Search

Insulin stimulates glucose transport in rat adipose cells through the translocation of glucose transporters from an intracellular pool to the plasma membrane. A detailed characterization of the morphology, protein composition and marker enzyme content of subcellular fractions of these cells, prepared by differential ultracentrifugation, and of the distribution of glucose transporters among these fractions is now described. Glucose transporters were measured using specific D-glucose-inhibitable [3H]cytochalasin B binding. In the basal state, roughly 90% of the cells' glucose transporters are associated with a low-density microsomal, Golgi marker enzyme-enriched membrane fraction. However, the distributions of glucose transporters and Golgi marker enzyme activities over all fractions are clearly distinct. Incubation of intact cells with insulin increases the number of glucose transporters in the plasma membrane fraction 4-5 fold and correspondingly decreases the intracellular pool, without influencing any other characteristics of the subcellular fractions examined or the estimated total number of glucose transporters (3.7 X 10(6)/cell). Insulin does not influence the Kd of the glucose transporters in the plasma membrane fraction for cytochalasin B binding (98 nM), but lowers that in the intracellular pool (from 141 to 93 nM). The calculated turnover numbers of the glucose transporters in the plasma membrane vesicles from basal and insulin-stimulated cells are similar (15 X 10(3) mol of glucose/min per mol of transporters at 37 degrees C), whereas insulin appears to increase the turnover number in the plasma membrane of intact cells roughly 4-fold. These results suggest that (1) the intracellular pool of glucose transporters may comprise a specialized membrane species, (2) intracellular glucose transporters may undergo conformational changes during their cycling to the plasma membrane in response to insulin, and (3) the translocation of glucose transporters may represent only one component in the mechanism through which insulin regulates glucose transport in the intact cell.

Three techniques have now been used to demonstrate that insulin stimulates glucose transport in isolated rat adipose cells through the translocation of glucose transporters from a large intracellular pool to the plasma membrane. By using a specific D-glucose-inhibitable cytochalasin B-binding assay, most of the basal cell's transporters are found associated with a low-density microsomal membrane fraction. However, although Golgi marker enzyme activities are also enriched in this fraction, their distributions over all fractions do not parallel that of the transporters. In response to insulin, more than half of the intracellular transporters are translocated to the plasma membranes without a corresponding redistribution of marker enzyme activities. Furthermore, although the Kd of the transporters in the plasma membranes remains constant at approximately 100 nM, that of the intracellular transporters decreases from approximately 140 to approximately 100 nM. Nevertheless, transport activity is reconstitutable from, and an affinity-purified rabbit IgG against the purified human erythrocyte transporter cross-reacts with a 45,000-dalton band in, both plasma membranes and the low-density microsomal membrane fraction in proportion to the number of glucose transporters determined by cytochalasin B binding. Thus, intracellular glucose transporters in the rat adipose cell appear to be 1) localized to a unique membrane species, 2) either compartmentalized in two distinguishable pools or processed during their cycling to the plasma membrane in response to insulin, but fully functional and indistinguishable when reconstituted into liposomes, and 3) immunologically similar to the human erythrocyte glucose transporter.

The possible role of hyperinsulinism in the generation and perpetuation of atherosclerosis is discussed. Although the hypothesis is an important one that merits intensive study, it is premature to approach the question through a clinical trial. Rather, a concerted effort should be made to gain more information through basic research in laboratory animals and clinical investigations in man. This basic information will form the foundation upon which future, sound clinical trials may be undertaken. The general characteristics of a sound clinical trial are also discussed. Topics highlighted include the need to define the question to be asked; the intervention to be used; the outcome to be measured; the patient population to be studied, including the necessity of a control group; the eligibility and exclusion criteria for entrance into a study; and basic study design. Each of these topics requires attention and careful planning before a trial is begun. Issues of randomizing patients into treatment groups, maximizing the likelihood of compliance, and the need for quality control and careful monitoring are also reviewed.

The effects of increasing cell size on glucose transport activity and metabolism and on the concentrations of glucose transport systems in both the plasma and low density microsomal membranes in isolated adipose cells from the aging rat model of obesity have been examined. Glucose transport activity was assessed by measuring l-arabinose transport and the concentration of glucose transport systems estimated by measuring specific d-glucose-inhibitable cytochalasin B-binding. Basal glucose transport activity increases from 0.3 to 1.4 fmol/cell/min with a 10-fold increase in cell size, but remains constant per unit cellular surface area and is accompanied by a constant 5 pmol of glucose transport systems/mg of membrane protein in the plasma membrane fraction. Maximally insulin-stimulated glucose transport activity, on the other hand, remains constant at 2.3 fmol/cell per min with increasing cell size, but markedly decreases per unit cellular surface area and is accompanied by a decrease from 30 pmol of glucose transport systems/mg of plasma membrane protein to the basal level. These diminished effects of insulin on glucose transport activity and the number of glucose transport systems in the plasma membrane fraction in enlarged cells are paralleled by an 80% decrease in the basal number of glucose transport systems/mg of membrane protein in the low density microsomal membrane fraction, the source of those glucose transport systems appearing in the plasma membrane in response to insulin. The effects of cell size on the metabolism of a low concentration of [1-(14)C]glucose (0.56 mM) directly parallel those on glucose transport activity and the concentration of glucose transport systems in the plasma membrane fraction, and are not associated with significant alterations in the cell's sensitivity to insulin. Thus, adipose cellular enlargement is accompanied by the development of a marked "insulin resistance" at the glucose transport level, which may be the consequence of a relative depletion of glucose transport systems in the intracellular pool.

Studies of adipose tissue cellularity were carried out in a group of nonobese adult male volunteers who gained 15-25% of their body weight as the result of prolonged high caloric intake. Adipose cell size (lipid content per cell) was determined in tissue obtained from three subcutaneous sites (gluteal, anterior abdominal wall, and triceps) and total adipose cell number estimated from measurement of total body fat. Five experimental subjects gained an average of 16.2 kg of body weight, of which 10.4 kg was determined to be fat. Expansion of the adipose mass was accompanied by a significant and relatively uniform increase in fat cell size in each subcutaneous site tested. Total adipose cell number did not change as a result of weight gain and expansion of the adipose depot in adult life. Subsequent loss of weight and restoration of original body fat was associated with a reduction in adipose cell size at each subcutaneous site, but no change in total number. In two control subjects who neither gained nor lost weight there were no changes in total adipose cell number or cell size. These observations suggest that expansion and retraction of the adipose depot in adult life is accompanied by changes in adipose cell size only. Significant differences in both the size and total number of adipose cells were observed between subjects in both the experimental and control groups. In addition, within individuals of both groups there were significant differences in cell size when adipose cells from the three subcutaneous sites were compared. These findings indicate that wide variations in adipose cell size and number exist in nonobese individuals having similar adipose depot sizes.

The cellular character of the adipose tissue of 21 nonobese and 78 obese patients has been examined. Adipose cell size (lipid per cell) was determined in three different subcutaneous and deep fat depots in each patient and the total number of adipose cells in the body estimated by division of total body fat by various combinations of the adipose cell sizes at six different sites. Cell number has also been estimated on the basis of various assumed distribution of total fat between the subcutaneous and deep fat depots. Obese patients, as a group, have larger adipose cells than do nonobese patients; cell size, however, varies considerably among the fat depots of individuals of either group. The variation in cell size exists not only between, but also within subcutaneous and deep sites. Estimates of total adipose cell number for a given individual based upon cell size can, therefore, vary by as much as 85%. On the basis of these studies it is suggested that the total adipose number of an individual is best and most practically estimated, at this time, by division of total body fat by the mean of the adipose cell sizes of at least three subcutaneous sites. IRRESPECTIVE OF THE METHOD BY WHICH TOTAL ADIPOSE CELL NUMBER IS ESTIMATED, TWO PATTERNS OF OBESITY EMERGE WITH RESPECT TO THE CELLULAR CHARACTER OF THE ADIPOSE TISSUE MASS OF THESE PATIENTS: hyperplastic, with increased adipose cell number and normal or increased size, and hypertrophic, with increased cell size alone. These two cellular patterns of obesity are independent of a variety of assumed distributions of fat among the subcutaneous and deep depots. When these different cellular patterns are examined in terms of various aspects of body size, body composition, and the degree, duration, and age of onset of obesity, only the latter uniquely distinguishes the hyperplastic from the hypertrophic: hyperplastic obesity is characterized by an early age of onset, hypertrophic, by a late age of onset. These studies indicate that there are two distinct periods early in life during which hypercellularity of the adipose tissue are most likely to occur: very early within the first few years, and again from age 9 to 13 yr.

The effect of insulin upon the lipid content, and the number and size of fat cells in the epididymal, retroperitoneal, and subcutaneous adipose tissue of a large number of rats were examined. Insulin administration began either in early life (birth, 1, or 3 wk of age) or during adulthood (age 10 wk). At different times during growth, groups of treated and control animals were killed and the size and number of fat cells in each of the three adipose depots were determined. Insulin-treated animals gained weight at an increased rate and had fatter epididymal, retroperitoneal, and subcutaneous adipose depots than untreated controls. In each site the expanded adipose tissue was accompanied by an increase in the lipid content per cell (cell size), but in no case was there an increase in the number of adipose cells. This was the case regardless of whether insulin treatment was initiated before weaning (birth, 1 wk of age), at weaning (3 wk), or post weaning (10 wk) and irrespective of the duration of the insulin treatment.

Glucose metabolism and insulin sensitivity of isolated human adipose tissue was studied as a function of adipose cell size and number. Glucose metabolism by these tissues was closely related to the number of cells in the fragment, irrespective of cell size. Adipose cells of obese individuals metabolized glucose to carbon dioxide and triglyceride at rates similar to adipose cells of nonobese subjects. In contrast, insulin responsiveness of adipose tissue was dependent upon adipose cell size. The larger its adipose cells the less insulin sensitive was the tissue. Thus, adipose tissue of obese subjects, with enlarged cells, showed a diminished response to insulin. After weight loss and reduction in adipose cell size, insulin sensitivity of the adipose tissue of obese patients was restored to normal. When adipose tissue of obese individuals showed impaired responsiveness to insulin, their plasma insulin levels, after oral glucose, were elevated. Weight loss and reduction in adipose cell size restored plasma insulin concentration to normal, concomitant with the return of normal tissue insulin sensitivity.

Glucose metabolism and insulin sensitivity of isolated rat epididymal fat cells and of their delipidated derivatives ("ghosts") was studied as a function of cellular lipid content (fat cell size), cellular protein content, animal age, and state of nutrition in an effort to examine the relationship of adipose cell size to adipose tissue insulin sensitivity.In ad libitum-fed rats, basal rates of glucose-1-(14)C incorporation into CO(2) and triglyceride are similar over a wide range of adipose cell size. In contrast, the insulin sensitivity of intact fat cells from rats fed ad libitum is inversely related to their lipid content: the larger the cell, the less the response to insulin. This "resistance" of the enlarged adipose cell to the action of insulin was demonstrated by a reduction in the per cent rise above the basal rate as well as in the absolute rate of glucose oxidation and lipogenesis caused by insulin. The protein content of fat cells was found to be relatively constant over a wide range of fat cell size. Thus, enlarged insulin "resistant" fat cells contained the same amount of protein as smaller insulin "sensitive" cells. These relationships between insulin sensitivity and cellular lipid or protein content were true regardless of whether cells of different sizes were obtained from animals of different body weights and ages, or from different portions of the epididymal fat pads of animals of the same weight and age.Acute delipidation of intact fat cells did not appear to alter these relationships between basal glucose metabolism, insulin sensitivity, and cell size. "Ghosts" prepared from fat cells of widely different sizes metabolized glucose to CO(2) and triglyceride at similar rates. The insulin sensitivity of the fat cell "ghost" appeared to be inversely related to the size of the intact cell from which it was derived: the larger the intact cell the less insulin sensitive its "ghost."Although the insulin "resistance" of adipose tissue was reversed by weight loss and reduction of fat cell size, these studies also demonstrate that the insulin sensitivity of adipose cells of similar sizes can vary widely depending upon the state of nutrition and growth of the animal. Thus, factors other than cell size can also influence the insulin sensitivity of the adipose tissue.