Portal glucose infusion exerts an incretin effect associated with changes in pancreatic neural activity in conscious dogs

Abstract We sought to determine whether an incretin effect could be observed when glucose was infused via the hepatic portal (Po) vein versus a peripheral (Pe) vein. Identical hyperglycemia (155 [plusmn] 7 and 154 [plusmn] 8 mg/dL, respectively) was produced in 2 groups (n = 9 each) of conscious dogs by Po or Pe glucose infusion. During glucose infusion, arterial plasma insulin levels increased by 28 [plusmn] 5 and 16 [plusmn] 3 [mu ]U/mL in Po and Pe, respectively ( P [lt ] .05 between groups). Pancreatic insulin output increased by 10.4 [plusmn] 3.2 and 6.7 [plusmn] 2.3 mU/min in Po and Pe, respectively ( P = .12 between groups). Arterial plasma glucagon levels and pancreatic glucagon output were similarly suppressed by Po and Pe glucose infusion. Pancreatic polypeptide (PP) output and norepinephrine (NE) spillover were measured as indices of pancreatic parasympathetic and sympathetic neural activity, respectively. During Pe, pancreatic PP output decreased from basal ([Delta ] [minus ]4.8 [plusmn] 2.5 ng/min, P [lt ] .05), with no significant change in NE spillover ([Delta ] +4.4 [plusmn] 4.0 ng/min). The PP output:NE spillover ratio decreased by 65% ( P [lt ] .05), suggesting a shift toward a dominance of sympathetic tone. During Po, there were no significant changes in PP output ([Delta ] [minus ]1.4 [plusmn] 3.1 ng/min) or NE spillover ([Delta ] +1.6 [plusmn] 1.2 ng/min), and consequently there was no significant change in the PP output:NE spillover ratio. Thus, activation of the Po glucose signal appears to inhibit the shift toward sympathetic dominance that would otherwise result, thereby causing an incretin effect.