SEARCH

SEARCH BY CITATION

Keywords:

  • glucagon;
  • glucose production;
  • hepatic insulin sensitivity

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Objective: Obese non-diabetic patients are characterized by an extra-hepatic insulin resistance. Whether obese patients also have decreased hepatic insulin sensitivity remains controversial.

Research Methods and Procedures: To estimate their hepatic insulin sensitivity, we measured the rate of exogenous insulin infusion required to maintain mildly elevated glycemia in obese patients with type 2 diabetes, obese non-diabetic patients, and lean control subjects during constant infusions of somatostatin and physiological low-glucagon replacement infusions. To account for differences in insulin concentrations among the three groups of subjects, an additional protocol was also performed in healthy lean subjects with higher insulin infusion rates and exogenous dextrose infusion.

Results: The insulin infusion rate required to maintain glycemia at 8.5 mM was increased 4-fold in obese patients with type 2 diabetes and 1.5-fold in obese non-diabetic patients. The net endogenous glucose production (measured with 6,6-2H2-glucose) and total glucose output (measured with 2-2H1-glucose) were ∼30% lower in the patients than in the lean subjects. Net endogenous glucose production and total glucose output were both markedly increased in both groups of obese patients compared with lean control subjects during hyperinsulinemia.

Discussion: Our data indicate that both obese non-diabetic and obese type 2 diabetic patients have a blunted suppressive action of insulin on glucose production, indicating hepatic and renal insulin resistance.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Obese type 2 diabetic patients are characterized by an impaired suppression of glucose production during hyperinsulinemia. Hepatic insulin resistance has been proposed as responsible for this enhanced glucose production (1, 2). Alternatively, increased glucagon secretion in type 2 diabetes, resulting in an elevation of the portal venous glucagon:insulin ratio may play a role (3). Such an impaired suppression by insulin of endogenous glucose production, together with decreased insulin secretion, extra-hepatic insulin resistance, and possibly decreased stimulation of glucose disposal by hyperglycemia itself, results in fasting and postprandial hyperglycemia (4).

Obesity per se is characterized by an extra-hepatic insulin resistance (5). Whether obese patients also have decreased hepatic insulin sensitivity remains a matter of debate. If that were the case, it may possibly be a factor predisposing to the development of diabetes in obese individuals. To estimate their hepatic insulin sensitivity, we measured the rate of exogenous insulin infusion required to maintain a mildly elevated glycemia (in the normal postprandial range) in obese non-diabetic and obese type 2 diabetic patients during constant infusions of somatostatin and physiological low-glucagon replacement infusions. The net endogenous glucose production (NEGP) was measured by means of a constant infusion of 6,6-2H2 glucose. In addition, because glucose-6-phosphatase and glucokinase are believed to play prominent roles in the regulation of glucose production by controlling the rate of glucose efflux and uptake in hepatocyte, we also measured total glucose output (TGO; i.e., total glucose flux through glucose-6-phosphatase) and the rates of glucose-glucose-6-phosphatase cycling (i.e., the difference between total fluxes through glucose-6-phosphatase and glucokinase) by simultaneously infusing 2-2H glucose (6).

Research Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Ten healthy human volunteers (six men and four women), six obese patients with impaired glucose tolerance(five men and one woman), and six type 2 diabetic obese patients (six men) were selected to participate in one of two experimental protocols. Healthy volunteers were all in good physical health and did not take any medication at the time the experiments were performed. None had a family history of diabetes or obesity among first-degree relatives. Obese type 2 diabetic patients were maintained in a good metabolic control with oral anti-diabetic agents [mean glycated hemoglobin(HbA1c), 6.2 ± 1.1%; normal values, 4% to 6%]. They discontinued all drug treatment 3 days before the study. The physical characteristics of the study participants are shown in Table 1. All subjects had maintained a constant body weight for at least 3 months before taking part in the study. They were placed on an isocaloric, 50% carbohydrate diet during the 3 days before each study. The experimental protocols were approved by the ethical committees of Lausanne and Liège Universities Schools of Medicine, and all participants provided informed, written consent.

Table 1.  Characteristics of study participants
 SexAge (years)Body weight (kg)Height (cm)Body mass index kg/m2Habitual medicationHbA1c (%)
  1. HbA1c, glycated hemoglobin.

Healthy subjects (n = 10) 6 men/4 women 23.9 ± 1.071.5 ± 4.8177.6 ± 4.622.7 ± 1.7  
Obese non-diabetic patients       
#1M379217530.0  
#2M469517830.0  
#3M469317331.1  
#4M5415018046.3  
#5F308615635.3  
#6M459818728.0  
Obese type 2 diabetics       
#1M5810116835.8Glibenclamide/metformin5.1
#2M528717229.3Gliclazide/metformin7.4
#3M709817532.0Metformin5.2
#4M5110517135.9Glibenclamide/metformin7.8
#5M5611716741.9Glibenclamide/metformin8
#6M638717129.8Gliclazide/metformin7.8

General Procedures

All experiments began between 7:00 am and 8:00am after an overnight fast. At their arrival in the metabolic investigation laboratory, the subjects were weighed and measured. They were placed in a bed in a semi-recumbent position and two venous cannulas were inserted, one in an antecubital vein for infusion of hormones, tracers and glucose, and the other in a wrist vein of the contralateral arm for collection of blood samples. This hand was placed in a thermostabilized box heated at 50 °C to achieve partial arterialization of venous blood.

Protocol 1: Hyperglycemic pancreatic clamp

Six healthy subjects, six obese patients, and six type 2 diabetic obese patients took part to this protocol. Cyclic somatostatin(Somatostatine, UCB, Brussels, Belgium) was continuously infused at a rate of 350 μg/h. Glucagon (Glucagen; Novo Nordisk, Copenhagen, Denmark) was infused at a rate of 0.5 ng/kg per minute for 180 minutes (time, 0 to 180 minutes) to obtain low physiological plasma glucagon concentrations. An infusion of crystalline insulin (Actrapid HM; Novo Nordisk) was adjusted during the initial 90 minutes of the protocol to achieve steady plasma glucose concentrations of ∼8 mM. Primed (1.6 μmol/kg), continuous (0.016μmol/kg per minute) infusions of 2-2H glucose and 6,6-2H glucose (both from Masstrace, Worcester, MA) were administered throughout the experiment to measure the rates of glucose production and glucose cycling (GC). Blood samples were taken at 5- and 10-minute intervals to measure plasma glucose using a glucose analyzer II (Beckman, Fullerton, CA) and at 30-minute intervals for determination of hormones, substrates, and tracer concentrations.

Protocol 2: Hyperglycemic-Hyperinsulinemic Clamp Studies

To account for differences in insulin concentrations between obese and lean subjects, seven healthy subjects (four men and three women) took part in this protocol. Somatostatin and glucagon were infused as in protocol 1, whereas crystalline insulin was administered at a rate of 1.5 pmol/kg per minute to reproduce plasma insulin concentrations similar to those obtained in obese patients in protocol 1. An infusion of dextrose 20% was periodically adjusted to maintain plasma glucose ∼8 mM. The exogenous glucose was labeled with 1.5% 6,6-2H glucose and 2-2H glucose to minimize changes in plasma glucose enrichments over time (7).

Analytical Procedures

Plasma 6,6- and 2-2H glucose were determined by gas chromatography-mass spectrometry (Hewlett Packard, Palo Alto, CA). Plasma insulin (Biodata, Guidonia Montecello, Italy) and glucagon(Linco Research, St. Charles, MO) were measured by radioimmunoassay using commercial kits.

Calculations

Glucose rates of appearance and disappearance were calculated using Steele's equations for steady-state conditions in protocols 1, and “hot infusate” equations in protocol 3 (7). TGO was obtained from 2-2H2 glucose enrichment and NEGP from 6,6-2H2 glucose enrichments (6, 8). In protocol 2, the exogenous glucose infusion was subtracted from the glucose rate of appearance. GC was calculated as the difference between TGO and NEGP (6).

Statistical Analysis

All measurements performed during the last 60 minutes of the protocol were averaged for data presentation and statistical analysis. Results in the text, figures, and tables are shown as means ± SEM unless otherwise stated. Between groups comparisons were performed by unpaired t tests with Bonferroni's correction.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

Changes in plasma insulin, C-peptide, and free fatty acids during protocol 1 are indicated in Table 2. Exogenous glucagon infusion resulted in plasma glucagon concentrations similar in obese type 2 diabetics patients, obese non-diabetics subjects, and lean controls. Somatostatin infusion efficiently inhibited endogenous insulin secretion, resulting in a suppressed C-peptide concentration (<0.4 μg/liter) in all three groups of subjects.

Table 2.  Plasma insulin, glucagon, and nonesterified fatty acid concentrations in basal state and after somatostatin-insulin-glucagon infusions (clamp) in lean subjects, obese non-diabetic subjects and type 2 obese diabetic subjects (protocol 1)
 Insulin (pM) Glucagon (ng/liter) Free fatty acid (mEq/liter) 
  • *

    p < 0.05 vs. lean subjects.

  • p < 0.01 vs. basal.

  • p < 0.05 vs. obese.

  • §

    p < 0.05 vs. basal.

 BasalClampBasalClampBasalClamp
Lean subjects64 ± 1167 ± 787 ± 972 ± 70.43 ± 0.130.21 ± 0.06§
Obese subjects119 ± 53*163 ± 60*81 ± 1669 ± 110.85 ± 0.2*0.65 ± 0.14
Type 2 diabetic patients128 ± 14*308 ± 70*,63 ± 450 ± 31.35 ± 0.2*,0.67 ± 0.20§

Basal plasma free fatty acid concentrations were slightly but significantly higher in type 2 diabetic patients than in obese non-diabetics and lean subjects. In type 2 diabetic patients and in lean subjects, plasma free fatty acid concentrations significantly decreased during exogenous insulin infusion, whereas in obese subjects, plasma free fatty acid concentrations decreased only slightly during the test (not significant). Compared with healthy lean volunteers, obese type 2 diabetic patients required a total insulin infusion∼4-fold higher to maintain a comparable plasma glucose concentration(lean subjects, 0.10 ± 0.03 mU/kg per minute; obese type 2 diabetic patients, 0.41 ± 0.07 mU/kg per minute; p < 0.01). The higher insulin infusion rate resulted in a plasma insulin concentration of 308 ± 70 pM in obese type 2 diabetic patients, which was markedly higher than the one obtained in healthy lean volunteers (67 ± 7 pM; Figure 1). In obese non-diabetic patients, the insulin infusion rates required to maintain glycemia at its target value (0.15 ± 0.05 mU/kg per minute) was 1.5-fold higher than in healthy lean subjects, and resulted in a plasma insulin concentration of 162 ± 76 pM (p < 0.05 vs. both lean and obese type 2 diabetics;Figure 1).

image

Figure 1. Plasma glucose, insulin, and free fatty acid concentrations observed during the last hour of the experiment in lean control subjects (C), in lean control subjects during hyperglycemic-hyperinsulinemic clamp studies (HI), in obese non-diabetic patients (O), and in obese type 2 diabetic subjects (D). *p < 0.05 vs. C; $p < 0.05 vs. HI.

Download figure to PowerPoint

In both obese type 2 diabetic patients and obese non-diabetic patients, TGO and NEGP were lower than in healthy lean volunteers (p < 0.05 in both cases). GC, however, was comparable in all three groups (Figure 2).

image

Figure 2. Total glucose output (TGO), net endogenous glucose production (NEGP), glucose cycling (GC), and glucose rate of disappearance (GRd) during the last hour of the experiment in lean control subjects (C), in lean control subjects during hyperglycemic-hyperinsulinemic clamp studies (HI), in obese non-diabetic patients (O), and in obese type 2 diabetic subjects (D). *p < 0.05 vs. C; #p < 0.05 vs. O; §p < 0.05 vs. D.

Download figure to PowerPoint

In protocol 2, a group of healthy lean volunteers was studied while receiving the same exogenous insulin infusion as the one required to maintain glycemia at its target value in obese subjects. The plasma insulin concentrations obtained during these experiments were not significantly different from those obtained in obese non-diabetic patients (Figure 1). This modest degree of hyperinsulinemia resulted in a 67% reduction of NEGP and 63% reduction in GC compared with lean control subjects (Figure 2). Glucose rate of disappearance, however, increased by 46% (from 13.3 ± 0.6 to 19.4 ± 2.8 μmol/kg per minute). When obese non-diabetic patients were compared with healthy subjects studied during these clamped conditions at similar insulinemia, they showed a significant increase of TGO (+149%), NEGP(+93%), GC (+204%), and a decrease in glucose rate of disappearance. Obese diabetic patients also showed comparable significant increases in TGO (+146%), NEGP (+93%), GC (+214%), and a decrease in glucose rate of disappearance (−55%), although their plasma insulin concentrations were ∼2.5-fold higher than in lean subjects.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

These experiments were designed to assess the sensitivity to insulin of glucose-producing tissues (essentially the liver with a possible contribution of the kidney) (9). For this purpose, somatostatin was infused while glucagon was administered to produce low physiological plasma glucagon concentrations in all three groups of subjects. Exogenous insulin infusion was adjusted to obtain steady, mild hyperglycemia (in the normal postprandial range). This design allowed us to estimate the sensitivity to insulin of glucose-producing tissues from the rate of insulin infusion required to maintain glycemia and to compare NEGP, TGO, and GC at similar glucose and glucagon concentrations. To account for differences in insulin concentrations among the three groups of subjects, an additional protocol was also performed in healthy lean subjects with higher insulin infusion rates and exogenous dextrose infusion(hyperinsulinemic clamp).

These data indicate that the insulin infusion rate required to maintain glycemia close to the target value of 8.5 mM was increased 4-fold in obese type 2 diabetic patients and 1.5-fold in obese non-diabetic patients. These 4- and 1.5-fold increases in insulin infusion rate in obese type 2 diabetic patients and obese non-diabetic patients resulted in 6.0- and 3.0-fold increases in plasma insulin concentrations. This reflects most likely a decreased hepatic insulin clearance associated with obesity (10, 11). The plasma insulin concentrations attained were consequently significantly higher in obese type 2 diabetic patients and in obese non-diabetic subjects than in healthy subjects. This clearly indicates insulin resistance in these two groups of patients (5, 12). The resistance to insulin actions affected whole-body glucose use, as indicated by a lower glucose rate of disappearance in type 2 diabetic and obese non-diabetic subjects, compared with healthy subjects. The decreased glucose use even underscores the degree of insulin resistance in these groups of patients because their plasma insulin concentrations were markedly higher than in healthy volunteers. In addition, our data clearly indicate that suppression of glucose production in response to insulin was impaired in obese type 2 diabetic and obese non-diabetic patients. NEGP and TGO were ∼30% lower in the patients than in lean subjects, while their plasma insulin concentrations were increased ∼6-fold and 3-fold, respectively. However, when glucose turnover data of obese type 2 diabetic and non-diabetic patients were compared with those measured in healthy subjects during moderate hyperinsulinemia (i.e., similar to those observed during protocol 1 in obese non-diabetic and approximately one-half that of obese type 2 diabetic patients), NEGP and TGO were both markedly increased in both groups of patients. This unequivocally indicates resistance of glucose-producing tissues to the suppressive effects of insulin in obese patients.

Insulin is known to suppress endogenous glucose production by both direct effects on liver cells and by indirect effects, involving reduction of adipose tissue lipolysis and reduction in plasma free fatty acid concentrations (13). In both groups of obese patients, plasma free fatty acid concentrations remained significantly higher than in healthy subjects, despite markedly higher insulinemia, suggesting that a decreased suppression of peripheral lipolysis played a role in the impaired suppression of glucose production. However, we cannot exclude at this point that a decreased direct effect of insulin on liver cells contributed as well.

Net glucose production occurs mainly in liver cells and depends on the balance between the activity of two opposing enzymes, glucokinase and glucose-6-phosphatase. The simultaneous activity of these two enzymes is responsible for the magnitude of the cycling between glucose and glucose-6-phosphate (14). In our experiments, hyperinsulinemia reduced both NEGP and TGO in healthy lean subjects and reduced drastically the rate of GC. This suggests that hyperinsulinemia essentially decreased glucose-6-phosphataseactivity because an insulin-induced increase in glucokinase would have been expected to increase GC. The mechanisms responsible for insulin resistance of glucose-producing tissues in obese non-diabetic patients cannot be delineated from our experiments. The observation of an increased GC compared with healthy subjects at comparable hyperinsulinemia is consistent with a higher activity of glucose 6-phosphatase. Such an increased activity may be secondary to a decreased suppression of the enzyme activity exerted by insulin at the level of the liver cell. Alternatively, it may possibly be secondary to the increased peripheral lipolysis and enhanced plasma free fatty acid concentrations because chronically elevated plasma free fatty acids have been shown to enhance liver glucose 6-phosphatase gene expression (15).

Obese type 2 diabetic patients also showed resistance to the suppressive actions of insulin on glucose production. Their degree of insulin resistance was even markedly higher than in obese non-diabetic patients, because obese diabetic patients required more than twice as much insulin to maintain the glycemia at ∼8 mM compared with obese non-diabetic patients. The mechanisms responsible for this worsening of hepatic/renal insulin resistance in type 2 diabetes remain unexplained. It is possible that factors distinct of obesity were present in type 2 diabetes. Alternatively, it is possible that the chronic hyperglycemia present in these patients worsened their insulin resistance. In support of this hypothesis, it has been observed that high-glucose concentrations decrease glucose-6-phosphatase activity in hepatocytes (16). Our observation of a similar GC activity in diabetic patients, although their plasma insulin concentrations were more than twice that of obese non-diabetic subjects, may reflect a further degree of impairment of the suppression of GC by insulin, possibly secondary to a higher glucose-6-phosphatase activity.

In these experiments, adjustment of the insulin infusion required to obtain a similar, 8 mM glycemia in all groups of subjects resulted in higher insulin concentrations in obese non-diabetics and obese patients with type 2 diabetes. Under such conditions, NEGP, TGO, and glucose rates of disappearance were all lower in both groups of obese patients compared with healthy subjects. This means that, in both groups of patients, the same glycemia as in healthy subjects was maintained with a lower rate of glucose production due to a marked reduction of insulin-mediated glucose disposal. This illustrates the complexity of the disturbance of glucose homeostasis in obesity and type 2 diabetes mellitus. In has, indeed, been observed that, in fasting conditions, both an increase in endogenous glucose production and a decreased glucose disposal operate to produce fasting hyperglycemia in type 2 diabetes mellitus (17). Accordingly, in type 2 diabetic patients, an increase in endogenous glucose production leads to a more important increase in glycemia than would be expected, due to the concomitant reduction in glucose use.

In conclusion, our data indicate that both obese non-diabetic and obese type 2 diabetic patients have a blunted suppressive action of insulin on glucose production, indicating hepatic and/or renal insulin resistance. The increased lipolysis and free fatty acid concentrations observed in both groups of patients and an increased glucose-6-phosphatase activity, suggested by the increased GC, may play a role in this hepatic/renal insulin resistance.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References

This study was supported by Grant 32-56700.99 from the Swiss National Science Foundation.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Acknowledgments
  8. References
  • 1
    Ferrannini, E., Locatelli, L., Jéquier, E., Felber, J. (1989) Differential effects of insulin and hyperglycemia on intracellular glucose disposition in humans. Metabolism 38: 459465.
  • 2
    Ferrannini, E., Groop, L. C. (1989) Hepatic glucose production in insulin-resistant states. Diabetes Metab Rev 5: 711725.
  • 3
    Lefèbvre, P., Paolisso, G., Scheen, A. (1991) The role of glucagon in non-insulin-dependent (type 2) diabetes mellitus. Sakamoto, N. Angel, A. Hotta, N. eds. New Directions in Research and Clinical Works for Obesity and Diabetes Mellitus 2529. Elsevier Science Publishers BV Amsterdam, The Netherlands.
  • 4
    DeFronzo, R. A., Bonadonna, R. C., Ferrannini, E. (1992) Pathogenesis of NIDDM: a balanced overview. Diabetes Care 15: 318368.
  • 5
    Felber, J. P., Acheson, K. J., Tappy, L. (1992) From Obesity to Diabetes John Wiley & Sons Chichester, UK.
  • 6
    Wajngot, A., Chandramouli, V., Schumann, W. C. (1989) Testing of the assumptions made in estimating futile cycling. Am J Physiol 256: E668EE75.
  • 7
    Finegood, D. T., Bergman, R. N., Vranic, M. (1987) Estimation of endogenous glucose production during hyperinsulinemic euglycemic glucose clamps: comparison of labeled and unlabelled glucose infusates. Diabetes 36: 914924.
  • 8
    Wajngot, A., Khan, A., Giacca, A., Vranic, M., Efendic, S. (1990) Dexamethasone increases glucose cycling, but not glucose production in healthy subjects. Am J Physiol 259: E626EE32.
  • 9
    Stumvoll, M., Meyer, C., Mitrakou, A., Nadkarni, V., Gerich, J. E. (1997) Renal glucose production and utilization: new aspects in humans. Diabetologia 40: 749757.
  • 10
    Robertson, D. A., Singh, B. M., Hale, P. J., Nattrass, M. (1992) Effects of morbid obesity on insulin clearance and insulin sensitivity in several aspects of metabolism as assessed by low-dose insulin infusion. Metab Clin Exp 41: 604612.
  • 11
    Escobar, O., Mizuma, H., Sothern, M. S., et al (1999) Hepatic insulin clearance increases after weight loss in obese children and adolescents. Am J Med Sci 317: 282286.
  • 12
    DeFronzo, R. A. (1988) The triumvirate:β-cell, muscle, liver: a collusion responsible for NIDDM. Diabetes 37: 667687.
  • 13
    Bergman, R. N., Ader, M. (2000) Free fatty acids and pathogenesis of type 2 diabetes mellitus. Trends Endocrinol Metab 11: 351356.
  • 14
    Vranic, M. (1992) Banting lecture: glucose turnover: a key to understanding the pathogenesis of diabetes (indirect effects of insulin). Diabetes 41: 11881206.
  • 15
    Massillon, D., Barzilai, N., Hawkins, M., Prus-Wertheimer, D., Rossetti, L. (1997) Induction of hepatic glucose-6-phosphatase gene expression by lipid infusion. Diabetes 46: 153157.
  • 16
    Mevorach, M., Giacca, A., Aharon, Y., Hawkins, M., Shamoon, H., Rossetti, L. (1998) Regulation of endogenous glucose production by glucose per se is impaired in type 2 diabetes mellitus. J Clin Invest 102: 744753.
  • 17
    Fery, F. (1994) Role of hepatic glucose production and glucose uptake in the pathogenesis of fasting hyperglycemia in type 2 diabetes: normalization of glucose. J Clin Endocrinol Metab 78: 536542.