Naltrexone does not prevent acquisition or expression of flavor preferences conditioned by fructose in rats.

The effects of the general opioid antagonist, naltrexone, on the acquisition and expression of flavor preferences conditioned by the sweet taste of fructose were examined. Food-restricted rats were trained over eight daily alternating one-bottle sessions (2 h) to drink an 8% fructose solution containing one novel flavor (CS+/F) and a less preferred 0.2% saccharin solution containing a different flavor (CS-/S). Four groups of rats were treated daily with either saline (control group) or naltrexone doses of 0.1, 1.0, or 5.0 mg/kg during training. Preferences were assessed in two-bottle tests with the CS+/S and CS-/S flavors presented in 0.2% saccharin solutions following saline injections. Naltrexone dose-dependently reduced fructose and saccharin intakes during training, confirming the drug's well-known suppressive effect on the intake of sweet solutions. Despite their reduced training intakes, the naltrexone groups displayed preferences for the CS+/S over the CS-/S (72-86%) that were similar to that of the control group (78%). The effect of naltrexone on the expression of the CS+/S flavor preference was evaluated by treating control rats with naltrexone (0.1-5 mg/kg) prior to CS+/S vs. CS-/S choice tests. The drug doses produced a dose-dependent reduction in CS+/S intake but did not significantly attenuate the CS+/S preference. These data are consistent with the relative inability of naltrexone to reduce flavor-flavor conditioning by sucrose in sham-feeding rats and flavor-nutrient conditioning in rats receiving intragastric sucrose infusions. In contrast, dopamine antagonists reduce both sucrose- and fructose-conditioned flavor preferences, which indicates the sensitivity of these conditioning paradigms to neuropharmacological manipulations. These data indicate that the endogenous opioid system, unlike the dopamine system, does not play a major role in either the acquisition or expression of flavor preference learning as measured in two-bottle choice tests.