Insulin Enhances Macrophage Scavenger Receptor-mediated Endocytic Uptake of Advanced Glycation End Products*

Hyperglycemia accelerates the formation and accumulation of advanced glycation end products (AGE) in plasma and tissue, which may cause diabetic vascular complications. We recently reported that scavenger receptors expressed by liver endothelial cells (LECs) dominantly mediate the endocytic uptake of AGE proteins from plasma, suggesting its potential role as an eliminating system for AGE proteins in vivo (Smedsrød, B., Melkko, J., Araki, N., Sano, H., and Horiuchi, S. (1997) Biochem. J. 322, 567–573). In the present study we examined the effects of insulin on macrophage scavenger receptor (MSR)-mediated endocytic uptake of AGE proteins. LECs expressing MSR showed an insulin-sensitive increase of endocytic uptake of AGE-bovine serum albumin (AGE-BSA). Next, RAW 264.7 cells expressing a high amount of MSR were overexpressed with human insulin receptor (HIR). Insulin caused a 3.7-fold increase in endocytic uptake of 125I-AGE-BSA by these cells. The effect of insulin was inhibited by wortmannin, a phosphatidylinositol-3-OH kinase (PI3 kinase) inhibitor. To examine at a molecular level the relationship between insulin signal and MSR function, Chinese hamster ovary (CHO) cells expressing a negligible level of MSR were cotransfected with both MSR and HIR. Insulin caused a 1.7-fold increase in the endocytic degradation of 125I-AGE-BSA by these cells, the effect of which was also inhibited by wortmannin and LY294002, another PI3 kinase inhibitor. Transfection of CHO cells overexpressing MSR with two HIR mutants, a kinase-deficient mutant, and another lacking the binding site for insulin receptor substrates (IRS) resulted in disappearance of the stimulatory effect of insulin on endocytic uptake of AGE proteins. The present results indicate that insulin may accelerate MSR-mediated endocytic uptake of AGE proteins through an IRS/PI3 kinase pathway.

Prolonged incubation of proteins with glucose leads, through the formation of early stage products, such as Schiff base and Amadori rearrangement products, to the formation of advanced glycation end products (AGE), 1 compounds that have unique properties, such as fluorescence, browning, and cross-linking. Accumulation of AGE proteins has been identified in several tissues in association with aging and age-enhanced disease states, including diabetic complications (1)(2)(3), atherosclerosis (4), hemodialysis-related amyloidosis (5), and Alzheimer's disease (6,7). AGE-modified proteins are known to induce several cellular responses, such as mitogenic activity for macrophages (8), and a chemotactic activity for vascular smooth muscle cells (9). Therefore, the presence and/or accumulation of AGE proteins in arterial walls (4,10) is thought to play an active role in the pathogenesis of diabetic microvascular and macrovascular complications (3,11). However, under physiological conditions, most AGE-modified proteins in plasma should undergo rapid plasma clearance. Thus, following intravenous injection in normal rats, AGE proteins are rapidly cleared from the circulation (12). Such clearance is largely achieved by active endocytic uptake by hepatic sinusoidal cells such as endothelial and Kupffer cells (12,13). It might be hypothesized therefore that the formation of AGE proteins beyond physiological levels or impairment of the AGE elimination system, potentially result in accumulation of AGE in tissues.
Insulin therapy reduces high plasma levels of early stage products, such as hemoglobin A1c and glycated albumin, as well as those of AGE proteins (14). Normalization of glycated proteins levels may thus delay the onset and slow the progression of diabetic complications (15). Insulin treatment may protect the tissues from AGE accumulation indirectly by reducing AGE formation due largely to normalization of plasma glucose levels. However, since AGE proteins are recognized as ligands by AGE receptors in vivo, it is also possible that insulin plays a more direct role in AGE elimination from plasma or tissues by regulating the AGE receptor system. Three types of AGE receptors have been identified so far, including MSR (16,17), RAGE (receptor for AGE) (18), and the receptor complex of OST-48, 80K-H, and galectin-3 (previously called as p60 and p90) (19). Our recent study using peritoneal macrophages from MSR knockout mice has indicated that the endocytic capacity for AGE-BSA by these cells was reduced to approximately 20 -30% of those of wild litter mate mice (20,21), suggesting a major role for MSR in endocytic uptake of AGE proteins by macrophages. Our recent study also showed that liver endothelial cells (LECs), which expressed a large amounts of MSR (20), were responsible for elimination of 60 -65% of intravenously injected AGE proteins from plasma and that the endocytic uptake of AGE proteins by cultured LECs was significantly inhibited by the ligands for MSR, suggesting a major contribution of MSR to endocytosis of AGE proteins by LECs (13). Therefore, in the present study, we focused on MSR as an elimination system for AGE proteins and investigated the effect of insulin on endocytic uptake of AGE proteins by MSR.

EXPERIMENTAL PROCEDURES
Chemicals and Materials-F-12 medium, RPMI 1640 medium, penicillin G, streptomycin, and G418 were purchased from Life Technologies, Inc., hygromycin B and wortmannin from Wako (Osaka, Japan), LY294002 and rapamycin from Biomol Research Laboratories (Plymouth Meeting, PA), and human recombinant insulin from Sigma. Na 125 I and 125 I-human insulin was purchased from Amersham Pharmacia Biotech (Little Chalfont, Buckinghamshire, United Kingdom). Other chemicals were of the best grade available from commercial sources. Culture dishes coated with type IV collagen (35 mm in diameter) were purchased from Becton Dickinson (Bedford, MA). Bovine type II MSR expression vector pXSRII was a kind gift from Dr. Tatsuhiko Kodama (Department of Molecular Biology and Medicine, Research Center for Advanced Science and Technology, University of Tokyo).
Ligand Preparation and Iodination-AGE-BSA was prepared as described previously (12) except for incubation for 40 weeks. Human LDL (d ϭ 1.019 -1.063 g/ml) was isolated by sequential ultracentrifugation of human plasma from normal lipidemic subjects after overnight fasting and dialyzed against 0.15 M NaCl and 1 mM EDTA (22). Acetylated LDL (acetyl-LDL) was prepared by chemical modification of LDL with acetic anhydride as described previously (23). AGE-BSA was labeled with 125 I by IODO-GEN (Bio-Rad) (12), and acetyl-LDL was labeled as described by McFarlane (24) to a specific radioactivity of 850 and 420 cpm/ng, respectively.
Isolation and Culture of LECs-The preparation of pure cultures of functionally intact LECs from a single rat liver has been detailed in the previous paper (25). After collagenase perfusion of the liver, and isopycnic centrifugation of the resulting dispersed cells through Percoll (Pharmacia), pure monolayer cultures of LECs were established by selective attachment to substrates of type IV collagen.
Isolation and Culture of Transfected Cell Lines-Human insulin receptor (HIR) cDNA nucleotides and amino acids were numbered according to the system of Ebina et al. (26). RAW 264.7 cells were transfected with the wild-type HIR expression vector (SR␣IR) and pSV2neo by the lipofection method according to the protocol recommended by the manufacturer (Lipofectin, Life Technologies, Inc.). Clones resistant to 0.4 mg/ml G418 were assayed for the expression of HIR by using 125 I-insulin binding assay as described previously (27). One of five positive clones selected was used for the experiment as HIR-transfected RAW (RAW-HIR) cells. To coexpress HIR and MSR in CHO cells, we used MSR-transfected CHO cells (CHO-MSR cells) as a starting cell, as described in our previous study (17). CHO-MSR cells were isolated according to the original method of Freeman et al. (28). Briefly, bovine type II MSR expression vector, pXSRII, was transfected to CHO cells using polybrene method. To select an MSR-positive clone, cells were cultured in F-12 medium containing 5% fetal calf lipoprotein-deficient serum, 250 M mevalonic acid, 3 g/ml acetyl-LDL, and 40 M compactin. We selected one of the positive clones with a high activity for incorporation of acetyl-LDL labeled by 1,1Ј-dioctadecyl-3,3,3Ј,3Ј-tetramethylindocarbocyanine perchlorate. CHO-MSR cells thus obtained were transfected with the wild-type HIR expression vector (SR␣IR) and pSV2hph by the lipofection method. Clones resistant to 0.4 mg/ml hygromycin B were selected for the expression of HIR in the same way. CHO-MSR cells were also transfected with mutant-type HIR expression vector in which Lys 1030 or Tyr 972 was replaced by methionine (29) and phenylalanine (30) (SR␣IR 1030M and SR␣IR 972F ), respectively. Clones expressing wild-or mutant-type HIR obtained by hygromycin B selection were named as CHO-MSR-HIR, CHO-MSR-HIR 1030M , and CHO-MSR-HIR 972F cells, respectively. HIR-transfected CHO (CHO-HIR) cells were obtained by transfecting SR␣IR and pSV2neo to parent CHO cells and selected against 0.4 mg/ml G418. Each clone expressing wild or mutant HIR thus obtained was subjected to the binding assay using 125 I-insulin. Briefly, cells in each well were incubated for 4 h at 4°C with 0.03 nM 125 I-insulin in the presence of various concentrations (0 -200 nM) of unlabeled insulin, washed three times, and the cell-bound radioactivity was measured as described previously (27). The number of cell surface receptors was calculated by Scatchard analysis. Clones expressing 1.7-3.1 ϫ 10 5 receptors/cell were used in the present experiments.
Cellular Assays-Except for the binding study, all cellular experiments were performed at 37°C in a humidified atmosphere of 5% CO 2 . The endocytic uptake of 125 I-AGE-BSA by LECs was measured as described previously (12) with some modifications. Briefly, 5 ϫ 10 6 of LECs were seeded and maintained in serum-free RPMI 1640 medium in 35-mm diameter wells, washed, and supplied with fresh medium containing 3% BSA and labeled ligands. The cells were incubated for 60 min with or without 10 nM insulin and indicated concentrations of 125 I-AGE-BSA (1.25-10 g/ml) with or without 100-fold unlabeled AGE-BSA in 1.0 ml of KRH buffer (136 mM NaCl, 4.7 mM KCl, 1.25 mM MgSO 4 , 1.25 mM CaCl 2 , 20 mM Hepes, 1 mg/ml glucose, and 30 mg/ml BSA). By contrast, the cells were incubated for 60 min with 2 g/ml 125 I-AGE-BSA and indicated concentrations of human insulin with or without 200 g/ml unlabeled AGE-BSA in 1.0 ml of KRH buffer. Additionally, the cells were incubated with 10 nM wortmannin (31) for 30 min before and after the addition of 2 g/ml 125 I-AGE-BSA and 10 nM insulin. After 60-min incubation with 125 I-AGE-BSA, 0.75 ml of the culture medium was taken from each well and mixed with 0.3 ml of 40% trichloroacetic acid in a vortex mixer. To this solution we added 0.2 ml of 0.7 M AgNO 3 , followed by centrifugation. The resulting supernatant (0.5 ml) was used to determine trichloroacetic acid-soluble radioactivity, which was taken as an index of cellular degradation. The remaining cells in culture dishes were washed three times with PBS containing 1% BSA and three more times with PBS. The cells were lysed at 37°C for 30 min with 1.0 ml of 0.1 N NaOH. One portion was used to determine the radioactivity as the cell-associated ligand, while the other portion was used to determine cellular proteins by BCA protein assay reagent (Bio-Rad).
The endocytic uptake of 125 I-AGE-BSA in each transfected cell line was measured. RAW-HIR cells and mock-transfected RAW (RAWmock) cells were cultured for 24 h on type IV collagen-coated wells (35 mm in diameter) in 1.0 ml of RPMI containing 10% fetal calf serum (FCS) at the final cell density of 8 ϫ 10 5

cells/well. CHO-MSR-HIR cells, CHO-MSR cells, CHO-HIR cells, parent CHO cells, CHO-MSR-HIR 1030M cells, and CHO-MSR-HIR 972F cells were cultured for 24 h on
type IV collagen-coated wells in 1.0 ml of F-12 medium supplemented with 10% FCS at a final cell density of 5 ϫ 10 5 cells/well. After serum starvation for 5 h, the cells were incubated for 60 min with 2 g/ml 125 I-AGE-BSA and selected concentrations of human insulin with or without 200 g/ml unlabeled AGE-BSA in 1.0 ml of KRH buffer. Cellassociated ligands and ligands degraded by those cells were measured as exactly described above.
To determine the effects of wortmannin, LY294002 (32), and rapamycin (33), CHO-MSR-HIR cells were pretreated for 30 min in 0.8 ml of KRH buffer with various concentrations of wortmannin, LY294002, rapamycin, and 0.1% of dimethyl sulfoxide as a vehicle, respectively. RAW-HIR cells were pretreated with these reagents in the same manner except for LY294002. In the next step, we added to the cells in each culture well 0.2 ml of KRH buffer containing labeled ligands to a final concentration of 2 g/ml 125 I-AGE-BSA in the absence or presence of 10 nM insulin. The endocytic degradations of 125 I-AGE-BSA and cell-associated 125 I-AGE-BSA were determined as described above. The endocytic uptake (degradation and cell association) of 125 I-acetyl-LDL by CHO-MSR-HIR cells was determined with 2 g/ml 125 I-acetyl-LDL similar to 125 I-AGE-BSA.
For the cellular binding study, the cells were cultured and subjected to serum starvation as described for the endocytic uptake assay. The cells in each well were preincubated for 60 min at 37°C with or without 10 nM insulin, washed with ice-cold KRH buffer, and replaced with 1.0 ml of KRH buffer containing 1.25-20 g/ml 125 I-AGE-BSA, in the absence or presence of 50-fold excess amounts of unlabeled ligand (50 -1000 g/ml). After incubation of the cells for 2 h at 4°C, each well was washed three times with 1.0 ml of ice-cold PBS containing 1% BSA and three more times with PBS. The cells were lysed, and the cell-bound radioactivity and cellular proteins were determined as described above.
Statistical Analysis-Values are expressed as means Ϯ S.E. Statistical significance was determined by unpaired Student's t test. Differences between insulin-stimulated and control groups were considered significant at p Ͻ 0.05.

RESULTS
Endocytic Uptake of AGE-BSA by LECs-As was shown previously (13), 125 I-AGE-BSA underwent effective receptor-mediated endocytosis by cultured LECs (Fig. 1, A and B). The endocytic uptake of 125 I-AGE-BSA by these cells was significantly enhanced by the presence of 10 nM insulin; at ligand concentrations from 1.25 to 10 g/ml, cell-associated 125 I-AGE-BSA was increased by 122-141%, and the amount of 125 I-AGE-BSA degraded was similarly increased by 113-134% above control (Fig.  1, A and B). When the dose-dependent effect of insulin was examined in the fixed amount of 125 I-AGE-BSA, the amounts of cell-associated 125 I-AGE-BSA, as well as those degraded by LECs, were increased with insulin concentrations up to 1 nM and became a plateau somewhere between 1 and 10 nM, followed by gradual decline at an insulin concentration of 100 nM; cell-associated 125 I-AGE-BSA was increased significantly from 775 to 1108 ng/mg of cell protein/h, and amounts of 125 I-AGE-BSA degraded were also increased significantly from 131 to 173 ng/mg of cell protein/h (Fig. 1, C and D). Under the identical conditions, the effect of 10 nM wortmannin, a reagent for PI3 kinase inhibitor, on the endocytosis was examined in the presence of 10 nM insulin. This drug completely reversed the insulin-induced increase of endocytosis to basal level, but slightly less than that of control ( Fig. 1, C and D). These results suggest a potential link of an insulin signaling pathway to the endocytic system of LECs. Recent studies indicated that MSR expressed by these cells play a major role in the endocytic uptake of chemically modified proteins such as acetyl-LDL (21) and AGE proteins (13). Therefore, to understand the mechanism of insulin-enhanced endocytic uptake of AGE-BSA, we thought it reasonable to use RAW cells, a macrophage-like cell line that is known to express a high level of endogenous MSR (34).
Endocytic Uptake of AGE-BSA by RAW-HIR Cells-We first introduced HIR expression vector to macrophage-like RAW 294.7 cells, since recent studies have shown the MSR-mediated endocytic uptake of AGE proteins by macrophages and macrophage-derived cells (17,20). The level of insulin receptor expressed in RAW-HIR cells was 1 ϫ 10 5 receptors/cell, about 100 times higher than RAW cells; the endogenous level of insulin receptor in RAW cells was negligibly low (Ͻ10 3 receptors/cell) (data not shown). Endocytic uptake of 125 I-AGE-BSA by RAW-HIR cells were examined in the presence of various concentrations of insulin. As shown in Fig. 2A, insulin increased the amount of 125 I-AGE-BSA associated with RAW-HIR cells from 128 to 154 ng/mg of cell protein, whereas RAWmock cells were not influenced by the presence of insulin. The amount of 125 I-AGE-BSA degraded by RAW-HIR cells increased proportionately with insulin concentrations up to 1 nM, but decreased sharply at insulin concentrations of 10 and 100 nM (Fig. 2B). The maximum level reached in the presence of 1 nM insulin was 90 ng/mg of cell protein, whereas the corresponding control level was 30 ng/mg of cell protein (Fig. 2B). RAW-mock cells did not show such an insulin-dependent enhancement of endocytic degradation of 125 I-AGE-BSA. These results indicate that endocytic uptake and degradation of 125 I-AGE-BSA by RAW-HIR cells is regulated by insulin. Since MSR is expected to mediate the endocytic degradation of AGE proteins by RAW cells, it is possible that insulin may act on MSR of RAW-HIR cells via transfected HIR. To test such possibility, CHO cells known to express a negligible level of MSR were coexpressed with MSR and HIR.
Endocytic Uptake of AGE-BSA by CHO-MSR-HIR Cells-CHO-MSR-HIR cells expressing 2.8 ϫ 10 5 insulin receptors/cell (Table I) were used for endocytic uptake study. these cells under identical conditions (Fig. 3A). The amount of cell-associated 125 I-AGE-BSA in CHO-MSR cells (36 ng/mg of cell protein) was significant; about 4-fold that of CHO and CHO-HIR cells (8 -10 ng/mg of cell protein). The amount of cell-associated 125 I-AGE-BSA in CHO-MSR cells was not influenced by insulin. Similar to endocytic degradation of 125 I-AGE-BSA (Fig. 3B), insulin increased the amount of 125 I-AGE-BSA associated with CHO-MSR-HIR cells from 38 to 51 ng/mg of cell protein (Fig. 3A). This phenomenon is directly related to the insulin-enhanced endocytic degradation of 125 I-AGE-BSA in CHO-MSR-HIR cells, since cell association and subsequent lysosomal degradation constitute a continuous set of the endocytic pathway.
Cellular Binding of AGE-BSA to CHO-MSR-HIR Cells-It is generally accepted that cell-associated ligands include those bound to cell surfaces as well as those internalized into the cells. Since insulin significantly increased the amount of cellassociated AGE-BSA in CHO-MSR-HIR cells (Fig. 3A), we performed the binding assay to test whether such increase was due to an increase in cell surface receptors. As shown in Fig. 4, the binding of 125 I-AGE-BSA to CHO-MSR-HIR cells exhibited a saturable binding, but the amount of cell-bound 125 I-AGE-BSA did not change when these CHO-MSR-HIR cells were incubated with 10 nM insulin (Fig. 4).
Endocytic Uptake of AGE-BSA by CHO-MSR-HIR 1030M and CHO-MSR-HIR 972F Cells-To examine whether HIR signaling is necessary for activating MSR function, we transfected two types of mutant HIR expression vectors to CHO-MSR cells. CHO-MSR cells were stably transfected with HIR 1030M in which lysine 1030 as an ATP binding site was replaced by methionine, thus exhibiting the kinase-deficient mutant (29). The other mutant used was HIR 972F , which possessed normal kinase activity but was unable to bind to IRS, because tyrosine 972 serving as a binding site for IRS was replaced with phenylalanine (30). In CHO-MSR-HIR cells, insulin enhanced the cell association (Fig. 5A) as well as endocytic degradation (Fig. 5B) of 125 I-AGE-BSA. However, no such response to insulin was evident in CHO-MSR-HIR 1030M or CHO-MSR-HIR 972F cells. These results suggest that the kinase activity of insulin receptor, especially IRS-1, -2, -3, or -4 pathway (35)(36)(37)(38)(39)(40), plays a key role in insulin-enhanced endocytic uptake of AGE proteins by MSR.

Effect of Wortmannin, LY294002, and Rapamycin on Insulin-enhanced Endocytosis by CHO-MSR-HIR and RAW-HIR
Cells-To examine the insulin signaling pathway to MSR, we examined the effects of several inhibitors on insulin-enhanced endocytic degradation of AGE-BSA in CHO-MSR-HIR cells and RAW-HIR cells. As shown in Fig. 6, insulin-enhanced endocytic degradation of 125 I-AGE-BSA by CHO-MSR-HIR cells was inhibited dose-dependently by wortmannin, a potent PI3 kinase inhibitor, with a 50% inhibitory concentration (IC 50 ) at 1 nM. Consistent with this, the phenomenon observed in CHO-MSR-HIR cells was similarly inhibited by another PI3 kinase inhibitor, LY294002 (Fig. 7A). Furthermore, the insulin-enhanced endocytic degradation of AGE-BSA by RAW-HIR cells disappeared completely by pretreatment with 10 nM wortmannin (Fig. 7C). These results suggest a possible involvement of insulin signaling via PI3 kinase in the insulin-enhanced MSRmediated endocytic uptake of AGE-BSA.
In addition to PI3 kinase, we tested the effects of pp70-S6 kinase (pp70 S6K ) on MSR function, since this kinase is located downstream of PI3 kinase (41). Similar to wortmannin and LY294002, rapamycin inhibits the activation of pp70 S6K , but has no effect on PI3 kinase (42). As shown in Fig. 7, rapamycin did not influence insulin-induced MSR function for endocytic uptake of AGE-BSA in CHO-MSR-HIR cells (Fig. 7A) as well as RAW-HIR cells (Fig. 7C). Insulin-enhanced endocytic uptake was also observed with 125 I-acetyl-LDL, an authentic ligand for MSR, in CHO-MSR-HIR cells (Fig. 7B) and RAW-HIR cells (data not shown). DISCUSSION More than 90% of intravenously injected AGE-BSA is eliminated within 15 min by sinusoidal liver cells such as LECs and  Kupffer cells (13), whereas native BSA is cleared from the circulation with a half-life of about 2 days in healthy rats (43). It is therefore evident that rapid elimination of AGE-BSA by rat sinusoidal cells is not due to the species specificity of the albumin, but to the specific recognition of the modified structure(s) of AGE-BSA by these cells. It is also shown that 60 -65% of the elimination of AGE-BSA is explained by LECs and 24 -28% by Kupffer cells, while contribution of hepatocytes is negligible (Ͻ0.1%) (13). Active endocytic activity for AGE-BSA was also confirmed by in vitro experiments using cultured LECs (13). The recent study using MSR knockout mice revealed that 80% of endocytic uptake of acetyl-LDL by cultured LECs is mediated by MSR (21). Moreover, the endocytic uptake of AGE proteins by cultured LECs was inhibited by 50 -70% by the ligands for MSR (13). These results likely suggest a significant role of MSR in endocytic uptake of these modified proteins by LECs. During the course of our studies of endocytic uptake of AGE proteins by rat cultured LECs, we came across the interesting observation that endocytic uptake of AGE-BSA by cul-tured LECs is significantly enhanced by insulin, which was completely reversed by 10 nM wortmannin (Fig. 1, C and D). For the reasons described above, we thought that a possible explanation would be some interaction of insulin signaling with MSR function. Therefore, we next tested with RAW cells, a macrophage cell line that has a high MSR activity, followed by CHO cells coexpressed with HIR and MSR.
The present study using the transfected cells has clearly shown that insulin up-regulates MSR-mediated endocytic uptake of AGE proteins through the IRS/PI3 kinase pathway in which IRS signaling following insulin receptor autophosphorylation plays an important role. Insulin-enhanced MSR-mediated endocytic uptake of 125 I-AGE-BSA was inhibited by wortmannin and LY294002, but not by rapamycin, indicating that PI3 kinase may be involved in this phenomenon. Thus, it is likely that, although PI3 kinase activity is required for both the activation of pp70 S6K and MSR-mediated endocytic uptake of AGE-BSA, the pathway leading to insulin-enhanced MSR-mediated endocytosis of AGE-BSA may branch at some point downstream of PI3 kinase but upstream of pp70 S6K . Insulinstimulated activation of PI3 kinase has been exemplified in several insulin-induced cellular responses, such as GLUT4 translocation (41,(47)(48)(49), activation of glycogen synthesis (50,51), membrane ruffling (52), trafficking of transferrin receptors (53), and activation of pp70 S6K kinase (41,42). However, no previous study has reported that insulin functionally regulates MSR-mediated endocytic uptake of AGE proteins or modified LDL. Thus, to our knowledge, this is the first demonstration of the presence of functional coupling between insulin-stimulated PI3 kinase and activation of MSR-mediated endocytic uptake of ligands.
MSR gene expression is enhanced via PU.1 and AP-1/ets transcription factors along with macrophage colony-stimulating factor-induced differentiation from human monocytes into macrophages (54 -56). Phorbol ester, platelets-derived growth factor-BB, and platelets-derived growth factor-AB are also able to induce MSR gene expression of vascular smooth muscle cells, but no such effect has been noted for insulin (57). The present study showed that insulin enhanced the endocytic uptake of AGE-BSA by CHO-MSR-HIR cells (Fig. 3), but it failed to increase the number of cell surface MSR expressed by these cells (Fig. 4), indicating that insulin-enhanced endocytic uptake of 125 I-AGE-BSA is not due to the increased number of cell-surface MSR, but rather to certain post-binding events. The exact mechanism is not known at present. However, it could be due to an insulin-induced acceleration of the rate of endocytic degradation of AGE-BSA by these cells (Fig. 8). A similar phenomenon was observed with insulin-induced GLUT4 recruitment from the intracellular membrane vesicular pool to the surface membrane (47)(48)(49).
In RAW-HIR and CHO-MSR-HIR cells, insulin stimulation may accelerate exocytosis of degraded products following the classical coated pit/endosomal pathway (58). Several receptors, such as LDL receptor and MSR, are known to be enriched in coated pits as clusters, from where ligand-receptor complexes are internalized into the cell, thus inducing the formation of endosomes. Endosomes are also formed constitutively to some extent even under nonstimulated conditions. Wortmannin and LY294002 did not affect the constitutive type of endocytic uptake both in CHO-MSR-HIR cells (Fig. 7) nor CHO-MSR cells (data not shown). This suggests that PI3 kinase is mainly coupled with an insulin-inducible type of endocytic uptake of ligands (Fig. 8), but not with a constitutive type. Further studies will help elucidate the mechanism by which insulin receptor signaling may enhance MSR-mediated endocytosis of ligands (Fig. 8). Insulin receptors are also known to form endosomes to be internalized on binding of insulin, which is inhibited by mutant insulin receptors or PI3 kinase inhibitors. So it still remains to be solved whether MSR utilize the HIR vesicles or insulin signal induce MSR vesicles trafficking independently of HIR vesicles. FCS. After serum starvation for 5 h, the cells were pretreated for 30 min with 10 nM wortmannin, 100 mM LY294002, or 20 ng/ml rapamycin, and 0.1% dimethyl sulfoxide as a vehicle, followed by incubation for 60 min with 2 g/ml 125 I-AGE-BSA, with or without 10 nM insulin. The specific cell association of 125 I-AGE-BSA (f) and specific degradation (o) were obtained after correcting for nonspecific cell association and degradation (A). Experiments were performed in parallel in the same way except that 125 I-AGE-BSA was replaced with 2 g/ml 125 I-acetyl-LDL in the absence or presence of 50-fold excess of acetyl-LDL (see above and "Experimental Procedures") (B). RAW-HIR cells were cultured in RPMI containing 10% FCS, and the assay was performed in the same way (see above and "Experimental Procedures") (C). Values of specific cell-association (f) and degradation A certain degree of insulin resistance may be explained by dysfunction of the insulin receptor signaling pathway. Cells transfected with mutant HIR, such as CHO-MSR-HIR 1030M or CHO-MSR-HIR 972F cells did not show insulin-enhanced endocytic uptake of AGE-BSA (Fig. 5). It is therefore possible that insulin resistance state in non-insulin-dependent diabetes mellitus could influence MSR function to some extent through its impaired-insulin signaling. Under such conditions, the insulininducible elimination of circulating AGE proteins by LECs in vivo may be reduced or impaired.
It can be hypothesized that atherosclerosis might develop as a consequence of a disability of the hepatic scavenger receptor to eliminate atherosclerotic substances from circulation. If the generation of ligands for the scavenger receptor(s) of LECs and Kupffer cells exceeds the clearance capacity of these cells, or if their endocytic activities are modulated, for example, by insulin resistance, atherogenic molecules might escape hepatic sequestration and reach cells of extra hepatic vessels, causing the development of atherosclerosis.
Recent immunological studies using anti-AGE antibodies demonstrated positive immune reactions in atherosclerotic lesions in human coronary arteries (10) and aorta (4) and aorta in streptozotocin-induced diabetic rats (59). AGE accumulations were found extracellularly as well as intracellularly in monocyte/macrophage-derived foam cells during the early stages of atherosclerosis as well as smooth muscle cell-derived foam cells in the advanced stages of atherosclerosis (4). This suggests that AGE proteins either infiltrate from the blood into the intimal layer or are formed in the subendothelial layer of the vascular wall. Speaking of chronic reaction, AGE proteins of subendothelial space may be endocytosed first by monocyte-derived macrophages and later by smooth muscle cells that migrate from the medial layer. In addition, AGE accumulation was also demonstrated in microvascular lesions, such as the nodular lesions of diabetic glomeruli (1), and intravenous administration of AGE proteins in normal rats induced focal glomerulosclerosis and albuminuria (60). Considered together, these findings suggest that AGE may play an active role in both macrovascular and microvascular diabetic complications either by inducing cellular responses (3,8,9) or by AGE accumulation in the tissues or cells (1,4,(5)(6)(7)10) following endocytosis of AGE proteins by cells of extra hepatic vessels mediated by AGE receptors (9,12,61).
In conclusion, although normalization of high glucose levels by insulin may help decrease AGE formation and subsequent accumulation in extrahepatic vessels, the present results obtained from in vitro cellular experiments suggest that insulin is directly involved in LECs-mediated elimination of plasma AGE proteins by enhancing MSR function of these cells.