Somatostatin receptor 5 and cannabinoid receptor 1 activation inhibit secretion of glucose-dependent insulinotropic polypeptide from intestinal K cells in rodents

Abstract

Aims/hypothesis: Glucose-dependent insulinotropic polypeptide (GIP) is an enteroendocrine hormone that promotes storage of glucose and fat. Its secretion from intestinal K cells is triggered by nutrient ingestion and is modulated by intracellular cAMP. In view of the proadipogenic actions of GIP, this study aimed to identify pathways in K cells that lower cAMP levels and GIP secretion.

Methods: Murine K cells purified by flow cytometry were analysed for expression of G(αi)-coupled receptors by transcriptomic microarrays. Somatostatin and cannabinoid receptor expression was confirmed by quantitative RT-PCR. Hormone secretion in vitro was measured in GLUTag and primary murine intestinal cultures. cAMP was monitored in GLUTag cells using the genetically encoded sensor Epac2-camps. In vivo tolerance tests were performed in cannulated rats.

Results: Purified murine K cells expressed high mRNA levels for somatostatin receptors (Sstrs) Sstr2, Sstr3 and Sstr5, and cannabinoid receptor type 1 (Cnr1, CB1). Somatostatin inhibited GIP and glucagon-like peptide-1 (GLP-1) secretion from primary small intestinal cultures, in part through SSTR5, and reduced cAMP generation in GLUTag cells. Although the CB1 agonist methanandamide (mAEA) inhibited GIP secretion, no significant effect was observed on GLP-1 secretion from primary cultures. In cannulated rats, treatment with mAEA prior to an oral glucose tolerance test suppressed plasma GIP but not GLP-1 levels, whereas the CB1 antagonist AM251 elevated basal GIP concentrations.

Conclusions/interpretation: GIP release is inhibited by somatostatin and CB1 agonists. The differential effects of CB1 ligands on GIP and GLP-1 release may provide a new tool to dissociate secretion of these incretin hormones and lower GIP but not GLP-1 levels in vivo.

Fig. 1

Enteroendocrine cells express Sstr2, Sstr 3 and Sstr 5. (a) Mean microarray RMA intensities for probes against Sstr1–Sstr5 in K cells (K+, black bars), small intestinal and colonic L cells (L+, dark grey bars; LC+, light grey bars, respectively) and non-fluorescent control cells from the same tissue preparations (K–, diagonal hatching; L–, horizontal hatching; LC–, vertical hatching) (n = 2–3 each). (b–d) Relative expression of Sstr2 (b), Sstr3 (c) and Sstr5 (d) mRNA relative to β-actin assessed by RT-PCR in FACS-sorted cell populations labelled as in (a) plus GLUTag cells (white bars). Data are presented as the geometric mean and upper SEM (n > 3 each). Significance comparisons between K+ and K– , L+ and L–, and K+ and L+ cells were calculated by one-way ANOVA with post hoc Bonferroni test performed on the log(base 2) data: **p < 0.01, ***p < 0.001

Fig. 2

Somatostatin (Sst) lowers cAMP in the GLUTag cell line. (a) Changes in cAMP concentration in response to G_αi activation. GLUTag cells transfected with Epac2-camps were perfused with forskolin (fsk; 2 μmol/l) with/without Sst (100 nmol/l), followed by fsk/IBMX (10 and 100 μmol/l respectively) as indicated. CFP and YFP emission was monitored in response to excitation with 435/10 nm. Three traces, each representing the CFP/YFP ratio of a single cell, are shown. (b) Mean changes in the CFP/YFP emission ratio in response to application of forskolin (2 μmol/l), Sst (100 nmol/l) or SSTR5A (100 nmol/l) in experiments performed as in (a). Responses to fsk/Sst, fsk/Sst/SSTR5A and the final application of fsk alone were normalised to the response triggered by the first application of fsk in the same cell. Data are means and error bars show 1 SEM, with the number of cells (monitored in between five and seven independent experiments per condition) indicated above each bar. Statistical significance was assessed by one-way ANOVA with post hoc Bonferroni test: ***p < 0.001 compared with first application of fsk and ^††p < 0.01 comparing the Sst response in the absence/presence of SSTR5A

Fig. 3

Inhibition of IBMX-stimulated GIP and GLP-1 secretion by somatostatin (Sst). (a) GIP secretion from primary small intestinal cultures treated with IBMX (100 μmol/l) with/without Sst (100 nmol/l) and SSTR5A (100 nmol/l), as indicated. (b,c) GLP-1 secretion from primary small intestinal (b) and colonic (c) cultures treated with IBMX (100 μmol/l) with or without Sst (100 nmol/l) and SSTR5A (100 nmol/l), as indicated. Data are expressed relative to the basal secretory rate measured in parallel on the same day. Data are means ± SEM of the number of wells indicated above each bar. Statistical significance was assessed by one-way ANOVA with post hoc Bonferroni test: **p < 0.01, ***p < 0.001 compared with control; ^†p < 0.05, ^††p < 0.01, ^†††p < 0.001 between conditions as indicated

Fig. 4

Expression of endocannabinoid receptors in enteroendocrine cells. (a) Mean microarray RMA intensities for probes against Cnr1 and Cnr2 in K cells (K+, black bars), small intestinal and colonic L cells (L+, dark grey bars; LC+, light grey bars, respectively), and non-fluorescent control cells from the same tissue preparations (K–, diagonal hatching; L–, horizontal hatching; LC–, vertical hatching) (n = 2–3 each). (b,c) Relative expression of Cnr1 (b) andCnr2 (c) mRNA relative to β-actin assessed by RT-PCR in FACS-sorted cell populations labelled as in (a), together with GLUTag cells (white bar). Data are presented as the geometric mean and upper SEM (n > 3 each). Significance comparisons between K+ and K– , L+ and L–, and K+ and L+ cells were calculated by one-way ANOVA with post hoc Bonferroni test performed on the log(base 2) data: ***p < 0.001

Fig. 5

Effects of CB1 ligands on K and L cells. (a) GIP secretion from primary small intestinal cultures treated with IBMX (100 μmol/l), mAEA (10 μmol/l) and AM251 (1 μmol/l), as indicated. (b, c) GLP-1 secretion from primary small intestinal (b) and colonic (c) cultures treated with IBMX (100 μmol/l), mAEA, (10 μmol/l) and AM251 (1 μmol/l), as indicated. In (a–c) Data are expressed relative to the basal secretory rate measured in parallel on the same day. Data represent the means ± SEM of the number of wells indicated above each bar. Statistical significance was assessed by one-way ANOVA with post hoc Bonferroni test: **p < 0.01, ***p < 0.001 compared with control buffer; ^††p < 0.01, ^†††p < 0.001 between conditions as indicated. (d) GLP-1 secretion from GLUTag cells treated with IBMX (100 μmol/l), mAEA (10 μmol/l) and AM251 (1 μmol/l), as indicated. GLP-1 secretion is expressed relative to the basal secretion measured in parallel on the same day. Data are means ± SEM of the number of wells indicated above each bar. Statistical significance was assessed by one-way ANOVA with post hoc Bonferroni test: ***p < 0.001 compared with control; ^††p < 0.01 between conditions. (e) GLUTag cells transfected with Epac2-camps were perfused with forskolin (fsk; 2 μmol/l) with or without mAEA (10 μmol/l), followed by fsk/IBMX (10 and 100 μmol/l respectively), as indicated. CFP and YFP emission was monitored in response to excitation with 435/10 nm. Three traces, each representing the CFP/YFP ratio of a single cell, are shown. (f) Mean changes in the CFP/YFP emission ratio in response to application of fsk, mAEA or AM251 (1 μmol/l), in experiments performed as in (e). Responses to fsk/mAEA, fsk/mAEA/AM251 and the final application of fsk alone were normalised to the response triggered by the first application of fsk in the same cell. Data are means and error bars show 1 SEM with the number of cells (monitored in between five and seven independent experiments per condition) indicated above each bar. Statistical significance was assessed by one-way ANOVA with post hoc Bonferroni test: **p < 0.01, ***p < 0.001 compared with first application of fsk; ^†p < 0.05 comparing the mAEA response in the absence and presence of AM251

Fig. 6

Effects of CB1 ligands on hormone secretion in vivo. (a–c) Plasma GIP levels in individual rats treated at −30 min with mAEA (a), vehicle control (squares represent controls for mAEA, diamonds represent controls for AM251) (b), or AM251 (c). Samples at 0 min were taken immediately prior to glucose gavage. (d–j) Plasma profiles for GIP (d), GLP-1 (g), insulin (h), glucose (i) and acetaminophen (j), and area under the curve (AUC) for GIP (e) or GLP-1 (f), following i.p. administration of mAEA (black circles and black bars; n = 8) or vehicle control (white squares and open bars; n = 5) at −30 min and glucose/acetaminophen gavage at 0 min. Data are means ±1 SEM of the plasma concentrations observed at the times indicated. Statistical significance was assessed in (d) and (g–j) by two-way ANOVA with post hoc Bonferroni test (**p < 0.01 vs vehicle control) and in (e) and (f) by unpaired Student’s t test (*p < 0.05)