Insulin Dose-Response Curves for Stimulation of Splanchnic Glucose Uptake and Suppression of Endogenous Glucose Production Differ in Nondiabetic Humans and Are Abnormal in People With Type 2 Diabetes

After approval from the Mayo Institutional Review Board, 14 nondiabetic subjects and 12 subjects with type 2 diabetes gave informed written consent to participate in the study. All subjects were in good health and at a stable weight. None regularly engaged in vigorous physical exercise. None of the parents, siblings, and children (first-degree relatives) of the nondiabetic subjects had a history of diabetes. At the time of screening, two of the diabetic subjects were being treated with lifestyle modifications alone and seven with either a sulfonylurea or metformin. One subject was on insulin alone, and two were on both insulin and metformin. Oral hypoglycemic agents were discontinued 3 weeks prior to study and long-acting insulin switched to regular insulin with meals, beginning with the midday meal 2 days prior to study. Subjects were on no medications at the time of the study other than either thyroxine or hormone replacement therapy. All subjects were instructed to follow a weight maintenance diet containing 55% carbohydrate, 30% fat, and 15% protein for at least 3 days prior to the day of study. Diabetic and nondiabetic subjects, respectively, were matched for age (65 ± 1 vs. 63 ± 2 years), sex (6/6 vs. 7/7 men/women), total body weight (89 ± 4.0 vs. 86 ± 3.3 kg), BMI (32.1 ± 1.2 vs. 30.1 ± 1.3 kg/m²), and body fat (39.1 ± 3.0 vs. 34.2 ± 2.5%). Fasting plasma glucose (8.0 ± 0.3 vs. 5.3 ± 0.1 mmol/l) and HbA_1c (9.1 ± 1.0 vs. 5.2 ± 1.0%), measured at the time of screening, were significantly (P < 0.001) higher in diabetic versus nondiabetic subjects.

Subjects were admitted to the Mayo Clinic General Clinical Research Center at 1700 on the evening before the study. A standard 10-cal/kg meal (55% carbohydrate, 30% fat, and 15% protein) was eaten between 1730 and 1800. After the meal, an 18-gauge catheter was inserted into a forearm vein and an infusion of insulin (100 units regular human insulin in 1 l of 0.9% saline containing 5 ml of 25% human albumin) was started in the diabetic subjects or saline in the nondiabetic subjects. The insulin infusion rate was adjusted to maintain glucose concentrations in the diabetic subjects at ∼5 mmol/l during the night (30). In addition, 0.9% saline along with 20 mEq of KCl per l was infused at 60 ml/min throughout the night in the diabetic subjects and during the insulin clamp in both the nondiabetic and diabetic subjects.

At 0600 on the morning after admission, a urinary bladder catheter was placed, after which the subjects were taken to an interventional radiology suite where femoral artery, femoral venous, and hepatic venous catheters were placed as previously described (6,7). The arterial catheter was used for blood sampling, and the arterial sheath was used for indocyanine green infusion (Akorn, Buffalo Grove, IL). The venous catheters were used for blood sampling. Although, the experimental design is similar to that previously reported by us (6,7), none of the subjects in the present experiments participated in those studies.

Subjects were then returned to the General Clinical Research Center, where at ∼0900, a primed-continuous infusion of [3-³H] glucose (12 μCi primed, 0.12 μCi/min continuous; New England Nuclear, Boston, MA) into a forearm vein was started. Infusions of somatostatin (60 ng · kg⁻¹ · min⁻¹), growth hormone (3 ng · kg⁻¹ · min⁻¹), and glucagon (0.65 ng · kg⁻¹ · min⁻¹) were also started (t = 0 min) and continued until the end of the study. Insulin was infused at a rate of 0.78 mU · kg lean body wt⁻¹ · min⁻¹ (∼0.5 mU · kg total body wt⁻¹ · min⁻¹) from 0 to 180 min, 1.56 (∼1.0) from 181 to 300 min, and 3.1 (∼2.0) from 301 to 420 min. A dextrose infusion also was begun and the rate adjusted so as to maintain plasma glucose concentrations at ∼9.3 mmol/l (∼165 mg/dl) over the next 7 h of study. All infused glucose contained [3-³H] glucose to minimize the change in plasma glucose specific activity (31,32). In addition, the rate of the “basal” [3-³H] glucose infusion also was reduced to mimic the anticipated changes in EGP (33).

All samples were placed in ice, centrifuged at 4°C, and separated. Plasma indocyanine green concentration was measured spectrophotometrically at 805 nm on the day of study as previously described (34). All other samples were stored at −20°C until analysis. Plasma glucose was measured by a glucose oxidase method using a YSI glucose analyzer (Yellow Springs, OH). Plasma insulin was measured using a chemiluminescence method with the Access Ultrasensitive Immunoenzymatic assay system (Beckman, Chaska, MN). C-peptide and glucagon concentrations were assayed by radioimmunoassay (RIA; Linco Research, St. Louis, MO). Growth hormone was measured with the Access hGH two-site immunoenzymatic assay (Beckman). Body composition, including percentage of fat and lean body mass, were measured using dual-energy X-ray absorptiometry (DPX-IQ scanner, SmartScan version 4.6; Hologic, Waltham, MA).

Splanchnic plasma flow was calculated by dividing the indocyanine green infusion rate by the arterial-hepatic venous concentration gradient of the dye, and leg plasma flow was calculated by dividing the dye infusion rate by the concentration gradient across the leg (34,35). The corresponding blood flows were calculated by dividing the respective plasma flow by (1 − hematocrit). Blood glucose concentrations were calculated by multiplying the plasma glucose concentrations by 0.85. Net splanchnic glucose balance was calculated as the product of the arterial-hepatic vein glucose difference and the median of quadruple determinations of splanchnic blood flow. The splanchnic-to-glucose extraction ratio was calculated as the difference of arterial and hepatic venous [3-³H] glucose concentrations divided by the arterial [3-³H] glucose concentration. SGU was determined by multiplying the arterial glucose concentration by splanchnic-to-glucose extraction ratio and splanchnic blood flow. Leg glucose uptake (LGU) was calculated as the difference between the femoral arterial and the femoral venous glucose concentration times the median of quadruple determinations of leg blood flow. The leg-to-glucose extraction ratio was calculated as the difference in arterial and femoral venous [3-³H] glucose concentration divided by the arterial [3-³H] glucose concentration. Rates of glucose appearance (R_a) and disappearance (R_d) were calculated using the steady-state equations of Steele et al. (36). EGP was determined by subtracting the glucose infusion rate from the tracer-determined rate of glucose appearance.

Data in the text and figures are expressed as mean ± SE. All rates (including SGU and LGU) are expressed as micromoles per kilogram of lean body mass per minute. Responses during the three dose insulin infusions were determined by meaning the results present, respectively, from 150 to 180 min, 270 to 300 min, and 390 to 420 min. Student’s unpaired t test was used to compare results between groups (e.g. diabetic versus nondiabetic subjects), and ANOVA was used to compare results within a group (e.g. low- versus mid- versus high-dose insulin infusion), followed by Student’s paired t test wherever appropriate. P < 0.05 was considered statistically significant.

Plasma glucose concentrations were higher (P < 0.001) in the diabetic than in the nondiabetic subjects (Fig. 1A) prior to the clamp (8.0 ± 0.3 vs. 5.5 ± 0.1 mmol), but did not differ during the clamp (9.3 ± 0.1 vs. 9.3 ± 0.1 mmol/l). Plasma insulin concentrations were slightly but not significantly higher in the diabetic than in the nondiabetic subjects prior to the clamp (48 ± 6 vs. 39 ± 5 pmol/l) and did not differ during either the low (161 ± 8 vs. 158 ± 10 pmol/l)-, middle (358 ± 31 vs. 326 ± 20)-, or high (727 ± 37 vs. 711 ± 46)-dose insulin infusions (Fig. 1B). Plasma C-peptide concentrations were slightly but not significantly lower in the diabetic than in the nondiabetic subjects prior to the clamp (0.42 ± 0.05 vs. 0.52 ± 0.04) and were comparably suppressed to almost undetectable levels in both groups (Fig. 1C). Plasma glucagon concentrations (Fig. 1D) did not differ in the diabetic subjects compared with the nondiabetic subjects either prior to (137 ± 9 vs. 128 ± 5 pg/ml) or during the clamp (118 ± 5 vs. 117 ± 3). Plasma growth hormone concentrations also did not differ in the diabetic and nondiabetic groups either prior to or during the clamps (data not shown).

The glucose infusion rate required to maintain plasma glucose at target concentrations was lower (P < 0.002) in the diabetic than nondiabetic subjects during the low (9 ± 2 vs. 31 ± 5 μmol · kg⁻¹ · min⁻¹)-, middle (25 ± 4 vs. 77 ± 7)-, and high (66 ± 11 vs. 110 ± 7)-dose insulin infusions, documenting the presence of insulin resistance (Fig. 2A). All exogenous glucose contained [3-³H] glucose in an effort to minimize change in glucose specific activity during the clamp. This caused plasma glucose specific activity to reach a plateau within 60 min, enabling accurate measurement of glucose production and disappearance thereafter (Fig. 2B).

Splanchnic blood flow did not differ in the diabetic and nondiabetic subjects during the low (1,157 ± 79 vs. 1,524 ± 288 ml/min)-, middle (1,205 ± 79 vs. 1,553 ± 271)-, or high (1,235 ± 65 vs. 1,687 ± 337)-dose insulin infusions. Despite hyperglycemia, net splanchnic glucose balance was negative (net release) in the diabetic subjects during the lowest-dose insulin infusion, becoming positive (net uptake) during the middle-dose insulin infusion (Fig. 3). On the other hand, net splanchnic glucose balance was positive in the nondiabetic subjects at all three insulin infusion rates. Net splanchnic glucose balance increased (P < 0.001, ANOVA) with increasing insulin concentrations in both groups. However, net splanchnic glucose balance remained lower (P < 0.05) in the diabetic than in the nondiabetic subjects during the low (−3.5 ± 1.4 vs. 2.6 ± 1.5 μmol · kg⁻¹ · min⁻¹)-, middle (2.6 ± 1.0 vs. 7.8 ± 1.2)-, and high (7.1 ± 1.2 vs. 11.6 ± 1.5)-dose insulin infusions.

Splanchnic (tracer) glucose extraction increased (P < 0.001, ANOVA) with increasing insulin concentrations in both groups. However, splanchnic glucose extraction remained lower (P < 0.02) in the diabetic than in the nondiabetic subjects during the low (2.0 ± 0.5 vs. 4.1 ± 0.5%)-, middle (3.2 ± 0.4 vs. 5.2 ± 0.4)-, and high (4.3 ± 0.5 vs. 7.2 ± 0.5)-dose insulin infusions.

The pattern of change in SGU paralleled that of splanchnic glucose extraction. SGU (Fig. 4A) increased (P < 0.001, ANOVA) with increasing insulin concentrations in both the diabetic and nondiabetic subjects. On the other hand, SGU was lower (P < 0.01) in the diabetic than nondiabetic subjects during the low (3.3 ± 0.9 vs. 8.5 ± 1.3 μmol · kg⁻¹ · min⁻¹)-, middle (5.6 ± 0.9 vs. 11.2 ± 0.8)-, and high (9.2 ± 0.9 vs. 14.9 ± 1.2)-dose insulin infusions. Of note, the increment in SGU observed when the insulin was increased from the low- to middle-dose (2.3 ± 1.2 vs. 2.6 ± 1.2 μmol · kg⁻¹ · min⁻¹) insulin infusions, and the middle- to high-dose (3.6 ± 1.1 vs. 4.2 ± 1.1) insulin infusions did not differ in the diabetic and nondiabetic subjects.

Leg blood flow did not differ in the diabetic and nondiabetic subjects during the low (514 ± 40 vs. 483 ± 53 ml/min)-, middle (457 ± 30 vs. 511 ± 45)-, or high (551 ± 49 vs. 630 ± 60)-dose insulin infusions. Leg (tracer) glucose extraction increased (P < 0.001, ANOVA) with increasing insulin concentrations in both groups. However, leg glucose extraction remained lower (P < 0.001) in the diabetic than in the nondiabetic subjects during the low (1.7 ± 0.0 vs. 7.1 ± 2.0%)-, middle (6.0 ± 1.5 vs. 17.1± 2.4)-, and high (15.0 ± 2.0 vs. 20.6 ± 2.0)-dose insulin infusions.

The pattern of change in LGU also paralleled that of leg glucose extraction. LGU (Fig. 4B) increased (P < 0.001, ANOVA) with increasing insulin concentrations in both the diabetic and nondiabetic subjects. On the other hand, LGU was lower (P < 0.001) in the diabetic than in the nondiabetic subjects during the low (1.2 ± 0.3 vs. 4.4 ± 1.0 μmol · kg⁻¹ · min⁻¹)-, middle (3.3 ± 0.7 vs. 14.1 ± 2.0)-, and high (11.2 ± 2.5 vs. 21.4 ± 2.0)-dose infusions. Although the increment in LGU observed when insulin was increased from the low to middle infusion rate (2.1 ± 0.6 vs. 9.7 ± 1.5 μmol · kg⁻¹ · min⁻¹) was lower (P < 0.001) in the diabetic than in the nondiabetic subjects, it did not differ when insulin was increased from the middle to high infusion rate (7.9 ± 1.9 vs. 7.3 ± 0.7).

Total glucose disappearance increased (P < 0.001, ANOVA) with increasing insulin concentrations in both the diabetic and nondiabetic subjects (Fig. 5A). Glucose disappearance was lower (P < 0.01) in the diabetic than nondiabetic subjects during the low (19 ± 2 vs. 37 ± 4 μmol · kg⁻¹ · min⁻¹)-, middle (31 ± 4 vs. 80 ± 7)-, and high (71 ± 11 vs. 114 ± 8)-dose insulin infusions. As with LGU, the increment in total body glucose disappearance observed when the insulin was increased from the low to middle infusion rate was lower (P < 0.001) in the diabetic than nondiabetic subjects (11.9 ± 3.1 vs. 43.1 ± 5.1), but did not differ when insulin was increased from the middle to high infusion rate (40.3 ± 8.0 vs. 33.3 ± 4.2).

EGP in the diabetic subjects progressively decreased (P < 0.001, ANOVA) with increasing insulin concentrations (Fig. 5B). On the other hand, EGP in the nondiabetic subjects decreased (P < 0.02) when insulin was increased from the low to middle insulin infusion rates but did not decrease thereafter. This resulted in EGP rates that were higher (P < 0.01) in the diabetic than nondiabetic subjects during the low but not the middle or high insulin infusion rates.

In order to determine whether SGU increased with increasing duration of hyperinsulinemia, insulin was infused at a rate of 0.78 mU · kg lean body wt⁻¹ · min⁻¹ in four nondiabetic subjects while glucose was clamped at ∼9.5 mmol/l for 7 h. As shown in Table 1, glucose, insulin, C-peptide, and glucagon concentrations reached a plateau within 1 h and did not change thereafter. The glucose infusion rate increased gradually for the first 5 h, then decreased slightly during the succeeding 2 h. Plasma [3-³H] glucose specific activity achieved a steady state within 1 h.

Splanchnic blood flow did not differ when measured from 150 to 180 min (1,066 ± 96 ml/min), 270 to 300 min (1,034 ± 95), or 390 to 420 min (1,197 ± 126). Splanchnic (tracer) glucose extraction (7.9 ± 0.4 vs. 7.6 ± 0.8 vs. 6.9 ± 0.1%, respectively) and SGU (12.3 ± 2.0 vs. 12.9 ± 2.1 vs. 12.6 ± 3.1 μmol · kg⁻¹ · min⁻¹, respectively) measured over the same intervals also did not differ (Fig. 6).

The present studies indicate that in the presence of hyperglycemia, an increase in insulin from ∼150 to ∼700 μmol/l resulted in a progressive increase in SGU in both diabetic and nondiabetic humans. On the other hand, EGP was maximally suppressed in the nondiabetic subjects at insulin concentrations of ∼350 pmol/l, but was further suppressed in the diabetic subjects at insulin concentrations up to ∼700 μU/ml. SGU was lower in the diabetic than nondiabetic subjects at all insulin concentrations tested, whereas glucose production only was higher in the diabetic than in the nondiabetic subjects during the lowest-dose insulin infusion. These data indicate that the insulin dose-response curves for stimulation of SGU and suppression of glucose production differ in nondiabetic humans and are abnormal in people with type 2 diabetes.

In the presence of hyperglycemia, SGU increased with increasing insulin concentrations in the nondiabetic subjects. This observation is consistent with that previously reported by Myers et al. (17) in nondiabetic dogs and by Petersen et al. (18) and ourselves (6,7) in humans. At the lowest insulin concentrations tested (∼150 pmol/l), SGU averaged ∼8.5 μmol · kg⁻¹ · min⁻¹, accounting for 23% of total body uptake. However, this comparison likely underestimates the potential contribution of hepatic glucose uptake to overall carbohydrate tolerance in the postprandial setting because portal insulin concentrations are several fold higher than peripheral insulin concentrations following food ingestion (i.e., when both glucose and insulin concentrations are normally increased). Assuming a portal venous-to-peripheral venous insulin gradient of ∼2.3 to 1 (13,15), the 37 μmol · kg⁻¹ · min⁻¹ of total glucose uptake observed in the presence of peripheral insulin concentrations of 150 pmol/l would be accompanied by portal venous insulin concentrations of ∼350 pmol/l and SGU of 11.2 μmol · kg⁻¹ · min⁻¹. Under these circumstances, the splanchnic bed would account for ∼30% of insulin-stimulated glucose uptake. These numbers are estimates of the potential rather than the actual effects of insulin observed after eating a carbohydrate-containing meal, because in that situation glucose and insulin concentrations are not maintained at a constant level, as was done in the current experiments, but rather are continuously changing. Therefore, the prolonged insulin infusions may have resulted in higher rates of SGU and extrasplanchnic glucose uptake than would occur following a briefer increase in insulin concentrations. On the other hand, glucose concentrations following food ingestion are higher in the portal than hepatic vein, likely leading to greater hepatic glucose uptake than occurs when all glucose is given intravenously, as was done in the present experiments (17).

In the presence of hyperglycemia and hyperinsulinemia, the majority of nonhepatic glucose uptake is believed to occur in muscle. In the present experiments, leg glucose extraction and uptake increased 3- and 4.8-fold, respectively, as insulin was increased from ∼150 to ∼700 pmol/l. The proportionately greater increase in LGU than leg glucose extraction resulted from a slight but nonsignificant increase in leg blood flow with increasing insulin concentrations. On the other hand, splanchnic glucose extraction and SGU only increased 1.75-fold, indicating a greater range of response to insulin in the leg than in the liver. The true dose-response curves for these tissues are not known because the present experiments began with insulin concentrations in the midphysiologic range and there was no evidence of a maximal response in either tissue at the highest insulin concentration tested. Nevertheless, the wider range of response to insulin in the leg than in the splanchnic bed indicates the relative contribution of each to glucose disposal; therefore, carbohydrate tolerance will change as insulin concentrations change.

Consistent with previous experiments (1,3,6,13,14), maximal suppression of glucose production in the nondiabetic subjects occurred at insulin concentrations between ∼150 and ∼350 pmol/l. In contrast, as noted above, SGU continued to increase at insulin concentrations up to ∼700 pmol/l. Therefore, the mechanism by which insulin regulates these processes must differ. The amount of glucose released by the liver is determined by the balance between the rate that intrahepatic free glucose is phosphorylated to glucose-6-phosphate by glucokinase and the rate at which glucose-6-phosphate is dephosphorylated to glucose by glucose-6-phosphatase. Insulin regulates flux through the glucose-6-phosphatase pathway both directly by altering the expression of the enzyme and indirectly by lowering other regulators such as free fatty acid concentrations (18,28,37–40). On the other hand, insulin increases hepatic glucose uptake by increasing the activity of glucokinase and glycogen synthetase (18,40,41). In the presence of hyperglycemia, insulin can also stimulate glycogen synthetase by increasing intrahepatic glucose-6-phosphate concentrations (24). Since glucokinase is believed (41–43) to be rate limiting for hepatic glucose uptake, the present experiments imply that insulin concentrations sufficient to maximally suppress flux through glucose-6-phosphatase are not sufficient to maximally stimulate flux through glucokinase. Therefore the signals regulating these pathways must differ.

Glucagon concentrations were maintained constant at all insulin concentrations. In addition, while we did not measure free fatty acid concentrations, previous experiments (14) have shown that they are maximally suppressed in nondiabetic subjects at insulin concentrations <150 pmol/l. Therefore neither likely played a role in the modulation of glucose production and uptake under the present experimental conditions. Insulin can increase both the amount and activity of hepatic glucokinase, leaving open the possibility that the 7-h insulin infusion stimulated glucokinase synthesis (28). However, the effects of insulin on SGU did not appear to change with time because the rates did not differ from 180 min onward during the 7-h 0.5-mU/kg insulin infusion. This observation is consistent with previous reports (16,17) wherein the effects of insulin on splanchnic glucose balance reach a plateau within 90 min.

Taken together, the present data indicate that the shape of the insulin dose-response curves for the regulation of hepatic glucose release and uptake differ, with the former being very sensitive to small changes in insulin and the latter continuing to increase at insulin concentrations spanning the physiologic range. They also indicate that in the presence of physiologic insulin concentrations, the splanchnic bed can account for ∼25–30% of total body glucose uptake.

The present data once again demonstrate that type 2 diabetes impairs insulin-induced suppression of glucose production and stimulation of SGU and LGU (1–3,6,7,13). However, although glucose production was eventually suppressed to nondiabetic rates, both SGU and LGU remained lower than nondiabetic rates at insulin concentrations spanning the physiologic range. As with the nondiabetic subjects, the relative contribution of the splanchnic bed and muscle (as reflected by LGU) to total body uptake in the diabetic subjects differed depending on the prevailing insulin concentration. In the nondiabetic subjects, leg glucose extraction was two- to threefold greater than splanchnic glucose extraction at all insulin concentrations examined. In contrast, splanchnic and leg fractional glucose extraction (both 2%) were virtually identical in the diabetic subjects at insulin concentrations of ∼150 pmol/l (Figs. 4 and 5). In addition, the absolute amount of glucose taken up in the diabetic subjects by the splanchnic bed at those insulin concentrations (∼3.3 μmol · kg⁻¹ · min⁻¹) exceeded that taken up by both legs (∼2.4 μmol · kg⁻¹ · min⁻¹, assuming equivalent uptake in each leg). This emphasizes the marked degree of muscle insulin resistance in people with type 2 diabetes as well as the importance of hepatic glucose uptake in the regulation of glucose tolerance in the presence of low physiologic insulin concentrations. On the other hand, the relative contribution of muscle and SGU changed as insulin concentrations increased. Leg fractional glucose extraction in the diabetic subjects was twofold greater than splanchnic fractional glucose extraction when insulin was increased to ∼350 pmol/l (6 vs. 3%) and almost four times that of splanchnic glucose extraction at insulin concentrations of ∼700 pmol/l (15 vs. 4%). This resulted in rates of SGU that were slightly lower than those of both legs at insulin concentrations of ∼350 pmol/l (∼5.6 vs. ∼6.7 μmol · kg⁻¹ · min⁻¹) and substantially lower than those of both legs at insulin concentrations of ∼700 pmol/l (∼9.2 vs. 22.4 μmol · kg⁻¹ · min⁻¹). Thus, although the ability of insulin to stimulate muscle glucose uptake is severely impaired in diabetic subjects, as with nondiabetic subjects, the range of response to insulin substantially exceeds that of the liver. Nevertheless, in the presence of insulin concentrations likely to be present in diabetic individuals under the conditions of daily living, SGU is likely to make a substantial contribution to total glucose uptake in people with type 2 diabetes.

The highest insulin concentration examined in the present experiments was ∼700 pmol/l. We therefore do not know whether SGU and LGU would have eventually reached the same maximal rates if the insulin concentrations were increased to sufficiently high concentrations. Previous studies (1,3,13,14) have shown that maximal rates of total body glucose disposal are lower in obese diabetic than nondiabetic subjects, similar to those who participated in the current experiments. Interestingly, the increment in both LGU and total body glucose disposal when insulin was increased from ∼150 to ∼350 pmol/l was lower in the diabetic than in the nondiabetic subjects, but did not differ when insulin was further increased to ∼700 pmol/l. This indicates that the insulin dose-response curve for both LGU and total body disposal was steeper in the nondiabetic than diabetic subjects at lower insulin concentrations but appeared to become parallel at higher concentrations. This pattern could have occurred if the step that was both defective and rate limiting for muscle glucose uptake at low insulin concentrations (e.g. glucose transport and/or phosphorylation) in the diabetic subjects was no longer rate limiting at higher insulin concentrations. In contrast, while the absolute rate of SGU was lower in the diabetic than nondiabetic subjects, the increment in uptake with each increment in insulin did not differ between groups. Glucokinase is believed to be rate limiting for hepatic glucose uptake, and the amount and activity of hepatic glucokinase has been reported (42–45) to be lower in animal models of diabetes and in people with diabetes. Therefore, the right-shifted (i.e. lower but parallel) insulin dose curve for stimulation of SGU could be explained if the amount of active glucokinase was reduced in the diabetic subjects in the basal state with preserved translocation and activation of glucokinase in response to an increment in insulin. It would be of interest in this regard if the insulin dose-response curve for the stimulation of hepatic glucose uptake also is shifted to the right in individuals known to have genetic defects (i.e., MODY2) in the activation of hepatic glucokinase (43).

In summary, the present experiments indicate that shape of the insulin dose-response curves for suppression of glucose production and stimulation of SGU differ in nondiabetic subjects. Whereas a relatively small increase in insulin results in maximal suppression of glucose production, SGU continues to increase at insulin concentrations spanning the physiologic range. Type 2 diabetes impairs both processes, with the defect in insulin-induced suppression of glucose production only being evident at lower insulin concentrations. In the presence of hyperglycemia, insulin-induced stimulation of SGU and muscle glucose uptake both contribute to overall glucose disposal in both diabetic and nondiabetic subjects. However, due to the presence of marked insulin resistance in muscle, the relative contribution of SGU to overall glucose disposal is greater in diabetic subjects, particularly at insulin concentrations likely to be present under conditions of daily living. Finally, the right shifted but parallel insulin dose-response curve for the stimulation of SGU observed in the diabetic subjects suggests a defect that lowers basal rates of uptake but does not alter the subsequent response to insulin. However, because the subjects who participated in the present studies were on average either overweight or obese and were predominantly 50–70 years of age, additional studies will be required to determine whether these conclusions also apply to people with type 2 diabetes whose age and degree of adiposity differs. Nevertheless, taken together, these data emphasize the importance of SGU as a contributor to overall glucose disposal. They also suggest that agents that enhance SGU in diabetic patients (e.g. glucokinase activators) are likely to have favorable effects on glucose tolerance.

This study was supported by the U.S. Public Health Service (DK29953 and RR-00585), a Novo Nordisk research infrastructure grant, and the Mayo Foundation. R.B. was supported by an American Diabetes Association mentor-based fellowship.

We thank B. Dicke, L. Heins, R. Rood, and C. Nordyke for technical assistance; J. Feehan, B. Norby, and the staff of the Mayo General Clinical Research Center for assistance in performing the studies; and M. Davis for assistance in preparation of the manuscript.