0021-972X/90/7106-1544$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1990 by The Endocrine Society
Vol. 71, No. 6 Printed in U.S.A.
Fasting Hyperglycemia Normalizes Oxidative and Nonoxidative Pathways of Insulin-Stimulated Glucose Metabolism in Noninsulin-Dependent Diabetes Mellitus* LAWRENCE J. MANDARINO, AGOSTINO CONSOLIt, DAVID E. KELLEY, JAMES J. REILLY, AND NURJAHAN NURJHAN Departments of Ophthalmology, Physiology, Surgery, and Medicine, University of Pittsburgh, and The Eye and Ear Institute of Pittsburgh, Pittsburgh, Pennsylvania 15213
ABSTRACT. The present studies were undertaken to determine whether fasting hyperglycemia can compensate for decreased insulin-stimulated glucose disposal, oxidation, and storage in noninsulin-dependent diabetes mellitus (NIDDM) as well as to determine whether hyperglycemia normalizes insulin-stimulated skeletal muscle glycogen synthase and pyruvate dehydrogenase (PDH) activities. To accomplish this, we used the glucose clamp technique with isotopic determination of glucose disposal and indirect calorimetry for measuring the pathways of glucose metabolism, and vastus lateralis muscle biopsies to determine the effects of insulin on glycogen synthase and PDH activities. Nine patients with NIDDM and eight matched nondiabetic subjects were infused with insulin (40 mU/m2-min) while plasma glucose was maintained at the prevailing fasting
I
NSULIN resistance is a characteristic feature of noninsulin-dependent diabetes mellitus (NIDDM). It is well established that when NIDDM subjects are infused with insulin and rendered euglycemic, their glucose disposal rates are significantly less than those of nondiabetic subjects infused with the same amount of insulin (1-4). However, when NIDDM subjects are studied without reducing their plasma glucose concentrations, that is at their prevailing fasting hyperglycemia, their insulinstimulated glucose disposal rates are not significantly different from those of nondiabetic subjects made comparably hyperinsulinemic and studied at their prevailing fasting plasma glucose concentration (5). This observation has been interpreted to indicate that fasting hyperglycemia exists in these patients as a compensatory Received April 6, 1990. Address all correspondence and requests for reprints to: L. Mandarino, Ph.D., The Eye and Ear Institute of Pittsburgh, 203 Lothrop Street, Pittsburgh, Pennsylavania 15213. * This work was supported in part by the American Diabetes Association, Juvenile Diabetes Foundation, Grant ROl-DK-41075, General Clinical Research Center Grant 5-MO1-RR-00056, Research to Prevent Blindness, Inc., the Pennsylvania Lions, and The Eye and Ear Institute of Pittsburgh. t On leave from the University of Chieti, Chieti, Italy.
concentration. During insulin infusion, rates of glucose disposal, storage, and oxidation were the same in the two groups. Insulin infusion significantly activated glycogen synthase fractional velocity to the same extent in NIDDM (0.210 ± 0.056 vs. 0.332 ± 0.079) and controls (0.192 ± 0.036 us. 0.294 ± 0.050). Insulin infusion increased PDH fractional velocity in controls (from 0.281 ± 0.022 to 0.404 ± 0.038), but not in NIDDM (from 0.356 ± 0.043 to 0.436 ± 0.060), although the activity of PDH during insulin infusion did not differ between the groups. We conclude that prevailing fasting hyperglycemia normalizes the nonoxidative and oxidative pathways of insulin-stimulated glucose in metabolism in NIDDM and may act as a homeostatic mechanism to normalize muscle glucose metabolism. {J Clin Endocrinol Metab 7 1 : 1544-1551, 1990)
homeostatic mechanism that normalizes glucose disposal (5). However, an equally possible alternative interpretation is that fasting hyperglycemia in NIDDM represents the combined effects of overproduction and underutilization of glucose, so that fasting plasma glucose concentrations are those at which mass action-induced increases in glucose utilization match the increased rates of glucose production, resulting in a new steady state for a given individual's metabolic defect. If this interpretation were correct, the pathways of intracellular glucose metabolism may not necessarily be normalized. However, if hyperglycemia were to serve a compensatory homeostatic function, one would predict that pathways of intracellular glucose metabolism as well as overall glucose disposal should be normal. Insulin-stimulated glucose disposal is composed of oxidative and nonoxidative metabolic pathways, the respective magnitude of which can be estimated by using indirect calorimetry in conjunction with isotopic determination of glucose flux (6). Investigators have usually defined whole body nonoxidative glucose metabolism as the difference between the overall rate of glucose disposal and the rate of glucose oxidation. As such, nonoxidative
1544
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 15:49 For personal use only. No other uses without permission. . All rights reserved.
HYPEROLYCEMIA NORMALIZES INSULIN-STIMULATED GLUCOSE METABOLISM
glucose metabolism includes glycogen synthesis and any pathway of glucose metabolism, such as lactate production, that does not involve gas exchange. Insulin stimulation of whole body rionoxidative glucose metabolism has been found to be Correlated with the activation of skeletal muscle glycogen synthase (GS), while increases in whole body glucose oxidation are correlated with activation of muscle pyfuvate dehydrogenase (PDH) in nondiabetic individuals (4, 7). Bogardus et al. (4) and Felber et al. (8), using the euglycemic clamp technique, demonstrated that the reduced insulin-stimulatdd glucose disposal found in obese NIDDM subjects rendered euglycemic could be accounted for primarily by decreased nonoxidative glucose metabolism, and that this was associated with reduced insulin-stimulated muscle GS activity. These observations suggest that decreased activation of this enzyme by insulin might account for the decreased insulin-stimulated nonoxidative glucose metabolism found under euglycemic conditions. Recently, it has been shown by using NMR technique$ that NIDDM subjects have decreased insulin-stimul&ted glycogen formation in leg muscle that can account for their decreased glucose disposal (9), providing additional evidence that there is an important defect in this pathway in NIDDM subjects. However, it must be borne in mind that the continuous insulin infusions used in these studies represent an unphysiological situation. On the other hand, in more physiological conditions (e.g. oral glucose tolerance test), Felber et al. (8), using indirect calorimetry, also found defects in both oxidative and nonoxidative glucose metabolism in NIDDM and concluded that hyperglycemia did not compensate for defects in either of these pathways. In addition, Wright et al. (10) found that activation of skeletal muscle glycogen synthase was reduced after ingestion of a mixed meal by patients with NIDDM, providing additional evidence that the pathway of nonoxidative glucose metabolism is abnormal in NIDDM. Unfortunately, in both of those studies the plasma insulin response in the NIDDM subjects was reduced, so that it is impossible to determine whether hypfcrglycemia was unable to compensate for reduced insulin-stimulated glucose metabolism or whether decreased plasma insulin concentrations were responsible for the defects. Furthermore, Thorburn et al. (11) recently reported that hyperglycemia during hyperinsulinemia normalized glucose disposal in NIDDM, primarily by increasing nonoxidative glucose disposal. Since plasma lactate concentrations increased more in the diabetic subjects than in the nondiabetic subjects during the insulin infusions, and since muscle glycogen synthase activity remained abnormally low, it was concluded that increased conversion of glucose to lacjtate, and not increased muscle
1545
glycogen synthesis, was primarily responsible for the increase in nonoxidative muscle glucose metabolism. However, this conclusion can be questioned on several grounds. First of all, in those studies it was necessary to double the average basal hyperglycemia from 9.9 to 20 mM (with glucose concentrations in some subjects being as high as 30 mM) to achieve glucose disposal rates matching those of the nondiabetic subjects. This must have increased glucose uptake by the gut and liver (1218), with consequent stimulation of glycogen deposition in the liver and of glycolysis in the liver and gut (19-21). Given the increase in nonoxidative glucose metabolism in nonmuscle tissues, it is not surprising that changes in muscle glycogen synthase activity might not reflect changes in systemic nonoxidative glucose metabolism under those conditions. The foregoing considerations indicate that it remains to be established whether fasting hyperglycemia can compensate for decreased insulin-stimulated glucose disposal in NIDDM and which pathways of insulin-stimulated glucose disposal are primarily affected by hyperglycemia. The present studies were, therefore, undertaken to answer these questions as well as to determine whether fasting hyperglycemia is associated with normal insulinstimulated muscle GS and PDH activities.
Materials and Methods Subjects All studies were approved by the Human Studies Committee of the School of Medicine of the University of Pittsburgh; informed written consent was obtained from each subject. The characteristics of the subjects are shown in Table 1. All subjects with NIDDM (n = 9) were being treated with oral agents or diet, otherwise were in good health, and were withdrawn from any diabetic treatment for 2 weeks before study. Control subjects (n = 8) were without a family history of diabetes and were not taking any medications known to affect glucose metabolism. Nondiabetic and NIDDM volunteers were matched with regard to gender and body mass index (Table 1). Both nondiabetic and NIDDM subjects were moderately obese.
Study design (Fig. 1) Subjects were admitted to the General Clinical Research Center of the University of Pittsburgh on the evening before the study. On the morning of the study at 0530 h, a primed (30 fid) continuous (0.3 juCi/min) infusion of [6-3H]glucose (Amersham, Arlington Heights, IL) was begun and continued throughout the study. After allowing 190 min for isotope equilibration, rates of glucose turnover were determined at 10-min intervals over the next 30 min. During this same 30-min period, subjects were connected to an indirect calorimeter for determination of respiratory exchange ratios. At the end of the 30-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 15:49 For personal use only. No other uses without permission. . All rights reserved.
MANDARINO ET AL.
1546
JCE & M • 1990 Vol71-No6
TABLE 1. Subject characteristics
NIDDM (n = 9) Mean SEM
Controls (n = 8) Mean SEM
Fasting plasma glucose
Fasting plasma insulin
(mM)
(pM)
27.1 0.9
12.5" 0.7
83 17
26.6 0.8
5.4 0.1
75 7
Age (yr)
Sex (M/F)
Ht (cm)
Wt (kg)
BMI (kg/m2)
59° 2
8/1
170 3
79.1 9.3
52 2
6/2
172 3
80.8 4.6
BMI, Body mass index. " P < 0.05, greater than control values. h P < 0.01, greater than control values. 6-3H-GLUCOSE FIG. 1. Study design. Tritiated glucose [6-3H]glucose was infused for 190 min before the basal sampling period to ensure isotopic equilibration. Indirect calorimetry was performed for 30-min periods basally and during insulin infusion, as described in the text. A percutaneous muscle biopsy was performed under basal conditions and during the final 20 min of insulin infusion. Biopsies were performed after blood sampling and indirect calorimetry.
INSULIN (40 TtiU/m2/min) BLOOD SAMPLING & INDIRECT CALORIMETRY
•
MUSCLE BIOPSY
-240
-50
-20
• 0
150
180 200
TIME (min) min period of basal determinations, a needle muscle biopsy of the vastus lateralis muscle was performed under local anesthesia, and the muscle sample was immediately frozen in liquid nitrogen. Immediately after the biopsy, a primed continuous infusion of insulin (40 mU/m2 • min) was started and continued for the next 180 min, during which time a variable infusion of 50% dextrose was used to maintain plasma glucose at the concentration observed during the basal period. During the last 30 min of insulin infusion, when glucose infusion rates had reached a near-steady state (2.5-7.5% change in glucose infusion rate between 150-180 min and
Vol. 71, No. 6 Printed in U.S.A.
Fasting Hyperglycemia Normalizes Oxidative and Nonoxidative Pathways of Insulin-Stimulated Glucose Metabolism in Noninsulin-Dependent Diabetes Mellitus* LAWRENCE J. MANDARINO, AGOSTINO CONSOLIt, DAVID E. KELLEY, JAMES J. REILLY, AND NURJAHAN NURJHAN Departments of Ophthalmology, Physiology, Surgery, and Medicine, University of Pittsburgh, and The Eye and Ear Institute of Pittsburgh, Pittsburgh, Pennsylvania 15213
ABSTRACT. The present studies were undertaken to determine whether fasting hyperglycemia can compensate for decreased insulin-stimulated glucose disposal, oxidation, and storage in noninsulin-dependent diabetes mellitus (NIDDM) as well as to determine whether hyperglycemia normalizes insulin-stimulated skeletal muscle glycogen synthase and pyruvate dehydrogenase (PDH) activities. To accomplish this, we used the glucose clamp technique with isotopic determination of glucose disposal and indirect calorimetry for measuring the pathways of glucose metabolism, and vastus lateralis muscle biopsies to determine the effects of insulin on glycogen synthase and PDH activities. Nine patients with NIDDM and eight matched nondiabetic subjects were infused with insulin (40 mU/m2-min) while plasma glucose was maintained at the prevailing fasting
I
NSULIN resistance is a characteristic feature of noninsulin-dependent diabetes mellitus (NIDDM). It is well established that when NIDDM subjects are infused with insulin and rendered euglycemic, their glucose disposal rates are significantly less than those of nondiabetic subjects infused with the same amount of insulin (1-4). However, when NIDDM subjects are studied without reducing their plasma glucose concentrations, that is at their prevailing fasting hyperglycemia, their insulinstimulated glucose disposal rates are not significantly different from those of nondiabetic subjects made comparably hyperinsulinemic and studied at their prevailing fasting plasma glucose concentration (5). This observation has been interpreted to indicate that fasting hyperglycemia exists in these patients as a compensatory Received April 6, 1990. Address all correspondence and requests for reprints to: L. Mandarino, Ph.D., The Eye and Ear Institute of Pittsburgh, 203 Lothrop Street, Pittsburgh, Pennsylavania 15213. * This work was supported in part by the American Diabetes Association, Juvenile Diabetes Foundation, Grant ROl-DK-41075, General Clinical Research Center Grant 5-MO1-RR-00056, Research to Prevent Blindness, Inc., the Pennsylvania Lions, and The Eye and Ear Institute of Pittsburgh. t On leave from the University of Chieti, Chieti, Italy.
concentration. During insulin infusion, rates of glucose disposal, storage, and oxidation were the same in the two groups. Insulin infusion significantly activated glycogen synthase fractional velocity to the same extent in NIDDM (0.210 ± 0.056 vs. 0.332 ± 0.079) and controls (0.192 ± 0.036 us. 0.294 ± 0.050). Insulin infusion increased PDH fractional velocity in controls (from 0.281 ± 0.022 to 0.404 ± 0.038), but not in NIDDM (from 0.356 ± 0.043 to 0.436 ± 0.060), although the activity of PDH during insulin infusion did not differ between the groups. We conclude that prevailing fasting hyperglycemia normalizes the nonoxidative and oxidative pathways of insulin-stimulated glucose in metabolism in NIDDM and may act as a homeostatic mechanism to normalize muscle glucose metabolism. {J Clin Endocrinol Metab 7 1 : 1544-1551, 1990)
homeostatic mechanism that normalizes glucose disposal (5). However, an equally possible alternative interpretation is that fasting hyperglycemia in NIDDM represents the combined effects of overproduction and underutilization of glucose, so that fasting plasma glucose concentrations are those at which mass action-induced increases in glucose utilization match the increased rates of glucose production, resulting in a new steady state for a given individual's metabolic defect. If this interpretation were correct, the pathways of intracellular glucose metabolism may not necessarily be normalized. However, if hyperglycemia were to serve a compensatory homeostatic function, one would predict that pathways of intracellular glucose metabolism as well as overall glucose disposal should be normal. Insulin-stimulated glucose disposal is composed of oxidative and nonoxidative metabolic pathways, the respective magnitude of which can be estimated by using indirect calorimetry in conjunction with isotopic determination of glucose flux (6). Investigators have usually defined whole body nonoxidative glucose metabolism as the difference between the overall rate of glucose disposal and the rate of glucose oxidation. As such, nonoxidative
1544
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 15:49 For personal use only. No other uses without permission. . All rights reserved.
HYPEROLYCEMIA NORMALIZES INSULIN-STIMULATED GLUCOSE METABOLISM
glucose metabolism includes glycogen synthesis and any pathway of glucose metabolism, such as lactate production, that does not involve gas exchange. Insulin stimulation of whole body rionoxidative glucose metabolism has been found to be Correlated with the activation of skeletal muscle glycogen synthase (GS), while increases in whole body glucose oxidation are correlated with activation of muscle pyfuvate dehydrogenase (PDH) in nondiabetic individuals (4, 7). Bogardus et al. (4) and Felber et al. (8), using the euglycemic clamp technique, demonstrated that the reduced insulin-stimulatdd glucose disposal found in obese NIDDM subjects rendered euglycemic could be accounted for primarily by decreased nonoxidative glucose metabolism, and that this was associated with reduced insulin-stimulated muscle GS activity. These observations suggest that decreased activation of this enzyme by insulin might account for the decreased insulin-stimulated nonoxidative glucose metabolism found under euglycemic conditions. Recently, it has been shown by using NMR technique$ that NIDDM subjects have decreased insulin-stimul&ted glycogen formation in leg muscle that can account for their decreased glucose disposal (9), providing additional evidence that there is an important defect in this pathway in NIDDM subjects. However, it must be borne in mind that the continuous insulin infusions used in these studies represent an unphysiological situation. On the other hand, in more physiological conditions (e.g. oral glucose tolerance test), Felber et al. (8), using indirect calorimetry, also found defects in both oxidative and nonoxidative glucose metabolism in NIDDM and concluded that hyperglycemia did not compensate for defects in either of these pathways. In addition, Wright et al. (10) found that activation of skeletal muscle glycogen synthase was reduced after ingestion of a mixed meal by patients with NIDDM, providing additional evidence that the pathway of nonoxidative glucose metabolism is abnormal in NIDDM. Unfortunately, in both of those studies the plasma insulin response in the NIDDM subjects was reduced, so that it is impossible to determine whether hypfcrglycemia was unable to compensate for reduced insulin-stimulated glucose metabolism or whether decreased plasma insulin concentrations were responsible for the defects. Furthermore, Thorburn et al. (11) recently reported that hyperglycemia during hyperinsulinemia normalized glucose disposal in NIDDM, primarily by increasing nonoxidative glucose disposal. Since plasma lactate concentrations increased more in the diabetic subjects than in the nondiabetic subjects during the insulin infusions, and since muscle glycogen synthase activity remained abnormally low, it was concluded that increased conversion of glucose to lacjtate, and not increased muscle
1545
glycogen synthesis, was primarily responsible for the increase in nonoxidative muscle glucose metabolism. However, this conclusion can be questioned on several grounds. First of all, in those studies it was necessary to double the average basal hyperglycemia from 9.9 to 20 mM (with glucose concentrations in some subjects being as high as 30 mM) to achieve glucose disposal rates matching those of the nondiabetic subjects. This must have increased glucose uptake by the gut and liver (1218), with consequent stimulation of glycogen deposition in the liver and of glycolysis in the liver and gut (19-21). Given the increase in nonoxidative glucose metabolism in nonmuscle tissues, it is not surprising that changes in muscle glycogen synthase activity might not reflect changes in systemic nonoxidative glucose metabolism under those conditions. The foregoing considerations indicate that it remains to be established whether fasting hyperglycemia can compensate for decreased insulin-stimulated glucose disposal in NIDDM and which pathways of insulin-stimulated glucose disposal are primarily affected by hyperglycemia. The present studies were, therefore, undertaken to answer these questions as well as to determine whether fasting hyperglycemia is associated with normal insulinstimulated muscle GS and PDH activities.
Materials and Methods Subjects All studies were approved by the Human Studies Committee of the School of Medicine of the University of Pittsburgh; informed written consent was obtained from each subject. The characteristics of the subjects are shown in Table 1. All subjects with NIDDM (n = 9) were being treated with oral agents or diet, otherwise were in good health, and were withdrawn from any diabetic treatment for 2 weeks before study. Control subjects (n = 8) were without a family history of diabetes and were not taking any medications known to affect glucose metabolism. Nondiabetic and NIDDM volunteers were matched with regard to gender and body mass index (Table 1). Both nondiabetic and NIDDM subjects were moderately obese.
Study design (Fig. 1) Subjects were admitted to the General Clinical Research Center of the University of Pittsburgh on the evening before the study. On the morning of the study at 0530 h, a primed (30 fid) continuous (0.3 juCi/min) infusion of [6-3H]glucose (Amersham, Arlington Heights, IL) was begun and continued throughout the study. After allowing 190 min for isotope equilibration, rates of glucose turnover were determined at 10-min intervals over the next 30 min. During this same 30-min period, subjects were connected to an indirect calorimeter for determination of respiratory exchange ratios. At the end of the 30-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 November 2015. at 15:49 For personal use only. No other uses without permission. . All rights reserved.
MANDARINO ET AL.
1546
JCE & M • 1990 Vol71-No6
TABLE 1. Subject characteristics
NIDDM (n = 9) Mean SEM
Controls (n = 8) Mean SEM
Fasting plasma glucose
Fasting plasma insulin
(mM)
(pM)
27.1 0.9
12.5" 0.7
83 17
26.6 0.8
5.4 0.1
75 7
Age (yr)
Sex (M/F)
Ht (cm)
Wt (kg)
BMI (kg/m2)
59° 2
8/1
170 3
79.1 9.3
52 2
6/2
172 3
80.8 4.6
BMI, Body mass index. " P < 0.05, greater than control values. h P < 0.01, greater than control values. 6-3H-GLUCOSE FIG. 1. Study design. Tritiated glucose [6-3H]glucose was infused for 190 min before the basal sampling period to ensure isotopic equilibration. Indirect calorimetry was performed for 30-min periods basally and during insulin infusion, as described in the text. A percutaneous muscle biopsy was performed under basal conditions and during the final 20 min of insulin infusion. Biopsies were performed after blood sampling and indirect calorimetry.
INSULIN (40 TtiU/m2/min) BLOOD SAMPLING & INDIRECT CALORIMETRY
•
MUSCLE BIOPSY
-240
-50
-20
• 0
150
180 200
TIME (min) min period of basal determinations, a needle muscle biopsy of the vastus lateralis muscle was performed under local anesthesia, and the muscle sample was immediately frozen in liquid nitrogen. Immediately after the biopsy, a primed continuous infusion of insulin (40 mU/m2 • min) was started and continued for the next 180 min, during which time a variable infusion of 50% dextrose was used to maintain plasma glucose at the concentration observed during the basal period. During the last 30 min of insulin infusion, when glucose infusion rates had reached a near-steady state (2.5-7.5% change in glucose infusion rate between 150-180 min and
