Duodenal GLP-1 signaling regulates hepatic glucose production through a PKC-δ-dependent neurocircuitry

Abstract

Intestinal glucagon-like peptide-1 (GLP-1) is a hormone that stimulates insulin secretion and acts as a neuropeptide to control glucose homeostasis, but little is known whether intestinal GLP-1 has any effect in the control of hepatic glucose production (HGP). Here we found that intraduodenal infusion of GLP-1 activated duodenal PKC-δ, lowered HGP and was accompanied by a decrease in hepatic expression of gluconeogenic enzymes and an increase in hepatic insulin signaling in rats. However, gut co-infusion of either the GLP-1 receptor antagonist Ex-9, or the PKC-δ inhibitor rottlerin with GLP-1, negated the ability of gut GLP-1 to lower HGP and to increase hepatic insulin signaling during clamps. The metabolic and molecular signal effects of duodenal GLP-1 were also negated by co-infusion with tetracaine, pharmacologic inhibition of N-methyl-d-aspartate receptors within the dorsalvagal complex, or hepatic vagotomy in rats. In summary, we identified a neural glucoregulatory function of gut GLP-1 signaling.

Figure 1

Gut GLP-1 inhibits liver glucose production through GLP-1 receptors. (a) Schematic representation of working hypothesis. GLP-1 with or without Ex-9 was infused through a duodenal catheter. GLP-1R, GLP-1 receptor; GP, glucose production. (b) Experimental procedure and clamp protocol. Duodenal catheter or venous and arterial catheters were implanted on day 1. The pancreatic (basal insulin) clamp studies were performed on day 5. (c) Glucose infusion rates (GIR) during the steady state of clamp (180–200 min). (d) Cumulative GIR during the steady state of clamp. (e) Hepatic glucose production (HGP). (f) Suppression of HGP during the clamp period expressed as the percentage reduction from basal HGP. (g) Glucose uptake. Data are means±S.E.M. (saline (n=8), GLP-1 (n=7), Ex-9 (n=6), GLP-1+Ex-9 (n=8)). *P<0.05, ^ΔP<0.01 versus saline and Ex-9; ^‡P<0.05, ^#P<0.01 versus GLP-1+Ex-9; ^▴P<0.01 versus all other groups. (h–j) The effect of GLP-1 receptors inhibition in the gut on glucose homeostasis during fasting–refeeding. (h) Schematic representation of experimental design. Duodenal catheter or venous and arterial catheters were implanted on day 1. Rats were subjected to a 40 h fasting from 1700 hours on day 5 until 0900 hours on day 7. Ten minutes before the completion of the 40 hour fast, rats were infused with intraduodenal saline or Ex-9 (n=6 for each group). Rats were refed on regular chow at time 0 min where food intake and blood glucose were monitored for 20 min. (i) Plasma glucose levels during refeeding. (j) Cumulative food intake during refeeding. Values are shown as mean±S.E.M. *P<0.05, **P<0.01 versus saline

Figure 2

Duodenal GLP-1 inhibits glucose production by activating a gut–brain–liver neurocircuitry. (a) Schematic representation of the working hypothesis. Gut GLP-1 was co-infused with tetracaine through a duodenal catheter, which abolishes the ascending neuronal signal to the brain. A subgroup of rats was given MK-801, an NMDA receptor inhibitor, directly into the NTS. In another studies, gut GLP-1 was infused into rats that underwent HVAG. HVAG, hepatic vagotomy; GP, glucose production; NMDA, N-methyl-d-aspartate; NTS, nucleus of the solitary tract. (b) Experimental procedure and clamp protocol. Stereotaxic surgeries were conducted on day 1. Duodenal catheter or venous and arterial catheters were implanted on day 7. HVAG was performed immediately before the implantation of the duodenal and vascular catheters. (c–e) Gut GLP-1 infusion increased glucose infusion rate and lowered GP. Rats that received tetracaine in the gut, MK-801 in the NTS or HVAG failed to respond to duodenal GLP-1 to increase the glucose infusion rate and lower GP. (f) Suppression of GP during the clamp expressed as the percentage decrease from basal GP. (g) Glucose uptake was unchanged in all groups. Values are shown as mean±S.E.M. (saline (n=6), GLP-1 (n=6), Tetracaine (n=5), GLP-1+tetracaine (n=7), NTS MK-801 (n=5), GLP-1+NTS MK-801 (n=8), HVAG (n=6), GLP-1+HVAG (n=8)).*P<0.05, **P<0.01, versus all other groups

Figure 3

GLP-1 receptor is required for duodenal lipids to suppress hepatic glucose production. (a) Schematic representation of the working hypothesis. Lipid with Ex-9 or saline was infused through a duodenal catheter. (b) Experimental procedure and clamp protocol. (c–f) Gut lipids infusion increased the GIR (c and d), and decreased GP (e and f). When duodenal lipid was co-infused with Ex-9, the effects of lipids on GIR and GP were abolished. (g) Glucose uptake was unchanged in all groups. Values are shown as mean±S.E.M. (Ex-9 (n=5), lipid (n=8), lipid+Ex-9 (n=8)). *P<0.05, **P<0.01 versus all other groups

Figure 4

Gut GLP-1 inhibits hepatic glucose production through PKC-δ activation. (a) Schematic of working hypothesis (left), experimental procedure and clamp protocol (right). GLP-1 was infused with or without rottlerin, a PKC-δ specific inhibitor, through a duodenal catheter. GLP-1R, GLP-1 receptor; GP, glucose production; ROT, rottlerin. (b) Representative western blots (n=6) of PKC-δ in the mucosal layer of the duodenum. (c–f) Gut GLP-1 infusion increased the GIR (c and d), and decreased GP (e and f). When duodenal GLP-1 was co-infused with rottlerin, the effects of GLP-1 on GIR and GP were abolished. (g) Glucose uptake was comparable in all groups. Values are shown as mean±S.E.M. (saline (n=6), GLP-1 (n=6), saline+ROT (n=5), GLP-1+ROT (n=5)). *P<0.05, **P<0.01 versus all other groups

Figure 5

Duodenal GLP-1 suppresses hepatic PEPCK and G6Pase expression through a gut–brain–liver neurocircuitry. Intraduodenal GLP-1 infusion in rats suppressed hepatic PEPCK and G6Pase mRNA (n=6) (a) and protein expression (n=6) (b–d). In contrast, rats that received tetracaine (b), MK-801 in the NTS (c) or HVAG (d) failed to respond to duodenal GLP-1 to suppress hepatic PEPCK and G6Pase expression. HVAG, hepatic vagotomy; NTS, nucleus of the solitary tract. Values are shown as mean±S.E.M. (saline (n=6), GLP-1 (n=6), Tetracaine (n=5), GLP-1+tetracaine (n=8), NTS MK-801 (n=5), GLP-1+NTS MK-801 (n=7), HVAG (n=5), GLP-1+HVAG (n=7)). *P<0.01 versus all other groups. (e–g) Duodenal GLP-1 augments hepatic insulin signaling through a gut–brain–liver neurocircuitry. Representative western blots (n=6) and ratios of protein levels of phosphorylated InsR, IRS-1, AKT and AMPK to total InsR, IRS-1, AKT and AMPK in livers from experiments shown in Figure 2b. HVAG, hepatic vagotomy; NTS, nucleus of the solitary tract. Values are shown as mean±S.E.M. (saline (n=6), GLP-1 (n=6), Tetracaine (n=5), GLP-1+tetracaine (n=5), NTS MK-801 (n=5), GLP-1+NTS MK-801 (n=5), HVAG (n=5), GLP-1+HVAG (n=5)). *P<0.01 versus all other groups