Ongoing ingestive behavior is rapidly suppressed by a preabsorptive, intestinal "bitter taste" cue

Abstract

The discovery that cells in the gastrointestinal (GI) tract express the same molecular receptors and intracellular signaling components known to be involved in taste has generated great interest in potential functions of such post-oral "taste" receptors in the control of food intake. To determine whether taste cues in the GI tract are detected and can directly influence behavior, the present study used a microbehavioral analysis of intake, in which rats drank from lickometers that were programmed to simultaneously deliver a brief yoked infusion of a taste stimulus to the intestines. Specifically, in daily 30-min sessions, thirsty rats with indwelling intraduodenal catheters were trained to drink hypotonic (0.12 M) sodium chloride (NaCl) and simultaneously self-infuse a 0.12 M NaCl solution. Once trained, in a subsequent series of intestinal taste probe trials, rats reduced licking during a 6-min infusion period, when a bitter stimulus denatonium benzoate (DB; 10 mM) was added to the NaCl vehicle for infusion, apparently conditioning a mild taste aversion. Presentation of the DB in isomolar lithium chloride (LiCl) for intestinal infusions accelerated the development of the response across trials and strengthened the temporal resolution of the early licking suppression in response to the arrival of the DB in the intestine. In an experiment to evaluate whether CCK is involved as a paracrine signal in transducing the intestinal taste of DB, the CCK-1R antagonist devazepide partially blocked the response to intestinal DB. In contrast to their ability to detect and avoid the bitter taste in the intestine, rats did not modify their licking to saccharin intraduodenal probe infusions. The intestinal taste aversion paradigm developed here provides a sensitive and effective protocol for evaluating which tastants-and concentrations of tastants-in the lumen of the gut can control ingestion.

Fig. 1.

Mean licks per minute for a NaCl solution at the sipper spout collapsed across five 30-min probe trials. A brief (6 min) intraduodenal infusion was yoked to licking at the beginning of the 30-min session (shaded in gray). Denatonium benzoate (DB; 10 mM) in NaCl was infused for one session in each trial, and plain unadulterated LiCl was infused for the alternative session in each trial (n = 7).

Fig. 2.

Mean licks per minute for a NaCl solution at the sipper spout collapsed across five 30-min probe trials. A brief (6 min) intraduodenal infusion was yoked to licking at the beginning of the 30-min session (shaded in gray). One group of rats (n = 7) received 10 mM of DB mixed into LiCl infusions for one session per trial and received plain NaCl infusions for the alternative session per trial (A). A second group of rats (n = 6) received plain LiCl infusions for one session in each trial and plain NaCl infusions for the alternative session in each trial (B).

Fig. 3.

Means ± SE difference in licks per minute in minutes 3–8 on the two probe sessions at each trial as a function of training group. For one group of rats (from experiment 1), the difference between licking for NaCl on intraduodenal DB in NaCl sessions from the alternate plain intraduodonal LiCl sessions is plotted (DB in NaCl). For a second group of rats (from experiment 2), the difference between licking for DB in LiCl sessions from the alternate plain NaCl sessions is plotted (DB in LiCl). For a third group (from experiment 2), the difference between licking for plain LiCl sessions from the alternate plain NaCl sessions is plotted (LiCl).

Fig. 4.

Mean licks per minute for a NaCl solution at the sipper spout collapsed across six 30-min probe trials. Rats received a brief yoked intraduodenal infusion during the first 6 min of each 30-min session (shaded in gray) of 9.75 mM sodium saccharin (Sacc) mixed into NaCl (n = 5), Sacc mixed in LiCl (n = 8), or plain LiCl (n = 7).

Fig. 5.

Mean licks per minute for a NaCl solution at the sipper spout on four 30-min test sessions. Rats (n = 18) were given an intraperitoneal injection of the CCK-1R antagonist devazepide (DEV) (75 μg/ml, 4 ml/kg) or its vehicle (VEH) 30 min prior to the start of the test session. Rats received an intraduodenal infusion (yoked to licking) of either 10 mM DB in NaCl or 9.75 mM Sacc in NaCl during the 6 min (shaded in gray) of each test session.

Fig. 6.

Mean licks per minute for NaCl at the sipper spout on two test sessions. Rats (n = 13) received intraduodenal infusions (yoked to licking) of either 10 mM DB in NaCl or 9.75 mM Sacc in NaCl, during minutes 8–13 of the session (shifted, shaded in gray).