Benzodiazepine-induced hyperphagia: a test of the hunger-mimetic model

Multiple processes underlie benzodiazepine-mediated increases in the consumption of accepted and avoided stimuli

Hyperphagia is a reported side effect of anxiolytic benzodiazepines such as chlordiazepoxide (CDP). Prior research has focused primarily on the ingestive responses to sweet or solid foods. We examined CDP effects on licking for normally accepted and avoided taste solutions across a range of concentrations. The effect of CDP (10 mg/kg) versus saline on the licking patterns of water-restricted rats for water and 3 concentrations of sucrose, saccharin, NaCl, monosodium glutamate (MSG), citric acid, and quinine (Q-HCl) solutions was evaluated during 1 h tests. CDP increased meal size for all tastants except citric acid. Analysis of licking microstructure revealed 3 dissociable effects of CDP. CDP affected oromotor coordination as indicated by a uniform increase in the modal interlick interval for all stimuli. CDP increased meal size as indicated by shorter pauses during consumption of water, MSG, and weaker saccharin concentrations, and by fewer long interlick intervals (250-2000 ms) for normally avoided tastants. CDP also increased meal size by increasing burst size, burst duration, and the initial rate of licking for most solutions, suggesting increased hedonic taste evaluation. CDP did not affect variables associated with postingestive feedback such as meal duration or number of bursts, and the results also suggest that CDP did not enhance the perceived taste intensity. We hypothesize that the reduction of pause duration is consistent with an increased motivation to sample the stimulus that synergizes with changes in taste-mediated responsiveness to some but not all stimuli to yield increases in the consumption of both normally accepted and avoided taste stimuli.

Opioid-dependent anticipatory negative contrast and binge-like eating in rats with limited access to highly preferred food

Binge eating and an increased role for palatability in determining food intake are abnormal adaptations in feeding behavior linked to eating disorders and body weight dysregulation. The present study tested the hypothesis that rats with limited access to highly preferred food would develop analogous opioid-dependent learned adaptations in feeding behavior, with associated changes in metabolism and anxiety-like behavior. For this purpose, adolescent female Wistar rats were daily food deprived (2 h) and then offered 10-min access to a feeder containing chow followed sequentially by 10-min access to a different feeder containing either chow (chow/chow; n=7) or a highly preferred, but macronutrient-comparable, sucrose-rich diet (chow/preferred; n=8). Chow/preferred-fed rats developed binge-like hyperphagia of preferred diet from the second feeder and anticipatory chow hypophagia from the first feeder with a time course suggesting associative learning. The feeding adaptations were dissociable in onset, across individuals, and in their dose-response to the opioid-receptor antagonist nalmefene, suggesting that they represent distinct palatability-motivated processes. Chow/preferred-fed rats showed increased anxiety-like behavior in relation to their propensity to binge as well as increased feed efficiency, body weight, and visceral adiposity. Chow/preferred-fed rats also had increased circulating leptin levels and decreased growth hormone and 'active' ghrelin levels. Thus, the short-term control of food intake in rats with restricted access to highly preferred foods comes to rely more on hedonic, rather than nutritional, properties of food, through associative learning mechanisms. Such rats show changes in ingestive, metabolic, endocrine, and anxiety-related measures, which resemble features of binge eating disorders or obesity.

FG 7142 specifically reduces meal size and the rate and regularity of sustained feeding in female rats: evidence that benzodiazepine inverse agonists reduce food palatability

Benzodiazepine receptor inverse agonists reduce food intake in males, but their actions in females, in whom stress-related eating disorders are more common, as well as their behavioral mode of action remain unclear. The consummatory effects of benzodiazepine receptor ligands have alternately been hypothesized to reflect changes in the hedonic evaluation of food or secondary effects of anxiety-related or cognitive properties. To test the anorectic mode of action of benzodiazepine inverse agonists, the effects of FG 7142 on feeding microstructure were studied in nondeprived female Wistar rats (n=32). Microstructure analysis used a novel meal definition that recognizes prandial drinking. On pharmacologically synchronized diestrus I, rats were pretreated (-30 min dark onset) with the benzodiazepine partial inverse agonist FG 7142 (i.p. 0, 3.75, 7.5, 15 mg/kg) in a between-subjects design. FG 7142 delayed the onset of (16-541%), decreased the amount eaten (36-52%) and drunk (63-87%), and reduced the time spent drinking (59-87%) within the first nocturnal meal. Dose-dependent incremental anorexia continued 6 h into the dark cycle, whereas FG 7142 did not suppress the quantity, duration or rate of drinking past the first meal. Treated rats ate smaller meals (17-42%) of normal duration. This reflected that FG 7142 slowed feeding within meals (9-38%) by decreasing the regularity and maintenance of feeding from pellet-to-pellet. FG 7142 did not influence postprandial satiety; meal frequency and inter-meal intervals were unaffected. FG 7142 anorexia was blocked by the benzodiazepine receptor antagonist flumazenil in a 2:1 molar ratio (n=17 rats). The very early, nonspecific (+10 min), but not subsequent (2.5, 4.5 h) feeding-specific phase, of FG 7142 anorexia was mirrored by anxiogenic-like behavior in FG 7142-treated (7.5 mg/kg) female rats (n=48) in the elevated plus-maze. Thus, benzodiazepine receptor inverse agonists preferentially lessen the maintenance of feeding in female rats, effects opposite to those of palatable food.

Low-dose diazepam primes motivation for alcohol and alcohol-related semantic networks in problem drinkers

Considerable research with animals indicates that the GABA-benzodiazepine (BZ) system plays a key role in alcohol reinforcement. However, only limited research appears to have assessed this issue directly in humans. The present study investigated whether low-dose diazepam would cross-prime motivation for alcohol in problem drinkers. Twelve male problem drinkers (Alcohol Dependence Scale; ADS score > or =9) received oral diazepam (5 mg) and placebo, in a counterbalanced manner on separate sessions. There were three measures of primed motivation for alcohol: self-reported desire for alcohol, consumption of placebo beer in an ostensible taste test procedure, and automatically executed vocal reading responses to Alcohol versus Neutral words on a computer-based task. Diazepam significantly increased beer consumption, and produced a marginally significant increase in reported desire for alcohol. On the reading task, diazepam significantly decreased response latency to Alcohol words relative to Neutral words. Latency to Alcohol words correlated significantly with beer consumption under the drug. Moreover, response latency to Alcohol words under the drug also predicted ADS scores. Thus, severity of dependence was directly linked with vulnerability to a BZ priming effect on motivation for alcohol. These findings provide direct evidence that the GABA-BZ system plays an important role in alcohol reinforcement in problem drinkers.

GABA(A) receptor modulation of the rewarding and aversive effects of ethanol

Ethanol has been shown to exert many of its biochemical and behavioral effects through an interaction with the gamma-aminobutyric acid (GABA) receptor system. This review focuses on a subset of studies that has used self-administration, as well as place and taste conditioning, procedures to investigate a role for the GABA(A) receptor system in modulating the rewarding and aversive effects of ethanol. Potential advantages and disadvantages of each procedure are also discussed. A significant amount of evidence supports the suggestion that GABA(A) receptors are important modulators of the motivational effects of ethanol, although most of the findings have been obtained from studies examining oral ethanol self-administration. Relatively fewer studies have investigated ethanol place and taste conditioning. All self-administration studies reviewed used rats, whereas most conditioning studies used mice. Results of these studies show that GABA(A) antagonists and inverse agonists reduce ethanol self-administration under limited-access conditions. The effect of GABA(A) agonists on ethanol self-administration is less clear due to their bidirectional effects. GABA(A) receptor antagonists have been shown to increase ethanol-induced conditioned place preference and conditioned taste aversion in mice and decrease ethanol-induced conditioned taste aversion in rats. Issues related to interpretation and integration of these findings across models and species are considered. The integration of data from self-administration and conditioning procedures is necessary to define the role of GABA(A) receptors in modulating the rewarding and aversive effects of ethanol and may lead to the development of pharmacotherapies that target GABA(A) receptors to treat alcoholism in human beings.

Bidirectional modulation of sweet and bitter taste by chlordiazepoxide and Ro 15-4513: lack of effect with GABA drugs

Five rats were trained to respond for 10% sucrose and 10% sucrose/0.006% quinine in an operant procedure. Both solutions were concurrently available on independent, variable-interval 5-s schedules of reinforcement. Rats reliably responded for both solutions throughout the sessions and made approximately 68% of their total daily responses for the sucrose solution. When injected prior to the sessions with 4 mg/kg of chlordiazepoxide, rats selectively increased quinine responding; injections of the benzodiazepine inverse agonist Ro 15-4513 (9 mg/kg) led to decreased quinine responding. The effects of both chlordiazepoxide and Ro 15-4513 were reversed by the benzodiazepine antagonist flumazenil. Presession injections of flumazenil, muscimol, baclofen, or picrotoxin all resulted in no changes in responding, or a decrease in responding for both solutions. These results are discussed in terms of a bidirectional modulation of sweet-bitter taste preference by drugs acting on the benzodiazepine receptor. Moreover, the data from these experiments suggest that any changes in the oral consumption of alcohol following administration of benzodiazepine drugs must be examined in light of their effects on taste palatability.

Food palatability and hunger modulated effects of CGS 9896 and CGS 8216 on food intake

The effect of food palatability and duration of food deprivation on the modulation of food intake by two benzodiazepine receptor (BDZR) ligands, CGS 9896 and CGS 8216, were investigated. Three diets differing in palatability (high, medium, or standard) and three different periods of food deprivation (0, 16, or 24 h) were used in all combinations to compare the effect of these variations on the observed modulation of food consumption by both BDZR ligands. Increasing diet palatability and/or food deprivation increased the baseline food consumption and reduced the sensitivity of the test to the detection of the hyperphagic effect of CGS 9896 but increased the sensitivity to detect the anorexic effect of CGS 8216. Only for the intermediate conditions of food deprivation (16 h) and for a standard or medium palatable diet were both significant hyperphagic effect of CGS 9896 and anorexic effect of CGS 8216 detected. Neither increased palatability nor hunger enhanced the modulation of feeding, indicating that neither "taste preference" nor "hunger" is the key factor in the mechanism of BDZR ligand-induced feeding response.

A contingent, conditioned suppression of eating following chronic benzodiazepine-induced hyperphagia

While the hyperphagic effect of chloradizepoxide (CDP) has been reported by some to be enhanced with chronic drug treatment, the processes underlying this phenomenon are not well understood. In the present study, it was predicted that following chronic exposure to CDP-induced hyperphagia, animals given a placebo in place of their usual drug injection might be expected to exhibit evidence of a conditioned, drug-like response. Such a finding would then be consistent with an underlying process of behavioral sensitization. In Experiment 1a, Male Sprague-Dawley rats were randomly assigned to one of two groups receiving intraperitoneal (IP) injections of either 5 mg/kg CDP (Group CDP) or physiological saline (1 ml/kg; Group SAL) administered over 15 drug treatment days. Thirty minutes after each injection, all animals were given 30 min access to sweetened condensed milk. A significant enhancement of CDP-induced hyperphagia was observed over treatment sessions, confirming an earlier report. Unexpectedly, in the Placebo Test, the CDP animals exhibited a supression of milk consumption relative to that of the SAL group. Using the same animals, this finding was successfully replicated in Experiment 1b. In Experiment 2, it was hypothesized that if this conditioned, drug-opposite response were to reflect the involvement of some underlying compensatory, homeostatic mechanism, then it should only be observable under food-contingent conditions of chronic drug treatment.(ABSTRACT TRUNCATED AT 250 WORDS)