Elevation of the number of cell-surface insulin receptors and the rate of 2-deoxyglucose uptake by exposure of 3T3-L1 adipocytes to tolbutamide.

Sulfonylurea compounds are hypoglycemic agents which by unknown mechanisms alter the amount of insulin receptor and the rate of glucose utilization in tissues exposed to the drugs. In this study the effects on insulin binding and uptake of 2-deoxyglucose by 3T3-L1 adipocytes were assessed after maintaining cell monolayers for 1-3 days in medium containing different concentrations of the sulfonylurea, tolbutamide. The amount of 125I-insulin bound by treated monolayers gradually increased to values 150-250% of those of control monolayers after 2-3 days of exposure to 1.5 mM tolbutamide. Such increases in insulin binding capacity arose primarily from an increase in receptor number and not from an alteration in the affinity of the receptor for insulin. Concomitant with the changes observed for the insulin receptor, tolbutamide-treated monolayers expressed 1.5-2-fold higher rates of uptake of 2-deoxyglucose relative to control monolayers at concentrations of insulin between 0 and 10(-10) M. This study thus demonstrates the responsiveness of adipocytes to tolbutamide and also establishes the usefulness of 3T3-L1 cells as a model system in which to study the mechanism of tolbutamide action, both as it relates to the use of sulfonylurea compounds in clinical applications and as possible probes for perturbing and studying relatively uncharacterized regulatory pathways controlling receptor level and biological responses to insulin.

Sulfonylurea compounds are hypoglycemic agents which by unknown mechanisms alter the amount of insulin receptor and the rate of glucose utilization in tissues exposed to the drugs. In this study the effects on insulin binding and uptake of 2-deoxyglucose by 3T3-Ll adipocytes were assessed after maintaining cell monolayers for 1-3 days in medium containing different concentrations of the sulfonylurea, tolbutamide. The amount of lasI-insulin bound by treated monolayers gradually increased to values 150-25090 of those of control monolayers after 2-3 days of exposure to 1.5 m~ tolbutamide. Such increases in insulin binding capacity arose primarily from an increase in receptor number and not from an alteration in the affinity of the receptor for insulin. Concomitant with the changes observed for the insulin receptor, tolbutamidetreated monolayers expressed 1.5-2-fold higher rates of uptake of 2-deoxyglucose relative to control monolayers at Concentrations of insulin between 0 and 10"' M. This study thus demonstrates the responsiveness of adipocytes to tolbutamide and also establishes the usefulness of 3T3-Ll cells 8s a model system in which to study the mechanism of tolbutamide action, both as it relates to the use of sulfonylurea compounds in clinical applications and as possible probes for perturbing and studying relatively uncharacterized regulatory pathways controlling receptor level and biological responses to insulin.
Administration of sulfonylurea drugs, alone or in combination with insulin therapy and diet control, has been shown to ameliorate hyperglycemia and restare insulin sensitivity in some patients (1)(2)(3). The initial effects of sulfonylurea administration appear to arise from acute stimulation of the rate of insulin release from the pancreas, thus providing a. sufficient increase in the concentration of circulating insulin to reduce plasma glucose concentrations (1,2,4). Some patients continue to demonstrate reduced plasma glucose concentrations and increased insulin binding to isolated cells after chronic * This work was supported by United States Public Health Service Grant AM 29717, the Robert A. Welch Foundation Grant 1-857, and in part by the American Heart Association Grant 83-1241, with funds contributed in part by the American Heart Association, Texas Affiliate. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked ''advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
$To whom correspondence should be addressed: Department of Biochemistry, University of Texas Health Science Center, 5323 Harry Hines Boulevard, Dallas, TX 75235. treatment with sulfonylurea agents (3, 5,. 6). Thus, these compounds possess the interesting additional property of increasing glucose uptake and receptor number in peripheral tissues after prolonged exposure when the concentrations of plasma insulin have subsided to pretreatment levels. Similar findings have been reported in studies with animals exposed for prolonged intervals to various sulfonylurea agents (4, 7).
An understanding of the mechanism of action of sulfonylurea drugs can provide both a better rationale for their clinical use or disuse and possibly important insight into unknown mechanistic aspects of the regulation of receptor concentration and distribution within the cell.
The demonstrated effects of sulfonylurea agents on cellular receptor level and glucose utilization in uitro have been vanable. Sulfonylurea drugs do elevate the level of insulin receptors expressed by certain cells in culture (8). However, other comprehensive studies with lymphocyte, mammary carcinoma, and hepatoma cell lines exposed to a variety of sulfonylurea agents demonstrated no alterations in insulin receptor level (9). Although the sulfonylurea, tolazamide, did not affect the receptor level for rat adipose tissue in organ culture, it did elevate the rate of insulin-stimulated 2-deoxyglucose uptake (10).
The variability of observed effects of sulfonylurea compounds in uitro on receptor level and glucose utilization may reflect the specificity of tissue response, the type of sulfonylurea utilized, or the inability to duplicate the extended intervals of exposure of cells to sulfonylureas attained in vim. Although sulfonylurea agents can be shown to have rapid short-term effects on some cellular processes, such as ion flux (ll), longer intervals of exposure appear to be required to observe their effects on insulin receptor levels and on the rate of glucose uptake in uitro.
We have investigated the effects that different concentrations and time of exposure to the sulfonylurea, tolbutamide, have on the rate of basal and insulin-stimulated uptake of 2deoxyglucose, and on the level of insulin receptor expressed by 3T3-Ll adipocytes. Consistent with the chronic effects of sulfonylurea drugs on extrapancreatic tissues in uiuo, this study demonstrates that exposure of 3T3-Ll adipocytes to tolbutamide for intervals in excess of 2 days induces both a change in surface insulin receptor number and the characteristics of glucose uptake. The magnitude, of these responses to tolbutamide by 3T3-Ll adipocytes is larger than that normally observed in other in vitro systems and should facilitate further mechanistic studies of tolbutamide action in uitro. medium (25 mM glucose) containing 10% (v/v) fetal calf serum (Gibco and KC Biological) as described (12). At 2 days post-confluence, the monolayers were exposed to fresh medium (1.5 ml) containing 10% fetal calf serum, 0.5 mM isobutylmethylxanthine, 1 p~ dexamethasone, and 10 pg/ml of insulin (day 0). On day 2, the isobutylmethylxanthine and dexamethasone were removed, and the medium was replaced every 2 days with medium containing 10% fetal calf serum and 10 pg/ml insulin. By day 6, >90% of the cells expressed the adipocyte phenotype.
For earlier experiments presented in Figs. 1-3, the monolayers were washed several times with phosphate-buffered saline prior to replacing the medium with Dulbecco's modified Eagle's medium (25 mM glucose) containing 10% fetal calf serum with or without tolbutamide. For the experiments presented in Figs. 4-7, a more extensive wash protocol was employed to ensure rapid attainment of basal glucose transport levels, and efficient reversal of the insulin induced state of receptor down-regulation (13). Prior to each experiment, the monolayers were washed with 2 X 1.5 ml of Dulbecco's modified Eagle's medium (25 mM glucose) containing 1% (v/v) fetal calf serum and then incubated at 37 "C (5% COz, 95% air) for three 20-min intervals. The medium was replaced with fresh medium prior to each 20-min incubation. After the last 20-min incubation, the medium was replaced with medium containing 10% (v/v) fetal calf serum. Twentyfour hours later fresh medium was added with or without tolbutamide and/or insulin as indicated in the figure legends.
In all experiments in which the monolayers were analyzed for a period exceeding 2 days (Figs. 1-4, 6B, 6C, 7A, and 7B), the medium was replaced on the 2nd day with fresh medium containing the appropriate additions of tolbutamide and/or insulin.
Reagents-Crystalline porcine insulin was the generous gift of Dr. Amersham and diluted to a specific activity between 500-700 dpm/nmol using unlabeled 2-deoxyglucose (Sigma). Insulin Binding Assays-The monolayers were washed five times with 1.5 ml of Krebs-Ringer phosphate buffer, pH 7.4, containing 1% (w/v) bovine serum albumin and 25 mM glucose, incubated for 20 min at 37 "C after each of the final three washes, and then stored for 30-60 min at 4 "C. The binding assay was initiated by replacing the buffer with an identical buffer containing the appropriate concentrations of 1251-insulin in the presence or absence of 4.56 pM unlabeled insulin. After 5 h at 4 "C unbound insulin was removed by washing the monolayers rapidly five times with 1.5 ml of cold phosphatebuffered saline, pH 7.4, containing 0.1% bovine serum albumin. The amount of bound 12'I-insulin was then determined using a y-counter after scraping the cells into a vial with two 0.5-ml aliquots of 1% (v/v) Triton X-100 in water. Specific binding was determined by subtracting the amount of insulin bound in the presence of 4.6 p~ unlabeled insulin from the total amount of binding activity determined in the absence of unlabeled insulin. Both total and nonspecific binding were determined on duplicate plates. Results are expressed as the mean of the specific binding f range of the two determinations.
Assays for 2-Deoxyglucose Uptake-Monolayers were washed three times with 1.5 ml of Krebs-Ringer phosphate buffer, pH 7.4, containing 1% (w/v) bovine serum albumin. The buffer was replaced with 0.9 ml of identical buffer containing the appropriate concentrations of insulin prior to placing the monolayers in a 37 "C water bath. After 30 min, 100 pl of 2-deoxy-[U-'4C]glucose (500-700 dpm/nmol) were added to each dish (0.2 mM 2-deoxyglucose, final concentration) to initiate 2-deoxyglucose uptake. After 4 min, uptake of 2-deoxyglucose was terminated by rapidly washing the monolayers five times with 1.5 ml of cold phosphate-buffered saline. The amount of 2-deoxyglucose associated with the monolayers was then determined in a scintillation counter after scraping the cells into a vial uqing two 0.5-ml aliquots of 0.2 N NaOH and solubilizing the cells by incubation at 55 "C for 1 h. Uptake assays were performed on duplicate monolayers for each condition tested and the results expressed as the mean rC_ range of the two determinations.
Cell Number Determinations-The number of cells contained in each dish was determined by incubating the monolayers for 2-3 h at 37 "C in 2 ml of Krebs-Ringer phosphate buffer, pH 7.4, lacking calcium and magnesium, and containing 1% (w/v) bovine serum albumin, 25 mM glucose, and 5 mM EDTA. The number of cells released from duplicate dishes was then determined using a hemocytometer or Coulter counter. The values of insulin binding or 2deoxyglucose uptake determined per dish were normalized to cell number.
Statistical Methods-The parametric paired-sample t test and the nonparametric Wilcoxon paired-sample test (15) were used to compare the significance of paired differences in insulin binding or 2deoxyglucose uptake between tolbutamide-treated and untreated monolayers. Means obtained from the pooled data points from multiple experiments assaying insulin binding or 2-deoxyglucose uptake for either tolbutamide-treated or untreated monolayers were compared using a two-sample t test (15). Statistical tests for the presence of differences in binding or transport values obtained between tolbutamide-treated and untreated monolayers were performed by testing the ratio of the values (tolbutamide-treated/untreated) for equality to 1, or the difference (tolbutamide-treated minus untreated) for equality to 0 using a single-tailed t test. Confidence limits for the mean, mean ratio, and mean difference were calculated as described (15). Calculations were performed using an interactive statistical program package provided by the Medical Computing Resources Center, University of Texas Health Science Center at Dallas.

Effects of Tolbutamide on the Amount of Surface Insulin
Receptor Expressed by 3T3-Ll Adipocytes-The results presented in Fig. 1 demonstrate a tolbutamide-dependent increase in 1251-insulin binding to 3T3-Ll adipocytes. Control monolayers maintained in the absence of sulfonylurea bound approximately 0.028 pmol of insulin/106 cells under these assay conditions. Exposure of 3T3-U adipocytes for 3 days to 0.5, 1.0, or 2.5 mM tolbutamide increased the amount of insulin bound to levels 1.7-, 2.1-, and 2.5-fold those of control values, respectively. A loss of cells from the monolayer was observed at the highest concentration of tolbutamide tested, and suggests that decreased cell viability (25% drop in cell number/dish) with 5 mM tolbutamide may have contributed to the fall in insulin binding observed at this concentration. The experiment presented in Fig. 1 is representative of three separate experiments which demonstrated a maximal effect on insulin binding by exposure of the cells to concentrations of tolbutamide between 1-3 mM. For this range of tolbutamide concentrations the mean of the ratios of the amount of insulin bound by tolbutamide-treated and untreated monolayers (treated/untreated) was 1.87 +-0.38 (S.D., n = 6). The 95% confidence interval for the mean ratio was 1.47-2.27.

Kinetics for the Alteration by Tolbutamide of Insulin Binding to 3T3-Ll Adipocytes-To determine the kinetics by which
tolbutamide alters the level of '251-insulin binding to the cell surface of 3T3-Ll adipocytes, differentiated cells were treated with or without 1.5 mM tolbutamide. The binding of insulin to 3T3-Ll adipocytes was measured (using 0.5 nM 1251-insulin) as described under "Materials and Methods" at intervals between 0 and 96 h of exposure to tolbutamide. The results from one of two similar experiments are presented in Fig. 2. The amount of 1251-insulin bound by tolbutamide-treated cells rose significantly within 24 h. This rise, however, was only slightly greater than that observed in the control monolayers and is attributed predominantly to the reversal of the "downregulated" state of cell-surface insulin receptors present prior to removal of insulin. The level of receptor measured in control monolayers decreased gradually during the following 2 days .in culture. In contrast, exposure of 3T3-Ll cells to 1.

Effects of Tolbutamide on the Isotherms for 1251-Insulin Binding to 3T3-Ll Adipocytes-To assess whether changes in
receptor number or affiniby were responsible for the observed increase in insulin binding after exposure of 3T3-Ll cells to tolbutamide, monolayers were incubated in the presence or absence of 1.5 mM tolbutamide for 3 days. Specific insulin binding was then measured at eight different insulin concentrations. One set of binding isotherms obtained from tolbutamide-treated and untreated monolayers is presented in Fig.  3. At saturation, control monolayers bound approximately 0.6 pmol of insulin/106 cells. This is approximately 2-fold the level of receptor measured previously for down-regulated cells maintained in the presence of high levels of insulin (16-18), and is consistent with the anticipated increase in receptor number following the removal of insulin from the medium. As shown by the values in parentheses (Fig. 3), relative to control monolayers, tolbutamide-treated monolayers bound 1.6-2.0-fold the amount of insulin at all concentrations of insulin tested. The possibility of small tolbutamide-induced alterations in affinity of the receptor for insulin therefore cannot be excluded due to the observed variation in the ratio of binding of insulin to untreated and treated monolayers at nonsaturating concentrations of insulin. At saturating concentrations of insulin, tolbutamide-treated monolayers bound approximately 1.2 pmol of insulin/106 cells compared to 0.6 pmol of insulin/106 cells for untreated monolayers. Thus, in this experiment 3T3-Ll adipocytes exposed to tolbutamide expressed a level of insulin receptors on the cell surface which is 2-fold that of control monolayers, and 4-fold that of downregulated cells.
The extrapolated values for insulin binding at saturating concentrations of insulin were obtained from 13 similar sets of isotherms. The mean k standard deviation, and (95% confidence interval) for insulin binding to control monolayers were: 0.58 & 0.18 (0.47-0.69) pmol of insulin bound/106 cells. The corresponding values for insulin binding to tolbutamidetreated monolayers were: 0.97 f 0.35 (0.76-1.18) pmol of insulin bound/106 cells. The mean, standard deviation, and 95% confidence interval for the ratio of insulin binding (treated/untreated) were 1.67 2 0.32 (1.48-1.87). When the means, paired ratios (treated/untreated), or paired differences (treated minus untreated) were compared, tolbutamidetreated monolayers were found to bind a statistically greater insulin than untreated monolayers (p < 0.002).

Antagonistic Effects of Insulin and Tolbutamide on the Expression of the Insulin Receptor on the Cell Surface-To
determine what effects the continued presence of high levels of insulin (a persistent down-regulated state, see Ref. 13) would have on the ability of tolbutamide to elevate receptor levels, differentiated 3T3-Ll monolayers were treated with either no additions, 1.5 mM tolbutamide, 10 pg/ml insulin, or 1.5 mM tolbutamide plus 10 pg/ml insulin. At the intervals presented in Fig. 4 the plates were washed to remove added insulin, and 1251-insulin binding was measured (using 5 nM 1251-insulin). Similar results were obtained from two separate experiments.
The gradual rise in receptor level (insulin binding) in control monolayers to amounts approximately twice those of insulin-treated monolayers confirms a previous demonstration (13) of the ability of this cell line to regulate its cellsurface receptor concentration when insulin is added to or removed from the culture media. When 3T3-Ll adipocytes were cultured in the absence of unlabeled insulin, tolbutamide increased the measured binding of 1251-insulin to values 130-140% of those observed in control cells. Treatment of the monolayers with insulin and tolbutamide also increased the level of 1251-insulin binding to 140% of that of monolayers exposed to insulin alone. Although the percentage by which tolbutamide elevated 1251-insulin binding was similar in either insulin-treated. or untreated monolayers, the absolute value of the increase was less in insulin-treated compared to untreated monolayers (0.3 pmol/106 cells versm 0.7 pmol/106 cells).
Concentration Dependence of the Effect of Tolbutamide ,on the Basal Rate of Uptake of 2-Deoxyglucose by 3T3-Ll Adipocytes-As chronic exposure of 3T3-Ll monolayers to tolbutamide markedly altered the level of surface receptor, it was of interest to determine if tolbutamide would also alter the rate of glucose utilization and/or the sensitivity of this process to insulin.
Differentiated 3T3-Ll adipocytes exhibit both rapid and long-term control of basal and maximal insulin-stimulated rates of glucose utilization (13,19,20). Under our culture conditions, as the monolayers were maintained in high levels of insulin to promote differentiation, the levels of both basal and insulin stimulated 2-deoxyglucose uptake were high in newly differentiated monolayers. Upon removal of insulin from the medium, concomitant dramatic losses in both the basal and insulin-stimulated rates of 2-deoxyglucose uptake were observed within the first several hours. Steady-state levels of both basal and insulin-stimulated rates of 2-deoxyglucose uptake were achieved during the ensuing 24 h when the monolayers were maintained in insulin-free medium (data not presented). The fall in transport activity was not simply a :eversal of the insulin-stimulated state, as both basal and maximal stimulated rates fell over the 24-h period, and -may have arisen from a reduction in the amount of total available transporter activity in addition to dissociation of insulin from the receptor. These processes were reversible, as prolonged exposure of the monolayers to insulin gradually increased the maximal Stimulated rate of glucose uptake far above that rate observed for the acute transport activation process normally occurring within minutes of eiposure of the cells to insulin (data not shown). Under conditions of chronic exposure to insulin, reduction of the level of cell-surface insulin receptors and a shift of the dose-response curve for sugar uptake to higher concentrations of insulin also occurs (13). Therefore, analysis of tolbutamide effects on glucose uptake were undertaken at intervals greater than 24 h after the removal of the insulin present in the media during differentiation, and insulin-stimulated rates were measured shortly after acute (30 min) exposure of the monolayers to insulin.
Thus, to determine the concentration dependence of the effects of tolbutamide on 2-deoxyglucose uptake by 3T3-Ll adipocytes, monolayers were incubated for 24 h in Dulbecco's modified Eagles's medium (25 mM glucose) containing 10% fetal calf serum and the concentrations of tolbutamide presented in Fig. 5. After 24 h of incubation in the presence or absence of tolbutamide, the monolayers were assayed for their basal rates of uptake of 2-deoxyglucose. As was observed for tolbutamide's effects on insulin binding, basal rates of 2deoxyglucose uptake by 3T3-Ll adipocytes were raised to values 160-180% of control rates by exposure to concentrations of tolbutamide between 0.8-2.5 mM.

Effect of Tolbutamide on the Dose-response Relationship between Insulin Concentration and the Rate of Uptake of 2-Deoxyglucose-
The previous experiments indicated that tolbutamide is capable of elevating both surface receptor number and the basal rate of 2-deoxyglucose uptake in cultured 3T3-L1 adipocytes. To determine how the dose-response relationship for insulin-stimulated 2-deoxyglucose transport might be altered by exposure of 3T3-U adipocytes to tolbutamide, monolayers were incubated without insulin in the presence or absence of 1.5 mM tolbutamide for a period of 1 day (Fig. 6A) or 3 days (Fig. 6B). The monolayers were then assayed for 2deoxyglucose uptake in the presence of the indicated concentrations of insulin as described under "Materials and Methods." After 1 day (Fig. 6A), elevated rates of basal transport appear in tolbutamide-treated monolayers. These differences are evident in assays using concentrations of insulin approaching 10-l' M. At higher concentrations of insulin the differences are statistically insignificant. After 3 days of exposure to tolbutamide (Fig. 6B) 1.9-fold differences in basal rate of 2-deoxyglucose uptake are observed in tolbutamidetreated monolayers. In this experiment the rates of 2-deoxyglucose uptake in treated monolayers were significantly higher than control cells throughout the range of insulin concentrations tested

M).
The experiment presented in Fig. 6B was repeated using cells grown and maintained under identical conditions only using a different lot of fetal calf serum. The results are presented in Fig. 6C. The rates of 2-deoxyglucose uptake were approximately 0.7-1.5 nmol/min/106 cells lower than those presented in Fig. 6B at all insulin concentrations tested. Tolbutamide nevertheless maintained its ability to stimulate the rate of 2-deoxyglucose uptake by 2-fold both in the absence and at low concentrations of insulin ( 10-12-10-" M). In the experiment presented' in Fig. 6C, some effect of tolbutamide exposure on the rate of 2-deoxyglucose uptake could also be observed at high insulin concentrations ( 10-s-10-6 M), but this was inconsistently observed in repeat experiments. To obtain a statistical comparison of the effects of tolbutamide on the rate of 2-deoxyglucose uptake in the two different sera, the data for rates of 2-deoxyglucose uptake observed in the two plateau regions of the dose-response curve, one at low (0-10-l' M) and one at high (10-s-10-6 M) concentrations of insulin, were analyzed. Five experiments with cells maintained in the serum used in Fig. 6B (serum A) and six experiments with cells maintained in the serum used in Fig.   6C (serum B) have been performed. Table I shows

Effects of tolbutamide on the rate of 2-deoxyg~ucose uptake measured at low or high inaulin concentrations in different sera
To determine the effects of serum on the tolbutamide-mediated increase in the rate of 2-deoxyglucose uptake at different insulin concentrations, statistical comparisons of the results obtained from six experiments with the serum used in the experiment presented in Fig. 6B (serum A) and five experiments with the one used for Fig. 6C  (serum B) were performed. The rates of 2-deoxyglucose uptake determined from multiple measurements at low (0-lo-" M) or high (10"-M) concentrations of insulin were averaged for monolayers maintained for 3 days in the presence (+) or absence (-) of tolbutamide. The 95% confidence intervals for the means are shown, as well as the p value necessary to reject equality of the means between tolbutamide-treated and untreated monolayers for each condition compared.  7. Kinetics for the development of enhanced rates of basal and insulin-stimulated uptake of 2-deoxyglucose in 3T3-Ll adipocytes exposed to tolbutamide. Monolayers were washed with medium containing 1% fetal calf serum and incubated for 24 h in medium containing 10% fetal calf serum (v/v) to remove insulin and return the monolayers to basal rates of 2-deoxyglucose uptake. The medium was replaced with fresh medium containing 10% fetal calf serum and either no additions or 1.5 mM tolbutamide. On each of the following 3 days of incubation, both control monolayers and those incubated with 1.5 mM tolbutamide were washed and assayed for both basal and insulin-stimulated (10" M insulin) rates of 2-deoxyglucose uptake as described under "Materials and Methods.'' (A) Basal rates of 2-deoxyglucose uptake in control (0) and tolbutamide-treated (W) monolayers. ( B ) Insulin-stimulated rates of 2-deoxyglucose uptake in control (U) and tolbutamide-treated (m) monolayers, measured in the presence of M insulin. The lot of fetal calf serum utilized in these experiments was identical to that used in Fig. 6C (serum B). Results are presented as the mean f range of duplicate determinations. the indicated intervals of exposure to tolbutamide, the monolayers were assayed for basal and insulin-stimulated rates of 2-deoxyglucose uptake as described under "Materials and Methods." Tolbutamide elicited a gradual rise in the basal rate of 2-deoxyglucose uptake (Fig. 7A). Maximal differences (2.4-fold those of control monolayers) were observed after 3 days of exposure to tolbutamide. Insulin-stimulated rates of 2-deoxyglucose uptake (Fig. 7B) were elevated slightly by exposure to tolbutamide, and attained levels 30% greater than those of control monolayers after 2 days of exposure to tolbutamide. No significant differences in the insulin-stimulated rates of uptake were observed after 3 days of exposure to tolbutamide, consistent with the statistical results (Table I) for serum B.

DISCUSSION
3T3-Ll adipocytes have proven to be a useful model system with which to continue an investigation of the mechanism of action of the hypoglycemic agent, tolbutamide, as chronic exposure to tolbutamide resulted in significant alterations of both surface receptor levels and the characteristics of sugar uptake in this cell line.
Concentrations of tolbutamide of 1 mM or greater increased the amount of receptor detectable on the cell surface of 3T3-L I adipocytes by insulin binding isotherms to values 150-250% of those of control. This process was not rapid and reflected the results of previous in vivo studies which demonstrated alterations in tissue receptor number after chronic exposure to tolbutamide (3)(4)(5)(6)(7). 3T3-Ll adipocytes in culture required 2-3 days of exposure to tolbutamide to express maximal increases i.n insulin binding and receptor number. Other studies with isolated adipocytes have previously demonstrated no effect of sulfonylurea treatment on the level of insulin binding or receptor number (10). The ability to demonstrate such effects with 3T3-Ll adipocytes in this study may reflect the difference in the structure of the sulfonylureas employed in the two studies (tolazamide uersus tolbutamide).
Previous relationships established between receptor number and the sensitivity to insulin of glucose uptake in 3T3-Ll adipocytes suggested a possible precedent for relating the effects of tolbutamide on receptor number to its effects on 2deoxyglucose uptake which would be consistent with the chronic hypoglycemic action of sulfonylurea agents observed in uiuo. When 3T3-Ll preadipocytes differentiate into adipocytes in culture, this process is accompanied by an approximate 10-fold increase in the level of insulin receptor, primarily through an increase in the rate of receptor synthesis (16,17). This increase in receptor number is paralleled by a marked increase in the sensitivity of glucose uptake by the cells to insulin (19). Removal of insulin present in the medium after differentiation of 3T3-Ll preadipocytes into adipocytes further elevates the level of insulin receptors by 2-fold. This process likewise is characterized by an increase in the sensitivity of glucose uptake to insulin in such up-regulated cells (13). From these results it was tempting to predict a priori that an increase in the capacity of 3T3-Ll adipocytes to bind insulin after exposure to tolbutamide would increase their sensitivity of glucose uptake to insulin, increase glucose utilization at submaximal stimulatory concentrations of insulin, and thus mimic the hypoglycemic action of the drug observed in uiuo. In an effort to relate tolbutamide-induced alterations in insulin binding capacity to enhanced glucose uptake, the effects of tolbutamide on the dose-response relationships between insulin concentration and 2-deoxyglucose uptake were analyzed. The results of five experiments identical to that presented in Fig. 6B demonstrated that, at all concentrations of insulin, the rates of 2-deoxyglucose uptake were higher in tolbutamide-treated cells, but there was no significant alteration of the concentration of insulin required for half-maximal stimulation of 2-deoxyglucose uptake relative to control monolayers. However, as there was a reasonably constant difference in rates of uptake of approximately 0.8 nmol/min/106 cells between tolbutamide-treated and control monolayers at each concentration of insulin tested (Fig. 6B), tolbutamide appeared to be activating a component of the uptake process that was independent of insulin-activated uptake. This alternate interpretation, however, was not entirely supported by the results of the six additional experiments identical to that presented in Fig. 6C, which utilized a different lot of fetal calf serum. In Fig. 6 c no differences in uptake rates were observed between tolbutamide-treated and untreated monolayers at insulin concentrations between lo-'' and 3 X lo-' M insulin.
Although some effect of tolbutamide could be observed at high insulin concentrations (10-8-10-6 M) in the particular experiments presented in Fig. 6C and 7B, the results of the six repeat experiments indicated that, when measured at high insulin concentrations, the effect of tolbutamide on the uptake of 2-deoxyglucose was statistically insignificant when the cells were maintained in the lot of serum identical to that used in Fig. 6C. The reason for the differences observed between the dose-response curves obtained from cells maintained in the two different lots of serum (A and B) are unknown. A comparison of the results obtained with the two sera can be summarized as follows: (a) at low insulin concentrations statistically significant higher rates of uptake were observed in tolbutamide-treated relative to control monolayers with either serum; (b) at high insulin concentrations statistically significant higher rates of 2-.deoxyglucose uptake arising from tolbutamide treatment were observed only in serum A; (c) in general, for untreated monolayers the rates of 2-deoxyglucose uptake were approximately 1.2 nmol/min/106 cells lower in cells maintained in serum B when compared to cells maintained in serum A (this relative difference in rates was ob-served at both high and low insulin concentrations); and ( 4 the two serum environments did not appear to markedly affect the ability of insulin to maximally alter the rate of 2-deoxyglucose uptake. When the rates of uptake were compared for the two plateau regions at low and high insulin concentrations (o-IO-~O M and 10-8-10-6 M, respectively), the increase in the rate of uptake arising from the higher insulin concentration was between 1.1 and 1.2 nmolfmin f lo6 cells regardless of the serum utilized or the presence or absence of previous exposure to tolbutamide.
It should be emphasized that the serum-related differences observed in transport rates arose from previous growth and maintenance of the cell monolayers in the different lots of serum. The serum was not present during the assay of 2deoxyglucose uptake. However, regardless of the serum lots utilized and the associated rates of basal transport, tolbutamide consistently increased by 1.5-1.7-fold the rates of basal 2-deoxyglucose uptake and the rate of uptake observed at physiological insulin concentrations between 10-l' and 10-l' M. Tolbutamide also increased the rate of uptake of 2-deoxyglucose measured at insulin concentrations of lo-' M or greater in some preparations of cells, but tbis effect was less consistently observed and appears to be modulated by other unidentified serum factors. Thus, the observed effects of tolbutamide on glucose uptake are complex and cannot be described simply by a shift of the dose-response curve to lower insulin concentrations arising from higher receptor levels as would be predicted by the previous examples (13,19), nor do these effects appear to arise solely from an increase in basal transport rates.
What then, is the mechanism by which tolbutamide elevates both insulin receptor levels and rates of glucose uptake in 3T3-Ll adipocytes? As mentioned previously, whether tolbutamide alters 2-deoxyglucose uptake in 3T3-Ll cells in a manner which is dependent or independent of the simultaneous increase in the number of insulin receptors is unknown. The ability of another sulfonylurea, tolazamide, to elevate insulin-stimulated glucose uptake in the absence of increased insulin binding would suggest the lack of such a dependence (10). It is interesting to note that the active concentration range in which tolbutamide maximally alters both the rate of 2-deoxyglucose uptake measured at low insulin concentrations and surface receptor number in 3T3-Ll adipocytes are quite similar (1-3 mM, Figs. 1 and 5). Furthermore, the respective changes in the rate of 2-deoxyglucose uptake and receptor number in tolbutamide-treated monolayers relative to control monolayers show similar kinetics for expression (Figs. 2 and 7A) as well as similar relative changes (approximately 1.7-fold). The similarity in the kinetics, concentration dependence, and magnitude of the changes in receptor number and the rate of 2-deoxyglucose uptake suggest that a common mechanism might be responsible for the effects of tolbutamide on the two processes of regulation of receptor number and transporter activity.
Regulation of cell-surface insulin receptor number and some forms of activation of glucose transport by insulin operate through a control of the cellular distribution of the respective receptor and transporter protein and thus potentially would share similar pathways of internalization and recycling of the transporters or receptors to and from external and internal cellular sites (21, 22). Tolbutamide could act by interrupting the normal traffic of both transporter and receptor proteins between internal and cell-surface sites in a fashion that could increase their steady-state surface concentration and thereby simultaneously increase the level of cellsurface receptor and the rate of glucose uptake. Tolbutamide has been suggested to inhibit internalization of some types of cell-surface molecules (23), and could presumably be inhibiting the internalization of both insulin receptors and glucose transporter units via a common mechanism. Whatever the mechanism, insulin appears to antagonize the effects of tolbutamide on receptor level (see Fig. 4), possibly by increasing the rate of receptor internalization and shifting the distribution of receptor to internal sites. Such antagonism of insulininduced down-regulation of the insulin receptor by the sulfonylurea, glyburide, has been previously noted in human fibroblasts (8).
The effects of tolbutamide on insulin receptor level and rates of glucose uptake in 3T3-Ll adipocytes were observed at concentrations of 1 mM or greater. How do these concentrations relate to those observed clinically? Patients receiving a dosage of 0.5 g of tolbutamide daily demonstrate lower plasma concentrations of tolbutamide (0.15 mM) than utilized in these studies (24). Those patients, however, receiving 2-3 g of tolbutamide daily (3) would approach plasma concentrations of tolbutamide similar to those concentrations utilized in this study, thus suggesting a clinical relevance for the in uitro effects of tolbutamide demonstrated with 3T3-Ll adipocytes.
In summary, chronic exposure of 3T3-Ll adipocytes to tolbutamide has been demonstrated to increase both the number of insulin receptors as well as the rate of glucose uptake measured at low insulin concentrations. Both processes would contribute towards the hypoglycemic action of the drug observed in uiuo, especially when tissues such as muscle and liver are also affected, as previously demonstrated (4, 7). The value of utilizing the 3T3-Ll adipocyte system for studying the mechanism of tolbutamide action in uitro is derived from: (a) the demonstration that the effects on glucose transport and receptor level are larger than those previously observed in other in vitro systems, and (b) the current availability of methods previously applied to this cell line which permit a measurement of altered rates of receptor internalization (25), synthesis (17, 26;27), degradation (17, 18), and their effects on total receptor number and distribution. We are currently measuring the effects of tolbutamide on these processes to assess which are affected by chronic exposure of the cell to tolbutamide.