Sugar-induced cephalic-phase insulin release is mediated by a T1r2+T1r3-independent taste transduction pathway in mice

Abstract

Sensory stimulation from foods elicits cephalic phase responses, which facilitate digestion and nutrient assimilation. One such response, cephalic-phase insulin release (CPIR), enhances glucose tolerance. Little is known about the chemosensory mechanisms that activate CPIR. We studied the contribution of the sweet taste receptor (T1r2+T1r3) to sugar-induced CPIR in C57BL/6 (B6) and T1r3 knockout (KO) mice. First, we measured insulin release and glucose tolerance following oral (i.e., normal ingestion) or intragastric (IG) administration of 2.8 M glucose. Both groups of mice exhibited a CPIR following oral but not IG administration, and this CPIR improved glucose tolerance. Second, we examined the specificity of CPIR. Both mouse groups exhibited a CPIR following oral administration of 1 M glucose and 1 M sucrose but not 1 M fructose or water alone. Third, we studied behavioral attraction to the same three sugar solutions in short-term acceptability tests. B6 mice licked more avidly for the sugar solutions than for water, whereas T1r3 KO mice licked no more for the sugar solutions than for water. Finally, we examined chorda tympani (CT) nerve responses to each of the sugars. Both mouse groups exhibited CT nerve responses to the sugars, although those of B6 mice were stronger. We propose that mice possess two taste transduction pathways for sugars. One mediates behavioral attraction to sugars and requires an intact T1r2+T1r3. The other mediates CPIR but does not require an intact T1r2+T1r3. If the latter taste transduction pathway exists in humans, it should provide opportunities for the development of new treatments for controlling blood sugar.

Keywords: T1r3; cephalic-phase insulin release; glucose tolerance; mice; sweet taste.

Fig. 1.

Plasma insulin levels in B6 and T1r3 knockout (KO) mice following oral (i.e., licking; A) or intragastric (B) administration of 2.8 M (50%) glucose (dosage = 2 mg/g mouse). *Insulin levels increased significantly (P < 0.01; one-sample t-test) above baseline within 5 min of the glucose challenge. Symbols represent means ± SE; n = 6 mice per mouse group and administration technique.

Fig. 2.

Glucose tolerance in B6 and T1r3 KO mice following oral (i.e., licking; A) or intragastric (B) administration of 2.8 M (50%) glucose (dosage = 2 mg/g). *Plasma glucose levels differed significantly (P < 0.05, unpaired t-test) between mouse groups. Symbols represent means ± SE; n = 6 mice per mouse group and administration technique.

Fig. 3.

Plasma insulin levels at baseline (i.e., 0 min) and 5 min after initiating licking for one of three sugar solutions (1 M glucose, 1 M fructose, and 1 M sucrose) in B6 and T1r3 KO mice. Each mouse took 4.3 licks/g mouse from each solution. *Plasma insulin levels increased significantly (P < 0.05, paired t-test) above baseline within 5 min. Bars indicate means ± SE; n = 6–14 per sugar for B6 and T1r3 KO mice.

Fig. 4.

Negative control experiments to determine whether licking alone or the sham gavage treatment altered plasma insulin levels. A: plasma insulin levels in B6 and T1r3 KO mice after licking for water vs. sitting in the cage (n = 4–6 mice per mouse group and treatment group). B: plasma insulin levels in B6 mice after the sham gavage treatment + licking for 1 M glucose vs. licking for 1 M glucose alone (n = 6–8 mice per treatment group). *Insulin levels increased significantly between 0 and 5 min (P < 0.02; paired t-test).

Fig. 5.

Licking by B6 and T1r3 KO mice for a sugar solution or water during 5-s trials. During each 30-min test session, mice were offered 2 sipper tubes serially. The tubes dispensed 1 M glucose or water (A), 1 M fructose or water (B), and 1 M sucrose or water (C). We compared the mean (±SE) number of licks per 5-s trial for the sugar solution vs. water, separately for each mouse group and sugar (*P < 0.05; paired t-test); n = 9 animals per mouse group; each mouse was tested with all 3 sugars in a counterbalanced design.

Fig. 6.

Chorda tympani (CT) nerve responses of B6 and T1r3 KO mice to lingual stimulation with various concentrations of glucose, fructose, and sucrose. A: relative responses reflect the ratio of the integrated response to a sugar solution divided by that to 0.1 M NH₄Cl. We tested 10 B6 and 11 T1r3 KO mice with glucose; 11 B6 and 18 T1r3 KO mice with fructose; and 7 B6 and 7 T1r3 KO mice with sucrose. We compared the magnitude of response across mouse groups, separately for each concentration, using Bonferroni's multiple comparison tests. *Significant mouse group differences at a given concentration (P < 0.05). B: typical whole nerve integrated CT nerve responses to 0.1 M NH₄Cl and different concentrations of glucose, fructose, and sucrose in B6 and T1r3 KO mice.