Effects of fasting on insulin binding, glucose transport, and glucose oxidation in isolated rat adipocytes: relationships between insulin receptors and insulin action.

UNLABELLED: Insulin binding, glucose transport, and glucose oxidation were studied in isolated adipocytes obtained from fasting rats. Fasting led to an increase in the overall binding affinity for insulin, while the number of receptor sites per cell remained constant. Glucose oxidation was markedly attenuated during fasting. Basal rates of oxidation decreased by about 50%, while insulin-stimulated rates decreased 6 to 10-fold. Glucose transport was assessed by measuring initial uptake rate of 2-deoxy-glucose. Fasting led to a 40-50% decrease in the apparent maximal transport capacity (Vmax) of 2-deoxy-glucose uptake with no change in apparent Km. A progressive decrease in basal and insulin-stimulated rates of 2-deoxy-glucose uptake was seen from 24-72 h of starvation and a significant correlation (r=0.85, P less than 0.001) existed between basal and maximal insulin-stimulated uptake rates in individual animals. When 2-deoxy-glucose uptake was plotted as a function of insulin bound, due to the decrease in maximal uptake capacity, cells from fasting animals took up less hexose for any amount of insulin bound. When the insulin bound was plotted as a function of the percent insulin effect on uptake, control cells and cells from 24-h-fasted rats gave comparable results, while cells from 48- and 72-h-fasted animals still took up less hexose for any amount of bound insulin. The effects of fasting on 3-O-methyl glucose uptake were comparable to the 2-deoxy-glucose data. IN CONCLUSION: (a) insulin binding is increased during fasting due to an increased overall binding affinity with no change in receptor number; (b) glucose oxidation is severely impaired during fasting; (c) 2-deoxy-glucose uptake decreases with fasting due to a decrease in maximal transport capacity (Vmax) with no change in Km; (d) the decrease in glucose oxidation is much greater than the decrease in glucose transport, indicating impaired intracellular oxidative metabolism; and (e) coupling between insulin receptors and the glucose transport system is normal after 24 h of fasting but is impaired at 48 and 72 h.

Effect of dexamethasone on insulin binding, glucose transport, and glucose oxidation of isolated rat adipocytes.

We have studied the in vitro effects of dexamethasone on isolated rat adipocytes at concentrations of dexamethasone therapeutically achieved in man. Glucose oxidation, glucose transport, and insulin binding were assessed. In dexamethasone-treated cells, glucose oxidation was decreased by 30-40% both in the absence of insulin (basal state) and at low insulin levels (less than 25 mu/ML). At maximally effective insulin levels (over 100 muU/ml) no differences existed between control and treated cells. If glucose transport were the rate-limiting step for glucose oxidation in the basal state and at low (submaximal) insulin levels, but not at maximally effective insulin concentrations, then these data could be explained by postulating that dexamethasone has a direct affect on glucose transport and does not affect intracellular oxidative pathways. We tested this hypothesis by directly assessing glucose transport in dexamethasone-treated cells. Glucose transport was assessed by measuring the uptake of [14C]2-deoxy glucose. These studies demonstrated a 30-40% decrease in 2-deoxy glucose uptake by treated cells both in the basal state and at all insulin concentrations. Thus, a direct glucocorticoid effect on the glucose transport system seems to account for the decreased ability of dexamethasone-treated cells to oxidize glucose. Since dexamethasone treatment leads to decreased insulin binding to adipocytes in vivo, we examined the possibility that the in vitro decreases in insulin-mediated glucose transport could be due to decreased insulin receptors. Insulin binding to control and treated adipocytes was measured, and no differences were found. Therefore, in cntrast to previously reported in vivo studies, adipocytes treated in vitro with dexamethasone retain a normal ability to bind insulin. Thus, these studies suggest that all of the in vitro effects of dexamethasone on glucose oxidation are due to direct inhibition of the glucose transport system.

The effects of spontaneous obesity on insulin binding, glucose transport, and glucose oxidation of isolated rat adipocytes.

UNLABELLED: We have studied insulin, binding, glucose transport, and glucose oxidation, using large adipocytes isolated from older, fatter rats (greater than 12-mo-old, greater than 550 g), and smaller cells obtained from younger, leaner animals (4-5-wk-old, 120-160 g). At media glucose levels less than 5 mM, basal (absence of insulin) rates of glucose oxidation are comparable in both groups of cells. However, in the presence of insulin, the increase in glucose oxidation is much greater in the smaller cells. Maximally effective insulin levels could not overcome the defect in glucose oxidation by larger cells, and thus, even though studies of insulin binding demonstrated a 30-40% decrease in insulin receptors on the larger cells, it is probable that the defect in glucose oxidation is distal to the insulin receptor. Glucose transport was assessed by direct measurement of 2-deoxy glucose uptake. Basal levels of uptake were greater for the larger cells, whereas at maximally effective insulin concentrations, rates of 2-deoxy glucose uptake were the same for both groups of cells. Thus, in the presence of maximally effective levels of insulin, the apparent Km (2.3-2.7 mM) and Vmax values (2.6 and 2.7 nmol/10(5) cells per min) of 2-deoxy glucose uptake were comparable, indicating that the glucose transport system of the larger cells was intact. However, at submaximal levels of insulin, small adipocytes took up more 2-deoxy glucose than larger cells. These findings represent a rightward shift in the insulin dose-response curve in the cells from the older, fatter animals, and this is the predicted functional sequelae of the observed decrease in insulin receptors. Finally, when the amount of insulin bound was plotted as a function of 2-deoxy glucose uptake, no difference was seen between both groups of cells. This indicates that coupling between insulin receptor complexes and the glucose transport system is intact in large adipocytes, and is further evidence that a defect(s) in intracellular glucose metabolism is responsible for the decrease in glucose oxidation of adipocytes from older, fatter rats. IN CONCLUSION: (a) insulin-mediated glucose oxidation is markedly decreased in large adipocytes from older, fatter rats, and since this decrease cannot be corrected by maximally effective insulin levels, the defect is probably distal to the insulin receptor; (b) the glucose transport system is basically normal in large adipocytes; (c) insulin binding to receptors is decreased in large cells and the functional sequelae of this decrease in insulin binding i.e., a rightward shift in the insulin dose-response curve for 2-deoxy glucose uptake, was observed, and (d) since the decreased rates of insulin-mediated glucose oxidation can not be attributed to changes in insulin receptors or to changes in glucose transport, an intracellular defect in glucose metabolism is suggested.

Decreased insulin binding to adipocytes and circulating monocytes from obese subjects.

UNLABELLED: Insulin binding to isolated adipocytes from 16 normal and 14 obese patients was studied. The data indicated that, as a group, adipocytes from the obese patients bound significantly less insulin than normal. However, of the 14 obese patients, 5 were not hyperinsulinemic and 4 of these 5 subjects had normal insulin binding. These subjects were also younger, and had the onset of obesity in childhood. When these five patients were separated from the original 14 obese patients, enhanced differences in insulin binding to adipocytes were observed when normals and the remaining 9 obese subjects were compared. Similar findings were obtained with isolated circulating mononuclear cells from these same patients. Presumably the five normoinsulinemic obese patients were not insulin-resistant, and, thus, the data indicate that insulin binding to adipocytes was decreased only in insulin-resistant obese patients. This conclusion was strengthened by finding a highly significant correlation (r=-0.71, p less than 0.001) between insulin binding to adipocytes and fasting plasma insulin level, while a weaker correlation (r=-0.49,p less than 0.01) existed between insulin binding and degree of obesity. Finally, when insulin binding to adipocytes and mononuclear cells from the same individual was compared, a significant positive correlation was found (r=0.53,p less than 0.01). IN CONCLUSION: (a) insulin binding to adipocytes and mononuclear cells is decreased in cells from insulin-resistant obese patients; (b) a significant inverse relationship exists between fasting plasma insulin level and insulin binding to adipocytes; and (c) in obesity, events that affect insulin receptors on adipocytes similarly affect insulin receptors on mononuclear cells.

In vivo kinetics of insulin action on peripheral glucose disposal and hepatic glucose output in normal and obese subjects.

To determine whether abnormal kinetics of insulin's biologic actions contribute to the overall insulin resistance in obesity, we compared the rate of activation and deactivation of insulin's effects to stimulate glucose disposal rate (Rd) and inhibit hepatic glucose output (HGO) in 12 nonobese and 10 obese subjects using the euglycemic clamp technique at insulin infusion rates of 15, 40, 120, and 1,200 mU/M2 per min. In both groups, stimulation of Rd was faster the higher the insulin infusion rate and the time to reach half maximal stimulation (A50 value) in normals was 52 +/- 4, 44 +/- 2, 29 +/- 3, and 21 +/- 2 min at infusion rates of 15, 40, 120, and 1,200 mU/M2 per min, respectively. In the obese subjects, the rate of activation was slower (higher A50 values) with A50 values of 74 +/- 6, P less than 0.001 (compared to normal), 64 +/- 8 min, P less than 0.001, and 28 +/- 3 min, P less than 0.01, at the 40, 120, and 1,200 mU/M2 per min insulin infusions. Deactivation of the insulin effect to stimulate glucose disposal rate (Rd) was faster in the obese group compared with normal individuals after all comparable insulin infusions. In summary: for both groups, the higher the insulin infusion rate, the higher the steady state Rd value, the faster the rate of activation and the slower the subsequent rate of deactivation. In insulin-resistant obese subjects, the rate of activation of insulin action was slower and the rate of deactivation faster at comparable insulin infusion rates. The rate of suppression of HGO was comparable in normal and obese subjects, but the rate of recovery of HGO back to basal values was faster in the obese group. And in view of the phasic manner in which insulin is normally secreted following meals, steady state insulin action is not normally achieved. Therefore, the abnormal kinetics of insulin action in insulin-resistant obese individuals may represent functionally important manifestations of the insulin resistance in this condition.

Direct in vitro effect of a sulfonylurea to increase human fibroblast insulin receptors.

We have studied the effects of the oral sulfonylurea agent glyburide to modulate insulin receptors on nontransformed human fibroblasts in tissue culture. When glyburide was added to monolayers of human fibroblasts, a dose-dependent increase in the number of cell surface receptors was observed with a maximum effect (19% increase) seen at 1 microgram/ml glyburide. Insulin can induce a loss of insulin receptors in these cells, and when fibroblasts are exposed to 100 ng/ml insulin for 6 h, approximately 60% of the initial complement of cell surface receptors are lost. When the process of insulin-induced receptor loss (or down regulation) was studied in the presence of glyburide, the drug exerted a marked inhibitory effect on this regulatory process. Thus, glyburide inhibited insulin-induced receptor loss in a dose-dependent fashion, and the maximally effective drug concentration (1 microgram/ml) inhibited 34% of the receptor loss. These studies demonstrate a direct in vitro effect of this oral hypoglycemic agent to increase the number of cell surface insulin receptors or prevent their loss, presumably by slowing the rate of receptor internalization. These findings may explain the well known extrapancreatic effect of sulfonylurea agents to improve insulin-mediated tissue glucose metabolism.

Effects of insulin incubation on insulin binding, glucose transport, and insulin degradation by isolated rat adipocytes. Evidence for hormone-induced desensitization at the receptor and postreceptor level.

We have examined the effect of in vitro hyperinsulinemia on insulin binding, glucose transport, and insulin degradation in isolated rat adipocytes. When cells were incubated with insulin for 2 or 4 h at 37 degrees C, followed by washing in insulin-free buffer to remove extracellular and receptor-bound insulin, a time and dose-dependent decrease in insulin receptors was observed, which was accompanied by a reduced ability of cells to degrade insulin. Furthermore, the quantitatively predicted rightward shift in the insulin-glucose transport dose-response curve could be demonstrated. In addition to this reduction in insulin sensitivity, a striking decrease in maximal insulin-stimulated glucose transport was observed in the 4-h insulin-treated cells, indicating an abnormality distal to the insulin receptor. Thus, in vitro insulin-induced insulin resistance in adipocytes is caused by both receptor and postreceptor abnormalities. The post-receptor defect is most likely at the level of the glucose transport system per se because the insulinlike agents, spermine and antiinsulin receptor antibodies, also had a markedly reduced ability to stimulate glucose transport in 4-h insulin-treated cells. On the other hand, when cells were incubated with 100 ng/ml insulin for up to 4 h, after which time 2-deoxy glucose uptake was measured without removing buffer insulin or allowing receptor-bound insulin to dissociate, no decrease in maximal insulin-stimulated glucose transport was found. In conclusion, (a) insulin leads to a dose-dependent loss of insulin receptors in freshly isolated adipocytes accompanied by the predicted functional consequence of decreased receptors, i.e., a rightward shift in the insulin-glucose transport dose-response curve, (b) prolonged incubation with insulin causes a marked postreceptor defect in the glucose transport system, (c) maintenance of the activated state of the glucose transport system prevents the expression of the post-receptor defect, (d) the location of the postreceptor abnormality is most likely in the glucose transport system per se, and (e) insulin-induced receptor loss is accompanied by a decrease in insulin degradation.

Decreased insulin binding to lymphocytes from diabetic subjects.

UNLABELLED: We have studied insulin binding to circulating lymphocytes isolated from 20 untreated adult, nonobese, nonketotic, diabetic subjects with fasting hyperglycemia, 20 normal subjects, and four patients with fasting hyperglycemia secondary to chronic pancreatitis. The results of these studies show that lymphocytes from diabetic patients have decreased ability to specificity bind insulin when compared to lymphocytes from normal subjects. For example, when lymphocytes from diabetic patients and a trace amount of [(125)I]insulin (3.3 x 10(-11) M) were incubated, binding was less than 50% of the value obtained with lymphocytes from normal subjects (2+/-0.2% vs. 4.2+/-0.4%). Furthermore, the data show that lymphocytes from diabetic patients have only 1,200 insulin receptor sites per cell compared to 2,200 sites per cell for lymphocytes from normal subjects. Competitive inhibition studies using unlabeled insulin indicate that the affinity for insulin of lymphocytes from both groups is comparable. Consequently the decreased insulin binding of diabetics' lymphocytes is primarily due to a decreased number of available receptors rather than decreased binding affinity. This decreased insulin binding is not due to chronic hyperglycemia since insulin binding to lymphocytes, obtained from four patients with fasting hyperglycemia secondary to chronic pancreatitis, was completely normal. The possibility that some factor present in the plasma of diabetic patients could cause decreased insulin binding also seems unlikely since we could demonstrate no in vitro effects of diabetics' plasma on insulin binding. Lastly, the proportion of lymphocytes which were thymus derived and bone marrow derived were the same in each of the study groups indicating that differences in lymphocyte subpopulations do not account for our results. IN CONCLUSION: (a) lymphocytes from nonobese, untreated, adult diabetic patients with fasting hyperglycemia demonstrate a decreased ability to bind insulin; (b) this decreased insulin binding to lymphocytes obtained from diabetic patients can be accounted for primarily by an absolute decrease in the number of available receptor sites per cell; and (c) these data suggest that this defect in insulin binding is a primary phenomenon.

Mechanisms of fasting-induced increase in insulin binding to rat adipocytes.

UNLABELLED: Fasting leads to an increase in the ability of adipocytes to bind insulin, and this was accounted for by an increase in the affinity of the receptors for insulin without any change in the number of receptors per cell. Binding affinity can increase because of a decrease in the dissociation rate constant (k(d)), an increase in the association rate constant (k(a)), or both. Kinetic studies demonstrated that fasting leads to a striking decrease in the rate at which insulin dissociates from its receptor, and the near two-fold prolongation of the time at which 50% of the bound (125)I-insulin dissociates (28+/-4 vs. 50+/-5 min) correlated quite well with the two-fold increase in binding affinity. On the other hand, the rate at which insulin associates with its receptor was essentially unchanged. Negatively cooperative interactions between receptors were readily demonstrated in cells from control and fasting animals, and the magnitude and sensitivity of this effect was the same in both groups of cells. It seemed likely that during fasting a change in the concentration of some substrate or hormone could lead to these effects on insulin binding. However, in vitro attempts to recreate the substrate and hormonal changes which occur in fasting produced no evidence to support this idea. IN CONCLUSION: (a) fasting leads to an increase in the ability of adipocytes to bind insulin because of an increase in binding affinity; (b) this increase in the affinity of the receptor for insulin was primarily accounted for by a decrease in the rate at which insulin dissociates from its receptors; and (c) fasting did not appreciably alter the negatively cooperative interactions displayed by adipocyte insulin receptors.

Long-term regulation of adipocyte glucose transport capacity by circulating insulin in rats.

We have tested the idea that the circulating plasma insulin level plays an important role in the long-term regulation, or maintenance, of the cellular glucose transport system, distinct from insulin's ability to acutely accelerate glucose transport. To study this hypothesis, groups of rats were made either hyperinsulinemic or hypoinsulinemic by daily insulin injections, or streptozotocin treatment, respectively. Different levels of hypoinsulinemia were produced by using different doses of streptozotocin (40 and 55 mg/kg). The mean (+/-SE) 9 a.m. plasma insulin level for each experimental group was: hyperinsulinemic animals, 65+/-5 muU/ml; controls, 32+/-3 muU/ml; low dose streptozotocin group, 18+/-3 muU/ml; and high dose streptozotocin group 5+/-2 muU/ml. Isolated adipocytes were prepared from each animal and glucose transport was assessed by measuring the initial rates of uptake of the nonmetabolyzable hexose 2-deoxy glucose. The V(max) and K(m) values for adipocyte glucose transport were calculated from the 2-deoxy glucose uptake data. The results demonstrated that in cells from control animals the V(max) of in vitro adipocyte glucose transport was 7.1+/-0.7 nmol/min per 10(6) cells in the basal state and 22.9+/-0.9 nmol/min per 10(6) cells in the presence of a maximally effective insulin concentration (25 ng/ml) in the buffer. In cells from the experimentally hyperinsulinemic animals these V(max) values were increased to 11.7+/-0.8 and 44.2+/-1.1 nmol/min per 10(6) cells. Using adipocytes from both groups of streptozotocin-treated (high dose, 55 mg/kg; low dose, 40 mg/kg) insulin-deficient diabetic animals, V(max) values were found to be progressively decreased. Thus, in the low dose group, basal-and insulin-stimulated V(max) values were 1.6+/-0.5 and 5.7+/-0.7 nmol/min per 10(6) cells, as compared to values of 0.9+/-0.2 and 1.7+/-0.6 in the high dose group. Thus, when considered as group data a positive relationship was found between circulating plasma insulin levels and adipocyte glucose transport V(max), with increased V(max) values in hyperinsulinemic rats and decreased V(max) values in hypoinsulinemic rats. Furthermore, when the individual data were analyzed, highly significant correlation coefficients were found between the height of the plasma insulin level and both the basal (r = 0.82, P < 0.001) and insulin-stimulated (r = 0.93, P < 0.001) V(max) values. The apparent K(m) for 2-deoxy glucose uptake was the same under all conditions. In conclusion, assuming that the V(max) of transport is some function of the number of glucose transport carriers per cell, then these results support the hypothesis that in addition to acute acceleration of glucose transport, insulin is also an important long-term regulator of the number of available adipocyte glucose transport carriers.

Glucose-induced phosphorylation of the insulin receptor. Functional effects and characterization of phosphorylation sites.

Elevated glucose concentrations have been reported to inhibit insulin receptor kinase activity. We studied the effects of high glucose on insulin action in Rat1 fibroblasts transfected with wild-type human insulin receptor (HIRcB) and a truncated receptor lacking the COOH-terminal 43 amino acids (delta CT). In both cell lines, 25 mM glucose impaired receptor and insulin receptor substrate-1 phosphorylation by 34%, but IGF-1 receptor phosphorylation was unaffected. Phosphatidylinositol 3-kinase activity and bromodeoxyuridine uptake were decreased by 85 and 35%, respectively. This was reversed by coincubation with a protein kinase C (PKC) inhibitor or microinjection of a PKC inhibitor peptide. Phosphopeptide mapping revealed that high glucose or PMA led to serine/threonine phosphorylation of similar peptides. Inhibition of the microtubule-associated protein (MAP) kinase cascade by the MAP kinase kinase inhibitor PD98059 did not reverse the impaired phosphorylation. We conclude that high glucose inhibits insulin action by inducing serine phosphorylation through a PKC-mediated mechanism at the level of the receptor at sites proximal to the COOH-terminal 43 amino acids. This effect is independent of activation of the MAP kinase cascade. Proportionately, the impairment of insulin receptor substrate-1 tyrosine phosphorylation is greater than that of the insulin receptor resulting in attenuated phosphatidylinositol 3-kinase activation and mitogenic signaling.

Endocytotic uptake, processing, and retroendocytosis of human biosynthetic proinsulin by rat fibroblasts transfected with the human insulin receptor gene.

The cellular itinerary and processing of insulin and proinsulin were studied to elucidate possible mechanisms for the observed in vivo differences in the biologic half-lives of these two hormones. A rat fibroblast cell line transfected with a normal human insulin receptor gene was used. Due to gene amplification, the cells express large numbers of receptors and are ideal for studying a ligand, such as proinsulin, that has a low affinity for the insulin receptor. Competitive binding at 4 degrees C showed that the concentration of unlabeled insulin and proinsulin that is needed to displace 50% of tracer insulin or proinsulin was 0.85-0.95 nM and 140-150 nM, respectively. Binding to surface receptors and internalization occur at rates that are four to five times faster in cells incubated with insulin compared with proinsulin. Chloroquine led to an increase in cell-associated radioactivity of approximately 1.4-fold in cells incubated with insulin or proinsulin, but inhibited the appearance of degraded insulin by 54% and degraded proinsulin by only 10%. To study the fate of internalized ligand, cells were incubated with insulin and proinsulin until steady state binding occurred. Surface bound ligand was removed by an acid wash and the remaining cell-associated radioactivity represented internalized ligand. Cells were then reincubated in 37 degrees C buffer and the cell-associated radioactivity and radioactivity released into the medium were analyzed by TCA precipitation, Sephadex G-50, and HPLC. The results demonstrated that proinsulin more readily bypasses the intracellular degradative machinery and is therefore released intact from the cell via the retroendocytotic pathway. These results may help to explain the prolonged metabolic clearance rate and biologic responsiveness of proinsulin in vivo.

Effects of aging on glucose-mediated glucose disposal and glucose transport.

To assess the effects of aging on glucose-mediated glucose disposal and glucose transport, glucose disposal rates were measured in 10 nonelderly (32 +/- 4 yr) and 11 elderly (64 +/- 4 yr) subjects at five different plasma glucose concentrations. Glucose disposal was decreased by 30-35% in the elderly at each level of glycemia (100-350 mg/dl) in the presence of similar levels of hyperinsulinemia (approximately 100 microU/ml), and the 50% effective concentration (EC50) was similar in both the nonelderly (100 +/- 9) and elderly (103 +/- 5 mg/dl). The Michaelis constant (Km) of 3-O-methyl glucose transport in adipocytes was unchanged with aging (3.8 +/- 0.5 vs. 3.2 +/- 0.2 mM) while the maximum velocity of insulin stimulated transport was reduced by 34% in the elderly (8.3 +/- 1.3 vs. 12.6 +/- 1.5 pmol/5 X 10(4) cells per s, P less than 0.05). The insulin resistance of aging is therefore due to a reduction in the capacity of the glucose uptake system, while the affinity of glucose utilization (EC50 and Km) is unchanged. This supports the hypothesis that a reduction in the number of glucose transport and metabolic units occurs with aging, but that each unit functions normally.

Insulin receptor down-regulation is linked to an insulin-induced postreceptor defect in the glucose transport system in rat adipocytes.

We have examined the relationship between insulin-induced receptor downregulation and the induction of a postreceptor defect in the insulin-stimulated glucose transport system in rat adipocytes, and found that downregulation was linked to the expression of the postreceptor defect. When recycling of insulin receptors was inhibited by 20 mM Tris, insulin pretreatment (100 ng/ml) for 4 h at 37 degrees C induced both net loss (65%) of cell-surface receptors and a 63% decrease in maximal insulin responsiveness. In contrast, when cells were treated with insulin alone for 4 h at 37 degrees C so that receptors could recycle, or treated at 16 degrees C with Tris plus insulin to inhibit receptor internalization, neither receptor downregulation nor a postreceptor defect was observed. Induction of the postreceptor defect was specific for insulin under conditions when downregulation would occur, since treatment of cells with Tris and the insulin mimicker spermine did not result in receptor loss or the postreceptor defect. Other experiments revealed that receptor downregulation occurred first without loss of insulin responsiveness, but, once the postreceptor defect appeared, its severity was correlated to the degree of further receptor loss, as a function of insulin dose and exposure time. Tris (20 mM) alone acutely decreased maximally stimulated glucose transport rates slightly (22%), but this effect was rapidly reversible after Tris removal and could not have been directly responsible for the lasting and profound postreceptor defect seen after pretreatment with insulin plus Tris. Taken together, these data suggest that insulin-induced receptor loss is linked to the induction of the postreceptor defect. The postreceptor defect was due to an inability to maximally increase the maximum velocity of glucose transport. Furthermore, the expression of the postreceptor defect depended upon the extent to which the glucose transport system was allowed to deactivate; maintaining the glucose transport system in an activated state prevented its expression. Thus, the mechanism could involve rapid inactivation or sequestration of glucose transporters during deactivation such that they become refractory to the subsequent stimulatory effects of insulin. In conclusion, (a) insulin does not acutely induce a postreceptor defect in the glucose transport system of adipocytes without loss of cell-surface insulin receptors; (b) the defect in stimulated glucose transport has been induced distal to the insulin receptor via a mechanism linked to receptor loss; and (c) the postreceptor lesion is due to decreased number of intrinsic activity of glucose transporters on the cell-surface in the presence of a maximally effective insulin concentration. These data suggest that insulin receptor downregulation and postreceptor defects in insulin action, which frequently co-exist both in vivo and in vitro, may be linked mechanistically.

Mechanism of the postreceptor defect in insulin action in human obesity. Decrease in glucose transport system activity.

We have studied insulin-stimulated 3-O-methyl glucose transport by isolated adipocytes prepared from 10 normal and 11 obese individuals. The results demonstrated that the insulin-glucose transport dose-response curves were shifted to the right in cells from the obese patients, and that the magnitude of this rightward shift was significantly correlated to the reduction in adipocyte insulin receptors in individual subjects (r = 0.48, P less than 0.01). In three obese patients a rightward shift in the dose-response curve could be demonstrated and there was no decrease in maximal insulin effect. This corresponded to in vivo glucose clamp results showing only a rightward shift in the insulin dose-response curve for overall glucose disposal in these three subjects (1980. J. Clin. Invest. 65: 1272-1284). In the remaining eight obese patients, the in vitro glucose transport studies showed not only a rightward shift in the dose-response curves but also a marked decrease in basal and maximally insulin-stimulated rates of transport, indicating a postreceptor defect in insulin action. Again, this was consistent with the in vivo glucose clamp studies demonstrating a marked postreceptor defect in these individuals. In conclusion, these results indicate that the mechanism of the postreceptor defect in insulin action, which exists in many obese patients, is related to a decrease in the activity of the glucose transport effector system.

Equivalence of the insulin sensitivity index in man derived by the minimal model method and the euglycemic glucose clamp.

Studies were done to determine whether the minimal model approach and the glucose clamp measure equivalent indices of insulin action. Euglycemic glucose clamps (glucose, G: 85 mg/dl) were performed at two rates of insulin (I) infusion (15 and 40 mU/min per m2) in 10 subjects (body mass index, BMI, from 21 to 41 kg/m2). Insulin sensitivity index (SI) from clamps varied from 0.15 to 3.15 (mean: 1.87 +/- 0.36 X 10(-2) dl/[min per m2] per microU/ml), and declined linearly with increasing adiposity (versus BMI: r = -0.97; P less than 0.001). SI from modeling the modified frequently sampled intravenous tolerance test varied from 0.66 to 7.34 X 10(-4) min-1 per microU/ml, and was strongly correlated with SIP(clamp) (r = 0.89; P less than 0.001). SI and SIP(clamp) were similar (0.046 +/- 0.008 vs. 0.037 +/- 0.007 dl/min per microU/ml, P greater than 0.35); the relation had a slope not different from unity (1.05 P greater than 0.70) and passed through the origin (P greater than 0.40). However, on a period basis, SI exceeded SIP(clamp) slightly, due to inhibition of hepatic glucose output during the FSIGT, not included in SIP(clamp). These methods are equivalent for assessment of overall insulin sensitivity in normal and insulin-resistant nondiabetic subjects.

Role of glucose transporters in the cellular insulin resistance of type II non-insulin-dependent diabetes mellitus.

To examine the role of glucose transport proteins in cellular insulin resistance, we studied subcutaneous adipocytes isolated from lean control, obese control (body mass index [BMI] 33.4 +/- 0.9), and untreated obese non-insulin-dependent diabetes mellitus (NIDDM) patients (BMI 35.2 +/- 2.1; fasting glucose 269 +/- 20 mg/dl). Glucose transporters were measured in plasma membrane (PM), low-density (LDM), and high-density (HDM) microsomal subfractions from basal and maximally insulin-stimulated cells using the cytochalasin B binding assay, and normalized per milligram of membrane protein. In all subgroups, insulin led to an increase in PM glucose transporters and a corresponding depletion of transporters in the LDM. Insulin recruited 20% fewer transporters to the PM in the obese subgroup when compared with lean controls, and this was associated with a decline in LDM transporters with enlarging cell size in the control subjects. In NIDDM, PM, and LDM, transporters were decreased 50% in both basal and stimulated cells when compared with obese controls having similar mean adipocyte size. Cellular depletion of glucose transporters was not the only cause of insulin resistance, because the decrease in rates of [14C]-D-glucose transport (basal and insulin-stimulated) was greater than could be explained by reduced numbers of PM transporters in both NIDDM and obesity. In HDM, the number of transporters was not influenced by insulin and was similar in all subgroups. We conclude that (a) in NIDDM and obesity, both reduced numbers and impaired activity of glucose transporters contribute to cellular insulin resistance, and (b) in NIDDM, more profound cellular insulin resistance is associated primarily with a further depletion of cellular transporters.

Reversibility of defective adipocyte insulin receptor kinase activity in non-insulin-dependent diabetes mellitus. Effect of weight loss.

Insulin-stimulated kinase activity of adipocyte-derived insulin receptors is reduced in subjects with non-insulin-dependent diabetes mellitus (NIDDM) but normal in obese nondiabetics. To assess the reversibility of the kinase defect in NIDDM, insulin receptor kinase activity was measured before and after weight loss in 10 NIDDM and 5 obese nondiabetic subjects. Peripheral insulin action was also assessed in vivo by glucose disposal rates (GDR) measured during a hyperinsulinemic (300 mU/M2 per min) euglycemic clamp. In the NIDDMs, insulin receptor kinase activity was reduced by 50-80% and rose to approximately 65-90% (P less than 0.01) of normal after 13.2 +/- 2.0 kg (P less than 0.01) weight loss; comparable weight loss (18.2 +/- 1.5 kg, P less than 0.01) in the nondiabetics resulted in no significant change in insulin receptor kinase activity. Relative to GDR measured in lean nondiabetics, GDR in the NIDDMs was 35% of normal initially and 67% (P less than 0.01) of normal after diet therapy; weight loss in the nondiabetics resulted in an increase in GDR from 53 to 76% of normal (P less than 0.05). These results indicate that the insulin receptor kinase defect that is present in NIDDM is largely reversible after weight reduction. In contrast, the improvement in GDR, in the absence of any change in insulin receptor kinase activity in the nondiabetics, suggests that the main cause of insulin resistance in obesity lies distal to the kinase.

Human adipocyte glucose transport system. Biochemical and functional heterogeneity of hexose carriers.

We have investigated glucose transport proteins in isolated human adipocytes. Using the cytochalasin B binding assay to measure glucose transporters in subcellular membrane subfractions, we found that insulin induced translocation of intracellular glucose transporters to the cell surface. Isoelectric focusing of glucose transporters photolabeled with [3H]cytochalasin B revealed two distinct glucose transporter isoforms in low density microsomes focusing at pH 5.6 and pH 6.4, but only the pH 5.6 isoform was detectable in plasma membranes and only the pH 6.4 form was found in the high density microsomes. Insulin recruited only the pH 5.6 glucose transporter from the low density microsomes to the plasma membrane with no effect on the pH 6.4 transporter isoform. The results suggest that the pH 6.4 species is an immature form of the glucose transporter initially located in the high-density microsome fraction, which then migrates to the low-density microsomes where it matures (converted to pH 5.6 species) and becomes available for insulin-mediated recruitment to the plasma membrane.

Mechanisms of insulin resistance in aging.

We have studied 17 elderly and 27 non-elderly, nonobese subjects (mean age 69+/-1 and 37+/-2 yr, respectively) to assess the mechanisms responsible for the abnormal carbohydrate tolerance associated with aging. Serum glucose and insulin levels were significantly elevated in the elderly subjects compared with the nonelderly subjects during a 75-g oral glucose tolerance test, suggesting an insulin resistant state. Peripheral insulin sensitivity was assessed in both groups using the euglycemic glucose clamp technique during an insulin infusion rate of 40 mU/m(2) per min. Similar steady-state serum insulin levels led to a peripheral glucose disposal rate of 151+/-17 mg/m(2) per min in the elderly compared with a value of 247+/-12 mg/m(2) per min in the nonelderly, thus documenting the presence of insulin resistance in the elderly subjects. Insulin binding to isolated adipocytes and monocytes was similar in the elderly and nonelderly groups (2.34+/-0.33 vs. 2.62+/-0.24% and 5.04+/-1.10 vs. 5.12+/-1.07%), respectively. Thus, insulin resistance in the presence of normal insulin binding suggests the presence of a postreceptor defect in insulin action. This was confirmed by performing additional euglycemic clamp studies using infusion rates of 15 and 1,200 mU/m(2) per min to assess the contours of the dose-response relationship. These studies revealed a 39 and 25% decrease in the glucose disposal rate in the elderly subjects, respectively. The results confirm the presence of a postreceptor defect as well as a rightward shift in the dose-response curve. Insulin's ability to suppress hepatic glucose output was less in the elderly subjects during the 15 mU/m(2) per min insulin infusion (77+/-5 vs. 89+/-4% suppression), but hepatic glucose output was fully and equally suppressed in both groups during the 40 and 1,200 mU/m(2) per min infusion. Finally, a significant inverse relationship was observed between the degree of glucose intolerance in the individual elderly subjects, as reflected by the 2-h serum glucose level during the oral glucose tolerance test, and the degree of peripheral insulin resistance as assessed by the glucose disposal rate during the 40 mU/m(2) per min insulin infusion (r = 0.59, P < 0.01).We conclude that carbohydrate intolerance develops as part of the aging process. This carbohydrate intolerance appears to be the consequence of peripheral insulin resistance caused by a postreceptor defect in target tissue insulin action, which causes both a decrease in the maximal rate of peripheral glucose disposal and a rightward shift in the insulin action dose-response curve. In elderly subjects, the severity of the abnormality in carbohydrate tolerance is directly correlated to the degree of peripheral insulin resistance.