The Relationship between the Insulin-binding Capacity of Fat Cells and the Cellular Response to Insulin

Insulin binding and effects in isolated soleus muscle of lean and obese mice

To get some insight into the mechanisms of insulin resistance in obesity, insulin binding and biological effects were investigated in soleus muscles isolated from normal and obese mice. Basal and insulin-stimulated 2-deoxyglucose uptake were measured at the steady state of insulin binding. The results were consistent with the concept of spare receptors, i.e., maximal insulin effect was achieved when only about 20% of total receptors was occupied. When similar studies were applied to muscles of gold thioglucose obese or genetically obese (ob/ob) mice, and compared to lean controls: a) insulin binding was decreased; b) the insulin dose-response curve of 2-deoxyglucose uptake was shifted to the right; c) maximally insulin-stimulated 2-deoxyglucose uptake, glycolysis, and glycogen synthesis were markedly decreased. Insulin binding and effects returned toward normal after a 40-h fast in obese mice. These results point to two loci for the insulin resistance of skeletal muscle in obesity: 1) a decrease in the number of insulin receptors, which results in a diminished insulin sensitivity; and 2) one or more alterations beyond receptor that are responsible for the decreased responsiveness of the tissue to insulin and appear to play a major role in the insulin resistance of muscle in obesity.

Insulin receptors on monocytes of young healthy persons correlated with glucose tolerance and insulin sensitivity

We have studied in normal man the interrelationships between insulin binding to monocytes, glucose tolerance and insulin sensitivity. In 25 young healthy persons we found a significant positive correlation between insulin binding and glucose disappearance rate both after glucose (R = 0.68, p less than 0.001) and insulin (R = 0.49, p less than 0.02) given intravenously.

Reduced glucose transport and increased binding of insulin in adipocytes from diabetic and fasted rats

1. Animals made diabetic by injection of streptozotocin or animals after 3 days of fasting show decreased insulin levels and a decrease in mean cell diameter of adipocytes from epidydymal fat pads in comparison with cells from normal animals. 2. 14CO2 production from D-[U-14C]glucose is impaired in diabetic and fasted animals both in presence or in absence of a concentration of insulin stimulating 14CO2 production maximally. 3. Insulin binding is increased in adipocytes from diabetic and fasted animals due to changes in affinity. 4. Transport studies show that basal and insulin stimulated 2-deoxy[1-14C]-glucose transport is decreased in absolute terms due to a decrease in V and an increase in Km. 5. The relative stimulatory effect of insulin is impaired in adipocytes of diabetic and fasted animals. 6. A shift of the maximal effect of insulin to lower insulin levels is seen in these cells.

Insulin receptors in adipocytes of non-diabetic and diabetic subjects. Preliminary report

We have measured insulin binding to human adipocytes isolated from subcutaneous adipose tissue removed during surgery in normal and insulin-independent diabetics. Collagenase digestion, 125I-monoiodoinsulin and Scatchard's plot were employed to analyze the results. Different kinetic patterns emerged together with differences in the dissociation constant and receptor numbers: in normal subjects K1 was 4 X 10(-9) moles/1 and K20.5 X 10(-8) moles/1, and in diabetic subjects K1 was 2.24 X 10(-9) moles/1 and K2 0.52 X 10 10(-8) moles/1; the two classes of receptors were 100,000 and 300,000 per cell in normals and 50,000 and 180,000 in diabetics. It was clear that even slight diabetes leads to receptor deficiency in adipocytes, though it could not be determined whether this a primary, perhaps genetic, defect or secondary to antibody damage, as suggested by some workers. Feed-back between circulating insulin and specific receptor availability in cells is another possibility.

Influence of trypsin on lipolysis in human fat cells. Comparison with rat adipocytes

1. Trypsin-treated human and rat fat cells were obtained by digestion of adipose tissue with collagenase plus trypsin and their lipolytic response to insulin, catecholamines and dibutyryl cyclic AMP were compared with the lipolytic response of human and rat fat cells isolated with collagenase only. 2. In both human and rat fat cells, no significant modification occurred in the intracellular lactate dehydrogenase content and in the basal release of glycerol after trypsination. 3. In rat fat cells, trypsin abolished the antilipolytic effect of insulin but maintained a normal lipolytic response to epinephrine, norepinephrine and isoproterenol. 4. In human fat cells, on the contrary, trypsin failed to modify the antilipolytic effect of insulin, but markedly potentiated the lipolytic response to epinephrine, norepinephrine and isoproterenol. Trypsin also increased the rate of intracellular 3' :5' cyclic AMP accumulation in response to catecholamines. Under these conditions, however, trypsin-treated human fat cells had a normal reponse to the lipolytic agent dibutyryl cyclin AMP. 5. These data suggest that human fat cells differ from the rat ones by the existence in human adipocyte membranes of a trypsin-sensitive component which inhibits the catecholamine induced lipolytic process and which is different from the alpha receptors.

Preparation of parenchymal and non-parenchymal cells from adult human liver--morphological and biochemical characteristics

By perfusion of the isolated human liver with collagenase and hyaluronidase a mixed suspension of various cell types was obtained. Pure parenchymal cells were prepared by differential centrifugation, pure non-parenchymal cells by the use of pronase and subsequent isopycnic centrifugation on metrizamide gradients (50-300 g/l). About 90% of the parenchymal and non-parenchymal cells were viable as judged by trypan blue staining. Non-parenchymal cells were not capable fo gluconeogenesis but at high rates. Parenchymal cells retained their ability to form glucose and to accumulate glycogen from fructose greater than lactate/pyruvate greater than alanine. Studies on binding of 125I-labelled insulin by isolated parenchymal cells were performed at 30 degrees C. The binding data may fit with a minimum of two classes of binding sites: (a) high affinity--low capacity sties (Kd approximately 6.6 nmol/l, capacity approximately 16 000 insulin molecules per cell) and (b) low affinity-high capacity sites (Kd approsimately 0.37 mumol/l, capacity approximately 646 000 molecules per cell).

Regulation of the D-glucose transport system in isolated fat cells

Recent technical advances have yielded considerable new biochemical insights into the hexose transport systems of both brown and white fat cells. In the present studies a novel filtration method was used to monitor initial rates of 3-O-(3H)methylglucose uptake in isolated white fat cells. Transport of 3-O-methylglucose, a non-metabolizable analogue of glucose, occurred by facilitated diffusion, was inhibited by glucose, phloridzin, cytochalasin B and dipyridamole, and was rapidly stimulated by insulin as well as lectins. Total 3-O-methylglucose uptake in white fat cells could be attributed to two kinetically distinct processes in addition to a certain degree of diffusion. Two important new features of glucose transport in fat cells have been discovered. First, in both brown and white fat cells transport per se does not appear to be necessarily rate-limiting for further glucose metabolism. Thus vitamin K5, which markedly increases glucose oxidation by brown fat cells, did not affect the glucose transport system activity. Glucose utilization can apparently be significantly enhanced in fat cells by agents which either increase transport system activity or intracellular enzyme activity. Second, the transport system itself, whether in the basal state or after activation by insulin, lectins, or oxidants, is resistant to sulfhydryl reagents such as N-ethylmaleimide, while the increase in transport activity due to these agents is exquisitely sensitive to sulfhydryl blockage. N-ethylmaleimide blocks the stimulatory effect of insulin on transport whereas addition of insulin to fat cells prior to the reagent completely protects against this inhibitory effect. Further, N-ethylmaleimide prevents the elevated rates of transport system activity due to insulin (or other agents) from returning to basal levels once the cells are washed free of hormone. These data are consistent with the concept that activation of the transport system involves oxidation of key membrane sulfhydryls to the disulfide form, but alternative models are also possible. In any case, these findings provide a possible biochemical clue for future studies designed to identify the specific component(s) involved in the regulatory mechanism which modulates transport of glucose in isolated fat cells.

Insulin receptors of skeletal muscle: specific insulin binding sites and demonstration of decreased numbers of sites in obese rats

A membrane preparation was obtained from rat striated muscle. The preparation used has been shown to contain plasma membranes by electron microscopy as well as by enrichment in specific activity of both a plasma membrane enzyme "marker" (5'-nucleotidase) and cell surface 125I-incorporated radioactivity. The characteristics of 125I-insulin binding to this striated muscle preparation were studied, and it was found that 125I-insulin readily and specifically binds to this membrane preparation. The binding reaction was time, pH, and temperature dependent with optimal steady-state binding conditions occurring at 20 degrees C and at pH 7.6. Under these conditions (20 degrees C, pH 7.6) skeletal muscle plasma membranes displayed little ability to degrade insulin. Binding of 125I-insulin was readily inhibited at physiologic concentrations of unlabeled insulin and the specificity of this receptor for insulin was demonstrated by finding that high concentrations of glucagon, b-LH, b-FSH, p-PRL, hCG, TSH, and HGH were without effect on 125I-insulin binding and that insulin analogues inhibited binding in proportion to their biologic activity. When membranes from older, fatter rats were compared to membranes from younger, lean animals, 5'-nucleotidase specific activity and insulin degrading activity were found to be comparable. On the other hand, insulin binding to membrane receptors was decreased 30%-40% in the older, fatter animals. Thus, these studies indicate that (1) specific insulin receptors exist in skeletal muscle plasma membranes, and (2) membranes from older, fatter rats have fewer receptors than those from younger, lean animals.

Interaction of carbohydrate binding sites on concanavalin A-agarose with receptors on adipocytes studied by buoyant density method

The interaction of concanavalin A (Con A) with isolated adipocytes was studied using Con A-Sepharose beads in the affinity binding buoyant density method previously used to study insulin receptors. Free Con A-Sepharose beads could be separated from the bound beads (cell-bead complexes) by sedimentation of the high density beads and floatation of the low density complexes. Sedimented and total beads could be determined by counting the radioactivity associated with [-125I]Con A coupled in tracer amounts to the beads. Various lines of evidence demonstrated the high specificity of binding. Soluble Con A, but neither insulin nor any of the other proteins tested, inhibited and reversed the binding of Con A-Sepharose to the cells. Whereas treatment of Con A- (and insulin-) derivatized beads with anti-insulin antiserum, and cells with trypsin, readily inhibited binding of insulin-Sepharose to cells, neither treatment inhibited Con A-Sepharose binding. According to the relative extents of inhibition and reversal of binding exhibited by 15 different carbohydrates, the saccharide binding sites on Con A-Sepharose appeared virtually identical with the known sites on free Con A. Protein-containing components of cell ghosts that were solubilized with Triton X-100 appeared to correspond to the Con A-Sepharose receptor sites on the basis of their ability to bind to Con A-Sepharose columns, be eluted with methyl alpha-D-mannopyranoside (MeMan) and be precipitated by the free lectin and redissolved by MeMan. According to (a) Normarski interference contrast microscopic examination of the topographical distribution of Con A-Sepharose beads and cells surrounding and bound to each other, and (b) absence of any apparent morphological changes in the cells due to binding, it is suggested that extensive clustering ("cap" or "macropatch" formation) of Con A receptors did not occur on the adipocyte as a consequence of the interaction of the cells with the Con A-Sepharose beads.

Binding of hormones to receptors. An alternative explanation of nonlinear Scatchard plots

Binding of 1251-labeled hormones to receptors can usually be inhibited by addition of unlabeled hormone. In analyzing such binding-inhibition data, it is commonly assumed that both labeled and unlabeled hormones are bound with equal affinity. When this assumption is made incorrectly, an artifactually nonlinear Scatchard plot results. Equations to describe these nonlinear Scatchard plots are derived. These results are discussed with regard to previously published observations of nonlinear Scatchard plots for binding of insulin to its receptor.

Decreased binding of insulin to liver plasma membrane receptors in hereditary diabetic mice

The interaction of insulin with its receptors was studied in liver plasma membranes of the young non-obese hereditary diabetic mouse (KK strain). Under identical conditions of preparation and incubation, the membranes of the KK mouse bind only 55-70% as much insulin per mg of protein as those of the control mouse (Swiss albino). Scatchard analysis suggests that this decrease in binding is due to a decrease in the number of receptor sites in the membrane of the diabetic mouse. However, the membranes of diabetic and control mice do not exhibit significant differences in hexosamine and sialic acid contents, enzyme activities, and protein and glycoprotein analysis. The decrease in insulin receptors in the KK mouse seems to correlate with the insulin resistance which they exhibit.

Leukocytic epinephrine receptors of normal and asthmatic individuals

To determine if the hypothesized beta adrenergic blockade in asthma is located at the level of the hormone receptor, we have compared the number and binding affinity of leukocytic epinephrine receptors in normal and asthmatic subjects. Human leukocytes, but not human erythrocytes, possess low-affinity epinephrine receptors. When saturated, each leukocyte binds approximately 1.0 times 10-6 molecules of epinephrine. Binding of 3H DL-epinephrine was largely inhibited by excess unlabeled L-epinephrine. The binding was reversible and involved both the D and L forms of 3H DL-epinephrine. Inhibition studies with nonradioactive isoproterenol, norepinephrine, propranolol, and tolazoline produced results consistent with the interpretation that the leukocytes contained both alpha and beta adrenergic receptors. Two procedures, subcellular fractionation of lymphocytes and incubation of leukocytes in 0.1 per cent trypsin, permitted the demonstration that most of the catecholamine binding occurred at the plasma membrane. Thin-layer chromatography of the bound 3H DL-epinephrine after its extraction from the leukocytes permitted the interpretation that the hormone had fully retained its chemical structure. In addition, epinephrine binding was associated with cAMP production. Leukocytic epinephrine receptors of 10 asthmatic and 9 normal individuals were compared and found not to be substantially different in number or binding affinity.

Cellular binding sites for insulin in rat liver

A study of the sites of insulin binding in subcellular fractions of rat liver is reported. A method for the isolation of liver plasma membranes, which permits one to follow quantitatively the distribution of all the parameters of interest, was modified and applied to the study of the cellular topography of insulin binding. The insulin binding capacity did not follow closely the enzyme marker (5'-nucleotidase) for plasma membranes when differential centrifugation schemes were used, and the divergence from this marker was more prominent when separations were performed on discontinous sucrose gradients. A significant amount of insulin binding capacity was always present in fractions with higher density than those containing the majority of 5'-nycleotidase. Results of studies on linear sucrose gradients have disclosed in some of the purified membrane fractions small but consistent differences in density of the insulin binding, and plasma membrane particles. It is suggested that there may be several types of intracellular membranes to which insulin can bind besides the plasma membranes.