Despite the importance of the gut microbiome to health, the role of human genetic variation in shaping its composition remains poorly understood. Here we report genome-wide association analyses of harmonized metagenomic data from 16,017 adults in four Swedish population-based studies, with replication in 12,652 people from the Norwegian HUNT study. We identified variants in the OR51E1-OR51E2 locus, encoding sensors for microbiome-derived fatty acids, associated with microbial richness. We further identified 15 study-wide significant genetic associations (P < 5.4 × 10-11) involving eight loci and 14 common bacterial species, of which 11 associations at six loci were replicated. The results confirm previously reported associations at LCT, ABO and FUT2, and provide evidence for new loci MUC12, CORO7-HMOX2, SLC5A11, FOXP1 and FUT3-FUT6, with supporting data from metabolomics and gene expression analyses. Our findings link gut microbial variation genetically to gastrointestinal functions, including enteroendocrine fatty acid sensing, bile composition and mucosal layer composition.

Glucose-dependent insulinotropic polypeptide (GIP) is an incretin hormone that promotes postprandial insulin release and euglycemia. Although GIP is the dominant incretin in humans, the precise mechanisms regulating GIP release from enteroendocrine K cells in vivo remain to be elucidated. Heterotrimeric G proteins of the Gi family regulate numerous key physiological functions. To explore the potential role of these proteins in modulating K cell function in vivo, we analyzed mutant mice that allowed us to selectively stimulate Gi signaling in K cells or studied mice lacking functional Gi in K cells only. Stimulation of K cell Gi signaling in vivo and in vitro led to reduced GIP release via an exchange protein directly activated by 3'-5'-cyclic adenosine monophosphate 2-dependent pathway and various metabolic deficits. In contrast, inactivation of this class of G proteins in K cells caused elevated plasma GIP levels and pronounced improvements in glucose homeostasis in lean and obese mice. These data provide important insights into the mechanisms of GIP release.

Central glucose-dependent insulinotropic polypeptide receptor (GIPR) signalling is required for the efficacy of GIP-based obesity therapeutics, yet how distinct subpopulations of GIPR neurons shape appetite remains undefined. Here we show that GIPR neurons in adjacent brainstem nuclei, the area postrema (AP) and nucleus tractus solitarius (NTS), exert opposing control over ingestion. We find GIPR AP neurons dampen post-ingestive satiation, permitting hyperphagia, whereas GIPR NTS neurons are anorectic. In line with this model, we show Gipr expression in AP, but not NTS, neurons is necessary for appetite suppression following GIPR antagonism. Additionally, we reveal that GIPR neurons in the AP and NTS occupy distinct gut-brain circuits, and are differentially sensitive to obesity-driven circuit remodelling. These data offer a framework for understanding how current GIPR agonist and antagonist strategies elicit weight loss.

Aims/hypothesisPostprandial glucagon-like peptide-1 (GLP-1) secretion by enteroendocrine L cells of the gut plays an important role in glucose homeostasis, thus representing a therapeutic option of ever-growing significance for type 2 diabetes. However, the precise mechanisms linking nutrient sensing and GLP-1 secretion are incompletely understood. In this study, we focused on a potential new role for endoplasmic reticulum (ER)-mitochondria contact sites, called mitochondria-associated membranes (MAMs), in nutrient-induced GLP-1 secretion by L cells, as they are dynamically regulated by nutrients, they influence cellular calcium homeostasis crucial for hormone secretion, and their miscommunication has been implicated in alterations of glucose homeostasis in several tissues.MethodsWe combined biochemical and imaging approaches to investigate nutrient-induced GLP-1 secretion, and ER-mitochondria interaction and calcium exchange in the STC-1 cell line, ex vivo ileal mouse organoids, and/or in vivo in gut enteroendocrine cells from Glu-Venus mice, both in acute conditions and after diet-induced obesity and type 2 diabetes.ResultsWe show here that ER-mitochondria interactions are dynamically induced by two GLP-1 secretagogues, glucose and deoxycholic acid (DCA), in STC-1 cells (1.8- and 2.1-fold, respectively), ileal mouse organoids (1.7- and 1.3-fold, respectively), and in vivo in colonic L cells of Glu-Venus mice (1.3- and 1.2-fold, respectively). In addition, glucose increased ER-mitochondria calcium exchange in STC-1 cells (1.2-fold). A paracrine action of secreted GLP-1 was also involved in the regulation of MAMs by glucose and DCA in STC-1 cells. Dynamic reinforcement of MAMs by glucose and DCA played a causal role in GLP-1 release, as both pharmacological and genetic disruption of organelle communication blocked L cell secretory response to the two stimuli in STC-1 cells. In agreement, depleting ER calcium levels or inhibiting mitochondrial calcium entry decreased glucose-induced GLP-1 secretion (-37.5% and -30.9%, respectively), whereas inducing ER or mitochondrial stress prevented it (-47.9% and -51.8%, respectively). Mechanistically, glucose induces ER-mitochondria communication through a sodium-glucose cotransporter 1-mediated electrogenic effect, whereas DCA acts through a Takeda G protein-coupled receptor 5 (TGR5)-cAMP-protein kinase A (PKA) pathway. Finally, we demonstrated in C57Bl/6J mice and in Glu-Venus mice that diet-induced obesity reinforced basal ER-mitochondria interactions in colonic L cells and blocked their ability to respond to oral glucose in terms of both GLP-1 secretion and MAM upregulation.Conclusions/interpretationThese results point to a new role for ER-mitochondria calcium coupling in glucose-induced GLP-1 secretion in L cells of the gut, which is impaired in obesity and type 2 diabetes, providing a novel target for the modulation of GLP-1 secretion. Therefore, these data reinforce the potential targeting of MAMs to improve glycaemic outcomes in metabolic diseases.

Growing evidence implicates gut microbiota-derived metabolites in metabolic homeostasis. Indole, a microbial tryptophan metabolite, has been reported to enhance glucagon-like peptide-1 (GLP-1) secretion in vitro, and its derivatives have been inversely associated with risk of type 2 diabetes. We hypothesised that indole acts via the gastrointestinal tract to modulate glucose homeostasis, and tested this hypothesis using in vitro and in vivo models. We measured GLP-1 secretion from cultured murine enteroendocrine cells, and evaluated intraperitoneal glucose tolerance and hormone secretion in mice following indole treatment. Subsequently, the impact of indole on intestinal epithelial cell fate and L cell number was examined using murine ileal organoid cultures and in vivo. Finally, we explored the effect of chronic indole administration on metabolic outcomes in a murine model of type 2 diabetes. Indole stimulated in vitro GLP-1 secretion in a concentration-dependent manner, and improved acute glucose management in vivo. Additionally, we demonstrate that indole drives enteroendocrine L cell differentiation in murine ileal organoids, resulting in increased L cell density and longer-term glucoregulatory benefits in vivo. Finally, sub-chronic indole administration improved glucose tolerance and insulin sensitivity in a diabetic mouse model. Our findings identify indole as a glucose-lowering molecule that acts on the gut, and raise the possibility of incorporating indole into nutraceutical supplements to aid in the treatment or prevention of type 2 diabetes. This study highlights the importance of gut microbiota-derived metabolites in metabolic health and opens new avenues for developing novel strategies to combat type 2 diabetes. RNA sequencing data are available from the Gene Expression Omnibus under accession number GSE306720.

The glucose-dependent insulinotropic polypeptide receptor (GIPR) is a class B1 G protein-coupled receptor (GPCR) that promotes glucose-dependent insulin secretion upon activation by GIP. Dual agonism of GIPR and glucagon-like peptide-1 receptor (GLP-1R) has emerged as a breakthrough therapeutic strategy for type 2 diabetes and obesity, improving glycaemic control and promoting weight loss. Although GIPR is widely expressed in the central nervous system, endogenous GIP expression in the brain is controversial, and peripherally-administered fluorescent GIPR/GLP-1R agonists predominantly localise to circumventricular organs, suggesting a lack of GIP-ligand for receptors shielded by the blood-brain barrier. Here, we considered the existence of alternative endogenous ligands for GIPR, potentially modulated by receptor activity-modifying proteins (RAMPs). Ligand activity was profiled at heterologously expressed human GIPR using cAMP accumulation, calcium mobilisation, and cAMP inhibition assays. Among the 42 ligands tested, glicentin was the only non-proGIP-derived peptide to elicit a cAMP response at GIPR (EC₅₀ = 877 nM). No ligand-induced calcium mobilisation or cAMP inhibition was observed. Glicentin also activated human GLP-1R and glucagon receptor (GCGR), with receptor-specific antibody blockade reducing potency by 10-20-fold at all three receptors. Glicentin's activity was conserved at rodent GIPRs, albeit with reduced potency. Co-expression with RAMP2 or RAMP3 attenuated cAMP responses to GIPR-active ligands, including glicentin, but did not affect other ligands tested. These findings show that the GIPR is highly selective and identifies glicentin as a low-potency, cross-reactive agonist at GIPR, GLP-1R and GCGR, although its high EC₅₀ makes it unlikely to act as a physiological ligand under basal conditions.

Background Insulin-like peptide 5 (INSL5) is an enteroendocrine hormone expressed in distal colonic ‘L cells’. Bile acid receptor agonists are known to stimulate INSL5 secretion in primary cell culture, and administration of an INSL5 analogue in animals promotes colonic motility. Objective This study used a new immunoassay to measure INSL5 in human blood samples, enabling assessment of whether rectal bile acids stimulate INSL5 release in humans and whether INSL5 levels are altered in patients with chronic diarrhoea. Design Serum/plasma samples from previously performed studies were used, including healthy volunteers (n=7) who received a rectal enema of taurocholic acid (TCA); fasting and post prandial samples from healthy volunteers (n=10); patients with bile acid diarrhoea (BAD) (n=19) or irritable bowel syndrome with diarrhoea (IBS-D) (n=8); and patients with IBS-D (n=64) treated with ondansetron or placebo. Results Rectal TCA but not a control enema promptly elevated plasma INSL5, with the increase in INSL5 correlating negatively with time to, and positively with desire to, defecate post enema. Healthy volunteers had low INSL5 levels (<100 pg/mL), with no change following a mixed meal. Patients with BAD had elevated INSL5 levels, with average stool consistency being positively correlated with serum INSL5 (p<0.001). In people with IBS-D, INSL5 was elevated (>100 pg/mL) in 42%, and this subgroup showed greater improvements in stool consistency with ondansetron therapy (p<0.05). Conclusion The study highlights that rectal bile acids stimulate INSL5 secretion in humans, and that INSL5 levels are associated with a colonic pro-motility response and pathophysiology of chronic diarrhoea.

Nausea arises from activation of specialized neurons in the area postrema (AP)1-8. The AP also mediates much of the satiety produced by GLP1R agonists9-12, suggesting a broader role in non-aversive physiology, yet the functions of AP cell types are not well understood. Here, we have used optical recordings in behaving mice to systematically define the natural regulation of an array of AP neurons, including the cell types that are principal targets of widely-used weight loss drugs. We discover that neurons expressing GFRAL, the receptor for the sickness-related hormone GDF15, are unexpectedly activated when mice consume food rich in fat. This fat-specific GFRAL neuron activation is required for fat satiation but does not involve GDF15 or canonical gut-brain pathways. Instead, "anti-nausea" neurons expressing GIPR, which directly inhibit GFRAL neurons, are selectively activated by sugar, enabling macronutrient-specific gating of GFRAL responses. In addition, we show that CALCR neurons link intestinal hyperosmolality to the suppression of feeding, whereas PRLHR neurons respond to changes in blood volume and pressure. These findings reveal a broad role for AP cell types in sensing and responding to physiologic signals unrelated to nausea. They also reveal that GFRAL and GIPR neurons, which are key targets of the weight-loss drug tirzepatide, have a natural function in sensing ingestion of fat and sugar, respectively.

Glucagon-like peptide-1 receptor (GLP-1R) activation in the brain strongly reduces appetite, but most brain GLP-1Rs are not accessible for systemically administered GLP-1R agonists. Acute activation of nucleus tractus solitarius (NTS) GLP-1 neurons, known as preproglucagon (PPG) neurons, strongly suppresses food intake separate from GLP-1R agonists. However, it is unknown if chronic stimulation of PPG neurons is a viable strategy for appetite suppression, or if obesity disrupts their function. Here we demonstrate that PPG neurons in the NTS and intermediate reticular nucleus (IRT) determine meal size, and that their total number is inversely correlated with bodyweight gain. We report that PPGNTS and PPGIRT neurons receive distinct monosynaptic inputs, but have convergent efferent projection targets throughout the brain, and that combined ablation of both populations delays the onset of physiological satiation to a degree sufficient to promote weight gain under ad libitum chow fed conditions. Crucially, chronic daily chemogenetic activation of PPGNTS+IRT neurons drives robust and sustained hypophagia and weight loss in obese mice without notable adverse effects, demonstrating their value as targets for obesity pharmacotherapy.

Incretin-based pharmacology has revolutionized the medical treatment of type 2 diabetes and obesity. The most effective drug to date is tirzepatide, a dual incretin receptor agonist that engages both the glucagon-like peptide 1 receptor (GLP-1R) and the glucose-dependent insulinotropic polypeptide receptor (GIPR). While the relative contributions of GIPR and GLP-1R actions to the clinical effects of tirzepatide have not been established, the potency of this agent has reignited interest in the clinical potential of GIPR agonism. Here, we discuss incretin biology as it relates to metabolic pharmacology and contextualize the mechanisms by which GIPR activity could contribute to the development of new and effective drugs. We explore current and future applications of GIPR agonists and antagonists, to underscore the potential that this signaling system could add to treatment of type 2 diabetes and obesity.

The development of dual agonists for the glucagon-like peptide-1 receptor (GLP-1R) and glucose-dependent insulinotropic polypeptide receptor (GIPR) has been a landmark moment in the treatment of type 2 diabetes and obesity. However, for reasons that are incompletely understood, in preclinical and clinical studies, adding either a GIPR agonist or GIPR antagonist to GLP-1R agonism causes additional weight loss1. Here we show that distinct brain regions mediate the appetite-suppressing effects of GIPR agonists and the synergistic weight loss effects conferred by GIPR antagonists. We knock out Gipr in either the area postrema (AP) or hypothalamus of mice (GiprAP-KO and Giprhypo-KO, respectively) and compare body weight and food intake responses to GIPR agonists and antagonists, alone and in combination with the GLP-1R agonist liraglutide. GiprAP-KO mice exhibit partial protection against diet-induced obesity, reduced responsiveness to the appetite-suppressing effects of acyl-GIP and a reduced ability of acyl-GIP to prevent avoidance triggered by peptide YY. Weight loss effects of liraglutide are comparable in GiprAP-KO and control mice, and the co-administration of a GIPR antagonist peptide causes similar additional weight loss in both groups. Giprhypo-KO mice, by contrast, exhibit normal appetite suppression by acyl-GIP but enhanced weight loss on liraglutide compared with control mice. Giprhypo-KO also abolishes the synergistic effect of a GIPR antagonist when combined with liraglutide-an effect that is not mediated by nucleus tractus solitarius preproglucagon neurons. GIPR antagonism and Giprhypo-KO also sensitise to cagrilintide-induced weight loss. Overall, our results suggest that the AP is responsible for the appetite-suppressing effects of GIPR agonism but that GIP receptors in the hypothalamus underlie the ability of GIPR antagonism to enhance the weight loss effects of GLP-1R and amylin receptor agonists.