James E. Foley

Enlarged subcutaneous abdominal adipocyte size, but not obesity itself, predicts type II diabetes independent of insulin resistance.

Aims/hypothesis. Cross-sectional studies indicate that enlarged subcutaneous abdominal adipocyte size is associated with hyperinsulinaemia, insulin resistance and glucose intolerance. To further explore the pathophysiological significance of these associations, we examined prospectively whether enlarged subcutaneous abdominal adipocyte size predicts Type II (non-insulin-dependent) diabetes mellitus. Methods. Body composition (hydrodensitometry), mean subcutaneous abdominal adipocyte size (fat biopsy), insulin sensitivity (hyperinsulinaemic clamp) and the acute insulin secretory response (25-g i. v. GTT) were assessed in 280 Pima Indians with either normal (NGT), impaired (IGT) or diabetic glucose tolerance (75-g OGTT). Subjects with NGT were then followed prospectively. Results. After adjusting for age, sex and per cent body fat, mean subcutaneous abdominal adipocyte size was 19 % and 11 % higher in subjects with diabetes and IGT, compared with those with NGT (p < 0.001). Insulin sensitivity was inversely correlated with mean subcutaneous abdominal adipocyte size (r = –0.53, p < 0.0001), even after adjusting for per cent body fat (r = –0.31, p < 0.001). In 108 NGT subjects followed over 9.3 ± 4.1 years (33 of whom developed diabetes), enlarged mean subcutaneous abdominal adipocyte size but not high per cent body fat, was an independent predictor of diabetes, in addition to a low insulin sensitivity and acute insulin secretory response [relative hazard 10th vs 90th centile (95 % CI): 5.8 (1.7–19.6), p < 0.005]. In 28 NGT subjects with a 9 % weight gain over 2.7 ± 1.3 years, changes in insulin sensitivity were inversely and independently related to changes in mean subcutaneous abdominal adipocyte size and per cent body fat. Conclusion/interpretation. Although enlarged mean subcutaneous abdominal adipocyte size is associated with insulin resistance cross-sectionally, prospectively, both abnormalities are independent and additive predictors of Type II diabetes. [Diabetologia (2000) 43: 1498–1506]

A high concentration of fasting plasma non-esterified fatty acids is a risk factor for the development of NIDDM

SummaryTo assess the role of fasting plasma non-esterified fatty acids (NEFA) in the development of non-insulin-dependent diabetes mellitus (NIDDM), data were analysed from annual examinations of 190 non-diabetic Pima Indians. Glucose tolerance was measured by a 75-g oral glucose tolerance test, insulin action by a euglycaemic hyperinsulinaemic (40 mU · m−2 · min−1) clamp and in vitro lipolysis using isolated abdominal fat cells. After a mean follow-up period of 4.0±2.4 years (mean ± SD), 47 subjects developed NIDDM. Risk factors for NIDDM were estimated by proportional-hazards analysis and risk ratios (RR) with 95% confidence intervals (95% CI) calculated at the 90th and 10th percentile of the predictor variables. A large average fat-cell volume was predictive of NIDDM (RR=2.4; 95% CI=1.2–4.8) independent of age, sex, percent body fat and body fat distribution. A high fasting plasma NEFA concentration was also a risk factor for NIDDM (RR=2.3; 95% CI=1.1–4.7) independent of sex, percent body fat, waist/thigh ratio, insulin-mediated glucose uptake and fasting triglyceride concentration. We conclude that large fat cells and the resulting increased plasma NEFA concentrations are risk factors for the development of NIDDM.

Kinetics of glucose disposal in whole body and across the forearm in man.

We reevaluated the concept that the in vivo glucose disposal rate in man is determined by the activity of the glucose transport system. Rates of glucose disposal were determined in whole body and across forearm at four insulin levels (approximately 9, approximately 50, approximately 160, and approximately 1700 microU/ml) and at each insulin level at four glucose levels (approximately 90, approximately 160, approximately 250, and approximately 400 mg/dl). At the lowest insulin level, the Michaelis constants (Ks:s) for glucose disposal in whole body (8.7 +/- 1.1 mM) and across forearm (7.4 +/- 1.4) mM) were compatible with a Ks determined in vitro for the transport system. At higher insulin levels, the apparent Ks increased significantly in whole body (16.2-37.7 mM) and across forearm (20.7-31.2 mM). We interpret the apparent increase of Ks by insulin to reflect a shift in the rate-limiting step from glucose transport to some step beyond transport.

Relationship between obesity and maximal insulin-stimulated glucose uptake in vivo and in vitro in Pima Indians.

Previous studies have left unanswered whether human obesity, independent of glucose intolerance, is associated with a postreceptor defect in insulin action. We have studied the relationship between the degree of obesity (as estimated by underwater weighing) and the maximal insulin-stimulated glucose disposal rate (M) in vivo in 52 glucose-tolerant Pima Indian males. The relationship was examined independently of differences in age and maximal oxygen uptake (an estimate of physical fitness). The maximal insulin-stimulated glucose transport rate (MTR) was also measured in isolated abdominal adipocytes from the same subjects to determine whether differences in M could be explained by differences in glucose transport. The results showed that there was a large variance in M and MTR among these glucose-tolerant subjects. M was better correlated with glucose storage rates than with oxidation rates, as estimated by indirect calorimetry. The most obese subjects had only a 20% lower mean M and 30% lower MTR than the most lean subjects. The lower M in the obese subjects was due to both lower glucose oxidation and storage rates. There was no significant, independent correlation between age or degree of obesity and M or MTR. The maximal oxygen uptake (VO2 max) appeared to independently account for 20% of the variance observed in M. MTR was only weakly correlated with M (r = 0.36, P less than 0.02). We concluded that differences in M in these glucose-tolerant subjects must be explained by factor(s) other than maximal oxygen uptake, age, maximal insulin-stimulated glucose transport in vitro, or degree of adiposity per se.

Free fatty acid metabolism and obesity in man: In vivo in vitro comparisons

We have examined the relationship of free fatty acid (FFA) turnover and lipid oxidation rates in vivo to the size of body triglyceride stores and compared these findings with the in vitro lipolytic rates of isolated abdominal fat cells. The studies were performed in 20 Pima Indian women 18 to 35 years of age, both lean and obese. FFA turnover rate was measured using a 1-14C-palmitate infusion, lipid oxidation rate by indirect calorimetry using a ventilated hood, body composition by underwater weighing with correction for residual lung volume, and fat cell lipolytic rates in vitro by published methods. Both FFA turnover and lipid oxidation rates, expressed per kg of body fat, decreased with increasing degree of obesity (as measured by percent body fat) (r = -0.90, and r = -0.75, P less than or equal to 0.0001, respectively). In contrast, the rate of lipolysis determined in vitro, expressed per kg of fat, increased with increasing degree of obesity (r = 0.58, P less than 0.01). A ratio of FFA turnover/lipolysis, which directly compares these in vivo and in vitro measurements, decreased significantly with increases in the degree of obesity (r = -0.81, P less than or equal to 0.0001). Furthermore, there were no positive correlations between the measures of in vivo FFA metabolism and in vitro lipolysis when both were expressed per fat mass, per fat cell number, or per fat cell surface area. The in vivo data also demonstrated that lipid oxidation could only account for 50% of the FFA disappearance rate. While lipid oxidation rate adjusted to the metabolic size increased with increasing plasma FFA concentration (r = 0.75, P less than 0.0003), the nonoxidative component of the FFA turnover failed to increase with increases in plasma FFA concentration (P = 0.5). We conclude that FFA is not available in vivo in proportion to the size of the triglyceride stores. The reason for this is not due to an inability of fat cells to release their stored triglyceride as assessed in vitro. Hence, in vitro measurements of fat cell lipolysis cannot be used to directly predict in vivo FFA metabolism. The large nonoxidative FFA disposal is likely to be important in the regulation of plasma FFA concentrations.

Comparison of body composition, adipocyte size, and glucose and insulin concentrations in Pima Indian and Caucasian children☆☆☆

Pima Indian adults with normal glucose tolerance have higher plasma glucose and insulin concentrations than Caucasian adults. To estimate the age of onset of these differences, and to assess their relationship to abdominal and gluteal adipocyte size, we measured adiposity, adipocyte size, and glucose and insulin concentrations during a glucose tolerance test in lean (less than 20% body fat), prepubertal children from each race. The Pima (n = 13) and Caucasian (n = 10) groups were of similar age, percent body fat, and weight. Pima Indian children had higher fasting glucose (101 +/- 2 v 94 +/- 2 mg/dL, P = .01) and insulin (22 +/- 2 v 15 +/- 2 microU/mL, P less than .01) concentrations and larger abdominal adipocytes (0.49 +/- 0.03 v 0.37 +/- 0.04 microgram lipid/cell, P less than .05) than the Caucasian children. Postprandial glucose and insulin concentrations and gluteal adipocyte size were similar in the two races. The higher plasma glucose and insulin concentrations found in Pima adults are present in lean Pima children, and are associated with increased abdominal adipocyte size. These increases may precede the development of obesity in this racial group.

Regulation of Plasma Lactate Concentration in Resting Human Subjects

We evaluated the relative contributions of glucose, insulin, and the rate of glucose disposal to the regulation of the plasma lactate concentration. Rates of glucose disposal were measured in 88 separate studies in whole body and across the forearm at varying plasma insulin (9, 50, 160, and 1,800 microU/mL) concentrations, and at each insulin concentration at four different glucose concentrations (90, 160, 250, and 400 mg/dL) in healthy male subjects. The rate of glucose disposal was positively correlated with the plasma lactate concentration (r = .83, n = 88, P less than .0001). When the plasma lactate concentration was adjusted for the rate of glucose disposal, plasma glucose or insulin concentrations did not contribute significantly to the residual variation in plasma lactate. When plasma lactate concentrations were compared at matched rates of glucose disposal, the lactate levels were similar regardless of whether glucose disposal was induced by hyperglycemia or hyperinsulinemia. At the lowest glucose and insulin concentrations, forearm tissues released lactate, but at all other glucose and insulin concentrations, no significant net lactate flux was observed. After subtraction of the rate of forearm glucose disposal from whole-body glucose disposal, the plasma lactate concentration correlated with the remaining, extramuscular, rate of glucose disposal (r = .60, P less than .0001). These data suggest that in resting normal subjects the plasma lactate concentration may be determined by the rate of glucose disposal in extramuscular tissues, rather than the ambient glucose or insulin concentration.

In vitro insensitivity of glucose transport and antilipolysis to insulin due to receptor and postreceptor abnormalities in obese Pima Indians with normal glucose tolerance

The in vitro sensitivities of glucose transport and antilipolysis to insulin and insulin binding were measured in adipocytes isolated from three groups of normal glycemic male Pima Indians--10 lean (11% to 22% body fat), 11 moderately obese (26% to 34% body fat), and 7 severely obese (37% to 40% body fat) subjects. Both a half-maximum concentration of insulin for the stimulation of glucose transport (ED50 [transport]) and a half-maximum concentration of insulin for the suppression of 25 nmol/L isoproterenol-stimulated lipolysis (ED50 [antilipolysis]) were significantly (P less than 0.05) greater in the moderately obese subjects than in the lean subjects as well as greater in the severely obese group than in the moderately obese group. Mono 125I-(Tyr A14)-insulin binding per cell in the presence of 25, 100, and 200 pm insulin was similar among lean, mildly obese, and severely obese subjects. 125I-insulin binding per cell surface area of adipocytes isolated from either moderately or severely obese Indians was significantly lower (P less than 0.005) than that of lean Indians. However, there was a similar insulin binding per cell surface area between mildly and severely obese subjects. These results indicate that diminished insulin binding per cell surface area may explain decreased sensitivity of transport and antilipolysis to insulin in moderately obese subjects relative to lean subjects. In contrast, these diminished sensitivities in the severely obese subjects relative to moderate obese subjects are not explained by a change in insulin binding and, therefore, are presumably induced by an abnormality of a postbinding step of insulin action.

Opposite effects of a β-adrenergic agonist and a phosphodiesterase inhibitor on glucose transport in isolated human adipocytes: Isoproterenol increases Vmax and IBMX increases Ks

Abstract Isoproterenol (2 μM) stimulated the basal rate of 3-0-methylglucose transport into isolated human adipocytes by increasing the Vmax without changing the dissociation constant (Ks), while isoproterenol had no consistent effect on maximum insulin-stimulated glucose transport. In contrast, 1 mM 3-isobutyl-1- methylxanthine inhibited both basal and maximum insulin-stimulated transport by increasing the Ks without a significant change in Vmax. Glucose transport, therefore, is affected oppositely and by different kinetic parameters by two agents which are known to increase cyclic AMP levels in isolated human adipocytes.

Glucose initiates but does not maintain acceleration of deoxyglucose transport

Glucose is the initiator of a cellular process which results in an acceleration of deoxyglucose uptake 10-15 min later. This acceleration is maintained for at least 30 min after glucose is removed from the medium and is independent of transcription and translation of mRNA.