As a regulator of food intake and energy metabolism, the role of ghrelin in glucose metabolism is still not fully understood. In this study, we determined the in vivo effect of ghrelin on incretin effect. We demonstrated that ghrelin inhibited the glucose-stimulated release of glucagon-like peptide-1 (GLP-1) when infused into the portal vein of Wistar rat. Hepatic vagotomy diminished the inhibitory effect of ghrelin on glucose-stimulated GLP-1 secretion. In addition, phentolamine, a nonselective α receptor antagonist, could recover the decrease of GLP-1 release induced by ghrelin infusion. Pralmorelin (an artificial growth hormone release peptide) infusion into the portal vein could also inhibit the glucose-stimulated release of GLP-1. And growth hormone secretagogue receptor antagonist, [D-lys3]-GHRP-6, infusion showed comparable increases of glucose stimulated GLP-1 release compared to ghrelin infusion into the portal vein. The data showed that intraportal infusion of ghrelin exerted an inhibitory effect on GLP-1 secretion through growth hormone secretagogue receptor 1α (GHS1α receptor), which indicated that the downregulation of ghrelin secretion after food intake was necessary for incretin effect. Furthermore, our results suggested that the enteric neural net involved hepatic vagal nerve and sympathetic nerve mediated inhibition effect of ghrelin on incretin effect.
Ghrelin is a 28-amino acid peptide isolated from human and rat stomach as an endogenous natural ligand of GHS1α receptor (growth hormone secretagogue receptor 1α) [
Although functions of ghrelin have been discovered more and more, as a gastric-intestine hormone, and secreted with a rhythm following food intake, the role of ghrelin on energy homeostasis is always a hot topic. Endogenous ghrelin has an important role in insulin secretion. Glucose-stimulated insulin secretion is reduced with exogenous ghrelin in humans [
In the study, we demonstrate the relationship of ghrelin and incretin effect. We investigate the change of glucose-stimulated GLP-1 concentration after exogenous ghrelin infused into the portal vein.
Normal male Wistar rats (age, 91 ± 8 d; body weight, 260 ± 21.3 g) were used in this study. All the rats were housed in wire-bottomed, stainless-steel cages and maintained in an ambient temperature of 20°C, with a light cycle between 0600 and 1800 h. The rats had free access to tap water and standard chow diet (Animal Nutrition Center, Harbin, China). The study was approved by the Ethics Committee of Harbin Medical University, China.
All experiments were performed after anesthetizing the animals by intraperitoneally injecting 50 mg/kg pentobarbital sodium after a 12 h starvation period.
After the induction of anesthesia was confirmed by the loss of the corneal reflex, polyethylene catheters (size, 0.5–1.0 mm) were inserted into the following veins: (1) left jugular vein for the collection of blood samples and (2) portal vein for the injection of agents.
After the polyethylene catheter was placed into the portal vein, the hepatic branch of the anterior vagal trunk was either sectioned below the diaphragm or subjected to a sham operation as described in detail in the previous report [
After 30-minute rest after the surgery, glucose was administered through stomach catheter by 1 gram per kilogram (1 g/kg).
A dose of 1 ng/kg/mL of rat acylated (active) ghrelin (AG; Peptide Institute, Osaka, Japan) was pumped into the portal vein from 0 to 60 min at a rate of 1 mL/h by using a micropump after glucose administered by stomach catheter.
KP-102 (GHRP-2, pralmorelin) (Peptide Institute, Osaka, Japan), a kind of artificial growth hormone release peptide, was infused at a dose of 1
Phentolamine, a nonselective α receptor antagonist, was infused at a dose of 15 ng/kg/mL into the portal vein from 0 to 60 min with or without ghrelin (1 ng/kg/mL) after glucose administered by stomach catheter.
[D-lys3]-GHRP-6, GHS1α receptor antagonist, was infused at a dose of 30 ng/kg/mL into the portal vein from 0 to 60 min with or without ghrelin (1 ng/kg/mL) infused after glucose administered by stomach catheter.
Blood samples were drawn from the jugular vein at 0, 5, 10, 15, 30, and 60 min and immediately transferred into polypropylene tubes containing 1 mg/mL EDTA-2Na, aprotinin (final concentration, 500 kallikrein-inhibiting units (KIU)/mL), and DPP-4 inhibitor (10
The parameters were compared among the groups with ANOVA and the Mann-Whitney
After glucose was loaded by stomach catheter, the plasma glucose concentrations were higher in the group of ghrelin infusion into the portal vein compared to those of saline infusion group (
Plasma concentrations of glucose (a), insulin (b), and GLP-1 (c) during ghrelin infusion in the OGTT. Ghrelin (1 ng/kg/mL) was infused into the portal vein (i.p.) from 0 to 60 min at a rate of 1 mL/h using a micropump. A 20% glucose solution (1 g/kg) was administered through stomach catheter orally. Data are represented as mean ± SE. Open squares, saline i.p. group (
When the hepatic branch of the anterior vagal trunk was sectioned, the insulin responses occurring at 10 and 15 min of OGTT were lower, and the plasma glucose levels were higher correspondingly (
Plasma concentrations of glucose (a), insulin (b), and GLP-1 (c) during ghrelin infusion in the OGTT after hepatic vagotomy. After a 30 min rest after hepatic vagotomy or sham operation, ghrelin (1 ng/kg/mL) was infused into the portal vein (i.p.) from 0 to 60 min at a rate of 1 mL/h using a micropump. A 20% glucose solution (1 g/kg) was administered through stomach catheter orally. Data are represented as mean ± SE. Open squares, saline i.p. with sham operation group (
Phentolamine, a nonselective α receptor antagonist, infusion at 15 ng/kg/mL recovered the inhibitory effect of ghrelin on insulin and GLP-1 release in OGTT (
Plasma concentrations of glucose (a), insulin (b), and GLP-1 (c) during ghrelin infusion with or without phentolamine in the OGTT; phentolamine (15 ng/kg/mL) and ghrelin (1 ng/kg/mL) were infused into the portal vein (i.p.) from 0 to 60 min at a rate of 1 mL/h using a micropump. A 20% glucose solution (1 g/kg) was administered through stomach catheter orally. Data are represented as mean ± SE. Open squares, saline i.p. closed squares, ghrelin i.p.; and closed triangles, ghrelin with phentolamine i.p.
After glucose was loaded by stomach catheter, there were great changes of the plasma glucose concentrations by the KP-102 infusion into the portal vein at the dose of 1
Plasma concentrations of glucose (a), insulin (b), and GLP-1 (c) during KP-102 infusion with or without ghrelin in the OGTT. KP-102 (1
Ghrelin coinfusion with [D-lys3]-GHRP-6, GHS1α receptor antagonist, at the dose of 30 ng/kg/mL did not induce the inhibitory effect on insulin and GLP-1 response compared with the intraportal ghrelin infusion alone group (
Plasma concentrations of glucose (a), insulin (b), and GLP-1 (c) during ghrelin infusion with or without GHRP-6 in the OGTT; GHRP-6 (30 ng/kg/mL) and ghrelin (1 ng/kg/mL) were infused into the portal vein (i.p.) from 0 to 60 min at a rate of 1 mL/h using a micropump. A 20% glucose solution (1 g/kg) was administered through stomach catheter orally. Data are represented as mean ± SE. Open squares, saline i.p. closed squares, ghrelin i.p.; and closed triangles, ghrelin with GHRP-6 i.p.
In the previous study, the acute infusion of ghrelin from portal vein actively inhibited the glucose-stimulated insulin secretion in IPGTT in rats [
The hepatic portal system is closely related to insulin secretion and glucose metabolism, including incretin effect. It is known that the vagus nerve organizes the hepatic portal system and the afferent firing of the vagus nerve is mediated by the glucose concentration in the portal vein. We have observed that exogenous ghrelin inhibited the glucose-induced insulin secretion via its action on the vagus nerve in IPGTT [
The functioning of the neuroendocrine loop mediates proximal nutrients induced GLP-1 secretion [
Ghrelin regulates GLP-1 secretion in portal system through its GHS1α receptor. The evidences were provided by the data of KP-102 and [D-lys3]-GHRP-6 administered groups. KP-102 infused into the portal also inhibited glucose stimulated GLP-1 secretion, which indirectly approved that ghrelin mediated GLP-1 secretion through its receptor GHS1α on hepatic vagus nerve. As a kind of artificial growth hormone release peptide, GHS1α was the only known receptor of KP102. Furthermore, the result that [D-lys3]-GHRP-6, GHS1α receptor antagonist, coinfusion with ghrelin recovered the inhibitory effect of ghrelin on GLP-1 secretion gave a direct evidence.
Many studies using animal models selected the time of OGTT longer (for example, until 120–150–180 minutes) than 60 minutes. We explained the choice of time to 60 minutes for OGTT and ghrelin infusion. First, in the study, we tested GLP-1 levels, which need relatively more blood samples for assay (1 mL for each point). If we take more time points to 120 minutes, even 180 minutes, the blood sample would not have been enough for a rat. Some rats could not last to 120 min or longer. Second, the animal selected in the study was in normal glucose tolerance and our results showed that 60 min of OGTT was enough to explain the changes of glucose, insulin, and GLP-1 by exogenous ghrelin infusion, especially for GLP-1. The time point selected is enough to reflect the physical changes.
In the prestudy, we had selected the glucose dose for OGTT. We found that, administering glucose at the dose of 1 g/kg compared to dose of 2 g/kg, there is no obvious difference for the results of plasma glucose and insulin levels. The mean body weight of rats in our study is about 250 g, not fat, so we choose the low dose. To avoid the direct insulinotropic effect of phentolamine, we had screened for the appropriate dose of phentolamine at gradient of 15 mg/kg/m, 1.5 mg/kg/mL, 150 ng/kg/mL, 15 ng/kg/mL, and 5 ng/kg/mL. At last, we choose the dose of 15 ng/kg/mL, at which there is no direct effect on insulin secretion. Phentolamine is a nonselective α receptor antagonist. Before the test, we do not know what result we could get. We think that it is better to choose a nonselective antagonist. In the later study, we plan to choose a selective one.
In summary, we observed that exogenous ghrelin inhibited incretin effect through the portal enteric nerve loop. The high concentration of ghrelin on fast and rapid downregulation after meal is one of a switch to initiate incretin effect. The functions of neuroendocrine loop in gut-insulin axis may be very helpful for understanding the role of this axis in pathogenesis of type 2 diabetes.
The authors declare that there is no conflict of interests regarding the publication of this paper.
This work was supported by funding from the National Nature Science Foundation of China (Grant no. 81100599).