Clinical Science and Molecular Medicine (1915) 49, 21-32.
Glucose metabolites in blood and adipose tissue of obese diabetic and non-diabetic subjects
M. A. PAGE
AND
D. J. GALTON
Diabetes Research Laboratory, Department of Medicine, St Bartholomew’s Hospital, London
(Received 30 Augurt 1974)
in pathological conditions such as diabetes the rate of synthesis varies (Galton & Wilson, 1970; Galton, Wilson & Kissebah, 1971), and such alterations may be reflected in the intracellular concentration of metabolites. Measurements of glycolytic intermediates have been made in rat epididymal fat-pads (Denton, Yorke & Randle, 1968; Saggerson & Greenbaum, 1970) but few studies have been made of the levels of intermediates in human adipose tissue. In this study, glucose 6-phosphate, fructose 6phosphate, fructose diphosphate, glycerol phosphate and uridine diphosphate glucose have been measured in human adipose tissue and in whole blood which, unlike adipose tissue, is insensitive to the action of insulin on glucose transport. The effect of fasting, diabetes and an oral glucose load on these metabolites were studied to see if these states are associated with alterations in the concentration of glycolytic intermediates in blood or adipose tissue.
Summary 1. Glucose 6-phosphate, fructose 6-phosphate, fructose diphosphate, glycerol phosphate and uridine diphosphate glucose have been measured in human adipose tissue and blood from obese subjects under fed and fasting conditions and in obese diabetic and non-diabetic subjects before and after an oral glucose load (100 g). 2. Adipose tissue metabolites expressed as nmollg wet weight correlated inversely with adipocyte diameter. 3. After fasting, fructose diphosphate and glycerol phosphate in adipose tissue decreased significantly. 4. The basal concentrations of metabolites in blood and adipose tissue were maintained at similar concentrations in diabetic and non-diabetic subjects despite very different blood glucose concentrations. 5. The significant increase in adipose tissue glucose 6-phosphate after the glucose load seen in the non-diabetic but not in the diabetic subjects suggests that glucose uptake is decreased in the diabetic adipocyte.
Materials and methods Four categories of patients were studied (Table 1). None of these patients was receiving dietary or drug therapy for diabetes. The purpose of the study was explained to the patients and each patient consented to the procedures carried out. Samples of subcutaneous adipose tissue were obtained from the abdominal wall of obese subjects by needle biopsy (Diengott & Kerpel, 1967) under local anaesthesia (1 ml of 1% lignocaine). Biopsies were obtained after an overnight fast of 10-14 h. The core of adipose tissue (200-600 mg) was immediately
Key words : adipocytes, diabetes, erythrocytes, fasting, glucose, obesity.
Introduction Triglyceride glycerol is readily synthesized from glucose by human adipose tissue. However, under different physiological conditions such as fasting and Correspondence: Dr D. J. Galton, Diabetes Research Laboratory, Department of Medicine, St Bartholomew’s Hospital, London, E.C.I.
27
M. A . Page and D. J. Galton
28
TABLE 1. Clinical details of patienfs in the study Data are mean values+ SEM. NO.
Patient group 20 Non-diabetic obese 8 Non-diabetic, fasted for 7-10 days Non-diabetic, obese, for glucose tolerance test 1 1 9 Diabetic obese, for glucose tolerance test
Age (years)
Height (cm)
Weight (kg)
% Ideal weight
39.0f3.8 44.5k4.9 34.0k3.7 56.1k4.8
162.0f 17.0 164.0+ 1.6 1645k 2.3 164*4+_3.9
107.0k4.9 108.6f4.5 90.3k3.8 79.2k3.3
183.0f 8.0 195.0+ 9.6 160.5f 7.0 137.0k 11.0
dropped into liquid nitrogen, except for a small portion which was kept for estimation of cell size. The frozen tissue was divided into 100-200 mg portions. Each piece was weighed before being homogenized in ice-cold perchloric acid (500 mmol/ 1) and light petroleum (boiling point 4O-6O0C), with a power-driven Teflon pestle in a glass mortar. The glass mortar was rinsed twice in cold perchloric acid (500 mmol/l) and light petroleum. The total volume of the perchloric acid phase was 2 ml and the total volume of light petroleum used was 8 ml. The cold homogenate was centrifuged for 5 min at 4000 g and the aqueous and light petroleum phases were separated. The light petroleum phase was evaporated and triglyceride in the residue was estimated (Cramp & Robertson, 1968). The aqueous layer was decanted from the precipitated protein and neutralized with KOH (1.8 mol/l). After cooling on ice the excess of perchlorate was removed by centrifugation and the supernatants were stored at -20°C for up to 48 h before assay. The concentration of themetabolites did not decrease during this storage period. Immediately after venepuncture 5 ml of blood was added to 10 ml of ice-cold perchloric acid (500 mmol/l) in a weighed bottle. The deproteinized blood was spun for 20 min at 6000 g at 4°C and the clear supernatant decanted and neutralized as described for adipose tissue. The results are presented for whole blood since it was considered that centrifugation before deproteinization may distort the concentrations of glycolytic intermediates in erythrocytes. Previous studies (Bosia, Pescarmona & Arese, 1971 ; Minakami, Suziki, Saito & Yoshikawa, 1965) have also ignored the contribution of leucocytes and plasma to erythrocyte metabolites. Metabolite concentrations were determined with an Aminco-Bowman Ratio Spectrofluorimeter by measuring the increase in fluorescence due to NADH and NADPH at excitation wavelength 340 nm and emission wavelength 460 nm. Metabolites were
weight
Adipocyte diameter (Pm) 110f3.9 116f48 109f4.3 112f5.8
assayed in duplicate on consecutive days and blank samples were run during each assay to correct for machine drift and any change in fluorescence due to the addition of enzyme. Standards for each intermediate were included in all assays. The concentrations of each standard were determined with a Hitachi 101 spectrophotometer. Glucose 6-phosphate, fructose 6-phosphate and fructose 1,6diphosphate were measured in the same cuvette, in 0.5 ml of tissue extract in a final volume of 1 ml, by the method of Racker (1965). Glycerol phosphate was determined with 0.2 ml of extract by the method of Hohorst (1965). Uridine diphosphate glucose was assayed in a final volume of 1 ml of glycine buffer (50 mmol/l, pH 8.7), containing NAD (2 mmol/l), tissue extract (0.2 ml) and uridine diphosphate glucose dehydrogenase (5 mg). Blood glucose was measured by an automated glucose oxidase method; plasma insulin was assayed by solid-phase radioimmunoassay. Determination of fat-cell diameter
Isolated fat-cells were prepared by the method of Rodbell (1964) and fat-cell diameters were determined by a direct microscopic method (di Girolamo, Mendlinger & Fertig, 1971). The adipose cell volume was calculated from the mean of the diameters of at least 300 cells, assuming that the cells were spherical (Goldrick, 1967). Under these conditions the diameter of cells liberated by collagenase correlates closely to the diameter of cells of intact tissue slices (r = 0.964, n = 11; Smith, Sjostrom & Bjorntorp, 1972). Statistics
Significances of differences were tested by use of Student’s paired t-test. Body weight (% of ideal) was calculated from tables derived by the Society of Actuaries, Chicago (1959). Linear correlation
Metabolites in blood and adipose tissue
29
TABLE 2. Concentrations of metabolites in human blood and adipose tissue from obese subjects The results are mean values* SEM and the number of subjects is shown in parentheses. Details of patients are given in the Materials and methods section. Adipose tissue was obtained before and after a fast from the same patients. Significance of differences by Student’s paired t-test: * P
Glucose metabolites in blood and adipose tissue of obese diabetic and non-diabetic subjects
M. A. PAGE
AND
D. J. GALTON
Diabetes Research Laboratory, Department of Medicine, St Bartholomew’s Hospital, London
(Received 30 Augurt 1974)
in pathological conditions such as diabetes the rate of synthesis varies (Galton & Wilson, 1970; Galton, Wilson & Kissebah, 1971), and such alterations may be reflected in the intracellular concentration of metabolites. Measurements of glycolytic intermediates have been made in rat epididymal fat-pads (Denton, Yorke & Randle, 1968; Saggerson & Greenbaum, 1970) but few studies have been made of the levels of intermediates in human adipose tissue. In this study, glucose 6-phosphate, fructose 6phosphate, fructose diphosphate, glycerol phosphate and uridine diphosphate glucose have been measured in human adipose tissue and in whole blood which, unlike adipose tissue, is insensitive to the action of insulin on glucose transport. The effect of fasting, diabetes and an oral glucose load on these metabolites were studied to see if these states are associated with alterations in the concentration of glycolytic intermediates in blood or adipose tissue.
Summary 1. Glucose 6-phosphate, fructose 6-phosphate, fructose diphosphate, glycerol phosphate and uridine diphosphate glucose have been measured in human adipose tissue and blood from obese subjects under fed and fasting conditions and in obese diabetic and non-diabetic subjects before and after an oral glucose load (100 g). 2. Adipose tissue metabolites expressed as nmollg wet weight correlated inversely with adipocyte diameter. 3. After fasting, fructose diphosphate and glycerol phosphate in adipose tissue decreased significantly. 4. The basal concentrations of metabolites in blood and adipose tissue were maintained at similar concentrations in diabetic and non-diabetic subjects despite very different blood glucose concentrations. 5. The significant increase in adipose tissue glucose 6-phosphate after the glucose load seen in the non-diabetic but not in the diabetic subjects suggests that glucose uptake is decreased in the diabetic adipocyte.
Materials and methods Four categories of patients were studied (Table 1). None of these patients was receiving dietary or drug therapy for diabetes. The purpose of the study was explained to the patients and each patient consented to the procedures carried out. Samples of subcutaneous adipose tissue were obtained from the abdominal wall of obese subjects by needle biopsy (Diengott & Kerpel, 1967) under local anaesthesia (1 ml of 1% lignocaine). Biopsies were obtained after an overnight fast of 10-14 h. The core of adipose tissue (200-600 mg) was immediately
Key words : adipocytes, diabetes, erythrocytes, fasting, glucose, obesity.
Introduction Triglyceride glycerol is readily synthesized from glucose by human adipose tissue. However, under different physiological conditions such as fasting and Correspondence: Dr D. J. Galton, Diabetes Research Laboratory, Department of Medicine, St Bartholomew’s Hospital, London, E.C.I.
27
M. A . Page and D. J. Galton
28
TABLE 1. Clinical details of patienfs in the study Data are mean values+ SEM. NO.
Patient group 20 Non-diabetic obese 8 Non-diabetic, fasted for 7-10 days Non-diabetic, obese, for glucose tolerance test 1 1 9 Diabetic obese, for glucose tolerance test
Age (years)
Height (cm)
Weight (kg)
% Ideal weight
39.0f3.8 44.5k4.9 34.0k3.7 56.1k4.8
162.0f 17.0 164.0+ 1.6 1645k 2.3 164*4+_3.9
107.0k4.9 108.6f4.5 90.3k3.8 79.2k3.3
183.0f 8.0 195.0+ 9.6 160.5f 7.0 137.0k 11.0
dropped into liquid nitrogen, except for a small portion which was kept for estimation of cell size. The frozen tissue was divided into 100-200 mg portions. Each piece was weighed before being homogenized in ice-cold perchloric acid (500 mmol/ 1) and light petroleum (boiling point 4O-6O0C), with a power-driven Teflon pestle in a glass mortar. The glass mortar was rinsed twice in cold perchloric acid (500 mmol/l) and light petroleum. The total volume of the perchloric acid phase was 2 ml and the total volume of light petroleum used was 8 ml. The cold homogenate was centrifuged for 5 min at 4000 g and the aqueous and light petroleum phases were separated. The light petroleum phase was evaporated and triglyceride in the residue was estimated (Cramp & Robertson, 1968). The aqueous layer was decanted from the precipitated protein and neutralized with KOH (1.8 mol/l). After cooling on ice the excess of perchlorate was removed by centrifugation and the supernatants were stored at -20°C for up to 48 h before assay. The concentration of themetabolites did not decrease during this storage period. Immediately after venepuncture 5 ml of blood was added to 10 ml of ice-cold perchloric acid (500 mmol/l) in a weighed bottle. The deproteinized blood was spun for 20 min at 6000 g at 4°C and the clear supernatant decanted and neutralized as described for adipose tissue. The results are presented for whole blood since it was considered that centrifugation before deproteinization may distort the concentrations of glycolytic intermediates in erythrocytes. Previous studies (Bosia, Pescarmona & Arese, 1971 ; Minakami, Suziki, Saito & Yoshikawa, 1965) have also ignored the contribution of leucocytes and plasma to erythrocyte metabolites. Metabolite concentrations were determined with an Aminco-Bowman Ratio Spectrofluorimeter by measuring the increase in fluorescence due to NADH and NADPH at excitation wavelength 340 nm and emission wavelength 460 nm. Metabolites were
weight
Adipocyte diameter (Pm) 110f3.9 116f48 109f4.3 112f5.8
assayed in duplicate on consecutive days and blank samples were run during each assay to correct for machine drift and any change in fluorescence due to the addition of enzyme. Standards for each intermediate were included in all assays. The concentrations of each standard were determined with a Hitachi 101 spectrophotometer. Glucose 6-phosphate, fructose 6-phosphate and fructose 1,6diphosphate were measured in the same cuvette, in 0.5 ml of tissue extract in a final volume of 1 ml, by the method of Racker (1965). Glycerol phosphate was determined with 0.2 ml of extract by the method of Hohorst (1965). Uridine diphosphate glucose was assayed in a final volume of 1 ml of glycine buffer (50 mmol/l, pH 8.7), containing NAD (2 mmol/l), tissue extract (0.2 ml) and uridine diphosphate glucose dehydrogenase (5 mg). Blood glucose was measured by an automated glucose oxidase method; plasma insulin was assayed by solid-phase radioimmunoassay. Determination of fat-cell diameter
Isolated fat-cells were prepared by the method of Rodbell (1964) and fat-cell diameters were determined by a direct microscopic method (di Girolamo, Mendlinger & Fertig, 1971). The adipose cell volume was calculated from the mean of the diameters of at least 300 cells, assuming that the cells were spherical (Goldrick, 1967). Under these conditions the diameter of cells liberated by collagenase correlates closely to the diameter of cells of intact tissue slices (r = 0.964, n = 11; Smith, Sjostrom & Bjorntorp, 1972). Statistics
Significances of differences were tested by use of Student’s paired t-test. Body weight (% of ideal) was calculated from tables derived by the Society of Actuaries, Chicago (1959). Linear correlation
Metabolites in blood and adipose tissue
29
TABLE 2. Concentrations of metabolites in human blood and adipose tissue from obese subjects The results are mean values* SEM and the number of subjects is shown in parentheses. Details of patients are given in the Materials and methods section. Adipose tissue was obtained before and after a fast from the same patients. Significance of differences by Student’s paired t-test: * P
