Gastrointestinal Nutrient Sensing in Obesity and Type 2 Diabetes: Role in Appetite Regulation and Glycaemic Control

Abstract

Background: The sensing of nutrients by the small intestine generates signals, including the secretion of gastrointestinal (GI) hormones, which are important determinants of subsequent energy intake and postprandial glycaemia. Recent studies have identified that specific free fatty acid (FFA) and sweet taste sensors/receptors, localised to enteroendocrine cells and/or absorptive cells, in the small intestine, play a central role in mediating nutrient-induced GI hormone release. Furthermore, studies in knock-out (KO) and diet-induced obese (DIO) animal models have revealed that altered expression of a number of these receptors attenuates GI hormone secretion, and consequently alters food intake and glycaemic control, thereby, providing evidence that intestinal nutrient sensing plays a significant role in the pathophysiology of obesity and type 2 diabetes (T2D). However, our understanding of the relationships between expression of nutrient receptors in the small intestine, nutrient-induced release of GI hormones, appetite regulation, and glycaemic control in human health and metabolic conditions such as obesity and T2D remains limited. Aims: The studies presented in this thesis aimed to characterise the expression and functional role of duodenal nutrient sensors for fats and carbohydrates in human health, obesity and T2D. Specifically, the aims were to investigate: 1) The effect of acute intraduodenal (ID) nutrient exposure (lipid or glucose) on duodenal nutrient sensor expression. 2) Relationships between the expression of nutrient sensors at baseline (fasted), and after nutrient infusion, with the secretion of GI hormones involved in regulating appetite, energy intake and glycaemia. 3) Relationships between the expression of nutrient sensors at baseline (fasted), and after nutrient infusion, with appetite perceptions, habitual energy and macronutrient intakes. Methods: For the studies presented in Chapter 3 and 4, 57 volunteers classified as lean (n = 20, body mass index (BMI) 18-24 kg.m2), overweight (n = 18, BMI 25-29 kg.m2) or obese (n = 19, BMI ≥ 30 kg.m2) underwent unsedated endoscopy. Duodenal biopsies were collected at baseline (following a 12 hour fast), and 30 min after an ID infusion of 10% Intralipid® (2 kcal/min). Duodenal expression of free fatty acid receptor 1 (FFAR1), FFAR4, G-protein coupled receptor 119 (GPR119), and the cluster-of-differentiation-36 (CD36) was assessed by quantitative reverse-transcription polymerase chain reaction (RT-PCR), relative to expression of the housekeeper gene β-2 microglobulin (β2M). On a separate visit, the effects of a 120 min ID infusion of Intralipid® (2 kcal/min) infusion on blood glucose, and plasma cholecystokinin (CCK), glucagon-like peptide-1 (GLP-1), glucose-dependent insulinotropic peptide (GIP), peptide-YY (PYY), insulin and leptin concentrations were evaluated, followed by an ad libitum buffet-meal, from which energy and macronutrient intake was quantified. Habitual dietary intake was assessed using food frequency questionnaires (FFQs). For the study presented in Chapter 5, 12 healthy control individuals (HC), 12 patients with well-controlled T2D (WC-T2D; HbA1c 6.3 ± 0.2%), and 9 patients with poorly-controlled T2D (PC-T2D; HbA1c 10.6 ± 0.5%) undertook an oral glucose tolerance test (OGTT) following an overnight fast, as previously described1. These participants were then studied during a euglycaemic clamp (5 ± 1 mmol/L), with duodenal biopsies collected at baseline (fasted) and after a 30 min ID glucose infusion (4 kcal/min). Copy numbers of taste receptor type 1, member 2 (T1R2), the sodium-glucose co-transporter 1 (SGLT-1) and glucose- transporter 2 (GLUT2) transcript were assessed at t = 0, 10 and 30 min by RT-PCR. Plasma concentrations of GIP, GLP-1, and C-peptide were measured at 10 min intervals from baseline (t = 0 min) for 60 min (t = 60 min). Plasma concentrations of 3-ortho-methylglucose (3-OMG) were measured at t = 30 and 60 min, using mass spectrometry, to assess glucose absorption. Results: Duodenal fatty acid sensing receptor expression in lean, overweight and obese individuals: During fasting, duodenal expression of FFAR1 and FFAR4 was lower (P ≤ 0.05), and CD36 higher (P ≤ 0.001), in obese, compared with lean and overweight, participants. ID lipid increased GPR119 and FFAR1 transcript levels independent of BMI (both P ≤ 0.05), while levels of CD36 and FFAR4 did not change. The lipid-induced change in FFAR1 was positively associated with the incremental area under the curve (iAUC) of GIP (r = 0.3, P ≤ 0.05). ID lipid induced the secretion of GIP, GLP-1, CCK, PYY and insulin, but there was no relationship between hormone levels with fat sensor expression. There was no relationship between acute energy and macronutrient intake at the buffet-meal and duodenal expression of fat sensors, however, habitual consumption of polyunsaturated fatty acids (PUFAs) was negatively associated with GPR119 in healthy, lean participants (r = -0.5, P ≤ 0.05) (Chapter 3, Chapter 4). Duodenal sweet taste receptor (STR) and glucose transporter expression in health, and patients with well- and poorly-controlled type 2 diabetes Blood glucose concentrations were higher in PC-T2D than WC-T2D and HC groups before and during the OGTT (P ≤ 0.001). Basal T1R2 transcript levels were similar across groups, while SGLT-1 transcripts were lower in PC-T2D than in the WC-T2D group (P ≤ 0.01), and GLUT2 transcripts lower in PC-T2D than in both WC-T2D and HC groups (P ≤ 0.01). Plasma GIP concentrations were higher in WC-T2D than in the HC group at baseline (P ≤ 0.01), with no group differences in GLP-1 and C-peptide concentrations. ID glucose increased SGLT-1 and decreased GLUT2 transcripts at 10 min (group × time interaction) in both HC and WC-T2D groups (both P ≤ 0.001, P ≤ 0.05 respectively), but had no effect on SGLT-1 or GLUT2 transcripts in the PC-T2D group. T1R2 transcripts were lower in PC-T2D at 10 min than in the WC-T2D group (P ≤ 0.05), while transcript levels of all targets were similar across groups at t = 30 min. ID glucose increased plasma GIP, GLP-1 and C-peptide concentrations (all P ≤ 0.001), with GIP higher in PC-T2D (iAUC, P ≤ 0.05) than in the HC group, GLP-1 higher in WC-T2D than the HC group (P ≤ 0.05), and C-peptide highest in HC compared to both WC-T2D and PC-T2D groups (P ≤ 0.01, P ≤ 0.001). T1R2 and GLUT2 transcripts at baseline, and in response to ID glucose, were unrelated to GIP, GLP-1 or C-peptide iAUC. GIP concentrations after 10 min were negatively associated with basal SGLT-1 transcripts (r = -0.6, P ≤ 0.05), and the degree of change in SGLT-1 during ID glucose (r = -0.5, P ≤ 0.05). Serum 3-OMG at 30 min was positively related to the change in T1R2 transcript level at 10 min in HC participants (r = 0.7, P ≤ 0.05) (Chapter 5). Conclusions: These studies have identified notable differences in the duodenal expression of the FFA sensors FFAR1, FFAR4 and CD36 in human obesity at baseline. GPR119 was linked to habitual PUFA consumption in health, indicating that dietary fatty acid composition, rather than high-fat diet (HFD) consumption per se, may influence fat sensor expression. Overall, the response of FFA sensors to acute ID lipid remained intact in obesity, with BMI-independent increases in FFAR1 and GPR119, but no association between FFA sensor expression and fat-induced secretion of gut hormones across the cohort. In separate studies we demonstrated that baseline expression of duodenal glucose transporters SGLT-1 and GLUT2 was lower in PC-T2D patients at euglycaemia. Incretin and transcriptional responses to glucose infusion, and 3-OMG absorption, was similar in WC-T2D and HC, however, PC-T2D patients showed a dysregulated T1R2 response, lack of transcriptional change in SGLT-1 and GLUT2 to ID glucose infusion, and exaggerated GIP secretion and 3-OMG absorption. Therefore, impaired glycaemic control in PC-T2D patients may be linked to impairment of luminal sweet sensing and its downstream signals. Further investigations are needed to define the functional connections between altered GI nutrient sensing and the pathophysiology of obesity and T2D.