Masashi KOBAYASHI, Makoto IWASAKI, and Yukio SHIGETA. The Third Department of Medicine, Shiga University of Medical Science, Ohtsu, Shiga

Transcription

1 J. Biochem. 88, (1980) Receptor Mediated Insulin Degradation Decreased by Chloroquine in Isolated Rat Adipocytes1 Masashi KOBAYASHI, Makoto IWASAKI, and Yukio SHIGETA The Third Department of Medicine, Shiga University of Medical Science, Ohtsu, Shiga Received for publication, January 7, 1980 The time-course of insulin binding to isolated adipocytes pretreated with chloroquine (0.1 mm) showed that no steady state condition was reached and that the binding kept increasing with the time of incubation up to three hours. Sixty-one percent of the bound labeled insulin could not be dissociated from these cells by 100ƒÊg/ml of unlabeled insulin, whereas 86 dissociated from the control cells. Chromatography revealed that the bound intact insulin, which had the ability to bind to insulin receptors of adipocytes, was increased and that de graded insulin decreased in the chloroquine treated cells. However, these cells showed normal insulin stimulation of 2-deoxy-glucose uptake. The data suggest that insulin is internalized in adipocytes after binding to insulin receptors and that insulin is degraded at the site, probably lysosomes, where chloroquine inhibits this degradation process. The first step of insulin action is binding to its receptors on plasma membranes. After binding to its receptors, insulin is thought to be internalized and then degraded at lysosomes of hepatocytes (1-3) and fibroblasts (4). Degraded insulin prod ucts may be necessary for insulin action (5) or the process after insulin binding to its receptor may not be necessary for the action. Chloroquine was shown to inhibit degrada tion of LDL (6) and epidermal growth factors (7) at lysosomes after these materials are inter nalized in cells. If this is the case with insulin, it is quite conceivable that chloroquine inhibits 1 This study was supported in part by a Research Grant for Intractable Diseases from the Ministry of Health and Welfare of Japan. A part of this study was pre sented at the Japan Endocrine Society Meeting in Tokyo on October 19, receptor mediated insulin degradation at lysosomes and that intact undegraded insulin may accumulate in the cells. To test this hypothesis, we performed insulin binding studies on rat isolated adipocytes treated with chloroquine. MATERIALS AND METHODS Materials-Porcine monocomponent insulin was a gift from Dr. Ronald Chance of Eli Lilly Company. Na[125I], [3H]2-deoxy-glucose and [3H] L-glucose were purchased from New England Nuclear Co., collagenase (Type 2) from Worthing ton Biochemicals, chloroquine from Sigma and Sephadex G-50 from Pharmacia. Preparati on of Isolated Adipocytes-Male Sprague-Dawley rats weighing g were used for all experiments. Isolated fat cells from Vol. 88, No. 1,

2 40 M. KOBAYASHI, M. IWASAKI, and Y. SHIGETA epididymal fat pads were prepared by shaking at 37 Ž for 60 min in Krebs-Ringer bicarbonate buffer containing collagenase (3mg/ml) and albumin (40mg/ml) according to the method of Rodbell (8). Adipocyte counts were performed according to the method of Hirsch and Gallian (9). Cell counting was performed using a model Industrial D Coulter Counter. Iodination of Insulin-125I-Insulin was pre pared with a specific activity of ƒÊCi/ƒÊg according to the method of Freychet et al. (10). Insulin Binding Studies-Isolated adipocytes were suspended in a buffer containing 50mM HEPES, 50mM Tris, 50mM NaCl, 10mM MgSO4, 5mM KCl, 10mM CaCl2, 2mM EDTA, 10mM glucose, and 1 % bovine serum albumin, ph 8.0, and incubated with 125I-insulin and unlabeled insulin in plastic flasks in a 24 Ž shaking water bath. Optimal steady state conditions were reached at 24 Ž after 90min of incubation. The incubations were terminated as described by Gammeltoft and Gliemann (11) by removing 200 ƒê1 aliquots from the cell suspension and rapidly the last portion of Peak 2. These groups of collected material were lyophilized and then dis solved in the buffer for binding assay. The radio active materials were incubated with rat adipocytes and binding studies were performed as described under "Insulin Binding Studies." 2-Deoxy-Glucose Uptake Studies-Isolated adipocytes were incubated with insulin at various concentrations for 60min at 24 Ž. Then, these cells were incubated with [3H]2-deoxyglucose at a concentration of 0.1mM in Krebs-Ringer bicar bonate buffer, ph 7.4, containing bovine serum albumin (10mg/ml) at 24 Ž. The assay was terminated at the end of 3 min by transferring 200 ƒêl aliquots from the assay mixture to plastic microtubes containing 100ƒÊ1 of silicone oil. The assay is considered terminated when centrifuga tion begins. The amount of sugar trapped in the extracellular water space of the cell layers was determined using [3H]L-glucose. Extracellular water space was measured in each experiment and all data of sugar uptake were corrected for this factor. centrifuging the cells in plastic microtubes to which 100ƒÊl of silicone oil had been added. The cells were then removed and the radioactivity was determined. Insulin binding studies of human erythrocytes were performed according to the method of Gambhir et al. (12). Chromatography-Cell solubilization and sub sequent chromatography were performed accord ing to the method of Gorden et al. (2). After adipocytes were incubated with 125I-insulin, cells were washed twice at 4 Ž and then solubilized in a solution containing 8 M urea, 0.1 % triton X-100 and 1.5mM acetic acid for 20min at 4 Ž. The mixture was then centrifuged for 5min at 10,000 ~ g at 4 Ž and the infranate was applied to a 48 ~1cm Sephadex G-50 column The column was eluted with 1M acetic acid and standardized with dextran blue, 125I-insulin and Na[125I] The recovery was greater than 90% of the total counts applied in all cases. Rebinding Studies-Eluted radioactive mate rials from the Sephadex column were grouped according the peak, i.e., Peak 1, Peak 2, and Peak 3. Peak 2 was further divided into three portions, Peak 2A, the fast eluted portion of Peak 2, Peak 2B, the center portion of Peak 2, and Peak 2C, RESULTS Binding Studies-The time-course of insulin binding to isolated adipocytes is summarized in Fig. 1. Steady state binding conditions were reached in 90min for control cells, whereas 1151 insulin continued to bind to cells up to 210min for chloroquine treated cells. In contrast to the total binding, the nonspecific binding of chloro quine treated cells did not increase with time of incubation. Therefore, increased insulin binding of chloroquine treated cells was not due to increased nonspecific binding. Insulin degradation in the buffer medium was comparable in both groups (7.8% in the control and 8.1 % in the chloroquine treated group). Thus, in chloroquine treated adipocytes, insulin seemed to accumulate in cells in a time dependent fashion. In order to see whether this accumulated 125I -insulin can be chased out by a large amount of unlabeled insulin, 100 jug of insulin was added to the ongoing binding assay system at 90 min after binding started, i.e. when steady state binding conditions were reached in control cells (Fig. 2). Eighty-six percent of bound 125I-insulin indicated nonspecific binding. In contrast to these control J. Biochem.

3 EFFECT OF CHLOROQUINE ON INSULIN DEGRADATION 41 Fig. 1. Time-course of 125I-insulin binding. Adi pocytes were incubated with ( ü) or without (9) chloroquine (0.1mM) for 20 min at 37 Ž. Then, cells were incubated at 24 Ž with 125I-insulin (0.2ng/ml) (-) or 125I-insulin (0.2ng/ml) and unlabeled insulin (100ƒÊg/ml) (--- ) (nonspecific binding). Bound 125Iinsulin was counted at the times indicated. Each point represents the mean of duplicate experiments. Percent bound was corrected for the cell number, 2.0 ~105 cells. Fig. 3. Effect of chloroquine on insulin binding. Cells were treated with chloroquine, 0.01mM ( ü) and 0.1 mm ( ), and without chloroquine ( œ) at 37 Ž for 20 min. Then, cells were incubated with 0.2 ng/ml of I'll-insulin and unlabeled insulin at various concen trations for 90 min at 24 Ž. Each set of binding data are specific binding which was calculated by subtracting nonspecific binding from total binding. Nonspecific binding was determined when unlabeled insulin (100 ƒêg/ml) was present in the incubation media. Each point represents the mean of triplicate experiments with standard error of the mean. t indicates statistical significance between control group and chloroquine treated groups (p<0.05). Fig. 2. Effect of addition of a large amount of un labeled insulin to bound 125I-insulin. The binding study was described in Fig. 1. At 90 min after starting the binding study, unlabeled insulin, 100ƒÊg, was added to each ml of total incubation medium. cells, 61% of I'll-insulin bound to chloroquine treated cells could not be chased out by a large amount of unlabeled insulin. Therefore, bound 125I-insulin remained in the chloroquine treated adipocytes, possibly within lysosomes of these cells. Next, we studied insulin binding at various insulin concentrations to see whether this increased insulin binding in chloroquine treated cells can be seen at all insulin concentrations. The results are summarized in Fig. 3, indicating that increased 125I-insulin binding in chloroquine treated cells is only seen at low insulin concentrations. A higher chloroquine concentration demonstrated higher insulin binding at low insulin concentrations. If the effect of chloroquine to increase insulin binding is due to decreased insulin degradation at lysosomes of these cells, this degradation process may be saturable and no increased insulin binding could be elicited by chloroquine treatment at high insulin concentrations. Human erythrocytes, which are assumed to have no lysosomes, were also studied for insulin binding. No significant difference was demon strated in insulin binding between control cells and chloroquine treated cells (Fig. 4). Chromatography-To examine the exact prod uct of bound insulin, we performed Sephadex G-50 column chromatography of bound 125I-materials after 90 min incubation. Adipocytes incubated with 125I-insulin for 90min were separated with Vol. 88, No. 1, 1980

4 42 M. KOBAYASHI, M. IWASAKI, and Y. SHIGETA Fig. 4. Effect of chloroquine on insulin binding in human erythrocytes. Cells were incubated with ( ü) and without ( œ) chloroquine (0.1mM) for 20min at 37 Ž. Then, cells were incubated with 125I-insulin (0.2 ng/ml) and unlabeled insulin at various concentrations for 180min at 24 Ž. Each set of binding data are specific binding which was corrected for nonspecific binding. There was no statistically significant differ ence between the two groups. Fig. 5. Gel filtration patterns of cell-bound 125I-radio activity at 90min of association at 24 Ž. Cells were initially treated with (--- ) and without (-) chlo roquine (0.1mM) for 20 min at 37 Ž and then were incubated with 125I-insulin for 90min at 24 Ž. After incubation, cells were separated from incubation media with silicone oil and were washed twice with cold (4 Ž) buffer. The cells were then solubilized with 1.5 M acetic acid, 8 M urea and 0.1% triton X-100 for 20 min at 4 Ž. The mixture was centrifuged for 5 min at 10,000 ~g (4 Ž) and the infranate was applied to a Sephadex G-50 column which was calibrated with dextran blue, insulin and 125I. silicone oil and solubilized materials were applied to the Sephadex G-50 column. Figure 5 shows the gel filtration pattern of solubilized materials. In control cells, 82% of the cell bound materials Fig. 6. Gel filtration patterns of the total accumulated radioactive materials that dissociated from the cells. Cells were treated with (--- ) or without (-) chlo roquine (0.1mM) and were incubated with 125I-insulin as described in Fig. 4. After 90min, cells were sepa rated and suspended in the buffer where bound radio active materials were dissociated. The original ex tracellular medium was diluted about 2,000 fold. The medium containing dissociated material was concen trated and applied to the column. is 125I-insulin, 12% is a low molecular weight component eluting just in front of the 125I-peak, and 6% is high molecular weight materials, pos sibly representing aggregated 125I-insulin. In cells treated with chloroquine, 91% is intact 125I-insulin and 6% is low molecular weight materials. Thus, the proportion of intact insulin in the bound total 125I-materials is increased and that of de graded insulin decreased in cells treated with chloroquine, indicating that chloroquine inhibits receptor mediated insulin degradation. In the next step, dissociated materials in the diluting buffer were examined. The gel filtration pattern of the dissociated materials is shown in Fig. 6, demon strating increased proportions of degraded ma terials during the process of dissociation after binding to receptors. The proportion of Peak 1 (high molecular weight materials) is comparable between the two groups, whereas that of Peak 2 (intact insulin) is increased (42% vs. 68%) and that of Peak 3 (low molecular weight materials) is decreased in the chloroquine treated group (41 vs. 18%). Therefore, receptor mediated insulin degradation is inhibited and internalized intact insulin may possibly be accumulated at lysosomes of these chloroquine treated cells. J. Biochem.

5 EFFECT OF CHLOROQUINE ON INSULIN DEGRADATION 43 Fig. 7. Rebinding studies of eluted radioactive mate rials from the column chromatography. Eluted mate rials of Fig. 5 were collected and grouped according to their peaks. Peak 2 was further divided into three portions, Peak 2A, Peak 2B, and Peak 2C. Since radioactivity of these materials was relatively low, three sets of column study experiments were necessary to accumulate enough radioactive materials. Binding studies were performed in the presence and absence of 200ƒÊg/ml of unlabeled insulin. Specific binding was calculated by subtracting percent binding in the presence of 200ƒÊg/ml of unlabeled insulin from percent binding in the absence of unlabeled insulin. 0.2 ng/ml of 125I -insulin was also incubated with adipocytes and percent binding of this control experiments was defined as 100 percent. Fig Deoxy-glucose uptake by adipocytes preincu bated with ( ü) or without ( œ) chloroquine. Adi pocytes were incubated with or without chloroquine (0.01mM) for 20min at 37 Ž. Then, the cells were incubated with insulin at various concentrations at 24 Ž for 60min. 2-Deoxy-glucose uptake was measured at 24 Ž at the end of 3-min incubation with 2-deoxy glucose (0.01mM). Data represent the mean ( }SE) of three experiments for each group. Rebinding Studies-Radioactive materials eluted from the Sephadex column were tested for binding ability. The materials of Peak 2, espe cially Peak 2B, are fully capable of binding to insulin receptors on rat adipocytes (Fig. 7). In contrast, materials from Peaks 1 and 3 had much less ability for binding to the receptors. Thus, the undegraded materials are structurally intact insulin, being capable of binding to the receptors. 2-Deoxy-Glucose Uptake Studies-In order to examine the relationship between receptor mediated degradation and insulin action, relatively early insulin action, namely glucose uptake, was studied. The results showed essentially no difference be tween chloroquine treated and control cells (Fig. 8), although cells treated with chloroquine showed slightly decreased glucose uptake. Thus, ac cumulated insulin trapped inside the cells did not result in increased insulin action. DISCUSSION It was recently suggested that insulin binds to its receptor forming an insulin-receptor complex followed by internalization of this complex to be degraded at lysosomes (1-4). These degraded fragments are probably dissociated out of cells or may become a second messenger to transmit insulin's signals (5). Chloroquine is thought to inhibit a degradative process of epidermal growth factors (7) and LDL (6) at lysosomes. We have presented data showing that chloroquine inhibits receptor mediated insulin degradation (Figs. 5 and 6) resulting in accumulation of intact 125I-insulin inside of (or firmly associated with) chloroquine treated adipocytes (Figs. I and 2). However, it can not be totally excluded that a nonspecific toxic effect of chloroquine indirectly caused firm as sociation of insulin with chloroquine treated cells. Although we do not have direct evidence that insulin-receptor complexes are degraded at lyso somes of adipocytes, from the data presented it is likely that insulin enters adipocytes to be degraded at lysosomes as in other cells such as hepatocytes (1-3) and fibroblasts (4). This receptor mediated degradative process is saturable in adipocytes (Fig. 3) and insulin binding curves at various insulin concentrations showed increased binding only at low insulin concentrations in chloroquine treated cells. In contrast to adipocytes, erythrocytes Vol. 88, No. 1, 1980

6 44 M. KOBAYASHI, M. IWASAKI, and Y. SHIGETA which have no lysosomes did not show any in creased insulin binding (Fig. 4). Recently, Duckworth et al. reported that the initial site of insulin cleavage was between B16 and B17 and that cleaved insulin is not able to bind to insulin receptors because B16 is an essen tial component for receptor binding (13). Unde graded insulin of chloroquine treated cells is able to bind to insulin receptors suggesting that it is structurally intact. The mechanism of insulin action is not clear. LeCam et al. reported that internalization is not a necessary step for insulin's action to stimulated amino acid transport in rat hepatocytes which requires a protein synthesis step (14). Further more, Van Obberghen et al. (15) showed anti-insul in-receptor antibodies were able to stimulate lipo protein lipase of 3T3-L1 fibroblasts, indicating that internalization was not necessary for the insulin action. These studies suggest that a slow action of insulin, i.e. protein synthesis or DNA synthesis, does not require internalization of insulin. Al though it is difficult to draw a conclusion from our glucose uptake studies because of partial suppression of insulin degradation in chloroquine treated cells, it may be suggested that accumulated insulin, probably inside the cells, did not result in increased glucose uptake, which was an early action of insulin not requiring protein synthesis, in these chloroquine treated cells. REFERENCES 1. Carpentier, J.L., Gorden, P., LeCam, A., Freychet, P., & Orci, L. (1977) Diabetologia 13, Gorden, P., Carpentier, J.L., Freychet, P., LeCam, A., & Orci, L. (1978) Science 200, Carpentier, J.L., Gorden, P., Freychet, P., LeCam, A., & Orci, L. (1979) J. Clin. Invest. 63, Schlessinger, J., Schechter, Y., Willingham, C., & Pastan, I. (1978) Proc. Natl. Acad. Sci. U.S. 75, Steiner, D.F. (1977) Diabetes 26, Goldstein, G. & Brown, M.S. (1974) J. Biol. Chem. 249, Carpenter, G. & Cohen, S. (1976) J. Cell Biol. 71, Rodbell, M. (1964) J. Biol. Chem. 239, Hirsch, J. & Gallian, E. (1968) J. Lipid Res. 9, Freychet, P., Roth, J., & Neville, D.M., Jr. (1971) Biochem. Biophys. Res. Commun. 43, Gammeltoft, S. & Gliemann, J. (1973) Biochim. Biophys. Acta 320, Gambhir, K.K., Archer, J.A., & Bradley, C.J. (1978) Diabetes 27, Duckworth, W.C., Stentz, F.B., Heinemann, M., & Kitabchi, A. (1979) Proc. Natl. Acad. Sci. U.S. 76, LeCam, A., Maxfield, F., Willingham, M., & Pastan, I. (1979) Biochem. Biophys. Res. Commun. 88, Van Obberghen, E., Spooner, P.M., Kahn, C.R., Chernick, S.S., Garrison, M.M., Karlsson, F.A., & Grunfeld, C. (1979) Nature 280, J. Biochem.