Decoy bypass for appetite suppression in obese adults: role of synergistic nutrient sensing receptors GPR84 and FFAR4 on colonic endocrine cells

Abstract

Objective: Colonic enteroendocrine cells (EECs) store and release potent anorectic hormones that are key regulators of satiety. EECs express multiple nutrient sensing receptors, particularly for medium-chain fatty acids (MCFAs): GPR84 and FFAR4. Here we show a non-surgical approach with targeted colonic delivery of MCFA, which induces EEC and neuronal activation leading to anorectic effects.

Design: A randomised, double-blind, placebo-controlled, cross-over study was performed in obese adults given combined GPR84 and FFAR4 agonists in colonic release capsules before meals. We measured serum hormones, energy intake and appetite perception. Cell type, activation by agonists and hormone/serotonin release were determined in human colonic explants. Mouse colonic afferent nerve responses to nutrients/mediators were recorded electrophysiologically.

Results: Subjects receiving GPR84 and FFAR4 agonists had reduced overall calorific intake and increased postprandial levels of PYY versus placebo. Receptors including GPR84 and FFAR4 were coexpressed on human colonic EEC. Activation of GPR84 exclusively induced intracellular pERK, whereas FFAR4 selectively activated pCaMKII. Coactivation of GPR84 and FFAR4 induced both phosphoproteins, and superadditive release of GLP-1 and PYY. Nutrients and hormones convergently activated murine colonic afterent nerves via GLP-1, Y2 and 5-HT3 receptors.

Conclusions: Colonic GPR84 and FFAR4 agonists reduce energy intake and increase postprandial PYY in obese adults. Human colonic EECs coexpress these receptors, which activate cells via parallel intracellular pathways and synergistically evoke hormone release. Further synergism occurs in sensory nerve responses to MCFA and EEC mediators. Thus, synergistic activation of colonic endocrine cells via nutrient receptors is an important target for metabolic regulation.

Trail registration number: NCT04292236.

Keywords: appetite; gut hormones; neuroendocrine cells; obesity.

Conflict of interest statement

Competing interests: None declared.

© Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY. Published by BMJ.

Figures

Figure 1

Colonic delivery of nutrient combination to obese subjects significantly reduces calorific intake. (A) Calorific intake at breakfast was unchanged between active versus placebo treatment. (B) Calorific intake at lunch was significantly lower in active versus placebo treatment (p=0.019). (C) Total calorific intake (including calorific intake from breakfast and lunch) was significantly lower in active versus placebo treatment arms (p=0.008). (D) VAS scores assessing feeling of hunger at each time-point in active versus placebo were unaltered. (E) VAS scores assessing desire to eat was not different between active versus placebo. (F) VAS scores assessing fullness showed no difference between active versus placebo. VAS, visual analogue scale.

Figure 2

Effect of acute coadministration of 3′3 diindolylmethane (DIM), alpha linolenic acid and lauric acid (LA) on gut hormone levels in obese subjects. (A) Circulating levels of PYY were significantly increased in the active treatment compared with placebo. (B) GLP-1 levels did not significantly differ between active and placebo. (C) 5-HT levels did not significantly differ between active and placebo. (D) Ghrelin levels did not significantly differ between active and placebo. Capsules were given at time 0 and 5 hours, and breakfast was given at 1 hour and lunch given at 6 hours as indicated by arrows.

Figure 3

L cells and enterochromaffin (EC) cells coexpress GPR84 and FFAR4 in human proximal colon. (A) GPR84 is expressed on GLP-1/PYY containing L cells and 5-HT containing EC cells in human proximal colon mucosa. Expression of GPR84 was highest in EC cells, with a smaller population of L cells expressing this GPCR. n=5. (B) FFAR4 expression was observed on both GLP-1 and PYY containing L cells, although in a small subpopulation. EC cells, however, expressed FFAR4 in majority of 5-HT containing cells. n=5. (C) Top panel: FFAR4 and GPR84 are coexpressed on GLP-1 containing L cells, as shown in the merged image. Analysis of coexpression is shown in the graphs where cells were analysed for expressing GLP-1 alone, FFAR4 alone, GPR84 alone and when all three were found expressed on the same cell as shown by ‘merge’. n=5. Bottom panel: FFAR4 and GPR84 are coexpressed on 5-HT containing EC cells. Analysis of coexpression is shown in the graphs where cells were analysed for expressing 5-HT alone, FFAR4 alone, GPR84 alone and when all three were found expressed on the same cell as shown by ‘merge’. n=5.

Figure 4

GPR84 and FFAR4 agonists activate separate intracellular pathways that converge to boost release of GLP-1 and PYY. (A) Stimulation of human colonic mucosa with GPR84 agonist, lauric acid (25 mmol−1) increases PERK expression in human colonic epithelial cells (green) compared with control buffer. n=4. (B) Lauric acid stimulation specifically induces expression of PERK (in a concentration dependent manner) and not pCaMKII, in human colonic mucosa. n=4. (C) FFAR4 agonist TUG891 (10 µM) increases expression of pCaMKII in human colonic epithelial cells (red) compared with control buffer. n=4. (D) TUG891 stimulation specifically induces expression of pCaMKII (in a concentration dependent manner) and not PERK, in human colonic mucosa. n=4. (E) Coapplication of lauric acid (25 mmol−1) and TUG891 (10 µM) enhances cell activation as measured by PERK immunoreactive cells. n=5. (F) quantification of cells positive for PERK showed co-application of Lauric acid (25 mmol−1) and TUG891 (10 µM) results in fourfold increase of positive cells compared with control and single nutrient application. co-stimulation doubled the number of pCaMKII cells compared with single nutrient application. n=5. (G) Immunoreactivity for pCaMKII increased in colonic crypts compared with buffer stimulation in human colonic mucosa. n=4. (H) Coapplication of lauric acid (25 mmol−1) and TUG891 (10 µM) to human mucosal tissue doubles the release of anorectic hormones GLP-1 and PYY compared with stimulation with single nutrient solution. Release of 5-HT is not enhanced by coapplication of GPR84 and FFAR4 agonists. n=5.

Figure 5

Synergistic effects are exerted on proximal colonic afferents by nutrients and appetite regulating hormones. (A) TUG891 evoked a slow-onset response in neuronal firing, whereas lauric acid (LA) had a faster onset effect with greater potency. (B) Combination of TUG891 and LA evoked a synergistic effect on afferent nerve firing rate. n=5/treatment. (C) Example rate histogram and RAW neurogram records of an afferent unit response to PYY (1 µM) and GLP-1 (1 µM), administered individually at subthreshold concentration and together. (D) Graph showing group data for responses to PYY and GLP-1 alone and in combination (n=7).