Dopamine D₁ and D₂ antagonists reduce the acquisition and expression of flavor-preferences conditioned by fructose in rats

Abstract

The effects of dopamine (DA) D₁ and D₂ receptor antagonists on the acquisition and expression of flavor-preferences conditioned by the sweet taste of fructose were examined. Food-restricted rats were trained over eight alternating one-bottle sessions to drink an 8% fructose solution containing one novel flavor (CS+) and a less preferred 0.2% saccharin solution containing a different flavor (CS−). Three groups of rats were treated daily with either vehicle (control group), SCH23390 (200 nmol/kg; D₁ group), or raclopride (200 nmol/kg; D₂ group) during training. Additional groups of vehicle-treated rats had their daily training intakes matched to that of the D₁ and D₂ groups. Preferences were assessed in two-bottle tests with the CS+ and CS− flavors presented in 0.2% saccharin solutions following doses of 0, 50, 200, 400, or 800 nmol/kg of either D₁ or D₂ antagonists. The D₁ and D₂ groups, unlike the control and yoked-control groups, failed to display a significant CS+ preference in the two-bottle tests following vehicle treatment. In addition, treatment with SCH23390 prior to the two-bottle tests blocked the expression of the CS+ preference in the control groups. Pretest raclopride treatment attenuated the CS+ preference at some dose levels. Raclopride also attenuated the preference for fructose in rats given two-bottle training with the CS+/fructose (CS+/F) and CS−/saccharin (CS−/S) solutions. These findings indicate that D₁ and D₂ antagonists block flavor-preference conditioning by sweet taste and that D₁, and to a lesser extent D₂, receptor antagonists attenuate the expression of a previously acquired preference.

Introduction

Animals use flavor cues (taste, odor, and texture) to guide their selection of nutritious foods and avoidance of toxic foods (Capaldi, 1996). Flavor-preferences and aversions, in turn, are based in part on learned associations between the various flavor elements in the foods, flavor–flavor conditioning, and between the flavor cues and postingestive consequences, flavor–nutrient, and flavor–toxin conditioning. A primary example of flavor–flavor conditioning is the acquired preference for an arbitrary flavor cue (e.g., banana extract) added to a sweet solution (e.g., saccharin solution; Holman, 1975). The naturally preferred sweet taste is considered to be an unconditioned stimulus (US) that reinforces the animal's preference for the added flavor, which represents the conditioned stimulus (CS).

The potent reward value of sweet taste may result, in part, because sweet taste activates mesolimbic dopamine (DA) circuits that are implicated in the mediation of natural as well as drug rewards. It has long been known that DA antagonists suppress the intake of sweet solutions in rats Geary and Smith, 1985, Muscat and Willner, 1989, Xenakis and Sclafani, 1981. Various findings suggest that this suppression results, in part, because DA antagonists reduce the reward value of sweet taste Schneider, 1989, Smith, 1995 although other explanations have been proposed to account for drug effects on food intake and reinforcement Berridge and Robinson, 1998, Ikemoto and Panksepp, 1999, Salamone et al., 1997. In addition to reducing the intake of sweet solutions, DA antagonists may also alter the ability of sweet solutions to reinforce the preference for other flavors. In particular, Hsiao and Smith (1995) reported that rats showed a reduced preference for a flavored 10% sucrose solution previously consumed while they were treated with the D₂ antagonist raclopride compared to a differently flavored sucrose solution previously consumed while they were treated with saline.

Sucrose can reinforce flavor-preferences based on its sweet taste as well as its postingestive nutritive actions through the processes of flavor–flavor and flavor–nutrient conditioning, respectively (Sclafani, 1995). Hsiao and Smith (1995) used brief (5 min) training sessions to minimize postingestive factors, but the amount of sucrose consumed in the training sessions may have had a postingestive reinforcing action. To separate the effects of drugs on flavor–flavor and flavor–nutrient learning, our laboratories have used sham-feeding and intragastric (IG) infusion procedures, respectively Azzara et al., 2000, Azzara et al., 2001, Yu et al., 1999, Yu et al., 2000a, Yu et al., 2000b. With sham feeding, the ingested sucrose solution drains out of an open gastric fistula thereby minimizing postingestive nutrient actions (Weingarten and Watson, 1982). With the IG procedure, on the other hand, the sucrose is infused into the stomach thereby eliminating the sugar's taste as a conditioning factor (Sclafani, 1995). Yu et al., 2000a, Yu et al., 2000b used the sham-feeding procedure to determine the effects of DA antagonists on the acquisition and expression of flavor conditioning by the sweet taste of sucrose. Rats were treated with a D₁ antagonist (SCH23390), a D₂ antagonist (raclopride), or saline during sham-feeding training trials with different flavors added to a 16% sucrose solution or a less preferred 0.2% saccharin solution. In subsequent drug-free choice tests with both flavors presented in sucrose+saccharin solutions, the D₁ and D₂ groups displayed comparable preferences for the sucrose-paired flavor to those of saline-control rats that had their training intake limited to that of the drug groups, indicating a negligible effect on acquisition (Yu et al., 2000b). However, both antagonists dose-dependently reduced the preference for the CS+ flavor when administered prior to the choice test, indicating strong expression effects (Yu et al., 2000a).

The finding of Yu et al. (2000b) that DA antagonists did not block the acquisition of flavor conditioning by the sweet taste of sucrose would appear inconsistent with Hsiao and Smith's (1995) conditioning results as well as other studies suggesting that DA antagonists reduce the reward value of sweet taste (see Schneider, 1989, Smith, 1995). There are several procedural differences between the two conditioning studies that may account for the discrepant findings. In particular, Hsiao and Smith (1995) used a higher dose of raclopride and exposed their rats to less sucrose during training compared to the Yu et al. (2000b) study. In addition, the rats in the Hsiao and Smith (1995) study were given matched amounts of the raclopride-paired flavor and vehicle-paired flavor, whereas the drug-exposed rats in the Yu et al. (2000b) study were given unlimited access to the CS+ and CS− flavors, and both flavors were paired with raclopride.

In view of these considerations, the present study further investigated the effect of DA antagonism on flavor-preference learning produced by sweet taste. In this case, a conditioning procedure developed by Sclafani and Ackroff (1994) was used in which rats are trained to drink matched amounts of differently flavored 8% fructose and 0.2% saccharin solutions. This training procedure produces a robust preference for the fructose-paired flavor in a two-bottle choice test when both flavors are presented in 0.2% saccharin. Fructose, rather than sucrose or glucose, is used in this conditioning procedure because, unlike these other sugars, fructose has little or no postingestive reinforcing action during the short-term sessions. This is demonstrated by the failure of IG fructose infusions to condition a CS+ preference as well as by other findings Sclafani and Ackroff, 1994, Sclafani et al., 1993, Sclafani et al., 1999. Thus, the preference for a flavor that is mixed into a fructose solution is considered to be a form of flavor–flavor learning based on the more preferred taste of 8% fructose relative to 0.2% saccharin (Sclafani and Ackroff, 1994).