Scavenger Receptor Class B Type I-mediated Reverse Cholesterol Transport Is Inhibited by Advanced Glycation End Products*

Cellular interactions of advanced glycation end products (AGE) are mediated by AGE receptors. We demonstrated previously that class A scavenger receptor types I and II (SR-A) and CD36, a member of class B scavenger receptor family, serve as the AGE receptors. In this study, we investigated whether scavenger receptor class B type I (SR-BI), another receptor belonging to class B scavenger receptor family, was also an AGE receptor. We used Chinese hamster ovary (CHO) cells overex-pressed hamster SR-BI (CHO-SR-BI cells). 125 I-AGE-bo-vine serum albumin (AGE-BSA) was endocytosed in a dose-dependent fashion and underwent lysosomal degradation by CHO-SR-BI cells. 125 I-AGE-BSA exhibited saturable binding to CHO-SR-BI cells ( K d 5 8.3 m g/ml). Endocytic uptake of 125 I-AGE-BSA by CHO-SR-BI cells was completely inhibited by oxidized low density lipoprotein (LDL) and acetylated for LDL and

In the Maillard reaction, proteins react with glucose to form Schiff base and Amadori products. After long term incubation, these early products are converted to advanced glycation end products (AGE), 1 which are characterized physicochemically by fluorescence, brown color, and intra-or inter-molecular crosslinking (1,2), and biologically by specific recognition by AGE receptors. The presence of AGE in several human tissues suggests that they may be involved in the aging process, diabetic complications, and atherosclerosis (3)(4)(5)(6)(7)(8)(9)(10)(11).
The physiological significance of AGE has been analyzed primarily using AGE structure(s) expressed in vivo and AGEbinding proteins or AGE receptors, through which AGE are believed to elicit several biological phenomena in monocytes/ macrophages (12)(13)(14)(15)(16)(17), endothelial cells (18,19), and mesangial cells (20,21). Several AGE receptors have been characterized (22)(23)(24)(25), one of which is a novel 35-kDa protein (called RAGE) from bovine lung endothelium that belongs to the immunoglobulin superfamily (23). Two AGE-binding proteins of 60-and 90-kDa (called p60 and p90) were also identified from the rat liver (24).Recently, galectin-3, a lectin-like protein with a high binding affinity for galactose-containing glycoproteins, was identified as a component of p90 (25). We have recently shown that the macrophage scavenger receptor class AI/AII (SR-A), which is known as a receptor for oxidized low density lipoprotein (Ox-LDL) (26), mediates the endocytic uptake and degradation of AGE-BSA by macrophages (26,27).
In contrast, although functionally related to SR-A, the class B receptors differ significantly in structure. CD36, the defining member of this class, binds Ox-LDL, fatty acids, and the proteins collagen and thrombospondin (28 -32). CD36 has a broad ligand specificity, and its multiple potential roles have been proposed. We recently discovered that the class B scavenger family member CD36 also serves as a receptor for AGE-BSA (33). Our study provided novel information that in addition to SR-A the class B scavenger receptor family seems to serve as AGE receptors in vivo and therefore might participate in the pathogenesis of diabetic macrovascular complications (33).
The scavenger receptor class B type I (SR-BI), a member of the class B scavenger receptor family, was first identified as a high density lipoprotein (HDL) receptor and was shown to mediate selective uptake of cholesteryl esters from HDL (HDL-CE) in vitro (34). A recent study demonstrated that SR-BImediated selective uptake of HDL-CE was much more efficient than CD36-mediated uptake, despite the high similarity in structure of SR-BI and CD36 (35). Immunohistochemical anal-yses in rodents revealed abundant SR-BI expression in the liver and steroidogenic tissues such as the adrenal glands and ovaries (34), where a selective uptake of HDL-CE is known to predominate (36 -39). Adenovirus-mediated overexpression of SR-BI in mouse liver is associated with reduced plasma HDL levels and increased cholesterol secretion into bile (40). Recent studies (40 -43) provided further solid evidence for a crucial role of SR-BI in selective uptake of HDL-CE by hepatocytes in vitro and in vivo.
In the present study, we investigated whether AGE-proteins could be recognized as ligands by SR-BI. Our results suggest that AGE proteins generated in vivo could affect SR-BI-mediated cholesterol metabolism. Ligand Preparation and Iodination-AGE-BSA was prepared as described previously (33). Briefly, 2.0 g of BSA was dissolved in 10 ml of 0.5 M sodium phosphate buffer (pH 7.4) with 3.0 g of D-glucose. Each sample was sterilized by ultrafiltration, incubated at 37°C for 40 weeks, and dialyzed against phosphate-buffered saline (pH 7.4) (PBS). The extent of lysine modification was 71% for AGE-BSA. The N ⑀ -(carboxymethyl) lysine content of AGE-BSA was 7.7 mol/mol of BSA (33). Nonglycated BSA was prepared as described previously (44). Briefly, 1.0 g of BSA was incubated with phenyl boronic acid resin (PBA-60, Amicon, Beverly, MA) in 800 ml of 0.5 M glycine/NaOH buffer containing 2% MgCl 2 (pH 8.5) for 2 h at room temperature. The filtrate through a glass filter was refiltered with another portion of the same resin. The final solution was dialyzed against PBS and used for the experiments. Human LDL (d ϭ 1.019 -1.063 g/ml) (33) and HDL (d ϭ 1.063-1.21 g/ml) (45) were isolated by sequential ultracentrifugation of human plasma from normolipidemic subjects after overnight fasting and dialyzed against 0.15 M NaCl and 1 mM EDTA (pH 7.4) (33). Traces of apoB and -E were removed from HDL by a heparin-agarose column. Acetylated LDL (acetyl-LDL) was prepared by chemical modification of LDL with acetic anhydride as described previously (46). To prepare Ox-LDL, LDL was dialyzed against PBS to remove EDTA. LDL (0.1 mg/ml) was then incubated for 16 h at 37°C with 5 M CuSO 4 , followed by addition of 1 mM EDTA and cooling (46). Electrophoretic mobility of AGE-BSA preparations toward the anode was 1.4 times higher than that of unmodified BSA. Under identical conditions, Ox-LDL showed a similar increase in electrophoretic mobility. Thus, modification of BSA with glucose is associated with a significant increase in net negative charge (33). AGE-BSA was labeled with 125 I by IODO-GEN (Pierce), and HDL was labeled as described previously (33) to a specific radioactivity of 850 and 500 cpm/ng, respectively.
Cell Culture and Isolation of a Transfected Cell Line-CHO-K1 cells were maintained at 37°C in medium A (Ham's F-12 supplemented with 100 units/ml penicillin and 100 units/ml streptomycin) containing 10% fetal calf serum (FCS). CHO-SR-BI cells (47) stably expressing hamster SR-BI were maintained in medium A supplemented with 50 g/ml G418 (medium B). CHO-mock cells were prepared by stably transfecting an expression vector pCR3 (Invitrogen) without a cDNA insert (empty control vector) into CHO-K1 cells. HepG2 cells were obtained from RIKEN cell bank (Japan) and maintained in DMEM containing 10% FCS supplemented with 100 units/ml penicillin and 100 units/ml streptomycin (medium C).
Immunoblotting-Rat liver parenchymal cells, as a positive control, were obtained from male Wistar rats. Isolation of liver parenchymal cells from rats was essentially similar to the method devised previously for preparation of rat liver cells (48). The portal vein was cannulated with a polyethylene catheter, and the liver was perfused with calciumfree Gey's balanced salt solution (GBSS) buffer (pH 7.5) for 10 min at a flow of 8 ml/min and then with GBSS buffer (pH 7.5) containing 0.05% collagenase and 4 mM CaCl 2 for 3 min at 37°C. The liver was minced and suspended in ice-cold PBS. The cell suspension was centrifuged at 100 ϫ g for 2 min at 4°C. The precipitate was suspended in ice-cold PBS, followed by centrifugation. The resulting precipitate was solubi-lized in lysis buffer (20 mM Tris, 150 mM NaCl (pH 7.4) (TBS) containing 2% Triton X-100, 1 mM phenylmethanesulfonyl fluoride, 1 g/ml leupeptin, 1 g/ml pepstatin A, and 1 g/ml aprotinin). Samples (30 g of protein/lane) were subjected to SDS-polyacrylamide gel electrophoresis (49) and then electrotransferred onto a nitrocellulose membrane (Schleicher & Schuell). The membrane was blocked for 2 h with 5% nonfat dry milk in TBS/Tween (20 mM Tris, 150 mM NaCl, and 0.1% Tween 20 (pH 7.4)) and washed with TBS/Tween three times. The membrane was then incubated for 1 h with a 1:1,000 dilution of anti-SR-BI antibody in 1.5% skim milk in TBS/Tween and then washed three times with TBS/Tween. Immunoreactive bands were detected by incubation for 30 min with a 1:5,000 dilution of horseradish peroxidaseconjugated goat secondary anti-rabbit IgG (H ϩ L) (Zymed Laboratories Inc., San Francisco, CA) in 1.5% skim milk in TBS/Tween and washed three times with TBS/Tween, followed by addition of ECL reagent.
Cellular Uptake of 125 I-HDL or 125 I-AGE-BSA by CHO-SR-BI Cells-All cellular experiments, except for the binding study, were performed at 37°C in a humidified atmosphere of 5% CO 2 in air. CHO-SR-BI cells and CHO-mock cells were cultured in medium B. Cells (8 ϫ 10 4 ) were seeded in a 24-well plate and cultured for 2 days in 1.0 ml of medium B, which was then replaced by DMEM containing 3% BSA (medium D). After culture for 1 h, each well received 0.5 ml of medium D containing various concentrations of 125 I-HDL or 125 I-AGE-BSA in the presence or absence of 20-fold excess amounts of the unlabeled ligand to be tested. After incubation for the indicated times, medium was taken from each well, and soluble radioactivity in trichloroacetic acid was determined as an index of cellular degradation, as described previously (33). After the cells were washed three times with 1 ml of medium D and then three more times with PBS, they were lysed with 1 ml of 0.1 N NaOH for 30 min at 37°C. After incubation, 125 I radioactivity of 0.1 N NaOH-soluble protein was determined (as the cell association of 125 I-HDL or 125 I-AGE-BSA), and cellular proteins were determined using BCA protein assay reagent (Bio-Rad). For the binding study, cells that had been seeded in each well as described above were incubated for 90 min at 4°C in 0.5 ml of medium D with various concentrations of 125 I-HDL or 125 I-AGE-BSA in the presence or absence of 20-fold excess amounts of the unlabeled ligands. Each well was washed with ice-cold PBS containing 3% BSA and PBS. The cells were lysed, and the cell-bound radioactivity and cellular proteins were determined as described above. Selective Uptake of HDL-CE by CHO-SR-BI Cells and HepG2 Cells-Cells (8 ϫ 10 4 ) were seeded in a 24-well plate and cultured for 2 days in 1.0 ml of medium B (for CHO-SR-BI and CHO-mock cells) or medium C (for HepG2 cells), which was then replaced by medium D. After culture for 1 h, each well received 0.5 ml of medium C containing various concentrations of 125 I-HDL-[ 3 H]CE in the presence (nonspecific uptake) or absence (total uptake) of 10-fold excess amounts of the unlabeled ligand to be tested. After incubation for the indicated times, the cells were washed three times with 1 ml of medium D and then three more times with PBS. The washed cells were lysed with 1 ml of 0.1 N NaOH for 30 min at 37°C. After incubation, 125 I radioactivity of 0.1 N NaOHsoluble protein (representing the cell association of 125 I-HDL) was determined, and cellular proteins were determined as described above. We then determined the 3 H radioactivity of 0.1 N NaOH-soluble protein (representing [ 3 H]CE uptake). HDL-CE-selective uptake was calculated by subtracting the amount of cell-associated HDL ( 125 I) from the amount of CE uptake ( 3 H) (51).

Preparation of 125 I-HDL-[ 3 H]Cholesteryl Oleoyl Ether-Human
[ 3  Statistical Analysis-All data were expressed as mean Ϯ S.D. Differences between groups were examined for statistical significance using the Student's t test. A p value less than 0.05 denoted the presence of a statistically significant difference.

Expression of SR-BI in CHO Cells Overexpressing Hamster SR-BI-CHO-SR-BI cells that stably expressed hamster SR-BI
were established. Immunoblotting with a polyclonal anti-SR-BI antibody yielded no bands in CHO-mock cells, whereas CHO-SR-BI cells exhibited a distinctive band. Parallel control experiments demonstrated that rat liver parenchymal cells expressing high levels of SR-BI exhibited a positive band with molecular weight indistinguishable from that obtained from CHO-SR-BI cells as reported previously (47) (Fig. 1). We further examined whether SR-BI expressed on the cells was functional; CHO-SR-BI cells exhibited cellular binding of 125 I-HDL with high affinity (K d ϭ 6.5 g/ml, B max ϭ 72.3 g/mg cell protein) but no significant capacity for endocytic uptake of 125 I-HDL (data not shown), as reported previously (34). Selective uptake of HDL-CE by CHO-SR-BI cells was 4-fold higher, at least, than that of CHO-mock cells (data not shown), as reported previously (47). These results indicated that SR-BI expressed on CHO-SR-BI cells functioned as the HDL receptor.
Endocytic Uptake and Degradation of 125 I-AGE-BSA by CHO-SR-BI Cells-We determined the endocytic uptake and degradation of 125 I-AGE-BSA by CHO-SR-BI cells at 37°C. The amount of cell association of 125 I-AGE-BSA with CHO-SR-BI cells increased in a dose-dependent manner and was almost competed away by a 20-fold excess of unlabeled AGE-BSA ( Fig.  2A). The specific cell association exhibited a dose-dependent saturation pattern with a plateau level of 640 ng/mg cell protein and an apparent K d ϭ 2.3 g/ml, and a maximal ligand association of 738 ng/mg cell protein ( Fig. 2A). Parallel experiments in CHO-mock cells revealed that the cell association of 125 I-AGE-BSA occurred at a negligible level ( Fig. 2A). In sharp contrast with 125 I-HDL (data not shown), endocytic degradation of 125 I-AGE-BSA by the transfected cells was significant (Fig. 2B). Specific degradation increased dose-dependently with a plateau level of 230 ng/mg cell protein; the apparent K d for degradation was 3.1 g/ml, and the maximal ligand degradation was 297 ng/mg cell protein, whereas CHO-mock cells did not degrade 125 I-AGE-BSA under the same conditions (Fig.  2B). This result indicates that SR-BI mediates the cell association of AGE-BSA and its endocytic degradation.
Binding of 125 I-AGE-BSA to CHO-SR-BI Cells-We next determined the cellular binding of 125 I-AGE-BSA to CHO-SR-BI cells at 4°C. Total binding of 125 I-AGE-BSA was inhibited by 70% by an excess amount of the same ligand. The specific binding, obtained by subtracting nonspecific binding from the total binding, yielded a saturation pattern. Scatchard analysis of the specific binding disclosed a binding site with an apparent K d of 8.3 g/ml and maximal binding of 85.7 ng/mg cell protein, indicating that CHO-SR-BI cells possess a high affinity binding site for AGE-BSA (Fig. 3). Since the parallel binding experiment of 125 I-AGE-BSA to CHO-mock cells failed to yield specific binding (data not shown), it is likely that the specific binding site for AGE-BSA on CHO-SR-BI cells is identical to SR-BI.
Effect of Modified LDL on Cellular Binding of 125 I-AGE-BSA to CHO-SR-BI Cells-To examine whether the recognition site for AGE-BSA in CHO-SR-BI cells was identical to that for Ox-LDL, acetyl-LDL, LDL, and HDL, we determined the effect of these lipoproteins on cellular binding of 125 I-AGE-BSA to CHO-SR-BI cells. Cellular binding of 125 I-AGE-BSA to CHO-SR-BI cells was effectively (Ͼ62%) replaced by unlabeled Ox-LDL and acetyl-LDL and by unlabeled AGE-BSA, whereas unlabeled LDL and HDL had a slightly weaker effect (Ͻ20%) (Fig. 4A). Non-glycated BSA, a negative control, had no effect on this process (Fig. 4A). Parallel experiments demonstrated that cellular binding of 125 I-HDL to CHO-SR-BI cells was effectively (60%) replaced by unlabeled HDL, whereas AGE-BSA had little effect (Ͻ15%) (Fig. 4B). These results suggest that the binding site of SR-BI for AGE-BSA might overlap with that for Ox-LDL and acetyl-LDL but not that for LDL and HDL.
Effect of Modified LDL on Endocytic Uptake of 125 I-AGE-BSA by CHO-SR-BI Cells-We determined the effect of Ox-LDL, acetyl-LDL, LDL, and HDL on endocytic uptake of 125 I-AGE-BSA by CHO-SR-BI cells. The cell association of 125 I-AGE-BSA was effectively (Ͼ85%) replaced by unlabeled Ox-LDL and acetyl-LDL and by unlabeled AGE-BSA, whereas the effect of unlabeled LDL and HDL was almost negligible (Ͻ20%), and nonglycated BSA, a negative control, had no effect on this process (Fig. 5A). Similarly, the endocytic degradation of 125 I-AGE-BSA was effectively inhibited (Ͼ85%) by the presence of unlabeled Ox-LDL, acetyl-LDL, and unlabeled AGE-BSA, whereas unlabeled LDL and HDL also exerted a much weaker effect (Ͻ25%) (Fig. 5B). The results of endocytic experiments are consistent with those of the binding experiments (Fig. 4).
Effect of AGE-BSA on Selective Uptake of HDL-CE by CHO-SR-BI Cells-We next examined the effects of AGE-BSA on the selective uptake of HDL-CE by CHO-SR-BI cells. AGE-BSA inhibited the selective uptake of HDL-CE by CHO-SR-BI cells in a dose-dependent manner, whereas non-glycated BSA had no effect (Fig. 6). SR-BI-mediated selective uptake of HDL-CE, which was obtained by subtracting the amount of selective uptake by CHO-mock cells from that by CHO-SR-BI cells, was inhibited completely by 100 g/ml AGE-BSA (Fig. 6). This inhibitory effect of AGE-BSA (IC 50 Ͻ10 g/ml) was more effective than that of unlabeled HDL (IC 50 ϭ 60 g/ml) (Fig. 6). AGE-BSA was ineffective in suppressing selective uptake by CHO-mock cells (data not shown). These data suggest that SR-BI-mediated selective uptake of HDL-CE is sensitively inhibited by AGE-BSA.
Effect of AGE-BSA on Selective Uptake of HDL-CE by HepG2 Cells-To investigate this notion further, we examined the effect of AGE-BSA on selective uptake of HDL-CE by human hepatocarcinoma cells, HepG2, model cells for human hepatocytes; the expression of human SR-BI (CLA-1) by these cells was confirmed by immunoblotting (data not shown). Selective uptake of HDL-CE by HepG2 cells was inhibited by HDL by 70% in a dose-dependent manner. The effect of AGE-BSA on this process was partial but significant (ϳ30%), whereas nonglycated BSA had no effect (Fig. 7). When the maximal effect of AGE-BSA is compared with that of HDL, it might follow that AGE-BSA may inhibit the specific selective uptake of HDL-CE by more than 40%. Thus, it is highly likely that selective uptake of HDL-CE by the liver could be inhibited by plasma AGE proteins in vivo, if it is actually present in the circulation as reported previously (53).

Effect of AGE-BSA on [ 3 H]Cholesterol
Efflux from CHO-SR-BI Cells to HDL-Finally, we examined the effect of AGE-BSA on SR-BI-mediated HDL-dependent cholesterol efflux from CHO-SR-BI cells. CHO-SR-BI cells showed a 3-4-fold increase in [ 3 H]cholesterol efflux from these cells to HDL in a time-and dose-dependent manner compared with CHO-mock cells (data not shown), as reported previously (52). Under the present conditions, the time-dependent [ 3 H]cholesterol efflux from CHO-SR-BI cells to HDL (50 g/ml) was inhibited by 40% by AGE-BSA (100 g/ml), whereas non-glycated BSA had no effect (Fig. 8A). Cholesterol efflux from CHO-SR-BI cells to 50 g/ml of HDL was inhibited dose-dependently by AGE-BSA (10 -200 g/ml), whereas AGE-BSA had no effect on cholesterol efflux from CHO-mock cells to HDL (Fig. 8B). Comparison based on the SR-BI-mediated HDL-dependent cholesterol efflux, which was obtained by subtracting the efflux level of CHO-mock cells from that of CHO-SR-BI cells, showed that it was inhibited by AGE-BSA dose-dependently with 50% inhibition at Ͻ30 g/ml of AGE-BSA, whereas non-glycated BSA had no effect (Fig. 8B). Taken together, these results suggest that AGE-BSA inhibits SR-BI-mediated HDL-dependent cholesterol efflux.

DISCUSSION
Our previous studies using CHO cells overexpressing the class A scavenger receptor (SR-A) (27) and peritoneal macrophages from SR-A knockout mice (26) and our recent study using CHO cells overexpressing the class B scavenger receptor CD36 (33) demonstrated that AGE ligands are effectively taken up by these scavenger receptors by a typical receptor-mediated endocytosis mechanism. These findings raised the general hypothesis that the scavenger receptor as the AGE receptor is not confined to the class A but also class B scavenger receptor family (33). We have now demonstrated this to be the case with another class B scavenger receptor, SR-BI, using CHO cells overexpressing hamster SR-BI. The unique function of SR-BI (not found in other scavenger receptors such as SR-A and CD36) is its major role as the HDL receptor in reverse cholesterol transport, from cholesterol efflux from peripheral cells into HDL particles to selective uptake of HDL-CE by the liver. Therefore, the novel findings of our present study were that SR-BI mediated the endocytic degradation of AGE ligands in a manner similar to SR-A and CD36, and also that the selective uptake of HDL-CE and cholesterol efflux mediated uniquely by SR-BI as an HDL receptor was inhibited by AGE ligands. These findings suggest a novel idea in AGE research that AGE proteins generated in situ could regulate SR-BI-mediated cholesterol metabolism in vivo.
SR-BI was first identified as an HDL receptor by Acton et al. (34). Immunohistochemical analyses of SR-BI in rodents showed its major sites of expression to be the liver and in steroidogenic cells such as adrenal glands and ovaries (34), where selective uptake of HDL-CE is significant (36 -39). Al- though belonging to the class B scavenger receptor family, SR-BI is functionally characterized by mediating selective uptake of HDL-CE in vitro (34) and in vivo (40 -43). The present study showed that this phenomenon mediated by CHO-SR-BI cells was effectively (almost completely) inhibited by AGE-BSA (Fig. 6), suggesting that AGE ligands could inhibit selective CE uptake in the liver and steroidogenic tissues. This finding in CHO-SR-BI cells was supported by results obtained from HepG2 cells, model cells for human hepatocytes; selective uptake of HDL-CE by these cells was partially but significantly inhibited by AGE-BSA (Fig. 7). Although immunoblot analysis showed that SR-BI was positively expressed in these cells (data not shown), the extent of contribution of human SR-BI to selective uptake of HDL-CE by HepG2 cells is unknown at present. Rhainds et al. (54) reported that SR-BI and/or CD36 of HepG2 cells are involved, at least in part, in this process. Therefore, the extent of inhibition by AGE-BSA of SR-BI-mediated-CE uptake by HepG2 cells could be lower than estimated. A recent study using transiently expressing cells demonstrated that selective uptake of HDL-CE by SR-BI is of more functional significance than that by CD36 (35). Therefore, it is likely that the extent of inhibition by AGE-BSA of selective uptake of HDL-CE by HepG2 cells could be significant. Thus, it is possible that AGE proteins generated in vivo may inhibit SR-BI-mediated selective uptake of HDL-CE, the final step of reverse cholesterol transport, implicating AGE ligands in regulation of plasma cholesterol levels.
In addition to selective uptake of HDL-CE by SR-BI, this receptor is also involved in cholesterol efflux from peripheral cells to HDL, the first step of reverse cholesterol transport (52). The present study using CHO-SR-BI cells showed that AGE-BSA effectively inhibited [ 3 H]cholesterol efflux from CHO-SR-BI cells to HDL (Fig. 8). Although Hirano et al. (55) reported that human SR-BI is expressed and markedly upregulated in differentiated human macrophages, the lack of in vivo experimental results means that the functional involvement of SR-BI in cholesterol efflux remains uncertain. It has been proposed, however, that SR-BI facilitates bidirectional cholesterol flux depending on the cholesterol gradient between HDL particles and the cell plasma membrane (52,56). There- fore, the present results suggest that AGE proteins inhibit both selective CE uptake and cholesterol efflux. Recent studies using genetically engineered mice, however, have shown that SR-BI plays an important role in the control of biliary cholesterol secretion (40,41). This notion was supported by the fact that SR-BI was expressed on the canalicular membrane of mouse hepatocytes overexpressing SR-BI (40,58). In addition, the immunohistochemical study by Ling et al. (59) using a monoclonal antibody against fluorolink, one of the AGE structures, demonstrated that AGE proteins are localized intracellularly and extracellularly in human hepatocytes. Therefore, it is possible that AGE proteins formed in hepatocytes may inhibit SR-BI-mediated biliary cholesterol secretion in vivo.
SR-BI is known to be a multifunctional molecule that recognizes many ligands of different structures including acetyl-LDL, Ox-LDL, LDL, maleylated BSA (60), HDL (34), VLDL (61), anionic phospholipids (62), and apoptotic cells (63). These findings do not provide a clear picture of the molecular basis for the recognition of these ligands by SR-BI. However, among these ligands, modified LDLs have in common a negatively charged nature, suggesting that SR-BI recognizes the negative charge of AGE proteins. Further studies are necessary to determine the AGE structure(s) required for recognition by SR-BI.
Several attempts were made to characterize the ligand binding domain of SR-BI. Temel et al. (64) prepared anti-SR-BI antiserum raised against the extracellular domain of mouse SR-BI (amino acid residues 174 -356) and demonstrated that it not only inhibited HDL binding and selective lipid uptake but also inhibited LDL binding and selective lipid uptake (65). Gu et al. (57) reported that a double substitution of arginine for glutamine at positions 402 and 418 (Q402R/Q418R) of mouse SR-BI led to a loss of ability to bind HDL but not LDL. Another mutant of mouse SR-BI with a Met-to-Arg substitution at position 158 (M158R) exhibited diminished binding ability to HDL and LDL, but no change in its binding capacity to acetyl-LDL (57). Furthermore, Acton et al. (60) demonstrated that the binding of 125 I-acetyl-LDL to COS cells transfected with hamster SR-BI was completely inhibited by Ox-LDL. Similarly, Calvo et al. (61) also showed that Ox-LDL and LDL effectively replaced binding of DiI-HDL to Sf9 cells transfected with human SR-BI. Taken together, these results support the notion that the region corresponding to amino acids 174 -356 of the extracellular domain of SR-BI is necessary for its HDL binding and selective CE uptake, and that the binding domain of Ox-LDL and acetyl-LDL to SR-BI might overlap with amino acids 174 -356 of its extracellular domain. In addition, it is likely that amino acids 402 and 418 in the extracellular domain of mouse SR-BI is important for binding HDL but not LDL. In the present study, Ox-LDL and acetyl-LDL completely inhibited endocytic uptake of 125 I-AGE-BSA by CHO-SR-BI cells, indicating that the binding domain of AGE-BSA to SR-BI might overlap with amino acids 174 -356 of its extracellular domain (Figs. 4 and 5). However, the extent of inhibition of HDL and LDL for 125 I-AGE-BSA binding was weak (Figs. 4 and 5), suggesting that the main binding site of AGE-BSA to SR-BI would be within amino acids 174 -356 of its extracellular domain but not its binding domain to HDL and LDL. The most novel finding in the present study was that AGE-BSA completely (Ͼ90%) inhibited selective uptake of HDL-CE by CHO-SR-BI cells (Fig. 6). In addition, AGE-BSA exhibited a significant inhibitory effect on HDL-induced cholesterol efflux from these cells (Fig. 8). It is not clear whether the HDL binding domain of SR-BI is involved in selective CE uptake. Williams and coworkers (66,67) recently proposed that SR-BI may form "a lipid channel" that facilitates transfer of lipid between cells and lipoproteins. Therefore, it is possible that AGE-BSA could inhibit SR-BI-mediated CE uptake by binding to the lipid channel rather than by binding to the HDL binding domain. Further studies are needed to elucidate the binding domain of SR-BI to AGE ligands.
Recently, it has been reported that CD36 and SR-BI are localized in caveolae of the plasma membrane and that Ox-LDL binding to CD36 or HDL binding to SR-BI affects the lipid composition of caveolae, which in turn modulates activation of endothelial nitric-oxide synthase in human microvascular endothelial cells (68). Stitt et al. (69) also reported that the R1 (OST-48), R2 (80K-H), and R3 (galectin-3) components of the AGE receptor complex were localized in caveolae of retinal microvascular endothelial cells. Therefore, it is possible that interaction of AGE ligands with CD36 and SR-BI modulates endothelial nitric-oxide synthase activity in the caveolae. Further studies are required to determine whether CD36 and SR-BI, acting as AGE receptors in caveolae, directly regulate the generation of nitric oxide. Chinetti et al. (70) recently showed that CLA-1/SR-BI expressed on macrophages is induced by ligands for PPAR␥ and PPAR␣ in human atherosclerotic lesions. Recent reverse transcriptase-polymerase chain reaction analyses by Iwashima et al. (71) demonstrated that AGE-BSA induced PPAR␥ expression and activation in cultured mesangial cells, and that SR-BI, SR-A, CD36, and Lox-1 were also up-regulated by AGE-proteins through activation of PPAR␥ in phorbol 12-myristate 13-acetate-treated THP-1 cells (72).
In conclusion, our finding that SR-BI-mediated functions were inhibited by AGE proteins suggests a potential link between SR-BI-mediated cholesterol metabolism and development of diabetic vascular complications.