Comparison of Class B Scavenger Receptors, CD36 and Scavenger Receptor BI (SR-BI), Shows That Both Receptors Mediate High Density Lipoprotein-Cholesteryl Ester Selective Uptake but SR-BI Exhibits a Unique Enhancement of Cholesteryl Ester Uptake*

Scavenger receptor BI (SR-BI) mediates the selective uptake of high density lipoprotein (HDL) cholesteryl ester (CE), a process by which HDL CE is taken into the cell without internalization and degradation of the HDL particle. The biochemical mechanism by which SR-BI mediates the selective uptake of HDL CE is poorly understood. Given that CE transfer will occur to some extent from HDL to protein-free synthetic membranes, one hypothesis is that the role of SR-BI is primarily to tether HDL close to the cell surface to facilitate CE transfer from the particle to the plasma membrane. In the present study, this hypothesis was tested by comparing the selective uptake of HDL CE mediated by mouse SR-BI (mSR-BI) with that mediated by rat CD36 (rCD36), a closely related class B scavenger receptor. Both mSR-BI and rCD36 bind HDL with high affinity, and both receptors mediate HDL CE selective uptake. However, SR-BI mediates selective uptake of HDL CE with a 7-fold greater efficiency than rCD36. HDL CE selective uptake mediated by rCD36 is dependent on HDL binding to the receptor, since a mutation that blocks HDL binding also blocks HDL CE selective uptake. These data lead us to hypothesize that one component of HDL CE selective uptake is the tethering of HDL particles to the cell surface. To explore the molecular domains responsible for the greater efficiency of selective uptake by mSR-BI, we compared binding and selective uptake among mSR-BI, scavenger receptor BII, and various chimeric receptors formed from mSR-BI and rCD36. The results show that the extracellular domain of mSR-BI is essential for efficient HDL CE uptake, but the C-terminal cytoplasmic tail also has a major influence on the selective uptake process.

The risk for developing atherosclerosis and cardiovascular disease is inversely related to plasma concentrations of high density lipoprotein (HDL) 1 cholesterol (1). Although the mechanism of this protective effect remains uncertain, it has been known for some time that HDL plays a pivotal role in the transport of free cholesterol and cholesteryl esters (CE) through the plasma. HDL participates in reverse cholesterol transport (2), a process involving the uptake of free cholesterol from peripheral tissue and its subsequent delivery (as free cholesterol or CE) to steroidogenic (for hormone synthesis) and hepatic (for bile acid synthesis) tissues. HDL provides CE to cells via the selective uptake pathway in which HDL CE is taken into the cell without the internalization and lysosomal degradation of the HDL particle (3)(4)(5)(6)(7)(8). Recent studies identified a cell surface receptor, scavenger receptor BI (SR-BI), which binds HDL particles and mediates the selective uptake of HDL CE in transfected cells (9). Immunochemical analysis of SR-BI in rodents indicates that it is expressed most abundantly in the liver and in steroidogenic cells of the adrenal gland and ovary (9 -11), where the selective uptake of HDL CE is greatest (4,7). SR-BI expression is regulated by gonadotropins and adrenocorticotropic hormone coordinately with the selective uptake of HDL CE and steroidogenesis (10,11). In addition, antibody blocking experiments show that SR-BI is the receptor responsible for the uptake of HDL CE and its delivery to the steroidogenic pathway in adrenocortical cells (12). Inactivation of the SR-BI gene in mice alters plasma HDL metabolism and reduces adrenal gland CE accumulation, results consistent with a major role for SR-BI in cholesterol metabolism in vivo (13). Taken together, these studies indicate that SR-BI is a physiologically relevant receptor for the selective uptake of HDL CE.
The biochemical mechanism by which SR-BI mediates the selective uptake of HDL CE is poorly understood. Given that CE transfer will occur to some extent from HDL or microemulsion particles to protein-free synthetic membranes (14), one hypothesis is that the role of SR-BI is primarily to tether HDL in close apposition to the cell surface to facilitate CE transfer from the particle to the plasma membrane. In the present study, we test this hypothesis by comparing the selective uptake of HDL CE mediated by mouse SR-BI (mSR-BI) with that mediated by rat CD36 (rCD36), a closely related receptor that also binds HDL with high affinity (15). SR-BI was originally defined as a class B scavenger receptor (16 -18) in a family that includes CD36, LIMPII, and SR-BII, a form of SR-BI with an alternate C-terminal cytoplasmic tail (19). The amino acid sequence homology between SR-BI and CD36 polypeptides has been reported to be 20 -33%; however, when the amino acid sequences are properly aligned, the proteins are strikingly similar. In fact, the predicted secondary structures and sizes of SR-BI and CD36 are nearly identical. Each protein contains two transmembrane and two cytoplasmic domains (the aminoand carboxyl-terminal domains) as well as a large extracellular domain containing a cysteine-rich region and nine putative sites for N-linked glycosylation. In addition, both proteins are palmitoylated and localized to caveolae (20 -22). The results of the present study comparing mSR-BI with rCD36 suggest that one component of HDL CE selective uptake is due to particle tethering. However, SR-BI-mediated selective uptake is a much more efficient process that combines tethering of the HDL particle to the cell surface and facilitated HDL CE movement into the cell. Furthermore, comparison of SR-BI, SR-BII, CD36, and chimeric receptors showed that the extracellular domain of mSR-BI is essential for efficient HDL CE uptake, but the C-terminal cytoplasmic tail also has a major influence on the selective uptake process.

EXPERIMENTAL PROCEDURES
Plasmids and Sequencing-PCR amplifications were performed using a Perkin-Elmer DNA Thermal Cycler model 480 or 9700. Oligonucleotides were either purchased from Integrated DNA Technologies or synthesized on a Beckman Oligo 1000 DNA synthesizer.
All plasmids were prepared using Endotoxin-free Qiagen Maxi-prep kits and sequenced throughout the coding region to confirm the correct fragment insertion and to ensure that no point mutations had been generated during the amplification process. DNA sequencing was performed by the automated sequencing facility at SUNY Stony Brook. Reactions were prepared using a dye termination cycle sequencing kit and analyzed on an Applied Biosystems model 373 DNA (Perkin-Elmer Applied Biosystems).
Immunoblots with several antibodies to different regions of SR-BI and CD36 were performed on lysates from COS-7 cells expressing chimeric receptors. As expected, these experiments confirmed the presence of each of the described regions of either mSR-BI or rCD36 in each of the expressed chimeras (data not shown).
Transient Transfection of COS-7 Cells-COS-7 cells were maintained in Dulbecco's modified Eagle's medium, 10% calf serum, 2 mM L-glutamine, 50 units/ml penicillin, 50 g/ml streptomycin, and 1 mM sodium pyruvate. Cells were seeded at a density of 2.0 ϫ 10 6 in 10-cm tissue culture dishes in 10 ml of fresh media. The cells were then incubated until approximately 80 -90% confluence (ϳ18 h). Transfections were performed with Fugene 6, a liposomal-like transfection reagent from Boehringer Mannheim, as directed by the manufacturer. The following day, the cells were trypsinized, resuspended in a total volume of 6 ml with fresh medium, and added to six wells of a six-well plate. After 18 -24 h, the cells were assayed for their ability to bind HDL particles and mediate the uptake of CE.
HDL Cell Association, Selective CE Uptake, and Apolipoprotein Degradation-HDL assays were performed ϳ48 h post-transfection. Cells were washed once with serum-free medium, 0.5% BSA. 125 I-Dilactitol tyramine-3 H-cholesterol oleolyl ether hHDL 3 particles were added at a concentration of 10 g protein/ml (unless otherwise indicated) in serum-free medium. After incubation for 1.5 h, the HDL-containing media was removed, and the cells were washed three times with 0.1% BSA in PBS (pH 7.4) and one time with PBS (pH 7.4). The cells were lysed with 1.1 ml of 0.1 N NaOH and pipetted 18 -20 times to fragment DNA. The lysate was then processed to determine trichoroacetic acid soluble and insoluble 125 I radioactivity and organic solvent-extractable 3 H radioactivity as described (25,26). The values for the cell-associated HDL apolipoprotein, the endocytosed and degraded HDL apolipoprotein, the total cell-associated HDL CE, and the selective uptake of HDL CE were obtained as described previously (25,26). Note that cell-associated HDL CE is derived from acid-insoluble 125 I radioactivity values and is believed to be primarily cell surface-bound, although it may include a small fraction of HDL particles that have been endocytosed but not yet degraded. In our experiments, endocytosed and degraded HDL is typically about 5% of cell-associated HDL. Thus, error due to endocytosed but not yet degraded HDL would be minimal.
Receptor Expression Levels-COS-7 cells (in 35-mm wells) were washed with 2 ml of cold PBS. Cells were removed from plates by the addition of 1 ml PBS, 0.5 mm EDTA and incubation for 5-7 min at room temperature. Cells were placed in a microcentrifuge tube and centrifuged at 200 ϫ g for 2-3 min and were resuspended in 100 l of PBS, 1% BSA. Anti-SR-BI primary antibody 356 (12) at a concentration of 0.48 mg/ml IgG or anti-CD36 39815 antiserum (27) at a 1:50 dilution was added to the cells and incubated for 1 h at 4°C. The cells were centrifuged at 200 ϫ g for 2-3 min, and the supernatant was aspirated. Cells were washed with 0.5 ml of PBS, 1% BSA before incubation with secondary antibody (4 l of fluorescein (Sigma) or phycoerythrin-conjugated anti-rabbit antibody (Molecular Probes, Inc., Eugene, OR) in 300 l of PBS, 1% BSA) for 30 min at 4°C. Cells were washed three times with 0.5 ml of PBS, 1% BSA and fixed with 0.5 ml of 1% formaldehyde in PBS, 1% BSA for 15 min at 4°C with gently shaking. Following incubation with fixative, the cells were centrifuged for 2-3 min and resuspended in 0.5 ml PBS, 1% BSA. Fluorescence intensities were measured using a Becton Dickinson FACSAdvantage cell sorter or FACScan flow cytometer.

The Role of HDL Tethering in HDL CE Selective Uptake-
The high degree of similarity between mSR-BI and rCD36 led us to ask whether rCD36 was capable of binding HDL particles and mediating the selective uptake of HDL CE. We first tested for HDL binding to rCD36 by transiently transfecting COS-7 cells with a vector that could drive high level receptor expres-sion. Like rCD36, the mSR-BI coding region was cloned into the pSG5 vector under the transcriptional control of the SV40 promoter. pSG5(mSR-BI), pSG5(rCD36), or vector alone were transfected into COS-7 cells using a liposomal mediated protocol that gave high transfection efficiencies. Cell surface receptor levels were monitored by flow cytometry using polyclonal antibodies to the extracellular domains of mSR-BI (12) and rCD36 (27). This analysis showed that 20 -60% of the cells expressed the receptors (data not shown). Western blot analysis also showed strong expression of both receptors. mSR-BI transiently expressed in COS-7 cells co-migrated with mSR-BI stably expressed in the ldlA[mSR-BI] cell (data not shown). Similarly, rCD36 transiently transfected in COS-7 cells comigrated with rCD36 stably expressed in Ob17PY fibroblasts (23) (data not shown). Fig. 1A shows that in comparison to vector-transfected cells, mSR-BI-expressing COS-7 cells bound HDL in a high affinity saturable manner. Nonlinear regression and Scatchard analysis of these data indicated an apparent K d of 10 g/ml HDL protein. This value is similar to that previously reported for HDL binding to mSR-BI in murine Y1-BS1 adrenocortical cells that naturally express SR-BI (12). Similar results were obtained with rCD36-expressing cells (Fig. 1B), which showed an apparent K d of 22 g/ml HDL protein. Thus, rCD36, like mSR-BI recognized HDL saturably and with high affinity in transiently transfected COS-7 cells.
In order to compare the capacities for HDL CE selective uptake, cells expressing mSR-BI or rCD36 or cells transfected with vector alone were incubated for 1.5 h with 10 g/ml double-labeled ( 125 I-apolipoprotein/ 3 H-cholesteryl oleolyl ether) human HDL 3 , after which HDL cell association ( Fig. 2A) and selective uptake of HDL CE (Fig. 2B) were determined. In contrast to the efficient binding of HDL by rCD36 relative to mSR-BI, rCD36 mediated a considerably reduced level of HDL CE selective uptake. In order to compare the relative efficiencies of mSR-BI and rCD36 for selective uptake, the contribution from vector-transfected cells was subtracted, and the amount of selective uptake was expressed relative to the amount of cell-associated HDL (Fig. 2C). In this way, HDL CE selective uptake is normalized to the quantity of HDL particles bound to the cell surface. This analysis showed that mSR-BI mediated HDL CE selective uptake with an approximately 7-fold greater efficiency than rCD36.
The reduced efficiency of HDL CE selective uptake with rCD36 suggested that this level of selective uptake might be attributable only to tethering of HDL particles on the cell surface. Alternatively, expression of rCD36 might indirectly alter HDL CE selective uptake by perturbing plasma membrane structure or lipid domains to facilitate the transfer of HDL CE to the plasma membrane independently of receptor-HDL particle binding. To test whether rCD36-mediated HDL CE selective uptake is due to HDL binding, we mutated the extracellular domain of rCD36 to disrupt HDL binding by inserting the 14-amino acid monoclonal M45 epitope within the N-terminal region of the extracellular domain (rCD36-X). Flow cytometry of cells expressing rCD36 or rCD36-X showed that both receptors were expressed on the cell surface to similar extents (data not shown). Comparison of HDL cell association (Fig. 3A) and HDL CE selective uptake (Fig. 3B) showed that the epitope tag reduced HDL binding and HDL CE selective uptake to the level seen with vector-transfected cells. This result indicates that the HDL CE selective uptake mediated by rCD36 requires cell surface HDL particle binding. These data support the "tethering hypothesis" by showing that as long as rCD36 can bind HDL particles close to the cell surface of COS-7 cells, HDL CE selective uptake will occur, albeit at a reduced efficiency compared with SR-BI.
Role of the C-terminal Cytoplasmic Tail in HDL Binding and HDL CE Selective Uptake-mSR-BII is identical to mSR-BI except that it contains a different C-terminal cytoplasmic tail as a result of alternate splicing of SR-BI pre-mRNA (19,29). The influence of the alternate C-terminal tail on HDL binding and HDL CE selective uptake was determined in transient transfection assays. Little or no difference was seen in the ability to bind HDL particles (Fig. 4A), but COS-7 cells expressing mSR-BII showed a greatly reduced ability to mediate selective uptake (Fig. 4B). When HDL CE selective uptake was expressed on the basis of cell-associated HDL particles (Fig.  4C), mSR-BI-mediated HDL CE selective uptake with an approximately 7-fold greater efficiency than mSR-BII. Therefore, rCD36 and mSR-BII both showed a similar low efficiency of HDL CE selective uptake in comparison with mSR-BI.
The comparison of mSR-BI and mSR-BII suggests that the C-terminal cytoplasmic tail has a significant influence on the selective uptake process. To further test the role of the cytoplasmic domains, chimeric receptors were constructed and expressed in COS-7 cells in comparison with mSR-BI and rCD36. One set of chimeric receptors contained the rCD36 extracellular domain with the C-terminal cytoplasmic tail (T) of mSR-BI with (CD/SRTM) or without (CD/SRT) the C-terminal transmembrane domain (M) of mSR-BI. Another chimeric receptor has the extracellular domain of CD36 with both the N-and C-terminal tails and transmembrane domains of mSR-BI (SR/ CD/SR) (Fig. 5). When these chimeric receptors were expressed in COS-7 cells, each bound HDL nearly as well as native rCD36 (Fig. 6A), and each showed a similar level of HDL CE selective uptake as compared with rCD36 (Fig. 6B). Comparison of the selective uptake efficiencies of these chimeras (Fig. 6C) showed that they were no different than CD36. Thus, the cytoplasmic tails and transmembrane domains of mSR-BI were not sufficient to confer efficient HDL CE selective uptake activity on the rCD36 extracellular domain despite its ability to bind HDL particles.
To determine whether the C-terminal cytoplasmic tail of mSR-BI is unique in facilitating HDL CE selective uptake in the context of SR-BI, the C-terminal tail of SR-BI was replaced with that of rCD36 (Fig. 5, SR/CDT). When expressed in COS-7 cells, this chimera bound HDL (Fig. 6A) and mediated HDL CE selective uptake (Fig. 6B) similar to native mSR-BI. Comparison of the selective uptake efficiency for SR/CDT (Fig.  6C) showed that the rCD36 C-terminal cytoplasmic tail supported full HDL CE selective uptake activity as seen with native mSR-BI. DISCUSSION Despite extensive studies of the HDL CE selective uptake process over the past 18 years, the molecular mechanism by which hydrophobic cholesteryl esters are transferred from the HDL particle to the plasma membrane has remained elusive. The finding that mSR-BI mediates HDL CE selective uptake provides a molecular link between a specific cell surface receptor and a widespread, but poorly understood, biological process. In the present study, we have examined the mechanism of mSR-BI-mediated HDL CE selective uptake by comparing mSR-BI (and mSR-BII) with the closely related class B scavenger receptor, rCD36, and with chimeric receptors formed from mSR-BI and rCD36. The results lead to three major conclusions.
First, it is clear that the high efficiency of mSR-BI-mediated HDL CE selective uptake is due to more than simply tethering HDL particles on the cell surface. Analysis of HDL binding and selective CE uptake showed that CD36 can mediate HDL CE selective uptake with a reduced efficiency. This component of selective uptake is probably due to tethering HDL particles to the plasma membrane, since disruption of HDL binding to rCD36 also disrupted HDL CE selective uptake (Fig. 3). Furthermore, reduced levels of selective uptake similar to that seen with rCD36 also were seen with mSR-BII and three chimeric receptors containing the extracellular domain of rCD36. Taken together, these results indicate that binding of HDL particles to a class B scavenger receptor is sufficient to produce HDL CE selective uptake, presumably by bringing the HDL particle in apposition to the plasma membrane. This finding is consistent with earlier studies in cell-free systems showing selective transfer of HDL CE to protein-free model membranes in a collision-mediated process (14). Thus, HDL CE transfer to membranes may occur at low levels in the absence of any receptor protein and is accelerated by tethering HDL particles to a class B scavenger receptor.
The enhancement of HDL CE selective uptake that is specific to mSR-BI, however, exceeds this tethering contribution by a factor of 7-fold. The mechanism by which mSR-BI facilitates HDL CE transfer is presently unclear, but it is interesting to note that the Arrhenius activation energy for mSR-BI-dependent HDL CE selective uptake in Y1-BS1 adrenocortical cells is very low (ϳ9 kcal/mole). 2 This suggests that CE molecules must move through a nonaqueous pathway or channel from the HDL particle to the plasma membrane. The high efficiency of mSR-BI-mediated selective uptake may be due to the formation of a nonaqueous pathway to permit HDL CE molecules to move efficiently into the plasma membrane. We hypothesize from these data that mSR-BI-mediated HDL CE selective uptake occurs through two steps: a generalized HDL tethering component that is shared with other class B scavenger receptors and an active facilitation of HDL CE uptake that is unique to mSR-BI. We note that it is formally possible that rCD36 also exhibits the facilitation of HDL CE uptake but does so at much reduced efficiency. However, the quantitative equivalence of the HDL CE selective uptake activity of rCD36, mSR-BII, and three chimeric receptors is more likely the result of a clear dissociation between HDL tethering and the facilitation of HDL CE uptake.
Second, comparison of mSR-BI and mSR-BII showed that mSR-BII mediated HDL CE selective uptake with a low efficiency similar to that seen with rCD36. Webb et al. (29) also noted that a stable CHO cell line expressing mSR-BII mediated HDL CE selective uptake to a lesser extent than a cell line expressing mSR-BI. Since mSR-BII and mSR-BI differ only in the C-terminal cytoplasmic tail, this result suggests that the tail is important for high efficiency HDL CE selective uptake as mediated by mSR-BI. The C-terminal tail of mSR-BI might be responsible for targeting the receptor to a plasma membrane domain or for interactions with cytoplasmic proteins that are necessary for selective uptake. With regard to the first possi-bility, mSR-BI has been found in membrane caveolae (22), a location that might have functional consequences for cholesterol flux (31). However, both mSR-BII and CD36 have also been found in caveolae (20,21,29), suggesting that caveolar localization per se may not explain the difference in selective uptake efficiency of mSR-BI versus mSR-BII and rCD36, unless there are uncharacterized differences in localization to lipid domains within caveolar fractions. Another possibility is that the mSR-BI C-terminal tail facilitates interactions with other membrane or cytoplasmic proteins that are necessary for efficient HDL CE uptake. If this is the case, these interactions must not be unique to the mSR-BI tail, since the rCD36 Cterminal tail yields full HDL CE selective uptake activity when swapped for the mSR-BI tail. Another possibility is that the C-terminal tail of mSR-BI has no active role in the selective uptake process but that the C-terminal tail of SR-BII is in some way inhibitory, perhaps by altering the conformation of the extracellular domain. Further studies will be necessary to resolve this point.
Third, the mSR-BI N-and C-terminal tails and transmembrane domains were not sufficient to confer high efficiency HDL CE selective uptake when expressed in a chimeric receptor with the extracellular domain of CD36 (Fig. 6). This result indicates that the extracellular domain of SR-BI does more than simply bind HDL with high affinity and tether the particles close to the plasma membrane. The extracellular domain appears to be required for the marked facilitation of HDL CE uptake that occurs in addition to the component that derives from HDL cell surface binding. One potential role of the extracellular domain is to create a nonaqueous pathway or channel for HDL CE transfer either by homomeric interactions or heteromeric interactions with other membrane proteins. This activity of the mSR-BI extracellular domain might also require the transmembrane domains of mSR-BI, although the transmembrane domains of mSR-BI themselves were not adequate to support high efficiency selective uptake when appended to the extracellular domain of rCD36.
An interesting result in the present study is the finding that rCD36 mediates HDL CE selective uptake although with a much reduced efficiency as compared with mSR-BI. This result raises the question of whether lipoprotein selective lipid uptake mediated by CD36 may be of physiological significance in some circumstances. Recent studies, for example, show that CD36 mediates the entry of lipid activators into macrophage from oxidized LDL. This results in transcriptional activation of PPAR␥ and increased expression of CD36 (32). This positive feedback pathway has been proposed to play a role in macrophage foam cell formation in the vascular wall during atherogenesis. Whether the selective uptake by CD36 of PPAR␥ activators from oxidized LDL contributes to this pathway remains to be determined. Furthermore, selective uptake of HDL CE or LDL-CE by CD36 may contribute to foam cell formation once this receptor has been up-regulated.
In summary, the present studies with mSR-BI, mSR-BII, CD36, and chimeric receptors support the idea that HDL CE selective uptake results from at least two steps. In the first, tethering the HDL particle to the cell surface via a class B scavenger receptor leads to a moderate increase in HDL CE uptake. The mechanism of this CE transfer is not clear, but it presumably reflects an enhancement of the process that can occur with model membranes in cell free systems as a result of bringing the HDL particle close to the plasma membrane. The second step is a marked facilitation of HDL CE selective uptake that is specific to mSR-BI. This facilitation step requires the extracellular domain of mSR-BI but can occur with the Cterminal cytoplasmic tail of either mSR-BI or CD36. We speculate that this mSR-BI-specific step is responsible for the formation of a nonaqueous pathway that permits the movement of the hydrophobic CE down its concentration gradient from the HDL particle to the plasma membrane.