Separate Contribution of Diabetes, Total Fat Mass, and Fat Topography to Glucose Production, Gluconeogenesis, and Glycogenolysis

Amalia Gastaldelli, Yoshinori Miyazaki, Maura Pettiti, Emma Buzzigoli, Srikanth Mahankali, Ele Ferrannini and Ralph A. DeFronzo J. Clin. Endocrinol. Metab. 2004 89: 3914-3921, doi: 10.1210/jc.2003-031941

To subscribe to Journal of Clinical Endocrinology & Metabolism or any of the other journals published by The Endocrine Society please go to: http://jcem.endojournals.org//subscriptions/

Copyright The Endocrine Society. All rights reserved. Print ISSN: 0021-972X. Online

0021-972X/04/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 89(8):3914 3921 Copyright 2004 by The Endocrine Society doi: 10.1210/jc.2003-031941

Separate Contribution of Diabetes, Total Fat Mass, and Fat Topography to Glucose Production, Gluconeogenesis, and Glycogenolysis
AMALIA GASTALDELLI, YOSHINORI MIYAZAKI, MAURA PETTITI, EMMA BUZZIGOLI, SRIKANTH MAHANKALI, ELE FERRANNINI, AND RALPH A. DEFRONZO
Metabolism Unit (A.G., M.P., E.B., E.F.), C.N.R. Institute of Clinical Physiology and Department of Internal Medicine, University of Pisa School of Medicine, 56100 Pisa, Italy; and Diabetes Division (A.G., Y.M., S.M., E.F., R.A.D.), University of Texas Health Science Center, San Antonio, Texas 78229-3900
The contribution of increased gluconeogenesis (GNG) to the excessive rate of endogenous glucose production (EGP) in type 2 diabetes (T2DM) is well established. However, the separate effects of obesity (total body fat), visceral adiposity, and T2DM have not been investigated. We measured GNG (by the 2 H2O technique) and EGP (with 3-3H-glucose) after an overnight fast in 44 type 2 diabetic and 29 gender/ethnic-matched controls. Subjects were classified as obese (body mass index 30 kg/m2 or greater) or nonobese (body mass index < 30 kg/m2); diabetic subjects were further subdivided according to the severity of fasting hyperglycemia [fasting plasma glucose (FPG) < 9 mM or > 9 mM]. EGP was similar in nondiabetic controls and T2DM with FPG less than 9 mM but was increased in T2DM with FPG > 9 mM (P < 0.001). Within the diabetic groups, obesity had an independent effect to further increase basal EGP (P < 0.01). In both nonobese diabetic groups, both the percent GNG and gluconeogenic flux were increased, compared with nonobese nondiabetic controls. In both diabetic groups, obesity further increased both percent GNG and gluconeogenic flux. In obese and nonobese T2DM, the increase in gluconeogenic flux was not accompanied by a reciprocal decrease in glycogenolysis, indicating a loss of hepatic autoregulation. By multivariate analysis, gluconeogenic flux was positively correlated with percent body fat, visceral fat, and the fasting plasma free fatty acid and glucose concentrations (all P < 0.02). We conclude that obesity per se, and visceral fat accumulation in particular, as well as poorly controlled diabetes are potent stimuli to augment gluconeogenic flux. (J Clin Endocrinol Metab 89: 3914 3921, 2004)

T IS WELL established that basal endogenous (primarily hepatic) glucose production (EGP) is increased in type 2 diabetes (T2DM) in proportion to the elevation in fasting plasma glucose concentration (FPG) (13). There is controversy, however, as to whether changes in gluconeogenesis (GNG) or glycogenolysis are responsible for the increase in EGP. We recently demonstrated that both total EGP and the fraction of plasma glucose derived from GNG were increased in both obesity and diabetes and contributed to their fasting hyperglycemia (3). Similar results have been reported by others (4), whereas Boden et al. (5) found that only T2DM patients with severe fasting hyperglycemia (10 mm) had an increased basal rate of EGP and GNG. However, in that study the T2DM patients were obese [average body mass index (BMI) 31.0 kg/m2], whereas the control group was nonobese (BMI 26.8 kg/m2). We have previously shown that basal GNG is increased in obese subjects with normal glucose tolerance (3), but total EGP is normal, most likely due to inhibition of glycogenolysis by the concomitant hyperinsuAbbreviations: BMI, Body mass index; C5, carbon 5; 2DM, type 2 diabetes; EGP, endogenous glucose production; FFA, free fatty acid; FFM, fat-free mass; FM, fat mass; FPG, fasting plasma glucose; GNG, gluconeogenesis; HbA1c, glycosylated hemoglobin; MRI, magnetic response imaging; Rd, glucose disappearance; T2DM, type 2 diabetes; VF, visceral fat. JCEM is published monthly by The Endocrine Society (http://www. endo-society.org), the foremost professional society serving the endocrine community.

linemia. This hepatic autoregulation (6) appears to be lost in T2DM subjects who, despite elevated basal plasma insulin levels, are unable to suppress appropriately glycogenolysis to maintain a normal basal rate of EGP (3). Thus, at present, the separate effects of obesity and T2DM on GNG and hepatic autoregulation remain unclear. A role for elevated plasma free fatty acid (FFA) levels has been implicated in the accelerated rate of GNG and total EGP in T2DM patients (7, 8). Increased FFA concentrations promote GNG in rat liver (9); similar observations have been made in humans (10, 11). In the postabsorptive state, circulating FFAs are derived from lipolysis, and the fat cell has been shown to be resistant to the antilipolytic effect of insulin in both T2DM and nondiabetic obese individuals (7, 12, 13). Because of these associations, one might expect that excess total body fat or altered fat distribution would be related to the accelerated rate of GNG and EGP in diabetic and obese individuals. With regard to this, increased visceral fat (VF) has been associated with resistance to the antilipolytic effect of insulin in adipocytes (14, 15), reduced rate of FFA reesterification (16), and decreased insulin-stimulated glucose disposal as measured by the euglycemic insulin clamp technique (17). It has been postulated that accelerated release of FFAs (18, 19) and/or other adipocytokines (20) by visceral adipocytes into the portal circulation can induce or augment hepatic insulin resistance and enhance GNG. The goal of this study was to examine the relationships among the rate of GNG, hepatic insulin resistance, and fat

3914

Gastaldelli et al. Obesity Is a Major Stimulus for Gluconeogenesis

J Clin Endocrinol Metab, August 2004, 89(8):3914 3921 3915

accumulation in normal glucose tolerant and T2DM individuals. Because visceral fat is related strongly to total adiposity (21), it is mandatory to account for influence of obesity when attempting to establish an independent role for visceral fat in glycemic control and the regulation of hepatic glucose metabolism. In the present study, we measured in the same individual total body fat content (by tracer), abdominal fat distribution [by magnetic response imaging (MRI)], EGP (by 3-3H-glucose infusion), and GNG (by the 2 H2O method) in a large group of diabetic and nondiabetic subjects spanning a wide range of adiposity.
Subjects and Methods Subjects
Forty-four T2DM patients (16 females, 28 males; 30 Mexican-Americans, 14 Caucasians) and 29 age-matched nondiabetic control subjects (12 females, 17 males; 16 Mexican-Americans, 13 Caucasians) participated in the study. Fifteen T2DM patients previously were studied (22). Based on World Health Organization criteria (23), subjects with a BMI greater than 30 kg/m2 were classified as obese. None of the diabetics had ever been treated with insulin or thiazolidinediones. For subjects who were taking metformin or sulfonylureas, the medication was stopped 2 wk before the study. Other than sulfonylureas or metformin, subjects were not taking any other drugs known to affect glucose tolerance. All control subjects had a normal 75-g oral glucose tolerance test (24). Body weight was stable in all subjects for at least 3 months before study. All studies were carried out at the Clinical Research Center of the University of Texas Health Science Center at San Antonio. The study protocol was approved by the Institutional Review Board of the University of Texas Health Science Center at San Antonio, and informed written consent was obtained from each subject before participation.

sample was obtained. The subjects were asked not to change their habitual dietary regimen, eat the last meal between 1800 and 1900 h on the night before study, and not eat or drink anything after the last meal. At 2200 h on the evening before study, all subjects drank 2H2O [5 g /kg of fat-free mass (FFM)] (Isotech, Williston, VT). A blood sample for the determination of baseline 2H2O enrichment was taken at 0800 h on the morning of the day before the study. Upon arrival, a polyethylene cannula was inserted into an antecubital vein for the infusion of all test substances. A second catheter was inserted retrogradely into an ipsilateral wrist vein on the dorsum of the hand for blood sampling, and the hand was kept in a heated box at 65 C. At 0700 h, blood was drawn for the determination of FPG. Then a primed (20 Ci FPG per 5 mmol/ liter) continuous (0.20 Ci/min) infusion of 3-3H-glucose (DuPont NEN Life Science Products, Boston, MA) was initiated and continued until the end of the study. During the last 30 min of the basal equilibration period (90 120 min after the start of 3-3H-glucose in nondiabetic subjects and 150 180 min for diabetic subjects), plasma samples were drawn at 5- to 10-min intervals for the determination of plasma glucose and insulin concentration and plasma tritiated glucose-specific activity. At the end of the basal equilibration period, a urine sample was obtained. After the basal equilibration period, insulin was administered as a prime-continuous (40 mU/m2min1) infusion for 120 min as previously described (27). The plasma glucose concentration was measured every 5 min after the start of the insulin infusion, and a variable infusion of 20% glucose was adjusted based on the negative feedback principle to maintain the plasma glucose level at 5 mm with a coefficient of variation less than 5%. In diabetic subjects the plasma glucose concentration was allowed to drop to 5 mm, at which level it was clamped. Plasma samples were collected every 15 min from 0 to 90 min and every 510 min from 90 to 120 min for the determination of plasma glucose and insulin concentrations and plasma tritiated glucose-specific activity. Plasma samples for the determination of deuterated glucose and water enrichment were taken before starting the tritiated glucose infusion and at the end of the basal equilibration period.

Study design
Within a 5- to 7-d interval, all subjects received: 1) measurement of lean body mass and fat mass (FM) using an iv bolus of 3H2O; 2) a euglycemic hyperinsulinemic clamp study in combination with 3-3Hglucose to measure basal EGP and hepatic and peripheral tissue sensitivity to insulin; 3) 2H2O given on the evening before the insulin clamp study to measure the contribution of GNG and glycogenolysis to EGP; and 4) in a subgroup of subjects (16 controls and 28 diabetics), quantitation of sc and intraabdominal visceral fat content at L4-L5 using nuclear MRI.

Analytical methods
Plasma glucose concentration was determined by the glucose oxidase method (Beckman II glucose analyzer, Fullerton, CA). Plasma insulin concentration was measured by RIA (Diagnostic Products Corp., Los Angeles, CA). Glycosylated hemoglobin (HbA1c) concentration was measured by affinity chromatography (biochemical methodology, Drower 4350; Isolab, Akron, OH). Plasma FFA concentration was measured spectrophotometrically (Wako Chemicals GmbH, Neuss, Germany). Plasma tritiated glucose-specific activity was determined on barium hydroxide/zinc sulfate deproteinized samples (Somogyis procedure). The pattern of 2H incorporation into plasma glucose after 2H2O ingestion was determined according to the method developed by Landau (28), as modified (22, 29). Briefly, the fraction of glucose produced via GNG from all precursors can be quantified from the ratio of 2H enrichment of carbon 5 (C5) to that of water. Because during glycogen breakdown there is no binding of hydrogen from body water to C5 of the glucose formed, enrichment at C5 in blood glucose vs. water reflects the fractional contribution of GNG to total EGP, i.e. from both phosphoenolpyruvate precursors and glycerol. Precision and accuracy of C5 have been reported previously (3). Water enrichment in the body water pool was monitored by reacting a sample of plasma or urine with calcium carbide (CaC2) to form acetylene (C2H2). The enrichment of acetylene was then determined by gas chromatography-mass spectrometry by monitoring peaks of mass 26 and 27 (30). All samples were run through the gas chromatography-mass spectrometry processing in duplicate or triplicate.

Lean and fat body mass

Subjects were admitted to the Clinical Research Center at 0800 h after a 10-h overnight fast. A catheter was placed into an antecubital vein and, after the withdrawal of a baseline blood sample, a 100-Ci iv bolus of 3 H2O was administered. Blood samples were drawn for the determination of plasma 3H2O radioactivity after 90, 105, and 120 min. Lean and fat body mass were calculated as described previously (25).

Abdominal fat distribution

Intraabdominal visceral and sc fat depots were measured in a subgroup of subjects by MRI, using imaging procedures that have been published previously (22, 26). Briefly, images were acquired on a 1.9 T Elscint Prestige MRI system (GE Medical System, Milwaukee WI). A sagittal localizing image was used to center transverse sections on the line through the space between L4 and L5, and the field of view was adjusted for body size to ensure a 2-mm pixel spacing. Signal averaging (four signals averaged) was used to reduce the effect of motion-related artifacts. Additionally, respiratory gating was used to combat motioninduced artifacts and reduce the blurring of fat boundaries in the anterior region of the abdomen.

Data analysis
Total body water was calculated from the mean plasma 3H2O radioactivity measured at 90, 105, and 120 min after the iv bolus of 3H2O. Plasma 3H2O-specific activity was calculated assuming that plasma water represents 93% of total plasma volume and FFM was calculated by dividing total body water by 0.73 (31). FM was calculated as the difference between body weight and FFM. Subcutaneous and VF areas were quantitated by magnetic resonance

Euglycemic hyperinsulinemic clamp

Subjects were admitted to the Clinical Research Center at 0700 h, after an approximately 13-h overnight fast, and a spontaneously voided urine

3916

J Clin Endocrinol Metab, August 2004, 89(8):3914 3921

Gastaldelli et al. Obesity Is a Major Stimulus for Gluconeogenesis

imaging at L4-L5 level as previously described (26). Briefly, images were processed using Alice software (Perceptive Systems Inc., Boulder, CO) to determine abdominal sc and intraabdominal visceral fat areas. The sc fat area was analyzed by selecting the outer and inner margins of sc adipose tissue as region of interest from the cross-sectional images and counting the number of pixels between the outer and inner margins of sc adipose tissue. The VF (intraabdominal) area was determined using histograms specific to the visceral regions. The histograms were summed over the range of pixel values designated as fat by fitting two normal analysis distribution curves to them. All glucose fluxes were expressed per FFM (in micromoles per minute per kilogramffm) because this has been shown to best account for differences due to sex, obesity, and age (32). During the last 30 min of the basal equilibration period, both plasma glucose concentration and 3-3Hglucose-specific activity were in steady-state in all subjects, and total EGP was calculated as the ratio of the 3-3H-glucose infusion rate to the plasma 3-3H-glucose-specific activity (mean of five determinations). During the euglycemic insulin clamp, non-steady-state conditions prevail and total glucose rate of appearance was calculated using Steeles equation. EGP during the last 30 min of the insulin clamp was obtained as the difference between rate of appearance and the exogenous glucose infusion rate. The insulin-stimulated rate of glucose disappearance (Rd) during the last 30 min of the insulin clamp was calculated from Steeles equation. The hepatic insulin resistance index was calculated as the product of EGP fasting plasma insulin. Over the range of insulin concentrations from 30 to 150 pmol/liter, i.e. those that exist under fasting conditions in T2DM and control subjects, the relationship between plasma insulin and EGP is linear (r 0.92, P 0.0001) (7). Data are given as the mean se. Proportions were compared by 2 analysis; comparison of mean group values was performed by two-way ANOVA, with obesity and diabetes as the factors. Partial correlation analyses were used to estimate associations among continuous variables in the whole data set.

not diabetes, was associated with an increase in plasma triglyceride concentration and decrease in plasma high-density lipoprotein-cholesterol concentration. The fasting plasma FFA concentration was increased modestly in both type 2 diabetic groups, and this increase was further enhanced by the presence of obesity (without reaching statistical significance). Fasting plasma insulin levels were significantly increased in diabetics and further increased by obesity in both diabetic and nondiabetic groups. With regard to body composition, FFM and VF area were increased in association with obesity alone, whereas percent FM and sc fat area were also independently increased in association with diabetes. FPG did not change significantly in nondiabetic subjects over the 2 h of tracer glucose infusion (5.4 0.1 to 5.3 0.1 mm, P ns). In contrast, in both diabetic groups, the fasting plasma glucose concentration decreased over the 3-h period of tritiated glucose infusion: from 8.7 0.3 to 7.4 0.3 mm (by 8%) in diabetics with FPG less than 9 mm and from 12.1 0.4 to 11.2 0.5 mm (by 15%) in diabetics with FPG 9 mm or more (both P 0.01).
Metabolic data (Table 2)

Results Anthropometric and metabolic data (Table 1)

In addition to subgrouping by obesity, individuals with T2DM were further classified according to severity of fasting hyperglycemia: FPG less than 9 mm (range 5.5 8.95 mm) or FPG 9 mm or more (range 9.114.4 mm). There were no significant differences between groups with respect to sex or ethnicity, but diabetics were older and heavier than nondiabetic subjects regardless of whether they were lean or obese. HbA1c and known duration of diabetes were not significantly different between obese and nonobese diabetics. Obesity, but
TABLE 1. Anthropometric and clinical characteristics
Control Nonobese Obese

In the basal state, EGP was similar in nondiabetic controls and diabetics with FPG less than 9 mm, whereas it was significantly increased in diabetics with FPG 9 mm or more. Within the diabetic groups, obesity had an independent effect to further increase basal EGP. The hepatic insulin resistance index increased progressively, and the basal glucose clearance decreased progressively, from groups of nondiabetic control to diabetic with FPG less than 9 mm to diabetic with FPG 9 mm or more (P 0.01 for both the effect of obesity and diabetes) (Fig. 1). In the nonobese, nondiabetic control subjects, the fraction of GNG derived from all precursors after an overnight fast averaged 57%, and was not influenced by age, sex, or ethnicity. In both nonobese diabetic groups, the percent GNG was increased, compared with the nonobese control group. Across all groups, obesity had a further significant effect to

T2DM-FPG (9 mM) Nonobese Obese

T2DM-FPG (9 mM) Nonobese Obese

No. Age (yr) HbA1c (%) Duration of diabetes (yr) FPG (mM) Fasting plasma insulin (pmol/liter) Fasting plasma FFA (mM) Triglycerides (mg/dl) HDL-cholesterol (mg/dl) LDL-cholesterol (mg/dl) Mean blood pressure (mm Hg) BMI (kg/m2) FFM (kg) FM (%) VF area (cm2) Subcutaneous area (cm2)

21 41 2 5.2 0.1 5.3 0.1 43 4 0.59 0.03 130 30 42 3 91 6 100 3 24.4 0.6 50 2 29 1 97 12 197 18

8 49 3 5.2 0.1 5.5 0.1 75 9 0.71 0.05 183 55 38 5 96 13 97 6 31.6 0.5 56 3 38 1 158 32 346 62

14 56 3 7.4 0.3 3.0 1.1 7.4 0.3 59 7 0.69 0.06 127 16 39 3 103 9 105 3 26.8 0.6 53 2 33 1 121 9 272 32

7 53 3 7.4 0.4 3.6 1.8 7.4 0.4 121 20 0.73 0.08 242 53 34 2 104 16 105 12 34.4 0.9 55 4 42 2 184 36 505 16

14 52 2 9.5 0.3 4.0 1.3 11.2 0.5 61 5 0.66 0.03 150 16 41 4 114 7 101 2 26.5 0.6 52 2 32 1 118 7 244 31

9 54 4 8.4 0.5 4.3 1.3 11.2 0.5 103 18 0.71 0.06 184 18 30 1 114 11 106 5 33.9 0.6 54 4 42 1 181 40 467 70

a a

ns
a a,b

ns
b b

ns
b a,b b a,b b a,b

HDL, High-density lipoprotein; LDL, low-density lipoprotein; ns, not significant. a P 0.05 or less for effect of diabetes and bP 0.05 or less for the effect of obesity by two-way ANOVA.

Gastaldelli et al. Obesity Is a Major Stimulus for Gluconeogenesis

J Clin Endocrinol Metab, August 2004, 89(8):3914 3921 3917

TABLE 2. Metabolic characteristics

Control Nonobese Obese T2DM-FPG (9 mM) Nonobese Obese T2DM-FPG (9 mM) Nonobese Obese P
a,b b a,b

Basal EGP (mol/minkgffm) Percent GNG (%) GNG flux (mol/minkgffm) GLY flux (mol/minkgffm) Clamp plasma insulin (pmol/liter) Clamp plasma FFA (mM) Rd (mol/minkgffm) Clamp EGP (mol/minkgffm)
a

16.3 0.3 56.5 1.6 9.1 0.3 7.0 0.3 299 18 0.16 0.02 31.7 3.1 4.5 1.0
b

17.7 0.3 66.2 1.8 11.7 0.5 6.0 0.3 368 31 0.29 0.04 31.8 4.5 4.8 0.8

15.5 0.6 64.4 3.0 10.0 0.6 5.5 0.5 362 27 0.22 0.02 20.3 2.2 5.5 0.9

17.1 0.8 72.5 4.6 12.2 0.6 4.9 1.0 458 47 0.31 0.03 16.1 0.7 7.1 0.7

17.7 0.8 63.9 4.3 11.3 0.9 6.4 0.8 364 24 0.20 0.01 17.3 0.7 8.4 1.0

19.2 0.8 69.0 3.7 13.3 0.9 5.9 0.6 446 50 0.26 0.02 15.9 1.1 8.8 1.4

ns
a,b b a,b a

P 0.05 or less for effect of diabetes and

P 0.05 or less for the effect of obesity by two-way ANOVA; ns, not significant.

FIG. 2. Relationship between the fasting plasma glucose concentration and gluconeogenic and glycogenolytic flux in obese (dashed lines) and nonobese (solid lines) individuals. See text for statistical analysis.

nonobese individuals. Clamp plasma FFA levels were less suppressed in obese than in nonobese subjects, the effect of diabetes per se not reaching statistical significance (Table 2).
Correlations

FIG. 1. Hepatic insulin resistance index (EGP FPI) (top) and basal glucose clearance (bottom) in nonobese (BMI 30 kg/m2) and obese (BMI 30 kg/m2) individuals. Nondiabetic control subjects are shown by the open bars. Diabetic subjects with mild or severe fasting hyperglycemia are represented by the cross-hatched and solid bars, respectively. See text for statistical analysis.

increase the percent GNG. Gluconeogenic flux was markedly increased by obesity and diabetes independently of one another (Fig. 2); the bivariate model predicted an increase in flux of 2.6 0.6 mol/min1kgffm1 in obese subjects (P 0.0001), one of 0.7 0.7 mol/min1kgffm1 in diabetics with FPG 9 mm or less (P ns), and one of 2.2 0.7 in diabetics with FPG greater than 9 mm (P 0.001 vs. nonobese, nondiabetic controls). Glycogenolytic flux was reduced slightly, although not significantly, in nonobese and obese type 2 diabetics with FPG less than 9 mm, compared with the other groups. Insulin-mediated Rd was significantly decreased in association with both obesity and diabetes (Table 2). EGP was suppressed by 75 4% in the control group, 61 5% in diabetics with FPG less than 9 mm, and 55 4% in diabetics with FPG 9 mm or more (P 0.005) similarly in obese and

In the whole data set, FPG was strongly and positively correlated with both the rates of EGP (r 0.42 and P 0.0001) and GNG (Fig. 3), whereas there was no significant relationship between FPG and glycogenolysis. None of the measured glucose parameters (EGP, percent GNG, GNG flux, glycogenolytic flux, hepatic insulin resistance index) was significantly associated with the plasma insulin concentration. Both percent GNG and gluconeogenic flux were strongly and positively related to the plasma FFA concentration (partial r 0.36, P 0.003 and partial r 0.31, P 0.01, respectively, after adjusting for sex, age, ethnicity, and BMI), whereas glycogenolysis was related to the plasma FFA concentration inversely (partial r 0.31, P 0.02). This result explained why no correlation was observed between plasma FFAs and EGP. Neither EGP nor glycogenolysis was related to VF or sc fat. In contrast, GNG, expressed as both the fractional contribution to GNG and total flux, was significantly and positively related to VF (but not sc fat), even after adjusting for percent FM (partial r 0.30 and P 0.05 for both). In a multivariate regression model, we correlated glucose fluxes with fasting plasma glucose, FFAs, and percent body fat, accounting also for age, gender, and ethnicity. GNG flux was independently related to percent body fat (partial r 0.30, P 0.01), fasting

3918

J Clin Endocrinol Metab, August 2004, 89(8):3914 3921

Gastaldelli et al. Obesity Is a Major Stimulus for Gluconeogenesis

FIG. 3. Relationship between gluconeogenic flux and FPG concentration, fasting plasma FFA concentration, percent body fat, and VF in nondiabetic control subjects (solid circles) and diabetic subjects (open circles).

plasma glucose (partial r 0.33, P 0.006), and FFA concentrations (partial r 0.29, P 0.02); in this model, glycogenolysis was negatively related only to plasma FFA concentration. The same relationships held by replacing percent body fat with VF area.
Discussion

Elevated basal EGP (2, 33) and impaired suppression of EGP by physiologic hyperinsulinemia (33, 34) are characteristic features of T2DM. The increase in EGP is the major determinant of fasting hyperglycemia, at least in poorly controlled diabetic patients (2, 33). The fraction of plasma glucose that is derived from GNG in T2DM patients has been reported to be increased (3, 4, 35, 36) or normal (5). However, in these previous studies, the number of subjects studied was small, and glycemic control and degree of obesity were not separately taken into account. Although Boden et al. (5)

found a difference in GNG flux between mildly and severely hyperglycemic patients, the diabetic and control subjects were not matched for obesity (seven controls with BMI of 26.8 kg/m2 vs. 14 T2DM with FPG 10 mm with BMI of 29.2 kg/m2 and 13 T2DM with FPG 10 mm with BMI of 32.9 kg/m2). Mismatch for adiposity is important because we have shown an independent effect of obesity on GNG when expressed both as percent contribution as well as total flux in normoglycemic subjects (3) and an effect of VF accumulation in T2DM subjects (22). Whereas the preponderance of evidence supports enhanced GNG as the cause of increased basal EGP in human T2DM (3, 4), there is some evidence to suggest that glycogenolysis also may be increased (36, 37). Most importantly, the separate impact of factors such as degree of obesity and fat distribution on GNG has not been examined. In the present study, we found that the fraction of EGP that is

Gastaldelli et al. Obesity Is a Major Stimulus for Gluconeogenesis

J Clin Endocrinol Metab, August 2004, 89(8):3914 3921 3919

derived from GNG was increased in obese patients at each degree of fasting hyperglycemia, whereas glycogenolytic flux tended to be decreased (although not significantly) in T2DM patients with mild fasting hyperglycemia and was unchanged in severely hyperglycemic diabetic subjects (Fig. 2). It is especially noteworthy that in our data a difference in BMI of only approximately 4 kg/m2 was associated with an increment in GNG flux similar to that associated with an increase in FPG of 6 mmol/liter. These results emphasize the quantitative effect of obesity to augment GNG flux, namely the impact of hyperglycemia, and the need to closely match diabetic and control subjects for adiposity when evaluating GNG. In addition to fasting plasma glucose and FFA levels, increased GNG flux was simultaneously related to percent body fat and VF area (independently of age, gender, BMI, and ethnicity). In nondiabetic subjects an increase in GNG flux is counterbalanced by a reduction in glycogenolysis and total EGP remains unchanged (6). In mildly diabetic subjects, there was a tendency (although not significant) for glycogenolysis to decrease as GNG increased (Fig. 2). However, in severely diabetic individuals, hepatic autoregulation clearly was lost and the marked increase in GNG flux was not associated with any compensatory decrease in glycogenolysis. Thus, these findings support the results of Boden et al. (10), who failed to observe the normal compensatory glycogenolytic response to stimulation (with lipid infusion) or inhibition (with nicotinic acid) of GNG in diabetic subjects. The hepatic insulin resistance index was influenced by both obesity and diabetes (i.e. severity of fasting hyperglycemia). Thus, for any level of glycemia, obesity approximately doubled the hepatic insulin resistance index, whereas, for any given level of obesity, diabetes increased the hepatic insulin resistance index by approximately 50% (Fig. 1). It is noteworthy that in both nonobese and obese diabetic individuals with severe fasting hyperglycemia (FPG 11.2 0.5 mmol/liter), the marked increase in both EGP and FPG, compared with the diabetic group with mild fasting hyperglycemia (FPG 7.4 0.3 mmol/liter and normal basal EGP), was associated with a failure of the fasting plasma insulin concentration to increase further to offset the worsening hepatic insulin resistance (Table 1). Moreover, during the clamp EGP was progressively less suppressed moving from control to diabetics with FPG less than 9 mm to diabetics with FPG 9 mm or greater (P 0.005) (Table 2). Consistent with previous results from our laboratory (33), these observations stress the role that -cell incompetence plays in determining when EGP begins to increase in absolute terms and how the liver is sensitive to the effect of insulin. In both nonobese and obese type 2 diabetic subjects, progressive increases in FPG were associated with a progressive decline in the basal glucose clearance (Fig. 1). This relation was observed, even after changes in the fasting plasma insulin concentration were taken into account. Likewise, Rd was reciprocally related to both obesity and diabetes. These observations confirm the inhibitory effect of hyperglycemia and obesity on glucose use and are consistent with previous publications from several laboratories (1, 39, 40).

Fat distribution and GNG

A significant body of evidence has accumulated to indicate that visceral, as contrasted with sc, adiposity is associated with hepatic and peripheral insulin resistance (41). Several hypotheses have been proposed to explain this association. One hypothesis proposes that increased FFA flux into the portal vein renders the liver resistant to the restraining effect of insulin on EGP (42). This hypothesis is based on the fact that visceral adipocytes are more resistant to the antilipolytic effect of insulin (18) than are sc fat cells. Several lines of evidence (10, 18, 20, 43, 44) support the concept that FFAs are an important regulator of EGP. In vitro studies have demonstrated that plasma FFAs increase the activity of pyruvate carboxylase and phosphoenolpyruvate carboxylase, the ratelimiting enzymes for GNG (9, 45, 46), and augment the activity of glucose-6-phosphatase, the enzyme that ultimately controls the release of glucose by the liver (47). In contrast, a decrease in plasma FFAs has been shown to inhibit glucose6-phosphatase, leading to a reduction in GNG, an effect also seen in man (48). In normal subjects, an increase in plasma FFA concentration stimulates GNG (10, 49, 50), whereas a decrease reduces it (10, 49, 50). With regard to the latter, it has been shown that a significant portion of the suppressive effect of insulin on hepatic glucose production is mediated via inhibition of lipolysis and a reduction in circulating FFA concentrations (44, 51, 52). Moreover, FFA infusion in normal humans under conditions that simulate the diabetic state (43) and in obese insulin-resistant subjects (53) enhances EGP, most likely secondarily to stimulation of GNG. In the present study, we found a strong linear relationship between GNG and the plasma FFA concentration. A strong correlation also was found between VF and both percent GNG and absolute gluconeogenic flux. These results are consistent with the possibility that increased release of FFAs by VF into the portal circulation may contribute to the stimulation of hepatic GNG. However, whether this stimulation results in overproduction of glucose by the liver (VF was not correlated with EGP) depends on the concomitant adjustment (or lack thereof) of the glycogenolytic rate. Because GNG is less sensitive to insulin inhibition than is glycogenolysis (34, 54), the increased fasting plasma insulin concentration in mildly hyperglycemic diabetic patients (Table 2) tends to down-regulate glycogenolysis, thereby maintaining EGP within normal limits. In the more hyperglycemic subjects, the ambient plasma insulin concentration is insufficient to restrain EGP, which consequently rises to levels that are elevated in absolute terms (Table 2). Thus, the combination of visceral adipose mass/visceral lipolytic activity, -cell incompetence, and insulin resistance could be viewed as a regulatory axis that controls EGP by independently regulating the individual components of EGP, GNG and glycogenolysis. The contribution of VF cell-derived adipocytokines to this visceral-portal-hepatic axis remains to be defined. This is in agreement with a recent paper (55), which observed that an increase in FM, primarily in the abdominal area, was associated with hepatic insulin resistance and inhibition of suppression of EGP during the clamp. In conclusion, in diabetic subjects obesity per se and VF accumulation in particular induce hepatic insulin resistance

3920

J Clin Endocrinol Metab, August 2004, 89(8):3914 3921

Gastaldelli et al. Obesity Is a Major Stimulus for Gluconeogenesis

and are potent stimuli to augment gluconeogenic flux. Poorly controlled diabetes and abdominal obesity are independent determinants of severe fasting hyperglycemia.
Acknowledgments
The authors thank Magda Ortiz, Dianne Frantz, Socorro Mejorado, Janet Shapiro, John Kincaid, John King, Norma Diaz, and Patricia Wolf for their assistance in performing the insulin clamp studies and S. Frascerra, Ph.D.; S. Baldi, Ph.D.; D. Ciociaro; and N. Pecori for their technical assistance in the measurement of GNG. Received December 16, 2003. Accepted May 6, 2004. Address all correspondence and requests for reprints to: Ralph A. DeFronzo, M.D., Diabetes Division, University of Texas Health Science Center, 7703 Floyd Curl Drive MS 7886, San Antonio, Texas 78229-3900. E-mail: albarado@uthscsa.edu. This work was supported by National Institutes of Health Grant DK-24092, General Clinical Research Center Grant M01-RR-01346, a Veterans Affairs Merit Award, and funds from the Veterans Administration Medical Research Service.

18. 19. 20. 21. 22. 23. 24. 25.

26. 27. 28. 29.

References
1. DeFronzo RA, Ferrannini E, Simonson DC 1989 Fasting hyperglycemia in non-insulin-dependent diabetes mellitus: contributions of excessive hepatic glucose production and impaired tissue glucose uptake. Metabolism 38:387 395 2. Jeng C, Sheu W, Fuh M, Chen Y, Reaven G 1994 Relationship between hepatic glucose production and fasting plasma glucose concentration in patients with NIDDM. Diabetes 43:1440 1444 3. Gastaldelli A, Baldi S, Pettiti M, Toschi E, Camastra S, Natali A, Landau BR, Ferrannini E 2000 Influence of obesity and type 2 diabetes on gluconeogenesis and glucose output in humans: a quantitative study. Diabetes 49:13671373 4. Magnusson I, Rothman D, Katz L, Shulman R, Shulman G 1992 Increased rate of gluconeogenesis in type II diabetes mellitus. A 13C nuclear magnetic resonance study. J Clin Invest 90:13231327 5. Boden G, Chen X, Stein TP 2001 Gluconeogenesis in moderately and severely hyperglycemic patients with type 2 diabetes mellitus. Am J Physiol Endocrinol Metab 280:E23E30 6. Moore MC, Connolly CC, Cherrington AD 1998 Autoregulation of hepatic glucose production. Eur J Endocrinol 138:240 248 7. Groop LC, Bonadonna RC, DelPrato S, Ratheiser K, Zyck K, Ferrannini E, DeFronzo RA 1989 Glucose and free fatty acid metabolism in non-insulindependent diabetes mellitus. Evidence for multiple sites of insulin resistance. J Clin Invest 84:205213 8. Boden G, Shulman GI 2002 Free fatty acids in obesity and type 2 diabetes: defining their role in the development of insulin resistance and -cell dysfunction. Eur J Clin Invest 32:14 23 9. Williamson J, Kreisberg R, Felts P 1966 Mechanism for the stimulation of gluconeogenesis by fatty acids in perfused rat liver. Proc Natl Acad Sci USA 56:247254 10. Boden G, Chen X, Capulong E, Mozzoli M 2001 Effects of free fatty acids on gluconeogenesis and autoregulation of glucose production in type 2 diabetes. Diabetes 50:810 816 11. Staehr P, Hother-Nielsen O, Landau BR, Chandramouli V, Holst JJ, BeckNielsen H 2003 Effects of free fatty acids per se on glucose production, gluconeogenesis, and glycogenolysis. Diabetes 52:260 267 12. Groop LC, Saloranta C, Shank M, Bonadonna RC, Ferrannini E, DeFronzo RA 1991 The role of free fatty acid metabolism in the pathogenesis of insulin resistance in obesity and noninsulin-dependent diabetes mellitus. J Clin Endocrinol Metab 72:96 107 13. Miles JM, Wooldridge D, Grellner WJ, Windsor S, Isley WL, Klein S, Harris WS 2003 Nocturnal and postprandial free fatty acid kinetics in normal and type 2 diabetic subjects: effects of insulin sensitization therapy. Diabetes 52:675 681 14. Meek SE, Nair KS, Jensen MD 1999 Insulin regulation of regional free fatty acid metabolism. Diabetes 48:10 14 15. Albu JB, Curi M, Shur M, Murphy L, Matthews DE, Pi-Sunyer FX 1999 Systemic resistance to the antilipolytic effect of insulin in black and white women with visceral obesity. Am J Physiol 277:E551E560 16. Zierath JR, Livingston JN, Thorne A, Bolinder J, Reynisdottir S, Lonnqvist F, Arner P 1998 Regional difference in insulin inhibition of non-esterified fatty acid release from human adipocytes: relation to insulin receptor phosphorylation and intracellular signalling through the insulin receptor substrate-1 pathway. Diabetologia 41:13431354 17. Brochu M, Starling R, Tchernof A, Matthews D, Garcia-Rubi E, Poehlman

30. 31. 32.

33. 34.

35.

36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46.

E 2000 Visceral adipose tissue is an independent correlate of glucose disposal in older obese postmenopausal women. J Clin Endocrinol Metab 85:2378 2384 Mittelman SD, Van Citters GW, Kirkman EL, Bergman RN 2002 Extreme insulin resistance of the central adipose depot in vivo. Diabetes 51:755761 Boden G, Cheung P, Stein TP, Kresge K, Mozzoli M 2002 FFA cause hepatic insulin resistance by inhibiting insulin suppression of glycogenolysis. Am J Physiol Endocrinol Metab 283:E12E19 Lam TK, Carpentier A, Lewis GF, van de Werve G, Fantus IG, Giacca A 2003 Mechanisms of the free fatty acid-induced increase in hepatic glucose production. Am J Physiol Endocrinol Metab 284:E863E873 Wajchenberg BL 2000 Subcutaneous and visceral adipose tissue: their relation to the metabolic syndrome. Endocr Rev 21:697738 Gastaldelli A, Miyazaki Y, Pettiti M, Matsuda M, Mahankali S, Santini E, DeFronzo RA, Ferrannini E 2002 Metabolic effects of visceral fat accumulation in type 2 diabetes. J Clin Endocrinol Metab 87:5098 5103 World Health Organization 1997 Obesity: preventing and managing the global epidemic of obesity. Report of the WHO Consultation of Obesity. Geneva: World Health Organization 1997 Report of the expert committee on the diagnosis and classification of diabetes mellitus. Diabetes Care 20:11831197 Bonora E, Del Prato S, Bonadonna R, Gulli G, Solini A, Shank M, Ghiatas A, Lancaster J, Kilcoyne R, Alyassin A, DeFronzo R 1992 Total body fat content and fat topography are associated differently with in vivo glucose metabolism in nonobese and obese nondiabetic women. Diabetes 41:11511159 Lancaster J, Ghiatas A, Alyassin A, Kilcoyne R, Bonora E, DeFronzo R 1991 Measurement of abdominal fat with T1-weighted MR images. J Magn Reson Imaging 1:363369 De Fronzo R, Tobin D, Andres R 1979 Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 237:E214 E223 Landau B, Wahren J, Chandramouli V, Schumann W, Ekberg K, Kalhan S 1996 Contributions of gluconeogenesis to glucose production in the fasted state. J Clin Invest 98:378 385 Schumann WC, Gastaldelli A, Chandramouli V, Previs SF, Pettiti M, Ferrannini E, Landau BR 2001 Determination of the enrichment of the hydrogen bound to carbon 5 of glucose on 2H2O administration. Anal Biochem 297: 195197 Previs S, Hazey J, Diraison F, Beylot M, David F, Brunengraber H 1996 Assay of the deuterium enrichment of water via acetylene. J Mass Spectrom 31:389 391 Hume R, Weyers E 1971 Relationship between total body water and surface area in normal and obese subjects. J Clin Pathol 24:234 238 Natali A, Toschi E, Camastra S, Gastaldelli A, Groop L, Ferrannini E 2000 Determinants of postabsorptive endogenous glucose output in non-diabetic subjects. European Group for the Study of Insulin Resistance (EGIR). Diabetologia 43:1266 1272 DeFronzo RA, Simonson D, Ferrannini E 1982 Hepatic and peripheral insulin resistance: a common feature of type 2 (non-insulin-dependent) and type 1 (insulin-dependent) diabetes mellitus. Diabetologia 23:313319 Gastaldelli A, Toschi E, Pettiti M, Frascerra S, Quinones-Galvan A, Sironi AM, Natali A, Ferrannini E 2001 Effect of physiological hyperinsulinemia on gluconeogenesis in nondiabetic subjects and in type 2 diabetic patients. Diabetes 50:18071812 Perriello G, Pampanelli S, Del Sindaco P, Lalli C, Ciofetta M, Volpi E, Santeusanio F, Brunetti P, Bolli GB 1997 Evidence of increased systemic glucose production and gluconeogenesis in an early stage of NIDDM. Diabetes 46:1010 1016 Consoli A, Nurjhan N, Capani F, Gerich J 1989 Predominant role of gluconeogenesis in increased hepatic glucose production in NIDDM. Diabetes 38:550 557 Cusi K, Consoli A, DeFronzo RA 1996 Metabolic effects of metformin on glucose and lactate metabolism in noninsulin-dependent diabetes mellitus. J Clin Endocrinol Metab 81:4059 4067 Deleted in proof. Hother-Nielsen O, Beck-Nielsen H 1991 Insulin resistance, but normal basal rates of glucose production in patients with newly diagnosed mild diabetes mellitus. Acta Endocrinol (Copenh) 124:637 645 Paquot N, Scheen AJ, Dirlewanger M, Lefebvre PJ, Tappy L 2002 Hepatic insulin resistance in obese non-diabetic subjects and in type 2 diabetic patients. Obes Res 10:129 134 Evans DJ, Hoffmann RG, Kalkhoff RK, Kissebah AH 1984 Relationship of body fat topography to insulin sensitivity and metabolic profiles in premenopausal women. Metabolism 33:68 75 Guo Z, Hensrud DD, Johnson CM, Jensen MD 1999 Regional postprandial fatty acid metabolism in different obesity phenotypes. Diabetes 48:1586 1592 Ferrannini E, Barrett E, Bevilacqua JS, DeFronzo RA 1983 Effects of fatty acids on glucose production and utilization in man. J Clin Invest 72:17371747 Cherrington A 1999 Banting Lecture. Control of glucose uptake and release by the liver in vivo. Diabetes 48:1198 1214 Bahl J, Matsuda M, DeFronzo R, Bressler R 1997 In vitro and in vivo suppression of gluconeogenesis by inhibition of pyruvate carboxylase. Biochem Pharmacol 53:6774 Exton J, Corbin J, Park C 1969 Control of gluconeogenesis in liver. IV. Dif-

Gastaldelli et al. Obesity Is a Major Stimulus for Gluconeogenesis

J Clin Endocrinol Metab, August 2004, 89(8):3914 3921 3921

47. 48. 49. 50.

51.

ferential effects of fatty acids and glucagons on ketogenesis and gluconeogenesis in the perfused rat liver. J Biol Chem 244:4095 4102 Massillon D, Barzailai N, Hawkins M, Prus-Wetheimer D, Rossetti L 1997 Induction of hepatic glucose-6-phosphatase gene expression by lipid infusion. Diabetes 46:153157 Puhakainen I, Yki-Jarvinen H 1993 Inhibition of lipolysis decreases lipid oxidation and gluconeogenesis from lactate but not fasting hyperglycemia or total hepatic glucose production in NIDDM. Diabetes 42:1694 1699 Chen X, Iqbal N, Boden G 1999 The effects of free fatty acids on gluconeogenesis and glycogenolysis in normal subjects. J Clin Invest 103:365372 Roden M, Stingl H, Chandramouli V, Schumann WC, Hofer A, Landau BR, Nowotny P, Waldhausl W, Shulman GI 2000 Effects of free fatty acid elevation on postabsorptive endogenous glucose production and gluconeogenesis in humans. Diabetes 49:701707 Rebrin K, Steil G, Mittelman S, Bergman R 1996 Causal linkage between

52. 53. 54.

55.

insulin suppression of lipolysis and suppression of liver glucose output in dogs. J Clin Invest 98:741749 Mittelman S, Bergman R 2000 Inhibition of lipolysis causes suppression of endogenous glucose production independent of changes in insulin. Am J Physiol Endocrinol Metab 279:E630 E637 Bevilacqua S, Bonadonna R, Buzzigoli G, Boni C, Ciociaro D, Maccari P, Giorico M, Ferrannini E 1987 Acute elevation of free fatty acid levels leads to hepatic insulin resistance in obese subjects. Metabolism 36:502506 Edgerton D, Cardin S, Emshwiller M, Neal D, Chandramouli V, Schumann W, Landau B, Rossetti L, Cherrington A 2001 Small increases in insulin inhibit hepatic glucose production solely caused by an effect on glycogen metabolism. Diabetes 50:18721882 Kim SP, Ellmerer M, Van Citters GW, Bergman RN 2003 Primacy of hepatic insulin resistance in the development of the metabolic syndrome induced by an isocaloric moderate-fat diet in the dog. Diabetes 52:24532460

JCEM is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.