THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1985 by The American Society of Biological Chemists, Inc.

Vol. 260, No. 7, Issue of April 10, pp. 41364144,1985 Printed in U.S.A.

Kinetics of Insulin Receptor Internalization and Recycling in Adipocytes SHUNTING OF RECEPTORS TO A DEGRADATIVE PATHWAY BY INHIBITORS OF RECYCLING*

(Received for publication, August 3,1984)

Stephen Marshall$ From the University of California at San Diego, Department of Medicine, M-023E, Division of Endocrinology and Metabolism, LA Jolla, California~92093

The kinetics of receptor internalization and recy- cling was directly determined in adipocytes by mea- suring '251-insulin binding to total, intracellular, and cell-surface insulin receptors. In the absence of insulin 90% of all receptors were on the cell-surface and 10% were intracellular. Insulin (100 ng/ml) rapidly altered this distribution by translocating surface receptors to the cell-interior through a temperature and energy- dependent process. Surface-derived receptors were seen within cells as early as 30 s and accumulated intracellularly at the rate of -20,00O/min (tn = 2.7 min). After 6 min the size of the intracellular receptor pool plateaued (for up to 2 h), with 30% of surface receptors residing within the cell. This plateau was due to the attainment of an equilibrium between receptor uptake and recycling, since removal of insulin (to stop receptor uptake) was followed by both a rapid deple- tion of intracellular receptors and a a concomitant and stoichiometric reappearance of receptors on the cell- surface. Receptors were efficiently recycled, with little or no net loss observed even after 4 h of insulin treat- ment; however, recycling could be partially inhibited (-10%) by several agents (e.g. chloroquine and Tris). Tris treatment of adipocytes in the presence of insulin led to 50% loss of surface and total receptors at 2 and 4 h, respectively. Since chloroquine prevented the de- crease in total receptors, but not the loss of surface receptors, it appears that Tris impairs recycling by diverting a portion of incoming receptors to a chloro- quine-inhibitable degradative site.

From these results we conclude that: 1) insulin trig- gers endocytotic uptake of insulin-receptor complexes; 2) internalized receptors are then rapidly reinserted into the plasma membrane, and the receptors can tra- verse this recycling pathway within 6 min; 3) pro- longed recycling does not normally result in measura- ble receptor loss, but when receptors are prevented from recycling, they become trapped intracellularly and are shunted to a chloroquine-sensitive degradative pathway; and 4) chloroquine and Tris are only par- tially effective inhibitors of receptor recycling.

* This work was supported by Research Grants AM 33647 and AM 34402 from the National Institutes of Health and by the Gail Patrick Velde Feasibility Grant from the American Diabetes Association. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "aduertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$Recipient of Research Career Development Award AM 01357 from the National Institutes of Health.

In earlier studies we (1-3) and others (4) postulated that insulin-induced receptor down-regulation in adipocytes is me- diated by endocytotic uptake of insulin-receptor complexes, and this conclusion was based on several lines of evidence. For example, when freshly isolated adipocytes were exposed to a high concentration of insulin (100 ng/ml) there was a rapid time-, temperature-, and energy-dependent decrease in the number of cell-surface receptors (50% loss by 2 h, 75% by 4 h), and such receptor loss could be prevented by conditions known to inhibit absorptive endocytosis of '251-insulin. Thus, incubation at 16 "C or depletion of metabolic energy was able to block both ligand internalization and insulin-induced down-regulation (l), and the extent of receptor loss and the amount of internalized '251-insulin were highly correlated as functions of time and temperature (1). Moreover, receptors lost from the cell-surface could be partially recovered in the cell-interior (4). The idea that uptake of insulin-receptor complexes mediates down-regulation was further strength- ened by photoaffinity-labeling studies in which the metabolic fate of endocytosed insulin receptors was followed by cova- lently attaching '251-insulin to cell-surface receptors at 16 "C and then warming cells to 37 "C. With this procedure insulin- receptor complexes were found to be degraded intracellularly through a chloroquine-sensitive pathway (5 , 6).

Although these studies convincingly demonstrated that binding leads to the rapid endocytotic uptake and subsequent degradation of insulin-receptor complexes which causes rapid down-regulation of surface receptors, we later discovered that when cells were incubated in a complete medium rather than a simple buffer, internalized receptors were recycled so a net loss of surface receptors was not observed (2). This observa- tion was later explained by our finding that tris(hy- droxymethyl)aminomethane, a constituent of the in vitro adipocyte incubation buffer, prevented endocytosed receptors from recycling back to the plasma membrane (3). Since the number of surface receptors during short-term insulin expo- sure represents an equilibrium between the rate of receptor uptake and the rate at which endocytosed receptors are recy- cled, by inhibiting recycling we disturbed this equilibrium and allowed the former process to predominate. Although this finding provided an explanation for the insulin-induced de- pletion of cell-surface receptors seen over several hours in earlier studies (1-3, 7), our understanding of receptor recy- cling, insulin-induced down-regulation, and degradation of insulin receptors under physiological conditions was limited.

A major obstacle in studying the early events in receptor processing has been the lack of an adequate method for quantitating insulin receptor uptake and recycling. In the past we attempted to estimate the rate of receptor internalization by inhibiting recycling with Tris and then measuring the

4136

Translocation and Recycling of Insulin Receptors 4137

extent of insulin-induced receptor loss as a function of time (1). From such data, inferences were drawn about the kinetics of receptor uptake and recycling; however, these experiments relied heavily on the assumption that recycling was com- pletely prevented. Since this was an unproven assumption and since recycling under physiological conditions could only be studied through the use of recycling inhibitors, we devel- oped an alternative and more direct technique for quantitating receptor uptake and recycling. Using this method we have now examined the kinetics of receptor uptake and recycling in detail and have found that the functional consequence of impaired recycling is the shunting of receptors to a degrada- tive pathway.

EXPERIMENTAL PROCEDURES'

RESULTS

Using our newly developed method for rapidly removing detergent and concentrating soluble insulin receptors (44), we followed insulin-induced translocation of functional insulin receptors (capable of binding insulin) from the cell-surface to the cell-interior. To perform such experiments it was first necessary to develop a low temperature (4 "C) procedure for removing extracellular and receptor-bound insulin and tryp- tically inactivating cell-surface receptors (to quantitate the intracellular receptor pool). Such methods were described in detail in the preceding paper and are now presented in the Miniprint Supplement. Performing all steps at 4 "C is impor- tant, since the processing of cells at a higher temperature (37 "C) results in a rapid repartitioning of receptors from the cell-interior to the cell-surface (Fig. 1) which obscures the original distribution of receptors. As seen in Fig. 1A, when adipocytes were exposed to insulin (100 ng/ml) at 4 "C and then warmed to 37 "C to initiate endocytotic uptake of insulin- receptor complexes, there was an initial lag period of 1-2 min before surface-derived receptors began to accumulate within cells. After this time intracellular receptors rapidly increased reaching a plateau by 8 min, at which time 45% of the total receptor complement was intracellular (uersus 10% in the absence of insulin). Attainment of this new steady state represents a dynamic equilibrium between receptor uptake and subsequent recycling, since removal of extracellular in- sulin after 15 min (at equilibrium) was quickly followed by a rapid depletion of intracellular receptors (Fig. 1A). An impor- tant distinction in the composition of the intracellular recep- tor pool in the absence and presence of insulin should be noted. In the absence of insulin approximately 10% of the total receptor pool can be found within cells, and most, if not all, of this pool consists of newly synthesized receptors enroute to the cell-surface (8). This nascent receptor pool is, therefore, referred to as the basal intracellular receptors (Fig. 1A). However, when insulin is added, receptors are translocated from the cell-surface to the cell-interior so that the intracel- lular pool now consists predominately of surface-derived re- ceptors which co-exist with the proportionately smaller nas- cent receptor pool.

Before continuing, it is important to bear in mind that adipocytes continually synthesize and degrade receptors in

' "Experimental Procedures" are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are available from the Journal of Biological Chemistry, 9650 Rockville Pike, Bethesda, MD 20814. Request Document No. 84M-2412, cite the authors, and include a check or money order for $1.20 per set of photocopies. Full size photocopies are also included in the microfilm edition of the Journal that is available from Waverly Press.

52

5 33,000 Basal Intracellular Receptor Pool

. ,J 120.00016' ' ' ' ' ' '

TIME (minutes)

FIG. 1. Ability of insulin to increase the size of the intracel- lular insulin receptor pool. A, cells were exposed to 100 ng/ml of insulin at 4 'C and then warmed to 37 "C. At the indicated times adipocytes were cooled to 4 "C, washed to remove extracellular and receptor-bound ligand, trypsinized, solubilized, and the soluble intra- cellular insulin receptors harvested by a polyethylene glycol receptor precipitation-reconstitution method as described under "Experimen- tal Procedures." Specific '251-insulin binding to soluble receptors was then measured after a 16-h incubation at 4 "C. The total receptor pool was determined by omitting the trypsinization step from the above procedure. The basal intracellular receptor pool was measured at time 0 and 2 h of incubation in the absence of insulin pretreatment. B, the number of surface-derived receptors translocated to an intra- cellular site as a function of time was determined from the binding data in panel A. B/F, bound/free.

order to keep the net number of cell-surface and total recep- tors constant. Therefore, the finding that net receptor number does not change over time cannot be construed as evidence that receptors are not being degraded. However, since receptor synthesis and degradation accounts for the turnover of only 2% of total receptor/h (8), these processes would be expected to have a minimal effect on acute studies of receptor trans- location and processing. With this caveat in mind, several other aspects of the data in Fig. 1A should be highlighted. First, after 2 h of continual insulin treatment a net loss of total receptors was not found, which suggests that transloca- tion of receptors to the cell interior does not lead to receptor degradation. Secondly, since the attainment of a new steady state in the size of the intracellular pool is most likely due to return of endocytosed receptors to the cell-surface (referred to as receptor insertion), it follows that the earliest time at which such an equilibrium is reached would reflect the time required for insulin receptors to traverse the entire recycling pathway. In other words, receptor internalization and subse- quent reinsertion appears to be completed within 8 min. Further, it is evident that translocation of receptors to the cell-interior is mediated by a mechanism which is initiated or triggered by insulin, consistent with an inducible, rather than a constitutive, receptor-mediated endocytotic process. Finally, and most importantly, it should be mentioned that insulin receptor translocation is being followed directly and under physiological conditions, whereas in previous studies reparti- tioning of cellular receptors between the cell-surface and the cell-interior was measured after receptor recycling was im-

4138 Translocation and Recycling of Insulin Receptors

paired by treating cells with either Tris or chloroquine (1-4). Fig. 1B depicts the number of surface-derived receptors

found within the cell-interior as a function of insulin treat- ment time. These values were calculated by converting the insulin binding data in Fig. lA to an equivalent receptor number and subtracting the number of receptors in the basal pool. This transformation is based on the fact that there are 300,000 cell-surface receptors/adipocyte (9), with an addi- tional 10% in the cell-interior (2,4,8). Thus, the total receptor pool consists of about 330,000 receptors, whereas the intra- cellular pool in the absence of insulin contains approximately 33,000 receptors (Fig. 1A). When insulin-induced receptor translocation is plotted in this manner (Fig. 1B) approxi- mately 110,000 receptors were found within the cell at steady state, and receptor uptake (calculated from 2 to 8 min) pro- ceeded at the rate of about 18,300 receptors/min.

The protocol used in Fig. 1 was designed specifically to assess receptor uptake for a large number of samples, since endocytosis could be started simultaneously in an assay by transferring an entire rack of cell samples to a 37 "C water bath. One disadvantage, however, is that this method under- estimates the time when receptors first appear in the cell because of the 1-2 min required for cells to warm from 4 to 37 "C. Although for most experiments this short time lag would not adversely affect the overall results, it does become important when determining the earliest time at which recep- tors can be found within cells. Therefore, in Fig. 2 we used a slightly modified protocol in which insulin (100 ng/ml) was added to cells already prewarmed to 37 "C. Under these con- ditions, receptors were found in the cell-interior as early as 30 s, and the time required to reach steady state was about 6 min ( tH - 2.7 min). In other words approximately 20,000 receptors/min were internalized and recycled, and receptors traversed this recycling pathway within 6 min. These values are in excellent agreement with the results obtained in Fig. 1 where receptor uptake was calculated after correcting for the initial 2-min lag period.

During insulin treatment the number of total receptors remained constant, while surface-derived receptors accumu- lated intracellularly (Fig. 1); therefore, there must have been a concomitant loss of cell-surface receptors. To verify this, we exposed adipocytes to insulin (100 ng/ml) for various times at 37 "C, washed cells to remove extracellular and receptor- bound ligand, and then measured specific '251-insulin binding

to intact adipocytes after a 3-h incubation at 16 "C. As shown in Fig. 3A, insulin rapidly produced a loss of cell-surface receptors, but upon insulin removal receptors reappeared on the cell-surface. Since only 10% of the total complement of receptors is intracellular and consists of nascent receptors which are reinserted into the plasma membrane relatively slowly (8), it is apparent that most receptors appearing on the cell-surface must be derived from the receptors translocated to the cell-interior during the first phase of the experiment. At each of the indicated times, insulin binding was measured in adipocytes that were exposed to insulin at 4 "C (to block endocytosis), and under these conditions no changes in recep- tor number were noted (data not shown). Thus, down-regu- lation was not due to residually bound insulin. When the loss of cell-surface receptors was calculated from the binding data in Fig. 3A, about 100,000 receptors were lost from the plasma membrane at equilibrium (Fig. 3B), and this value is in excellent agreement with the number of receptors recovered in the cell-interior (Fig. 1B). Thus, when Figs. 1-3 are consid- ered together, it is apparent that insulin causes a stoichio- metric redistribution of receptors, and in the presence of insulin the size of the intracellular receptor pool is greatly increased due to a dynamic equilibrium between receptor uptake and receptor recycling.

In Fig. 4A the redistribution of receptors from the cell- surface to the cell-interior was followed over time in adipo- cytes exposed to insulin at 4 "C and warmed to the indicated temperatures. The temperature-dependent slowing of receptor uptake at 24 "C, compared to 37 "C, and the prolonged time to reach a new equilibrium suggest both receptor internaliza- tion and recycling were affected at the lower temperature.

I A C'ELL-~URF~CE ~ E C E P ~ O R S ' 1

m

0 1 2 3 4 5 6 7 8 9 1 0 TIME (minutes)

FIG. 2. Intracellular accumulation of insulin receptors after the addition of insulin. An experiment protocol identical to that described in the legend to Fig. 1 was used with one important difference. Rather than adding insulin a t 4 "C and warming cells to 37 "C, insulin was added to adipocytes prewarmed to 37 "C. Thus, the lag time due to temperature was eliminated. B/F, boundlfree.

1 1 1 1 I

0 5 10 15 20 25 30

TIME (minutes) FIG. 3. Effect of insulin on the number of cell-surface in-

sulin receptors. A, cells were exposed to 100 ng/ml of insulin at 4 "C and then warmed to 37 "C. At the indicated times adipocytes were cooled to 4 "C, washed to remove extracellular and receptor- bound ligand, and then incubated at 16 "C with '"1-insulin (0.2 ng/ ml) for 3 h in the absence (total binding) or presence of 10 Fg/ml unlabeled insulin (nonspecific binding). E, loss of cell-surface recep- tors as a function of time was determined from the insulin-binding data in panel A. B/F, boundlfree.

Translocation and Recyc :ling of Insulin Receptors 4139 I l I l I I I I *

45 - (Total Receptor Pool)

- 0 0 - X Y \

E. L3

0

m

z z z

I

1 a

2 4 6 8 10 12 15 20 30 TIME (minutes)

FIG. 4. Effect of temperature and energy depleters on the size of the intracellular receptor pool as a function of insulin treatment time. A, cells were exposed to 100 ng/ml of insulin and warmed to 16,24, or 37 'C for the indicated times. Specific '251-insulin to soluble intracellular receptors was then determined as in Fig. 1. B, adipocytes were preincubated at 37 "C in the absence (control) or presence of 1 mM dinitrophenol (DNP) and then exposed to insulin for the indicated times. B/F, hound/free.

Further, since the actual size of the intracellular receptor pool at equilibrium was smaller at 24 than at 37 "C, it may be that the rate of receptor uptake was slowed proportionally more than recycling. At 16 "C receptor uptake was negligible, and this agrees well with our previous finding that both ligand uptake (10, 11) and insulin-induced down-regulation (1) are minimal at 16 "C compared to the rate at 37 "C. Also in agreement with earlier studies (1) is the current finding that endocytosis of insulin-receptor complexes is an energy-de- pendent process inhibited by depleters of metabolic energy (Fig. 4B).

Both chloroquine and Tris have previously been shown to inhibit the recycling of insulin receptors in adipocytes (2, 3). Therefore, we used these agents to confirm the time required for receptors to trasverse the recycling pathway. These results are shown in Fig. 5A and reveal that after 10 min of insulin treatment the intracellular receptor pool was greater in both Tris and chloroquine-pretreated cells compared to control cells. The most likely explanation for this difference is that in control cells internalized insulin receptors began appearing on the cell surface by about 8 min, thus establishing a new steady state. When recycling was impaired, a greater number of receptors accumulated intracellularly, and the time when these pool sizes diverged (8-10 min) confirms our earlier estimates of the time required for receptors to traverse the recycling pathway. It is particularly important to note that since neither Tris nor chloroquine effected the initial uptake of insulin-receptor complexes, these agents have a selective

n z m z 35

L3 0 2 4 6 8 1 0 1 2 15 20

z z n

l l l l l l 1 1 I 1 B Insulin Removed

.)

E 1 I , , , 1 1 1 1 1 1 1 Ly

v3 Lz 2 4 6 8 1 0 1 2 1 5 2 0 25 30

TIME (minutes)

FIG. 5. Effect of Tris and chloroquine on insulin receptor recycling. A, the intracellular pool of insulin receptors was measured at various times after cells were pretreated with 35 mM Tris or 0.2 mM chloroquine (CQ; 15 min at 37 "C) and then exposed to insulin (100 ng/ml). B, adipocytes were pretreated at 37 "C for 15 min in the absence (controls) or presence of 0.2 mM chloroquine, exposed to insulin for 10 min, cooled to 4 'C, washed to remove extracellular and receptor-bound insulin, and then (after replacing chloroquine) warmed to 37 "C for various times. Specific insulin binding to intra- cellular receptors was measured as in Fig. 1. B/F, bound/free.

effect on receptor recycling at a site distal to the endocytosis of insulin-receptor complexes.

The concentrations of inhibitor (Tris or CQ) used in the above studies were previously shown to be maximally effective in blocking recycling as measured by the rapidity and extent of insulin-induced down-regulation (12). However, based on our current findings that these agents do not alter the rate of receptor uptake (Fig. 5A) and that receptor uptake is ex- tremely rapid (Figs. 1-3), one would expect a 30% loss of surface receptors within 6-8 min rather than the observed 1 h (1). To reconcile these differences, we hypothesized that recycling inhibitors such as chloroquine are only partially effective in blocking recycling. This idea was tested by pre- treating adipocytes for 15 min with a maximally effective concentration of chloroquine to inhibit recycling and then treating cells with insulin for 10 min to preload the cell- interior with surface-derived receptors. Further influx of re- ceptors was prevented by removing insulin at 4 "C. After readding chloroquine and warming cells to 37 "C, it can be clearly seen in Fig. 5B that intracellular receptors were lost from the cell-interior despite the continued presence of chlor- oquine. In other studies we found that receptors reappeared on the cell-surface concomitant with this loss (data not shown); thus, the finding that most of the internalized recep- tors were recycled back to the plasma membrane in the

4140 Translocation and Recycling of Insulin Receptors

presence of chloroquine supports our hypothesis that inhibi- tors of recycling are, at best, only partially effective.

Although 35 mM Tris and 0.2 mM chloroquine impair insulin receptor recycling to an equal extent as estimated by the extent of insulin-induced loss of cell-surface receptors (12), we recently highlighted an important difference in the cellular action of these two agents (12). Tris was found to selectively impair receptor recycling while leaving the insulin- degradative pathway intact, whereas chloroquine at low doses markedly inhibited insulin degradation and only at higher doses did it affect recycling. Since the inability of Tris to impair insulin degradation indicates that adipocytes are still capable of degrading incoming proteins, we next determined whether or not insulin receptors were degraded under condi- tions where receptor recycling was impaired.

As shown in Fig. 6A, after 4 h of insulin treatment (100 ng/ ml) the number of total cellular receptors (cell-surface + intracellular receptors) remained constant. In other words, although about 5 million receptors were calculated to have been internalized and recycled (20,00O/min), no receptor deg- radation was detected. However, when Tris was added, the total receptor pool decreased at a linear rate of 600 receptors/ min (Fig. 6B), indicating receptors were routed to a degrada- tive pathway. Over the time frame of the experiment, it is

u LL

n V W

v)

v)

0 I-

V W

W

a

n

a LL 0

WJ v)

s

l o / 5

150,000 -

200,000 -

250,OOOL I I 0.5 1 2 3 4

TIME (hours) FIG. 6. Ability of Tris to divert incoming insulin receptors

to a degradative pathway. A, specific '=I-insulin binding to the total pool of insulin receptors (cell-surface + intracellular) was mea- sured at various times over 4 h after exposing cells (at 37 " C ) to 100 ng/ml of insulin either alone or in the presence of 35 mM Tris. B, in a separate experiment 'Z61-insulin binding to cell-surface receptors was measured after incubating cells for various times at 37 "C in the presence of insulin plus Tris. From this binding data, loss of cell- surface receptors was calculated and is plotted for each time. The

the binding data in Fig. 7A. BIF, boundlfree. loss of total receptors was similarly calculated and plotted based on

unlikely that de mu0 receptor synthesis played a significant role in maintaining the receptor number constant since ex- periments performed in the presence of cycloheximide yielded similar results (data not shown). Further, it is likely that receptor degradation was occurring intracellularly and related to insulin-induced receptor uptake, since no net receptor loss was seen over 4 h when cells were exposed to Tris alone (data not shown).

To examine the relationship between receptor degradation and the loss of cell-surface receptors in Tris-treated cells exposed to insulin, we measured the loss of receptors from the plasma membrane over 4 h and plotted these results together with the loss of total receptors (Fig. 6B). Although there may be other interpretations, it appears that many more receptors are lost from the cell-surface than are degraded, indicating that the number of trapped incoming receptors destined for degradation exceeds the capacity of the receptor catabolic pathway. During the first hour of insulin treatment about 1,700 receptors/min were lost from the cell-surface due to the inability of such receptors to recycle, which is relatively slow compared to the rate at which receptors were internalized and recycled (about 20,00O/min, Fig. 2). In fact, when these rates are compared it appears that Tris prevents only 8.5% of all endocytosed receptors from returning to the cell-surface. In other words, in previous studies (2) we underestimated the number of insulin receptors that were undergoing internali- zation by about 10-fold, and this was due to the fact that greater than 90% of receptors were still able to recycle in the presence of maximally effective concentrations of Tris or chloroquine.

The site of receptor degradation was examined by pretreat- ing adipocytes with chloroquine before exposing cells to Tris plus insulin. Since chloroquine had no effect on the insulin- induced loss of cell-surface receptors when recycling was inhibited by Tris, but did effectively prevent the loss of total cellular receptors (Fig. 7B), these results reveal that receptor degradation occurs intracellularly (Fig. 7). In other words, receptor degradation is minimal following insulin-mediated receptor uptake, due to the efficiency of recycling; however, when recycling is impaired, receptors are catabolized intra- cellularly through a chloroquine-sensitive degradative path- way.

DISCUSSION

In earlier adipocyte studies we found that internalized receptors are rapidly recycled back into the plasma membrane following endocytotic uptake of insulin-receptor complexes (2, 3). However, we were unable at this time to examine the kinetics of these processes without the use of recycling inhib- itors, such as chloroquine or Tris, because of the rapidity of receptor internalization and recycling. In retrospect, our ina- bility to detect changes in receptor distribution in the absence of inhibitors is clear. After insulin treatment our standard procedure for removing extracellular and receptor-bound in- sulin entailed washing cells several times in insulin-free buffer and then further incubating adipocytes for 1 h at 37 "C to dissociate surface-bound ligand. It is now obvious that this method allowed endocytosed receptors to be reinserted back into the plasma membrane so that a net loss of cell-surface receptors was not seen. Recently, Knutson et al. (13) in their studies on 3T3-Ll cells have highlighted this idea that repar- titioning of receptors between the cell-surface and the cell- interior during cell preparation could obscure potential differ- ences in receptor distribution.

Two methodological procedures have made the current studies possible. The first is a method for removing both

Translocation and Recyc

1 A CELL-SURFACE RECEPTORS I 4

3

2

z 1

z m INSULIN INSULIN INSULIN INSULIN

L3

0

z s "r0 TRlS , TRlS , 2- 0.2 mM CHLOROOUINE

CONTROL + CONTROL +

z x

- E - n u uJ\ 30 B TOTAL RECEPTORS N

u u- 25

CL 20

15

10

5

0 w

v)

INSULIN INSULIN INSULIN INSULIN CONTROL + CONTROL +

TRlS , TRlS , 0.2 mM CHLOROOUINE

FIG. 7.. Inhibition of receptor degradation by chloroquine. A, cells were preincubated at 37 "C for 2 h with insulin (100 ng/ml) in the absence (black bars) or presence (hatched bars) of 35 mM Tris. After washing cells at 37 "C to remove extracellular and receptor- bound insulin, specific 1261-insulin binding was determined after a 3- h incubation period at 16 "C. B, a similar protocol was used to measure the total receptor pool except after the washing procedure cells were solubilized and the binding to soluble receptors was measured after a 16-h incubation period at 4 "C. B/F, bound/free.

extracellular and receptor-bound insulin and tryptically in- activating surface receptors at low temperatures (4 "C). Thus, repartitioning of receptors during cell handling is prevented. The second involves a new method to quantitate solubilized receptors using a rapid one-step method to remove Triton X- 100 and concentrate receptors from control and trypsinized adipocytes after cell solubilization (44). When used together the kinetics of insulin receptor uptake and recycling in adi- pocytes can be accurately determined.

In applying these techniques we found that in the absence of insulin 90% of the total receptor pool was on the cell- surface and 10% was intracellular; however, when cells were exposed to 100 ng/ml insulin surface-derived receptors accu- mulated within cells at the rate of -20,00O/min and could be found at an intracellular site as early as 30 s after the addition of ligand. After 6 min the insulin-induced repartitioning of receptors reached a new equilibrium in which 30% of cell- surface receptors could, at any one time, be found in the cell- interior. Most probably, this new distribution represents a dynamic steady state between receptor internalization and subsequent recycling, since removal of extracellular insulin (to halt receptor uptake) was rapidly followed by a decrease in the size of the intracellular receptor pool and a concomitant and stoichiometric reappearance of receptors on the cell- surface. Further, since equilibrium was reached within 6 min after the addition of insulin and since we believe that the

ding of Insulin Receptors 4141

rapid plateau in the size of the intracellular receptor pool is due to recycling of receptors back to the cell-surface at a rate equal to receptor uptake, we conclude that receptors traverse the entire recycling pathway from endocytosis to reinsertion within 6 min. It should be noted that our estimated time for the recycling of insulin receptor in adipocytes is in excellent agreement with those times reported for the recycling of low density lipoprotein receptors in fibroblasts (14) and asialo- glycoprotein receptors in both hepatocytes (15) and HepG2 hepatoma cells (16-18).

Despite the general consensus on the rapidity of receptor recycling in various cell types, there is considerable contro- versy as to whether receptors are recycled constitutively or through an inducible process. Specifically, are receptors con- tinually internalized and recycled independent of occupancy (constitutive), or does ligand binding trigger endocytotic up- take of ligand-receptor complexes and subsequent receptor recycling (inducible process)? By far, the most convincing evidence for constitutive recycling comes from the observation that weak bases (e.g. chloroquine) or carboxylic ionophores (e.g. monensin) inhibit recycling and deplete greater than 50% of cell-surface receptors within 20 min, even in the absence of ligand. This has been convincingly shown for low density lipoprotein receptors in fibroblasts (19), mannose receptors in macrophages (20, 21), and asialoglycoprotein receptors on isolated hepatocytes (22) and hepatoma HepG2 cells (18). If receptors are indeed recycled constitutively, then it follows that even in the absence of ligand, a portion of the intracel- lular receptor pool would consist of receptors undergoing recycling which, therefore, would be capable of being rapidly reinserted into the plasma membrane. In many of the receptor systems mentioned above, this has been clearly shown by trypsinizing cells at 4 "C to destroy cell-surface receptors and demonstrating that receptors reappear on the cell surface within minutes after warming to 37 "C (20,23).

Arguments for an inducible mechanism whereby ligand binding actually initiates endocytotic uptake have been de- rived mostly from kinetic experiments. For example, it has been shown by Ciechanover et al. (16) that asialoglycoprotein receptors internalize and recycle independently of transferrin and insulin receptors. In addition, the loss and reappearance of cell-surface receptors is specific and dependent on the prebinding of ligand to receptors. Interestingly, in these stud- ies HepG2 hepatoma cells were used to gather evidence for an inducible mechanism of recycling, and this is the same cell-type used in experiments to support constitutive recycling (18). As a way of reconciling these apparent differences it has been postulated by several investigators (14, 16) that ligand binding does indeed trigger the redistribution of receptors from the plasma membrane to the cell-interior; however, this could occur through a constitutive recycling mechanism that returns occupied receptors to the surface more slowly than unoccupied receptors. In other words, ligand-induced receptor redistribution is due to a slowing of recycling rather than initiation of endocytotic uptake.

For the insulin receptor system in isolated adipocytes, all our past and present data are consistent with the idea that receptors undergo little or no constitutive recycling. Rather, insulin appears to trigger receptor internalization through an inducible process. This conclusion is based on several lines of evidence. First, in the absence of insulin inhibitors of receptor recycling (e.g. Tris or chloroquine) have little or no effect on the number of cell-surface receptors, even after several hours of treatment (2, 3, 12). In the presence of insulin, however, either Tris or chloroquine produces a 50% loss of cell-surface receptors within 2 h (1-4), thus revealing that these agents

4142 Translocation and Recycling of Insulin Receptors

can indeed impair recycling. The second line of evidence is derived from the fact that adipocytes, in the absence of insulin, do not possess a pool of intracellular receptors which can be rapidly reinserted into the plasma membrane (8). This would be expected if recycling were occurring constitutively. Rather, these cells contain a small receptor pool (10%) con- sisting mostly, if not entirely, of newly synthesized receptors enroute to the cell-surface, and these receptors are inserted into the plasma membrane relatively slowly at a rate incom- patible with the rapidity of recycling (8). Lastly, these current studies provide kinetic evidence consistent with the inducible uptake and recycling of insulin receptors. This is based on the finding that insulin rapidly triggers the uptake and recy- cling of 20,000 receptors/min which alters receptor distribu- tion such that at any one time about 100,000 surface receptors can be found within the cell-interior. Upon removal of extra- cellular insulin (to halt receptor uptake) these receptors rap- idly reappear on the cell-surface. When considered together, these data provide compelling evidence that most receptors on the surface of adipocytes are internalized and recycled through an inducible, rather than a constitutive, process.

An important area of receptor research undergoing inten- sive investigation is the correlation of receptor traffic with the morphological features of various cells. One aspect of such studies is the role of coated pits and vesicles in receptor uptake and recycling. For example, the majority of low density lipoprotein receptor (5040%) on normal human fibroblasts are found localized to coated pits on the cell-surface (24), and the role of coated pits and coated vesicles in the shuttling and sorting of membrane proteins is well established (25). There- fore, the selective concentration of low density lipoprotein in these structures, even in the absence of ligand (24), provides a possible mechanism for the continuous uptake and recycling of receptors independent of occupancy. Adipocytes, however, have few coated pits as revealed by the elegant morphological and ultrastructural studies of Smith and Jarett (26). More- over, localization or clustering of insulin into specialized coated pit regions on the plasma membrane appears not to play a role in insulin-mediated receptor uptake as evidenced by the absence of monomeric ferritin-labeled insulin in such unique structures. Therefore, the question arises as to whether cell-surface proteins, such as receptors, can be concentrated and selectively internalized in structures other than the tra- ditional clathrin-coated pits and coated vesicles. If so, then the morphological features of cells may play a role in the way receptors are endocytosed (inducible versus constitutive up- take), processed, or routed within the cell.

In general, endocytosis of insulin-receptor complexes is agreed to be rapid in almost all cell types studied, regardless of the experimental methods used to quantitate endocytotic uptake. For example, subcellular fractionation studies have biochemically identified insulin-receptor complexes within liver cells as early as 1-2 min after insulin treatment whether uptake of the receptor itself was followed (27, 28) or whether the internalization rate of '251-insulin was determined (29). In adipocytes, comparably rapid internalization times were found by Smith and Jarett (30) in morphological studies where they determined the rate at which ferritin-labeled insulin was internalized by using quantitative electron microscopy. Sim- ilarly, in photoaffinity-labeling studies internalization of co- valently linked insulin-receptor complexes was seen after 2- 10 min regardless of whether receptor partitioning was as- sessed by the insensitivity of intracellular receptors to tryptic digestion (5, 31) or through localization and identification of receptors in various subcellular compartments (32). The cur- rent finding that receptors are found within adipocytes within

30 s after the addition of ligand highlights the rapidity of receptor uptake and leads to the conclusion that endocytosis of insulin-receptor complexes can now be listed among the earliest events following the binding of insulin to cell-surface receptors.

Once internalized, it appears that insulin receptors are processed through one of three alternative, but not mutually exclusive, pathways. Specifically, following endocytotic up- take, intercellular receptors can be: 1) reinserted back into the plasma membrane (recycling pathway); 2) sequestered within the cell with unaltered binding capacity (translocation pathway); or 3) degraded or inactivated (catabolic pathway). Adipocytes clearly represent the best example of a cell type that can process internalized receptors predominately through a recycling pathway. Thus, we observed no loss of either total or cell-surface receptors after 3 h of insulin treatment (Fig. 6A and Ref. 2), although during this time the equivalent of 12 times the total receptor complement would have been endo- cytosed. The sequestration pathway is best exemplified by the studies of Krupp and Lane (331, who found that treatment of chick liver cells with insulin for 18 h resulted in a 60% loss of cell-surface receptors with no change in the number of total receptors. This data, coupled with the fact that insulin af- fected neither insulin receptor synthesis nor degradation, led these authors to conclude that insulin-induced receptor down- regulation, at least in these cells, occurs through translocation of receptors from the plasma membrane to the cell-interior where they are sequestered. Evidence for a third pathway (receptor degradation/inactivation pathway) is derived from the finding that insulin can accelerate insulin receptor turn- over by increasing the overall rate of receptor degradation (34, 35); however, few studies have convincingly localized the cellular site where such receptor degradation occurs. Of those few studies that have localized the degradative site photoaf- finity-labeling techniques were used to show that receptor catabolism occurred at an intracellular chloroquine-sensitive site (5, 6).

In view of the fact that endocytosed insulin receptors appear to be processed by various cells in different ways, the question arises as to whether each cell type has the potential to process receptors through several different pathways. Or, is each cell specialized in its ability to handle incoming receptors exclu- sively through one pathway to the exclusion of the others? Available evidence tends to support the idea of multiple pathways and is derived from studies showing that adipocyte insulin receptors can be either recycled or degraded under different circumstances. For example, in the current study we followed receptor transit between the cell-surface and the cell- interior under physiological conditions (where insulin binding and intracellular dissociation is not impaired) and found receptor uptake tightly coupled to an efficient recycling mech- anism so there was no measurable receptor degradation. How- ever, a catabolic pathway does exist as clearly demonstrated by adipocyte photoaffinity-labeling experiments in which "'1- insulin was covalently attached to cell-surface receptor so that the receptor processing could be followed without the problem of dissociation of the radiolabeled ligand. In these studies, a large percentage of photolabeled receptors was degraded at an intracellular site through a chloroquine-sen- sitive process (5, 6). Receptor degradation was also seen in the present study after pretreating cells with Tris to block recycling and then exposing cells to insulin. Thus, different methodological approaches have convincingly demonstrated the existence of two receptor-processing pathways in adipo- cytes: one for the recycling of receptors and the other for receptor degradation.

Translocation and Recycling of Insulin Receptors 4143

Although the co-existence within one cell type (adipocytes) of two receptor pathways is inherently interesting, it is im- portant to understand why different techniques and various agents lead to the differential processing of insulin receptors (recycling versus degradation). One possible explanation re- lates to the ability of insulin to effectively dissociate from the receptor after endocytotic uptake of the insulin-receptor com- plex. In photoaffinity labeling studies insulin cannot disso- ciate simply because the ligand is covalently attached to the receptor, whereas pretreatment of cells with lysosomotropic agents such as chloroquine inhibits acidification of endocy- totic vesicles (36) in which the pH-dependent dissociation of receptor-ligand complexes is thought to occur (37). S‘ mce we know that at some point after endocytotic uptake of insulin- receptor complexes the ligand separates from the receptor and is degraded, while the receptor is recycled back to the plasma membrane (12), it is possible that inhibition of this dissociation process diverted the receptor from its normal recycling route and shunted it to the degradative pathway of the ligand. Indeed, our current findings provide experimental evidence consistent with such a hypothesis. In the presence of a high concentration of insulin a net loss of total cellular receptors was not seen, even after 4 h, presumably due to efficient recycling of internalized receptors. However, when cells were treated with Tris, an agent shown to selectively impair recycling without altering insulin degradation (12), receptors were trapped within cells and subsequently degraded by a chloroquine-sensitive process. These results reveal that the functional consequence of impaired recycling is the shunt- ing of receptors to a catabolic compartment and suggest that this may be mediated by the inability of insulin receptors to dissociate after internalization. Additional studies will be necessary to determine whether this is the same catabolic pathway normally used by adipocytes to degrade receptors under physiological conditions or whether this route exists only as a result of abnormal insulin-receptor interactions during intracellular transit.

Thus far, emphasis has been placed on the idea that there are several distinctive pathways for the processing of insulin receptors (recycling, sequestration, and degradation); how- ever, the issue of multiple pathways for receptor movement and the effects of agents that perturb receptor traffic can have additional levels of complexity. This is illustrated by the studies of Tietze et al. (23) which show that within the recycling pathway there are at least several routes through which internalized receptors can recycle to the cell-surface. These investigators found two functionally distinct pools of asialoglycoprotein receptors within macrophages. The first, referred to as the cycling pool, carried intact ligand back to the cell-surface, presumably still bound to the receptor, and this route was unaffected by lysosomotropic agents. The sec- ond receptor pool was postulated to recycle back to plasma membrane unoccupied, after it delivered the ligand to low pH vesicles for eventual transfer to lysosomes. In this case since ligand delivery was pH dependent, lysosomotropic agents such as chloroquine inhibited the dissociation process, thereby preventing the subsequent return of receptors.

Although the above studies highlight the complex nature of receptor traffic and reveal possible difficulties in interpreting data in which various agents are used to disrupt receptor processing, they also provide a foundation for interpreting our present data which reveals that chloroquine and Tris inhibit the recycling of insulin receptors by only 10%. Thus, this finding could be explained by postulating that insulin recep- tors in adipocytes can also recycle back to the cell-surface by alternative pathways. One such pathway would be dependent

upon ligand dissociation and consequently inhibitable by ly- sosomotropic agents and Tris, while the other pathway recy- cles receptors to the plasma membrane together with intact ligand by a mechanism not inhibitable by lysosomotropic agents. In fact, data has recently been obtained’ which strengthens this idea. We found that 25% of endocytosed 1251- insulin is released intact by adipocytes, and this retroendo- cytosis of intact ligand is unaffected by chloroquine. In addi- tion, it has recently been reported that when insulin-receptor complexes on adipocytes are covalently linked through pho- toaffinity-labeling techniques, occupied receptors are still ef- fectively recycled, while the remainder are degraded (38). Although these findings suggest dual pathways for the recy- cling of insulin receptors, additional studies are underway to more critically examine these hypotheses.

Although we have elucidated the characteristics and kinet- ics of insulin receptor uptake and recycling in adipocytes, the functional role of these processes in modifying biological responsiveness to the action of insulin remains to be deter- mined. Generally speaking, internalization of insulin-receptor complexes could influence cell function in at least two ways: an indirect effect through regulation of cell-surface insulin receptors that, in turn, would alter tissue sensitivity to cir- culating insulin; or by a more direct mechanism in which incoming insulin-receptor complexes alter post-binding events along the insulin action pathway. Although at this point the relationship between receptor uptake and recycling and the biological action of insulin is only speculative, infor- mation has been recently obtained on the role of insulin in mediating down-regulation of surface receptors. When we studied insulin-mediated receptor regulation in isolated adi- pocytes which were maintained in primary culture under physiological conditions where recycling was operative (39), we found that insulin “triggered” down-regulation such that receptor loss continued even in the absence of ligand. There- fore, since only unoccupied receptors were on the cell-surface during the time receptors were actually undergoing down- regulation, it is unlikely that receptor uptake and recycling played a direct role in modulating receptor number. However, we did not rule out the possibility that uptake of insulin- receptor complexes at early times preceding receptor loss may have been involved in the actual triggering of this regulatory mechanism. Hopefully, with the ability to directly measure both receptor uptake and recycling under physiological con- ditions, we can begin more detailed studies on the regulation and function of insulin receptor recycling.

Acknowledgment-I would like to thank Cleon B. Tate for his secretarial assistance.

REFERENCES

1. Marshall, S., and Olefsky, J. M. (1981) Diabetes 30, 746-753 2. Marshall, S., Green, A., and Olefsky, J. M. (1981) J. Biol. Chem.

3. Marshall, S., and Olefsky, J. M. (1982) Eiochem. Eiophys. Res.

4. Green, A., and Olefsky, J. M. (1982) Proc. Natl. Acad. Sci. U. S.

5. Berhanu, P., Kolterman, 0. G., Baron, A., Tsai, P., Olefsky, J. M., and Brandenburg, D. (1983) J. Clin. Inuest. 728, 1958- 1970

6. Heidenreich, K. A., Berhanu, P., Brandenburg, D., and Olefsky, J. M. (1983) Diabetes 32,1001-1009

7. Marshall, S., and Olefsky, J. M. (1980) J. Clin. Inuest. 66, 763- 772

8. Marshall, S. (1983) Diabetes 32, 319-325

* S. Marshall, manuscript in preparation.

256,11464-11470

Commun. 102,646-653

A. 79,427-431

4144 Translocation and Recycling of Insulin Receptors

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21. 22.

23.

24.

25.

Olefsky, J. M., Jen, P., and haven, G. M. (1974) Diabetes 23,

Marshall, S., and Olefsky, J. M. (1979) J. Biol. Chem. 254,

Marshall, S., and Olefsky, J. M. (1980) Endocrinology 107,1937-

Marshall, S., and Olefsky, J. M. (1983) J. Cell. Physiol. 117, 195-

Knutson, V. P., Ronnett, G. V., and Lane, M. D. (1983) J. Biol.

Brown, M. S., Anderson, R. G. W., and Goldstein, J. L. (1983)

Geuze, H. J. J. W., Slot, J. W., Strous, G. J. A. M., Lodish, H. F.,

Ciechanover, A. J., Schwartz, A. L., and Lodish, H. F. (1983) Cell

Schwartz, A. L., Fridovich, S. E., and Lodish, H. F. (1982) J. Biol.

Schwartz, A. L., Bolognesi, A,, and Fridovich, S. E. (1984) J. Cell

Basu, S. K., Goldstein, J. L., Anderson, R. G. W., and Brown, M.

Tietze, C., Schlesinger, P., and Stahl, P. (1980) Biochem. Biophys.

Kaplan, J., and Keogh, E. A. (1981) Cell 24,925-932 Tolleshaug, H., and Berg, T. (1979) Biochern. Phurmacol. 28,

Tietze, C., Schlesinger, P., and Stahl, P. (1982) J. Cell Biol. 92,

Anderson, R. G. W., Vasile, E., Mello, R. J., Brown, M. S., and

Helenius, A,, Mellman, I., Wall, D., and Hubbard, A. (1983)

565-571

10153-10160

1945

203

Chem. 258, 12139-12142

Cell 32,663-667

and Schwartz, A. L. (1983) Cell 32,277-287

32,267-275

Chem. 257,4230-4237

Biol. 98, 732-738

S. (1981) Cell 24,493-501

Res. Commun. 93,l-8

2919-2922

417-424

Goldstein, J. L. (1978) Cell 33, 273-285

Trends Biochem. Sci. 8, 245-249

26. Smith, R. M., and Jarett, L. (1983) J. Cell. Physiol. 115, 199- 203

27. Desbuquois, B., Willeput, J., and Huet de Froberville, A. (1980)

28. Desbuquois, B., Lopez, S., and Burlet, H. (1982) J. Biol. Chem.

29. Khan, M. N., Posner, B. I., Khan, R. J., and Bergeron, J. J. M.

FEBS Lett. 106,338-344

257,10852-10860

(1982) J. Biol. Chem. 257,5969-5976 30.

31.

32.

33.

34.

35.

36.

Smith, R. M., and Jarett, L. (1982) Proc. Nutl. Acud. Sci. U. S.

Fehlmann, M., Carpentier, J. L., Van Obberghen, E., Freychet, P., Thamm, P., Saunders, D., Brandenburg, D., and Orci, L. (1982) Proc. Natl. Acud. Sci. U. S. A. 79,5921-5925

Wang, C., Sonne, O., Hedo, J. A., Cushman, S. W., and Simpson, I. A. (1983) J. Biol. Chem. 258, 5129-5134

Krupp, M., and Lane, M. D. (1981) J. Biol. Chem. 256, 1689- 1694

Ronnett, G. V., Knutson, V. P., and Lane, M. D. (1982) J. Biol. Chem. 257,4285-4291

Kasuga, M., Kahn, C. R., Hedo, J. A., Van Obberghen, E., and Yamada, K. (1981) Proc. Nutl. Acud. Sci. U. S. A . 78, 6917- 6921

A. 79,7302-7306

Maxfield, F. R. (1982) J. Cell Biol. 95,676-6n1 37. Harford, J., Bridges, K., Ashwell, G., and Klaiiner, R. D. (1983)

38. Heidenreich, K. A., Brandenburg, D., Berhanu, P., and Olefsky,

39. Marshall, S., Garvey, W. T., and Geller, M. (1984) J. Bid. C!hem.

J. Biol. Chem. 258,3191-3197

J. M. (1984) J. Biol. Chern. 259,6511-6515

259,6376-6384 40. Deleted in proof. 41. Rodbell, M. (1964) J. Biol. Chern. 239, 375-380 42. Hirsch, J., and Gallian, E. (1968) J. Lipid Res. 9, 110-119 43. Cuatrecasas, P. (1972) Proc. Nutl. Acud. Sci. U. S. A. 69, 318-

44. Marshall, S. (1985) J. Biol. Chem. 260, 4128-4135 323