Based on our earlier work, a 2.5-fold increase in insulin secretion should completely inhibit hepatic glucose production through the hormone's direct effect on hepatic glycogen metabolism. The aim of the present study was to test the accuracy of this prediction and to confirm that gluconeogenic flux, as measured by three independent techniques, was unaffected by the increase in insulin. A 40-min basal period was followed by a 180-min experimental period in which an increase in insulin was induced, with euglycemia maintained by peripheral glucose infusion. Arterial and hepatic sinusoidal insulin levels increased from 10 ؎ 2 to 19 ؎ 3 and 20 ؎ 4 to 45 ؎ 5 U/ml, respectively. Net hepatic glucose output decreased rapidly from 1.90 ؎ 0.13 to 0.23 ؎ 0.16 mg ⅐ kg ؊1 ⅐ min ؊1 . Three methods of measuring gluconeogenesis and glycogenolysis were used: 1) the hepatic arteriovenous difference technique (n ‫؍‬ 8), 2) the [ (1) showed that hepatic glucose production (HGP) can be inhibited by selective increases in the arterial or portal vein insulin concentration. In response to a 14-U/ml increase in arterial insulin (no change in portal insulin), a Ͼ50% reduction in net hepatic glucose output (NHGO) was observed. Likewise, a 14-U/ml increase in portal insulin (no change in arterial insulin) also resulted in a Ͼ50% reduction in NHGO. In addition, the above studies showed that insulin acted directly on the liver, with a rise in hepatic sinusoidal insulin quickly inhibiting HGP by reducing net hepatic glycogenolysis. The indirect effect of insulin on HGP, on the other hand, resulted from a decrease in gluconeogenic flux rate caused by a reduction in the flow of gluconeogenic amino acids and glycerol to the liver and diversion of carbon derived from glycogenolysis to lactate rather than glucose. The reduction in HGP in this group was also, in part, the result of a decrease in net hepatic glycogenolysis, which occurred as a result of a slight rise in the hepatic sinusoidal insulin level, which, in turn, occurred as a result of the rise in hepatic artery insulin. It took 1 h to detect a significant indirect effect of insulin on HGP.Sindelar et al.(1) created selective changes in the arterial or portal insulin level by infusing somatostatin to inhibit insulin secretion and replacing insulin by infusion through a peripheral and/or portal catheter. Stimulation of pancreatic insulin secretion, on the other hand, results in an increase in both portal and arterial levels of the hormone. Therefore, in the present study, our aim was to determine if a two-to threefold increase in insulin, occurring simultaneously in portal and peripheral blood, would inhibit HGP primarily through an effect on glycogen metabolism. Although Sindelar et al. (1) reported that portally delivered insulin did not affect gluconeogenic flux, their estimate of the latter relied solely on the measurement of the net hepatic uptake (arteriovenous [AV] difference) of gluconeogenic precursors. In the present study, we combined the hepatic AV difference technique, along...

No abstract

The aim of these studies was to investigate the effect of hyperglycemia with or without hyperinsulinemia on hepatic gluconeogenic flux, with the hypothesis that inhibition would be greatest with combined hyperglycemia/hyperinsulinemia. A glycogen phosphorylase inhibitor (BAY R3401) was used to inhibit glycogen breakdown in the conscious overnight-fasted dog, and the effects of a twofold rise in plasma glucose level (HI group) accompanied by 1) euinsulinemia (HG group) or 2) a fourfold rise in plasma insulin were assessed over a 5-h experimental period. Hormone levels were controlled using somatostatin with portal insulin and glucagon infusion. In the HG group, net hepatic glucose uptake and net hepatic lactate output substantially increased. There was little or no effect on the net hepatic uptake of gluconeogenic precursors other than lactate (amino acids and glycerol) or on the net hepatic uptake of free fatty acids compared with the control group. Consequently, whereas hyperglycemia had little effect on gluconeogenic flux to glucose 6-phosphate (G-6- P), net hepatic gluconeogenic flux was reduced because of increased hepatic glycolytic flux during hyperglycemia. Net hepatic glycogen synthesis was increased by hyperglycemia. The effect of hyperglycemia on gluconeogenic flux to G-6- P and net hepatic gluconeogenic flux was similar. We conclude that, in the absence of appreciable glycogen breakdown, the increase in glycolytic flux that accompanies hyperglycemia results in decreased net carbon flux to G-6- P but no effect on gluconeogenic flux to G-6- P.

Hyperglycemia in type 2 diabetes is due, in part, to elevated gluconeogenesis (GNG). This has been attributed, based largely on in vitro models, to increased mRNA transcription of supposed rate‐limiting enzymes PEPCK and G6Pase. However, in vivo metabolic substrate analysis in man and dog indicate that GNG flux to G6P varies minimally within the physiological range of insulin. We propose that GNG mRNA and protein expression would be decreased during hyperinsulinemic (2.0 mU/kg/min), euglycemic conditions in 24 h fasted conscious dogs subjected to a pancreatic clamp, but that the time course of expression would not correlate with functional GNG flux to G6P. Arterial and hepatic sinusoidal insulin levels increased 8‐fold and glucagon was clamped at basal levels. At 0.5 h, GNG flux to G6P was decreased 50% due to the suppression of lipolysis and a switch from net hepatic lactate uptake to output. Liver FOXO1 phosphorylation was increased 1.9‐fold and GNG enzyme mRNA expression was decreased ~40‐50% at 0.5 h but PEPCK protein levels did not change. By 4 h, GNG mRNA expression was decreased ~80%, and PEPCK protein was reduced by 45%, but GNG flux to G6P had stabilized at near‐basal levels. Thus, while insulin can inhibit GNG gene transcription and protein expression in vivo, these changes do not result in functional changes in substrate flux in the GNG pathway, indicating PEPCK was not a rate‐limiting marker for GNG in vivo.