Molecular mechanisms underlying glucose-dependent insulinotropic polypeptide secretion in human duodenal organoids.

AIMS/HYPOTHESIS: Glucose-dependent insulinotropic polypeptide (GIP) is an incretin hormone secreted by enteroendocrine K cells in the proximal small intestine. This study aimed to explore the function of human K cells at the molecular and cellular levels.

METHODS

CRISPR-Cas9 homology-directed repair was used to insert transgenes encoding a yellow fluorescent protein (Venus) or an Epac-based cAMP sensor (Epac-S-H187) in the GIP locus in human duodenal-derived organoids. Fluorescently labelled K cells were purified by FACS for RNA-seq and peptidomic analysis. GIP reporter organoids were employed for GIP secretion assays, live-cell imaging of Ca using Fura-2 and cAMP using Epac-S-H187, and basic electrophysiological characterisation. The G protein-coupled receptor genes GPR142 and CASR were knocked out to evaluate roles in amino acid sensing.

RESULTS

RNA-seq of human duodenal K cells revealed enrichment of several G protein-coupled receptors involved in nutrient sensing, including FFAR1, GPBAR1, GPR119, CASR and GPR142. Glucose induced action potential firing and cytosolic Ca elevation and caused a 1.8-fold increase in GIP secretion, which was inhibited by the sodium glucose co-transporter 1/2 (SGLT1/2) blocker sotagliflozin. Activation of the long-chain fatty acid receptor free fatty acid receptor 1 (FFAR1) induced a 2.7-fold increase in GIP secretion, while tryptophan and phenylalanine stimulated secretion by 2.8- and 2.1-fold, respectively. While CASR knockout blunted intracellular Ca responses, a CASR/GPR142 double knockout was needed to reduce GIP secretory responses to aromatic amino acids.

CONCLUSIONS/INTERPRETATION: The newly generated human organoid K cell model enables transcriptomic and functional characterisation of nutrient-sensing pathways involved in human GIP secretion. Both calcium-sensing receptor (CASR) and G protein-coupled receptor 142 (GPR142) contribute to protein-stimulated GIP secretion. This model will be further used to identify potential targets for modulation of native GIP secretion in diabetes and obesity.