remained significantly different from basal values or FAUC 213 alone [Figure 2B; meal 1 F(3,30) = 3.81, P = 0.021, post-hoc P < 0.05 vs. VEH/FAUC; meal 2 F(3,30) = 3.96, P = 0.018; meal 3 F(3,30) = 3.01, P = 0.047; post-hoc meals 2 and 3 P < 0.05 vs. FAUC].

The effects of bupropion and antagonists on meal initiation

Regarding initiation time of meals, three-factor mixed ANOVA revealed a significant main effect for antagonist pre-treatment [F(1.80,12.61) = 12.96, P = 0.001], bupropion treatment [F(1,7) = 7.62, P = 0.028] and meal number 1 to 6 [F(1.61,11.28) = 9.82, P = 0.005]. There was a significant pre-treatment × meal number interaction [F(2.90,20.30) = 36.82, P < 0.001], but no significant pre-treatment × treatment or treatment × meal number interactions. There was a significant pre-treatment × treatment × meal number interaction [F(2.90,20.30) = 4.62, P = 0.013]. Bupropion-treated rats thus initiated meals significantly later than vehicle-treated rats (P = 0.028), and the antagonist-pre-treated rats showed varying effects on initiation of meals, but depending upon which meal it was and if the antagonist was given prior to bupropion.

Bupropion (30 mg/kg) postponed initiation of meals, such that the first meal was initiated 1.2 h later than the first meal during baseline, which consequently postponed initiation of following meals as intermeal intervals remained unaltered (Figure 2). The AR antagonists had no significant effects on meal initiation on their own but had varying effects on it when given prior to bupropion (Figure 2A). Pre-treatment with WB4101 (α1-AR; 2 mg/kg) and BRL 44408 (α2A/D-AR; 2 mg/kg) partially counteracted bupropion's effect on initiation of meals. They advanced initiation from the first or second meal onwards and shortened meal intervals during the first 3 h independent of the size of a prior meal (Figure 2A). On the contrary, pre-treatment with atipamezole (α2-AR; 1 mg/kg) or imiloxan (α2B-AR; 5 mg/kg) further enhanced bupropion's effect on initiation of meals. Atipamezole + bupropion significantly postponed initiation of first and second meal, as compared to baseline, bupropion or atipamezole alone [Figure 2A; meal 1 F(3,28) = 12.79, P < 0.001, post-hoc P < 0.001 vs. VEH/ATI, P < 0.01 vs. BUP; meal 2 F(3,28) = 4.08, P = 0.017, post-hoc P < 0.05 vs. VEH/ATI]. Also imiloxan+bupropion significantly further postponed initiation of the first meal as compared to baseline or imiloxan alone [Figure 2A; F(3,29) = 7.83, P = 0.001, post-hoc P < 0.01 vs. VEH/IMIL].

The effects of DA antagonists on meal initiation were similar on their own and in combination with bupropion (Figure 2B). After raclopride (D2/3; 0.5 mg/kg) + bupropion, initiation of meals was postponed from the third meal onwards and the effect on initiation of the third meal reached significance, as compared to baseline [Figure 2B; F(3,30) = 3.05, P = 0.035, post-hoc P < 0.05 vs. VEH]. Pre-treatment with FAUC 213 (D4; 0.5 mg/kg) even modestly further postponed initiation of three meals from that by bupropion alone, such that the first meal was initiated significantly later than the corresponding meal during baseline or after FAUC 213 alone [Figure 2B; F(3,30) = 5.72, P = 0.004, post-hoc P < 0.05 vs. VEH/FAUC]. SKF 83566 (D1/5; 0.3 mg/kg) somewhat advanced initiation of the first meal, and data analyses revealed no significant effects for it, alone or in combination with bupropion (Figure 2B).

The effects of bupropion and antagonists on energy expenditure

A two-way ANOVA was conducted to examine the effect of antagonist pre-treatment and bupropion treatment on locomotion measured during the first 2 h (16:00 h-18:00 h) after returning the food. There was a significant effect for antagonist pre-treatment [F(7,105) = 8.40, P < 0.001] and bupropion treatment [F(1,105) = 14.10, P < 0.001], and a significant pre-treatment × treatment interaction on activity [F(7,105) = 3.47, P = 0.002]. Simple main effects analysis showed for antagonists that raclopride (D2/3; 0.5 mg/kg) on its own significantly inhibited locomotion as compared to baseline [Figure 3A; F(3,31) = 18.02, P < 0.001, post-hoc P < 0.05]. Atipamezole (α2-AR; 1 mg/kg) somewhat increased and SKF 83566 (D1/5; 0.3 mg/kg) somewhat reduced locomotion, but their effects did not reach significance as compared to baseline (Figure 3A).

image

Figure 3. The relative changes in locomotor activity during 2 h (16:00 h–18:00 h) after administration of α1- or α2-AR (adrenoceptor) or dopamine (DA) D1/5, D2/3, and D4 receptor antagonists, on their own (A) or prior to bupropion (B; BUP, 30 mg/kg). Below are shown antagonists and their primary target receptors and BUP treatment (B). AR antagonists were: WB4101 (WB, 2 mg/kg, α1), atipamezole (ATI, 1 mg/kg, α2), BRL 44408 (BR, 2 mg/kg, α2A/D), imiloxan (IMIL, 5 mg/kg, α2B). DA antagonists were: SKF 83566 (SKF, 0.3 mg/kg, D1/5), raclopride (RAC, 0.5 mg/kg, D2/3), and FAUC 213 (FAUC, 0.5 mg/kg, D4). Food was removed for the last 6 h of the light period, and the antagonists were administered via an i.p. cannula 15 min before BUP and 45 min before returning the food in the beginning of the dark period. Locomotor activity was measured via telemetry transmitters during the first 2 h after returning the food (16:00–18:00). Baseline values (=100%) were measured for each rat after vehicle administration. Data expressed as percentage change from the baseline, mean ± SEM, n = 7–10 for all groups. ANOVA followed by Dunnett's post-hoc test: * P < 0.05, ** P < 0.01 versus vehicle baseline (the first column), ## P < 0.01 versus BUP. Part of the 2-h locomotor data for antagonists alone was reported in Ref. [19].

Download figure to PowerPoint

Bupropion (30 mg/kg) significantly stimulated locomotion up to 200% as compared to baseline [Figure 3B; F(3,31) = 6.07, P = 0.003, post-hoc P < 0.05]. Pre-treatment with the AR antagonists WB4101 (α1-AR; 2 mg/kg), atipamezole (α2-AR; 1 mg/kg) and imiloxan (α2B-AR; 5 mg/kg) attenuated the bupropion-induced increase in locomotion and it no longer differed from the baseline (Figure 3B). Pre-treatment with BRL 44408 (α2A/D-AR; 2 mg/kg) however did not alter bupropion's effect on locomotion and it significantly differed from baseline [Figure 3B; BRL+BUP F(3,30) = 8.32, P < 0.001, post-hoc P < 0.01 vs. VEH]. Pre-treatment with the DA antagonists SKF 83566 (D1/5; 0.3 mg/kg) and raclopride (D2/3; 0.5 mg/kg) completely attenuated bupropion-induced stimulation and locomotion was significantly reduced as compared to bupropion [Figure 3B; SKF+BUP F(3,31) = 18.22, P < 0.001, post-hoc P < 0.01 vs. BUP; RAC+BUP F(3,31) = 18.02, P < 0.001, post-hoc P < 0.01 vs. BUP]. Pre-treatment with FAUC 213 (D4; 0.5 mg/kg) modestly inhibited bupropion-induced hyperactivity (Figure 3B).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

This study shows that a moderate, non-sedative dose of bupropion markedly reduces feeding. Moreover, we present novel findings of pre-treatment with AR and DA antagonists on bupropion's effects on meal patterning in rats. We here show that bupropion reduced 2-h caloric intake by reducing the size of the first three meals and by postponing initiation of meals. Different AR and DA receptor subtypes mediate bupropion's effects on size of meals and on meal initiation. Pre-treatment with WB4101 (α1-AR) and imiloxan (α2B-AR) partially counteracted bupropion's effect on the size of the first meal and 2-h caloric intake. In general, pre-treatment with the AR antagonists altered bupropion's effect on meal initiation rather than altering the size of a meal. WB4101 (α1-AR) and BRL 44408 (α2A/D-AR) advanced initiation of meals, whereas atipamezole (α2-AR) and imiloxan (α2B-AR) further postponed initiation of meals and even somewhat further enhanced hypophagia. Pre-treatment with DA antagonists counteracted bupropion's effect on both size and initiation of meals. Raclopride (D2/3) reversed bupropion's effect completely, SKF 83566 (D1/5) reversed it partially and FAUC 213 (D4) reversed it modestly. Our data thus point towards a role for α1-ARs, α2B-ARs, D1/5, and D2/3 in the regulation of the size of a meal after bupropion. Furthermore, α1-ARs, α2A/D-ARs, D1/5, and D2/3 mediate bupropion's effect on initiation of meals (for overview see Table 1). Interestingly, blocking pre-synaptic α2-ARs by atipamezole or α2B-ARs by imiloxan further postponed initiation of meals and thus enhanced long-term 18-h hypophagia by bupropion. Our data suggests that the hypophagic effect of bupropion was enhanced by simultaneously blocking pre-synaptic α2(A)-ARs by atipamezole. This combined treatment further postponed initiation of meals and did not interfere with the hypophagic effect of bupropion on the size of meals. Bupropion increased 2-h locomotion, which supports previous findings that bupropion increases energy expenditure [8]. The hyperactivity by bupropion was counteracted by WB4101 (α1-AR), atipamezole (α2-AR), imiloxan (α2B-AR) and all three DA antagonists, particularly by SKF 83566 (D1/5) and raclopride (D2/3) (Table 1).This supports a role for several AR and DA subtypes in the regulation of locomotion.

Table 1. Summary of the effects of adrenoceptor (AR) and dopamine (DA) receptor antagonists on the effects of bupropion on size of the first meal or average meal, intermeal intervals between the first three meals, and locomotion
AntagonistReceptor subtypeParameter
  1. Abbreviations: Bupropion' effect was attenuated [DOWNWARDS ARROW] = modestly, [DOWNWARDS ARROW][DOWNWARDS ARROW] = markedly, [DOWNWARDS ARROW][DOWNWARDS ARROW][DOWNWARDS ARROW] = completely; Bupropion's effect was enhanced [UPWARDS ARROW] = modestly, [UPWARDS ARROW][UPWARDS ARROW] = markedly.

WB4101α1-ARSize of first meal [DOWNWARDS ARROW]; Meal initiation [DOWNWARDS ARROW]; Intermeal interval [UPWARDS ARROW]; Locomotion [DOWNWARDS ARROW][DOWNWARDS ARROW]
Atipamezoleα2-ARMeal initiation [UPWARDS ARROW][UPWARDS ARROW]; Intermeal interval [UPWARDS ARROW]; Locomotion [DOWNWARDS ARROW][DOWNWARDS ARROW]
BRL 44408α2A/D-ARMeal initiation [DOWNWARDS ARROW]; Intermeal interval [UPWARDS ARROW]; Locomotion [UPWARDS ARROW]
Imiloxanα2B-ARSize of first meal [DOWNWARDS ARROW]; Meal initiation [UPWARDS ARROW]; Intermeal interval [UPWARDS ARROW]; Locomotion [DOWNWARDS ARROW][DOWNWARDS ARROW]
SKF 83566D1/5Size of first meal [DOWNWARDS ARROW]; Meal size [DOWNWARDS ARROW]; Intermeal interval [DOWNWARDS ARROW]; Locomotion [DOWNWARDS ARROW][DOWNWARDS ARROW][DOWNWARDS ARROW]
RacloprideD2/3Size of first meal [DOWNWARDS ARROW][DOWNWARDS ARROW]; Meal size [DOWNWARDS ARROW][DOWNWARDS ARROW]; Intermeal interval [DOWNWARDS ARROW][DOWNWARDS ARROW]; Locomotion [DOWNWARDS ARROW][DOWNWARDS ARROW][DOWNWARDS ARROW]
FAUC 213D4Meal size [DOWNWARDS ARROW] ; Intermeal interval [DOWNWARDS ARROW] ; Locomotion [DOWNWARDS ARROW][DOWNWARDS ARROW]

Bupropion (30 mg/kg) reduced caloric intake by reducing the size of meals, particularly during the first 4h after its administration, whereas it did not alter meal frequency or relative intermeal intervals. Bupropion postponed initiation of the first meal after 6-h restriction, and the following meals were consequently similarly postponed. This postponed meal initiation after NE and DA reuptake inhibition is in line with the physiological regulatory role of NE and DA in meal initiation [26]. Moreover, NE release increases during feeding onset and this is even more pronounced when the animals have been restricted to food prior to a meal [26]. Indeed, the present and previous data suggest that bupropion inhibits caloric intake robustly after food restriction. This is likely the result of more pronounced increase of NE in the synapse as response to food restriction in combination with bupropion treatment. Bupropion reduced chow intake at high dose in meal-fed male rats; doses above 60 mg/kg i.p. were required to reduce 2-h chow intake similarly to our study [11]. In non-deprived female rats, bupropion (10-30 mg/kg p.o.) did not reduce feeding at all [12]. In fasted mice however, bupropion reduced acute chow intake [7]. The dependence of bupropion-induced hypophagia upon food can, at least partly, explain these inconsistent previous findings from bupropion's effects on feeding. Moreover, since the rats were food-restricted prior to drug administration in the present study, the smaller size and postponed initiation of the first meal may also suggest that bupropion attenuated hunger. Interestingly, postponed initiation of meals and limited size of meals may explain bupropion's efficacy to reduce binge eating in bulimia [27].

Regarding AR blockade in food intake, AR antagonists had no significant effect on caloric intake on their own, while in combination with bupropion their effects varied. In general, AR antagonism altered bupropion's effect on initiation of meals and intermeal intervals rather than size of a meal. Pre-treatment with WB4101 (α1-AR) or imiloxan (α2B-AR) counteracted bupropion's effect on the size of the first meal and thus on the short-term 2-h hypophagia (Table 1). Pre-treatment with WB4101 (α1-AR) and BRL 44408 (α2A/D-AR) advanced initiation of meals that were postponed by bupropion (Table 1). Our data suggests that α1-ARs and α2B-ARs mediate bupropion's effect on size of a meal, while α1-ARs and α2A/D-ARs mediate effects on initiation of meals. There are no previous reports which AR subtypes mediate bupropion's or a NE reuptake inhibitors effect on food intake. Our findings are however, consistent with previous reports that α1-ARs, α2B-ARs, and α2A/D-ARs mediate effects of the NE and 5-HT reuptake inhibitor sibutramine on meal size and satiety [19]. All AR antagonists used also shortened intermeal intervals between the first three meals regardless of the size of prior meal, suggesting that activation of α1- and α2-ARs is required for maintaining normal relative intermeal intervals after bupropion.

In contrast to WB4101 and BRL 44408, pre-treatment with atipamezole (α2-AR) and to lesser extent imiloxan (α2B-AR) postponed initiation of the first meal even further, and reduced hunger and enhanced long-term 18-h hypophagia as compared to bupropion alone (Table 1). This implies that blocking specific α2-ARs can further reduce feeding. This agrees with our previous findings from atipamezole given prior to sibutramine [19], and presumably results from atipamezole blocking adrenergic α2A-autoreceptors,which further enhanced NE release in the brain resulting in enhanced hypophagia. In contrast to atipamezole, the selective α2A/D-AR antagonist BRL 44408 did not enhance bupropion's anorectic effect, which is also in line with our previous findings [19]. BRL 44408 was found to elicit less profound effects on NE release [28] than atipamezole did [29], and it acts as an antagonist, whereas atipamezole was reported to act as inverse agonist at α2A/D-ARs [15, 30]. Also imiloxan postponed initiation of meals when given prior to bupropion, but it also counteracted bupropion's effect on the size of a meal, which counteracts hypophagia. This was not the case with atipamezole and thus, blocking pre-synaptic α2A-ARs by atipamezole beneficially enhances bupropion's or other NE reuptake inhibitors hypophagic and weight-reducing effect and offer a novel combination treatment for obesity.

Bupropion preferentially enhances DA transmission to NE transmission [31]. DA plays a pivotal role in initiation of feeding and regulation of food intake through enhancing “wanting” food [26]. In addition to food consumption, DA in nucleus accumbens is suggested to relate to sensory input, feeding reflexes, food reward, and memory processes [32]. Consistently with the pivotal role of DA in feeding, the DA antagonists had marked effects on feeding both on their own and in combination with bupropion. In contrast to AR antagonists, DA antagonists clearly attenuated bupropion's effect on size of a meal and tended to increase intermeal intervals. On its own, SKF 83566 caused hyperphagia during the first 2 h and the third meal was significantly larger than the corresponding meal during baseline. When given prior to bupropion, SKF 83566 (D1/5) did not cause hyperphagia, although it markedly counteracted bupropion's effect on the size of a meal, advanced initiation of the first meal and somewhat increased intermeal intervals between the first three meals (Table 1). Also FAUC 213 (D4) tended to increase feeding, the size of meals and intermeal intervals on its own. When given prior to bupropion, FAUC 213 modestly counteracted bupropion's effect on size of the second and third meal (Table 1). For both SKF 83566 and FAUC 213, the size of the first three meals was significantly smaller when compounds were given in combination with bupropion than when given alone, and bupropion's hypophagic effect was thus not completely blocked by these antagonists.

Raclopride (D2/3) increased the size of meals on its own, and completely attenuated bupropion-induced hypophagia by counteracting the effects on the size of meals (Table 1). The combination of bupropion and raclopride did not induce any hypophagia and caloric intake was comparable to that after raclopride alone. Large meals were followed by longer intermeal intervals, suggesting that raclopride did not alter relative intermeal intervals. Thus DA antagonists, given prior to bupropion, counteracted effects on the size of a meal, which suggests impaired satiation and satiety. Also impaired food reinforcement may explain the tendency to consume more food after administration of DA antagonists [18]. In spite of the pivotal role of DA in regulation of feeding, the impact of DA blockade on bupropion's or a selective DA reuptake inhibitors effect on food intake has not previously been reported. However, in line with our findings, raclopride and SKF 83566 were previously found to increase chow intake at similar doses to those used here [18], and raclopride increased size of a meal without affecting intervals between meals [33].

The effects of FAUC 213 or other D4 antagonists on feeding and meal patterning have not previously been explored. However, reduced function of D4 receptor has been linked to increased obesity susceptibility and food craving, making it an interesting target in control of abnormal feeding [34]. The present data show that blocking D4 receptors somewhat increased size of a meal suggesting tendency to impair satiety, which may also contribute to obesity risk in individuals with a hypofunctional D4 receptor. All in all, all DA receptor types, particularly D2/3, seem to regulate caloric intake in the form of regulating size of a meal.

Bupropion-induced hypophagia was accompanied by a 2-fold increase in locomotion 2 h following returning the food. This increase is still relatively modest and likely not an explanation for the observed hypophagia. Indeed, a previous report suggests that bupropion-induced hypophagia is independent of its ability to increase locomotion in mice [7]. Bupropion's effect on locomotion was clearly inhibited by several AR and DA antagonists, except FAUC 213 (D4) or BRL 44408 (α2A/D-AR). SKF 83566 (D1/5) and even more raclopride (D2/3), alone or prior to bupropion, resulted in marked hypoactivity (Table 1). The reduced locomotor activity by D1- and D2-antagonists was not so marked that it would have impaired feeding behavior, because the hypophagic rats ate normal or even larger amounts during the same 2-h period.

Consistently, a robust increase in synaptic level of NE is known to reduce locomotion in rodents [8]. Our data suggests that NE does so via activation of several AR subtypes, including α1-AR and α2B-ARs, which is in line with previous reports [35]. In agreement with the present findings, we previously reported that the α1-AR antagonist WB4101 significantly inhibited sibutramine-induced hyperactivity, while BRL 44408 did not [19]. Furthermore, blocking α1- or α2-ARs has been repeatedly described to attenuate locomotion induced by monoamine reuptake inhibition [36, 37]. Our findings with DA antagonists also agree with previous reports. D1- and D2-type antagonists reduced hyperactivity after stimulants, like amphetamine and cocaine [38, 39]. Interestingly, α1-AR subtypes were suggested to act synergistically with D2 receptors in the modulation of locomotion [40]. This could partly explain the robust effect of the DA and NE reuptake inhibition on locomotion. All in all, both bupropion-induced hypophagia and hyperactivity likely contribute to bupropion's negative impact on energy balance and weight loss in rodents and humans [7, 11]. The present findings imply that particularly D1-type and D2/3 receptors, as well as α1-ARs and several α2-AR subtypes are involved in enhanced energy expenditure by bupropion, which is also in line with the role of these receptors in stimulant-induced locomotion [35, 38].

To summarize, bupropion reduced caloric intake, particularly during the first 4 h after its administration, because it reduced the size of meals and postponed initiation of meals. Partly different subtypes of AR and DA receptors mediate bupropion's effects on the size of meals and on meal initiation. The α1-ARs, α2B-ARs, D1/5, and D2/3 receptors mediate bupropion's effects on the size of a meal, suggesting that these receptors mediate effects on satiety and satiation. Alpha1-ARs, α2A/D-ARs, D1/5, and D2/3 receptors mediate bupropion's effects on initiation of meals. Bupropion also doubled locomotion 2 h following administration. D1-type and D2/3 receptors, α1-ARs and several α2-AR subtypes underlie the enhanced locomotion by bupropion. Interestingly, pre-treatment with atipamezole (which blocks pre-synaptic α2A-ARs), enhanced the hypophagic effects of bupropion. The limited size of a meal and postponed initiation of meals by such drug combination could help to normalize meal patterns and to control body weight in individuals with eating disorders and obesity.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References