Mapping glucose-mediated gut-to-brain signalling pathways in humans

Abstract

Objectives: Previous fMRI studies have demonstrated that glucose decreases the hypothalamic BOLD response in humans. However, the mechanisms underlying the CNS response to glucose have not been defined. We recently demonstrated that the slowing of gastric emptying by glucose is dependent on activation of the gut peptide cholecystokinin (CCK1) receptor. Using physiological functional magnetic resonance imaging this study aimed to determine the whole brain response to glucose, and whether CCK plays a central role.

Experimental design: Changes in blood oxygenation level-dependent (BOLD) signal were monitored using fMRI in 12 healthy subjects following intragastric infusion (250ml) of: 1M glucose+predosing with dexloxiglumide (CCK1 receptor antagonist), 1M glucose+placebo, or 0.9% saline (control)+placebo, in a single-blind, randomised fashion. Gallbladder volume, blood glucose, insulin, and GLP-1 and CCK concentrations were determined. Hunger, fullness and nausea scores were also recorded.

Principal observations: Intragastric glucose elevated plasma glucose, insulin, and GLP-1, and reduced gall bladder volume (an in vivo assay for CCK secretion). Glucose decreased BOLD signal, relative to saline, in the brainstem and hypothalamus as well as the cerebellum, right occipital cortex, putamen and thalamus. The timing of the BOLD signal decrease was negatively correlated with the rise in blood glucose and insulin levels. The glucose+dex arm highlighted a CCK1-receptor dependent increase in BOLD signal only in the motor cortex.

Conclusions: Glucose induces site-specific differences in BOLD response in the human brain; the brainstem and hypothalamus show a CCK1 receptor-independent reduction which is likely to be mediated by a circulatory effect of glucose and insulin, whereas the motor cortex shows an early dexloxiglumide-reversible increase in signal, suggesting a CCK1 receptor-dependent neural pathway.

Keywords: Cholecystokinin (CCK); Dexloxiglumide; Glucose; nutrient; physMRI.

Copyright © 2014. Published by Elsevier Inc.

Figures

Fig. 1

(A) Gastric emptying (n = 3/12) and (B) change in gallbladder volume (n = 12) following intragastric glucose (1000 mOsmol), glucose (1000 mOsmol) + dexloxiglumide (600 mg) or saline/water control. Data are mean (SEM). *glucose vs. water/saline, ^glucose + dex vs. glucose, p 

Fig. 2

Brain images showing areas exhibiting significant effect of time (pFDRc 

Fig. 3

Brain images showing areas exhibiting significant effect of time (pFDRc 

Fig. 4

Changes in blood-oxygenation-level-dependent (BOLD) signal over time in the medulla (A), pons (B), midbrain (C), and hypothalamus (D), and motor cortex (E), following bolus intragastric glucose (1000 mOsmol) and glucose + dex, relative to saline.

Fig. 5

Blood glucose (A) and plasma insulin (B), cholecystokinin (CCK) (C) and glucagon-like peptide-1 (GLP-1) (D) concentrations following bolus intragastric glucose (1000 mOsmol), glucose + dexloxiglumide, or saline (0.9%). *glucose vs. saline, p 

Fig. 6

Effects of saline, glucose and glucose + dexloxiglumide on perceptions of nausea (A), hunger (B) and fullness (C). There were no differences in baseline ratings, nor any effect of any treatment, on perceptions of nausea or hunger. There was an effect of treatment for fullness scores (p = 0.05). Glucose increased fullness when compared with saline (p = 0.009), with no differences between saline and glucose + dex. *P