Ghrelin Does Not Directly Stimulate Secretion of Glucagon-like Peptide-1

Abstract

Context: The gastrointestinal hormone ghrelin stimulates growth hormone secretion and appetite, but recent studies indicate that ghrelin also stimulates the secretion of the appetite-inhibiting and insulinotropic hormone glucagon-like peptide-1 (GLP-1).

Objective: To investigate the putative effect of ghrelin on GLP-1 secretion in vivo and in vitro.

Subjects and methods: A randomized placebo-controlled crossover study was performed in eight hypopituitary subjects. Ghrelin or saline was infused intravenously (1 pmol/min × kg) after collection of baseline sample (0 min), and blood was subsequently collected at time 30, 60, 90, and 120 minutes. Mouse small intestine was perfused (n = 6) and GLP-1 output from perfused mouse small intestine was investigated in response to vascular ghrelin administration in the presence and absence of a simultaneous luminal glucose stimulus. Ghrelin receptor expression was quantified in human (n = 11) and mouse L-cells (n = 3) by RNA sequencing and RT-qPCR, respectively.

Results: Ghrelin did not affect GLP-1 secretion in humans (area under the curve [AUC; 0-120 min]: ghrelin infusion = 1.37 ± 0.05 min × nmol vs. saline infusion = 1.40 ± 0.06 min × nmol [P = 0.63]), but induced peripheral insulin resistance. Likewise, ghrelin did not stimulate GLP-1 secretion from the perfused mouse small intestine model (mean outputs during baseline/ghrelin infusion = 19.3 ± 1.6/25.5 ± 2.0 fmol/min, n = 6, P = 0.16), whereas glucose-dependent insulinotropic polypeptide administration, used as a positive control, doubled GLP-1 secretion (P < 0.001). Intraluminal glucose increased GLP-1 secretion by 4-fold (P < 0.001), which was not potentiated by ghrelin. Finally, gene expression of the ghrelin receptor was undetectable in mouse L-cells and marginal in human L-cells.

Conclusions: Ghrelin does not interact directly with the L-cell and does not directly affect GLP-1 secretion.

Trial registration: ClinicalTrials.gov NCT01209416.

Figure 1.

Effects of intravenous ghrelin infusion on GLP-1 secretion. Plasma GLP-1 (total) concentrations are shown in response to intravenous acyl-ghrelin infusion (started after collection of 0 min sample; 1 pmol/min × kg; grey dots/bar) or saline infusion (black dots/white bar). Data are presented as means ± SEM. A: plasma GLP-1 total (pmol/l)and B: plasma GLP-1 total AUC (min × pmol/min). AUCs are from 0 to 120 min. n = 8 (A,B). Statistical significance was tested by paired t test. P-values **P < 0.01. P-values over 0.05 are indicated by respective pars. Individual values in B are indicated with dots. Plasma glucose and acyl-ghrelin data have been published previously as mentioned in the material and method section.

Figure 2.

Effects of acyl-ghrelin and glucose-dependent insulinotropic polypeptide on GLP-1 secretion from isolated perfused mouse small intestine. A: GLP-1 output (fmol/min) in response to intravascular acyl-ghrelin or intra-vascular glucose-dependent insulinotropic peptide 1–42 (GIP; both 200 nM). B: Mean GLP-1 outputs (fmol/min) within ghrelin and GIP administration periods or at preceding time periods (baseline 1 and 2). C: GLP-1 output (fmol/min) in response to intraluminal glucose (20% w/v) administration with or without simultaneously intravascular acyl ghrelin infusion (100 nM). D: Mean GLP-1 outputs (fmol/min) at within respective stimulus periods or at time periods leading up to the stimulations (baseline 1 and 2). E: Mean incremental GLP-1 outputs (fmol/min) in response to luminal glucose or luminal glucose+ghrelin. F-H: as (C-E) but reversed order of glucose/glucose+ghrelin administration. In both experiments glucose concentration in the perfusion buffer was 3.5 mM. Statistical significance was tested by one-way analysis of variance for repeated measurements followed by Bonferroni multiple comparison test (B, D and G) or by paired t test (E and H). Data are presented as means ± SEM; dots in B, D, and E indicate values from different animals. Non-significant P-values are indicated above respective bars. **P < 0.05, **P < 0.01, ***P < 0.001. n (A-E) = 6.

Figure 3.

Expression of growth hormone secretagogue receptor (ghrelin receptor) and glucose-dependent insulinotropic polypeptide receptor in L-cells. Expression is shown in A: human L-cells, enteroendocrine cells (EEC, not including L-cells) and nonenteroendocrine cells. Growth hormone secretagogue receptor (GHSR) expression is shown with white bars and glucose-dependent insulinotropic polypeptide receptor (GIPR) with grey bars. B: Ghsr and Gipr expression in L-cells and non-L-cells from mouse duodenum and ileum. Ghsr expression are shown with white bars; Gipr with grey bars. Data are presented as mean values ± SEM and are (A) log2 transformed normalized read counts and (B) – relative gene expression to β-actin (Actb) expression. Dots are values from cell populations sorted from different individuals/mice; a value of zero indicates lack of detection. Dotted line in (B) indicates detection limit. n (A) = 11, n (B) = 3.