Mechanisms of activation of mouse and human enteroendocrine cells by nutrients

Abstract

Objective: Inhibition of food intake and glucose homeostasis are both promoted when nutrients stimulate enteroendocrine cells (EEC) to release gut hormones. Several specific nutrient receptors may be located on EEC that respond to dietary sugars, amino acids and fatty acids. Bypass surgery for obesity and type II diabetes works by shunting nutrients to the distal gut, where it increases activation of nutrient receptors and mediator release, but cellular mechanisms of activation are largely unknown. We determined which nutrient receptors are expressed in which gut regions and in which cells in mouse and human, how they are associated with different types of EEC, how they are activated leading to hormone and 5-HT release.

Design and results: mRNA expression of 17 nutrient receptors and EEC mediators was assessed by quantitative PCR and found throughout mouse and human gut epithelium. Many species similarities emerged, in particular the dense expression of several receptors in the distal gut. Immunolabelling showed specific colocalisation of receptors with EEC mediators PYY and GLP-1 (L-cells) or 5-HT (enterochromaffin cells). We exposed isolated proximal colonic mucosa to specific nutrients, which recruited signalling pathways within specific EEC extracellular receptor-regulated kinase (p-ERK) and calmodulin kinase II (pCAMKII), as shown by subsequent immunolabelling, and activated release of these mediators. Aromatic amino acids activated both pathways in mouse, but in humans they induced only pCAMKII, which was colocalised mainly with 5-HT expression. Activation was pertussis toxin-sensitive. Fatty acid (C12) potently activated p-ERK in human in all EEC types and evoked potent release of all three mediators.

Conclusions: Specific nutrient receptors associate with distinct activation pathways within EEC. These may provide discrete, complementary pharmacological targets for intervention in obesity and type II diabetes.

Keywords: Gut Hormones; Obesity.

Figure 1

Regional expression in mouse GI tract of the G-protein coupled receptors (GPCR) for carbohydrate-sweet T1R2 (A), bile acids, TGR5 (B), amino acids, GPR93, T1R3, CaSR, mGluR4 and T1R1 (C), fatty acids GRP120, FFAR3, FFAR1, FFAR2, GPR84 and GPR119 (D), relative to GAPDH (n=7 for all expression studies). (E and F) Immunolabelling in mouse proximal colon for GPR93, FFAR3, GLP-1 and PYY in mouse proximal colon. Charts represent the proportion of individual GPCRs colocalised with cells expressing GLP-1 or PYY (yellow), those with GPCR alone (green) and those with hormone alone (red).

Figure 2

Regional expression in human GI tract of the G-protein coupled receptors (GPCR) for carbohydrate-sweet T1R2 (A), bile acids, TGR5 (B), amino acids, GPR93, T1R3, CaSR, mGluR4, T1R1 (C), fatty acids GRP120, FFAR3, FFAR1, FFAR2, GPR84 and GPR119 (D) relative to β-actin (n=6 for all expression studies). (E) Immunolabelling in human proximal colon for CaSR, GLP-1, PYY and 5-HT. Chart represents the proportion of individual cells colocalised with PYY, GLP-1 or 5-HT with and without CaSR.

Figure 3

Effects of phenylalanine and tryptophan (1 and 10 mmol/L) on enzyme phosphorylation in sheets of mouse proximal colonic mucosa. There was a significant increase in the number of pERK-IR (A) and pCamKII-IR (B) cells following stimulation at both concentrations (*p<0.05) compared with control. pERK-IR cells were commonly observed in groups towards the apical aspect of mucosal villi. Within pERK-positive cell groups, one or two cells often stained more positively than surrounding IR cells (C (i), arrows). Comparatively fewer pERK-positive cells appeared in crypts. The opposite was true for pCamKII-IR cells (C (ii) and schematic in (iii)). Insets show 40× magnification of activated cells. Scale bar 15 μm.

Figure 4

(A) Human ascending colonic mucosa was stimulated with 25 and 50 mmol/L Phe/Trp or high calcium Krebs as control. There was a significant increase in the number of pCamKII-IR cells following stimulation at both concentrations (*p<0.05) compared with control. pERK-IR was not induced (not shown). Scale bar 15 μm. (B) In human colon, Phe/trp at 25 mmol/L resulted in specific activation of 5-HT-IR pCaMKII cells, while GLP-1 cells were not found to colocalise with pCamKII. 5-HT-IR and GLP-IR were found predominantly in the basolateral membrane (insets), suggesting that the peptide was located in vesicles in anticipation of release following CaSR stimulation. Scale bar 15 μm. (C) Effect of lauric acid (12.5 and 25 mmol/L) on pERK-IR in sheets of human proximal colonic mucosa. There was a concentration-dependent increase in the number of IR cells with increasing concentrations (*p<0.05) compared with control. Scale bar 15 μm.

Figure 5

Phe/trp stimulation of human ascending colon mucosa pretreated with pertussis toxin (+PT) resulted in a significant decrease in the number of pCamKII-IR cells/crypt (arrows) compared with untreated controls (−PT, *p<0.05). Scale bar 20 µm.

Figure 6

Stimulation with nutrients phenylalanine/tryptophan (P/T) and lauric acid (LA) induces release of appetite regulating hormones in the human proximal colon. (A) Colonic biopsies stimulated with 50 mM P/T induce a significant increase in GLP-1 release but not PYY or 5-HT. (B) 25 mM LA significantly increased release of GLP-1, PYY and 5-HT compared with buffer control (*p<0.05).