Dana K. Sindelar

The Role of Fatty Acids in Mediating the Effects of Peripheral Insulin on Hepatic Glucose Production in the Conscious Dog

We investigated the mechanism by which a selective increase in arterial insulin can suppress hepatic glucose production in vivo. Isotopic (3-3H-glucose) and arteriovenous difference methods were used in overnight-fasted, conscious dogs. A pancreatic clamp (somatostatin, basal portal insulin, and glucagon infusions) was used to control the endocrine pancreas. Equilibration (100 min) and basal (40 min) periods were followed by a 180-min test period. In control dogs (n = 5), basal insulin delivery was continued throughout the study. In the other two groups, peripheral insulin was selectively increased at the beginning of the test period by stopping the portal insulin infusion and infusing insulin peripherally at twice the basal portal rate. One group (INS + FAT; n = 6) received an infusion of 20% intralipid + heparin (0.5 U · kg−1 · min−1) to clamp the nonesterified fatty acid (NEFA) levels during hyperinsulinemia; the other group (INS; n = 7) received only saline during the experimental period. In the INS group, a selective increase in peripheral insulin of 84 pmol/l was achieved (36 ± 6 to 120 ± 24 pmol/l, last 30 min) while portal insulin was unaltered (84 ± 18 pmol/l). In the INS + FAT group, a similar increase in peripheral insulin was achieved (36 ± 6 to 114 ± 6 pmol/l, last 30 min); again, portal insulin was unaltered (96 ± 12 pmol/l). In the control group, basal insulin did not change. Glucagon and glucose remained near basal values in all protocols. In the INS group, NEFA levels dropped from 700 ± 90 (basal) to 230 ± 65 μmol/l (last 30 min; P > 0.05), but in the INS + FAT group changed minimally (723 ± 115 [basal] to 782 ± 125 μmol/l [last 30 min]). In the INS group, net hepatic glucose output dropped by 6.7 μmol · kg−1 · min−1 (P < 0.05), whereas in the INS + FAT group it dropped by 3.9 μmol · kg−1 · min−1 (P < 0.05). When insulin levels were not increased (i.e., in the control group), net hepatic glucose output dropped 1.7 μmol · kg−1 · min−1 (P , 0.05). In all groups, the net hepatic glucose output data were confirmed by the tracer-determined glucose production data. In the INS group, net hepatic gluconeogenic substrate uptake (alanine, glutamine, glutamate, glycerol, glycine, lactate, threonine, and serine) fell slightly (10.4 ± 1.3 [basal] to 7.2 ± 1.3 μmol · kg−1 · min−1 [last 30 min]), whereas in the INS + FAT group it did not change (7.3 ±1.5 [basal] to 7.4 ± 0.6 μmol · kg−1 · min−1 [last 30 min]), and in the control group it increased slightly (9.6 ± 1.3 [basal] to 10.3 ± 1.4 μmol · kg−1 · min−1 [last 30 min]). These results indicate that peripheral insulins ability to regulate hepatic glucose production is partially linked to its inhibition of lipolysis. When plasma NEFA levels were prevented from falling during a selective arterial hyperinsulinemia, ∼55% of insulin%s inhibition of net hepatic glucose output (NHGO) was eliminated. The fall in NEFA levels brings about a redirection of glycogenolytically derived carbon within the hepatocyte such that there is an increase in lactate efflux and a corresponding decrease in NHGO.

A Comparison of the Effects of Selective Increases in Peripheral or Portal insulin on Hepatic Glucose Production in the Conscious Dog

We investigated the mechanisms by which peripheral or portal insulin can independently alter liver glucose production. Isotopic ([3-3H]glucose) and arteriovenous difference methods were used in conscious overnight-fasted dogs. A pancreatic clamp (somatostatin plus basal insulin and basal glucagon infusions) was used to control the endocrine pancreas. After a 40-min basal period, a 180-min experimental period followed in which selective increases in peripheral (PERI group, n = 5) or portal-vein (PORT group, n = 5) insulin were induced. In control dogs (CONT group, n = 10), insulin was not increased. Glucagon levels were fixed in all studies, and basal euglycemia was maintained by peripheral glucose infusion in the two experimental groups. In the PERI group, arterial insulin rose from 36 ± 12 to 120 ± 12 pmol/l, while portal insulin was unaltered. In the PORT group, portal insulin rose from 108 ± 42 to 192 ± 42 pmol/l, while arterial insulin was unaltered. Neither arterial nor portal insulin changed from basal in the CONT group. With a selective rise in peripheral insulin, the net hepatic glucose output (NHGO; basal, 11.8 ± 0.7 µmol · kg−1 · min−1) did not change initially (11.8 ± 2.1 µmol · kg−1 · min−1, 30 min after the insulin increase), but eventually fell (P < 0.05) to 6.1 ± 0.9 µmol · kg−1 · min−1 (last 30 min). With a selective rise in portal insulin, NHGO dropped quickly (P < 0.05) from 10.0 ± 0.9 to 5.6 ± 0.6 µmol · kg−1 · min−1 (30 min after the insulin increase) and eventually reached 3.1 ± 1.1 µmol x kg−1 · min−1 (last 30 min). When insulin levels were not increased (CONT group), NHGO dropped progressively from 10.1 ± 0.6 to 8.3 ± 0.6 µmol · kg−1 · min−1 (last 30 min). Conclusions drawn from the net hepatic glucose balance data were confirmed by the tracer data. Net hepatic gluconeogenic substrate uptake (three carbon precursors) fell 2.0 µmol · kg−1 · min−1 in the PERI group, but rose 1.2 µmol · kg−1 · min−1 in the PORT group and 1.2 µmol · kg−1 · min−1 in the CONT group. A selective 84 pmol/l rise in arterial insulin was thus associated with a fall in NHGO of ∼ 50%, which took 1 h to manifest. Conversely, a selective 84 pmol/l rise in portal insulin was associated with a 50% fall in NHGO, which occurred quickly (15 min). From the control data, it is evident that in either case ∼ 30% of the fall in NHGO was due to a drift down in baseline and that 70% was due to the rise in insulin. In conclusion, an increment in portal insulin had a rapid inhibitory effect on NHGO, caused by the suppression of glycogenolysis, while an equal increment in arterial insulin produced an equally potent but slower effect that resulted from a small increase in hepatic sinusoidal insulin, from a suppression of gluconeogenic precursor uptake by the liver, and from a redirection of glycogenolytic carbon to lactate rather than glucose.

The direct and indirect effects of insulin on hepatic glucose production in vivo

IntroductionGlucose production by the liver is controlled on aminute to minute basis by the plasma insulin and glu-cagon concentrations. Glucagon increases glycogenbreakdown and stimulates the gluconeogenic path-way and insulin inhibits both glycogenolysis and glu-coneogenesis [1]. Until ten years ago it was generallyaccepted that the insulin level within the hepatic sinu-soids was responsible for the hormone’s inhibitory ef-fect on the liver. Recently, however, that concept hasbeen challenged.In 1987 this concept was first called into question[2]. It was noted that in obese non-diabetic humanssuppression of glucose production could occur in re-sponse to insulin infusion even when the estimatedportal insulin concentrations did not rise. Insulin wasinfused into a peripheral vein and euglycaemia wasmaintained with a glucose infusion. Endogenous in-sulin secretion, estimated from C-peptide concentra-tions, decreased by about 50% over the course ofthe experiment. Based on the fall in endogenous insu-lin release and the prevailing peripheral insulin con-centration, the authors estimated that the portal veininsulin concentration probably remained unchanged.Nevertheless glucose production was suppressed byabout 80% at the end of their study. It should be re-membered, however, that about 20% of liver bloodflow is derived from the hepatic artery, so a rise in ar-terial insulin would have increased the liver sinusoi-dal insulin concentration somewhat, even though theportal insulin level did not change. Likewise, al-though the authors assumed that their tracer datayielded an estimate of hepatic glucose production, itis now clear that it reflects both hepatic and renal glu-cose output [3]. After an overnight fast renal glucoseproduction is not great (5–25% of total glucose pro-duction) but it is insulin sensitive [4]. It is likely there-fore that a small portion of the insulin induced fall inglucose production which they observed was due toa decrease in renal glucose production. Despite theabove caveats, the study [2] was of importance be-cause it was the first to focus attention on the indirecteffect of insulin to inhibit glucose production by theliver.To further investigate the effect of peripherally de-livered insulin on glucose production, experimentswere carried out [5] in which conscious dogs were giv-en insulin intraportally (at 0.3 to 10 pmol × kg

Comparison of the direct and indirect effects of epinephrine on hepatic glucose production.

To determine the extent to which the effect of a physiologic increment in epinephrine (EPI) on glucose production (GP) arises indirectly from its action on peripheral tissues (muscle and adipose tissue), epinephrine was infused intraportally (EPI po) or peripherally (EPI pe) into 18-h-fasted conscious dogs maintained on a pancreatic clamp. Arterial EPI levels in EPI po and EPI pe groups rose from 97 +/- 29 to 107 +/- 37 and 42 +/- 12 to 1,064 +/- 144 pg/ml, respectively. Hepatic sinusoidal EPI levels in EPI po and EPI pe were indistinguishable (561 +/- 84 and 568 +/- 75 pg/ml, respectively). During peripheral epinephrine infusion, GP increased from 2.2 +/- 0.1 to 5.1 +/- 0.2 mg/kg x min (10 min). In the presence of the same rise in sinusoidal EPI, but with no rise in arterial EPI (during portal EPI infusion), GP increased from 2.1 +/- 0.1 to 3.8 +/- 0.6 mg/kg x min. Peripheral EPI infusion increased the maximal gluconeogenic rate from 0.7 +/- 0.4 to 1.8 +/- 0.5 mg/ kg x min. Portal EPI infusion did not change the maximal gluconeogenic rate. The estimated initial increase in glycogenolysis was approximately 1.7 and 2.3 mg/kg x min in the EPI pe and EPI po groups, respectively. Gluconeogenesis was responsible for 60% of the overall increase in glucose production stimulated by the increase in plasma epinephrine (EPI pe). Elevation of sinusoidal EPI per se had no direct gluconeogenic effect on the liver, thus its effect on glucose production was solely attributable to an increase in glycogenolysis. Lastly, the gluconeogenic effects of EPI markedly decreased (60-80%) its overall glycogenolytic action on the liver.

Effect of a selective rise in sinusoidal norepinephrine on HGP is due to an increase in glycogenolysis

To determine the effect of a selective rise in liver sinusoidal norepinephrine (NE) on hepatic glucose production (HGP), norepinephrine (50 ng.kg-1.min-1) was infused intraportally (Po-NE) for 3 h into five 18-h-fasted conscious dogs with a pancreatic clamp. In the control protocol, NE (0.2 ng.kg-1.min-1) and glucose were infused peripherally to match the arterial NE and blood glucose levels in the Po-NE group. Hepatic sinusoidal NE levels rose approximately 30-fold in the Po-NE group but did not change in the control group. The arterial NE levels did not change significantly in either group. During the portal NE infusion, HGP increased from 1.9 +/- 0.2 to 3.5 +/- 0.4 mg.kg-1.min-1 (15 min; P < 0.05) and then gradually fell to 2.4 +/- 0.4 mg.kg-1.min-1 by 3 h. HGP in the control group did not change (2.0 +/- 0.2 to 2.0 +/- 0.2 mg.kg-1.min-1) for 15 min but then gradually fell to 1.1 +/- 0.2 mg.kg-1.min-1 by the end of the study. Because the fall in HGP from 15 min on was parallel in the two groups, the effect of NE on HGP (the difference between HGP in the two groups) did not decline over time. Gluconeogenesis did not change significantly in either group. In conclusion, elevation in hepatic sinusoidal NE significantly increases HGP by selectively stimulating glycogenolysis. Compared with the previously determined effects of epinephrine or glucagon on HGP, the effect of NE is, on a molar basis, less potent but more sustained over time.To determine the effect of a selective rise in liver sinusoidal norepinephrine (NE) on hepatic glucose production (HGP), norepinephrine (50 ng ⋅ kg-1 ⋅ min-1) was infused intraportally (Po-NE) for 3 h into five 18-h-fasted conscious dogs with a pancreatic clamp. In the control protocol, NE (0.2 ng ⋅ kg-1 ⋅ min-1) and glucose were infused peripherally to match the arterial NE and blood glucose levels in the Po-NE group. Hepatic sinusoidal NE levels rose ∼30-fold in the Po-NE group but did not change in the control group. The arterial NE levels did not change significantly in either group. During the portal NE infusion, HGP increased from 1.9 ± 0.2 to 3.5 ± 0.4 mg ⋅ kg-1 ⋅ min-1(15 min; P < 0.05) and then gradually fell to 2.4 ± 0.4 mg ⋅ kg-1 ⋅ min-1by 3 h. HGP in the control group did not change (2.0 ± 0.2 to 2.0 ± 0.2 mg ⋅ kg-1 ⋅ min-1) for 15 min but then gradually fell to 1.1 ± 0.2 mg ⋅ kg-1 ⋅ min-1by the end of the study. Because the fall in HGP from 15 min on was parallel in the two groups, the effect of NE on HGP (the difference between HGP in the two groups) did not decline over time. Gluconeogenesis did not change significantly in either group. In conclusion, elevation in hepatic sinusoidal NE significantly increases HGP by selectively stimulating glycogenolysis. Compared with the previously determined effects of epinephrine or glucagon on HGP, the effect of NE is, on a molar basis, less potent but nore sustained over time.

Insulin action on the liver in vivo

Insulin has a potent inhibitory effect on hepatic glucose production by direct action at hepatic receptors. The hormone also inhibits glucose production by suppressing both lipolysis in the fat cell and secretion of glucagon by the alpha-cell. Neural sensing of insulin levels appears to participate in control of hepatic glucose production in rodents, but a role for brain insulin sensing has not been documented in dogs or humans. The primary effect of insulin on the liver is its direct action.

Interaction of equal increments in arterial and portal vein insulin on hepatic glucose production in the dog

We have previously shown that a selective increase of 84 pmol/l in either arterial or portal vein insulin (independent of a change in insulin in the other vessel) can suppress tracer-determined glucose production (TDGP) and net hepatic glucose output (NHGO) by approximately 50%. In the present study we investigated the interaction between equal increments in arterial and portal vein insulin in the suppression of TDGP and NHGO. Isotopic ([3-3H]glucose) and arteriovenous difference methods were used in conscious overnight fasted dogs. A pancreatic clamp was used to control the endocrine pancreas. A 40-min basal period was followed by a 180-min test period, during which arterial and portal vein insulin levels were simulataneously and equally increased 102 pmol/l. Hepatic sinusoidal glucagon levels remained unchanged, and euglycemia was maintained by peripheral glucose infusion. TDGP was suppressed approximately 60% by the last 30 min of the experimental period. In contrast, NHGO was suppressed 100% by that time. Coincidentally, hepatic glucose uptake (net hepatic [3H]glucose balance) increased significantly (approximately 4 mumol.kg-1.min-1). The effects of simultaneous equal increases in peripheral and portal venous insulin were not additive in the suppression of TDGP. However, they were additive in decreasing NHGO as a result of an increase in the uptake of glucose by the liver.We have previously shown that a selective increase of 84 pmol/l in either arterial or portal vein insulin (independent of a change in insulin in the other vessel) can suppress tracer-determined glucose production (TDGP) and net hepatic glucose output (NHGO) by ∼50%. In the present study we investigated the interaction between equal increments in arterial and portal vein insulin in the suppression of TDGP and NHGO. Isotopic ([3-3H]glucose) and arteriovenous difference methods were used in conscious overnight fasted dogs. A pancreatic clamp was used to control the endocrine pancreas. A 40-min basal period was followed by a 180-min test period, during which arterial and portal vein insulin levels were simultaneously and equally increased 102 pmol/l. Hepatic sinusoidal glucagon levels remained unchanged, and euglycemia was maintained by peripheral glucose infusion. TDGP was suppressed ∼60% by the last 30 min of the experimental period. In contrast, NHGO was suppressed 100% by that time. Coincidentally, hepatic glucose uptake (net hepatic [3H]glucose balance) increased significantly (∼4 μmol ⋅ kg-1 ⋅ min-1). The effects of simultaneous equal increases in peripheral and portal venous insulin were not additive in the suppression of TDGP. However, they were additive in decreasing NHGO as a result of an increase in the uptake of glucose by the liver.

Hepatic and Gut Clearance of Catecholamines in the Conscious Dog

Our aim was to assess hepatic and gut catecholamine clearance under normal and simulated stress conditions. Following a 90-minute saline infusion period, epinephrine ([EPI] 180 ng/kg x min) and norepinephrine ([NE] 500 ng/kg x min) were infused peripherally for 90 minutes into five 18-hour fasted, conscious dogs undergoing a pancreatic clamp (somatostatin plus basal insulin and glucagon). Arterial plasma levels of EPI and NE increased from 44 +/- 9 to 2,961 +/- 445 and 96 +/- 6 to 6,467 +/- 571 pg/mL, respectively (both P < .05). Portal vein plasma levels of EPI and NE increased from 23 +/- 8 to 1,311 +/- 173 and 79 +/- 10 to 3,477 +/- 380 pg/mL, respectively (both P < .05). Hepatic vein plasma levels of EPI and NE increased from 5 +/- 2 to 117 +/- 33 and 48 +/- 10 to 448 +/- 59 pg/mL, respectively (both P < .05). Net hepatic and gut EPI uptake increased from 0.5 +/- 0.1 to 30.0 +/- 3.0 and 0.4 +/- 0.1 to 26.3 +/- 4.0 ng/kg x min, respectively (both P < .05). Net hepatic and gut NE uptake increased from 1.5 +/- 0.4 to 74.7 +/- 8.4 and 0.8 +/- 0.2 to 57.9 +/- 7.6 ng/kg x min, respectively (both P < .05). Neither the net hepatic (0.86 +/- 0.05 to 0.93 +/- 0.02) nor gut (0.45 +/- 0.10 to 0.55 +/- 0.04) fractional extraction of EPI changed significantly during the simulated stress condition. Net hepatic and gut spillover of NE increased from 0.8 +/- 0.2 to 3.5 +/- 1.3 and 0.6 +/- 0.2 to 8.8 +/- 2.0 ng/kg x min, respectively, during catecholamine infusion (both P < .05). These results indicate that (1) approximately 30% of circulating catecholamines are cleared by the splanchnic bed (16% and 14% by the liver and gut, respectively); (2) the liver and gut remove a large proportion (approximately 86% to 93% and 45% to 55%, respectively) of the catecholamines delivered to them on first pass; and (3) high levels of plasma catecholamines increase NE spillover from both the liver and gut, suggesting that the percentage of NE released from the presynaptic neuron that escapes the synaptic cleft is increased in the presence of high circulating catecholamine levels.

Portal adrenergic blockade does not inhibit the gluconeogenic effects of circulating catecholamines on the liver

This study was undertaken to determine the impact of portal adrenergic blockade on the gluconeogenic effects of epinephrine (EPI) and norepinephrine (NE). Experiments were performed on 18-hour fasted conscious dogs and consisted of a 100-minute equilibration, a 40-minute basal, and two 90-minute test periods. A pancreatic clamp was used to fix insulin and glucagon levels at basal values. Propranolol (1 microgram/kg.min) and phentolamine (2 micrograms/kg.min) were infused intraportally during both test periods. Portal infusion of alpha- and beta-adrenergic blockers alone (first test period) slightly increased hepatic glucose production from 2.4 +/- 0.4 to 2.8 +/- 0.5 mg/kg.min (nonsignificant [NS]) NE (500 ng/kg.min) and EPI (180 ng/kg.min) were infused peripherally during the second test period. Arterial NE and EPI increased from 186 +/- 63 to 6,725 +/- 913 pg/mL and 76 +/- 25 to 2,674 +/- 344 pg/mL, respectively. Portal NE and EPI increased from 135 +/- 32 to 4,082 +/- 747 pg/mL and 28 +/- 8 to 1,114 +/- 174 pg/mL, respectively. Hepatic glucose production, the maximal gluconeogenic rate, and gluconeogenic efficiency increased from 2.8 +/- 0.5 to 3.8 +/- 0.4 mg/kg.min (P < .05), 0.7 +/- 0.3 to 2.1 +/- 0.6 mg/kg.min (P < .05), and 21% +/- 8% to 60% +/- 13% (P < .05), respectively, in response to catecholamine infusion. Net hepatic lactate balance changed from output (1.5 +/- 3.3 mumol/kg.min) to uptake (-11.0 +/- 3.8 mumol/kg.min, P < .05). Net hepatic glycerol uptake increased from -1.5 +/- 0.7 to -5.5 +/- 2.0 mumol/kg.min (P < .05). Net hepatic uptake of gluconeogenic amino acids did not change significantly. Similarly, hepatic glycogenolysis did not increase during catecholamine infusion. In conclusion, portal delivery of adrenergic blockers selectively inhibits the glycogenolytic effects of EPI and NE on the liver, but allows a marked gluconeogenic response to the catecholamines.

Effect of a selective rise in hepatic artery insulin on hepatic glucose production in the conscious dog

In the present study we compared the hepatic effects of a selective increase in hepatic sinusoidal insulin brought about by insulin infusion into the hepatic artery with those resulting from insulin infusion into the portal vein. A pancreatic clamp was used to control the endocrine pancreas in conscious overnight-fasted dogs. In the control period, insulin was infused via peripheral vein and the portal vein. After the 40-min basal period, there was a 180-min test period during which the peripheral insulin infusion was stopped and an additional 1.2 pmol. kg-1. min-1 of insulin was infused into the hepatic artery (HART, n = 5) or the portal vein (PORT, n = 5, data published previously). In the HART group, the calculated hepatic sinusoidal insulin level increased from 99 +/- 20 (basal) to 165 +/- 21 pmol/l (last 30 min). The calculated hepatic artery insulin concentration rose from 50 +/- 8 (basal) to 289 +/- 19 pmol/l (last 30 min). However, the overall arterial (50 +/- 8 pmol/l) and portal vein insulin levels (118 +/- 24 pmol/l) did not change over the course of the experiment. In the PORT group, the calculated hepatic sinusoidal insulin level increased from 94 +/- 30 (basal) to 156 +/- 33 pmol/l (last 30 min). The portal insulin rose from 108 +/- 42 (basal) to 192 +/- 42 pmol/l (last 30 min), whereas the overall arterial insulin (54 +/- 6 pmol/l) was unaltered during the study. In both groups hepatic sinusoidal glucagon levels remained unchanged, and euglycemia was maintained by peripheral glucose infusion. In the HART group, net hepatic glucose output (NHGO) was suppressed from 9.6 +/- 2.1 micromol. kg-1. min-1 (basal) to 4.6 +/- 1.0 micromol. kg-1. min-1 (15 min) and eventually fell to 3.5 +/- 0.8 micromol. kg-1. min-1 (last 30 min, P < 0.05). In the PORT group, NHGO dropped quickly (P < 0.05) from 10.0 +/- 0.9 (basal) to 7.8 +/- 1.6 (15 min) and eventually reached 3.1 +/- 1.1 micromol. kg-1. min-1 (last 30 min). Thus NHGO decreases in response to a selective increase in hepatic sinusoidal insulin, regardless of whether it comes about because of hyperinsulinemia in the hepatic artery or portal vein.In the present study we compared the hepatic effects of a selective increase in hepatic sinusoidal insulin brought about by insulin infusion into the hepatic artery with those resulting from insulin infusion into the portal vein. A pancreatic clamp was used to control the endocrine pancreas in conscious overnight-fasted dogs. In the control period, insulin was infused via peripheral vein and the portal vein. After the 40-min basal period, there was a 180-min test period during which the peripheral insulin infusion was stopped and an additional 1.2 pmol ⋅ kg-1 ⋅ min-1of insulin was infused into the hepatic artery (HART, n = 5) or the portal vein (PORT, n = 5, data published previously). In the HART group, the calculated hepatic sinusoidal insulin level increased from 99 ± 20 (basal) to 165 ± 21 pmol/l (last 30 min). The calculated hepatic artery insulin concentration rose from 50 ± 8 (basal) to 289 ± 19 pmol/l (last 30 min). However, the overall arterial (50 ± 8 pmol/l) and portal vein insulin levels (118 ± 24 pmol/l) did not change over the course of the experiment. In the PORT group, the calculated hepatic sinusoidal insulin level increased from 94 ± 30 (basal) to 156 ± 33 pmol/l (last 30 min). The portal insulin rose from 108 ± 42 (basal) to 192 ± 42 pmol/l (last 30 min), whereas the overall arterial insulin (54 ± 6 pmol/l) was unaltered during the study. In both groups hepatic sinusoidal glucagon levels remained unchanged, and euglycemia was maintained by peripheral glucose infusion. In the HART group, net hepatic glucose output (NHGO) was suppressed from 9.6 ± 2.1 μmol ⋅ kg-1 ⋅ min-1 (basal) to 4.6 ± 1.0 μmol ⋅ kg-1 ⋅ min-1(15 min) and eventually fell to 3.5 ± 0.8 μmol ⋅ kg-1 ⋅ min-1(last 30 min, P < 0.05). In the PORT group, NHGO dropped quickly ( P < 0.05) from 10.0 ± 0.9 (basal) to 7.8 ± 1.6 (15 min) and eventually reached 3.1 ± 1.1 μmol ⋅ kg-1 ⋅ min-1(last 30 min). Thus NHGO decreases in response to a selective increase in hepatic sinusoidal insulin, regardless of whether it comes about because of hyperinsulinemia in the hepatic artery or portal vein.